Topics in Mining, Metallurgy and Materials Engineering

Series Editor: Carlos P. Bergmann

Stefano Gialanella
Alessio Malandruccolo

Aerospace
Alloys
Topics in Mining, Metallurgy and Materials
Engineering

Series editor
Carlos P. Bergmann, Federal University of Rio Grande do Sul, Porto Alegre,
Rio Grande do Sul, Brazil
“Topics in Mining, Metallurgy and Materials Engineering” welcomes manuscripts in
these three main focus areas: Extractive Metallurgy/Mineral Technology;
Manufacturing Processes, and Materials Science and Technology. Manuscripts
should present scientific solutions for technological problems. The three focus
areas have a vertically lined multidisciplinarity, starting from mineral assets, their
extraction and processing, their transformation into materials useful for the society,
and their interaction with the environment.

More information about this series at http://www.springer.com/series/11054

Stefano Gialanella • Alessio Malandruccolo

Aerospace Alloys
Stefano Gialanella Alessio Malandruccolo
Industrial Engineering Department Metallurgy Industrial Consultant
University of Trento Bolzano, Italy
Trento, Italy

ISSN 2364-3293 ISSN 2364-3307 (electronic)

Topics in Mining, Metallurgy and Materials Engineering
ISBN 978-3-030-24439-2 ISBN 978-3-030-24440-8 (eBook)
https://doi.org/10.1007/978-3-030-24440-8

© Springer Nature Switzerland AG 2020

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims
in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To my son Giulio Carlo
(SG)
To my friends and to all those who supported
me
(AM)
Preface

The aerospace field is traditionally strategic for the development of novel materials
needed for the specific requirements of aircraft structures and propulsion systems. A
large number of advanced materials and relevant processing routes have been
implemented first for aerospace needs. On many occasions, aerospace materials
and technologies have subsequently been transferred successfully to a wider range
of applications, like chemical industry, energy production, nuclear plants, terrestrial
and marine transportation, and many more.
The aim of this book is to present an overview of the main classes of metallic
materials, like light alloys (Al-, Mg-, and Ti-based, including Ti-aluminides), steels,
superalloys, refractory alloys, oxide dispersion strengthened alloys, ordered inter-
metallics, shape memory alloys, and related systems, e.g., laminate composites and
coatings.
The basic concepts concerning the structure of aircrafts and the working princi-
ples of gas turbine engines are recalled to provide the reader with indications on the
specific needs and possible issues in this field, so that the adoption of given materials
systems and relevant processing routes are better understood.
The content of the book is organized according to the main classes of materials
that are of actual and potential interest for aerospace applications. In each chapter,
materials properties and relevant technological aspects, particularly as concerns
processing, are presented.
The motivations that have guided the research efforts for the development of new
materials systems have changed over the comparatively short life of aeronautics.
Nowadays, energy-saving and environmental issues are paramount: these stand
points and their role in materials development are also highlighted.
A deliberate attempt, pursued in writing Aerospace Alloys, has been to include an
updated bibliography and indication of further readings, both important in a contin-
uously evolving field. A list of general references concerning materials science and
engineering and physical metallurgy is included among the “Further Readings” lists
at the end of each chapter. In this way, the fundamental aspects of these disciplines,

vii
viii Preface

which may contribute to a better understanding of the book content, can easily be
recovered.
We believe the book can be used in advanced undergraduate courses in materials
and mechanical engineering, considering that the starting point for writing this book
has been a collection of lecture notes of courses that SG ran over several years at the
University of Trento and the University of Florence in Italy. However, we would
expect that the subject and relevant materials information may also be of interest to
researchers working in R&D industrial laboratories, considering that in this perspec-
tive, we have included compositional tables, some in the appendices, so that the
textbook turns out be reasonably self-consistent. We believe that the mixed aca-
demic (SG’s) and industrial (AM’s) background may provide original and comple-
mentary stand points on the different subjects, attracting the interest of researchers
and professionals working in the different but still strongly interacting development
laboratories in the academia and industry.
We conclude this preface with the acknowledgments. In the first place, we wish
to thank all the students (AM has been one of them at some stage!), who during the
lecture classes placed questions and contributed to a stimulating discussion, pro-
moting in this way the improvement and correction of the lecture notes. We also
thank our colleagues for stimulating discussions and suggestions on several subject
dealt in the book.

Trento, Italy Stefano Gialanella

Bolzano, Italy Alessio Malandruccolo
Contents

1 A Brief Introduction to Aerospace Applications . . . . . . . . . . . . . . . . 1

1.1 Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Novel Design Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Aerospace-Related Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2 Gas Turbine Aero-Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 Types of Aircraft Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3 Compressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.4 Combustors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.5 Turbines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.6 State of the Art and Future Trends . . . . . . . . . . . . . . . . . . . . . . . 33
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3 Alloys for Aircraft Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.2 Aluminum Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.2.2 Cast Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.2.3 Wrought Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
3.2.4 Aluminum-Lithium Alloys . . . . . . . . . . . . . . . . . . . . . . . 73
3.2.5 Aluminum Alloys in Aircraft Structures: Present State
and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.3 Magnesium Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.3.2 Aerospace Magnesium Alloys: Metallurgy, Main
Designation System, and Tempers . . . . . . . . . . . . . . . . . . 95
3.3.3 Magnesium Alloy Processing . . . . . . . . . . . . . . . . . . . . . 98

ix
x Contents

3.3.4 Development of Aerospace Magnesium Alloys . . . . . . . . 99

3.3.5 Current Aerospace Applications . . . . . . . . . . . . . . . . . . . 106
3.4 Composite Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
3.4.2 Fiber-Metal Laminates: Development and Designation . . . 112
3.4.3 Main Production Steps and Resulting Characteristics . . . . 113
3.4.4 Mechanical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . 118
3.4.5 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
4 Titanium and Titanium Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
4.2 Physical Metallurgy Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4.3 Main Groups of the Titanium Alloys . . . . . . . . . . . . . . . . . . . . . 145
4.3.1 Conventional Titanium Alloys . . . . . . . . . . . . . . . . . . . . 146
4.3.2 Advanced Titanium Alloys . . . . . . . . . . . . . . . . . . . . . . . 164
4.4 Titanium Processing Technologies . . . . . . . . . . . . . . . . . . . . . . . 169
4.5 Defects in Titanium Alloys and Diagnostics . . . . . . . . . . . . . . . . 179
4.6 Main Aerospace Applications of Titanium Alloys . . . . . . . . . . . . 182
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
5 Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
5.2 Iron and Steel Metallurgical Base Concepts . . . . . . . . . . . . . . . . 193
5.2.1 Annealing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
5.2.2 Quenching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
5.2.3 Tempering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
5.2.4 Age Hardening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
5.2.5 Steel Designation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
5.3 Standard and Special Production Techniques . . . . . . . . . . . . . . . 200
5.3.1 Arc Melting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
5.3.2 Vacuum Induction Melting (VIM) . . . . . . . . . . . . . . . . . . 203
5.3.3 Argon Oxygen Decarburization (AOD) . . . . . . . . . . . . . . 203
5.3.4 Vacuum Oxygen Decarburization (VOD and VODC) . . . . 207
5.3.5 Electroslag Remelting (ESR) . . . . . . . . . . . . . . . . . . . . . 209
5.3.6 Vacuum Arc Remelting (VAR) . . . . . . . . . . . . . . . . . . . . 212
5.4 Steels for Aerospace Applications . . . . . . . . . . . . . . . . . . . . . . . 216
5.4.1 Maraging Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
5.4.2 AerMet Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
5.4.3 Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
5.5 Steel-Based Aerospace Components . . . . . . . . . . . . . . . . . . . . . . 252
5.5.1 Bearings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
5.5.2 Brake Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
5.5.3 Gas Turbine Jet Engine Parts . . . . . . . . . . . . . . . . . . . . . 258
5.5.4 Gearboxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
Contents xi

5.5.5 Undercarriage Applications . . . . . . . . . . . . . . . . . . . . . . . 260

5.5.6 Wings and Engine Pylons . . . . . . . . . . . . . . . . . . . . . . . . 260
5.5.7 Aerospace Applications of Steels: An Overview . . . . . . . . 261
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
6 Superalloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
6.2 Cobalt-Based Superalloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
6.2.1 Main Constituent Elements and Relevant Strengthening
Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
6.2.2 Latest Developments of Co Superalloys . . . . . . . . . . . . . . 286
6.3 Iron-Based Superalloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
6.3.1 Strengthening Mechanisms and Classification
of Iron-Based Superalloys . . . . . . . . . . . . . . . . . . . . . . . 290
6.3.2 Alloy 718 and Properties of Iron-Based Superalloys . . . . . 297
6.4 Nickel-Based Superalloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
6.4.1 Relevant Metallurgical Issues . . . . . . . . . . . . . . . . . . . . . 311
6.4.2 Selection Criteria for Alloying Elements
in Nickel-Based Superalloys: The γ Matrix . . . . . . . . . . . 321
6.4.3 Selection Criteria for Alloying Elements
in Nickel-Based Superalloys: The γ0 Precipitates . . . . . . . 331
6.5 Single-Crystal and Directionally Solidified Superalloys . . . . . . . . 343
6.5.1 Properties of DS and SC Superalloys . . . . . . . . . . . . . . . . 345
6.5.2 Principles and Technological Aspects of Directional
Solidification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
6.6 Heat Treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
6.7 Applications of Superalloys in the Aerospace Field . . . . . . . . . . . 378
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
7 Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
7.1 General Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
7.2 Coating Deposition Techniques for Gas Turbine Engines . . . . . . . 388
7.2.1 Thermal Spraying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
7.2.2 Physical Vapor Deposition . . . . . . . . . . . . . . . . . . . . . . . 396
7.2.3 Chemical Vapor Deposition . . . . . . . . . . . . . . . . . . . . . . 398
7.3 Main Coating Systems in Gas Turbine Engines . . . . . . . . . . . . . . 401
7.3.1 Coatings Resistant to High-Temperature Corrosion . . . . . 402
7.3.2 Thermal Barrier Coatings . . . . . . . . . . . . . . . . . . . . . . . . 409
7.3.3 Failure Mechanisms of Corrosion Resistant
and Thermal Barrier Coatings . . . . . . . . . . . . . . . . . . . . . 419
7.3.4 Clearance Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
7.3.5 Erosion- and Fretting-Resistant Coatings . . . . . . . . . . . . . 428
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
xii Contents

8 Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
8.1.1 General Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 444
8.1.2 Pitting Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
8.1.3 Crevice Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
8.1.4 Selective Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
8.1.5 Galvanic Corrosion (Two-Metal Corrosion) . . . . . . . . . . . 451
8.1.6 Mechanochemical Corrosion . . . . . . . . . . . . . . . . . . . . . . 451
8.1.7 Hydrogen Embrittlement . . . . . . . . . . . . . . . . . . . . . . . . 456
8.2 High-Temperature Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
8.2.1 High-Temperature Oxidation: General Thermodynamic
and Kinetic Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
8.2.2 Alloy Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
8.2.3 Surface Nanocrystallization . . . . . . . . . . . . . . . . . . . . . . 472
8.2.4 Gas Phase Corrosion: Sulfur-Induced Corrosive
Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 478
8.2.5 Hot Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480
8.2.6 Hot Corrosion Testing . . . . . . . . . . . . . . . . . . . . . . . . . . 492
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 496
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499
9 Other Interesting Alloys for Aerospace and Related Applications . . . 501
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
9.2 Refractory Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
9.2.1 General Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
9.2.2 Refractory Metal Alloys . . . . . . . . . . . . . . . . . . . . . . . . . 504
9.2.3 Other Refractory Alloys: Chromium and the Platinum
Group Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509
9.3 ODS: Oxide Dispersion Strengthened Alloys . . . . . . . . . . . . . . . 511
9.4 Intermetallic Compounds and Ordered Alloys . . . . . . . . . . . . . . . 520
9.5 Shape Memory Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
9.5.1 Main Phenomenological Aspects and Relevant
Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
9.5.2 Aerospace Applications of Shape Memory Alloys . . . . . . 531
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545

Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Abbreviations

ACARE Advisory Council for Aviation Research and Innovation in Europe

AISI American Iron and Steel Institute
ALPS Advanced low-pressure systems
AOD Argon oxygen decarburization
APB Antiphase boundary
APS Air plasma spraying
ASM American Society of Materials
ASTM American Society for Testing and Materials
ATC Atmosphere and temperature controlled plasma spraying
AUW All-up weight
BC Before Christ
bcc Body centered cubic
bct Body-centered tetragonal
BVI Blade-vortex interaction
CBN Cubic boron nitride
CCT Continuous cooling transformation
CEVAM Consumable electrode arc melting
CEVAR Consumable electrode vacuum arc remelting
CF Casting factor
CFRP Carbon fiber reinforced polymer
CGDS Columnar-grained directionally solidified
CHM Cold heart melting
CMAS Calcium-magnesium-aluminosilicate
CP Commercially pure
CROR Counter-rotating open rotor
CTE Coefficient of thermal expansion
CTOL Conventional take-off and landing
CV Carrier variant
CVD Chemical vapor deposition
DAS Dendrite arm spacing
DB Diffusion bonding
xiii
xiv Abbreviations

DBT Ductile to brittle transition

DBTT Ductile to brittle transition temperature
DC Direct current
D-Gun Detonation gun
DIN Deutsche Institut für Normung
DMHT Dual microstructure heat treatment
DOS Density of states
DRX Direct recrystallization
DS Directionally solidified or directional solidification
EAF Electric arc furnace
EB Electron beam
EBCHR Electron beam cold heart remelting
EBM Electron beam melting
EB-PVD Electron beam physical vapor deposition
EDXS Energy-dispersive X-ray spectroscopy
EFTA European Free Trade Association
ELI Extra low interstitial
EN European Norms (Euronorm)
ESR Electroslag remelting
ESRR Electroslag rapid remelting
ETEM Environmental transmission electron microscopy
F/M Ferritic martensitic
F1 Formula 1
fcc Face-centered cubic
FML Fiber-metal laminate
GAR Grain aspect ratio
GB Grain boundary
GE General Electric
GFRP Glass fiber reinforced polymer
GPZ Guinier-Preston zone
GT Gas turbine
GTE Gas turbine engine
GTEs Gas turbine engines
HAZ Heat-altered zone or heat-affected zone
hBN Hexagonal boron nitride
hcp Hexagonal close packed
HCP High-pressure compressor
HDI High density inclusion
HE Hydrogen embrittlement
HEAC Hydrogen environmental assisted cracking
HEDE Hydrogen-enhanced decohesion
HELP Hydrogen-enhanced localized plasticity
HID Hard interstitial defects
HIP Hot isostatic pressing
Abbreviations xv

HSLA High-strength low-alloy steels

HTHC High-temperature hot corrosion
HTLA High-temperature low activity
HVOF High-velocity oxy-fuel flame spraying
IADS International Alloy Designation System
IHAC Internal hydrogen-assisted cracking
IUPAC International Union of Pure and Applied Chemistry
LBW Laser beam welding
LCF Low-cycle fatigue
LDI Low-density inclusions
LPPS Low-pressure plasma spraying
LPSO Long-period stacking ordered
LPT Low-pressure turbine
LTHA Low-temperature high activity
LTHC Low-temperature hot corrosion
MA Mechanical alloying
MISS-NC Mechanically induced surface self-nanocrystallization
Ms Martensite start
NDT Nondestructive testing
NEO Near Earth objects
NEP Nuclear electric propulsion
NS Nanostructured
NTP Nuclear thermal propulsion
ODS Oxide dispersion strengthened
OEW Operating empty weight
OPR Overall pressure ratio
OTB Oxygen top and bottom blowing
OWSME One-way shape memory effect
PAM Plasma arc melting
PCA Process control agent
PFZ Precipitation-free zone or precipitate-free zone
PGM Platinum group metal
PH Precipitation hardening or precipitation hardenable
PHSS Precipitation hardening stainless steels or precipitation hardenable
stainless steels
PM Powder metallurgy
PS Plasma spraying
PSZ Partially stabilized zirconia
PTC Plasma transferred arc
PVD Physical vapor deposition
R&D Research and development
RA Reduced activation
RRB Reconfigurable rotor blade
xvi Abbreviations

RS Rapid solidification
SAE Society of Automotive Engineers
SAP Sinter-aluminum pulver
SC Single crystal
SCC Stress corrosion cracking
SE Superelasticity
SEM Scanning electron microscopy
SFC Specific fuel consumption
SFE Stacking fault energy
SFH Simulated flying hours
SHE Standard hydrogen electrode
SIM Stress-induced martensite
SMA Shape memory alloy
SMAT Surface mechanical attrition treatment
SME Shape memory effect
SNC Surface nanocrystallization
SPF Superplastic forming
SRZ Secondary reaction zone
SSCP Stress-strain calculation program
STOVL Short take-off and vertical landing
TBC Thermal barrier coating
TBCs Thermal barrier coatings
TBO Time between overhaul
TCP Tetragonally close packed
TD Thoriated
TEM Transmission electron microscopy
TET Turbine entry temperature
TGO Thermally grown oxide
TISS-NC Thermally induced surface self-nanocrystallization
TPS Thermal protection system
TTT Time-temperature-transformation
TWSME Two-way shape memory effect
UNFCCC United Nations Framework Convention on Climate Change
UNS Unified Numbering System
USD US dollar
UT Ultrasonic testing
VAFN Variable area fan nozzle
VAR Vacuum arc remelting
VARTM Vacuum assisted resin transfer molding
VARTMFML Vacuum-assisted resin transfer molding fiber-metal laminate
VARTMPCL Vacuum-assisted resin transfer molding-infused plasma coated
laminates
VGC Variable geometry chevron
Abbreviations xvii

VIM Vacuum induction melting

VOCD Vacuum oxygen converter decarburization
VOD Vacuum oxygen decarburization
VPS Vacuum plasma spraying
VSTOL Vertical and/or short take-off and landing
XRD X-ray diffraction
YSZ Yttria-stabilized zirconia
About the Author

Stefano Gialanella is an Associate Professor in materials science and technology at

the Department of Industrial Engineering, University of Trento, Italy. He is respon-
sible for delivering courses to undergraduate students in industrial engineering,
Master’s students in materials engineering and cultural heritage, and PhD students
in materials, mechatronics, and system engineering. In addition, he has given
lectures and courses in materials science at the University of Florence, University
of Pisa, Politecnico di Milano, and Universitat Autonoma de Barcelona (Spain). Dr.
Gialanella’s research interests include intermetallics, oxidation of structural alloys,
nondestructive testing of power plant components, shape-memory alloys, tribology
with particular reference to automotive brake systems, and environmental monitor-
ing of cultural heritage with particular reference to phase transformations and
relevant structural and microstructural aspects. Based on this activity, he has
coauthored more than 130 scientific papers in international refereed journals. He is
also a member of the editorial boards of Materials Science Foundations and Journal
of Metallurgy.

Alessio Malandruccolo is a plant metallurgist at an Italian steel and nickel alloy

company, a position held since the time he received his Master degree in Materials
Science and Engineering at the University of Trento. He is currently involved in
R&D, quality control and improvement, failure analysis, and revamping and design
of new installations for production of new alloys. He has also worked as a teaching
assistant in Materials Technology and Applied Chemistry in the Master course in
Architecture and Building Engineering at the University of Trento (Italy), collabo-
rating also with the Faculty of Science and Technology of the University of Bolzano
(Italy), with lectures and seminars regarding his professional sector. He developed
part of his research experience and skills at the European Synchrotron Radiation
Facility – ESRF (Grenoble, F) – and at the Diamond Light Source (Didcot, UK),
focusing mainly on X-ray diffraction applied to inorganic materials characterization.
He cultivates a personal interest in product design and perception, user-oriented
design, and cognitive sciences.

xix
Chapter 1
A Brief Introduction to Aerospace
Applications

1.1 Historical Background

Aerospace applications represent an extremely challenging field, for the develop-

ment of new materials, and for the improvement of the existing ones, with the
important additional requirement of complying with highest safety standards, par-
ticularly as concerns civil and commercial aviation. A successful design and
manufacturing of components, relevant performances, certification, maintenance,
and safe operation are all directly influenced by the properties of materials used for
engines and airframe structures. Therefore, mechanical strength, creep, corrosion,
and oxidation resistance must be optimized for any specific application in this field
(Mouritz 2012). Process technologies for materials and production of components
are also subjected to a continuous evolution, with important reciprocal interaction.
Aerospace materials technology proceed very rapidly, in order to keep up with the
momentum of changes in the air transportation that over the years have been driven
by different forces. Nowadays, compositional refining and novel processing routes
attain better properties in already existing material systems so that more demanding
servicing conditions and higher safety standards can be met. The development of
new materials and processing technologies, needed for the new engines, airframes,
and control systems, not only resulted in improved aircraft performances but was
also transferred and applied in other technological fields (King et al. 2009), as
recalled in the next chapters.
The first engine-operated aircraft was the Flyer I (1903, Fig. 1.1), built by Wilbur
and Orville Wright. In order to achieve an adequate rigidity, the wing structure was
realized using a triangulated frame (see Fig. 1.2).
The triangulated structure of the wing provides several advantages, in addition to
rigidity, especially interesting for the aircraft:
• High maneuverability.
• Low weight as compared to bulk structures.

© Springer Nature Switzerland AG 2020 1

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_1
2 1 A Brief Introduction to Aerospace Applications

Fig. 1.1 The Flyer I airplane of the Wright brothers (NASA 2003)

Fig. 1.2 Schematic of triangulated structures under two different loading regimes. The letter C
indicates compressive load, the letter T indicates tensile load, the number 0 indicates the load-free
condition, and P is the overall load acting on the structure. (Redrawn from Curtis 1997)

• Possibility to swap from tension to compression of the bracing wires to sustain the
spars in all operational conditions.
At the time the Flyer I was built, no specific materials were available for the
construction of an aircraft, mostly due to the absence of an industrial field dedicated
to these applications. The materials used for the fabrication of the Flyer I were:
• Aluminum. An aspect not always mentioned is that the engine of the Flyer I
airplane, to meet the weight reduction requirements, was made of cast aluminum.
• Cotton fabric used as coating for the wing surface.
• Steel used for the critical structural parts, like the bracing wires (i.e., tension-
compression wires) and the wing struts.
1.1 Historical Background 3

Fig. 1.3 Schematic of the overall structure of the Flyer I wing and detail of the wing rib. (Redrawn
from Curtis 1997)

• Wood for the fabrication of the ash (inner part the wing), the ribs, the spars, and
the propeller.
The maximum speed of the Flyer I was 56 km/h, due to the limitations caused by
the aerodynamic drag associated with the shape of the aircraft, in particular with the
structure of the double-wing with bracing wires (see Fig. 1.3). To overcome the drag
effect of the double-wing surface and relevant wiring, needed to achieve a suitable
rigidity of the structure, as a possible approach to reduce this effect, a cantilever wing
was developed.
A sort of intermediate step toward this innovative structure can be found in the
Blériot monoplane (1909). The Blériot monoplane had a single wing, although not
fully self-sustaining, so that steel wires were needed to ensure the rigidity of the
structure (see Fig. 1.4). This airplane was the first to feature a particularly rigid
fuselage structure, a choice made to improve further its stability and maneuverabil-
ity. The airframe was made of tension and compression wires (see Fig. 1.5) with a
similar geometry to the wings of the Flyer I (see Fig. 1.3). The maximum speed of
this aircraft was still below 100 km/h, due to the poor aerodynamics of the whole
structure.
The situation was improved with the introduction of real cantilever wing struc-
tures. The main difference with respect to past geometries was that this new design
4 1 A Brief Introduction to Aerospace Applications

Fig. 1.4 1909 Blériot monoplane and overall schematic of the structure. (Redrawn from Curtis
1997)

Fig. 1.5 Schematic of the Blériot’s fuselage. (Redrawn from Curtis 1997)

was based on a self-standing structure, not requiring bracing wires and parts other
than the wing itself, to guarantee an adequate rigidity. An example, probably the
first, certainly the most popular, of the application of cantilever wings was the
Fokker DR 1 (1917) (Fig. 1.6). The main feature of this kind of wing is its thickness
that is the key factor as concerns its strength and rigidity. The internal structure of the
wing, characterized by composite ribs, is sufficient to avoid any additional external
structure (e.g., bracing wires). Thanks to the cantilever wing, the maximum speed
increased up to 160 km/h, still maintaining substantially the same materials as the
Flyer I. The situation evolved with the availability of more performing piston
engines and, subsequently, toward the end of World War II, with the development
of gas turbine engines, affording even higher speed and in principle capable to
overcome the speed of sound.
This evolution required, in the first place, new wing geometries and, conse-
quently, the selection of new materials.
The higher velocity had another important consequence: the overheating of the
aircraft skin, particularly at the wing leading edge and other frontal parts of the
1.1 Historical Background 5

Fig. 1.6 The cantilever wing of the Fokker DR 1 (Redrawn from Curtis 1997)

aircraft in direct contact with the static air film that forms along the splitting line of
the air flux. Therefore, the materials traditionally used for aircraft skin, especially
fabric and wood, turned out to be absolutely inadequate to face the temperature
reached under the new operational conditions.
The search for new materials was also encouraged by the contemporary shortage
of wood, a factor that turned to be critical, particularly in the late years of World War
II, so that alternative structural materials were increasingly needed. The solution to
these two problems was the production of metallic aircrafts, like the Junkers J-1
(1915). This airplane was fully built using steel: quite an important demonstration of
what might have been achieved with metal in aircraft structures. On the other hand,
the exceedingly high density of the ferrous alloy, notwithstanding the adopted
design solutions, rendered the airplane not very practical and flexible in operational
maneuvering. Therefore, few years later (1917), the availability of a novel aluminum
alloy was very much welcome, and a full-aluminum airplane was produced: the
Junkers J-4. No pure aluminum but an Al-Cu alloy was used (see Sect. 3.2). The
alloy had been developed a few years earlier by the German metallurgist Alfred
Wilm, who could only confirm empirically the hardening effect of thermal treat-
ments, without having the possibility of a direct observation of the precipitates
responsible for this improvement in the mechanical properties (see Sect. 3.2.3.3).
It can be noted that this was quite a fast technological transfer, from materials
research to product development, of particular specific interest, considering the
6 1 A Brief Introduction to Aerospace Applications

importance that Al-alloys were bound to have in the aerospace field, still lasting until
the present times. These two cases are particularly important, since they have
demonstrated for the first time the possibilities of metallic materials in the fabrication
of aircrafts: airframe, fuselage, and engines. This was certainly an enabling factor for
the development of commercial aviation that in fact, in between World War I and
World War II, started its history. In this regard, a year that is worth mentioning is
1933. In February the first Boeing 247, a full-metal airliner, capable to carry ten
passengers, took off for the first time, being regarded as the first modern transport
airliner. Although on board comfort was very much limited by structural restrictions,
the orders started to arrive soon, so numerous that the company was not able to keep
up with the pace. The situation was better exploited by Douglas Aircraft, which
launched in the second half of the same year 1933 the 12-passenger DC-1 airliner,
more spacious and comfortable than the Boeing 247. Other DCs (DC stays for
Douglas Commercial) were delivered in the following years: DC-2 (1934, 14 pas-
sengers) and DC-3 (1935, 21 passengers). The DC-3, thanks to its flight perfor-
mances and payload capabilities, gained enormous success among the starting up
airline companies that also thanks to this airplane, realized the possibility of making
profit out of air transportation.
Although, particularly in the war years, the technological advancements in the
military field were transferred to civilian air transportation, in some cases, the needs
were rather divergent. This is the case of the maximum operational speed that with
the advent of gas turbine engines disclosed the possibility to overcome the intrinsic
limit for a propeller aircraft, i.e., the speed of sound. The supersonic flight implies a
change both in the profile and in the dimension of the wings, in order to reduce air
drag, being the lift anyway guaranteed by the higher accessible speed. A good
example of this new philosophy was the Lockheed F-104 “Starfighter” (see
Fig. 1.7) capable to reach a maximum speed of Mach 2.2. Shorter and thinner
wings require a change not only in their design but also in the selected materials.
Titanium alloys for the fabrication of structural components became an obvious
choice. The development of aircraft structures, considering the interplay with the
development of new materials, is actually quite a complex process, since a great
number of variables must be considered. When increasing the maximum speed, the
skin temperature becomes an issue. In this regard, the application of titanium alloys
reached a particularly impressive level in the Lockheed YF 12 A and its successor,
the Lockheed SR-71 “Blackbird” (see Fig. 1.8), in which not only structural parts but
also the skin and control surfaces were made of newly developed β-Ti-alloys (see
Chap. 4 and Sect. 4.3.1.4).
The adoption of titanium alloys also for the fabrication of the skin of the SR-71
was a necessary choice to stand the peak temperatures at its highest speed of Mach
3.35. It is worthwhile to note that conventional aluminum alloys would not have had
a sufficient strength to resist above Mach 2.0 and steel would be too heavy. On this
occasion too, the experience and knowledge developed for a military aircraft were
important also for the design and its implementation in supersonic civil aircrafts. The
Concorde structure and skin were made of aluminum alloys (see Chap. 3), suitable
for a peak speed not exceeding Mach 2.
1.1 Historical Background 7

Fig. 1.7 Lockheed F-104 Starfighter. (Friend and Sefic 1972, NASA 2017, reproduced with
permission of the Publisher)

Fig. 1.8 Lockheed SR-71 “Blackbird.” (NASA 2008, reproduced with permission of the
Publisher)
8 1 A Brief Introduction to Aerospace Applications

If higher temperatures of the aircraft skin are to be stood, considering, for

instance, the conditions faced in atmosphere reentry after orbital flight, like in case
of the Space Shuttle, a composite, reusable, ceramic-based protection system is
needed, made of tiles of zirconia-glassy silica felt and, for the hottest parts,
carbon-carbon composites. Thanks to this protective shield, surface temperatures
in excess of 1300
C do not provoke any harm to the underlying Al-alloy frame,
unless faults in the protection system would intervene, as occurred in case of a tragic
accident (Savino et al. 2005). The engine temperature is another parameter that has
been sensitively influenced by the selected materials properties. The trend in the
operational temperatures and capabilities of aircraft engines has always been increas-
ing, starting from the in-line piston engines of the early propeller-operated aircrafts,
reaching material temperature in the 200
C range, going through the radial high-
efficiency engines, with peak temperatures of 500
C, arriving to the extreme
thermal challenge of the newly developed gas turbine engines. The hottest structural
parts, i.e., the high-pressure stage turbine blades, nowadays made of nickel-based
superalloys (see Sect. 6.4), are operated at gas temperatures above the material
capability, i.e., its incipient melting point. It is only thanks to the specific micro-
structure of the components, i.e., single crystal, and optimized cooling design, i.e.,
hollow blades, that these conditions can be safely sustained.

1.2 Novel Design Criteria

In recent years, the development of aircraft structures has become increasingly

complex in terms of metrics and design drives, with respect to the early times,
when technical performances were by far the dominant aspects. Nonetheless, as in
the past, contemporary materials are still a decisive enabling factor, since only
suitable materials properties will render new design requirements feasible. A remark-
able evolution has occurred in the selection criteria that, depending on the specific
aircraft structure, will anyway be a trade-off choice.
The improvements achieved in aircraft structures have been continuously accom-
panied by a growing concern for safety. This aspect becomes increasingly important
with the development of civil aviation, where the protection of passengers’ lives is
not only an ethical issue but also an economic parameter, considering the impact of
insurance costs on the business plan budget of any company, not to mention the
indirect costs that air accidents may provoke. This aspect appeared soon dramatically
evident, for the accidents occurred at two de Havilland Comet aircrafts during
take-off at the airport of Ciampino in Rome (Italy) in 1954. The de Havilland
Comet (1952, see Fig. 1.9) was a commercial plane derived from a military cargo
that originally had no windows in the fuselage that were present instead in the case of
the commercial version.
The reason for the failure, occurring under the effect of pressurization-
depressurization cycles, was an inappropriate design of the window frames, whose
corners acted as initiation spots for the fracture, propagating to the rest of the
1.2 Novel Design Criteria 9

Fig. 1.9 RCAF de Havilland Comet. (a) Internal view of the aircraft showing the rupture around
the window; (b) external view showing the same rupture in the skin of the fuselage (see arrows for
the indication of the position of the window and the crack nearby); (c) schematic of the airplane
(Darling 2001; The National Archives 2019)

fuselage. This failure could not occur in the original military model since no
windows were there in the fuselage. The disasters blocked the fast expansion of
British Airways that at the time was gaining a leading position in civil aero-
transportation. Important consequence of this sort of structural failures was the
adoption of two general design criteria for aircraft structures: safe life and fail safe,
which can be regarded as the main guidelines for the development of passenger
airliners and in general of all aircrafts for commercial transportation. Safe life
criterion means the possibility to predict the useful lifetime of structures, compo-
nents, and systems of an aircraft, in order to schedule maintenance, to proceed with
limited repairs, and to replace heavily damaged parts. As to the fail safe criterion, it is
paramount in aircraft structures since it should allow the airplane to retain sufficient
operational capabilities, notwithstanding any possible failure, allowing, for instance,
emergency landing procedures.
Looking at the present and past use of aluminum alloys in aircraft structures, it
sounds strange that after the Comet disasters, Al-Zn alloys (see Fig. 1.10 and also
Chap. 3) were banned, since they were considered the main responsible for the
structural failure. Once the real reasons for the accidents were made clear, the use
and consequent development of these alloys was resumed and is continuing until
10 1 A Brief Introduction to Aerospace Applications

Fig. 1.10 Evolution of the yield strength of various Al-alloys as a function of the 1st year they were
used in relevant commercial aircrafts. (Redrawn from Starke and Staley 1996)

present, involving also other systems like: Al-Li and Al-Sc alloys, the latest families
of Al-alloys to be developed.
The success of aluminum in aerospace applications is due to the excellent specific
material properties, considering its comparatively low density, also as concerns its
alloys. For this same reason, magnesium alloys are attracting renovated interest, after
having had an important role particularly in military aircraft soon after the end of
World War II and a decline afterward, for their limitations (see Sect. 3.3) but also for
the strong competition by Al-alloys. The absolute leading role of Al-alloys in aircraft
industry has been “spoiled” by a real breakthrough brought about by the latest
Boeing airliner: the 787 Dreamliner (see Fig. 1.11). The whole project has been
developed on the basis of completely new approaches, redesigning all parts and
systems, according to new standards. One of the most innovative and revolutionary
aspects regards the massive use of structural composites, adopted also for the
fabrication of large single-piece components, like the wing box and the forward
fuselage piece including the cockpit, and the barrel section. These are polymer
matrix composites with reinforcing carbon fiber felts, manufactured directly with
the component shape, thus requiring particularly large infrastructures.
Interestingly the extended use of structural composites in the Dreamliner, sum-
ming up to 50% of the operating empty weight (OEW, see Fig. 1.11), has occurred in
association with a further increase in the percentage of titanium alloys. These
materials had already been extensively used in the previous Boeing airliner, the
B777 for structural components and other parts, previously largely made of steel (see
Table 1.1).
1.2 Novel Design Criteria 11

Fig. 1.11 Materials used in Boeing 787 Dreamliner. (Redrawn from Boeing 2013)

Table 1.1 Titanium alloys used on the Boeing 777

Alloy Condition UTS [MPa] Forms
Commercially pure Annealed 345–550 Bar, plate, sheet
Ti-3Al-2.5 V Cold worked and stress relieved 860 Hydraulic tubing
Annealed 690 Honeycomb core
Ti-6Al-4 V Annealed 895 All forms
β-annealed 895 Forgings
Solution treated and annealed 1100 Fasteners
Ti-10V-2Fe-3Al Solution treated and annealed 1190 Forgings
Ti-15V-3Cr-3Al-3Sn Solution treated and annealed 1035 Sheet
Solution treated and annealed 1140 Castings
Ti-3Al-8V-6Cr-4Mo-4Zr Cold drawn and aged 1240–1450 Springs
Beta-21S Solution treated and annealed 860 Nacelle area

This choice is coherent with the deliberate strategy, which drove the design of
these more recent airliners, for an extreme weight reduction. This is of course
generally important for aircraft structures. As concerns composites, also in the
past, their early use in aeronautics turned out to be decisive on some notorious
occasions. The vertical takeoff option of the Sea Harrier could only be implemented
once some structural parts were made using polymer matrix composites, as illus-
trated in Fig. 1.12, with a consequent decisive weight reduction. However, the
choice of lighter materials is nowadays driven also by the need to reduce fuel
consumption, with relevant benefits not only on budget savings but also environ-
mental sustainability. For a large airliner, a weight reduction of 1000 kg reduces the
fuel consumption by 1.1–1.5% (King et al. 2009). This can be achieved by an
intelligent combination of the different classes of available materials, as shown by
the data in Table 1.2 pertaining to popular aircrafts.
12 1 A Brief Introduction to Aerospace Applications

Fig. 1.12 Sea Harrier and some parts replaced by structural composites to achieve an adequate
weight reduction. Values among brackets refer to the weight saving obtained by means of
composite materials. (Redrawn from Calvert 1990)

Table 1.2 Weight percentages of structural materials used in the airframes of civilian and military
aircrafts (Mouritz 2012)
Aluminum Other
Aircraft alloys Composites Steel Titanium materials
Boeing 737 81% 3% 6% 3% 7%
Airbus A330-340 68% 17% 9% 4% 2%
Airbus A380 61% 25% 10% 4%
McDonnell Douglas F-18 51% 9% 16% 13% 11%
Hornet
Lockheed F-22 Raptor 11% 35% 5% 33% 16%

As it is for other transportation fields, also aviation companies have to consider

their own economical sustainability and competitiveness. A successful aerospace
product should be valuable for the customers, with a low environmental impact, and
still retain elevated safety standards. The cost for the end user, the passenger in case
of civil aviation, results from a number of parameters, including initial cost and cost
of ownership. The total cost of ownership includes several terms, according to the
scheme in Fig. 1.13. It is worth saying that a more efficient management affords not
only money savings to the companies but also a reduction of the impact of aviation
on the environment and, thereby, on human health. Better designed and lighter
aircraft structures will have better aerodynamic performances, lower exhaust, and
noise emissions. Of course, for these latter aspects, an improved efficiency of the
engines is paramount. This is the reason why also for the engines, a continuous
search for new materials and part design have been pursued, as reflected by the
evolution over the years of the turbine inlet temperature, even named turbine entry
temperature (TET).
1.2 Novel Design Criteria 13

Fig. 1.13 Total cost of ownership and its split in the main components.  refers to parameters that
are influenced by custom features such as aircraft type, age of the aircraft, operating conditions, and
owner (Cantor et al. 2001)

Fig. 1.14 Evolution of the turbine inlet temperature as a function of the in-service date. See main
text for the legend of the processing techniques recalled in the graph. (Redrawn from King et al.
2009)

As it will be illustrated in Chap. 2, the turbine inlet temperature is also a very

important parameter to assess the efficiency of a GT engine. An increase by 50
C of
the turbine inlet temperature can increase by 1% the engine efficiency, with a
consequently lower fuel consumption and reduced exhaust emission. Considering
that fuel costs are the most important expenses for airline companies (see Fig. 1.13),
the interest in better performing engines becomes evident and may explain the
remarkable development of the processing routes for the manufacturing of turbine
blades. This is shown in Fig. 1.14, in which the main enabling process technologies
used for the production of high-pressure blades, regarded as the most critical parts of
a gas turbine engine, are associated with the relevant achieved TETs. Casting and
vacuum induction melting (VIM, see Sect. 5.3.2) were used to produce
14 1 A Brief Introduction to Aerospace Applications

polycrystalline equiaxed blades, in general solid or with cooling channels drilled

after solidification. These components suffered from creep-limited performances,
since the elevated concentration of grain boundaries determined comparatively high
deformation rate, mainly due to diffusive creep (see Sect. 6.5.1). A possible
approach to improve creep life was through the reduction of the metal temperature,
attained with the deposition of a thermal barrier coating (TBC, see Sect. 7.3.2).
A lower blade temperature would display a lower creep deformation rate. This
was actually the case, although the durability of TBC systems and the prediction of
their useful lifetime were soon regarded as critical parameters for key components,
as high-pressure turbine blades are. The creep problem was therefore tackled acting
directly on the structural reason for its occurrence in turbine blades, i.e., the presence
of grain boundaries, particularly those oriented normally to the maximum stress,
applied along the blade main axis. Therefore, using directional solidification
(DS) and, subsequently, single-crystal (SC) solidification (see Sect. 6.5), blades
with grain boundaries aligned along the main axis of the blade, or missing at all,
were obtained. These components, since the 1980s, are the standard for commercial
aircrafts and are being adopted also for land-based gas turbines, again for enhancing
their performances at the acceptable costs afforded by the now mature technology of
the directional solidification and single-crystal processes.

1.3 Aerospace-Related Fields

The mentioned land-based gas turbine engines are just one of the many technologies
and materials developed for the aerospace field and then transferred to terrestrial
applications. The situation is actually not always so clear-cut: in fact, gas turbine
engines exploit some of the working principles of turbo-charged piston engines,
originally developed for automotive applications. It is therefore advisable thinking in
terms of two-way exchange between aerospace and other technological fields.
It is certainly true that the aerospace field has always had a particular focus on the
development of materials and systems complying with strict limitations, as concerns
safety and reliability. These standards require important systemic research efforts to
study problems in depth and to find optimal solutions. Therefore, materials and
systems developed and/or optimized for the aerospace fields are most interesting
candidates for applications also in other fields. In the present book, only some of
those more directly involved with the use of aerospace alloys will be considered.
The field of energy production is certainly benefitting from the processing
technologies for the production of components made of superalloys (see Chap. 6)
and special steels (see Chap. 5), used in land-based gas turbines, vapor turbines, and
other systems working at comparatively high temperatures, although generally lower
than the aerospace counterpart, but still demanding for the alloy materials. This is so
even because the environmental conditions that are faced by components of terres-
trial plants may be more demanding in terms of corrosion resistance, considering that
References 15

the quality of the fuels used in this field is definitely worse than the extremely refined
aerospace ones.
Light alloys, in particular aluminum and magnesium (see Chap. 3), are being
employed to increasing extent in the automotive field, for the same reason just
discussed for aerospace applications, to reduce the weight of the vehicle in order
to reduce fuel consumption and consequently the environmental pollution associated
with the production of particulate matter and gaseous emissions.
Titanium alloys (see Chap. 4) are very important for biomedical applications, as
concerns particularly their biocompatibility. Shape-memory alloys (see Sect. 9.5),
which have been applied for the first time as cryo-fit joints in a military aircraft, are
now having an important set of applications in the aerospace field, in some case
really strategic, like the deployment system Frangibolt (see Sect. 9.5.2). At the same
time, still remaining in the field of products with high added value, numerous
biomedical applications are now available, like stents, orthodontic wires, surgery
tools, and many more.
Structural polymer matrix composites are now finding their way in aerospace,
already widely employed for terrestrial application, and are gaining more interest.
The aircraft brake systems, based on carbon-carbon composites, have provided
fundamental knowledge and performance data for the application of similar princi-
ples and technical design to brakes of top and high-performance cars, like the F1
racing.

References

Boeing (2013) 787 Aircraft Rescue & Firefighting Composite Structure. Available via DIALOG.
http://www.boeing.com/assets/pdf/commercial/airports/faqs/787_composite_arff_data.pdf.
Accessed 18 Aug 2017
Calvert D J (1990) Harrier. Ian Allan Publishing Ltd
Cantor B, Assender H, Grant P (2001) Aerospace Materials. Institute of Physics Publishing
Curtis H D (1997) Fundamentals of aircraft structural analysis. Irwin Publ., Chicago
Darling K (2001) Airliner Tech Volume 7 – De Havilland Comet. Specialty Press Publishers and
Wholesalers
Friend E L, Sefic W J (1972) Flight Measurements of Buffet Characteristics of the F-104 Airplane
for Selected Wing-Flap Deflections. Nasa Technical Note D-6943. Available via DIALOG.
https://www.nasa.gov/centers/dryden/pdf/87804main_H-666.pdf. Accessed 05 May 2019
King D et al (2009) Advanced Aerospace Materials: Past, Present and Future. Aviation and the
Environment 3(9): 22–27
Krishnan K S (2006) Materials and the Aerospace Industry. In: Enhancing Innovation and Com-
petitiveness Through Investments in Fundamental Research, Westin Hotel, Arlington, VA,
3–5 December 2006
Mouritz A P (2012) Introduction to Aerospace Materials. Woodhead Publishing Ltd
NASA (2003) Learning to Fly: The Wright Brothers’ Adventure – A Guide for Educators and
Students With Activities in Aeronautics. NASA Educational Division. Available via DIALOG.
https://www.nasa.gov/audience/foreducators/topnav/materials/listbytype/Learning.to.FlyThe.
Wright.Brothers.Adventure.html. Accessed 09 Jun 2017
16 1 A Brief Introduction to Aerospace Applications

NASA (2008) SR-71 Blackbird. NASA Facts. Available via DIALOG. https://www.nasa.gov/
centers/dryden/pdf/495839main_FS-030_SR-71.pdf. Accessed 10 Jun 2017
NASA (2018) F-104 Starfighter Image Gallery. https://www.nasa.gov/centers/armstrong/multime
dia/imagegallery/F-104/index.html. Accessed 05 May 2019
Savino R et al (2005) Aerothermodynamic Study of UHTC-based Thermal Protection Systems.
Aerospace Science and Technology 9 (2): 151–160
Starke E A, Staley J T (1996) Application of Modern Aluminum Alloys to Aircraft. Progr
Aerospace Sci 32:131–172
The National Archives (2019) Public Information Films | 1945 to 1951. Available via DIALOG.
http://www.nationalarchives.gov.uk/search/results?_q¼de+havilland+comet. Accessed 05 May
2019
Torenbeek E (2013) Advanced Aircraft Design. John Wiley & Sons Ltd

Further Reading

Bowman M W, Vogelsang M (2000) Lockheed F-104 Starfighter. The Crowood Press Ltd,
Ramsbury
Cutler J (2005) Understanding Aircraft Structures. Blackwell Science
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook – Airframe
Vol. 1. Federal Aviation Administration
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook – Airframe
Vol. 2. Federal Aviation Administration
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook – Powerplant
Vol. 1. Federal Aviation Administration
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook – Powerplant
Vol. 2. Federal Aviation Administration
Gates D (2004) Sky Wars: A History of Military Aerospace Power. Reaktion Books Ltd
Megson T (2013) Aircraft Structures for Engineering Students. Elsevier
Withey P (2019) The Real Story of the Comet Disaster: De Havilland Comet – Structural Fatigue.
In: Hamburg Aerospace Lecture Series. Available via DIALOG. https://zenodo.org/record/
2551089#.XNHwSI5LiUl. Accessed 05 May 2019
Wixey K E (1987) Lockheed Constellation. Ian Allan Ltd
Chapter 2
Gas Turbine Aero-Engines

2.1 Introduction

Gas turbine engines are mechanical systems able to provide an extremely efficient
conversion of the energy, produced by internal combustion, into mechanical work,
to power an electric generator or to provide thrust for an aircraft to fly. Several
features of gas turbine engines have been especially developed for aircraft propul-
sion, since the pioneering studies by René Lorin in 1913 (Rolls-Royce 1996),
followed by Frank Whittle in the UK and Hans von Ohain in Germany, in the
years preceding the burst of the World War II (Gunston 1995). A jet propulsion
engine is a particular type of reaction engine, in which a fast-moving airflow (jet)
generates a thrust, according to Newton’s third law of motion. Several different types
of jet engines do exist, named according to the relevant working principles. Strictly
speaking, a pure jet propulsion occurs in turbojets, ramjets, pulse jets, and rockets
too. In turbo-propeller and turbofan engines, the thrust is completely or largely
produced by the compression waves generated by the propeller or ducted fan,
respectively. Gas turbines have moving parts, which pose important challenges to
materials owing to the combined effects of mechanical stresses, corrosive atmo-
sphere, and high temperatures. In these air-breathing systems, a rotary compressor,
powered by the turbine, provides high-pressure air to the combustor. Here, air is
mixed up with the fuel and spark ignited to produce the flame and enthalpy output
that through the gas is transferred to the turbine rotor. Although, working under
continuous gas flux, the efficiency of these thermal machines can be evaluated
referring to a closed Brayton cycle, thermal efficiency is not the only parameter
that qualifies different classes of jet aero-engines. In fact, specific fuel consumption;
specific thrust, particularly during takeoff; noise; and other issues pertaining to the
environmental impact are all other important parameters to be considered.

© Springer Nature Switzerland AG 2020 17

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_2
18 2 Gas Turbine Aero-Engines

2.2 Types of Aircraft Engines

The development of gas turbine engines stems from the need to overcome the
intrinsic limits of piston engines, used since the early flight of the Wright brothers
(see Chap. 1) to operate a standard wooden propeller. The two main drawbacks that
have emerged in applying these engines to aircraft propulsion are efficiency reduc-
tion for speeds in excess of about 600 km/h and reduction in the propulsion
efficiency determined by the so-called air starvation at high altitude flights. The
first limit is due to the fact that the propeller cannot reach, and overtake, the speed of
the pressure waves it produces. Therefore, the propulsion efficiency of the engine
would progressively decrease as the speed of sound is approached. The second
phenomenon is caused by the reduction of air density in the upper layers of the
Earth’s atmosphere, with a consequent effect on the compression ratio that can be
reached inside the piston chamber.
The first patent for a jet propulsion system was issued by René Lorin (1913, see
Fig. 2.1), although the early application of this new technology appeared more than
20 years later, due to the limitations in technological processes and materials
properties. The main idea behind the development of jet propulsion engines can be
retrieved in the working principle of a turbocharger engine (see Fig. 2.2).

Fig. 2.1 Scheme of the first

jet propulsion system
developed by René Lorin,
1913. (Redrawn from Rolls-
Royce 1996)

Fig. 2.2 Turbocharger engine with centrifugal compressor. (Redrawn from Meetham 1981)
2.2 Types of Aircraft Engines 19

Fig. 2.3 Structure of a two-stage gas turbine jet engine

The pressure of the air entering the piston chamber is raised by a centrifugal
compressor connected to the turbine. The turbine is operated by the exhaust emission
ejected from the combustion chamber, so that further energy can be recovered from
the gas before it is emitted from the engine. Similarly, in gas turbine engines, a
compressor is meant to increase the pressure of the air entering the combustors to
optimize the combustion under all operational conditions. Part of the energy carried
by the gas flux is transferred back to the compressor by the turbine stages through
coaxial multiple shafts. The rest of the combustion energy output is used to generate
thrust or other forms of energy in case of non-aero applications, e.g., marine and
terrestrial engines. The main operation stages of a gas turbine (GT, Fig. 2.3) can be
summarized as follows:
• Air inlet into the compressor
• Air compression
• Compressed air inlet into the combustion chamber
• Fuel addition
• Ignition
• Entrance of the exhaust gas into the turbine
• Generation of thrust work
20 2 Gas Turbine Aero-Engines

Fig. 2.4 Comparison between a gas turbine jet engine and a four-stroke piston engine. (Rolls-
Royce 1996, images courtesy of Rolls-Royce plc)

GT engines may reach extremely high efficiencies, close to 100%. They operate
according to a continuous process with a free expansion in the combustion area.
Piston engines are characterized by a lower efficiency, if compared with GTs,
since just one out of the four strokes is active (see Fig. 2.4). Furthermore, they must
be more massive in order to withstand an instantaneous increase in pressure associ-
ated with the constrained, fixed combustion volume.
These two features contribute to make these engines a non-ideal choice for
aerospace applications. Notwithstanding the “open” character of a GT engine, it is
common practice to analyze its working principles referring to a basic thermody-
namic approach, using an equivalent thermodynamic cycle to estimate its efficiency
and to evaluate the associated work output from the relevant cycle area, as depicted
by the general Carnot cycle in Fig. 2.5.
As anticipated in the introductory paragraph of this chapter, the GT engine
working principles can be described according to the Brayton cycle. It is evident
that the thermodynamic cycle is different if compared to the piston engine, consid-
ering that this latter is better represented by the Otto cycle (Fig. 2.6).
One of the most important parameters to take into account while analyzing GT
engine thermodynamic cycles is the turbine entry, or inlet, temperature (TET). This
2.2 Types of Aircraft Engines 21

Fig. 2.5 P-V diagram of the

Carnot cycle

Fig. 2.6 Comparison between Otto and Brayton cycles using P-V and P-T diagrams

is a very important parameter for evaluating the efficiency of the propulsion system.
The most severe operating conditions inside the jet engine are those corresponding to
point 3 in Figs. 2.6 and 2.7. In fact, this point of the cycle features the highest
temperature of the hot gases, impinging directly on the vanes and blades of the
turbine.
22 2 Gas Turbine Aero-Engines

Fig. 2.7 Brayton cycle and correspondence with different GT operational stages. (Rolls-Royce
1996, images courtesy of Rolls-Royce plc)

In addition to the combustion gases and relevant thermomechanical stresses,

blades are also under the effect of the centrifugal force generated by the disc and
blade high-speed rotation.
The efficiency of the propulsion system is improved by increasing the difference
in temperature between points 1 and 3 (see Fig. 2.6, Brayton cycle). Inlet air
temperature is a parameter that cannot be changed straightaway, depending on the
atmospheric conditions. Conversely, T3, i.e., the turbine entry temperature, can be
modified and, in particular, increased with an appropriate design of the engine and
selection of suitable materials. However, the improvement in thermodynamic effi-
ciency is accompanied by an unwanted increase in the specific fuel consumption
(SFC). SFC can be regarded as a fuel-related efficiency parameter of the engine, and
it is given by the amount of fuel consumed per unit thrust. This increase in SFC can
be compensated for by a corresponding increase in the overall pressure ratio (OPR),
i.e., the ultimate air pressure value that can be achieved in the latest stage of the
compressor, before air injection into the combustion chamber. The situation is
2.2 Types of Aircraft Engines 23

Fig. 2.8 Dependence of the specific fuel consumption and engine thrust as a function of the turbine
entry temperature (TET) and overall pressure ratio (OPR). (Redrawn from Reed 2006)

Fig. 2.9 Propulsive efficiency vs aircraft speed for different engines (1 mph ¼ 1.609 km/h).
(Redrawn from Rolls-Royce 1996)

depicted by the graph in Fig. 2.8, displaying the combined effect of TET and OPR on
the engine performances. In state-of-the-art gas turbine engines, e.g., Rolls-Royce
Trent XWB, the OPR has reached values of 50 or so. Concerning the efficiency of
the propulsion system, turbojets show an increase in their efficiency with the aircraft
speed, since this reduces the dissipation of the energy of the outlet gas flux against
the surrounding static air (see Fig. 2.9). The peak followed by a decreasing trend at
24 2 Gas Turbine Aero-Engines

Fig. 2.10 Rolls-Royce Avon single-shaft engine. (Rolls-Royce 1996, images courtesy of Rolls-
Royce plc)

higher speeds observed in the efficiency curve of the propeller-operated aircrafts is

dependent instead on the reduction in efficiency that occurs once the advancement
speed of the propeller in air is approaching the speed of sound.
Differences in gas turbine engines for aircraft applications can be found in the
architecture and in the range of performances. Turbojet engines use the combustion
gases directly for the propulsive thrust. In turbofans, the propulsion is provided by
the fan and to minor extent only by the jet of the exhaust gas. This is the basis for the
by-pass principle, which implies a split in the airflow. The whole air intake is given
an initial low compression, and a large fraction is then ducted to the bypass, the rest
being delivered to the combustion chamber (Fig. 2.11). The idea behind the devel-
opment of by-pass engines is related to the following specific aspects: better fuel
consumption, higher thrust (even at low-speed regimes), noise reduction, and
reduced relative weight of the engine. These aspects have been achieved thanks to
the evolution of the engine design, starting from a comparatively simple single-shaft
turbojet, as shown by the schematic of the Avon engine (see Fig. 2.10), to reach the
complexity of the multiple-shaft engines in Fig. 2.11, featuring also higher by-pass
ratios.
The by-pass ratio, i.e., the ratio between cool air by-passed through the outer duct
and the air flowing through the high-pressure inner part of the engine, is an important
feature of these engines. With low bypass ratios (e.g., 1:1), the two air streams are
generally mixed before being emitted through the exhaust duct of the engine. In
turbofan engines requirements for higher bypass ratios, up to 5:1, are met by large
front fans and multiple-stage compressor, using a twin- or triple-shaft configuration
(see Fig. 2.11). In this case the front fan acts also as a low-pressure compressor stage.
Bypass ratios in the order of 15:1 are achieved using prop-fans (Rolls-Royce 1996).
Custom design has been developed for complying with the specific requirements of
the different aircrafts and relevant operational conditions. Correspondingly, different
pressure and temperature profiles are present along the engine like those shown in
2.2 Types of Aircraft Engines 25

Fig. 2.11 Rolls-Royce engines with a multiple-shaft design. (Rolls-Royce 1996, images courtesy
of Rolls-Royce plc)

Fig. 2.12 Air speed, pressure, and temperature profiles in a turbojet engine. (Rolls-Royce 1996,
image courtesy of Rolls-Royce plc)

Fig. 2.12 for a turbojet engine. These fundamental changes in the structure of the
aircraft engines have been possible thanks to the availability not only of better
performing materials but also of new technical choices in the design of the engine
components. These aspects are considered in the following sections.
26 2 Gas Turbine Aero-Engines

2.3 Compressors

In a gas turbine engine, the compressor converts the power from the turbine into kinetic
energy of the airflow, by increasing its pressure. An upstream rotating compressor is
coupled to a downstream turbine. The combustion chamber is placed in between these
two components. Energy is added to the gas flow through the ignition of the fuel-air
mixture that forms inside the combustor by spraying fuel into the compressed air. The
resulting enthalpy output raises the gas temperature, the flame temperature, which in
the most severe operating conditions can even reach values of 3000  C.
The work output of the combustion is used to drive, through coaxial components
of the shaft, the different stages of the turbine and the corresponding stages in the
compressor. All energy that is not used for the shaft operation leaves the engine as an
exhaust gas flux and contributes to the thrust of the aircraft.
Actually, a multi-stage compressor is rather common in case of axial geometry, but
not so much in centrifugal compressors, the earliest to be used in aerospace gas
turbines. In this regard, compare typical schemes for centrifugal (Fig. 2.13) and axial
(Fig. 2.14) compressors, respectively. The complex trajectories of the air flux inside
centrifugal compressors (see Figs. 2.13 and 2.15) lead to instabilities that can be
regarded among the main reasons for their decline. In jet engines equipped with axial
compressors, the pressure raise is due to the progressive reduction of the volume
available to the fluid as it proceeds inside the engine (see Fig. 2.15). Axial compression
produces a nearly linear velocity profile of the gas, and it is definitely more stable than
in centrifugal compression. One important parameter is the compressor exit tempera-
ture, increasing over the years (see Fig. 2.16), following the trend of the OPR, also
quoted in the same graph for different engines. The related increasing trend of the
pressure ratio for both aerospace and industrial applications is also displayed (Fig. 2.16).
The two main components of the compressor rotor are the discs and the blades.
Disc materials should exhibit adequate values of the following properties:
• Resistance to high-frequency fatigue
• Resistance to low-frequency fatigue (engine start and shutdown phases)
• Tensile strength, toughness, and creep resistance

Fig. 2.13 Two examples of centrifugal compressors used in early aero gas turbine engines. (Rolls-
Royce 1996, images courtesy of Rolls-Royce plc)
2.3 Compressors 27

Fig. 2.14 Twin-shaft compressor. (Rolls-Royce 1996, image courtesy of Rolls-Royce plc)

Fig. 2.15 Air velocity and pressure for centrifugal and axial compression systems. (Rolls-Royce
1996, images courtesy of Rolls-Royce plc)

Compressor blades have some specific requirement too. In particular, they must
be characterized by good values of:
• Erosion resistance
• Resistance to high-frequency fatigue
• Tensile strength
• Fracture toughness
Erosion resistance can be improved using protective composite coatings; Co-
WC-based coatings show excellent erosion resistance and tribological properties
28 2 Gas Turbine Aero-Engines

Fig. 2.16 Compressor exit temperature vs in-service date for several GT engine models (for each
one of them, the OPR value is also indicated) and compression ratio vs in-service date for aircraft
and industrial applications. (Redrawn from Boyce 2012; Meetham 1981)

(see Sect. 7.3.5). In Table 2.1 a list of materials used for the fabrication of compres-
sor rotor parts of a commercial GT engine is presented.
Titanium alloys (see Chap. 4) are used for the first outer stages, both discs and
blades, since they provide a good compromise of high specific mechanical strength,
creep, and oxidation resistance. Where higher temperatures are achieved and render
diffusion-assisted phenomena more important, nickel- and iron-based superalloys,
like Inconel 718 (see Sect. 6.3), and precipitation hardened austenitic stainless steels
(PHSS), like A286 (see Sect. 5.4.3.3), are preferably used. These materials indeed
provide adequate mechanical stability even against creep deformation, and the
definitely higher densities, as compared to Ti-alloys, are not resulting in too large
inertial momenta, due to high-speed rotation, for the comparatively smaller dimen-
sions of these parts. The intrinsically good oxidation resistance can be improved
further with suitable coating systems, like Ni-aluminides diffusion coatings or
MCrAlY overlay coatings (see Sect. 7.3.1).

2.4 Combustors

The combustor is the part of the GT engine in which the conversion of the chemical
energy content of the fuel is turned into the thermal energy of the gas flow. This is
achieved through the combustion of the air-fuel mixture with the as higher as
possible efficiency, also to keep the emission of undesired pollutants at the lowest
levels. To meet this target, a highly turbulent gas flow is very much welcome
(Cumpsty 2003). In early GT engines, the combustor was made of separated
chambers, with several inlet hole access for the compressed air, in order to improve
the homogeneity of the composition of the combustion mixture (see Fig. 2.17).
2.4 Combustors 29

Table 2.1 Compressor rotor materials, coating systems, and part finish
Part name Material Finishing Part name Material Finishing
Blades, stage Ti-6Al-4V WC Blades, stages Ti-6Al-4V Colloidal
1 4–5 graphite
Adapter AMS 5754 None Blades, stages Ti-6Al-2Sn- Colloidal
retainer 6–9 4Zr-2Mo graphite
Air duct A286 Cu-Ni-In Bumper AMS 5612 Ni graphite
bearing
Ag plate Disc, stage 1 Ti-6Al-4V Ni-aluminide
Assembly, Inconel 718 Ni-aluminide Molydisulfide
air seal Aluminum Disc, stage 2 Ti-6Al-4V Molydisulfide
oxide
Blades, stage Ti-6Al-4V Cu-Ni-In Discs, stages Inconel 718 None
1 10 and 14
Molydisulfide Hub tie clamps Inconel 718 Ni-aluminide
Blades, stage Ti-6Al-4V Cu-Ni-In Cu-Ni-In
2 Ti-6Al-2Sn- Molydisulfide Hub tie Inconel 718 None
4Zr-2Mo supports
Blades, stage Ti-6Al-2Sn- Colloidal Rear shaft Inconel 718 Cr plate
3 4Zr-2Mo graphite
Cu-Ni-In Ni-aluminide
Molydisulfide Spline adapter AMS 6381 None
Blades, A286 Colloidal Spools, stages Ti-6Al-2Sn- Ni-aluminide
stages 10–12 graphite 3–9 4Zr-2Mo
Blades, A286 Colloidal Spools, stages Inconel 718 Ni-aluminide
stages 13–14 Inconel 718 graphite 11–13

Nowadays, annular combustors have been developed (see Fig. 2.18), although a
multiple-chamber design afforded higher safety standards, since the failure or the
malfunctioning of one component element is not necessarily jeopardizing the whole
combustor. Air enters into the chamber at different places along the combustor,
thanks to the presence of calibrated gaps left among the metal sheets of the annular
chamber walls. A double-wall architecture is mandatory for this part of the engine
where the peak temperatures of the whole thermodynamic cycle are achieved and an
effective cooling is thus paramount, and in fact a fraction of the air flux from the
compressor is used to cool down the combustor walls. Different operating regimes
such as take-off, cruise, and landing would feature specific ranges for the actual air
mass flow and of the air/fuel ratio. Requirements for combustor materials are:
• Oxidation/corrosion resistance. In the steady cruise regime, gas phase corrosion,
i.e., oxidation and vanadation, may occur. During the engine shutdown and start-
up phases, deposition of sodium sulfate may occur, introducing an additional
surface instability process, i.e., hot corrosion (see Sect. 8.2.5).
• Thermal fatigue resistance.
• Thermal stability.
30 2 Gas Turbine Aero-Engines

Fig. 2.17 Scheme of an early multiple-chamber system with detail regarding the single combustion
chamber component and indication of the main functions of the inlet airflow coming from the
compressor. (Rolls-Royce 1996, images courtesy of Rolls-Royce plc)

Fig. 2.18 Scheme of an annular combustion chamber. (Rolls-Royce 1996, image courtesy of
Rolls-Royce plc)

According to these requirements, it is possible to identify some solutions that are

highly reliable and, thereby, recommended for the fabrication of combustion
chambers.
2.5 Turbines 31

• Co-based superalloys (see Sect. 6.2). They usually display a lower strength than
the companion materials, Ni-based superalloys (see Sect. 6.4). However, their
strengthening mechanisms, mostly based on carbide precipitation, together with a
comparatively lower coefficient of thermal expansion, render these materials
preferable for the combustion chambers.
• Ceramic-based thermal barrier coatings (see Sect. 7.3.2). Coatings made of ZrO2
with suitable stabilizer additions, like Y2O3 and ScO2, are a standard choice in
this respect; thanks to the thermomechanical compatibility with the underlying
superalloy, they guarantee for these critical parts.

2.5 Turbines

The turbine converts the kinetic energy of the gases coming from the combustor into
mechanical energy used to drive possible appliances and, most importantly, the
compressor rotor stages and the fan, if present. In the turbine, a slight reduction, with
respect to the combustion chamber, in the gas pressure and temperature takes place.
The hot gases are at temperatures above 1500  C, and the flow rate may exceed
750 m/s. The energy conversion process involves high stresses and high rotational
speeds. Altogether, this renders the working conditions in the high-pressure stage of
the GT extremely harsh. Vanes (static blades) and turbine blades of the first high-
pressure stage are the most critical components of the whole engine, as concerns
materials requirements (see Fig. 2.19). In fact, they are directly exposed to the jet of
hot gases leaving the combustor, whose temperature (TET, see Fig. 2.20) has
continuously increased over the years.
As depicted in Figs. 2.10 and 2.11, different architectures for jet engines are
possible, with a different number of shafts and therefore of turbine stages. High
bypass ratio fan engines may be equipped with a third intermediate-pressure turbine
stage to operate the fan. This further stage is located between the high- and
low-pressure turbines (vide infra, Fig. 2.22), according to a triple-spoon arrangement
(Rolls-Royce 1996).
High peak temperatures and high-temperature gradients are not the only critical
conditions for these turbine components, which have to withstand the high level of
stresses, related to both the rotational motion and differential heating. Nickel-based
superalloys (see Sect. 6.4) are the obvious choice, not only for the blades but also for
other critical components of the turbine, requiring excellent strength associated
with creep and fatigue resistance. Optimal performances have been achieved thanks
to the use of newly developed casting techniques, leading to the manufacturing of
directionally solidified and single-crystal components (see Sect. 6.5). It is important
to highlight also the role of inner cooling channels. In cooled blades, the energy
transferred from the flame to the material is removed so fast and efficiently that the
alloy can even stand working temperatures that are important fractions of its
incipient melting temperature.
32 2 Gas Turbine Aero-Engines

Fig. 2.19 Twin turbine and relevant double-shaft arrangement. (Rolls-Royce 1996, images cour-
tesy of Rolls-Royce plc)

Fig. 2.20 Temperature capability of the blade superalloys and dependence also on the blade
design: cooled or solid blades. In the graph the evolution of the take-off TET of civil aero-
engines is considered. (Redrawn from Mukherji et al. 2011)
2.6 State of the Art and Future Trends 33

Eventually, surface reactions, induced by the interactions with the aggressive

atmosphere, require an intrinsic resistance of the alloy against high-temperature
corrosion (see Sect. 8.2). This feature can be improved with suitable coating
systems. The already mentioned diffusion coatings, based on nickel aluminides, or
the MCrAlY overlay coatings (M being Ni and/or Co and/or Fe) are excellent
choices (see Sect. 7.3.1). As concerns coatings, turbine high-pressure blades have
been in some engines also protected with thermal barrier coatings, using specific
deposition techniques based on electron beam physical vapor deposition (see Sects.
7.2.2 and 7.3.2).

2.6 State of the Art and Future Trends

The aerospace industry has a great impact in terms of employees and output value.
As an example, 14.9 billion USD in 2004 were invested in aviation gas turbine
engines, and 3.7 billion USD have been used for military aircraft applications, while
11.2 billion USD have been invested in the field of commercial aviation (Giampaolo
2006). Aerospace industry permeates several other industries and is used as a
fundamental benchmark for experimenting and testing new materials and technolo-
gies. Some of the most important driving forces pushing the efforts for innovation
are associated with the improvement of aircraft performances, concerning safety and,
in more recent times, pollution reduction. The reduction in emissions has become a
major issue, considering that commercial aviation alone is responsible for the release
in the atmosphere of 705 million of metric tons of CO2 per year, corresponding to
about 2% of the yearly global carbon emissions. Moreover, there is evidence that the
greenhouse gases in jet fuel have greater effects on the atmosphere, since they are
released at elevated altitudes (Martin 2016). The typical emission profile of a
two-engine aircraft during 1 h flight with 150 passengers is given in Fig. 2.21.
Pollutant substances and greenhouse gases in engine exhaust are present: CO2, CO,
H2O, NOx, SO2, various hydrocarbons, and particulate matter, just to mention the
main constituents. Some of these substances had attracted already investigation
interests and research efforts, mainly in connection with their influence on corrosion
(see Sect. 8.2) and erosion (see Sect. 7.3.5) aspects in gas turbine engines.
The European market is one of the largest for commercial aviation. According to
the data reported by international agencies, like EU28 and European Free Trade
Association (EFTA), and by the United Nations Framework Convention on Climate
Change (UNFCCC), CO2 emissions in the atmosphere have experienced a 77%
growth between 1990 and 2005. Starting from the present aviation traffic forecast,
notwithstanding the improvement rate of innovative technologies, an unavoidable
44% increase in CO2 emissions from 2005 to 2035 period (European Commission
2016) is expected. For this reason, several international programs aiming at reducing
atmospheric pollution are being conducted. ACARE (Advisory Council for Aviation
and Innovation in Europe) Vision 2020 and Flightpath 2050 are two examples of
European collaborative programs that set the targets for the future of the air
34 2 Gas Turbine Aero-Engines

Fig. 2.21 Typical emission profile of a two-engine aircraft during 1 h flight with 150 passengers.
(Redrawn from European Commission 2016)

transportation, also as concerns climate change and atmospheric pollution reduction.

The two programs will set ambitious goals in terms of emissions cut:
• 90% reduction in NOx emissions (15% achieved through operational efficiency
and 75% from technology improvements)
• 75% reduction in CO2 emissions per passenger per kilometer
• 25–30% overall CO2 emissions reduction
For meeting these targets, multidisciplinary approaches are necessary. The main
pillars on which research efforts are based can be summarized as follows:
• Average improvement in fuel efficiency, by at least 1.5% per year from 2009 to
2020
• Development of a more efficient air traffic management system
• Improvement in aircraft operational efficiency
• Technology improvement including sustainable low-carbon fuels and biofuels
As concerns low-carbon fuels, in 2009 the incidence of biofuels in the total fuel
consumption was 0.05% only. A widespread use of biofuels for aircraft engines is
not that easy to achieve, first of all due to the competitive demand by other
transportation sectors. Furthermore, the hesitation of the industry concerning invest-
ments in dedicated production facilities for biomass to liquid fuels transformation
can be justified by the limited foreseen supply of 0.05 million tons by 2020
(European Commission 2016). Therefore, biofuels introduction is still far from its
potential target, and the main approach behind the development of more
environmental-friendly jet engines is rather through the improvement of their effi-
ciency. This has been highlighted already in the previous sections of this chapter,
2.6 State of the Art and Future Trends 35

with reference to the improvement in the engine performances demonstrated by the

evolution of parameters like TET, OPR, SFC, and bypass ratio.
However, increasing TET will result in higher emissions that can be limited by
the adoption of catalytic combustors, capable to reduce the NOx formation in the
combustion chamber directly. Furthermore, higher TET would require the develop-
ment of more resistant thermal barrier coatings and more efficient cooling design for
the components, in order to withstand the more demanding operating temperatures.
There are several examples of newly developed engines with improved perfor-
mances. One of these is the Pratt & Whitney F135 engine for the F35 military
aircraft. This engine is characterized by an OPR of 28 and is capable of 190 kN of
takeoff thrust and, in general, maximum efficiency under different operational
conditions: conventional take-off/landing (CTOL), carrier variant (CV), and short
take-off/vertical landing (STOVL) (Pratt and Whitney 2012). In commercial aviation
the latest Airbus A380 is powered by the General Electric-Pratt & Whitney GP 7000
or by the Rolls-Royce Trent 900 engines (Giampaolo 2006). The GP 7000 has a
maximum thrust of 340 kN with an OPR of 43.9 and a bypass ratio of 8.7:1 (MTU
Aero Engines 2016). The Trent 900 is a three-shaft turbofan engine capable of a
maximum thrust of about 342 kN. Its architecture consists of an eight-stage inter-
mediate-pressure compressor, a six-stage high-pressure compressor, a single-stage
high-pressure turbine, a single-stage intermediate-pressure turbine, and a five-stage
low-pressure turbine. Its OPR is 39, and the bypass ratio reaches a value of 8.5:1
(Rolls-Royce 2015). A further evolution of this engine, the Rolls-Royce Trent XWB
(see Fig. 2.22) is a three-shaft turbofan engine capable of a thrust value between

Fig. 2.22 Modern Rolls-Royce Trent XWB. (Rolls-Royce 2016, images courtesy of Rolls-Royce
plc)
36 2 Gas Turbine Aero-Engines

374 and 432 kN (84,000–97,000 lbf). A large diameter swept fan and an eight-stage
intermediate-pressure compressor followed by a six-stage high-pressure compressor
characterize the front part of the engine. The combustor has an innovative design,
incorporating advanced ceramic coatings that can withstand combustion tempera-
tures exceeding 2700  C. A six-stage low-pressure turbine follows a single-stage
high-pressure turbine. The power generated by the turbine system can reach values
as high as 50,000 hp. Its overall pressure ratio (OPR) can climb up to 50, well above
those given in Fig. 2.8. The bypass ratio of the engine is 9.6:1. These features allow
the Trent XWB to improve several performances, including a fuel consumption
reduction by 10% (Rolls-Royce 2016).
Future developments are already on their way, and the engine of the future should
be lighter, less noisy, and more energy-efficient than conventional engines used
today. Pratt & Whitney developed a new engine that might save up to 20% of fuel,
with respect to the present average engine. One of the features that enable to achieve
this result is a gearbox that slows down the fan. Meanwhile, CFM International is
working on a conventional turbofan architecture engine that will reach similar results
but without adding weight and drag through a gearbox. The increase in efficiency is
obtained thanks to lightweight composite materials used in several components, such
as carbon fiber-based fan blades (Martin 2016). Eventually, Rolls-Royce is devel-
oping an innovation plan, based on Advance® and UltraFan® engines (see Fig. 2.23).
Advance® will be ready for the market by 2020 and can be defined as a hybrid
engine that will be manufactured with novel materials, like the light ceramic matrix
composites (CMCs), 3D-printed components, and the innovative low-pressure sys-
tem with carbon-titanium (C-Ti) composite fan system, including fan casing and
blades. The engine will provide a 20% better efficiency as compared to the early
Trent engine, fewer parts, and lower weight. UltraFan® engine will be available
starting from 2025. One of the most impressive features of this engine will be its
OPR, reaching a value of 70:1. It will be equipped with an innovative gearing
system, i.e., geared architecture, located between the fan and the intermediate-
pressure compressor. This system will be able to ensure that the fan runs at the
optimum speed in any condition. Further improvements in the C-Ti composites will
allow the removal of the thrust reverser and, thereby, a slim-line nacelle system.
When compared with the first Trent engine, the UltraFan® engine will provide a 25%
increase in fuel efficiency (Rolls-Royce 2017).
Aircrafts and their components are usually subjected to research and development
cycles, whose duration can reach 20 years. For this reason, research efforts are based
on a long-term planning approach that provides continuity over several years. The
European Union Clean Sky research program comprises several projects aiming to
meet both performance increase and environmental challenges. Two examples of
R&D projects in this context are:
• Advanced low-pressure systems (ALPS). These engines are intended to demon-
strate the feasibility of innovative lightweight materials for the production of
engine components such as C-Ti composite fan blades and coatings. The use of
these materials will be expected to lead to a weight saving of around 700 kg on a
twin-shaft engine.
2.6 State of the Art and Future Trends 37

Fig. 2.23 Future Rolls-Royce engines. (Rolls-Royce 2017, images courtesy of Rolls-Royce plc)

• Counter-rotating open rotor engines (CROR). This aero-engine family should

reach a 30% reduction in emissions compared to a year 2000 benchmark engine.
The relevant noise level should be comparable to that of a modern turbofan
engine. Airbus and Rolls-Royce are currently involved in research programs on
this propulsion system (European Commission 2016).
CROR and ALPS are just two of the many research programs currently under
development. Engine producers are now facing many other challenging tasks in the
search for the “best engine:”
• Advanced cycles.
• Airframe optimization.
• Hybrid systems with separate power and thrust generation. They
are turboelectrical hybrid propulsion system with a gas turbine for electricity
generation and electrically driven fans coupled to batteries for energy storage.
• Pressure gain combustion.
38 2 Gas Turbine Aero-Engines

Considering the above scenario, it is possible to conclude that gas turbine engines
will still have a lot of potential in the future of aircraft propulsion. Furthermore, they
will generate the power for aircraft propulsion to a great extent, possibly through
hybrid systems and electrically driven engines (Wenger 2014).

References

Boyce M P (2012) Gas Turbine Engineering Handbook 4th edn. Elsevier, Oxford
Cumpsty N (2003) Jet Propulsion: A Simple Guide to Aerodynamic and Thermodynamic Design
and Performance of Jet Engine 2nd edn. Cambridge University Press
European Commission (2011) Flightpath 2050 Europe’s Vision for Aviation. Available via DIA-
LOG. http://ec.europa.eu/transport/modes/air/doc/flightpath2050.pdf. Accessed 17 Aug 2017
European Commission (2016) European Aviation Environmental Report 2016. Available via
DIALOG. https://ec.europa.eu/transport/sites/transport/files/european-aviation-environmental-
report-2016-72dpi.pdf. Accessed 17 Aug 2017
Giampaolo A (2006) Gas Turbine Handbook Principles and Practices. The Fairmont Press Inc.,
Lilburn
Gunston B (1995) The Development of Jet and Turbine Aero Engines. Patrick Stephens Ltd Publ
Kear B H (1986) Advanced Metals. Scientific American 220: 130–139
Martin R (2016) The Race for the Ultra-Efficient Jet Engine of the Future. In: Sustainable Energy.
MIT Technology Review. https://www.technologyreview.com/s/601008/the-race-for-the-ultra-
efficient-jet-engine-of-the-future/. Accessed 15 Aug 2017
Meetham G W (1981) The Development of Gas Turbine Materials. Applied Science Publishers Ltd,
London
MTU Aero Engines (2016) GP7000 turbofan engine. Available via Dialog. http://www.mtu.de/
fileadmin/EN/7_News_Media/2_Media/Brochures/Engines/GP7000.pdf. Accessed 07 Sept
2017
Mukherji D et al (2011) Beyond Ni-based Superalloys: Development of CoRe-based Alloys for Gas
Turbine Applications at Very High Temperatures. International Journal of Materials Research
102(9): 1125–1132
Pratt & Whitney (2012) F135 Specs Charts. Available via DIALOG. http://www.pw.utc.com/
F135_Engine. Accessed 06 Sept 2017
Reed R C (2006) The Superalloys-Fundamentals and Applications. Cambridge University Press,
New York
Rolls-Royce (1996) The Jet Engine 5th edn. Rolls-Royce plc, Derby
Rolls-Royce (2015) Trent 900. Available via DIALOG. https://www.rolls-royce.com/products-and-
services/civil-aerospace/airlines/trent-900.aspx#training. Accessed 07 Sept 2017
Rolls-Royce (2016) Trent XWB. Available via DIALOG. http://www.rolls-royce.com/site-ser
vices/images/trent-xwb-infographic.aspx. Accessed 17 Aug 2017
Rolls-Royce (2017) Pioneering intelligent innovation for our customers. https://www.rolls-royce.
com/products-and-services/civil-aerospace/future-products.aspx#/. Accessed 23 Aug 2017
The Washington Post (2017) The next generation of eco-friendly airplanes has arrived. . .quietly.
http://www.washingtonpost.com/sf/brand-connect/wp/enterprise/the-next-generation-of-eco-
friendly-airplanes-has-arrivedquietly/. Accessed 17 Aug 2017
Wenger U (2014) Rolls-Royce Technology for the Future Aircraft Engines. Lecture for RAeS,
DGLR, VDI at the Hamburg University of Applied Sciences, Hamburg, 20 March 2014
References 39

Further Reading

Connors J (2010) The Engines of Pratt & Whitney. American Institute of Aeronautics and
Astronautics, Inc.
Hill P G, Peterson C R (1992) Mechanics and Thermodynamics of Propulsion 2nd edn. Pearson
Kerrebrock J L (1992) Aircraft Engines and Gas Turbines 2nd edn. The MIT Press
Soares C (2015) Gas Turbines: A Handbook of Air, Land and Sea Applications. Elsevier
Chapter 3
Alloys for Aircraft Structures

3.1 Introduction

As seen in Chap. 1, the Flyer I of the Wright brothers was mostly made of wood and
fabric. However, in addition to steel, used in the wing struts and the bracing wires, an
aluminum alloy was used for the crankcase of the piston engine, in order to reduce its
weight. So, aluminum started to fly since the very early beginning of modern
aviation. After this start, aluminum attracted further interest and research efforts,
since the need for new materials capable to resist the mechanical and thermal stresses
associated with the increasing flight performances emerged soon (see Chap. 1).
Moreover, the shortage of wood, particularly spruce and willow tree, caused by
the massive aircraft production during World War I, created the conditions for the
development and use of light alloys in aircraft structures. Light refers to the com-
paratively low densities of alloys based on elements like aluminum, magnesium, and
titanium. Although definitely higher than the average density of wood, which was
the main structural material of the early airplanes, the densities of aluminum alloys
are still absolutely acceptable, even considering their effect on the specific structural
properties, i.e., properties averaged by the material density. Aluminum alloys were
among the first metallic alloys used in aircraft structures, and they still are the most
used alloys in this field. Magnesium alloys are other interesting systems, particularly
now that compositions displaying acceptable corrosion resistance and interesting
mechanical properties have been developed. These alloys are nowadays gaining an
increasing importance in the aeronautic field, where actually had already been used
in the past. For them, some issues are still standing, like the search for mechanically
workable compositions, which are introducing some obstacle to a more rapid
widespread use of these materials. Titanium alloys, owing to their specific proper-
ties, are used not only for aircraft structures but also for engine parts. Therefore, they
will be treated separately in Chap. 4. Laminate composite materials instead are
included in this chapter. Many of them are based on aluminum alloys and represent
a very important class of aerospace structural materials.

© Springer Nature Switzerland AG 2020 41

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_3
42 3 Alloys for Aircraft Structures

Fig. 3.1 Structure and main components of a commercial aircraft. (Redrawn from Boeing 2008)

The structure of modern aircrafts, although recalling to some extent early air-
planes, exhibits obvious aerodynamic optimization. Commercial and military air-
crafts share the airframe shape and the denomination of the main components
(Fig. 3.1), although major differences in size, geometry, and structural properties
are necessary to comply with the specific operational requirements.
The main structural parts of an airframe are the fuselage, the wings, the landing
gear, and the flight control surfaces.
The wings are meant to provide the necessary lift but are also subjected to other
mechanical stresses, like, for instance, the weight of fuel and engines. Their design
and internal structure are therefore conceived to fit the specific needs of any aircraft
(Fig. 3.2). Similar principles apply, obviously with different parameters, to fuselage,
vertical stabilizers, and all the other parts of an aircraft (Fig. 3.3). In general terms,
the components whose failure would endanger the safety of the whole aircraft are
called primary structures, being secondary structures all the other. Fuselage, wings,
empennage, and landing gears are typical primary structures, and for some of them,
aluminum alloys are an excellent choice.

3.2 Aluminum Alloys

3.2.1 Introduction

Aluminum is a post-transition metal with atomic number 13 (13th group, 3rd period)
of the periodic table. It is characterized by a face-centered cubic ( fcc) crystal lattice,
a density of 2.70 g/cm3, and a Young’s modulus of 70 GPa. The name aluminum
derives from the term alum, used by Greeks and, afterward, by the Romans with
3.2 Aluminum Alloys 43

Fig. 3.2 Wing components and structure. (Redrawn from Federal Aviation Administration 2012)

Fig. 3.3 (a) Semimonocoque tail part of the fuselage; (b) structure of the vertical stabilizer.
(Redrawn from Federal Aviation Administration 2012)

reference to a bitter salt, based on aluminum double sulfate hydrated (e.g., KAl
(SO4)2 H2O), used in medicine as an astringent but also for skin tanning and glass
production. The first who actually identified the presence of aluminum in alum salt
was Sir Humphry Davy at the beginning of 1800, although he was not able to isolate
it. In 1807, he proposed the name aluminium for the metal. Shortly after, the name
aluminium was adopted by the IUPAC. However, in 1925 the American Chemical
Society introduced the name aluminum, still used in the American English literature
(Totten et al. 2003). The driving force for the development of Al-alloys for aerospace
structures has been the improvement in static strength, as well as fracture toughness
and resistance to crack growth, particularly in fatigue regimes, damage tolerance,
and stress corrosion. As shown next, an incremental progress and an alloy-targeted
design have been particularly successful so that Al-alloys are still strategic materials
44 3 Alloys for Aircraft Structures

Fig. 3.4 Pourbaix diagram of pure aluminum. (Redrawn from McCafferty 2010)

for aerospace industry. This is also due to common features of these alloys, which
can be largely ascribed, and to some extent retrieved, in pure aluminum, by far their
majority component:
• Acceptable corrosion resistance.
• Good fatigue resistance, due to the high intrinsic ductility of the fcc structure
given by the high number of slip systems.
• Good oxidation resistance.
• Good specific mechanical resistance.
• High specific stiffness (E ¼ 70 GPa, average density of the alloys ¼ 2.768 g/cm3).
In standard conditions, the presence of a self-passivating and self-healing Al2O3
surface layer is the main reason for the fairly good surface stability of aluminum
alloys, resulting in their good resistance to dry oxidation.
For similar reasons, it is possible to see from the Pourbaix diagram in Fig. 3.4 that
Al-alloys still provide a positive response in not too aggressive corrosive environ-
ments, like aqueous solutions with a pH in the 3.9–8.6 range that do not determine
localized corrosion phenomena, like pitting and crevice corrosion, discussed in
Chap. 8 (ASM International 2003). As concerns fatigue resistance, Al-alloys are
extremely sensitive to the presence of hard inclusions, which may act as stress
concentrators. An intrinsic beneficial factor, as concerns fatigue resistance, is the
face-centered cubic structure of aluminum and relevant large number of slip systems
that help in accommodating local stresses by plastically straining, thus keeping their
level relatively low.
3.2 Aluminum Alloys 45

The availability of aluminum and its alloys has remarkably influenced the
aerospace industry since the beginning. Aluminum alloys were used during the
World War I for the production of structural parts of the Zeppelin dirigibles and
during the World War II for military and civil aircrafts. In these early applications,
the hardenable Al-3.5Cu-0.5Mg-0.5 Mn alloy, named Duralumin, developed in the
1906 by the German chemist Alfred Wilm, was used. Hardening of Duralumin was
attained thanks to a precipitation treatment, which induced in the alloy properties
interesting not only for aerospace but also automotive applications (Lumely 2011).
Notwithstanding the intrinsic corrosion resistance, a critical aspect of this alloy, that
emerged soon in outdoor applications, was its tendency to exfoliation corrosion (see
Sect. 8.1.4), which of course would not be acceptable for any application involving a
long-term exposure to external atmosphere and meteorological events.
The problem was successfully solved with two alternative approaches, developed
in the late 1920s: cladding the aluminum alloy with pure aluminum sheets and alloy
protection through the formation of a surface film made of reaction products, using
an anodizing process. Concerning this latter process, by means of appropriate bath
composition and deposition parameters, it is possible to produce protective surface
layers thicker than those forming spontaneously by the self-passivation tendency of
the alloy. Moreover, by tuning the composition of the bath, it is possible to obtain
deposits containing chemical agents, actively enhancing the corrosion resistance
(Thompson et al. 1999). Thanks to these fundamental improvements, aluminum
alloys rapidly became, and still are, the most used materials for aircraft structures, as
outlined in the following.
Concerning the production of Al-alloys, raw materials usually are oxide-based
ores with various grades of purity and levels of hydration. Bauxite, from the French
village Les Baux, where alumina was first discovered, is the most important exam-
ple. It is a sedimentary formation containing different aluminum hydroxides, like
AlO(OH) and Al(OH)3. Chemical and mechanical enrichment of the ores is a
common practice to increase the concentration of Al2O3 available to extraction
(Totten et al. 2003). The main steps of aluminum extraction, still based on the
Bayer process, developed and patented by Karl Bayer in 1888, are illustrated in
Fig. 3.5. The output of the first step is hydrated alumina: Al2O3 ∙ H2O. Pure alumi-
num requires the electrolytic reduction of Al2O3 dissolved in molten cryolite
(Na3AlF6, Hall-Héroult process, see Fig. 3.5). Charles Hall and Paul Héroult devel-
oped independently this process in 1886, in Ohio (USA) and France, respectively
(ASM International 1991). The overall electrochemical reaction involved in the
extraction of aluminum is:

1 3 3
Al2 O3 ðdissolvedÞ þ CðsÞ ! AlðlÞ þ CO2 ðgÞ ð3:1Þ
2 4 4

The electrolytic dissolution process is conducted at about 960  C in the Hall-

Héroult cells. Typically, for each kilogram of primary aluminum, 1.93 kg of Al2O3 is
needed, while 0.40–0.45 kg of C and 1.5 kg of CO2 are eliminated as secondary
products (Lumely 2011).
46 3 Alloys for Aircraft Structures

Fig. 3.5 Aluminum production route from ores to primary aluminum (Carter and Norton 2007;
Lumely 2011)
3.2 Aluminum Alloys 47

Fig. 3.6 Percentages of different materials used for the A380 Airbus, involving and promoting the
development of the new materials and relevant processing routes. (Redrawn from Williams and
Starke 2003)

Components made of aluminum alloys can sum up to more than a half of the total
weight of a commercial airliner (see Fig. 3.6). Moreover, the possibility of varying
and tailoring alloy’s properties by suitable compositional changes, processing routes
and post-processing thermomechanical treatments, is regarded as an attractive aspect
of these materials, in addition to low density, a comparatively low cost, good
availability, ease of manufacturing, reliability, and predictability of the in-service
performances. Indeed, the leading role of Al-alloys in aerospace applications is
mainly due to very well established alloy design principles. It is possible to obtain
materials with specific properties for the different part requirements (see Table 3.1).
The density of Al-alloys, a primary selection criterion for these and other
aerospace materials, varies from 2.643 g/cm3 for commercially pure aluminum up
to 2.803 g/cm3 for highly alloyed compositions, like Al-Zn alloys. Metal working
technologies, like milling, extrusion, and forging, lead to work-hardened products
with higher specific strength (Starke and Staley 1996). The development occurring
over the years that followed the end of World War II required a change of strategy
after the accidents involving the Comet airliners at the Ciampino airport in Rome
(1954, see Chap. 1). A premature fatigue failure was the main reason for the two
accidents. However, an insufficient damage tolerance of the Al-Zn alloy, namely, the
7075 material used for the airframe of the Comet, was regarded as an additional
factor behind the disasters. Therefore, these 7XXX alloys (see Sect. 3.2.3) were
banned for several years from aerospace applications, and lower strength, but
more damage-tolerant, Al-Cu alloys (precisely the 2024-T3 and 2014-T6) were
used instead, with the exception for those components requiring highest tensile
strength. The progress of commercial aviation has been accompanied by the request
for larger payloads, larger number of passengers, and, thereby, larger airliners.
Thicker sections of the structural parts started to be produced. At the same time,
48 3 Alloys for Aircraft Structures

Table 3.1 Summary table of the main properties of Al-alloys that can be attained by changing the
composition and/or the microstructure of the alloy
Property Desired microstructural feature Function
Creep Thermally stable particles within the Inhibition of grain boundary sliding
resistance matrix and on the grain boundary and coarse microstructure formation
Ductility and Fine structure with clean grain bound- Encourage plasticity and work-
toughness aries and absence of shearable precipi- hardening. Inhibition of void forma-
tates or large particle tion and growth
Fatigue crack Fine grain size with absence of Prevent strain stress concentration,
initiation shearable particles. No surface defects strain localization, and surface slip
resistance steps
Fatigue crack Large grain size with shearable particles Encourage crack closure, branching,
propagation and no anodic phases or hydrogen traps deflection, and slip reversibility
resistance
Pitting Absence of anodic phases Prevent preferential dissolution of
second phase particles
Strength Fine grain size with uniform dispersion Inhibition of dislocation motion
of small, hard particles. Fine structure
with clean grain boundaries and absence
of shearable precipitates or large
particles
Stress corro- Hard particles and no anodic phases or Homogenization of slip and preven-
sion cracking interconnected hydrogen traps tion of crack propagation due to
and HE anodic dissolution or HE
HE hydrogen embrittlement (Williams et al. 2003)

the incidence of damages due to stress corrosion started to be observed, with micro-
cracks developing particularly when the load was applied along the short transverse
direction. The most used alloys were 2014-T3, 7075-T6, and 7079-T6, where T3 and
T6 represent two specific tempers, used to refine microstructure and relevant prop-
erties of aluminum alloys, as described in upcoming sections (Tables 3.9 and
Table 3.10). Subsequently, thanks to alternative tempers for the 7075 alloy,
codenamed T73 and T76, its stress and exfoliation corrosion problems were largely
solved. These two tempers involve combinations of artificial aging at low and, as a
subsequent step, at high temperature. The high-temperature treatment would gener-
ally result in a slight overaging of the alloy, with a consequent softening. However,
the alternative beneficial effect of the treatment is the strengthening of the grain
boundary region that enhances the fracture toughness of the alloy. This effect is
determined by the enrichment of the grain boundary with Cu-rich precipitates.
Nobler precipitates are less prone to microstructural changes and provide more
effective obstacles to crack propagation. As an indirect toughening mechanism,
there is also the lower affinity of the grain boundary precipitates and of the metallic
matrix itself to hydrogen, whose embrittling action is thus neutralized or largely
limited. In the A380, for assembling parts of the fuselage, laser beam welding
(LBW) is used, with consequent requirements for a good weldability of the alloys.
For this reason, Al-Mg-Si (6XXX alloys, see Sect. 3.2.3) materials have been
considered. They exhibit excellent welding properties, and, in addition to that,
3.2 Aluminum Alloys 49

they have a comparatively lower density and price, as compared to Al-Cu materials.
The reference alloy 6056-T78, featuring a relatively low density, i.e., 2.702 g/cm3, in
comparison with 2.781 g/cm3 of the 2024 alloy, is also a laser-weldable product,
developed at a later stage to comply with the elevated damage tolerance require-
ments. The outcome of the evolution of this material was the 6156 alloy, extensively
used in lower shell fuselage applications. It is worth saying that also the temper
(T78) developed for the 6XXX alloys played an important role in overcoming the
problem of intergranular corrosion, which was observed in T6-treated parts made of
the same materials. In the A380, notwithstanding the availability and actual use of
other materials, like composites used already in the Boeing 787 Dreamliner (see
Fig. 1.11), Al-alloys still retain a leading role, as shown in Fig. 3.6.
An indirect proof of the technological reliability of standard aluminum alloys,
like the 2024 and 7050 materials, is provided by the improvements attained thanks to
new processing and manufacturing steps, still using the same alloy compositions.
This choice had to be made on several occasions to comply with requests from
aircraft producers in terms of better properties but also for a faster and, therefore,
cheaper production. Larger monoliths, machined from bulk thick plates and beams,
have replaced assemblies and built-up structures. Machining may involve the
removal by up to 95% of the original volume of the part. On the other hand,
stress-free components can be obtained in this way, and the scraps can be recycled
straightaway by remelting. Through the optimization of the alloy composition, in
association with controlled conditions during temper treatments, important enhance-
ment in alloy response to rolling can be achieved. Larger and thinner Al-alloy sheets
are very much interesting for the process and manufacture optimization that they can
bring about (Heinz et al. 2000).
Another issue that promoted important research efforts for developing special
aluminum alloys was the expected heating associated with air drag in supersonic
flight, which started to be considered also for commercial airliners. The Concorde
project, started in 1962, required the development of quite stable alloys, not affected
by any kind of long term, in-service aging, due to peak temperatures close to 100  C,
for exposure times of 20,000 h, out of the 50,000 h expected total service lifetime, at
the highest Mach 2 speed. As seen in Chap. 1, based on the experience gained with
military aircraft, Mach 2 is still a speed affordable to Al-alloys. The starting point of
this material development was an alloy, codenamed RR58, originally designed for
components of the compressor of the early gas turbine jet engines. The alloy, that in
the end was employed for the fuselage panels and other structural parts of the
Concorde, has a modified composition with respect to the parent material, i.e., the
2618 alloy, in which copper was the main alloying element. In particular, the ratios
of copper to iron and nickel were fixed in the deliberate attempt to limit the formation
of equilibrium intermetallics, typical of the overaging condition of these hardenable
alloys. Moreover, an accurate microstructural refinement, inducing large grain size,
was pursued to optimize creep resistance and to reduce exfoliation and stress
corrosion. Subsequently, the alloy development regarded other critical mechanical
properties, like strength and fracture toughness, leading to the design of several
alloys not only of the 2XXX (Al-Cu, see Sect. 3.2.3) but also of the 7XXX (Al-Zn,
50 3 Alloys for Aircraft Structures

Fig. 3.7 General design aspects for the A380 Airbus aircraft, involving and promoting the
development of the new materials and relevant processing routes. (Redrawn from Lequeu et al.
2007)

see Sect. 3.2.3) series, like the 7050-T74, 7050-T76, and 7150-T6, particularly
recommended for the upper wing skin. For the Al-Cu series, it is worth mentioning
the 2324-T39 and 2224-T351X materials, developed for the lower wing structures.
The more recent developments of these alloys have been triggered by the design
requirements for large size airliners, like the Airbus A380. Larger size and loads
involved important improvements in static mechanical strength and damage toler-
ance, to be developed according to the leading selection criteria summarized in
Fig. 3.7 for each specific component (Lequeu et al. 2007).
Moreover, thanks to a progressive implementation approach in alloy develop-
ment, novel material products came up that could be processed relying on the
knowledge background accumulated over the years. This is a fundamental prereq-
uisite for a rapid technological transfer from research to applications. In this regard,
to complete the picture of the main milestones in the development of aluminum-
based materials, Al-Li alloys will be mentioned as well, together with the most
recent class of aluminum alloys, containing scandium among alloying elements, and
laminate composites, using aluminum as main metallic constituent, as illustrated in
Sect. 3.4.
Al-alloys are generally divided into two large families: cast (see Sect. 3.2.2) and
wrought (see Sect. 3.2.3) alloys. A great number of both cast and wrought alloys
respond to thermal treatments that are essentially based on phase solubility, i.e.,
solutioning, quenching, and precipitation, and age hardening. These alloys are said
to be heat treatable (or hardenable), since different phases and, thereby, different
properties can be induced into the alloy through the control of the solid-state
precipitation. Duralumin, the Al-3.5Cu-0.5Mg alloy (alloy 2017, according to the
code that will be introduced in Sect. 3.2.3), already mentioned with reference to the
early use of an aluminum alloys in aircraft structures (see Sect. 1.1), can be regarded
3.2 Aluminum Alloys 51

as the prototype of the heat treatable alloys. If the composition of the alloy does not
offer this chance and no precipitate phases can be formed in the alloy matrix, the
alloy is said to be non-heat treatable. Both wrought and cast alloys can be classified
according to an alloy nomenclature system. This coding system was developed by
the Aluminum Association (ASM International 1991). The nomenclature of both
wrought and cast alloys, used already in some part of this section dealing with
specific alloys, will be discussed in details in the relevant sections of this chapter.
The aluminum parts and components of aircrafts are supplied either as wrought or
cast products, these latter typically requiring some machining and surface finishing
before installation. As already mentioned, the fuselage, the wings, the empennage,
and the supporting structure of an aircraft are regarded as primary structures, since
their failure would endanger the whole aircraft. Usually, wrought alloys are used for
these parts, although for military airplanes highly controlled casting products are on
some occasion employed. For secondary structures both cast and wrought products
are used, depending on the shape and required part finishing.
Before entering this broad subject, a few bibliographic notes, which might
interest the reader to widen the knowledge on this most important class of alloys,
concern some of the properties of aluminum alloys that render these materials
excellent candidates for aerospace applications and to complete the picture emerging
from the former sections. Metal handbooks, such as that of ASM International
(2015), offer a wide description of Al-alloy properties, relevant compositions, and
tempers developed to attain alloy properties suitable for specific applications.
Another essential reading is the handbook edited by Davis (2001). The textbook
of Polmear (2006) is particularly recommended for a comprehensive treatment of the
main themes concerning aluminum but also other light alloys, like magnesium,
titanium, and derived alloys. A particular focus on the aerospace applications of
all these light alloys can be found in the monographic chapter by Prasad and
Wanhill (2017).

3.2.2 Cast Alloys

Aluminum and its alloys can virtually be subjected to any casting and forming
process. However, early casting of aluminum has been characterized by several
problems. Because of gas pick-up during melting, castings were often characterized
by the presence of porosity and other defects, easily visible also on eye inspection.
Porosity in aluminum castings is still an issue, and several techniques have been
developed over the years to control its impact on the final products, if not to
eliminate it at all. Some examples include degassing, filtration, improved handling
of the molten alloy, even using low pressures, and solidification simulation, for
predicting the potential effects of technological changes. Thanks to these contribu-
tions, aluminum is now regarded as one of the more versatile among the foundry
metals, even considering the continuous progress in alloy design (ASM International
1991; Lumely 2011; Polmear 2006). Since casting techniques allow for the
52 3 Alloys for Aircraft Structures

Fig. 3.8 Aluminum cast alloys nomenclature. Designation on the left is valid for 1XX.X alloys.
Designation on the right holds for all the other cast aluminum alloys (ASM International 1991;
Kaufman 2000). See also specifications in Table 3.2

Table 3.2 Aluminum cast alloy classification (ASM International 1991)

Series Composition and brief description
1XX.X Commercially pure aluminum with different grades
2XX.X Copper is the main alloying element. Eventually, other alloying elements may be
specified
3XX.X Silicon is the main alloying element. Other alloying elements are present as
specified, e.g., Cu and Mg
4XX.X Silicon is the main alloying element
5XX.X Magnesium is the main alloying element
6XX.X The name of this series is not in use
7XX.X Zinc is the main alloying element. Other elements, such as Mg, may be specified
8XX.X Tin is the main alloying element
9XX.X The name of this series is not in use

production of complex shapes and geometries, the choice of these materials is quite
often driven by net shape considerations. Concerning the coding system for alumi-
num cast alloys, nine main groups do exist, although not all of them have been used
so far: nXX.X, with n ranging from 1 to 9. The 1XX.X series is specific for the
commercially pure (CP) aluminum alloys, whereas different codes have been devel-
oped for the other groups of alloys. The rationale of this coding system is provided in
Fig. 3.8, together with the specification for the other alloy groups, listed in Table 3.2.
The selection of the best casting technique depends on the shape, the size, and the
number of parts that have to be produced. High-pressure die-casting is used in more
than 50% of the applications, whereas sand casting and permanent mold casting
techniques are more often employed for the production of thick wall parts or for
those characterized by internal hollow cavities for which high-pressure die-casting is
3.2 Aluminum Alloys 53

not a suitable choice. High-pressure die-casting implies that the molten aluminum
alloy is forced into a steel die at high speeds, ranging from 20 to 100 m/s. This value
can be achieved thanks to a piston-cylinder system, based on a hydraulic ram that is
able to reach pressures up to 100 MPa on the metal (Polmear 2006). Aluminum
castings are widely spread in the automotive industry, like for the fabrication of
wheels, engine blocks, and pistons. About 2% of the products obtained via high-
pressure die-casting is used in aerospace applications. The percentage raises up to
10% of the total when considering aerospace applications of castings obtained via
permanent mold casting and sand casting techniques. Permanent mold casting is the
preferred process for the production of wheels and suspension components (Lumely
2011). The chemical composition of some cast alloys is given in Table 3.3.
Usually, castings have mechanical properties that are lower if compared with the
wrought counterparts, except for the creep resistance. Furthermore, properties tend
to have a certain variability throughout the component (Polmear 2006). However,
the ratio of cast to wrought aluminum alloys is increasing thanks to the push of the
automotive industry (ASM International 1991).

3.2.3 Wrought Alloys

Wrought products include extrusions, foils, forgings, plates, sheets, stampings, and
wires. The most common form is that of the non-heat treatable sheet, largely used in
the construction, packaging, and transportation industry (Lumely 2011). The feed-
stock for metal working is typically an alloy ingot produced with a direct chill
casting. Wrought aluminum sheet products have usually a thickness in between 0.15
and 6.3 mm. Plates feature a thickness above 6.3 mm. Concerning specific aerospace
applications of the wrought alloys, the fuselage skin can be made of sheets and plates
with thickness ranging from 1 up to 10 mm. Otherwise, wing covers, that are load-
bearing structures, are generally 25–50 mm thick. The largest thicknesses can be
found in bulkheads, wing spars, and other structural parts. The classification of
extruded products is still made referring to the part thickness or to the diameter of the
smallest circle in which the cross section of the piece is contained. Open- or close-die
forgings are the two main options for these products. Open-die forged parts are
obtained by repeatedly deforming the alloy using flat dies or with very simple
geometries. The alternative approach (close-die) is used for parts with more complex
geometries, for which impression dies are needed. Wrought alloys can be classified
in nine different families according to their composition and in accordance with the
International Alloy Designation System (IADS), introduced in 1970 and now shared
worldwide. The classification of these Al-alloys is presented in Table 3.4.
Apart from the 1XXX family, the international coding is essentially referred to
the main alloying elements (see Fig. 3.9), according to the specifications in
Table 3.4. The second figure indicates the degree of development of a specific
alloy and, indirectly, the “age” of the alloy. For instance, the 2124 alloy has been
Table 3.3 Cast alloys chemical composition. Compositions are intended as maximum weight percentage unless shown as a range of values. Except for 150.1
54

alloy, Al content is balance (Polmear 2006)

International designation Si Fe Cu Mn Mg Cr Zn Ti Others
a a
150.1 0.10 – – – 0.05 – Al ¼ 99.50 min
201.0 0.10 0.15 4.0–5.2 0.20–0.50 0.15–0.55 – – 0.15–0.35 Ag ¼ 0.40–1.0
208.0 2.5–3.5 1.2 3.5–0.45 0.50 0.10 – 1.0 0.25 Ni ¼ 0.35
213.0 1.0–3.0 1.2 6.0–8.0 0.6 0.10 – 2.5 0.25 Ni ¼ 0.35
238.0 3.5–4.5 1.5 9.0–11.0 0.6 0.15–3.5 – 1.5 0.25 Ni ¼ 1.0
242.0 0.7 1.0 3.5–4.5 0.35 0.15–0.35 0.25 0.35 0.35 Ni ¼ 1.7–2.3
295.0 0.7–1.5 1.0 4.0–5.0 0.35 0.03 – 0.35 0.25 –
308.0 5.0–6.0 1.0 4.0–5.0 0.50 0.10 – 1.0 0.25 –
319.0 5.5–6.5 1.0 3.0–4.0 0.50 0.10 – 1.0 0.25 Ni ¼ 0.35
328.0 7.5–8.5 1.0 1.0–2.0 0.20–0.60 0.20–0.60 0.35 1.5 0.25 Ni ¼ 0.25
355.0 4.5–5.5 0.60b 1.0-1.5 0.50b 0.40-0.60 0.25 0.35 0.25 –
356.0 6.5–7.5 0.6 0.25 0.35 0.20–0.40 – 0.35 0.25 –
357.0 6.5–7.5 0.15 0.05 0.03 0.45–0.60 – 0.05 0.20 –
360.0 9.0–10.0 2.0 0.60 0.35 0.40–0.60 – 0.50 – Ni ¼ 0.50
380.0 7.5–9.5 2.0 3.0–4.0 0.50 0.10 3.0 – Ni ¼ 0.50
390.0 16.0–18.0 1.3 4.0–5.0 0.10 0.45–0.65 – 0.10 – –
413.0 11.0–13.0 2.0 1.0 0.35 0.10 – 0.50 – Ni ¼ 0.50
443.0 4.5–6.5 0.8 0.6 0.50 0.05 0.25 0.50 0.25 –
514.0 0.35 0.50 0.15 0.35 3.5–4.5 – 0.15 0.25 –
518.0 0.35 1.8 0.25 0.35 7.5–8.5 – 0.15 – Ni ¼ 0.15
520.0 0.25 0.30 0.25 0.15 9.5–10.6 – 0.15 0.25 –
535.0 0.15 0.15 0.05 0.10–0.25 6.2–7.5 – – 0.10–0.35 –
3 Alloys for Aircraft Structures
 705.0 0.20 0.80 0.20 0.40–0.60 1.4–1.8 0.20–0.40 2.7–3.3 0.25 –
707.0 0.20 0.60 0.20 0.40–0.60 1.4–1.8 0.20–0.40 4.0–4.5 0.25 –
712.0 0.15 0.50 0.25 0.10 0.50–0.65 0.40–0.60 5.0–6.5 0.10–0.25 –
713.0 0.25 1.18 0.40–1.0 0.60 0.20–0.50 0.35 7.0–8.0 0.25 Ni ¼ 0.15
850.0 0.70 0.70 0.70–1.3 0.10 0.10 – – 0.20 Ni ¼ 070–1.3 Sn ¼ 5.5–7.0
a
Ratio Fe/Si minimum 2:1
b
3.2 Aluminum Alloys

In case Fe content exceeds 0.45%, Mn content must be less than 0.5 times the Fe content
55
56 3 Alloys for Aircraft Structures

Table 3.4 Wrought alloys classification (ASM International 1991; Lumely 2011)
Series Information about the composition Heat treatment
1XXX Aluminum. CP – commercially pure with a purity Non-treatable
degree higher than 99%
2XXX Copper is the main alloying element. However, Treatable
other elements may be present (Mg in particular)
3XXX Manganese is the main alloying element Non-treatable
4XXX Silicon is the main alloying element Generally non-treatable. Some
exceptions do exist (e.g., 4032)
5XXX Magnesium is the main alloying element Non-treatable
6XXX Magnesium and silicon are the principal alloying Treatable
elements
7XXX Zinc is the main alloying element. However, other Treatable
spices can be present. It is the case of Cr, Cu, Mg,
and Zr
8XXX Alloys in which elements such as lithium and tin are Both treatable (e.g., 8090) and
present non-treatable
9XXX The name of this series is reserved for future use –

Fig. 3.9 Wrought alloy codename. Designation on the left is only valid for 1XXX alloys.
Designation on the right holds for all the other wrought alloys (ASM International 1991; Kaufman
2000)

developed later, i.e., it is more recent, than 2024 alloy. However, both alloys have
similar composition.
The latest two figures on the right indicate a serial number providing information
on the specific set of alloying elements and relevant concentrations. For instance,
alloy 7049 and 7050 are slightly different in composition. In case of 1XXX alloys,
the last two figures of the code refer to the level of impurities of the alloy: e.g., 1155
alloy has a purity of at least 99.55% (Fig. 3.9). As seen in Table 3.4, concerning their
composition, except for 1XXX series, wrought aluminum alloys can contain a wide
range of alloying elements. The composition of selected wrought alloys of 1XXX
series is given in Table 3.5.
A wider set of alloy compositions is given in Appendix 1; the indication of the
main effect of any element on the alloy properties is given in Appendix 2. Aluminum
wrought alloys can be split into two main subgroups: heat treatable and non-treatable
alloys.
Heat treatments of aluminum alloys are used to improve two main properties:
hardness and mechanical strength. This is attained through precipitation hardening,
although other effects of thermomechanical treatments, e.g., strain hardening,
 3.2 Aluminum Alloys

Table 3.5 1XXX alloys chemical composition. Compositions are intended as maximum weight percentage unless shown as a range of values
International designation Si Fe Cu Mn Mg Cr Zn Ti Others
1050 0.25 0.40 0.05 0.05 0.05 – 0.05 0.03 Al ¼ 99.50 V ¼ 0.05
1070 0.20 0.25 0.04 0.03 0.03 – 0.04 0.03 Al ¼ 99.70 V ¼ 0.05
1098 0.01 0.006 0.003 – – – 0.015 0.003 Al ¼ 99.98
1100 – – 0.05–0.20 0.05 – – 0.10 – Al ¼ 99.00
1275 0.08 0.12 0.05–0.10 0.02 0.02 – 0.03 0.02 Al ¼ 99.75 Ga ¼ 0.03 V ¼ 0.03
1420 – – – – 5.2 – – – Li ¼ 2.1 Zr ¼ 0.11
1421 – – – – 5.2 – – – Li ¼ 2.1 Sc ¼ 0.17 Zr ¼ 0.11
1440 – – 1.5 – 0.80 – – – Li ¼ 2.4 Zr ¼ 0.11
1441 – – 1.65 – 0.90 – – – Li ¼ 2.4 Zr ¼ 0.11
1450 0.25 0.40 0.05 0.05 0.05 – 0.07 0.10–0.20 –
1460 – – 2.9 – – – – – Li ¼ 2.25 Sc ¼ 0.09 Zr ¼ 0.11
If not specified, Al content is balance (Polmear et al. 2017; Prasad et al. 2014; The International Aluminum Association 2015)
57
58 3 Alloys for Aircraft Structures

Table 3.6 Main families of wrought alloys and relevant strengthening mechanisms (ASM Inter-
national 2015)
International designation Alloys system Strengthening mechanism
1XXX Pure Al Work-hardenable alloys
3XXX Al-Mn
4XXX Al-Si
5XXX Al-Mg
8XXX Al-Fe
8XXX Al-Fe-Ni
2XXX Al-Cu Precipitation-hardenable alloys
2XXX Al-Cu-Mg
2XXX Al-Cu-Li
6XXX Al-Mg-Si
7XXX Al-Zn
7XXX Al-Zn-Mg
7XXX Al-Zn-Mg-Cu
8XXX Al-Li-Cu-Mg

contribute to different extent to the overall characteristics of the alloys. Work-

hardening is the main strengthening approach for non-heat treatable alloys. In
Table 3.6, a further classification is proposed for the wrought alloys according to
the main strengthening mechanisms and with the indication of the subgroups of the
alloys depending on the reference compositional system.

3.2.3.1 Wrought Non-heat Treatable Alloys

Alloys of the series 1XXX, 3XXX, 4XXX, and 5XXX are non-treatable. This means
that no alloy structural changes can be induced by thermal treatments so that the
relevant properties are substantially insensitive to the thermal history of the alloy. In
some respect, this is a positive feature, for instance, in applications involving
uncontrolled thermal conditions, possibly inducing fast melting and solidification,
like welding. On the other hand, cold working processes, even conducted on the
component surface, like shot pinning and similar procedures, can drastically change
the mechanical properties. Although not particularly used in aircraft structures, it is
worth remembering some of their main application fields that are quite numerous
and diverse (Davis 2001; Polmear 2006):
• 1XXX: this family comprises un-alloyed aluminum alloys, with different grades
of purity: commercially pure and super-pure aluminum (Belov et al. 2002). This
series is used for the fabrication of several components, some interesting exam-
ples being architectural and decorations for structural parts in buildings, compo-
nents of chemical plants, and electric wires and cables, considering the good
electrical conductivity of pure aluminum.
3.2 Aluminum Alloys 59

Fig. 3.10 Al-Si phase diagram. (Redrawn from ASM International 1992)

• 3XXX: interesting characteristics of this family of alloys, having manganese as

main addition, are a low mechanical resistance and a high ductility, in association
with a good corrosion resistance. The most popular product is the 3003 alloy,
whose strengthening mechanism relies on the formation of intermetallic com-
pounds, e.g., Al6Mn, dispersed within the alloy matrix. This intermetallic phase
forms directly from the melt, and the structure of the phase diagram does not
allow solutioning and thermal treatments inducing solid-state precipitation, nec-
essary to refine the alloy microstructure and optimize its mechanical properties.
Some common applications of the 3XXX alloys are production of multipurpose
sheets, including building structures; furniture; cooking utensils, in particular
pottery; cans for food and drinks; and thin foils.
• 4XXX: the main alloying element is silicon, and these are two-phase alloys used
for welding processes thanks to the presence of a low-temperature eutectic
transformation (see Fig. 3.10). Silicon is not soluble in aluminum in the solid
state, and this leads to the solidification of a two-phase mixture, consisting of
mostly pure silicon grains dispersed into the aluminum matrix. The grain size and
morphologies will depend upon the kinetic aspects of the transformation,
although eutectic microstructures are very likely.
• 5XXX: this series of alloys is characterized by the presence of magnesium as
main alloying element. A particular feature of these alloys, interesting for some
applications, is the fact that the fcc austenitic phase exhibits a very low ductile-to-
brittle transition temperature. Therefore, these alloys are very much suitable for
cryogenic use, like inner pressurized vessels for liquid nitrogen dewar containers,
cryogenic sample holders, etc. They show a good weldability, so that complex
60 3 Alloys for Aircraft Structures

Table 3.7 Designation of the main temper condition used for all Al-alloys
Name Main features
F As fabricated. Corresponds to the strain-hardened condition due the production process.
The letter F stands alone
O Annealed condition to obtain the lowest strength temper, aiming at increasing the alloy
workability. O can be applied to cast products that are annealed to improve dimensional
stability and ductility. The letter O may be followed by a digit to indicate an annealed
condition with special characteristics
H Strain-hardened wrought products. Additional thermal treatment can be employed if a
reduction of the alloy strength is needed. H is followed by two or more digits (see also
Table 3.8)
W Solution heat-treated alloys. This temper is applied only to alloys that undergo sponta-
neous aging after solution heat treatment. It induces a sort of metastable condition,
followed by natural aging at room temperature. Digits to indicate the natural aging
period, e.g., W1/2 h, are commonly used
T Thermally heat-treated to produce stable tempers other than F, H, O. Supplementary
strain hardening may be employed as well. The letter T is always followed by one or
more digits (see Tables 3.9 and 3.10)
This scheme is adopted by the main international standards referring to these alloys (Benedyk 2010;
Kaufman 2000)

structures and/or comparatively large pieces can be assembled: large tanks for
milk but also petrol, oil, and other organic liquids. Small boat hulls have been also
made using Al-Mg alloy sheets.
More information on wrought non-heat treatable alloys can be found in the
references quoted at the end of the chapter.

3.2.3.2 Heat-Treatable Alloys and Designation of Tempers

Heat-treatable aluminum alloys are particularly interesting in aerospace applications.

Different combinations of mechanical and corrosion resistance properties can be
imparted to the same alloy thanks to suitable heat treatments. An international
designation code, registered by the Aluminum Association, does exist for these
tempers, and it is usually added to the alloy designation number. All changes
induced into heat-treatable alloys are based on the solid-state formation of stable
or metastable precipitates, as described in Sect. 3.2.3.3. The heat treatments are
typically a combination of a solution stage and quenching, followed by natural and
artificial aging. Considering the role that relevant defects may have in solid-state
precipitation, cold-working is another aspect involved in the temper design. As
shown by Table 3.7, the T temper is only one of the possible tempers used for
Al-alloys. The designation, used since 1948 and updated over the years, is meant to
assign to any heat-treatable alloy product the information on the thermal treatment
used to set its properties.
Table 3.8 gives information about the codes used for strain-hardened, temper H,
alloys.
3.2 Aluminum Alloys 61

Table 3.8 Description of the different “H” tempers. X indicates the residual strain hardening
percentage, remaining in the alloy after each treatment, except that for H1X, where X is the
percentage of imparted strain hardening
Type Main features
H1X Strain hardened only
H2X Strain hardened and partially annealed. The strength acquired during strain hardening is
reduced at the desired level by means of a partial annealing process. In this case X
corresponds to the percentage of residual strain hardening after the partial annealing
H3X Strain hardened and stabilized. Stabilization is used for improving ductility and can be
carried out either thanks to heat introduced during the fabrication or thanks to a low
temperature thermal treatment. Again, X corresponds to the percentage of residual strain
hardening after the stabilization treatment
H4X Strain hardened and lacquered or painted. It indicates the products that are strain hardened
and subsequently subjected to heat during painting or lacquering. X corresponds to the
percentage of residual strain hardening after lacquering or painting
The maximum value for X ¼ 8 (Kaufman 2000)

As concerns the T tempers, a brief description is given in Table 3.9, with full
details in Benedyk (2010). All T-treatments start with alloy solutioning. The subse-
quent quenching aims at freezing down to room temperature a super-saturated solid
solution. The two main temper parameters are solution temperature and cooling rate.
Precipitation occurs next, induced by either artificial or natural aging. Natural aging
occurs spontaneously at room temperature, an aspect to be considered not only
during the intermediate steps of the heat treatment but also as concerns the micro-
structural evolution occurring under operating conditions. Natural aging determines
the useful lifetime of the alloy, corresponding to the time required to reach the
overaging conditions, when material properties are no longer suitable for the relevant
application.
The T tempers described in Table 3.9 can be rendered more complex by intro-
ducing further steps, mainly designed for relieving the stress accumulated into the
alloy piece. Therefore, other digits can be added to the temper identification code, as
specified in Table 3.10. In particular, stress relief by stretching is codenamed T51, by
compressing T52, a combination of stretching and compressing T54. Stress relief
can be performed either after solution heat treatment or after quenching, following a
shaping process conducted at elevated temperatures. More digits can be used to
provide additional information on other relevant parameters of the temper, like
percentage of nominal permanent set, level of stress relieved, etc. (Table 3.10).
The selection of the right temper conditions is often the result of a compromise of
diverging requirements. For example, tempers interesting for aircraft alloys of the
7XXX family are the T6 and T7. As shown by Fig. 3.11, T6 would induce excellent
mechanical strength but a correspondingly lower corrosion resistance of the alloy.
Intermediate conditions can be attained using suitable variants of the T7 temper.
62 3 Alloys for Aircraft Structures

Table 3.9 Description of the T temper treatments (Kaufman 2000)

Type Main features
T1 Cooling from elevated temperature shaping process and natural aging. Natural aging leads
to a substantially stable condition. Applications: products that do not undergo cold
working after cooling from an elevated temperature shaping process
T2 Cooling from elevated temperature shaping process, cold working, and final natural aging
process. Natural aging leads to a substantially stable condition. Applications: products
that undergo cold working after cooling from an elevated temperature shaping process
T3 Solution treatment, cold working, and final natural aging process. Natural aging leads to a
substantially stable condition. Applications: products that undergo cold working to
improve the strength after being solution treated
T4 Solution treatment, natural aging. Natural aging leads to a substantially stable condition.
Applications: products that do not undergo cold working after solution treatment
T5 Cooling from elevated temperature shaping process and artificial aging. Applications:
products that do not undergo cold working after being cooled from elevated temperature
shaping process
T6 Solution treatment and artificial aging. Applications: products that do not undergo cold
working after solution treatment
T7 Solution treatment and overaging/stabilization. Applications: cast products that undergo
artificial aging after solution treatment in order to give them stability in strength and
dimension, wrought products that undergo artificial aging after solution treatment in order
to give them an increase in strength beyond the maximum value achievable to control
some specific characteristics or properties
T8 Solution treatment, cold working, and artificial aging. Applications: products that undergo
cold working for increasing their strength
T9 Solution treatment, artificial aging, and cold working. Applications: products that undergo
cold working operations for increasing their strength
T10 Cooling from elevated temperature shaping process, cold working, and artificial aging.
Applications: products that undergo cold working for increasing their strength
For all T tempers, the solution heat treatment involves heating the alloy up to the solution
temperature and holding at that temperature for a long enough time, so to obtain a complete
solutioning of the alloying elements. The subsequent cooling (quenching) should be sufficiently
fast to retain a supersaturated solid solution down to the quenching temperature, in order to have
precipitation to take place during the subsequent artificial or natural aging

3.2.3.3 Precipitation Hardening

Solid-state precipitation is the phase transformation on which the hardening of

heat-treatable Al-alloys is based. The same transformation and resulting strength-
ening mechanisms hold also for other classes of alloys that will be considered next:
magnesium and titanium alloys (Sect. 3.3 and Chap. 4), PH stainless steels (Sect.
5.4.3.3), and iron- and nickel-based superalloys (Sects. 6.3 and 6.4). The fundamen-
tal aspects of these treatments, which will be illustrated herewith with reference to an
aluminum alloy, are meant to provide indications, from the physical metallurgy point
of view, for the heat treatments of the other alloy systems too, with obvious
important differences as concerns both the process parameters and the precipitates
that actually form in each specific alloy system.
3.2 Aluminum Alloys 63

Table 3.10 Additional digits for stress relief treatments in T-tempered aluminum alloy products
(Kaufman 2000)
Additional digits for T Permanent
tempers Wrought product type Stress relief process set
TX51a Plate Stretching 1.5–3.0%
TX51a Rolled or cold finished bars Stretching 1.0–3.0%
and rods
TX51a Die or ring forgings and rolled Stretching 1.0–5.0%
rings
TX510a Extruded bars, profiles, rods, Stretching 1.2–3.0%
and tubes
TX510a Drawn tubes Stretching 0.5–3.0%
TX511b Extruded bars, profiles, rods, Stretching 1.0–3.0%
and tubes
TX511b Drawn tubes Stretching 0.5–3.0%
TX52 All products Compressing 1.0–5.0%
TX54 Die forgings Restriking cold in –
finish die
a
No further straightening after stretching
b
Minor straightening after stretching. This may be required by tolerance limitations

Fig. 3.11 Dependence of mechanical strength and corrosion resistance of the T6 and different T7
tempers. These tempers are typically used for alloys of the 7XXX group, including those for aircraft
parts (redrawn from Benedyk 2010)

The main aspects of solid-state precipitation in hardenable aluminum alloys can

be discussed considering the binary Al-Cu system, being an important reference also
for alloys with more complex compositions. For technological purposes, the alloy
composition never goes above the solubility limit of copper in aluminum (see
Fig. 3.12); therefore the aluminum-rich corner is the interesting part of the phase
diagram, as shown by Fig. 3.13. Solutioning of copper into the aluminum crystalline
lattice results in a single-phase alloy, the α-phase in the phase diagram in Fig. 3.13.
Quenching freezes the supersaturated alloy down to room temperature, and the
subsequent aging, conducted either at room (natural aging) or higher (artificial
aging) temperatures, is designed to reach the target conditions, in agreement with
the requirements of the specific applications. The intermetallic compound θ-Al2Cu,
present in the Al-Cu equilibrium phase diagram, is not really interesting for the
64 3 Alloys for Aircraft Structures

Fig. 3.12 Al-Cu phase diagram. (Redrawn from ASM International 1992)

improvement of the mechanical properties of the alloy and it is actually a typical

phase observed in overaged conditions. Similar equilibrium phases, with different
compositions, will form in other heat-treatable aluminum alloys (see Table 3.11).
Several important metastable structures can be obtained indeed in real alloys, thanks
to the kinetic constrains introduced by the heat treatment parameters. These struc-
tures, not present in the equilibrium phase diagram, are very important as concerns
the alloy properties. In the Al-Cu system, these structures are θ0 , θ", and GPZ
(Guinier-Preston zone), described in the following.
The heat-treated condition of the alloy can be predicted from TTT curves
(Fig. 3.14), once the main parameters of the treatment, like quenching rate, aging
temperature, and time, have been selected. As illustrated in Fig. 3.14, there are two
possibilities for obtaining GPZ. The first consists in quenching the alloy directly to
the relevant GPZ field as shown by the cooling trajectory 1. A second approach is to
cool down the alloy at a sufficiently high rate not to cut the GPZ directly from the
solutioning temperature and then start an aging treatment, at constant temperature
until the transformation is fully accomplished, according to the trajectory 2 in
Fig. 3.14.
3.2 Aluminum Alloys 65

Fig. 3.13 Part of the Al-Cu phase diagram showing the stable θ-Al2Cu precipitate and the
metastable θ0 , θ", and GPZ. (Redrawn from Porter et al. 2009)

Table 3.11 Solid-state precipitation in some alloy systems (Porter et al. 2009)
Base metal Alloy Precipitation sequence
Al Al-Ag GPZ (spheres) ! γ 0 (plates) ! γ (Ag2Al)
Al-Cu GPZ (discs) ! θ" (discs) ! θ0 (plates) ! θ (Cu Al2)
Al-Cu-Mg GPZ (rods) ! S0 (laths) ! S (CuMgAl2, laths)
Al-Zn-Mg GPZ (spheres) ! η0 (plates) ! η (MhZn2, plates or rods)
Al-Mg-Si GPZ (rods) ! β0 (rods) ! β (Mg2Si, plates)
Cu Cu-Be GPZ (discs) ! γ 0 ! γ (CuBe)
Cu-Co GPZ (spheres) ! β (Co, plates)
Fe Fe-C ε carbide (discs) ! Fe3C (plates)
Fe-N α" (discs) ! Fe4N
Ni Ni-Cr-Ti-Al γ 0 (cubes or spheres)
66 3 Alloys for Aircraft Structures

Fig. 3.14 Al-Cu phase diagram showing the stable θ precipitate and the metastable θ0 , θ", and GPZ
and related TTT diagram with different cooling trajectories. See main text for details. (Redrawn and
adapted from Porter et al. 2009)

Similarly, the precipitation of other phases can be induced. Different precipitates

feature different crystallographic structures that induce different kinds of interface
with respect to the hosting aluminum matrix: coherent, semicoherent, or fully
incoherent (Porter et al. 2009 – see also Sect. 6.4.1.1).
Figure 3.15 summarizes the situation for the precipitates’ features in Al-Cu
alloys. An important parameter that characterizes the matrix-precipitate interface is
the lattice misfit δ (Geddes et al. 2010; Reed 2006):

aθ00  aα
δ¼ 2 ð3:2Þ
aθ00 þ aα

where
• δ is the lattice misfit.
• aα is the lattice parameter of the matrix.
• aθ" is the lattice parameter of the θ" precipitate.
The lattice misfit depends on the difference of the lattice parameter, or interatomic
distance, along specific crystallographic planes that are facing each other across the
precipitate-matrix interface (see also Sect. 6.4.1.1). In the present example (Eq. 3.2),
the lateral plane for the θ" and cube plane for the α Al-fcc structure are considered.
Other important features of the precipitates, like size, shape, composition, etc., are
dependent on the overall alloy composition and on the elastic anisotropy along
crystallographic directions. These intrinsic alloy parameters and the thermal treat-
ment conditions act as kinetic constraints, so that not only stable but also metastable
phases can form.
3.2 Aluminum Alloys 67

Fig. 3.15 α matrix, stable θ precipitate and metastable θ0 , θ00 structure and morphology. (Redrawn
from Porter et al. 2009)

In Table 3.11, the precipitation transformations for different alloy systems are
summarized. In case of heat-treatable aluminum alloys, the formation of metastable
structures is quite a general feature. This is the case of the β0 precipitates in the 6063
alloy (Al-Mg-Si group, Fig. 3.16). β0 (Mg2Si composition) is a coherent precipitate
with preferential orientation along the <100> direction of the aluminum matrix. The
precipitates are aligned along this particular crystallographic direction since it turns
out to be the more energetically favored, as compared to the others. In fact, along this
direction there is the lowest atomic linear density, the interatomic distances are
larger, and so the elastic constant is lower. The shape of the precipitate is sensitive
to the elastic anisotropy as well. In fact, the interfaces with the aluminum matrix
having a lower elastic strain energy term will be wider than those with a higher
elastic energy term. The hardening induced by the alloy heat treatments will depend
on the resulting microstructure that influences the way in which precipitates and
dislocations interact.
In agreement with the TTT curves (Fig. 3.14), at an aging temperature of 190  C,
the GPZs cannot form, since the relevant aging trajectory is not cutting the stability
domain of these structures, meaning that the aging temperature is larger than the
GPZ solvus. Direct precipitation of θ00 can take place instead, and these precipitates
for longer aging times may transform into θ0 precipitates. The peak hardness value
scales up with the concentration of copper and, thereby, with the fraction of pre-
cipitates that may form. The hardening curves corresponding to a lower aging
temperature (Fig. 3.17, T ¼ 130  C) may reach higher hardness peaks, for each
one of the investigated alloys. This can be ascribed to the formation of finely
68 3 Alloys for Aircraft Structures

Fig. 3.16 Transmission electron microscopy micrograph of β0 precipitate in 6063 alloy. The image
contrast is due to the strain ε at the precipitate-alloy matrix coherent interface, as represented by the
bottom scheme. (Photo property of the Authors, bottom scheme redrawn from Porter et al. 2009)

dispersed GPZs, not forming at the higher aging temperature. The evolution of
hardness on aging is determined by the change of the dominant interaction mecha-
nisms between dislocations and precipitates. In the early stages of the lower tem-
perature aging, GPZs are the main strengthening agents. Dislocations are hindered in
their movement by the elastically strained aluminum matrix. The same mechanism is
active when the θ" coherent precipitates start to form. Moreover, the coherency of
the θ" particles and their larger size, as compared to GPZs, allows also for the
dislocation cutting mechanism to happen (Fig. 3.18). The resulting strengthening is
due to the drag that the dislocations are subjected to when moving through the
precipitates and to the resistance determined by the strain field at the precipitate-
matrix interface.
The mechanisms described so far are responsible for the rising branch of the
hardening curves in Fig. 3.17. Each one of these curves reaches a maximum and then
starts to decrease. This is due to a change in the dislocation-precipitate interaction
mechanism. The ripening of the precipitates with aging time results in a progressive
alloying elements depletion from aluminum alloy matrix.
3.2 Aluminum Alloys 69

Fig. 3.17 Effect of the aging temperature on the hardness of binary Al-Cu alloys. (a) 130  C aging;
(b) 190  C aging. Note that for the higher temperature, the precipitation of GPZ is nearly fully
suppressed. (Redrawn from Porter et al. 2009)

This reduces the solid solution strengthening, and dislocations can now move
more easily. Actually, they get pinned by the precipitates but can still continue to
move if the applied stress is sufficient to increase their length. This is the strength-
ening mechanism active in the so-called dislocation bowing regime (also known as
Orowan mechanism, Fig. 3.19; see Sect. 6.4.3 for further details).
Actually, additional strengthening may also come from the piling up of disloca-
tion loops around each pinning precipitate. The Orowan mechanism is active in the
Al-Cu alloys when incoherent θ0 and equilibrium θ precipitates form. The occur-
rence of equilibrium precipitates in all hardenable aluminum alloys, i.e., those
belonging to the 2XXX, 6XXX, and 7XXX groups (see Table 3.4), is often
70 3 Alloys for Aircraft Structures

Fig. 3.18 Dislocation cutting through a spherical precipitate

Fig. 3.19 Dislocation bowing. (Redrawn from Hull and Bacon 2011)

associated with the condition of overaging (Fig. 3.20 referring to a 6063 alloy with
equilibrium Mg2Si precipitates).
At this stage, which can be reached either for an incorrect heat treatment or for the
prolonged exposure of the alloy to demanding servicing conditions, an important
decay of the alloy hardness is observed. The transition from cutting to bowing as
concerns the dominant dislocation-precipitate interaction mechanism is observed in
other precipitation-hardened alloys, like Mg-alloys (see Sect. 3.3) and Ni-based
3.2 Aluminum Alloys 71

Fig. 3.20 Mg2Si precipitate

in an overaged 6063 alloy.
(Photo property of the
Authors)

superalloys (Chap. 6). For these latter, an important role as concerns strengthening is
played by the atomic crystallographic order of the γ 0 precipitates (see Chap. 6 for
further details). The heat treatments of aluminum alloys are important for the effects
they have on the structure, composition, spatial distribution, and size of the strength-
ening precipitates. Also the microstructure of the alloy in the grain boundary regions
is influenced by the heat treatment parameters and relevant precipitation kinetics.
Figure 3.21 shows some examples of the so-called precipitation-free zone (PFZ) and
relevant scheme.
Heterogeneous nucleation, driven by the grain boundary surface energy term,
determines an anticipated precipitation along grain boundaries, which depletes the
surrounding regions of alloying elements. When nucleation, mostly homogeneous,
and subsequent growth of precipitates inside the grains start, they can no longer
occur in the alloy-depleted regions that give place to the mentioned PFZs. As shown
by the scheme in Fig. 3.21, PFZs become a preferential path for the movement of
dislocations, resulting in a general weakening of the alloy. To get rid or at least to
reduce the incidence of the PFZs, several approaches are possible. A higher
undercooling, with respect to the solution temperature, enhances the homogeneous
nucleation rate inside the grains. Alloying with insoluble atoms, like silicon, may
introduce preferential nucleation sites in the alloy, which again favor a uniformly
distributed precipitation within the alloy grains, by reducing the incidence of the
preferential precipitation along grain boundaries. Another defect structure that can
72 3 Alloys for Aircraft Structures

Fig. 3.21 Examples of precipitate-free zones. (Left picture redrawn from Polmear 2006; images on
the right are photo property of the Authors)

Fig. 3.22 (a) Transmission electron micrograph showing hydrogen inclusions development at the
grain boundary of a thin foil of an artificially aged Al-Zn-Mg alloy; (b) blisters on the surface of an
Al-alloy component after solution treatment in a humid atmosphere (Polmear 2006, reproduced
with permission of the Publisher)

be observed along grain boundaries and subsurface regions is hydrogen micro-

inclusions (Fig. 3.22). This phenomenon, called blistering for the particular mor-
phology that induces on the alloy surface, occurs during water quenching as a
consequence of the hydrolysis of H2O catalyzed by the oxidation of aluminum,
according to the reaction: 2Al + 3H2O ! Al2O3 + 3H2.
The grade of the alloy, depending on the concentration of impurity elements, is
very important because it determines the formation of intermetallic particles within
the alloy itself (see Fig. 3.23). The size and intrinsic brittleness of these phases
render their presence potentially detrimental for the properties of the alloy, in which
they act as stress concentrators.
3.2 Aluminum Alloys 73

Fig. 3.23 Fe-rich inclusion

in a 6063 alloy,
isomorphous to the Al3Fe
intermetallic. (Photo
property of the Authors)

3.2.4 Aluminum-Lithium Alloys

Lithium is present in several Al-alloys, but since it is not necessarily the majority
alloying element, wrought Li-containing Al-alloys can be found in different alloy
groups: 1XXX, 2XXX, and, finally, 8XXX, being the alloys of this latter group,
those featuring larger Li concentrations (Table 3.12).
Although regarded as the most recently developed of the Al-alloys, actually,
Al-Li alloys started their metallurgical history back in the 1920s, and lithium
additions were used to enhance the strengthening of age-hardenable alloys (Starke
et al. 1981). The very first Li-containing Al-alloy was developed in Germany in 1924
with the commercial name Scleron and nominal composition Al-12Zn-3Cu-0.6Mn-
0.1Li (Prasad et al. 2014). The first results were very encouraging indeed, since the
expected improvement in strength was accompanied also by a higher stiffness and,
obviously, lower density (see Fig. 3.24). The production of Al-Li alloys requires a
careful working condition control, in order to avoid the losses of lithium during
melting, for which vacuum induction melting is regarded as a safe route (for
additional information see Sect. 5.3.2) (ASM International 2008). Lithium content
in the main Al-Li alloys is given in Table 3.12, while the complete chemical
composition of the main alloys of the family can be found in Table 3.5 and Appendix
1. Three main different development stages can be identified. The first-generation
74 3 Alloys for Aircraft Structures

Table 3.12 Designation, Li content, density, and introduction of some relevant Al-Li alloys
wt% Density
Alloy Generation Li [g/cm3] Producer (introduction year)
2020 1 1.00 2.71 Alcoa (1958)
1420 2.10 2.47 Soviet (1965)
1421 2.10 2.47 Soviet (1965)
2090 2 2.10 2.59 Alcoa (1984)
2091 (Li  2%) 2.00 2.58 Pechiney (1985)
8090 2.40 2.54 EAA (1984)
1440 2.40 2.55 Soviet (1980s)
1441 1.95 2.59 Soviet (1980s)
1450 2.10 2.60 Soviet (1980s)
1460 2.25 2.60 Soviet (1980s)
2195 3 1.00 2.71 Lockheed Martin Corporation/Reynolds (1992)
2196 (Li  2%) 1.75 2.63 Lockheed Martin Corporation/Reynolds/McCook
Metals (2000)
2297 1.40 2.65 Lockheed Martin Corporation/Reynolds (1997)
2397 1.40 2.65 Alcoa (1993)
2098 1.05 2.70 McCook Metals (2000)
2198 1.00 2.69 Reynolds/McCook Metals/Alcan (2005)
2099 1.80 2.63 Alcoa (2003)
2199 1.60 2.64 Alcoa (2005)
2050 1.00 2.70 Pechiney/Alcan (2004)
2296 1.60 2.63 Constellium Alcan (2010)
2060 0.75 2.72 Alcoa (2011)
2055 1.15 2.70 Alcoa (2011)
2065 1.20 2.70 Constellium (2012)
2079 1.50 2.64 Constellium (2012)
The complete chemical composition is given in Table 3.5 and Appendix 1 (Prasad and Wanhill
2017)

Fig. 3.24 Dependence of the Young’s modulus and density of aluminum as a function of the
concentration of different alloying elements. (Redrawn from Polmear et al. 2017)
3.2 Aluminum Alloys 75

Table 3.13 Main precipitate phases encountered in commercial Al-Li alloys (Prasad et al. 2014)
Lattice
Crystal parameters
Phase Chemistry structure [nm] Main features
δ AlLi Cubic (NaCl) a ¼ 0.638 Equilibrium phase with a plate morphology
δ0 Al3Li Cubic (L12) a ¼ 0.401 Non-equilibrium ordered phase, usually
with spherical shape and coherent interface
with the aluminum matrix
T1 Al2CuLi Hexagonal a ¼ 0.4965 Equilibrium phase. Partially coherent
c ¼ 0.9345 interfaces
T2 Al6CuLi3 Cubic a ¼ 1.3914 Please note: the cubic structure displays
icosahedral symmetry
S0 Al2CuMg Orthorhombic a ¼ 0.401 Semicoherent phase
b ¼ 0.925
c ¼ 0.715
θ0 Al2Cu Tetragonal a ¼ 0.404 –
c ¼ 0.580
– Al3Zr Cubic (L12) a ¼ 0.405 Coherent, ordered, spherical dispersoids

alloys were developed over a decade across 1960. The potential of these mate-
rials was demonstrated and the critical aspects of their processing emerged. The
second-generation Al-Li alloys featured a Li content in excess, or equal to, 2 wt%
and were produced either via ingot metallurgy, mainly, or powder metallurgy. The
development of the third-generation was initiated in the late 1980s, when the Martin
Marietta Corporation was engaged in the design of a weldable, low-density alumi-
num alloy for cryogenic tanks and other components of aerospace vehicles (Prasad
et al. 2014).
Lithium alloying reduces aluminum density and increases its stiffness (Fig. 3.24).
The higher stiffness and, most importantly, the enhancement of the alloy strength are
due to the formation of the ordered δ0 -Al3Li precipitates. These precipitates have the
same L12 structure as the ordered γ 0 precipitates in Ni-based superalloys (see
Chap. 6). The partial covalent character of the chemical bonds in the intermetallic
phase enhances the Young’s modulus of the alloy and, as counterpart, reduces the
alloy ductility and its fracture toughness. Moreover, fracture toughness of Al-Li
alloys is extremely sensitive to the content of alkali metal impurities. A few tens of
ppm of Na and K, introduced during the ingot metallurgy casting for the production
of the primary alloy, are sufficient to reduce the alloy fracture toughness below the
safety values compatible with structural applications (Polmear 2006).
The main strengthening mechanism active in the early compositions (first gener-
ation, Table 3.12) of the Al-Li alloys is based on the dislocations cutting through the
δ0 -Al3Li precipitates (Table 3.13). This intermetallic, not present in the relevant
equilibrium phase diagram (Fig. 3.25), can be stabilized though using suitable
thermal treatments.
The ordered structure of the δ0 -Al3Li precipitates introduces a further strengthen-
ing factor, associated with their interaction with dislocations and with the formation
76 3 Alloys for Aircraft Structures

Fig. 3.25 Al-Li phase diagram. (Redrawn from ASM International 1992)

of complex defect structures, involving antiphase boundaries and super-dislocations,

as in Ni-based superalloys (see Sect. 6.4.3). However, the localized shearing across
the ordered precipitates may produce excessive strain hardening concentrated close
to grain boundary, leading in the long run to diffuse cracking. This was one of the
main reasons for the withdrawal of the first-generation alloys, e.g., the A2020 (see
Table 3.12), used for the wing skin of the Northrop RA-5C Vigilante aircraft. The
alloy compositions and relevant annealing treatments were developed further,
aiming at the formation of other precipitates and dispersoids, capable to homogenize
the dislocation structures. The reinforcing phases that formed in the novel alloy
formulations were either finely distributed inside the alloy matrix, like the θ0 -Al2Cu,
introducing lattice strain as obstacle to dislocation movement, or were semicoherent
grains with a limited number of slip systems. This is the case of the T1-Al2CuLi and
S0 -Al2CuMg intermetallics, not easily cut by dislocations, that thereby remain stuck
at the precipitate-alloy interface. The formation of these phases is favored by the
presence of alloying elements, like silver and zirconium, introduced in several of the
third-generation Al-Li alloys. Another feature of these alloys is the optimum
concentration of lithium, typically ranging from 1.0 to 1.3 wt%, anyway lower
than 2.0%, or more, typical of the second-generation Al-Li alloys, for this reason
featuring extremely low-density values and regarded as a promising metallurgical
response to the early encouraging results obtained with the extra-light polymer
matrix structural composites. Notwithstanding the interest in these alloys, their
elevated brittleness and low toughness lead to a rapid abandon of these formulations
3.2 Aluminum Alloys 77

and stopped any further development. A definitely better outcome was faced by the
third-generation Al-Li alloys. Although relatively heavier, for the smaller concen-
trations of lithium, these alloys exhibited the right balance between workability and
in-service performances. The development of these alloys received an important
boost also from the results of the research conducted by metallurgists in the Soviet
Union in the 1970s. The most recent Al-Li products, featuring the constant presence
of copper, have been qualified for high strength and high damage-tolerant applica-
tions in aircraft structures, exhibiting even interesting corrosion resistance properties
(Lequeu et al. 2010). Altogether the Al-Li alloys represent an important step in the
development of light alloys. In fact, several Al-Li products have reached similar or
better properties than the 2XXX and 7XXX traditional alloys, of which they
represent excellent candidates for substitution (Table 3.14).

3.2.5 Aluminum Alloys in Aircraft Structures: Present State

and Perspectives

The capability to meet quite complex and diverse requirements is an important

aspect of the aluminum alloys and probably one of the main reasons for their success
in aircraft structures. The situation is well depicted by the scheme in Fig. 3.26,
summarizing the main properties for the principal parts and components of an
aircraft structure, for which aluminum alloy represents an excellent choice.
Possible candidates like magnesium alloys (see Sect. 3.3) and structural polymer
matrix composites that in the long run might take over the role of aluminum alloys
are emerging. The B787-Dreamliner has demonstrated the possibility of using
structural composites as reliable replacements for aluminum alloys (see Fig. 1.11
in Chap. 1). This result has been possible thanks to an integrated production
approach, involving materials development and the implementation of novel
manufacturing technologies. In some respect, this is what has also been made for
the Airbus A380, this time with a priority for structural aluminum alloys. In this
aircraft, an intelligent mix of improved new composition and workhorse alloys have
been largely employed, as indicated by the list of aluminum parts given in Table 3.15
(Lequeu et al. 2007).
The large dimensions of the A380 (Fig. 3.27) and the optimization of the
production costs, have required the development of new equipment for casting the
primary alloys, for machining, for hot-rolling, and for handling the super-size slabs
and plates (Lequeu et al. 2007). The use of several Al-Li alloys is coherent with the
Airbus strategy that over the years has progressively reduced the percentage of
conventional aluminum alloys used in their airplane structures and at the same
time has increased the fraction of the lighter Al-Li alloys.
In other commercial airliners, like the B777 and A460-600, high strength and
high damage-tolerant aluminum alloys belonging to the 2XXX and 7XXX groups,
as those listed in Table 3.16, have been selected for the airframes, using products
Table 3.14 Products, requirements, applications of selected Al-Li alloys, with reference also to the substitution of alloys from other groups (Prasad et al. 2014)
78

Generation Alloy Temper Product Alloys to substitute Requirements Applications

2 8090 T3 Sheet – – Floor installations, brackets, stiffeners,
frames, spars, stringers, longerons, ribs,
bulkheads
8090C T8 2014-T6 Medium strength Flying control systems, avionics bay struc-
tures, cabin roof frames
8090C T81 2024-T3 Damage Lower fuselage skin panels, flat roof panels,
2024-T42 tolerance flying control structures
8090 T852 Forgings 7010-T7451 Medium/high Cabin roof and side frames
strength
8090 T8511 Extrusions 7075 Medium strength Frames, brackets, stringers, bulkheads, door
rails
3 2098 T8511 Sheet 2024-T3 Damage toler- Fuselage/pressure cabin skin
2524-T3 ance, medium
2198 T8 2524-T351 strength
2199 T8E74
2060 T8E30
2199 T86 Plate 2024-T351 2324-T39 Damage Lower wing covers
2050 T84 tolerance
2060 T8E86 2624-T351
2624-T39
2098 T82P 2024-T62 Medium strength F-16 fuselage panels
2297 T87 2124-T851 F-16 fuselage bulkheads
2397 T87
2050 T84 7150-T7751 7055-T7751 7055-T7951 High strength Upper wing covers
2055 T8X 7255-T7951
2195 T82P
3 Alloys for Aircraft Structures
2050 T852 Forgings 7175-T7351 High strength Wing/fuselage attachments, windows frames,
7050-T7452 crown frames
2099 T81 Extrusions 2024-T3511 2026-T3511 2024-T4312 Damage Lower wing stringers, fuselage/pressure cabin
6110-T6511 tolerance stringers
2099 T83 7075-T73511 7075-T79511 7150- Medium/high Fuselage/pressure cabin stringers and frames,
2055 T8R83 T6511 7175-T79511 7055-T77511 strength upper wing stringers, Airbus A380 floor
7055-T79511 beams and seat rails
3.2 Aluminum Alloys

2065 T8511
79
80 3 Alloys for Aircraft Structures

Fig. 3.26 Materials property requirements for a commercial airliner. Material properties abbrevi-
ations: COR corrosion, CYS compressive yield strength, E modulus, FAT fatigue, FCG fatigue
crack growth, FT fracture toughness, SS shear strength, TS tensile strength. Material properties in
brackets ( ) are regarded as important, but not critical, design requirements. (Redrawn from Polmear
et al. 2006)

obtained by rolling, extrusion, or forging. In most cases, the starting batch is an ingot
semi-continually direct-chill cast.
The extreme reliability of the conventional Al-alloys and the knowledge that has
been accumulated over the years in more than one century represent also an
important benchmarking for the development of new wrought alloys, like the recent
formulations and tempers of Al-Li alloys discussed in Sect. 3.2.4 (Lequeu et al.
2010), particularly as concerns those mechanical properties important both for
processing and in-service performances (Table 3.17).
Some structural aircraft components made of aluminum alloys are produced also
by casting: sand casting, investment casting, and permanent mold casting. In this
way, single-part components have been obtained, thus reducing or eliminating at all
joining steps. The cast components are particularly well suited for replacing parts
with complex geometries, requiring time-consuming and expensive machining.
Another advantage of cast components is the elimination of all those defects that
are typically associated with joints, like structural flaws and corrosion-initiation
regions. However, since the resulting mechanical properties of cast products are
on average lower than those of the corresponding wrought counterparts (see Sect.
3.2.2), the so-called casting factor (CF) has been introduced. This is a sort of safety
factor, ranging from 1.0 to 2.0 used to scale down the reliable strength of the cast
components, any time the material has an assigned CF value in excess of 1.0. A
3.2 Aluminum Alloys 81

Table 3.15 Aluminum alloy products used for different parts and components of the A380
(Lequeu et al. 2007)
Alloy Temper Form A380 application
2024A T351 Plates Lower wing reinforcement
2027 T351 Lower outer wing panel (A380-800F)
2050 T84 Lower wing reinforcement
7010 T7651 Upper outer wing panel, heavier gauge wing ribs
7040 T7451 Fuselage main frames, cockpit, window frames, beams,
fittings
7040 T7651 Wing spars (inner front and inner center part)
7056 T7951 Upper wing panels (A380-800F)
7449 T7951 Upper wing
7449 T7651 Lower gauge wing rib
2027 T3511 Heavy Lower wing stringers
2196 T8511 sections Floor beams
7449 T79511 Upper wing stringers
2024HS T432 Small Fuselage frames
2196 T8511 sections Floor structure, fuselage stiffeners
6056 T78 Fuselage stiffeners
6056 T6 Fuselage stiffeners
7349 T6511 Seat rails, stiffeners of center wing box
7349 T76511 Fuselage stiffeners
6056 T78 Sheet Pressure bulkhead below cockpit floor
6156Cl T6 Fuselage panels

consolidated experience in using castings has been developed for military aircrafts.
A few examples of quite an extensive use of Al-alloy castings in military aircrafts
and helicopters are:
• The inlet duct of the General Dynamics F-16 Fighting Falcon.
• The pave tack pod of the General Dynamics F-111 and of the McDonnel Douglas
F-4.
• The main support of the sensor for night vision in the Apache helicopter (Starke
and Staley 1996).
The experience with military aircrafts has been exploited in the first place to
reduce the impact of CF. Therefore, the use of castings has become a valuable and
reliable choice also for primary structures. This has indeed happened for several
Airbus aircrafts, starting from the A320 onward with the following main
components:
• Baggage compartment door.
• Flap tracks.
• The inner structure of the passenger door.
The same choice has been adopted by Boeing too and has now become a standard
(Rendigs 1994).
82 3 Alloys for Aircraft Structures

Fig. 3.27 Schematic of the Airbus A380

To complete the picture concerning the perspectives of aluminum alloys, a further

alloying element is being considered for improving their structural properties: this is
scandium. Scandium (Z ¼ 21) is a light metal, having a density of 2.985 g/cm3, and it
is often associated with the group of RE elements, although it is not really part of this
group. The early reported results on aluminum-scandium alloys date back to the
early 1970s, both in the USA and in the Soviet Union. In 1971, the early Al-Sc alloys
were patented. Meanwhile, several studies started at the VILS (All-Russian Institute
of Light Alloys), demonstrating that it was possible to produce rather thick ingots of
Sc-bearing Al-Mg alloys with non-dendritic structures. Subsequently, the investiga-
tions proceeded to assess the potential of these alloys in the aerospace field
(Djukanovic 2017; Filatov et al. 2000; Röyset and Ryum 2005). The interest for
Sc-containing alloys relies on the improved mechanical properties and processing
3.2 Aluminum Alloys 83

Table 3.16 Conventional wrought Al-alloys used for commercial aircraft parts (Polmear 2006;
Staley and Lege 1993)
Alloy Temper Application Alloy Temper Application
7090 T7E71 Aircraft parts 2024 T3 Fuselage skin
7091 T7E69 7075 T6
2618 T61 Aircraft parts and structures for high- 7475 T6
2219 T62 temperature applications 7075 T6 Fuselage
2219 T87 7075 T73 stringers
2025 T6 Aircraft propellers, forgings 7475 T76
6013 T4 Aircraft sheet 7150 T77
2014 T6 Aircraft structures 6005 T5 General purpose
extrusions
2124 T8 6060 T5 General purpose
extrusions
2048 T85 6151 T6 Medium-
strength forgings
7001 T6 2024 T3 Ribs and spars
7009 T6 7010 T76
7010 T6 7150 T77
7049 T73 2017 T4 Screw machine
fittings
7050 T736 2011 T6 Screw machine
parts
7075 T6 6061 T6 Welded
structures
7075 T73 2024 T3 Wing lower
7075 T76 7075 T6 panels
7475 T7651 7175 T73
7178 T6 2024 T3 Wing lower skin
7055 T7751 7475 T73
7085 T7651 2024 T3 Wing lower
2024 T4 Aircraft structures and sheets 7075 T6 stringers
2024 T6 2224 T39
2024 T73 Empennage (tail) 7075 T6 Wing upper skin
7075 T6 7150 T6
7050 T76 7055 T77
6063 T6 Extrusions 7075 T6 Wing upper
2024 T6 Fuselage frames/bulkheads 7150 T6 stringers
7075 T6 7055 T77
7050 T6 7150 T77

possibilities. In fact, scandium additions determine the highest specific increment of

tensile strength in aluminum alloys than any other alloying element. Scandium has
also attracted interest because its presence can lead to:
84 3 Alloys for Aircraft Structures

Table 3.17 Tensile properties of selected wrought Al-alloys and relevant tempers, widely used in
aircraft structures (Polmear 2006)
Alloy Temper Tensile strength [MPa] Yield strength [MPa] %Elongation in 50 mm
2011 T6 390 295 17
2014 T6 480 410 13
2017 T4 425 275 22
2618 T61 435 330 10
2219 T62 415 290 10
2219 T87 475 395 10
2024 T4 470 325 20
2024 T6 475 395 10
2124 T8 490 440 8
2025 T6 400 255 19
2048 T85 480 440 10
6005 T5 305 270 12
6060 T5 215 195 17
6063 T6 240 215 12
6061 T6 310 275 12
6151 T6 330 295 17
6013 T4 315 185 25
7001 T6 675 625 9
7009 T6 535 470 12
7010 T6 545 485 12
7049 T73 530 470 11
7050 T736 550 510 11
7075 T6 570 500 11
7075 T73 500 430 13
7075 T76 540 470 12
7475 T7651 590 560 12
7178 T6 610 540 10
7055 T7751 630 610 12
7085 T7651 510 475 7
7090a T7E71 620 580 9
7091a T7E69 590 545 11
a
Refers to alloys produced via powder metallurgy

• Better quality of welded joints due to the reduction in the incidence of cracks,
with a consequent increase, up to 200%, in the fatigue life of the joints.
• Grain refinement during both casting and welding processes.
• Precipitation hardening.
• Recrystallization inhibition.
As concerns recrystallization, its temperature can be as high as 600  C when
scandium is added. This value is definitely above the typical heat treatment
3.2 Aluminum Alloys 85

temperatures of aluminum alloys. A better understanding of grain refinement and

hardening can be inferred from the Al-Sc phase diagram (see Fig. 3.28), featuring an
eutectic on the Al-rich side, at 659  C and 0.56 wt% scandium. On cooling from the
melt a hypereutectic composition, the Al3Sc intermetallic forms. This phase contrib-
utes to the strengthening of the alloy and also to the grain refinement during the
solidification process. In fact, the intermetallic grains are preferred nucleation sites for
the solidifying alloy, owing to the crystallographic affinity to aluminum. This aspect
can be improved by the addition of titanium or zirconium, resulting in the formation of
Al3(Sc, Ti) or Al3(Sc, Zr) particles, that, as shown by high resolution atom probe
investigations, improves the control of the intermetallic formation kinetics, in agree-
ment with the relevant transformation curves in Fig. 3.29 (Röyset and Ryum 2005).
Scandium has been added to non-heat treatable alloys of the 1XXX, 3XXX,
4XXX, and 5XXX series (Röyset and Ryum 2005). The percentage of added
scandium goes from 0.1% to 0.45% (Scandium International Mining Corp 2017).
The best combination of strength and other structural properties has been obtained
with the Al-Mg alloys (5XXX series – Filatov et al. 2000). Thanks to the homog-
enous, fine-grained microstructure of the alloys, tensile strengths in the
300–450 MPa range are obtained, still retaining a 12–15% elongation, depending
on the actual scandium concentration. The fine grain size has also positive effects on
the extrusion flow rate (Scandium International Mining Corp 2017). Sc-containing
Al-alloys are interesting for aerospace applications, with particular regard to
non-treatable alloys, like welded gas tanks, structures for dashboard panels and
compartments, fuselage stringers, and large stamped and welded structures
(Djukanovic 2017; Röyset and Ryum 2005). The Russian military aircraft MIG
29 has the largest use of Sc-containing Al-alloys so far (Djukanovic 2017; Röyset
and Ryum 2005). Airbus Group APWORKS Gmbh, in association with Airbus
Group R&D, has developed recently the Scalmalloy®, a high-performance Al-Mg-
Sc alloy specifically designed for the production of high-strength aerospace struc-
tures via laser powder bed additive manufacturing (see Table 3.18). The result is a
material with exceptional high fatigue strength and specific strength approaching
that of titanium alloys (APWORKS 2017; Djukanovic 2017).
It has been estimated that the use of high-strength Sc-containing alloys in aircrafts
would afford a 10–15% weight reduction (Djukanovic 2017). Unfortunately, the
extremely high cost of scandium (now higher than 2000 USD/kg) and its relatively
low abundance and the relevant extraction difficulties (Djukanovic 2017) seem to be
an important obstacle to a widespread application of this element, unless novel
extraction technologies will maintain their improvement promises (Röyset and
Ryum 2005). One approach is based on using, as raw material, ScO3 from the “red
mud” coming from the Bayer process (see Sect. 3.2.1). Alternatively, ScO3 may be
reduced directly in the liquid aluminum alloy. In this way the production cost is cut
by one order of magnitude approximately (Röyset and Ryum 2005). Sc-containing
alloys already on the market are the 1460, 2023, 5024, and 5025 (for full composi-
tion see Table 3.5 and Appendix 1). Other alloys, whose designation is not present in
the IADS, are 1515 and 1570 (5XXX series; 0.30–0.50% and 0.15–0.35% Sc,
respectively) and 1970 (7XXX series; 0.25% Sc).
86 3 Alloys for Aircraft Structures

Fig. 3.28 Al-rich part of the Al-Sc phase diagram. (Redrawn from Röyset and Ryum 2005)

Fig. 3.29 Transformation curves for the formation of the intermetallic phases in Al-0.4Sc and
Al-0.4Sc-0.15Zr (see main text for details). In the plot, the completion percentage of the transfor-
mation is also indicated. (Redrawn from Röyset and Ryum 2005)
3.3 Magnesium Alloys 87

Table 3.18 Typical mechanical properties of Scalmalloy®

Young’s Yield Hardness
Density modulus Ultimate tensile strength % vickers
[g/cm3] [GPa] strength [MPa] [MPa] Elongation (HV 0.3)
2.67 70 520 470 13.0 180
Values are obtained according to the DIN EN ISO 2002-001 standard and measured from speci-
mens that have been heat treated and machined. Values are typical of products obtained via ALM
(Additive Layer Manufacturing) and are measured in the least strong direction (APWORKS 2017)

3.3 Magnesium Alloys

3.3.1 Introduction

Magnesium (atomic number Z ¼ 12) is in the second group, third period of the
periodic table. It is characterized by a density of 1.738 g/cm3 and a Young’s modulus
of 45 GPa. Magnesium has several applications in the modern industry (e.g.,
steelmaking processes, Kroll’s process for titanium extraction and alloying element
in Al-alloys), and it is also an interesting metal for aerospace applications. Although
aluminum and its alloys are holding a leading role in aerospace structures, this has
not always been the case. Indeed, magnesium can be regarded as the less dense of the
metals interesting for structural applications, leaving aside lithium and beryllium,
not suited for bulk structural uses. The name magnesium derives from the Greek
name Magnesia, a district of Thessaly, a traditional geographic and modern region of
Greece (Davy 1808; Perkguleryuz et al. 2013). The existence of elemental magne-
sium was first claimed in 1755 by Joseph Black, who pointed out that magnesia
mineral (MgO) contained a new element that, unfortunately, he was not able to
isolate. In the past, magnesia was known as white earth or white stone. Magnesium
was discovered by Sir Humphry Davy in 1775 (Davy also discovered aluminum, see
Sect. 3.2.1). Sir Davy also suggested the name magnium for the metal, but magne-
sium was adopted instead. First extraction of the metal by Davy occurred in 1808,
thanks to the electrolysis of moistened magnesium sulfate. A mercury cathode was
employed in the process. The resulting product was a magnesium amalgam, from
which magnesium was distilled by removing mercury. In 1828, Antoine Alexandre
Brutus Bussy was able to isolate magnesium thanks to the fusion of dehydrated
MgCl2 with potassium at high temperature. Some years later, in 1833, Michael
Faraday obtained metallic magnesium from the electrolysis of impure magnesium
chloride in a molten state. However, it took two more decades to achieve production
capabilities of commercially relevant quantities of metallic magnesium. The first
who achieved this result was Robert Bunsen, who obtained pure magnesium using a
small laboratory cell starting from anhydrous MgCl2. At the beginning of its history,
magnesium found considerable interest in Germany only, that in 1868 was the only
producer in the world. The application of magnesium was not focused on structural
components but on the production of powder for flash-lights and fireworks. Indus-
trial production of magnesium started around the late 1920s in several countries.
88 3 Alloys for Aircraft Structures

Magnesium Elektron, a still existing company, was founded in 1934 and started its
production 2 years later, using magnesia coming from Greece as raw material.
However, when World War II was about to start, magnesia was directly extracted
from seawater. In Italy, the SAMIS company built a production plant in Sardinia in
1935. The first owner was very much related to the aerospace field. His name was
Giovanni Caproni regarded as one of the originators of the newborn Italian aircraft
industry. In 1939 the production of magnesium also started in Bolzano, a town
located in the Dolomites in the northern part of Italy. The plant was owned by the
company Società Magnesio e Leghe di Magnesio. One interesting peculiarity of the
Bolzano plant was its silicothermic production process, named Bolzano process
(Horst and Mordike 2006; Perkguleryuz et al. 2013).
Magnesium is one of the most common elements on the Earth: according to the
estimated total abundance, it is in the fourth place, after iron, oxygen, and silicon.
Magnesium occupies the eighth position in the list of the most abundant elements in
the Earth’s crust (lithosphere), considering its actual availability to mining and
extraction processes, using the so-called Clarke number. This is the concentration
of each element referred to a surface crustal layer, and although magnesium is
preceded by metals like iron and aluminum, it still has a larger Clarke number
than other strategic raw materials, like nickel, chromium, and so on (Parker 1967).
Furthermore, magnesium is the second most abundant metal in seawater, the first
being sodium (Perkguleryuz et al. 2013). Primary magnesium can be extracted from
very common carbonate ores, like magnesite (MgCO3) and dolomite (CaCO3 ∙
MgCO3). In other words, the availability of magnesium as raw material seems not
to be an issue, even in view of future developments and increase in the production.
The extraction metallurgy of metallic magnesium is based on two main processes:
electrolysis and thermal reduction of the ores.
The extraction of magnesium requires a relatively high energy consumption for
the strong chemical bonds it has with other elements in the ores. From the 1970s to
the 1990s, the production of Mg was primarily performed according to the electro-
lytic route (Table 3.19). Since the chemical reaction that takes place during the
electrolytic process is a redox process, it can be considered as made of two semi-
reactions, the cathodic and the anodic one (see Chap. 8). The electrolytic route for
the production of magnesium is well established for both MgCl2 coming from
traditional raw materials (see Table 3.19) and for the recycled molten anhydrous
magnesium chloride deriving from the Kroll’s process used for the extraction of
titanium (see Sect. 4.1). A few details on the thermal reduction of magnesium ores
and major process parameters are given in Table 3.20.
As concerns the silicothermic extraction process, it is based on two main
approaches: the Bolzano process and the Pidgeon process. In both cases, silicon
coming from a ferroalloy, named ferrosilicon (FeSi), is used as reducing agent. This
same agent is used for steel deoxidation and in the reduction stage during the steel
conversion process (see Sect. 5.3) (Gasik 2013). The Bolzano process, using dolo-
mite as raw material, was initiated by Edoardo Ravelli, who industrialized it on the
basis of the silicothermic reduction process developed by Amati in 1938 at the
University of Padua (Italy). The plant, based in Bolzano (Italy, from which the name
3.3 Magnesium Alloys 89

Table 3.19 Electrolytic processes for the production of magnesium

Raw Raw material Chemical reactions Temperature Pressure
Process material preparation involved [ C] [kPa]
AM Magnesite Mining, leaching Cathode semi- 700–800 101.325
with HCl, reaction
dehydration 2Cl ! Cl2 + 2e
Dow Seawater/ Neutralization, Anode semi-
desalination purification, reaction
brine dehydration Mg2+ + 2e ! Mg
IG Neutralization, Global redox
Farben prilling, dehydra- reaction
tion chlorination MgCl2
(l) ! Mg(l) + Cl2(g)

AM indicates a process based on the chlorination of MgO obtained from thermal decomposition of
magnesite (MgCO3). In the IG Farben process, the dehydration step is optimized by prilling
(Wulandari et al. 2010)

of the process comes), used ferrosilicon to reduce calcined dolomite. The production
reached a peak in 1981 with 10,700 metric tons and continued until the closing down
of the plant in 1992, when the costs of magnesium production became too high, thus
not profitable any longer. In the Bolzano process, still used in Brazil by a company
named Bramag, dolomite undergoes a calcination first, and it is then mixed and
compacted with the ferrosilicon reduction agent. Bauxite may be employed as an
additive in the process. The resulting material is prepared in the form of large blocks
that are connected to the electric-heating conductors. Thanks to this particular
design, the blocks only are heated up, thus avoiding heat dispersion over the entire
furnace. The process is carried out at a pressure of about 0.4 kPa and at a temperature
of 1200  C, a relatively low value if compared to the typical values for non-
silicothermic-based processes (Table 3.20), so that also the energy consumption is
comparatively lower with respect to other thermal reduction processes. The alterna-
tive silicothermic extraction process for magnesium, i.e., the Pidgeon process, was
developed by Lloyd Montgomery Pidgeon in the late 1940s. The reduction of
dolomite, like in the Bolzano process, is carried out in a reactor using ferrosilicon.
The process uses a furnace (retort) in which a briquetted mixture of ferrosilicon and
calcined dolomite powder are heated in order to obtain the reduction of magnesium
oxide coming from the carbonate decomposition. High-purity magnesium is pro-
duced and then remelted and cast in form of ingots. The process is carried out at a
temperature of about 1500  C, thus higher than that of the Bolzano process, and over
a pressure range of 0.013–0.4 kPa. The retort can be coal-fired, gas-fired, or
electrically heated (Horst and Mordike 2006). Notwithstanding specific differences,
all magnesium extraction processes can be traced back to the same main functional
stages, according to the schematic in Fig. 3.30.
Magnesium has a hexagonal close-packed (hcp) structure, whose limited number
of slip systems (three slip systems) renders this metal and its alloys intrinsically
difficult to be deformed, definitely more difficult than austenitic fcc alloys.
 90

Table 3.20 Processes involving thermal reduction of Mg-containing ores (Horst and Mordike 2006; Wulandari et al. 2010)
Raw material Temperature
Process Raw materials preparation Main chemical reactions involved [ C] Pressure [kPa]
Aluminothermic Dolomite, Al scrap Calcination 4MgO + 2Al ! 3Mg + MgAl2O4 1700 86.126–101.325
Carbothermic Magnesite, carbon Calcination and MgO + C ! Mg + CO 1700 101.325
pelleting
Magnetherm Dolomite, bauxite, Calcination and FeSi 2CaO ∙ MgO + (xFe) 1550 5.066
FeSi formation Si + nAl2O3 ! 2CaO ∙ SiO2 ∙ nAl2O3 + 2Mg + xFe
Mintek Dolomite, bauxite, Calcination 2CaO ∙ MgO + (xFe) 1700 86.126
FeSi, Al scrap Si + nAl2O3 ! 2CaO ∙ SiO2 ∙ nAl2O3 + 2Mg + xFe
4MgO + 2Al ! 3Mg + MgAl2O4
Silicothermic Dolomite, FeSi Calcination, FeSi for- MgO + CaO + FeSi ! Mg + Ca2SiO + Fe 1160–1500 0.013–0.4
mation, pelleting
3 Alloys for Aircraft Structures
3.3 Magnesium Alloys 91

Fig. 3.30 Flowchart for a general magnesium production process (Perkguleryuz et al. 2013)

Table 3.21 A comparison among properties of key elements for structural alloys, including
magnesium
Property Mg Al Ti Fe
Density [g/cm3] 1.74 2.70 4.51 7.87
Melting T [ C] 650 660 1670 1538
Boiling T [ C] 1105 2470 3287 2862
Thermal conductivity (RT) [W/  C ∙ m] 160 235 22 80
Specific heat (RT) [J/kg  C] 1025 904 520 449
Young’s modulus [GPa] 45 70 120 210
UTS [MPa] 90 98 150 275
UTS indicates the ultimate tensile strength (Donachie 2000; Pepperhoff and Acet 2001;
Perkguleryuz et al. 2013)

Table 3.21 gives some properties of this metal, as compared with the same properties
of other key metals: aluminum, titanium, and iron. The low density of magnesium
stimulated the interest for this metal in the early days of the aerospace industry.
Starting from the military field, the first important example of the use of
Mg-alloys in aeronautic structures was probably the prototype concept of
all-magnesium aircraft, the Northrop XP-56, the so-called Black Bullet. It was tested
in experimental flights in the latest years of World War II. The Black Bullet was
important, since it demonstrated the real potentials of Mg-alloys in aircraft struc-
tures. Another full magnesium structure was there in the Lockheed F-80C. Magne-
sium and its alloys started to be used in large military transportation airplanes, like
the Convair B-36, with a total amount of 8600 kg of magnesium, of which about
5600 kg of magnesium sheet and 700 kg of forged parts and 300 kg of castings
(Ostrovsky and Henn 2007). Another notable example of application of Mg-alloys to
aircraft structures is the Convair XC 99, which serviced from 1949 until 1957, and is
considered as the first double-deck airliner. This airplane had several components of
the structure and fuselage skin made of Mg-alloys. The exceptional payload and
passenger capacity (i.e., 400 fully equipped troupes), allowed by the double-deck
design, rendered this aircraft particularly attractive also for civil and commercial
92 3 Alloys for Aircraft Structures

Fig. 3.31 Pourbaix diagram of pure magnesium. (Redrawn from ASM International 2003)

purposes. In fact, Pan Am ordered 15 of these aircrafts for its fleet, although no one
of them was delivered and never started its service, because the program was first
suspended and then cancelled. Among the reasons for this change of plans, there
were the emerging limitations and critical aspects exhibited by magnesium alloys
available at the time, particularly as concerns their surface durability.
The poor corrosion resistance is particularly evident when analyzing the Pourbaix
diagram of pure Mg in aqueous environment (Fig. 3.31), where corrosion takes place
over a wide range of pH and potential. Since it was not an issue in the military field,
this feature was initially overlooked, although it emerged soon as an important
aspect for commercial fleets in terms of reliability and costs associated with main-
tenance and repair. These aspects and the contemporary rapid, solid evolution of new
aluminum alloys determined the progressive abandon of magnesium in air transpor-
tation. Quite different was the situation in the Soviet Union, where military air-
craft structures were systematically used for prototyping solutions to be used for civil
transportation purposes, with the possibility of faster development and validation
loops for new materials. The Boeing 737 (produced since 1967) features an
extremely limited number of magnesium components, summing up to a total weight
of a few tens of kilograms, much lower if compared to the Tupolev TU-134,
3.3 Magnesium Alloys 93

Fig. 3.32 Evolution of the weight of magnesium used for the different models of Tupolev planes,
developed over the years (Gupta and Gupta 2017; Ostrovsky and Henn 2007)

developed over the same years (early production date 1963). In this case, the
components made of magnesium alloys had a total weight of 780 kg. The Tupolev
planes can be taken as important witnesses of the use of magnesium alloys for
civilian air transportation. After the golden years for magnesium, 1960s and 1970s, a
progressive reduction has been observed also in the Soviet Union, as can be seen in
Fig. 3.32, referring to the use of magnesium in Tupolev airplanes. The situation in
Russia in the 1990s was substantially similar to that of the Western aerospace
industry, whose main players never used magnesium in their commercial and
passenger airplanes intensively. The same is not true for helicopters, for which
magnesium is a standard choice for gearboxes, transmission between engine and
rotor, and other components, mainly nonstructural. In this case, weight saving is a
decisive selection parameter. The comparatively limited applications of magnesium
in aeronautic structures, if compared to aluminum alloys and notwithstanding the
interesting properties and potential of this class of materials, can be ascribed to a set
of concurrent reasons, in addition to the already mentioned poor corrosion resis-
tance. The elevated oxidation kinetics, resembling combustion under particular
conditions, has been obviously regarded as a negative issue in applications for
which safety and reliability are paramount, although the magnesium combustion
issue is to be regarded as an example of misconception. In real terms, combustion
may become an actual risk for magnesium powder samples only and requiring some
precautions in handling, for instance, scraps from machining of magnesium parts.
94 3 Alloys for Aircraft Structures

All commercial magnesium alloys are compliant and fully certified as concerns
the resistance to ignition, although limitations to their use in specific components
have been imposed, owing to the strict safety standards and legislation adopted for
aerospace products. However, in recent times, a ban on the use of magnesium alloys
for aircraft seats has been lifted, once the presumed flammability risk of candidate
magnesium alloys has been ruled out with lab size and full-scale experiments
(Marker 2013; Nature Materials 2016).
In addition to the limited corrosion resistance, the inadequate mechanical strength
was identified as another key property to be improved in order to enable a wider use
of magnesium alloys also in the aerospace industry. The important research efforts
resulted in the developments of the main classes of alloys, described in the following
with regard to specific aerospace applications. As concerns processing, casting
appeared from the beginning as the obvious choice, considering the comparatively
low-melting temperature of magnesium and its low latent heat of fusion, which
renders solidification and cooling rather fast. The role of casting techniques was
strengthened further by the intrinsically low deformability of magnesium for its
limited number of slip systems. The development of more ductile alloys with novel
compositions has incremented the number of wrought magnesium alloys available
on the market. In this regard, the mechanical properties achieved by the state-of-the-
art magnesium alloys, like the high-strength Elektron 21 and Elektron 675 materials,
make them potential valid replacements for aluminum alloys, laminate, and polymer
matrix composites in aeronautic structures (Lyon et al. 2007). This trend is encour-
aged by the always standing issue of the weight reduction, that is, nowadays
considered also for the important beneficial effects, concerning fuel saving and
consequent reduction of environmental pollution (see Chap. 2). Since the 1990s,
there has been renewed interest in magnesium alloys for weight-sensitive applica-
tions, including the aerospace field. At the beginning of that period, there were a few
commercial magnesium alloys systems: Mg-Al, Mg-Al-Zn, Mg-Zn-RE, and
Mg-REs. Most magnesium alloys usage for structural applications was in
die-casting, and a single alloy (AZ91D) accounted for around the 90% of the total
market. A clear indication of the increasing, renewed interest for magnesium can be
also inferred by the growing trend of magnesium production, started in the early
1990s, under the incisive drive by China, that is holding nowadays 80% of the global
supply of this element (see the historical development in Fig. 3.33). When China
entered the global economy in the early 1990s, it made magnesium one of the
fundamental items of the raw materials and metallurgy industry portfolio. The
base approach is the silicothermic Pidgeon production process, facilitated by the
large availability of low-cost ferrosilicon and using coal as main heat source. This
evolution has probably fostered the interest in magnesium also in the rest of the
world, particularly as concerns its use in civilian transportation, as reported, for
instance, in the research documents by the European Community AEROMAG
program for the aerospace application of magnesium alloys (Perkguleryuz et al.
2013).
3.3 Magnesium Alloys 95

Fig. 3.33 The global magnesium industry in 2011. Quantities are expressed in thousands of metric
tons. (Redrawn from The CM Group 2012)

3.3.2 Aerospace Magnesium Alloys: Metallurgy, Main

Designation System, and Tempers

The ASTM (American Society for Testing and Materials) international designation
system is the most widely accepted for both cast and wrought magnesium alloys,
although it is not the only one, being a possible alternative of the so-called British
system. The alphabetic part of the alloy name refers to the main alloying elements,
according to the symbols listed in Table 3.22. The first digit of the code represents
the rounded off concentration (in wt%) of the majority alloying element, identified
by the first letter. The second figure corresponds to the concentration of the second
element and so on. Additional letters, following the concentration reference num-
bers, represent possible modifications to the same alloy material. Some discrepancies
are there, but it is not difficult to sort them out and identify the actual alloy product,
also thanks to the conspicuous available literature on this matter (Cahn et al. 2005;
Cardarelli 2008).
The AZ91 code indicates the Mg-9Al-1Zn cast alloy, with the relevant compo-
sition range and minority elements indicated in Table 3.23. The ZMC711 indicates
the wrought alloy with composition: for Zn-6.5-Mn 0.75-Cu-1.25 (Polmear 2006).
Therefore, there is no specific differentiation between the cast and wrought products,
as seen instead for aluminum alloys. Other magnesium alloys are simply identified
by the relevant trademark, like the Elektron alloys produced by Magnesium
Elektron, and codenamed using a reference number, like the Elektron 21 alloy. A
96 3 Alloys for Aircraft Structures

Table 3.22 ASTM designation code for the main classes of magnesium alloys (ASM International
1991)
Code letter Alloying element Code letter Alloying element
A Aluminum M Manganese
B Bismuth N Nickel
C Copper P Lead
D Cadmium Q Silver
E Rare earths R Chromium
F Iron S Silicon
G Magnesium T Tin
H Thorium W Yttrium
K Zirconium Y Antimony
L Lithium Z Zinc

Table 3.23 Chemical composition of some magnesium alloys

Alloy Al Mn Th Zn Zr Other elements
AM50 5.0 0.30 – – – –
AZ31 3.0 0.20 – 1.0 – –
AZ61 6.5 0.15 – 1.0 – –
AZ81 7.0–8.0 0.13–0.35 – 0.40–1.0 – Be ¼ 0.0008 max
AZ91 8.1–9.2 0.17–0.35 – 0.40–1.0 – –
AZ91E 9.0 0.10 – 0.50 – –
AZ92 8.3–9.7 0.15–0.50 – 1.7–2.3 – Be ¼ 0.0008 max
AZ92A 9.0 0.10 – 2.0 – –
ELECTRON General chemical composition is property of Magnesium Elektron
675
ELEKTRON – – – 0.50 0.20–0.50 Gd ¼ 1.0–1.7
21 Nd ¼ 2.6–3.1
EZ33 – – – 2.0–3.0 0.50–1.0 RE ¼ 2.5–4.0
EZ33A – – – 2.0–3.0 0.40–1.0 RE ¼ 2.5–4.0
HK31 3.0 – 3.25 0.30 0.70 –
HM21 – 0.45 2.0 – – –
LA141 1.0 – – – – Li ¼ 14.0
LS141A 0.50–0.80 – – – – Li ¼ 13–15
QE22 – – – – 0.40–1.0 Ag ¼ 2.0–2.7
RE ¼ 2.0–3.0
WE43A – – – – 0.70 RE ¼ 3.4; Y ¼ 4.0
WE54 – – – – 0.50 Ag ¼ 5.25 Nd ¼ 3.5
ZE41 – – – 4.2 – RE ¼ 1.2
ZE63 – – – 5.5–6.0 0.40–1.0 Cu ¼ 0.10 max
Nd ¼ 2.1–3.0
ZK60 – – – 5.5 0.45 –
ZMC711 – 0.75 – 6.5 – Cu ¼ 1.25
Chemical compositions can slightly change according to the type of product (i.e., sheets, forgings,
extrusions) (ASM International 1991; Czerwinski 2011; Magnesium Elektron 2012; Magnesium
Elektron 2014, 2015; Perkguleryuz et al. 2013; Polmear et al. 2017; Prasad and Wanhill 2017)
3.3 Magnesium Alloys 97

Fig. 3.34 Alloying behavior of magnesium. Values among brackets are the solid solubility limits of
the element in Mg. (Redrawn from Pekguleryuz et al. 2013)

list of some of the most common magnesium alloys is given in Table 3.23, including
their compositions.
The compositions of the magnesium alloys are designed according to the standard
physical metallurgy principles for the formation of solid solutions and intermetallic
phases, aiming at the optimization of the structural properties and corrosion resis-
tance under different operating conditions. The graph in Fig. 3.34 shows how atomic
size, electronegativity, and valence influence the solid solubility of several elements
in magnesium, providing different effects, as specified in Appendix 3. Elements to
the left of the curve have at least 0.5% maximum solid solubility in magnesium. The
highest solubility belongs to those elements showing favorable size factor and
valence, meaning a low difference in atom size and a valence equal to 2 or 3. Con-
siderable solid solution strengthening occurs at around 12–15% size factor values,
for instance, with Al, REs, and Zn. When Hume-Rothery rules are not satisfied,
second phase formation occurs (Perkguleryuz et al. 2013). An important difference
with respect to the tempers of Al-alloys is that the equilibrium phases foreseen by the
relevant phase diagrams are not to be regarded as detrimental, as in the case of
overaged aluminum alloy, but absolutely beneficial, being important strengthening
agents. The equilibrium phases that are interesting as concerns precipitation hard-
ening of magnesium alloys belong to three main intermetallic types. The first is the
simple cubic CsCl structure such as that of MgTl, MgAg, CeMg, and SnMg. The
second is that of the Laves phases, with an AB2 stoichiometry like MgCu2, MgZn2,
and MgNi2, each featuring a different stacking sequence of the closest packed
planes. Finally, the third type is that of the compounds with the antifluorite structure,
like Mg2Si, and with the fluorite structure, like Mg2Sn. The equilibrium phases used
as strengthening agents demonstrate the intrinsic stability of the alloy conditions
induced by treatments in these alloys.
The equilibrium intermetallic phases are important in magnesium alloys also for
their beneficial effect on creep resistance and for hindering alloy recrystallization
(Pekguleryuz et al. 2013). Concerning the heat treatments of Mg-alloys, the temper
condition is indicated in the alloy designation code, according to the specifications
98 3 Alloys for Aircraft Structures

Table 3.24 Designation letters of the main tempers for magnesium alloys (ASM International
1991; Pekguleryuz et al. 2013)
Temper
Code description Details
F As-fabricated –
O Annealed –
H Strain H1: strain-hardened only
hardened H2: strain-hardened and partially annealed
H3: strain-hardened and stabilized
The number following the letter H indicates the degree of strain hard-
ening remaining in the alloy. Additional digits (H2X) indicate a sub-
division of the H tempers giving information about the final degree of
strain hardening between 0 (annealed) and 8 (totally hard)
T Tempered T4: solution heat treatment
T5: artificial age hardening
T6: solution heat treatment followed by artificial aging
T7: solution heat treatment and subsequent stabilization
T8: solution heat treatment, cold working, and artificial aging
W Solution heat –
treated

contained in Table 3.24 with the main temper treatments for magnesium alloys.
Those more commonly used for aerospace alloys are the T4 (solution-treated alloy),
T5 (direct cooling and artificially aged), and T6 (solution-treated and artificially
aged) (Eswara Prasad and Wanhill 2017).

3.3.3 Magnesium Alloy Processing

The alloy design and development have essentially concerned cast products, which
still cover the majority of magnesium alloy production, notwithstanding the contin-
uous progress in the development of deformable alloys, as discussed in the next
section. Casting, using different techniques, with some restrictions and recommen-
dations determined by the target grade and by other properties required by specific
applications, is still the most flexible processing route for magnesium alloys. The
main casting methods that are used for processing magnesium alloy parts are:
• Investment and die-casting.
• Permanent and semi-permanent mold casting.
• Sand casting.
A full description of these technologies, specifically applied to magnesium and its
alloys, can be found in Polmear (2006), in the ASM Specialty Handbook: Magne-
sium and Magnesium Alloys (1999) and in the ASM Handbook: Volume 15 – Casting
(2008). Wrought magnesium alloys, featuring in general better mechanical proper-
ties than cast alloys, although more challenging to be produced for the objective
3.3 Magnesium Alloys 99

difficulties concerning their formability and therefore production costs, are gaining
increasing importance. In fact, the possibility of using magnesium alloys as wrought
products is paramount to sustain the general tendency to increase further the overall
volume of magnesium products. As concerns aerospace applications, with particular
regard to civil and commercial aircrafts, it is clear that the enhancement in the
workability of magnesium alloys is a necessary step forward to widen the possibil-
ities of these alloys in this field. Considering the standard parts of a commercial
airliner, like the B757, it turns out that only 1% in weight are castings, whereas
worked parts are the majority: 28% extruded components, 20% sheets, 18% forg-
ings, 16% plates, and 13% bars (Kettner et al. 2007). The interesting aspect is that the
main materials of which magnesium alloys are potential replacements, aluminum
alloys and structural composites, sum up in this case to 78% and 3%, respectively,
which is an extremely large volume. Continuous research efforts, backed by indus-
trial and scientific projects, are actively working to find reliable solutions to this
strategic issue. The EU project AEROMAG (6th Framework Program of the
European Commission 2005) can be regarded as an interesting example of this
challenge. The program was aimed at developing new magnesium wrought products
(sheets and extrusions), particularly for aerospace structures and components, with
excellent static strength and fatigue resistance. The benchmarks for these novel
materials were two aluminum alloys: 5038 for nonstructural applications and 2024
for secondary structures. The three independent slip systems of the hcp Mg crystal-
lographic structure are not sufficient to guarantee the homogeneous polycrystalline
plastic deformation, since, according to the von Mises criterion, at least five of these
slip systems are necessary (Hosford 2010). To overcome this intrinsic obstacle,
higher temperatures are required to achieve adequate workability. Accordingly,
temperatures close to 300  C are absolutely recommended, and, indeed, magnesium
alloys used in the aircraft industry are mechanically formed over a temperature range
between 290 and 450  C (Davis et al. 2007). Similarly, to make the process more
effective, accurate, and fast, a preliminary heating up of the working tools at
temperatures ranging from 250 to 350  C is adopted, if necessary. In general,
these temperatures are kept constant during the entire process, which is thus
conducted under nearly isothermal conditions.

3.3.4 Development of Aerospace Magnesium Alloys

As seen, magnesium alloys experienced an alternating interest in the field of

aerospace applications since the beginning of the aerospace era. Early compositions
showed, in fact, several features that, in the long run, limited their application in
aircraft structures. The main development lines of these alloys have been focused on
the improvement of the following properties:
• Corrosion resistance, that in the early formulations was relatively poor.
100 3 Alloys for Aircraft Structures

• High-temperature creep resistance, a property that is particularly interesting for

the relatively high-temperature applications of magnesium alloys, ranging from
150 up to 350  C.
• Strength and elongation (Pekguleryuz and Celikin 2010).
A galvanic attack (see Sect. 8.1) is often observed at the contact region of a
magnesium alloy part with dissimilar alloys. Heavy metals, reaching the magnesium
surface from atmospheric deposition, may produce a generalized pitting attack, due
to a localized galvanic coupling (Esmaily et al. 2017). The protection of the alloy
surface by suitable coating systems, namely, synthetic and electrical insulating ones,
is probably the most effective and practical strategy. Corrosion failures are usually
associated with internal or surface contamination. In the first case, copper, iron, and
nickel contamination can be introduced during casting operations. Another internal
contamination phenomenon can be due to the inclusions coming from salt fluxing
employed during casting and recycling operations. On the other side, concerning
surface contamination, blast residues can have detrimental effects on the corrosion
resistance of the material. These residues would typically come from blasting
operations, used for improving the surface uniformity and appearance (Horst and
Mordike 2006). In this case, a twofold corrosion protection strategy can be
implemented: firstly, through the removal of the surface contaminants by surface
treatments (e.g., chemical etching and sand blasting), possibly in association with the
deposition of protective coatings. Secondly, a better control of the processing
conditions, particularly during casting operations, will reduce the occurrence of
contaminants and impurities automatically. The improvement in the alloy grade
has generally resulted in better performing materials, also as concerns corrosion
resistance. Of course, the search for better magnesium alloys, in terms of corrosion
resistance, has been also the opportunity for developing novel coating systems and
surface treatments. The surface protection strategies developed for these alloys are
based on:
• Anodizing.
• Cold-spraying deposition.
• Conversion coatings.
• Electrochemical plating.
• Gas-phase deposition.
• Laser surface alloying and/or cladding.
• Organic coatings and paints.
• Plasma gel coatings.
Organic and composite coatings have also been specifically designed for aero-
space industry and then transferred to other fields of applications. These coatings,
being directly exposed to external conditions and undergoing structural and chem-
ical evolution under environmental and operating conditions, are prone to release
degradation products. Of course, also for organic coatings, the weight issue is
important. Therefore, single rather than multiple layers are preferable. Possible
3.3 Magnesium Alloys 101

alternatives are the wet methods, like conversion coatings, electrochemical plating,
anodizing treatments, etc. (Blawert et al. 2006).
They are in general less expensive than advanced organic coatings, although still
requiring particular precautions for the presence in their formulations of chromates
and cyanides, both toxic to the environment and human health. In this regard, dry
methods, using thermal spraying, laser melting, physical and chemical vapor depo-
sition, and solid diffusion, are certainly more appealing. The best choice, as usual, is
a trade-off of several factors, including the costs of the infrastructures and consum-
ables. Considering the complexity of the requirements and conditions, composite
coatings, like the Gardobond® coating, and surface treatments, like PGA ALGAN
2M, are attracting specific research interests in this field (Ostrovsky and Henn 2007).
These coatings would not only make magnesium alloys as stable as aluminum but
provide also an excellent surface finish for subsequent application of paints for an
improvement of the surface barrier effect and also for decorative purposes.
In the aerospace industry, the first magnesium alloys that were developed
contained Al, Mn, and Zn. AZ81, AZ91, and AZ92 are important early representa-
tives of this group (Table 3.23). Referring the reader to Appendix 3 for a complete
description of role of alloying elements, it is herewith reminded, as concerns these
important aerospace alloys, that aluminum has positive effects on the mechanical
properties of the alloy, enhancing its ultimate strength. Furthermore, aluminum
improves creep resistance and the mechanical properties at elevated temperatures.
Zinc has a strengthening effect too, although it is also important for the grain
refinement, in particular, by preventing excessive grain growth during solidification
and cooling. Zinc-containing alloys also show improved workability, quite a critical
issue for these alloys. Zinc is generally associated to manganese. This latter element,
in addition to the enhancement of the creep resistance of the alloy, has a central role
in improving its corrosion resistance. It is now clear that even very little concentra-
tions of iron (a tolerance level of 170 ppm of Fe is reported), generally deriving from
contamination during primary casting, have an essential role in the corrosion behav-
ior of magnesium alloys (Atrens et al. 2015). By controlling the alloy grade and the
formation of stable Mn-Fe intermetallic phases, that neutralize the galvanic action of
iron, a significant improvement in the corrosion resistance of the alloy can be
attained. Very much the same effect of iron has been observed also with other
impurities, particularly transition metals, determining detrimental galvanic cou-
plings with the alloy. The intermetallic precipitates and secondary phases, like
those obtained with temper treatments (Table 3.24) and having in general a higher
electrochemical potential than the magnesium alloy matrix, affect the corrosion
behavior of the alloy. Also in this case, the prevailing mechanism is a localized
corrosion attack. Actually, the effect of impurity and secondary phases can be
inhibited if their average grain size is kept below a critical value, estimated in
4 nm for standard aluminum alloys. The same can be assumed for magnesium
alloys, considering that on this length scale, the effects of corrosion can be mended
by the growth of surface reaction layers. The beneficial effects of a controlled
refinement of the alloy microstructure will be recalled next with reference to
Mg-Li alloys, being both these elements particularly reactive, and with a low
102 3 Alloys for Aircraft Structures

Fig. 3.35 Mg-Gd phase diagram. (Redrawn from ASM International 1992)

position in the galvanic series. Another important class of Mg-alloys, developed

next, featured as main alloying elements rare earth (REs) elements and zirconium, in
addition to zinc that is still present.
RE elements are interesting for their good solubility in magnesium (see
Fig. 3.34). This aspect can be seen in the Mg-Gd phase diagram in Fig. 3.35,
exhibiting also a low-temperature eutectic. A technological consequence of this
behavior is an excellent castability of RE-containing Mg-alloys. To reduce the
costs of the raw materials, RE elements are used in these alloys in the form of the
so-called misch metals. These are mixtures of several RE elements based either on
cerium (Ce55-La20-Nd15-Pr5) or neodymium and praseodymium. This latter mix-
ture is often designated as didymium misch metal. This name derives from that of
didymium, a mixture of neodymium and praseodymium that, in a first attempt, was
considered as a new metal when Gustav Carl Mosander discovered it in 1840. Few
years later, it was demonstrated that didymium was not really a new element in the
periodic table but a mixture of oxides of the two elements, i.e., neodymium and
praseodymium (Meyer 2001). The optimization of the mechanical properties,
strength and creep resistance in particular, required alloying with zinc and zirco-
nium, as anticipated. The already attested beneficial effect of zinc is enhanced by
zirconium, which is added both as a grain refiner and a precipitation hardener. A
refined and homogeneous microstructure, attained thanks to the right selection of the
composition and subsequent temper, can stabilize the alloy microstructure and
prevent fast recrystallization, an important aspect when the component is bound to
be serviced in creep conditions. To this group of alloys, whose development started
quite early (1940s) and continued until recently, belong several materials used in
3.3 Magnesium Alloys 103

Fig. 3.36 Mg-Li phase diagram. (Redrawn from ASM International 1992; Perkguleryuz et al.
2013)

aircraft structures, like the EZ33, ZE41, ZE63, and the more recent ELEKTRON
21 (see Table 3.23). Strictly associated with REs-Mg-alloys are the alloys of the
Mg-Ag system. The central feature of these alloys is the remarkable improvement of
the mechanical properties, in particular tensile strength, observed, particularly in
Mg-REs-Zr materials, with silver addition in concentrations close to 2.0%, which
seems to be a critical value to differentiate Mg-Ag-REs-Zr alloys from standard Mg-
REs-Zr (Polmear 2006). In fact, in the QE22 alloy (Table 3.23), one of the most used
of this family, a concentration of silver equal to 2.5% is present. For silver-
containing alloys, the use of the didymium misch metal is particularly recommended.
Alloys combining excellent mechanical properties, also in creep regime up to
300  C, and corrosion resistance have been obtained using yttrium as majority
alloying element. In the WE54 alloy, for instance, a concentration of yttrium equal
to 5.1% is present, attained using a particular misch metal, dominated by the
presence of yttrium (75% approx.), in association with gadolinium and erbium. In
analogy with Al-alloys, for which Al-Li compositions represent the latest develop-
ment of their design; lithium has also been selected as alloying element for magne-
sium alloys. The addition of lithium reduces further the already low density of these
systems, which represent at the present time the lightest metallic structural materials
(Wu et al. 2015).
In the binary Mg-Li phase diagram (Fig. 3.36), the α-phase, i.e., the hexagonal
solid solution of lithium in magnesium, is stable up to a concentration of 5.5 wt% Li
104 3 Alloys for Aircraft Structures

in the phase diagram. Above this limit and up to 10.3 wt% Li, a mixture of α + β
phases is present, being β a bcc solid solution of magnesium in lithium, (Li) in the
phase diagram. The β-phase is the only one present at room temperature above
10.3 wt% up to pure lithium axis. Any raise in the concentration of β-phase improves
the ductility of the alloy, an aspect that renders the Mg-Li alloys ideal candidates for
metal forming, although strength and creep resistance tend to be reduced by the
presence of the β-phase. Moreover, the high chemical reactivity of lithium, partic-
ularly in association with magnesium, requires a particular attention in managing the
oxidation and corrosion resistance of these alloys. Initially, the alloy design regarded
compositions in which standard alloying elements, already encountered in magne-
sium alloys, are present. Several families of alloys were developed: LA (Mg-Li-Al),
LZ (Mg-Li-Zn), and LAZ (Mg-Li-Al-Zn), strengthened by the formation of inter-
metallics, like MgLi2Al, MgLi2Zn, AlLi, and MgLiZn (see Sect. 3.3.2). Specific
additions have been explored in the attempt to improve the alloy properties and
compensate for the partial instability exhibited by some of the above strengthening
intermetallic phases. Starting from the reference compositions of the LA and LAZ
alloys, the formation of stable intermetallic phases was attained thanks to the
addition of Ag, Ca, Cu, and REs (Zhang et al. 2013). These alloy developments
indeed achieved the expected results, although some limitations, pertaining surface
stability, are still there and hinder the widespread, safe use of these alloys. However,
strengthening mechanisms, which already proved successful in other Mg-alloys,
have been applied to improve the properties of Mg-Li alloys too. Mg-RE-Zn alloys,
in particular those with RE concentration in excess of the zinc concentration, exhibit
an improved room- and high-temperature mechanical strength for the presence of the
so-called long-period stacking ordered (LPSO) structures (Zhu et al. 2010). LPSO
structures are based on intermetallic compositions, like Mg96Y2Zn2, featuring a
peculiar crystalline superlattice, as results from electron diffraction investigations.
TEM observations have also demonstrated that the LPSO determine a fine lamellar
structure within the relevant domains. The formation of the LPSO structures has
been reported to occur in as-cast alloys, although not with optimized properties. For
a complete exploitation of the benefits introduced by these ordered phases, solution
heat treatments, possibly followed by extrusion, are recommended. The first reported
application of this approach to Mg-Li alloys is the study on the cast Mg-8Li-6Y-2Zn
(wt%) alloy, solution treated at 500  C for 6 h in an inert atmosphere and then
hot-extruded with an extrusion ratio of 14:1 (Zhang et al. 2013). The resulting
mechanical properties turned out to be particularly good, as concerns the ultimate
tensile strength, the tensile yield strength, and the elongation to fracture. If extrusion
is preceded by the above mentioned thermal treatment, the refined microstructure
shows mechanical properties up to 243 MPa, 187 MPa, and 30%, starting from
202 MPa, 142 MPa, to 19%, respectively, in alloy samples that were just extruded
but not heat-treated. The microstructural refinement, resulting from the
thermomechanical treatments imparted to the alloy samples, has also an indirect
beneficial influence on its corrosion behavior, since it reduces the detrimental micro-
galvanic effect associated with presence in the alloy of intermetallic particles and
unwanted impurities. A similar approach has been applied also to fully β Mg-Li
3.3 Magnesium Alloys 105

alloys, as reported in (Xu et al. 2015). Again, using specific thermomechanical

treatments, remarkable improvements in the alloy mechanical strength and corrosion
resistance have been achieved retaining at the same time an excellent ductility. The
main processing steps adopted for Mg-Li alloy materials comprise hot-extrusion and
solution heat treatments, followed by water quenching. Furthermore, aging treat-
ments, conducted at low temperatures (<100  C), and cold-rolling are imparted to the
alloy in order to stabilize the microstructure and relevant properties. The newly
developed alloy exhibits also a very attractive corrosion behavior that overcomes the
intrinsic weakness of α- and α + β alloys, which is determined by combined effect of
the alloy composition, in particular the elevated concentration of lithium, and the
microstructural refinement induced by the processing treatments. The investigated
Li-rich alloy is characterized by a single-phase bcc microstructure containing a fine
Li-rich solute nanostructure. The solute nanostructure is likely to be formed thanks
to the high diffusivity of Li in the Mg lattice at low temperatures. Furthermore, the
most interesting feature of this alloy is probably its corrosion behavior, being the
Li-rich solute nanostructure substantially electrochemically homogeneous, thus
increasing the intrinsic corrosion resistance of the alloy, in which inhomogeneities
acting as preferential corrosion nucleation sites have been eliminated (Xu et al.
2015). A comparative picture of the way Mg-Li alloys perform in air is shown in
Fig. 3.37. The scheme illustrates how the surface region of different Mg-alloys reacts
to standard atmospheric conditions. The poor oxidation resistance of Mg-rich
α-alloy is due to the incomplete coverage by the oxide layer that spontaneously
forms on the alloy surface (Fig. 3.37, incomplete coverage). Therefore, the alloy is
not insulated from the surrounding environment, and oxidation can proceed, possi-
bly accompanied by the hydration of the outer oxide. Higher concentrations of
lithium, like in two-phase α + β alloys, result in the oxidation of this element too,
to form Li2O (Fig. 3.37, preferential attack). The α-grains, which are still present,
behave similarly to what is observed in the case of a fully α-alloy. From the scheme,
it is also evident that the finer is the microstructure of the alloy, the more homoge-
neous is the coverage of the alloy itself produced by the outer reaction layer. The
interesting novelty introduced by these advanced Mg-Li alloys is the active role of
lithium, which, thanks to the refined microstructure of the alloy, better if with a
nanometric grain size, forms together with Mg a compact mixed oxide layer.
Moreover, lithium oxide (Li2O), reacting with carbon dioxide to form Li2CO3,
improves the protection performances of the outer reaction layer. Thanks to the
combined oxidation and carbonation reactions, a complete coverage of the metallic
surface can be achieved in case of the fully bcc nanostructured Mg-Li alloy
(Fig. 3.37, complete coverage). This latter alloy is definitely better performing as
concerns corrosion resistance, and this is due both to the complete surface coverage
and to the underlying matrix with the nanostructured features avoiding corrosion to
initiate at inhomogeneities contained in the material (Xu et al. 2015).
Moreover, it turns out that this reaction layer has excellent self-healing features,
so that it can also stand possible damages, thus behaving very much like the self-
passivating film of native oxides, which form on the surface of aluminum, stainless
steels, and titanium alloys. These novel Mg-Li alloys exhibit structural properties
106 3 Alloys for Aircraft Structures

Fig. 3.37 Schemes of the surface reactions of different Mg-Li alloys when exposed to air with
reference to three matrix conditions: conventional hcp Mg-rich α-alloy, bcc + hcp Mg-Li extruded
alloy, and bcc nanostructured Mg-Li alloy. (Redrawn from Xu et al. 2015)

and surface stability that render them potentially interesting for real applications,
including aerospace ones.
The research efforts that have been spent for this task indicate anyway the
growing, in some respect, renovated interest for Mg-alloys, as proved also by the
ever-growing trend in Mg production shares (Fig. 3.33) and by the variety of
available mechanical properties, and related tempers, provided by state-of-the-art
magnesium alloys (Table 3.25).

3.3.5 Current Aerospace Applications

The aircraft components in Fig. 3.38 provide an interesting overview of the potential
afforded by metal working technologies of magnesium alloys. These encouraging
results are promoting new investigations for the development of fully enabled
production routes, suitable also for the aerospace field. The explorative manufactur-
ing of aircraft brackets, with different geometries, has attained excellent results in
terms of part control and resulting properties, absolutely comparable, if not better
than those of the benchmark cast counterparts, that still represent the larger part of
Table 3.25 Selected Mg-alloys and tempers with resulting mechanical properties
Thermal Tensile strength Tensile yield strength Compressive yield strength Elongation Product
Alloy treatment [MPa] [MPa] [MPa] [%] Hardness form
AM50 F 268 168 – 18.0 – Extrusions
AZ31 F 245 165 – 12.0 – Extrusions
AZ31 H24 325 180 – 15.0 – Sheets
AZ61 F 280 165 – 14.0 – Extrusions
3.3 Magnesium Alloys

AZ91A F 230 150 165 3.0 63 HRB Castings

AZ92A T6 275 150 150 3.0 84 HRB Castings
ELEKTRON T6 248–280 145–170 168 2.0–5.0 – Castings
21
ELEKTRON T5A 300–350a 200–230a – 1.0–5.0a – Extrusions
675
ELEKTRON T5B 340–400a 230–260a – 1.0–3.0a – Extrusions
675
HK31 H24 285 160 – 9.0 – Sheets
HM21 T8 270 130 – 11.0 – Sheets
LA141 T7 132–165 105–160 – 11.0–24.0 – Sheets
WE43A T6 250 165 – 2.0 75–95 Castings
HRB
WE54A T6 250 172 172 2.0 75–95 Castings
HRB
ZE41A T5 205 140 140 3.5 62 HRB Castings
ZE63A T6 300 190 195 10.0 60–85 Castings
HRB
(continued)
107
Table 3.25 (continued)
108

Thermal Tensile strength Tensile yield strength Compressive yield strength Elongation Product
Alloy treatment [MPa] [MPa] [MPa] [%] Hardness form
ZK60 F 325 240 – 13.0 – Extrusions
ZK60 T5 330 268 – 12.0 – Extrusions
ZK60A T5 365 305 250 11.0 88 HRB Extrusions
ZMC711 T6 325 300 – 3.0 – Extrusions
References and details about temper treatments are given in Table 3.24 (ASM International 1991; Magnesium Elektron 2014, 2016; Pekguleryuz et al. 2013)
a
Properties of samples up to 100 mm in diameter. Lower value is measured in transverse direction, while higher value is measured in longitudinal direction
3 Alloys for Aircraft Structures
3.3 Magnesium Alloys 109

Fig. 3.38 Some aircraft components obtained by different forming techniques and made of
different Mg-alloys, as specified in the following. (a) WE43 impeller with twisted blades; (b)
AZ31B support of the Eurocopter antenna – deep drawing; (c) AZ80 door stop fitting – forging; (d)
WE43 compressor upper case for air conditioning plant – closed-die forging; (e) AZ80 window
frame for Airbus plane – forging (Dziubinska and Gontarz 2015, reproduced with permission of the
Publisher)

magnesium alloy components (Dziubinska and Gontarz 2015; Dziubinska and

Gontarz 2016). The weight saving introduced by magnesium alloys encouraged
their widespread and successful use for helicopters. Some examples of applications
of magnesium alloys for helicopter and airplane components are given in Table 3.26.
The future applications of magnesium alloys in the aerospace field seem to be
liable of important expansion. In fact, as suggested by the examples in Fig. 3.38 and
Table 3.26, the consolidation of metal working technologies for the manufacture of
magnesium alloy components is widening the application possibilities of these
materials. This process route is certainly benefitting from the established use of
these alloys in the automotive industry and in other large-scale production industrial
fields, leading to the creation of knowledge and solutions for high-quality
components.
110 3 Alloys for Aircraft Structures

Table 3.26 List of some magnesium alloys used in the fabrication of components of helicopters
and aircrafts (Gupta and Gupta 2017)
Alloy Applications
AZ19E Sikorsky CH33D-Seastallion with a part weight of 238 kg
After the T6 temper, it is used for the sump of the six-cylinder 540 Lycoming
engine used to power the Piper Chieftain
Fabrication of some components for the Piper Comanche
Fabrication of some components for the Sikorsky CH53D
AZ92A Thrust reverser panels obtained by cascade casting used in the Boeing B737,
B747, B757, and B767 airliners
ELEKTRON 21 Parts of the Boeing AH64D Apache
EZ33A Parts of the gearbox of the Rolls-Royce turbofan engine RB211
QE22 Compressor casing of the ALF 507s engine
Gearbox of the General Electric F110 engine
WE43 Parts of the MD 600 and Sikorsky S92
WE43 Pratt & Whitney F119 auxiliary casing
WE43 Castings used in the gearbox of the Bell Augusta 609 twin engine tilt rotor
VTOL (vertical take-off and landing) aircraft
ZE41 Parts of the Sud-Aviation Super Frelon helicopter
Parts of the Boeing AH 64 Apache
Parts of the Boeing CH47 (Chinook)
Parts of the Sikorsky UH60 (Black Hawk)
Parts of the Pratt & Whitney PW 305 turbofan
Castings used in the Rolls-Royce Allison AE-3007 engine

3.4 Composite Materials

3.4.1 Introduction

Composites are materials made of at least two phases, which are physically distinct.
In general terms, any two or multiphase material can be classified as a composite,
including the precipitation-hardened aluminum and magnesium alloys, discussed in
Sects. 3.2 and 3.3, and other multiphase alloy systems that will be discussed next. A
particular group of composites will be considered in this section, belonging to the
broader family of structural composites. Structural composites are materials
designed to respond better to specific requirements regarding mechanical properties
and realized by mixing together two or more phases in order to achieve these
properties, which each single component would not be able to reach by itself (Hull
1981). This aspect together with the low-density values of the main composite
systems are the main reasons for the strong interest for structural composites in
aircraft applications. The main classes of composites used in aircraft structures are
based on epoxy polymer matrix, reinforced with carbon, glass, or aramid fibers.
Since the early use of composites in aircraft structures, dating back to the end of the
3.4 Composite Materials 111

1990s, their shares have continuously increased, starting, as usual, from military
fighter airplanes and spreading to any aircraft and spacecraft. Actually, the path was
not straight, and some diverting accident occurred. This is the case of the low
toughness and impact resistance exhibited by the carbon fiber fans and blades of
the early stages of the compressor of the RB211 Rolls-Royce engine, which did not
resist bird strikes and for this reason had to be temporarily abandoned (see fail safe
concept, Sect. 1.2). Moreover, the quest for a reliable assessment of the useful
lifetime (life safe criterion, Sect. 1.2), especially, of primary structures, like wings
and empennages, was not always met, particularly in case of early composites based
on carbon fibers, not displaying any fatigue limit. The situation evolved positively,
following two parallel paths: materials based on alternative reinforcements and
better grade and microstructural control on carbon fibers. One interesting example
of novel design component is the single-shell monocoque that provides higher
strength and lower total weight, as compared to the same part assembled by joining
together several parts. In addition to the improvement in the specific mechanical
properties obtained both from the adoption of new materials and of the new
structures that they allow to realize, the rapid spread of epoxy matrix structural
composites has been sustained by other important advantages. With reference to the
B787-Dreamliner (Sect. 1.2, Fig. 1.11), the use of composites requires also novel
processing and manufacturing technologies that result in important cost reduction,
thanks to the optimization of the material use. In fact, once the target properties
(strength, stiffness, toughness, etc.) have been set, the lay-up design indicates
precisely the necessary production steps and relevant materials quantities. The
composite parts produced in this way can easily be assembled using lay-up machin-
ery and rotational molding process. In general, composites exhibit a larger mechan-
ical stability as a consequence of their relatively low coefficient of thermal
expansion. Lighter structures obviously result in a reduction in fuel consumption
and an increase in the payload and/or passenger number. The latest generation of
composite materials, featuring high damage tolerance, has also improved the acci-
dent survivability of the components. Eventually, the electrically insulating character
of the polymer matrix structural composites eliminates all drawbacks pertaining
galvanic corrosion, typically initiated at the contact regions between dissimilar metal
alloys, particularly in marine environments, featuring high concentrations of water
vapor and salty aerosols. For the same reasons, the incidence of fatigue/corrosion
problems is virtually eliminated. These latter aspects, influencing materials and
structural durability, are of course very positive for the maintenance programs that
may potentially result in important reduction in the fleet management costs.
For the military field, other benefits can be considered, related not only, or not
necessarily, to the increase the payload, but also operational range, flight perfor-
mances, and maneuverability under extreme conditions. Polymer matrix structural
composites definitely have a number of interesting properties, and this can explain
their continuously increasing usage in aerospace structures. However, a specific
aspect descending from the electrically insulating character of these materials has
to be underlined. Whereas an aircraft with fuselage and skin made of aluminum
alloys is quite conductive so that it can dissipate the electrical currents produced by
112 3 Alloys for Aircraft Structures

lightning strikes, the same would not be possible for a polymer matrix composite,
unless some alternative solutions are implemented. In fact, all structural composites
used for the outer surfaces of an airplane, those directly exposed to possible lightning
strike, have in their lay-up a protection system, consisting of a ply of metallic mesh
or other kinds of conductive layers. These metallic components have essentially a
functional role, i.e., improving electrical conductivity, and their relative concentra-
tions are comparatively small. A class of composites that has been specifically
developed for aircraft structures and that features major fractions of metallic com-
ponents are the fiber-metal laminates (FMLs). These materials are composed of
stacks of metal sheets, made of aluminum or titanium alloys, and prepreg layers,
made of continuous felts of aramid, glass, or carbon fibers. The potential of FMLs in
aerospace structures appeared evident since the beginning, even considering the low
density of their component phases: metallic alloys, reinforcing fibers (see next), and
epoxy resin. The possibility to produce anisotropic parts through a suitable fiber
orientation and to optimize the stress distribution within the load-bearing section is
another positive feature of these important structural composites (Botelho et al.
2006; Vermeeren 2003).

3.4.2 Fiber-Metal Laminates: Development and Designation

The early developed FML was the aramid-fiber-reinforced Al-laminate, ARALL, for
which the Dutch University of Delft and the AKZO company filed two patents in
1981. One of the patents covered the general concept of the FML, based on the
combination of metal sheets and prepreg fiber felts. The other patent regarded the
so-called stretch forming (stretching) process. Although flying components of this
prototype FML were actually produced, any further development of ARALL was
interrupted at the time, due to an insufficient fatigue resistance, particularly under
compressive and tensile load conditions. A new class of laminates was developed
next, whose name was GLARE, i.e., glass-fiber-reinforced Al-laminate. Also for this
material, there is a patent by AKZO, now AKZO Nobel (1987). The general layout
of an FML is shown in Fig. 3.39.
Further developments of FML were composites reinforced by carbon fibers:
• CARALL – carbon fiber reinforced aluminum laminate.
• Ti-Gr (titanium-graphite composites).
In this latter, developed by the Boeing company, aluminum is replaced by
titanium. Other materials systems have been considered, but their actual use in
aircraft structures is still far to come. This is the case of the composites like
ALLIC (Al-Li reinforced by carbon fibers), VIRALL (Al-alloy reinforced by
Vinylon), BRALL (Al-alloy reinforced by bamboo), and RAFAL (Al-alloy
reinforced by ramie) (Laliberte et al. 2000).
For the identification of the different products, an international designation code
has been established, according to the scheme illustrated in Fig. 3.40.
3.4 Composite Materials 113

Fig. 3.39 General layout of a FML composite. On the right the cross-sectional view of a real FML.
Please note the different orientation of each layer of fibers (Vermeeren 2003, reproduced with
permission of the Publisher)

Fig. 3.40 Scheme for the FML international designation code, with an example

For a full qualification of FMLs, other features are important, in particular the
orientation of the fibers (Table 3.27).

3.4.3 Main Production Steps and Resulting Characteristics

The processing of FML components typically involves the following steps: surface
preparation, lay-up preparation, preparation of the lay-up for the curing treatment,
curing treatment, inspection, and post-processing operations (Botelho et al. 2006).
During the first step (surface preparation), the metallic surfaces are pre-treated in
order to improve the bonding efficiency with the prepreg layer. Different approaches
114 3 Alloys for Aircraft Structures

Table 3.27 Specifications of the GLARE composites (Wu and Yang 2005)

Typical Al layers characteristics Fiber layer characteristics

density Thickness Thickness
Type (g/cm3) Alloy Temper [mm] Orientation [mm]
GLARE 1 2.52 7475 T76 0.3–0.4 Unidirectional 0.25
GLARE 2 2.52 2024 T3 0.2–0.5 Unidirectional 0.25
GLARE 3 2.52 2024 T3 0.2–0.5 0 /90 cross-ply 0.25
(50%)
GLARE 4 2.45 2024 T3 0.2–0.5 0 /90 /0 cross- 0.375
ply (67%–33%)
GLARE 5 2.38 2024 T3 0.2–0.5 0 /90 /90 /0 0.50
cross-ply (50%)
GLARE 6 2.52 2024 T3 0.2–0.5 +45 /45 cross- 0.25
ply (50%)

have been developed in this regard, which can be grouped into the following main
categories: mechanical, chemical, and electrochemical (Sinmazçelik et al. 2011).
The effects of a chemical etching, carried out on an aluminum alloy (2024-T3)
foil, using a sulfuric acid solution, are visible in the SEM micrographs in Fig. 3.41.
The increase in the surface irregularity and, consequently, in the specific surface area
is an important parameter used to improve the adhesion between the aluminum foil
and the prepreg fiber felt.
After treating the metallic surfaces, the lay-up of the composite is assembled. At
this stage, the choice of the right orientation of the fiber layers is paramount to obtain
the expected properties.
Some indications have been developed over the years as concerns this aspect,
depending on the different loading conditions:
• 0 /0 to enhance fatigue resistance, ultimate tensile strength and yield strength;
• 0 /90 to obtain fatigue resistance, impact resistance;
• +45 /45 to obtain shear resistance.
Concerning the metallic alloy layers, they are integrated into the composite
starting from already formed foils. The alternative involves the direct deposition of
Al-foil directly from the melt using a method that will be fully discussed next. Once
the composite lay-up has been prepared, it is transferred to a vacuum bag for the
curing treatment (Fig. 3.42). The vacuum bag is typically made of nylon with rubber
seals. A release film, e.g., fluorinated ethylene propylene, wraps the composite
stacks and helps its detachment from the vacuum bag and other parts of the set up
at the end of the curing. A bleeder, that might be woven fabric or felts, adsorbs the
excess of resin, possibly ejected during the treatment (Gutowski 1997). The auto-
clave curing occurs at pressures and temperatures suitable for a full consolidation of
the composite. Temperature and pressure profiles suitable for ARALL (2024 alloy,
with a T3 temper) composites are shown by the graph in Fig. 3.42, as a function of
3.4 Composite Materials 115

Fig. 3.41 SEM micrograph showing the surface of a 2024 T3 alloy before (untreated) and after a
H2SO4 chemical etching used to improve the adhesion with the prepreg fibers. (Bothelo et al. 2009,
reproduced with permission of the Publisher)

Fig. 3.42 Vacuum-bag setup for the impregnation and curing treatment of the composite lay-up.
(Redrawn from Bothelo et al. 2006)

the time of curing, which is the most critical part of the production process. Thanks
to the removal of air via a vacuum system, this procedure allows to press the
multilayer and helps the spread of the resin and the elimination of excess residues
and volatile components. Limits and possible problems at this stage of the process
are essentially two: firstly, curing temperature can be comparable with the temper
treatment temperature of the selected aluminum alloy, being the use of aged and
stabilized alloys a possible remediation. Secondly, an optimal distribution of the
resin can be achieved only if dimensions of the workpiece are not too big.
To overcome these limitations and to render the whole production process less
expensive and applicable to relatively larger pieces too, the vacuum-assisted resin
transfer molding (VARTM) process has been developed (Jensen et al. 2009). In this
process, the resin is injected into the fiber preform and spreads through the metal and
fiber layers. This is a fundamental step of the process, since a great deal of the final
mechanical properties of the FML is dependent on the homogeneous spread of the
resin within the fiber felt and at the interfaces with the metal foils. During the
VARTM process, any excess resin is sucked away from the opposite side, with
respect to the inlet duct of the vacuum bag (Fig. 3.43).
116 3 Alloys for Aircraft Structures

Fig. 3.43 Scheme of the fabrication set up for the VARTM. (Redrawn from Jensen et al. 2010)

Fig. 3.44 Sequence of resin infiltration and curing for the VARTMFML. (Redrawn from Jensen
et al. 2009)

To promote the homogeneous distribution of the resin within the composite, holes
are drilled into the metal foils core, although this step requires a particular care, since
an important influence of the hole features on the mechanical properties of the
composite has been reported, with particular regard to fatigue life. Porous aluminum
foils have been considered as a possible alternative approach (Fig. 3.44). In the
process, named VARTM-infused plasma coated laminates (VARTMPCL), the fiber
fabric is coated with a porous metal layer, obtained by depositing the Al-alloy via
plasma-spraying (see Sect. 7.2.1). In Fig. 3.45 the main steps of the deposition
process are illustrated. The fabric, fixed on a rotating drum, is coated with the
metallic layer by repeated scans of the torch.
3.4 Composite Materials 117

Fig. 3.45 Deposition of porous Al-alloy layer onto the fabric to produce FML via the VARTMPCL
processing (Jensen et al. 2009, reproduced with permission of the Publisher)

Fig. 3.46 Comparison of the microstructure of the VARTMFLM (left) and VARTMPCL (right) –
0.38 mm thick foils (Jensen et al. 2009, reproduced with permission of the Publisher)

The thickness and the other coating features can be tuned by selecting the torch
deposition parameters and the rotating and translational speed of the drum. It is also
possible to coat the fabric on both sides by just turning it over and replacing it back
onto the drum and then going ahead with the deposition. The alternate layers of fiber
felts and porous alloy layers are subjected to the VARTM process in the same way as
the “hole-drilled” VARTMFML systems.
Figure 3.46 shows the microstructure of FML with the same stratigraphy and
composition but with a difference consisting in the VARTM technique used for their
production. Considering the rather different microstructures and, thereby, properties
of the two materials, they may have different applications: the VARTMFML
typically exhibits good mechanical properties, whereas the VARTMPCL products
are preferable for their functional properties, like enhancing electrical and thermal
conductivity, interesting, for instance, to improve the resistance of the composite to
lightning strikes. From the structural point of view, the VARTMPCL materials
suffer from the extreme irregularity of the aluminum layers, generally resulting in
118 3 Alloys for Aircraft Structures

an early crack initiation. Although the intrinsic damage tolerance of these materials
renders them capable to withstand also relatively high defect concentrations, still,
safety criteria may suggest more conservative choices.
For a complete exploitation of FMLs, several post-production treatments have
been developed to manufacture composite components with the right geometry
required by any specific application. Joining techniques are very important to
produce components with complex shapes and geometries. An alternative approach,
largely used even in the preparation of the GLARE fuselage panels of the A380, is
the stretch forming. The only precaution that has to be taken when stretch forming
FML components is the delamination that may occur when excessive bending and
straining are imparted to the component, favored by the elastic modulus mismatch
between the metal and prepreg layers. Notwithstanding the possibility of forming
FMLs using different approaches, joining still remains a fundamental aspect for
obtaining complex piece geometries. Adhesives are used to join different parts, but
in doing this, a continuity of the fiber felts should be preserved, in order to avoid an
excessive weakening of the final pieces. Riveting is also used, exploiting the good
deformability of the FML, and still sticking to the recalled precautions. The com-
plexity and critical dependence of the FML properties, namely, the mechanical
properties, which will be discussed in Sect. 3.4.4, on their actual microstructure,
renders all production steps extremely critical.

3.4.4 Mechanical Properties

With reference to the values in Table 3.28, the generally anisotropic character of the
mechanical properties of FMLs can be clearly seen, depending on the direction of the
applied load with respect to the composite symmetry. The tensile and compression
behavior of FMLs is influenced by the plasticity of the metallic component,
displaying a nonelastic trend, depending on the load application direction: either
longitudinal (L) or transverse (T). Although the metallic component behaves as a
monolith and its properties cannot be modulated and adapted that much, the orien-
tation of the fibers can be optimized, to avoid a too heterogeneous and aniso-
tropic material to be obtained and to maximize the performances along the most
demanding directions in terms of mechanical stress field.
The composite design requires a thorough analysis of its complex mechanical
behavior. For this purpose the Stress-Strain Calculation Program (SSCP) provides
excellent indications, at least in static loading conditions (Hagenbeek et al. 2003).
The key factor for a good fatigue resistance of FMLs can be identified in the
relatively thin layers building up the composite. This condition determines, within
each layer, a biaxial plane stress, to be regarded as a beneficial factor for a slow crack
propagation (Dieter 1988). In addition to this mechanism, crack bridging, i.e., the
transfer of the load from the cracked metallic foil to the still undamaged fibers,
3.4 Composite Materials 119

Table 3.28 Tensile and compressive mechanical properties of ARALL and GLARE materials
along longitudinal (L) and transverse (T) directions (Laliberte et al. 2000)
Material
Parameter Direction A-2 A-3 G-2 G-3 G-4 G-5
Ultimate tensile strength [MPa] L 717 821 1213 745 1027 683
LT 317 780 317 745 607 683
Yield strength [MPa] L 366 607 393 303 352 –
LT 228 331 228 303 25 –
Young’s modulus [GPa] L 66 66 66 58 57 70
LT 53 49 50 58 50 70
Ultimate strain [%] L 2.5 2.2 4.7 4.7 4.7 –
LT 12.7 8.6 10.8 4.7 4.7 –
Compressive yield strength [MPa] L 255 345 413 310 365 283
LT 234 360 236 310 285 283
Compressive elastic modulus [GPa] L 65 64 67 59 60 –
LT 53 50 52 59 54 –
A ARALL, G GLARE, L longitudinal, T transverse

contributes to enhance the fatigue resistance of the composite. An appropriate

adhesion strength between fibers and metal foil completes the picture and determines
the right balance among different mechanisms, all of them very much dependent on
the layered structure of the FML. This aspect, i.e., the central role of the layered
structure in determining the mechanical properties and in particular the fatigue
behavior of FMLs, is clearly demonstrated by the curves in Fig. 3.47. In the case
of FMLs, the crack propagation rate tends to decrease with increasing crack length,
whereas in a bulk Al-alloy, the opposite situation is usually observed.
This behavior, exhibited by the laminate composites, is most interesting for
aerospace applications, since the integrity of the component is preserved all through
its useful lifetime. This can be achieved thanks to the effective action of the
mentioned toughening mechanisms, capable to actively counteract the fatigue dam-
ages introduced into the material structure by service conditions. The damage
tolerance of FMLs is proved by the fatigue data in Fig. 3.48. The behavior of
ARALL is compared to that of bulk aluminum alloys. Interestingly, the number of
cycles for the appearance of first fatigue crack (Ni) in the case of the ARALL
material is lower or, at most, comparable with Ni of the bulk alloys. The maximum
number of cycles (Nf), i.e., the fatigue life, of the FML is order of magnitudes larger
than the fatigue life of all the reference Al-alloys (Lanciotti and Lazzeri 2009). The
damage tolerance of FMLs has been also confirmed with reference to notched
samples: the data for the FMLs are comparatively better than those for aluminum
alloys (Wu and Yang 2005). This situation is particularly important for materials to
be used for those parts more directly exposed to impacts, like the leading edges of the
wings, the engine entrance ducts, etc., in which notch-like defects can be produced
120 3 Alloys for Aircraft Structures

Fig. 3.47 Crack growth rate vs crack length for a bulk aluminum alloy (7075 – T6) and an ARALL
composite. The maximum applied loading spectrum twist was truncated at 1.3 times the Smf. Three
different compression residual stress levels in the Al sheets are shown: 95 MPa, 120 MPa, and
145 MPa. (Redrawn from Buhl 1992)

by the bird strikes, hailstorms, etc. Other kinds of defects can be rather ascribed to
structural features, like riveting, joint edges, and screws. Not always, the impact
damages are immediately visible, and radiographic investigations are recommended,
to prevent possible severe consequences, particularly as concerns the strength and
stiffness of the structure. In addition to mechanical damages, changes in the FML
properties may derive from environmental interactions, in the form of corrosion
phenomena and water absorption. Water may enter the composite as air humidity,
which may permeate the fibrous part of the composite.
Moreover, condensation on the cold metallic surfaces may result in liquid water
infiltration, although the presence of outer aluminum foils and epoxy-sealed layers
reduce significantly these phenomena with respect to the standard structural com-
posites. Water absorption may produce the delamination of the prepreg layers and
their detachment from the metallic foils with a consequent reduction in strength,
stiffness, and fatigue life (Botelho et al. 2007, 2009). Surface corrosion may be
3.4 Composite Materials 121

Fig. 3.48 Comparative fatigue data for bulk Al-alloys and an ARALL composite. For each graph
the relevant fatigue test conditions (a–c) are indicated. (Redrawn from Buhl 1992)
122 3 Alloys for Aircraft Structures

avoided by anodizing the aluminum foils and by protecting the surface with a primer
layer containing inhibitor substances (Wu and Yang 2005).

3.4.5 Applications

The initial focus, as concerns the application of FMLs was to use ARALL in the
wing structures, with particular regard for a panel with a tank inspection hatch of the
Fokker F-27 aircraft (Vermeeren 2003). This aircraft was selected for this study,
since at the time, i.e., the early 1980s, a complete background information was
available. The idea was discussed by the involved research parties, i.e., Fokker, Delft
University, and ALCOA. The outcome of the study was a full-size panel
(2.0  1.5 m2), which achieved the target of 25% weight reduction with respect to
the same part made in aluminum alloy. A series of products followed the early panel
in ARALL 1, using different alloys:
• ARALL 1: with 7075 alloy layers, stretched after curing;
• ARALL 2: with 2024 alloy layers, in the as-cured condition.
In the following years, other products were made available:
• ARALL 3: same as ARALL 1 but with 7475 alloy in the place of 7075 alloy
layers;
• ARALL 4: produced using a high-temperature adhesive, interesting for military
applications.
An important indication emerging from the qualification tests carried out on these
early products was the sensitivity of ARALL to the presence of drilled holes, that
reduced significantly the strength of the material. Moreover, notwithstanding general
excellent material properties as concerns fatigue crack growth, specific piece geom-
etries displayed premature fatigue cracks. This is the case of the areas close to
doublers bonded on the structure to increase its strength. However, these cracks
regarded the outer aluminum layer only, and the residual strength of the component
was still within the safety levels, thanks to the still integer fibers. Interestingly,
similar cracks would have certainly resulted in a premature failure of an aluminum
component, and in the end, the wing panel was very successful. In addition to wing
panels, ARALL provided an effective solution to the excessive weight problems of
the C-17 military transport aircraft that limited the loading capacity. For this reason,
ARALL was chosen for the cargo door (5.6  9.7 m2) with a relevant weight
reduction of 26%. However, the complex production steps and joining necessary
to obtain a complete door resulted too demanding, so that in the end, only 30 of these
doors were actually produced. Furthermore, the cost-effective, highly performing
aluminum alloys initially hindered the use of the FMLs in the fuselage. However,
when GLARE 3 and 4 were made available, most of the limitations, which were
there in ARALL materials and in GLARE 1 and 2, mainly because of the
unidirectionality of the fibers, were overcome. The cross-ply structure of the fibers
and the asymmetry in the properties, associated with the peculiar fiber distribution,
3.4 Composite Materials 123

Fig. 3.49 The position of GLARE panels in the fuselage of the A380. (Redrawn from Wu and
Yang 2005)

Fig. 3.50 The ECOS3™

container before and after a
blast test (Vlot and Gunnink
2001, reproduced with
permission of the Publisher)

could be exploited for the fabrication of large fuselage parts. In this fundamental
development for FMLs in aerospace applications, the studies and tests conducted by
Airbus for several years, in some cases on full size components, have been funda-
mental. Figure 3.49 gives the distribution of GLARE panels used for the fuselage of
the Airbus A380. This laminate composite is employed in association with other
structural composites, namely, graphite-fiber reinforced polymer, with a significant
overall weight reduction.
The combination of aluminum and glass fibers in GLARE induces in the com-
posite an extremely high fire resistance. In fact, although fire resistance of aluminum
is rather poor, when an aluminum alloy is coupled with the glass fiber felt, the
situation is completely different. A GLARE sample, exposed to a fire source, might
still exhibit the melting of the aluminum outer layer. However, the glass fibers felt
remains in place, thus preventing the heat to propagate toward the inner layers of the
composite, that in this way are preserved. Fire resistance, coupled with the excellent
impact resistance of FMLs in general and particularly of GLARE, has fostered the
use of these materials for cargo and luggage containers capable to resist possible
explosions. Figure 3.50 shows the effect of a blast test on the ECOS3™ container
made of GLARE 3.
As a final example of application of FMLs, these materials are also used as
patches for repairing cracks in polymer matrix composites. In this regard, the smaller
124 3 Alloys for Aircraft Structures

thermal mismatch between aluminum and GLARE, as compared to aluminum with

respect to polymer matrix composites, is exploited. An example of this application is
the use of a GLARE patch to repair a crack in the fuselage of a C-5A Galaxy
transport aircraft (Vermeeren 2003).
New systems are continuously investigated, as concerns both the metallic part
and the prepreg felt, in order to improve the already excellent properties achieved so
far. Processing is also paramount to make these materials really attractive also for
large-scale, low-cost productions. One of the latest aspects that is under consider-
ation is the use of thermoplastics rather than thermosetting polymers as binder
materials. In the future, renovated and growing interest can be foreseen for FMLs
featuring natural reinforcing fibers, like the mentioned BRALL and RAFAL with
bamboo and ramie natural reinforcement, respectively.

References

Aleris Corporation (2015) Aerospace Aluminum A5028 AlMgSc – The Strong Lightweight.
Available via DIALOG. https://www.aleris.com/wp-content/uploads/2016/02/AL-2342_012-
Aktualisierung-BR-AlMgSc-2015-06-03-WEB.pdf. Accessed 04 January 2018.
APWORKS (2017) Material Data Sheet – Scalmalloy®. Available via DIALOG. http://www.
apworks.de/en/scalmalloy/. Accessed 04 January 2018
APWORKS (2017) Scalmalloy ®. Available via DIALOG. http://www.apworks.de/en/scalmalloy/.
Accessed 04 January 2018
ASM International (1991) ASM Handbook Vol. 2 – Properties and selection: Nonferrous Alloys
and Special-Purpose Materials. ASM International, Materials Park, Ohio
ASM International (1992) ASM Handbook Vol. 3 – Alloy phase diagrams. ASM International,
Materials Park, Ohio
ASM International (2003) ASM Handbook Vol. 13A – Corrosion: Fundamentals, Testing and
Protection. ASM International, Materials Park, Ohio
ASM International (2015) ASM Subject Guide – Aluminum and Aluminum Alloys. ASM Inter-
national, Materials Park, Ohio
Atrens A et al (2015) Review of Recent Developments in the Field of Magnesium Corrosion.
Advanced Engineering Materials 17(4): 400–453
Belov et al (2002) Iron in Aluminium Alloys: Impurity and Alloying Element. CRC Press
Benedyk J C (2010) International Temper Designation Systems for Wrought Aluminum Alloys:
Part II – Thermally Treated (T Temper) Aluminum Alloys. Light Metal Age: 16–22
Blawert C et al (2006) Anodizing Treatments for Magnesium Alloys and Their Effect on Corrosion
Resistance in Various Environments. Advanced Engineering Materials 8 (6): 511–533
Boeing (2008) AERO Magazine 06. Available via DIALOG. http://www.boeing.com/commercial/
aeromagazine/articles/qtr_4_06/index.html. Accessed 8 October 2017.
Botelho E C et al (2006) A Review on the Development and Properties of Continuous fiber/epoxy/
aluminum Hybrid Composites for Aircraft Structures. Materials Research 9 (3): 247–256
Botelho E C et al (2007) Influence of Hygrothermal Conditioning on the Elastic Properties of Carall
Laminates. Applied Composite Materials 14: 209–222
Botelho E C, Da Silva D A, Rexende M C (2009) Hygrothermal Aging Effect on Fatigue Behavior
of GLARE. Journal of Reinforced Plastics and Composite Materials 28: 2487–2499
Buhl H (1992) Advanced Aerospace Materials. Springer Verlag, Berlin
Cahn R et al (2005) Structure and Properties of Nonferrous Alloys. Materials Science and
Technology 8–9. Wiley & sons
References 125

Cardarelli F (2008) Materials Handbook 2nd edn. Springer Verlag, London

Carter C, Norton M (2007) Ceramic Materials – Science and Engineering. Springer
Czerwinski F (2011) Magnesium Alloys – Design, Processing and Properties. InTech
Davis J R (2001) Aluminum and Aluminum Alloys. ASM International
Davy H (1808) Electro-Chemical Researches, on the Decomposition of the Earths: With Observa-
tions on the Metals Obtained from the Alkaline Earths, and on the Amalgam Procured from
Ammonia. Philosophical Transactions of the Royal Society of London 98: 333–370.
Dieter G E (1988) Mechanical Metallurgy. McGraw Hill
Djukanovic G (2017) Are Aluminium-Scandium Alloys the Future? http://aluminiuminsider.com/
aluminium-scandium-alloys-future/. Accessed 1 January 2018
Donachie M J (2000) Titanium – A Technical Guide. ASM International, Materials Park, Ohio
Dziubinska A, Gontarz A (2015) A new method for producing magnesium alloy twin-rib aircraft
brackets. Aircraft Engineering and Aerospace Technology: An International Journal 87(2):
180–188
Dziubinska A, Gontarz A (2016) A new technology for producing AZ31 magnesium alloy aircraft
brackets with a triangular outline. Aircraft Engineering and Aerospace Technology: An Inter-
national Journal 88(3): 452–457
Esmaily M et al (2017) Fundamentals and advances in magnesium alloy corrosion. Progress in
Materials Science 89: 92–193
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook-Airframe
Volume 1. Federal Aviation Administration
Filatov Y A et al (2000) New Al-Mg-Sc alloys. Materials Science and Engineering A 280: 97–101
Gasik M (2013) Handbook of Ferroalloys: Theory and Technology. Butterworth Heinemann
Gupta M, Gupta N (2017) The Promise of Magnesium Based Materials in Aerospace Sector.
International Journal of Aeronautical Science & Aerospace Research 4 (1):141–149
Gupta M, Gupta N (2017) Utilizing Magnesium based Materials to Reduce Green House Gas
Emissions in Aerospace Sector. Aeronautics and Aerospace Open Access Journal 1 (1): 1–6
Gutowski T G (1997) Advanced Composites Manufacturing 1st edn. John Wiley & Sons,
New York
Hagenbeek M, Van Hengel C, Bosker O J, Vermeeren C A J R (2003) Static Properties of Fiber
Metal Laminates. Applied Composite Materials 10: 207–222
Heinz A et al. (2000) Recent developments in aluminum alloys for aerospace applications. Mat Sci
Eng A 280: 102–107
Horst F, Mordike B (2006) Magnesium Technology-Metallurgy, Design Data, Applications.
Springer Verlag Berlin-Heidelberg
Hosford W F (2010) Mechanical Behavior of Materials. Cambridge University Press
Hull D (1981) An Introduction to Composites Materials. Cambridge University Press
Hull D, Bacon J (2011) Introduction to Dislocations. Elsevier
Jensen B J et al (2009) Fiber Metal Laminates made by the VARTM process. In: 17th International
Conference on Composite Materials, British Composite Society, Edinburgh, 27–31 July 2009
Kaufmann G (2000) Introduction to Aluminum Alloys and Tempers. ASM International
Kettner M et al (2007) The InnMag Project – Processing Mg for Civil Aircraft Application.
Advanced Engineering Materials 9(9): 813–819
Laliberte J F, Poon C, Straznicky P V(2000) Applications of fiber-metal laminates. Polymer
Composites 21: 558–567
Lanciotti A, Lazzeri L (2009) Fatigue resistance and residual strength of riveted joints in FML.
Fatigue & Fracture of Engineering Materials & Structures 32: 837–846
Lequeu P, Lassince P, Warner T (2007) Aluminum alloy development for the Airbus A380-Part 1 &
2. Adv Materials & Processes July: 33–35 & 41–44
Lequeu P, Smith K P, Daniélou A (2010) Aluminium-Copper-Lithium Alloy 20150 Developed for
Medium to Thick Plate. Journal of Materials Engineering and Performance 19: 841–847
Lumely R (2011) Fundamentals of Aluminium Metallurgy. Woodhead Publishing Ltd
126 3 Alloys for Aircraft Structures

Lyon P, Syed I, Heaney S (2007) Elektron 21 – An Aerospace Magnesium Alloy for Sand Cast and
Investment Cast Applications. Advanced Engineering Materials 9(9): 793–798
Magnesium Elektron (2012) Magnesium Alloy Welding Rod Datasheet. Available via DIALOG.
https://www.magnesium-elektron.com/wp-content/uploads/2016/10/Magnesium-Alloy-Weld-
Rod_0.pdf. Accessed 31 December 2017.
Magnesium Elektron (2014) Elektron ® 675 Datasheet. Available via DIALOG. https://www.
magnesium-elektron.com/wp-content/uploads/2016/10/Elektron-675_0.pdf. Accessed
28 November 2017.
Magnesium Elektron (2014) Elektron ® ZREI Datasheet. Available via DIALOG. https://www.
magnesium-elektron.com/wp-content/uploads/2016/10/Elektron-ZRE1_0.pdf. Accessed
31 December 2017)
Magnesium Elektron (2015) Elektron ® QE22 Datasheet. Available via DIALOG. https://www.
magnesium-elektron.com/wp-content/uploads/2016/10/Elektron-QE22_2.pdf. Accessed
28 November 2017.
Magnesium Elektron (2016) Elektron ® 21 Datasheet. Available via DIALOG. https://www.
magnesium-elektron.com/wp-content/uploads/2016/10/Elektron-21_1.pdf. Accessed
28 November 2017.
Marker T R (2013) Evaluating the flammability of various magnesium alloys during laboratory and
full-scale aircraft fire tests. US Department of Transportation
McCafferty E (2010) Introduction to Corrosion Science. Springer Verlag, New York
Meyer G (2001) Die (ungleichen) Dydimium-Zwillinge. Chemie in unserer Zeit 35(2):116–123
Nature Materials (2016) No Easy Solutions for Aerospace (editorial). Nature Materials 15 (8): 803
Ostrovsky I, Henn Y (2007) Present state and future of magnesium application in aerospace
industry. In: International Conference “New Challenges in Aeronautics” ASTEC’2007, Mos-
cow, 19–22 August 2007
Parker R L (1967) Data of Geochemistry – Composition of the Earth’s crust. US Government
Printing Office
Pekguleryuz M O, Celikin M (2010) Creep resistance in magnesium alloys. International Materials
Science Reviews 55: 197–217
Pepperhoff W, Acet M (2001) Constitution and Magnetism of Iron and its Alloys. Springer, Berlin
Heidelberg
Perkguleryuz M O, Kainer K U, Arslan Kaya A (2013) Fundamentals of Magnesium Alloy
Metallurgy. Woodhead Publishing
Polmear I J (2006) Light Alloys – From Traditional Alloys to Nanocrystals, 4th end. Butterworth-
Heinemann
Polmear I J et al. (2017) Light Alloys – Metallurgy of the Light Metals. Butterworth-Heinemann
Porter D, Easterling K E, Sherif M Y (2009) Phase Transformations in Metals and Alloys, 3rd end.
CRC Press
Prasad E N, Gokhale R, Wanhill R (2014) Aluminum – Lithium Alloys. Elsevier
Prasad E N, Wanhill R (2017) Aerospace Materials and Material Technologies. Springer
Rendigs K H (1994) Metallic structures used in aerospace during 25 years and prospects. In:
50 years of Advanced Materials or Back to the Future, Proceedings of the 15th International
European Chapter Conference of the Society for the Advance of Material and Process Engi-
neering, Toulouse, 1994
Röyset J, Ryum N (2005) Scandium in Aluminium Alloys. International Materials Reviews 50 (1):
19–44
Scandium International Mining Corp (2017) The Aluminum-Scandium Alloy Advantage. Avail-
able via DIALOG. http://www.scandiummining.com/i/pdf/Scandium-Alloy-Fact-Sheet.pdf.
Accessed 01 January 2018
Sinmazçelik T et al (2011) A review: Fibre metal laminates, background, bonding types and applied
test methods. Materials and Design 32: 3671–3685
Staley J T, Lege D J (1993) Advances in aluminum alloy products for structural applications in
transportation. Journal of Physics IV, Colloque C7, 3: 179–190
References 127

Starke Jr E A, Sanders Jr T H, Palmer I G (1981) New Approaches to Alloy Development in the

Al-Li System. Journal of Materials Aug 1981: 24–33
Starke Jr E A, Staley J T (1996) Application of modern aluminum alloys to aircraft. Progr
Aerospace Sci 32: 131–172
The Aluminum Association (2015) International Alloy Designation and Chemical Composition
Limits for Wrought Aluminum and Wrought Aluminum Alloys. The Aluminum Association
Inc, Arlington
The CM Group (2012) The Global Mg Industry in 2011 – The Impact of Chinese Production, Costs
and Shipments. IMA Annual Conference, San Francisco, 21 May 2012
Thompson G E et al (1999) Anodizing of Aluminium Alloys. Aircraft Engineering and Aerospace
Technology 71: 228–238
Totten E G, MacKenzie S, Scott D (2003) Handbook of Aluminum Volume 1 – Physical Metallurgy
and Processes. Marcel Dekker Inc.
Vermeeren C A J R (2003) An Historic Overview of the Development of Fiber Metal Laminates.
Applied Composite Materials 10: 189–205
Vlot A, Gunnink J W (2001) Fibre Metal Laminates. Springer–Science+Business Media
Williams J C, Starke Jr E A (2003) Progress in structural materials for aerospace systems. Acta
Materialia 51: 5775–5799
Wu G, Yang J M (2005) The Mechanical Behavior of GLARE Laminates for Aircraft Structures.
Journal of Materials 57(1): 72–79
Wu R et al (2015) Recent progress in magnesium-lithium alloys. International Materials Review 60
(2):65–100
Wulandari W et al (2010) Magnesium: current and alternative production routes. Chemeca:
Australasian Conference on Chemical Engineering. Available via DIALOG. http://ro.uow.
edu.au/engpapers/1254. Accessed 07 Nov 2017
Xu W et al (2015) A high-specific-strength and corrosion-resistant magnesium alloy. Nature
Materials 14: 1229–1237
Zhang J et al. (2013) Experimental study on strengthening of Mg-Li alloy by introducing long-
period stacking ordered structure. Scripta Materialia 68: 675–678
Zhu Y M, Morton A J, Nie J F (2010) The 18R and 14H long-period stacking ordered structures in
Mg-Y-Zn alloys. Acta Materialia 58(8): 2936–2947

Further Reading

ASM International (1999) ASM Specialty Handbook: Magnesium and Magnesium Alloys. ASM
International, Materials Park, Ohio
ASM International (2008) ASM Handbook Vol. 15 – Casting. ASM International, Materials Park
Ohio
Baker et al (2004) Composite Materials for Aircraft Structures 2nd edn. American Institute of
Aeronautics and Astronautics Inc.
Davis J R (2001) ASM Specialty Handbook: Aluminum and Aluminum Alloys. ASM International,
p 351–416
Rana S, Fangueiro R (2016) Advanced Composite Materials for Aerospace Engineering:
Processing, Properties and Applications. Woodhead Publishing Limited
Zhang S, Zhao D (2012) Aerospace Materials Handbook. CRC Press
Chapter 4
Titanium and Titanium Alloys

4.1 Introduction

Titanium is a chemical element with atomic number 22, occupying the fourth group,
fourth period of the periodic table of elements. It is classified as a transition metal,
with a density of 4.51 g/cm3 and a Young’s modulus of 120 GPa. Titanium exhibits
two polymorphs (see Sect. 4.2): α titanium (hcp structure), stable up to 882
C, a
temperature named β-transus. The other polymorph is indeed β-titanium (bcc struc-
ture), stable from 882
C up to the melting point (1670
C). The widespread use of
titanium and its alloys can be ascribed to a number of interesting properties of these
materials, like:
• Exceptional corrosion resistance, in many cases better than in stainless steels,
owing to the spontaneous formation of a stable, adherent, and protective TiO2-
based surface layer.
• High specific strength. The tensile strength of some alloys is comparable to that of
low-strength martensitic and austenitic stainless steels, like the A286 material
(PHSS, see Sect. 5.4.3.3) and Co-based superalloys, such as the L-605 alloy (see
Sect. 6.2).
• Low density, as compared to the densities of Ni-based superalloys and steels.
• Processing capabilities. Titanium can be forged, hot-rolled, and welded but also
machined and cast. For this latter process, the best choice is investment casting
(see Sect. 4.4). Finally, titanium can be processed by powder metallurgy
techniques.
Aerospace applications played a major role in the development of titanium alloys,
and the driving force for the massive production of titanium came directly from the
highly demanding and rapidly growing aerospace industry in the late 1950s
(Donachie 2000). Of course, the great variety of properties render titanium alloys
an attractive choice for other fields too, like architectural, automotive, and biomed-
ical. Titanium accounts for about the 0.6% of the Earth’s crust, and it is the fourth

© Springer Nature Switzerland AG 2020 129

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_4
130 4 Titanium and Titanium Alloys

Fig. 4.1 The distribution of the global titanium ores deposits. Quantities are expressed in billions
of metric tons. (Data from Leyens and Peters 2003)

most abundant structural metal, following iron, aluminum, and magnesium. Like
other most important metals used for technological purposes, it is present on Earth in
form of oxides. The most important minerals containing titanium are TiO2 (rutile),
TiO2 (brookite), TiO2 (anatase), FeTiO3 (ilmenite), CaTiO3 (perovskite), MgTiO3
(geikielite), and CaSiTiO5 (sphene). Of all these raw materials, the three polymorphs
of titanium oxide are those exhibiting the higher concentration of titanium and
therefore to be preferred in the extraction metallurgy process (Froes and Imam
2018). Figure 4.1 shows the geographical distribution of the main provenance
regions for titanium ores.
The reactivity of titanium with oxygen and other common gaseous species, like
hydrogen and nitrogen, and the stability of the resulting compounds are the main
reasons why this metal started to be used only in relatively recent times. In 1791, the
English clergyman and amateur mineralogist, William Gregor, proposed the exis-
tence of a new unknown element contained in a dark magnetic iron ore (ilmenite).
Some years later, in 1795 the German chemist Martin H. Klaproth analyzed rutile
from a Hungarian site and pointed out the presence of an unknown element, the same
as the one reported by Gregor. The new element was then named titanium after the
4.1 Introduction 131

Fig. 4.2 Schematic of the Kroll’s reactor plant for the extraction of primary titanium (Lüterling and
Williams 2007)

Titans, the members of the second generation of divine creatures, according to the
Greek mythology. Matthew A. Hunter was the first able to extract pure metallic
titanium in 1910, starting from the thermal decomposition of TiCl4 (titanium tetra-
chloride) induced by sodium in a steel reactor. This process was named the Hunter
process after its developer, although William J. Kroll is unanimously recognized as
the father of titanium extractive industry. In fact, in 1932 in Luxemburg he produced
significant amounts of titanium, starting from TiCl4 combined with calcium, using
what can be regarded as the prototype of the modern pyrometallurgical Kroll’s
process. Afterward, the process was modified, by replacing the reducing agent:
magnesium instead of calcium, leading to the modern Kroll’s process.
The primary output of the Kroll’s process is the so-called titanium sponge. The
name derives from the porous and spongy appearance of titanium obtained from the
purification process that removes magnesium chloride residuals through evaporation.
The density of titanium sponge ranges from 1.2 to 3.0 g/cm3. Therefore, before any
usage, it is densified in a hydraulic press. Once densified, pieces of titanium sponge are
assembled to form electrodes for the melting processes. Generally, multiple remelting
steps are necessary, for the morphology of the sponge, for adjusting the chemical
composition of the target alloy, and for eliminating, at the same time, any possible
remaining contaminants (Lüterling and Williams 2007; Leyens and Peters 2003).
The Kroll’s process consists of six main stages that can be described as follows,
with reference to the fluidized bed reactor plant in Fig. 4.2, by far the most used for
the extraction of primary titanium.
132 4 Titanium and Titanium Alloys

1. Chlorination. A mix of raw materials (titanium ores) and coke undergo chlorina-
tion in the reactor chamber to produce titanium tetrachloride (TiCl4). The process
is carried out at temperatures in between 800
C and 1000
C, reached thanks to
the resistor bed supporting the raw material and carbon mix. The formation of
TiCl4 occurs according to the following reaction:

2TiO2 ðsÞ þ 3CðsÞ þ 4Cl2 ðgÞ ! 2TiCl4 ðgÞ þ 2COðgÞ þ CO2 ðgÞ ð4:1Þ

Carbon particles, residues of the ores and quartz fragments, and fly ashes are
the main impurities typically present in the gas flux leaving the reactor chamber.
Since their presence in the final product (sponge) would be extremely detrimental,
they are eliminated using a cyclone dust collector.
2. Distillation. The tetrachloride flux is conveyed to a fractional distillation unit for
its complete purification, since the starting grade, as deriving from the chlorinator
directly, would not guarantee a sufficient purity level of the final product. The
process is carried out in two steps. The first is meant to remove the impurities with
low boiling points, e.g., CO and CO2. The second step removes the high boiling
point impurities, such as SiCl4 and SnCl4.
3. Condensation of TiCl4. At the end of the distillation, the purified TiCl4 is
collected as crude product and stored under an inert gas atmosphere.
4. Reduction of TiCl4. This is the actual Kroll’s reaction. Titanium tetrachloride is
reduced by means of molten magnesium. The reduction using magnesium is
performed according to the reaction:

TiCl4 þ 2Mg ! Ti þ 2MgCl2 ð4:2Þ

The reaction temperature is kept in between 712 and 920
C. Magnesium is

used in excess with respect to the reaction stoichiometry in order to ensure the
complete elimination of chlorine from the products. Incidentally, the Hunter
process, mentioned at the beginning of this section, differs from the Kroll’s
approach for the use of sodium as reducing agent. The sodium-based process
was not only more expensive but involved also the risk of a higher flammability
due to the presence of residues of this element, sodium, trapped into the final
titanium sponge. As concerns the reducing reaction, its efficiency is limited by the
following competitive reactions:

3TiCl4 þ Ti ! Ti þ 4TiCl3
ð4:3Þ
2TiCl4 þ Ti ! 3TiCl2

5. Extraction of titanium from the sponge cake. The process aims at eliminating
MgCl2 (see Eq. 4.2) residuals. This can be achieved by acidic washing of the
crushed sponge and exploiting the preferential solubility of the MgCl2 in
HCl-based solutions. This process requires the extraction of the sponge cake
from the reactor. An alternative approach, which is carried out in situ, i.e., without
removing the titanium sponge from the Kroll’s reactor, is a preferable alternative.
4.1 Introduction 133

Fig. 4.3 Titanium sponge global production expressed in metric tons. (Data from Prasad and
Wanhill 2017)

The evaporation of MgCl2 is induced by the high vapor pressure of this chloride
that evaporates from the sponge and, after recondensation, can be removed from
the reactor. A vacuum condensation process can alternatively be used: the sponge
is heated under vacuum without being removed from the Kroll’s vessel. The
vapor flux that is emitted from the sponge condensates in another vessel, from
which it is recovered.
6. Crushing and sizing. This final stage is required for obtaining pieces of the sponge
to be welded to form electrodes for the remelting processes (see Sect. 4.4).
The aerospace interest for titanium started after the World War II, between the
late 1940s and the early 1950s. In that period, titanium and its alloys were rapidly
recognized as key materials for aircraft engine parts. In the USA, the government-
sponsored programs for the construction of large production plants of titanium
sponge, such as those at TIMET (1951) and RMI (1958). The industrial production
in Europe started in 1951 in the UK at the Metals Division of Imperial Chemical
Industries that later became IMI and Deeside Titanium (see Fig. 4.3). The Soviet
Union started the production of Ti sponge in 1954. The dramatic increase in the
production of titanium led Soviet Union to become the world’s largest producer in
1979. Innovation and alloy development in the USA rapidly progressed thanks to the
attempts of alloying titanium with other elements such as Al, Mo, and Sn (see Sect.
4.3.1). Contemporarily, there was also an increase in the production of commercially
pure (CP) titanium for non-aerospace applications, with a similar trend also in Japan
(Leyens and Peters 2003; Lüterling and Williams 2007).
134 4 Titanium and Titanium Alloys

 Condensers
 Cooling systems
 Agitators  Airframes  Evaporators
 Heat exchangers  Engines  Heat exchangers
 Piping and tubing  Fittings  Nuclear waste
 Storage tanks  Rockets disposal
 Valves  Satellites  Piping and tubing

Chemical processing Aerospace Power generation

Construction Titanium Miscellaneous

 Anodes for cathodic  Armor

protection  Bicycle frames
 Door plates  Camera shutters
 Facing and roofing  Eyeglass frames
 Fountain bases  Helmets
 Monument Biomedical engineering Marine engineering Automotive industry  Jewelry
refurbishment  Bone plates  Cathodic  Camshafts  Knives
 Dentures protection anodes  Chassis  Ornaments
 Heart valves  Deep-sea pressure  Connecting rods  Pens
 High-speed hulls  Crankshafts  Scissors
centrifugal  Desulfurizers  Gears  Tennis rackets
separators for blood  Flanges  Suspension  Watches
 Hip-joints  Propellers assemblies
 Insulin pumps  Regenerators  Valves
 Pacemaker housings  Structural
components
 Submarine ball
valves
 Vapor heaters

Fig. 4.4 Typical markets and main applications for titanium and titanium alloys (ASM Interna-
tional 1991; Donachie 2000)

The combination of mechanical properties and surface stability renders the use of
titanium alloys interesting and ubiquitous in aircraft structures, aero-engines, and
other aerospace applications. These have been for quite some time the main refer-
ence market for titanium alloy products, until recent, when titanium and titanium
alloys are being massively used also in other fields, like power industry, civil
engineering, and general consumer goods and commodities. The continuously
growing trend of the primary titanium production (Fig. 4.3) is coherent with its
expansion to new market shares (Fig. 4.4). The percentage of titanium used in other
applications than aircraft structures and engines is increasing, and this trend seems to
be growing further (see Fig. 4.5). At the same time, the link between the primary
titanium production (sponge production) and the aerospace market has become more
tenuous. Furthermore, the lower demand for titanium due to the reduction of the
military budgets, both in the USA and in the Countries of the former Soviet Union,
has led to an estimated 25% decrease in the world production capacity over the
1990–1995 interval. After this drop, the worldwide titanium sponge production
resumed, mainly due to the rise in commercial airplanes sales and the increasing
demand from other new markets (Lüterling and Williams 2007). In the early 1990s, a
series of low-cost titanium alloys was introduced into the market. One example of
these alloys is the Ti-6Al-1.7Fe-0.1Si (TIMETAL 62S), with comparable properties
as those of Ti-6Al-4V. The lower cost of these alloys, mainly based on recycled
titanium scraps, was attained also thanks to changes in the production and subse-
quent fabrication steps. The substitution of some alloying elements contributed to a
further cost reduction. Notwithstanding the success of these low-cost alloys, the
4.1 Introduction 135

Fig. 4.5 Industrial titanium demand forecast expressed in metric tons. Aerospace applications are
intended to belong to secondary markets. Data do not include shipment within China and Russia.
(Redrawn from Cain 2016)

Fig. 4.6 Commercial jet engine deliveries and relevant forecast for the main Producers. (Redrawn
from Leach 2016)

workhorse Ti-6Al-4V still remains the most used titanium alloy, with particular
regard for aerospace applications (Donachie 2000).
As discussed in Chap. 1 (Fig. 1.11), a state-of-the-art commercial airliner, like the
Boeing B787, has a significant number of components made of titanium and titanium
alloys, more than in the past, as indirectly indicated by with the trend shown
in Fig. 4.6, referring to the deliveries of commercial jet engines and forecast of the
main Producers, from 2010 onward.
As in other fields, also titanium industry has experienced a consolidation of
material specifications, with benefits for both producers and customers.
136 4 Titanium and Titanium Alloys

Fig. 4.7 Crystallographic structures of titanium polymorphs with the indication of slip planes (see
also Table 4.1) and lattice parameters. The α phase has hexagonal close packed (hcp) structure,
while the β phase is body centered cubic (bcc). (Redrawn from Leyens and Peters 2003)

This has contributed to widen the field of applications of titanium alloys, which
can also be exploited in the production of innovative materials, such as titanium
matrix composites and advanced alloys based on intermetallic phases, like gamma
titanium aluminides (see Sect. 4.3.2).

4.2 Physical Metallurgy Aspects

The allotropic transformation, α-hcp to β-bcc (see Fig. 4.7) occurring at 882
C in
pure titanium, is also present in titanium alloys, with variations in temperature
depending on their actual composition. Specific alloying elements can promote the
stabilization of one polymorph preferentially, rather than the other. Moreover,
through the balanced combination of stabilizing elements of the two allotropic
phases, it is possible to have both of them in the same alloy in different concentra-
tions. These metallurgical aspects will be recalled herewith, considering the impor-
tance they have in determining the properties of titanium alloys.
Some of the main features of α-Ti and β-Ti are as follows:
• α phase. This hcp (hexagonal close packed) phase, stable up to the β-transus
temperature, is characterized by good mechanical properties with some limita-
tions as concerns formability and toughness, deriving from the limited number of
slip systems, just three in the hcp structure (see Table 4.1). The c/a ratio between
the main axes of the crystallographic cell (see Fig. 4.7) is 1.587 for the α-Ti
polymorph. This value, which is slightly lower than the ideal value for a hexag-
onal cell (1.633), can be varied when interstitially dissolved atoms are present
 4.2 Physical Metallurgy Aspects

Table 4.1 Some characteristics of the crystallographic structures of the two polymorphs of pure titanium (Leyens and Peters 2003)
Slip planes Slip directions
Slip systems per unit Atomic density in slip
Polymorph N CN P Indices Number Indices Number cell Lattice parameters planes
α 6 12 74% {0001} 1

<1120> 3 3 a ¼ 0.295 nm 91%
c ¼ 0.468 nm
β 2 8 68% {110} 6 <110> 2 12 a ¼ 0.332 nm 83%
N number of atoms per unit cell, CN coordination number, P packing density
137
138 4 Titanium and Titanium Alloys

(e.g., hydrogen, carbon, nitrogen, and oxygen) or when substitutional atoms with
smaller than titanium radii are incorporated into the host lattice. An important
example of this latter situation is provided by alloying with aluminum. The
reduced c/a ratio induces a larger spacing between prism planes, and an increase
in their packing density, so that they become possible slip planes, alternative to
the basal planes. With reference to Fig. 4.7, prism planes are the (10 1
0) type

and the basal plane is (0001). According to
planes; the pyramidal plane is (1011),
the von Mises criterion, at least five independent slip systems are required for a
homogeneous plastic deformation. This explains the limited deformability of
polycrystalline α-Ti and why its limited ductility may result in the formation of
mechanical twinning, as an additional plastic deformation mode together with the
activation of secondary slip systems.
• β phase. This bcc polymorph is characterized by 12 slip systems, and this makes
plastic deformation easier. However, the no-close packed character of the bcc
lattice limits other properties, creep resistance in first place. Other interesting
features of the β polymorph are given in Table 4.1.
The transition from the β phase to the α phase upon cooling can occur according
to two mechanisms, depending on the cooling rate and the alloy composition:
martensitic transformation or diffusion-controlled nucleation and growth.
A cooperative movement of atoms in a shear-type process accompanies the
martensitic transformation. The hexagonal martensite of titanium is usually named
α0 and may exhibit two microstructures, featuring either massive or acicular grains.
Massive martensite is also named lath or packed martensite. This type of martensite
occurs in pure titanium, in low-alloyed materials, and in alloys in which the
martensitic transformation occurs at relatively high temperatures. Massive martens-
ite would usually feature a coarse-grained microstructure, with an average grain
size in the 50–100 μm range. Within the coarse grains, small packets of parallel
α plates or lathes with a thickness of 0.5–1.0 μm are present. Acicular martensite
is characterized by an intimate mixture of α plates that generally contain a
high density of dislocations. This is the so-called basketweave microstructure. The
α plates may also contain twins. In titanium alloys, the α0 martensite is supersatu-
rated in β-stabilizing elements. When α0 martensite is annealed in the α + β field
(see Figs. 4.8 and 4.17), it decomposes into α + β mixture, either by the
precipitation of β phase layers at the plate boundaries, or by precipitating β grains
in regions with higher dislocation density. An increasing concentration of solute
atoms hosted in martensite would tend to distort the lattice upon cooling, so that
the hexagonal symmetry is lost and an orthorhombic martensite would form instead.
The designation for this latter structure is α00 . In Ti-X binary systems, the α0 /α00
boundary concentrations are the following (in brackets, wt%): Mo (4), Nb (10.5),
Ta (26.5), V (9.4), W (8). For the sake of completeness, another quenching
structure should be mentioned as concerns β-Ti-alloys (see Sect. 4.3.1.4): the ω
phase. The formation of the ω phase occurs for high solute concentrations, which
inhibit the standard martensitic transformation. Upon quenching from the β field,
the ω phase precipitates athermally, and it is characterized by a uniform dispersion
4.2 Physical Metallurgy Aspects 139

Fig. 4.8 Schematic of the influence of different alloying elements on the Ti binary systems.
(Redrawn from Leyens and Peters 2003)

of extremely fine particles. The ω phase shows a trigonal symmetry in heavily

β-stabilized alloys and a non-close-packed hexagonal symmetry in low-alloyed
compositions, being the transition from hexagonal to trigonal continuous, as a
function of the content of the alloying elements.
Alternatively, diffusion-controlled nucleation and growth of the α phase occurs
when the titanium alloy is cooled from the β phase field to the α + β field (see Figs. 4.8
and 4.17 in Sect. 4.3.1) at a moderate rate. The nucleation of the α phase occurs
at the β grain boundaries preferentially, evolving in a continuous layer of α phase.
α grains may nucleate inside the β grains in form of parallel platelets, that continue
their growth until they get in contact with other α colonies. This process is often
defined as sympathetic nucleation and growth. α plates within the α colonies are
separated by retained β, remnant of the high-temperature phase. The resulting
microstructure features alternating α and β lamellae. An example of this structure
can be obtained by slow cooling the Ti-6Al-4V alloy from the β field. The size
of the α colonies, as well as their thickness, is reduced by increasing the cooling
rates.
The new α plates that nucleate on existing ones tend to grow perpendicularly to that
plate in order to minimize the overall elastic strain. This particular type of nucleation and
growth usually leads to the formation of a Widmannstätten structure (see Fig. 4.23)
(Boyer et al. 1994; Lüterling and Williams 2007).
The stability range of the two polymorphs, α and β, can be varied by means of
suitable alloying elements. The alloying elements are usually classified into three
main families; according to the polymorph, they are able to stabilize: α-stabilizers,
β-stabilizers, and neutral elements. β-stabilizers can also be split in other two sub-
groups: isomorphous and eutectoid. A schematic representation of the influence of
each one of them can be inferred from the relevant binary phase diagrams in Fig. 4.8.
An important aspect of alloying is the appearance of a two-phase region (α + β),
140 4 Titanium and Titanium Alloys

Fig. 4.9 Ti-Al phase diagram. Aluminum is an α-stabilizer element. (Redrawn from ASM Inter-
national 1992)

extremely interesting for the development of titanium alloys. Elements like Al, C, N,
and O are classified as α-stabilizers. Additions of aluminum stabilize the alpha
polymorph of titanium also at temperatures higher than 882
C, i.e., they raise the
alloy β-transus (Tβ). In general, aluminum is deliberately added to titanium and its
alloys to obtain the stabilization effect and related improvements in the mechanical
properties. In addition to the α and β phases, a number of intermetallics are present in
the binary Ti-Al phase diagram. Among them, α2-Ti3Al and γ-TiAl are of specific
technological interest, since on these ordered phases the advanced titanium
aluminides alloys are based (see Sect. 4.3.2) (see Fig. 4.9).
The other α-stabilizers would generally originate from unwanted contamination,
coming, for instance, from the working atmosphere, like nitrogen and oxygen, and
production plant, i.e., carbon from graphite electrodes of nonconsumable arc melting
(see Fig. 4.34). Alloy properties are very sensitive even to comparatively low
concentrations of these elements, which in fact are used to qualify the grade of the
alloy itself (see Sect. 4.3).
This aspect is particularly important for oxygen that, as shown by the binary Ti-O
phase diagram (Fig. 4.10), has quite a high solid-state solubility and can be intersti-
tially hosted in the titanium lattice, up to relatively high concentrations. This, of
course, poses severe restrictions and control requirements on the different production
steps, particularly to avoid air contamination (see Sect. 4.4). Obviously, the impurity
4.2 Physical Metallurgy Aspects 141

Fig. 4.10 Ti-O phase diagram. Oxygen is a α-stabilizer element. (Redrawn from Massalski et al.
1986)

content of the raw materials, i.e., titanium and alloying elements, is another impor-
tant aspect to consider.
Substantially no effect on the β-transus is introduced by the neutral elements, like
tin (Fig. 4.11) and zirconium, that are soluble both in α and β phases. They act as
solid solution strengtheners and guarantee this mechanism also at high temperatures
and in creep regime. This is an aspect that has been exploited for the development of
high-performance alloys (see Sect. 4.3.1.3).
β-stabilizers are those elements that widen the stability temperature range of the
high-temperature polymorph of titanium, the β phase, by lowering the β-transus.
These alloying elements can be divided into two main categories: isomorphous and
eutectoid. Isomorphous β-stabilizers, e.g., Nb and V (Fig. 4.12), have the same bcc
crystalline structure as β-titanium. Moreover, they are characterized by a relatively
limited solubility in the α phase and a complete solid-state solubility in β-titanium
above given temperatures. β phase can be stabilized down to room temperature for
sufficiently high concentrations of β-stabilizers. Mo, Ta, Nb, and V are the most
important β-isomorphous elements, showing relatively high solubility in Ti. The
melting temperature of a β-isomorphous-containing alloy may be either lower or
higher than the melting temperature of pure titanium. This aspect, together with the
142 4 Titanium and Titanium Alloys

Fig. 4.11 Ti-Sn binary phase diagram. Tin additions do not change significantly the β-transus, and
in fact tin is classified as neutral alloying element. (Redrawn from Massalski et al. 1986)

Fig. 4.12 Two examples of binary phase diagram involving titanium and two isomorphous β phase
stabilizers: niobium and vanadium. See main text for the comments on the general features of these
diagrams. (Redrawn from Massalski et al. 1986)
4.2 Physical Metallurgy Aspects 143

Fig. 4.13 Ti-Cr phase diagram. Chromium is a passive eutectoid β phase stabilizer. (Redrawn from
Massalski et al. 1986)

presence of a liquid-solid two-phase field, has to be considered from the technolog-

ical point of view, in relation to possible segregation effects during the solidification
of the alloy from the melt.
The binary phase diagrams involving titanium and eutectoid β-stabilizers do not
display a complete solid-state solubility range (Figs. 4.13 and 4.14). The solubility of
these elements is limited also in the α phase, and the common feature to all relevant
systems is the presence of an eutectoid transformation. Upon cooling, the decom-
position of the β phase into a mixture of α phase and an intermetallic is observed.
Depending on the kinetics of this transformation, these alloying elements can be
classified as either active or passive β-stabilizers. For active β-stabilizers, like copper
and silicon, the relevant decomposition reactions are β-Ti to α-Ti + Ti2Cu and β-Ti to
α-Ti + Ti3Si. Elements that can be classified as passive β-stabilizers are, for instance,
chromium (bcc), iron (bcc), and manganese (bcc). In this case, the decomposition
reaction is more sluggish, if compared to active elements. In the Ti-Cr system
(Fig. 4.13), the reaction is β-Ti to α-Ti + TiCr2. TiFe and TiMn are other
144 4 Titanium and Titanium Alloys

Fig. 4.14 Ti-Cu phase diagram. Copper is an active eutectoid β phase stabilizer. (Redrawn from
Massalski et al. 1986)

intermetallics that may precipitate in the α phase in the Ti-Fe and Ti-Mn binary
systems. As a general comment on the active or passive character of the eutectoid
β-stabilizers, some clues can be inferred from the relevant binary phase diagrams,
with the necessary precautions recommended when drawing conclusions on kinetic
aspects from thermodynamic tools, like phase diagrams are. The stoichiometry of the
intermetallic phases are an indirect indication of the compositional changes involved
with the eutectoid transformations. The larger is the compositional change (TiCr2,
TiFe, TiMn, etc.), the longer time it will take for the transformation to be fully
accomplished. A sluggish eutectoid transformation of the passive titanium alloying
elements would be coherent with this aspect. Intermetallics with titanium-rich
composition, closer to the eutectoid, result from the transformation involving active
β-stabilizers. From the alloy design point of view, an interesting information is the
concentration of each one of the β-stabilizer elements, needed to retain the high-
temperature polymorph down to room temperature. This critical concentration (βc)
for some of the following transition metals is given among brackets (wt%): Co (7),
Cr (8), Cu (13), Fe (4), Mn (6), Mo (10), Nb (36), Ni (9), Ta (45), V (15), and W
(23) (Boyer et al. 1994). To complete the picture of the β-stabilizers, it is worth
4.3 Main Groups of the Titanium Alloys 145

Fig. 4.15 Ti-H phase diagram. (Redrawn from Massalski et al. 1986)

mentioning hydrogen, an important gaseous element that is classified as a passive

β-stabilizer (Fig. 4.15). The eutectoid reaction leads to the formation of the hydrogen
rich titanium δ-hydride: β-Ti to α-Ti + δTiH2. This hydride may cause an important
embrittlement of the alloy; the more severe, the higher is its concentration. For the
same reasons quoted with reference to carbon and the gaseous α-stabilizer, i.e.,
nitrogen and oxygen, hydrogen can possibly be present in titanium and its alloys out
of contamination. This raises the need for a strict control on the processing and
fabrication conditions, all through the production cycle, starting from extraction or
recycling of titanium up to the final forming operations.

4.3 Main Groups of the Titanium Alloys

Several classifications can be found in the literature as concerns titanium alloys,

generally coherent with each other, just introducing some naming differences.
Titanium alloys can be grouped into two main families: conventional titanium alloys
and advanced titanium alloys, as displayed by the diagram in Fig. 4.16.
146 4 Titanium and Titanium Alloys

Fig. 4.16 The main classes of titanium alloys and relevant short descriptions as concerns their
compositions

The classification of the conventional titanium alloys, is based on their phase

composition, with reference to the α and β polymorphs. This is certainly an impor-
tant aspect to be considered for materials selection purposes, although the actual
microstructure and relevant properties of any specific Ti-alloy are very much
dependent also on its thermomechanical history.
The advanced titanium alloys are based on the intermetallic aluminide phases:
γ-TiAl and α2-Ti3Al. Several systems have been taken into account, single and
multiphase alloys, although at the present time, the only important aircraft applica-
tion has been achieved with TiAl-based alloys (see Sect. 4.3.2). In the following
sections, both families of conventional and advanced titanium alloys will be
presented, with particular regard to the metallurgical aspects and important charac-
teristics that are interesting for the applications in the aerospace field.

4.3.1 Conventional Titanium Alloys

The main groups of conventional titanium alloys are α-, near-α, α + β, and β-alloys.
Table 4.2 gives the compositions of selected titanium alloys, of the first three groups
of the classification proposed in Fig. 4.16. In Sect. 4.3.1.4, similar information is
provided in Table 4.4, for the main titanium β-alloys. Another information that is
specified for each alloy is the relevant beta transus, above which the alloy is fully β.
This temperature is most important for setting up the conditions of the thermal
treatments and hot-working operations.
The classification adopted for the conventional titanium alloys can be visualized
using a pseudo-binary phase diagram (Fig. 4.17), in which the alloy phase fields are
given as a function of temperature and equivalent β-stabilizer concentration.

4.3.1.1 α-Alloys

These alloys are characterized by the absence of the ductile-to-brittle transition,

which is typical of the bcc alloys. For this reason, α-alloys can be employed also for
4.3 Main Groups of the Titanium Alloys 147

Table 4.2 Composition and classification of some important conventional Ti-alloys

Alloy Alloy β-transus
type designation Alloy composition [wt%] [
C]
α Grade 1 CP-Ti (0.2Fe, 0.18O) 890
Grade 2 CP-Ti (0.3Fe, 0.25O) 915
Grade 3 CP-Ti (0.3Fe, 0.35O) 920
Grade 4 CP-Ti (0.5Fe, 0.40O) 950
Grade 6 Ti-5Al-2.5Sn 1040
Grade 7 Ti-0.2Pd 915
Grade 12 Ti-0.3Mo-0.8Ni 880
IMI 230 Ti-2.5Cu –
Ti-3-2.5 Ti-3Al-2.5V 935
Near-α IMI 685 Ti-6Al-5Zr-0.5Mo-0.25Si 1020
IMI 834 Ti-5.8Al-4Sn-3.5Zr-0.5Mo-0.7Nb-0.35Si- 1045
0.06C
Ti6–2–4-2-S Ti-6Al2Sn-4Zr-2Mo-0.1Si 995
Ti-811 Ti-8Al-1V-1Mo 1040
TIMETAL 1100 Ti-6Al-2.7Sn-4Zr-0.4Mo 1010
α+β Ti-6-4 Ti-6Al-4V (0.20O) 995
Ti-6-4 ELI Ti-6Al-4V (0.13O) 975
IMI 315 Ti-2Al-2Mn –
IMI 550 Ti-4Al-2Sn-4Mo-0.5Si 975
IMI 679 Ti-2Al-11Sn-4Zr-1Mo-0.25Si –
Ti-17 Ti-5Al-2Sn-2Zr-4Mo-4Cr 890
Ti-6-2-2-2-2 Ti-6Al-2Sn-2Zr-2Mo-2Cr-0.25Si –
Ti-6-2-4-6 Ti-6Al-2Sn-4Zr-6Mo 940
Ti-6-6-2 Ti-6Al-6V-2Sn 945
For the β-alloys the reader is referred to Table 4.4. The β-transus for each alloy is also indicated
(Cotton et al. 2015; Leyens and Peters 2003; Lüterling and Williams 2007; Meetham 1981)

cryogenic applications, e.g., spacecraft tanks, for which Ti-5Al-2.5Sn is used.

α-Alloys exhibit good strength, creep resistance up to intermediate temperatures,
and toughness. Owing to their plain composition and microstructure, α-alloys are
generally well weldable. Unalloyed titanium, also known as CP (commercially pure)
titanium, is generally classified as a conventional α-alloy, since its structure at room
temperature is fully hcp. There are several grades of CP-titanium (see Table 4.2), and
the main difference among them consists in the oxygen and iron content.
CP-titanium is characterized by an excellent corrosion resistance, as it can be
inferred from the Pourbaix diagram for pure titanium (Fig. 4.18). The stability
field of a protective passivation layer is very wide and the corrosion resistance of
CP-Ti has made of this material an interesting choice for the fabrication of compo-
nents of chemical and petrochemical industry. Although CP-Ti is more expensive
than stainless steel, still, the products made of this alloy may have a lower lifecycle
cost thanks to their better durability. Therefore, CP-Ti, whose tensile properties are
148 4 Titanium and Titanium Alloys

Fig. 4.17 Portion of binary phase diagram describing the classification and phase composition of
the main titanium alloys, with the indication of some of them. (Redrawn from Boyer et al. 1994;
Lüterling and Williams 2007)

listed in Table 4.3, is often used as substitute material of various grades of stainless
steel (Lüterling and Williams 2007).
Grade 1 CP-Ti has the lowest strength level, but an excellent cold formability.
Thanks to its properties, it can be used as cladding alloy for steel reactors and deep
drawing applications. Grade 2 has a higher strength and it is the most used CP-Ti
grade, while Grade 3 is quite exclusively used for the fabrication of pressure vessels.
Grade 4 has the highest strength and it is used for fittings and mountings (Leyens and
Peters 2003). α-Alloys were the first type of titanium alloys to be developed. The
IMI 317 (Grade 6) belongs to this group and also alloys with limited concentrations
of β-stabilizers, like the IMI 230, containing 2.5% of Cu (see Table 4.2). However, in
this alloy the main role of copper is not much the stabilization of the β phase, but
rather a contribution, through the formation of the ordered intermetallic phase Ti2Cu,
to precipitation strengthening. In α-alloys, alloying elements are those that can be
solutioned to some extent into the hcp lattice, either as substitutional (Al, Sn, Zr) or
interstitial (C, N, O) atoms. The main strengthening mechanisms for these alloys rely
on solid solution hardening in the first place, with possible contribution from
4.3 Main Groups of the Titanium Alloys 149

Fig. 4.18 Pourbaix diagram for pure titanium. (Redrawn from McCafferty 2010)

Table 4.3 Tensile properties of CP-titanium

Alloy E [GPa] UTS [MPa] Yield strength [MPa] Elongation [%]
Grade 1 105 240 170 24
Grade 2 105 345 275 20
Grade 3 105 445 380 18
Grade 4 105 550 480 15
E ¼ Young’s modulus; UTS ¼ ultimate tensile strength (Lüterling and Williams 2007)

precipitation hardening. Substitutional strengthening relies on aluminum mainly,

neutral elements (Sn and Zr in particular) and limited concentrations of β-stabilizers.
Oxygen enters the hexagonal lattice interstitially, and its concentration determines
the grade of the alloy (see Table 4.2), influencing its mechanical properties. The
precipitation of coherent α2-Ti3Al particles contributes to precipitation hardening, in
a similar way as the γ0 strengthening in nickel-based superalloys (see Sect. 6.4).
Dislocations that form in the alloy matrix are blocked by the precipitate boundaries,
resulting in extensive piling-up. Alternatively, dislocations may cut through the
precipitates and in this case their movement is hindered by the DO19 ordered
structure of the intermetallic phase. For these reasons, although a generally good
weldability and structural stability of α-alloys promoted their use for noncritical
components, multiphase systems responded better to the production and service
150 4 Titanium and Titanium Alloys

requirements of aerospace components (Leyens and Peters 2003; Lüterling and

Williams 2007).

4.3.1.2 α + β Alloys

The first “aeronautic” α + β alloy was the IMI 315 (Ti-2Al-2Mn), used for the
compressor blades of the Avon Rolls-Royce engine, starting from 1954 (Froes
2015). α + β alloys turned out to be excellent candidates for overcoming the
limitations of single-phase α-alloys. When molybdenum was introduced as
β-stabilizer, the Ti-7Al-4Mo and Ti-4Al-4Mo-2Sn-0.5Si (IMI550) alloys were
developed, this latter with a particular compositional focus on improving creep
resistance. The IMI550 alloy, introduced in 1958, together with the notorious and
most used Ti-6Al-4V (IMI318), became the state-of-the-art reference alloys for
aircraft applications at the beginning of 1960s. The Ti-6Al-4V alloy was developed
in the USA in the early 1950s and it is characterized by an excellent balance of
mechanical properties. Aerospace industry is its largest market, but not the only one.
In fact, Ti-6Al-4V (also known as Grade 5) is largely used in biomedical applica-
tions, especially in its cleaner extra low interstitial (ELI) form: Ti-6Al-4V ELI
(Donachie 2000; Leyens and Peters 2003; Lüterling and Williams 2007). Other
examples of α + β alloys are Ti-17 and Ti-6-2-2-2-2 (Table 4.2). The main difference
between α + β and near-α alloys, considered in the next section (Sect. 4.3.1.3), is
essentially in the maximum fraction of β phase that can form. For instance, in the
Ti-6Al-4V, the β phase content is equal to 15% at most. By adding β-stabilizers a
β-transus reduction is observed (see Table 4.2). The α + β alloys can be strengthened
and their microstructure modified by solution and aging treatments. The Ms and Mf
curves in Figs. 4.17 and 4.19 represent the temperature below which the β phase
starts and completes, respectively, the transformation to the α0 hexagonal martensite,
if cooled down at a sufficiently high rate (see Sect. 4.2). Upon aging, quenched from
beta field alloys show a continuous raise of the mechanical strength (Fig. 4.19),
determined by the solid solution hardening of both polymorphs.
The introduction of the β phase is also important for the increased forgeabilty of
the alloys. Hot forging, either in α + β or β field, becomes in fact the main processing
approach for producing structural and engine components (Leyens and Peters 2003;
Lüterling and Williams 2007).
The service temperature for α + β alloys usually does not exceed 440
C. Above
this limit, creep deformation rate becomes too high and reduces intolerably the
useful lifetime of the involved components. This behavior is determined by too a
high concentration of β phase, that is not only more deformable at room temperature,
but features also a diffusivity two orders of magnitude higher than the hexagonal
counterpart (Leyens and Peters 2003, Lüterling and Williams 2007).
4.3 Main Groups of the Titanium Alloys 151

Fig. 4.19 The effects of

different thermal treatments
on different alloy
compositions. (Redrawn
from Polmear 2006)

4.3.1.3 Near-α Alloys

An obvious approach to overcome creep limitations of the α + β alloys was to

develop alloys with a lower concentration of the cubic titanium polymorph, the
so-called near-α (N-α) alloys. These alloys are characterized by comparatively low
concentrations of β-stabilizers in an α-alloyed chemistry. Therefore, the β phase that
forms within the α phase matrix, would not generally exceed 10% (see Fig. 4.17)
(Lüterling and Williams 2007). At 800
C, the β phase content in the IMI834 top
performing near-α alloy is 5 vol% only, whereas the value for the Ti-6Al-4V α + β
alloy is three times higher (Leyens and Peters 2003; Lüterling and Williams 2007).
For the stabilization of the β phase, refractory elements, like molybdenum and
niobium, are preferentially used rather than vanadium, so that lower diffusivity
alloys are obtained. At the same time, the grade of the raw materials can be
improved, in order to reduce the content of unwanted impurities, like, in particular,
iron. Comparatively high concentrations of neutral elements, like tin and zirconium,
are also there. These elements enhance the room temperature strength of the alloys,
thanks to a solid solution strengthening mechanism. Zirconium is also contributing,
152 4 Titanium and Titanium Alloys

Fig. 4.20 Scheme of a thermomechanical treatment capable to induce a bimodal (duplex) structure
in a near-α alloy, like IMI834. (Redrawn from Lüterling and Williams 2007)

with titanium, to the formation of silicide intermetallic compounds, like Ti5Si3 and
(Ti, Zr)5Si3. These incoherent precipitates, that form preferentially along grain and
α-β lamellae boundaries, stabilize the alloy microstructure, particularly at high
temperatures. The excellent creep resistance of the hcp α phase, together with the
combination of strengthening, thermal stability and toughness associated with the
presence of a limited bcc β phase, allow using the near-α alloys up to 600
C
(Donachie 2000). This temperature still represents the upper limit for the applica-
tions involving conventional titanium alloys under high stress conditions, like it is
very often the case of the aircraft engine applications. The grain structure of N-α
alloys can be changed through the recrystallization obtained via cold-working and
annealing. The very first alloy developed for high-temperature applications was the
Ti-811 (Table 4.2). The high Al content used to obtain the N-α structure may result
in the formation of the α2-Ti3Al intermetallic, which acts as a barrier to dislocation
movement. The stability field of these phases is interesting for the definition of
acceptable service conditions but also to design thermomechanical treatments, like
the one illustrated in Fig. 4.20. The sequence of operations highlights the possibility
to carry out the treatments both in α + β and β field, forming α0 martensite on cooling.
The microstructures obtained from these treatments are intrinsically more stable,
considering the higher values of the β-transus of these materials, in comparison with
all the other groups of conventional Ti-alloys (Table 4.2). The alloy development
process led to a couple of alloys that are still regarded as at the top of the perfor-
mance ranking. The IMI 834 alloy (year of issue into the market: 1984. The alloy
became later on: TIMETAL 834, when TIMET acquired IMI Titanium Ltd.) and the
companion alloy Ti 1100 (1988), although no longer on the market, had comparable
peak performances and its codename indicates clearly the maximum operating
temperature, i.e., 1100
F corresponding to 600
C approximately. Although other
4.3 Main Groups of the Titanium Alloys 153

alloys, mostly prototypal though, have been developed afterwards, the above limit
still remains, depending on the constitutional aspects of these materials (Zhao et al.
2017). Subsequent research, concerning the improvement of high-temperature prop-
erties of these alloys, pointed out that the addition of silicon in small amounts (up to
0.1%) would significantly improve their creep resistance. Following a similar
approach, RMI developed an alloy named Ti-6-2-4-2-Si (see Table 4.2), featuring
a fine dispersion of titanium silicide intermetallic particles, acting as dislocation
pinning centers, thus reducing the creep deformation rate. According to these results,
all newly developed high-temperature alloys contained up to 0.5% of silicon. The
IMI 685 was the first alloy employed after a β-annealing, a treatment resulting in a
superior creep resistance (Leyens and Peters 2003).
The highly alpha stabilized N-α alloys may contain an excess of the α2-Ti3Al
intermetallic, particularly after long-term exposure to high-temperature conditions.
This has a twofold detrimental effect: increasing the alloy brittleness and, under
particular environmental conditions, triggering hot-salt stress corrosion cracking
(see Sect. 8.2). Therefore, real N-α alloys would generally feature an aluminum
content evaluated on the basis of the aluminum equivalent concentration, or order-
ing parameter (Eq. 4.4, all concentration in wt%).

1 1
%Al þ %Sn þ %Zr þ 10ð%O þ %C þ 2%NÞ < 9% ð4:4Þ
3 6

4.3.1.4 β-Alloys

β-Alloys contain a sufficient amount of β-stabilizers so to suppress the martensitic

transformation during cooling the alloy from the beta field (Eylon et al. 1994). The
as-quenched condition may be metastable, since the precipitation of the α-Ti phase
may still occur upon aging the alloy or by exposing it at relatively high operational
temperatures. Fully stable β-alloys do exist above a critical beta stabilizer concen-
tration (Fig. 4.17). To evaluate the level of constitutional stability of a β-Ti-alloy, a
parameter called the molybdenum equivalent concentration, Moeq, can be used. It is
given by Eq. 4.5 (relevant element concentrations in wt%).

Moeq ¼ 1:0Mo þ 0:67V þ 0:44W þ 0:28Nb þ 0:22Ta þ 2:9Fe þ 1:6Cr  1:0Al

ð4:5Þ

Values of Moeq in excess of 8 are necessary to stabilize the β phase upon

quenching (β-metastable alloys). The limit for fully stable β-alloys is not so well
clear-cut. In fact, even the nominally β-stable Ti-30Mo alloy turns out to be unstable
at elevated temperatures under applied stresses (Bania 1994). In Table 4.4, ranked
according to relevant Moeq, a list of the main β-alloys is provided, with the indication
of the β-transus values and of the more common applications. In some respect,
Table 4.4 Commercial name, composition, and applications of main β-alloys sorted by increasing
Mo equivalent (Leyens and Peters 2003)
Alloy Composition [wt Mo β-transus
designation %] equivalent [
C] Main applications Developer
13-13 Ti-13Nb-13Zr 3.6 Biomedical (orthope- Smith &
dic implants) N. (1992)
Beta-CEZ Ti-5Al-2Sn-2Cr- 5.1 890 High-strength applica- CEZUs
4Mo-4Zr-1Fe tions at medium (1990)
temperature
SP 700 Ti-4.5Al-3V- 5.3 900 High-strength NKK
2Mo-2Fe applications (1989)
Ti-17 Ti-5Al-2Sn-2Zr- 5.4 885 High-strength applica- GEAE
4Mo-4Cr tions at medium (1968)
temperature
VT 22 Ti-5V-5Mo-1Cr- 8 850 High-strength forgings –
1Fe-5Al
Ti-10-2-3 Ti-10V-2Fe-3Al 9.5 800 High-strength forgings Timet
(1971)
15-3 Ti-15V-3Cr-3Sn- 12 760 Airframe castings, USAF
3Al plates, sheets (1978)
Beta III Ti-11.5Mo-6Zr- 12 745 High-strength Crucible
4.5Sn applications (1969)
TIMETAL Ti-15Mo-2.6Nb- 13 807 Corrosion/oxidation Timet
21S 3Al-0.2Si resistance, TMCsb (1989)
8-8-2-3 Ti-8V-8Mo-2Fe- 15 775 High-strength forgings Timet
3Al (1969)
IMI 205 Ti-15Mo 15 727 Corrosion resistance IMI
(1958)
Beta-C Ti-3Al-8V-6Cr- 16 795 Fasteners, oils fields, RMI
4Mo-4Zr springs (1969)
VT 35 Ti-15V-3Cr- 16 – High-strength airframe –
1Mo-0.5Nb-3Al- castings
3Sn-0.5Zr
TIMETAL Ti-4.5Fe-6.8Mo- 18 810 High-strength Timet
LCBa 1.5Al applications (1990)
TMZF Ti-12Mo-6Zr-2Fe 18 743 Biomedical (orthope- How-
dic implants) medica
(1992)
1–8-5 Ti-1Al-8V-5Fe 19 825 Fasteners RMI
(1957)
B120VCA Ti-13V-11Vr-3Al 23 700 Airframe (SR-71), RemCru
landing gears, springs (1952)
TIMETAL Ti-6V-6Mo- 24 704 High-strength Timet
125 5.7Fe-2.7Al fasteners (1990)
– Ti-30Mo 30 Corrosion resistance RemCru
(1952)
– Ti-40Mo 40 Corrosion resistance RemCru
(1952)
Alloy C Ti-35V-15Cr 47 – Burn-resistant alloy Pratt &
Whitney
(1990)
a
LCB low cost beta
b
TMCs titanium matrix composites
4.3 Main Groups of the Titanium Alloys 155

β-alloys can be regarded as the most versatile class of titanium alloys, featuring the
highest strength to weight ratios and very attractive combinations of toughness,
strength, and fatigue resistance, even when produced in large cross-section parts. In
the past, the β-alloy B120VCA had been extensively used for the supersonic SR-71
Blackbird project. The α phase can precipitate from the metastable β phase in form of
very fine and extremely hard platelets with tensile strength exceeding 1400 MPa. For
this reason, β-alloys can be hardened up to much higher yield strength values than
α + β alloys. β-alloys can be processed at lower temperatures than those required for
α + β alloys, and they are also more hydrogen-tolerant than the other conventional
titanium alloys, with particular regard for α-alloys. This latter feature allows their use
in environments where hydrogen pickup is an issue.
Metastable β-alloys have gained both interest and importance because of their
combination of mechanical properties. They are characterized by a complex micro-
structure that enables the alloy designer to optimize not only strength but also
toughness. Some examples coming from this family of β-alloys are Ti-6246,
Ti-17, Beta-CEZ, SP-700, and Ti-10-2-3. Beta-CEZ is an alloy specifically designed
for applications at moderate temperatures in gas turbine engines. The Japanese
SP-700 is characterized by an excellent superplastic deformability, even at temper-
atures as low as 700
C. This result is possible thanks to a very fine-grained duplex
microstructure. The high-strength β-alloy Ti-10-2-3 combines strength, toughness,
and high-cycle fatigue life. However, these performances can be achieved if the alloy
has been properly processed, including an appropriate thermomechanical cycle. For
instance, Fig. 4.21 shows a TTT diagram for the Ti-10-2-3 alloy, with the indications
of the microstructures developed using different cooling rates. Heavily stabilized
β-alloys contain higher fractions of β-stabilizing elements and consequently a lower
volume fraction of α phase, resulting in a reduction of mechanical strength. Most of
the commercial β-alloys have a Moeq in the β-metastable range, i.e., 10–24% (see
Table 4.4) (Leyens and Peters 2003; Lüterling and Williams 2007).
The metastability of the β-alloys is responsible for their strengthening. Processing
of β-alloys usually comprises a hot-working stage followed by heat treatments (see
Sect. 4.3.1.5).
Specific characteristic properties of β-alloys are:
• Critical processing parameters, i.e., small processing window of some alloys.
• Good fatigue resistance.
• Good mechanical resistance up to 400
C.
• Good workability and weldability.
• High cost due to the high concentration of alloying elements.
• High springback.
• Low elastic modulus (application: springs and biocomponents).
• Presence of impurities and segregation in the structure.
These properties are important selection criteria for suitable applications of these
alloys that are mostly employed in structural parts at medium low temperatures apart
a few exceptions (see Sect. 4.6).
156 4 Titanium and Titanium Alloys

Fig. 4.21 TTT diagram for the high-strength Ti-10-2-3 alloy (redrawn from Leyens and Peters
2003)

4.3.1.5 Thermal Treatments and Properties of Conventional Titanium

Alloys

Alloy composition is not the only parameter that determines its properties, very
much dependent also on the microstructural features induced by suitable thermal
treatments and deformation steps. The working principles behind these aspects are
presented with reference to the Ti-6Al-4V alloy. The main aspects and alloy features
that are involved in the treatment concern:
• α-Ti, α0 -Ti, and β-Ti transformations.
• Intermetallic phase precipitation.
• Microstructure, e.g., grain morphology, grain size, composition.
• Phase composition.
The principal actions potentially interesting for a thermomechanical treatment
protocol are summarized in Fig. 4.22. The atmosphere during the treatments is either
inert or reduced down to relatively low pressure. In this way, air contamination is
avoided or kept under control, considering that gaseous species, like oxygen and
4.3 Main Groups of the Titanium Alloys 157

Fig. 4.22 Scheme of the four main phases of thermomechanical treatments that can be carried out
on Ti-alloys. AR Arc-remelted state of the alloy, SHT solutioning heat treatment

nitrogen, have an active role in the stabilization of the α polymorph. Thanks to

specific treatments, it is possible to obtain various microstructures and therefore
properties best suited for any application. The Ti-6Al-4V alloy has an estimated
Tβ ¼ 995
C (see Table 4.2). In order to understand differences arising from process
parameters variations, four different thermal treatments will be discussed.
Treatment 1 The first treatment comprises a β-solutioning followed by air cooling
down to room temperature. The main steps and relevant transformations are:
• Solutioning: 1037
C for 1 h in order to obtain a fully β phase alloy.
• Air cooling inducing the transformation: β-Ti ! β-Ti + α-Ti.
Air cooling occurs with a relatively slow rate, and a Wiedmanstätten structure
(Fig. 4.23a) is obtained from the diffusive decomposition of β phase into the mixture
of α + β. A semiquantitative estimation of the phase composition can be inferred
from the relevant pseudo-binary phase diagram (Fig. 4.23b). This grain morphology
induces in the alloy a good fatigue resistance, since it tends to hinder crack
propagation.
Treatment 2 The second treatment still involves a β-solutioning, followed this time
by an aging step at intermediate temperature in the α + β range:
• Solutioning: 1037
C for 1 h to obtain a fully β phase alloy.
• Aging: 704
C for 2 h.
158 4 Titanium and Titanium Alloys

Fig. 4.23 (a) Typical Wiedmanstätten structure obtained from treatment 1 and its effect on the
crack propagation (Polmear 2006, reproduced with permission of the Publisher); (b) a semiquan-
titative estimation of the phase composition can be obtained from the pseudo-binary phase diagram

During aging, the primary β phase decomposes into alternate α and β lamellar
grains. The resulting alloy properties are not that different from those just discussed
for treatment 1, although the aging step introduces a higher stability of the alloy
microstructure in particular as concerns the β component.
Treatment 3 In the third treatment, a faster cooling rate from solutioning is
experienced. In fact, after β-solutioning, the alloy is water-quenched and then aged
in the α + β range:
• Solutioning: 1066
C for 1 h in order to obtain a fully β phase alloy.
• Water cooling: upon rapid quenching, the α0 martensite is forming with acicular
grain morphology.
• Aging: 550
C for 24 h.
• Air cooling.
This treatment induces the formation of the so-called basketweave structure
(Fig. 4.24), preserving the martensite grain morphology but with the α + β equilib-
rium phase composition. The alloy is characterized by a good mechanical strength,
toughness, and fatigue resistance, enhanced by the fine-grained martensitic
microstructure.
Treatment 4 The main difference of this treatment concerns the substitution of
β-solutioning with an annealing in the α + β field, i.e., at a temperature below
β-transus. The other stages of the temper treatment are similar to the equivalent
ones discussed for the former treatment (n.3): water quenching and aging in the α + β
range:
• Annealing: 954
C for 2 h. In this case the alloy remains in the two-phase field
also during the high-temperature annealing. So the phase mixture of β + α
titanium is obtained.
4.3 Main Groups of the Titanium Alloys 159

Fig. 4.24 The basketweave

grain microstructure
mentioned in the description
of Treatment 3 (Polmear
2006, reproduced with
permission of the Publisher)

• Water cooling: the β phase present in the initial phase mixture, transforms into the
α0 martensite:
β-Ti + α-Ti ! α0 -Ti + α-Ti.
• Aging: 538
C for 4 h. During aging the α0 martensite component undergoes a
similar transformation as that induced by the Treatment 3. Therefore, the single-
phase α0 grains turn into a phase mixture, β + α, exhibiting a basketweave
microstructure, according to the following transformation:
α0 -Ti + α-Ti ! α-Ti + β-Ti (both from martensite decomposition) + α-Ti (from
the high-temperature annealing step, commonly named primary α phase).
This microstructure (Fig. 4.25) can be identified as duplex, since it features two
different kinds of grain morphologies. Equiaxed coarse grains of the primary α
phase, which form during the high-temperature annealing treatment and remain
substantially unchanged during the following steps of the treatment. In between
these equiaxed grains, a two-phase region still exhibits a basketweave microstruc-
ture, featuring finely intermixed laths of α and β grains, whose actual size depends of
the cooling rate. The material is characterized by a good strength, good creep
resistance, and toughness.
Tensile properties of selected alloys are given in Table 4.5, with the indication of
their specific temper condition.
As an example of the dependence of mechanical properties on the specific
microstructural features, some data for the Beta-CEZ alloy are given in Table 4.6.
A very important issue for the aerospace high-temperature applications of tita-
nium alloys is their creep resistance. The Larson-Miller curves in Fig. 4.26 summa-
rize the creep behavior of several Ti-alloys in terms of stress needed to induce 0.2%
plastic deformation. The Larson-Miller parameter (LMP) is defined as:

LMP ¼ T  ½C 1 þlog10 ðt Þ ð4:6Þ

Where:
• T is the temperature expressed in Kelvin.
• t is, in the present context, the time needed to reach 0.2% plastic deformation.
• C1 is a material constant that in case of metallic alloys can be taken equal to
20–22.
160 4 Titanium and Titanium Alloys

Fig. 4.25 The duplex grain

microstructure obtained
from the Treatment 4
(Polmear 2006, reproduced
with permission of the
Publisher)

The LM curves are widely used for the evaluation of the creep performance and
are particularly useful for comparative studies (Kutz 2002). Reliable indications can
be obtained performing the test according to standard conditions and testing times
not shorter than 1% of the expected service lifetime are recommended.
The creep curves plotted in Fig. 4.26 are fully coherent with the timeline of the
development of conventional titanium alloys. The performances of the early α-alloys
were ameliorated by the α + β alloys first and, subsequently, by state-of-the-art near-
α alloys. In agreement with this evolution, the alloys of the β-group are not present in
the Larson-Miller plot, since their higher concentrations of β phase render the
relevant creep deformation rates exceedingly faster. In the plot, the dotted curves
for two intermetallic phases, Ti3Al and TiAl, are present. These ordered compounds
have represented, and still are, a major breakthrough in the field of titanium alloys.
High strength at room and high temperatures, creep resistance, and acceptable
fracture toughness, particularly for the improved compositions and microstructures,
are some of the interesting properties of titanium aluminide alloys, especially those
based on the γ-TiAl intermetallic. The advanced Ti-aluminide alloys are now
competing with the nickel-based superalloys for the fabrication of blading of
low-pressure stages of commercial gas turbine engines (see Sects. 4.3.2 and 4.6).
4.3 Main Groups of the Titanium Alloys 161

Table 4.5 Tensile properties of selected Ti-alloys (ASM International 1991)

Alloy type Alloy denomination Tensile strength [MPa] Yield strength [MPa]
α-Alloys Grade 6 790 760
Grade 12 480 380
Near-α IMI 685 790 690
IMI 834 1030 910
Ti-811 900 830
α + β alloys Ti-6-4a 900 830
Ti-6-4 ELIa 830 730
IMI 550 1100 960
Ti-17b,c 1125 1055
Ti-6-2-2-2-2c 1030 970
Ti-6-2-4-6b 1170 1100
Ti-6-6-2a 1030 970
β-Alloys Beta-C1,3 900 830
Beta IIIa 690 620
B120VCAb 1170 1100
Ti-10-2-3a,c 1170 1100
Beta 21S 862 793
Ti-15-3b 1000 965
a
Annealed condition, may be solution treated and subsequently aged to increase the strength
b
Solution treated and aged
c
Mechanical properties and composition limits are subjected to negotiation with suppliers

Table 4.6 Tensile properties of Beta-CEZ alloy as a function of its microstructural features
Alloy UTS [MPa] Yield strength [MPa] Elongation [%]
Beta-CEZ beta annealed 1280 1180 4.0
Beta-CEZ duplex 1275 1200 13.0
Beta-CEZ beta annealed (400 μm) 1335 1275 1.0
Beta-CEZ beta annealed (60 μm) 1350 1275 8.0
Among brackets the values of the average grain size of the two alloy tempers (Lüterling and
Williams 2007)

As can be inferred from the thermogravimetric curves in Fig. 4.27, titanium

aluminide alloys are also those performing better as concerns oxidation resistance.
The Ti-6Al-4V alloy displays an extremely fast oxidation kinetics, and the close-to-
linear trend indicates that the surface reaction products are not protective, so that
oxidation may proceed and, in the long run, propagate to the whole alloy. The
reaction kinetics is slower in the case of the two state-of-the-art N-α alloys: Ti-1100
and IMI834. However, the trend of the thermogravimetric curves still suggests a
non-protective surface oxide growth. Similar comments can be applied to α2-Ti3Al
and the derived Super-α2 alloy (see Sect. 4.3.2). Definitely better is the behavior
exhibited by the TiAl intermetallic compound that in fact at this temperature
develops a protective mixed oxide layer, based on TiO2 and Al2O3, which becomes
protective at some stage (see Sect. 8.2.3).
162 4 Titanium and Titanium Alloys

Fig. 4.26 The Larson-Miller parameter for several conventional Ti- alloys of the α, α-β, and near-α
families, these latter being still the best-performing conventional alloys: IMI 834 and Ti 1100.
(Redrawn from Polmear 2006)

Fig. 4.27 Thermogravimetric curves for different Ti-alloys exposed to 1 atm air at 700
C
4.3 Main Groups of the Titanium Alloys 163

Under particular conditions, i.e., high air pressure and elevated temperature, the
kinetics of oxidation of titanium alloys may proceed at a very high rate, resembling
rather a combustion process, and indeed the term “ignition” is often used in this
regard. In a gas turbine engine, the following situations may be envisaged for the
occurrence of the alloy ignition:
• Aerodynamic stall, with a localized increase of the part temperature, since heat
removal is no longer effective.
• Impact with particles entering the engine and with the debris that come from outer
stages of the compressor. In both cases, the alloy surface will be directly exposed
to the outer atmosphere for the removal of the surface protective oxide layer.
• A similar damage can be induced by an occasional rubbing between static and
rotary parts of the compressor.
In all these situations, oxidation can proceed and involves also the inner part of
the component (Fig. 4.28) for the lack of any protective layer on the alloy surface.
Moreover, the oxidation process is sustained by the enthalpy output associated with
its exothermic character. In general, two stages can be identified during this cata-
strophic oxidation: initiation, when the process starts with the accidental damage of
the protective oxide surface layer; propagation, accompanied by the spread of the
oxidation process to the bulk component.
The oxidative reaction may structurally weaken the component and/or produce
loose debris that may contribute to the damage of the other components. Not all
titanium alloys exhibit the same behavior as concerns ignition and a testing rig
(Fig. 4.29) can be used to qualify the different materials in this respect. The test
specimen is kept under a high-pressure hot air flux and heated locally by a laser
beam, which is meant to trigger the possible ignition. As displayed by the graph in
Fig. 4.29, ignition is sustained by the alloy over specific temperature-pressure ranges
and a ranking of titanium alloys, as concerns their resistance to ignition can be drawn
from this approach:
• β alloys;
• TiAl and Ti3Al-based alloys;
• α + β alloys;
• α and near-α alloys.
Ti-aluminide and β-alloys, particularly those fully β-stabilized, resist ignition
better than the other groups of titanium alloys, for an intrinsically lower diffusivity.
In titanium aluminide alloys, in fact, the comparatively higher concentration of slow
diffusing aluminum and correspondingly lower concentration of titanium result in
lower effective diffusion coefficients than in conventional titanium rich alloys. The
significant presence of refractory elements has a similar effect in β-alloys.
The alloy C (Ti-35V-15Cr), whose ignition behavior is displayed by the diagram
in Fig. 4.29, has been developed on purpose by Pratt & Whitney to resist to burning
also when used in afterburners and in the hottest stages of the gas turbine
compressor.
164 4 Titanium and Titanium Alloys

Fig. 4.28 Schematic of the ignition process involving Ti-alloys at high temperature and air
pressure, once the protective oxide surface layer has been damaged

Fig. 4.29 Schematic of the test rig used to qualify the ignition behavior of an alloy. The graph
shows the P-T fields in which ignition is sustained or not for the α + β alloy, Ti-6Al-6V, and the
highly stabilized β-alloy, codenamed Alloy C. The plot refers to conditions with a flow in the
50–120 m/s. (Redrawn from Donachie 2000)

4.3.2 Advanced Titanium Alloys

In the search for alloys capable to overcome the limitations of the conventional
titanium alloys, systems based on the intermetallic phases such as Ti3Al and TiAl
have attracted important research interests and efforts. Table 4.7 illustrates the
potentials of the binary intermetallic Ti-Al phases, as compared to average reference
properties of conventional Ti-alloys. A common feature to all intermetallic com-
pounds is their ordered crystallographic structure (see Sect. 9.4). The ordered
character refers to the selective occupation of the crystalline sites by each atomic
species, as indicated in Fig. 4.30. This ordered atomic configuration induces
4.3 Main Groups of the Titanium Alloys 165

Table 4.7 Main properties of the binary intermetallics γ-TiAl and α2-Ti3Al compared to the
average corresponding properties of conventional Ti-alloys
Young’s RT RT fracture Creep
Density Modulus ductility toughness limit T Oxidation
[g/cm3] [GPa] [%] [MPa√m] [
C] limit T [
C]
Ti- 4.5 110 10–25 35–60 600 600
alloys
Ti3Al 4.3 145 2–10 25 800 650
TiAl 3.8 176 1–4 25 950 900

Fig. 4.30 Crystallographic structure of the α2-Ti3Al and γ-TiAl. The positions of aluminum and
titanium atoms for a fully stoichiometric composition are also indicated

particular properties in the alloy. The mechanical behavior of an intermetallic alloy is

in fact strongly dependent on the peculiar character of the chemical bonding and the
relevant influence on the dislocation structure and dynamics.
As concerns Ti-aluminides, the early alloys that have been developed for aero-
space applications are those based on the α2-Ti3Al phase, with hexagonal DO19
structure. This intermetallic was well known already as a strengthening precipitate
phase in titanium α-alloys (Sect. 4.3.1). The composition of the alloys belonging to
this family typically includes 23–25 at% Al, 11–18 at% Nb, up to 3 at% V, and
approximately 1 at% Mo (Donachie 2000). The first generation of Ti3Al-based
alloys was named near-α2, and it was characterized by a Nb concentration between
10 at% and 12 at%. Subsequent research suggested that an increased Nb content, i.e.,
13–20 at%, would have a beneficial influence on the balance between room and
high-temperature properties. The top product of this development process was the
extensively investigated TIMET Super Alpha 2 alloy (Ti-25Al-10Nb-3V-1Mo at%),
mentioned already as concerns oxidation resistance of Ti-alloys (see Fig. 4.27).
166 4 Titanium and Titanium Alloys

Table 4.8 Tensile properties of Super Alpha 2 alloy as a function of the microstructure (Lüterling
and Williams 2007)
Structure UTS [MPa] Yield strength [MPa] Elongation [%]
Bimodal 1235 1010 5.2
Lamellar 615 585 0.3
Equiaxed 900 835 1.8

The complex composition of Super Alpha 2 alloy was specifically designed to

enhance the intrinsic properties of the binary intermetallic that otherwise suffered
from poor room temperature ductility and fracture toughness (Table 4.7), quite a
common drawback for ordered compounds (see Sect. 9.4). The tensile properties of
Super Alpha 2 alloy are given in Table 4.8, with reference to different microstruc-
tures achievable with appropriate thermal treatments.
In Super Alpha 2 alloy, vanadium is used as “low-cost” replacement of niobium,
although its addition also leads to a reduction in the oxidation resistance of the alloy.
Molybdenum is used to enhance the Young’s modulus, the creep resistance, and the
high-temperature strength and, together with niobium and vanadium, contributes
also to the stabilization of the β phase, providing a decisive contribution to the
excellent ductility and fracture toughness of this alloy. Actually, in this alloy the
cubic phase has an ordered CsCl structure, designated by β0, or β2 in analogy with α2
(Leyens and Peters 2003; Proske et al. 1992).
The stabilization of the β cubic phase, in association with both, or just one, of the
Ti-aluminide phases, is rather a constant feature of these alloys. In fact, the alloy
design process resulted in important improvements in the composition of the Super
Alpha 2 and other companion alloys. Interesting mechanical resistance and a rea-
sonable workability were attained also thanks to specific processing routes, devel-
oped on purpose for these materials, like the pack-rolling process, capable to
produce 0.2 mm thick sheets of Super-α2 (Peters and Blank-Berwersdorff 1992).
However, the oxidation and creep performances of α2-Ti3Al-based alloys at temper-
ature in excess of 750
C (Table 4.7) were not adequate for the long-term structural
requirements of the hot parts of the compressor of a gas turbine engine. Therefore,
alloys based on another Ti-aluminide phase, i.e., the tetragonal L10 γ-TiAl, were
developed instead (Donachie 2000). Several properties of this intermetallic are
particularly attractive in view of the structural applications at elevated temperatures,
required by a gas turbine aerospace engine:
• Excellent oxidation resistance (see Fig. 4.27).
• Good creep properties, better than any other Ti-based alloy (see Fig. 4.26).
• High ignition resistance when compared to conventional titanium alloys.
• High specific yield strength.
• High specific Young’s modulus.
The high concentration of aluminum lowers the alloy density below the typical
values of the other Ti-alloys (Table 4.7). The ordered structure, which results in
limited ductility and fracture toughness, contributes on the other hand to the intrin-
sic high strength and creep resistance. Oxidation resistance is also positively
4.3 Main Groups of the Titanium Alloys 167

Fig. 4.31 Portion of the Ti-Al phase diagram interesting to discuss the phase stability and the alloy
microstructures as a function of the thermal treatments of γ-TiAl-based alloys. The gray-shadowed
area refers to the equivalent composition of these alloys and relevant temperature ranges for the
treatments: Te, the eutectoid transition temperature, and Tα, the alpha transus. The main micro-
structures that can be obtained with specific thermal treatments are also displayed on the right. They
have been obtained with an alloy with composition: Ti-48Al-2Cr-2Nb-1B (at%) and the treatment
parameters listed in Table 4.5 (micrographs properties of the authors)

influenced by the ordered structure, that limits the interdiffusion rate, and, most
importantly, the high concentration of the low diffusive aluminum atoms is a
prerequisite for the formation of a protective oxide layer. Single-phase alloys have
been developed with compositions like Ti-(50  53)Al-(1  2)x1, being x1 one or a
combination of the following β-stabilizers: Cr, Mn, or V. Multiphase alloys have
been also considered, including the ordered intermetallics only (α2-γ alloys), or the
three-phase systems including the β-Ti polymorph (α2-γ-β alloys). The general
description of these multiphase compositions can be parametrized as Ti-(44  49)
Al-(1  3)x1-(1  4)x2-(0.1  1)x3. The term x1 is the same as above, x2 is one or a
combination of isomorphous β-stabilizing elements like Mo, Nb, Ta, and W. The
term x3 is referred to B, C, or Si. Boron and carbon are present in the alloy in the form
of several different compounds, like Ti3AlC (P-type carbide with perovskite struc-
ture), Ti2AlC (H-type carbide with hexagonal structure), and a variety of titanium
borides, like TiB and TiB2 with the combined role of alloy strengtheners, in both
static and creep loading regimes, and grain boundary stabilizers (Liu et al. 2017;
Wang et al. 2017). As for conventional Ti-alloys, the microstructure and, thereby,
the mechanical properties of these advanced Ti-alloys can be tailored by suitable
thermomechanical treatments, capable to induce different microstructures. In
Fig. 4.31 the midsection of the Ti-Al phase diagram is displayed. The composition
range for the γ-TiAl alloys is marked by the shadowed region. Te, the eutectoid
transition temperature, and Tα, the alpha transus, are two temperatures interesting
for setting the heat treatment parameters of these alloys.
168 4 Titanium and Titanium Alloys

Table 4.9 Thermal treatments carried out on the alloy: Ti-48Al-2Cr-2Nb-1B (at%), resulting in the
microstructures in Fig. 4.31.
Treatment T [
C] Annealing time [h] Cooling rate Final microstructure
1200 4 Slow Equiaxed
1300 3 Slow Duplex
1380 1 Slow Lamellar
For this alloy, Tα ¼ 1340
C and Te ¼ 1120
C, respectively

Figure 4.31 shows also the reference microstructures for these alloys, obtained
from the thermal treatments described in Table 4.9. The alloy Ti-48Al-2Cr-2Nb-1B
is a three-phase alloy, comprising in addition to the two aluminide phases (γ-TiAl,
α2-Ti3Al) also a very limited amount of the cubic β phase, an important ingredient to
optimize the fracture toughness and the ductility of the alloy. In the present discus-
sion, the evolution of the β phase, and its ordered variants, is not considered
herewith, requiring ternary phase diagram sections for a satisfactory description of
the relevant transformations (Hashimoto et al. 1998) (Table 4.9).
The equiaxed microstructure exhibits grains of the three phases that form and are
stabilized by the annealing at 1200
C, above the eutectoid temperature. Upon
cooling down to room temperature, the phase composition and relevant microstruc-
ture do not change significantly, and the only transformation that occurs is the
ordering reaction from the α solid solution to the ordered intermetallic α2-Ti3Al.
The duplex structured alloys are obtained from annealing treatments just below the
α-transus. During this treatment too, the constituent phases are initially stabilized at
the annealing temperature, but upon cooling down to room temperature, their
concentration changes, as indicated by the lever rule. In particular, the concentration
of the α2-Ti3Al decreases, whereas that of γ-TiAl increases. This determines the
formation of colonies of alternate γ-α2 lamellar grains of the two phases inside the
grains that at the annealing temperature were fully α2-Ti3Al. These lamellar grains
are typical also of the fully lamellar microstructure that forms from a fully
α-stabilized alloy, obtained when annealing above α-transus. Upon cooling from
high temperature, the following decomposition occurs: α ! α + γ. This decompo-
sition is followed by the disorder-to-order transformation of the α phase, to form the
α2-Ti3Al intermetallic: α + γ ! α2 + γ. In this case no important changes in the
lamellar grain morphology occur.
The resulting properties are very much dependent on the microstructure of the
alloys, as exemplified by the stress-strain curves for the Ti-48Al-2Cr-2Nb (at%)
alloy in Fig. 4.32, with two different microstructures induced by suitable thermal
treatments, to be selected also in view of the possible subsequent steps of mechanical
processing. In fact, extrusion, hot-rolling, and forging are usually conducted above
Te (eutectoid temperature), whereas for superplastic and conventional hot-forming of
sheet materials, and residual stress relief heat treatments, moderate temperatures,
below Te, are more suitable.
The level of knowledge and technological development achieved with Ti-
aluminide-based alloys has slowed down the research on other intermetallic alloys,
4.4 Titanium Processing Technologies 169

Fig. 4.32 Stress-strain curves for the Ti-48Al-2Cr-2Nb (at%) alloy, with different microstructures:
duplex and lamellar. (Redrawn from Dimiduk et al. 1992)

having in principle very close target applications as those of γ-TiAl alloys. This is
the case of the alloys based on the orthorhombic Ti2AlNb intermetallic Heusler
phase (Partridge and Shelton 2001). Development steps for these alloys are mainly
concentrated on the design of new refined compositions and on the improvement of
process related aspects (Chen et al. 2014).

4.4 Titanium Processing Technologies

Some aspects of the production cycle of titanium alloy components have been
discussed already, with regard to primary titanium extraction (Sect. 4.1), types of
available alloys, and some aspects of thermomechanical treatments, developed to
induce specific structural properties (see Sect. 4.3). The production of a master alloy,
involving melting technologies, is indeed particularly critical for titanium, owing to
its elevated reactivity with elements like oxygen, nitrogen, and hydrogen, which
might be present in the working atmosphere. Moreover, titanium tends to react with
carbon, when nonconsumable electrodes of arc melting plants and graphite con-
tainers and liners are in contact with the molten alloy. Similarly, refractory ceramic
crucibles and casting molds, used in this field, suffer from important thermochemical
erosion phenomena, that may induce the release of brittle particles into the liquid
alloy. To eliminate contamination from gaseous species, titanium alloy melting is
conducted under vacuum or inert atmosphere, using argon or helium. To prevent
contamination from the mold and lining materials, a system involving water-cooled
copper containers is used. By far the most common approach is based on cold hearth
170 4 Titanium and Titanium Alloys

Fig. 4.33 Qualitative

steady-state temperature
profile for the skull-melting
process. (Redrawn from
Sears 1990)

melting (CHM), also known as “skull” melting. Cold heart melting has become an
industrial process for the production of rotor grade titanium starting from 1985. A
water-cooled copper mold assures a rapid extraction of the solidification enthalpy, so
that a solidified alloy layer forms inside the mold, at the interface with the rest of the
solidifying alloy (Fig. 4.33). Therefore, a direct contact between the liquid alloy and
the copper mold is avoided, and the solidification of the inner part of the ingot can
proceed in contamination-free conditions. At the end, the outer part of the resulting
ingot can be eliminated with all the contaminant burden, coming from the refinement
that the CHM has on the master alloy batch. The skull-melting principle is quite
flexible, and it can be employed in different production plants, involving secondary
melting for the production of ingots and powders, like in the case of a gas atomiza-
tion stage.
Vacuum arc-remelting (VAR), using CHM, is classified for the production of
ingots of CP-titanium and other titanium alloys, characterized by high purity and
chemical homogeneity. These features are a prerequisite to guarantee adequate
standard of the resulting mechanical properties. Differently from VAR applied to
steels (see Sect. 5.3.6) and nickel base superalloys (see Sect. 6.4), in the production
of titanium alloys it may be a primary melting process, whereas for special steels and
superalloys, the primary melting is performed either by electric arc melting (EAF,
see Sect. 5.3.1) or by vacuum induction melting (VIM, see Sect. 5.3.2). In addition to
VAR, other CHM processes are becoming more and more important for titanium
alloy melting (Lüterling and Williams 2007). The production path for titanium alloys
includes single, double, or even triple VAR processing, to form, in the first place
cylindrical ingots, i.e., the (consumable) electrodes of the process. The very first
electrode is actually obtained by compacting and welding together pieces and pellets
of titanium sponge and alloying elements. Since there are strong incentives in using
titanium scraps also in VAR, this source of titanium can be incorporated into the
starting electrode. Each electrode will have the desired alloy nominal composition.
Compacted blocks have sufficient strength to remain intact during handling and
melting. If needed, mechanical compaction by means of a hydraulic press can be
used to strengthen the electrode. A twin cone mixer can be used to optimize the
blending of the different components: titanium sponge fragments, scraps, and
4.4 Titanium Processing Technologies 171

alloying elements (Lüterling and Williams 2007). The process is carried out using a
cylindrical crucible under vacuum. Initially, an arc is sparked between the electrode
and a small amount of the same alloy to be remelted, positioned at the bottom of the
crucible. Ideally, the melting rate is constant all through the process. The main
advantages of the VAR technology (also known as CEVAM or CEVAR, consum-
able electrode vacuum arc melting/vacuum arc remelting) are:
• Considerable reduction of dissolved gases, such as oxygen, nitrogen, and hydro-
gen, since under vacuum conditions these gases are released from the liquid metal
and pumped away by the vacuum system.
• High degree of control on the solidification microstructure, thanks to the possi-
bility of tuning and adapting the melting rate of the alloy to the specific process
requirements.
• Minimization of segregations.
• Alloy chemical homogeneity across the ingot, improved by the successive
remelting.
The main critical aspects concern the production of electrodes and the segregation
of alloying elements within the electrodes. Remediation approaches and other details
about the technique can be found in Sect. 5.3.6. However, the electrode issue can be
solved by another melting process, involving the use of the non-consumable elec-
trode vacuum melting (Fig. 4.34).
In this process, the electrode is not made of the same alloy that has to be melted
but of another material, like graphite or tungsten. The working principle is similar to
gas tungsten arc welding (GTAW). The non-consumable electrode rotates above the
molten pool, and in the meantime its temperature is kept at acceptable values by a
dedicated water cooling system. The technique ensures good flexibility and allows
for the production of both powders and ingots. Moreover, since there is no con-
sumption of the primary electrode and no need to prepare it, melting can even
proceed from scraps directly, as shown by the scheme in Fig. 4.34. In this regard,
the technique is an excellent choice for recycling purposes. Two other melting
processes interesting for titanium alloys processing are electron beam melting
(EBM) and plasma arc melting (PAM). In EBM plants, melting, carried out under
vacuum (104–106 torr) is induced by an electron beam bombardment of the alloy
charge. The general scheme of an EBM equipment is given in Fig. 4.35. EBM
furnaces employ one or more electron beam guns (power up to 1.2 MW).
To enhance the homogeneity of the final product and to reduce the density of
defects and the incidence of ballistic effects, several melting stages (two in the
scheme in Fig. 4.35) are usually employed. EBM can be used for the production
of both primary electrodes, finished ingots, and slabs starting from sponge and
scraps. Owing to the highly controlled working conditions needed by the electron
beam guns, this technique guarantees elevated purity standards, although it is not
recommended for large-scale production and as melting stage of powder atomization
and rapid solidification plants. These latter processes, in fact, require comparatively
higher gas pressures to facilitate the rapid solidification of the powders (Sears 1990).
172 4 Titanium and Titanium Alloys

Fig. 4.34 Schematic of a non-consumable electrode vacuum arc remelting (VAR) equipment used
for the production of titanium and titanium alloy ingots. (Redrawn from Sears 1990)

Fig. 4.35 Schematic of an electron beam melting (EBM) apparatus used for the production of
ingots. (Redrawn from Sears 1990)
4.4 Titanium Processing Technologies 173

Fig. 4.36 Schematic of a plasma arc melting (PAM) apparatus used for the production of ingots.
(Redrawn from Sears 1990)

Plasma arc melting (PAM) features an architecture which is similar to that of

EBM but based on plasma torches to melt the alloy. This latter feature is illustrated
by the scheme of the plant in Fig. 4.36. A series of plasma torches are present for the
same reason as in EBM plants. The composition of the plasma gas and of the
working atmosphere are chosen considering the high reactivity of titanium and the
possibility of limiting any possible contamination of the alloy.
Titanium sponge fragments, scraps, and pieces of recycled ingots constitute
the feed material for the PAM process. The molds used in both EBM and PAM
exploit the skull-melting principle to eliminate contamination from the crucibles
and retorts (Sears 1990). Technical data of real titanium-melting plants are listed
in Table 4.10
Several post-melting processes may be used for turning the titanium master alloys
into any final product. Rapid solidification (RS) is one option that finds application
in the production of powders for powder metallurgy (PM) routes. RS is known to
have had an impact on the metallurgy of several alloys for the changes, mostly far
from equilibrium conditions, introduced by the very fast removal of the solidification
enthalpy from the melt and cooling down of the solid phase (Das and Davis 1988).
174 4 Titanium and Titanium Alloys

Table 4.10 Technical specifications of real plants for melting and casting of titanium alloys (Sears
1990)
Efficiency
Technique Type Total powera [MW] kWh/kg kg/h
Consumable arc (VAR) Direct melt 1.0 0.85 1200
Nonconsumable arc Tilt-pour 0.6 2.0 225
Electron beam (EBM) – 4 guns Cold hearth 2.0 1.1–2.2 450–1360
Electron beam (EBM) – 2 guns Cold hearth 2.4 1.6–3.0 450–900
Plasma arc (PASER) Direct melt 0.75 1.1 690
Plasma arc (PAM) – 4 torches Cold hearth 2.1 2.5b 900b
Plasma arc (PAM) – 4 torches Cold hearth 2.25 2.5b 900b
a
Data based on arc or beam power only. The power value can vary greatly due to the type of feed
material
b
Predicted values

RS of titanium and its alloys is interesting as concerns the enhancement of the

solubility limits and the obtainment of either micro- or nanocrystalline structures. On
some occasion, this has resulted in improvements of the physical and mechanical
properties of the alloys and has afforded the possibility of developing novel com-
positions (Sastry et al. 1983, Suryanarayana et al. 1991). Thanks to the higher
concentrations of reactive elements that can be solutioned into the titanium alloy
matrix through RS, the formation of a fine dispersion of oxide particles can thereafter
be induced by post-processing thermal treatments. The oxide particles hinder the
motion of grain boundaries, so that a micrometric grain size is retained in the
consolidated bulk alloy. The oxide dispersion has also beneficial effects on the
creep resistance of the alloy that behaves in this respect like an oxide dispersion
strengthened (ODS) material (see Sect. 9.3) (Sastry et al. 1983; Suryanarayana and
Froes 1990).
Powder metallurgy (PM) represents an important class of processing for titanium
alloys, which is recently gaining more importance, even as concerns additive
manufacturing. Powder metallurgy is regarded as an economically advantageous
method for the production of near-net shape components. Mechanical properties of
PM titanium products can reach values that may be even better than those of cast
alloys and comparable to those of forged alloys. At the present time, powder
metallurgy is experiencing a sort of renewed interest, in particular, thanks to
innovative 3D printing techniques, offering the possibility of producing stiff and
wear resistant components based on reinforced titanium PM alloys. The production
of high-quality powders is of course paramount for PM processing, and, in this
regard, the high surface area may critically enhance the reactivity of titanium
powders. The expertise developed over the years in powder atomization and rapid
solidification methods is very useful indeed for setting up the production of high-
grade titanium alloy powders. Before consolidation, for instance, through hot iso-
static pressing (HIP), the powders are usually encapsulated in austenitic steel cans
and then degassed and evacuated. These steps are absolutely recommended in order
to avoid surface gas absorption, with the risk of forming oxide defects on the alloy
4.4 Titanium Processing Technologies 175

surface and inside the open pores. The HIP process conducted to consolidate a
Ti-6Al-4V powder has a duration of 1–3 h, at a temperature between 920 and
970
C, while the compaction pressure is about 2000 bar (Leyens and Peters
2003). Titanium powder production might benefit soon from the consolidation and
widespread use of an alternative approach to the Kroll’s extraction method: the
so-called Meltless Ti Process (Dargusch and Keay 2009). The substantial difference
with respect to the classical and still dominant Kroll’s extraction method is that the
Meltless approach, since producing sponge powders, instead of sponge cakes,
requires PM approaches for the production of the bulk components. At the present
time, the interest for this novel technique is sustained by a number of several positive
aspects, for instance, an estimated 50% reduction in energy consumption and a
comparable reduction of greenhouse gas emission. A similar reduction concerns
also the emission of pollutant gases, like SO2 and NOx.
The development of γ-TiAl-based alloys has prompted research activities on the
casting technology, aiming at obtaining components with enhanced mechanical
properties. A particular research interest has been dedicated to the directional
solidification of γ-TiAl alloys, using a technology in some respects similar to
directional solidification of polycrystalline columnar grained or single crystal super-
alloys (see Sect. 6.5.2). A major difficulty of directional solidification applied to
γ-TiAl alloys is controlling the formation of an optimal lamellar microstructure (see
Fig. 4.31), resulting from the solid-state phase decomposition of the alloy. More-
over, the apparatus employed for this casting technology, considering the required
solidification kinetics, cannot rely on skull melting, typical of titanium casting
techniques. In fact, ceramic molds would be the obvious choice, considering the
comparatively long solidification times caused by the involved low growth rates.
This implies a prolonged contact between the molten alloy and the ceramic, with a
consequent erosion of the mold that may lead to contamination of the melt by
ceramic inclusions. The solidification parameters, in particular the growth rate, can
be tuned to reduce this contamination and at the same time induce the formation of a
better dendritic microstructure, with particular regard to improved creep resistance
(Ding et al. 2011; Lapin et al. 2002; Lapin 2006).
Solidification using mold casting is just one of the possible processing
approaches available, used for both conventional and advanced Ti-alloys, including
the manufacturing of aerospace near-net shape components. Referring to investment
casting, different procedures are available: centrifugal, tilt, and counter-gravity
casting. Casting techniques are preferred for those alloys exhibiting poor workabil-
ity. In all titanium alloys, this feature is to some extent dependent on the presence of
the β phase, either as a stable constitutional component of the alloy or as a phase
present at the metal working temperature only, like for α and near-α alloys. How-
ever, the high performances and elevated reliability levels required for aerospace
components can be fully accomplished only through complex processing routes. For
instance, the complete manufacturing sequence of the low-pressure blades, first
mounted on the GEnX engine (Bewlay et al. 2016), made of the General Electric
176 4 Titanium and Titanium Alloys

alloy: Ti-48Al-2Cr-2Nb (see Sect. 4.6), involves the following main processing
steps:
• Master alloy ingot production, by casting.
• Blade production by melting and casting.
• Pre-HIP (hot isostatic pressing) solution heat treatment.
• HIP, to eliminate casting porosity.
• Recovery post-HIP heat treatment.
• Radiographic part control and inspection.
• Machining for part finishing and oxidation resistant coating deposition.
• Assembly, in the low-pressure turbine (LPT) stages.
The success of the Ti-48Al-2Cr-2Nb alloy in the LPT blades of commercial
engines is now fully established, with remarkable advantages regarding the overall
performances of the engine, also in terms of fuel saving. Recent developments in the
directional solidification process have demonstrated the possibility to obtain a single
crystal TiAl alloy with exceptional creep resistance (Chen et al. 2016). This trend
and the promising results may suggest to extend the use of these alloys to other
structural components of the gas turbine engines.
To complete the picture as regards the processing of titanium alloy components
for aerospace applications, metal forming still maintains an important role, notwith-
standing some critical aspect existing in the manufacturing of some titanium prod-
ucts, like sheets. The difficulties in processing titanium and its alloys are primarily
due to some specific material features, such as strong springback effect, fret tendency
when in contact with steel tools, high strength, and anisotropic sheet properties. For
these reasons, deep drawing and stretch forming are not suitable for titanium alloys
due to the low strain that can be achieved and the correspondingly high required
stresses. Hot-forming is an important option. Warm calibration, i.e., creep forming,
is used for sheet forming and hot-working, which is usually performed at around
850
C (see Figs. 4.20 and 4.22). After hot-working, an oxide layer may form on the
surface of the material at 500
C. This layer should be removed by means of pickling
processes, since an incomplete removal may result in a surface embrittlement and a
consequent reduction in the fatigue strength and toughness of the alloy.
Superplastic forming (SPF), i.e., the possibility for some materials, under specific
loading conditions, to undergo extremely large tensile strains without necking
(Perez-Prado and Kassner 2009) has been considered for titanium alloy too. SPF
for the manufacturing of aerospace components begun with two projects for the
fabrication of supersonic aircrafts: the Concorde and the Rockwell B-1. An adequate
knowledge on the superplastic behavior of the relevant materials was developed,
including Ti-6Al-4V alloy. At that time, when both the Concorde and the B-1 were
under development, composite sandwich structures were selected for several com-
ponents. These structures were obtained by the combination of SPF and diffusion
bonding (DB). DB is a welding technique, based on the solid-state interdiffusion of
the two materials to be joined, pressed together at relatively high temperatures
(0.5∙Tm of the alloy approx.). The Ti-6Al-4V alloy, during isothermal tensile test
performed at 0.5∙Tm, shows a superplastic behavior, with a tensile elongation of
4.4 Titanium Processing Technologies 177

Table. 4.11 Superplastic properties of selected Ti-alloys (ASM International 1991)

Alloy Common alloy Test temperature Elongation
type name [
C] Strain rate [s1] m [%]
α+β Ti-6-4 840–870 1.3  104 to 0.75 750–1170
103
β B120VCA 800 2  104 0.50 229

1000% while developing no necking (Leyens and Peters 2003; Nieh et al. 1994). To
rank the SPF behavior of material, the strain rate sensitivity factor, m, has been
introduced, using the Eq. 4.7 (ASM International 1991):

∂ ln σ
m¼ ð4:7Þ
∂ ln ε_

where:
• m is the strain rate sensitivity factor. The higher the factor, the greater the
superplastic deformation exhibited by the alloy.
• σ is the flow stress, expressed in MPa.
• ε_ is the strain rate expressed in s1.
The large strain values associated with superplasticity can be explained by means
of three mechanisms occurring at the same time: grain boundary diffusion, grain
boundary sliding, and grain rotation. Nowadays, it is assumed that more or less all
polycrystalline materials may exhibit a superplastic behavior, under appropriate
temperature and strain rate conditions, if it has a sufficiently fine-grained micro-
structure and a good resistance to grain growth during SPF processes (Leyens and
Peters 2003; Nieh et al. 1994). Since some Ti-alloys exhibit a superplastic behavior
in their standard production state, no particular modifications in the alloy chemistry
or in the high-temperature processing is needed (ASM International 1991). Details
regarding the superplastic properties of the Ti-6Al-4V and B120VCA alloys are
given in Table 4.11. Further details about SPF of materials can be found both in the
further readings, and in the references at the end of this chapter.
Among the different plastic deformation processes, forging is by far the most
used for titanium and its alloys, although rolling is used as well for the production of
specific products. A comparison among the main plastic deformation processes, with
relevant deformation rates, is given in Fig. 4.37. One of the advantages of forging is
that it allows for a precise tailoring of the resulting alloy properties and
microstructures.
Forging of titanium alloys can be conducted either in α/β or β field. In α/β forging,
the alloy is heated up at a temperature 30–100
C below its β-transus. The final
cooling process is usually performed in air. In case of β forging, the alloy is heated
above its β-transus, and the process has to be completed before the start of the β-to-α
transition. β-forged alloys tend to be much more sensitive to process conditions,
especially as concerns deformation temperature. For β forging, the cooling rate
needs to be selected on the basis of the characteristics of the alloy: β-alloys can be
178 4 Titanium and Titanium Alloys

Fig. 4.37 Deformation processes, equipments, and relevant strain rates. (Redrawn from Leyens
and Peters 2003)

slowly cooled down in air, while near-α alloys need a faster cooling, for which water
quench can be used (Leyens and Peters 2003).
The main forging techniques are hammer forging and press forging. In the case of
hammers, the following three main families can be considered:
• Gravity hammers.
• Counter-gravity hammers (counterblow hammers).
• Double action hammers.
The main features of the processes using hammers are:
• Ease of piece positioning during forging processes (especially gravity hammer).
• High fraction of scrap.
• High strain rates.
• Post-finishing and excess material removal are needed.
• Series of blows and heating of the workpiece.
Press forging is different from standard forging process. In fact, while in forging
the force is applied impulsively, press forging works by slowly applying a contin-
uous pressure, so the strain rate is lower if compared to that of hammer forging
processes (see Fig. 4.37). The main features of the press forging are:
• Deformation rate is controlled.
• The piece can be subjected to multidirectional stresses.
• Material is processed in the right quantity without and excessive scrap at the end.
4.5 Defects in Titanium Alloys and Diagnostics 179

Hot forging is usually performed in Ni-based superalloy dies that are preheated at
a temperature in a range between 760 and 1050
C, rather than 300–500
C, like in
traditional forging processes. Isothermal forging is performed at same temperatures
for die and workpiece, in inert atmosphere and with low deformation rates, to avoid
overheating. For Ti-alloys, the temperature of the process can be in the 900–1050
C
range, depending on the forging process that the material has to undergo. Of course,
also during the hot-forming steps, the control of the atmosphere to avoid surface
reaction is paramount, requiring the use of environmental chambers, or inert gas flux,
to protect the part surface.

4.5 Defects in Titanium Alloys and Diagnostics

Notwithstanding the sophisticated control and monitoring systems operated all along
the production cycle of titanium casting and metal working, titanium alloy products still
require quality controls and inspection tests both during their processing and at the end
of the production cycle. These controls are needed to guarantee, with a high level of
confidence, that the materials and the components have actually reached the targeted
properties, in order to comply with the relevant performance and safety requirements.
The main kinds of defects and inhomogeneities present in titanium products have been
classified according to their formation mechanisms. This is particularly useful, since
starting from defect analysis, the identification of possible faulty aspects of the process
can be straightforwardly identified and, hopefully, fixed. The main kinds of defects that
can be encountered in titanium alloys are:
• Type I (hard α) defects, or hard interstitial defects (HID), are mostly associated
with the presence of oxygen. These defects may be originated during various
steps of the alloy processing: sponge production, first melt electrode formation,
and melting and remelting stages. Oxygen is able to contaminate the titanium
sponge during its production. The same may happen during the production of the
VAR electrodes, in particular during welding operations, handling, especially
under fast oxidation conditions, contamination introduced by scraps, contami-
nated master alloy, air, or water leaks. These defects are typically constituted by
highly interstitially alloyed titanium, featuring a particularly brittle behavior.
Since these defects are generally due to contamination by gaseous, and therefore
light, elements, like oxygen but also nitrogen, they are also alternatively identified
as low-density inclusions (LDIs).
• Type II (α-stabilized) defects are regions in which an excess of α-stabilizers is
present so to change locally the alloy phase composition. These defects can be
produced by the ballistic drop in EBM or PAM plants from scrap source and
segregation during the final stages of solidification during VAR processes.
• High-density inclusions (HDIs) are due to debris and fragments from
nonconsumable electrodes and grains of unmelted refractory metals, e.g., iso-
morphous β-stabilizers. Another source of HDIs is the scraps and debris from
machining tools.
180 4 Titanium and Titanium Alloys

• β-flecks. Like Type II defects, these defects too are generated by the segregation
in the melt, but of β-rather than α-stabilizers. Alternatively, an excess of β phase
can be trapped into the alloy if presumed α/β treatments actually occur in the β
field, possibly for the effects of adiabatic heating during metal working
operations.
• Voids and residual porosity from solidification volume contraction. These pores
remain trapped into the alloy and should be eliminated. HIP is a typical and
effective approach to reduce or to get completely rid of these detrimental defects,
which may significantly reduce the actual bearing section of the component and
act as stress concentrators (Lüterling and Williams 2007).
Several inspection methods are available for a full characterization of the defect
structures that may be present in titanium alloys, starting from the as-cast conditions
up to the real component and, if necessary, even after a prolonged period of service.
Actually, these same methods can be used for other alloys and, if electrical conduc-
tivity is not required, like for the eddy currents, also for other materials, like ceramics
and structural polymer matrix composites, all relevant to aerospace structures.
Considering that investigations on real components are required, nondestructive
methods (NDTs, nondestructive tests) are definitely preferred. NDTs are defined
as examination, evaluation, or test performed on any type of test object without
changing or altering that object in any way, in order to determine the absence or the
presence of conditions or discontinuities that may have an effect on the usefulness or
serviceability of that object (Hellier 2001). There are several NDT methods, and a
possible classification can be made considering the main working principle the
method is based on:
• Acoustic emission testing.
• Eddy current testing.
• Magnetic particle testing.
• Penetrant liquid testing.
• Radiographic testing (including X-ray, gamma rays, and neutron radiographic
testing).
• Thermal/infrared testing.
• Ultrasonic testing.
• Visual and optical testing.
For bulk analyses, very effective approaches are based on radiographic inspections
and ultrasonic tests. In the first case, using high-energy electromagnetic radiation, i.e.,
X- or gamma rays, if available, with tomographic instrumentation, 2D or 3D images of
the inner structure of the ingot, bar, or component can be obtained. The main working
principle of the radiographic approach is based on, concerns the different absorption
coefficients of the radiation in the alloy, the inclusions, pores, etc. Radiographic testing
provides information about the shape, size, and position of defects in the material/
component. Usually, gamma ray analyses are performed on larger components, while
X-rays are used for smaller ones. The energy associated with the electromagnetic
radiation that has to be used for the analysis of the specific component must be
4.5 Defects in Titanium Alloys and Diagnostics 181

calibrated according to several parameters, for instance, the thickness of the compo-
nent to analyze and the material of which it is made.
Ultrasonic testing (UT) is also sensitive to bulk defects, since the ultrasonic wave
is reflected with different amplitudes by sample regions with different densities. In
this way, using different ultrasonic wave transducers, capable to sample the test
specimen to different depths, the shape, position, and distribution of the defects can
be mapped within the component. Since the method is essentially based on the
reflection of an ultrasonic wave, defects that are paralleled to the wave propagation
direction are difficult to be detected. An interesting development in this field, also for
the imaging capabilities that affords, is the ultrasonic phased array technique. In this
case, the transducer is composed of a matrix of several elements, so the ultrasonic
wave can be dynamically focused at variable depths within the tested material. This
can be achieved by sequentially pulsing the elements of the transducer using a
specific combination of them in the array. Furthermore, multiple wave fronts are
used in combination, in order to form a beam with a particular shape, optimized for
the geometry of the region to analyze (Hellier 2001; Mix 2005).
A characterization of surface defects can be obtained using the distribution of the
eddy currents induced into the alloy by a suitable electromagnetic probe, whose
geometry is usually designed to be compliant with the shape and geometry of the
component to be tested. The spatial distribution of the induced currents is determined
in the first place by the geometry of the component itself and, most importantly for
this methodology, by the value of the local electrical resistivity. This is the parameter
that, being influenced by the defects localized in the subsurface regions, is providing
a very powerful tool for their detection and characterization.
Surface etching is classified as a visual testing method, and it is based on the same
working principles of the metallographic etching: the regions with different micro-
structures of the examined components, like rotors, blades, and airfoils, possibly
machined for surface finishing, will be selectively attacked by a suitable chemical
etching solution, inducing chromatic changes on the alloy surface. The approach is in
principle very simple and can be applied directly onto the surface of the components,
new or already serviced. Any chemical inhomogeneity in the alloy results in a specific
reaction kinetics, that modifies the surface topography and can be examined by visual
and microscopy inspection to reconstruct the spatial distribution of the surface defects.
An alternative analysis uses dye penetrant liquids, whose absorption in the alloy
component is dependent on the open porosity and surface microcracks. These defects
can therefore be visualized, since retaining a larger concentration of the coloring agent
(Hellier 2001; Mix 2005). Another technique, particularly interesting for analyzing the
microstructure of a component, is based on the replication of the surface features using
film-forming substances, like cellulose acetate, aldehyde acetate, etc. Once consoli-
dated, the film is lifted off from surface, of which the film retains the morphology. The
replica prepared in this way can then be examined with microscopy techniques,
without the need to cut off real specimens from the component, an operation not
always feasible. Moreover, using an appropriate sequence of chemical etchings, it is
possible to remove from the sampled surface also traces of the underlying material
(extraction replica), with the possibility of obtaining also chemical and structural
information acting in a fully nondestructive way (ASM International 1992).
182 4 Titanium and Titanium Alloys

4.6 Main Aerospace Applications of Titanium Alloys

When dealing with titanium in aircraft structures, it is in a way mandatory to mention

what is still the case of the most significant use of titanium for aircraft applications,
i.e., the Lockheed SR-71 “Blackbird” (see Chap. 1, Fig. 1.8). An estimated 95% of
its total mass was made of titanium alloys, including engine’s components and most
of the airframe and fuselage, using the B120VCA alloy (see Table 4.4) (Boyer
1994). Irrespective of the comparatively limited number of these aircrafts, the SR-71
is a clear demonstration of the great flexibility of titanium alloys, in principle usable
in many parts of any aircraft. Although the SR-71 is an unprecedented example of
titanium applications to aircraft structures, still a significant percentage, in the
35–50% range, of the whole weight of a fighter aircraft is titanium. Furthermore,
titanium alloys have been, and still are, widely used in rocket manufacturing (Leyens
and Peters 2003). Some examples coming from the former Soviet Union are the
Vostok and Soyuz spacecraft and the unpiloted Luna, Mars, and Venera. For these
particular applications, titanium and its alloys were employed for the fabrication of
liquid-propellant rocket engines, skin, solid-fuel engines, fasteners, and high-
pressure vessels (Moiseyev 2006).
When coming to material selection for commercial aircrafts, other parameters
than top performances, like peak speed, a priority in the military field, must be
considered.
Weight saving, which may also concern the replacement of part originally made
of aluminum alloys, particularly when service temperatures in excess of 130
C are
possible, is one of the reasons for the success of Ti-alloys. In case of replacement of
steels and superalloys, the lower density of titanium alloys is sufficient to compen-
sate for the higher absolute strength of the replaced alloys. Weight saving and
reduction of component size become essential criteria for selecting titanium alloys.
Structural weight savings provide the possibility of increasing the payload, a most
interesting parameter when dealing with valuable items, like satellites and other
deployable objects carried by extra-atmospheric flights.
The main applications of titanium and its alloys in aircraft structures, aero-
engines, and aerospace field are herewith presented, with the deliberate intention
to highlight the extreme flexibility of these alloys. The alloys that will be considered
are those from the groups discussed in the former sections, i.e., the conventional and
advanced titanium alloys, the latter mostly based on γ-TiAl intermetallic.
Additional applications of titanium (containing) alloys will be considered in
Chap. 9 (see Sect. 9.5.2), with reference to shape memory alloys, based on the
NiTi intermetallic compound. However, even though these systems actually contain
significant concentrations of titanium, they are rather characterized by the particular
functional properties associated with the phenomenology of the shape memory effect
(see Sect. 9.5.1).
Airframe Titanium alloys can have a central role in the production of the fuselage,
whereas aluminum alloys are not suitable, due to aero-kinetic heating of the surface
skin. Titanium sheets and rivets have been extensively used in the aft end of the
McDonnell Douglas F-15 and F-15E “Strike Eagle” fighters. This latest model also
4.6 Main Aerospace Applications of Titanium Alloys 183

Fig. 4.38 Some of the parts made of titanium alloys in the Lockheed Martin-Boeing F22 Raptor
advanced fighter aircraft. (Redrawn from Donachie 2000)

employs components made using advanced SPF-DB (superplastic forming and

diffusion bonding) technique. In this way up to 726 part details and 10,000 fasteners
could be eliminated with a consequent weight reduction and an enhanced maintain-
ability. The Ti-6Al-2Zr-2Sn-2Mo-2Cr-0.25Si alloy was used in the airframe of the
Lockheed Martin F-22 (see Fig. 4.38) and in the JSF (Joint Strike Fighter) project.
This alloy is characterized by a moderate temperature capability and it is used in the
fabrication of engine bay bulkheads, where fuselage temperatures reach the highest
values. Several advantages derive from the use of this alloy in the mid-fuselage
bulkhead of the F-22, resulting in a final weight of 150 kg only.
An important application of titanium alloys in aircraft structures is as reinforcing
bands, strengthening the fuselage against fatigue cracks, that might propagate from
the inner towards the outer skin layer, with obviously potentially catastrophic
effects. The surface stability of titanium alloys, ensured by the self-passivating
titanium oxide surface layer, is interesting for piping applications, particularly for
corrosive fluids, like systems for air conditioning and de-icing. In this case, an
additional burden from temperatures up to 200
C must be considered. For these
applications, the equivalent steel tubes would be 40% heavier. The easily deformable
α + β Ti-3Al-2.5V alloy is a standard choice. Still corrosion resistance and weight
saving recommend the use of CP-titanium for the aircraft floors of the onboard
kitchens and toilets. The higher electrochemical compatibility of Ti-alloys, coupled
with higher stiffness and strength, and also a better match in the coefficient of
thermal expansion have recommended the use of these material as vertical and
horizontal reinforcing and stabilizer structures for empennage tail, made of CFRP
structural composites (Leyens and Peters 2003).
Landing Gear To meet volume constraints and, at the same time, the high-strength
requirements, titanium alloys have been extensively used for the landing gear
assembly of the Boeing 777 and 787 and Airbus A380. Boeing introduced on the
B777 the metastable-β Ti-10V-2Fe-3Al alloy (Boyer 2010). In the landing gear,
titanium alloy replaced steel (Fig. 4.39), with an estimated weight saving of 580 kg.
184 4 Titanium and Titanium Alloys

Fig. 4.39 Scheme of the main landing gear of the B 777 airplane. All underline-labeled parts are
potential candidates for being manufactured using forgings made of Ti-5553 alloy. (Redrawn from
Boyer 2010; Polmear 2006)

In the B787 a higher-strength alloy is used: Ti-5Al-5V-5Mo-3Cr. Using titanium

alloys in landing gear structure, in addition to weight saving, reduces the relevant
maintenance costs, mainly for an intrinsically better corrosion resistance.
Strictly related to the landing gear is the beam running between the wing and the
fuselage, which actually supports the landing gear itself. Usually this part, as most of
the airframe in commercial aircraft, is made of an aluminum alloy. In the Boeing
aircraft models, B737, B747, and B757, due to a significantly larger applied load, a
titanium alloy had necessarily to be employed. Indeed, the standard Al-alloy choice
would not be feasible, since the section of the beam would not fit within the envelope
of the wing. The alternative choice, i.e., a high-strength steel, would excessively
increase the total weight of the aircraft. For the window frames of the pilot cockpit of
commercial aircrafts, a forged titanium alloy is preferred to aluminum, suitable for
the other widow frames. The reason for this is the requirement to resist bird strikes,
an aspect that is not so critical in case of lateral fuselage surfaces.
Engine The front fan blades are other parts that have benefitted from the
availability of advanced titanium alloys, although this is not the only class of
materials that has been used for this part. Larger fan blades, up to 1 m long, required
by larger by-pass ratio engines, would exhibit serious flutter problems, due to the
shock waves generated by their tips when reaching a velocity close to that of sound.
Several approaches have been implemented. In the first place, with the intention
of increasing the structural stiffness of these components, mid-span shrouds
were used to connect one blade to the other. However, this solution, which
4.6 Main Aerospace Applications of Titanium Alloys 185

was certainly successful in solving the flutter vibration problem, presents some
adverse aerodynamic effects. Novel design criteria were developed and lead to the
so-called wide-chord fan blade design. These larger blades, lower in number, are
intrinsically stiffer. Therefore, they do not require any structural reinforcement and
guarantee a better engine efficiency. On the other hand, they would have been too
heavy if made of solid titanium alloy forgings, as it was in case of shrouded blades.
Lighter material systems were necessary and two solutions were actually proposed.
After the pioneering attempts by Rolls-Royce, which were abandoned for excessive
erosion problems, General Electric introduced in its workhorse engine, GE90, fan
blades made of carbon fiber reinforced composites with strengthened leading edges
made of titanium alloy. A full titanium alternative approach was adopted by Rolls-
Royce and Pratt & Whitney. The problem of the excessive weight of a solid blade
was overcome by adopting hollow components. Nowadays, titanium fan blades are
produced by superplastic forming and solid-state diffusion bonding of titanium alloy
sheet (Leyens and Peters 2003).
In Table 2.1 (Chap. 2), the parts of the compressor rotor of a commercial gas
turbine engine are listed. Titanium alloys are the best choice for the compressor
blades, as emerged since the early use of fan and compressor blades made of the
α + β IMI550 alloy in the Pratt & Whitney JT3D gas turbine engine (1960). This
resulted in remarkable improvements with respect to the same engine with steel
blades: the take-off thrust increased by 42%; the SFC and total weight of the engine
were reduced by 13% and 18%, respectively. The near-α alloys, which have been
mainly developed to improve creep resistance, in order to widen the application field
of titanium alloys, cannot overcome the 600
C temperature. Therefore, the hottest
parts of the compressor require the usage of steels and nickel-based superalloys. This
aspect has become increasingly important with the raise of the air compression ratio
reached in the last compressor stages in the state-of-the-art gas turbine engines
(Fig. 2.8). The development of stronger titanium alloys allowed also using these
materials for the compressor discs and shafts, thus reducing further the overall
weight of the engines.
Until recent, compressor was the only internal part of a gas turbine engine in
which titanium and its alloys could be safely used (see Table 2.1). In the latest
General Electric gas turbine engine, GEnX, developed for commercial airliners and
installed already on Boeing 747–8 and Boeing 787 aircrafts, the low-pressure turbine
blades have been made of a γ-TiAl-based alloy: Ti-48Al- 2Cr-2Nb (at%) that, as
discussed (Sect. 4.3.2), has adequate oxidation and creep resistance to be operated in
the demanding turbine conditions. In this engine, thanks to this material and to the
new design solutions, remarkable reductions in fuel consumption (20%), peak noise
(50%), and NOx emissions (80%) compared with prior engines have been achieved.
The use of a γ-TiAl alloy for low-pressure turbine blades would render this material
suitable also for the blading of the hot part of the compressor, afterburners, and
nozzle case in military engines, which at the present time are realized with steel and
superalloys. However, since it is still too expensive, the use of the Ti-aluminide
intermetallic alloy is so far restricted to the mentioned parts of the gas turbine engine
(Linger 2009). Interesting to note, which concerns Ti-aluminide-based alloys, the
186 4 Titanium and Titanium Alloys

very early applications involved alloys based on the α2-Ti3Al intermetallic (Sect.
4.3.2). However, although tested for components and parts of military aircraft
prototype engines, like afterburner sheets and low-pressure turbine blades, the
alloy never reached a sufficient level of technological development and reliability
to justify its employment in commercial aircraft. In particular, oxidation resistance
resulted still too poor to guarantee a sufficiently long lifetime.
A better success was achieved by replacing Inconel 625 with the metastable
β-alloy TIMETAL B21S for the plug and nozzle assembly of the Rolls-Royce Trent
400 engine, mounted on Airbus 340 and Boeing 777 aircrafts, resulting in more than
160 kg weight reduction. The positive results, obtained in this case, have encouraged
research efforts aiming at using titanium alloys also for the thrust reverser for the
Boeing 737MAX. The use of a titanium alloy is intended in this specific case, not
only to reach the expected weight reduction but also an implementation in the engine
design, namely, longer fan blades, without the need of a proportional increase in the
size of the engine nacelle. As reported for similar situations, the wider usage of
titanium determines a significant reduction in the maintenance costs (Cotton et al.
2015). New concept design, like an integrally bladed disc, conventionally called
“blisk,” has contributed to push further the benefits associated with the use of lighter
alloys for compressor components. The “blisk” is a single part in which disc and
blades are bonded together by linear friction welding. In this way, the typical
nucleation sites for fatigue fracture are eliminated, and, thereby, the relevant inspec-
tion intervals are extended. Moreover, friction welding can be used for replacing
individual blades, if damaged during service (Leyens and Peters 2003).
Springs The elastic moduli of titanium and α + β Ti-alloys fall in the 100–110 GPa
range that is definitely smaller than the elastic modulus of stainless steel (210 GPa).
Therefore, one of the more interesting areas in which titanium alloys may replace
steels is in spring applications. Indeed, considering the lower value of the Young’s
modulus, only half of the coils of the same size are needed to replace an equivalent
steel spring. Moreover, the lower density of titanium, 60% approximately the density
of the steel, results in up to 70% weight reduction for these parts. Titanium aircraft
springs can be made of titanium alloys, and β-alloys, like the β-metastable Ti-15-3
(Ti-15V-3Cr-3Sn-3Al, Table 4.3) and Beta-C (Ti-3Al-8V-6Cr-4Mo-4Zr, Table 4.3),
are particularly recommended. Interestingly, the low cost beta (LCB) alloy (meta-
stable-β, with composition Ti-4.5Fe-6.8Mo-1.5Al, see Table 4.4) had been origi-
nally developed for car springs. Although this application was not feasible, since the
material and production costs were still too high for the automotive field, this
research made an alloy available to aerospace spring manufacturing, having half
the price of another all-around alloy, the Beta-C (see above), and exhibiting very
much the same mechanical properties.
Accessories Cargo handling systems, crew microphone holder (Boeing 737 and
757), access door clips, interior window clips, and other similar precision forged
parts are all made of titanium alloys in commercial airliners starting from the early
1990s (Boyer et al. 1994). Structural parts, like brake torque tubes of aircrafts
(military fighter and commercial), can be made of castings of metastable β-alloys,
like the Ti-15-3 alloy.
References 187

Containers and Tanks A peculiar application of titanium alloys in aerospace

structures is for high-pressure containers, ranging from fire extinguisher bottles to
pressure tanks for space transportation systems (Boyer et al. 1994). In both cases, the
very well-predictable behavior of titanium alloys and their low density are para-
mount selection criteria. A further plus of titanium alloys for aerospace fuel tanks is
the possibility to be nondestructively tested before being launched with a level of
statistical reliability not achievable with polymer matrix structural composites. This
is an aspect which is sometimes overlooked, even though it is of absolutely impor-
tant for these applications, in which any kind of critical defects may jeopardize the
success of the entire mission.
Helicopters The high strength and fatigue resistance afforded by β-Ti-alloys have
been very important for improving the performance of top-class helicopters (Boyer
et al. 1994). From this point of view, the progressive increase in the all-up weight
(AUW) of the Westland Lynx helicopter is an interesting case. The semirigid rotor
head, the hub disc, the sleeve, and the mast were originally made of the α + β
classical Ti-6-4 alloy, which was no longer adequate to support an increase in the
AUW from 38.6 tons up to the 56.0 tons. The Ti-6-4 alloy was replaced by a
metastable β-alloy: Ti-10-2-3 (Ti-10V-2Fe-3Al, see Table 4.4) that exhibited
enhanced high-cycle fatigue, lower stiffness, and superior strength. Other helicopter
companies adopted similar alloys and technological development process for their
products.

References

ASM International (1991) ASM Handbook Vol. 2 – Properties and Selection: Nonferrous Alloys
and Special-Purpose Materials. ASM International
ASM International (1992) ASM Handbook Vol. 17 – Nondestructive Evaluation and Quality
Control
ASM International (1992) ASM Handbook Vol. 3 – Alloy phase diagrams. ASM International
Bania P J (1994) Beta Titanium Alloys and Their Role in the Titanium Industry. Journal of
Materials 46 (7): 16–19
Bewlay BP et al (2016) TiAl Alloys in Commercial Aircraft Engines. Materials at High Temper-
atures 33 (4–5): 549–559
Boyer R (1994) Aerospace Applications of Beta Titanium Alloys. Journal of Materials 46(7): 20–23
Boyer R (2010) Application of Titanium and Its Alloys. Journal of Materials 62(5): 21–24
Boyer R, Collings E W, Welsch G (1994) Materials Properties Handbook: Titanium Alloys. ASM
International
Cain K J (2016) Industrial Titanium Demand Forecast 2016. In: Titanium USA 2016, the 32nd
annual conference and exhibition produced by the International Titanium Association (ITA),
J. W. Marriot Desert Ridge Resort, Scottsdale, 25–28 September 2016
Chen G et al (2016) Polysynthetic twinned TiAl Single Crystals for high-temperature Applications.
Nature Materials 15: 876–881
Chen W et al (2014) Development of Ti2AlNb Alloys: Opportunities and Challenges. Advanced
Materials & Processes 172(5): 23–27
Cotton J et al (2015) State of the Art in Beta Titanium Alloys for Airframe Applications. Journal of
Materials 67 (6):1281–1303
188 4 Titanium and Titanium Alloys

Dargusch M S, Keay S M (2009) Meltless Ti-Al New Light Metals Industry. In: Dargusch M S,
Keay S M (eds) Light Metals Technology, Trans Tech Publications Ltd, p 135–138
Das S K, Davis L A (1988) High Performance Aerospace Alloys via Rapid Solidification
Processing. Materials Science and Engineering 98: 1–12
Dimiduk D M et al (1992) Development of Intermetallic materials for Aerospace Systems.
Materials Science and Technology 8(4): 367–375
Ding X F et al. (2011) Microstructure evolution of directionally solidified Ti-45Al-8.5Nb-(W, B, Y)
alloys. Journal of Alloys and Compounds 509(9): 4041–4046
Donachie M J (2000) Titanium – A Technical Guide, 2nd edn. ASM International
Eylon D et al (1994) Issues in the Development of Beta Titanium Alloys. Journal of Materials
46 (7): 14–15
Froes F H (2015) Titanium: Physical Metallurgy, Processing and Applications. ASM International,
Materials Park, Ohio
Froes F H, Imam A (2018) Titanium: A Historic and Current Perspective-Part I. Advanced
Materials and Processes 176: 19–25
Hashimoto K et al. (1998) Alloy design of gamma titanium aluminides based on phase diagrams.
Intermetallics 6(7–8): 667–672
Hellier C (2001) Handbook of Nondestructive Evaluation. McGraw-Hill
Kutz M (2002) Handbook of Materials Selection. John Wiley and Sons
Lapin J (2006) Creep behavior of a cast TiAl-based alloy for industrial applications. Intermetallics
14(2): 115–122
Lapin J, Ondrúš L, Nazmy M (2002) Directional Solidification of Intermetallic Ti-46Al-2W-0.5Si
Alloy in Alumina Moulds. Intermetallics 10 (10): 1019–1031
Leach W (2016) Titanium Demand and Trends in the Aero Engine Market. In: Titanium USA 2016,
the 32nd annual conference and exhibition produced by the International Titanium Association
(ITA), J. W. Marriot Desert Ridge Resort, Scottsdale, 25–28 September 2016
Leyens C, Peters M (2003) Titanium and Titanium Alloys – Fundamentals and Applications. Wiley
VHC Verlag, Weinheim
Linger D (2009) Titanium Utilization & Vision for The Next Decade: An Aircraft Engine OEM
Perspective. In: Titanium 2009, the 25th annual conference and exhibition produced by the
International Titanium Association (ITA), Hilton Waikoloa Village, Hawaii, 13–16 September
2009
Liu B G et al. (2017) Structural stability and the alloying effect of TiB polymorphs in TiAl alloys.
Intermatallics 90: 97–102
Lütjering G, Williams J C (2007) Titanium, 2nd edn. Springer Berlin Heidelberg, Berlin
Massalski T B et al (1986) Binary Alloys Phase Diagrams Vol. 2. ASM International
McCafferty E (2010) Introduction to Corrosion Science. Springer Verlag
Meetham G W (1981) The Development of Gas Turbine Materials. Applied Science Publishers Ltd,
London
Mix P E (2005) Introduction to Nondestructive Testing: A Training Guide. John Wiley & Sons
Moiseyev V N (2006) Titanium Alloys-Russian Aircraft and Aerospace Applications. Taylor &
Francis
Nieh T G et al. (1994) Superplasticity in Metals and Ceramics. Cambridge University Press
Partridge A, Shelton E F J (2001) Processing and Mechanical Property Studies of Orthorhombic
Titanium-Aluminide-based Alloys. Air & Space Europe 3 (3–4): 170–173
Pérez-Prado M T, Kassner ME (2009) Superplasticity. In: Kassner M E (ed) Fundamentals of Creep
in Metals and Alloys, 2nd edn. Elsevier, p 137–149
Peters J M, Blank-Berwersdorff M (1992) Titanium Aluminide Foil. Materials and Design 13 (2):
83–92
Polmear I J (2006) Light Alloys – From Traditional Alloys to Nanocrystals, 4th edn. Butterworth-
Heinemann
Prasad E N, Wanhill R (2017) Aerospace Materials and Material Technologies Volume 1: Aero-
space Materials. Springer
References 189

Proske G et al. (1992) The Microstructure and Mechanical Properties of the Intermetallic Compund
Super Alpha 2. Materials Science and Engineering A 152: 310–316
Sastry S M L et al (1983) Rapid Solidification Processing of Titanium Alloys. JOM 35 (9): 21–28
Sears J W (1990) Current Processes for the Cold-Wall Melting of Titanium. Journal of Minerals,
Metals and Materials Society 42 (3):17–21
Suryanarayana C et al. (1991) Rapid Solidification Processing of Titanium Alloys. International
Materials Review 36 (1): 85–123
Suryanarayana C, Froes F H (1990) The Current Status of Titanium Rapid Solidification. Journal of
Materials 42 (3): 22–25
Wang L et al. (2017) Influence of Alloy Composition and Thermal History on Carbide Precipitation
in γ-based TiAl Alloys. Intermetallics 89:32–39
Zhao Z B et al. (2017) Effect of heat treatment on the crystallographic orientation evolution of a
near-α titanium alloy Ti60. Acta Materialia 131: 305–314

Further Reading

Boyer R R (1996) An overview on the use of titanium in the aerospace industry. Materials Science
and Engineering: A 213 (1–2):103–114
Gillo G (2011) Superplastic Forming of Advanced Metallic Materials – Methods and Applications.
Woodhead Publishing Ltd
Joshi V A (2006) Titanium Alloys – An Atlas of Structures and Fracture Features. Taylor & Francis
Moore H D (1981) Materials and Processes for NDT Technology. The American Society for
Nondestructive Testing
Seong S et al. (2009) Titanium – Industrial Base, Price Trends and Technology Initiatives. RAND
Corp., CA (USA)
Chapter 5
Steels

5.1 Introduction

A very simple definition for steels is “iron-carbon-based alloys”. In the common

language, iron and steel are sometimes used to describe the same material, and this
is mainly due to the fact that people tend to associate each metallic object to the most
notorious and better known metal, i.e., iron, a chemical element with the atomic
number equal to 26.
The name directly comes from the Latin word ferrum, from which the Italian term
( ferro) derives. Concerning the English term, iron derives from the Scandinavian
term iarn. On the Earth, the most abundant sources of this metal are the Fe3O4
(magnetite) and Fe2O3 (hematite)-containing ores. Pure iron is only available in
meteoritic and particular native form. Native iron, also known as telluric, can be
found in basaltic rocks, as small grain inclusions, with a maximum Ni content of 3 wt
%, and resulting from a thermal reduction occurring within the silicic matrix. Nickel
is also present in meteoritic iron, in which higher concentrations of this metal have
been measured, ranging from 4 to 26 wt% (Buchwald and Mosdal 1985). Although
rather scarce, it seems that the first iron that mankind actually used was the meteoritic
type. It was regarded as a magic “metal from the stars,” and this marked the
vocabulary of iron-related activities. Iron metallurgy in Latin languages, like the
French sidérurgie, is named after sidus-sideris, i.e., star. Examples of both meteoric-
and telluric-based iron objects have been found in Greenland. The main source of
meteoric iron in Greenland was the Cape York meteorite that collided with the Earth
approximately 10,000 years ago, with fragments falling out over a wide area (see
Fig. 5.1). The early reported examples of iron production from ores date back to
2000 BC in India, and by 1200 BC, iron production processes were widespread both
in China and in the Near East. This was the time when iron metallurgy appeared also
in Mediterranean area, and it is taken as the starting time for the Iron Age (Ashby
2005; Tylecote 2002). Nowadays, the reference material has become steel, named
after the Sanskrit word stakati, (Stahl in German, stalin in Russian), identifying a

© Springer Nature Switzerland AG 2020 191

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_5
192 5 Steels

Fig. 5.1 Part of the Cape

York meteorite at the
American Museum of
Natural History in
New York (USA). The
polished part from which the
Widmanstätten structure can
be clearly seen is several
centimeters in width. (Photo
property of the authors)

class of iron alloys featuring carbon as component. Carbon content can strongly vary
depending on the steel grade. The lowest concentrations are those of the low-carbon
steels, containing no more than 0.06 wt% C.
Until relatively recent times, in the Western metallurgy, the production of steel
was mainly the result of the skills of blacksmiths, heating bloomery iron in charcoal
fires. Blooming was an important stage, as concerns both the alloy refinement and
the optimization of the carbon content. The situation was subsequently ameliorated
thanks to several improvements. Pushed by the shortage of wood and legislation
restrictions, the process to eliminate sulfur from the fossil coal was developed,
resulting in the production of coke coal, by the end of the seventeenth century AD
in England. The coke coal boosted any metallurgical production, with particular
regard for iron metallurgy, since the mechanically stronger fuel afforded the possi-
bility to realize larger furnaces and raw material stacks. The process lead in the end
to blast furnaces, and cast iron was the main product of these plants. Thereafter, the
production cycle required the treatment of cast iron to reduce the excessive carbon
content. This was achieved thanks to the development of the puddling furnaces first
and then of the Bessemer converter, unanimously recognized as the beginning of the
modern steelmaking in 1856 and of the so-called open-heart process shortly after-
ward (Hosford 2012). Steel is nowadays present in almost every technological field,
including aerospace industry, of which some relevant aspects are considered in this
chapter.
5.2 Iron and Steel Metallurgical Base Concepts 193

5.2 Iron and Steel Metallurgical Base Concepts

Steels are characterized by a variety of properties and microstructures, which can be

achieved through solid-state transformations and processing. Like titanium (see
Chap. 4), iron shows allotropic forms, three in this element, stable at increasing
temperatures: α (bcc), γ ( fcc), and δ (bcc). Iron exhibits also a magnetic transition,
occurring at 768
C (the Curie temperature) above which it becomes paramagnetic,
still retaining a bcc structure. This form is called β-iron. At atmospheric pressure, the
α polymorph (bcc) transforms into the close packed fcc γ-iron above 911
C. This
phase is stable up to 1392
C, above which iron turns back to the bcc structure of
δ-iron. This phase is stable up to the melting temperature (1538
C). The stability of
the different polymorphs can be justified in terms of the distribution of the density of
electronic levels, with some complication coming from the partially filled d-bands
and relevant magnetic properties (Ashcroft and Mermin 1988). For instance, if
nonmagnetic iron was stable at room temperature, it would have had an hcp instead
of the bcc structure (Pepperhoff and Acet 2001).
The polymorphism of iron is maintained in steels, with stability fields now
dependent on temperature and composition too. This can be clearly seen from the
steel prototype phase diagram, i.e., the Fe-C binary phase diagram in Fig. 5.2.
The combination of alloying elements, including undesired impurities, that might
be present in lower-grade alloys and heat treatments lead to a variety of microstruc-
tures and, thereby, properties. Some alloying elements will dissolve into the iron
matrix; others will form precipitates and secondary phases.
Concerning steels, there are two main criteria for the classification of alloying
elements. The first is based on the relevant role in the stabilization of austenite; the
second regards the tendency for the alloying element to form carbides. It is worth
noting that carbide formation needs the concentration of the alloying element to
overcome a threshold value, having to do with its solubility in iron. According to
these criteria, in Appendix 4 the most important alloying elements and their effects
on steel properties are given. The microstructure of steel products resulting from any
composition can be modified by specific heat and thermomechanical treatments, as
described in full details in the cited literature at the end of the chapter.

5.2.1 Annealing

Annealing treatments are conducted after either high-temperature (casting,

hot-rolling, forging) or cold-forming processes. The main effects induced by the
annealing regard hardness reduction, grain refinement, increased microstructural
stability, and better machinability. The driving force for these changes is the energy
stored into the material as a consequence of previous processing steps, like casting
(internal solidification stresses) and deformation processes, dealing with wrought
products. The most important categories of annealing treatments are summarized in
Table 5.1.
194 5 Steels

Fig. 5.2 Fe-C phase diagram, iron rich corner. (Redrawn from Callister and Rethwisch 2012;
Hosford 2012)

5.2.2 Quenching

Quenching involves a sufficiently fast cooling of a steel from its austenitizing

temperature field, capable to induce the formation of a non-equilibrium structure,
called martensite. The martensite formation occurs via an athermal transformation
that starts once the alloy has been taken below the so-called martensite start, Ms,
temperature, and it is accomplished below the martensite finish temperature, Mf. Due
to its athermal character, the transformation requires a continuous cooling to pro-
ceed. Since the rate of change propagates at the speed of sound, the transformation
would typically take place in a relatively short time, so that no carbon diffusion
throughout the crystal lattice or any other compositional change may occur. Carbon
remains trapped into the crystalline lattice, which is fcc in the initial state (austenite),
and transforms into body-centered tetragonal (bct), owing to the deformation asso-
ciated with the excessive concentration of carbon into the iron lattice. The cooling
rate necessary to attain the martensite varies according to the chemical composition
of the steel. Optimal values can be determined from the continuous cooling trans-
formation (CCT) and time-temperature-transformation (TTT) diagrams (Totten
2007). An excessively high cooling rate will cause tension cracks in the material.
5.2 Iron and Steel Metallurgical Base Concepts 195

Table 5.1 A summary of the most important annealing treatments for steels (Campbell 2008;
Totten 2007)
Name Aim Description
Diffusion Elimination of chemical inhomogenei- Diffusion annealing is carried out in
annealing ties in alloy composition, in particular the austenite stability region between
along grain boundaries, to reduce 1100 and 1300
C, to reduce inho-
liquation phenomena mogeneity in alloying elements dis-
tribution that may result in liquation.
Therefore, microhardness gradients
are eliminated as well. Finally, the
overall hardness of the steel decreases
with respect to the initial value
Normalization Grain size refinement, better micro- It is carried out at a temperature which
structure, improved machinability, and is not much higher than the ferrite-to-
enhanced response to hardening pro- austenite transformation temperature,
cesses. Normalization is generally car- this for avoiding grain coarsening
ried out for producing a uniform phenomena. Normalization tempera-
structure made of ferrite and pearlite ture usually lays 20–50
C above the
ferrite-to-austenite transformation
temperature. Cooling is usually
performed in air. The higher the
cooling rate, the finer the final
microstructure
Softening Production of a globular structure, Softening is performed by heating the
known as pearlite, in steels with more materials up to a temperature slightly
than 0.5% C. Production of a uniform above its eutectoid temperature and
fine microstructure with finely dis- holding the condition for a reasonable
persed carbides after quenching long time. Cooling after softening can
process be performed in air as well but only
when the temperature has been
decreased below 600
C
Solution Dissolution of particular phases pre- The alloy is heated above the solvus
annealing cipitated during high-temperature pro- temperature of the secondary phases
cesses, such as hot-rolling. Since (either carbides, or intermetallics) and
solution annealing is the most used cooled down sufficiently fast, in order
heat treatment for austenitic stainless to avoid re-precipitation. Stainless
steels, phases that have to be dissolved steels like AISI 303, 304, and 304L
are the Cr-rich M23C6 carbides grades are usually solution annealed
(M ¼ Cr, Fe, Mo) or the Mo- and in a temperature range in between
Cr-rich deleterious σ phase 1010 and 1120
C. Mo-containing
stainless steels such as AISI
316, 316L, 317, and 317L are treated
between 1040 and 1120
C. For
Ti-stabilized austenitic stainless steels
such as AISI 321 grade, the tempera-
ture range is between 955 and
1065
C. Complete dissolution of
carbides has to be carried out, and this
usually takes considerable time to
happen. As a general rule, the higher
the treatment temperature, the faster
carbide dissolution takes place.
(continued)
196 5 Steels

Table 5.1 (continued)

Name Aim Description
Duration of the treatment depends on
the component size and on the furnace
type. The maximum annealing tem-
perature and duration are limited by
the need to avoid abnormal grain
growth. Concerning the cooling
media, the type used is related to the
desired cooling rate. For instance,
water is a stronger cooling medium if
compared to oil, and oil is stronger
than air. Most common cooling
medium is water; however oil and
solutions containing polymers are also
used
Spheroidizing Production of a spheroidized (globular) In order to obtain spheroidizing, the
structure with maximum ductility and steel has to be heated at a temperature
minimum hardness just below its ferrite-to-austenite
transformation for a sufficiently long
time. Another possibility is to perform
spheroidizing by cyclic oscillation
above and below the ferrite-to-aus-
tenite transformation temperature. The
treatment produces small cementite
(Fe3C) spheroids in a ferrite matrix. It
is often performed after a normaliza-
tion annealing
Stabilization Maximization of intergranular corro- The material is heated up to the tem-
annealing sion resistance of stabilized austenitic perature range 845–955
C in order to
stainless steels promote the formation of the thermo-
dynamically stable MC carbides, in
place of the less stable M23C6 carbide.
MC carbide precipitation temperature
is higher than that of the Cr-rich
M23C6. Furthermore, M23C6 carbide
formation leads to Cr depletion in the
stabilized austenitic stainless steels.
This phenomenon is a particular type
of intergranular corrosion named sen-
sitization. The formation of MC rather
than M23C6 leads to a lower decrease
in Cr content of the bulk and therefore
a better corrosion resistance, particu-
larly along grain boundaries
Stress relief Elimination of macroscopic stresses in The process temperature strongly
annealing the material due to cold-working depends on the preliminary deforma-
tion the material has been subjected
to. The higher the degree of deforma-
tion, the lower is the process temper-
ature necessary for stress relief. Two
(continued)
5.2 Iron and Steel Metallurgical Base Concepts 197

Table 5.1 (continued)

Name Aim Description
different phenomena lead to stress
relief: recovery and recrystallization.
The material is slowly cooled down
from the annealing temperature (typi-
cally in the 425–550
C range for
austenitic stainless steels) in order to
reduce stresses. However, this can
cause sensitization of the steel, i.e.,
M23C6 carbide precipitation, with
the consequent removal of chromium
from the alloy and consequent wors-
ening of the corrosion resistance. Fast
cooling causes residual stresses lead-
ing to a structure which is more prone
to stress corrosion cracking. However,
it is generally preferable to have a
small sensitization rather than cata-
strophic failures caused by stress cor-
rosion cracking (SCC – see Sect.
8.1.5) that can indeed be reduced by
the present treatment

On the other hand, a too low cooling rate will not allow martensite to form at all or
just to a limited extent. There are two main types of quenching processes: normal and
thermochemical quenching. The first one involves the treatment of the bulk volume
and the second the surface. This means that using processes, like surface carburizing
for low-carbon steels, an external layer only is transformed into martensite. In this
way, the surface properties, e.g., hardness, would be improved, without excessively
embrittling the bulk. Since martensite has a high hardness and a low toughness,
quenching is usually followed by tempering treatments (see Sect. 5.2.3). After
quenching, retained austenite can be observed, i.e., a residual fraction of the initial
austenite phase that does not undergo any transformation. If required, the retention
of austenite can be avoided by means of a cryogenic, or subzero temperature,
treatment (Totten 2007). Concerning steels that will be discussed in this chapter,
martensitic transformation, with material-specific differences, regards maraging
steels (Sect. 5.4.1), AerMet steels (Sect. 5.4.2), martensitic stainless steels (Sect.
5.4.3.2), and martensitic PH stainless steels (Sect. 5.4.3.3).

5.2.3 Tempering

Tempering treatments are meant to optimize the properties of the metastable mar-
tensite, as concerns particularly its low toughness. Depending on the material
composition, part size, and residual stress distribution, tempering can be performed
over either low (e.g., 100–200
C) or relatively higher (e.g., 450–700
C)
198 5 Steels

temperature intervals, however, still well below the ferrite-to-austenite transition

temperature. The tempering process can be split into four main stages (ASM
International 1992; Totten 2007):
• Stage I (T < 250
C). At this stage, a reduction of the carbon content in the
martensite takes place with the formation of ε carbide (Fe2.4C). The martensite is
still a supersaturated solid solution and may undergo further transformations at
higher temperatures.
• Stage II (200
C < T < 300
C). During the II stage, retained austenite decom-
poses into ferrite and cementite.
• Stage III (250
C < T < 350
C). The third stage of tempering implies the
replacement of transition carbides, like the ε carbide, and low-temperature mar-
tensite substitution by ferrite and cementite. Cementite precipitates heteroge-
neously, as needles and platelets at the inter-lath boundaries (Campbell 2008).
The formation of cementite results in a further reduction of the carbon level in the
martensite, and, thereby, the bct structure will lose its tetragonal symmetry.
• Stage IV (350
C < T < 700
C). This stage is typical of highly alloyed steels.
During the fourth stage of tempering, further equilibrium carbides, like M7C3 and
M23C6, precipitate, resulting in the so-called secondary hardening.
A detrimental phenomenon associated with tempering is temper embrittlement,
involving a rise in the ductile-to-brittle transition temperature and, consequently, a
reduction in the low-temperature fracture toughness. Temper embrittlement is asso-
ciated with two phenomena, depending on the temperature range: tempered mar-
tensite (irreversible) embrittlement and the temper embrittlement. Tempered
martensite embrittlement typically occurs over the 230
C–400
C temperature
range, corresponding to II and III tempering stages. It is observed in plain carbon
steels containing phosphorous, and low-alloy steels subjected to heat treatments to
reach high-strength levels. In steel materials relatively free from nitrogen and
phosphorous, a decrease in toughness is expected due to the formation of cementite,
which is a preferential path for fracture propagation. In case the steel contains a high
amount of phosphorous (e.g., 0.03 wt% or more), the embrittlement results in an
intergranular failure along initial austenite grain boundaries, for the decrease in their
cohesive strength (Campbell 2008; Totten 2007).
Temper embrittlement occurs at higher temperatures (450
C–600
C). It is also
known as two-step embrittlement, since it is observed either on slow cooling through
the critical temperature range or upon aging the alloy in the same temperature
interval. Since a heating cycle above 600
C followed by rapid cooling can eliminate
temper embrittlement, differently from tempered martensite embrittlement, this
phenomenon is reversible and is characterized by an intergranular fracture morphol-
ogy. Impurities, like As, P, Sb, and Sn, play an important role in this phenomenol-
ogy, since affecting the ductile-to-brittle transition temperature of the alloy. The
effect of these elements can be enhanced further by the presence of Cr, Mn, Ni, and
Si (Campbell 2008; Totten 2007).
5.2 Iron and Steel Metallurgical Base Concepts 199

Tungsten and molybdenum (see Appendix 4) slow down temper embrittlement,

by reducing the alloy diffusivity. Temper embrittlement can also be prevented if the
critical temperature interval is crossed on cooling at a sufficiently fast rate (Campbell
2008; Totten 2007).
Strictly related to the above treatments is austempering, not involving modifica-
tions of the martensite but still interesting in several respects. This thermal treatment,
developed by Eduard Devenport and Edgar Bain (US Patent 1,924,099), can be
performed on both cast irons (ADI, austempered ductile irons) and steels and involves
an isothermal transformation of the alloy at a temperature above its martensite start
temperature (Ms) but below that of pearlite transformation. The alloy is cooled down to
this temperature interval after homogenization in the austenitic field. The most used
cooling medium is a molten salt bath, and the main improvements introduced by this
treatment are increased ductility, strength, and toughness; reduction in the distortions
associated with heat treatment; and possibility to rapidly obtain hardness values in the
35–55 HRC range (ASM International 2016).

5.2.4 Age Hardening

Age (or precipitation) hardening treatments produce an increase of the steel mechan-
ical properties due to the solid-state precipitation of secondary phases. Age-hardened
materials are of particular interest for aerospace applications, and precipitation
treatments are interesting not only as concerns steels but also aluminum and mag-
nesium alloys (see Chap. 3), titanium alloys (see Chap. 4), iron- and nickel-based
superalloys, and also the latest generation of cobalt-based superalloys (see Sect.
6.2.2). The main steel families that exhibit precipitation hardening are AerMet,
maraging, and PH (PH ¼ precipitation hardening or precipitation hardenable) steels.
Details on age-hardening treatments of maraging, AerMet, and PH stainless steels
can be found in Sects. 5.4.1, 5.4.2, and 5.4.3.3, respectively.

5.2.5 Steel Designation

The most common steel designation system is based on their chemical composition,
although other classifications do exist, based on different intrinsic and technological
aspects:
1. Crystallographic structure, e.g., ferritic, martensitic, austenitic, duplex, bainitic,
and pearlitic
2. Heat treatment condition and thermomechanical processing, like air cooled,
solution annealed, etc.
3. Manufacturing methods: vacuum arc remelting, electric slag remelting, etc.
4. Main application fields: stainless steels, tool steels, structural steels, etc.
200 5 Steels

5. Oxidation protocol used for their production: killed and semi-killed

6. Finishing method: hot-rolled, cold-rolled, and continuously cast
7. Product shapes: bars, coils, plates, and sheets
8. Required strength level according to ASTM standards
9. Quality indications and classification: commercial quality, forging quality, etc.
A detailed description of each classification system can be found in specific
references (ASM International 1993; Totten 2007). In the USA, the AISI (American
Iron and Steel Institute) system is the most common, together with UNS (unified
numbering system), that is widely accepted also in the rest of North America. In
Europe, the EN (Euronorm) and the DIN (Deutsche Institut für Normung) standard,
including the Werkstoff number (WN), are common designation systems. For
instance, the austenitic stainless steel AISI 316 is the 1.4401 (WN) and the
X5CrNiMo17-12-2 according to the EN designation system (ASM Interna-
tional 1993; Totten 2007).

5.3 Standard and Special Production Techniques

This section is dedicated to a brief overview on the principal steel production

techniques. The aim is to provide a general information on the production and
processing of particular steel grades, especially those employed in the aerospace
field that will be discussed in Sects. 5.4 and 5.5. Steel metallurgy is either primary or
secondary, depending on the feedstock type used in the first melting process. Iron-
rich ores are used as feedstock for extractive metallurgy, using a blast furnace.
Alternatively, steel scraps can be used as feedstock in the alternative metallurgy
processing, for which the electric arc furnace (EAF) is the most common production
plant. The output of an EAF melting may be an alloy whose chemical composition
differs from that of the final product. This is the case of highly alloyed steels,
that need a further refining step to set the correct chemical composition. They are
usually characterized by a carbon content in the 1–2 wt% range, lower than a typical
EAF product. Therefore, to remove the excess carbon, an oxidation is required.
Other refining approaches can be employed as concerns the tuning of the concen-
tration of other alloying elements, like Al, Cr, Mo, Ti, etc. The final chemical
composition of the products and the desired cleanliness level are achieved through
refining processes (e.g., argon oxygen decarburization, AOD), before casting and
subsequent solidification, following a standard production route. Plain carbon steels
and low-alloyed grades can be attained directly from the EAF thanks to a careful
selection of the scraps and to specific practices during the process. Therefore, the
composition of the scraps charge must be calibrated according to the steel grade that
has to be produced and to the melting and refining steps that are needed. Concerning
the steel grades that are interesting to the aerospace industry, the main processes
discussed in this section are:
• Arc melting
• Vacuum induction melting
5.3 Standard and Special Production Techniques 201

• Argon oxygen decarburization, vacuum oxygen decarburization (VOD and

VODC)
• Electroslag remelting
• Vacuum arc remelting
These same processes are also used in the production of superalloys and, as
concerns vacuum processes, in the production of titanium alloys (see Chaps. 6 and 4,
respectively).

5.3.1 Arc Melting

The EAF is one of the most important plants in foundries and steel production
factories. It was first introduced at the beginning of the twentieth century, and
nowadays it is worldwide used by more than 80% of steel producers. The rest will
use induction furnaces (around 17%) and other melting practices (ASM International
2008; Schifo and Radia 2004; Toulouevski and Zinurov 2013).
As shown in Fig. 5.3, the EAF is composed of three main parts: a vessel, a roof, and
the electrodes. The steel vessel has an internal refractory layer either made of bricks or
of monolithic material. The roof is movable, in order to allow feeding the scrap charge.
Its inner surface is also protected by a refractory layer. The EAFs used for steel
production are in general alternating current furnaces, featuring three graphite-based
electrodes. Direct current furnaces have only one electrode. The electrodes are
mounted vertically through the roof of the furnace and float above the solid steel
scrap, or the molten bath, depending on the specific step of the production cycle and on
the distance between the electrodes and the charge/melt, which varies during the
melting process. The temperature is increased until charge melting is achieved thanks
to the energy transfer through the electric arc sparking between the charge and the
electrodes. Usually, due to the high volume of the scraps, the whole charge is added to
the furnace in several steps. For a 50t EAF, there are usually three charging steps. After
the first portion is introduced into the furnace chamber, the roof is repositioned onto
the vessel, and the electrodes are lowered into the charge. As melting initiates, the
voltage is gradually increased and the process continues until the charge is totally
molten. At this stage, since the volume occupied by the initially solid, now molten
scraps is lower, it is possible to add a second batch of metallic scraps. This process is
repeated until the nominal volume capacity of the furnace is reached. In addition to the
metal scraps, it is possible to introduce into the furnace other materials, such as DRI
(direct reduced iron) and substances, like mixtures of oxides and fluorine compounds,
such as CaF2, which promote the formation of the slag and optimize its viscosity. The
slag is an important by-product of the melting process, playing an important role in
embedding impurities and undesired substances present in the melt. Since the slag,
mainly consisting of metal oxides, has a lower density than liquid steel, it can float on
the liquid alloy surface. The chemical equilibrium between the slag and the liquid alloy
is an important parameter, influencing the partition between liquid and slag of several
elements (Turkdogan 2010). After the complete melting and homogenization, the
202 5 Steels

Fig. 5.3 Schematic of an electric arc furnace with traditional architecture and detailed components
list. (Redrawn from ASM International 2008; Donachie and Donachie 2002): (1) transformer; (2)
flexible cable connection; (3) electrode arms; (4) electrodes and electrodes clamping system; (5)
arms; (6) cooled off-gas duct; (7) cooled panes (furnace shell); (8) structure; (9) basculating
structure; (10) rack; (11) cooled roof; (12) basculating device; (13) hydraulic group (Madias
2013, reproduced with permission of the Publisher)

liquid steel is poured into a ladle and sent to the next production step, e.g., a conversion
process. The slag is removed, either completely or partially, thus allowing its refor-
mation and promoting further reactions to occur during the conversion process (see
Sects. 5.3.3 and 5.3.4).
5.3 Standard and Special Production Techniques 203

5.3.2 Vacuum Induction Melting (VIM)

The development of the vacuum induction melting (VIM) process started around
the mid-nineteenth century, and its industrial use started in the early 1950s. The
driving force for the development of VIM was the need of producing materials with a
particular compositional cleanliness and controlled mechanical properties. One
example is the production of steels for rolling bearings. It has been observed that
the improvement in the steelmaking process, using VIM, results in significantly
better performances of the material, particularly as concerns contact fatigue life
(Meetham 1981). Compared to traditional air-melting processes (e.g., EAF plus
AOD, see below), VIM ensures a considerable reduction in both nitrogen and
oxygen content in the final products. Also other impurity elements, like Sb and
Pb, can be eliminated through their evaporation during melting, favored by the
low-pressure environment. As compared to EAF + AOD, VIM affords a better
control of the alloying element content and thereby improved stress rupture and
fatigue properties.
Nonetheless, some drawbacks are there in vacuum induction melting too:
• The composition of the molten charge is essentially determined and fixed by the
choice of the charge materials, and no compositional changes, such as C, Si, S
reductions, can be implemented.
• The process is definitely more expensive as compared to EAF + AOD.
• The compositional control of elements with high vapor pressure, like manganese,
might be critical.
Figure 5.4 shows the general schematic of a VIM furnace. It consists of a melting
vessel inside a steel shell kept under vacuum using a high pumping rate vacuum
system. The heart of the furnace is the crucible, and the induction heating is obtained
by a magnetic field that induces a current into the metal charge (ASM International
2008). The induction coils, wrapped around the melting vessel, are made of water
cooled copper tubing.
One important process parameter is the induction coil frequency, to be varied
according to the composition and the mass of the alloy charge. For aerospace
applications, VIM is often a first melting step, followed by further remelting stages,
such as ESR and VAR, as illustrated by the scheme in Fig. 5.5.

5.3.3 Argon Oxygen Decarburization (AOD)

Argon oxygen decarburization (AOD) is a refining process that was originally

developed for the production of stainless steel and successively extended to other
metal alloys:
• Carbon steels.
• Superalloys (see Chap. 6).
204 5 Steels

Fig. 5.4 Schematic of a VIM crucible and of a top-opening, double-chamber vacuum induction
melting furnace. (Redrawn from ASM International 2008)

Fig. 5.5 Possible processing routes used for VIM casting products (ASM International 2008)
5.3 Standard and Special Production Techniques 205

Fig. 5.6 AOD converter with the oxygen top blowing (OTB) system. (Redrawn from The AISE
Foundation 1998)

• Refractory alloys (see Sect. 9.2).

• HSLA (high-strength low-alloy) steels.
• Low-alloy steels.
• Tool steels.
The process is conducted using the AOD converter, i.e., a processing vessel
comprising a refractory-lined steel shell mounted on a tiltable trunnion ring.
Concerning the refractory material, most AOD converters use dolomite refractories
(ASM International 2008, Fig. 5.6).
The AOD converter looks very much as a Bessemer converter and has tuyeres,
located in the lower sidewalls, which are used for the injection of argon, nitrogen,
and oxygen. The most recent AOD converters use a combined blowing system,
called OTB (oxygen top blowing), in which gases are injected into the molten bath
also through a top lance (see Fig. 5.6). The inert component of the gas flux is used for
stirring the molten mass with a consequent better chemical and thermal homogeneity
of the liquid. In the AOD process, the heat transferred to the liquid alloy is supplied
by exothermic oxidation reactions taking place into the liquid alloy and involving the
injected oxygen, so that no external heating source is required. The temperature is
controlled by adjusting the amount of oxygen, alloying elements, and solid steel
scraps introduced into the converter during the refinement. The reactions occurring
206 5 Steels

during the refining process involve a preliminary reduction of the oxides, formed in
the melt, and subsequent release of carbon monoxide. Two decarburization reac-
tions, involving chromium, are:

1 3
Cr3 O4 þ C ! Cr þ COðgÞ ð5:1Þ
4 4

and

Cr2 O3 þ 3C ! 3COðgÞ þ 2Cr ð5:2Þ

The relevant equilibrium constants, K, can be expressed as:

3
ðaCr Þ4  ðPCO Þ ðaCr Þ  ðPCO Þ3
K¼ and K ¼ ð5:3Þ
ðaCr3 O4 Þ  ðaC Þ ðaCr2 O3 Þ  ðaC Þ3

where:
aCr is the activity of chromium.
PCO is the partial pressure of the carbon monoxide.
aCr3 O4 and aCr2 O3 are the activities of chromium oxides.
aC is the activity of carbon.
Equation 5.3 suggests that a reduction in the partial pressure of CO leads to an
increase in the equilibrium concentration of chromium with carbon in the molten
steel (Fig. 5.7).
Similar reactions may involve other oxygen reactive elements, which are indeed
considered in the refinement of steels. The carbon content at meltdown can typically
vary between 0.5% and 3.0%, depending on the composition of the furnace charge
(ASM International 2008). The AOD refining process implies a stepwise injection of
oxygen and inert gases, whose ratio is varied during the process (see inset in
Fig. 5.8). The schematic of a typical stainless steel refining cycle for a 30 t vessel
charge is shown by Fig. 5.8.
According to these curves, as the oxygen to inert gas, Ar or N, ratio increases, the
carbon content decreases, owing to its progressive oxidation. Other processing
parameters for the AOD decarburization refinement of an AISI 304 stainless steel
are given in Tables 5.2 and 5.3.
During the AOD refinement, the chemical composition of the bath is checked at
regular intervals and when the target composition is reached, the alloy is either cast,
e.g., ingot or continuous casting products, or poured into a ladle furnace for further
refining treatments. The charge composition control during AOD process is
extremely accurate and involves weight control of the heat and sampling of the
liquid steel.
5.3 Standard and Special Production Techniques 207

Fig. 5.7 C-Cr equilibrium curves in the AOD converter. The data show that a reduction in the
partial pressure of CO corresponds to lower carbon levels at higher chromium concentrations, also
at lower temperatures. (Redrawn from ASM International 2008)

The dilution of oxygen with inert gases results in an increased carbon removal
efficiency, resulting in carbon concentrations below 0.01%. At the same time, the
desulfurization process can reduce the sulfur content down to 0.01% or less (ASM
International 2008).

5.3.4 Vacuum Oxygen Decarburization (VOD and VODC)

Vacuum oxygen decarburization is another refining process conducted under vac-

uum in a converter vessel (VODC vacuum oxygen decarburization converter) or in a
ladle (VOD). The conversion process is not that different from the AOD (-OTB)
process, although operating under a vacuum environment requires a thorough
control of the bottom blowing of the reaction gases (Fig. 5.9).
208 5 Steels

Fig. 5.8 Schematic of a stainless steel refining for a small AOD converter (30 t charge) showing the
relationship between C and Cr contents as a function of time and temperature. (Redrawn from ASM
International 2008)

Table 5.2 Processing parameters for AOD refinement of an AISI 304 austenitic stainless steel
(ASM International 2008)
Partial pressure
Gas ratios [atm]
Carbon CRE CRR Temperature
level O2 N2 Ar Ar N2 CO %a [ppm/min]b [
C]
3.0–0.7 4 1 – – 0.14 0.86 80 1200 1510
1–0.25 3 1 – – 0.20 0.80 65 830 1700
0.25–0.12 1 1 – – 0.53 0.47 45 430 1700
0.12–0.04 1 – 3 0.83 – 0.17 30 133 1730
0.04–0.01 1 – 8 0.96 – 0.04 15 26 1730
Total gas flow rate ¼ about 1.5 m3/min/ton
a
CRE carbon removal efficiency
b
CRR carbon removal rate

VODC ensures low chromium losses during refining and higher carbon removal
rates thanks to the very low CO partial pressures (Fig. 5.10). Sulfur levels can be
kept in the 10–30 ppm range for any steel grade, and the resulting hydrogen content
is generally below 2 ppm. A common feature to all the converter treated products,
including VODC, is gas inclusions, which remain trapped into the liquid during
5.3 Standard and Special Production Techniques 209

Table 5.3 Composition control of an AOD-refined high-strength low-alloy steel (ASM Interna-
tional 2008)
Element Target [%] Mean [%] Standard deviation [%]
Aluminum 0.05 0.043 0.009
Carbon 0.27 0.264 0.006
Chromium 0.50 0.56 0.07
Manganese 0.80 0.83 0.03
Molybdenum 0.28 0.28 0.01
Nickel – 0.13 0.03
Phosphorous – 0.019 0.002
Silicon 0.50 0.49 0.04
Sulfur – 0.002 0.001

Fig. 5.9 Schematic of a VODC with the oxygen top blowing (OTB) system. (Redrawn from The
AISE Foundation 1998)

tapping into the teeming ladle. To overcome or reduce the incidence of this phe-
nomenon, if critical, a successive ladle treatment may be employed.

5.3.5 Electroslag Remelting (ESR)

Electroslag remelting (ESR) is a refining process (Fig. 5.11a), highly recommended

if better alloy homogeneity, cleanliness, and control of the solidification structures
are paramount. The early design of ESR dates back to the 1930s and successively
patented in 1945. Nowadays, it is a fully established process, with a steel production
210 5 Steels

Fig. 5.10 Chromium oxidation rates and carbon removal rates for the VODC, OTB, and AOD
processes. (Redrawn from ASM International 2008)

Fig. 5.11 (a) ESR furnace with retractable bottom baseplate. A fixed mold design is a possible
alternative. (b) Schematic of an ESR furnace with multiple electrodes and detail of the ingot
solidification. (Redrawn from ASM International 2008)

capacity of more than 1 million tons per year worldwide (ASM International 2008).
The ESR process is not restricted to steels, and it is actually employed in the
processing Ti-alloys (see Chap. 4) and Ni-based superalloys (see Chap. 6) as well.
As concerns steels, ESR is particularly suited for HSLA steels, stainless steels, tool
5.3 Standard and Special Production Techniques 211

steels, bearing, and heat-resistant steels. ESR requires a consumable electrode,

which is melted by immersion into the molten slag. Melting temperature is reached
thanks to an electric current flowing from the base plate through the slag and the
electrode (Fig. 5.11b). ESR process is used for the production of cylindrical ingots,
with diameters of 50–75 cm. A variant of the process, named ESRR (electroslag
rapid remelting), is used for the production of smaller ingots with slab, square, or
round cross section that can be directly processed, using conventional rolling mills
and presses.
Possible alternatives to the ESR process are VAC-ESR and pressurized ESR. In
the first, remelting is conducted under a low gas pressure, possibly slightly reducing,
with additions of nitrogen and hydrogen. The conventional ESR instead is
performed in air, with a worse control of the impurity level in the final products.
In pressurized ESR, a gas, e.g., nitrogen, overpressure promotes its own diffusion
into the alloy and modifies the resulting mechanical properties (see Appendix 4).
Most ESR plants have AC furnace, using multiple electrodes for the production of
single ingots (Fig. 5.11b). At the beginning of the remelting process, the slag induces
the melting of the tip of the electrode, and metal droplets falling through the liquid
slag are collected into a water-cooled copper mold. The solidification rate is constant
since the mold is subjected to a constant heat extraction through the cooling water
circuit. The metal undergoes a refinement process, and it is purified from unwanted
substances, like oxide particles and other kinds of inclusions that are trapped into the
slag. The main parameters that control the metal/slag reaction are:
• Superheating of both slag and metal that promotes metal/slag reactions and
consequent cleaning up of the liquid metal.
• Slag chemical composition, which can be tuned according to specific applica-
tions. General purpose slag has composition: 70% CaF2, 20% CaO, and 10%
Al2O3. A slag made of 30% CaO and 70% CaF2 has a high thermal conductivity,
recommended when alumina is not allowed and there is a risk of hydrogen pickup
(ASM International 2008).
• Metal-slag interaction. Metal droplets fall into the mold passing through the
liquid slag. The large specific surface area of the metal/slag interface enhances
the kinetics of the reactions between the liquid alloy and the slag.
The inclusions that remain trapped into the solidifying ingot are usually very fine
grained, and their distribution is rather homogeneous, considering the kinetics of the
remelting process, featuring a comparatively limited thickness of the liquid pool.
Concerning the slag chemical composition, it can be adjusted in order to improve the
dephosphorization and desulfurization processes. The final desulfurization that can
be attained for a given a slag composition is usually evaluated by means of the
following reactions:

FeS þ CaO ! CaS þ FeO ð5:4Þ

212 5 Steels

3
CaS þ O2 ðgÞ ! CaO þ SO2 ðgÞ ð5:5Þ
2

These two reactions describe the transfer of sulfur from the metal to the slag and
from the slag to the process gas, collected and removed from the working atmo-
sphere. In this way, the desulfurization capacity of the slag remains substantially
unaltered with time. When the index of basicity (also known as IB) of the slag,
evaluated as the CaO/SiO2 ratio, is larger than 3, more than 80% of sulfur can be
removed from the slag during remelting. Reactions between the liquid alloy and
oxygen cannot be avoided when the process is carried out in air. Therefore, oxidation
of the alloy occurs to some extent, either by a direct reaction of the electrode surface
above the liquid slag or through the oxidation of elements, such as iron and
manganese, at the slag surface. These oxides are subsequently reduced by alloy
elements that have high affinity to oxygen, e.g., aluminum and silicon, resulting in
the unwanted depletion of these elements in the ingot. In order to limit this draw-
back, a continuous deoxidation of the slag can be carried out by means of aluminum
additions that can even lead to an oxygen content below 20 ppm in unalloyed steels
(ASM International 2008). ESR ingots are characterized by a very clean and smooth
surface: this helps reducing the production costs, since there is no need for grinding
prior to hot-working. The homogeneity of ESR ingots, as concerns macro-
segregation, and the uniform distribution of inclusions result in improved mechan-
ical properties, with particular regard to yield strength and toughness. Uniformity in
transverse and longitudinal direction after hot-working is another important techno-
logical advantage. Concerning process parameters, the control of the molten pool
depth and shape has direct influence on the solidification structures, which are
dependent on the temperature gradient at the solid/liquid interface. A relatively
high-temperature gradient at the solidification front leads to the development of
directional dendritic structures. Micro-segregation in ESR ingots increases with
increasing dendrite arm spacing (DAS). Minimum micro-segregation can be
achieved through a dendrite growth direction as closer as possible to the ingot
solidification axis. This latter condition cannot be achieved in all cases, particularly
in larger ingots. ESR ingots are prone to the formation of specific defects, like white
spots, freckles, and tree-ring pattern, all described in the next section (Sect. 5.3.6),
since the same kind of defects can be found in VAR ingots.

5.3.6 Vacuum Arc Remelting (VAR)

Vacuum arc remelting (VAR) has been specifically developed for the production of
superalloys (see Chap. 6), and it was first used in the late 1950s for aerospace
materials. The use of the VAR has been thereafter extended to other alloys (Mitchell
2005):
• Steels.
• Titanium alloys (see Chap. 4).
• Refractory alloys (see Sect. 9.2).
5.3 Standard and Special Production Techniques 213

VAR can be used for refinement and purity improvement of air-melted or VIM
ingots, and it is also a fundamental step in the superalloy-specific triplex process,
i.e., VIM-ESR-VAR (see Sect. 6.3.2) (ASM International 2008; Cobb 2010). The
refined alloys, produced by vacuum arc remelting, are characterized by improved
mechanical properties, in particular fracture toughness and fatigue resistance. The
main reasons for this improvement can be ascribed to the following factors or to a
combination of them:
• Elimination of porosity.
• No macro-segregations, thanks to micro-fusion and layer-by-layer solidification
of the ingot.
• Reduction of impurity content, thanks to their preferential evaporation under
vacuum, as observed for lead and tin. Unfortunately, other alloying elements,
like manganese, with high vapor pressure, intentionally added to the master batch
for their beneficial effects, may undergo the same process so that their concen-
tration is unwillingly reduced, thus requiring an adequate compensation in setting
the composition of the initial batch.
• Reduction of micro-segregations.
• Removal of dissolved gaseous elements, e.g., N2, O2, CO, and H2, that are easily
outgassed from the melt.
• Reduction of incidence of oxides and consequent alloy purity improvement.
Oxides, and also nitrides, are eliminated from the melt by thermal dissociation or
for the reducing effect of carbon and subsequent elimination via gas phase (ASM
International 2008). Very stable oxides, like Al2O3, and titanium carbonitrides are
removed by floatation, favored by their lower density (ASM International 2008).
The overall design of a VAR furnace has not undergone particular changes over
the years (scheme in Fig. 5.12). The mold is placed into a water-cooled vessel,
similarly to ESRs. In Fig. 5.12 it is possible to see that the electrode is driven into the
crucible by a ram, while the electrode itself is being consumed, and its position is
continuously controlled in order to maintain the optimal arc gap, i.e., the distance
between the electrode tip and the molten bath. Reproducibility of the process and
homogeneity of the produced ingots strongly depend on a close control of the
process parameters, which has been considerably improved in fully automatized
modern plants.
VAR ingots exhibit a microstructure influenced by both the temperature gradient
of the liquid-solid interface and the local solidification rate. A relatively high-
temperature gradient at the solidification front is a necessary requirement to obtain
an oriented dendritic primary structure, whose growth direction, always perpendic-
ular to the solidification front, is parallel to the direction of the temperature gradient.
Vacuum pressure typically ranges in between 0.1 and 1.0 Pa, although remelting
can also be carried out under an inert gas pressure that can be as high as 10 kPa, in
order to reduce the evaporation losses of the more volatile alloying elements. On the
other hand, an increase in the working pressure corresponds to a decrease in the
214 5 Steels

Fig. 5.12 Schematic of a

VAR furnace with its key
components. (Redrawn from
ASM International 2008)

melting rate, due to the thermal losses induced by the residual gas pressure (Boxman
et al. 1996). The peak direct current can be as high as 40 kA, remelting rate can reach
19 kg/min, and a typical ingot’s diameter can be up to 1.5 m.
Figure 5.13 shows the scheme of the molten metal pool-electrode region of a
VAR ingot during remelting. The electrical current flowing through the electrode
determines the depth of molten pool, mainly influenced by the thermal buoyancy and
Lorentz stirring. An excessive depth of the liquid inhibits the migration of the
impurity oxides toward the outer surface and favors instead their segregation along
the centerline of the ingot. ESR and VAR ingots share similar defect structures,
which may importantly affect the quality and properties of the relevant products. The
main kinds of defects are:
• Freckles. They are regions of the alloy containing an excess of carbide-forming
elements or carbides, resulting from segregations within the liquid pool, induced
by its excessive depth. The situation may be worsened by turbulences induced by
external causes, such as stray magnetic fields.
• “Tree rings.” They are series of comparatively small equiaxed grains that form
within the elongated dendritic microstructure for a locally higher grain nucleation
rate, due to thermal and compositional fluctuations inside the melt (Xu et al.
2002).
5.3 Standard and Special Production Techniques 215

Fig. 5.13 Schematic of the liquid pool region of a VAR ingot during remelting process. (Redrawn
from ASM International 2008)

• White spots. They can be easily recognized on the external surface of the ingot, as
white-etched spots, featuring compositional fluctuations, in particular a depletion
of alloy elements, originating from unmolten dendrites that fall into the metal
pool directly from the electrode, thus retaining the original electrode composition.
Solid pieces, contributing to the formation of white spots, may detach from other
parts of the VAR too, like the crown (see below) that forms on the inner wall of
the mold, along the ingot-electrode gap (see Fig. 5.13).
Freckles and white spots act as stress concentrators and tend to reduce the fatigue
life of structural components. In parts, like compressor discs of the final stages and
turbine superalloy disc (see Table 2.1, Sect. 5.5.3, and Chap. 6), these defects have
been identified as one of the main causes of premature failure. To reduce the
incidence of these defects, in particular of the white spots, a careful control of the
following remelting conditions and parameters is required:
• High melting rates, capable to rapidly melt any solid particle that may fall into the
molten metal pool.
216 5 Steels

• High-quality electrodes, with no chemical and microstructural heterogeneities,

like cracks and solidification shrinkage voids, from which a preferential particle
detachment may more easily occur.
• Reduction of the electrode-ingot gap, so to limit the length of the arc and the
relevant tendency to form the so-called crown, which is a possible source of solid
fragments contributing to the formation of white spots.

5.4 Steels for Aerospace Applications

Selected kinds of steels, interesting for the aerospace industry, are described here-
with, also mentioning criteria for their selection. Together with the following main
families of steels, i.e., maraging, AerMet, and stainless steels, others are used in
aerospace applications, like:
• Advanced carburizing steels (e.g., Pyrowear 53).
• Carburizing/nitriding steels (e.g., X30).
• Conventional carburizing steels (e.g., AMS 6265).
• Conventional nitriding steels, such as Nitralloy N.
• High fracture toughness steels such as AF1410 and HP 9-4-30.
• Medium-carbon low-alloy steels (e.g., AISI 4130, AISI 4140, AISI 4340, and
300M) (Campbell 2006; Flower 1995).

5.4.1 Maraging Steels

The total use of steel for structural components in aircrafts has progressively
decreased, due to the development and availability of aluminum (Chap. 3) and
titanium (Chap. 4) alloys. However, there are still components requiring high tensile
strength, wear resistance, and stiffness that render steels still the best-candidates,
particularly when ultrahigh tensile strengths are required. Since high strength is often
accompanied by a reduced ductility (Megson 2013), the development of a new class
of steels began at the end of the 1950s at the INCO research laboratories, to
overcome this limitation (Würzinger et al. 2004). This class of steels is named
maraging, i.e., martensite aging or martensite age hardening, from the main
strengthening mechanism. It involves the age hardening of the low-C, Fe-Ni lath
martensite structures present in the steel after quenching. The main phase trans-
formations occurring in these alloys are illustrated in the phase diagrams in Fig. 5.14.
Maraging steels are characterized by chemical compositions rich in nickel, cobalt,
and molybdenum and relatively poor in carbon, as shown by Table 5.4. In fact,
intermetallic compounds precipitate during the age-hardening treatments and do not
involve any solid-state reaction with carbon (ASM International 1993). Except 18Ni
(180), all maraging steels are based on the 18Ni-Co-Mo type, and the available
5.4 Steels for Aerospace Applications 217

Fig. 5.14 (a) Equilibrium phase relationships in the Fe-Ni system; (b) metastable phase relation-
ships in the Fe-Ni system. (Redrawn from ASM International 1993)

Table 5.4 Nominal composition (wt%) of maraging steels. Fe concentration balance (ASM
International 1993; Novotny and Maurer 2007)
Grade C Ni Mo Co Ti Al Nb
18Ni(200) – 18.0 3.3 8.5 0.2 0.1 –
18Ni(250) 0.03 max 18.0 5.0 8.5 0.4 0.1 –
18(Ni300) 0.03 max 18.0 5.0 9.0 0.7 0.1 –
18(Ni350) 0.05 max 18.0 4.2 12.2 1.6 0.10 –
18Ni(Cast) – 17.0 4.6 10.0 0.3 0.1 –
Co-free 18Ni(200) – 18.5 3.0 – 0.7 0.1 –
Co-free 18Ni(250) 0.03 max 18.5 3.0 – 1.4 0.1 –
Co-free 18Ni(300) 0.03 max 18.5 4.0 – 1.85 0.1 –
Low-Co 18Ni(250) 0.03 max 18.5 2.6 2.0 1.2 0.1 0.1
12-5-3(180) – 12.0 3.0 – 0.2 0.3 –

commercial grades are 18Ni(200), 18Ni(250), 18Ni(300), 18Ni(350), 18Ni(cast),

and 12-5-3(180) (Prasad and Wanhill 2017). The shortage of cobalt during the late
1970s and 1980s, with the consequent cobalt price fluctuations, led to the develop-
ment of several low-Co grades and Co-free grades of maraging steels: Co-free 18Ni
(200), Co-free 18Ni(250), low-Co 18Ni(250), and Co-free 18Ni(300). Designation
of maraging steels refers to their nominal yield strength, which is the value (in ksi)
among brackets. For example, the standard 18Ni(350) steel has a yield strength of
350 ksi (2420 MPa) (ASM International 1993).
218 5 Steels

Table 5.5 Effect of the most important alloying elements in maraging steels (ASM International
1993; Prasad and Wanhill 2017)
Name Symbol Features and alloying effects
Aluminum Al It is usually kept below 0.2–0.3% in order to avoid a decrease in ductility
Chromium Cr Chromium increases the strain hardening coefficient of martensite as
well as the corrosion resistance
Cobalt Co Cobalt rises the martensite start temperature (Ms) so that it is possible
to increase the concentration of hardening elements without increasing
the retained austenite content. Furthermore, its presence leads to an
increase in the volume fraction of Ni3Mo and Ni3Ti precipitates
Molybdenum Mo Molybdenum, in addition to forming the reinforcing Ni3Mo phase,
promotes the plasticity and ductility in the aged condition. This is so,
since molybdenum is able to reduce the diffusion coefficient of several
elements, particularly at the grain boundary, with a consequent
reduction in precipitation of second phases during age hardening
Nickel Ni Nickel can be regarded as the most important alloying element in
maraging steels. Its presence leads to the formation of a lath martens-
itic structure
Titanium Ti The main effect of titanium is to harden the material after heat treat-
ment. If enough carbon is present in the alloy, it can react with titanium
leading to the formation of the deleterious, since embrittling TiC

In Table 5.5 the main effects of the most important alloying elements present in
maraging steels are indicated. Although not included in the table, other alloying
elements too, like Cu, Nb, Si, Ta, and V, may contribute to the steel hardening (Sha
and Guo 2009).
Figure 5.14a shows the Fe-rich region of the Ni-Fe equilibrium phase diagram
and indicates that the equilibrium phases at low temperature for high nickel content
are austenite (γ) and ferrite (α). Figure 5.14b shows the metastable phase diagram,
including the austenite-to-martensite transformation upon cooling and the reverse
transformation on heating. From the diagram in Fig. 5.14b, it turns out that there are
no phase transformations occurring until the martensite start temperature is reached
(i.e., martensite formation on cooling region in Fig. 5.14b). One interesting feature
of maraging steels is the possibility of obtaining a fully martensitic structure even
with a very slow cooling rate and also when treating large sections. In other words,
maraging steels would exhibit some hardenability anyway. Ms can be tuned through
a suitable choice of the alloy chemical composition that does not introduce any
change in the character of the martensitic transformation, which remains indepen-
dent from the cooling rate. Nickel and the other alloying elements, except cobalt,
tend to lower Ms. Cobalt raises Ms, and it can be exploited to compensate for the
addition of the other alloying elements in order to maintain a martensitic transfor-
mation still fully accomplished at room temperature. In fact, the Ms of maraging
steels lays in the range of 200
C–300
C. The resulting martensite is relatively soft
with hardness of 30–35 HRC, so that it can be machined and deformed quite easily
(Campbell 2006). The martensitic transformation is not sufficient to reach the peak
properties, which require instead an aging treatment, to induce precipitation
strengthening mechanisms (see Chap. 6).
5.4 Steels for Aerospace Applications 219

Table 5.6 Characteristics of intermetallic phases in maraging steels (Moshka et al. 2015; Tewari
et al. 2000)
Intermetallic phase Unit cell Composition/stoichiometry Lattice parameters [Å]
H Hexagonal Ni3(Ti, Mo) a ¼ 5.101
c ¼ 8.307
Laves Hexagonal Fe2Mo a ¼ 4.755
c ¼ 7.767
Ni3Mo Orthorhombic Ni3Mo a ¼ 4.224
b ¼ 4.448
c ¼ 5.064
S Hexagonal A8B a ¼ 7.04
c ¼ 2.48
μ Rhombohedral Fe7Mo a ¼ 4.751
c ¼ 8.988
(γ ¼ 120
)
Χ Hexagonal A3B a ¼ 2.55
c ¼ 8.30

Several studies have been performed for the identification of the main reinforcing
intermetallic phases in maraging steels; it has been observed that the precipitation
process occurs preferentially at dislocations and within the martensite laths, leading
to the formation of a fine and uniform distribution of coherent precipitates (see Sect.
6.4.1.1) (Moshka et al. 2015; Pardal et al. 2005; Tewari et al. 2000). Thanks to their
nanometric size and coherent interface with the alloy matrix, the otherwise embrit-
tling intermetallic precipitates ensure a beneficial strengthening effect (Leitner et al.
2011; Yang et al. 2015). Phases observed in maraging steels are given in Table 5.6.
The alloying element providing the highest contribution to hardening is molybde-
num, forming the metastable Ni3Mo phase. Coherency strains tend to limit the
growth of Ni3Mo, and, therefore, further aging results in the transformation of
Ni3Mo into the equilibrium Fe2Mo precipitate. Titanium promotes further hardening
with the formation of Ni3Ti, and another intermetallic precipitate, FeTi, may form in
the nickel-depleted regions, after the formation of the Ni3Mo and Ni3Ti precipitates
(ASM International 1993). During the hardening treatments, detrimental phases may
also form, like elongated nonmetallic inclusions (NMIs), visible on the fracture
surfaces (Fig. 5.15).
Energy dispersive X-ray (EDX) analysis has been used to identify the composi-
tion of these NMIs, of which two main types have been recognized: blocky titanium
nitrides (Fig. 5.15) and titanium sulfides (Fig. 5.16) (Würzinger et al. 2004). Low
carbon concentration (Table 5.4) is also meant to inhibit the formation of other NMIs
phases, like TiC. These and similar phases may have deleterious consequences on
the toughness, ductility, and strength of the alloy (ASM International 1993;
Würzinger et al. 2004).
220 5 Steels

Fig. 5.15 (a) NMIs inhomogeneous band; (b) blocky titanium nitride. (Würzinger et al. 2004,
reproduced with permission of the Publisher)

Fig. 5.16 SEM investigation on inclusions showing (a) and (b) to be titanium sulfides. (Würzinger
et al. 2004, reproduced with permission of the Publisher)
5.4 Steels for Aerospace Applications 221

Fig. 5.17 Vickers hardness of 18Ni(250) maraging steel versus aging time for various aging
temperatures. (Redrawn from ASM International 1993)

The melting process used for maraging steels depends on the target grade of the
final product and relevant applications. There are two main options: either EAF or
VIM, followed in both cases followed by VAR (ASM International 1993). EAF/VAR
route is mainly used for the production of tool steel grades (Würzinger et al. 2004).
Maraging steels that belong to the family of cast grades can be melted and cast in air, in
view of their adequate fluidity and castability (Prasad and Wanhill 2017). In Ti- and
Al-containing grades, the affinity of these elements for nitrogen and oxygen may result
in an increase in the inclusion content (Würzinger et al. 2004).
The relatively high titanium and molybdenum concentrations render the steel
more prone to micro-segregations during solidification. Therefore, premium grades,
used for the more demanding applications, where homogeneity and high mechanical
properties are paramount, are prepared using the so-called triple process, EAF
melting, VIM, and final VAR, since in this way, it is possible to reach very low
concentrations of gaseous, i.e., N2, H2, and O2, and of other impurities, like Mn, C,
P, and S (ASM International 1993).
As concerns the post-processing heat treatments of the maraging steels, before
age hardening, the steel is usually solution annealed in the austenite field; 1 h is an
average time for 25 mm section parts. Subsequently, the material is cooled down to
room temperature before aging. Properties and microstructure are only slightly
affected by the cooling rate after solution annealing. Age hardening requires heat
treatment at temperatures in the 455
C–530
C range, lasting from 3h up to 12 h,
depending on the grade and desired properties (Fig. 5.17). Specific applications,
such as die casting, require an aging temperature of about 530
C. If it is maintained
for too a long time, coarsening of precipitates leads to a sharp softening and to the
martensite to austenite reverse transformation. The refinement of coarse-grained
microstructure can be obtained in maraging steels thanks to a thermal cycling
222 5 Steels

Table 5.7 Mechanical properties and heat treatments of standard maraging steel grades (ASM
International 1993)
Tensile Yield % % Fracture
Heat strength strength Elongation Reduction toughness
Grade treatment [MPa] [MPa] in 50 mm in area [MPa√m]
18Ni A 1500 1400 10.0 60 155–240
(200)
18Ni A 1800 1700 8.0 55 120
(250)
18Ni A 2050 2000 7.0 40 80
(300)
18Ni B 2450 2400 6.0 25 35–50
(350)
18Ni C 1750 1650 8.0 35 105
(Cast)
Heat treatment A: solution treatment 1 h at 820
C. Age hardening 3 h at 480
C
Heat treatment B: solution treatment 1 h at 820
C. Age hardening 12 h at 480
C
Heat treatment C: annealing 1 h at 1150
C. Age hardening 1 h at 595
C. Solution treatment 1 h at
820. Age hardening 3 h at 480
C

Table 5.8 Mechanical properties of three variants of the heat-treated 18Ni(250) steel (ASM
International 1993)
Tensile Yield % % Charpy V-notch Fracture
strength strength Elongation Reduction impact toughness
Grade [MPa] [MPa] in 25 mm in area toughness [J]a [MPa√m]a
18Ni 1870 1825 12.0 64.5 37 138
(250)
Co-free 1895 1825 11.5 58.5 34 127
18Ni
(250)
Low-Co 1835 1780 11.0 63.5 43 149
18Ni
(250)
Heat treatment: solution treatment 1 h at 815
C. Age hardening 5 h at 480
C
Samples: 63.5  88.9 mm2 billets originated from Ø200 mm VIM/VAR billet
a
Orientation of the test: L-S (longitudinal-short transverse orientation)

between Mf and a temperature which is significantly higher than the precipitate

solutioning temperature. For instance, an ASTM grain size of 1/1 in 18Ni(300) has
been refined by Saul et al. (1970) by means of three cycles between room temper-
ature and 1025
C. Prolonged aging treatments induce the formation of the equilib-
rium phases, i.e., austenite and ferrite (Saul et al. 1970; ASM International 1993).
The mechanical properties of various maraging steel grades are given in
Table 5.7, with the indication of the relevant heat treatments.
The Table 5.8 shows the main mechanical properties of three variants of the 18Ni
(250) maraging steel: standard composition, Co-free, and low-Co bearing grade (See
Table 5.4).
5.4 Steels for Aerospace Applications 223

Fig. 5.18 Plane-strain fracture toughness of several steel materials: maraging steel family and
ultrahigh strength steels as a function of tensile strength. (Redrawn from ASM International 1993)

A comparison of the plane-strain fracture toughness of maraging steels and

conventional tempered martensite steels is given in Fig. 5.18 as a function of their
tensile strength.
The evolution with temperature of tensile strength, yield strength, ductility, and
hardness of several maraging steels is given in Fig. 5.19.
From the plots in Fig. 5.19, it turns out that the maraging steels can be used for
long times at temperatures up to 400
C, considering that the tensile and yield
strengths at this temperature are 80% of the room temperature values approximately.
Concerning fatigue properties, the behavior of maraging steels and other high-
strength steels is shown by Fig. 5.20.
Fatigue crack growth rate is comparable to the rates observed in conventional
steels. An overview of most important technological processes applied to maraging
steels is given in Table 5.9.
The corrosion resistance of maraging steels is slightly better than that of tempered
martensitic steels: in industrial and marine atmospheres, it is about one and a half of
that observed in conventional steels. Maraging steels have essentially the same
behavior of conventional steels when in contact with static or flowing seawater.
Nickel and cobalt improve their high-temperature oxidation resistance, which is
indeed better than that of conventional martensitic alloy steels (ASM International
1993). Maraging steels do not undergo passivation (see Sect. 8.1.6) and are
subjected to stress corrosion cracking (SCC, see Sect. 8.1) in most aqueous envi-
ronments: the higher the yield strength, the higher the susceptibility to SCC (Flint
et al. 2010).
224 5 Steels

Fig. 5.19 Effect of the temperature on the mechanical properties of 18Ni maraging steel family: (a)
tensile and yield strength, (b) ductility, and (c) Rockwell hardness C. (Redrawn from ASM
International 1993)
5.4 Steels for Aerospace Applications 225

Fig. 5.20 Rotating-beam fatigue properties of 18Ni maraging steels family. (Redrawn from ASM
International 1993)

Table 5.9 Overview table of the main processing routes used for maraging steels (ASM Interna-
tional 1993; Campbell 2008)
Process Features
Hot-working Hot-working can be carried out by conventional forging and steel mill tech-
niques. Precipitation of TiC at the grain boundary must be avoided; therefore a
slow cooling process in the range of 750–1025
C after hot-working is unde-
sirable. The same holds for hot-working processes above 1260
C. Further-
more, Ti-containing grades have greater hot strength if compared with
conventional steels; therefore they need either higher working temperatures or
higher hot-working loads. However, for mechanical properties maximization,
maraging steels should be hot-worked at temperatures as lower as possible
Cold forming When in un-aged condition (i.e., solution annealed or as-transformed),
maraging steels can be cold-worked by any conventional technique. They show
very low work-hardening rates and can accept heavy reductions (50%) with
only slight increase in hardness
Machining Maraging steels can be machined by means of conventional techniques both in
solution-annealed and age-hardened condition. Machinability of maraging
steels is similar, and sometimes better, as that of conventional steels of
comparable hardness
Welding Maraging steels show an excellent weldability. They can be readily welded by
means of GTAW (gas tungsten arc welding) in both the age-hardened and
annealed condition
Powder Maraging steels produced via powder metallurgy of elemental powders show
metallurgy reasonable ductility combined with a yield strength around 1600 MPa.
Gas-atomized alloy powders canned and extruded are able to generate a
material with excellent ductility and yield strength up to 1900 MPa. Contam-
ination of the surface of the powder during atomization and subsequent col-
lection is one of the major issues of this process
226 5 Steels

5.4.2 AerMet Steels

There has been a growing interest for ultra-high strength steels over the last decades,
in particular for those alloys exhibiting secondary hardening, due to the precipitation
of carbides during tempering. After the development of commercial maraging steel
grades, AerMet steels have been introduced. AerMet steels show both outstanding
toughness and ultra-high strength. These steels were developed at the McDonnell-
Douglas, as candidate materials for the fabrication of the landing gear of the F/A
18 E/F military aircraft. The most demanding requirement for this material was a
toughness value at least twice that of the 300M steel (AISI 4340M) but still retaining,
or even improving, its strength. AerMet 100 had met these requirements, considering
its minimum toughness of 110 MPa√m and 1930 MPa ultimate tensile strength.
After its very first application for the F/A 18 E/F fighter, AerMet 100 was used as
replacement for other steels, such as AF1410 (Air Force 14Co 10Ni), HP-9-4-30,
and AISI 4340. As compared to maraging steels and titanium alloys, AerMet steels
show better fatigue life and excellent combination of toughness and strength
(Novotny and Maurer 2007). They can be considered as a family of secondary
hardening steels based on the Ni-Co-Cr-Mo alloying system (Prasad and Wanhill
2017). In addition to AerMet 100, other varieties of this family have been developed
over the years, with increasingly higher strength. This is the case of the 488
C
(875
F) age-hardened AerMet 100, the AerMet 310 (UTSmin ¼ 2140 MPa), and the
AerMet 340, this latter featuring the highest strength among all of the AerMet alloys
developed so far (UTSmin ¼ 2344 MPa) (Novotny and Maurer 2007). The nominal
compositions of the three AerMet steels are listed in Table 5.10.
Like maraging steels, AerMet alloys have an age-hardened Fe-Ni-Co-C lath
martensite matrix. They attain their peak strength thanks to the precipitation of
M2C carbides, where M can be either a Mo/Cr combination or just one of the two
elements (Novotny and Maurer 2007). The important role of carbides in strength-
ening AerMet steels is coherent with the higher carbon concentration in these alloys
(Table 5.10). The M2C carbides replace M3C (i.e., Fe3C, cementite), thus removing
an important grain boundary embrittling agent. Moreover, the fine-grained M2C
carbides, being better distributed within the alloy matrix, contribute to increasing
both hardness, enhanced by age-hardening thermal treatments (Novotny and Maurer
2007), and fracture toughness (Graves 1994). The melting process used for the
production of AerMet steels is generally a combination of VIM and VAR, suitable
to achieve the required mechanical properties and low impurity levels. The amount
of inclusions and the impurity concentration are kept to a minimum thanks to the
double vacuum process, as concerns particularly silicon and manganese that tend to
reduce the grain boundary cohesion of the austenite phase. After ingot production,
the AerMet steels are processed to obtain different products: billets, plates, sheets,
strips, die forgings, bars, and wires (Campbell 2008). The maximum starting tem-
perature for forging (primary breakdown) is 1230
C. Finishing operations can be
carried out starting from 980
C. Air cooling and subsequent annealing must be
performed on workpieces after forging; a normalization treatment is generally
needed to restore the properties of the dead zones, not involved in the plastic
 5.4 Steels for Aerospace Applications

Table 5.10 Nominal compositions (wt%) of AerMet alloys (Novotny and Maurer 2007)
Grade C Cr Ni Mo Cu Co V Ti Al B Zr Ca Fe
AerMet 100 0.23 3.10 11.50 1.20 – 13.50 – 0.015 max 0.015 max – – – Balance
AerMet 310 0.25 1.40 11.00 1.40 – 15.00 – 0.015 max 0.015 max – – – Balance
AerMet 340 0.33 2.25 12.00 1.85 – 15.60 – 0.015 max 0.015 max – – – Balance
227
228 5 Steels

Table 5.11 Standard heat treatment sequence for AerMet alloys (Prasad and Wanhill 2017)
AerMet 100 AerMet 310 AerMet 340
Normalization Austenitization: Austenitization: 970
C. Austenitization: 968
C.
900–925
C. Soaking Soaking time: at least 1 h Soaking time: at least
time: at least 1 h each each 25 mm of thick- 1 h each 25 mm of
25 mm of thickness. ness. Final cooling: air thickness. Final cooling:
Final cooling: air air
Stress relief Stress relief is performed after normalization annealing. Processing
temperature ¼ 675
C
Annealing Annealing is performed at 677
C with a soaking time of 16 h followed by air
cooling. The process leads to a maximum hardness equal to HRC 40
Quenching/ Austenitization: Austenitization: Austenitization:
hardening 870–900
C. Soaking 900–925
C. Soaking 955–988
C. Soaking
time: at least 1 h each time: at least 1 h each time: at least 1 h each
25 mm of thickness. 25 mm of thickness. 25 mm of thickness.
Cooling process either in Cooling process either in Cooling process either in
air or oil in order to air or oil in order to air or oil in order to
reach 65
C in 1–2 h. reach 65
C in 1–2 h. reach 65
C in 1–2 h.
Cryogenic treatment for Cryogenic treatment for Cryogenic treatment for
1 h at 75
C and final 1 h at 75
C and final 1 h at 75
C and final
warming to room warming to room warming to room
temperature temperature temperature
Age Age hardening can be carried out between 475 and 488
C as a function of the
hardening final desired properties. Age-hardening processes carried out below 468
C
result in a material with very low toughness values

deformation during the metal forming process (Prasad and Wanhill 2017). Joining
processes, e.g., welding, are not a problem for AerMet alloys, and no preheating is
generally required (Campbell 2008). AerMet steels can undergo machining,
although the process might be rather challenging and requires tungsten carbide
tools (Prasad and Wanhill 2017).
As concerns age-hardening treatments, the optimum temperature ranges from
450
C to 600
C (Graves 1994; Shi et al. 2016). In Table 5.11 the main indications
for the heat treatments of each grade of AerMet alloys are provided.
AerMet alloys are very sensitive to the age-hardening treatments, as concerns
microstructure and resulting properties. For this reason, specific standard treatments
have been developed.
Figure 5.21 confirms that the optimum combination of strength and fracture
toughness is achieved with an age-hardening treatment carried out at 482
C,
resulting in carbide precipitation. This transformation is accompanied by the forma-
tion of a reverted austenite film at the martensite lath boundaries. The combination of
M2C carbides and the reverted austenite film present along the boundaries of
martensite laths are responsible for the optimum balance of mechanical strength
and toughness/ductility (Shi et al. 2016). The heat treatments of AerMet steels
should be performed under vacuum, inert atmosphere, or salt bath due to the
decarburization tendency of the material, a feature that they share with other high-
strength steels with relatively high carbon concentration (Prasad and Wanhill 2017).
5.4 Steels for Aerospace Applications 229

Fig. 5.21 (a) Tensile properties; (b) fracture toughness of AerMet 100 as a function of the
tempering temperature. RA % reduction in area, UTS ultimate tensile strength, YS yield strength,
EL % elongation to fracture. (Redrawn from Shi et al. 2016)

Table 5.12 Mechanical properties of AerMet alloys (Novotny and Maurer 2007)
482
C aged 454
C aged AerMet AerMet
AerMet 100 AerMet 100 310 340
UTS [MPa] 1978 2082 2172 2427
Yield strength [MPa] 1696 1779 1896 2165
% Elongation 16.0 14.0 14.5 11.0
% Reduction in area 67.0 64.0 63.0 55.0
Charpy V-notch impact test 47.5 40.7 27.1 14.9
[J]
Fracture toughness 132 109 71.4 35.2
[MPa√m]
Density [g/cm3] 7.889 7.889 7.972 7.861
Fatigue strength 107 cycles 945 944.6 1034.2 986
[MPa]
Specific strength [(MPa 251 264 272 309
cm3)/g]

The reference values of the main mechanical properties of aged AerMet alloys are
listed in Table 5.12.
AerMet ductility is generally higher than that of maraging steels. As concerns
hardness, AerMet 100 steel exhibits values in between 53 and 54 HRC, 55 HRC for
AerMet 310, and 56.5 HRC for AerMet 340 (Prasad and Wanhill 2017). As concerns
surface stability, AerMet steels do not contain a sufficient Cr concentration for
developing a protective self-passivating layer. For this reason, applications, includ-
ing aerospace ones, usually require a coating or a plating process, necessary to
prevent in-service corrosion. Like other high-strength materials, AerMet steels can
be subjected to SCC (stress corrosion cracking, see Sect. 8.1.6), with the exception
of AerMet 100 steel, which is generally regarded as better SCC resistant (Campbell
2008). The application of AerMet steels in the aerospace field is extended to a wide
range of components, as described in detail in Table 5.24 (Sect. 5.5.7).
230 5 Steels

5.4.3 Stainless Steels

The main alloying element in stainless steels is chromium, since the specific feature
of these materials, i.e., corrosion resistance, is determined by a passivation surface
layer based on chromium oxide. The first time chromium was identified as a new
chemical element was more than 200 years ago, being extracted from PbCrO4
(crocoite). The name chromium derives from the Greek word khrôma-atos, which
means color (Totten 2007). At the present time, the main source of metallic chro-
mium is FeCr2O4 (chromite). The process to extract chromium from this compound
was used for the first time in 1798 by the German chemist Tobias Lowitz. The low-C
ferrochromium processing (aluminothermic process), needed in view of the
low-carbon stainless steel production, was patented in 1891 in Germany by Hans
Goldschmidt.
Even Michael Faraday and James Stodart had been involved in this metallurgical
research on new alloy steels. In their paper “Experiments on the Alloy Steel, made
with a view to its improvements,” they described possible approaches for producing
better cutting tools and alloys less prone to oxidation (Stodart and Faraday 1820). In
1821, the French Pierre Berthier found out that chromium-alloyed iron had a better
corrosion resistance against acids (Cobb 2010). In the following years, the produc-
tion of stainless steels started quite simultaneously in different countries in Europe
and also in the USA (Totten 2007). Lion Guillet, professor of metallurgy and metal
processing at the Conservatoire National des Arts et Métiers, published in 1904 a
series of papers on Fe-Cr alloys characterized by carbon and chromium concentra-
tions comparable to modern ferritic and martensitic stainless steels, such as AISI
421, AISI 446, AISI 410, AISI 420, and AISI 440C (Cobb 2010). In 1913, the
English metallurgist Harry Brearley, head of the Brown-Firth Research Laboratories
in Sheffield, coordinated the production of the first heat of an alloy containing
0.24% C, 0.44% Mn, 0.22% Si, and 12.86% Cr. At the same time in the USA,
Frederick Becket and Christian Dantsizen were starting the production of alloys
containing from 0.007% to 0.015% C and from 14% to 16% Cr, obtained thanks to
the low-C ferrochromium precursor and produced via the aluminothermic process
(Totten 2007). The well-known 18–8 austenitic stainless steel first appeared in 1920,
although its wide-scale production just started a few years later. Nowadays, stainless
steels are classified into five main groups, according to their microstructure and
composition: austenitic, martensitic, ferritic, duplex, and PH (precipitation hard-
ened). The remarkable corrosion resistance, together with a very wide range of
mechanical properties, is the main reason for the massive use of these alloys in
many different fields: food industry, cryogenic applications, cutlery, engine valves,
bearings, medical applications, power generation, and aerospace industry, of course.
This paragraph presents the three main types of stainless steels used in the aerospace
field, austenitic, martensitic, and PH, highlighting their most important features,
particularly as concerns their properties and applications for aircraft components.
5.4 Steels for Aerospace Applications 231

5.4.3.1 Austenitic Stainless Steels

Benno Strauss and Edward Maurer at the Krupp Research Laboratories developed
the first commercial austenitic stainless steel in between 1909 and 1912. This very
year, a stainless steel with a “modern” composition (18% Cr 8% Ni) codenamed
AISI 304 was developed. At the beginning of their era, austenitic stainless steels
were prone to intergranular corrosion, i.e., sensitization (see Sect. 8.1.4), because of
the precipitation of the Cr-rich M23C6 carbides, occurring quite often upon cooling
the alloy from the melt. Two remediation approaches were adopted. The first was a
reduction in the carbon content, in the so-called L-type austenitic stainless steels,
such as AISI 304L and AISI 316L. The second approach was the addition of specific
alloying elements featuring a higher affinity to carbon than chromium. This is the
case of titanium and niobium, which resulted in the production of the stabilized
grades AISI 321 and AISI 347, respectively. Thanks to these successful develop-
ments, austenitic stainless steels are nowadays regarded as benchmark alloys, as far
as mechanical strength coupled with oxidation and corrosion resistance are
concerned (Cobb 2010; Demeri 2013; Totten 2007). Austenitic stainless steels can
be classified as Fe-based alloys containing from 16 to 26 wt% chromium and
8–35 wt% nickel. Other alloying elements are also present but in lower percentage.
Austenitic stainless steels belong to the AISI 2xx and 3xx groups (see Sect. 5.2.5).
According to the international classification, the AISI 2xx are Cr-Mn-Ni- or Fe-Cr-
Mn-Ni-N-based alloys. The other AISI 3xx series have a general composition
referring to the Fe-Cr-Ni-C system.
Austenitic stainless steels are essentially nonmagnetic in the annealed condition
and can be hardened by cold-working. Their chemical composition is designed so to
suppress the austenite to ferrite phase transformation on cooling, thus keeping the
steel fully austenitic also at room temperature (ASM International 1993; Demeri
2013; Hosford 2010; Totten 2007). The chemical compositions of the main austen-
itic stainless steels are given in Table 5.13, while the “family tree” of austenitic
stainless steels is illustrated in Fig. 5.22.
Chromium is essential for the formation of the protective passivation layer. Other
elements are added to improve the quality of the protective film, but no other element
can fully replace chromium in this role. Nickel is a γ-stabilizer, and therefore its
presence is absolutely necessary for these steels. It promotes re-passivation, partic-
ularly in chemically reducing environments. Furthermore, it enhances the mechan-
ical properties and the workability. Manganese can replace nickel partially: the 2xx
series contains manganese from 5.5 to 15.5 wt% and nickel up to 6 wt%. Concerning
the formability, it decreases upon decreasing the nickel and chromium content. Other
elements, such as niobium, tantalum, titanium, and also carbon, may introduce some
limitations due to their tendency both to form precipitates and to contribute to grain
refinement (see Appendix 4 for further details). For these reasons, the formability of
the AISI 321 and AISI 347 is lower than that of the AISI 302 and AISI 304. Nitrogen
has several positive effects on the properties of austenitic stainless steels: it increases
mechanical strength, enhances pitting corrosion resistance, and retards the formation
Table 5.13 Chemical composition of the principal austenitic stainless steels. Fe concentration balance (ASM International 1993)
232

Type C Mn Si Cr Ni P S N Others
AISI 201 0.15 5.5–7.5 1.0 16.0–18.0 3.5–5.5 0.06 0.03 0.25 –
AISI 202 0.15 7.5–10 1.0 17.0–19.0 4.0–6.0 0.06 0.03 0.25 –
AISI 205 0.12–0.25 14.0–15.5 1.0 16.5–18.0 1.0–1.75 0.06 0.03 0.32–0.40 –
AISI 301 0.15 2.0 1.0 16.0–18.0 6.0–8.0 0.045 0.03 – –
AISI 302 0.15 2.0 1.0 17.0–19.0 8.0–10.0 0.045 0.03 – –
AISI 303 0.15 2.0 2.0–3.0 17.0–19.0 8.0–10.0 0.20 0.15 min – Mo ¼ 0.6
AISI 303Se 0.15 2.0 2.0–3.0 17.0–19.0 8.0–10.0 0.20 0.06 – Se ¼ 0.15 min
AISI 304 0.08 2.0 1.0 18.0–20.0 8.0–10.5 0.045 0.03 – –
AISI 304H 0.04–0.10 2.0 1.0 18.0–20.0 8.0–10.5 0.045 0.03 – –
AISI 304L 0.03 2.0 1.0 18.0–20.0 8.0–10.5 0.045 0.03 – –
AISI 304LN 0.03 2.0 1.0 18.0–20.0 8.0–10.5 0.045 0.03 0.10–0.16 –
AISI 304N 0.08 2.0 1.0 18.0–20.0 8.0–10.5 0.045 0.03 0.10–0.16 –
AISI 309 0.20 2.0 1.0 22.0–24.0 12.0–15.0 0.045 0.03 – –
AISI 309S 0.08 2.0 1.0 22.0–24.0 12.0–15.0 0.045 0.03 – –
AISI 310 0.25 2.0 1.5 24.0–26.0 19.0–22.0 0.045 0.03 – –
AISI 310S 0.08 2.0 1.5 24.0–26.0 19.0–22.0 0.045 0.03 – –
AISI 316 0.08 2.0 1.0 16.0–18.0 10.0–14.0 0.045 0.03 – Mo ¼ 2.0–3.0
AISI 316F 0.08 2.0 1.0 16.0–18.0 10.0–14.0 0.20 0.10 min – Mo ¼ 1.75–2.5
AISI 316L 0.03 2.0 1.0 16.0–18.0 10.0–14.0 0.045 0.03 – Mo ¼ 2.0–3.0
AISI 316LN 0.03 2.0 1.0 16.0–18.0 10.0–14.0 0.045 0.03 0.10–0.16 Mo ¼ 2.0–3.0
AISI 316N 0.08 2.0 1.0 16.0–18.0 10.0–14.0 0.045 0.03 0.10–0.16 Mo ¼ 2.0–3.0
AISI 317 0.08 2.0 1.0 18.0–20.0 11.0–15.0 0.045 0.03 – Mo ¼ 3.0–4.0
AISI 317L 0.03 2.0 1.0 18.0–20.0 11.0–15.0 0.045 0.03 – Mo ¼ 3.0–4.0
AISI 321 0.08 2.0 1.0 17.0–19.0 9.0–12.0 0.045 0.03 – Ti ¼ 5  %C min
AISI 330 0.08 2.0 0.75–1.5 17.0–20.0 34.0–37.0 0.04 0.03 – –
AISI 347 0.08 2.0 1.0 17.0–19.0 9.0–13.0 0.045 0.03 – Nb ¼ 10  %C min
5 Steels
AISI 348 0.08 2.0 1.0 17.0–19.0 9.0–13.0 0.045 0.03 – Co ¼ 0.2
Nb ¼ 10  %C min Ta ¼ 0.10
Nitronic 50 0.06 4.0–6.0 1.0 20.5–23.5 11.5–13.5 0.040 0.03 0.2–0.4 Mo ¼ 1.5–3.0
Nb ¼ 0.1–0.3
V ¼ 0.1–0.3
Nitronic 60 0.10 7.0–9.0 3.5–4.5 16.0–18.0 8.0–9.0 0.040 0.03 0.08–0.18 –
253MA 0.05–0.10 0.8 1.4–2.0 20.0–22.0 10.0–12.0 0.040 0.03 0.14–0.20 Al ¼ 1.0
Ce ¼ 0.03–0.08
5.4 Steels for Aerospace Applications
233
234 5 Steels

Fig. 5.22 Austenitic stainless steels family tree, AISI, and UNS designation (ASM International
1993)

of the detrimental Cr-Mo-based σ phase. In association with molybdenum, nitrogen

enhances also the halide pitting corrosion resistance. Molybdenum is interesting not
only for its solid solution strengthening of the alloy but also because it promotes the
formation of the protective passivation layer (see Appendix 4).
The resistance to stress corrosion cracking (SSC, see Sect. 8.1.6) is improved by
the addition of silicon and copper. Selenium and sulfur are added to improve the
steel machinability. The majority of stainless steel, including austenitic ones, is
produced using scrap based metallurgy processes. The typical melting practice
implies electric arc furnace (EAF, see Sect. 5.3.1) and argon oxygen decarburization
(AOD see Sect. 5.3.3). The casting of austenitic stainless steels is based on contin-
uous casters and ingot casting processes. Aerospace applications often require
products subjected to double melting processes: the first is conducted via EAF and
AOD; then the starting ingot becomes the electrode for ESR (see Sect. 5.3.5) or for
VAR (see Sect. 5.3.6). Double melting leads to a much cleaner material, in terms of
inclusions and homogeneity, and produces ingots with improved hot-workability
(ASM International 1993; ASM International 2008).
Austenitic stainless steels are typically supplied in the annealed condition, since
they are not strengthened by thermal treatments but rather by cold-working. Cooling
from the annealing temperature (see Sect. 5.2.1) must be sufficiently rapid to avoid
the trajectory to cross the carbide transformation curves in the relevant TTT diagram
(Fig. 5.23); otherwise a precipitation of carbides occurs (e.g., Cr23C6), with poten-
tially negative effects on the corrosion resistance of the steel, determined by chro-
mium depletion in the alloy (see Sect. 8.1.4).
5.4 Steels for Aerospace Applications 235

Fig. 5.23 Main thermal treatments and relevant transformations occurring in austenitic stainless.
(Redrawn from Totten 2007)

Austenitic stainless steels do not form martensite on cooling because their Ms

(martensite start) is well below room temperature. However, martensite can still be
induced by mechanical strain (see Sect. 9.5.1), the so-called strain-induced martens-
ite (SIM). Two types of martensite can occur, α0 (bcc and ferromagnetic) and ε (hcp,
non-ferromagnetic), depending on the composition, stacking fault energy, deforma-
tion rate, and temperature range (Kaieda and Oguchi 1985; Totten 2007). The
transitions from γ to α0 and from γ to ε involve 2.57% volume increase or 0.8%
shrinkage, respectively. The temperature below which martensite can be induced by
mechanical strain is defined as Md (see also Sect. 9.5.1). It is given by empirical
expressions, based on the chemical composition of the alloy and on the percentage of
strain-induced martensite, like Eq. 5.6, referring to a 30% true tensile strain needed
to obtain 50% α0 martensite (Angel 1954):

M d ð30=50Þ ¼ 413  13:7ð%CrÞ  9:5ð%NiÞ  8:1ð%MnÞ

ð5:6Þ
 18:5ð%MoÞ  9:2ð%SiÞ  462ð%C þ %NÞ

The annealing of stress induced martensite may cause its reversion to austenite for
treatments carried out at temperatures and times lower than those required by
recrystallization treatments (Demeri 2013, Totten 2007). The most common heat
treatments for these steels are solution annealing, stabilization annealing, and stress
236 5 Steels

Table 5.14 Annealing treatments for austenitic stainless steels (Totten 2007)
Treatment Temperature range
Solution annealing 1010–1120
C. For grades that are more subjected to abnormal grain growth
and secondary recrystallization (e.g., stabilized grades): 1065
C max
Stabilization 845–955
C
annealing
Stress relief 425–550
C
annealing

Table 5.15 Typical mechanical properties at room temperature of some austenitic stainless steels
used in the aerospace field (ASM International 1993)
AISI Tensile strength Yield strength Elongation Reduction in area
type [MPa] [MPa] [%] [%]
304 640 235 76.0 82.0
304L 660 405 78.0 81.0
310 585 340 50.0 76.0
316 595 275 60.0 –
321 675 430 55.0 79.0
347 670 340 57.0 76.0

relief annealing (see Sect. 5.2.1). In Table 5.14 the description for the most important
heat treatment parameters for austenitic stainless steels is provided.
Important room temperature mechanical properties of some austenitic stainless
steels used in the aerospace field (see also Table 5.25) are listed in Table 5.15. These
steels do not exhibit any ductile-to-brittle transition, and for this reason, they are
extensively used in cryogenic applications, even down to temperatures below
200
C. Actually, the yield and ultimate tensile strengths of austenitic steels
increase substantially as the testing temperature decreases. Obviously, cold-working
greatly increases the yield strength while reduces the alloy ductility (ASM Interna-
tional 1993).
High-temperature plastic deformation is generally not an issue for austenitic
stainless steels. Most grades can be subjected to forging operations in a wide
range of temperatures above 925
C, with no equipment damage and degrading
quality of the material. Chemical composition differences affect the generally good
weldability of austenitic stainless steels. The carbon content plays an important role
too, since it influences carbide precipitation in the HAZ (heat-affected zone). AISI
304L is known to minimize this deleterious phenomenon when compared with
standard AISI 304. Grades with improved machinability, the so-called free-machin-
ing steel grades, contain phosphorous, sulfur, and selenium, that, on the other hand,
reduce weldability since their presence can cause severe hot cracking of the metal.
The weldability of AISI 318, AISI 321, AISI 347, and AISI 348 is guaranteed by
alloying with Ti, Nb, and Ta that preventing the formation of chromium carbides
avoids sensitization phenomena (see Sect. 8.1.4).
5.4 Steels for Aerospace Applications 237

Fig. 5.24 Martensitic stainless steels. The family tree of the martensitic stainless steels. AISI and
UNS designation (ASM International 1993)

In addition to sensitization, austenitic stainless steels are prone to SCC in chloride

containing environments. In order to decrease the susceptibility to SCC, a high
nickel content is recommended. In high-temperature applications, austenitic stain-
less steels with 10–20% Ni and 20–25% Cr can be used, still exhibiting acceptable
performances as concerns oxidation and mechanical properties.

5.4.3.2 Martensitic Stainless Steels

The very first commercial martensitic stainless steel was probably developed by
Harry Brearley in 1913. The present family of stainless steels is based on the Fe-Cr-
C system, where Cr ranges in between 11.5 and 18 wt %, while carbon content is
kept between 0.1 and 1.2 wt %, with little or no additions of other alloying elements
(Totten 2007). Martensitic stainless steels belong to the AISI 4xx series (see Sect.
5.2.5) and are ferromagnetic and resistant to corrosion in relatively mild environ-
ments. Their corrosion resistance is lower if compared to that of austenitic and
ferritic grades. Most commonly used martensitic grades are those with low-carbon
content and up to 12 wt% Cr. AISI 410 can be regarded as the reference, general
purpose martensitic stainless steel. It is possible to classify the standard martensitic
stainless steel grades according to the scheme in Fig. 5.24.
The application of martensitic stainless steels in the aerospace field is limited to
specific components that benefit from the combination of specific properties of these
alloys. Of course, tribological (i.e., wear resistance) features together with remark-
able tensile strength, as well as their corrosion resistance, are the main reasons for
their adoption (see Table 5.26 in Sect. 5.5.7). Martensitic stainless steels can be
238 5 Steels

Fig. 5.25 Projection of the ternary Fe-Cr-C diagram on the temperature vs. %Cr plane. (Redrawn
from Totten 2007)

subdivided into three main subgroups, according to their carbon content and relevant
applications: low-C steels for gas turbines, medium-C steels for cutlery, and high-C
grades for wear-resistant applications. Martensitic stainless steels with high carbon
content are those able to reach the highest strength. The carbon percentage influences
the microstructure of the alloy. Low-C alloys typically show the formation of
martensitic needlelike microstructure; medium-C alloys usually have a fine martens-
itic structure, while high-C grades show ultrafine martensitic structure with primary
carbides. Ni-containing grades (e.g., AISI 431) and high-C grades (e.g., AISI 440C)
may show the presence of a high volume fraction of retained austenite after
quenching, due to a very low Mf (martensite finish), that may fall in the cryogenic
range (Totten 2007). Cr is kept as high as possible in order to attain the maximum
corrosion resistance. The concentrations of chromium and carbon must be balanced
to ensure a martensitic structure after a quenching hardening treatment (see Sect.
5.2.2). As chromium level is increased, carbon has to be increased as well, in order to
stabilize the austenite field (see Fig. 5.25). If other elements other than C and Cr are
present, their concentration does not usually exceed 2–3 wt%. This is the case of
nickel, a γ-stabilizing element (see Fig. 5.26). Martensitic steels containing nickel
(e.g., AISI 414 and AISI 431) will also feature lower C content. Therefore, they will
exhibit higher toughness and better corrosion resistance. A similar effect can be
obtained with nitrogen, another γ-stabilizer. Its influence on the γ phase stability
5.4 Steels for Aerospace Applications 239

Fig. 5.26 Projection of the ternary Fe-Cr-C diagram on the temperature vs. %Cr (in wt%) plane
and influence of Ni content. (Redrawn from Totten 2007)

field is shown by Fig. 5.27. The chemical compositions of the most common
martensitic stainless steels are listed in Table 5.16.
High chromium contents have also influence on the tempering response of the
material, since chromium slows down the softening, allowing for the removal of
residual stresses without any appreciable decrease in the strength and hardness.
Some martensitic grades contain up to 1% Mo for the improvement of both
corrosion resistance and mechanical properties. Nb, Si, V, and W are added in
order to modify the tempering response of the material after the hardening process.
Sulfur and selenium are added for improving machinability (ASM International
1993; Campbell 2008; Totten 2007). Martensitic stainless steels may be subjected
to the precipitation of several carbides: MC, M2C, M3C, M7C3, and M23C6. σ phase
precipitation is usually not an issue for this class of stainless steels (Totten 2007).
Concerning melting and casting practices, martensitic stainless steels can be
produced via EAF + AOD route or, in some cases, via arc melting, AOD refining,
and final VAR process. This approach is very much recommended for those appli-
cations requiring homogeneity and cleanliness, in order to ensure a safe fatigue life.
This aspect can be particularly interesting for bearing systems for aerospace uses
(e.g., AISI 440C). Martensitic stainless steels can be produced via ingot casting or
continuous casting, depending on the chemistry and on the final product
specifications.
Martensitic stainless steels can be heat-treated in the same way as plain carbon,
low alloy, and alloy steels. This implies the presence of at least three main stages:
austenitization, quenching, and tempering. Heat treatment of these alloys is very
240 5 Steels

Fig. 5.27 Shift of the γ loop with carbon and nitrogen content. (Redrawn from Totten 2007)

Table 5.16 Chemical composition of martensitic stainless steels using the AISI code. Fe concen-
tration balance (ASM International 1993)
Type C Mn Si Cr Ni P S Others
403 0.15 1.0 0.5 11.5–13.0 – 0.04 0.03 –
410 0.15 1.0 1.0 11.5–13.5 – 0.04 0.03
410S 0.08 1.0 1.0 11.5–13.5 0.60 0.04 0.03 –
414 0.15 1.0 1.0 11.5–13.5 1.25–2.50 0.04 0.03 –
416 0.15 1.25 1.0 12.0–14.0 – 0.06 0.15 min Mo ¼ 0.60
416Se 0.15 1.25 1.0 12.0–14.0 – 0.06 0.06 Se ¼ 0.15 min
420 0.15 min 1.0 1.0 12.0–14.0 – 0.04 0.03 –
420F 0.15 min 1.25 1.0 12.0–14.0 – 0.06 0.15 min Mo ¼ 0.60
422 0.20–0.25 1.0 0.75 11.5–13.5 0.5–1.0 0.04 0.03 Mo ¼ 0.75–1.25
W ¼ 0.75–1.25
V ¼ 0.15–0.30
431 0.20 1.0 1.0 15.0–17.0 1.25–2.50 0.04 0.03 –
440A 0.60–0.75 1.0 1.0 16.0–18.0 – 0.04 0.03 Mo ¼ 0.75
440B 0.75–0.95 1.0 1.0 16.0–18.0 – 0.04 0.03 Mo ¼ 0.75
440C 0.95–1.20 1.0 1.0 16.0–18.0 – 0.04 0.03 Mo ¼ 0.75
5.4 Steels for Aerospace Applications 241

Table 5.17 Typical Type Austenitizing [
C] Tempering [
C]

austenitizing and tempering
403 925–1010 150–370
temperatures for martensitic
AISI stainless steels used in 410 925–1010 150–370
the aerospace field (Totten 431 980–1065 205–370
2007) 440C 1010–1065 150–370

important, since the material is purchased either in the annealed or in the fully treated
(i.e., hardened and tempered) condition. Table 5.17 presents some typical values for
the austenitizing and tempering temperatures for martensitic stainless steels used in
the aerospace field.
The high chromium content leads to a significant shift of the time-temperature-
transformation curves to the right (see Fig. 5.28) (Verhoeven 2007).
For this reason, martensitic stainless steels are said to be highly hardenable, using as
cooling media either air or oil. These alloys exhibit a much lower thermal conductivity
if compared with plain carbon and low-alloy steels. This results in higher-temperature
gradients over the cross section and higher probability of warpage and cracking, even
on heating. Moreover, metal carbides tend to dissolve slowly in the austenite, and
carbide density is affected by the austenitizing temperature in such a way that an
increase in this temperature determines a decrease in the carbide density. An example
of this behavior is given in Fig. 5.29 where the carbide density vs. austenitizing
temperature is given for the AISI 440C material (Verhoeven 2007).
Since these grades harden when cooled off the mill after hot-processing, the
material requires an annealing at 650–760
C for about 4 h. This treatment differs
from full annealing that is usually carried out in between 800
C and 880
C and
followed by a slow cooling in the furnace down to about 450
C, before air cooling.
Oil quenching is preferred for avoiding carbides precipitation, while air cooling is
indicated to avoid part distortion. Low-C grades (e.g., AISI 410) show a relatively
high Ms and Mf temperatures, 350
C and 250
C, respectively, and are fully
martensitic at room temperature. High-C grades may retain a relatively high con-
centration of austenite down to room temperature, thus requiring cryogenic treatment
to avoid this phenomenon (vide infra).
The evolution of retained austenite as a function of the austenitizing temperature
is given in Fig. 5.30 for the AISI 440C steel. Figure 5.31 shows that for the AISI
440C at the austenitizing temperature of 1100
C, a significant increase of the
C-saturation line occurs. This requires a higher Cr concentration in order to obtain,
upon quenching, a martensite with a Cr content above the threshold value of 13 wt%,
necessary to be in the self-passivation range. This higher Cr concentration results in a
lower both Ms and Mf. Consequently, a higher amount of retained austenite is
present in the quenched alloy. Therefore, the completion of martensite transforma-
tion requires cryogenic quenching treatments, down to temperatures as low as
96
C (ASM International 1993; Campbell 2008; Totten 2007; Verhoeven 2007).
242 5 Steels

Fig. 5.28 Isothermal transformation diagram for AISI 410 stainless steel. (Redrawn from
Verhoeven 2007)

Fig. 5.29 Carbide density as a function of the austenitizing temperature for AISI 440C. (Redrawn
from Verhoeven 2007)
5.4 Steels for Aerospace Applications 243

Fig. 5.30 Evolution of retained austenite as a function of the austenitizing temperature for room-
temperature quench for the AISI 440C. (Redrawn from Verhoeven 2007)

Fig. 5.31 The Fe-Cr-C isothermal sections at 1000 and 1100
C for AISI 440C. (Redrawn from
Verhoeven 2007)
244 5 Steels

Table 5.18 Mechanical properties of several martensitic stainless steels (Campbell 2008)
Tensile Yield Elongation
AISI strength strength in 50 mm Reduction
type Condition [MPa] [MPa] [%] in area [%] Hardness
403 Annealed 517 276 35.0 70.0 82 HRB
Tempered 765 586 23.0 67.0 97 HRB
410 Oil quenched from 1089 1006 13.0 70.0 –
980
C and tempered at
540
C
414 Annealed 793 621 20.0 60.0 235 HB
Cold drawn 896 793 15.0 58.0 270 HB
Oil quenched from 1006 800 19.0 58.0 –
980
C and tempered at
650
C
420 Annealed 655 345 25.0 55.0 195 HB
Annealed and cold 758 690 14.0 40.0 228 HB
drawn
431 Annealed 862 655 20.0 55.0 260 HB
Annealed and cold 896 758 15.0 35.0 270 HB
drawn
Oil quenched from 834 738 20.0 64.0 –
980
C and tempered at
650
C
440C Annealed 758 448 14.0 25.0 97 HRB
Annealed and cold 862 690 7.0 20.0 260 HB
drawn
Hardened and tem- 1965 1896 2.0 10.0 580 HB
pered at 315
C

Mechanical properties of selected martensitic stainless steels are summarized in

Table 5.18 for different temper conditions.
Concerning processing, martensitic stainless steels afford several options. Due to
their mechanical properties and structures, hot-working must be performed when
they are in the austenitic field, since the material is close to a ductility maximum.
Hot-working temperature, typically in excess of 1100
C, for both forging and
rolling processes, must be selected according to the chemical composition of the
specific steel grade. Other process parameters, such as strain rate, temperature during
hot-working, are crucial for the fabrication of flawless material.
Martensitic stainless steels can be machined, but the process has to be carried out
with some precaution. In comparison with carbon steels, the required power for the
process is higher, the cutting speed must be lower, and particular care must be taken
to ensure correct cooling and adequate lubrication.
The weldability of martensitic stainless steels depends on their hardenability,
considering that the metallurgical changes occurring during hardening may also take
place during welding, regardless of the initial condition of the alloy: hardened,
annealed, or tempered. The HAZ (heat-altered zone) is therefore characterized by
5.4 Steels for Aerospace Applications 245

a reduced toughness and becomes more susceptible to cracking. Pre-heating is

recommended, and it is usually performed at temperatures in the 205
C–315
C
range (ASM International 1993).
The high carbon concentration in martensitic stainless steels results in a lower
corrosion resistance of these alloys if compared to austenitic and ferritic grades,
since M23C6 (M ¼ Cr, Fe) carbide may form, thus reducing chromium concentration
in the alloy, even below the sensitization limit (Cr 13 wt%). Martensitic stainless
steels are also prone to SCC (see Sect. 8.1.6) when exposed to marine environments,
particularly in the presence of high chloride concentrations (ASM International
1993; Verhoeven 2007).

5.4.3.3 PH Stainless Steels

Until the 1930s there were three classes of stainless steels only: austenitic, martens-
itic, and ferritic. The first precipitation hardenable stainless steel was developed in
Luxemburg in 1929 thanks to the work of William Kroll, who is nowadays notorious
mainly for the invention of the first commercial extraction process for titanium (see
Sect. 4.1) and zirconium. Precipitation hardening stainless steels were introduced
into the market more or less at the same time on both sides of the Atlantic. In
Sheffield, Firth Vickers was working on FV520B, while in 1948 Armco (USA) first
introduced in the market the well-known 17-4PH, that was followed by 17-7PH and
subsequently by 15-7Mo. Other PH stainless steels, named A286, AM350, and
AM355, were developed by Allegheny Ludlum Steel Company. Finally, the Car-
penter Steel Company developed the Custom 450, Custom 455, and Custom
630 (Cobb 2010; Farrar 2004).
Precipitation hardening stainless steels (PHSS or simply PH) are a family of
alloys that exhibit a secondary hardening for the precipitation of compounds that are
not carbides. PH stainless steels can be defined as a Cr-Ni steels, containing specific
alloying elements, that induce precipitation hardening. Indeed, PH steels were
introduced for both overcoming the limits of solution strengthening of the austenitic
stainless steels and the low toughness and ductility of high-C martensitic grades. The
PH steels retain good mechanical properties in association with an excellent corro-
sion resistance, so to be attractive for several aircraft components with quite a broad
range of requirements, as concerns service temperatures; mechanical strength, also in
creep and fatigue regime; and corrosion resistance (see Table 5.27 in Sect. 5.5.7).
This section is devoted to the description of the main families of PH stainless steels,
with a particular focus for those features interesting in aerospace applications. These
alloys can be grouped into three main classes, according to the parent phase from
which precipitation hardening occurs: martensitic, semiaustenitic, and austenitic.
Martensitic PHSS are also called single-treatment alloys, since their peak strength
condition is attained after a single low-temperature aging treatment, carried out after
a solutioning annealing. The microstructure is austenitic at the annealing temperature
and becomes martensitic on cooling, so that a martensite matrix is attained at the end
of the final aging treatment. Semiaustenitic PHSS are also called double-treatment
246 5 Steels

alloys. Their starting structure is austenitic, and the two-stage heat treatment includes
austenite conditioning (heating at a temperature that allows martensite formation
on cooling) and final age hardening after intermediate cooling. Austenitic PHSS
feature an austenitic structure in both the solution-annealed and age-treated
conditions. Strength improvement of these alloys is achieved by means of single
or double precipitation hardening. The chemical composition of PHSS is the key
factor for the development of their final microstructure and mechanical properties.
Classification and chemical composition of the main PH stainless steels are shown in
Table 5.19.
Figure 5.32 shows the Cr-Ni composition of a martensitic (17-4PH), semiaustenitic
(17-7PH), and austenitic (A286) PHSS. 17-4PH lies in the region where rapid cooling
produces martensite, and precipitation hardening occurs after a final aging treatment.
17-7PH requires a double treatment, since the first is needed to produce an austenite
matrix with minor concentrations of δ-ferrite, while the second is needed for the
martensitic transformation of the alloy matrix. A286 instead lies in the stable austenite
region, and this implies that precipitation hardening occurs in a fully austenitic matrix
(Campbell 2008; Everhart 1971; Verhoeven 2007; Totten 2007).
Precipitation hardening involves the formation of intermetallics, carbides,
nitrides, and carbonitrides. Cu-, Ni-, and Al-rich intermetallics, with composition
of the Ni3M type (M ¼ Al, Mo, Nb, Ti), are the most common phases (Campbell
2008; Chandler 1996; Farrar 2004; Mouritz 2012). Titanium and niobium tend to
form also the very stable MCN and MN carbonitrides and nitrides (Miller and Burke
1991; Yeli et al. 2017). CrN precipitates have been reported to form in 17-4PH steel,
along dislocation lines and microstructural defects preferentially (Yeli et al. 2017).
However, the complex transformation intervening during precipitation treatments of
the 17-4PH is still attracting major research efforts to be fully characterized.
The EAF + AOD process is commonly used for PHSS. After melting and
refining, the material can be either continuously or ingot cast, depending on the
grade and on the subsequent production stages, i.e., forging or hot-rolling. For
specific applications, including aerospace, ESR, ESRR, or VAR processes can be
used for improving the quality of the material in terms of microstructural homoge-
neity and segregation reduction. The 17–4PH steel can be produced with
EAF + ESR/ESRR; for the martensitic 13-8Mo, a combination of VIM and VAR
processes is a common choice. Different production routes are recommended for
5-5PH martensitic PH stainless steel: EAF + AOD, EAF + AOD + ESR, or
EAF + AOD + VAR. Finally, the semiaustenitic 17-7PH is EAF melted and AOD
refined.
Figure 5.33 shows the typical microstructure of 17-4PH and A286 in different
conditions of treatment. Figure 5.33a shows the martensitic microstructure of the
17-4PH steel. The microstructure of the semiaustenitic 17-7PH would look rather
similar, since the Al- and Ni-rich intermetallic precipitates, forming in the martens-
ite, would not change that much the situation as compared to 17-4PH. Figure 5.33b
shows the microstructure of the austenitic A286, after solution annealing and aging
treatments: the austenitic grains can be clearly observed.
Table 5.19 Chemical composition of some important PHSS
Grade Designation Type %C %Mn %Ni %Cr %Mo %Si %Others
17-10P – A 0.07 0.75 10.0 17.0 – 0.5 P ¼ 0.3
17-14CuMo AISI 653 A 0.12 0.75 14.0 16.0 2.5 0.5 Cu ¼ 3.0
Nb ¼ 0.4
Ti ¼ 0.3
A286 AISI 660 A 0.08 2.0 24.0–27.0 13.5–16.0 1.0–1.5 1.0 Al ¼ 0.35
B ¼ 10–100 ppm
Ti ¼ 1.90–2.35
V ¼ 0.1–0.5
5.4 Steels for Aerospace Applications

HNM – A 0.3 3.5 9.5 18.5 – – –

13-8Mo – M 0.05 0.2 7.5–8.5 12.25–13.25 2.0–2.5 0.1 Al ¼ 0.90–1.35
N ¼ 0.01
15-5PH – M 0.07 1.0 3.5–5.5 12.5–14.0 – 1.0 Cu ¼ 2.5–4.5
Nb ¼ 0.15–0.45
17–4PH AISI 630 M 0.07 1.0 3.0–5.0 12.5–17.5 – 1.0 Cu ¼ 3.0–5.0
Nb ¼ 0.15–0.45
Custom 450 – M 0.05 1.0 5.0–7.0 14.0–16.0 0.5–1.0 1.0 Cu ¼ 1.25–1.75
Nb  8  %C
Stainless W AISI 635 M 0.08 1.0 6.0–7.5 16.0–17.0 – 0.5 Al ¼ 0.4
Ti ¼ 0.4–1.2
15-7Mo AISI 632 SA 0.09 1.0 6.50–7.75 14.0–16.0 2.0–3.0 1.0 Al ¼ 0.75–1.50
17-7PH AISI 631 SA 0.09 1.0 6.50–7.75 16.0–18.0 – 1.0 Al ¼ 0.75–1.50
AM355 AISI 634 SA 0.13 0.95 4.3 15.5 2.75 – N ¼ 0.1
A austenitic, M martensitic, SA semiaustenitic
Fe concentration balance (Campbell 2008; Chandler 1996; Mouritz 2012; Totten 2007; Verhoeven 2007)
247
248 5 Steels

Fig. 5.32 Position of A286, 17-4PH, and 17-7PH steel in the relevant portion of the Fe-Cr-Ni
ternary phase diagram. The diagram indicates the position of the three PH stainless steels after
cooling from the 1100
C austenitizing temperature. (Redrawn from Verhoeven 2007)

Fig. 5.33 (a) 17–4PH microstructure after 1 h aging at 470
C (Viswanathan et al. 1988,
reproduced with permission of the Publisher); (b) A286 microstructure after annealing at 930
C
and subsequent aging at 730
C. (De Cicco et al. 2004, reproduced with permission of the
Publisher)
5.4 Steels for Aerospace Applications 249

Fig. 5.34 Heat treatments for three types of PHSS: martensitic 17-4PH, semiaustenitic 17-7PH,
and austenitic A286 (Campbell 2008; Chandler 1996; Totten 2007)

Since there are important constitutional differences among the three families of
PHSS, heat treatments conducted for their microstructural refinement and property
optimization must be designed accordingly. Regarding the heat treatment code, the
solution treated condition is designed by “A” (see scheme in Fig. 5.34), while the
aged condition is identified by “H” followed by the treatment temperature expressed
in Fahrenheit degrees. Figure 5.34 shows a few examples of heat treatments for the
main representatives of each type of PH steel: martensitic, semiaustenitic, and
austenitic.
Martensitic PHSS are solution annealed and transform on cooling, leading to a
fully martensitic structure. Afterward, aging is performed to induce the desired phase
precipitation, without the need for austenite conditioning, which is necessary instead
for semiaustenitic PHSS. The solution annealing (condition A) of the martensitic
17-4PH (see Fig. 5.34) is conducted at 1040
C, usually followed by air cooling,
resulting in the formation of a fully martensitic structure, i.e., no retained austenite,
since Mf is above room temperature, although some δ-ferrite may be present. Age
hardening, to obtain peak properties, is performed at 480
C (900
F), the so-called
H900 condition. Another common heat-treated condition is H1095 (590
C for 4 h).
Whereas H900 is used to obtain peak hardness and strength, the H1095 improves
ductility and toughness, through a suitable increase of the aging temperature. In both
these main aging conditions, the strengthening phases are NbN-, CrN-, and Cu-rich
precipitates (CRPs), these latter forming to different extent in the H900 and H1050
materials and providing the main strengthening contributions.
250 5 Steels

Semiaustenitic PHSS are also subjected to a solution annealing process

(condition A, see Fig. 5.34). With reference to 17-7PH (Fig. 5.34), solution
annealing is performed at 1065
C, resulting in the formation of thermodynamically
unstable austenite. Austenite conditioning, necessary to induce the subsequent
martensite formation, can be performed at various temperatures, that will influence
the kinetics of the transformation. At high temperatures (e.g., 955
C, condition
A1750), a reduced carbide precipitation takes place, and this leads to the need of a
subzero treatment (73
C, condition R100) to obtain the formation of martensite.
Otherwise, austenite conditioning at 760
C for 90 min (condition A1400) is
sufficient and does not require subzero cooling for the conversion into martensite,
since Cr and C depletion from the matrix takes place as a consequence of precipi-
tation, mainly of the M23C6 type of carbides. The material is then cooled down to
about 15
C and held for 30 min leading to condition T. In case this procedure is
followed by aging at 565
C for 90 min with subsequent cooling process down to
room temperature, the final condition is named TH1050. During age-hardening
treatment of 17-7PH, Al in the martensitic matrix combines with Ni, thus strength-
ening the alloy through the precipitation of the Ni3Al and NiAl intermetallics (Totten
2007). It is worth reminding that in these alloys too, like in austenitic stainless steels
(see Sect. 5.4.3.1), martensite can also be induced by mechanical strain (SIM, see
Sect. 5.4.3.1).
The heat treatments of austenitic PHSS usually involve a solution annealing
followed by an aging. Figure 5.34 shows the typical sequence used for solution
annealing and age hardening of the austenitic A286 steel. Solution annealing is
performed at 900–1050
C. For high stress rupture properties, the recommended
treatment temperature is 980
C, and cooling is performed either in oil or water,
although no martensitic transformation would actually occur. Aging treatment is
carried out in the 700–760
C temperature range and leads to the precipitation of
carbides and γ0 phase. For γ0 to form, the chemical composition of the alloy should
contain a sufficient amount of aluminum and titanium, as it is in iron- and nickel-
based superalloys (see Sects. 6.3.1 and 6.4.3 and Appendix 6 for further details on γ0 )
(Campbell 2008; Chandler 1996; Totten 2007; Verhoeven 2007; Yeli et al. 2017). A
two-step aging treatment can also be used, to control the size and the distribution of
the γ0 precipitates, thus increasing the alloy yield strength (see Sect. 6.6). This kind
of aging treatment is also used to control the morphology and size of grain boundary
carbides, particularly the MC type.
Both martensitic and semiaustenitic PHSS after aging are able to reach tensile
strength that exceeds 1300 MPa, although they also exhibit a significant ductility
decrease when shifting from the annealed to the aged condition. Austenitic PHSS are
not characterized by that great improvement in tensile strength after aging, reaching
maximum values of around 950 MPa, again with a reduced ductility. However, in
austenitic steels this effect is markedly lower as compared to martensitic and
semiaustenitic PH steels. In Table 5.20 the mechanical properties of selected
PHSS, as resulting from different heat treatments, are quoted.
In addition to heat treatments, cold-working can be used to modify further the
microstructure and, thereby, the mechanical properties of semiaustentic PHSS.
5.4 Steels for Aerospace Applications 251

Table 5.20 Properties of some PHSS as resulting from the specified aging treatments (Campbell
2008)
Tensile strength Yield strength Elongation
Type Treatment [MPa] [MPa] [%] Hardness
13-8Mo Aged at 1520 1410 6.0–10.0 45 HRC
510
C min.
15-5PH Aged at 1310 1170 – –
480
C
17-4PH Aged at 1310 1170 5.0–10.0 40 HRC
480
C min.
Custom Aged at 1240 1170 3.0–5.0 40 HRC
450 480
C min.
Custom Aged at 1530 1450 4.0 44 HRC
455 510
C min.
15-7Mo Aged at 1650 1590 1.0 46 HRC
480
C min.
17-7PH Aged at 1450 1310 1.0–6.0 43 HRC
510
C min.
AM355 Aged at 1170 1030 12.0 37 HRC
540
C min.
A286 Aged at 860–965 655 4.0–15.0 24 HRC
730
C min.

Table 5.21 Mechanical properties of the semiaustenitic 17-7PH resulting from heat treatments and
cold-working processes (Totten 2007)
Tensile Yield
strength strength Elongation
Condition Treatment [MPa] [MPa] [%]
A 1040–1080
C, air or water cooled <1030 <380 18.0
TH1050 A + 90 min at 955
C air cooled +60 min 1170–1240 960–1030 5.0–6.0
at room temperature + 90 min at 565
C
RH950 A + 10 min at 955
C + 8 h at 1370–1450 1240–1310 4.0–5.0
73
C + 60 min at 510
C
C A + 50% cold-working >1400 >1200 1.0
CH900 C + 60 min at 480
C >1650 >1580 1.0

Table 5.21 illustrates how some heat treatments and cold-working processes may
determine specific mechanical properties of the 17-7PH steel (Campbell 2008;
Everhart 1971; Totten 2007).
Hot-working is another important processing option for PHSS, that can be forged
and hot-rolled. Heating cycles in furnaces prior to hot-working must be carefully
calibrated according to the grade and to the final quality requirements for any
specific product. PHSS can effectively be joined by resistance and fusion welding
techniques. The main critical aspects to consider are the same as for the other
families of stainless steels, i.e., the HAZ and relevant metallurgical issues. The
252 5 Steels

machinability of PHSS is good in the annealed and decreases in the aged condition,
especially for martensitic and semiaustenitic grades, due to the strength improve-
ment that results from the thermal treatments. PHSS show better corrosion resistance
than traditional high-strength steels, as a direct consequence of their chemical
compositions and effect on the self-passivation behavior. However, PHSS are rather
prone to hydrogen embrittlement (see Sect. 8.1.7), especially in alloys with increas-
ing yield strength. Like low-carbon steels, PH stainless steels exhibit a dependence
of the corrosion resistance from strength level and microstructure: a decrease in the
corrosion resistance is observed at increasing yield strengths. Although the mechan-
ical properties of several PHSS are comparable to those of martensitic stainless
steels, their corrosion resistance is definitely better. The 17-4PH has a corrosion
resistance that is very close to that of the austenitic AISI 304L. The corrosion
resistance of Custom 450 is similar to that of 17-4PH, although somehow better as
concerns pitting corrosion (see Sect. 8.1.2). This is due to the Mo content of the
Custom 450 alloy, higher than that of 17-4PH steel (see Table 5.19). Among the
austenitic PHSS, A286 displays a good corrosion resistance thanks to its chromium-
rich austenitic structure. Improvements in SCC resistance can be attained by aging
the alloy at the maximum treatment temperature, compatible with the achievement of
an adequate strength level (Savoie et al. 2007; Campbell 2008; DeCicco et al. 2004;
Farrar 2004).

5.5 Steel-Based Aerospace Components

The Flyer I (see Sect. 1.1) had some parts made of steel, which was indeed used since
the beginning of the aerospace era, although for structural parts only. Steel was the
standard choice for bracing wires and wing struts, being wood preferred for the
fuselage and other important parts, like propellers, wing ribs, etc. This situation
lasted until the mid-1930s, when wood started to be replaced by other materials due
to the emerging limitations associated with its thermal stability at high speeds and for
moisture absorption. The material universally used for most engineering structures
was ready to replace wood and indeed steel tubes started to be used for the
fabrication of the main fuselage parts, as well as for the wing spars (Cutler 2005).
Afterward, the percentage of steel used in aircraft structures and engines was
progressively reduced, passing from about 40% down to 5–15%, depending on the
aircraft, as a consequence of the increased availability of light alloys (see Fig. 1.11,
Chaps. 3 and 4) (Prasad and Wanhill 2017). On average, helicopters and aircrafts
account for 5–10 wt% of their structure to be made of steel. Designers use steels in
applications where there is no alternative valuable choice, especially for components
that are regarded as highly critical, both for safety and performances. The reason for
the limited structural aerospace applications of steel is mainly due to the compara-
tively high density of these alloys, being on average 3.5 times more than that of
carbo-epoxy composites, 1.5 times that of titanium and its alloys, and 2.5 times the
density of aluminum alloys.
5.5 Steel-Based Aerospace Components 253

However, there are still aircraft parts for which steel is an excellent choice, as it
will be illustrated in the upcoming sections on:
• Bearings.
• Brake systems.
• Engine parts.
• Gearings.
• Undercarriage (landing gear).
• Wings (root attachments, carry trough, slot track, fasteners) and parts of engine
pylons.
A detailed list of general aerospace applications of steels in provided at the end of
this section.

5.5.1 Bearings

Bearings are rolling components usually comprising a cage, inner and outer race-
ways, and balls (or rollers) (Fig. 5.35). Bearing systems in aerospace must withstand
rather severe conditions. Moreover, they have to guarantee reliable performances
and predictable lifetimes, particularly those in very important positions, like the
engine shaft bearings. Steels are still the preferred choice, either low or highly
alloyed materials (Flower 1995). Helicopters’ bearings are usually made of
low-alloy steels that have been case-hardened. For applications where high strength
is required in combination with corrosion resistance, a suitable choice is the AISI
440C martensitic stainless steel. Other examples of bearing steels are SAE S2100,
AISI M50, M50NiL, and X30 steels (Flower 1995).
Two main factors may affect the performances of bearings:
• Grain flow, i.e., forging flow orientation with respect to the rolling contact
surface. In this respect, a parallel, rather than perpendicular, orientation is
recommended since it increases the fatigue life.
• Melting procedures, influencing the homogeneity, microscopic cleanliness (e.g.,
carbide and inclusion distribution) of the product.
Bearing life is limited by the high rotational and vibrational levels reached under
service conditions and relevant temperature rise. Depending on the specific applica-
tions, they may reach peak values in the 350–400
C in case of most demanding
conditions, like turbine bearings (Flower 1995). Of course, much higher flash
temperatures may be reached at asperity contact regions, although for comparatively
much shorter times. Another important element is the corrosive environment, due to
the lubricants, that may turn out to be an important lifetime limiting factor. The
development of more performing engines would pose more challenging require-
ments to bearing systems too, in terms of higher rotational contact speeds and load
capacity, longer life, and lower weight. Steel processing is continuously evolving, so
that inclusion-free and cleaner materials can be produced, with generally positive
254 5 Steels

Fig. 5.35 Roller-bearing and ball-bearing structures. (Soares 2015, reproduced with permission of
the Publisher)

effects on the expected lifetime of the bearing (Scott 1969). In this regard, the main
specific requests for bearing materials are:
• Adequate rolling fatigue resistance.
• Corrosion resistance.
• High hardness, which may guarantee an adequate resistance to wear.
• Structural stability at the working stress and temperature conditions.

5.5.2 Brake Systems

In the early days of the aircraft era, airplanes had no brakes, and, once landed, during
taxing they just exploited the frictional forces resulting from the contact between the
aircraft wheels and tail and the grassy ground of the early runways. The very first
aircraft braking systems were piston brakes derived from road vehicles. They
became absolutely necessary when hard, concrete or asphalt, runways started to be
used. These brakes were made of pads and drums, which looked very similar to
brake systems of contemporary cars. However, this system is characterized by some
disadvantages that were not compatible with the increasing needs in terms of braking
performances of the newly developed aircrafts. In particular, the thermal expansion
of both brake fluid and drum occurring during braking leads to the fading of the
brake action, thus reducing brake performances and increasing the space needed to
stop the aircraft. To overcome this drawback, expander tube brakes were introduced
and became the standard for any kind of aircrafts over the 1930s–1950s period.
5.5 Steel-Based Aerospace Components 255

Expander tube brakes were then replaced progressively by diverse kinds of new disc
brake systems: single disc, dual and multiple discs, and segmented-rotor disc brakes.
Single disc brakes were installed on aircrafts with nonretractable landing gear (Jetley
2007). The system is built in such a way that the disc rotates together with the wheel
assembly. Meanwhile, when the braking pressure is applied, a stationary caliper
resists the rotation producing a frictional torque against the disc (Federal Aviation
Administration 2012). Increasing flight speeds called for retractable landing gear
systems, and for this reason, wheels had to become smaller and lighter, to save space
and weight. The need for limiting the wheel and brake assembly size, still preserving
an adequate stopping power, promoted the design of dual and, afterward, multiple-
disc systems. A dual disc brake system is composed of two discs for each wheel,
with lining on each side of the disc. A frictional force is developed when the linings
are pressed against the discs. The multiple-disc concept is an obvious extension of
the above brakes. An alternate stack of stators, usually made of steel, and bronze
alloy discs provide a very effective frictional system, capable to dissipate large
amounts of kinetic energy. Multiple-disc brakes were designed for large and heavy
aircrafts and can still be found in cargo aircrafts. However, in multiple-disc brakes,
heat dissipation is not optimized, and they have high-warp tendency, considering the
very large frictional heat concentration that can be reached in these highly dense
packed systems. This has suggested the development of the so-called segmented-
rotor disc design. Segmented-rotor disc brakes are multiple-disc braking systems
with an improved design, capable to overcome the cooling limitations of multiple-
disc brakes, with obvious benefit in terms of brake efficiency and part durability
(Jetley 2007). Their use is widely spread in air carrier and high-performance aircrafts
and commercial airliners (Federal Aviation Administration 2012). The brake unit is
constituted by a set of discs that include alternating rotors and stators built with slots
or spacings among them, resulting in a better heat dissipation (Federal Aviation
Administration 2012). Rotor discs rotate with the wheel, while stator discs are
stationary with respect to the landing gear. When the brake fluid hydraulic pressure
increases, pistons press stator discs against rotors, thus generating a friction between
the two sets of discs, leading to a reduction in the rotational speed and eventually to
stop the aircraft (Young 2018). The force generated by the braking system can be
determined according to Eq. 5.7 (Young 2018):


F ¼ μb W  cos γ g  L ð5:7Þ

where:
F is the braking force.
μb is the braking coefficient. It is experimentally determined and depends on the
airplane’s weight and speed: when the speed decreases, the braking coefficient
increases. The braking coefficient depends also on maximum slip that occurs
between tire and runway, maximum braking torque, and tire-runway interface
conditions (dry or wet runway, runway surface condition and profile, tire pressure
and wear).
256 5 Steels

W is the airplane weight.

γ g is the runway slope angle expressed in rad.
L is the instantaneous lift.
The lift is calculated according to Eq. 5.8.

1
L ¼ ρSCL v2 ð5:8Þ
2

where:
L is the instantaneous lift force, expressed in N.
ρ is the air density expressed in kg/m3.
S is the wing reference area expressed in m2.
CL is the lift coefficient (dimensionless).
v is the speed in m/s.
Thanks to the small angle approximation, cosγ g can be taken equal to 1. Equation
5.7 can therefore be rearranged using the definition of instantaneous lift given by
Eq. 5.9:
 
1
F ¼ μb W  ρSC L v2 ð5:9Þ
2

There are of course different braking regimes, the most severe one being the
emergency braking, i.e., the braking action necessary to abort take-off once the
aircraft has reached a sub-maximum speed. In this case, in fact, the whole burden of
stopping the aircraft relies on the wheel braking system, which cannot be backed by
any additional stopping contribution, like thrust reversers, if available. Severe
braking regimes may also result in the melting down of disc surface, with a
consequent significant reduction in the braking power, the so-called brake fade
(Young 2018). Although, disc replacement is in general necessary after an emer-
gency braking event, still it is important that even these extreme conditions are not
producing any catastrophic effect that may jeopardize the passenger and crew safety.
Other elements that may influence the brake system performances are the additional
and uneven loads on wheel, that may be acting in case of crosswinds and consequent
compensation through wheel steering. The kinetic energy generated by the aircraft
engines will also vary depending on the altitude at which the airport is located.
Higher altitude leads to an increased landing and take-off speed owing to the less
dense atmosphere (Jetley 2007). Material requirements for brake applications con-
cern wear and heat resistance, heat dissipation capacity, and adequate toughness to
withstand rapidly changing loading conditions. Steels exhibit properties that render
them suitable candidates for this undercarriage subsystem, as proved by the impor-
tant role they have played in different types of aircraft brake systems. Stators are
steel plates having the lining material (usually a Cu-based alloy) on both sides, either
riveted or deposited. Rotors are made of steel, and they may have a frictional surface
5.5 Steel-Based Aerospace Components 257

Table 5.22 Chemical composition (wt%) of AMS 6320 (Jetley 2007)

C Mn Cr Mo Si S P V
0.28–0.33 0.45–0.65 1.0–1.5 0.40–0.60 0.55–0.75 0.04 0.04 0.2–0.3

made of sintered metal or cermet systems, to improve brake efficiency (Federal

Aviation Administration 2012). Wear resistance directly depends on the hardness
and, therefore, on the microstructure and on possible thermal treatments. Road
vehicles, mainly cars and heavy duty vehicles, have cast iron or austenitic stainless
steel discs. Stainless steel, like the AISI 410 steel, is the typical choice for motor-
cycles (Sarip et al. 2013). For aircraft applications one possibility is heat-resistant
low-alloy steels. These steels have chemical compositions that allow heat treatment
to be carried out to induce structural transformations in order to meet the severe
requirements of aircrafts service conditions. One example of heat-resistant low-alloy
steel used in the production of brake discs is the AMS 6320. Its standard chemical
composition is given in Table 5.22.
The standard heat treatment for this steel is austempering (see Sect. 5.2.3).
Austempering consists in heating the steel up to the relevant austenitic field. After
homogenization, the steel is cooled down to a temperature above its Ms and held
until isothermal transformation into bainite has occurred (Jetley 2007).
The increased dimension of the multiple-disc brakes rendered particularly critical
the weight issue, considering the densities of the metallic alloys used for discs and
stators. Carbon-carbon (C–C) composites turned out to be excellent candidates for
the substitution of steels and bronze alloys. C–C composites afford an excellent
combination of high strength, retained also at the highest peak temperatures, and
tribological properties, once the oxidation and humidity problems have been
addressed and solved. The early important aircraft application of C–C composites
was in the brakes of the Concorde supersonic airliner in the mid-1970s (Stimson and
Fisher 1980). By replacing a steel counterpart, this choice introduced an estimated
600 kg weight saving, being the density of C–C composites equal to 2.3 g/cm3, i.e.,
one third of the density of steels. Another important material property for brake
applications, in which discs behave by all means as heat sinks, is their heat capacity,
which is 2.5 higher in C–C composites than in steels. Since then, these brake systems
became the main choice of the aircraft industry, particularly for large airliner,
considering the remarkable weight saving that a C–C composite design introduces
with respect to the use of steel and bronze alloy (Gatt et al. 2017; Windhorst and
Blount 1997). In the Boeing 737 airliner, the weight reduction is 300 kg approxi-
mately, involving a significant fuel saving over the whole lifetime of the aircraft. The
better performances of the C–C brakes, capable to manage higher landing and take-
off energies, are fundamental to keep up with the tendency to develop larger and,
thereby, heavier airliners, like, for instance, the Airbus 380 and the Boeing 747-8.
Moreover, these brakes would also allow shorter landing and take-off distances.
Interestingly, C-C brakes have been developed and deployed on commercial air-
crafts first, with respect to military ones, which actually on several occasions have
benefited from this technological transfer. For instance, the brakes of the F-16 fighter
258 5 Steels

are derived from the brakes of the Boeing 747-400; those of the C-130 Hercules are a
derivative of the brake systems of the Airbus A320 and Boeing 777LR. Although out
of the reach of the monograph, it is worth mentioning that at the present time, new
materials, i.e., the C/SiC composites, originally developed for top cars, are now
under investigation in view of their possible usage for aircraft brakes too (Fan et al.
2016; Narottam and Boccaccini 2012).

5.5.3 Gas Turbine Jet Engine Parts

AISI 403 and AISI 403 Cb stainless steels have been the base materials for
compressor blades until the 1980s, when a new steel for these components, Custom
450 (see Table 5.19), was introduced. Custom 450 (also known as GTD-450),
features better tensile strength, stress corrosion resistance and both higher fatigue
corrosion and high-cycle fatigue resistance, than AISI 403 (Boyce 2012). Other
steels used for the fabrication of compressor blades of GTs are the FV458, FV46
7, FV520, and 15-5 PH stainless steel and the Rex 706 steel (Meetham 1981). As
concerns compressor discs, despite steel-based discs are still in production, the use of
this material has been reduced over the years mainly for the increasing temperature
requirements. Concerning the combustor, some parts, which are not directly in
contact with high-temperature gases, are made of AISI 309 stainless steel (Boyce
2012). Both ferritic and austenitic stainless steels have been used for turbine
components, in particular for turbine discs. The austenitic PH stainless steel A286
has been successfully used for several years in aircraft applications (see Sect.
5.4.3.3), and starting from 1965, this same material has been used in the fabrication
of discs for land-based gas turbines. Occasionally, A286 has been replaced by the
M152 steel (12% Cr and 2–3% Ni), exhibiting a higher tensile and an intermediate
rupture strengths, this latter roughly in between the rupture strength of Cr-Mo-V
steel (1%Cr-1.25%Mo-0.25%V) and that of A286 (Boyce 2012). Stainless steels are
used in the exhaust ducts of the engine too, where they are exposed to relatively high
gas temperatures, in particular when the afterburner is enabled. For these parts, an
important request is not to transfer excessive heat to the surrounding aircraft
structure. A possible approach is to protect the hot parts with insulating “blankets”
(Prasad and Wanhill 2017).

5.5.4 Gearboxes

Gearboxes are particularly critical in helicopters, since they have to match the
rotational speed of the turbine with that of the rotors (Flower 1995). Concerning
aircrafts, gearing systems are important too and are used to drive accessory units
5.5 Steel-Based Aerospace Components 259

Fig. 5.36 Position of gearing systems in a triple-shaft front fan turbo jet and assembly of an
external gearbox with auxiliary gearbox drive. (Rolls-Royce 1996, images courtesy of Rolls-Royce
plc)

connected to a rotating engine shaft, as illustrated in Fig. 5.36. Accessory units have
to provide the power for aircraft hydraulic, electrical and pneumatic systems, pumps
and other control systems, all of great importance, both for performance and safety
reasons. Therefore, they require an extreme reliability, also as concerns materials
properties (Rolls-Royce 1996).
Materials used for gearing purposes must meet the following main requirements:
• Adequate resistance to surface degradation phenomena, such as wear, pitting, and
scuffing.
• Corrosion resistance in oil environments over a temperature range of 90
C–
130
C.
• Sufficient strength for withstanding both static and fatigue loads.
• Toughness for resisting against shock loads.
For these reasons, gearing steels are usually subjected to case-hardening, either
nitriding or carburizing. Typical nitriding steels for these applications are Nitralloy
N and 32CDV13 (see Sect. 5.4). Two examples of conventional carburizing steels
are AMS 6265 and BS S156, replaced in some engines by advanced carburizing
steels, like VASCO X-2M and Pyrowear 53.
The performances of gearing materials may be affected by metallurgical factors,
the most important being:
260 5 Steels

• Core microstructure. Since surface hardening may alter the properties of the
surface and substrate, it is important to check for an adequate toughness of
the core.
• Grain size. Fine and uniform grain size is preferred, and usually an ASTM grain
size of 6–8 is required for matching fatigue strength and ductility requirements.
• Steel cleanliness in terms of type, size, and shape of inclusions.

5.5.5 Undercarriage Applications

As regards undercarriage components, the landing gear is of course particularly

important, considering its central role in aircraft operations, and steels have been a
standard choice over the years (Fig. 5.37). Landing gears are subjected to repeated
high dynamic loads on landing: civil aircrafts are designed for 30,000–60,000
flights, while military aircrafts are designed for an expected 5000–10,000 flights,
which are spread over a 20–30 year operational time. The landing gears have also to
sustain the full weight of the aircraft, i.e., masses up to several hundreds of tons (e.g.,
380 t for Boeing 747), during land operations, e.g., taxing, and parking. Another
important requirement is a design that reduces not only the weight but also the
volume of the undercarriage in the retracted condition, since this volume, and weight
too, competes with the volume available to the payload.
High-strength low-alloy steels are excellent candidates for landing gear applica-
tions. A few examples of this kind of alloys are the high-strength steels (see labels in
Fig. 5.37): S99 (WN 1.6745), AISI 4340, AISI 4340M (alternatively known as
300M), and AMS 6487. In addition to conventional steels, the family of advanced
AerMet steels, like AerMet 100 (see Sect. 5.3.2), are also used for undercarriage
applications.

5.5.6 Wings and Engine Pylons

Steels are used for other important structural aircraft components, like wing root
connections, slat tracks (i.e., part of the leading edge of the wings), and joints/fittings
of engine pylons (Cantor et al. 2001). Wing roots and parts of the engine pylons are
particularly critical, and the use of steels is still recommended. Wing roots connect
wings to the fuselage and are subjected to diverse stress conditions, particularly
static, dynamic, and fatigue loads. Slat tracks are moving beams that allow the
relative displacement between the wing and the front part of the wing itself that
enhances the lift during take-off and landing (see Fig. 3.1). Owing to their high
strength and damage tolerance, maraging steels are an excellent choice for these
parts. Fittings exhibit quite a broad range of odd design and relevant stress condi-
tions, depending on the specific parts of the aircraft they are connected to. Fasteners
are relatively small items, although often in critical positions for the safety and the
5.5 Steel-Based Aerospace Components 261

Fig. 5.37 Airbus A330 undercarriage with the indication of steel components. (Flower 1995,
reproduced with permission of the Publisher)

efficient operation of aircrafts. Typical steels used for the fabrication of fasteners are
17-4PH and A286 (Speck 2015). Dealing with fasteners, an important issue is the
way in which they are connected to the main part, bolting and riveting being the most
common approaches. In this context, AerMet 100 can be a good candidate (Speck
2015). All parts of the joint are to be selected in order to have compatible mechanical
properties, with no excessive increase of internal stresses, and electrochemical
compatibility.

5.5.7 Aerospace Applications of Steels: An Overview

Steels introduced in this chapter are used for several aircraft components. They are
selected on the basis of safety, critical loading, and environmental conditions.
Maraging steels are the best choice for a large variety of components, as shown by
the information in Table 5.23.
AerMet steels are employed in the fabrication of components that require partic-
ularly high strength, toughness, and ductility. They have the advantage over titanium
alloys, of which AerMet steels can be regarded as direct competitors, of reduced
dimensions of the parts, due to their higher density (Novotny and Maurer 2007). This
aspect may be a decisive selection criterium in case space saving is an issue.
262 5 Steels

Table 5.23 Aircraft components made of maraging steels (ASM International 1993; Llewellyn and
Hudd 1995; Megson 2013)
Component/part Component/part
Aircraft arrest hooks Hydraulic hoses and bolts
Aircraft forgings Landing gear components
Bearings Missile cases
Cannon recoil springs Slat tracks
Cargo systems Transmission shafts
Fan shafts of commercial jet engines Undercarriage pivot brackets
Fasteners Wing lateral panels and wing roots attachments

Table 5.24 Aircraft components made of AerMet steels (Steel Service Aerospace Corp. 2016;
Novotny and Maurer 2007; Prasad and Wanhill 2017)
Component/part Steel
Actuators AerMet 100, AerMet 310
Armor AerMet 100, AerMet 310
Connecting rods AerMet 340
Crank shafts AerMet 340
Drive shafts AerMet 100, AerMet 310, AerMet 340
Fasteners AerMet 100, AerMet 310
Jet engine shafts AerMet 100, AerMet 310
Landing gear components AerMet 100, AerMet 310
Ordnance AerMet 100, AerMet 310
Springs AerMet 340
Structural tubing AerMet 100, AerMet 310

The main applications of AerMet alloys are summarized in Table 5.24.

Depending on the atmosphere, the temperature range for austenitic stainless steel
applications can reach 900–1100
C (Bhadeshia and Honeycombe 2006; Young
2015). Therefore, AISI 321 is an ideal candidate for the production of some
aerospace components, such as piston engine exhaust manifolds, exhaust ducts,
and flanges. In fact, this steel retains stable and adequate mechanical properties
over a temperatures interval between 400 and 820
C also resisting a to high-
temperature oxidation. Other applications of austenitic stainless steels are listed in
Table 5.25.
The components for which martensitic stainless steels are typically used are
listed in Table 5.26. Other possible applications for martensitic stainless steels are
axels cams, shafts and springs, blading in steam, and gas turbines, operating at
temperatures up to 540
C (ASM International 1993; Campbell 2008; Meetham
1981; Prasad and Wanhill 2017).
The 58 HRC hardened condition is the standard for the AISI 440C used in the
production of rolling bearings operating in corrosive environments, where there is no
other way to protect the material from corrosion (ASM International 1993; Campbell
2008; Harnoy 2003; Totten 2007).
5.5 Steel-Based Aerospace Components 263

Table 5.25 Aircraft components made of austenitic stainless steels (Boyce 2012; Prasad and
Wanhill 2017)
Component/part Steel
Exhaust components and medium-temperature engine AISI 309, AISI 310, AISI 316, AISI
structural parts 316L, AISI 321
Fuel tanks AISI 304, AISI 304L
Jet engine parts AISI 321, AISI 347
Parts of the combustor that are not directly in contact AISI 309
with high temperatures
Pins for compressor variable vanes Nitronic 60
Piston engine exhaust manifolds and exhaust ducts and AISI 321
flanges

Table 5.26 Aircraft Component/part Steel

components made of
Ball bearings AISI 440C
martensitic stainless steels
(ASM International 1993; Bolts AISI 410
Boyce 2012; Campbell 2008; Compressor blades AISI 403, AISI 410, AISI 414
Harnoy 2003; Meetham 1981; Spray nozzles AISI 440
Prasad and Wanhill 2017) Valves and valve needles AISI 410, AISI 431

Table 5.27 Aircraft components made of PH stainless steels (Boyce 2012; Speck 2015)
Component/part Steel
Afterburner parts A286
Aircraft bulkheads 15-7Mo
Compressor blades A286, 15-5PH
Door surrounds 15-5PH, 13-8Mo
Fasteners A286, 13-8Mo, 17-4PH
Landing gear components 15-5PH, 13-8Mo
Nose gear door 15-5PH, 13-8Mo
Pylon and engine mounting parts 15-5PH
Rockets and missiles propulsion systems 15-5PH
Slat tracks 17-4PH, 15-5PH
Springs A286, 17-7PH, 15-7Mo
Tail cones A286
Turbine wheels A286
Valves 15-5Mo, 13-8Mo
Welded and brazed honeycomb panels 15-7Mo
Wing flap tracks 17-4PH, 15-5PH

Thanks to their unique combination of mechanical properties and corrosion

resistance, PH stainless steels are a class of alloys that is massively used in several
parts and componets of aircraft structure and engine, the main ones being listed in
Table 5.27.
264 5 Steels

References

Angel T J (1954) Formation of martensite in austenitic stainless steels. ISIJ International (Japan)
177: 165–174
Ashby M F (2005) Materials – A brief History. Philosophical Magazine Letters 88 (9–10): 749–755
Ashcroft N W, Mermin N D (1988) Solid State Physics. HRW International Editions, Hong Kong
ASM International (1992) ASM Handbook Vol. 3 – Alloy phase diagrams. ASM International,
Materials Park, Ohio
ASM International (1993) ASM Handbook Vol. 1 – Properties and Selection: Iron, Steel and High
Performance Alloys. ASM International, Materials Park, Ohio
ASM International (2008) ASM Handbook Vol. 15 – Casting. ASM International, Materials Park,
Ohio
ASM International (2016) ASM Handbook Vol. 4E – Heat treating of Nonferrous Alloys. ASM
International, Materials Park, Ohio
Bhadeshia H. K. D. H, Honeycombe R. W. K. (2006) Steels – Microstructure and Properties, 3rd
edn. Butterworth-Heinemann, Oxford
Boxman R L et al (1996) Handbook of Vacuum Arc Science and Technology: Fundamentals and
Applications. Noyes Publications, Park Ridge, New Jersey
Boyce M P (2012) Gas Turbine Engineering Handbook, 4th edn. Butterworth-Heinemann, Oxford
Buchwald V F, Mosdal G (1985) Meteoric Iron, Telluric Iron and Wrought in Greenland. Man &
Society 9: 3–49
Callister W, Rethwisch D (2012) Fundamentals of Materials Science and Engineering: An Inte-
grated Approach, 4th edn. John Wiley & Sons Ltd
Campbell F (2006) Manufacturing Technology for Aerospace Structural Materials. Elsevier
Campbell F (2008) Elements of Metallurgy and Engineering Alloys. ASM International
Cantor B et al (2001) Aerospace Materials. Institute of Physics Publishing
Chandler H (1996) Heat Treater’s Guide – Practices and Procedures for Nonferrous Alloys. ASM
International
Cobb H M (2010) The History of Stainless Steel. ASM International
Cutler J (2005) Understanding Aircraft Structures, 4th edn. Wiley-Blackwell
De Cicco H, Luppo M, Gribaudo L et al (2004) Microstructural Development and Creep Behavior
in A286 Superally. Materials Characterization 52: 85–92
Demeri M (2013) Advanced High-strength Steels: Science, Technology and Applications. ASM
International
Donachie M, Donachie S (2002) Superalloys, a Technical Guide. ASM International
Everhart J L (1971) Engineering Properties of Nickel and Nickel Alloys. Plenum Press, New York
Fan s et al (2016) Progress of Ceramic Matrix Composites Brake Materials for Aircraft Applica-
tions. Reviews on Advanced Materials Science 44 (4): 313–325
Farrar J (2004) The Alloy Tree – A guide to Low-Alloy Steels, Stainless Steels and Nickel-base
Alloys. Woodhead Publishing Limited and CRC Press
Federal Aviation Administration (2012) Aviation Maintenance Technician Handbook-Airframe
Volume 2. Federal Aviation Administration
Flint G, Oldfield J, Dautovich D P (2010) Corrosion of Iron Nickel Alloys and Maraging Steel.
Shreir’s Corrosion: 1789–1801
Flower H (1995) High Performance Materials in Aerospace. Springer-Science+Business Media
Gatt A et al (2017) Hydro-Mechanical Coupling in Unstable Aircraft Braking Systems. In: Harvie
J M, Bawersad J (ed) Shock & Vibration, Aircraft/Aerospace, Energy Harvesting, Acoustics &
Optics, Volume 9 – Proceedings of the 35th IMAC, A Conference and Exposition on Structural
Dynamics, Garden Grove, California, 2017
Graves J (1994) Effect of Heat Treatment on the Microstructure and Properties of AerMet 100 Steel.
Army Research Laboratory. Available via DIALOG. https://apps.dtic.mil/dtic/tr/fulltext/u2/
a284387.pdf Accessed 30 March 2019
Harnoy A (2003) Bearing Design in Machinery. Marcel Dekker Inc
References 265

Hosford W F (2010) Physical Metallurgy. CRC Press, Boca Raton

Hosford W F (2012) Iron and Steel. Cambridge University Press, Cambridge
Jetley S (2007) Martempering to Improve Wear Properties of Aircraft Brake Steel Rotors. Journal of
Industrial Technology 23 (2)
Kaieda Y, Oguchi A (1985) Strain-induced transformation and plastic deformation behavior of a
17Cr-7Ni-1Al steel at high hydrostatic pressure. Journal of Materials Science 20: 1847–1858
Leitner H et al. (2011) Strengthening behavior of Fe–Cr–Ni– Al–(Ti) maraging steels. Materials
Science and Engineering A, Structural Materials Properties, Microstructure and Processing 528
(15): 5264–5270
Llewellyn D T, Hudd R C (1995) Steels: Metallurgy and Applications 3rd edn. Butterworth-
Heinemann
Madias J (2013) Electric Arc Furnace. In: Seetharaman S (ed) Treatise on Process Metallurgy
Volume 3: Industrial Processes. Elsevier, p 271–300
Meetham G W (1981) The Development of Gas Turbine Materials. Applied Science Publishers Ltd,
London
Megson T (2013) Aircraft Structures for Engineering Students, 5th edn. Butterworth-Heinemann,
Oxford
Miller M K, Burke M G (1991) Characterization of copper precipitation in a 17/4PH steel: a
combined APFIM/TEM study. In: Proc. 5th Intern Conf Environ Degrad Mater Nucl Power Syst
React, LaGrange Park, 1991
Mitchell A (2005) Solidification in Remelting Processes. Materials Science and Engineering A
413–414: 10–18
Moshka O et al (2015) Adressing the Issue of Precipitates in Maraging Steels – Unambiguous
Answer. Materials Science and Engineering A 638: 232–239
Mouritz A (2012) Introduction to Aerospace Materials. Woodhead Publishing Limited
Narottam B, Boccaccini A (2012) Ceramic and Compostites Processing Methods. John Wiley &
Sons Ltd
Novotny P, Maurer G (2007) Ultra High Strengh Steels vs Titanium Alloys. Advanced Materials
and Processing 165(11): 37–40
Pardal JM et al (2005) Modeling of precipitation hardening during the aging and overaging of
18Ni–Co–Mo–Ti maraging 300 steel. Journal of Alloys Compounds 393(1–2):109–113
Pepperhoff W, Acet M (2001) Constitution and Magnetism of Iron and its Alloys. Springer, Berlin
Heidelberg
Prasad E N, Wanhill R (2017) Aerospace Materials and Material Technologies Volume 1: Aero-
space Materials. Springer
Prasad E N, Wanhill R (2017) Aerospace Materials and Material Technologies Volume 2: Aero-
space Material Technologies. Springer
Rolls-Royce (1996) The Jet Engine 5th edition. Rolls-Royce plc, Derby
Sarip S et al (2013) Analysis of the Transient Thermomechanical Behaviour of a Ligthweight Brake
Disc for a Regenerative Braking System. In: SAE-China, FISITA (eds) Proceedings of the
FISITA 2012 World Automotive Congress Volume 8: Vehicle Design and Testing (II). Lecture
Notes in Electrical Engineering, vol 196. Springer, Berlin, Heidelberg, p 1075–1088
Saul G, Roberson J A, Adair A M (1970) The effects of thermal treatment on the austenitic grain
size and mechanical properties of 18 Pct Ni maraging steels. Metallurgical and Materials
Transactions B 1 (2): 383–387
Savoie M, Esnouf C, Fournier L et al (2007) Influence of Ageing Heat Treatment on Alloy A-286
Microstructure and Stress Corrosion Cracking Behavior in PWR primary water. Journal of
Nuclear Materials 360; 222–230
Schifo J F, Radia J T (2004) Theoretical/Best Practice Energy Use In Metalcasting Operations.
U.S. Department of Energy Industrial Technologies Program, Washington, DC
Scott D (1969) The Effect of Steelmaking, Vacuum Melting and Casting Techniques on the Life or
Rolling Bearings. Vacuum 19(4):167–169
266 5 Steels

Service Steel Aerospace Corp. (2016) AerMet ® 100 Alloy Data Sheet. Available via DIALOG.
https://www.ssa-corp.com/documents/Data%20Sheet%20AerMet100.pdf. Accessed 16 April
2017
Sha W, Guo Z (2009) Maraging steels – Modeling of microstructure, properties and applications.
CRC Pess
Shi X, Zeng W et al (2016) Study of the Microstructure and Mechanical Properties of AerMet
100 Steel at the Tempering Temperature around 482
C. Journal of Alloys and Compounds 679:
184–190
Soares C (2015) Gas Turbines: A Handbook of Air, Land and Sea Applications, 2nd edn.
Butterworth-Heinemann
Speck J A (2015) Mechanical Fastening, Joining and Assembly. CRC Press
Stimson I L, Fisher R (1980) Design and Engineering of Carbon Brakes. Philosophical Transactions
of the Royal Society A Mathematical Physical and Engineering Sciences 294 (1411): 583–590
Stodart J, Faraday M (1820) Experiments on the Alloys of Steel, made with a View to its
Improvement. Philosophical Magazine 56 (257): 26–35
Tewari R et al (2000) Precipitation in 18 wt% Ni Maraging Steel of Grade 350. Acta Materialia
48 (5): 1187–1200
The AISE Steel Foundation (1998) The Making Shaping and Treating of Steel, 10th edn. The AISE
Foundation, Pittsburg
Totten G E (2007) Steel Heat Treatment – Metallurgy and Technologies, 2nd edn. CRC Press
Toulouevski Y, Zinurov I (2013) Innovation in Electric Arc Furnaces, 2nd edn. Springer
Turkdogan E T (2010) Fundamentals of Steelmaking. Maney Publishing
Tylecote R F (2002) A History of Metallurgy, 2nd edn. Maney Publishing
Verhoeven J D (2007) Steel Metallurgy for the Non-Metallurgist. ASM International
Viswanathan U et al (1988) Effects of Aging on the Microstructure of 17-4 PH Stainless Steel.
Materials Science and Engineering A 104: 181–189
Windhorst T, Blount G (1997) Carbon-Carbon Composites: a Summary of Recent Developments
and Applications. Materials & Design 18 (1): 11–15
Würzinger P, Rabitsch R, Meyer W (2004) Production of Maraging Steel Grades and the Influence
of Specified and Nonspecified Elements for Special Applications. Journal of Materials Science
39 (24): 7295–7302
Xu X et al (2002) Tree-ring Formation during Vacuum Arc Remelting of INCONEL 718: Part
I. Experimental Investigation. Metallurgical and Materials Transactions A 33 (6): 1795–1804
Yang S et al (2015) Phase transformation and its effect on mechanical properties of C300 weld
metal after aging treatment at different temperatures. Journal of Iron and Steel Research
(International) 22(6): 527–533.
Yeli G, et al (2017) Sequential Nucleation of Phases in a 17-4PH Steel: Microstructural Charac-
terization and Mechanical Properties. Acta Materialia 125: 38–49
Young D (2015) High temperature oxidation and corrosion of metals. Elsevier, Oxford
Young T M (2018) Performance of the Jet Transport Airplane. John Wiley & Sons Ltd

Further Reading

ASM International (1994) ASM Specialty Handbook – Stainless Steels. ASM International,
Materials Park, Ohio
Faivre D, Frankel R (2016) Iron Oxides: from Nature to Applications. Wiley-VHC, Weinheim
Gale W F, Totemeyer T C (2004) Smithells Metals Reference Book 8th ed. Butterwoth-Heinemann
Chapter 6
Superalloys

6.1 Introduction

In the first half of the twentieth century, stainless steels were employed in high-
temperature structural applications, including some components of the newly devel-
oped jet engines (see Sects. 2.1 and 2.2). Materials used in hot parts of jet engines
must retain elevated performances and structural stability for prolonged periods
under service conditions involving high mechanical stresses and corrosive agents.
The increasing performances of the gas turbine engines evidenced some limitations
of steels. This fostered the development of a new class of metallic alloys, named
superalloys in view of their remarkable structural properties retained even up to
relatively high temperatures. The main features of superalloys can be summarized as
follows:
• Capability to withstand loads at operating temperatures that are important frac-
tions of their incipient melting point, expressed in terms of the ratio Top/Tm, i.e.,
operating temperature versus incipient melting temperature ratio, with tempera-
tures expressed in Kelvin. In the hot spots of gas turbine components, like the
blades of the turbine high-pressure stage, values in excess of 0.70 can be
achieved.
• Mechanical stability over long times. For high-temperature structural applica-
tions, materials with low creep rates are desirable. Other mechanical properties,
such as ultimate tensile strength, tensile yield strength, and fracture toughness, are
important too and should be preserved over the time, to guarantee safe perfor-
mances of the structural components.
• Surface stability in aggressive environments. The highly corrosive environments
generated by the combustion of jet fuels and ingestion of contaminants into the
gas turbine engine (see Sect. 8.2.4) may induce important gas phase high-
temperature corrosion phenomena, enhanced by the possible condensation of
salt deposits, either solid or liquid (see Sect. 8.2.5).

© Springer Nature Switzerland AG 2020 267

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_6
268 6 Superalloys

The early superalloys were stainless varieties of the existing steels, subsequently
officially named superalloys. Patents for Al and Ti additions to the Nichrome-type
alloys were issued in the 1920s. The superalloy industry really emerged with the
introduction of a Co-based alloy, initially used in dentistry: the Vitallium. It was
selected as a candidate material capable to satisfy the high-temperature requirements
of the emerging gas turbine aircraft engines. At the same time, other alloy systems
attracted research interests. This led to the widening of the superalloy family tree.
Nowadays, a widely accepted definition of superalloys refers to a group of austenitic
Fe, Co, or Ni (group VIII A)-based alloys that are extensively used for applications
involving operating temperature exceeding 600
C (Donachie and Donachie 2002;
Reed 2006). Interestingly, although all of them are austenitic, i.e., they are face-
centered cubic ( fcc), nickel only exhibits this crystalline structure from room
temperature up to the melting point. The crystallographic structure of cobalt at
room temperature and below is hexagonal close packed (hcp). Upon heating, the
structure changes martensitically into the fcc-Co polymorph, stable up to the melting
temperature. The transition temperature for pure Co from the low- to the high-
temperature polymorph, i.e., austenite start (As, see Sect. 9.5.1), is 430
C. The
reverse transformation occurs, with some hysteresis, at the martensite start temper-
ature (Ms), equal to 370
C. Iron is body-centered cubic (bcc) up to 911
C (see Sect.
5.1). Above this temperature, the stable phase is the fcc austenite (γ). Therefore, the
common austenitic phase on which all superalloys are based needs to be stabilized in
the cobalt and iron alloys, using suitable stabilizers elements.
The austenitic phase exhibits several interesting features:
• The close-packed atomic arrangement would generally feature intrinsically lower
diffusivities, than, for instance, a bcc structure.
• A lower diffusivity results in a good stability of the precipitate phases and of the
alloy microstructure. Creep resistance too is beneficially influenced by a low
diffusivity (see Sect. 6.5.1).
• The hexagonal close-packed structure, although sharing the same packing factor
of the fcc structure, still has the important limitation of an intrinsic brittleness due
to the limited number of slip systems (see Sects. 3.3 and 4.2), just 3 as compared
to the 12. Actually, also the bcc structure has 12 slip systems, but none of them
comprises close-packed planes, like fcc. This feature has important consequences
on the stability of the dislocation in bcc structures and, thereby, on the ductility of
the alloy that turns out to be generally not so high as in fcc structures.
• The close-packed structure and the large number of slip systems of austenite are
prerequisites for better mechanical properties, as concerns also toughness, duc-
tility, and mechanical resistance.
The austenitic structure displays along close-packed directions elevated stiffness
that can be modulated by selecting alternative directions on purpose. This is the case
of turbine blades of the high-pressure stage. To enhance their thermal fatigue
resistance, their main axis is aligned parallel to the more compliant <001> direction
(see Sect. 6.5.1).
6.1 Introduction 269

Fig. 6.1 A scheme with the main application fields of superalloys. (Donachie and Donachie 2002)

One of the main drives for the development of superalloys has been their use in
aircraft GTEs. Thanks to their properties, superalloys represent nowadays the
highest percentage of all materials used in the fabrication of gas turbine engines.
However, the aerospace industry is not the only field where superalloys find suc-
cessful applications. An overview of the most important applications of superalloys
is shown in Fig. 6.1 (Donachie and Donachie 2002; Reed 2006).
Considering the application of superalloys in the production of gas turbine
components, material capabilities alone would not be sufficient to bear the extreme
conditions of the hottest stages. The situation is well represented by the curves of the
Larson-Miller parameter in Fig. 6.2 (see also Sect. 4.3.1.5), for different classes of
alloys interesting for high-temperature structural applications. The nickel-based
superalloys SRR99, CMSX-4, and the RR3000 all show a creep resistance that is
definitely better than other titanium alloys, intermetallic alloys, like the TiAl and
Ti3Al, in addition to iron and nickel aluminide alloys. The refractory alloys only,
based on niobium, tungsten, and rhenium, can do better (see Sect. 9.2), although the
elevated density is certainly a detrimental parameter as far as aerospace applications
are concerned.
Indeed, specific processing routes have been developed on purpose to produce
high-pressure hollow, air-cooled blades with single-crystalline superalloy, having
remarkable mechanical properties in association with extremely high surface stabil-
ity against corrosive attack. Superalloys have played a key role in the development
of high-performance gas turbine engines since the 1950s. At that time, 10% only of
the total weight of an aircraft engine was made of superalloys; the percentage
reached 50% in 1985.
An obvious approach to increasing the efficiency of a jet engine is to raise its
operating temperature (see Fig. 2.6, Brayton cycle). An important limiting factor are
270 6 Superalloys

Fig. 6.2 Larson-Miller parameter plot for several high-temperature alloys. (Redrawn from Reed
2006)

the materials used for the most demanding parts of the engine (see point 3 in the
Brayton cycle in Fig. 2.7). In this regard, the high-pressure blades all through the
development history of GTEs have been a most challenging component as concerns
material capabilities, which can be ranked, in the first place, with reference to the
turbine entry temperature (TET, see Figs. 1.14, 2.8, and 2.20). Indeed, the operating
temperatures are affecting the strength of the materials but also the deformation rate
in the relevant creep regime, largely determined by diffusive phenomena (see Sect.
6.5.1). Still, diffusion is paramount for the corrosive phenomena occurring at high
temperatures, contributing to reduce the service time of turbine blades. The success
of the superalloys is based on the combination of high-temperature structural
properties with very good resistance to high-temperature corrosion (see Sect. 8.2).
This aspect is well illustrated by Fig. 6.3, in which the behavior of several alloy
groups is comparatively considered (Geddes et al. 2010). The damages, particularly
on the blade surfaces, induced by creep and oxidation, can be important nucleation
sites for micro-cracks and flaws. In order to avoid catastrophic propagation to the rest
of the component, a further requirement for superalloys used in the turbine blading is
an excellent damage tolerance and high fracture toughness, particularly under a
fatigue regime.
Several superalloy parts of a GTE, in addition to high-pressure stage turbine
blades, would usually operate at temperatures in excess of 600
C, a fundamental
limit for titanium alloys (see Chap. 4). The success of the superalloys is to be
ascribed to the wide range of properties that can be attained through a suitable
selection of rather complex compositions and relevant processing routes, including
6.1 Introduction 271

Fig. 6.3 Interplay between environmental resistance and creep strength for several families of
metallic alloys. (Redrawn from Geddes et al. 2010)

the post-processing thermal treatments. All these aspects contribute to optimize the
diverse strengthening mechanisms (Fig. 6.4). Appendix 5 illustrates the main spe-
cific effects of the most important alloying elements in each superalloy family.
Appendix 6 presents a set of information concerning the phases that may form in
different superalloys. The superalloy design has been importantly guided by the need
for better high-temperature corrosion resistance. The general strategy of forming a
protective scale of reaction products, which for all superalloy families are based on
chromium and aluminum oxides (see Sect. 8.2), has been refined through the
selection of minority elements, like yttrium, scandium, cerium, lanthanum, etc.
(the so-called reactive elements), that resulted in an improved barrier effect of the
surface reaction layer (see Sect. 7.3.3).
At the present time, the environmental advantages that can be achieved through
alloy density reduction and consequent improved engine efficiency are becoming
major drives for further developments (see Chap. 2 and particularly Sect. 2.6).
As concerns processing, the early steps of the production route, common to the
three families of superalloys, are melting and refining. A possible process flowchart
for the production of superalloy components is given in Fig. 6.5. During primary
melting of the raw material or scrap, the chemical composition of the alloy is set.
Common primary melting routes for superalloys are VIM (see Sect. 5.3.2) and AOD
(see Sect. 5.3.3) of material coming from EAF melting (EAF + AOD or arc AOD
process, see Sects. 5.3.3 and 5.3.1, respectively). Once melting and conversion
272 6 Superalloys

Fig. 6.4 Effect of different strengthening mechanisms on creep-rupture life of cobalt-, iron-, and
nickel-based superalloys. (Redrawn from ASM International 2000)

processes are accomplished, the melt is sent to the casting unit. Castings of super-
alloys produced via arc AOD process are poured ingots, electrodes for secondary
melting, or continuously cast products. VIM is especially employed for the produc-
tion of electrodes for secondary melting, when higher purity standard is an issue.
ESR or VAR secondary processes, applied to superalloys produced via arc AOD,
can compensate to some extent for the lower quality of the produced material by
increasing the cleanliness and improving the solidification structures. As concerns
secondary melting, superalloys are generally processed either through electroslag
(see Sect. 5.3.5) or vacuum arc (see Sect. 5.3.6) remelting.
For obtaining finished or semifinished products, casting is definitely the most
common and suitable process, when maximum strength levels are required, being
typically associated with an optimum microstructural control. High volume frac-
tions of ordered precipitates, i.e., γ0 and γ00 , as well as the control of carbide
concentration and morphology can be attained using multistep post-processing
thermal treatments (see Sect. 6.6). Concerning the grain size of a cast alloy, it is
generally coarser than that obtained from forging processes, unless rapid chill
casting is used. Investment casting allows to control the grain alignment, using
directional solidification techniques (see Sect. 6.5), resulting in exceptional high-
temperature strength and creep resistance, when the alloy component is subjected to
stresses along the grain growth direction (Geddes et al. 2010). Hot-rolling and
6.1 Introduction 273

Fig. 6.5 Process flowchart for the fabrication of superalloy components. (Campbell 2006; Prasad
and Wanhill 2017)

forging are other interesting approaches to the production of superalloys compo-

nents. Concerning hot-rolling, material temperature, as well as process parameters,
like rolling speed and in-line cooling, must be carefully tuned in order to avoid the
formation of both surface and internal defects. Selecting the right concentrations of
aluminum and titanium is paramount when using forging to produce high toughness
alloys. These alloys may be very difficult to forge when the sum of the two elements
overcomes 6% approximately, with a consequent formation of an excessive fraction
of the γ0 -Ni3(Al,Ti) strengthening phase. A solution would be a forging process
carried out above γ0 -solvus. Owing to the proximity of the γ0 -solvus to the incipient
melting temperature, temperature control of the component is paramount, in order
to avoid unwanted local melting. The input for forging process can be either ingot
metallurgy (usually triplex process) or powder metallurgy (PM) products. The
choice of the production route is mainly a function of the chemistry of the selected
superalloy. For instance, Alloy 718 and Waspaloy are generally produced via ingot
metallurgy. The introduction of alloys like Astroloy and René 95 made it evident
that conventional ingot metallurgy was not suitable, thus leading to the develop-
ment of PM processing (Reed 2006). These products are also interesting for the
274 6 Superalloys

specialized thermal treatments that have been developed on purpose (see DMHT in
Sect. 6.6) to enhance further their already excellent properties through a pinpointed
development of specific microstructural features. Effective strengthening, due to
intermetallics and carbides, the ability to retain hardness also at high temperature,
and the work hardening tendency are all aspects that render the superalloys forming
via machining not always that easy. Cubic boron nitride (CBN) tools are generally
required.
Additional complications are faced when machining parts with an intrinsic
anisotropy due to directionally solidified (DS) or single-crystal (SC) casting (Geddes
et al. 2010). In the next sections, the main aspects concerning the cobalt- and iron-
based superalloys will be introduced. Nickel-based superalloys will follow, includ-
ing the description of specific processing routes, like directional and single-crystal
solidification (see Sect. 6.5) and relevant thermal treatments (see Sect. 6.6).

6.2 Cobalt-Based Superalloys

Cobalt is a transition metal (fourth period, ninth group) with atomic number 27. Its
name derives from the Latin Cobaltum, following the names of malicious spirits of
the Nordic mythology, the Kobolds. There was the belief that Kobolds, inhabitants of
the provenance mines of the cobalt ores, were used to steal good minerals,
containing more precious metals like gold and nickel, substituting them with other
useless ones, among which cobalt itself. Indeed, Co-containing minerals were often
mistaken for those containing more noble metals. Moreover, cobalt-containing
minerals released toxic vapors during smelting (Fontani et al. 2015).The element
was identified by Torbern Bergman in 1780, although it had been already unknow-
ingly isolated by Georg Brandt in 1742. In the 1920s, intensive research in the field
of magnetic materials for the electrical industry fostered a major interest for cobalt
alloys, in particular as concerns the production of permanent magnets. Several alloys
emerged from these studies, that explored the effects of alloying cobalt with alumi-
num, titanium, and nickel, resulting in the production of new alloys, such as
Alcomax, Alnico, and Titanal (Tylcote 2002). Nowadays, the largest suppliers of
cobalt ores are located in the so-called “Copper Belt” region in Africa. This area
comprises the Democratic Republic of Congo (the former Zaire) and Zambia. The
distribution of the world’s cobalt mining capacity is shown in Fig. 6.6. The main
refiners of cobalt ores, like Australia, Belgium, Canada, China, Finland, Japan,
Madagascar, Norway, Russia, and Zambia, purchase raw material concentrate of
this strategic and critical metal and ship it to their own production sites (ASM
International 2000).
Cobalt extraction is generally subsidiary to that of other metals, like copper and
nickel. Therefore, the extraction process varies according to the ores in which Co is
contained. The main ones are listed in Table 6.1.
6.2 Cobalt-Based Superalloys 275

Fig. 6.6 Cobalt world mine production 2016. Data are expressed in metric tons of Co content.
(Data from US Geological Survey 2017)

Table 6.1 The main cobalt ores and relevant main geographical origin (ASM International 2000)
Ore Composition Geographical origin
Asbolite Mixed Mn-Fe oxides + Co Congo, New Caledonia,
Zambia
Carrolite CuCo2S4, CuSCo2S3 Congo, Zambia
Cattierite CoS2 Congo
Cobaltite CoAsS Idaho (USA), Ontario
(Canada)
Heterogenite 2Co2O3CuO∙6H2O Congo
Laterites Weathered igneous rock containing limonite, Mg Australia, Brazil, Moa
oxide, silicates and 2–3% Ni+Co Bay (Cuba), Russia
Linnaeite Co3S4 (+Cu, Fe, Ni) Congo, Mississippi Valley
(USA), Zambia
Nickel (Co, Ni)S2 Congo
cattierite
Pyrrhotite Ni, Cu, (Co) sulfides Ontario (Canada)
Skutterudite (Co, Ni)As3 Bou-Azzer (Morocco),
Ontario (Canada)
276 6 Superalloys

Table 6.2 An overview of the cobalt extraction and refining processes (ASM International 2000)
Source Process details
Arsenide ores Arsenic-containing concentrate undergoes roasting in a fluidized bed
reactor in order to remove from 60% to 70% As. The resulting product can
be treated with either chlorine or sulfuric acid, resulting in a leach solution
that can be purified by hydrometallurgical processes. Cobalt is usually
recovered by carbonate precipitation or electrolysis
Cu-Co oxide The mineral undergoes crushing and milling before the metal-bearing
concentrates fraction undergoes concentration by froth-flotation to separate the valu-
able mineral from the unwanted gangue. Co-rich oxides are sintered in
pellets and then fed to electric smelting furnaces. The concentrate is then
mixed up with coal and lime and is melted in a reducing atmosphere. This
leads to the production of Co-Cu-Fe alloys. The light white alloy obtained
from this process contains approximately 40% Co. The compound is then
subjected to further hydrometallurgical refining to produce 99% purity
cobalt
Cu-Co sulfide The ores are treated by a sulfatizing roast in a fluidized bed furnace leading
concentrates to the conversion of Cu and Co sulfides into soluble oxides. If present, Fe
is turned into hematite. The calcine is subsequently leached with sulfuric
acid from the spent Cu recovery electrolyte, which partially enters the Co
recovery circuit. Purification is then performed prior to Co precipitation,
as hydroxide, which is then redissolved. Final Co refining is carried out by
electrolysis
Laterite ores Ni-Co laterite ores can be treated by either hydrometallurgical or pyro-
metallurgical processes. Hydrometallurgical processes can use ammonia
or sulfuric acid leach solutions. Pyrometallurgical processes such as
ferronickel or matte smelting require the entire ore to be melted to separate
the metal from the other residual components
Ni-Co sulfide These compounds can be treated by either roasting or flash smelting. The
concentrates resulting Co- and Ni-containing matte can be recovered
hydrometallurgically or treated by pressure leach with an ammonia solu-
tion. The Sherritt Gordon process uses a feed matte that contains approx-
imately 0.4% Co and 30% S, and the pressure leaching process is carried
out in an ammonia solution. The resulting compound contains in addition
to Co, also Cu and Ni. After Cu has been separated, the remaining solution
undergoes a further separation by selective reduction processes leading to
the production of a Co powder with a purity of 99%. Temperature,
pressure, and pH value are critical process parameters

The main extraction processes for cobalt are electrolytic, hydrometallurgical,

pyrometallurgical, vapormetallurgical extraction, or a combination of them.
Depending on the mineralogy of the ores, Table 6.2 provides synthetic information
on the relevant preferred processes.
Cobalt is used in several products, as shown by the block diagram in Fig. 6.7.
Cemented carbide tools, corrosion- and wear-resistant alloys, high-speed tool steels,
orthopedic implants, and, of course, superalloys are the main metallurgical products
involving cobalt. Cobalt-based compounds are interesting for quite a broad range of
products, like catalysts for chemical industry, pigments, food industry, and batteries.
6.2 Cobalt-Based Superalloys 277

Fig. 6.7 Most important uses of cobalt. (ASM International 2000)

Cathodes for lithium ion batteries are made of the cobalt mixed oxide, LiCoO2
(Fergus 2010). The main applications of cobalt are summarized by the scheme in
Fig. 6.7.
The development of cobalt-based superalloys dates back to the early 1900s. In
1907, Elwood Haynes patented (No. 873,745 US patent) a Co-Cr binary alloy and
for the production of this and other cobalt-based alloys created the Haynes Stellite
Company in Kokomo, Indiana (Kracke 2010). Several commercial cobalt-based
alloys nowadays available derive from Co-Cr-Mo and Co-Cr-W ternary systems
that were investigated first by Haynes. Both high strength and corrosion resistance
characterized the Co-Cr-based alloys developed by Haynes. The mechanical strength
was improved further thanks to the introduction of molybdenum and tungsten.
Haynes named this family of alloys Stellites, due to their star-like shiny luster.
Later on, in the 1930s and 1940s, Co-based alloys for high-temperature applications
featuring corrosion resistance were developed further, leading to the production of a
Co-Cr-Mo alloy with moderately low C content, used for biocompatible investment
casting applications: the Vitallium (Sjöberg 2010), a still existing tradename (ASM
International 2000). In the same period, General Electric (GE) was particularly
concentrated in the innovation process of high-strength aerospace alloys. A casting
process, originally used for dentistry implants, was extended to the production of
turbochargers components, using the newly developed Stellite 21 (Table 6.3),
directly derived from the prototype alloy Vitallium. This alloy exhibited, in addition
to high-strength level, interesting tribological, namely, wear resistance, properties.
Cobalt-based alloys can be regarded as the first set of alloys deliberately designed for
aerospace gas turbine components, exploiting the well-established carbide-based
strengthening mechanisms, which were boosted by the stronger cobalt-rich alloy
matrix. Stellites are by all means the prototypes of modern cobalt-based superalloys,
developed and still used for tribological applications, also as wear-resistant coatings,
deposited onto cutting tools and turbine blades to enhance their erosion resistance
against particulate matter carried by the hot gas flux (Antony 1983; Riddihough
Table 6.3 Nominal composition of cobalt-based superalloys. Concentrations are expressed in wt% (ASM International 2000; Donachie and Donachie 2002;
278

Geddes et al. 2010)

Alloy Other designations Cr Ni Co Mo W Al Fe C Others Form
AiResist 13 AR-13 21.0 – 62.0 – 11.0 3.4 – 0.45 Ta ¼ 2.0; Y ¼ 0.1 Cast
AiResist 213 AR-213 20.0 0.5 64.0 – 4.5 3.5 0.5 0.2 Ta ¼ 6.5; Y ¼ 0.1; Zr ¼ 0.1 Cast
AiResist 215 – 19.0 0.5 63.0 – 4.5 4.3 0.5 0.35 Ta ¼ 7.5; Y ¼ 0.1; Zr ¼ 0.1 Cast
FSX-414 – 29.0 10.0 52.5 – 7.5 – 1.0 0.25 B ¼ 0.01 Cast
FSX-418 – 29.5 11.0 52.0 – 7.0 – 2.0 max 0.25 B ¼ 0.012; Cast
Mn ¼ 1.0 max;
Si ¼ 1.0 max
Y ¼ 0.2
FSX-430 – 29.5 10.0 51.0 – 7.5 – – 0.4 B ¼ 0.027; Cast
Zr ¼ 0.9
Jetalloy 1570 – 20.0 28.0 46.0 – – – 2.0 0.2 Ti ¼ 4.0 Cast
Jetalloy 1650 – 19.0 27.0 36.0 12.0 0.2 B ¼ 0.02;
Ta ¼ 2.0;
Ti ¼ 3.8
Haynes 188 HA188 22.0 22.0 37.0 – 14.5 – 3.0 max 0.10 La ¼ 0.9 Wrought
HS-188
Haynes 21 HS21 27.0 3.0 64.0 5.0 – – 1.0 0.25 – Cast
Stellite 21
Haynes 25 L-605 20.0 10.0 50.0 – 15.0 – 3.0 0.10 Mn ¼ 1.5 Wrought
Stellite 25
J-1650 – 19.0 27.0 36.0 – 12.0 – – 0.2 B ¼ 0.02; Ta ¼ 2.0; Cast
Ti ¼ 3.8
MAR-M 302 – 21.5 – 58.0 – 10.0 – 0.5 0.85 B ¼ 0.005; Ta ¼ 9.0; Zr ¼ 0.2 Cast
MAR-M 322 – 21.5 – 60.5 – 9.0 – 0.5 1.0 Ta ¼ 4.5; Ti ¼ 0.75; Cast
Zr ¼ 2
6 Superalloys
MAR-M 509 MM-509 23.5 10.0 54.5 – 7.0 – – 0.6 Ta ¼ 3.5; Ti ¼ 0.2; Cast
Zr ¼ 0.5
MAR-M 918 – 20.0 20.0 52.0 – – – – 0.05 Ta ¼ 7.5; Zr ¼ 0.1 Cast
MP159 – 19.0 25.0 36.0 7.0 – 0.2 9.0 – Nb ¼ 0.6; Wrought
Ti ¼ 3.0
MP35-N MP35N 20.0 35.0 35.0 10.0 – – – – – Wrought
NASA Co-W-Re – 3.0 – 67.5 – 25.0 – – 0.4 Re ¼ 2.0; Cast
Ti ¼ 1.0
S-816 – 20.0 20.0 42.0 4.0 4.0 – 4.0 0.4 Mn ¼ 1.2; Nb ¼ 4.0; Cast
6.2 Cobalt-Based Superalloys

Si ¼ 0.4
Stellite 23 – 24.0 – 65.5 – 5.0 – 1.0 0.4 Mn ¼ 0.3; Si ¼ 0.6 Cast
Stellite 27 – 25.0 32.0 35.0 5.5 – – 1.0 0.4 Mn ¼ 0.3; Si ¼ 0.6 Cast
Stellite 30 – 26.0 15.0 50.5 6.0 – – 1.0 0.5 Mn ¼ 0.6; Si ¼ 0.6 Cast
Stellite 31 X-40 22.0 10.0 57.5 – 7.5 – 1.5 0.50 Mn ¼ 0.5; Si ¼ 0.5 Cast
Stellite B – 30.0 1.0 61.5 – 4.5 – 1.0 1.0 – Wrought
UMCo-50 Haynes 150 28.0 – 49.0 – – – 21.0 0.12 – Wrought
X-45 – 25.5 11.0 56.0 – 7.0 – 2.0 max 0.25 B ¼ 0.01; Cast
Mn ¼ 1.0 max
V-36 – 25.0 20.0 42.0 4.0 2.0 – 3.0 0.27 Mn ¼ 1.0; Nb ¼ 2.0; Cast
Si ¼ 0.4
WI-52 – 21.0 – 63.5 – 11.0 – 2.0 0.45 Ta+Nb ¼ 2.0 Cast
279
280 6 Superalloys

1970). Stellite 21, based on the Co-Cr-Mo system, was followed by other two
important alloys: S-816 (Co-Ni-Cr) and X-40 (Co-Ni-Cr-W) for gas turbine blades
and vanes, respectively (see Table 6.3).
The leading position of cobalt alloys was rapidly overtaken by nickel-based
superalloys, and this situation is lasting until present for different reasons, in the
first place, the price, lower for nickel-based alloys mainly for a lower impact of the
raw material costs. The top properties achieved by the nickel-based superalloys are
by far better than those of cobalt alloys, except for a newer, recently developed group
of alloys (see Sect. 6.2.2). Eventually, on the eve of the potential extensive spread of
cobalt alloys in the aerospace market, the end of the Colonialism era led to consider,
in some way, unreliable supply of several raw materials, including cobalt, whose
main provenance was, and still is, the African state of Congo, former Zaire (see
Fig. 6.6). This set of reasons, and the contemporary achievements attained with
the design of novel nickel alloys, reduced the interest for cobalt alloys. Nonetheless,
for some specific applications in gas turbine engines, they are still the preferred
choice.
Cobalt-based superalloys feature incipient melting temperature larger than that of
nickel-based alloys, considering that cobalt has a melting temperature that is 40
C
higher than that of nickel. The diffusivity of alloying elements and the stacking fault
energy (SFE) are comparatively lower (Neumeier et al. 2016). These two features
favor good creep performances, considering that a lower SFE is a prerequisite to
achieve higher concentrations of dislocations and, thereby, higher frequency of
dislocation locks. Carbides, on which the main strengthening mechanisms of these
alloys are based, are stable up to higher temperatures than γ0 precipitates, the main
strengthening agents of nickel superalloys, on which they rely for their exceptional
mechanical properties. Moreover, cobalt alloys exhibit a good weldability and
excellent hot corrosion resistance in particular conditions, i.e., sulfate-based deposits
(see Sect. 8.2.4).
Concerning the composition, Co superalloys usually contain more than 60 wt%
of cobalt. Nickel and iron are present as stabilizers of the austenitic, high-
temperature polymorph of cobalt. Chromium is usually present in the concentration
range of 20–30%, while molybdenum, tungsten, and other refractory metals sum up
to a maximum value of about 10%, whereas carbon is present in concentrations not
exceeding 1 wt% approximately. The chemical compositions of some cobalt-based
superalloys are given in Table 6.3.
The right combination of alloying elements in cobalt alloys is intended to attain
both solid solution and carbide strengthening and surface and γ phase stabilization.
Cobalt-based superalloys do not generally require vacuum melting, and the process
can be carried out by air induction melting, except for those alloys, like MAR-M
509, MAR-M 302, and MAR-M 918, containing Al, Ta, Ti, and Zr. For these
formulations, VIM is required (Campbell 2006).Vacuum melting of other Co alloys
would lead to enhancement of their strength and ductility, thanks to an improved
cleanliness and microstructural control. The charge that should undergo the primary
melting process is generally the mixing of virgin material with selected scraps. The
amount of virgin material is usually around 40%. Concerning argon oxygen
6.2 Cobalt-Based Superalloys 281

Fig. 6.8 A comparison between Ni and Co superalloys. (a) Creep properties (stress rupture) where
PLM is the Larson-Miller parameter; (b) ultimate tensile strength. (Redrawn from Sims et al. 1987)

decarburization (AOD), the process can be successfully used for the reduction of
the C, Si, and S content in Co-based superalloys, as well as for the reduction of the
dissolved gases. The control on sulfur is performed through the use of artificial slags,
rich in CaO, or other basic oxides, capable to reduce sulfur levels down to 10 ppm, or
less (ASM International 2008). Co-based superalloys are frequently remelted by
means of ESR or VAR.
According to the graphs in Fig. 6.8, adapted from the classical textbook Super-
alloys II (Sims et al. 1987), cobalt alloys still have interesting stability and associated
structural properties at high temperatures, although not achieving the top mechanical
strength and creep resistance of nickel-based superalloys. This defines the typical
application fields for these alloys, generally used in components that operate under
low stresses but still need excellent hot corrosion resistance. Incidentally, in the
Larson-Miller plot in Fig. 6.8a, a set of data for two cobalt alloys, not strictly
classified as “superalloys,” are shown. The first is an oxide dispersion strengthened
(ODS see Sect. 9.3) alloy named TD-Co-Cr. TD means “thoriated”, i.e., strength-
ened by a dispersion of thorium oxide (ThO2) particles. Incidentally, it is worth
making clear the pure historical interest of TD alloys, all abandoned for the radio-
active character of the oxide particles containing thorium. As regards the DS-CoTaC
alloy, it is classified as an in situ composite and consists of a metallic matrix, a Co-Cr
alloy in this case, reinforced by tantalum carbides (TaC), that form directly from the
melt during the alloy solidification. The process is conducted following the direc-
tional solidification principles (see also Sect. 6.5). Therefore, the carbides rather
resemble to long fibers that provide exceptional structural properties to the
282 6 Superalloys

component. Both ThO2 particles and TaC elongated grains can guarantee remarkable
strengthening mechanisms up to temperatures much higher than those accessible to
the metallic counterparts, as a further confirmation of the typical application fields of
these cobalt alloys.

6.2.1 Main Constituent Elements and Relevant Strengthening

Mechanisms

Each Co-based superalloy contains different alloying elements (see Table 6.3),
whose main roles in the alloy are summarized in Table 6.4.
Carbide dispersion and solid solution are the main strengthening mechanisms
in Co-based superalloys that, together with the high-temperature corrosion
requirements, determine the main families of constituent elements of these
alloys: γ ( fcc)-phase stabilizers and strengthener, surface stabilizers, and carbide
formers. Nickel and iron, added to stabilize the high-temperature fcc polymorph
of cobalt down to room temperature, tend to reduce the corrosion resistance of the
alloy. This is among the main limiting factors to the upper concentration of these
elements.
Chromium definitely is one of the most important alloying elements in cobalt
superalloys, in view of its positive influence on the high-temperature corrosion
resistance, through the formation of a Cr2O3-based protective layer on the alloy
surface. Moreover, chromium is an important carbide former. Other elements that
have beneficial effects on the corrosion resistance of Co-based superalloys are
aluminum and the so-called reactive elements (see Sec. 6.1). Yttrium is particularly
effective when added to alumina-forming alloys, and in fact, the AiResist alloys may
contain up to 4.3 wt% aluminum (see Table 6.3). Lanthanum is another reactive
element, particularly suited for chromia-forming alloys, like Haynes 188.

Table 6.4 Alloying elements in Co-based superalloys and relevant role in the alloy – see also
Appendix 5 for a comparative overview, referring to all superalloy families (Geddes et al. 2010)
Expected effect Alloying elements
Carbide formation C, Cr, Mo, Nb, Ta, W
Austenite stabilization Fe, Ni
Grain boundary refinement B, C, Zr
Hot corrosion resistance La, Th, Y
Intermetallics and/or other hardening precipitates Al, Mo, Ta, Ti, W
Carbonitride formation C, N
Oxidation resistance Al, Cr
Rupture strength B, Zr
Solid solution strengthening Cr, Mo, Nb, Ni, Ta, W
Sulfidation resistance improvement Cr
Workability improvement B, C, Fe
6.2 Cobalt-Based Superalloys 283

Fig. 6.9 Isothermal section at 1200
C of the ternary Fe-Co-Cr phase diagram. The occurrence of
the σ phase is possible at concentrations in excess of 30 wt% of chromium. (ASM International
2000)

A maximum concentration of 30 wt% chromium is not exceeded in commercial

cobalt superalloys (Table 6.3), since above this limit the formation of the σ phase
becomes increasingly likely, at the operating conditions of the gas turbine engines. σ
phase is a topologically close-packed phase (TCP, see Sect. 6.4.2), and its stability
field is visible in the isothermal section at 1200
C of the Co-Cr-Fe phase diagram in
Fig. 6.9. The relevant degrading effects of TCP phases are discussed in Sect. 6.4.2.
The σ phase is a Cr-rich phase, whose formation may be due to carbide transfor-
mation, with the preferential replacement of carbon atoms by nickel and cobalt.
Other intermetallic phases may form as indicated in Table 6.5, as alternatives to
carbides.
Thus, Mo, Nb, Ta, W, and Zr are other constituents of carbides that can be found
in a cobalt alloy in association with Cr carbides. The optimum concentrations of
carbides can be attained thanks to the comparatively high carbon content of cobalt
alloys, as compared to other technological alloys:
• Co-based superalloys have typical concentration in the 0.25–1.00% range
(MAR-M 322).
• Ni-based superalloys have typical concentration in the range 0.05–0.20% range.
284 6 Superalloys

Table 6.5 Carbides and intermetallics in Co-based superalloys; see Appendix 6 for a comparative
overview on intermetallic phases that may form also in other superalloy families (ASM Interna-
tional 2000; Geddes et al. 2010)
Element Carbides Intermetallics Effects
B – – Increase stress-rupture strength
C MC, M7C3, M6C, – Strengthening through carbides precipitation
M23C6
Cr M7C3, M23C6 – Oxidation and sulfidation resistance; solid
solution strengthener
La – – Oxidation resistance
Mo M6C Co3M Solid solution strengthener
Nb MC, M6C Co3M Solid solution strengthener
Ta MC, M6C Co2M Solid solution strengthener
W M6C Co3M Solid solution strengthener
Y – – Oxidation resistance
Zr MC – Increase stress-rupture strength
M metal

Fig. 6.10 Portion of the periodic table of the elements with the indication of the main carbide formers
in Co-based superalloys and relevant structure (left). In the plot (right), the typical stoichiometry of the
main carbides forming in different alloys, as a function of the content of the carbide forming elements
and relative electronegativity difference with cobalt. (Redrawn from Sims et al. 1987)

• Stainless steels show typical concentration in the 0.02–0.2% range (although

some tool steels and some martensitic stainless steels such as AISI 440C may
contain up to about 1.0% C).
The formation of carbides is ruled by the difference in electronegativity of each
carbide-forming element, as shown by the graph in Fig. 6.10. In particular, elements
sitting in the columns of the periodic table to the left of cobalt, being more
electronegative than cobalt, would preferentially tend to form carbides.
Typical refractory metal carbides are MC and M6C types, generally with complex
compositions.
6.2 Cobalt-Based Superalloys 285

MC are primary carbides, characterized by cubic structure, equiaxed microstruc-

ture, and incoherent interface, and tend to form directly from the melt. Changes in
the composition and thereby in the crystallographic structure and relevant morphol-
ogy may occur during service, according to the reactions with the γ matrix of the
alloy:


MC þ γ ! M6 C at T > 1000 C

Example TaC þ ðCo; Ni; Cr; CÞ ! ðCo; NiÞ4 ðCr; TaÞ2 þ ðCo; Ni; Cr; CÞ ð6:1Þ

M6C carbides also tend to form from the melt, particularly in W- and Mo-rich
alloys. Similarly to MC carbides, they may undergo transformations, reacting with
the γ matrix of the superalloy:


M6 C þ γ ! M23 C6 at T > 800  900 C

Example : ðCo; NiÞ3 Mo3 C þ γðCo; Ni; Cr Þ

! Cr21 Mo2 C6 þ γðCo; Ni; Cr; MoÞ ð6:2Þ

Concerning chromium, there are three types of carbides that form in cobalt
superalloys, for which the binary Cr-C phase diagram in Fig. 6.11 provides relevant
indications. These carbides are:
• Cr3C2, forming from a peritectic reaction.
• Cr7C3 that forms during alloy solidification from the melt. In this case, the carbide
is classified as primary.
• Cr23C6, usually obtained from other carbides’ transformation.
Like refractory metal carbides, also Cr carbides are not strictly binary but feature
more complex compositions that may evolve during the alloy lifetime, with changes
similar to those described already for refractory metal carbides, including the
formation of the TCP σ phase.
Carbides, particularly primary ones, produce several beneficial effects on the
alloy stability and microstructure, with positive consequences on their mechanical
properties: blockage of grain boundary sliding, grain size stabilization against
recrystallization, and hindering of the dislocation movement (Orowan mechanism,
see Sects. 3.2.3.3 and 6.4.3).
The evolution of carbides, upon exposure to operating conditions, would gener-
ally determine a downgrading of the mechanical properties of the alloy, since both
carbide and solid solution strengthening mechanisms are reduced (see Table 6.6).
The formation of secondary carbides, and their coarsening and coalescence along
grain boundary regions, leads to an embrittlement and to reduction in the alloy’s
fracture toughness. At the same time, the depletion of the alloy matrix of the solid
286 6 Superalloys

Fig. 6.11 Chromium-carbon binary phase diagram with the indication of the main carbide phases
interesting also for cobalt alloys. (Redrawn from Massalski et al. 1986)

solution strengthening elements results in a constitutional weakening of the alloy

(see also Sect. 6.4.2 on the weakening effects of TCP phases). To counterbalance
this effect, cobalt alloys in which the precipitation of ordered intermetallic as
strengthening agents occurs have been designed at different stages and lately (see
Sect. 6.2.2) with extremely promising outcomes. Refining heat treatments of cobalt
alloys are an additional factor which can be exploited for improving their perfor-
mances. A complete overview on the main aspects of these thermal treatments can be
found in Chandler (1996).

6.2.2 Latest Developments of Co Superalloys

Considering the enormous success of nickel-based superalloys, mainly relying on

the very effective strengthening due to the precipitation of the γ0 phase with an L12
structure (see Appendix 6), a few cobalt alloys have been developed in which a
similar mechanism has been implemented. Since an analog of the Ni3Al intermetal-
lic, the prototype of the γ0 phase in Ni alloys, is missing in the Co-Al phase diagram,
Table 6.6 Tensile properties of selected Co-based superalloys at three temperatures. E% ¼ fracture strain (ASM International 2000)
UTS at UTS at E% at UTS at E% at
RT Yield strength E% at 650
C Yield strength at 650
C 870
C Yield strength at 870
C
Alloy [MPa] at RT [MPa] RT [%] [MPa] 650
C [MPa] [%] [MPa] 870
C [MPa] [%]
AiResist 13 600 530 1.5 475 385 4.5 290 275 21.0
AiResist 213 1200 625 14.0 960 455 28.0 315 220 55.0
FSX-414 740 440 11.0 485 215 15.0 310 165 23.0
6.2 Cobalt-Based Superalloys

Haynes 25 1005 475 51.0 745 330 60.0 415 250 45.0
Haynes 188 (sheet) 945 465 53.0 710 275 59.0 415 250 64.0
Haynes 188 (plate) 985 470 56.0 755 265 73.0 450 250 77.0
MAR-M 302 930 690 2.0 785 450 – 450 310 11.0
MAR-M 322 825 625 3.2 655 415 6.0 550 345 12.0
MAR-M 509 785 570 4.0 560 370 7.0 350 290 20.0
MP159 (annealed) 850 400 60.0 – – – – – –
MP159 (48% cold- 1585 1415 12.0 – – – – – –
worked)
MP159 (48% cold- 1895 1825 8.0 – – – – – –
worked + aged)
MP35-N (25% 1172 1034 28.0 – – – – – –
work strengthened)
MP35-Na 1282 1207 24.0 – – – – – –
UMCo-50 (as-cast) 550 315 8.0 – – – – – –
UMCo-50 925 610 10.0 – – – – – –
(wrought)
X-40 745 525 9.0 515 260 12.0 325 – –
a
25% work strengthened +538
C aged and air cooled
287
288 6 Superalloys

the early attempts were made with alloy formulations resulting in the precipitation of
the ordered Co3Ti intermetallic compound, indeed with the same L12 structure of the
γ0 phase in Ni superalloys (Sims et al. 1987). These early attempts were subsequently
abandoned for several reasons: the comparatively low solution temperature of
Co3Ti, ranging from 815 to 900
C, depending on the actual alloy composition.
Moreover, the lattice mismatch between Co3Ti and the austenitic alloy matrix
resulted to be larger in absolute terms than 1%, a value recommended both for an
adequate stability of the precipitates against recrystallization and for better high-
temperature properties (Rogister et al. 1967; Viatour et al. 1973). Moreover, the
maximum volume fraction of the reinforcing intermetallic phase that can be obtained
is just 20%, due to the limited solubility of Ti in Co (12 at% approx.). This value is
definitely low, if compared with 70% precipitate percentage attained in state-of-the-
art nickel superalloys (see Sect. 6.4) (Murray 1982, 1987). A further drawback of the
Co-Ti system emerged when Ni additions have been attempted to produce Ni-Co
superalloys. The formation of plate-like η-Ni3Ti (D024 structure) precipitates was
observed, whose cellular structure influenced negatively the fracture toughness of
the alloy (Cui et al. 2006). All these factors cooled down, but not eliminated, the
interest for cobalt alloys. The next step in the development of new alloys consisted in
the stabilization, through ternary additions, of a more suitable than the Co3Ti ordered
intermetallic as strengthening phase of the austenitic γ-matrix. The goal was
achieved with a new generation of cobalt alloys, based on the ternary Co-Al-W
system, having as strengthening phase the ternary L12 ordered γ0 phase: Co3(Al,W).
The mechanical properties of these alloys turned out to be extremely promising for
structural applications at high temperatures, although some critical features required
some amendment. In fact, the newly developed alloys had a relatively high density,
considering that the investigated compositions have tungsten concentrations in the
15–30 wt% (5–11 at%) range, corresponding to a maximum density of the alloy of
9.4 g/cm3 (a reference alloy composition in at% is Co-9Al-8W). This density value is
definitely higher than the density of the conventional MAR-M 509, i.e., 8.8 g/cm3
(Bauer et al. 2010; Sato et al. 2006; Suzuki and Pollock 2008). Moreover, the γ0
solvus of 990
C only and a narrow compositional stability range of the two-phase
γ-γ0 field represent other important limitations to any further alloy development.
Alloys with slightly better properties were developed more or less at the same time,
with more complex compositions, still featuring a γ-γ0 structure and a relatively high
concentration of nickel. The best performing material was the alloy with composi-
tion Co-28.6Ni-10.9Ti-8.0Cr-1.3Al-1.5Mo-0.6W wt% (Co-27.6Ni-12.9Ti-8.7Cr-
2.6Al-0.8Mo-0.2W at%) (Cui et al. 2006). So far, the latest step of the alloy
development is represented by γ-γ0 alloys of the Co-Ti-Cr system. The alloy
Co-11Ti-15Cr (at%, alloy codename CTC-1), reinforced by the L12 Co-Ti14.1-
Cr11.3 γ0 phase, has a density of 8.1 g/cm3, i.e., 14% lower that the density of the
Co-9Al-8W alloy. Moreover, this material has a γ0 solvus in excess of 1100
C and
the possibility of precipitating up to 60 vol.% of this reinforcing phase.
Figure 6.12 displays the evolution of the yield strength with temperature of
various Co-based superalloys. The Co-11Ti-15Cr (code-named CTC-1) alloy results
to be stronger than the other alloys in the graph over the entire temperature range.
The relevant curve displays a yield strength peak, typical of the γ0 strengthened
6.3 Iron-Based Superalloys 289

Fig. 6.12 Yield strength as a function of the temperature for the Co alloys listed in the graph
legend. The composition of alloy CTC-1 is Co-11Ti-15Cr at %; that of the Co-based superalloy
Mar-M 509 is given in Table 6.3. (Redrawn from Zenck et al. 2017)

alloys (see Sect. 6.4), that is not exhibited by the conventional, carbide, and solid
solution strengthened MAR-M 509 Co superalloy (see Sect. 6.2.2). In the same
graph, the yield strength curves for the alloys Co-9Al-8W and Co-12Ti (Sato et al.
2006) are drawn for a comparison.

6.3 Iron-Based Superalloys

As concerns composition and microstructural features, iron-based superalloys are

somehow intermediate between austenitic stainless steels and nickel-based superal-
loys. Therefore, it is not surprisingly that Fe-based superalloys are characterized by
the same production and processing routes considered for austenitic stainless steels
(see Sect. 5.4.3.1). The era of austenitic stainless steels began in the 1910–1915,
fostered by the increasing needs, as concerns structural stability at higher tempera-
tures, of the aerospace industry. This led to the birth of early Fe-based superalloys
with peculiar chemistry and improved properties. Initially, these materials, in addi-
tion to iron and nickel, contained a sufficient amount of chromium for improving the
oxidation resistance. Afterward, small amounts of aluminum, niobium, and titanium
started to be added, in order to produce a coherent, creep-resistant phase, named γ0
(Sims 1984). Most of Fe-based superalloys contain high concentrations of nickel,
which sometimes, like in the most popular Alloy 718, is still the majority element.
290 6 Superalloys

This aspect justifies the nomenclature occasionally adopted in the literature and in
the present text too of Ni-Fe superalloys. Iron is present in concentrations up to 75 wt
%. Nickel is not only meant to stabilize the austenitic fcc crystallographic structure
but also contributes to the formation of strengthening ordered precipitates, like in
nickel-based superalloys (see Sect. 6.4.3) and in the newly developed cobalt-based
superalloys (see Sect. 6.2.2). The compositions of the main Fe-based superalloys are
given in Table 6.7.
Chromium is an essential component of iron-based superalloys, interesting for
carbide formation and, although to a limited extent, solid solution hardening.
Moreover, chromium provides a protective Cr2O3-rich native oxide scale, which is
an essential protection barrier against corrosion at high temperatures (see Sect. 8.2).
The different effects of other alloying elements are given in Table 6.8.

6.3.1 Strengthening Mechanisms and Classification of Iron-

Based Superalloys

This class of superalloys is perhaps the one that is nowadays receiving the lowest
research interest worldwide, if compared to nickel and, more recently, cobalt-based
superalloys. Nonetheless, iron-based superalloys are still very important for aero-
space applications, in view of the very well-established knowledge available on the
main products that are used for an extremely wide range of components, present in
almost all parts of a gas turbine engine. Some examples are: compressor discs and
blading, casings, heat exchangers, discs, and blading of the intermediate and
low-pressure stages of the turbine, exhaust system, and cans. Fe-based superalloys
are used in the wrought condition only, since in this way a better control of the grain
size and morphology can be attained (Geddes et al. 2010). One of the reasons for the
long-lasting success of these alloys is that they are relatively cheaper, if compared to
the other superalloy families. The lower cost is mainly due to the high Fe content.
Good mechanical strength at room and high temperatures, including also a good
creep resistance, and wear, oxidation, and corrosion resistance are other properties
that not surprisingly have considerably widened the fields of application of iron-
based superalloys since the early times of their development. Furthermore, they are
characterized by high fracture toughness and ductility. Just to mention a few
applications of iron-based superalloys, often in association with steels (see
Chap. 5): steam turbine power plants (as bolts and stack of gas reheaters), recipro-
cating engines (exhaust valves, hot plugs, turbochargers), metal processing (hot
work tools, forging dies), and pollutant abating systems, like scrubbers for large
diesel engines.
Precipitation hardening is the strengthening mechanism peculiar of Fe- and
Ni-based superalloys, and in some of them, two hardening phases may form
(see Fig. 6.13). One is the γ0 phase (Appendix 6 and Sect. 6.4.2), having an ordered
L12 structure based on the face-centered cubic lattice and reference stoichiometry
Table 6.7 Chemical compositions of selected Fe-based superalloys
Alloy Type Cr Ni Co Mo Ti Al Fe C Others
19-9DL SSH – 9.0 – 1.3 0.3 – 66.8 0.3 Mn ¼ 1.1
Nb ¼ 0.4
Si ¼ 0.6
W ¼ 1.3
Alloy 718 PH 19.0 53.0 – 3.0 0.9 0.5 18.5 0.08 max Cu ¼ 0.15 max
Nb ¼ 5.1
6.3 Iron-Based Superalloys

D-979 PH 15.0 45.0 – 4.0 3.0 1.0 26.44 min 0.05 B ¼ 0.01
Mn ¼ 0.75 max Si ¼ 0.75 max
W ¼ 4.0
Discaloy SSH 14.0 26.0 – 3.0 1.7 0.25 55.0 0.06 –
Hastelloy X SSH 22.0 49.0 1.5 max 9.0 – 2.0 15.8 0.15 W ¼ 0.6
Haynes N-1655 SSH 21.0 20.0 20.0 3.0 – – 32.2 0.15 La ¼ 0.2
N ¼ 0.15
Nb ¼ 1.0
W ¼ 2.5
Zr ¼ 0.002
Haynes 556 SSH 22.0 20.0 20.0 3.0 – 0.3 29.0 0.4 La ¼ 0.02
N ¼ 0.2
Nb ¼ 0.1
Ta ¼ 0.9
W ¼ 2.5
Incoloy 800 SSH 21.0 32.5 – – 0.38 0.38 45.7 0.05 –
Incoloy 800H SSH 21.0 33.0 – – 0.15–0.6 0.15–0.6 45.8 0.05–0.1 –
Incoloy 800HT SSH 21.0 33.0 – – 0.25–0.6 0.25–0.6 46.0 0.06–0.1 Cu ¼ 0.4
Mn ¼ 0.8
Si ¼ 0.5
291

(continued)
Table 6.7 (continued)
292

Alloy Type Cr Ni Co Mo Ti Al Fe C Others

Incoloy 801 SSH 20.5 32.0 – – 1.13 – 46.3 0.05 –
Incoloy 802 SSH 21.0 33.0 – – – – 44.8 0.35 –
Incoloy 901 PH 12.5 43.0 – 6.0 2.7 – 36.2 0.10 max –
Incoloy 903 PH 0.1 max 38.0 15.0 0.1 1.4 0.7 41.0 0.04 Nb ¼ 3.0
Incoloy 907 PH – 38.0 13.0 – 1.5 0.03 42.0 0.01 Nb ¼ 4.7
Si ¼ 0.15
Incoloy 909 PH – 38.0 13.0 – 1.5 0.03 42.0 0.01 Nb ¼ 4.7
Si ¼ 0.4
Incoloy 925 PH 20.5 44.0 – 2.8 2.1 0.2 29.0 0.01 Cu ¼ 1.8
Inconel 706 PH 16.0 42.0 – – 1.8 0.2 37.0 min 0.01 Nb ¼ 2.9
Si ¼ 0.05
Inconel 783 PH 3.0 29.0 34.0 – 0.1 5.4 25.5 – Nb ¼ 3.0
MA-956 ODS PM 20.0 – – – – 4.5 75.0 0.05 Y2O3 ¼ 0.5
Pyromet CTX-1 PH 0.1 max 37.7 16.0 0.1 1.7 1.0 39.0 0.03 Nb ¼ 3.0
REX-78 SSH 14.0 18.0 – 4.0 0.6 – 60.0 0.01 B ¼ 0.015
Thermo-span PH 5.5 25.0 29.0 – 0.85 0.45 34.5 0.004 Nb ¼ 4.8
Si ¼ 0.35
V57 PH 14.8 27.0 – 1.25 3.0 0.25 48.6 0.08 max B ¼ 0.01
V ¼ 0.5 max
W-545 PH 13.5 26.0 – 1.5 2.85 0.2 55.8 0.08 max B ¼ 0.05
PH precipitation hardening, SSH solid solution hardening, ODS oxide dispersion strengthened, PM powder metallurgy (ASM International 1993; Donachie and
Donachie 2002; Geddes et al. 2010)
6 Superalloys
6.3 Iron-Based Superalloys 293

Table 6.8 Alloying elements in iron-based superalloys and relevant effects. See also Appendix
5 for a comparative overview, concerning the compositions of all families of superalloys (Geddes
et al. 2010)
Desired effect Element required
Carbide formation Cr, Mo, Nb, Ta, Ti, W
Fcc matrix stabilization C, Ni, W
Grain boundary refinement B, C
Hot corrosion resistance improvement La, Y
Intermetallics and/or other hardening precipitates Al, Nb, Ni, Ta, Ti
Carbonitride precipitation C, N
Oxidation resistance Cr
Retarded formation of η phase Al, B, Zr
Rupture strength improvement B
Solid solution strengthening Cr, Mo
Sulfidation resistance improvement Cr
Workability improvement B, C

Fig. 6.13 Crystallographic structures of the γ0 Ni3(Al,Ti) and of the γ00 Ni3Nb intermetallics.
(Redrawn from Reed 2006)
294 6 Superalloys

Ni3(Al,Ti), although in real alloys much more complex compositions have been
experimentally measured (see Sect. 6.4.3).
The second precipitate phase is the body-centered tetragonal (bct) ordered struc-
ture, D022, designated γ00 , and with a reference composition Ni3Nb. The ordered
atomic occupation in both γ0 and γ00 has important effects on the dislocation
structure, inducing the formation of the so-called super-dislocations (see Sect.
6.4.1) and influencing their dynamics.
This results in the peak of the yield strength as a function of the temperature,
similar to the trend illustrated for cobalt alloys in Fig. 6.12. Several classifications of
the iron-based superalloys are available. Sims et al. (1987) proposed that they can be
grouped into four families:
• γ0 - Ni3(Al,Ti) strengthened alloys. This group comprises Ni-rich compositions
(Ni% > 40 wt%), such as Incoloy 901. Other important members of the group are
Fe-rich alloys, with a lower content of nickel, ranging in between 25 and 35 wt%.
An example belonging to this group of alloys is V57. Mostly for historical
reasons, the Tinidur alloy is also to be mentioned (Fe-30Ni-15Cr-1.8Ti-0.08C).
Friedrich Krupp developed it in 1936, and as suggested by its denomination, an
elevated strength was achieved by adding to an austenitic steel, a relatively high
concentration of titanium and aluminum. Due to its poor weldability, another
alloy replaced it for specific applications: Cromadur (Fe-12Cr-18Mn) (Kracke
2010). Although the actual strengthening mechanism, as it happened for
precipitation-hardened Al-alloys (see Sect. 3.2.3.3), was fully understood at a
later stage, through transmission electron microscope observations, its remark-
able effect on the alloy performances was an important boost for the development
of iron- and nickel-based superalloys. An elevated concentration of titanium is a
common feature to all these alloys. This choice is also determined by the
necessity to attain a better matching between the lattice parameter of the austenitic
γ matrix and the γ0 ordered precipitates, in order to retain a coherent interface (see
Sect. 6.4.1), as required to achieve an effective strengthening effect. On the other
hand, a critical side effect of the elevated titanium content is that the exposure at
too elevated temperatures, in excess of 650
C, results in the rapid formation of
the hexagonal η-Ni3Ti phase, to be accurately avoided considering its detrimental
effects on the alloy properties, particularly a reduction in ductility and fracture
toughness.
• γ0 - Ni3(Al,Ti) strengthened alloys with compositions on the Fe-rich range of
the Ni-Co-Fe system. In these alloys, mechanical strength is enhanced by
increasing the fraction of the precipitate phase, whose formation is favored by
the high aluminum concentrations (>1 wt%). The limiting application tempera-
ture is still 650
C, although an insufficient oxidation resistance rather than the
formation of embrittling phases sets the application limits. The need to overcome
the 650
C temperature has driven the development of iron superalloys in which
the strengthening is due to other phases than ordered precipitates, as described
further on, concerning the fourth group of iron alloys. Members of this group are
Incoloy 903, 907, 909, and Pyromet CTX-1 (see Table 6.7). These alloys are able
6.3 Iron-Based Superalloys 295

to combine high strength up to 650
C with a low thermal expansion coefficient

obtained thanks to the elimination of the ferrite-stabilizing elements. A small
amount of titanium is added to favor, during the aging treatments, the formation
of γ0 -Ni3(Ti,Al). In the Pyromet alloys, the stable hcp plate-like η phase may also
form (see Appendix 6). Table 6.7 shows that these alloys are essentially Cr-free
and therefore with a relatively limited oxidation resistance (ASM International
1993).
• γ00 - Ni3Nb strengthened alloys. In this group of alloys, the prevailing, although
not only, strengthening precipitate phase is γ00 -Ni3Nb, whose formation is ensured
by appropriate concentrations of niobium: 5.1 wt% in Alloy 718, 3 wt% in
Inconel 706, just to mention a few important examples, considering that Alloy
718 is still the most used superalloy ever. In this group of alloys, the prevailing
precipitate phase depends also on the concentration of aluminum. In fact, γ-
00
-Ni3Nb would preferentially form in alloys with lower concentrations of alumi-
num (<0.2 wt%). Aluminum, a component of the γ0 -Ni3(Al,Ti), contributes to the
precipitation of this phase if present with a concentration in excess of 0.5 wt%.
Similarly to the previous group of alloys, a limiting temperature of 650
C is set
by the possible formation of orthorhombic δ-Ni3Nb, evolving from the bct
γ00 -Ni3Nb. The formation of the δ-Ni3Nb has similar detrimental effects on the
alloy properties as the formation η-Ni3Ti phase (see also Appendix 6).
• Alloys strengthened by carbides and solid solution. These alloys, all featuring
a good oxidation resistance, can be divided into two main groups, according to
the strengthening mechanism. The first group includes alloys strengthened by the
precipitation of carbides, nitrides, and carbonitrides. The higher stability of these
secondary phases raises the maximum application temperature up to 800
C.
However, since the active strengthening mechanisms are not so effective as in
case of ordered precipitates, particularly as concerns the ultimate tensile strength,
lower applied stresses are recommended.
The second group of alloys is reinforced by solid solution, a reinforcing
mechanism that allows the materials to be used up to temperatures in excess of
1000
C. The main representatives of this family are described herewith.
Hastelloy X (see Table 6.7) is characterized by an exceptional stress corrosion
cracking resistance in specific environments, as well as by a remarkable oxidation
resistance. Furthermore, it possesses good ductility also after prolonged exposure
at temperatures at 650–870
C (Haynes International 2017a). Haynes N-155, also
known as Multimet, whose first production dates back to the late 1940s,
(Table 6.7), can be used in high-stress applications at temperatures up to
800–1100
C. Incoloy 800, in its three versions 800, 800H, and 800HT, has a
similar composition as the austenitic stainless steel AISI 330 (see Table 5.13).
They are usually supplied in the solution treated condition followed by
quenching. Quenching is necessary for giving the alloy a stable austenitic struc-
ture. Second phases that may form in these alloys are titanium carbides and
nitrides, as well as chromium carbides. Alloy 800 was first developed by INCO
in the 1950s aiming at the production of a heat- and corrosion-resistant alloy with
a relatively low nickel content, to save this strategic material. Incoloy 800 is not
296 6 Superalloys

Table 6.9 Typical heat treatments for precipitation hardening of Fe-based superalloys. The
treatment time intervals, in the 1–4 h range, depend on the plant features and part size and shape
(ASM International 2000; Chandler 1996)
Solution treatment temperature Aging temperature
Alloy [
C] [
C] Cooling process
Incoloy 903 845 620–720 Air or furnace
cooling
Incoloy 907 980 775–620 Air or furnace
cooling
Incoloy 909 980 620–720 Air or furnace
cooling
Incoloy 925 1010 620–730 Air or furnace
cooling
Alloy 718 980 620–718 Air or furnace
cooling
Pyromet 845–870 720 Air cooling
CTX-1

prone to the formation of the detrimental σ phase, even after long exposure times
at 650–870
C. Furthermore, it shows good resistance to carburization, oxidation,
and sulfidation (see Sect. 8.2), combined with creep and rupture strength up to
temperatures of about 800
C. It was found that high-temperature properties could
be improved thanks to a better control of the carbon content, and this led to the
development of the 800H composition (Everhart 1971; Farrar 2004). Alloy
800HT was developed next as a consequence of both the need for tight control
on aluminum and titanium content and to limitations in the grain size.
γ0 precipitation in Fe-based superalloys occurs for relatively low concentrations
of aluminum only, so that titanium turns out to be the main constituent of the
precipitate. For this reason, long-term exposure at temperatures above 650
C
would result in the conversion of the γ0 -Ni3Ti into the detrimental η phase (see
Appendix 6), which has an hcp structure and causes a decrease in the strength of the
alloy. Examples of thermal treatments used for precipitation hardening of Fe-based
superalloys are given in Table 6.9.
An alternative classification of iron-based superalloys can be found in Geddes
et al. (2010). Three groups of alloys are proposed:
• Precipitation hardenable alloys.
• Low coefficient of thermal expansion (CTE) alloys.
• Modified stainless steels, featuring solid solution hardening as primary strength-
ening mechanism.
The first family includes γ0 - and γ00 -strengthened alloys, irrespective of the γ
phase actual composition, i.e., Fe- or Ni-rich compositions. The second group
includes alloys like Incoloy 903, Incoloy 907, Incoloy 909, Inconel 783, and
Thermo-span, featuring the lowest CTE among all superalloys, with the additional
6.3 Iron-Based Superalloys 297

beneficial aspect that the CTE remains relatively constant even at higher tempera-
tures (Geddes et al. 2010). The low CTE results from the so-called Invar effect,
observed in diverse materials, including austenitic iron and nickel, ordered and
random, as well as amorphous alloys. In 1897 Charles Edouard Guillame discovered
austenitic iron alloys, having a nickel content around 35 at%, with a particularly low
thermal expansion over a wide range of temperatures. Materials exhibiting the Invar
effect also show anomalies in other physical properties such as atomic volume, Curie
temperature, elastic modulus, heat capacity, and magnetization. It has been found
that the Invar effect is related to magnetism. In practice, the contraction associated
with the loss of ferromagnetism upon heating the alloy counterbalances the normal
thermal expansion. In general chromium content is strongly reduced in these alloys,
since it depresses the Curie temperature (Table 6.7). Above the Curie temperature,
the material shows paramagnetic behavior, and therefore the thermal expansion is no
more compensated by any magnetic effect. A lower chromium content renders these
alloys more sensitive to corrosion phenomena (Van Schilfgaarde et al. 1999).
Finally, the third group, of modified stainless steels, is primarily strengthened via
solid solution hardening and, to a minor extent, via the precipitation of carbides and
ordered intermetallics. 19-9DL and Inconel 800H are two examples of alloys
belonging to this family. The first is essentially an 18–8 austenitic stainless steel
(see Table 5.13) with higher C content. The second has 21% of Cr and 33% of Ni
with additions of Ti and Al to promote the γ0 precipitation (Geddes et al. 2010).

6.3.2 Alloy 718 and Properties of Iron-Based Superalloys

Dealing with iron-based superalloys, a few words have to be spent on the most
popular alloy of this family the family: the γ00 -Ni3Nb strengthened Alloy 718, also
known as INCO 718, IN 718, or Inconel 718. It is regarded as one of the most
versatile and successful superalloys ever. Alloy 718 is used not only in the aerospace
field but also for the fabrication of components used in other industries, such as oil
and gas, power generation, and critical applications in hot corrosive environments.
Alloy 718 was developed to meet the needs of the jet engine market and to overcome
limitations of existing alloys, such as the PHSS A286 and the Ni-based superalloy
René 41. Alloy 718 was patented on July 24, 1962, by Herb Edelstein, and it has
been used in the aerospace field ever since, both for commercial and military
applications. The Rolls-Royce Trent 1000 and XWB (see Sect. 2.6) engines use
this alloy for several components. The GTE of the next generation of the defense
military aircraft, the JSF, is equipped with components made of Alloy 718. Its
versatility is demonstrated also by the high number of component elements (almost
20), melt practices (seven different schemes having multiple varieties), and
hot-working processing, such as hot-rolling and forging employed in the production
and processing of this alloy (Kracke 2010). Concerning the melt practices, critical
components made of Alloy 718 are produced via the triplex process (VIM-ESR-
VAR – see Sect. 5.3.6) and usually would not feature freckles or whites spots,
298 6 Superalloys

Fig. 6.14 TTT diagram for Alloy 718. (Redrawn from ASM International 2000)

possibly associated with the remelting practice (see Sect. 5.3.6). During solidifica-
tion, niobium, molybdenum, and titanium tend to segregate in the inter-dendritic
regions, where Nb-rich Laves phases (see Sect. 6.4.2) may form. A solution treat-
ment is required for dissolving these phases and homogenizing the alloy composi-
tion. γ00 precipitates are regarded as the main strengthening agents in Alloy 718. It
has been observed that an increasing niobium content results in a higher solvus
temperature for the δ- Ni3Nb phase, with obvious consequences on the design of the
solution treatment. γ00 can transform into γ0 and δ on increasing temperature and
exposure times. The formation of large amounts of δ phase at the expense of γ00 leads
to a heavy drop in the mechanical properties of the alloy. The TTT diagram for Alloy
718, with the relevant stability field of the alloy phases, is shown in Fig. 6.14. MC
and TiN do not change with thermal cycles at low temperatures and are considered as
inert phases, differently from γ00 , γ0 , δ, and Laves phases. The use of Alloy 718 in
critical applications promoted the development of two other variants: HS718 (high
strength) and DA718 (direct age). Both alloys employ processing improvements,
such as cleaner input stock for the melting process and a dedicated chemistry. In fact,
both HS718 and DA718 are characterized by a higher niobium content in the
5.0–5.5% range.
DA718 is characterized by a production process that avoids solution treatment
and uses a double aging treatment followed by water quenching. The double aging
process implies a first treatment at 718
C followed by a second at 621
C.
As-quenched material is characterized by retained strain that increases the γ00 and
γ0 precipitation kinetics (Radavich 1989; Schafrik et al. 2001).
One of the many reasons for the success of Alloy 718 are its exceptional
mechanical properties, whose evolution with temperature, as concerns UTS, yield
6.4 Nickel-Based Superalloys 299

Fig. 6.15 Evolution of tensile properties as a function of the temperature for a hot-rolled, annealed,
and aged 13 mm round bar of Alloy 718. (Redrawn from ASM International 2000)

strength, and elongation, is given in Fig. 6.15. Table 6.10 shows how this alloy is
ranked with respect to other iron-based high-strength superalloys (ASM Interna-
tional 1993).

6.4 Nickel-Based Superalloys

Nickel is a transition metal (fourth period, tenth group) with atomic number 28. The
melting temperature for pure nickel is 1453
C; its Young’s modulus is 207 GPa,
with a density of 8.902 g/cm3 (ASM International 2000), relatively high and due to
the combined effect of atomic mass and close packing bonds induced by the outer
shell d electrons, typical of transition metals (Reed 2006).
Apart from nickel of meteoritic origin (see also Sect. 5.1), the earliest reported
example of an artificial alloy containing nickel was the Pai-Thong (white copper).
This Chinese alloy was obtained by adding zinc to nickel-copper ores. Owing to
its aspect, nickel was used for the production of several decorative pieces, imported
to Europe by the East India Company in the seventeenth century. In the early
eighteenth century, attempts of smelting a newly discovered, presumed, copper ore
were performed in the German region of Saxony. Although unaware, miners, had
actually found niccolite, a nickel arsenide containing traces of Co, Fe, S, and Sb. The
result of the extraction attempts was a white metal, too hard to be effectively
hammered. For this reason, German miners thought the material was somehow
cursed, blaming for this Nickel, a deceptive demon in the German mythology.
Therefore, the name given to the new alloy was the German Kupfer-Nickel, literally
“copper-demon,” which was abandoned soon, since not regarded as technologically
 300

Table 6.10 Mechanical properties of selected Fe-based superalloys at room and high temperatures (650
C and 870
C). E% ¼ fracture strain (ASM
International 1993)
UTS at E% at
UTS at RT Yield strength E% at RT 650
C Yield strength at 650
C 1000 h rupture strength 1000 h rupture strength
Alloy [MPa] at RT [MPa] [%] [MPa] 650
C [MPa] [%] at 650
C [MPa] at 870
C [MPa]
19-9 DL 815 570 43.0 517 360 30.0 – –
Alloy 1350–1530 1105–1365 16.0–21.0 1130–1235 860–1090 19.0–24.0 405–600 –
718
Discaloy 1000 730 19.0 720 630 19.0 275 –
Haynes 815 410 48.0 590 225 52.0 275 55
556
Haynes 815 400 40.0 545 295 32.0 295 70
N-155
Incoloy 595 250 44.0 405 150 51.0 165 30
800
Incoloy 785 385 30.0 540 305 26.0 – –
801
Incoloy 690 290 44.0 525 200 25.0 170 69
802
Incoloy 1205 895 14.0 960 760 13.0 – –
901
Incoloy 1310 1105 14.0 1000 895 18.0 510 –
903
Incoloy 1365 1110 12.0 1035 895 10.0 – –
907
Incoloy 1310 1020 16.0 1025 870 24.0 345 –
909
V57 1170 830 26.0 895 745 22.0 – –
6 Superalloys
6.4 Nickel-Based Superalloys 301

Table 6.11 Main Ni-containing ores used in the modern production of primary nickel, with
relevant geographic provenance (ASM International 2000)
Ni-
containing
Type ore Chemistry Impurities Deposit
Laterite (hydrated mag- Garnierite (Ni, Co Central America
nesium-iron-silicates) Mg)6Si4O10(OH)8 Cr2O3 Cuba
MgO Indonesia
New Caledonia
Philippines
Laterite (limonitic- Goethite (Ni, Fe)2O3∙H2O Co South America
type) Cr2O3
Fe
Sulfide ore Pentalandite (Ni, Fe)9S8 Al2O3 Kola Peninsula
(Russia)
CaO Norlisk (Siberia,
Russia)
CuFeS2 South Africa
(chalcopyrite)
Fe7S8 Sudbury
(pyrrhotite) (Ontario,
Canada)
MgO Voisey’s Bay
(Labrador,
Canada)
SiO2 Western
Australia

promising. Nickel was identified by the Swedish mineralogist Axel Fredrick

Cronstedt in 1751, after 5 years of work using an instrument that previously was
utilized only by jewelers and metal workers: the blowpipe. Cronstedt named the
metal simply nickel after the Swedish word: kopparnickel, the translation of the
original German Kupfer-Nickel (ASM International 1991; Fontani et al. 2015;
Tylcote 2002).
Massive nickel production did not start until 1865, when an ever-increasing
market request for this metal, determined a major upscaling of the production plants.
In Table 6.11 the main nickel ores are recalled, with the indication of the main
mining sites that have been identified over the years.
Among the different Ni-bearing minerals, sulfides and laterites (oxides and
silicates) are the most important ones and represent the majority of the ores from
which primary nickel is nowadays extracted.
The industrial production of nickel was initially based on the so-called Orford
process, a technique introduced in 1890. The process involves the gravity segrega-
tion of nickel from a mixture of sodium sulfide and Cu-Ni-containing matte fused
together in a smelter with the help of a suitable coke charge. The separation occurs
during solidification when Cu-Na sulfide segregates above 99.8% pure Ni sulfide.
This intermediate product can be refined further using, for instance, a Bessemer
302 6 Superalloys

Table 6.12 Processes used for the refinement of Ni concentrates (ASM International 2000)
Process type Description
Hydrometallurgical Pure nickel cathode sheets are produced thanks to electrorefining and
electrowinning, two electrolytic processes. Electrorefining uses a nickel
electrolytic refining cell consisting of a thin nickel starting sheet (cathode)
and a slab of crude nickel that must be refined (anode). Both are immersed
into an aqueous electrolyte, and a direct current supply is fed into the cell.
Nickel, together with some impurities, undergoes dissolution in the elec-
trolyte. The solution obtained from this process is pumped out from the cell,
and then impurities are chemically removed. The electrolyte is purified and
returns to the cell depositing as pure nickel on the cathode. Electrowinning
is also known as electroextraction and consists in the electrodeposition of
metallic nickel
Pyrometallurgical Nickel concentrates are calcined using a roasting furnace. Then, the com-
pounds undergo smelting in a reverberatory furnace or in a similar equip-
ment, using natural gas or coal for fuel before being blown with air in a
converter. Further refining occurs in order to increase the purity level. In
particular, refinement is used to reduce the content of elements that may
have detrimental effects on the performances of the final alloy. Removed
elements are, for instance, As, Bi Cu, Fe, P, Pb, Sb, Sn, and Zn
Vapometallurgical The extraction of pure nickel from nickel oxide implies the use of a gas-to-
metal transformation according to the carbonyl process. Nickel oxide is
reduced with hydrogen, and the resulting nickel selectively reacts with CO
to form gaseous Ni-carbonyl. Subsequently, Ni-carbonyl is thermally
decomposed to obtain pure nickel, in the form of pellets or powders with a
very high purity degree

process, in order to obtain metallic nickel. The Orford process remained the only
used for the industrial production of nickel until 1948, when the development of new
and more efficient techniques led to its progressive replacement (ASM International
1991; Tylcote 2002). Instead of being smelted using sodium sulfide, the Cu-Ni matte
is allowed to slowly solidify in order to obtain the crystallization of the two separate
compounds, Ni3S2 and Cu2S, both solidified in form of blocks. Blocks are then
crushed and separated by means of froth flotation or magnetic methods. Ni3S2 is
oxidized to NiO and then reduced to metallic Ni (Tylecote 2002).
Several processes can be employed for the extraction of nickel from laterite ores.
One possibility is to dry up and preheat the ore, commercially known as ferronickel,
under reducing conditions, followed by melting in an electric arc furnace. The crude
metal is then refined and cast into ferronickel pigs (ASM International 2000; Davis
2000). Hydrometallurgical, pyrometallurgical, and vapormetallurgical were, and still
are, the main processes to refine the nickel concentrates (Table 6.12) (ASM Inter-
national 1991). The main market volumes of nickel, accounting for 47.1% of the
total, belong to China (33%), the USA (7.3%), and Japan (6.8%) (Nakajima et al.
2018). From an industrial point of view, the most important use of nickel, accounting
for more than 50% of the total usage, is as alloying element in ferrous alloys, i.e.,
steels (see Fig. 6.16).
6.4 Nickel-Based Superalloys 303

Fig. 6.16 Most important applications of nickel. (Davis 2000)

Nickel-based superalloys, in view of their strength and temperature capabilities

under different loading regimes, static, creep, and fatigue, in association with
corrosion resistance (see Figs. 6.2, 6.3, 6.4, and 6.8), are the most used materials
in gas turbine engines. This results from a combination of sophisticated alloy design
methodologies, capable to optimize rather complex compositions, and microstruc-
tural refinement, achieved with processing techniques developed on purpose for
specific part manufacturing. An excellent example regards single-crystal turbine
blades, with operating peak temperatures that can be raised up to 70% of their
incipient melting temperatures in Kelvin.
Thanks to vacuum melting, including VIM and VAR technologies (see Sects.
5.3.2 and 5.3.6), new alloy compositions were developed. Elwood Haynes, known
for the development of the cobalt-based superalloys (see Sect. 6.2), investigated the
Ni-Cr alloys too, used for wear- and oxidation-resistant applications and in 1907
patented a Ni-Cr binary alloy used in cutting tools for machining automotive
components. Albert Leroy Marsh was experimenting Ni-Cr alloys for electrical
applications, resulting in the patent of alloys of the Nichrome series. Furthermore,
the research effort made by Marsh paved the way for the development of Ni-Cr-Ti
alloys: the Nimonic series. The work on Ni-Mo alloys led to the development of the
Hastelloy family, known for their remarkable corrosion resistance. In 1919 the
International Nickel Company Research Laboratory in Bayonne (NJ, USA) started
a development project on high-strength Monel 400, used for steam turbine blades.
The chemical composition was tuned by increasing the Al content up to 5%, in order
to enhance age hardening, an important step forward in the development of the
modern nickel-based superalloys. The composition of this alloy was patented in
1926 (US Patent 1,572,744) and became later known as Monel K-500. Age hard-
ening was then patented by the International Nickel Company in 1930 (US Patent
304 6 Superalloys

1,755,554). Finally, Nimonic 80 and Nimonic 80A were developed during the
1940s. Nimonic 80 was developed in 1946 by Mond Nickel Company Ltd.: the
first steps of a success story that is still going on until the present times (ASM 1991;
Kracke 2010; Sims 1984).
Nickel-based superalloys generally contain at least 50 wt% of Ni, together with
different combinations of Cr, Al and Ti (all together summing up to 8 wt%), and Co
and refractory elements, like Mo, W, Nb, and Re, in lower concentrations. Ni-based
superalloys can be produced in both wrought and cast forms. The chemical compo-
sitions of some important wrought Ni superalloys are given by Table 6.13.
Wrought alloys are usually produced by remelting VIM ingots by means of VAR,
ESR, or electron beam cold heart remelting (EBCHR) processes. This leads to the
production of secondary ingots, which undergo deformation processing. The second-
ary processing is needed because Ni-based superalloys are very sensitive to chemical
and microstructural inhomogeneities and to the presence of inclusions. Furthermore,
large ingots produced via VIM are often characterized by macro-segregations and
shrinkage cavities. This is mainly due to the intrinsic low thermal conductivity of the
solidifying alloy, which limits the heat transfer out of the ingot. Defects are therefore
originated by large-scale solute segregations associated with dendritic solidification
under low thermal gradients. Special processing routes, such as powder metallurgy
(PM) or isothermal forging, or a combination of the two, can be employed for the
production of wrought highly alloyed compositions. For instance, Astroloy is an alloy
that can be processed in this way (ASM 2000; Prasad and Wanhill 2017), and, at the
present time, three generations of PM superalloys are present on the market. The
René 95 alloy belongs to the first generation of PM superalloys. Their most important
features are the fine grain size (in the 10 μm range) and high volume fraction of γ0
precipitates (around 55%), a combination used to achieve maximum tensile strength.
Therefore, sub-solvus heat treatments are employed in order to prevent modifications
in both the γ0 volume fraction and grain size that might degrade the resulting alloy
properties. The second generation of PM superalloys, like René 88DT, features lower
γ0 volume fraction and a larger grain size: they have been developed with task of
increasing the resistance against fatigue crack propagation. The LSHR and ME3
alloys belong to third-generation PM superalloys. They are characterized by a
generally higher volume fraction of γ0 , as compared to second-generation alloys,
and recently developed heat treatments (e.g., DMHT, see Sects. 6.4.3 and 6.6) are
used on these alloys in order to induce dual microstructure and, consequently,
improved properties within the same component (Reed 2006).
The chemical composition of wrought Ni-based superalloys has to comply with
the hot-workability requirements. The development of cast Ni-based superalloys has
been supported by two casting technologies, developed on purpose for high-
temperature components, in particular the blading of the high-pressure gas turbine:
directional solidification (DS) and single-crystal (SC) casting (see Sect. 6.5).
The chemical compositions of selected cast Ni-based superalloys are listed in
Table 6.14, whereas the compositions specifically designed for DS and SC solidifi-
cation processes are provided in Tables 6.15 and 6.16.
Table 6.13 Chemical analysis of selected wrought Ni-based superalloys
Alloy Strengthening mechanism Ni Cr Co Mo Al Ti Fe Mn Si C B Zr Others
Astroloy PH 55.0 15.0 17.0 5.3 4.0 3.5 – – – 0.06 0.03 – –
Hastelloy C-22 PH 51.6 21.5 2.5 13.5 – – 5.5 1.0 0.1 0.01 – – V ¼ 0.3 W ¼ 4.0
Hastelloy C-276 SSH 59.0 15.5 2.5 16.0 – – 5.5 1.0 0.1 0.01 – – V ¼ 0.3 W ¼ 3.7
Hastelloy S SSH 67.0 15.5 – 14.5 0.3 – 1.0 0.5 0.4 0.009 – La ¼ 0.05
Hastelloy W SSH 61.0 5.0 2.5 24.5 – – 5.5 – – 0.12 – – V ¼ 0.6
Haynes 214 SSH 75.0 16.0 – 4.5 – 2.5 – – – – Y ¼ 0.01
Haynes 230 SSH 57.0 22.0 2.0 0.3 0.5 0.4 0.1 La ¼ 0.02 W ¼ 14.0
6.4 Nickel-Based Superalloys

– – – – –
Inconel 617 SSH 54.0 22.0 12.5 9.0 1.0 0.3 – – – 0.07 – – Nb ¼ 3.6
Inconel 625 SSH 61.0 21.5 – 9.0 0.2 0.2 2.5 0.2 0.2 0.05 – – Nb ¼ 1.0
Inconel 700 PH Bal. 14.0 29.5 3.0 3.0 2.5 – – – 0.1 – – –
Inconel X750 PH 73.0 15.5 – – 0.7 2.5 7.0 0.5 0.2 0.04 – – –
LSHRa PH Bal. 13.0 21.0 2.7 3.5 3.5 – – – 0.03 0.03 0.05 Nb ¼ 1.5
Ta ¼ 1.6
W ¼ 4.3
M-252 PH 55.0 20.0 10.0 10.0 1.0 2.6 – 0.5 0.5 0.15 0.005 – –
ME3a PH Bal. 13.1 18.2 3.8 3.5 3.5 – – – 0.03 0.03 0.05 Nb ¼ 1.4
Ta ¼ 2.7
W ¼ 1.9
Nimonic 105 PH 53.0 15.0 20.0 5.0 4.7 1.2 – 0.3 0.3 0.16 0.008 0.1 –
Nimonic 115 PH 60.0 14.3 13.2 – 4.9 3.7 – – – 0.15 0.16 0.04 –
Nimonic 263 PH 51.0 20.0 20.0 5.9 0.5 2.1 – 0.4 0.3 0.06 0.001 0.02 –
Nimonic 75 SSH 76.0 19.5 – – – 0.4 3.0 0.3 0.3 0.1 – –
Nimonic 80A PH 76.0 19.5 – – 1.4 2.4 – 0.3 0.3 0.06 0.003 0.06 –
Nimonic 86 SSH 65.0 25.0 – 10.0 – – – – – 0.05 – Ce ¼ 0.03
Mg ¼ 0.015
(continued)
305
Table 6.13 (continued)
306

Alloy Strengthening mechanism Ni Cr Co Mo Al Ti Fe Mn Si C B Zr Others

Nimonic 90 PH 59.0 19.5 16.5 – 1.5 2.5 – 0.3 0.3 0.07 0.003 0.06 –
René 41 PH 55.0 19.0 11.0 1.0 1.5 3.1 – – – 0.09 0.005 – –
René 88DTb PH Bal. 16.0 13.0 4.0 2.1 3.7 – – – 0.03 0.015 0.03 Nb ¼ 0.7
W ¼ 4.0
René 95c PH 61.0 14.0 8.0 3.5 3.5 2.5 – – – 0.15 0.01 0.05 Nb ¼ 3.5 W ¼ 3.5
Udimet 500 PH 54.0 18.0 18.5 4.0 2.9 2.9 – – – 0.08 0.006 0.05 –
Udimet 520 PH 57.0 19.0 12.0 6.0 2.0 3.0 – – – 0.05 0.005 – W ¼ 1.0
Udimet 700 PH 55.0 15.0 17.0 5.0 4.0 3.5 – – – 0.06 0.03 – –
Udimet 710 PH 55.0 18.0 15.0 3.0 2.5 5.0 – – – 0.07 0.02 – W ¼ 1.5
Udimet 720 PH 55.0 17.9 14.7 3.0 2.5 5.0 – – – 0.03 0.033 0.03 W ¼ 1.3
Unitemp AF2-1DA6 PH 60.0 12.0 10.0 2.7 4.0 2.8 – – – 0.04 0.015 0.1 Ta ¼ 1.5 W ¼ 6.5
Waspaloy PH 58.0 19.5 13.5 4.3 1.3 3.0 – – – 0.08 0.006 – –
Custom age 625 Plus PH 61.0 21.0 – 8.0 0.2 1.3 5.0 – – 0.01 – – Nb ¼ 3.4
PH precipitation hardening, SSH solid solution hardening, PM powder metallurgy (ASM International 2000; Donachie and Donachie 2002; Reed 2006; Zhang
and Zhao 2012)
a
Third-generation PM superalloy
b
Second-generation PM superalloy
c
First-generation PM superalloy
6 Superalloys
Table 6.14 Chemical compositions of selected cast Ni-based superalloys. (ASM International 2000; Donachie and Donachie 2002; Kracke 2010; Zhang and
Zhao 2012)
Alloy Ni Cr Co Mo W Nb Al Ti C B Zr Others
Alloy 713C Bal. 12.5 – 4.2 – 2.0 6.1 0.8 0.12 0.012 0.10 –
C 130 Bal. 21.5 – 10.0 – – 0.8 2.6 0.04 – – –
C1023 Bal. 15.5 10.0 8.0 – – 4.2 3.6 0.15 0.006 – –
Hastelloy S Bal. 16.0 – – 15.0 – 0.40 – 0.01 0.009 – Fe ¼ 3.0
Mn ¼ 0.55
La ¼ 0.02 Si ¼ 0.65
6.4 Nickel-Based Superalloys

IN-939 Bal. 22.4 19.0 – 2.0 1.0 1.9 3.7 0.15 0.009 0.10 Ta ¼ 1.4
Inconel 625 Bal. 21.5 – 8.5 – 4.0 0.2 0.2 0.06 – – Fe ¼ 2.5
Inconel 700 Bal. 14.0 29.5 3.0 – – 3.0 2.5 0.1 – – –
M-22 Bal. 5.7 – 2.0 11.0 – 6.3 – 0.13 – 0.6 –
MAR-M 200 Bal. 9.0 10.0 – 12.5 1.8 5.0 2.0 0.15 0.015 0.05 –
MAR-M 200 Hf Bal. 9.0 10.0 – 12.5 1.0 5.0 2.0 0.14 0.015 – Hf ¼ 2.0
MAR-M 421 Bal. 15.8 9.5 2.0 3.8 4.3 1.8 0.14 0.015 0.05 –
MAR-M 432 Bal. 15.5 20.0 – 3.0 2.0 2.8 4.3 – 0.015 0.05 Ta ¼ 2.0
MM 002 Bal. 9.0 10.0 – – – 5.5 1.5 0.15 0.015 0.05 Hf ¼ 1.5 Ta ¼ 2.5
Nimocast 100 Bal. 11.0 20.0 5.0 – – 5.0 1.5 0.20 0.015 0.03 –
Nimocast 80 Bal. 19.5 – – – – 1.4 2.3 0.05 – – Fe ¼ 1.5
NX 188 Bal. – – 18.0 – – 8.0 – 0.04 – – –
René 41 Bal. 19.0 10.5 9.5 – – 1.7 3.2 0.08 0.005 0.01 –
René 77 Bal. 15.0 18.5 5.2 – – 4.25 3.5 0.08 0.015 – –
René 80 Bal. 14.0 9.5 4.0 4.0 – 3.0 5.0 0.17 0.015 0.03 –
(continued)
307
Table 6.14 (continued)
308

Alloy Ni Cr Co Mo W Nb Al Ti C B Zr Others
René 80 Hf Bal. 14.0 9.5 4.0 4.0 – 3.0 4.7 0.15 0.015 – Hf ¼ 0.8
Udimet 500 Bal. 18.5 16.5 3.5 – – 3.0 3.0 0.08 0.006 – –
Udimet 700 Bal. 14.3 14.5 4.3 – – 3.0 3.0 0.08 0.015 0.02 –
Udimet 710 Bal. 18.0 15.0 3.0 1.5 – 2.5 5.0 0.13 – 0.08 –
UDM 56 Bal. 16.0 5.0 1.5 6.0 – 4.5 2.0 0.02 0.07 0.03 V ¼ 0.5
Waspaloy Bal. 19.0 12.3 3.8 – – 1.2 3.0 0.06 0.005 0.01 Mn ¼ 0.45
6 Superalloys
Table 6.15 Chemical compositions of selected DS Ni-based superalloys. (ASM International 2000)
Alloy Generation Ni Cr Co Mo W Al Ti C B Zr Others
MAR-M 002 1 Bal. 8.0 10.0 – 10.0 5.5 1.5 0.15 0.015 0.03 Hf ¼ 1.5
Ta ¼ 2.6
MAR-M 200 Hf 1 Bal. 8.0 9.0 – 12.0 5.0 1.9 0.13 0.015 0.03 Nb ¼ 1.0
Hf ¼ 2.0
6.4 Nickel-Based Superalloys

MAR-M 247 1 Bal. 8.0 10.0 0.6 10.0 5.5 1.0 0.15 0.015 0.03 Hf ¼ 1.5
Ta ¼ 3.0
PWA 1422 1 Bal. 9.0 10.0 – 12.0 5.0 2.0 0.14 0.015 0.1 Nb ¼ 1.0
Hf ¼ 1.5
René 80 H 1 Bal. 14.0 9.0 4.0 4.0 3.0 4.7 0.16 0.015 0.01 Hf ¼ 0.8
CM 186 LC 2 Bal. 6.0 9.0 0.5 8.4 5.7 0.7 0.07 0.015 0.005 Hf ¼ 1.4
Re ¼ 3.0
Ta ¼ 3.4
CM 247 LC 2 Bal. 8.0 9.0 0.5 10.0 5.6 0.7 0.07 0.015 0.01 Hf ¼ 1.4
Ta ¼ 3.2
PWA 1426 2 Bal. 6.5 10.0 1.7 6.5 6.0 – 0.10 0.015 0.10 Hf ¼ 1.5
Re ¼ 3.0
Ta ¼ 4.0
René 142 2 Bal. 6.8 12.0 1.5 4.9 6.15 – 0.12 0.015 0.02 Hf ¼ 1.5
Re ¼ 2.8
Ta ¼ 6.35
309
310 6 Superalloys

Table 6.16 Chemical compositions of selected SC Ni-based superalloys. (ASM International

2000; Prasad and Wanhill 2017)
Alloy Generation Ni Cr Co Mo W Al Ti Ta Others
AF 56 1 Bal. 12.0 8.0 2.0 4.0 3.4 4.2 5.0 –
(SX 792)
AM1 1 Bal. 8.0 6.0 2.0 6.0 5.2 1.2 9.0 –
AM3 1 Bal. 8.0 6.0 2.0 5.0 6.0 2.0 4.0 –
CMSX- 1 Bal. 12.5 7.0 0.5 5.0 3.6 4.2 5.0 Nb ¼ 0.1
11B Hf ¼ 0.04
CMSX- 1 Bal. 14.9 3.0 0.4 4.5 3.4 4.2 5.0 Nb ¼ 0.1
11C Hf ¼ 0.04
CMSX-2 1 Bal. 8.0 5.0 0.6 8.0 5.6 1.0 6.0 –
CMSX-3 1 Bal. 8.0 5.0 0.6 8.0 5.6 1.0 6.0 Hf ¼ 0.1
CMSX-6 1 Bal. 10.0 5.0 3.0 4.8 4.7 2.0 Hf ¼ 0.1
PWA 1 Bal. 10.0 5.0 – 4.0 5.0 1.5 12.0 –
1480
PWA 1 Bal. 12.8 9.0 1.9 3.8 3.6 4.0 4.0 –
1483
René N4 1 Bal. 9.0 8.0 2.0 6.0 3.7 4.2 4.0 Nb ¼ 0.5
RR 2000 1 Bal. 10.0 15.0 3.0 – 5.5 4.0 – –
SC 16 1 Bal. 16.0 – 3.0 – 3.5 3.5 3.5 –
SRR 99 1 Bal. 8.0 5.0 – 10.0 5.5 2.2 3.0 –
CMSX-4 2 Bal. 6.5 9.0 0.6 6.0 5.6 1.0 6.5 Hf ¼ 0.1
Re ¼ 3.0
MC2 2 Bal. 8.0 5.0 2.0 8.0 5.0 1.5 6.0 –
PWA 2 Bal. 5.0 10.0 2.0 6.0 5.6 – 9.0 Hf ¼ 0.1
1484 Re ¼ 3.0
René N5 2 Bal. 7.0 8.0 2.0 5.0 6.2 – 7.0 Hf ¼ 0.2
Re ¼ 3.0
SC 180 2 Bal. 5.0 10.0 2.0 5.0 5.2 1.0 8.5 Hf ¼ 0.1
Re ¼ 3.0
CMSX- 3 Bal. 2.0 3.0 0.4 5.0 5.7 0.2 8.0 Nb ¼ 0.1
10 Hf ¼ 0.03
Re ¼ 6.0
René N6 3 Bal. 4.2 12.5 1.4 6.0 5.75 – 7.2 C ¼ 0.05
Hf ¼ 0.15
Re ¼ 5.4
TMS-75 3 Bal. 3.0 12.0 2.0 6.0 6.0 – 6.0 Hf ¼ 0.1
EPM- 4 Bal. 2.0 16.5 2.0 6.0 5.6 – 8.25 B ¼ 0.004
102 C ¼ 0.03
Hf ¼ 0.15
Re ¼ 5.95
Ru ¼ 3.0
Y ¼ 0.001
TMS- 4 Bal. 3.2 5.8 2.9 5.9 5.8 – 5.6 Hf ¼ 0.1;
138 Re ¼ 5.0
(continued)
6.4 Nickel-Based Superalloys 311

Table 6.16 (continued)

Alloy Generation Ni Cr Co Mo W Al Ti Ta Others
Ru ¼ 2.0
TMS- 4 Bal. 3.2 5.8 2.9 5.6 5.7 – 5.6 Hf ¼ 0.1
138A Re ¼ 5.8
Ru ¼ 3.6
TMS- 5 Bal. 3.0 5.8 3.9 5.8 5.8 – 5.6 Hf ¼ 0.1
162 Re ¼ 4.9
Ru ¼ 6.0
TMS- 5 Bal. 3.0 5.6 2.8 5.6 5.6 – 5.6 Hf ¼ 0.1
173 Re ¼ 6.9
Ru ¼ 5.0
TMS- 5 Bal. 4.6 5.6 2.4 5.0 5.6 – 5.6 Hf ¼ 0.1
196 Re ¼ 6.4
Ru ¼ 5.0
TMS- 6 Bal. 4.6 6.5 1.1 4.0 5.9 – 7.6 Hf ¼ 0.1
238 Re ¼ 6.4
Ru ¼ 5.0

The design of new alloy compositions, the refinement of their microstructure

through controlled thermal treatments, and the prediction of the relevant properties
under different operating conditions involve some metallurgical aspects, which will
be shortly revised in Sect. 6.4.1, with particular regard to solid interfaces and
dislocation dynamics. Sections 6.4.2 and 6.4.3 will illustrate the main selection
criteria of alloying elements in nickel-based superalloy, including the way to avoid
the formation of topologically close-packed (TCP) phases, like the σ phase,
discussed already with reference to cobalt-based superalloys (see Sect. 6.2.2).
The large variety of superalloy compositions affords a wide range of mechanical
properties, regarded as valuable selection criteria in view of any specific structural
application. A summary of the main mechanical properties of wrought and cast
Ni-based superalloys is provided in Tables 6.17 and 6.18, respectively.

6.4.1 Relevant Metallurgical Issues

The strengthening mechanisms interesting for structural alloys to be used in the

hottest part of a gas turbine involve phenomena occurring at different length scales:
from nanometer up to millimeter. Precipitation strengthening in nickel-based super-
alloys is mainly based on the formation of ordered the γ0 -Ni3(Al,Ti) precipitates (see
Sect. 6.3 and Appendix 6). Their ordered atomic arrangement determines peculiar
dislocation structures and, thereby, dynamics that will be considered and discussed
in details. Beforehand, the main kinds of interfaces that can form between crystalline
phases will be recalled, considering their central role in the strengthening
 312

Table 6.17 Mechanical properties of selected wrought Ni-based superalloys. E% is the fracture elongation
UTS at UTS at E% at 1000 h rupture 1000 h rupture
RT Yield strength E% at 760
C Yield strength at 760
C strength at 650
C strength at 870
C
Alloy Type [MPa] at RT [MPa] RT [%] [MPa] 760
C [MPa] [%] [MPa] [MPa]
Astroloy PH 1415 1050 16.0 1160 910 21.0 770 170
Hastelloy SSH 845 455 49.0 575 310 70.0 – 25
S
Inconel SSH 965 490 50.0 550 415 45.0 370 50
625
Inconel PH 1200 815 27.0 – – – 470 50
X750
Nimonic PH 1000 620 39.0 600 505 17.0 420 –
80A
Nimonic PH 1235 810 33.0 655 540 12.0 455 60
90
René 41 PH 1420 1060 14.0 1105 940 11.0 705 115
René 95 PH 1620 1310 15.0 1170 1100 15.0 860 –
Udimet PH 1310 840 32.0 840 730 39.0 760 110
500
Udimet PH 1185 910 7.0 1020 815 25.0 870 200
710
AF2- PH 1560 1015 20.0 1290 995 16.0 885 –
1DA6
Waspaloy PH 1275 795 25.0 650 675 28.0 615 110
PH precipitation hardening, SSH solid solution hardening. Tensile properties of all alloys are obtained from bar samples (Donachie and Donachie 2002; Geddes
et al. 2010)
6 Superalloys
Table 6.18 Mechanical properties of selected cast Ni-based superalloys. E% refers to the elongation at rupture. RT is referred to room temperature condition,
meaning 21
C
UTS at E% at UTS at high E% at high 1000 h rupture 1000 h rupture
RT Yield strength RT temperature Yield strength at high temperature strength at 815
C strength at 980
C
Alloy [MPa] at RT [MPa] [%] [MPa] temperature [MPa] [%] [MPa] [MPa]
Alloy 850 740 8.0 860a 740a 10.0a 305 70
713C
M-22 730 685 5.5 780a 685a 4.5a 385 130
b
6.4 Nickel-Based Superalloys

MAR- 930 840 7.0 325b – –b 415 125

M200
Nimocast 730 520 15.0 – – – – –
80
Udimet 930 815 13.0 895a 725a 13.0a 240 –
500
Udimet 1075 895 8.0 – – – 325 76
710
CMSX-2c 1185 1135 10.0 1295d 1245d 17.0d – 170
MM 002 1035 825 7.0 550e 345e 12.0d – 125
MAR- 1035 825 5.0 540e 345e 10.0d – 125
M200 Hf
PWA- – – 4.0 685e 495e 20.0d – –
1480c
UDM 56 945 850 3.0 945d 725d 5.0d – 125
Donachie and Donachie (2002) and Geddes et al. (2010)
a
¼ 538
C
b
¼ 1093
C
c
¼ Single crystal [001]
d
¼ 760
C
e
¼ 980
C
313
314 6 Superalloys

mechanisms in alloys, including those regarding the mechanical behavior of

Ni-based superalloys and also the latest generation of Co-based (see Sect. 6.2.2)
and Fe-based superalloys (see Sect. 6.3.1 and Appendix 6).

6.4.1.1 Solid Phase Interfaces

Interfaces, like the twins and grain boundaries separating the same phase (although
with different crystallographic orientations), are called homo-phase interfaces.
Hetero-phase interfaces are present instead, between a precipitate phase and its
hosting parent phase (i.e., the matrix). In this regard, grain boundaries separating
grains of two or more different phases would belong to this category. Since many
properties of engineering materials are determined by the structure of these solid-
solid interfaces, a classification has been developed. It is based on the relationship, if
any, between the crystallographic structures of the two phases across the interface.
Three main types of interfaces can be classified as coherent, semicoherent, and
incoherent.
A coherent interface is defined as a plane across which there is a continuity in the
crystallographic structures of the two phases (Fig. 6.17). This implies a correspon-
dence in the atomic occupation in the crystallographic planes facing each other
across the interface. Coherent interfaces can form if the two phases have the same
crystallographic structure. In this case, a coherent interface may form when equiv-
alent crystallographic planes face each other along specific crystallographic direc-
tions. The same may happen between phases with different structures but still having
planes with an equivalent atomic occupation geometry. A typical example is the
coherent interface which may form between the basal plane of the hcp structure and

Fig. 6.17 Scheme of a

coherent interface, featuring
a small misfit elastic strain
6.4 Nickel-Based Superalloys 315

the diagonal plane of the fcc structure, both close-packed planes. The coherency
orientation relationship can be written as:

f111gcubic ==f0001ghexagonal ð6:1Þ

< 110 > cubic ==

 >
< 1120 hexagonal ð6:2Þ

According to Eqs. 6.1 and 6.2, coherent interfaces are identified by parallel
crystallographic planes (Eq. 6.1), mating across the interface itself, and directions
(Eq. 6.2), indicating the orientation of the crystallographic plane. In Ni-based
superalloys, coherent interfaces separate the ordered γ0 -Ni3(Al,Ti) precipitates and
the austenitic γ matrix. Indeed, the difference between the lattice parameters of the
two phases is specifically designed, through an appropriate compositional selection,
to remain below the coherency limit over the whole temperature range interesting for
their applications, irrespective of any differential expansion induced by different
CTEs in the two phases. A coherent interface may tolerate differences in the
interatomic distances of the facing planes up to 0.05, i.e., 5%, which is regarded as
the coherency limit. Since both the γ matrix and the γ0 precipitate have cubic
structure, lattice strain is usually quantified in terms of the lattice misfit, δ, that can
be defined, in terms of the lattice parameters, aγ and aγ 0 of the two phases: γ matrix
and γ0 precipitate (Eq. 6.3) (Geddes et al. 2010; Reed 2006).

aγ 0  aγ
δ¼ 2 ð6:3Þ
aγ 0 þ aγ

Since it depends on the lattice elastic strain, the lattice misfit influences the
interface surface energy. In case of coherent interfaces, this term is comparatively
smaller than the surface energy term (Γch) due to the different chemical compositions
across the interface itself, which is indeed prevailing. Therefore, the surface energy
of a coherent interface (Γcoherent) is relatively low at room temperature, and it can be
expressed as Γcoherent ¼ Γch, with values usually lower than 200 mJ/m2. Considering
Ni-based superalloys, there exists a coherency orientation relationship between the
planes of the γ matrix and the γ0 precipitate, the so-called cube-cube orientation
relationship (Eqs. 6.4 and 6.5).

f100gγ ==f100gγ0 ð6:4Þ

< 100 > γ == < 100 > γ0 ð6:5Þ

In Sect. 6.3, γ00 -Ni3Nb has been introduced as another strengthening phase in
iron-based superalloys (Fig. 6.13). The bct D022 structure of the γ00 phase features an
ordered atomic arrangement with a symmetry similar to the ordered fcc L12 structure
316 6 Superalloys

of the γ0 phase. Also γ00 precipitates can form coherent interfaces with the γ matrix in
iron-based superalloys. Because of the tetragonal structure of the γ00 cell, two
coherency relationships are needed to describe the situation:

f100gγ ==f100gγ00 ð6:6Þ

< 100 > γ == < 100 > γ00 ð6:7Þ

and

f001gγ ==f001gγ00 ð6:8Þ

< 001 > γ == < 001 > γ00 ð6:9Þ

The coherency relationships influence the morphology and dimensions of the

precipitates. In γ0 -strengthened superalloys, the typical precipitate morphology is
cube like, whereas with γ00 phase, owing to the anisotropy of the tetragonal structure,
disc-shaped precipitates would usually form. The interest for coherent interfaces in
superalloys strengthened by the precipitation of ordered intermetallics is twofold. In
the first place, a low surface energy term is a prerequisite for a better stability of the
alloy microstructure against recrystallization and precipitate coalescence phenom-
ena, promoted by diffusion through chemical potential gradients (see Sect. 6.6 on
thermal treatments). Secondly, coherent interfaces have an important role in the
strengthening mechanisms, as concerns both static and creep loading regimes (see
Sects. 6.4.3 and 6.5.1 on precipitation hardening and creep mechanisms).
At higher coherency strains, resulting from larger differences in the interatomic
distances between precipitate phase and alloy matrix, in excess of the coherency
limit (5%), a complete coherency at the interface is lost. This determines the
formation of coherency dislocations (Fig. 6.18). A surface energy term due to the
lattice elastic strain at the interface (Γst), including also the dislocation contribution,
raises the overall interface surface energy, and the interface is now semicoherent.
The surface energy of a semicoherent interface is given by both a chemical and strain
contributions (Γsemicoherent ¼ Γch + Γst) with larger values than in coherent interfaces:
from 200 to 500 mJ/m2 (Smallman and Ngan 2014).
Two phases are separated by an incoherent interface (Fig. 6.19) when there is
neither continuity, nor correspondence between the relevant crystallographic struc-
tures. The surface energy contribution is mainly due to the volume mismatch
between precipitate and matrix. Carbides, topologically close-packed (TCP) phases
in steels and superalloys, are typical examples of incoherent precipitates. The surface
energy of an incoherent interface (Γincoherent) can reach values as large as 1000 mJ/
m2 (Smallman and Ngan 2014).
6.4 Nickel-Based Superalloys 317

Fig. 6.18 Scheme of a semicoherent interface. Note the presence of a misfit dislocation

Fig. 6.19 Scheme of an incoherent interface

318 6 Superalloys

6.4.1.2 Dislocations Structure and Dynamics in Ordered Alloys

From a microscopic standpoint, plastic deformation in crystalline materials can be

described in terms of dislocation movement and interactions with the microstructural
elements present in the materials, like precipitates, grain boundaries, and other
defects (Callister and Rethwisch 2012; Hull and Bacon 2011). The dislocation vector
notation provides a very useful tool, when dealing with these aspects. In fcc metals,
the elementary atomic displacement, i.e., the movement of each single atom (of the
dislocation line) from one to another lattice site, so that the symmetry of the crystal is
not broken, corresponds to the displacement along the cube diagonal, <110> direc-
tion, lying in the cube diagonal planes {111}, by half of its length. The
corresponding Burgers vector of this dislocation, describing an elementary displace-
ment, and named complete dislocation is:

a
h110i ð6:10Þ
2

where
• a is the lattice parameter of the cubic cell. Sometimes a is omitted in this
expression.
The movement of the atoms belonging to a complete dislocation line does not
occur in a straight line but rather with a zigzag movement. In other terms, a
dislocation dissociation occurs that can be expressed as:

a a a
h110i ! h211i þ h121i ð6:11Þ
2 6 6

Dislocations on the right side of the expression are called partial (Shockley),
since they do not represent a complete displacement between two equivalent atomic
positions. Nonetheless, their formation is favored since they reduce the lattice strain
energy. This energy term is proportional to the squared modulus of the Burgers
vector, and the above dissociation reaction confirms that the complete dislocation
has a larger energy than that of the sum of the relevant partials (Reed-Hill 1973). fcc
crystals result from an ABC stacking of {111} planes. The atomic displacement due
to the movement of a partial dislocation takes the atoms to nearest neighbor
positions, thus changing the correct stacking sequence, since atoms in the B layer
become C-atoms and so forth. A repulsive force is acting on the two partials, which
tends to increase their separation. An equilibrium distance is achieved when the
repulsion energy is counterbalanced by the stacking fault energy (SFE). The defect
structure consisting of two partials separated by a stacking fault band is called
extended dislocation, and it is, by all means, a planar defect (Fig. 6.20).
The dislocation structure becomes slightly more complex in case of the ordered
L12 structures. Slip systems are still the same as in an austenitic random alloy,
although the minimum displacement required to preserve the lattice symmetry and
6.4 Nickel-Based Superalloys 319

Fig. 6.20 Dislocation structures: dissociation of a complete dislocation into the extended disloca-
tions. Structure of a super-dislocation, typical of ordered intermetallics, like γ0 -Ni3(Al, Ti)

the correct atomic occupation of lattice sites is now doubled. The relevant Burgers
vector (Eq. 6.12) is equal to the entire cube diagonal, and it is named super-
dislocation:

a h110i ð6:12Þ

Similarly to the movement of a partial dislocation that creates stacking

faults in an
fcc lattice, the movement of an elementary complete dislocation a2h110i within an
ordered lattice induces the formation of a planar defect, featuring a wrong atomic
occupation. This defect is named antiphase boundary (APB). In analogy with
complete dislocations in fcc lattice, super-dislocations exhibit the tendency to
dissociate. The dissociation scheme is slightly more complex, as shown by
Fig. 6.20. The two complete dislocations, possibly dissociated into the relevant
Shockley dislocations according to Eq. 6.11, are now separated by a band of
antiphase boundary. The width of the APB is the results of the repulsive interaction
between the two complete dislocations and their attraction, which limits the positive
APB energy term.
An important effect determined by the super-dislocation dynamics is the anom-
alous dependence of the yield strength on temperature. In fact, as long as the leading
and trailing dislocation glide together on {111} slip planes, under the effect of a
shear stress, the width of the APB present in between the two component disloca-
tions remains substantially constant. Upon increasing the temperature, a cross-slip
mechanism, involving the leading super-partial dislocation, occurs, as a result of a
mechanical torque, the so-called Yoo torque, acting on each component of the super-
320 6 Superalloys

Fig. 6.21 Evolution of the

yield strength with
temperature for an austenitic
stainless steel (AISI 316), an
Fe-based superalloy
(Hastelloy X), the γ0 -Ni3Al
binary intermetallic with
0.2% B, and the advanced
Ni3Al-based intermetallics.
(Redrawn from Jozwik et al.
2015)

dislocation (Yoo 1987). In particular, the cross-slip from {111} octahedral planes to
{100} cube planes becomes more and more probable at increasing temperature.
Once the leading dislocation has moved to a {100} plane, the trailing one remains
blocked at the intersection between the two planes. These dislocation structures are
named Kear-Wilsdorf locks, after the names of their discoverers (Dimiduk 1991;
Kear and Wilsdorf 1962). Any further movement of the leading dislocation
becomes a difficult process and requires a widening of the APB and, thereby, an
increase of the associated energy. This is the reason for the increased strength of the
γ0 -Ni3(Al, Ti) phase with temperature, a phenomenon that is paramount as con-
cerns precipitation hardening of the nickel and of the new-generation cobalt
superalloys (see Sect. 6.2.2) and has been one of the main driving forces for the
development of novel alloy compositions. Figure 6.21 shows the yield strength
curves for the binary intermetallic compound, Ni3Al, which can be regarded as the
prototype of the gamma prime phase, compared to advanced Ni3Al alloys obtained
from the binary compound with specific additions (see Sect. 9.4). These curves are
displayed together with the yield strength curves for Hastelloy X, an iron-rich
Fe-based superalloy (see Table 6.7), and the austenitic stainless steel AISI 316 (see
Table 5.13). This latter displays a monotonous softening with temperature. A
similar trend is there in the Hastelloy X material, mitigated though in the high-
temperature range by the presence of ordered γ0 -Ni3(Al, Ti) precipitates. Since they
6.4 Nickel-Based Superalloys 321

become stronger at higher temperatures, a positive derivative change in the yield

strength is observed, although not to the same extent as in pure γ0 materials, i.e.,
γ0 -Ni3Al and advanced Ni3Al-based intermetallic alloys.
The peak in the yield strength, more or less intense, depending on the actual
composition of the alloy, is reached when the thermally activated unlocking of the
trailing dislocations occurs. At this stage, the activation of additional slip systems
results in a general softening of the alloy.
The yield strength dependence on temperature is strictly related to another
“anomaly,” observed in the creep behavior of the Ni3Al-based (Hemeker et al.
1991) and also of other intermetallic alloys (Smallman et al. 2002). In fact, over
an intermediate temperature range, close to the peak temperature of the yield strength
vs temperature curve (see Fig. 6.21), the dislocation structures that have been
considered under static loading conditions become important also to explain the
creep deformation behavior. The primary creep is sustained by the gliding of super-
dislocations (see Fig. 6.20) on the {111} octahedral planes. The coordinated move-
ment of these dislocation couples tends to saturate at the end of the primary creep
regime for the cross-slip that produces sessile dislocations on cube planes, on which
they are substantially blocked, i.e., the Kear-Wilsdorf locks (Hull and Bacon 2011).
This results in a slowing down of creep deformation. Subsequently, the cross-slipped
dislocations start both to bow out, generating more dislocation segments, and to
glide on the cube planes. The consequent increase in the density of mobile disloca-
tions on cube planes leads to an increase in the creep rate, that is, exactly the inverse
creep rate effect.

6.4.2 Selection Criteria for Alloying Elements in Nickel-

Based Superalloys: The γ Matrix

A common feature to superalloys in general and, in particular, to nickel-based ones is

the complexity of their compositions, required to satisfy the demanding operating
conditions typically faced in gas turbine engines and other structural applications at
high temperatures and in potentially corrosive environments. Adopting a composite-
like approach, a nickel-based superalloy can be regarded as made of the following
main components:
• The alloy matrix, i.e., a Ni-based solid solution with austenitic structure. Its
intrinsic strength and properties depend on the concentrations of alloying ele-
ments in solution.
• Reinforcing phases. γ0 -Ni3(Al,Ti) and γ00 -Ni3Nb ordered precipitates are the main
reinforcement of these alloys. In case of polycrystalline alloys, carbides are very
important too, with beneficial effects on the grain boundary stability, hindering
recrystallization and grain boundary slip. Direct carbide strengthening is based on
their interaction with dislocation lines, mainly through the Orowan bowing
mechanism (see Sects. 6.4.3 and 6.6). Carbides, indeed interesting for
322 6 Superalloys

polycrystalline alloys, are not even present in alloys developed for single-crystal
castings, although carbon is still having a role in the latest formulations of
Ni-based superalloys (see Table 6.16).
Moreover, the request for an intrinsic resistance against high-temperature corro-
sion determines the selection of specific alloying elements capable to produce
protective native mixed oxide scales. Although this requirement is mitigated by
the possibility of depositing suitable overlay coatings (see Sect. 7.3.1), still the
design of an alloy composition stable against diffusion-assisted phenomena, includ-
ing high-temperature corrosion, has beneficial effects also on creep resistance, for
which diffusion plays a central role. Controlling diffusion is also interesting as
regards the stability of the microstructure of the superalloys when exposed to service
conditions. The retention of an optimized microstructure depends also on preventing
the formation of unwanted phases, like the topologically closed packed (TCP) ones,
including the σ phase introduced already dealing with Co-based superalloys (see
Sect. 6.2.1 and Appendix 6). As concerns the nickel solutes, i.e., those elements that
enter the γ matrix of the nickel-based superalloys, they should exhibit the following
three main features: a good solid solubility in nickel, a high hardening coefficient,
and they should also contribute to the improvement of creep resistance of the
superalloy. The solubility of alloying elements in nickel can be assessed on the
basis of the difference of their atomic diameters with respect to that of nickel itself.
In Fig. 6.22, the diameters of several atoms, some of which potentially interesting
for being solid solution elements in nickel-based superalloys, are shown. The solid

Fig. 6.22 Atomic diameter of several elements. The shadowed area marks a range of variability of
15% with respect to the atomic diameter of nickel. For each element the maximum solid solubility
in nickel is also quoted. (Redrawn from Jena and Chaturvedi 1984; Reed 2006)
6.4 Nickel-Based Superalloys 323

Table 6.19 Some parameters relevant to solid solution hardening of nickel (Jena and Chaturvedi
1984)
%Δda Groups
%Δ in the lattice with separation with Change in yield
Melting parameter of nickel respect to respect to strength per at%
Element T [
C] per at% solute atom nickel nickel of solute [MPa]
Co 1493 0.005 0.2 1 +0.3
Cr 1907 0.032 0.2 4 +4.9
Cu 1085 0.025 2.5 1 +1.6
Fe 1538 0.031 1.3 2 +3.3
Mo 2623 0.116 9.5 4 +14.2
Ti 1670 0.095 16.0 6 +22.7
W 3422 0.135 10.0 4 +16.5
a
d indicates the atomic diameter

solubility limit of each element is also indicated. In general, a better solubility is

exhibited by those elements whose atomic diameter differs by no more than 15%
from the atomic diameter of nickel, i.e., if it falls within the shadowed band in
Fig. 6.22. For structural applications, it is important not only the concentration that
each solute element can reach but also how effective this element is in solid solution
strengthening the alloy matrix. This mechanism relies on the increase in the resis-
tance to dislocation movement, determined by the lattice distortions and changes in
the shear modulus.
The effects of changes in lattice parameter of nickel, introduced by alloying
elements, on the yield strength are more pronounced when there is a larger separa-
tion between the groups of the periodic table between nickel and the solute.
With reference to Table 6.19, titanium introduces a lattice expansion equal to
0.095% and a corresponding hardening of 22.7 MPa per at%. Molybdenum,
although exhibiting a slightly larger change in the lattice parameter (0.116% per at
%), has a strengthening factor of 14.2 MPa per at% only. This can be ascribed to the
different position with respect to the group of nickel of the two elements; 6 and 4 are
for Ti and Mo, respectively. A larger separation between elements in the periodic
table corresponds to a valence difference, which influences the stacking fault energy
(SFE) of the alloy. In particular, solute atoms with a larger valence difference with
respect to nickel lead to a greater reduction in the SFE of the solid solution. This
stabilizes the dislocations on the slip planes, inhibiting cross-slip. This condition
leads to the deformation behavior shown by Fig. 6.23, referring to a classical study
on a Ni-Co binary system (Cahn and Haasen 1996). The shear stress vs shear strain
curves for pure nickel and for the Ni-Co solid solutions, with the composition values
indicated in Fig. 6.23, exhibit an initial easy glide regime, whose maximum strain
can reach larger values when cobalt concentration is higher. The larger deformation
that can be reached for higher Co concentrations is in agreement with the possibility
of stabilizing a higher density of dislocations, since they are more stable in case of a
solute-richer alloy. The same is true for strain hardening, which is enhanced by the
raise in the dislocation density and consequent entanglement. Eventually, when the
324 6 Superalloys

Fig. 6.23 Deformation behavior of pure nickel and Ni-Co solid solutions. Compositions are
indicated in the plot. The extent of the three deformation regimes (easy glide, strain hardening
and dynamic recovery) is influenced by the SFE values determined by the solute, i.e., cobalt,
concentration. (Redrawn from Cahn and Haasen 1996)

dislocation density becomes unsustainably high, an abrupt softening of the alloy

occurs. This is due to its dynamic recovery, resulting from the recombination and
annihilation of dislocations.
Low SFE and high elastic modulus are important selection parameters also as
concerns creep resistance, a fundamental property of nickel-based superalloys,
particularly in view of their applications in gas turbines. Considering that creep
deformation mechanisms and relevant deformation rates are very much dependent
on diffusion, an excellent selection parameter for solute elements that may benefi-
cially affect the creep resistance of the alloy is the melting temperature (see
Fig. 6.24). In fact, the higher is the melting temperature of the pure element, the
lower is its solid-state diffusivity in nickel and, thereby, the relevant creep deforma-
tion rate. Therefore, refractory elements, like Mo, Rh, Ta, and W, are good candi-
dates to improve the creep resistance of the γ matrix of the superalloy and, to some
extent, also to prevent other diffusion-assisted phenomena, like high-temperature
oxidation and corrosion (see Sect. 8.2) and alloy recrystallization.
However, upper limits do exist for the concentration of refractory elements. If
present to an excessively high extent, these elements may result in the formation of
the so-called topologically close-packed (TCP) phases (see also Appendix 6). The
denomination comes from the fact that the space filling principle is the optimization
of the atomic packing factor with coordination polyhedral (the Frank-Kasper
6.4 Nickel-Based Superalloys 325

Fig. 6.24 Diffusivity vs

melting temperature for
solute species in nickel.
(Redrawn from Jena and
Chaturvedi 1984)

Fig. 6.25 Frank- Kasper coordination polyhedra typical of TCP phases. (Redrawn from Seiser
et al. 2011)

polyhedra, see Fig. 6.25), allowing for coordination numbers, i.e., number of nearest
neighbors atoms, equal to 12, 13, 14, 15, and 16 and resulting in very tight packing
(Barrett and Massalski 1980). Concerning the standard geometrically packed phases
like body- and face-centered cubic (bcc and fcc), hexagonal close packed (hcp), etc.,
they feature different coordination polyhedra that can be thought of as composed of
suitable combinations of tetrahedral and octahedral voids. This is shown in
326 6 Superalloys

Table 6.20 Types and relevant number of coordination polyhedral of the main TCP structures
together with geometrically packed ones. <Z> is the average coordination number for each one of
the listed structures. f12 is the fraction of the 12-fold polyhedra featuring each structure, including
bcc, fcc, and hcp (Seiser et al. 2011)
Coordination polyhedra
Structure Atoms in basis <Z> f12 Z12 Z13 Z14 Z15 Z16
fcc/hcp 1 12.00 1.00 (1) 0 0 0 0
C14 12 13.33 0.67 8 0 0 0 4
C15 24 13.33 0.67 16 0 0 0 8
C36 24 13.33 0.67 16 0 0 0 8
M 52 13.36 0.54 28 0 8 8 8
μ 13 13.39 0.54 7 0 2 2 2
R 53 13.39 0.51 27 0 12 6 8
δ 56 13.43 0.43 24 0 20 8 4
P 56 13.43 0.43 24 0 20 8 4
χ 58 13.10 0.41 24 24 0 0 10
σ 30 13.47 0.33 10 0 16 4 0
A15 8 13.50 0.25 2 0 6 0 0
bcc 1 14.00 0.00 0 0 (1) 0 0

Table 6.20, in which the most common TCP phases are listed together with the
number and types of relevant Frank-Kasper polyhedra, ranked on the basis of the
fraction content (f12) of the 12-fold coordination polyhedra. The situation for bcc,
fcc, and hcp structures is also displayed for a comparison. Since they are not made up
of Frank-Kasper polyhedra, the relevant structures in the table are indicated among
brackets. TCP phases can reach comparatively high atomic packing and rather
complex geometries, as can be seen in Fig. 6.26 for the R, μ and χ (or C36 Laves)
phase. Another important feature of these structures is the lack of active slip systems,
which justifies their high intrinsic brittleness that may degrade the overall fracture
toughness of the superalloy. Fracture nucleates at the stress concentration points
introduced by the hard TCP grains and may propagate more easily through the brittle
intermetallic grains (intersigmatic fracture). However, this is not the only effect these
phases are able to produce. Their compositions, particularly rich in refractory ele-
ments, definitely richer than the alloy matrix, can be achieved thanks to a diffusive
flow, occurring during the alloy exposure at high-temperature conditions, not only
during service but also during the pre-service thermal treatments, or welding oper-
ations, particularly if the process conditions are not fully under control.
A consequence of these heat-induced transformations is the formation in the alloy
matrix, nearby the TCP phase grains, of regions in which the depletion of the
refractory elements has significantly reduced solid solution strengthening. This
sort of constitutional softening, influencing both mechanical strength and creep
resistance of the superalloys, is another important effect of the formation of TCP
phases.
6.4 Nickel-Based Superalloys 327

Fig. 6.26 The complex structures of the TCP phases are obtained from the combination of the
coordination polyhedra (Fig. 6.25, Table 6.20), as exemplified in the figure by the R-phase, μ-phase,
and C36-Laves phase. (Redrawn from Barrett and Massalski 1980)

Therefore, all through the evolution of novel compositions of superalloys, par-

ticular attention has been devoted to the design of the composition with a reduced
tendency to form TCP phases. A major help came from alloy design and phase
prediction programs implemented within PHACOMP – involving phase computa-
tion programs. These programs evaluate the average number of holes in the d-bands
to define the stability range of the TCP phases, considering their formation in the
austenitic γ matrix of the superalloy (Seiser et al. 2011). The improvement
implemented in the new PHACOMP was the introduction of the average energy
level of the d-orbital, obtained from electronic structure calculations. In this way,
both atomic size and electronegativity differences are accounted for in the definition
of the stability range of the TCP intermetallic phases, including a better temperature
sensitivity. Thanks to the results of these calculations, notwithstanding the simpli-
fication of the approach if compared to the complexity of the superalloy composi-
tions, reliable criteria for new alloy design have been identified. The starting point,
that to some extent provided the first important guideline, was based on the obser-
vations that the detrimental σ phase in Co-Cr and Cr-Fe alloy forms at a fixed value
of the valence electron concentration. Therefore, this can be assumed as a cut-off
parameter for the newly developed alloys. According to this criterion, the main TCP
phases that can form in superalloys have been predicted and identified, and the main
ones are listed in Table 6.21.
328 6 Superalloys

Table 6.21 Main TCP phases that may form in superalloys, with a selection of stoichiometric
compositions and relevant crystallographic parameters (Krakow et al. 2017; Kuo et al. 1985; Rae
and Reed 2001; Rae et al. 2000)
Phase Structure and space group Lattice parameters [nm] Possible stoichiometry
A15  221
Cubic, Pm3m, a ¼ 0.456 Cr3Si
C14 Laves Hexagonal, P63/mmc, 194 a ¼ 0.515 MgZn2
c ¼ 0.848
C15 Laves  227
Cubic, Fd3m, a ¼ 0.708 MgCu2
C36 Laves Hexagonal, P63/mmc, 194 a ¼ 0.527 MgNi2
c ¼ 1.33
M Orthorhombic, Pnma, 62 a ¼ 0.930 Nb10Ni9Al3,
b ¼ 1.62 Nb48Ni38Al13
c ¼ 0.493
μ  166
Rhombohedral, R3m, a ¼ 0.476 Mo6Co7,
c ¼ 2.56 Mo6Fe7
P Orthorhombic, Pnma, 62 a ¼ 1.70 Mo42Cr18Ni40
b ¼ 0.475
c ¼ 0.907 nm
R  148
Rhombohedral, R3, a ¼ b ¼ 1.09 Mo31Cr18Co51
c ¼ 1.93
σ Tetragonal, P42/mnm, 136 a ¼ b ¼ 0.880 Cr46Fe54
c ¼ 0.456

The limits of the PHACOM approaches became more evident when dealing with
new generations of superalloys, like those containing particularly high concentra-
tions of rhenium. To overcome these limitations, the latest approaches are based on
the development of phase stability maps, including two main coordinates: valence
electron concentration and size factor difference (Seiser et al. 2011).
Several mechanisms may lead to the formation of the TCP phases. One is the
transformation of carbides, a phenomenon favored by the similarities of their
crystallographic structures. TCP phases originate from the degeneration of carbides,
during which carbon is replaced by refractory elements. TCP phases have also been
observed in high-temperature creep studies of single-crystal superalloys for the local
compositional changes determined by the diffusive processes occurring under these
loading regimes (Reed et al. 2007).
The formation of the TCP phases is not only due to a prolonged exposure to high
operational temperatures but also to a too slow cooling from high-temperature
stages, i.e., solutioning, of microstructural refinement heat treatments.
This risk was particularly important for early generation superalloys, in which the
compositions were not yet sufficiently optimized in this respect. Figure 6.27 displays
this aspect for a general-purpose nickel-based superalloy in conventionally cast and
wrought conditions. The inhomogeneity, that is typical of castings, makes these
products rather prone to the formation of the TCP phases, unless a careful control of
the heat treatment conditions is implemented.
6.4 Nickel-Based Superalloys 329

Fig. 6.27 A classical study showing TTT curves for a cast (upper) and wrought (lower) nickel-
based superalloys, with the indication of the formation fields for the TCP σ phase. (Redrawn from
Sims et al. 1987)

Another important selection rule for the components of the γ matrix of a nickel-
based superalloy concerns the improvement in both outer and inner surface stability.
Concerning the outer surface, specific elements tend to promote the formation of a
native protective oxide layer, usually several tens of micrometer in thickness, whose
main task is to avoid, or limit, further oxidation. Important elements for outer surface
stability are Cr, Al, Ce, Cs, Mo, La, Mo, Sc, Ta, W, and Y (see also Sect. 6.1 and
Appendix 5).
The first, and maybe the most important element, is chromium, which contributes
to the formation of a surface protective layer including Cr2O3 as fundamental
constituent. Its stability is generally guaranteed up to 950
C, the actual temperature
depending on the alloy and, thereby, on the native oxide layer composition. Above
950
C, chromium oxide tends to become unstable and to evolve in a gaseous phase,
according to:

3 ∘
Cr2 O3 ðsÞ þ O2 ðgÞ ! 2CrO3 ðgÞ at T > 950 C ð6:13Þ
2

Aluminum forms a number of oxide phases, the most interesting in this context
being the α-Al2O3 phase that, unlike Cr2O3, is stable also at temperatures in excess
of 950
C. The presence of cobalt improves the resistance to hot corrosion under salt
330 6 Superalloys

(Na2SO4) deposit, due to a lower diffusivity of sulfur in cobalt than in nickel, with a
two order of magnitude difference. Moreover, the melting temperature of an eutectic
mixture, like Co4S3-Co, is higher than that of Ni3S2-Ni: 877
C vs 645
C, and thus
the corrosive attack is sustained by solid, rather than liquid, phase diffusion (see
Sect. 8.2.5).
Refractory elements, like molybdenum, tantalum, tungsten, and rhenium, reduce
both inward and outward diffusion rates, for oxygen and metallic alloying elements,
respectively. Moreover, molybdenum and tungsten improve the adhesion of the
surface oxide layer to the substrate, through the formation of intermetallic precipi-
tates at the interface that produce a sort of “pegging” effect (see Sect. 7.3.3). A
similar effect can be achieved with the addition of rare earths (La, Ce) and reactive
elements (Cs, Sc, Y) (see Sect. 6.1). On the other hand, these elements may react
with sulfur to form eutectics with low-melting temperatures and in this regard may
reduce the alloy resistance to hot corrosion in salt deposit (see Sect. 8.2.5).
The main category of inner surfaces is the grain boundaries of the alloy matrix
and of secondary phases, like carbides. Actually, some carbides are specifically
localized along the alloy GBs and usually would stabilize them against sliding and
migration under elevated stresses at high temperatures. Some elements, possibly
impurities, are not soluble in nickel and therefore segregate along grain boundaries
with detrimental or beneficial effects, depending in the first place on the element but
also on its concentration. Elements that affect the stability of the inner surfaces are S,
P, Bi, Pb, Te, and Th. These elements are classified as embrittling impurities and can
reduce grain boundary strength, even if present in very little concentration. Manga-
nese is able to compensate the S-induced embrittlement through the formation of
MnS, which segregates at the grain boundary regions, thus sequestrating the detri-
mental element. Boron has a positive effect on grain boundary stability, since it
increases the grain boundary cohesion, enhances the ductility, and improves oxida-
tion resistance, being able to reduce the diffusional phenomena at the grain bound-
ary. Its concentration is usually kept below 1 wt%, since higher values may result in
the formation of coarse borides along the alloy grain boundaries that, in the long run,
may reduce the grain boundary cohesion. Also hafnium and zirconium reduce
diffusional phenomena at the grain boundaries, including the segregation of alloying
elements. Table 6.22 recalls the main alloy elements for the γ phase and relevant role
in a Ni-based superalloy.

Table 6.22 Alloying elements and relevant roles in Ni-based superalloys (Geddes et al. 2010)
Role Element required
Carbides formation Cr, Hf, Mo, Nb, Ta, Ti, W
Grain boundary refinement and stregthening B, C, Hf, Zr
High-temperature corrosion resistance La, Th
Oxidation resistance Al, Ce, Cr, La, Y
Rupture strength improvement B
Sulfidation resistance improvement Co, Cr, Si
Workability improvement B, C, Fe
6.4 Nickel-Based Superalloys 331

6.4.3 Selection Criteria for Alloying Elements in Nickel-

Based Superalloys: The γ 0 Precipitates

The main strengthening mechanism active in nickel-based superalloys is precipita-

tion hardening, based on the formation of γ0 precipitates. They have an ordered L12
structure (see Sects. 6.3.1 and 6.4.1.1 and Appendix 6), based on the fcc lattice.
Although in real alloys γ0 precipitates have rather complex compositions, the binary
Ni3Al intermetallic compound can be taken as a benchmark. The choice of γ0
precipitates has essentially been determined by their thermal stability, the possibility
of forming coherent interfaces, and the existence of common slip systems with the
γ-matrix. These selection criteria have restricted the set of possible candidate phases.
Figure 6.28 shows the unit cells of the austenitic fcc γ phase and the fcc cell with the
ordered atomic occupation pattern for the L12 and D022 structures, corresponding to
the γ0 and γ00 precipitate phases, respectively (see also Fig. 6.13). The tetragonal D022
structure, introduced dealing with iron-based superalloys (see Sect. 6.3), can actually
be regarded as consisting of two stacked fcc cells, with a clear atomic occupation
similarity with the other ordered γ0 phase. In fact, assuming a binary composition for
γ00 , e.g., Ni3Nb, the only difference with respect to γ0 is that the ordered atomic
sequence exhibits, in the site at the center of the base and central crystallographic
planes, the alternate occupation of Ni and Nb atoms, respectively. The structural
similarity can be straightforwardly extended to slip planes too, as evidenced by
Fig. 6.28.
The small coherency strains and relevant low surface energies of the interfaces
among the ordered precipitates and the γ matrix are important prerequisites for the

Fig. 6.28 Structural correspondences among the unit cells of the (a) fcc γ; (b) L12 γ0 , and (c) D022
γ00 phases. (Redrawn from Jena and Chaturvedi 1984)
332 6 Superalloys

Fig. 6.29 Scheme illustrating the interaction among a dislocation line, length LI, and ordered γ0
precipitates of average radii r0

alloy stability against recrystallization. The uniform spatial distribution and an

optimized cuboidal shape of the precipitates are additional factors, which can be
improved through suitable thermal treatments (see Sect. 6.6).
For the exploitation of the anomalous dependence of the yield strength of γ0
precipitates on temperature (Sect. 6.4.1.1), it is important that the average size and
reciprocal separation of the precipitates are sufficiently small, so that the prevailing
interaction mechanism among them and the incoming dislocations is cutting. Dislo-
cations generated in the γ matrix, being just partials for the ordered structure of γ0
precipitates (see Fig. 6.20), when cutting through them require an additional energy
term, needed to create the antiphase boundaries (vide infra). Therefore, either they
are blocked at the matrix-precipitate interface or they are hindered in their movement
through the precipitates.
To quantify this contribution, a simple model can be considered (Fig. 6.29).
Ordered γ0 precipitates are cut through by a dislocation line of length L1, equal to
the average distance between pairs of these precipitates, assumed to be of spherical
shape, for the ease of computation, and with an average size of r0.
The force per unit length, t, of a dislocation line with a Burgers vector b, moving
under the effect of a shear stress, τ, is given by:

t ¼ τb ð6:14Þ

As noted above, the dislocation that forms in the alloy matrix is an incomplete
dislocation (partial) for ordered γ0 precipitates. This implies that when the
6.4 Nickel-Based Superalloys 333

dislocation moves across an ordered γ0 precipitate, it creates an antiphase boundary

(APB) with a specific surface energy (ΓAPB). By equating the work required to move
the dislocation line from one to the other end of the precipitates (see Fig. 6.29), to the
energy term required for the creation of the APB across the precipitates, it is possible
to obtain:

πr 20 ΓAPB ¼ L1 τb2r 0 ð6:15Þ

The Eq. 6.16 is obtained for precipitation strengthening associated with the
cutting interaction mechanism:

πr 0 ΓAPB
τCUT ¼ ð6:16Þ
2bL1

The expression for τCUT shows clearly the importance of the APB energy term
(ΓAPB) that is being actually adopted as a guideline for the design of novel alloys,
including γ0 precipitates with an optimized composition, increasing their strength-
ening efficiency. As shown in Table 6.23, the real compositions of the γ0 precipitates
turn out to be rather complex, definitely more complex than those of the just binary,
Ni3Al, or, at most, ternary Ni3(Al, Ti) intermetallics, usually taken as reference com-
positions for γ0 .
To infer how different alloying elements tend to replace Ni and Al in the relevant
atomic positions, it is possible to refer to the pseudo-ternary phase diagram in
Fig. 6.30. Cobalt and copper are preferential substitutes of nickel; molybdenum,
chromium and iron replace indifferently both nickel and aluminum; the other
elements in the diagram, replace aluminum.
Still referring to the dislocation-precipitate model in Fig. 6.29, a simple calcula-
tion provides the contribution of dislocation cutting to the superalloy strengthening.
Considering γ0 precipitates with an average size r0 ¼ 250 nm and an average distance
L1 ¼ 25 nm, being b ¼ 0.25 nm and ΓAPB ¼ 0.1 J/m2, τCUT can be calculated using
Eq. 6.16:

πr 0 ΓAPB π ∙ 2:5 ∙ 107 ∙ 0:1

τCUT ¼ 0 ¼ ¼ 6000 MPa ð6:17Þ
2bL1 2 ∙ 0:25 ∙ 2:5 ∙ 108

A competitive interaction mechanism to cutting is bowing, once the dislocation

line is pinned by the precipitates, as illustrated in Fig. 6.31. This situation can be
faced either in early superalloys, with a comparatively low fraction of γ0 phase, or in
case of overaged microstructures in which the ripening of the precipitates, for
instance, after long-term exposure to the operating conditions, determines an
increase in their average distance. The strengthening by the Orowan (bowing)
mechanism is on average lower than cutting. It is therefore interesting to evaluate
the conditions corresponding to the transition from one to the other dislocation
interaction mechanism.
 334

Table 6.23 Average compositions (at%) of selected nickel-based superalloys and actual compositions of the relevant γ0 precipitates (Sims et al. 1987)
Chemical composition [at%]
Alloy Component analyzed Ni Al Ti Cr Co Mo W Nb V C
Alloy 713C Alloy 68.77 12.03 0.92 13.96 – 2.58 – 1.24 0.50
γ0 73.36 17.95 1.26 3.73 – 1.21 – 2.49 –
MAR-M200 Alloy 59.60 10.97 2.47 10.24 10.95 – 4.02 0.64 0.74
γ0 64.19 18.01 3.42 4.02 6.71 – 3.16 0.49 –
Udimet 500 Alloy 49.62 6.22 2.33 20.84 18.11 2.50 – – 0.37
γ0 50.00 – 4.00 1.60 4.00 1.30 – – –
Udimet 520 Alloy 55.35 4.45 3.58 20.87 11.63 3.57 0.31 – 0.24
GB γ0 73.89 – 9.49 4.31 7.22 1.80 0.31 – –
6 Superalloys
6.4 Nickel-Based Superalloys 335

Fig. 6.30 Pseudo-ternary phase diagram, Ni-Al-X, indicating the tendency of ternary alloying
elements to preferentially replace Ni, Al, or both atomic in the binary Ni3Al reference compound.
(Redrawn from Sims et al. 1987)

Fig. 6.31 Scheme illustrating the bowing, or Orowan, interaction mechanism of a dislocation line
with a couple of precipitates. (Redrawn from Hull and Bacon 2011)
336 6 Superalloys

Fig. 6.32 Hardness of a Ni-

Cr-Al superalloys as a
function of the average
particle size of the
strengthening γ0
precipitates. (Redrawn from
ASM International 2000)

The strengthening due to bowing (τBOW) of dislocations can be expressed as

(Hull and Bacon 2011):

2Gγ b
τBOW ¼ ð6:18Þ
L1

where:
Gγis the shear modulus of the γ matrix.
b is the Burgers vector.
L1 dislocation line length (equal to the distance between precipitates).
The transition from cutting to bowing of dislocations will occur when the
condition: τCUT ¼ τBOW is satisfied. From Eqs. 6.18 and 6.16, a critical size of the
γ0 precipitates can be evaluated:

2Gb
r 0 ¼ ð6:19Þ
πΓAPB

Above this size, dislocations, once blocked at the matrix-precipitate interface,

start bending and remain within the γ matrix. The situation is illustrated by the plot
in Fig. 6.32 of the hardness as a function of the average size of γ0 precipitates of a
Ni-Cr-Al Ni-based superalloy (ASM 2000). The trend reminds the age-hardening
curves discussed for heat-treatable aluminum alloys in Sect. 3.2.3.3 (Fig. 3.17),
with a similar evolution in the dislocation-precipitate interaction mechanisms. The
only important difference is that in case of aluminum alloys, metastable and
non-equilibrium structures are involved, whereas in case of nickel-based superal-
loys, the equilibrium γ0 phase is present.
6.4 Nickel-Based Superalloys 337

The phase and microstructural stabilities are paramount in view of the high-stress
field and temperatures structural components of a gas turbine engine have to stand.
Adequate stability conditions can be achieved through a suitable selection of the
alloy composition by refining the microstructure and tuning the lattice misfit, δ (see
Sect. 6.4.1.1), a crucial parameter for the long-term in-service behavior of nickel-
based superalloys. If γ0 particles are small in size and with low lattice misfit, they
tend to have a spherical shape, i.e., an isotropic geometry not affected by any
directionality in the elastic strain field at the interface with the alloy matrix. A
cubic morphology is generally regarded as the more desirable for γ0 precipitates in
Ni superalloys. In fact, this geometry exhibits an intrinsic stability, resulting from a
combination of low-energy interfaces, thanks to a low lattice misfit, and for being
substantially planar. This aspect and relevant implications will be discussed further
on, since they affect also the response of superalloys to the thermal treatments carried
out for their microstructural refinement (see Sect. 6.6).
As concerns the interface geometry of the precipitates, it can be noted soon, with
reference Fig. 6.33, that the recrystallization and precipitate coarsening propagate
starting from the corners of the cubes, the regions with the higher curvature. Since
crystal lattice misfit plays an important role in determining the precipitate stability

Fig. 6.33 Shapes that the ordered γ0 precipitates can assume during microstructural refinement
thermal treatments or due to the exposure to real working conditions. (Redrawn from Ricks et al.
1983; Reed 2006)
338 6 Superalloys

and, thereby, the performances of the superalloy, research efforts have been invested
in developing effective tools and models, based on PHACOMP, Thermo-Calc,
DICTRA, and other similar modeling approaches, resulting in the optimization of
this parameter through the design of appropriate alloy compositions (Reed 2006).
For a fast prediction of the effects of different alloying elements on the lattice
parameter of the γ matrix and γ0 precipitates, a general expression of the Vegard’s
law, capable to treat multicomponent systems, has been developed. The lattice
parameters of the γ and γ0 phases would change according to the expressions:
X
aγ ¼ a
γ þ μi, γ X i, γ ð6:20Þ
i

X
aγ0 ¼ a
γ0 þ μi, γ0 X i, γ0 ð6:21Þ
i

In Eqs. 6.20 and 6.21, a
γ and a
γ 0 are the lattice parameter values of pure nickel
and of the binary Ni3Al intermetallic compound, respectively (see Fig. 6.34a). The
values obtained from Eqs. 6.20 and 6.21 depend on the molar fraction of the alloying
elements (Xi) and on the relevant Vegard’s coefficients (μi) given in Fig. 6.34b.
The total fraction of the γ0 precipitates has been, and still is, a very important
parameter in the development of novel superalloy compositions, considering the
effect this parameter has on the mechanical properties of the superalloys. Still
referring to a very simple model (Fig. 6.35), the dependence of precipitation
strengthening on the precipitate fraction is derived.
The number of precipitates (nL) that are interacting with a dislocation line of
length L is given by:

nL ¼ 4πr 20 L ∙ n ð6:22Þ

where n in the concentration of precipitate, i.e., the number of precipitates per unit
volume. From the definition of volume fraction of precipitate phase, f:

V tot:precipitate
f ¼ ð6:23Þ
V alloy

An expression for n can be obtained by dividing f by the volume of a single

precipitate. Again, for the ease of calculation, spherical precipitates with r0 radius are
assumed. Therefore, Eq. 6.22 becomes:

f 3Lf
nL ¼ 4πr 20 L ∙ 4 3
¼ ð6:24Þ
3nπr 0
r0
6.4 Nickel-Based Superalloys 339

Fig. 6.34 Coefficients to be used in the generalized Vegard’s law to evaluate the lattice parameters
of (a) the ordered L12 γ0 precipitates, having as reference composition Ni3Al, and (b) a nickel-based
fcc solid solution (the γ matrix of the superalloy). (Redrawn from Reed 2006)
340 6 Superalloys

Fig. 6.35 A dislocation line interacting with two precipitates. Schematic of the model employed to
derive the dependence of the strengthening effect on the volume fraction of precipitates

The average distance between couples of precipitates, LI, can be written as:

L r0
LI ¼ ¼ ð6:25Þ
nL 3f

Therefore, the Eq. 6.16 for the strength enhancement of the alloy due to disloca-
tions cutting through ordered precipitates can be expressed as a function of the
volume fraction of precipitate phase:

πr 0 ΓAPB 3πf ΓAPB

τCUT ¼ ¼ ð6:26Þ
2bLI 2b

Equation 6.26 confirms the expected increasing trend of the alloy strengthening
with the volume fraction of ordered precipitates. A practical evidence of this result is
represented in Fig. 6.36.
In agreement with these results, novel superalloys, particularly those for single-
crystal turbine high-pressure blades (see Sect. 6.5), feature an increasing fraction of
γ0 phase and a corresponding improvement in the mechanical properties, including
creep resistance (Fig. 6.37). However, the volume fraction of γ0 phase has leveled off
at 70%, in state-of-the-art materials, since higher values turned out to induce worse
structural properties. This is an interesting aspect, which emerged in the pioneering
paper by Murakumo et al. (2004) and was successively confirmed by other investi-
gations (Murakumo et al. 2004).
Single-crystal superalloys guarantee the best creep performances, as far as
vacancy diffusion creep is the dominating mechanism (see Sect. 6.5). However,
for sufficiently high applied stress and elevated temperatures, diffusion-assisted
6.4 Nickel-Based Superalloys 341

Fig. 6.36 Improvement of yield stress with increasing volume fraction of γ0 and γ00 phases in state-
of-the-art nickel-containing superalloys at 650
C. (Redrawn from Reed 2006)

dislocation motion may provide an important contribution to creep deformation. In

this situation, the best way to improve the creep performances of the alloy, without
excessively reducing other important structural properties, like fracture toughness
and fatigue life, is to restrict creep to the so-called γ channel. In the γ austenitic
matrix, which is a minority phase in the latest-generation superalloys and it is
localized between the ordered γ0 precipitates, dislocations may form under the effect
of the applied load. The results of creep tests and experiments demonstrate the
fundamental role played by the γ/γ0 interfaces in preventing dislocations to enter
the γ0 precipitates and confirm that an optimized microstructure still requires a minor
fraction of γ matrix in which creep is confined and limited by the surrounding γ0
precipitates.
To complete the picture of the microstructural aspects of nickel-based superal-
loys, a few notes on the other secondary phases that can be found in the alloy γ
matrix in association with the main reinforcing phase, i.e., γ0 , like carbides and
borides with different stoichiometries (see also Appendix 6). The main carbides are
the same as those of cobalt-based superalloys (see Sect. 6.2.1). They may hinder
dislocation movement, directly, through an Orowan-like interaction, and indirectly,
through the lattice strain they introduce into the alloy. Other important roles that the
carbides share with borides concern the optimization of grain boundaries, which are
stabilized by these phases against sliding and grain recrystallization. For these
reasons, the optimization of carbon and boron concentrations, together with the
concentrations of carbide- and boride-forming elements, has been investigated with
particular regard for polycrystalline alloys, usually obtained using casting technol-
ogies. As concerns borides, the main stoichiometry that is encountered in Ni
342 6 Superalloys

Fig. 6.37 Dependence of the creep life of TMS-75 single-crystal Ni-based superalloy and Inconel
713 C polycrystalline alloy on different volume fractions of γ0 phase. (Redrawn from Murakumo
et al. 2004)

superalloys is the M3B2, being M one or a combination of the following main metals:
Cr, Hf, Mo, Ta, and Ti. The superalloys obtained by powder metallurgy (PM) routes
are another class of products for which grain boundary engineering is an issue. As
concerns aerospace gas turbines engines, turbine discs have recently attracted a
remarkable technological interest of PM processing. The control of grain boundary
structure has been improved with the development of specifically designed produc-
tion steps. Differential thermal treatments have been applied to optimize the micro-
structure of different regions of the same critical components, like turbine discs. A
suitable gradient of grain size, depending on the specific local requirements and part
specifications, can be induced thanks to this approach (Reed 2006). The dual
microstructure heat treatment (DMHT) technology, developed and patented by the
NASA (see Sect. 6.6), has been applied to the third-generation PM superalloys, such
as LSHR and ME3 (see Table 6.13) to induce different grain size and morphology in
the bore and the rim of the same turbine disc (Reed 2006, see also Sects. 6.6 and 6.7).
6.5 Single-Crystal and Directionally Solidified Superalloys 343

6.5 Single-Crystal and Directionally Solidified Superalloys

The continuous progress in the aircraft GTEs has also been driven by the will to
increase their efficiency and, consequently, to reduce fuel consumption, with obvi-
ous advantages not only from the economical but also the environmental point of
view. Raising the operating temperatures is an effective strategy for achieving this
goal (see Fig. 2.6). Over the years, the TET (see Sect. 2.2 and Fig. 2.20) has greatly
increased, and this improvement can be ascribed to two main factors: more efficient
design of the cooling system of turbine blades and vanes and improved superalloys
chemistry and relevant casting processes. In this respect, high-pressure turbine
blades, discs, and vanes are critical components because, in addition to the high
temperature, they are subjected also to a combination of corrosive environment and
mechanical stresses. In the early days of the GTEs history, the use of wrought alloys
was predominant over castings, which where tolerated when strictly necessary only
(castings had a bad reputation due to some failures associated with casting defects).
However, higher-strength requirements continued to increase over the years, and, for
the fabrication of some components, casting became the only choice because high-
strength alloys cannot be subjected to forging processes successfully (Giamei 2013).
Also for cast superalloys, some limiting factors emerged, related to fatigue and creep
resistance, surface stability against oxidation, and corrosion phenomena.
Columnar-grained superalloys, with crystalline grains oriented parallel to the
longitudinal axis or principal stress axis, were thought to be the ideal choice for
components subjected to the axial loads produced by high-speed rotation (Giamei
2013). The directional solidification (DS) casting process was implemented to obtain
columnar-grained, still polycrystalline, superalloy components. Further technologi-
cal developments afforded the possibility to produce single-crystal (SC) components
that still represent the state-of-the-art achievement of this process. Pratt & Whitney
Aircraft (PWA) was the pioneering company in the development of both DS and SC
superalloy products. Indeed, the history of both DS and SC alloys started at Pratt &
Whitney Advanced Materials Laboratory in the mid-1960s (ASM International
1993; Giamei 2013). CGDS (columnar-grained directionally solidified) structures
have been produced, allowing unidirectional heat flow during the solidification
process of the alloy. The heat of solidification is extracted along one single direction,
which is parallel to the main axis of the component. Although the application of the
DS process to already-existing polycrystalline alloys, e.g., the MAR-M200 alloy, led
to important improvements in their performances, yet only with the design of
specific alloy compositions, the improved microstructure could be better exploited
(Donachie and Donachie 2002). The same SC casting process is, in some respect, the
natural evolution of the DS technology, and the optimization of the alloy composi-
tion became even more critical. The research and development at PWA proved that
the removal of grain boundary strengthening elements, such as B, C, Hf, and Zr,
resulted in the increase in the incipient melting temperature of the alloy (see
Table 6.16 for the chemical composition of SC superalloys). Actually, a minor
amount of carbon may still have beneficial effects, since it helps reducing sulfur
344 6 Superalloys

content, and this is the reason why a few SC alloy formulations contain limited
amounts of this element (Table 6.16). Tantalum turned out to be a strongly positive
addition (see PWA 1480 alloy – Table 6.16), since it induces strengthening and
oxidation resistance and improves also the alloy castability. Rhenium additions are
responsible for reducing the alloy diffusivity, with beneficial effects on a slower
precipitate coarsening and improved creep resistance (see PWA 1484 alloy –
Table 6.16).
Notwithstanding the excellent performances achieved by DS and SC components
(see Sect. 6.5.1), still the directional solidification technology introduced specific
processing issues:
• Higher tendency to inclusion formation.
• Separately nucleated grains.
• Grain misorientation.
• Tendency of freckle grains formation (nucleation of grains as a consequence of
inverse segregation caused by dendrite erosion).
They were addressed not only with the alloy design optimization but also with the
evolution of the production apparatuses and solidification procedures (see Sect. 6.5.2).
Turbine blades made of a DS superalloy were first tested in the small PT6 engine,
typically used for helicopters. Since the components did not fulfill the operational
requirements, the test was suspended. Afterward, a replacement set of SC turbine
blades was subjected to the same trial, fully successful this time. Progress in DS and
SC superalloys brought very important positive consequences in terms of achievable
engine performances, both as concerns the thrust increase and the improvement in
the time between overhaul (TBO). At the beginning of the 1980s, SC alloys started
to be employed in the engines of large commercial airliners, as well as in military
aircrafts (Giamei 2013). The first reported commercial application of single-crystal
turbine high-pressure blades dates back to 1981, on an Airbus aircraft, and these
components are still to be regarded as the only structural parts featuring a single-
crystalline microstructure (ASM International 1993). There are two generations of
DS alloys and six of the SC ones (see Tables 6.15 and 6.16). The two generations of
DS Ni-based superalloys differ for the presence of Re in their composition. In
particular, the first-generation alloys do not contain this element, which is present
instead, in concentrations of about 3 wt%, in the second generation of DS alloys.
Furthermore, the first-generation alloys derived essentially from wrought products,
following early attempts, made in the 1950s, to produce them by vacuum melting
and casting processes. These alloys were developed with the intention of improving
strength through the elimination of the deleterious Laves and σ phases, as well as by
increasing the γ0 volume fraction. In these alloys, Mo and W additions were meant to
improve high-temperature capabilities, while Hf in concentrations around 2 wt%
was used for increasing rupture stress ductility.
As concerns SC superalloys, the first generation has the same features as those of
the first DS superalloy generation, while the second generation of SC superalloys was
characterized by better oxidation resistance coupled with higher-temperature capa-
bilities, about 30
C more than the average peak temperatures of the first-generation
alloys. This result was possible thanks to the addition of Re (3 wt% approx.) that
6.5 Single-Crystal and Directionally Solidified Superalloys 345

contributed to the general increase of the refractory elements content. Moreover, the
γ0 volume fraction exceeded 60%. The third generation of SC superalloys was
characterized by lower Cr concentrations and even higher Re content for improving
both environmental resistance and creep strength. The fourth generation was essen-
tially characterized by the addition of platinum-group metals, Ru in particular. A
further increase in the refractory elements content aimed at improving the alloy
strength. High-temperature low-stress resistance was also improved thanks to a
higher negative γ/γ0 misfit. The fifth generation of SC superalloys was characterized
by even higher Ru concentrations, higher lattice misfit, and additional strengthening
obtained through additions of Mo, Re, and Ru itself. Ru also improved the phase
stability thanks to a reduction in the alloy diffusivity, having an indirect beneficial
effect on the superior creep resistance than the alloys of the four previous generations.
Finally, sixth-generation superalloys show structural properties that are in some
respect comparable to those of fifth-generation SC superalloys but with a better
oxidation and hot corrosion resistance (Prasad and Wanhill 2017). Still concerning
gas turbine components, starting from high-pressure blading, there is a growing trend
to apply the SC casting technology to a wider range of aero-engine components,
which undergo complex loading conditions involving a combination of multiaxial
and centrifugal stresses. At the same time, SC and DS blading has been considered for
other than aerospace field: industrial natural gas-fueled and large land-based gas
turbine engines used for electricity production (ASM International 1993).

6.5.1 Properties of DS and SC Superalloys

As shown in Fig. 6.38, the main benefits of SC and DS over polycrystalline

superalloys derive from a combination of improved creep behavior, excellent intrin-
sic corrosion and thermal fatigue resistance.
The improved corrosion resistance displayed by these materials depends on
several factors. The main ones are the absence, or the reduction, of grain boundaries,
a controlled and homogeneous composition, both giving the alloy an intrinsic
stability against inward and outward diffusion of chemical species. Thermal fatigue
life benefits from the compliant <001> direction, along which the components are
typically oriented, thanks to the directional solidification processing. In this way, as
shown by Fig. 6.39, the thermal stress, σTH, associated with the temperature gradient
along the hot spot regions, remains below critical values. The search for better creep
performances has been the main driving force for developing the DS and SC
technology, considering in particular the central role of time-dependent deformation
in limiting the useful lifetime of the hot parts of gas turbine engines. Owing to the
operational conditions, the creep is mainly diffusive in character, involving vacancy
and, in general, point defect diffusion.
These aspects have been treated in a comprehensive approach by Ashby (1972)
and Frost and Ashby (1982), in their pioneering works, who grouped together in
stress-temperature maps, the main deformation modes in polycrystalline materials.
In these maps the stress (tensile or shear), normalized by a suitable elastic modulus,
346 6 Superalloys

Fig. 6.38 Relative lifetime

for polycrystalline,
directionally solidified
(DS) and single-crystal
(SC) components, with
regard to creep, thermal
fatigue, and corrosion
resistance. (Redrawn from
Gell et al. 1987)

Fig. 6.39 Schematic of the

temperature distribution
along the axis of a gas
turbine high-pressure blade.
The expression for the
thermal stress, σTH,
indicates that the lower is
the elastic modulus, E, the
lower is the stress in the
alloy resulting from a
temperature gradient. α is
the coefficient of thermal
expansion (CTE) of the
alloy

is plotted against the homologous temperature (T/Tm). According to this approach,

there are six main different deformation mechanisms that have been acknowledged
as essential to describe the behavior of real materials:
• Defectless flow.
• Dislocation glide.
• Diffusion creep along grain boundaries (Coble creep).
• Diffusion creep within grain boundaries (Nabarro-Herring creep).
• Dislocation creep.
• Twinning.
6.5 Single-Crystal and Directionally Solidified Superalloys 347

Fig. 6.40 Ashby deformation-mechanism maps: (a) general scheme; (b) deformation map for iron;
(c) deformation map for tungsten. (Redrawn from Ashby 1972)

In Fig 6.40a the deformation map including the above mechanisms, but twinning,
is shown. Twinning in general accounts for minor deformations, although it may be
very important when dealing with some classes of materials, like shape memory
alloys (see Sect. 9.5). In these particular alloys, twinning accounts for large strains
(up to 8%), induced by applying comparatively low stresses, as compared to the
yield stress. In Sect. 9.3, dealing with the applications of oxide dispersion strength-
ened (ODS) alloys, more extended Ashby deformation maps are considered, includ-
ing also the effects of irradiation-induced deformations.
Apart from defectless flow, occurring in ideal conditions when the applied stress
exceeds the theoretical shear strength of the material, all the other mechanisms
involve different kinds of defects. The Ashby deformation maps are built starting
from constitutive equations that depend on the dominant deformation mechanism
and the specific microstructural features of the material. This is exemplified in the
series of maps for polycrystalline iron and tungsten (Fig. 6.40b and c), respectively,
and nickel (Fig. 6.41).
The transition borderline from one regime to the other is determined by equating
the deformation rate due to each mechanism. Other important specifications
 quoted
for each map are the minimum deformation rate (e.g., ε_ ¼ 108 s1 and the specific
microstructure, usually the average grain size (e.g., d ¼ 32 μm, see Fig. 6.40). When
348 6 Superalloys

Fig. 6.41 Ashby deformation-mechanism maps for nickel, with constant deformation rate curves.
(Ashby 1972)

the minimum deformation rate for a map is quoted, then an elastic regime region
appears. This is used to indicate conditions that are not suitable to induce plastic
deformation, at least not with the specified rate. Alternatively, a set of curves
corresponding to constant deformation rates are introduced in the map (Fig. 6.41),
so that the stress-temperature data point can be calculated under different loading
conditions.
Dislocation glide, i.e., the movement of dislocation on slip planes along specific
slip directions, may result in extensive plastic deformations in real materials, partic-
ularly metals and metallic alloys. The actual deformation depends on several factors,
including the stacking fault energy, whose effect on dislocation stability and,
thereby, plastic deformation has been considered in Sects. 6.4.1.2 and 6.4.2. At
higher temperatures, dislocation climb, a typical diffusion-assisted phenomenon,
may occur. This is the mechanism for dislocation creep. Also point defects may
provide their contribution to the high-temperature deformation of materials by
diffusing through the grains (Nabarro-Herring creep) and along the grain bound-
aries (Coble creep). These latter mechanisms will be considered next, and a simpli-
fied expression for their constitutive equations is derived.
The approach can be applied also to nonmetallic, polycrystalline materials as
shown in Fig. 6.42, in which the deformation maps for germanium, magnesium
oxide, and uranium oxide are shown (Ashby 1972).
6.5 Single-Crystal and Directionally Solidified Superalloys 349

Fig. 6.42 Ashby deformation-mechanism maps for nonmetallic materials: (a) germanium; (b)
magnesium oxide; (c) uranium oxide. (Ashby 1972)

The stress and temperature conditions faced by the high-pressure turbine blades
would typically fall in the diffusive creep regions, as displayed by the Ashby map in
Fig. 6.43, referring to the nickel-based superalloy MAR-M200., featuring a com-
paratively coarse grain size, i.e., 10 mm.
Grain coarsening results in a significant reduction of the deformation rate
resulting from diffusive creep mechanisms (Coble and Nabarro-Herring creep),
with a prevailing contribution from dislocation creep. To get a better understanding
of the important effect that the grain coarsening has on diffusive creep, let us
consider the schemes in Fig. 6.44, showing qualitatively the trajectories of the
diffusive flow of atoms and counterflow of vacancies in a polycrystalline alloy
under an applied stress and at sufficiently high temperature. The driving force for
these flows is the concentration gradient of excess vacancies, i.e., those vacancies
created by the atomic displacement from the equilibrium position induced by the
applied stress. These excess vacancies give a net contribution to the vacancy
concentration along the grain boundary regions perpendicular to the applied stress.
In the long run, this process leads to the accumulation of diffusing atoms in the
grain boundary regions normal to the applied stress. These atoms leave behind
350 6 Superalloys

Fig. 6.43 Ashby deformation maps for the nickel-based superalloy, MAR-M200. Note the migra-
tion of the constant deformation rate curves to the dislocation creep region, determined by the
comparatively coarse (10 mm) grain size. (Redrawn from Ashby and Frost 1982)

Fig. 6.44 Scheme of a polycrystalline sample with the diffusion mechanisms activated by the
applied stress: bulk and grain boundary diffusion for the Nabarro-Herring and Coble creep,
respectively. Diffusive flows of atoms and vacancies result in the accumulation of atoms in the
grain boundary regions normal to the applied stress
6.5 Single-Crystal and Directionally Solidified Superalloys 351

vacancies, piling up along the grain boundaries parallel to the applied stress. A
permanent deformation will then appear, accompanied by the reduction of the load-
bearing section of the sample due to the creation of micropores for the coalescence of
vacancies. This specific damaging mechanism may take the creeping sample to a
tertiary creep stage, resulting in its rapid failure. The simple microstructural model
depicted in Fig. 6.44 can be used to evaluate the deformation rate under diffusive
creep conditions, with particular regard to the dependence of strain rate on the
average grain size, d. Considering stationary conditions, the change of the grain
size due to Nabarro-Herring creep occurs at a rate that can be expressed in terms of
the Fick’s first law:
 
Cex
d_ ¼ Dl ð6:27Þ
d

where:
d_ is the rate of change of the grain size.
d is the average grain size.
Dl is the lattice diffusivity, i.e., the diffusivity inside the grains.
Cex is the concentration of excess vacancies, i.e., those vacancies produced by the
atomic displacement due to the applied stress (Nabarro and de Villers 1995).
From (6.27) the strain rate for the Nabarro-Herring creep (ε_ NH ) can be evaluated
as:
 
d_ Cex
ε_ NH ¼ ¼ Dl ð6:28Þ
d d2

A similar expression can be obtained for the strain rate due to the Coble creep. For
this deformation mechanism, it has to be considered that diffusion occurs along the
grain boundaries with a much larger diffusivity (DGB). On the other hand, the
effective diffusion path is now limited by the grain boundary section (dδ), whereas
the diffusion path for the Nabarro-Herring creep was the whole grain section (d2).
Therefore, the same functional relationship for the strain rate derived for the
Nabarro-Herring creep can be assumed for Coble’s, although scaled by the ratio
between the relevant effective diffusion area:

Cex dδ Cex
εC_ ¼ DGB ¼ DGB 3 δ ð6:29Þ
d2 d2 d

where:
δ is the grain boundary thickness.
DGB is the diffusivity along grain boundary.
352 6 Superalloys

Fig. 6.45 Creep curves at

750
C for the CMSX4
nickel-based superalloy, at
the constant stress values
indicated in the graph.
(Redrawn from Reed 2006)

From these two expressions for the diffusive creep strain rates and from the
exponents that the average grain size exhibits, the reason why DS and SC compo-
nents have been developed turns out to be rather clear. The lack of grain boundaries
normal to the main load axis and the extremely large grain size, as compared to
conventionally cast alloys, are very effective in reducing the diffusive creep strain.
Obviously creep still remains an issue for superalloys, in particular as concerns
materials of the latest generations, which have been developed with the deliberate
intention to improve the engine performances acting on the turbine entry temperature
(see Fig. 1.14 and Chap. 2). In this regard, it is important to consider that once the
creep strain accumulated in service by a high-pressure turbine blade goes above a
few percent, the involved tolerances are so strict that the part replacement becomes
necessary, even if no other damages are present. Moreover, real conditions, featuring
thermal gradients and corrosive and turbulent gases, influence the creep mechanisms
in not always an easily predictable way. These are aspects that suggest some
precautions, when transferring results from laboratory experiments to real compo-
nents, even in view of the complex phenomenology reported for the creep behavior
of nickel-based superalloys.
Primary creep is observed at lower temperatures (700–850
C) for sufficiently
high stresses, capable to induce the nucleation, typically occurring in the γ matrix,
and movement of dislocations. In the low-stress range, complete dislocations in the
austenitic γ phase are observed (see Sect. 6.4.1.2, Fig. 6.20). The movement of these
a/2<110> dislocations, mostly by glide, is limited to the γ channels. The strain curve
is monotonous with time, showing a trend like that in Fig. 6.45 for the 450 MPa
applied stress. In the same graph, other curves are displayed, corresponding to higher
applied stresses. The primary creep strain increases with increasing stress. Moreover,
a transition to a secondary creep regime, following an inflection in the strain versus
time curve, is observed. Transmission electron microscopy (TEM) investigations
have demonstrated that this behavior is determined by the interaction of the dislo-
cations with the ordered γ0 precipitates. Thanks to dislocation decomposition and
subsequent recombination, with other dislocations, complex dislocation structures
would form, in particular, dislocation reactions like the one given in Eq. 6.30:
6.5 Single-Crystal and Directionally Solidified Superalloys 353

Fig. 6.46 γ0 cutting by paired a/2 <112> dislocations. See main text for the detailed description of
the involved dislocation reactions. (Redrawn from Reed 2006)

a a   a   a  
½011 þ 101 ! 112 þ 112 ð6:30Þ
2 2 3 6
 

As shown in Fig. 6.46a, the leading dislocation a=3 112 can cut through the γ0
precipitate if the shear component of the applied stress is sufficiently high. In doing
this, it leaves behind a superlattice intrinsic stacking fault (SISF).
When a second similar decomposition reaction occurs, new dislocation compo-
nents may interact with those that entered already the ordered precipitates. The
formation of a super-dislocation may now occur, a line defect capable to move
inside the precipitate conservatively (Fig. 6.46b). In the process another planar
defect is formed, the counterpart of the SISF, i.e., a superlattice extrinsic stacking
fault (SESF). The dislocation cutting and, thereby, the primary creep regime can
continue as long as new dislocations are not blocked at the γ/γ0 interface, preventing
the conservative cutting through the precipitate to go ahead. At this stage, the
secondary creep regime would initiate. The consequent restriction of the disloca-
tions’ movement and their accumulation inside the γ channel only, with a consequent
increase of their concentration, may drive the superalloy to the tertiary creep regime
that can be directly accessed from the primary one at higher temperatures. The
density of dislocations that can now move also by diffusion-assisted climb and
cross-slip in the γ channel can be taken as an indication of the increasing damage
of the alloy. The strain rate, after an incubation time needed to the dislocation density
to reach a critical value, goes up abruptly leading the superalloy rapidly to unac-
ceptable strain values, as shown by Fig. 6.47a. At higher temperatures, a mechanism
typical of single-crystal superalloys is observed. Under the effect of the applied
stress, a progressive coalescence of the γ0 precipitates occurs, resulting in plate-
shaped rafts, so that the mechanism is called “rafting.” This kind of directional
recrystallization determines the formation of rafted structures, oriented with the
basal planes orthogonal to the applied stress direction in case of negatively misfitting
alloys, which is the case in most commercial SC superalloys. If positive lattice misfit
354 6 Superalloys

Fig. 6.47 Creep curves for the CMSX4 nickel-based superalloy. (a) 950
C and 185 MPa, typical
tertiary creep behavior; (b) 1150
C and 100 MPa, the deformation behavior of the alloy is
dominated by the rafting mechanism. (Redrawn from Reed 2006)

is present, rafting determines the formation of needlelike rafts, oriented parallel to

the stress direction. Like for the tertiary creep, also rafting creep (Fig. 6.47b) displays
an initial incubation time, required for the recrystallization of the γ0 precipitates and
for the accumulation of dislocation at the γ/γ0 interface. Controversial aspects are still
standing, as concerns the effect of rafting on the creep behavior of superalloys,
reported to be beneficial in some respects. It is likely that an important contribution
to weakening the rafted superalloy comes from the increase in dislocation density
within the γ phase, similarly to the tertiary creep regime.

6.5.2 Principles and Technological Aspects of Directional

Solidification

DS and SC superalloy components generally have complex geometry, surface

topography, and a hollow internal structure, essential for an efficient cooling of the
component. The main production approach is the investment casting, or lost-wax,
process. This method implies the use of a “wax” model that ensures an adequate
surface finishing and dimensional accuracy, in view of the final machining and
polishing steps. The cooling of the blade is more and more crucial, considering the
operating temperatures that can be reached in the hot stages of the turbine. To
increase the heat exchange efficiency between the blade wall and the compressed
air circulating through it, complex patterns of the cooling internal channels and film-
cooling holes have been designed, much more complex than in early blades, for
which extensive mechanical and laser drilling were used, with obvious limitations on
the geometries that could be realized (see Fig. 6.48). The cooling patterns are
typically obtained by inserting ceramic cores inside the wax model. Differently
from the wax model, which is eliminated from the mold before its sintering and
6.5 Single-Crystal and Directionally Solidified Superalloys 355

Fig. 6.48 Comparison between the cooling systems in turbine blades. On the left and in the middle,
single-pass internal cooling; the straight channel and the film-cooling holes were originally just
drilled in the solid blade. The cooling efficiency was limited by the limited contact surface area
between the air flux and the alloy. On the right, a multi-pass serpentine cooling system, obtained by
inserting in the casting mold ceramic cores, removed after the alloy solidification. (Rolls-Royce
1996, images courtesy of Rolls-Royce plc)

subsequent casting of the alloy, the ceramic cores remain in place until the alloy
solidification is complete. Afterward, they are removed by ultrasonic crushing and
chemical leaching (see Fig. 6.49).
The investment shell usually comprises multiple, up to 20 or so, chambers for
producing several parts in each single run. For obtaining the investment shell, the
multiple-wax model is dipped repeatedly into a ceramic slurry, containing a mixture
of ZrSiO4, Al2O3, and SiO2 powders, suspended in colloidal silica or ethyl-silicate
alcohol. To improve the strength of the shell, the dipping step in the slurry may be
alternated with fluidized bed coating and stuccoing, using the same ceramic powders
contained in the slurry. Once the wanted wall thickness is reached, the mold is sent to
the consolidation thermal treatments. The first step is usually conducted in a steam
autoclave and is needed to melt the wax model away and to promote the evaporation
of the less volatile residues of the solvent of the slurry.
Subsequently, a proper firing treatment is carried out for a complete consolidation
and sintering of the ceramic mold that is now ready for the casting. The production of
reliable shells for investment casting to be used for the directional solidification
process has been a challenging issue to solve for the full development of this
technology. This is due to the more demanding conditions, as concerns the duration
356 6 Superalloys

Fig. 6.49 Schemes of the different stages of the production of the high-pressure stage turbine blade
of a gas turbine aero-engine: (1) wax model; (2) as-cast component; (3) finished blade after
machining

of the process and the thermal gradients across the advancing liquid-solid interface,
if compared to other conventional casting and solidification technologies.
For nickel-based superalloys, casting occurs under vacuum usually at a temper-
ature of 1550
C. The microstructure obtained in the solidified pieces is strictly
dependent on the adopted solidification mode. In the early applications of investment
casting to the production of gas turbine blades, polycrystalline components were
obtained using the “power down” method, consisting in switching off the furnace
after casting and leaving the solidification to occur spontaneously. The master alloy
melting typically occurs in a disposable crucible made of refractory fibers. The
furnace usually consists of an upper melting chamber, a central mold chamber
maintained at the correct temperature by a graphite resistance heating element, and
a lower withdrawal chamber (Fig. 6.50) (Reed 2006).
The directionality in the way heat is removed during solidification has been
achieved with the Bridgman method using vacuum furnaces like the one illustrated
in Fig. 6.50. To obtain the directional solidification, the mold that sits on a Cu cooled
plate (water cooled in the scheme in Fig. 6.50) is extracted from the bottom part of
the furnace. Typically, thermal gradients in the 4000
C/m range are achieved. The
time needed to pass from 1400 to 1000
C is typically reached within 20 min (Reed
2006). The directionality of the solidification is thus assured, and in this way DS
components can be obtained.
To obtain a SC component, it is necessary to eliminate from the solidifying cast
all grains but one. The most used technology in this field exploits a particular design
of the mold, also known as grain selector, which in the lower part, in contact with the
6.5 Single-Crystal and Directionally Solidified Superalloys 357

Fig. 6.50 Layout of an investment casting plant for the production of SC components, using the
Bridgman method. (Redrawn from Reed 2006)

cooled plate where the solidification begins, is characterized by a spiral (see

Fig. 6.51) or angular geometry. This particular design allows only one of the grains
that nucleate on the cold plate at the bottom of the mold to enter its upper part, i.e.,
the part having the desired component shape.
A spontaneous growth direction does exist for nickel-based superalloys, parallel
to the cube axis <001>. If, for any reason a different texture is interesting, the
solidification can be initiated by a “seed” that is placed at the bottom of the mold,
having the required crystallographic orientation, which will preferentially grow in
the rest of the solidified component. Considering the microstructure of the base block
of the SC blade shown in Fig. 6.51, it turns out that in case of directionally solidified
358 6 Superalloys

Fig. 6.51 (a) Working principle of the grain selector, used to obtain SC component, with sketch of
the real part; (b) a “seed” with a specific orientation can be used to induce that orientation, possibly
different from the spontaneous one, in the solidified component

components, no grain selector is necessary. The uniaxial orientation of the grains is

simply caused by the growth direction induced by the heat extraction direction.
The spontaneous growth direction for nickel-based superalloys (actually, in
general, for any fcc alloy) is the <001>, since it is perpendicular to the cube planes
{001}, featuring the lowest atomic surface density for an fcc structure (see Fig. 6.52).
Therefore, in case of a controlled solidification process, in which the solid phase
would actually form one atomic layer at the time, as it occurs in the Bridgman
solidification method, the <001> growth direction is naturally favored. The SC and
DS solidification of superalloy components leads to interesting properties: low
porosity, crystallographic texture, homogeneous microstructure and composition.
Low porosity castings are favored not only by the limited shrinkage associated with
the solid-liquid volume mismatch but also by the low-pressure atmosphere in the
solidification environment, preventing any significant gas trapping in the melt. As
shown in Sect. 6.6, the homogeneity in microstructure and composition is an
important prerequisite to carry out the refinement thermal treatments needed to
optimize the alloy properties. In fact, particularly the newest alloy compositions,
featuring high fractions of γ0 phase, exhibit a comparatively narrow solutioning
temperature range. The main challenge of DS and SC technology is to obtain over
the full component length the same level of chemical homogeneity that is usually
attained within each single grain of a polycrystalline alloy, i.e., over a few tens of
micrometers (Porter et al. 2009). Ruling out the equilibrium solidification, interest-
ing for geological or planetary solidification processes, but totally unsuitable for a
technologically relevant process, two alternative solidification regimes are in prin-
ciple feasible. For the treatment of these solidification regimes, it is practical to refer
6.5 Single-Crystal and Directionally Solidified Superalloys 359

Fig. 6.52 Comparison between the atomic surface density, ρhkl, for the lowest and highest atomic
packing planes in an fcc structure: the cube planes {001} and diagonal planes {111}

to the pseudo-binary phase diagram in Fig. 6.53, featuring a partition coefficient of

the solute atoms between solid and liquid given by:

CS
K¼ ð6:31Þ
CL

where:
CS is the concentration of the solute in the solid.
CL is the concentration of the solute in the liquid.
K in the present context is <1. Referring to the pseudo-binary phase diagram in
Fig. 6.53, when dealing with the solidification of multicomponent superalloys, is still
an acceptable choice, since superalloys solidify in a single phase (γ) and γ0 forms
through solid-state precipitation afterward.
In the first solidification regime, the composition of the liquid phase is perfectly
homogenized by mixing, e.g., by convection and stirring, two phenomena typically
present in conventional casting processes. On the other hand, diffusion, occurring in
the solid state, is not sufficiently fast to induce any meaningful homogenization of
the alloy composition. In agreement with these assumptions, Fig. 6.54 shows the
transient concentration profile together with the complete profile observed at the end
of the solidification process that can be assumed to proceed from left to right as
indicate by the arrow in Fig. 6.54.
The composition in the solid is not uniform, and a eutectic fraction is present in
the final segment of the profile. This is coherent with the phase diagram in Fig. 6.53
360 6 Superalloys

Fig. 6.53 Schematic of the pseudo-binary phase diagram used for the discussion of different
solidification regimes

Fig. 6.54 Transient concentration profile in the liquid and solid phase during solidification
occurring with perfect mixing in the liquid and no mixing in the solid. (a) T ¼ T1; (b)
T2 < T < T1; (c) T ¼ eutectic temperature (see Fig. 6.53)

and with the continuous solute enrichment of the liquid all through the solidification
range. With reference to the single-phase region (no eutectics), the concentrations of
both liquid and solid can be determined by solving the equation:
6.5 Single-Crystal and Directionally Solidified Superalloys 361

df S ðC L  C S Þ ¼ f L dCL ð6:32Þ

where:
fS is the solid fraction.
fL is the liquid fraction.
CL is the concentration of the solute in the liquid.
CS is the concentration of the solute in the solid.
Assuming the same molar volume for the liquid and the solid phase, it is possible
to rearrange Eq. 6.32 using the expressions:
(
f S þ f L ¼ 1 ! f S ¼ 1  f L ! df S ¼ df L
CS ð6:33Þ
K¼ ! C S ¼ KCL ! C L  C S ¼ CL ð1  KÞ
CL

So that Eq. 6.32 becomes:

df L dCL
¼ ð6:34Þ
fL C L ð1  K Þ

After the integration, being C0 the composition of the molten alloy, expressions
for CL and CS can be obtained as:
Z fL Z CL    
df L 1 dCL 1 1 CL 1
 ¼ ! ln ¼ ln !
fL 1K C
8L f 1  K C 0 f
1 0
L 1K L
 1 >
< CL ¼ C0 1 ð6:35Þ
1K fL
¼ CCL0 !  1K
>
: C S ¼ KC0 1
f L

The concentration profile in the solid phase can be easily homogenized through
thermal treatments, if it refers to crystalline grains with average size not larger than a
few hundreds of micrometers. Therefore, this concentration profile is not suitable
when dealing with very large grain size, in the range of several centimeters,
interesting for SC and DS products.
The solidification regime that provides a constant concentration in the solidified
alloy is the one in which the solid is not having any significant solute homogeniza-
tion via mixing phenomena in the liquid, being diffusion the dominant solute
transport phenomenon in the liquid. The concentration profile, at the beginning of
the solidification process, is shown in Fig. 6.55a. The progressive solute enrichment
of the liquid at the interface with the solid results in a progressive increase of the
solute concentration in the solid too (Fig. 6.55b). At some stage the composition of
the solid phase equals the C0 composition of the initial melt (Fig. 6.55c). From this
point onward, the composition in the solid remains constant up to the end of the
process, when a limited eutectic solidification would in general occur. Figure 6.55d
362 6 Superalloys

Fig. 6.55 Concentration profiles at different stages of the alloy solidification occurring with no
homogenization in the solid and diffusion in the liquid phase. (a) Initial stage; (b) transient; (c)
steady-state solidification; (d) solidification complete, with the formation of the final eutectics

shows the complete resulting concentration profile in the solid. Assuming a constant
advancement velocity (v) for the solid-liquid interface, the expression for the solute
concentration in the two phases over the steady-state solidification regime (station-
ary solidification) can be evaluated. By equating the amount of solute that is leaving
the solid on its formation to the amount that is diffusing away from the solid-liquid
interface into the liquid, with a rate given by the first Fick’s law, the resulting
expression can be written as:

∂CL
ðC L  C S Þv ¼ DL ð6:36Þ
∂x

DL being the diffusivity of the solute in the liquid phase. For the stationary state,
where CS¼C0, by introducing the variables:
6.5 Single-Crystal and Directionally Solidified Superalloys 363

y ¼ CL  C 0 and dy ¼ dCL ð6:37Þ

and assuming the following boundary condition:

 
C0 1k
yx¼0 ¼  C0 ¼ C0 ð6:38Þ
K K

For the steady-state solidification stage, Eq. 6.36 becomes:

∂CL
ðC L  C 0 Þv ¼ DL ð6:39Þ
∂x

Introducing the variables from Eq. 6.37, the Eq. 6.39 becomes:

dy v
¼  dx ð6:40Þ
y CL

By integrating Eq. 6.40:

Z   Z x !
y
1  K dy v y v
¼ dx ! ln
1K  ¼  x ð6:41Þ
C0 K y 0 D L C 0 K D L

Equation 6.42 provides the concentration profile in the liquid phase, during the
stationary solidification stage:

   
1K v
CL ¼ C0 1 þ exp  x ð6:42Þ
K DL

The constitutional temperature profile (TL, see Fig. 6.56) depends on the diffusion
coefficient in the liquid and on the advancing velocity (v) of the solid-liquid
interface.
The processing parameters involved in the former expressions strongly influence
the features of the solidifying alloy. Therefore, major efforts have been devoted to
the development of processing equipment and relevant procedures, capable to assure
high reliability and productivity associated with flexible operational parameters
(Pollock and Tin 2006; Pollock 2016). This resulted in the development of processes
capable to comply with the specific processing requests of different products, like
small and large size high-pressure blades for aircraft and terrestrial heavy duty gas
turbines, respectively.
Two parameters are regarded as particularly critical in the directional solidifica-
tion practice; they are thermal gradient in the liquid phase at the solid-liquid interface
and growth rate, affecting very much the microstructure of the resulting products.
Figure 6.57 gives the evolution of these two parameters with reference to three
casting technologies, all of them used in the production of superalloys. The “power
364 6 Superalloys

Fig. 6.56 Temperature profile in the solidifying liquid and constitutional supercooling. The liquid
ahead of the solidification front has the TL profile. Te indicates the equilibrium liquidus temperature
adjacent to the solid-liquid interface. Constitutional supercooling is obtained when the TLline lies
below the critical temperature gradient

down” approach has been mentioned already as the first used in investment casting
of Ni-based superalloys. The “withdrawal” corresponds to the Bridgman method
(see Fig. 6.50), and its performances have been remarkably improved by the liquid
metal cooling (LMC) method. The main novelty of this latter technology relies on
the substitution of the heat removal strategy from the investment casting shell and
solidifying alloy. The combined system, involving water-cooled chill plate, in direct
contact with the casting mold, associated with a radiative cooling system and
boosted by a gas flow, has been replaced by a metal bath (tin or aluminum),
extracting the heat by conduction in a more efficient and controllable way. In this
way, the variation range of the operational parameters has significantly been
expanded. This has resulted in the production of alloys with improved microstruc-
tural features, like a reduced fraction of inter-dendritic eutectics and the size and
shape of the γ0 precipitates, which render more effective the post-processing thermal
treatments (see Sect. 6.6).

6.6 Heat Treatments

Some specific aspects concerning the heat treatments of superalloys are discussed in
this section. Generally, these treatments are classified into four main groups: stress
relief, annealing, solutioning (or solution annealing), and age hardening (or just
aging); and a short description of each one of them is given in Table 6.24.
6.6 Heat Treatments 365

Fig. 6.57 (a) Variation ranges of the thermal gradients and growth rates accessible with different
setup for directional solidification processes (see main text for figure description); (b) schematic
illustration of the conventional Bridgman process; (c) schematic illustration of liquid metal cooling
(LMC) process. (Redrawn from Reed 2006)

The first two types of treatments are dealing with the microstructural refinement
of the alloy, with the elimination of defects and residual stresses. The focus of this
section is rather on the second pair, i.e., solution and aging treatments, that still
modify the microstructure of the superalloy but with the primary aim of inducing
the precipitation and refinement of the strengthening phases, like γ0 , carbides,
borides, etc. (see Appendix 6). In this respect, some aspects of these treatments
are similar to those considered already with regard to other precipitation and
age-hardenable alloys, like aluminum alloys (see Sects. 3.2.3.2 and 3.2.3.3) and
steels (see Sect. 5.4).
366 6 Superalloys

Table 6.24 Thermal treatments performed on superalloys – a summary (ASM International 2000,
2016)
Name Description
Stress relief Stress relief is often a compromise between the relief of residual stresses in the
material and effects that can be deleterious to corrosion resistance and/or high-
temperature properties. The stress relieving temperature typically falls within
the upper temperature range for precipitation hardening and is generally below
the recrystallization or annealing temperature. For Ni-based superalloys, it is
typically used to reduce/eliminate residual stresses in work-hardened no-age-
treatable alloys, but avoiding the recrystallization of the microstructure. A typ-
ical temperature range for Ni alloys is in between 425
C and 870
C,
depending on the chemical composition. Stress equalizing can be considered as
a subgroup of stress relieving. In this latter case, a low-temperature treatment is
used for homogenizing stresses after cold-working operations, without
decreasing the mechanical strength gained via cold-working
Annealing Annealing treatment of superalloys implies a full recrystallization and complete
softening when dealing with no-age-hardenable wrought alloys. When applied
to precipitation hardenable alloys, the treatment is essentially a solution
annealing (see below). The most important outcomes of annealing are ductility
increase, hardness reduction, solutioning of second phases. Annealing is in
general used as preliminary to forming, machining, and welding operations
Solution A solution annealing treatment can be regarded as the first step of the standard
annealing heat treatment of superalloys. Treatments performed at lower temperatures are
generally indicated for improving the resistance to notch-rupture sensitivity, to
improve fatigue resistance thanks to smaller grain size, better tensile properties
at elevated temperatures. No significant carbide solutioning or grain growth is
taking place at low temperature. Higher temperature treatments lead to optimum
creep-rupture properties thanks to coarser grain size, as well as the dissolution
of carbides. When followed by age hardening, the treatment results in large
grains containing aging phases (γ0 , γ00 ) and high concentrations of carbides
mainly at the grain boundary. The main goal of solution treatments is to
solubilize the γ0 precipitates and carbides that after solutioning may
re-precipitate with an optimized shape, more uniform size, and spatial distri-
bution. A quenching step is usually employed after solution annealing. Cooling
media depend on the alloy composition and can be water, oil, air, and also some
inert gas cooling
Age hardening Aging treatment is meant to induce the precipitation of one or more strength-
ening phases from a supersaturated alloy matrix. Important factors in this regard
are combination of desired strength and ductility, type and number of precip-
itating phases, precipitate size

Dealing with the thermal treatments of superalloys and also as concerns the
stability of precipitate phases, it is paramount to consider the effect of their surface
energy term and its dependence on the curvature radius of the precipitate itself. Let
us consider a two-phase alloy, made of α phase (parent phase) in which the β phase
precipitates. To show how the stability of the phase, β, depends on the size and shape
of the precipitates, in Fig. 6.58 two β-precipitates are shown: a spherical precipitate
having a finite radius r and another precipitate with a planar interface, corresponding
to an infinite radius.
6.6 Heat Treatments 367

Fig. 6.58 Scheme used to

discuss the stability of the β
phase in the α matrix, as a
function of the curvature
radius

The difference in free energy between the two precipitates is ΔGΓs, due to the
different contribution from the surface energy term, being the composition of the two
precipitates the same. The variation of the free energy of the system for a transfer of
dn moles of β from the spherical to the infinite radius precipitate can be expressed as:

dG ¼ ΔGΓS dn ð6:43Þ

Assuming constant the free energy of the infinite radius precipitate that, being
much larger than the other, is substantially unaffected by the infinitesimal mass
change, the free energy change of the system is entirely due to the decrease in the
surface area (dA) of the spherical precipitate:

dG ¼ ΓS dA ð6:44Þ

By equating the two expressions for the energy change:

ΓS dA
ΔGΓS dn ¼ ΓS dA ! ΔGΓS ¼ ð6:45Þ
dn

In case of a spherical precipitate, simple expressions for its surface area and
volume can be introduced in order to evaluate the relevant free energy terms:

A ¼ 4πr 2 ! dA ¼ 8πrdr ð6:46Þ

As concerns the number of moles contained in the spherical precipitate, the value
is given by:

4πr 3
n¼ ð6:47Þ
3V mβ

where V mβ is the molar volume of the β phase. Differentiating Eq. 6.47:

368 6 Superalloys

4πr 2 dr
dn ¼ ð6:48Þ
V mβ

In this way the following expression for ΔGΓS is obtained:

2
ΔGΓS ðrÞ ¼ ΓS V βm ð6:49Þ
r

Equation 6.49 is the Gibbs-Thomson equation and has very important conse-
quences on the free energy equilibrium curve of the α-β alloy.
As shown in Fig. 6.59, the extra energy term ΔGΓ(r) determines the rise of the
minima of the free energy curves for the β phase, depending on the different average
size of the precipitates. This has an effect on the equilibrium concentration in the α
matrix that can be evaluated, at a constant temperature, using the common tangent
construction. As the radius of the precipitate decreases, the tangent to the free energy
curve of α moves toward solute-richer concentrations.
A concentration gradient tends to be established in the alloy and activates a
diffusive flow, resulting in the so-called Ostwald ripening of the precipitates. With
reference to the scheme in Fig. 6.60, showing two β-precipitates with different radii,
the diffusive flow, triggered by the concentration gradient, results in the progressive
shrinking of the smaller one and coarsening of the larger one. The homogeneous
distribution of the size and planar interfaces of the precipitates are essential features
to obtain an intrinsically stable alloy microstructure. This latter feature is certainly

Fig. 6.59 Gibbs free energy stability curves for the α-β alloy binary, modified by the surface
energy terms for a finite radius of the β-precipitate phase
6.6 Heat Treatments 369

Fig. 6.60 Concentration profile in the α phase in between two β precipitates. The different
equilibrium concentrations at the α-β interface are due to the Gibbs-Thomson effect. The arrow
indicates the direction of the diffusive flow

one of the main aspects that the design of a successful thermal treatment should take
into account.
Alloy-specific parameters, like γ0 solvus, incipient melting, and liquidus temper-
ature, are essential information to set the right conditions of the treatments. The
pseudo-binary phase diagram in Fig. 6.61 shows the general trend of the superalloy
characteristic temperatures as a function of an increasing concentration of γ0 phase,
observed with the development of nickel-based superalloys, leading to the γ0 -richer
compositions of the latest-generation SC superalloys.
The complexity of the treatment is dependent on the target microstructure and
alloy properties, but also on the size of the component, ingot, bar, or semi-product, to
be treated. This is an important aspect for the correct evaluation of the time necessary
to reach any set temperature and for ensuring a controlled quenching rate.
Examples of real thermal cycles are shown by Fig. 6.62, concerning two conven-
tionally cast superalloys: Inconel 700 and Udimet 700 featuring different fractions of
γ0 (20% and 45%, respectively – see Fig. 6.61; for the alloy compositions see
Table 6.14).
The treatments of both Inconel 700 and Udimet 700 are discussed hereafter with
reference to the relevant temperature-transformation-time diagram in Fig. 6.62. For
the Inconel 700 superalloy, with a 20% γ0 phase, the heat treatment features two
main stages, involving the transformations shortly described by the scheme in
Fig. 6.63.
The Udimet 700 alloy has a higher percentage of γ0 phase. Therefore, a multiple-
stage treatment is recommended, in order to obtain the full precipitation of the γ0 and
carbide phases and relevant microstructural refinement. The heat treatment described
370 6 Superalloys

Fig. 6.61 Pseudo-binary

phase diagram referring to
different superalloys,
featuring different
concentrations of γ0 - phase.
(Redrawn from Guttmann
1981)

by Fig. 6.64 involves four main steps. The treatment sequence proposed for the
Udimet 700 alloy comprises two solutioning treatments. A two-step solutioning,
involving two different temperatures, is meant to obtain a complete solutioning of
those γ0 precipitates already present in the alloy, possibly featuring an inhomoge-
neous size distribution for the uncontrolled precipitation conditions and thereby a
different stability (see Eq. 6.49). The precipitates are subsequently stabilized by a
two-step aging. As detailed in the scheme in Fig. 6.64, the microstructural refine-
ment during the treatment involves also carbide phases and the grain morphology of
the superalloy.
Based on these principles, a broad range of heat treatment have been developed,
according to the alloy chemistry and target properties (ASM International 2016;
Donachie and Donachie 2002). Table 6.25 contains the main steps and parameters
for a number of heat treatments used for different groups of superalloys.
Also DS and SC alloys, notwithstanding the extreme control of the investment
casting process, still require post-processing treatments for the optimization of their
microstructure. The main aspects that are managed through these treatments concern
the optimization, usually the reduction, of the average size of the γ0 precipitates and
the elimination of the inter-dendritic eutectic regions. These latter are unavoidable
features of the as-solidified alloy and result from the solute enrichment in the liquid
phase, close to the solid-liquid interface (Figs. 6.54 and 6.55). The average size of
6.6 Heat Treatments 371

Fig. 6.62 TTT curves referring to the heat treatments for the microstructural refinement of two
polycrystalline superalloys with different fractions of γ0 phase: (a) Inconel 700, 20% γ0 ; (b) Udimet
700, 45% γ0 . (Redrawn from Guttmann 1981)

the γ0 precipitates tends to exceed the optimal value, since an excessive coarsening
may be favored by the long-term exposure to comparatively high temperatures,
required by the directional solidification process. The typical processing duration
involved with the directional solidification process is, in fact, significantly longer
than that of conventional casting processes. The refinement of the γ0 phase is a
fundamental aspect considering that it is the majority phase in superalloys belonging
to the latest generations (see Tables 6.15 and 6.16). The increasing fraction of γ0 in
372 6 Superalloys

Fig. 6.63 Heat treatment steps for Inconel 700. Related TTT curves are given in Fig. 6.62a.
(Guttmann 1981)

Fig. 6.64 Heat treatment steps for Udimet 700. Related TTT curves are given in Fig. 6.62b.
(Guttmann 1981)
Table 6.25 Examples of thermal treatments for selected polycrystalline superalloys (ASM International 2000, 2016; Donachie and Donachie 2002)
Stress relief Annealing Solution annealing Aging
Temperature Temperature Time Cooling Temperature Time Cooling
Group Alloy Temperature [
C] [
C] [
C] [h] media [
C] [h] media
Co- Haynes – – 1175 0.5 Rapid air – – –
6.6 Heat Treatments

based 188 cooling

Haynes Approx. 55
C below annealing 1230 1230 1 Rapid air Aging occurs at elevated tempera-
25 temperature (1) cooling tures during service
S-816 Approx. 55
C below annealing 1205 1175 1 (4) 760 12 Air
temperature (1)
Fe- 19-9DL 675 (2) 980 – – – – – –
based Alloy 718 (3) 955 980–1065 1 Air or water 718 8 Furnace
cooling
Discaloy (3) 1035 1010 2 Oil 650 20 Air
730 20 Air
Incoloy – – 980 1 Air 620 8 Air
907 775 12 Furnace
cooling
Incoloy – – 980 1 Air 620 8 Air
909
Incoloy – – 1010 1 Air 620 8 Air
925 730 8 Furnace
cooling
Inconel – – 925–1010 – – 620 8 Air
706 845 3 Air
N-155 Approx. 55
C below annealing 1230 1175 1 Water 815 4 Air
temperature (1)
(continued)
373
Table 6.25 (continued)
374

Stress relief Annealing Solution annealing Aging

Temperature Temperature Time Cooling Temperature Time Cooling
Group Alloy Temperature [
C] [
C] [
C] [h] media [
C] [h] media
Ni- Astroloy (3) 1135 1175 4 Air 845 24 Air
based Hastelloy – – 1065 0.5 Rapid – – –
S quench
Inconel – – 1175 2 (4) – – –
617
Inconel 870 980 1150 2 (4) – – –
625
Nimonic (3) 1080 1080 8 Air 705 16 Air
80A
Nimonic – – 1080 8 Air 705 16 Air
90
René 41 – – 1065 0.5 Air 760 16 Air
Udimet (3) 1080 1080 4 Air 845 24 Air
500
Waspaloy (3) 1010 1080 4 Air 760 16 Air
845 24 Air
(1) If further fabrication is needed, full annealing is recommended
(2) 675
C is the nominal temperature, the 650–705
C range is allowed
(3) Since intermediate temperatures lead to aging, full annealing is recommended
(4) Rapid cooling below 540
C is necessary to allow adequate quench and to avoid precipitation in the intermediate temperature range. For thin sections rapid
air cooling is enough; oil or water quenching is used for heavier sections that do not undergo cracking while cooling
6 Superalloys
6.6 Heat Treatments 375

the alloy results in an increasing γ0 solvus and, correspondingly, involves a reduction

of the temperature range available to the solutioning. This aspect renders an accurate
control of the treatment conditions particularly critical.
In Table 6.26 a series of heat treatments adopted for conventionally cast, DS and
SC superalloys are illustrated.
The importance of an optimized alloy microstructure is illustrated in Fig. 6.65,
with the creep life of the PWA1480 SC alloy as a function of the average size of the
ordered γ0 precipitates, exhibiting a maximum value close to 0.4 μm (Fig. 6.65a). In
Sect. 6.5.1, it has been demonstrated that the creep behavior of SC superalloys is
very much dependent on mechanisms involving dislocations and, thereby, their
interaction modes with γ0 precipitates. If the ordered precipitates have too a small
average size, dislocations may climb over them. Therefore, the obstacle provided by
the precipitates to dislocation movement is not that effective as it would be in case
dislocations were forced to cut through them, as shown by direct observations and
relevant modeling (Fig. 6.46). This is the condition corresponding to the peak in the
lifetime of the alloy, since the trailing dislocations remain blocked at the interface
between the γ matrix and the γ0 precipitates as long as their concentration becomes
very high. At this stage, the alloy enters the high deformation rate tertiary creep
regime, unless, as it is typically the case, the component is replaced. A reduction in
the creep life of the alloy is observed when too coarse precipitates render the bowing
(Orowan) mechanism relatively more important. The optimum solution temperature
of the PWA1480 SC alloy is restricted to a narrow range, which, through the
optimization of the precipitate shape and size, affords the most effective strength-
ening of the alloy (Fig. 6.65b). The softening of the alloy, observed for solution
treatments performed at lower temperatures, is due to residual of γ0 precipitates
formed during the solidification and cooling of the alloy during primary investment
casting solidification process. The size of these precipitates is too large to provide the
best mechanical behavior. On the other hand, if an excessive solution temperature is
used, defects in the microstructure of the alloy are introduced as a consequence of
unwanted melting. Since the subsequent solidification would occur in an
uncontrolled regime, the formation of defects and freckles would be unavoidable.
Concerning superalloy heat treatments, it is worth recalling the DMHT (dual
microstructure heat treatment) (see Sect. 6.4.3) developed by NASA for turbine disc
obtained via powder metallurgy. A sub-solvus heat treatment is conducted in the
bore region, while the rim is subjected to a super-solvus heat treatment. This is
achieved thanks to a particular apparatus, which is able to keep bore and ring at two
different temperatures at the same time, while the whole disc is kept inside the
treatment furnace. The bore region is encapsulated by a thermal insulating jacket, so
that it remains at lower temperatures with respect to the disc rim. The thermal
gradients present in the part during the heat treatment are enhanced by the heat
sink placed in contact with the bore inside the insulation jacket. The heat sinks are
usually solid metal cylinders. Thanks to this particular architecture of the furnace,
finer grain size and larger γ0 volume fractions can be attained in the bore region,
while the rim, being subjected to a super-solvus heat treatment, is characterized by
 376

Table 6.26 Thermal treatments for different kinds of nickel-based superalloys (Donachie and Donachie 2002)
Step 1 Step 2 Step 3 Step 4
Temp. Time [h] + Temp. Time [h] + Temp. Time [h] + Temp. Time [h] +
Alloy [
C] cooling [
C] cooling [
C] cooling [
C] cooling
Polycrystalline conven- IN-939 1160 4 – rapid air 1000 6 – rapid air 900 24 – air 700 16 – air
tional castings cooling cooling
René 41 1065 3 – air 1120 0.5 – air 900 4 – air – –
René 77 1163 4 – air 1080 4 – air 760 16 – air – –
René 80 1220 2 – gas furnace 1095 4 – gas furnace 1050 4 – air 845 16 – air
quench quench
Udimet 1150 4 – air 1080 4 – air 760 16 – air – –
500
Waspaloy 1080 4 – air 845 4 – air 760 16 – air – –
DS castings MAR-M 1230 2 – gas furnace 980 5 – air 870 20 – air – –
247 quench
MAR-M 1230 4 – gas furnace 1080 4 – air 870 32 – air – –
200Hf quench
René 80H 1190 2 – gas furnace 1080 4 – air 870 16 – air – –
quench
SC castings CMSX-2 1315 3 – gas furnace 980 5 – air 870 20 – air – –
quench
PWA 1480 1290 4 – gas furnace 1080 4 – air 870 32 – air – –
quench
René N4 1270 2 – gas furnace 1080 4 – air 900 16 – air – –
quench
6 Superalloys
6.6 Heat Treatments 377

Fig. 6.65 Creep curves of

the PWA1480 alloy for 1%
deformation and rupture, as
a function of the average
size of the γ0 precipitates, at
(a) 760
C and (b) 950
C.
(c) Yield strength at 600
C
as a function of the
solutioning temperature.
The reason for the decay of
the alloy strength, at
temperatures above and
below the optimum
temperature range, is
indicated in the graph

larger grain size and lower volume fraction of γ0 precipitates. The disc with this dual
microstructure exhibits excellent performances when exposed to the complex work-
ing condition of state-of-the art GTEs (Reed 2006).
378 6 Superalloys

6.7 Applications of Superalloys in the Aerospace Field

All through this chapter, the properties of the superalloys belonging to the three main
groups, cobalt-, iron- and nickel-based, have been discussed, and on some occasion,
a direct reference to specific applications has been made, like in case of directionally
solidified turbine blades and PM turbine discs. Considering the importance of these
alloys in the manufacture of aerospace gas turbine engines, an overview of their
specific applications is presented in this section, using the structure of a GTE as
guideline, moving from the compressor to the low-pressure stages of the gas turbine
(see Fig. 2.3).
As concerns compressor, in the inner stages where the highest air compressed
temperatures are reached, nickel- and iron-based superalloys are used for stationary
blades (vanes). Nickel-based superalloys are employed also for the blades and the
discs of the last compressor stages, closer to the combustor. In modern engines the
compression ratio can reach values close to 50 (see Fig. 2.8 and Sect. 2.6), and the
relevant gas temperatures require such choice. Forged blades and discs are used,
although the development of reliable powder metallurgy processes for the turbine
discs may extend this technology to compressor parts too. The blades, as those used
in the turbine, can be also produced by investment casting, although not requiring
directionally solidified microstructure.
For the combustor chamber, wrought cobalt superalloys are an excellent choice,
for their high-temperature stability and corrosion resistance, coupled with ease of
manufacturing, considering the good weldability of these materials. This is an
interesting feature both for the initial assembling of the engine and also for possible
repair and maintenance. The inner surface of the combustion chamber requires
protection by thermal barrier coatings (see Sect. 7.3.2), so that the hottest gases in
the whole engine are not in direct contact with the inner surfaces of the metallic wall.
An essential constituent of the TBC system is the bond coat, needed to assure a good
matching of the mechanical and thermophysical properties of the outer ceramic part
of the TBC and the underlying cobalt alloy. The composition of the bond coat has to
be selected accordingly.
TBC coatings are also used for the high-pressure blades and vanes in the turbine.
These parts are directionally solidified or single-crystal components. TBCs have to
sustain more demanding conditions than in the combustor chamber, considering that
the exhaust gases impinge directly onto the stationary and rotating blades, whereas in
the combustor, although hotter, the flux is nearly tangent to the metal parts. More-
over, the rotating blades are stressed by centrifugal forces. For these reasons, the
coating reliability is paramount, and this is one of the main reasons why electron
beam physical vapor deposition (EB PVD) is used in the place of PS, as discussed in
Sect. 7.2.2.
Investment-cast nickel-based superalloys are used for the blading and discs of the
intermediate and low-pressure stages of the turbine. Considering the less demanding
conditions, equiaxed alloys are used for the disc, either produced by forging or by PM
approach. The turbine disc can be regarded as one of the most critical components in a
6.7 Applications of Superalloys in the Aerospace Field 379

GTE. This becomes particularly evident when analyzing the incidence of the turbine
disc on both the weight and the cost of a modern aircraft engine. For example, in the
Rolls-Royce Trent 800, the turbine disc accounts for about the 20% of the total engine
weight while representing about 10% of its value. In a military engine, such as the
EJ200 (produced by Eurojet Turbo Gmbh) mounted on the Eurofighter Typhoon jet,
the turbine disc accounts for about the 5% of the total engine weight, but for about
25% of its entire value. The main task of turbine discs is to provide fixturing to the
turbine blades, that receive kinetic energy from the gas stream coming from the
combustor (see Fig. 2.3). The disc-blade assembly employs this energy to operate the
front fan and the compressor by means of single or multiple shafts, depending on the
engine’s architecture (see Figs. 2.10 and 2.11) (Reed 2006). The design of a turbine
disc is optimized, so to satisfy the requirements of dimensional stability, weight
minimization, mechanical resistance, and integrity, in order to sustain general yield-
ing and low-cycle fatigue cracking in the region near the bore. Furthermore, turbine
blades may transfer part of the heat from the hot gas flux so that the temperature in the
disc’s rim can reach peak values close to 650
C. In these conditions, centrifugal
forces due to the high rotational speed, even higher than 10,500 rpm, can have severe
consequences on the material integrity (Reed 2006). Creep and high-temperature
dwell crack propagation resistance are assuming an increasing importance in newly
developed engines. In this regard, PM started to be used for the production of this
critical component and the possibility to have locally optimized microstructure, and,
thereby, properties in the bore and the rim of the same turbine disc have gained great
interest (see Sect. 6.6). The use of PM was initially limited to GTEs used in the
military field, like in the Pratt & Whitney F100 engine employed for the propulsion of
the McDonnell Douglas F-15 “Eagle” jet fighter (early operational flight in 1974).
Subsequently, the experience developed in the military field was transferred to
civilian transportation, and PM discs are used in the engines for the Boeing
747 and 777 airliners, the General Electric CF6-80 and GE90, respectively (Reed
2006).
A list of superalloy components used in aerospace applications is presented
herewith with the same order used in this chapter for the description of each family
of superalloys. Concerning Co-based superalloys, they are used for those compo-
nents subjected to working conditions where hot corrosion is a major concern at
moderate-to-high temperatures and for low-stress structural applications. Moreover,
Co-based superalloys have superior hot corrosion resistance in environments
containing sulfur, lead oxides, and other compounds produced from the combustion
of jet fuels. Superior stress-rupture properties and hot corrosion resistance make
cobalt alloys a suitable choice for vanes and other stationary components (Geddes
et al. 2010; Mouritz 2012). With reference to the alloys presented in Sect. 6.2 (see
also Table 6.3), a list of components made of Co-based superalloys is given in
Table 6.27.
Iron-based superalloys (see Sect. 6.3) provide a very wide combination of
properties that are interesting for the fabrication of several components in the
aerospace field and also for other structural, possibly at high temperature, applica-
tions. In Table 6.28, the main applications of these superalloys are listed.
380 6 Superalloys

Table 6.27 Uses of selected Co-based superalloys (ASM International 2000; Geddes et al. 2010;
Mouritz 2012; Zhang and Zhao 2012)
Alloy Application
AiResist 213 Gas turbine components
AiResist 215 Gas turbine engines (stator vanes)
FSX-414 Gas turbine stator vanes
FSX-418 Gas turbine stator vanes
FSX-430 Gas turbine stator vanes
Haynes 25 Bearing material (balls and races)
Combustion chamber
Flame holder components
Liners
Haynes 188 Afterburner components
Combustion cans
Military and commercial GTE components
Transition ducts
MAR-M 302 Buckets in GTE
Nozzle guide vanes
Turbine vanes
MAR-M 322 Turbine blades
Turbine vanes
MAR-M 509 Gas turbine engines blades and vanes
MP35-N Fasteners for aerospace applications
MP159 Aircraft landing gear
Airframes
Fasteners used in GTE
Space shuttle propulsion system
NASA Co-Re-W High-temperature space applications
S-816 Gas turbine engines blades, bolts, and springs
WI-52 Gas turbine engines stator vanes
X-40 Gas turbine engine stator vanes
X-45 Gas turbine engine stator vanes

All through the chapter, Ni-based superalloys have been described as materials
possessing an exceptional combination of properties and high-temperature capabil-
ities, and for this reason, their use is indicated for the fabrication of the most
demanding components for aerospace turbine engines. In Table 6.29 the most
important applications of Ni-based superalloys (see Sect. 6.4) restricted to the
aerospace field are provided.
6.7 Applications of Superalloys in the Aerospace Field 381

Table 6.28 Applications in aerospace and related high-temperature fields of selected Fe-based
superalloys (Geddes et al. 2010; Haynes International 2017a, b; Schafrik et al. 2001)
Alloy Application Additional information
19-9DL Afterburners components High strength, oxidation, and corro-
Casings sion resistance up to about 670
C.
Industrial gas turbine blades Relatively low cost
Turbine and supercharger wheels
Alloy 718 Compressor blades and discs, casings, It is the most widely used Fe-based
fasteners, locking lugs, other com- superalloy. The creep-rupture strength
pressor components (rear shaft, hub tie is maintained up to 650–700
C
clamps, and supports)
Turbine discs and wheels
Discaloy Bolts –
GTE parts
Hastelloy X GTE combustors (combustor cans, Hastelloy X is characterized by good
flame holders, spray bars, transition oxidation resistance and good high-
ducts) temperature strength. It also shows
GTE fabricated parts exceptional resistance to stress corro-
GTE parts (afterburners, tailpipes) sion cracking
Haynes 556 GTE sheet components (afterburners, The alloy is similar to N-155, but the
combustors, transition ducts) compositional control has been
Heat treating improved. It has improved oxidation
Waste incineration and hot corrosion resistance, high-
temperature ductility, fatigue resis-
tance, thermal shock resistance, and
weldability
Haynes N-155 Afterburner parts, exhaust manifolds, The alloy can be used in applications
combustion chambers, nozzles, turbine involving high stress at temperatures
blades up to 800–1100
C. This alloy is also
used in the fabrication of heat treat-
ment equipment, where high-
temperature strength and corrosion
resistance are two essential features
Incoloy 800 Heat exchangers Excellent carburization and oxidation
GTE parts resistance. Good strength
Incoloy 901 GTE rotors –
Incoloy 903 GTE rings and casings Low and constant thermal expansion
Rocket components coefficient up to 430
C. High thermal
fatigue and thermal shock resistance
Incoloy 907 GTE seals Low and constant thermal expansion
GTE shafts coefficient and improved notch-
GTE casings rupture properties at elevated temper-
atures in comparison with Incoloy 903
Incoloy 909 GTE casings Low and constant thermal expansion
GTE shafts coefficient. In comparison with
GTE shrouds Incoloy 903 and 907, the addition of
Si leads to improved tensile properties
GTE vanes
and notch-rupture properties, while
Turbine shafts heat treatments can be shorter and
Turbine supports and other processing becomes less restrictive
components
(continued)
382 6 Superalloys

Table 6.28 (continued)

Alloy Application Additional information
MA-956 GTE combustion chamber This is an ODS alloy (see Sect. 9.3)
with a ferritic matrix. Exceptional
resistance to carburization, hot corro-
sion and oxidation at temperatures
above 1100
C
N-155 GTE sheet components (afterburners, It has been developed before Haynes
combustors, transition ducts). Exten- 556. It can be used in high-stress
sive use in older aircraft GTEs applications up to 820
C and in
moderate-stress applications up to
1090
C
V-57 GTE rotors –
GTE bolts
REX-78 Blading in the Whittle engine (see –
Sect. 2.1)
W-545 GTE parts –

Table 6.29 Applications in the aerospace field of selected Ni-based superalloys (Geddes et al.
2010; Mukherji et al. 2011; Reed 2006)
Alloy Application Additional information
Astroloy Forgings for high –
temperatures
CMSX-4 Turbine blades Second-generation SC superalloy
CMSX-10 Turbine blades Third-generation SC superalloy
Hastelloy B Old rocket engines and GTEs Good strength up to 1095
C, oxidation resis-
tance up to 650
C
Hastelloy S GTE components subjected to Relatively low coefficient of thermal expan-
low stress sion. Good fatigue resistance and high thermal
stability
Hastelloy W Combustors –
GTE fabricated parts
Haynes 230 Combustion environments Properties in terms of oxidation resistance,
strength, and stability up to 1150
C
High-temperature bellows
Inconel 617 Combustion cans in GTE Exceptional combination of oxidation resis-
Ducting an transition liners in tance, metallurgical stability and strength at
GTE high temperature
Inconel X750 GTEs High-stress relaxation resistance. High creep-
High-temperature springs and rupture and tensile properties maintained up to
bolts 700
C
Rocket engines
Aircraft structures
LSHR Turbine discs Third-generation PM superalloy
M-252 GTE blades, parts, sheets –
ME3 Turbine discs Third-generation PM superalloy
(continued)
References 383

Table 6.29 (continued)

Alloy Application Additional information
Nimonic 75 GTE sheet metal fabrications –
Nimonic 80A GTE blades, bolts, discs, rings Properties in terms of corrosion resistance,
tensile and creep-rupture strength up to 815
C
Nimonic 86 GTE sheets (afterburners and The addition of Ce leads to scaling and oxida-
combustion chamber) tion resistance up to 1050
C. Good formability
and weldability
Nimonic 90 GTE blades and discs High resistance against hot corrosion and oxi-
dation. High creep and stress-rupture properties
up to 920
C
PWA 1480 Turbine blades First-generation SC superalloy
PWA 1484 Turbine blades Second-generation SC superalloy
René 41 GTE blades and other parts –
René 88DT Turbine discs Second-generation PM superalloy
René 95 Discs for compressor and tur- First-generation PM superalloy
bine in GTEs
René N5 Turbine blades Second-generation SC superalloy
TMS 162 Turbine blades Fifth-generation SC superalloy
TMS 196 Turbine blades Fifth-generation SC superalloy
Udimet 720 Discs for compressor and tur- High metallurgical stability. Good impact, oxi-
bine in GTEs dation, and sulfidation resistance
Unitemp Turbine discs –
AF2-1DA
Waspaloy Airframe assemblies Used for the fabrication of highly loaded com-
Fasteners ponents with service temperature up to 650
C.
GTE components (casings, High strength and oxidation resistance at high
rotor discs, shafts, spacers, temperature
rings)

References

Antony K C (1983) Wear-resistant Cobalt-base Alloys. JOM 35 (2): 52–60

Ashby M F (1972) A first Report on Deformation-mechanism Maps. Acta Metallurgica 20 (7):
887–897
Ashby M F, Frost H J (1982) Deformation Mechanisms Maps. Pergamon Press
ASM International (1991) ASM Handbook Vol. 2 – Properties and Selection of Nonferrous Alloys
and Special Purpose Materials. ASM International, Materials Park, Ohio
ASM International (1993) ASM Handbook Vol. 1 – Properties and Selection: Iron, Steel and High
Performance Alloys. ASM International, Materials Park, Ohio
ASM International (2000) ASM Specialty Handbook – Nickel, Cobalt and Their Alloys. ASM
International, Materials Park, Ohio
ASM International (2008) ASM Handbook Vol. 15 – Casting. ASM International, Materials Park,
Ohio
384 6 Superalloys

ASM International (2016) ASM Handbook Vol. 4E – Heat treating of Nonferrous Alloys. ASM
International, Materials Park, Ohio
Barrett C S, Massalski T (1980) Structure of Metals 3rd edn. Pergamon Press
Bauer A et al. (2010) Microstructure and Creep Strength of different γ/γ0 -strengthened Co-base
superalloy variants. Scripta Materialia 63 (12): 1197–1200
Cahn R W, Haasen P (1996) Physical Metallurgy 4th edn. North Holland
Callister W, Rethwisch D (2012) Fundamentals of Materials Science and Engineering: An Inte-
grated Approach, 4th edn. Wiley John & Sons
Campbell F C (2006) Manufacturing Technology for Aerospace Structural Materials. Elsevier
Chandler H (1996) Heat Treater’s Guide: Practices and Procedures for Nonferrous Alloys. ASM
International, Materials Park, Ohio
Cui C et al (2006) A New Co-Base Superalloy Strengthened by γ0 Phase. Materials Transactions
47 (8): 2099–2102
Davis J R (2000) The Nickel Industry: Occurrence, Recovery, and Consumption. In: ASM
Specialty Handbook – Nickel, Cobalt and Their Alloys. ASM International, Materials Park,
Ohio, p 3–6
Dimiduk D (1991) Dislocation Structures and Anomalous Flow in L12 Compounds. Journal de
Physique III 1 (6): 1025–1053
Donachie M J, Donachie S J (2002) Superalloys – A Technical Guide. ASM International
Everhart J (1971) Engineering Properties of Nickel and Nickel Alloys. Springer US
Farrar J (2004) The Alloy Tree – A guide to Low-Alloy Steels, Stainless Steels and Nickel-base
Alloys. Woodhead Publishing Limited and CRC Press
Fergus JW (2010) Recent developments in cathode materials for lithium ion batteries. Journal of
Power Sources, 195(4): 939–954
Fontani M et al (2015) The Lost Elements: The Periodic Table’s Shadow Side. Oxford University
Press
Frost J H, Ashby M F (1982) Deformation-Mechanism Maps: The Plasticity and Creep of Metals
and Ceramics. Pergamon Press
Geddes B et al (2010) Superalloys: Alloying and Performance. ASM International, Materials Park,
Ohio
Gell M et al (1987) Advanced Superalloy Airfoil. JOM 39 (7): 11–15
Giamei A. F. (2013) Development of Single Crystal Superalloy: A brief History. Advanced
Materials and Processes 171 (9): 26–30
Guttmann V (1981) Phase Stability in High Temperature Alloys. Applied Science Publishers LTD,
London
Haynes International (2017a) HASTELLOY® X alloy. Available via DIALOG. http://www.
haynesintl.com/alloys/alloy-portfolio_/High-temperature-Alloys/HASTELLOY-X-alloy/
HASTELLOY-X-principal-features.aspx. Accessed 27 Dec 2018
Haynes International (2017b) MULTIMET® alloy. Available via DIALOG. https://www.
haynesintl.com/alloys/alloy-portfolio_/High-temperature-Alloys/MULTIMET-alloy/principal-
features.aspx. Accessed 28 Dec 2018
Hemeker K J et al (1991) An Investigation of the Mechanisms that Control Intermediate Temper-
ature Creep of Ni3Al. Acta Metallurgica et Materialia 39 (8): 1901–1913
Hull D, Bacon J (2011) Introduction to Dislocations 5th edn. Butterworth-Heinemann
Jena A K, Chaturvedi M C (1984) The Role of Alloying Elements in the Design of Nickel-base
Superalloys. Journal of Materials Science 19 (10): 3121–3139
Jozwik P et al (2015) Applications of Ni3Al Based Intermetallic Alloys – Current Stage and
Potential Perspectives. Materials 8 (5): 2537–2568
Kear B H, Wilsdorf H G F (1962) Dislocation Configurations in Plastically deformed Cu3Au
Alloys. Transactions of the Metallurgical Society of AIME 224: 382–386
Kracke A (2010) Superalloys, the most successful Alloy System of modern Times – Past, Present
and Future. In: Ott E A et al (eds) Superalloy 718 and Derivatives – Proceedings of the 7th
International Symposium on SUPERALLOY 718 and DERIVATIVES, Pittsburgh, 2010
References 385

Krakow R et al (2017) On the Crystallography and Composition of topologically close-packed

phases in ATI 718Plus ®. Acta Materialia 130: 271–280
Kuo K H et al (1985) Tetrahedrally close-packed Phases in Superalloys: new Phases and Domain
Structures observed by high-resolution Electron Microscopy. Journal of Materials Science
21 (8): 2597–2622
Massalski T B et al (1986) Binary Alloys Phase Diagrams Vol. 2. ASM International
Mouritz A (2012) Introduction to Aerospace Materials. Woodhead Publishing Limited
Mukherji D et al (2011) Beyond Ni-based Superalloys: Development of CoRe-based Alloys for Gas
Turbine Applications at Very High Temperatures. International Journal of Materials Research
102(9): 1125–1132
Murakumo T et al (2004) Creep Behaviour of Ni-base Single-crystal Superalloys with various γ0
Volume Fraction. Acta Materialia 52 (12): 3737–3744
Murray J L (1982) The Co-Ti (Cobalt-Titanium) System. Bulletin of Alloy Phase Diagrams 3:
74–85
Murray J L (1987) Phase Diagrams of Binary Titanium Alloys. ASM International
Nabarro F R N, de Villers F (1995) The Physics of Creep and Creep-resistant Alloys. CRC Press
Nakajima K et al (2018) Global Distribution of Material Consumption: Nickel, Copper and Iron.
Resource, Conservation and Recycling 133: 369–374
Neumeier S et al (2016) Diffusion of Solutes in fcc Cobalt investigated by Diffusion Couples and
First Principles Kinetic Monte Carlo. Acta Materialia 106: 304–312
Pollock T M (2016) Alloy Design for Aircraft Engines. Nature Materials 15 (8): 809–815
Pollock T M, Tin S (2006) Nickel-Based Superalloys for Advanced Turbine Engines: Chemistry,
Microstructure, and Properties. Journal of Propulsion and Power 22 (2): 361–374
Porter D, Easterling K E, Sherif M Y (2009) Phase Transformations in Metals and Alloys, 3rd end.
CRC Press
Prasad E N, Wanhill R (2017) Aerospace Materials and Material Technologies Volume 1: Aero-
space Materials. Springer
Radavich J F (1989) The Physical Metallurgy of Cast and Wrought Alloy 718. In: Loria E A
(ed) Superalloys 718 Metallurgy and Applications. TMS, p 229–240fi
Rae C M F et al (2000) Topologically close-packed Phases in an experimental Rhenium-containing
Single Crystal Superalloy. Superalloys 2000: 767–776
Rae C M F, Reed R C (2001) The Precipitation of Topologically close-packed Phases in Rhenium-
containing Superalloys. Acta Materialia 49 (19): 4113–4125
Reed R C (2006) The Superalloys – Fundamentals and Applications. Cambridge University Press,
Cambridge (UK)
Reed R C et al. (2007) Damage Accumulation during Creep Deformation of a Single Crystal
Superalloy at 1150
C. Materials Science and Engineering A 448 (1–2): 88–96
Reed-Hill R E (1973) Physical Metallurgy Principles 2nd edn. PWS-KENT Publ., Boston
Ricks R A et al (1983) The Growth of γ Precipitates in Nickel-Base Superalloys. Acta Materialia 31:
43–53
Riddihough M (1970) Stellite as a wear-resistant material. Tribology 3 (4): 211–215
Rogister C et al (1967) Improvement of Heat Resisting Cobalt Base Alloys by Precipitation
Hardening. Cobalt 34: 3–9
Rolls-Royce (1996) The Jet Engine 5th edition. Rolls-Royce plc, Derby
Sato J et al (2006) Cobalt-Base High-Temperature Alloys. Science 312 (5770): 90–91
Schafrik R E et al (2001) Application of Alloy 718 in GE Aircraft Engines: Past, Present and Next
Five Years. In: Fifth International Symposium on Superalloys 718, 625, 706 and Derivatives,
Embassy Suites Hotel, Pittsburg, Pennsylvania, 17–20 June 2001
Seiser B et al (2011) TCP Phase Predictions in Ni-based Superalloys: Structure Maps Revisited.
Acta Materialia 59 (2): 749–763.
Sims C T (1984) A History of Superalloy Metallurgy for Superalloy Metallurgist. Superalloys 1984:
Proceedings of the Fifth International Symposium on Superalloys, Seven Springs Mountain
Resort, Champion, Pennsylvania, 7–11 Oct 1984
386 6 Superalloys

Sims C T et al (1987) Superalloys II. John Wiley & Sons, New York
Sjöberg G (2010) Casting Superalloys for Structural Applications. In: Ott E A et al (ed) Superalloy
718 and Derivatives – Proceedings of the 7th International Symposium on SUPERALLOY
718 and DERIVATIVES, Pittsburgh, 2010
Smallman R E et al (2002) Inverse Creep in Intermetallics. Materials Science and Engineering A
329–331: 852–855
Smallman R E, Ngan A H W (2014) Modern Physical Metallurgy 8th edn. Butterworth-Heinemann
Suzuki A, Pollock T M (2008) High-temperature Strength and Deformation of γ/γ0 two Phase
Co-Al-W-base alloys. Acta Materialia 56 (6): 1288.1297
Tylcote R F (2002) A History of Metallurgy. Maney Publishing
U.S. Geological Survey (2017) Mineral Commodity Summaries. Available via DIALOG. https://
minerals.usgs.gov/minerals/pubs/commodity/cobalt/mcs-2017-cobal.pdf. Accessed Jun 2018
Van Schilfgaarde M et al (1999) Origin of the Invar Effect in Iron-Nickel Alloys. Nature 400: 46–49
Viatour P et al (1973) Stability of the Gamma Prime Co3Ti Compound in Simple and Complex
Cobalt Alloys. Cobalt 3: 67–74
Yoo M H (1987) Stability of Superdislocations and Shear Faults in L12 ordered Alloys. Acta
Metallurgica 35 (7): 1559–1569
Zenck C H et al (2017) A novel Type of Co-Ti-Cr-base γ/γ0 Superalloys with low Mass Density.
Acta Materialia 135: 244–251
Zhang S, Zhao D (2012) Aerospace Materials Handbook. CRC Press

Further Reading

Durand-Charre M (1997) The Microstructure of Superalloys. CRC Press

Gessinger G H (2013) Powder Metallurgy of Superalloys. Butterworth-Heinemann
Kazantseva N V et al (2019) Superalloys – Analysis and Control of Failure Process. CRC Press
Ott E et al (2010) Superalloy 718 and Derivatives – Proceedings of the 7th International Symposium
on SUPERALLOY 718 and DERIVATIVES, Marriott Pittsburgh City Center, Pittsburgh,
Pennsylvania, 10–13 Oct 2010
Srivastava R R et al (2014) Resource recycling of superalloys and hydrometallurgical challenges.
Journal of Materials Science 49 (14): 4671–4686
Chapter 7
Coatings

7.1 General Aspects

Coatings are extensively used in gas turbine engines, as illustrated by Fig. 7.1. They
are designed to improve the capabilities of the different components of the engine,
with specific regard to surface stability against high-temperature corrosion and
thermally activated transformations. Particularly for the compressor, erosion and
fretting resistance are other aspects that can be tackled using suitable coating
systems.
Coatings are also important to ensure and retain high engine efficiency by
reducing the mechanical tolerances and gas leakages, for which abradable and seal
coatings are used.
As concerns surface stability, the main degradation phenomena are high-
temperature corrosion, involving oxidation and hot corrosion (see Chap. 8), wear,
and erosion. High-temperature corrosion is particularly important in the hottest
stages of the turbine and requires the use of overlay and diffusion coatings, capable
to form on their own surface protective layers, mainly made of native oxides. Wear is
rather a complex phenomenon, involving both mechanical stresses, possibly cyclic,
like fretting, and chemical reactions, like oxidation. For this reason, the surface
treatments and coating systems used for these purposes require specific design for
each part to be protected. Erosion is quite an issue for the front part of the engine, i.e.,
fan and early compressor stages, considering the ingestion of particulate matter and
sands from the outer atmosphere. Since the erosion rate depends on the direction of
the impinging particles, composite cermet coatings are the best candidates, consid-
ering the wide range of impact angles faced in real conditions. For ground-based gas
turbines, generally using not so refined fuels as in aerospace engines, erosion is also
observed in the high-pressure stages of the turbine as an effect of the combustion
soot carried by the high-velocity exhaust gas flow coming out from the combustion
chamber.

© Springer Nature Switzerland AG 2020 387

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_7
388 7 Coatings

Fig. 7.1 Coating systems in a gas turbine aero-engine, with the indication of the main surface
degrading mechanisms (Rajendran 2012; Rolls-Royce 1996, image courtesy of Rolls-Royce plc)

Clearance control and seal coatings are obviously interesting for retaining an
excellent engine efficiency over its entire lifetime. These systems are there to
optimize the mechanical matching between rotary and stationary parts. Wear-
resistant and abradable coatings are generally facing one another to achieve this.
Still, in some way related to efficiency improvement are the thermal barrier coatings,
because they allow increasing operating temperatures of the engine, beyond the
limits of the component alloys.
For a correct design and selection of an appropriate coating system, it is para-
mount to know the variation range of the operating parameters. Reference values of
the more critical of these parameters for gas turbine aero-engines are listed in
Table 7.1. They are classified with reference to the dominant character of the attack:
mechanical, chemical, and thermal. These parameters are of course fundamental not
only to identify and select the most suitable coating for any specific application but
also the relevant deposition techniques, discussed in Sect. 7.2.

7.2 Coating Deposition Techniques for Gas Turbine

Engines

Several techniques have been developed over the years and employed to deposit coatings
onto components of gas turbine engines in order to protect them from the attack and
damaging phenomena described in the previous section. Two of the most important
requirements for these techniques are the good reproducibility of the coating features
and the flexibility as concerns the geometry of the substrate on which the coating is
deposited. These features are particularly important for aero-engine applications and
7.2 Coating Deposition Techniques for Gas Turbine Engines 389

Table 7.1 Indication of the main working conditions and relevant variation range interesting for
the design of coating systems for aerospace (and terrestrial) gas turbine engines (Hocking et al.
1989)
Mechanical
Load or aggressive
phenomenon Maximum value (terrestrial) Damage
Centrifugal stresses 170 MPa Fatigue
Non-stationary 30 MPa (16 MPa) Micro-cracking
stresses (gas T and P)
Non-stationary 600 m/s (300 m/s) Coating removal, erosion
stresses (gas flux)
Shocks – Coating removal
Chemical
Oxidizing – Oxidation and corrosion, composi-
atmosphere tional changes, localized attack
Combustion Up to 0.3 wt% S Corrosion, compositional changes
products
Air pollution Marine salt, industrial pollution, Corrosion, erosion, surface
fly ash, particulate matter roughening
Thermal
Exhaust gas 1400–2000
C Phase and microstructural
transformations
Components 1000–1300
C Phase and microstructural
transformations
Thermal gradients vs 100
C/s Thermomechanical fatigue
time
Thermal gradients vs 200
C/mm Thermomechanical fatigue
space

represent a prerequisite for the achievement of adequate safety and reliability levels. All
coatings are deposited, or directly form, like diffusion coatings, on engine parts sometimes
having rather complex shapes and surface topography. Nonetheless, the thickness and
adhesion strength have to be uniform all over the coated surface. The same uniformity
requirement can be extended to the main coating characteristics, like composition,
microstructure, and internal stress fields. The deposition techniques herewith discussed
are very well established and optimized to produce coatings that for gas turbine engines
have thickness typically ranging from tens to hundreds of micrometers, i.e., much larger
than the average thickness of thin films used in other fields, like microelectronics and
optical and aesthetic coatings, just to mention a few major categories. The different
operating conditions present in different parts of the engine require different coating
systems and, consequently, different deposition techniques. The main ones used for gas
turbine aero-engines belong to the following groups: thermal spraying deposition, chem-
ical vapor deposition (CVD), and physical vapor deposition (PVD). In the next sections,
the main features of the aforementioned techniques will be discussed with particular
reference to the most important process parameters and resulting properties.
390 7 Coatings

7.2.1 Thermal Spraying

Quite a large number of thermal deposition techniques are available, the most
important and interesting for the aerospace field being (Fauchais et al. 2014):
• Flame spraying
• High-velocity oxy-fuel flame spraying
• Detonation gun and super detonation gun
• Electric arc spraying
• Plasma spraying
All these coating deposition techniques are characterized by a specific heat
source, after which they are named, needed to melt the feedstock to be deposited.
Combustion is the approach to generate heat in the first three techniques of the list,
with differences in the fuel and carrier gases. The energy transferred by an electrical
discharge is exploited by the electric arc deposition system, whereas in plasma
spraying, the energy input to the feedstock powders comes from a high-temperature
ionized gas. The droplets are produced by torches or guns and carried by a gas
stream, sometimes comprising also an auxiliary carrier gas flow, towards the part to
be coated and splat onto it. Depending on the deposition system, the feedstock can be
in the form of powders, wires, or rods. In all cases, the complete deposition of a
thermally sprayed coating requires the controlled movement of the spraying nozzle
over the surface of the component with an appropriate overlapping in order to
achieve the complete coverage of the surface. The temperature rise in thermal
spraying techniques is meant to promote the melting of the source material in
order to improve the coating formation process. In this regard, the bonding mech-
anisms, their control, and their optimization are very important indeed. Initially, the
interaction occurs essentially between the substrate and the impinging particles.
Subsequently, once the first layer of the coating has formed, the incoming particles
will interact and try to adhere to this layer. Mechanical bonding is one of the main
adhesion mechanisms, and surface roughness of the substrate and of the already
formed layers is therefore paramount. Diffusion and chemical bonding are other
possible interaction mechanisms (Fauchais et al. 2014).
The thermal spraying techniques are line-of-sight processes. Therefore, those
parts only that can be directly reached by the particle jet will have the chance to
be coated. To reach the final thickness, repeated passes are generally required, and
this can be achieved by combining together controlled relative displacements and
rotations of the “source” of material droplets/particles and the substrate. For the
controlled movement of the spraying nozzle, depending on the complexity of the
geometry and topography of the surface to be coated, options ranging from a plain
x-y system up to more sophisticated robotic arms can be selected. At the same time,
the part is also displaced and rotated relatively to the particle flux, in order to obtain a
homogeneous coverage of the substrate and uniform thickness of the deposit. An
optimum surface finishing of the as-deposited coating, typically obtained using sand
blasting or mechanical machining, may have a very important effect on the coating
7.2 Coating Deposition Techniques for Gas Turbine Engines 391

Fig. 7.2 Schematic of two flame spraying systems: (a) wire and (b) powder fed. (Redrawn from
Hocking et al. 1989)

performances and durability. Of course, each deposition technique has specific

features, important to be considered for the selection of the most suitable for each
coating system.

7.2.1.1 Flame Spraying

It is regarded as a relatively cheap technique (Fig. 7.2). The torch is operated in air
and employs an oxy-acetylene combustion gas mixture, capable to reach tempera-
tures in excess of 2500
C and flame velocities of 100 m/s or so. Methane and
hydrogen can alternatively be used; the choice is made on the basis of the features of
the material to be deposited and of the target properties of the coating (Hermanek
2001).
The feedstock material can be in the form of powders, rods, or wires. It is
important to note that in case of rod and wire sources (Fig. 7.2a), the material is
completely melted by the flame. This is not necessarily the case for powder particles
(Fig. 7.2b) that are fed into the flame and can melt down only if they reach the hot
part of the plume for a sufficiently long time. The main features of the flame spraying
process are:
• Coatings are characterized by a relatively high porosity.
• Complex geometries can be easily coated, but not narrow channels, which cannot
effectively be accessed by the material flow.
• Deposition can only involve materials that do not decompose during the melting
or, in general, heating up.
• High deposition rate.
• Poor adhesion of the coating on the substrate.
• Relatively low cost of process and equipment.
• Low-temperature rise of the substrate.
392 7 Coatings

Fig. 7.3 Schematic of the HVOF system, based on the Jet Kote gun. (Redrawn from Thorpe and
Richter 1992)

The main systems for which flame spraying is used are metal-ceramic (CERMET)
mixed coatings and wear-resistant coatings, based on carbide phases, e.g., WC and
Cr3C2, embedded into a metallic matrix, for which Co-based alloys have been a
widespread and common choice, so far.

7.2.1.2 High-Velocity Oxy-fuel Flame Spraying

HVOF can be regarded as a more energetic version of the flame spraying process, as
can be inferred also from the denomination. A higher velocity of the powder
particles can be reached since the combustion is triggered inside a pressurized
chamber (Fig. 7.3). As a consequence of this improved combustion, HVOF super-
sonic speeds can be reached, even as high as 2000 m/s, whereas the particle velocity
in flame spraying is typically in the 100–200 m/s range and anyway never above the
speed of sound. To sustain combustion, the HVOFs are fed with elevated flow rates
of both fuel and oxygen, which are typically one order of magnitude higher than in
flame spraying equipment. Thanks to the elevated velocity of the powder particles,
high-density coatings can be obtained. As an alternative to powder, HVOF can also
have a wire source that is progressively melted down by the hot gas flow, and the
droplets are carried away towards the substrate.

7.2.1.3 Detonation Gun (D-Gun) and the Derived Super-Detonation

Gun (Super D-Gun)

Both are based on controlled, alternate explosions of a mixture of fuel, generally

acetylene, and oxygen, mixed with the powder of the coating material. This material
7.2 Coating Deposition Techniques for Gas Turbine Engines 393

Fig. 7.4 Schematic of a detonation gun apparatus. (Redrawn from Hocking et al. 1989)

charge is thereby melted and ejected from the D-Gun barrel pointed toward the
substrate (Fig. 7.4) (Hermanek 2001).
After each detonation, sparked by a plug located close to the bottom of the barrel,
the inner chamber is purged with a nitrogen gas flux, to eliminate any combustion
residue. A new filling of combustion gases and powders is prepared next, ready for
another shot. The sequence continues with a frequency of 4–8 explosions per
second, with particle velocities up to 3000 m/s, until the deposition is fully accom-
plished. Thanks to the high velocity of the material flux, extremely high compaction,
good adhesion, and elevated densities, i.e., low residual porosity, are typical features
of the coatings obtained with this technique. Other interesting features of the D-Gun
products are good mechanical and tribological properties and presence of internal
compressive stresses due to a sort of shot pinning effect induced by the particles
arriving and hitting the already deposited layers. An appropriate setting of the
process parameters allows for an optimum tuning of the temperature and deposition
rate, in order to avoid overheating and unwanted phase transformations to occur
within the coating.

7.2.1.4 Electric Arc Spraying

It is a suitable deposition technique to obtain electrically conductive coatings. The

source features two wires, also acting as consumable electrodes, between which an
arc is struck to melt the coating material (Fig. 7.5). The molten material reaches the
substrate carried by a flux either of compressed air or, in case of particularly reactive
alloys, inert gas. The gas velocities and consequently the flux of molten particles can
reach comparable velocities as those of a flame spray torch, i.e., 100–200 m/s.
Therefore, the deposited coatings will generally feature a relatively high porosity
and may contain oxide grains, although the complete melting of the feeding
394 7 Coatings

Fig. 7.5 Main parts of an electric arc spraying technique (schematic). (Redrawn from Hocking
et al. 1989)

materials results in better adhesion and higher deposition rates than in the case of
flame spraying, notwithstanding a certain fraction of metallic droplets may solidify
during the flight. The carrier gas is mostly unaffected by the arc heating, and
consequently it does not heat up the substrate. For this reason, electric arc spraying
can be used also to deposit coatings on low melting or heat-sensitive substrate
materials, for instance, polymers.

7.2.1.5 Plasma Spraying

It is a coating technique that uses a plasma torch (Fig. 7.6) as heat source. As for the
previous technique, an arc is also present in plasma spraying. The arc is not meant to
melt directly the material to be deposited but to ionize a plasma gas, in which the
particles of the coating material are injected, melted, and carried to the workpiece.
Several plasma gas mixtures are used: Ar, Ar–H2, Ar–He, Ar–He–H2, N2, and N2–
H2, just to mention the most common ones. In the design and selection of these
mixtures, two main parameters are considered: a primary heavy gas (Ar or N2)
carrier, to assure an adequate flow and particle entrainment, and a secondary gas, like
H2 and He, which is generally introduced to improve the heat transfer. In the case of
metallic powders, the above gas selections are also very effective in preventing
oxidation. At 1 atm pressure, a stable plasma torch is attained for temperatures in
excess of 7500
C approximately. Under particular operating conditions, tempera-
tures as high as 20,000
C can be achieved, so that the technique is potentially
capable to melt any kind of powder, once it is injected into the plasma-flame dart.
This allows different types of coatings to be deposited: metallic, particularly wear-
and corrosion-resistant coatings, ceramic coatings, cermets, and thermal barrier
coatings.
7.2 Coating Deposition Techniques for Gas Turbine Engines 395

Fig. 7.6 Main parts of a plasma torch used in plasma spraying coating deposition. (Redrawn from
Reed 2006)

Plasma spraying coatings are used in many technological fields: furnaces for
thermal treatments, petrochemical plants, biomedical applications, ground-based gas
turbine plants, and aero-engines (Clarke et al. 2012; Driver 2012; Heimann 2006).
Thermal barrier coatings (TBCs), originally developed for increasing the ignition
temperature of diesel engines of heavy duty vehicles, have been paramount for
improving the efficiency of terrestrial and aircraft gas turbine engines (see Sect.
7.3.2).
Depending on the deposition environment, plasma spraying (PS) process can be
classified as atmospheric (or air) PS (APS) and vacuum PS (VPS), alternatively
defined as low-pressure PS (LPPS). APS is by far the most used technique. Depo-
sition is carried out in air at atmospheric pressure inside a chamber, mainly meant to
reduce the environmental noise and to contain the dispersion of the airborne pow-
ders. The main reasons why powder particles may not reach the target or not get
embedded into the forming coating can be either the unwanted escape from the
plasma dart or their bouncing off after hitting the surface of the substrate target.
According to the torch conditions, for any specific powder, the deposition rate can be
optimized in order to control and improve the features of the coating. In general, the
high melting efficiency that is achieved with the plasma torches coupled with a
relatively high particle velocity that can reach the supersonic limit of 300 m/s results
in the formation of coatings that are relatively dense and well adherent to the
substrate. However, notwithstanding the inert or slightly reducing plasma gas, the
presence of spurious oxides in a metallic coating deposited via APS cannot be fully
avoided for the interaction of the molten alloy powders with the air in the atmo-
sphere. Even to prevent this effect, the VPS has been implemented. In this case, the
whole process occurs inside a sealed chamber connected to a vacuum system,
typically a mechanical roughing pump. Once the limiting pressure has been
achieved, the deposition starts. Alternatively, the vacuum chamber is backfilled
396 7 Coatings

with a low pressure (50–250 mbar) of inert gas, i.e., argon, which still assures cleaner
and more controlled conditions than in APS. Thanks to the reduced gas pressure
used in VPS, the plasma flame can be longer, and higher particle velocity
(500–800 m/s) and higher deposition efficiency can be attained. This latter aspect
is favored also by a reduced fraction of losses in the sprayed powder. The possibility
of controlling the processing atmosphere is a prerequisite to implement, if available,
additional operating PS modes, like the plasma-transferred arc (PTA) process and
the atmosphere and temperature controlled (ATC) plasma spraying process
(Firdaouss et al. 2017; Wilden et al. 2006).
Vacuum and controlled atmosphere plasma spraying are characterized by some
important advantages with respect to APS:
• Possibility to deposit refractory and hard metals. These materials would otherwise
tend to oxidize, and vacuum avoids or, at least, limits this effect.
• In VPS, including PTA, the substrate may be heated before starting the deposi-
tion, with limited risks of oxidizing it. This operation is useful both for removing
any impurity from the substrate surface and to induce compressive stress fields
into the coating.
On the other hand, VPS is more expensive, owing to the higher costs of the
spraying equipment and running expenses, considering that each deposition requires
high volumes of inert gas. Another important restriction to the use of VPS is the
limited possibilities to move the component inside the vacuum chamber, although
these latter aspects have been profitably managed with the introduction of rotating
stages and robotic arms, used to move the plasma torch on the sample surface in a
controlled way.

7.2.2 Physical Vapor Deposition

Quite a large number of deposition techniques are based on physical methods to

produce the vapor phase from which the deposit, thin film, coating, and multilayer
systems, will form (Bose 2007). The condensation of the coatings from the vapor
phase requires an apparatus in which it is possible to adjust and control the working
atmosphere, as concerns its pressure and composition, during the process. This in
general requires a vacuum system and connections to pure gas pipelines. In this way,
all needed procedures to remove unwanted gaseous impurities can be accomplished
in order to control better all the reactions accompanying the deposition. In the field of
coatings for gas turbine applications, electron beam physical vapor deposition
(EB-PVD) is certainly the most important technique. EB-PVD derives from the
electron beam melting technology (see also Sect. 4.4). The use of EB-PVD for large-
scale productions began in the early 1980s, especially thanks to the improvements
attained in electron guns, relevant performances, advances in the control, and
vacuum systems. In the EB-PVD apparatus, focused high-energy electron beams,
generated by several electron guns, are used to evaporate the ingots made of the
coating material, and also for preheating the substrate, e.g., turbine blades, inside the
7.2 Coating Deposition Techniques for Gas Turbine Engines 397

Fig. 7.7 Scheme of an EB-PVD plant with details of the vapor source and component movements
during deposition. (Redrawn from Wolfe et al. 2005)

vacuum chamber. This step is important to promote the formation of the so-called
thermally grown oxide (TGO, see Sect. 7.3.2) on the surface of the bond coat already
present on the blades, thanks to a controlled amount of pure oxygen added to the
working atmosphere. In the setup in Fig. 7.7, six electron guns are present (45 kW
each). More powerful units, with gun power in excess of 100 kW, are also available,
affording high deposition rates, in the range of 5–10 μm/min.
Although completely different as concerns the working principles from thermal
deposition still, also in EB-PVD the coating material reaches the substrate following
straight trajectories from the source. Therefore, the component needs to be rotated
and moved appropriately with respect to the vapor source (Fig. 7.7), in order to
obtain a homogeneous coverage, in terms of thickness, microstructure, and
composition.
Early experimental works were conducted to deposit overlay coatings (Sect. 7.3.1),
with a composition based on the FeCrAlY system with 10–15% aluminum and 125 μm
thickness (Talboom and Grafwallner 1970). Although not particularly successful, for the
occurrence of too fast interdiffusion reactions at the interface with the Ni-based superalloy
substrate, the results of this research demonstrated the feasibility of using protective
coatings having a composition independent from the substrate. Compositions more
suitable to protect Ni- and Co-based superalloys against oxidation and hot corrosion
were successively developed. Nickel, cobalt, or a combination of the two became the
major constituents, together with Cr, Al, and Y, this latter as reactive element. Other
reactive elements, added in concentrations usually not exceeding 1 wt% approx., are Si,
Hf, Zr, and Ta (see Table 7.2). They improve the adhesion of the surface oxide scale,
particularly under cyclic oxidation conditions (Gupta and Duvall 1984). The control of the
concentration of these elements is crucial, and in this respect the limits of EB-PVD
emerged. In fact, a constant quantity of reactive elements in the vapor phase during the
398 7 Coatings

Table 7.2 Compositions of some overlay coatings (Nicoll et al. 1986; Reed 2006)
Overlay coating
type Ni Co Cr Al Y Ti Si Hf Others
NiCrAlY Bal. – 25 6 0.4 – – – –
NiCrAlY Bal. – 22 10 1.0 – – – –
NiCrAlY Bal. – 31 11 0.6 – – – –
NiCrAlY Bal. – 35 6 0.5 – – – –
CoNiCrAlY 32 Bal. 21 8 0.5 – – – –
CoCrAlY – Bal. 25 14 0.5 – – – –
NiCoCrAlTaY Bal. 23 20 8.5 0.6 – – – Ta ¼ 4
NiCoCrAlYSi Bal. 0– 12.5–20 2–8 0–0.25 0– 2– – Fe ¼ 0–
40 10 10 20
Mn ¼ 0–
5
Mo ¼ 0–
5
Nb ¼ 0–
4
NiCrAlTi Bal. – 30–40 1–10 – 1–5 – – –
NiCoCrAlHf Bal. 0– 10–40 6–25 – – – 0– –
40 10

deposition run is not easy to maintain, particularly with a sufficient degree of reproduc-
ibility, owing to the comparatively low vapor pressure of these elements. This critical
aspect is still there and limits the possibility of using EB-PVD for overlay coatings (Gupta
and Duvall 1984). The contemporary development of vacuum plasma spraying (VPS)
technology established this technique as preferred for the deposition of overlay coatings.
Considering that the compositions of overlay coatings and, thereby, the technological
aspects related to their deposition are rather similar to those of bond coats in TBCs, the
VPS approach became the standard choice for them too. Still concerning TBCs, EB-PVD
introduced a major breakthrough in the deposition of the ceramic, zirconia-based top coat
(Duvall 1981; Anderson and Sheffler 1983). Once similar reproducibility issues, like those
faced for MCrAlY coatings, were settled, the columnar grain morphology of the zirconia
layers and the associated, remarkable improvement in their durability established TBCs
deposited using EB-PVD as the best choice for the protection of vanes and blades of the
gas turbine high-pressure stages (Sect. 7.3.2) (Lammermann and Kienel 1991; Meier et al.
1990; Strangman 1987).

7.2.3 Chemical Vapor Deposition

Still vapor, but originated from chemical reactions, is used in the so-called chemical
vapor deposition (CVD) methods. In the field of gas turbine aero-engines, most of
7.2 Coating Deposition Techniques for Gas Turbine Engines 399

the coatings obtained from CVD techniques are diffusion coatings, interesting for
the protection of gas turbine parts against corrosion at high temperature and bond
coats in TBCs. Indeed, this specific application triggered a renewed interest for
diffusion coatings that, on the other hand, as anti-corrosion barriers, have largely
been substituted by overlay coatings (Strangman 1996).
Two main kinds of CVD techniques are used in gas turbine applications: pack-
cementation and out-of-contact processes (Goward 1998). Actually, the early use of
diffusion coatings via pack cementation was not intended for gas turbines but for
enhancing the oxidation resistance of heating elements made of iron wires, copper
steam condenser tubes for power plants, steel furnace fixtures, and nickel combus-
tion screens. In all cases, it was clear that the improved oxidation resistance resulted
from the preferential formation of a protective alumina scale. On the basis of this
know-how, the main technological aspects of the CVD processes have been
extended to the gas turbine engines, including those for aerospace applications.
The main components of the pack mixture used for this process are the source of
the metal to be deposited, e.g., aluminum powder; the so-called activator compo-
nents, e.g., NaF, NaCl, and NH4Cl; and an inert filler, like alumina powder, whose
main role is to control the kinetics of the reaction and avoid sintering of the metal
source.
The main reactions, occurring during the aluminizing of base alloy components,
can be summarized as follows: ammonium chloride (or one of the alternatives) acts
as a reaction activator gas, and its decomposition determines the availability of
hydrochloric acid that reacts with aluminum to form aluminum chloride, AlCl3,
according to reactions described by Eq. 7.1:

3
3HCl þ Al ! AlCl3 þ H 2 ð7:1Þ
2

Aluminum chloride vapor condensates at temperatures of 600–650
C on the

surface of the alloy components, which becomes enriched with aluminum, whereas
HCl is released as exhaust, and it is eliminated from the reaction chamber (Eq. 7.2):

AlCl3 þ Ni ! Ni  Al þ HCl ð7:2Þ

The phase composition and the microstructure of the coating that will form on the
alloy depend on the interdiffusion kinetics and, consequently, on the aluminum
activity. For high Al activities, the Ni2Al3 intermetallic compound would preferen-
tially form, with a prevailing inward diffusion of aluminum. This regime is usually
identified as low-temperature high activity (LTHA), to indicate also the typical
temperature range for the diffusion treatment, carried out at 870
C for 20 h or
so. Alternative to this is the high-temperature low activity regime (HTLA, see also
Sect. 7.3.1), during which significantly higher temperatures are employed. The
diffusion treatment can be conducted at 1100
C for 1 h, followed by an intermediate
temperature step at 875
C for 16 h (Reed 2006). In this case, the prevailing
formation of NiAl is observed. In Sect. 7.3.1, the specific features of the two kinds
400 7 Coatings

of diffusion coatings are discussed with regard to their oxidation protection proper-
ties. In the present context, it is interesting to note that the compositions of diffusion
coatings are strictly dependent on the composition of the base alloy. Specific coating
systems have been developed for other than nickel-based superalloy. For instance,
diffusion coatings based on the FeAl and CoAl intermetallic phases are used for the
protection of Fe- and Co-based superalloys, respectively (Naji et al. 2015; Pint et al.
2001; Priest and Zhang 2015). Typical thicknesses of aluminide coatings range from
30 to 100 μm, selected on the basis of the composition, component to be protected,
relevant service conditions, and expected lifetime.
The development of diffusion coatings for hot parts of gas turbine aero-engines,
like inner walls of the combustors, vanes, and blades, was pursued since the early
times. In fact, the first aircraft components using diffusion coatings were the inner
walls of the combustion chambers made of low-alloy steel and, most probably, also
the blades, made of the same steel, of the Messerschmitt ME 262 fighter aircraft that
started its service in 1944 (Goward 1998). Pack-cementation aluminizing was used
to protect gas turbine vanes made of cobalt-based superalloy (see Sect. 6.2) in 1957
for the first time. The process was subsequently extended to turbine blades made of
nickel-based superalloys (see Sect. 6.4). Actually, in the early reported examples, no
pack-cementation but hot-dip (Seelig and Steuber 1978) and slurry-fusion (Joseph
1960) processes were used. Their working principles have some affinity to pack
cementation, considering that the surface metal enrichment occurs by dipping the
components into hot molten metal and into a slurry containing metal powders and
halides suspended in an organic binder, which is subsequently evaporated away.
Diffusional thermal treatments are carried out afterward.
Starting from the early 1970s, the standard choice became aluminizing vanes and
blades using pack-cementation processes (Gauje and Morbioli 1983; Parzukowski
1977; Warnes and Punola 1997). With the development of hollow blades, a limit of
the technique emerged. Although aluminum is deposited onto the alloy surface
through the condensation of gaseous AlCl3, for the presence of the pack, this
condensation and, thereby, aluminum surface enrichment occur preferentially at
the contact region between the pack and the component, so that the inner parts and
cooling channels of the blades remain substantially unprotected, since no aluminum
surface enrichment takes place during the aluminizing treatment (Smith et al. 1999;
Squillace et al. 1999). To tackle this problem, several approaches were considered:
slurry aluminizing, vacuum pulse aluminizing, and finally forced flow gas phase
aluminizing, which can be regarded as a precursor of the out-of-contact processes
(Galmiche 1975; Restall and Wood 1986; Warnes and Punola 1997). The reactors
illustrated in Fig. 7.8 rely on two slightly different working principles, as concerns
the way in which aluminum is added to form the active gas phase, AlCl3. In Fig. 7.8a
AlCl3 develops internally to the reactor, from the reaction of the gas phase with the
source material. In the alternative approach illustrated by Fig. 7.8b (“above pack”),
AlCl3 is formed outside the reactor and can be introduced into the reaction chamber
under more controlled flux rate and temperature. Surface aluminum enrichment and
7.3 Main Coating Systems in Gas Turbine Engines 401

Fig. 7.8 Scheme of two alternative out-of-contact aluminizing reactors: (a) the vapor flux is
enriched in aluminum inside the reaction chamber for the reduction of alumina powder and (b)
the gas phase entering into the reactor chamber contains the aluminum precursor already. (Redrawn
from Reed 2006)

subsequent diffusion into the alloy are the main steps to produce aluminide-based
coatings capable to improve the high-temperature corrosion resistance of the under-
lying alloy. Several secondary additions have been considered in order to enhance
the protective performances of the coating and its durability. As discussed in details
in Sect. 7.3.1, these compositional changes have been pursued using different
approaches. Controlled amounts of alloying elements, like silicon and chromium,
have been added to the vapor source. In case of platinum-containing or platinum
aluminide coatings, featuring remarkably long lifetimes, the aluminizing CVD step
is preceded by the deposition of a few micrometer thick platinum layers, which take
part to the diffusion reactions. Other coating systems have been developed using
different diffusing elements. CVD surface treatments fully based on silicon and
chromium have also been explored (Goward 1998). Although out of the scope of the
present textbook, alternative applications, like coal gasification and future generation
nuclear power plants, emerged as interesting for these coating systems.

7.3 Main Coating Systems in Gas Turbine Engines

After having considered the main functional aspects of coating systems in gas
turbine engines, both for aircraft and terrestrial applications (Sect. 7.1), the main
deposition techniques have been outlined in Sect. 7.2. In this last section of Chap. 7,
the main coating systems in gas turbine aero-engines are considered.
402 7 Coatings

7.3.1 Coatings Resistant to High-Temperature Corrosion

A complete discussion on the main phenomenological aspects concerning high-

temperature corrosion, i.e., oxidation and hot corrosion, can be found in Chap. 8.
A general approach adopted to improve the resistance of gas turbine components to
these attacks relies on promoting the formation of a dense, adherent, and continuous
protective layer of native oxides and other secondary reaction phases, like nitrides.
These layers may form on the base alloys directly or on the protective coatings
present on the alloy surface. Diffusion and overlay coatings are the main systems
used for improving the high-temperature corrosion resistance of aero-engine parts,
with compositions targeted at obtaining the formation of an oxide scale rich in
alumina, in particular the α-Al2O3 polymorph. Diffusion coatings (Sect. 7.2.3) for
nickel-based superalloys are based on two nickel aluminide phases, either δ-Ni2Al3
or β-NiAl.
Both phases are rich in aluminum, with concentrations significantly higher than
those of nickel-based superalloys. Therefore, they can provide an adequate outward
aluminum flux for the formation of protective layer rich in α-Al2O3. Moreover, the
ordered crystalline structures of these intermetallic phases (see Sect. 9.4) slow down
all diffusive phenomena involved with oxidation, so to reduce the relevant reaction
kinetics. The Al-Ni phase diagram in Fig. 7.9 shows further important aspects,
explaining the reason for the success of nickel aluminide diffusion coatings, with
particular regard to the β-NiAl phase. This intermetallic phase is named “perma-
nently ordered” since it retains an elevated degree of long-range order (see Sect. 9.4)
up to the melting point, with no solid-state order-disorder transformation that in this
case would be a second order one. The higher stability temperature and the wider
compositional range make β-NiAl definitely preferable to δ-Ni2Al3. Even the larger
aluminum content in the δ-Ni2Al3, that can be regarded as a plus in some respects,
on the other hand, determines higher interdiffusion kinetics with the underlying
nickel-based superalloy, having a different composition and, particularly, a signifi-
cantly lower aluminum concentration than the coating itself. Together with the
compositional changes resulting from the formation of the outer oxide scale, the
coating-substrate interdiffusion is indeed another major lifetime limiting factor for
the coating. Therefore, β-NiAl can be regarded in several respects as the preferable
choice as concerns nickel-based superalloys. Similarly, for alloy systems based on
iron, e.g., steels and iron-based superalloys, and cobalt, e.g., cobalt-based superal-
loys, overlay coating based on the FeAl and CoAl intermetallic phases have been
developed. Again, the selection of iron and cobalt aluminides is suggested by their
compositional affinity to the relevant substrate alloys (Lee and Kuo 2005; Pint and
Zhang 2011; Ren et al. 2014; Zhang et al. 2004, 2007).
As concerns β-NiAl-based coatings, compositional changes have been introduced
to enhance the coating resistance to high-temperature corrosion, also under cyclic
oxidation conditions, which better represents real service requirements. Significant
improvements in corrosion resistance performances have been achieved in IN100
and AISI 304 austenitic stainless steel, using mixed aluminum and chromium
7.3 Main Coating Systems in Gas Turbine Engines 403

Fig. 7.9 Binary Al-Ni phase diagram

diffusion coatings (Pérez et al. 1999a, b). Chromium additions can also be attained
by preparing the superalloy surface with a chromizing treatment preceding alumi-
nizing (Luthra and LeBlanc 1987; Mévrel et al. 1986). Thus, the interdiffusion
treatment, needed to promote the formation of the coating, involves both elements,
resulting in a mixed composition coating. The advantage of having both aluminum
and chromium as oxidation active elements is based on the reduction of the mini-
mum concentration of aluminum necessary to the formation of a protective outer
alumina layer. This same approach is exploited even more effectively in MCrAlY
overlay coatings (vide infra). Chromium has also been considered in aluminide
coatings in association with silicon (Nicholls 2003). Silicon can be co-deposited
with aluminum, using a mixed slurry method, involving the application onto the
component surface of a uniform layer of a slurry containing both elements. The
coating forms afterward, during the diffusion heat treatment, carried out in an inert or
slightly reducing atmosphere. In this regard, the Sermaloy™ coatings should be
mentioned, like Sermaloy J and Sermaloy 1515. Their excellent behavior can be
ascribed to the co-presence of silicon and chromium, this latter coming from the base
alloy, inducing the formation of the CrSi2 intermetallic. The microstructural refine-
ment is very important too. In the Sermaloy 1515 coating, the alternate surface layers
of NiAl and CrSi2 are both acting as reservoirs of oxidation and corrosion active
elements, i.e., Cr and Si, and thus they guarantee a continuous outward flux of these
404 7 Coatings

elements to form a protective surface oxide scale (Berry et al. 1995; Van Roode and
Hsu 1989). Although these coatings have been successfully applied also to single-
crystal superalloy blades, chromium- and silicon-alloyed coatings have primarily
been investigated and are still used for improving the aluminide coating protection
performances, when corrosive agents are present in the working environments
(Meelu et al. 1992; Prater et al. 1981). This is the case of land-based and marine
turbines, burning not very refined and pure fuels, particularly as concerns the sulfur
content (Luthra and LeBlanc 1987). Silicon-alloyed coatings, featuring up to 10 wt%
of this element, provide particularly good hot corrosion, both Type I and Type II,
resistance (see Sect. 8.2.5) (Berry et al. 1995; Van Roode and Hsu 1989). The most
successful and widely used addition to nickel aluminide coatings is platinum, a noble
metal whose application to protect engine components is justified by their high
added value (Allam et al. 1979; Felten 1976; Felten and Pettit 1976; Fountain
et al. 1976). This element improves the oxidation resistance of nickel aluminide
coatings, influencing positively some of their critical aspects. In the first place,
platinum additions result in the reduction of the coating-alloy substrate interdiffu-
sion, due to the different compositions of nickel aluminide coatings and superalloy
substrates. A strong driving force is there for an inward diffusion of aluminum,
which may be rather fast at elevated service temperatures. For the same reason, the
outward diffusion of minority alloying elements, like titanium, niobium, tantalum,
etc., is slowed down too. The consequent phase and compositional changes experi-
enced by the coating determine eventually the loss of its protective action. Moreover,
the progressive depletion of aluminum, so that precipitates of the γ’-Ni3Al and of the
martensitic polymorph of β-NiAl start to form, renders the coating particularly
brittle. Concerning the mechanical stability of the coating, platinum alloying turned
out to reduce also the formation of voids from vacancy coalescence at the interface
with the alloy substrate. This feature turns out to be particularly important to resist
better oxidation under cyclic conditions, when the interplay of mechanical stresses
and stress concentrator defects may result in a premature failure, i.e., cracking and
detachment, of the coating. The diffusion control operated by the platinum additions
has also a positive effect on the quality of the oxide scale, that results to be made of
particularly pure, compact, and adherent α-Al2O3, the most preferred oxide phase in
this field. Platinum alloying is obtained starting with electroplating the engine
components with a layer 5–10 μm thick of this element. Alternatively, the platinum
layer can be deposited using EB-PVD, a choice to be made in case a particular
control on composition and microstructure is required. A diffusion heat treatment,
performed under inert atmosphere, e.g., argon, for several hours at about 1000
C is
conducted before the CVD aluminizing process. Depending on the overall deposi-
tion procedure, three main kinds of coatings have been reported: single platinum
aluminide phase (generally PtAl2) coating; two-phase coating, featuring both β-NiAl
and PtAl2; and single-phase coating, i.e., (Pt,Ni)Al, with platinum atoms substitu-
tional in the β-NiAl lattice (Bose 2007; Smith and Boone 1990). The development of
advanced aluminide diffusion coatings resulted in a significant increase in the
component lifetime. These coatings received a particular boost to meet the require-
ments posed by the development of novel single-crystal nickel-based superalloys
7.3 Main Coating Systems in Gas Turbine Engines 405

Fig. 7.10 Isothermal weight evolution of several aluminide coatings oxidized in air at 1200
C.
(Redrawn from Pint et al. 1998)

(see Sect. 6.5), oxide dispersion strengthened (ODS, see Sect. 9.3) alloys, and
relevant rise of the turbine entry temperature (TET, see Sects. 2.2 and 2.5) (Giggins
and Pettit 1979; Goward 1998; Nicholls 2000; Pint 1997). With particular regard to
advanced superalloys, the role of protective coatings became critical, once it became
clear that the novel compositions, mainly for the reduction of chromium content and
overestimated beneficial effects of refractory elements, like rhenium, exhibited a
lower intrinsic resistance to high-temperature corrosion (Pint et al. 2006).
Several compositions were explored to optimize all contributions to the coating
durability. For instance, as concerns a better adhesion of the alumina scale, Fig. 7.10
shows the evolution of the parabolic kinetic constant (see Sect. 8.2.1, Wagner’s
model) for the oxidation at 1200
C in air of several NiAl-Pt diffusion coatings with
different additions of Zr, Hf, and Si (Pint et al. 1998). Incidentally, the beneficial
effects on the oxidation resistance of the coatings, of reactive elements, like hafnium
and zirconium, are similar to the effects that these same and other reactive elements
introduce in MCrAlY overlay coatings (see Sect. 8.2.2).
An alternative approach to control the degradation kinetics of diffusion coatings
is based on the reduction of their interdiffusion with the underlying alloy. The
negative effect of this phenomenon depends not only on the aluminum depletion
in the coating but also on the possibility that the compositional changes occurring in
the so-called secondary reaction zone (SRZ), forming in the alloy subsurface
regions, may result in the formation of TCP phases (see Sect. 6.4.2). In this respect,
the positive influence of other alloying elements, like ruthenium, added in a similar
way as platinum to the NiAl coatings, can be cited (Tyron et al. 2006, 2007; Wang
et al. 2011). Major improvements of the coating lifetime have been attained thanks to
406 7 Coatings

the slowing down of the diffusive phenomena occurring in the SRZ: inward diffu-
sion of aluminum from the coating and outward diffusion of refractory elements, like
tungsten, molybdenum, and tantalum, from the substrate.
The research efforts made to improve the durability of the diffusion coatings,
through the optimization of their critical parameters, have also been exploited for the
development of high-performing bond coats in TBCs (see Sect. 7.3.2), based on
these very same systems. Nonetheless, diffusion coatings have been progressively,
although not completely, replaced by the MCrAlX overlay coatings, as protective
coatings against high-temperature corrosion phenomena (Sect. 8.2). One of the main
reasons for the success of the overlay coatings can be traced to the freedom of
designing their composition independently from the composition of the base alloy.
This is particularly interesting for a better control of the optimal concentration of
reactive elements, like yttrium, hafnium, zirconium, etc., whose central role in
enhancing the corrosion protection capabilities is discussed in the following. Dif-
ferently from diffusion coatings, based on the intrinsically brittle ordered nickel
aluminide phases (β-NiAl and δ-Ni2Al3), overlay coatings exhibit their ductile to
brittle transition at temperatures (DBTT) below the operational range of the gas
turbine engines. This is shown by Fig. 7.11, in which the dependence on temperature
of the strain for crack initiation of several diffusion and overlay coatings is shown
(Hocking et al. 1989). An alternative approach, based on the measure of the room-
and high-temperature hardness of the coating, using especially designed apparatus,
provides similar results (Dryepondt and Pint 2010).
The brittleness and the low fracture toughness of the aluminide coatings represent
a potential factor of risk for the whole component. In fact, cracks that may form on
the coating surface as a consequence of the thermal mismatch with the substrate or

Fig. 7.11 Ductile to brittle transition temperature (DBTT) for diffusion coatings, based on the NiAl
and Ni2Al3 intermetallics and MCrAlY overlay coatings. (Redrawn from Hocking et al. 1989)
7.3 Main Coating Systems in Gas Turbine Engines 407

erosion impact with particulate matter carried by the hot gas flow may propagate to
the underlying superalloy. To reduce this possibility, in addition to improving the
composition and microstructure of the coating to enhance its fracture toughness,
another approach was to reduce the thickness and, thereby, the maximum length of
the cracks that may form. In this way, the probability to have a spontaneous
propagation to the substrate was automatically reduced.
An alternative approach to overcome the limits of the intrinsically brittle β-NiAl
intermetallic is its substitution with tougher systems. This is precisely the route
followed in the development of the MCrAlX overlay coatings, described in the
following. As a sort of bridging step between diffusion and overlay coatings, it is
interesting to mention an explorative study conducted on two-phase diffusion
coatings based on the γ and γ’ austenitic phases (Pint 2004). Although largely
replaced by overlay coatings, diffusion coatings are still having an important role
as bond coats for TBCs (Sect. 7.3.2). For this application, aluminide coatings are not
directly exposed to the hot gases flux, since protected by the ceramic top coat.
Moreover, still thanks to the presence of the top coat, lower temperatures and
temperature gradients are there in the aluminide coatings, thus reducing the relevant
thermomechanical stresses and the associated risk of crack formation. The situation
is different for overlay coatings that, according to the Fig. 7.11, are reasonably more
ductile and, therefore, tougher in-service conditions. The general composition of the
overlay coatings can be expressed as MCrAlX, where “M” can be either nickel,
cobalt, iron, or a combination of two of these, selected on the basis of the specific
applications. “X” refers to the already introduced reactive elements, like Y, Hf, Zr,
Si, and Ta, present in concentrations typically not exceeding 1 wt%. Reactive
elements optimize the corrosion protection action of the coating through the
improvement of the character and adhesion of the surface oxide scale. In Sect.
8.2.2, regarding the main high-temperature corrosion mechanisms, the way in
which reactive elements may improve the surface protection will be illustrated. It
will be also made clear why the presence of both aluminum and chromium is
recommended, although alumina is still the preferred oxide in order to have a fully
protective scale (See Sect. 8.2.2). In Table 7.2 the compositions of some commercial
overlay coatings are listed (Nicoll et al. 1986; Reed 2006). Compositions are
designed so to have austenitic γ-γ’ matrix alloy, in which β-NiAl grains are dispersed
and act as reservoirs for the aluminum needed to sustain the outward diffusive flux to
form and subsequently mend, in case of spallation and detachment, the protective
oxide scale.
The preferred deposition system for these coatings is plasma spraying (see Sect.
7.2.1), although, for standard applications, HVOF can still provide excellent results.
Both thermal spraying processes are suitable for controlling better the concentrations
of minority elements, i.e., the reactive elements, which have a fundamental role in
determining the actual protective properties of the coating. In this regard, Fig. 7.12
shows the progressive degradation behavior, involving several damage stages,
observed in overlay coatings as a function of the exposure time to a burner rig (see
Sect. 8.2.6) cyclic oxidation conditions at a temperature of 1149
C. Together with a
reference composition, NiCoCrAlY, deposited using both EB-PVD and VPS,
408 7 Coatings

Fig. 7.12 Comparative degradation behavior of NiCoCrAlY, deposited with EB-PVD and vacuum
plasma spraying (VPS in the figure). (Redrawn from Gupta and Duvall 1984)

different additions of Si and/or Hf were explored, using VPS deposition only (Gupta
and Duvall 1984). As concerns the reference samples (i.e., NiCoCrAlY), the better
performances of the plasma-sprayed coating are determined by its optimized micro-
structure with respect to EB-PVD sample. Further improvements have been
obtained, thanks to the ameliorated compositions, those involving Si and/or Hf, of
the plasma-sprayed coatings. This study had also demonstrated that among the
different plasma spray modes, i.e., air, inert gas, and vacuum, this latter is able to
produce coatings with a better microstructure and compositional homogeneity, so
that the relevant trends observed in Fig. 7.12 are fully justified.
In Chap. 8 a general overview on the reaction products forming on the surface of
overlay coatings will be presented (Sect. 8.2.2), depending on their compositions
and testing conditions. The efficiency of overlay coatings as protective barriers
against high-temperature corrosion phenomena has been optimized further through
the development of functionally gradient systems (SMARTCOAT design). Layers
of different compositions have been deposited, according to a design suggested by
the different operating conditions, pertaining not only aircraft gas turbines but also
industrial and marine land-based plants (Nicholls et al. 2002). For these latter
applications, while peak temperatures are lower than in aero-engines (see
Table 7.1), still harsh working conditions may be determined by lower purity fuels
and critical environments, like marine and polluted industrial atmospheres. There-
fore, beyond the SMARTCOAT approach, the compositions of the overlay coatings
have been tuned to better respond to specific requirements, including the combined
action of gas phase and salt deposit corrosion phenomena, like oxidation, Type I and
Type II hot corrosion (see Sect. 8.2.5) (Goward 1998; Rajendran 2012). Thus, higher
chromium concentrations, i.e., 25–40%, for situations involving Type II corrosion
7.3 Main Coating Systems in Gas Turbine Engines 409

and lower, 12–20%, for Type I, are recommended (Mevrel 1989). At higher tem-
peratures, in excess of 900
C, where chromium oxide starts to be unstable, it is
advisable to increase the concentration of aluminum to favor the formation of the
protective α-Al2O3. A good choice is 18–22% Cr and 8–12%Al (Nicholls et al.
2002). The corresponding concentrations of nickel and cobalt have to be selected
accordingly. Nickel guarantees a higher ductility and toughness of the coating, while
cobalt controls Type I hot corrosion. As a general rule, MCrAlX overlay coatings are
not suitable for long-term exposures at temperatures in excess of 1100
C, since a too
fast oxidation kinetics leads to a too rapid growth of the surface oxide layers. The
stresses building up inside the oxide scale, associated with those due to thermal
cycling, would result in a local detachment of the scale itself and the consequent loss
of protection by the oxide layer. High-temperature conditions are normally present in
new gas turbine aero-engines and have suggested using overlay and diffusion
coatings as bond coats in thermal barriers coating (TBCs) systems. In TBCs, the
ceramic top coat reduces the temperatures of the metallic coating and of the
components down to acceptable values for a sufficient lifetime, as discussed in the
next section (Sect. 7.3.2).

7.3.2 Thermal Barrier Coatings

Thermal barrier coatings have had a central role in the evolution of the turbine entry
temperature (see Sects. 2.2 and 2.5). In fact, they are meant to thermally insulate the
substrate, allowing higher service temperatures to be reached, thus increasing the
engine efficiency (see Sect. 2.2), without increasing excessively the kinetics of
thermally induced transformations that might otherwise shorten the useful lifetime
of the structural components.
In Fig. 7.13 the main components of a TBC are shown (Padture et al. 2002).
Although deposited onto a gas turbine blade, the structure is quite general and
comprises, from the inner to the outer part, the bond coat that can be either an
MCrAlX overlay coating or a diffusion coating, based on intermetallic aluminide
phases; the thermally grown oxide (TGO) layer that results from the oxidation of the
bond coat; and the ceramic top coat, very often made of a ZrO2-based ceramic. Since
the TGO layer should provide an important protection against further oxidation and
corrosion of the bond coat, elevated standard of adhesion and compactness is crucial.
This may recommend a controlled oxidation treatment prior to installation of the
coated component (Wolfe et al. 2005).
An additional advantage of having an oxide layer already present on the surface
of the bond coat is the reduction of the stresses in the top coat, produced by
in-service oxidation. This is a central issue as concerns the TBC failure mechanisms
(Sect. 7.3.3) and justifies the research efforts in developing better performing bond
coats (Hazell et al. 2008). The combination of TBCs and internal cooling of
superalloy blades has allowed gas turbine engines to operate even at higher than
the incipient melting temperatures of the superalloy itself.
410 7 Coatings

Fig. 7.13 Structure of a TBC deposited on a gas turbine blade. (Redrawn from Padture et al. 2002)

This aspect is clearly shown by Fig. 7.14, and TBCs can be regarded as one of the
most important innovation factors in this field, considering the enormous impact on
the performances and efficiency of the aero-engines. TBCs were first developed by
NACA (National Advisory Committee for Aeronautics) and by NBS (National
Bureau of Standards) (Garrett and Gyorgak 1953; Harrison 1947). The early
reported aerospace applications of TBCs are probably the zirconia-calcia stabilized
flame-sprayed coating deposited onto the walls of the thrust chamber of the X-15
rocket plane in the 1960s (Davies 1963; Hjelm and Bornhorst 1961). Initially, TBCs
were used on static parts of the engine only, like combustors and afterburners, not
subjected to excessively high mechanical stresses. For the deposition of TBCs,
atmospheric plasma spray (APS) is used for both the NiCrAlY bond coat and the
top coat, for which yttria-stabilized zirconia (YSZ) was identified as an excellent
choice (Rigney et al. 1997). The comprehension of the main failure mechanisms and
the relevant remediation approaches afforded the possibility of using these ceramic
coatings also for selected regions of the high-pressure gas turbine. Starting from the
1980s, TBCs were used to protect the nozzles in the gas turbine, still using plasma
spraying for the deposition. However, more demanding requirements regarding the
oxidation resistance performances of the bond coat suggested the use of a VPS
(or LPPS, see Sect. 7.2.1) deposition system. A major breakthrough in the coating
production technology came up with the EB-PVD, which started to be used for the
deposition of the top coats. The columnar microstructure of the zirconia grains,
7.3 Main Coating Systems in Gas Turbine Engines 411

Fig. 7.14 Evolution of the materials (superalloys and TBCs) and gas temperatures in a gas turbine
engine. (Redrawn from Clarke et al. 2012)

visible in Fig. 7.13, increased considerably the effective lifetime of the coating.
Therefore, these coatings started to be used for the protection of turbine vanes and
blades and still are, even though good progresses have been made in the plasma-
sprayed counterpart. Nowadays, APS is still the main technique used to deposit the
top ceramic part of TBCs on stationary components of the engine, like combustors,
shrouds, vanes, and also blade baseplates. For the bond coat, either diffusion or
plasma-sprayed coatings are used. In this case, vacuum plasma spraying is preferred,
for a better control of the impurity concentrations. For the top coat of the rotating
parts and, in general, of the hot section of the turbine, EB-PVD has become the
standard choice. The particular coating deposition dynamics induce a typical colum-
nar grain morphology and a much smoother surface finish than plasma spraying.
This feature is paramount, particularly for turbine vanes and blades of the first high-
pressure stage, in relation to their aerodynamic efficiency and, consequently, the
efficiency of the whole engine. This aspect is shown by Fig. 7.15, which also
demonstrates how the dependence of the aerodynamic efficiency of the turbine
vanes and blades on their surface finish is less critical for the second and further
stages, being exposed to hot gases with lower pressure and velocity. The same can be
applied to the inner walls of combustors and, if present, afterburners, exposed
tangentially to the gas flow. On these bases, the best combination of coatings and
deposition technologies can be selected for each part, to meet both performances and
412 7 Coatings

Fig. 7.15 Dependence of the aerodynamic efficiency of the first and second stage turbine blades
and vanes on their surface finish. (Redrawn from Rigney et al. 1997)

cost requirements. When depositing protective coatings onto the surface of gas
turbine blades and vanes, it is important to avoid that the deposited layer would
obstruct the cooling holes, with obvious detrimental effects on the component
lifetime and integrity. Also in this regard, EB-PVD coatings are preferable, since
the parameters of the atom-by-atom physical deposition process can be controlled so
to obtain a smoothly tapered coating at the hole entrance. The same cannot be
achieved with plasma spraying, considering that the coating builds up through
discrete single droplets, each one of which can significantly reduce the hole’s
opening, having diameters varying from 0.25 to 0.5 mm approximately.
Important features for the top coat of a TBC are:
• Chemical stability and corrosion resistance
• Erosion resistance
• Good adhesion to the substrate
• Resistance to high- and low-frequency fatigue
• Thermal expansion coefficient similar to that of the substrate in order to reduce
the thermal mismatch stress value
• Thermal stability over all the temperature range
In addition to these properties, low thermal conductivity is obviously paramount,
considering the expected main role of TBCs. In Fig. 7.16 the thermal conductivity
curves, as a function of the temperature, of several ceramics, mainly oxides, are
shown. ZrO2 seems to be an excellent candidate for TBC, considering its lowest
7.3 Main Coating Systems in Gas Turbine Engines 413

Fig. 7.16 Thermal

conductivity of several
ceramic compounds.
(Redrawn from Clarke
2003; Kingery 1955)

value of thermal conductivity, substantially temperature-independent over a wide

range, which is not the case for other candidate materials. However, the presence of
three ZrO2 polymorphs requires suitable alloying for phase stabilization. Pure
zirconium oxide, at 1 atm, can indeed exist in one of the following phases, depending
on the temperature:
• Monoclinic up to 1170
C on heating
• Tetragonal from 1170 to 2370
C on heating
• Cubic from 2370
C up to melting (2715
C)
All the transformations involved are martensitic and are characterized by impor-
tant volume changes. For instance, the displacive tetragonal to monoclinic phase
transformation is accompanied by a molar volume increase of 3–5% and up to 8%
shear strain (Sato et al. 1985). The temperature changes and fluctuations under
typical gas turbine servicing conditions may trigger these transformations with
unavoidable consequences on the mechanical stability of the TBC. The situation is
rendered even more complex because of the dependence of the transition tempera-
tures on the microstructural parameters of the oxide samples, like defect structures
and grain size that depend also on the coating deposition conditions (Garvie and
Goss 1986; Tu et al. 1994). This is the reason why zirconia-based materials, with
selected concentrations of phase stabilizer oxides, like Y2O3, MgO, CaO, are used in
real coatings. The addition of the above oxides is capable to change the stability
fields of the zirconia polymorphs, so to render the top coat more suitable to operate
over the whole range of servicing conditions. Y2O3 has been identified as an
excellent candidate for improving the zirconia coating performances, as concerns
particularly the resistance to cyclic heating and cooling stages. This is shown by
414 7 Coatings

Fig. 7.17 Burner rig tests on TBC. The best performances, i.e., the largest number of cycles to
failure, are achieved by the coating with composition: ZrO2-7Y2O3. (Redrawn from Stecura 1986)

Fig. 7.17 (Stecura 1986). Data points are from burner rig tests (see Sect. 8.2.6) and
illustrate the dependence of a zirconia coating lifetime on its yttria content. In
particular, a concentration of Y2O3 equal to 7 wt% turns out to be the best choice,
and, indeed, the ZrO2-7Y2O3 composition became the standard for these applica-
tions. The reason can be inferred from the portion of the ZrO2-Y2O3 phase diagram
in Fig. 7.18, in which the findings of a former investigation on this same system have
been applied to the definition of the phase composition of plasma-sprayed zirconia
coatings. For Y2O3 concentrations in the 6–11 wt%, quenching from the high-
temperature cubic ZrO2 field results in the formation of the metastable “non-trans-
formable” tetragonal ZrO2 (Miller et al. 1981; Scott 1975). The peculiar aspect of
this tetragonal polymorph of ZrO2, as obtained by EB-PVD and identified as t’, is its
microstructure, with two main features relevant to the observed improved coating
durability. A <100> crystallographic texture means that the tetragonal c-axis is
parallel to the metal-ceramic interface. Moreover, a diffuse intergranular porosity
is present. Since the coating grains have a typical columnar morphology, also the
pores feature an asymmetric elongated shape, aligned along a direction perpendic-
ular to the metal-ceramic interface (Alperine and Lelait 1994; Lelait et al. 1993).
The combination of these two microstructural and crystallographic features
determines the remarkable improvement in the coating lifetime for two main
reasons:
• Cracks propagating parallel to the ceramic-metal interface are stopped by the
intrinsic coating porosity, through the twofold action of crack blunting and crack
deflection along directions perpendicular to crack propagation.
7.3 Main Coating Systems in Gas Turbine Engines 415

Fig. 7.18 Portion of the phase diagram showing the composition range for the formation of the t’
tetragonal phase at room temperature. (Redrawn from Miller et al. 1981)

• Cracks propagating parallel to the ceramic-metal interface would very unlikely

meet easy cleavage planes for the tetragonal structure that are perpendicular to the
<100> direction and, thereby, also perpendicular to the mentioned interface. In
this respect, not only crack propagation is hindered by the lack of easy cleavage
planes, but an additional toughening contribution is provided by crack deflection
along perpendicular directions to the top-bond coat interface.
It is evident from this description that the excellent thermal cycling resistance of
the EB-PVD zirconia top coats cannot be attained with plasma-sprayed TBCs. The
molten zirconia droplets are splatted either onto the substrate or on previously
deposited ceramic layer, where they rapidly solidify, without any possibility of
controlling microstructure and crystallographic texture.
The zirconia, in which the tetragonal t’ phase is stabilized by alloying with the
optimal 7 wt% concentration of yttria, is usually code-named 7YSZ. As seen, the
elevated performances and reliability required in TBCs by the extreme working
416 7 Coatings

conditions result from the combination of different aspects, concerning the compo-
sition, microstructure, and crystallographic texture. An important factor, discussed in
the next section on the failure mechanisms (Sect. 7.3.3), is the coupling of the top
with the bond coat through the TGO layer. A further aspect contributes to the
intrinsic fracture toughness of EB-PVD ceramic top coat; the microstructure of the
zirconia layer exhibits very fine twinning structures that are extremely effective in
accommodating the thermomechanical strains. This behavior, which has also been
reported for zirconia coatings stabilized with other oxides, for instance, HfO2, is
based on the strain compliance through the reversible re-orientation of the tetragonal
variants, with a consequent release of the local stresses that in this way are kept
below critical values (Ibegazene et al. 1993; Mercer et al. 2007). This mechanism is
quite similar to the alignment of variants observed in shape-memory alloys (see Sect.
9.5.1), resulting in the partial or complete de-twinning of the alloy when in the
martensitic phase or in stress-induced austenite (superelastic behavior). In TBCs, the
mechanism is active also at high temperatures, so that high fracture toughness is
guaranteed even at the more demanding operating conditions. It is worth saying that
this mechanism, even named ferroelastic, is different from transformation toughen-
ing that is active in partially stabilized zirconia. This latter mechanism involves the
irreversible stress-induced transformation of zirconia from tetragonal into mono-
clinic phase, and the toughening effect is associated with the relevant volume
expansion that hinders the propagation of the crack (Green et al. 1989), thus
providing a further contribution to the fracture toughness of PSZ. In this system, a
fine distribution of crystalline domains also of the other two ZrO2 polymorphs, i.e.,
cubic and monoclinic, is present (Garvie et al. 1972). The irreversibility of the
martensitic transformation on which this toughening mechanism is based has
suggested the definition of “ceramic steels” for these zirconia-based materials
(Garvie et al. 1975).
The thickness of the top coat is primarily selected considering the component to
be protected. As concerns the blades, the coating thickness is a trade-off between
achieving the target temperature drop in the base metal and avoiding an excessive
weight increase, due to the contribution from the coating. The typical thickness of
the top coat ranges from 100 to 250 μm. On the inner walls of the combustors, the
shrouds, the high-pressure vanes, being all stationary components, and thicker
coatings are acceptable, and indeed thicknesses ranging from 500 μm to 1 mm are
typically used.
Notwithstanding the excellent performances of 7YSZ-based TBCs, new systems
are being explored in order to sustain the continuous development of higher effi-
ciency gas turbine engines, in the first place dependent on the maximum operating
temperatures. Considering the primary scope of TBCs, thermal conductivity is to be
regarded as a valuable selection parameter (Winter and Clarke 2007). An early
development route was aimed at overcoming the limitations of the 7YSZ material,
particularly when used at temperatures in excess of 1200
C. At high temperatures,
the t’ tetragonal phase tends to separate into the equilibrium tetragonal (t) and cubic
(c) phases. Subsequently, the tetragonal phase, under the effect of thermal cycling,
will transform into the monoclinic low-temperature polymorph. At this stage, the
7.3 Main Coating Systems in Gas Turbine Engines 417

integrity of the coating may be highly jeopardized. The time required to detect the
early signs of the monoclinic phase is relatively short, i.e., in the 100 h range at
T ¼ 1350
C (Lughi and Clarke 2005). In order to slow down this process,
alternatives to yttria stabilizers have been considered. Heavier dopants have been
selected, so to hinder phonon propagation and reduce the intrinsic thermal conduc-
tivity of the coating. Several rare earth oxides like Gd2O3, Yb2O3, and Nd2O3 have
been considered in this respect. The approach was only partially successful though.
As expected, the larger molecular masses of the dopants, in association with the
elevated density of constitutional point defects, mostly oxygen vacancy, resulted in
lower thermal conductivities. However, the novel oxide systems displayed even
lower stability of the t’ phase, as compared to the reference 7YSZ system, with the
only exception of Yb2O3. The situation has been improved to some extent using a
combination of different dopants in the same material. For instance, to the reference
ZrO2-Y2O3 system, with the trivalent yttrium, a given concentration of pentavalent
cation oxides, like Nb2O5 or Ta2O5, were added (Raghavan et al. 2001; Raghavan
et al. 2004). The complementary valence of the additions reduced the concentration
of constitutional point defects, particularly oxygen vacancies. As expected, the
resulting structures featured a lower thermal conductivity and a good level of
stability of the t’ phase. The trend was taken to the limit for ceramic oxide compo-
sitions in which the trivalent and pentavalent ions concentrations are substantially
the same. This determines the nominal zeroing of the constitutional oxygen vacan-
cies, resulting in a significantly reduced thermal conductivity, as compared to the
reference value for 7YSZ. Moreover, the t’ phase demonstrates an extraordinary
stability (Shen et al. 2010). The combination of multiple dopants has been adopted
also to improve the performances of 7YSZ, with the additions of Nd2O5 (or Gd2O5)
and Yb2O3 or Sc2O3 (Jarligo et al. 2010; Song et al. 2011; Vassen et al. 2000, 2010;
Wuench and Eberman 2000; Zhu and Miller 2004). Interestingly, in these systems,
owing to the elevated concentrations of high molar weight dopants, nanometric
oxide clusters tend to form, having the additional effects of stabilizing the micro-
structure of the coating against recrystallization and sintering, phenomena typically
occurring during servicing and resulting in a progressive reduction of the thermal
insulating capabilities of the coating, for the elimination of the initial porosity.
Molecular dynamic simulations applied to the investigation of the zirconia-based
systems revealed the defect structures that appear at increasing temperatures, resem-
bling those observed in amorphous, rather than crystalline structures, with their
important effects on the thermal conductivity of the ceramic oxide. These studies
indicated promising directions to be taken in order to develop more efficient TBCs
(Allen et al. 1999; Pan et al. 2012). It is clear that increasingly complex crystalline
structures would involve lower thermal conductivities, owing to the more difficult
propagation of the acoustic phonons, a paramount aspect in heat transfer phenom-
ena. Moreover, the higher concentrations of optical phonons encountered in low
symmetry, complex crystalline structures determine an indirect reduction of thermal
conductivity, owing to the reduced contribution of these phonons to heat transfer.
The thermal conductivity curves, as a function of the temperatures, in Fig. 7.19a, b,
illustrate the possible strategies for the development of ceramic oxides for TBC
418 7 Coatings

Fig. 7.19 New ceramics for TBCs: dependence of their thermal conductivities on temperature, also
in comparison with the thermal conductivity of 7YSZ. (Redrawn from Pan et al. 2012)

applications. Following the direction marked by molecular dynamic investigations,

several families of superlattice structures have been identified and demonstrated to
have a range of similar vibrational modes, as seen with highly dense packed crystal
lattice and capable of extensive solid solutioning (Cao et al. 2004; Guo et al. 2009;
Qu et al. 2011; Shen et al. 2008; Sparks et al. 2010; Yang et al. 2008). This is an
important prerequisite for tuning the coating composition in order to respond better
to the requirements of specific applications. The main groups of ceramics that are
being investigated are perovskite-structured oxides (ABO3 type), lanthanide ortho-
phosphates (LnPO4), and rare earth silicates, like the apatite structured
Gd8Ca2(SiO4)6O, this latter exhibiting a particularly low thermal conductivity,
substantially independent from temperature.
One of the systems that is providing already very promising results, in compar-
ison to the standard 7YSZ coating composition, is based on lanthanides (reference
composition Ln2Zr2O7, with Ln ¼ La, Nd, Sm, Gd). Compounds, like Gd2Zr2O7,
turn out to be stable up to 1500
C, i.e., 300
C more than 7YSZ. A comparative
investigation (see Fig. 7.20) conducted also with field tests demonstrated a definitely
better durability and at the same time an excellent reliability of the novel coatings,
ready to replace the classical systems (Levi et al. 2012).
7.3 Main Coating Systems in Gas Turbine Engines 419

Fig. 7.20 Comparative

engine test showing the
different behaviors of
conventional 7YSZ and
novel Gd2Zr1xO2x/2
(x ¼ 0.5) pyrochlore
structure TBCs, after
simultaneous exposure to
identical working conditions
(Levi et al. 2012,
reproduced with permission
of the Publisher)

7.3.3 Failure Mechanisms of Corrosion Resistant

and Thermal Barrier Coatings

Diffusion and overlay coatings, when not used as bond coats in TBCs, are directly
exposed to the flux of hot gases that through chemical reactions and erosive
phenomena may produce their progressive disruption. The ordinary sequence of
events involves the initial formation of a protective surface oxide layer, compact, and
adherent to the alloy of which either diffusion or overlay coatings are made. In the
case of aluminide coatings (see Sect. 7.3.1), the surface oxide is essentially com-
posed of α-Al2O3. This most stable oxide prevents the metallic substrate to get in
contact with the corrosive atmosphere, so that a stationary state can be reached. In
overlay, i.e., MCrAlX, coatings, the so-called synergic effect (see also Sect. 8.2.2) of
chromium and aluminum is exploited. Thanks to an optimized design of the coating
compositions, the formation of nickel oxide and other spinel oxide phases, still
containing nickel, is significantly inhibited, except for the very early stages of
oxidation. The microstructure of these phases prevents them to provide any effective
protection to the underlying alloy. The occasional damage of the surface oxide layer
may occur for the impact of particles carried by the high-velocity gas flux. Another
reason for the local cracking and possible partial detachment of the oxide scale is the
building up of internal stresses during the oxide growth process. These stresses are
usually produced by the formation of the reaction products within the already
existing oxide layer, a process that results from the prevailing outward diffusion of
the metallic cations that can in this way react with the incoming oxygen (see Sect.
8.2.1). Thermal cycling too influences the stability of the surface oxide layer.
Therefore, major improvements concerning the oxidation-resistant coatings have
especially been focused on developing strategies to enhance their durability also
under cyclic conditions. Reactive elements, like yttrium, zirconium, etc., present in
the overlay coatings, provide several contributions for a longer lifetime (see also
Sect. 8.2.2). Through their reaction with oxygen, a fine dispersion of oxide
nanoparticles forms on the coating surface. These particles act as heterogeneous
420 7 Coatings

nucleation sites and induce a fine-grained microstructure into the growing oxide
layer. This is regarded as a positive feature to resist better to thermomechanical
stresses, since grain boundary sliding becomes feasible, thus reducing the incidence
of cracking and spallation. As oxidation proceeds, reactive elements tend to react
preferentially with the oxygen flux diffusing through the oxide layer toward the
metallic substrate. This mechanism avoids an excessive internal oxidation of the
alloy. Another beneficial effect of this same process is the formation of an interme-
diate oxide barrier, which not only limits inner oxidation but also the outward
diffusion of metallic cations from the underlying alloy. This function is quite similar
to the barrier effect operated by the platinum additions in aluminide coatings (Das
2013; Tawancy et al. 1991). The oxide particles of the reactive elements that form at
the interface between the outer scale and the coating result in a better adhesion of the
oxide scale, thanks to the pegging effect called in this way since it exploits also a
mechanical grip component. The interface bonding is strengthened by the reduction
of sulfur impurities that are captured by RE atoms to form relevant sulfide species.
The high concentration of interfaces, deriving from all these secondary phases,
promotes the capture of vacancies created by the Kirkendall effect as a consequence
of the outer diffusion of oxide-forming elements (Khanna 2002). When this process
cannot occur any longer for the saturation of the available sites, vacancy clustering
becomes predominant, determined by their diffusion-assisted coalescence and lead-
ing to the formation of macroscopic voids within the bulk alloy. This phenomenon is
occurring also in aluminide coatings, and it is typically observed in the final stage of
the degradation process.
To summarize, the life span of oxidation-resistant coatings, i.e., diffusion and
overlay coatings, can be divided into the following main phases:
• Initial state. The surface oxide scale starts to form and initiates its protective
action of the underlying coating, thus inhibiting further oxidation. This stage may
have an initial transient when the gaseous phases are in direct contact with the
metallic components, so that reactions can occur directly, i.e., not assisted by
diffusion mass transport. From the kinetic point of view, this corresponds to a
linear dependence on time of the specific mass increase associated with the oxide
growth. At this stage, the oxide scale exhibits also non-protective phases, e.g.,
NiO or NiAl2O4.
• Stationary state. A steady-state condition is reached when a protective oxide
layer has formed on the metallic coating surface and the coating composition is
still such, that in case of scale damage, a sufficient outgoing flux of metal is
mending it with the formation of a further layer of protective oxide. This produces
anyway compositional changes in the coating, leading to the formation of the
so-called Al-depleted zone. At the same time, the interdiffusion between the
coating and the underlying alloy results in the formation of the already mentioned
Kirkendall voids and intermetallic phases. The formation of these brittle phases is
promoted by the increased local concentrations of slow-diffusing atomic species,
like refractory elements, present in the superalloys. At this stage, the oxidation
kinetics is usually well represented by a parabolic law, according to the Wagner
oxidation model (see Sect. 8.2.1).
7.3 Main Coating Systems in Gas Turbine Engines 421

• Damage propagation. When the concentration of aluminum in the overlay or

diffusion coatings falls below the critical value, necessary to the formation of an
outer protective scale based on α-Al2O3, other non-protective oxides start to form
on the coating surface. This stage can be regarded as the end of the coating useful
lifetime. The composition of the aluminide diffusion coatings may determine the
formation of a martensite phase, exhibiting volume changes upon thermal
cycling, with consequent stress generation in the coating itself (Zhang et al.
2003). This situation may be also responsible for the so-called rumpling of the
coating. “Rumpling” can be regarded as a mechanical plastic instability resulting
in the formation of a corrugated surface of the coating (Deb et al. 1987). In
addition to the martensitic transformation occurring during thermal cycling,
another drive to this phenomenon comes from the β to γ’ transformation, occur-
ring in the aluminum-depleted coating, concurrently to the martensitic transfor-
mation of the β-phase. These transformations are not the only reasons for
rumpling that is observed also in overlay coatings. The phenomenon is still
attracting research efforts aiming at understanding the relevant critical aspects
in order to develop coatings less prone to this kind of degradation (Chen et al.
2017; Jorgensen et al. 2016; Lesnikova and Lesnikov 1986; Smialek 1971).
The phenomenology described so far with regard to the failure and degradation
mechanisms of diffusion and overlay coatings is interesting also as concerns the use
of these same coating systems as bond coats in TBCs (Sect. 7.3.2). In TBCs, the
surface directly exposed to the hot gas atmosphere is that of the ceramic top coat.
Since the ceramic layer is permeable to air, oxidation takes place (Sect. 7.3.2) with
the formation of the TGO layer at the bond-top coat interface. Although the kinetics
of oxidation occurring at this location is mitigated by the lower gas pressure and
temperature, still, it has very important consequences on the stability and durability
of the whole TBC. TBCs, particularly those protecting vanes and blades of gas
turbine high-pressure stages, require an absolutely reliable estimation of their actual
lifetime, since an extensive and unexpected failure would result in catastrophic
effects for the structural integrity of the components, considering that gas operating
temperatures can be far above the material capabilities (Fig. 7.14). High-temperature
spikes may result in accelerated kinetics of diffusion-assisted phenomena and, in the
most severe cases, in local melting.
The layered structure of a TBC poses the important issue of the mechanical
matching and accommodation of the strain misfit and relevant residual stresses present
at each one of the different interfaces: alloy-bond coat and bond-top coat. This second
interface actually features the presence of the intermediate layer of the TGO. Consid-
ering the compositional affinity and the common metallic character, the alloy-bond
coat interface is mainly influenced by the strains generated by interdiffusion and
relevant phase and microstructural transformations. The formation of Kirkendall
voids and consequent swelling, precipitation of intermetallic phases, outward diffusion
of metallic atoms from the alloy substrate (Chen M et al. 2003; Tolpygo and Clarke
2004a, b), are all phenomena intervening to a different extent in the failure of TBCs,
which has to be rather regarded as the result of several concurrent causes, very much
dependent on the specific systems and working conditions involved.
422 7 Coatings

Fig. 7.21 Failure mechanisms in TBCs deposited via EB-PVD: a comprehensive overview (Levi
et al. 2012, reproduced with permission of the Publisher)

Notwithstanding this complexity, still, the main mechanisms for TBC failure can
be classified in two broad categories: intrinsic and extrinsic (Fig. 7.21) (Evans et al.
2008; Levi et al. 2012; Wu et al. 2017). This classification is very useful also for
designing reliable lab tests, particularly as regards the intrinsic mechanisms, and, in
general, to spot the reasons for observed failures in serviced components. The
intrinsic failure mechanisms can be identified by analyzing the morphology of the
detached coating. The detachment exposes the metallic surface of the bond coat,
showing, as a general feature, the imprint of the TGO grain boundaries, scattered
patches of alumina, still from the TGO layer, and some limited zirconia fragments
from the top coat. The spallation has been determined by the weakening of the metal
(bond coat)-TGO interface, along which crack propagates preferentially. In corre-
spondence of heterogeneities and surface protrusions, the fracture may also cut
through the TGO, and this is the reason for the limited zirconia residues observed
on the detached scale (Xu et al. 2004). No metallic surface of the bond coat is
exposed when the detachment of the TBC is determined by the rumpling of the TGO,
with the involvement of the top coat, and producing the delamination through cracks
advancing parallel to the bond-top coat interface (Davis and Evans 2006; Ruud et al.
2001; Tolpygo and Clarke 2004a, b). The detachment mechanisms of the TGO layer
that have been presented herewith are fully compliant with the results achieved in a
research on NiAl-Hf bond coats modified with additions of Pt and Pd (Jackson et al.
2014). The different compositions resulted in different adhesion strengths of the
TGO to the bond coat. In the case of comparatively weaker adhesion, the failure of
the TBC, tested under cyclic conditions, starts at the TGO-bond coat interface. If a
good grip is realized by the improved bond coat composition, the TBC failure would
be this time triggered by a diffuse rumpling of the bond coat, although occurring at a
later stage than the former mechanism. According to the above description, the
cracks initiate and propagate within the top coat and at the TGO-top coat interface.
Other compositional changes of β-NiAl-based bond coats, using carbon and yttrium,
7.3 Main Coating Systems in Gas Turbine Engines 423

together with the mentioned hafnium, have been investigated in the successful
attempt to improve their durability (Tolpygo et al. 2008).
Longer times and/or higher temperature exposures may result in the formation of
Kirkendall voids. Their growth can proceed, until their average size becomes so big
to determine the detachment of the coating. The fracture surface, propagating within
the thickness of the bond coat, would exhibit a characteristic void morphology
(Mumm et al. 2001). Before achieving the final lifetime stage, a significant thickness
increase, up to 10–20%, of the bond coat can be observed. It is due to the combined
effect of void formation and compositional changes consequent to superalloy-bond
coat interdiffusion. A net increase in nickel concentration in the coating is the
consequence of the higher diffusivity of nickel, from the alloy substrate toward the
bond coat, than that of aluminum in the opposite direction (Shankar and Seigle
1978). The Kirkendall voids may be further enlarged by the cavitation induced by
external stress and pressure acting on the component surface. This aspect can be
listed among the extrinsic failure mechanisms. In general, they are due to the
interaction of the TBCs with damaging agents depending on the external environ-
ment, like particles impacting on the surface of the top coat, producing erosion and
foreign object damage (FOD). The effects on the coating integrity depend on their
size and mass and their velocity and impinging angle (Bruce 1998; Chen X et al.
2003; Nicholls et al. 1999). Erosion produces the progressive thinning of the top
coat, particularly in the regions close to the leading edges of rotary blades. Detach-
ment of the top coat fragments, micro-crack formation within the top coat, and
plastic deformation of the bond coat in regions just beneath the impact spot are the
main effects of impacting particles.
An important damage mechanism of TBCs is due to the penetration into the top
coat of liquid deposits of calcium-magnesium-aluminosilicate (CMAS) (Kakuda
et al. 2015; Jackson et al. 2015; Poerschke et al. 2016). This contaminant is
originated by the ingestion into the engine of ash, dust, and sands present in the
outer atmosphere (Borom et al. 1996; Krämer et al. 2008; Mercer et al. 2005). The
deposition of CMAS occurs preferentially along the pressure surface of the airfoils
and on the shrouds, where indeed this kind of damage is more frequently observed.
For the difficulty of reproducing the phenomenology involved with CMAS attack
using laboratory tests, the main studies and relevant remediation strategies have been
developed mostly using data from diagnostics of really serviced components. This
kind of damage turns out to be rather important in aircrafts flying at low altitudes
over marine and desert regions. In the case of marine atmosphere, sulfide contam-
ination has been often observed (Pokluda and Kianicová 2010). The molten CMAS
deposit that forms at temperatures above its melting point, i.e., 1240
C approx.,
penetrates into the ceramic top coat preferentially through the grain boundaries, with
an important contribution by the capillary forces acting on the inner surface of the
columnar pores. The solidification, upon cooling, determines a densification of the
CMAS-penetrated parts of the top coat, with a consequent increase in the effective
elastic modulus and reduction in fracture toughness and damage tolerance. Coher-
ently with this picture, also the reduction in the porosity of the ceramic coating,
deriving from long-term exposure to high temperatures, provides a further contribu-
tion to increasing the stiffness of the coating itself. A complex stress field is
424 7 Coatings

generated inside the top coat, for the mismatch of the thermal expansion coefficients
induced by the top coat densification, its contamination by CMAS material, occur-
ring inside the pores, in the intergranular regions, and on the free surface of the
coating. In the long run, the stress field in the top coat may determine the nucleation
of micro-cracks and its partial detachment.
The selective dissolution of the zirconia stabilizers in the top coat may be induced
by the surface contamination with a layer of molten sulfate, according to the
principles of the high-temperature hot corrosion in salt deposit (see Sect. 8.2.5),
and may result in the destabilization of the t’- ZrO2 phase and its transformation into
the low-temperature monoclinic polymorph. The relevant volume change generates
stresses that, if sufficiently intense, may induce the delamination of the ceramic
coating.
To complete the picture, a quick glance at a terrestrial application: the land-based
gas turbine plants fired by coal-derived synthesis gas (i.e., syngas) or hydrogen. In
these engines, the content of water vapor is definitely higher than in conventional gas
turbines, and therefore its effect on the durability of the TBC becomes an issue. The
TGO observed on single-phase β-(Ni,Pt)Al and two-phase γ-γ’ bond coats, oxidation
tested under cyclic conditions, exhibited higher oxidation kinetics when several
percentages (10, 50, 90%) of water vapor were added to the initially dry oxygen
atmosphere (Unocic and Pint 2013; Wright and Gibbons 2007). Reduced lifetimes,
with respect to the “dry” values, were observed, although with large fluctuations for
the different test conditions. The best durability was guaranteed by the TBC with the
two-phase (γ-γ’) bond coat, most likely for their more stable structure, with no
reported coating rumpling, as observed instead in the samples with single-phase
β-(Ni,Pt)Al bond coat, after exposure to the same testing conditions. These results
confirmed the findings of a former study with a similar system, tested using the same
experimental protocol (Pint et al. 2011).

7.3.4 Clearance Coatings

The design of clearance coating systems stems from two opposite requirements: to
avoid reciprocal rubbing between rotary and stationary components and to attain a
sufficiently small level of mechanical clearance capable to ensure an adequate
efficiency and thrust to the engine. For these reasons, a clearance is left beyond
the tip of the blades and vanes, so that under conventional steady working condi-
tions, they should not come into contact with the case. However, there are several
reasons that may change the reciprocal rotor-stator distance, resulting in unwanted
rubbing and collisions: differential thermal expansion; sudden changes in the engine
working conditions; particular loading associated with specific operational requests,
like during hard landing and surge; rapid shutdowns due to sudden fuel starvation;
and high acceleration maneuvering. Other important aspects that may influence the
correct clearance alignment are casing out-of-roundness, bearing damage, mechan-
ical vibrations due to instabilities, and changes in the gas flow. On the other hand,
having a tight sealing clearance has beneficial effects on the engine efficiency and,
7.3 Main Coating Systems in Gas Turbine Engines 425

thereby, on fuel and other running costs. As an example, it has been demonstrated
that 25 μm reduction in the clearance of the high-pressure turbine blades results in a
reduction by 0.1% in the specific fuel consumption (SFC, see Sects. 2.2 and 2.6)
(Lattime and Steinetz 2004). Mechanical clearance is an issue not only in the gas
turbine but in the compressors too. Here, a good control of clearance results in a
lower number of stages necessary to achieve a target value of the compression ratio
(see Sect. 2.2), with a consequent reduction in the total weight of the engine.
The loss in engine specific power and thrust may be determined by a non-optimal
management of the mechanical clearances, which leads to higher fuel consumption
in order to achieve the desired engine performance. However, the lower efficiency of
the thermal cycle is accompanied by a higher heat generation with immediate
consequences on the exhaust gas temperatures. The kinetics of degradation and
microstructural evolution of the hot structural components is enhanced, and their
expected lifetime reduced. In case temperature certified limits do exist for aero-
engines parts, and these are overcome, a complete, costly overhaul may be required
to comply with the relevant international safety regulations. Thus, there are several
important reasons for reducing mechanical clearance and for improving sealing in
gas turbine engines, issues that over the years have been approached adopting
several strategies. Active solutions are based on the revision of the existing part
design, with the introduction of actuators and labyrinth fin seals (Dobek 1973;
Scharrer and Pelletti 1995; Van Ness et al. 2006). In this way dimensional changes
can be compensated either by varying in situ the part geometry or by limiting the
effects of gas heating through local cooling of the components. Shrouded blades are
a typical example of parts designed for the implementation of an efficient sealing.
The T-shaped tips of the blades are in contact with one another for a better stiffness
of the whole part, being the unshrouded blades used as a lighter alternative. As all the
rest of the blade body, also the shroud is cooled, and the outer part of this ring-
shaped component features one or a series of fins that cut through the seal coating
deposited onto the outer rubbing surface. The tips of the blades cut directly into the
abradable counterpart. In this regard, the choice of the materials for this tribological
system is important to guarantee both the sealing performances and adequate blade
durability. The following properties and parameters should be considered when
selecting materials for abradable systems:
• Deposition technologies that guarantee repeatable results
• Good adhesion to the substrate, in order to resist adequately against the shear
stresses by the rubbing tips of the blades or by the fins of shrouded components
• Production of small-sized wear debris that should not provoke any major erosion
damage to the parts found along the path out of the engine
• Resistance to thermal shock and thermal cycling
• Surface stability under the normal servicing conditions, as concerns both hot
corrosion and erosion resistance, in order to retain the above properties during the
whole servicing lifetime
These requirements are rather different in case of land-based with respect to aero
turbomachinery, just considering the different environments and standard operational
cycles that these latter would rather face. For instance, large land-based gas turbines on
426 7 Coatings

average will have a stop and start once every 3–5 years, whereas commercial aircraft gas
turbine engines have on average 1500 stop and start cycles per year (Markou et al. 2011).
Moreover, an important aspect of thermal cycling is the possibility for liquid and solid
layers of contaminants to condensate on the exposed surface, like sulfates and silicates,
i.e., CMAS (Sect. 7.3.3), inducing hot corrosion phenomena (Sect. 8.2.5), involving
important changes in the materials properties.
As a general rule, the abradable coating materials should have a lower hardness
than the material of the rubbing counterpart. As seen in Chaps. 3, 4, 5 and 6, the main
classes of materials for structural components in an aero-gas turbine engine are metal
laminate composites (see latest fan material), titanium alloys, steels, and superalloys.
The main wear phenomena that may intervene and whose incidence should be kept
under control are essentially related to the wearing out of the rotating components
(e.g., the blade tip or the fins on the outer rim of shrouded blades) due to abrasive or
oxidation wear. In the case of titanium alloys, overheating, in association with high
gas pressure conditions, may trigger ignition (Sect. 4.3.1.5). The frictional
overheating in association with the enthalpy output coming from the rapid oxidation
of the titanium alloy components may determine the complete loss of an acceptable
aerodynamic geometry. Wear mechanisms may also involve adhesive contributions.
In this case, material transfer from the abradable coating to the rubbing component
can be observed, the phenomenon being known as adhesive transfer or “galling.” An
excessive material transfer to the tip of the blades or to the fins can result in deep
grooving on the inner shroud wall accompanied by a temperature rise due to
frictional heating. To avoid these drawbacks, a composite material approach is
mandatory, and indeed all coating systems especially developed for improving
mechanical clearances in different parts of a gas turbine aero-engine are based on
this principle. The thermal stability requirements have been fulfilled by choosing
suitable combinations of matrix and strengthening secondary phases. Compressor
coatings are based on a polymeric matrix, usually a polyester, reinforced with Al-Si
or Al-bronze additions. The strength of the shearable polymer matrix is tuned to the
specific working conditions by the inorganic additions, in order to reduce the
frictional heating and at the same time to prevent excessive adhesive mass transfer.
Higher working temperatures render the polymeric matrix inadequate as concerns its
mechanical stability, worsened by the faster kinetics of the oxidation phenomena.
Using oxygen-acetylene flame spraying deposition (Sect. 7.2.1), nickel and nickel
alloy-based systems are adopted for higher temperature use and relevant more
demanding stability requirements. On the other hand, to avoid that these mechani-
cally stronger than polymer-based materials may damage the rotating counterparts,
the deposition parameters, the powder grain morphology, and size are selected so to
obtain relatively porous coatings. Powders used for these coatings are:
• Nickel-graphite. In this case graphite is encapsulated inside a nickel shell using
an autoclave-based process. These coatings are typically coupled with steels and
nickel superalloy blades or fins of shrouded parts. However, selecting a suitable
hardness value and playing with the coating porosity, they can also be used with
titanium alloy components.
7.3 Main Coating Systems in Gas Turbine Engines 427

• NiCrAl-bentonite. A similar microstructure as in the former system is present in

these powders. Bentonite grains are embedded into pre-alloyed Ni-Cr-Al parti-
cles. The coatings, with suitable degree of porosity and, thereby, hardness, are
used in association with steel and superalloys rubbing components. A graded
structure can be attained by changing the deposition parameters and thereby the
hardness and porosity across the deposit thickness. The aluminosilicate compo-
nent renders these coatings too abrasive for titanium alloys.
• NiCrFeAl-hBN (hexagonal boron nitride). CERMET composite coatings are
obtained from powders made of pre-alloyed NiCrFe mixed with aluminum and
boron nitride. The reaction of aluminum with the pre-alloyed Ni-Cr-Fe solid
solution produces an enthalpy output, which sustains the melting of the metallic
powders. Boron nitride (hexagonal polymorph) reduces grain boundary diffusiv-
ity, thus enhancing the oxidation resistance and promoting the incursion by the
rotating components, since it helps the cutting through of the abradable matrix
acting as a solid lubricant, with the additional beneficial effect of keeping the
temperature rise within acceptable values. This functional aspect and the overall
performances of these coatings can be optimized for the different blading mate-
rials: titanium alloys, steel, and nickel-based superalloys.
An important boost to the technology of abradable coatings came from the
introduction of APS (Sect. 7.2.1) and its flexibility in modulating the deposition
conditions and to adapt them to each specific material system. According to the
scheme proposed in Wilson (2012), the following main coating systems can be
deposited by APS. The first is obtained from powder mixtures containing one of the
following combinations of materials: aluminum-polymer, aluminum-hexagonal
boron nitride, and aluminum-bronze-polymer. Polymers are typically heat-resistant
polyesters, and the resulting coatings are by all means similar to those described
above (Al-Si in a polymer matrix). They are suitable for fan and low-pressure
compressor (LPC) stages, operating at temperatures up to 325
C. Higher tempera-
tures, up to 450
C, require more stable matrix and solid lubricants, like hBN. At this
stage the shearing of the coating matrix is ruled by the lubrication action of boron
nitride, having a similar role as graphite in the Al-Si-graphite and Ni-graphite flame-
sprayed coatings. A further group of APS coatings are those based on the MCrAlY-
hBN-polymer system, designed for higher temperatures, up to 750
C. The compo-
sition of the metallic components is similar to the composition of the overlay
coatings (see Sect. 7.3.1). In these coatings the role of the polymer is paramount in
determining the porosity, which becomes the ruling factor of the abrasion behavior.
In fact, a full density MCrAlY material would have a comparable or higher hardness
than the materials of the rubbing counterparts, and a 35–60% porosity is necessary to
render these coatings actually abradable. Therefore, a controlled amount of polymer
is deliberately entrapped into the coating material and afterward eliminated with a
post-deposition heat treatment. This process leaves behind the wanted percentage of
porosity that can even be distributed in a graded way within the coating thickness,
from the top, higher porosity values, down to the interface with the substrate (lower
porosity and therefore higher density). In these higher-temperature abradable
428 7 Coatings

coatings, the polymer component has a role in the manufacturing stage only, since it
is exploited to attain the right degree of porosity, whereas in the abradable coatings
for lower-temperature applications, a significant polymer fraction is retained also in
the coating material. This induces two different wear mechanisms which are para-
mount to ensure an effective tolerance and sealing effect. In the low-temperature
coatings, the comparatively softer polymer matrices are shear cut by the intruding
components. The wearing out of the high-temperature abradable systems is
guaranteed by a sort of friable cutting of the highly porous surface layer favored
by the presence of a solid lubricant, like hBN. This same approach is adopted for the
refractory ceramic abradable coatings used for the hot parts of the gas turbine engine,
like the high-pressure stage of the turbine, where temperatures as high as 1200
C
can be reached. Ceramic coatings are necessary, and the obvious candidate is ZrO2,
already providing excellent performances in TBCs (see Sect. 7.3.2). In fact, yttria
partially stabilized zirconia (PSZ) is used in abradable coatings, adopting the
microstructural features recommended for these applications (Zhao et al. 2012).
However, specific compositions have also been developed, since the PSZ materials
are not fully compliant with the specific requirements recommended for these
applications. A significant porosity is still interesting, to have reasonably abradable
coatings that should be stable against sintering under the extreme temperature
conditions, up to 1200
C that they may face in service. In addition to that, ceramic
abradable coatings require thicknesses up to 2 mm, which are definitely higher than
the typical thicknesses of TBCs, usually 50–200 μm. This enhances the spallation
probability, particularly under cyclic thermal conditions, so that adequate fracture
toughness and thermal shock resistance are paramount (Vassen and Stöver 2007). To
tackle these critical aspects, alternative stabilizer oxides have been considered, and
important improvements have been achieved with dysprosia (Dy2O3) and, although
to a lower extent, also with ytterbia (Yb2O3) (Dorfman et al. 2008; Oerlikon Metco
2014; Schmid 1997; Sporer et al. 2011). Both of them have introduced significant
improvements in the high-temperature durability, with respect to the YSZ coatings.
Similarly to TBCs, zirconia-based abradable coatings are deposited not on the bare
alloy surface but onto a bond coat, necessary for protecting the underlying compo-
nents by an excessive oxidation and corrosion. Nickel aluminide coatings or overlay
coatings are used in this regard. In Table 7.3 the main abradable coating systems are
listed with the indication of the relevant temperature limits and corresponding engine
part in which they can be used. The abrasive counterpart material systems are also
indicated together with the dominant wear mechanisms.

7.3.5 Erosion- and Fretting-Resistant Coatings

As concerns the coating systems for gas turbines involving metallic components,
two more surface degradation mechanisms have to be considered. The first is
erosion, and the other one is fretting. Erosion can be defined as the progressive
removal of material from the surface of a structural component, due to the
7.3 Main Coating Systems in Gas Turbine Engines 429

Table 7.3 Overview on the main abradable coating systems

Temperature GT Intruding blading material Cutting
Materials systems limits [
C] part compatibility mechanisms
Polymers 200 Fan Composites Softer
Al-Si + polymer 325 LPC Titanium alloys matrices:
Steels shear
cutting
Ni-based Superalloys
Al-Si +graphite 450 LPC Titanium alloys
Al-Si+hBN Steels
Ni+graphite Ni-based Superalloys
Al-bronze+polymer 550 LPC Titanium alloys Harder
CoNiCrAlY+hBN+polymer HPC Steels matrices:
NiCrFeAl + hBN Ni-based Superalloys friable
cutting
NiCrAl + bentonite 650 HPC Ni-based Superalloys
NiCrFeAl + hBN 750 HPC Ni-based Superalloys
CoNiCrAlY+hBN+polymer
YSZ + hBN + polymer 1100 HPT Ni-based Superalloys
DySZ + hBN + polymer 1200 Recommended h(c)BN
on blade tip or coating
porosity >30% with bare
superalloys
h(c)BN hexagonal (cubic) boron nitride, Y(Dy)SZ yttria (dysprosia) stabilized zirconia, LPC and
HPC low- and high-pressure compressor, respectively, HPT high-pressure turbine (Wilson 2012)

interactions with solid and even liquid particles/droplets contained in the gas flux.
This kind of attack is rather important for compressors, since they are the parts of the
engines directly exposed to the incoming air flux, which may contain sand, fly and
volcanic ashes, salt, and ice crystals (Helm and Olaf Roder 2007). The wearing out
of the compressor blades due to erosion may lead to an increase in the aerodynamic
losses with a consequent decrease in the compression ratio and thereby a reduction in
the overall engine efficiency. Similar issues are there for land-based gas turbine
engines (Alman et al. 2013; Muboyadzhyan 2009). Considering that erosion is by far
a critical aspect for compressors, coatings for titanium alloys and steels have been
developed the most. The successful approach is based on the combination of very
hard, brittle layers alternate to tough metallic ones. These combined brittle-tough
systems in general guarantee an optimal protection against erosion and its depen-
dence on the impinging angles of the eroding agents. Material removal by the
impinging particles is reduced by the hard, brittle components of the coating, like
nitrides (e.g., TiN) or carbides (e.g., Cr3C2). If cracks nucleate in this brittle layer,
their propagation is stopped by the tough metallic layer, like Ti, Cr, etc., that is also
present in these coating systems. The actual coating composition to be selected will
depend, in the first place, on compatibility criteria with the substrate and, secondly,
on the required servicing conditions. These may be rather demanding in case of
inner, high-pressure stages of the compressor. Therefore, the erosion protective
coatings are also required to have a good thermal stability and to provide adequate
430 7 Coatings

protection against oxidation. Several categories of these coatings can be found in the
literature. One possibility, are just two-layer coatings, like Ti-TiN, Ti-TiC, Zr-ZrN,
Cr-Cr3C2, and V-VC, with a total thickness of 70–100 μm. The alternative approach
is based on the same material systems but with coatings that are now multilayered,
with a typical thickness of each single layer equal to 1–2 μm, up to an overall
thickness of 25–30 μm (Muboyadzhyan 2009; Rajendran 2012). More systems, like
TiN-Zr and TiN-Nb, have been considered, and from the characterization tests, it
turns out the paramount role of the ductile phase in maximizing the erosion resis-
tance (Brendel et al. 2008; Borawski et al. 2011). The latest evolution, influenced by
the positive results obtained in the tribological field, is still based on multilayered
coatings but with the thickness of each layer far below 1 μm. The finer structures
exhibit intrinsic toughness even when composed of alternate layers of nominally
brittle phases, like TiN-CrN and TiC-CrC (Kablow and Muboyadzhyan 2017).
The second surface degradation mechanism considered herewith, fretting, is
associated with the standard operating conditions and is due to the unavoidable
contact between surfaces under cyclic loading with typical oscillation amplitudes of
a few tenths of millimeter (Cortez et al. 1999; Kermapur et al. 2015; Park et al.
2006). When fretting occurs in a corrosive environment, it would be classified as a
mechanochemical corrosion phenomenon (see Sect. 8.1.6). Critical parts for fretting
are the dovetail joints in compressors, where the blades are connected to the discs.
The sequence of events that may lead to damaging the parts, even macroscopically,
can be summarized as follows. The very early stage of fretting features the progres-
sive removal of the thin oxide layers that are initially present on the surface of the
metallic components, when they start rubbing each other. When these oxide layers
are interrupted, metal-metal adhesive junctions start to form. Intense plastic defor-
mation can occur, accompanied by material transfer when the junctions get frac-
tured. The process continues and protrusions start to form on the mating surfaces.
The increasing roughness of the mating surfaces enhances the formation of more
junctions, hillocks interlocking, and a reciprocal plowing. Eventually, these phe-
nomena may lead to the formation of macroscopic cracks and to the detachment of
wear fragments. Wear debris may either be released or remain trapped in between the
contacting surfaces and get oxidized by the reactive atmosphere. The hard oxide
particles may both abrade the metallic surface and initiate further fatigue cracks, so
that the damage would progressively propagate to subsurface regions of the compo-
nent. To prevent or to reduce the incidence of fretting, triballoy coatings are
deposited onto the blade dovetail root, via thermal spraying deposition routes, like
APS, flame, and arc spraying (Sect. 7.2.1). Common coating compositions are based
on the Cu-Ni-In system, Cu-38Ni or Cu-36Ni-5In, deposited with thickness ranging
from 10 to 100 μm (Oerlikon Metco 2016; Praxair Surface Technologies 2000). To
optimize the mechanical coupling between the inner surface of the dovetail on the
disc, the surface of the coating is usually covered with a layer of a composite paste,
made of a polymeric matrix with a suspension of MoSi2 particles (Molydag) (Zhou
1999).
References 431

References

Allam I M et al (1979) Influence of small Pt additions on Al2O3 Scale Adherence. Oxidation of

Metals 14: 517
Allen P B et al (1999) Diffusion, Locons and Propagons; Character of atomic Vibrations in
amorphous Si. Philosophical Magazine B 79 (11–12): 1715–1731
Alman et al (2013) Erosion-resistant Nanocoatings for improved Energy Efficiency in Gas Tur-
bines. U.S. Department of Energy
Alperine S, Lelait L (1994) Microstructural Investigations of Plasma-sprayed Yttria partially
stabilised Zirconia TBC. Transactions ASME: Journal of Engineering for Gas Turbines and
Flows 116: 258–265
Anderson P N, Sheffler K D (1983) Development of Strain Tolerant Thermal Barrier Coating
Systems. Available via DIALOG. https://ntrs.nasa.gov/search.jsp?R¼19840004244. Accessed
25 Jul 2017
Berry D et al (1995) Enhancing Performance of Silicon-modified Slurry Aluminides on Turbine
Components operating in Marine Environments. In: International Gas Turbine and Aeroengine
Congress and Exposition, Huston, Texas, 5–8 June 1995
Borawski B et al (2011) The Influence of ductile interlayer Material on the Particle Erosion
Resistance of Multilayer TiN based Coatings. Wear 271 (11–12): 2890–2898
Borom M P et al (1996) Role of Environmental Deposits and operating Surface Temperature in
Spallation of Air Plasma sprayed Thermal Barrier Coatings. Surface and Coatings Technology
86–87 Part 1: 116–126
Bose S (2007) High Temperature Coatings. Butterworth Heinemann
Brendel et al (2008) MTU Solutions against Erosive Attack and Loss of EGT Margin in Turbo
Engines-ER coat. MTU Technical Publication
Bruce R W (1998) Development of 1232
C (2250 F) Erosion and Impact Tests for Thermal Barrier
Coatings. Tribology Transactions 41 (4): 399–410
Cao X Q et al. (2004) Ceramic Materials for Thermal Barrier Coatings. Journal of the European
Ceramic Society 24 (1): 1–10
Chen M W et al (2003) Characterization and Modeling of a Martensitic Transformation in a
Platinum modified Diffusion Aluminide bond Coat for Thermal Barrier Coatings. Acta
Materialia 51 (14): 4279–4294
Chen X et al (2003) Foreign Object Damage in a Thermal Barrier System: Mechanisms and
Simulations. Materials Science and Engineering A 352 (1–2): 221–231
Chen Y et al (2017) A Mechanistic Understanding on Rumpling of NiCoCrAlY bond Coat for
Thermal Barrier Coating Applications. Acta Materialia 128: 31–42
Clarke D R (2003) Materials Selection Guidelines for low thermal Conductivity Thermal Barrier
Coatings. Surface and Coatings Technology 163–164: 67–74
Clarke D R et al (2012) Thermal-barrier Coatings for more efficient Gas-turbine Engines. MRS
Bulletin 37 (10):891–898
Cortez R et al (1999) Investigation of variable Amplitude Loading on Fretting Fatigue Behavior of
Ti-6Al-4V. International Journal of Fatigue 21 (7): 709–717
Das D K (2013) Microstructure and High Temperature Oxidation Behavior of Pt-modified
Aluminide bond Coats on Ni-base Superalloys. Progress in Materials Science 58 (2): 151–182
Davies H (1963) The Design and Development of the Thiocol XRL99 Rocket Engine for the X-15
Aircraft. Journal of the Royal Aeronautical Society 167: 79–91
Davis A W, Evans A G (2006) Effects of bond Coat Misfit Strains on the Rumpling of thermally
grown Oxides. Metallurgical and Materials Transactions A 37 (7): 2085–2095
Deb P et al (1987) Surface Stability of Platinum Modified Aluminide Coatings during 1100
C
Cyclic Test. Journal of Vacuum Science and Technology A 5 (6): 3366–3372
432 7 Coatings

Dobek L J (1973) Labyrynth Seal Testing for Lift Fan Engines. Available via DIALOG. https://ntrs.
nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19730007797.pdf. Accessed 4 September 2017
Dorfman M R et al (2008) Recent Development of High Temperature Thermal Spray Abradable
Coatings. Ceramics Japan, The Ceramic Society of Jaan 43 (5): 389–395
Driver M (2012) Coatings for Biomedical Applications. Woodhead Publishing Ltd, Cambridge
(UK)
Dryepondt S, Pint B A (2010) Determination of the Ductile to Brittle Temperature Transition of
Aluminide Coatings and its Influence on the mechanical Behavior of coated Specimens. Surface
Coatings and Technology 205: 1195–1199
Duvall D S (1981) Processing Technology for Advanced Metallic and Ceramic Turbine Airfoil
Coatings. In: Proceedings of 2nd Conference of Advanced Materials for Alternative-Fuel-
Capable Heat Engines, Electric Power Research Institute, 24–28 August 1981
Evans et al (2008) The Influence of Oxides on the Performance of Advanced Gas Turbines. Journal
of the European Ceramic Society 28 (7): 1405–1419
Fauchais P L et al (2014) Thermal spray Fundamentals: From Powder to Part. Srpinger
Felten E J (1976) Use of Platinum and Rhodium to improve Oxide Adherence on Ni-8Cr-6Al
Alloys. Oxidation of Metals 10 (1): 23–28
Felten E J, Pettit F S (1976) Development, Growth, and Adhesion of Al2O3 on Platinum-Aluminum
Alloys. Oxidation of Metals 10 (3): 189–223
Firdaouss M et al (2017) Tungsten Coating by ATC Plasma Spraying on CFC for WEST Tokamak.
Physica Scripta T170
Fountain J G et al (1976) The Influence of Platinum on the Maintenance of α-Al2O3 as a protective
Scale. Oxidation of Metals 10 (5): 341–345
Galmiche P M (1975) US Patent 3,900,613
Garrett E G, Gyorgak C A (1953) Adhesive and Protective Characteristics of Ceramic Coating
A-417 and its Effects on Engine Life of Forged Refractory-26 (AMS 5760) and Cast Stellite
(AMS 5385) Turbine Blades. Available via DIALOG. https://ntrs.nasa.gov/search.jsp?
R¼19930082087. Accessed 20 July 2017
Garvie R C et al (1972) Structure and Thermomechanical Properties of Partially Stabilized Zirconia
in the CaO-ZrO2 System. Journal of the American Ceramic Society 55 (3) 152–157
Garvie R C et al (1975) Ceramic Steel? Nature 258: 703–704
Garvie RC, Goss M F (1986) Intrinsic Size Dependence of the Phase Transformation in Zirconia
Microcrystals. Journal of Materials Science 21 (4): 1253–1257
Gauje G, Morbioli R (1983) Vapor Phase Aluminizing to Protect Turbine Airfoils. In: Singhal S C
(ed) High Temperature Protective Coatings. The Metallurgical Society of AIME, Atlanta, GA, p
13–26
Giggins C S, Pettit F S (1979) Hot Corrosion Degradation of Metals and Alloys – A Unified Theory.
Pratt & Whitney
Goward G W (1998) Progress in Coatings for Gas Turbine Airfoils. Surface and Coatings
Technology 108-109: 73–79
Green D J et al (1989) Transformation Toughening of Ceramics. CRC Press, Boca Raton, Florida,
USA
Guo H et al (2009) Thermo-physical and Thermal Cycling Properties of Plasma-sprayed
BaLa2Ti3O10 Coating as Potential Thermal Barrier Materials. Surface and Coatings Technology
204 (5): 691–696
Gupta D K, Duvall D S (1984) A Silicon and Hafnium modified Plasma Sprayed MCrAlY Coating.
In: Bricknell R H et al (eds) Superlloys. The Metallurgical Society of AIMR, Warrendale, PA, p
711–720
Harrison W N (1947) Review of an Investigation of Ceramic Coatings for Metallic Turbine Parts
and Other High Temperature Applications. Available via DIALOG. https://ntrs.nasa.gov/
archive/nasa/casi.ntrs.nasa.gov/19930087630.pdf. Accessed 20 July 2017
References 433

Hazell B et al (2008) Development of improved bond Coat for enhanced Turbine Durability.
Superalloys 2008: 753–760
Heimann R B (2006) Plasma Spray Coating: Principles and Applications 2nd edn. Wiley VHC
Helm D, Olaf Roder O (2007) Recent Titanium Research and Development in Germani. In: Ninomi
M et al (ed) Ti-2007 Science and Technology. The Japan Institute of Metals, p 25–32
Hermanek F J (2001) Thermal Spray Terminology and Origins. ASM International, Materials Park,
OH
Hjelm L N, Bornhorst B R (1961) Research-Airplane-Committee on Conference on the Progress of
the X-15 Project. Available via DIALOG. https://ntrs.nasa.gov/search.jsp?R¼19710070129.
Accessed 20 July 2017
Hocking M G et al (1989) Metallic and Ceramic Coatings. Longman Scientific & Technical
Ibegazene H et al (1993) Microstructure of Yttria stabilized Zirconia-Hafnia Plasma sprayed
Thermal Barrier Coatings. Journal De Phisique IV 3: 1013–1016
Jackson R W et al (2014) Thermal Barrier Coatings Adherence to Hf-modified B2 NiAl bond
Coatings. Acta Materialia 80: 39–47
Jackson R W et al (2015) Interaction of molten Silicates with Thermal Barrier Coatings under
Temperature Gradients. Acta Materialia 89: 396–407
Jarligo M O et al (2010) Atmospheric Plasma Spraying of High Melting Temperature Complex
Perovskites for TBC Application. Journal of Thermal Spray Technology 19 (1–2): 303–310
Jorgensen D J et al (2016) Bond Coatings with high Rumpling Resistance: Design and Character-
ization. Surface and Coating Technology 300: 25–34
Joseph A D (1960) US Patent 3,102,044
Kablow E N, Muboyadzhyan S A (2017) Erosion-Resistant Coatings for Gas Turbine Engine
Compressor Blades. Russian Metallurgy 2017 (6): 494–504
Kakuda T R et al (2015) The Thermal Behavior of CMAS-infiltrated Thermal Barrier Coatings.
Surface and Coatings Technology 272: 350–356
Kermapur A et al (2015) Failure Analysis of Ti6Al4V Gas Turbine Compressor Blades. Engineer-
ing Failure Analysis 15 (8): 1052–1064
Khanna A S (2002) Introduction to High Temperature Oxidation. ASM International
Kingery W D (1955) Thermal Conductivity: Temperature Dependence of Conductivity for Single-
phase Ceramics. Journal of the American Ceramic Society 38: 251–255
Krämer S et al (2008) Mechanism of Cracking and Delamination within Thermal Barrier Systems in
Aeroengines subject to Calcium-Magnesium-Alumino-Silicate (CMAS) Penetration. Materials
Science and Engineering A 490 (1–2): 26–35
Lammermann H, Kienel G (1991) PVD Coatings for Aircraft Turbine Blades. Advanced Materials
and Processes 140 (6): 18–23
Lattime S B, Steinetz B M (2004) Turbine Engine Clearance Control Systems: Current Practices
and Future Directions. Journal of Propulsion and Power 20 (2): 302–311
Lee J W, Kuo Y C (2005) Cyclic Oxidation Behavior of a Cobalt Aluminide Coating on Co-base
Superalloy AMS 5608. Surface and Coatings Technology 200 (5–6): 1225–1230
Lelait L et al (1993) Microstructural Investigations of EBPVD Thermal Barrier Coatings. Journal
De Physique IV 3: 645–654
Lesnikova E G, Lesnikov V P (1986) Influence of the β-Phase of the Aluminide Coating on the
Condition and Scale Resitance of the Surface Layer of NBi-Al alloys. Metal Science and Heat
Treatment 28 (5): 372–376
Levi C G et al (2012) Environmental Degradation of Thermal Barrier Coatings by molten Deposits.
MRS Bulletin 37: 932–941
Lughi V, Clarke D R (2005) High Temperature Aging of YSZ Coatings and subsequent Transfor-
mation at low Temperature. Surface and Coatings Technology 200 (5–6): 1287–1291
Luthra K L, Leblanc O H (1987) Low Temperature Hot Corrosion of CoCrAl Alloys. Materials
Science and Engineering 87: 329–335
434 7 Coatings

Markou C et al (2011) IATA Maintenance Cost Executive Commentary. Available via DIALOG.
https://www.iata.org/whatwedo/workgroups/Documents/MCTF/AMC-Exec-Comment-FY14.
pdf. Accessed 12 November 2017
Meelu M C et al (1992) Sermaloy J (Silicon modified Aluminide) Coating mechanical Improve-
ments. Processing, Properties and Applications of Metallic and Ceramic Materials 2: 1247–1253
Meier S M et al (1990) Status of Ceramic Thermal Barrier Coatings – Gas Turbine Applications and
Life Prediction Methods. In: Proceedings of Coatings for Advanced Heat Engines Workshop,
US Department of Energy, 6–9 August 1990
Mercer C et al (2005) A Delamination Mechanism for Thermal Barrier Coatings subject to Calcium-
Magnesium-Alumino-Silicate (CMAS) Infiltration. Acta Materialia 53 (4): 1029–1039
Mercer C et al (2007) On a ferroelastic Mechanism governing the Toughness of Metastable
tetragonal-prime (t’) Yttria-stabilized Zirconia. Proceedings of the Royal Society A:
Mathematica, Physical and Engineering Sciences 463 (2081): 1393–1408
Mévrel R (1989) State of Art on High Temperature Corrosion resistant Coatings. Materials Science
and Engineering A 120–121 Part 1: 13–24
Mévrel R et al (1986) Pack Cementation Process. Materials Science and Technology 2: 201–206
Miller R A et al. Phase Stability in Plasma-sprayed partially stabilised Zirconia-Yttria. In: Heuer
A H, Hobbs L W, editors. Advances in Ceramics Volume 3 – Science and Technology of
Zirconia. The American Ceramic Society, Columbus, Ohio; 1981. p. 241–253
Muboyadzhyan S A (2009) Erosion Resistant Coatings for Gas Turbine Compressor Blades.
Russian Metallurgy 2009 (3): 183–186
Mumm D R et al (2001) Characterization of a Cyclic Displacement Instability for a thermally grown
Oxide in a Thermal Barrier System. Acta Materialia 49 (12): 2329–2340
Naji A et al (2015) Improvements in the Thermodynamic and Kinetic Considerations on the
Coating Design for Diffusion Coatings formed via Pack Cementation. Materials and Corrosion
66 (9): 863–868
Nicholls et al (1999) A Comparison between the Erosion Behavior of Thermal Spray and Electron-
beam Physical Vapour Deposition Thermal Barrier Coatings. Wear 233: 352–361
Nicholls J R (2000) Designing Oxidation-resistant Coatings. JOM 52 (1): 28–35
Nicholls J R (2003) Advances in Coating Design for High-Performance Gas Turbines. MRS
Bulletin 28 (9): 659–670
Nicholls J R et al (2002) Smart Overlay Coatings – Concept and Practice. Surface and Coatings
Technology 149: 236–244
Nicoll A R et al (1986) Future Developments in Plasma Spray Coating. Materials Science and
Technology 2 (3): 214–219
Oerlikon Metco (2014) Solutions Flash: Improve Efficiency and Reduce Emissions with High
Pressure Turbine Abradable Coatings for Industrial Gas Turbines. Available via DIALOG.
https://www.oerlikon.com/ecomaXL/files/metco/oerlikon_SF-0015.1_IGT_HPT_Abradables_
EN.pdf&download¼1. Accessed 12 November 2017
Oerlikon Metco (2016) Material Product Data Sheet – Copper Nickel and Copper Nickel Indium
Thermal Spray Powders. Available via DIALOG. https://www.oerlikon.com/ecomaXL/files/
metco/oerlikon_DSMTS-0061.5_CuNi_CuNiIn_Powders.pdf&download¼1. Accessed
12 November 2017
Padture N P et al (2002) Thermal Barrier Coatings or Gas-Turbine Engine Applications. Science
296: 280–284
Pan W et al (2012) Low Thermal Conductivity Oxides. MRS Bulletin 37: 917–922
Park Y W et al (2006) Effect of Fretting Amplitude and Frequency on the Fretting Corrosion
Behavior of Tin plated Contacts. Surface and Coatings Technology 201: 2181–2192
Parzukowski R S (1977) Gas-Phase Deposition of Aluminum on Nickel Alloys. Thin Solid Films
45: 349–355
Pérez F J et al (1999a) Aluminizing and chromizing Bed Treatment by CVD in a fluidized Bed
Reactor on Austenitic Stainless Steels. Surface and Coatings Technology 120–121: 151–157
References 435

Pérez F J et al (1999b) Kinetic Studies of Cr and Al Deposition using CVD-FBR on different

metallic Substrates. Surface and Coatings Technology 122: 281–289
Pint B A (1997) Study of the Reactive Element Effect in ODS Iron-Base Alumina Formers.
Materials Science Forum 251–254: 397–404
Pint B A (2004) The Role of Chemical Composition on the Oxidation Performance of Aluminide
Coatings. Surface and Coatings Technology 188–189: 71–78
Pint B A et al (1998) Substrate and Bond Coat Compositions: Factors affecting Alumina Scale
Adhesion. Materials Science and Engineering A 245 (2): 201–211
Pint B A et al (2001) Evaluation of Iron-Aluminide CVD Coatings for High Temperature Corrosion
Protection. Materials at High Temperatures 18 (3): 185–192
Pint B A et al (2006) Oxidation Resistance: One Barrier to moving beyond Ni-base Superalloys.
Materials Science and Engineering A 415 (1–2):255–263
Pint B A et al (2011) Effect of increased Water Vapor Levels on TBC Lifetime with Pt-containing
bond Coatings. Surface and Coatings Technology 206: 1566–1570
Pint B A, Zhang Y (2011) Performance of Al-rich Oxidation-resistant Coatings for Fe-base Alloys.
Materials and Corrosion 62 (6): 549–560
Poerschke D L et al (2016) Equilibrium Relationships between Thermal Barrier Oxides and Silicate
Melts. Acta Materialia 120: 302–314
Pokluda J, Kianicová M (2010) Damage and Performance Assessment of Protective Coatings on
Turbine Blades. Available via DIALOG. https://www.intechopen.com/download/pdf/12092.
Accessed 28 October 2017
Prater J T et al (1981) Proceedings of the second Conference on Advanced Materials for alternate
Fuels capable Heat Engines. Palo Alto, CA 1981
Praxair Surface Technologies (2000) Praxair and TAFA Arc Spray Copper-Nickel-Indium Wire-
58T. Available via DIALOG. http://tatiscia.com/wp-content/pdf/tafa/1.9.1.2-58T%20-%20Cop
per%20Nickel%20Indium%20Wire.pdf. Accessed 13 November 2017
Priest M S, Zhang Y (2015) Synthesis of Clean Aluminide Coatings on Ni-based Superalloys via
modified Pack Cementation Process. Materials and Corrosion 66 (10): 1111–1119
Qu Z et al (2011) Thermal Conductivity of the Gadolinium Calcium Silicate Apatites: Effect of
Different Point Defect Types. Acta Materialia 59 (10): 3941–3850
Raghavan S et al (2001) Thermal Properties of Zirconia co-doped with trivalent and pentavalent
Oxides. Acta Materialia 49 (1): 169–179
Raghavan S et al (2004) Ta2O5/Nb2O5 and Y2O3 Co-doped Zirconias for Thermal Barrier Coatings.
Journal of the American Ceramic Society 87 (3): 431–437
Rajendran R (2012) Gas Turbine Coatings – An Overview. Engineering Failure Analysis 26:
355–369
Reed R C (2006) The Superalloys – Fundamentals and Applications. Cambridge University Press,
Cambridge (UK)
Ren W et al (2014) Oxidation and Microstructure Evolution of Cobalt Aluminide Coatings on
directionally-solidified Superalloys during long term Exposure at 1000
C. Materials Research
Innovations 18 (S4):945–951
Restall J E, Wood M I (1986) Alternative Processes and Treatments. Materials Science and
Technology 2: 225–231
Rigney D V et al (1997) PVD Thermal Barrier Coating Applications and Process Development for
Aircraft Engines. Journal of Thermal Spray Technology 6 (2): 167–175
Rolls-Royce (1996) The Jet Engine 5th edition. Rolls-Royce plc, Derby
Ruud J A et al (2001) Strength Degradation and Failure Mechanisms of Electron-beam Physical-
vapor-deposited Thermal Barrier Coatings. Journal of the American Ceramic Society 84 (7):
1545–1552
Sato T et al (1985) Transformation of Yttria partially stabilized Zirconia by low Temperature
Annealing in Air. Journal of Materials Science 20 (4): 1466–1470
436 7 Coatings

Scharrer and Pelletti (1995) Leakage and Rotordynamic Effects of Compressor annular Seals. In:
Proceedings of 24th Turbomachinery Symposium, Huston, Texas, 26–28 September 1995, p
175
Schmid R (1997) New High Temperature Abradable Coatings for Gas Turbines. PhD Thesis, Swiss
Federal Institute of Technology
Scott H G (1975) Phase Relationship in the Zirconia-Yttria System. Journal of Materials Science
10 (9): 1527–1535
Seelig R P, Steuber R J (1978) High-Temperture-resistant Coatings for Superalloys. High Temper-
atures – High Pressures 10 (2): 207–213
Shankar S, Seigle L L (1978) Interdiffusion and intrinsic Diffusion in the NiAl (δ) Phase of the
Al-Ni System. Metallurgical Transactions A 9 (10): 1467–1476
Shen et al (2008) Anisotropic thermal Conductivity of the Aurivillus Phase, Bismuth Titanate
(Bi4Ti3O12): A natural nanostructured Superlattice. Applied Physics Letters 93 (10) 102907
Shen Y et al (2010) Low thermal Conductivity without Oxygen Vacancies in equimolar
YO1.5+TaO2.5- and YbO1.5+TaO2.5- stabilized Zirconia Ceramics. Acta Materialia 58 (13):
4424–4431
Smialek J L (1971) Marteniste in NiAl Oxidation Resistant Coatings. Metallurgical Transactions
2 (3): 913–915
Smith A B et al (1999) Vapour Aluminide Coating of Internal Cooling Channels in Turbine Blades
and Vanes. Surface Coatings Technology 120–121: 112–117
Smith J S, Boone D H (1990) Platinum Modified Aluminides-Present Status. In: Gas Turbine and
Aeroengine Congress and Exposition, Brussels, Belgium, 1980
Song X W et al (2011) Influence of the partial Substitution of Y2O3 with Ln2O3 (Ln¼Nd, Sm, Gd)
on the Phase Structure and thermophysical Properties of ZrO2-Nb2O3-Y2O3 Ceramics. Acta
Materialia 59 (10): 3895–3902
Sparks T D et al (2010) Anisotropic Thermal Diffusivity and Conductivity of La-doped Strontium
Niobate Sr2Nb2O7. Journal of the American Ceramic Society 93 (4): 1136–1141
Sporer D R et al (2011) US Patent 7981530 B2
Squillace A et al (1999) The Control of the Composition and Structure of Aluminide Layers formed
by Vapour Aluminising. Surface and Coatings Technology 120–121: 118–123
Stecura S (1986) Advanced Thermal Barrier System bond Coatings for Use on Nickel-, Cobalt- and
Iron-base Alloy Substrate. Thin Solid Films 136 (2): 241–256
Strangman T E (1987) Development and Performance of Physical Vapor Deposition Thermal
Barrier Coating Systems. In: Proceedings of Coatings for Advanced Heat Engines Workshop,
US Department of Energy, 27–30 July 1987
Strangman T E (1996) US Patent 5,514,482
Talboom F P, Grafwallner J (1970) US Patent 3,542,5330
Tawancy H M et al (1991) Role of Platinum in Aluminide Coatings. Surface and Coatings
Technology 49 (1-3):1–7
Thorpe M L, Richter J (1992) A pragmatic Analysis and Comparison of HVOF Processes. Journal
of Thermal Spray Technology 1 (2): 161–170
Tolpygo V K et al (2008) Effect of Hf, Y and C in the underlying Superalloy on the Rumpling of
diffusion Aluminide Coatings. Acta Materialia 56 (3): 489–499
Tolpygo V K, Clarke D R (2004a) On the Rumpling Mechanism in Nickel-Aluminide Coatings:
Part I: An experimental Assesment. Acta Materialia 52 (17): 5115–5127
Tolpygo V K, Clarke D R (2004b) On the Rumpling Mechanism in Nickel-Aluminide Coatings:
Part II: Characterization of Surface Undulations and bond Coat Swelling. Acta Materialia
52 (17): 5129–5141
Tu J et al (1994) The Size Effect of the martensitic Transformation in ZrO2-containing Ceramics.
Journal of Materials Science 29 (6): 1662–1665
Tyron B et al (2006) Multilayered Ruthenium-modified bond Coats for Thermal Barrier Coatings.
Metallurgical and Materials Transactions A 37 (11): 3347–3358
References 437

Tyron B et al (2007) Hybrid Intermetallic Ru/Pt modified bond Coatings for Thermal Barrier
Systems. Surface and Coatings Technology 2020 (2): 349–361
Unocic K A, Pint B A (2013) Effect of Water Vapor on thermally grown Alumina Scales on bond
Coating. Surface and Coatings Technology 215: 30–38
Van Ness D et al (2006) Turbine Tip Clearance Flow Control Using Plasma Actuators. Paper
presented at the 44th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, 9–-
12 January 2006
Van Roode M, Hsu L (1989) Evaluation of the Hot Corrosion Protection of Coatings for Turbine
hot Section Components. Surface and Coatings Technology 37 (4): 461–481
Vassen R et al (2000) Zirconates as New Materials for Thermal Barrier Coatings. Journal of the
American Ceramic Society 83 (8): 2023–2028
Vassen R et al (2010) Overview on advanced thermal Barrier Coatings. Surface and Coatings
Technology 205 (4): 938–942
Vassen R, Stöver D (2007) Influence of Microstructure on the Thermal Cycling Performance of
Thermal Barrier Coatings. In: Marple B R et al (ed) Thermal Spray 2007: Global Coating
Solutions: Proceedings of the 2007 International Thermal Spray Conference. ASM Interna-
tional, Materials Park, Ohio, 2007
Wang et al (2011) Diffusion Barrier behaviors of (Ru, Ni)Al/NiAl Coatings on Ni-based Superalloy
Substrate. Intermetallics 19 (2): 191–195
Warnes B M, Punola D C (1997) Clean Diffusion Coatings by Chemical Vapor Deposition. Surface
and Coatings Technology 94–95: 1–6
Wilden J et al (2006) Plasma transferred Arc Welding – Modeling and experimental Optimization.
Journal of Spray Technology 15 (4): 779–784
Wilson S (2012) Thermally sprayed abradable Coating Technology for Sealing in Gas Turbines.
Oerlikon Metco White Paper – Thermally Sprayed Abradable Coatins 10: 1–9
Winter M R, Clarke D R (2007) Oxide Materials with Low Thermal Conductivity. Journal of the
American Ceramic Society 90 (2): 533–540
Wolfe D E et al (2005) Tailored Microstructure of EB-PVD 8YSZ Thermal Barrier Coatings with
low thermal Conductivity and high thermal Reflectivity for Turbine Applications. Surface and
Coatings Technology 190: 132–149
Wright I G, Gibbons T B (2007) Recent Developments in Gas Turbine Materials and Technology
and their Implications for Syngas Firing. International Journal of Hydrogen Energy 32 (16):
3610–3621
Wu L T et al (2017) A prominent Driving Force for the Spallation of Thermal Barrier Coatings:
Chemistry dependent Phase Transformation of the bond Coat. Acta Materialia 137: 22–35
Wuench B J, Eberman K W (2000) Order-Disorder Phenomena in A2B2O7 Pyrochlore Oxides.
JOM 52 (7): 19–21
Xu T et al (2004) Observations and Analyses of Failure Mechanisms in Thermal Barrier Systems
with two Phase bond Coats based on NiCoCrAlY. Acta Materialia 52 (6): 1439–1450
Yang S et al (2008) Anisotropic Thermal Conductivity of the Aurivillus Phase, Bismuth Titanate
(Bi4Ti3O12): A Natural Nanostructured Superlattice. Applied Physics Letters 93 (10):
102907–102907-3
Zhang Y et al (2003) Martensitic Transformation in CVD NiAl and (Ni, Pt)Al bond Coatings.
Surface and Coatings Technology 163–164: 19–24
Zhang Y et al (2004) Effect of Cycle Length on the Oxidation Performance of Iron Aluminide
Coatings. Surface and Coatings Technology 188–189: 35–40
Zhang Y et al (2007) Interdiffusional Degradation of Oxidation-resistant Aluminide Coatings on
Fe-base alloys. Materials and Corrosion 58 (10): 751–761
Zhao M et al (2012) Properties of Yttria-Stabilized-Zirconia Based Ceramic Composite Abradable
Coatings. Key Engineering Materials 512–515: 1551–1554
Zhou Z R (1999) Lubrication in Fretting – A Review. Wear 225–229: 962–967
Zhu D M, Miller R A (2004) Development of Advanced Low Conductivity Thermal Barrier
Coatings. International Journal of Applied Ceramic Technology 1 (1): 86–94
438 7 Coatings

Further Reading

Davis J R (2004) Handbook of Thermal Spray Technology. ASM International

Dorfman M R et al (2013), Thermal Spray Technology Growth in Gas Turbine Applications. In:
Tucker R C (ed) Thermal Spray Technology vol. 5A, ASM International, pp 280–286
Xu H and Guo H (2011) Thermal Barrier Coatings. Woodhead publishing
Kanna A S (2016) High Temperature Corrosion. World Scientific
Tamarin Y (2002) Protective Coatings for Turbine Blades. ASM International
Chapter 8
Corrosion

8.1 Introduction

Corrosion can be defined, in general terms, as a deterioration process that metallic

materials undergo in specific environmental conditions, involving chemical reac-
tions that may significantly change the original properties of the base alloys,
including those used in aerospace applications. The concept of corrosion can be
applied beyond the limits of metallic materials, and it is nowadays used to describe
deterioration of a wider range of nonmetallic materials too, like rubbers, polymers,
and ceramics (ASM International 2003). Liquid metal corrosion and deterioration of
paints exposed to sunlight are just a few examples of the relevant phenomenology
(Fontana 1987). In some respects, corrosion of metals can be regarded as the
opposite transformation to the metallurgical extraction process. Apart from the
so-called noble metals, the other metallic elements are usually found in nature as
compounds, like oxides, sulfides, sulfates, silicates, and carbonates, just to mention
the main ore components. These same compounds are often present as products of
the corrosion reactions.
Corrosion has a large impact in several technological fields, like transportation,
food industry, energy production, petrochemical, and aerospace. Effects of corrosion
phenomena can be observed on a great variety of products, objects, and piece of
equipment, like metallic structures, pipelines, vehicles, biomedical implants, chem-
ical and nuclear reactors, structural parts of aircrafts, and parts of engines (Pedeferri
2010). An important step in the development of products that need to resist corrosion
is their accurate structural and functional design, taking into account the environ-
mental working conditions, which will thus complete the set of parameters to be
accounted for in view of a reliable life cycle and life cycle cost assessment. However,
it is not always that simple to predict and anticipate all service conditions and
relevant protection measures against corrosion phenomena that may even lead to

© Springer Nature Switzerland AG 2020 439

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_8
440 8 Corrosion

Table 8.1 Frequency of failure mechanisms for engineering and aircraft components (Findlay and
Harrison 2002)
Percentage of component failures
[%]
Mechanism Aircraft General Engineering
Fatigue 55 25
Corrosion 16 29
Overload 14 11
Mechanochemical corrosion and hydrogen embrittlement 7 6
Abrasion/erosion/wear 6 3
High-temperature corrosion 2 7
Brittle fatigue – 16
Creep – 3

part failure, when the component has lost its load-bearing capabilities. Failures in
aircraft structures are commonly associated with stress concentration (see Sect. 1.2)
due to:
• Design mistakes, e.g., holes in the wrong position, notches, and too tight fillet
radii
• Corrosion attack
• Microstructural features of the material, e.g., segregations and voids and unfitted
grain size
Records and case histories data show that for aircraft components the dominant
failure mode is fatigue (Findlay and Harrison 2002). Table 8.1 shows a comparison
between the failure modes for general engineering and aircraft components with the
relative incidence of each one of them. After fatigue (55%), corrosion-related
mechanisms turn out to be the second reason for failure (25% in total). Moreover,
the efforts for detection and repair of corrosion damages are prevailing over other
similar activities, like those involved with fatigue.
Different from fatigue damage, corrosive phenomena are generally slower and
their effects easier to be detected and, if possible, mended with routine maintenance
activities (Findlay and Harrison 2002). Corrosion protection involves both thermo-
dynamic and kinetic issues. The strategy aims not much at avoiding corrosion at all
but rather slowing it down, so to guarantee a reasonable and well-predictable service
lifetime. Thermodynamics provides a quantitative indication of the driving force for
the corrosion process to occur. The rate of reactions can be determined by means of
kinetic laws that help distinguishing among different corrosion processes, since they
mainly depend on the environment in which corrosion takes place (ASM Interna-
tional 2003).
8.1 Introduction 441

Fig. 8.1 The Daniell battery: a reference electrochemical system, useful to discuss the main
reactions involved with electrochemical (aqueous) corrosion. The zinc plate is the anode, while
the copper plate is the cathode

Two main groups of corrosion phenomena can be identified: electrochemical

corrosion and high-temperature corrosion. Electrochemical corrosion is quite a
common process, which depends on reactions involving transfer of electrons from
and to metallic alloys. In this regard, batteries can be considered as prototypes of
electrochemical systems. Referring to the copper-zinc cell developed by John
Frederic Daniell, Cu and Zn electrically connected plates are immersed in solutions
containing their salts (copper and zinc sulfate in the present example) (see Fig. 8.1).
A salt bridge between the two solutions allows the electrochemical cell circuit to be
completed. The two metals exhibit a different tendency to be corroded. In particular,
zinc is oxidizing, according to the reaction: Zn2+ + 2e
resulting in the dissolution of
zinc in the zinc sulfate solution. Copper is reducing, i.e., copper ions (Cu2+)
contained in the copper sulfate solution turn into metallic Cu by capturing the
electrons produced by zinc oxidation and reaching the copper plate. Common
features to all electrochemical corrosion processes are:
• The overall corrosion reaction can be split into two “half-reactions”: anodic and
cathodic.
• The two half reactions occur either in the same region at the interface between the
metal and the environment or in separate spots, depending on the local electro-
chemical potential.
• Since the half reactions are sustained by electron release and trapping, they
proceed at the same rate.
• Liquid water or alternative electrolytes are necessary.
442 8 Corrosion

Fig. 8.2 Corrosion reactions in basic and acid aqueous solutions. M is the base metal; ne
is the
number of electrons generated during the process

For aqueous solutions, it is important to consider also the water electrolysis

reactions, since they have an active role in the corrosion process. With reference
to the scheme in Fig. 8.2, depicting a metal in contact with pure water, the oxidation
reaction involving the metal is:

M ! Mzþ þ ze
ð8:1Þ

The reduction reactions involving water are:

1
O2 þ H2 O þ 2e
! 2OH
ð8:2Þ
2
2Hþ þ 2e
! H2 ð8:3Þ

Equations 8.2 and 8.3 describe the reduction reactions that take place in a basic
and acid environment, respectively.
Considering the oxidation reaction, z indicates the oxidation number, i.e., the
valence of the metal involved in the process, influencing the density of charge
carriers. The electron flow is generated by a difference in electrical potential between
cathodic and anodic regions.
The chemical equilibrium reached by a metal in contact with an electrolyte can be
described in terms of an electrical potential of the relevant half reaction. The
electrical potential of each half reaction, E0, is usually measured with respect to a
standard electrode potential, taken equal to zero. This is the standard hydrogen
electrode (SHE) potential. The E0 series in Table 8.2 is obtained in standard
conditions, i.e., considering each single chemical species in a 1 mole concentration
solution of the relevant ion. The series is obtained under equilibrium conditions and
cannot account for the kinetic aspects of the corrosive process. The kinetics of the
corrosive reaction strongly depends on the passivation phenomena. Passivation
8.1 Introduction 443

Table 8.2 Selected standard reduction potentials, E0, referred to the standard hydrogen electrode,
SHE at 298 K. The concentration of each aqueous solution is 1.0 mol/dm3. The pressure of a
gaseous component is 105 Pa (1 atm) (Housecroft and Constable 2006)
Reduction half-reaction E0 [V]
Li+(aq) + e– $ Li(s)
3.04
K+(aq) + e– $ K(s)
2.93
Ca2+(aq) + 2e– $ Ca(s)
2.87
Na+(aq) + e– $ Na(s)
2.71
Mg2+(aq) + 2e– $ Mg(s)
2.37
Al3+(aq) + 3e– $ Al(s)
1.66
Zn2+(aq) + 2e– $ Zn(s)
0.76
S (s) + 2e
$ S2
(aq)
0.48
Fe2+(aq) + 2e
$ Fe(s)
0.44
Cr3+(aq) + e
$ Cr2+(s)
0.41
Co2+(aq) + 2e
$ Co(s)
0.28
Pb2+(aq) + 2e
$ Pb(s)
0.13
Fe3+(aq) + 3e
$ Fe(s)
0.04
2H+(aq) + 2e
$ H2(gaseous at 1 bar) 0.00
2H+(aq) + S(s) + 2e
$ H2S(s) +0.14
Cu2+(aq) + e
$ Cu+(s) +0.15
Cu2+(aq) + 2e
$ Cu(s) +0.34
O2(g) + 2H2O(l) + 4e
$ 4[OH]
(aq) +0.40
Cu+(aq) + e
$ Cu(s) +0.52
I2(aq) + 2e
$ 2I
(aq) +0.54
Fe3+(aq) + e
$ Fe2+(aq) +0.77
Ag+(aq) + e
$ Ag(s) +0.80
Br2(aq) + 2e
$ 2Br
(s) +1.09
O2(g) + 4H+(g) + 4e
$ 2H2O (l) +1.23
[Cr2O7]2
(aq) + 14H+ (aq) + 6e
$ 2Cr3+(aq) + 7H2O(l) +1.33
Cl2(aq) + 2e
$ 2Cl
(aq) +1.36
[MnO4]
(aq) + 8H+ (aq) + 5e
$ Mn2+(aq) + 4H2O(l) +1.51
Ce4+(aq) + e
$ Ce3+(aq) +1.72
F2(aq) + 2e
$ 2F
(aq) +2.87
aq aqueous, g gaseous, l liquid, s solid

involves the formation of surface layer of metal native oxides and hydroxides. These
reaction products interfere with the electrochemical degradation process, creating an
insulating barrier on the metal surface. This leads to a protective condition, generated
by the presence of corrosion products on the corroding metal surface. The actual
protection of the surface corrosion layer strongly depends on the characteristics of
the environment in which the corrosion process is taking place. For instance, iron
undergoes passivation only in alkaline environments (pH > 7). The passivation
capability of metallic materials can be accounted for in the definition of an electro-
chemical classification. Namely, those metals that exhibit passivation would occupy
places in the upper part of this ranking. Real alloys do not have a single precise but
444 8 Corrosion

rather a range of electrochemical potentials. The potential range is determined by the

presence of alloying elements and impurities. Moreover, the way the material
undergoes corrosion is a function of both the potential and of the environment in
which the corrosive phenomena take place (McCafferty 2010). A specific combina-
tion of environment and stress field, both applied and internal, will determine the
actual corrosion behavior of any material. The surface conditions, concerning
finishing, defect structures, actual composition, and relevant evolution with time,
would strongly influence the kinetics and thermodynamics of corrosion processes.
The combination of these factors determines diverse corrosion types that can be
summarized as follows:
• General corrosion (uniform or nonuniform)
• Pitting
• Crevice
• Selective corrosion
• Galvanic corrosion (two metal corrosion)
• Mechanochemical corrosion, including fretting and stress corrosion cracking
(SCC)
• Hydrogen embrittlement (HE)
The most important corrosion types, fully discussed in (McCafferty 2010), will be
herewith illustrated with reference to specific cases, relevant to the aerospace field.

8.1.1 General Corrosion

General corrosion occurs either on the entire surface of a metal component or on a

very wide part of it, on which the general attack causes the metal to be corroded
homogeneously. General corrosion occurs in aircraft structures under normal service
conditions, in particular, in those parts of the airplane where condensation of water
may intervene. Cathodic and anodic reactions change dynamically over the time,
with reaction products that determine a uniform weight change all over the involved
surface area. A weight loss is observed as a consequence of non-adherent corrosion
products that may spall off after some time. Nonuniform general corrosion is
characterized by an irregular spatial distribution of the corrosive attack and conse-
quent weight change. Uniform general corrosion can be tolerated if it proceeds at a
sufficiently low rate to guarantee an adequate lifetime to the component. Actually,
general corrosion is not necessarily to be regarded as detrimental, since it might also
have positive effects; therefore it can be deliberately induced, like in the following
application cases:
• Increasing the roughness of the surface for specific applications, e.g., osteogen-
esis improvement in orthopedic implants.
• Metallographic attack for enhancing the surface contrast of the microstructure of
crystalline materials.
8.1 Introduction 445

• Pickling of the surface as preparation treatment for further surface operations, like
painting.
However, if prolonged in time, general corrosion may lead to a weakening of the
base metal, with a load redistribution causing local stress concentration and, possi-
bly, secondary bending (Benavides 2014).
If the areas subjected to corrosion damage are accessible to visual inspection,
once the attack has been detected, protection and remediation activities can be
initiated. On the other hand, when the corrosive attack involves inner parts, where
direct visual inspection may be difficult or impossible at all, the corrosion damage
may become serious. Therefore, inspection methods, such as radiographic- and
ultrasonic-based techniques (see Sect. 4.5), are recommended for an early detection
and diagnostics, followed by a proper maintenance intervention, if needed. Alumi-
num and magnesium alloys are prone to general corrosion. Concerning Al-alloys,
general corrosion can affect stringers when the chromate pre-treatment and the
primer coat fail, due to aging and exposure to moisture. General corrosion of
Mg-alloys can occur on external parts, such as the skin, again for the localized
failure of the chromate pre-treatment (Wallace et al. 1985). As concerns general
corrosion, relative humidity, environmental, and relevant pH conditions are all
extremely important parameters. Their effects can be discussed considering the
relevant Pourbaix diagrams (see Fig. 3.4, Fig. 3.31, and Fig. 4.18 for the corrosion
behavior of Al, Mg, and Ti).

8.1.2 Pitting Corrosion

Pitting corrosion is a localized degradation process, regarding those alloys usually

showing a self-passivation behavior, like stainless steels, aluminum, and titanium
alloys. Reactive metals, e.g., magnesium, may suffer from this kind of corrosive attack
too, as a result of local wetting of their surface. The possible reasons for pitting
corrosion are defects and cracks in the protective self-passivation surface layer,
which allow a very localized exposure of the alloy surface to the outer environment.
Halide ions, like Cl
, highly soluble and present in diverse environments, like
seawater and ground, render the attack more aggressive. Indeed, the formation of
metal chlorides slows down or inhibits at all the re-formation of the protective oxide
layer on the alloy surface, which progressively loses its self-passivating attitude. The
occurrence of pitting corrosion depends on the characteristics of the passivation layer:
if its thickness is not uniform on the entire surface, and microstructural and lattice
defects are present, this renders specific areas preferential sites for a localized corro-
sion. Figure 8.3 illustrates the main mechanisms involved with the pitting corrosion
process of a metal in contact with an electrolyte containing chlorine. If the concentra-
tion of oxygen inside the surface cavity is comparatively low, this may locally inhibit
the passivation of the metal. Consequently, two types of reactions occur inside the
defect: an anodic reaction involving the metal and the reduction of hydrogen, this latter
446 8 Corrosion

Fig. 8.3 Schematic of pitting corrosion

being favored by the increasing concentration of metallic ions (M+). The metal ions
diffuse out of the defect, whereas a counter-flux of chlorine ions tends to increase their
concentration inside the defect itself, thanks to a mass transport mechanism sustained
by the electric field (electrophoresis) associated with the ionic charges. This situation
enhances metal dissolution and hinders its passivation further. The anodic reaction
brings metallic ions into solution, sustaining a localized corrosion and subsequent
extension of the pit. The alloy resistance to pitting corrosion depends on its chemical
capability to re-form a protective layer in case it gets damaged. The AISI 316 stainless
steel, thanks to its molybdenum content, exhibits a better pitting resistance than AISI
304. In fact, molybdenum acts as a stabilizer for the passivation layer and slows down
the outward metal diffusion.
There are several spots in aircraft structures that may be subjected to pitting
corrosion, in particular those exposed to local contamination by highly corrosive
substances. Some examples are battery compartments, galley areas, and toilets.
Aircrafts operating in marine environments are particularly prone to pitting corrosion
phenomena because of the pervasive presence of chlorine ions. It has been reported
that the failure of an AISI 422 stainless steel (see Sect. 5.4.3.2) component caused
the loss of an engine’s thrust reverser door from a large commercial airliner during
cruise flight. The door driver link, made of AISI 422, broke down, allowing the door
to lower in the 1000 km/h airstream. More detailed inspections showed that small
fatigue cracks were growing from each of the many pits on the part surface. The root
cause for pitting corrosion was then identified in the entrapment of moisture and fuel
combustion by-products collected in fissures in the coating in direct contact with the
stainless steel surface (Wallace et al. 1985).
8.1 Introduction 447

Fig. 8.4 Schematic of the

P-180 aircraft main landing
gear wheel flange with
indication of the crack path.
(Redrawn from Bagnoli and
Bernabei 2008)

Aluminum alloys too can be subjected to pitting corrosion. The high-strength

7075-T6XXX aluminum alloy (see Chap. 3), extensively used in the fabrication of
fuselage skin and airframes (see Tables 3.14 and 3.15), is susceptible to both pitting
and stress corrosion cracking (see Sect. 8.1.4). Pitting corrosion of 7075 alloy may
have serious consequences on the fatigue life of the component. Experimental
observations reported by DuQuesnay et al. (2002) have shown that, owing to the
defects introduced by pit corrosion, fatigue life of the material, in simulated flying
hours (SFH), can decrease from 25,000 SFH with a defect depth of 0.1 mm down to
5000 SFH, when the defect has a depth of 0.75 mm. Furthermore, it has been shown
that the average depth of the pit can be regarded as the critical factor influencing the
phenomenon (DuQuesnay et al. 2002). Still concerning the effect of pitting on the
failure of Al alloy component, Bagnoli and Bernabei (2008) reported on the crack of
the inner wheel flange of the main landing gear of a Piaggio P-180 Avanti aircraft
after landing, during taxing on the runway. At that very moment, the crew heard a
sudden loud bang from the right side of the aircraft. An inspection carried out
afterwards showed that a part of the inner flange wheel fillet had detached from
the rest of the landing gear (Fig. 8.4). The failed wheel was made of a 2014-T6 Al
alloy forging and had been in service for 22 years before the accident.
The failure analysis revealed the presence of three different regions on the
fracture surface: bright and smooth surface, containing the typical “beach marks”
(fatigue crack growth region); rough and dull surface alternated to a bright one
(unstable crack propagation); and a dull, rough with a coarse-grained region (final
unstable fracture). The crack initiation area was identified to be along the inner
surface of the flange, near the fillet. In this area, corrosion pits were found,
suggesting multiple initiation sites. Moreover, a little amount of corrosion products
were found on the fracture surface (Bagnoli and Bernabei 2008). It has been reported
over the years, on several occasions, that aluminum and magnesium alloys, used in
aircraft wheels, had similar cracks. The wheels had failed similarly to what is
observed in this specific case history, and many of these failures were clearly
initiated by cracks due to pit corrosion that damaged the protective coatings or
passivation layer on the wheel part (ASM International 1992).
448 8 Corrosion

Fig. 8.5 Schematic of the crevice corrosion mechanism

8.1.3 Crevice Corrosion

Crevice corrosion is another type of localized corrosion, somehow similar to pitting

corrosion. The mechanism is indeed triggered by a surface defect, typically a tiny
gap between two alloy surfaces (crevice, Fig. 8.5). Any alloy may suffer from this
corrosion attack, irrespective of the composition of the electrolyte. It is also known
as “concentration cell corrosion,” and it may take place in interstices of mechanical
couplings, where water stagnates. Concerning aircrafts, crevice corrosion is usually
associated with small volumes of stagnant aqueous solutions trapped in between
gasket surfaces, lap joints, and holes; under rivets and loose detached paint; and in
crevices under fasteners heads, within delaminated bonds or unsealed joints. It is the
most common type of corrosion phenomenon that can be found in old aircrafts, when
maintenance had not been performed correctly (Mouritz 2012). Crevice corrosion is
triggered by variations in the composition of the electrolyte between the inner and
the outer part of the interstice (similarly to pitting corrosion), due to either corrosion
reactions, or obstacles to ionic transport phenomena, which would otherwise homog-
enize the composition of the electrolyte, thus reducing the crevice corrosion inci-
dence. The way the corrosive process happens is very similar to pitting corrosion.
However, one important difference is that crevice corrosion can be prevented with an
accurate component design that would minimize the presence of interstices, and
using appropriate rubber gaskets, if necessary.
Figure 8.5 shows a schematic of the crevice corrosion mechanism, involving two
metal parts in contact with an aqueous electrolyte. At the initial stage, there is the
oxidation of part of the metal, leading to the formation of Mz+ ions. The resulting
8.1 Introduction 449

electron flow allows OH
to form at the meta-electrolyte interface. After an initial
stage, the oxygen within the interstice is consumed for the restrictions to mass
transport from and to the crevice. Therefore, the electrolyte compositions inside
and outside the interstice become progressively different. In particular, inside the
crevice, with a very low oxygen concentration, the cathodic semi-reaction cannot
occur any longer. Hence, the anodic semi-reaction is localized into the crevice, while
the cathodic can still occur outside. OH
ions balance the production of Mz+ and H+
ions so that the corrosive process is sustained. The volume and reaction surface in the
interstice, available to the anodic semi-reaction, are relatively smaller than for the
cathodic one. Therefore, the corrosive phenomenon is largely localized in between
the two metal parts and is characterized by the severity of the anodic process. Due to
its particular phenomenology, crevice corrosion is also known as differential aera-
tion corrosion. Depending on the alloy composition and microstructure, and, obvi-
ously, on the environmental conditions, crevice corrosion can quickly evolve into
exfoliation corrosion (see Sect. 8.1.4) or pitting, if it is not adequately counteracted
(Mouritz 2012). This corrosion type may be particularly important in Mg-alloys,
since they are the most anodic among the structural alloys (see Table 8.2), although
this corrosive attack is reported also in Al-alloys. In fact, crevice corrosion was
observed in the outer wing coupling flange of old small transport aircrafts that had
been operated in the Mediterranean region. The structure was predominantly made
of Al-alloy sheet and flanges with steel fasteners. Moreover, a protective TiO2-based
painting was applied onto the structure for protective purposes. Crevice corrosion
was evident at the aluminum lap joints and between the aluminum flanges and sheet.
Furthermore, corrosion cavities were found under the nuts and bolts, and the
corrosive attack extended beyond the area of contact of the fasteners (Wallace
et al. 1985).

8.1.4 Selective Corrosion

Selective corrosion of metallic alloys implies the preferential dissolution of the

alloying elements, or alloy phases, with lower electrochemical potentials. Selective
corrosion occurs in environments that are not particularly aggressive and typically
results in the formation of a porous surface layer. An interesting example of selective
corrosion is intergranular corrosion, i.e., a corrosion process that occurs at the alloy
grain boundaries. The driving force for intergranular corrosion is the difference in
electrochemical potential between the inner and boundary regions of the same grain.
In the corrosive process, the inner region of the grain becomes cathodic, and the
grain boundary becomes anodic. The anodic character of the grain boundary can
mainly be explained through some of its features, which imply an excess of energy
and, therefore, a higher reactivity. In particular, the chemical composition at the
grain boundary regions is different from the grain core because of segregation of
impurities and precipitation of secondary phases. Grain boundary precipitation is
often accompanied by the formation of a precipitation-free zone (PFZ, see Fig. 3.21)
450 8 Corrosion

due to the depletion of alloying elements along the grain boundaries (Mouritz 2012).
A common intergranular corrosion process is the sensitization occurring in austenitic
stainless steels (see Sect. 5.4.3.1), often as a consequence of welding processes. The
increase in temperature near the welded zone causes the precipitation of Cr-rich
carbides (M23C6) at the grain boundary with a consequent depletion of chromium in
the surrounding region, where a lowering of the corrosion resistance is observed.
This leads, in the worst condition, to the absence of the passivation layer on the steel
surface and to the localization of corrosive phenomena in this area. There are
different ways to avoid sensitization of stainless steels:
• Addition of alloying elements, like titanium, that have higher than chromium
affinity to carbon, thus avoiding chromium carbides formation and consequent
chromium depletion from the bulk.
• Fast cooling after the welding in order to avoid crossing the carbide transforma-
tion curves.
• Low-carbon content. This reduces the possibility for chromium to combine with
carbon during cooling after the welding.
• Welding process involving a reduced heating of the part, using, for instance, laser
welding.
Another particular type of intergranular corrosion is exfoliation corrosion. The
corrosive attack occurs at the grain boundaries, and the resulting corrosion products
exert a pressure on the surface grains. This pressure may lift off surface alloy grains
that then are removed by abrasion or other mechanical actions, which in this way
may sustain the process (Mouritz 2012). In the mid-1960s, exfoliation corrosion was
one of the main corrosive problems faced by the US Navy with the 7075-T651
aluminum alloy extrusions, used for the wing spars of air-sea rescue aircrafts.
Metallographic analyses of failed specimens highlighted that the deteriorated lami-
nar paths were preferential sites for fatigue crack nucleation. For this reason, the
fatigue life of the exfoliated material was strongly reduced, even down to the 70%.
Exfoliation corrosion originated deep inside rivet holes under crevice corrosion
conditions (see Sect. 8.1.3). The resulting deterioration was not detectable by
standard visual inspection, and in all cases, the intergranular corrosion propagated
along corrosion pits preferentially (Mills et al. 2014). Intergranular corrosion was
also detected in an aluminum alloy, used for wing attachment fitting of a fighter
aircraft during periodic inspections. The component was made of an Al-Zn-Mg-Cu
alloy with a chemical composition similar to that of 7075 alloy but with a little
amount of Ag. The mechanical properties were those induced by a solution treatment
followed by aging (T6 temper condition, see Table 3.9) and shot peening on all
external surfaces. Chemical conversion coatings, using Alodine 1200 and Zn chro-
mate primer, were also employed as surface finishing. Intergranular corrosion
occurred at the wing attachment fitting due to a combination of several adverse
conditions that were not inhibited by the surface protective treatment and coating
(Wallace et al. 1985).
8.1 Introduction 451

8.1.5 Galvanic Corrosion (Two-Metal Corrosion)

Galvanic corrosion takes place when metals and alloys with different electrochem-
ical potentials (see Table 8.2) are in contact in a corrosive medium. The resulting
corrosive attack is the consequence of the galvanic coupling among the involved
metals and alloys. However, galvanic corrosion does not necessarily require the
electrical contact between different alloys, since a galvanic coupling can also be
established within the same material between regions with different electrochemical
potentials. Quite common examples are the inner parts and the boundaries of
crystalline grains; precipitates in a nobler parent phase matrix; and regions with
different residual stress values, resulting from heat treatments and mechanical
working. In fact, the only strict requirement for galvanic corrosion to occur is a
difference in electrochemical potential and the presence of a suitable electrolyte (see
Daniell battery in Fig. 8.1). The less noble metal will be subjected to the anodic
semi-reaction, thus becoming the anode of the equivalent cell. The nobler metal is
instead subjected to the cathodic semi-reaction and acts as the cathode of the
electrolytic cell equivalent. The metal, or part of it, subjected to the anodic process
will show a progressive corrosive damage.
Systems in which galvanic corrosion is observed are, for instance, bronze pro-
peller in contact with steel hull, copper water pipes coupled with iron pipes, and
cold-worked alloy in contact with the same alloy but in a fully annealed condition.
One of the most common examples in aircraft structures is the corrosion occur-
ring at fasteners holes of Al- or Mg-alloy skin coupled with steel bolts and rivets or
in contact with a conductive cement. The use of graphite-based products or com-
posite materials, if not properly designed, can lead to galvanic corrosion too.
Galvanic corrosion has been reported for an Al-alloy wing skin, as a consequence
of the use of a graphite-based marker, i.e., a pencil, used for identifying a small nick
in the lower wing skin of a fighter aircraft (Wallace et al. 1985).

8.1.6 Mechanochemical Corrosion

Mechanochemical corrosion, also known as mechanically-assisted corrosion, is the

degradation process that takes place when mechanical stresses are combined with a
corrosive environment (McCafferty 2010). The intensity of the overall corrosive
attack would strongly depend on the balance between mechanical stress and the main
characteristics of the corrosive environment. Fretting is a typical example of a
mechanochemical corrosion process, resulting from the combined action of the
corrosive environment and a relative oscillatory movement between two mating
contacting surfaces (Fig. 8.6). Fretting may even occur between ball bearings and
their cases and in flexible mechanical couplings (Wallace et al. 1985). Although the
phenomenology of fretting is rather variate, two important mechanisms are inter-
vening in most systems:
452 8 Corrosion

Fig. 8.6 Schematic of fretting corrosion initiated by the damage of the passivation layer

• Damage of passivation layer on the alloy surface. Mechanical damage of the

passivation layer may be induced by the relative in contact movement of two
surfaces, with the partial removal of the protective oxide layer and consequent
localization of the corrosive phenomenon at the freshly exposed metallic surfaces.
• Oxidation wear. A mechanical damage of the surface is observed as a conse-
quence of the removal of the products of the oxidation induced by the localized
temperature rise.
Fretting damage can be particularly severe, since it may evolve into a sudden and
unexpected fatigue failure of the components, notwithstanding a correct design to
sustain normal service conditions.
Fretting corrosion has been identified as the main reason for the in-flight engine
failure of a piston engine helicopter. The failure was caused by fretting-induced
fatigue of an engine connecting rod (in this case, it is rather more appropriate to talk
of fretting fatigue). Fretting was caused by the rotational movement of the shell
bearing within the connecting rod big end bore (Wallace et al. 1985). One of the
most important reported cases of fretting corrosion in the aerospace field was the
corrosion of electrical contacts in the fuel control system of the General Dynamics/
Lockheed Martin F-16 Flying Falcon. The fretting damage caused an automatic shut
off of the valves of the main fuel supply without any preliminary warning. Before the
failure mechanism could be identified and the problem fixed, this anomaly was the
8.1 Introduction 453

Fig. 8.7 Schematic of the compressor blade with the detail of the crack at the tang and deformed
retainer pin. (Redrawn from Lee et al. 2011)

main reason for a high number, at least six, of F-16 crashes (Mouritz 2012). Fretting
is also an important issue in gas turbine engine compressor and turbine discs, in
particular at the dovetail joint between discs and blades. In those regions, contact
stresses are very high during flight operations, and, even in view of the relatively low
damage tolerance of the parts, the failure consequences are potentially serious (Mills
et al. 2014). In this regard, Lee et al. (2011) reported on the fracture of a Ti-6Al-4 V
(see Chap. 4) compressor blade installed on a J85 engine and connected to the
rotating disc through retainer pins made of Alloy 718 (see Sect. 6.3.2), inserted
through the tang-type root of the blade. The failure occurred at the blade tang, as
evidenced in Fig. 8.7.
The blade-disc connection is subjected to particularly severe service conditions,
involving both high rotational frequencies, vibrations, and centrifugal loads. The
highest centrifugal loads are localized in the first stages of the compressor at the
blade-disc connection regions, owing to the high rotational speed, the large radius of
the rotating disc, and the length of the blades. The retainer pins of the failed blade
resulted in severe deformation, and the detected diameter was smaller than that of
new parts (see Fig. 8.7), probably for the prolonged oscillating motion during the
engine operation. The damage of the retainer pin produced limited oscillatory
movements of the pin in contact with the inner surface of the pinhole in the blade
tang. The diagnostic investigation that was conducted on the damaged blade and pin
confirmed fretting corrosion to be the main cause for the observed failure. Debris
loss and mud cracks were observed at the crack initiation spot. The mud crack was an
evidence of dried liquid lubricant, probably leaked out once the crack had nucleated.
454 8 Corrosion

The damaged retainer pin accelerated the propagation of the fretting damage because
of the increased clearance between the two contacting surfaces and consequent
increase of mechanical tolerance (Lee et al. 2011).
Another type of damage depending on mechanochemical corrosion is the stress
corrosion cracking (SCC). SCC implies the failure induced by a slow, environmentally
driven crack propagation, occurring as a consequence of the synergistic interaction
between corrosion reaction and mechanical stress. Concerning the aerospace field,
routine maintenance and airframe inspection activities often include the search for
SCC signs. Although SCC is not the most common form of corrosive attack in
aerospace components, it accounts for 5–10% of all failures of aircraft components
(Mouritz 2012). Therefore, it is important to map the materials that can be subjected to
it, the type of stresses involved, and the environments and relevant interactions that
lead to SCC. Here is the summary list of the main representatives of these SCC factors:
• Materials: α-brass; several steels, like AerMet steels, austenitic stainless steels,
carbon and high-strength steels in general, maraging steels, and martensitic and
PH stainless steels; magnesium, aluminum, and titanium alloys; some nickel-
based superalloys; zirconium alloys.
• Stress conditions: tensile loads; external applied stresses, such as structural or
aerodynamic loads; overtightening of fasteners heads causing stresses in the
material underneath. Internal stresses, resulting from processing like bending or
rolling, nonuniform cooling after heat treatment (Mouritz 2012).
• Environmental agents: aqueous electrolytes, especially when containing H2S
(high-strength steels). Ammoniacal solutions (α-brass), solutions containing
chlorine compounds (austenitic stainless steels, Mg- and Zr-alloys), high purity
steam (Ni-alloys), chloride, bromides, and iodide solutions (Al- and Ti-alloys).
These contaminants may be present in the outer atmosphere, in lubricants and
products deriving from exhaust condensation, just to mention a few main sources.
• Interactions: it is possible to distinguish between corrosion-dominated failure and
stress-dominated failure. In the first case, either intergranular or transgranular
fracture occurs. Solution requirements are alloy specific: NO3
solutions for
carbon steels; chloride solutions for Al-Zn-Mg alloys and for Fe-Cr-Ni alloys;
NH3 solutions for Cu-Zn alloys. In case of stress-dominated failure, brittle
fracture would generally occur, and the influence of the electrolyte is not that
critical, if not just as a precursor that promotes the formation of reaction products.
They may act as nucleation sites for fracture and also favor its subsequent
propagation, although under the prevailing action of the mechanical drive
(Jones 1992).
SCC presents an early stage, usually propagating slowly and corresponding to
crack initiation. Afterward, a steady-state crack propagation follows, rapidly leading
to the third and final stage, featuring a sudden complete failure. The unpredictability
of the whole time evolution of SCC, owing to the extended hidden period of the
process, renders this phenomenon rather serious as concerns the assessment of a
component lifetime. For this reason, prevention is very much recommended. Several
strategies and relevant practical approaches are nowadays available (Table 8.3) and
are systematically implemented.
8.1 Introduction 455

Table 8.3 Prevention strategies against stress corrosion cracking

Mechanical approach Metallurgical approach Environmental approach
Compression stresses on Changes in the microstructure Anodic or cathodic protection
the surface
Limitation of the stress Optimization of the composition Environmental control (e.g.,
concentration (i.e., optimi- by choosing a proper alloy (e.g., limitation of chlorides pres-
zation of the geometry) duplex stainless steels instead of ence in the environment)
AISI 316)
Maintenance and control- Use of surface coatings Temperature control (the
oriented design lower the temperature, the
slower the corrosive
phenomenon)
Reduction of the applied – Use of inhibitors of corrosive
loads (e.g., increase in phenomena
dimensions of the
component)
Relaxation of residual – –
stresses

Regarding Al-alloys used in aerospace applications, SCC has been reported for the
front and rear spars of a fighter aircraft vertical fin. The spars were made of 7079-T6
Al-alloy forgings, age tempered to the peak strength condition. Stress corrosion
cracking was detected during standard inspections, carried out with the eddy current
NDT method. The spars were actually coated. Nonetheless, the coating was damaged
during the part assembly, using bolts. This provoked the exposure of the bare metal to
the outer environment and SCC developed in the material as a consequence of the
humidity combined with high tensile stresses of the installed fasteners (Wallace et al.
1985). SCC has also been reported to occur in a 7075-T6 fuselage frame flange of a
fighter aircraft. Cracks were detected during routine inspection and were identified
along a row of rivets in the upper flange, linking together adjacent rivet holes. The
microstructural characterization of the damaged part evidenced a heavily wrought
microstructure characterized by pancake-shaped grains. The crack plane turned out to
be coincident with the plane containing the longitudinal-long transverse load direction,
whereas the transverse-short direction was normal to the crack plane. Further analyses
confirmed the absence of sealants or protective substances, so that the damaged alloy
was actually free from any corrosion protection. SCC turned out to have initiated at
corrosion pits in the bore of the fastener holes. Stresses were originated both by the
part assembling and by the in-service operating loads, especially in the short-
transverse direction close to the fastener holes. In this latter area, stresses were higher
than the threshold value for SCC (Wallace et al. 1985).
Concerning Mg-Alloys, stress corrosion cracking was identified as the main
reason for the failure of an AZ91C-T6 brake housing casting of a popular light
aircraft. The failure was induced by cracks and leaks at the brake-liner union. SCC
may occur also in Mg-alloys, containing more than 2.5% Al (see Table 3.23), and
when the applied loads are higher than 90% of the yield strength.
456 8 Corrosion

Regarding SCC in aircraft steels, an interesting case history concerns the failure of a
Nitronic 60 (see Table 5.13) pin in a rear compressor variable vane lever arm assembly
of an F-16 fighter, reported to be occurred in February 1992. The engine failure caused
the aircraft crash in the Dutch city of Hengelo. At the time, the susceptibility of
Nitronic 60 to SCC was unknown, and, therefore, the failure was absolutely
unpredictable and unexpected. Subsequent analyses showed that the failure began
with crevice corrosion, and the fracture was caused by SSC. At the same time, many
more pins were subjected to investigations, and it was found that a large amount of
them contained cracks, caused by a combination of salt solutions in crevices between
the lever arms and the pins and by the presence of residual stresses in the pins caused
by the production process. In order to prevent further failures, the engine manufacturer
adopted two remediation measures: change of the pin material and progressive
worldwide replacement of all old lever arm assemblies with new ones. The Ni-based
Inconel 625 superalloy, immune to SCC in salt solutions, replaced the austenitic
stainless steel Nitronic 60 (Wanhill et al. 2011). Still regarding high-strength steels,
the failure of a cylindrical pressure vessel made of a 18Ni maraging steel (see Sect.
5.4.1) has been reported. The cylinder, containing He-gas, failed and fragmented while
having an internal pressure of 55 MPa. The component was the auxiliary power unit of
a missile and was supposed to leak and not to crack down if and when subjected to
overpressure conditions. Crack initiation occurred on the inner surface of the cylinder
and propagated through about 80% of the wall thickness before the final overload
rupture was achieved. Before this final stage, the crack propagation occurred in a slow
growth mode, typical of SCC. The slow crack growth region was characterized by the
presence of corrosion products and intergranular fracture. These latter factors
suggested an environmentally induced failure. In fact, the inner surface of the cylinder
showed the presence of a thin layer of hydraulic fluid that provoked the corrosive
attack to the vessel wall (Wallace et al. 1985).

8.1.7 Hydrogen Embrittlement

Hydrogen embrittlement is a particular type of degenerative phenomenon, occurring

in metals as a consequence of hydrogen permeation, resulting in a catastrophic
failure. There are actually several types of hydrogen-related degradation processes
occurring in metals, in addition to hydrogen embrittlement, like hydrogen attack and
hydrogen blistering (ASM International 2003). Hydrogen attack consists in the
solutioning of this element in the lattice of the host metal, up to the solubility
limit. At this stage, hydride precipitation may occur and a general embrittlement is
also observed. Blistering refers to the formation of gaseous hydrogen, generally
localized on the material surface, resulting from metal catalyzed hydrolysis, as
observed in aluminum alloys (see Fig. 3.22). Since hydrogen is present in the
environment of several processes, e.g., hydrocarbon processing plants, melting and
welding processes without controlled atmosphere, etc., there are many possibilities
for metals to be subjected to hydrogen inward diffusion. Hydrogen embrittlement
8.1 Introduction 457

implies a sort of ductile to brittle transition (DBT), caused by the interaction of

hydrogen with metal, changing the chemical bond character from exclusively metal-
lic into partially covalent.
Although the process of hydrogen embrittlement has been extensively analyzed
in the literature, the nature of the atomistic mechanism behind the phenomenon is
still controversial, and three main models can be considered:
• Hydrogen-enhanced decohesion (HEDE). This process results from the reduction
in the cohesive strength of the atomic bonding, produced by hydrogen accumu-
lation in the crystalline lattice. This mechanism is coherent with the fact that the
relevant damage occurs at the crack tip fracture preferentially. In this region, the
local tensile load exceeds the maximum value of the local cohesion strength,
which was lowered by the presence of hydrogen. HEDE is emerging as the
dominant mechanism for hydrogen environmental assisted cracking (HEAC)
and for internal hydrogen assisted cracking (IHAC) in high-strength alloys that
are not prone to the internal formation of metal hydrides (Popov 2015).
• Hydrogen-enhanced localized plasticity. Different from hydrogen embrittlement,
HELP (hydrogen-enhanced localized plasticity) is a phenomenon in which hydro-
gen may act as an enhancer of the dislocation mobility, resulting in a localized
plasticity and a consequent subcritical crack growth, still exhibiting, though,
brittle features on the macroscopic scale. HELP model is based on the effect of
hydrogen either in reducing the stacking fault energy or preventing the disloca-
tions from interacting with other obstacles, allowing them to move under lower
applied stresses. The evidence for HELP has been obtained through in situ ETEM
(environmental transmission electron microscopy) of specimens subjected to
plastic deformation either in vacuum or in hydrogen atmosphere. These observa-
tions revealed at a microscopic level the dislocation movement and their
increased number in pile-ups, when hydrogen was introduced in the microscope
sample region, coherently with a reduction in the stacking fault energy. A plastic
deformation similar to the one described above, accompanied crack growth
observed in the TEM in situ tests (Popov 2015).
• Hydride-induced embrittlement. Stress-induced hydride formation and cleavage
are established hydrogen embrittlement mechanisms, as confirmed by a large
theoretical and experimental support. Hydrides nucleate and grow in the volume
ahead of a propagating crack, according to a sort of autocatalytic process. The
brittle fracture of carbides, if present, may contribute to the process (ASM
International 2003).
Hydrogen embrittlement was analyzed by Lee and Krakkainen (2018), in con-
nection with the discovery of a 1 m long crack in the inboard trailing edge of the
horizontal stabilizer of a jet aircraft during the post flight routine inspection (see
Fig. 8.8). The stabilizer is a control surface (see also Fig. 3.1), and its failure can lead
to serious consequences for the loss of aircraft maneuverability. The jet fighter,
whose stabilizer failed, belongs to a group of aircrafts that had been operated for
more than 40 years. The failed stabilizer structure consisted of an aluminum alloy
square honeycomb core with high-strength steel skin that, before failing, accounted
458 8 Corrosion

Fig. 8.8 Schematic of the stabilizer and of the location of the detected crack. (Redrawn from Lee
and Karkkanien 2018)

for more than 700 h of time-flight. Before this failure, there were only two reported
cases of a complete loss of the inboard stabilizer trailing edge. The cracking pattern
usually associated with this type of structures starts from a bolt hole edge and
extends along adjacent holes, due to fatigue. In this case, fastener holes were not
characterized by any crack propagation, but the crack was localized at a certain
distance from the fasteners holes line with a propagation path that was almost linear
across the entire panel. The analyses of the fracture surface were the key point in
determining the real cause of the failure. SEM observations revealed various mor-
phologies, although the dominant fracture mode, accounting for more than 75% of
the total fracture surface, was an intergranular fracture. The crack initiation region
was characterized by a fatigue morphology, while the transition region, between
fatigue cracking and intergranular fracture, showed a quasi-cleavage morphology.
The fracture surface close to the overload area exhibited typical ductile micro-void
coalescence. This fracture morphology is typical of hydrogen-induced cracking in
high-strength steels.
Samples were taken from the fracture surface and from the opposite side of the
skin panel, in order to measure the concentration of hydrogen. The results indicated a
concentration of 154 ppm at the fracture surface, while on the opposite side, the
concentration was 9.5 ppm only. It is worth mentioning that a concentration of
hydrogen in the 100–1000 ppm range is sufficient for degrading mechanical prop-
erties of several steels significantly. The part failure is induced by the piling up
of residual internal stresses, resulting from maintenance activities, missing shims,
and aerodynamic loads, generated during the flight. Moreover, the notorious sensi-
tivity of honeycomb structures to moisture intake, like that coming from fastener
holes and trapped into the core cells, may have worsened further the situation (Lee
and Karkkainen 2018).
8.2 High-Temperature Corrosion 459

8.2 High-Temperature Corrosion

High-temperature corrosion occurs in several technological fields. Similar to elec-

trochemical corrosion, the relevant phenomenology influences the in-service perfor-
mances, particularly of the engine components, and represents an important lifetime
limiting factor. In aircraft and aerospace structures, high-temperature corrosion is
particularly interesting for the metallic alloys used in the engines, although it is an
issue also for the skin and fuselage of special structures, like the thermal protection
system of the Space Shuttle Orbiter (Buckley et al. 1981; Korb et al. 1981). In this
section, the main principles of oxidation will be illustrated with reference to simple
models, based on the Ellingham diagrams, as concerns the thermodynamics, and on
the Wagner theory for the main kinetic aspects. Although high-temperature oxida-
tion will be the leading topic, a few aspects of mixed gas phase corrosion, with
particular regard to sulfidation, will be also recalled. These principles are useful to
investigate the reactions occurring in real systems and laboratory testing rigs.
Moreover, they provide indications and first principles for the development of
high-temperature corrosion protection systems, like the coatings, discussed in Sect.
7.3, and relevant strategies. An alternative protection approach will be presented,
based on the nanocrystallization of the alloy surface. Examples of the systems for
which this protection approach has provided good results will be illustrated. The last
part of the section is devoted to the important phenomenon of hot corrosion. This
complex process involves not only a gaseous environment but also sulfate deposits
that condensate on the alloy surface. This introduces a further attack mechanism,
based on the formation of non-protective reaction products and the dissolution of
potentially protective oxide layers, present on the alloy surface before salt deposition
or even forming in the presence of the condensate deposit. The complexity of these
systems renders effective characterization protocols, using laboratory tests, very
important to interpret real-system phenomena, which in fact are presented in the
final part of the section.

8.2.1 High-Temperature Oxidation: General Thermodynamic

and Kinetic Aspects

From a chemical point of view, an oxidation is a reaction in which a metal forms more
stable chemical compounds, like oxides, nitrides, carbides, etc. A general high-
temperature oxidation reaction between a metal, M, and oxygen, O2, can be written as:

MðsÞ þ O2 ðgÞ ¼ MO2 ðsÞ ð8:4Þ

460 8 Corrosion

The condition for this reaction to occur depends on the higher stability of the
oxide phase with respect to the reagents in the initial state. This condition can be
expressed in terms of the standard free energy of the oxidation reaction:

ΔG0 ¼ G0MO2 ðsÞ
G0O2 ðgÞ
G0M ðsÞ ¼ RTlnðPO2 Þ ð8:5Þ

where:
• ΔG0 is the standard Gibbs free energy variation of the oxidation reaction.
• G0MO2 ðsÞ is the standard Gibbs free energy of the oxide.
• G0O2 ðgÞ is the standard Gibbs for oxygen.
• G0M ðsÞ is the standard Gibbs free energy of the reacting metal.
• R is the gas constant.
• T is the absolute temperature.
• PO2 is the partial pressure of oxygen.
Equation 8.5 can be made simpler by the assumption that the activities of the
condensed phases, i.e., the oxide product and the metal, both considered to be solid
in the present example, can be taken equal to unity. The first outcome of Eq. 8.5 is
the value of the oxygen pressure in equilibrium with the oxide, i.e., the minimum
oxygen partial pressure needed to form a stable oxide.
The British chemist, Harold Johann Thomas Ellingham, created a graphical
representation of experimentally determined variations of the standard Gibbs free
energy (ΔG0) versus temperature for the oxidation and sulfidation of several metals
as a function of the temperature. In Fig. 8.9, the Ellingham diagram for the oxidation
reactions is shown. Similar diagrams are available also for other gas phase reactions
and are extremely useful to predict the system behavior as a function of the
temperature and relevant gas phase pressures. Starting from a general simplified
expression for ΔG0 in terms of the standard enthalpy, ΔH0, and entropy, ΔS0, of the
reaction, the Eq. 8.6 can be written:

ΔG ¼ ΔH 0
TΔS0 ð8:6Þ

This expression is coherent with the linear trend of the curves in Fig. 8.9. The
intercepts of these lines are the standard enthalpies of the oxidation reactions; the
slopes are the standard entropies. The slope changes observed in some of the reaction
lines correspond to the melting (M points) and boiling (B points) points. The positive
slope usually displayed by the free energy lines indicates that a decrease in entropy
would accompany the oxidation reactions. This is so when the oxide is a condensate
phase, since the corresponding configurations are definitely less than the configura-
tions of the reagents, including also a gas phase. Of course, the carbon dioxide
formation occurs at constant entropy, since 1 mole of gaseous reagent (oxygen) is
replaced by 1 mole of another gas phase, i.e., CO2. A negative slope is there for the
CO formation reaction, corresponding to an entropy increase, since 1 mole of
oxygen is replaced by 2 moles of carbon monoxide, both phases being gaseous.
8.2 High-Temperature Corrosion 461

Fig. 8.9 Ellingham diagram for the formation of oxides. The point “O” is used as reference to draw
constant oxygen pressure lines, useful for graphical thermodynamic calculations, like the
equilibrium partial pressure of any metal-metal oxide couple. Alternatively, the pressure of oxygen,
as a function of the temperature, may be established by considering the H2/H2O and CO/CO2
activity ratios in water vapor and carbon oxides mixtures. The points “H” and “C”, instead of point
“O”, are used for the relevant graphical constructions (Kofstad 1988)

The position of the lines in the Ellingham diagram indicates the stability of the
relevant oxides, as a function of the temperature. The higher stability oxides
correspond to lower laying curves in the diagram. CaO turns out to be the most
stable oxide among those in Fig. 8.9; whereas the curves for copper and iron oxides
462 8 Corrosion

are located in the upper end of the diagram, indicating that they are the less stable
oxides in this set. This information provides an important indication on the oxide
species to be preferred for alloy protection. Indeed, a common approach is to
promote the formation of a protective scale of native oxides either on the alloy itself
or on the oxidation-resistant coating. The oxide phases of nickel, cobalt, and iron, the
elements on which the main families of superalloys are based (see Chap. 6), are by
far less stable than aluminum and chromium oxides. Suitable alloy additions (see
Sect. 6.4) or protective coatings (Sect. 7.3.1) are designed so to induce the prefer-
ential formation of the mentioned, more stable oxide phases. For the same reason,
the position of SiO2 turns out to be a good choice to protect niobium-based alloys, on
which the formation of a silica-rich surface layer, can be obtained with specific
alloying strategies or coating systems with the appropriate composition (see Sect.
9.2.2). In addition to the thermal stability, other features are fundamental for an
effective protection of the alloy, like adhesion to the substrate, compact microstruc-
ture, and good thermomechanical compatibility. These aspects are also influenced by
the kinetics of the oxidation reactions, which are paramount in most of real appli-
cations, including those involving high-temperature structural alloys. The actual
kinetics of the process is fundamental in the evaluation of the real life of the
component or to select the right coating system, capable to achieve the target
performances. Although rather simple and, in many respect, basic, the Wagner’s
theory of oxidation is considered herewith. Wagner’s theory is based on a purely
diffusive model of the reacting species through a comparatively thick oxide,
resulting in a parabolic kinetics. Its validity depends on some important
assumptions:
• Electrostatic interaction range is little if compared to the thickness of the oxide
layer.
• Negligible solubility of oxygen in the base metal.
• The oxidation process is assisted by the diffusion of ions and electrons through
the oxide layer.
• The oxide is perfectly adherent, i.e., there are no gaps at the interface with the
base metal.
• The reactions produce exclusively stoichiometric oxides.
• Thermodynamic equilibrium at the oxide-atmosphere interface and within the
oxide.
Wagner’s model was found to be qualitatively and quantitatively accurate for
many real cases, like sulfidation of silver and iron and oxidation of nickel, cobalt,
and copper. However, also for those alloys that are far too complex for an accurate
description in terms of Wagner’s model, still, the predictions of this approach can
profitably be used as reference for the interpretation of the real results. A condition
for diffusion to be the rate-controlling process is that a local equilibrium must be
achieved at the scale boundaries. This latter condition requires that the reactions
occurring at the metal-oxide and oxide-atmosphere interfaces are fast enough, and
are not the rate control processes (Young 2015).
8.2 High-Temperature Corrosion 463

Fig. 8.10 Oxidation reactions according to the Wagner’s model. External oxidation occurs when
the diffusion rate (νdiff) of the metal cations through the oxide layer is higher than that of the oxygen
from the outer atmosphere. On the other hand, internal oxidation occurs when the diffusion rate of
oxygen through the oxide layer is larger than that of the metal cations

Wagner’s model is illustrated with reference to Fig. 8.10, which describes a

general oxidation process: yM + x½O2 ¼ MyOx. Depending on material properties
and environmental conditions, metal oxidation generally involves two main mech-
anisms: external and internal oxidation (Fig. 8.10).
External oxidation is driven by the outward diffusion of metal cations through the
surface oxide layer; the alternative process, internal oxidation, occurs when the
inward oxygen diffusion prevails on the metal outward flux. External and internal
oxidation should be regarded as limit situations that in real systems would generally
coexist, to different extent depending on the oxidation regime and degree of
advancement of the process. In fact, three main stages can be identified for the
high-temperature corrosion phenomena, including oxidation. The initial stage, fea-
turing the metal or alloy surface in direct contact with the outer atmosphere, does not
require mass transport, and the products that start to form are those involving the
more reactive metals present in the alloy. At this stage, the rate of reaction is directly
proportional to the time. After this initiation stage, if the reaction products are
thermodynamically stable, adherent to the substrate, and reasonably compact, a
surface reaction layer starts to form on the alloy surface, and mass transport
mechanisms are needed to sustain the reactions. The kinetics, initially linear, starts
to level up, and a logarithmic law (direct or inverse) would typically describe this
further regime, and the scale thickness (or specific mass change), x, as a function of
time, t, can be expressed as:

x ¼ K logðt Þ þ A and x
1 ¼ B þ K 0 logðt Þ ð8:7Þ

where:
• K and K0 are the rate constants of the reactions.
464 8 Corrosion

• A and B are integration constants.

The logarithmic kinetics is compatible with the presence of a thin oxide scale on
the alloy surface, through which the transport of ions and electrons takes place. The
transport is not purely diffusive and includes also the electrostatic attraction among
opposite charges. As the oxide thickness increases, the system may get the features
described with reference to the Wagner’s model, which can therefore be used to
describe the steady- state condition. According to Fick’s first law of diffusion, the
kinetic law of the oxide scale growth under steady-state oxidation conditions is:

dx dC
¼D ð8:8Þ
dt dx

where:
• D is an equivalent diffusion coefficient, taking into account the contributions
from all reagents.
• dC
dx is the concentration gradient across the oxide layer thickness.

From the integration of Eq. 8.8, the following parabolic expression is obtained:

x2 ðt Þ
x2 ðt 0 Þ ¼ K 0p ðt
t 0 Þ ð8:9Þ

Assuming that x represents the oxide scale thickness increase starting from the
initial time, t ¼ 0, when the diffusive model starts to be applicable, the above
equation becomes:

x2 ðt Þ ¼ K 0p t ð8:10Þ

This is a very practical form for the interpretation and modeling of the experi-
mental data. Similar expressions are used for the other oxidation kinetics, i.e., linear
and logarithmic kinetics (Fig. 8.11).
It is common practice to consider the specific weight, i.e., the weight of reaction
products per unit surface area of the alloy, in order to obtain kinetics parameters
easily comparable. This can be done by multiplying Eq. 8.10 by the oxide scale
density (ρ), so to obtain an expression for the parabolic oxidation law that can be
written as:


ΔM 2
2
ðxρÞ ¼ ρK 0p t ! ¼ Kpt ð8:11Þ
A

where Kp is the parabolic constant of the oxidation reaction, ΔM is the weight

change due to oxidation, and A is the alloy sample free surface. In case of laboratory
tests, the alloy free surface area is directly measured on the test sample before
8.2 High-Temperature Corrosion 465

Fig. 8.11 The linear,

logarithmic, and parabolic
oxidation kinetics.
(Redrawn from McCafferty
2010)

starting the oxidation experiment. Since Kp includes a diffusion coefficient (see

Eq. 8.8), it may be expressed with an Arrhenius-like relation:

ERT
act
Kp ¼ K0 ð8:12Þ

from which the equivalent activation energy, Eact, of the process can be evaluated.
Although the solution of Wagner’s model has been obtained starting from very
simple assumptions, still, as it was the case for Ellingham diagram, also Wagner
approach provides interesting information concerning the oxidation kinetics of real
systems. The parabolic constants for the formation reactions of several oxides are
shown by Fig. 8.12. Interestingly, the Al2O3 and Cr2O3, mentioned already as
thermodynamically more stable than nickel, cobalt, and iron oxides, would also
feature slower formation kinetics, which rationalize further the important role of
these phases in the oxidation protection strategies.
The steady-state condition would last as long as the outward metallic ion flux is
sufficient to guarantee the formation of a protective oxide scale also in case of
damages and spallation phenomena. The thermogravimetric tests conducted with
thermo-analytical instrumentation under conditions simulating those faced under
real service can be used to predict the real alloy behavior. For instance, in Fig. 8.13,
the ideal “theoretical” parabolic behavior (dotted curve) is compared with a “real”
system oxidation curve, in which two different oxidation stages can be identified,
each one featuring a specific parabolic constant. This behavior can be explained with
the compositional changes in the base alloy during the process. Another possible
phenomenon is illustrated by the other “real” curve, describing a kinetics involving
cracking and partial spallation of the surface oxide layer.
466 8 Corrosion

Fig. 8.12 Parabolic

constants of several oxides
as a function of the
temperature. (Redrawn from
Okafor and Reddy 1999)

Fig. 8.13 Schematic of

possible real kinetic curves
for the oxidation of alloys
under different conditions:
specific weight change
(mass change over sample
area) versus time

The partial detachment of the oxide scale, occurring in correspondence of the

sharp drops visible in the relevant thermogravimetric curve, may be originated by
mechanical stresses associated with thermal cycling. In the long run, the alloy, or the
protective coating, is no longer capable to provide a sufficient amount of reagents to
mend the damaged scale. The damage can therefore propagate to the alloy and this
would correspond to the end of life of the component. This latter phenomenon is
known as “propagation” or final breakdown stage. When partial detachment of the
8.2 High-Temperature Corrosion 467

surface oxide scale occurs, the kinetics may resume the linear trend, if the metal gets
back in direct contact with the gaseous atmosphere.
Oxidation-resistant alloys should exhibit prolonged steady-state regimes, char-
acterized by a slow progression of the corrosion damage. Although high-temperature
oxidation phenomena are complex, experimental approaches and modelling tools are
available, affording important inputs to design more effective protection strategies,
as illustrated in the following (Young 2015). Wagner’s model discussed so far would
strictly refer to the gas phase oxidation of pure metals, which is actually quite a
peculiar situation, certainly not interesting for real high-temperature applications,
including aerospace engines. Actually, as discussed with reference to titanium alloys
(see Chap. 4), steels (see Chap. 5), superalloys (see Chap. 6), and refractory alloys
(see Sect. 9.2), a common requirement for these materials, in order to extend at
higher temperatures their usability range, has always been to have adequate high-
temperature corrosion resistance capabilities. The common protection strategy, as
emerging also from the above discussion, relies on the design of alloy bulk compo-
sition or surface coating capable to form on their own surface a layer of reaction
products (initiation stage) suitable to protect the underlying alloy for a conveniently
long time, corresponding to the steady-state oxidation regime. These aspects will be
considered next with reference to important examples in the field of aerospace
alloys.

8.2.2 Alloy Oxidation

In the previous section, Wagner’s model has been introduced and discussed with
reference to gas phase oxidation of pure metals, which is not the situation usually
faced in real applications, including the aerospace field, where alloys are typically
employed. Therefore, the study and control of high-temperature oxidation of real
systems require accounting for the more complex reactions that multi-elemental
alloys necessarily involve. However, the main principles introduced as regards pure
metals are still valid, and also the protection strategy aiming at reducing the effects of
high-temperature gas phase reactions is still the same as those discussed in the
previous section. In particular, obtaining on the alloy, or coating, surface, a protec-
tive scale made of reaction products of the base alloy elements with the outer
environment is still the standard protection approach, whose effectiveness depends
upon thermodynamic stability, slow growth rate, low diffusivity, compactness, good
adhesion strength, and capacity of self-repairing, in case of damage. With the help of
fundamental parameters, like energy of formation, reaction constants, and relevant
activation energies, which can be obtained from laboratory tests, the design of alloys
and protective coatings has been possible, resulting in the extension of the lifetimes
of real components. A typical example in this regard is the development of
oxidation-resistant steel alloys, in which the interplay of chemical, i.e., chromium
alloying, and structural, i.e., austenite stabilization with nickel addition, factors have
been profitably exploited. In Fig. 8.14, the progressive reduction in the oxidation rate
468 8 Corrosion

Fig. 8.14 Effect of the Cr

content on the oxidation
resistance of steel alloys at
1000  C. (Redrawn from
ASM International 2003)

at 1000  C of diverse steel alloys, featuring increasing chromium concentrations, is

shown. Higher concentrations of chromium modify the character of the surface
oxide scale. The formation of spinel Fe(FeCr)2O4 reduces the diffusivity of the
iron oxide scale, which is typically composed of all three polymorphs: wüstite
(FeO), magnetite (Fe3O4), and hematite (Fe2O3).
A further improvement is introduced by alloying with nickel and relevant stabi-
lization of the γ phase (see Appendix 4 and Chap. 5), whose closest packed structure
has an intrinsically lower diffusivity than the “open” bcc structure of ferrite. This
effect is still there if nickel is either the majority, like in nickel-based superalloys (see
Sect. 6.4), or is added in significant concentrations, above 20 wt% (Fig. 8.14), like in
iron-based superalloys (see Sect. 6.3).
Dealing with the high-temperature oxidation behavior of superalloys (see Sect.
6.4.2) and oxidation-resistant coatings, like diffusion and overlay coatings (see Sect.
7.3.1), the importance of Al2O3 and Cr2O3 has been discussed already. The thermo-
dynamic aspects emerging from the Ellingham diagram (Fig. 8.9) confirm the
elevated stability of these oxides, as compared to the oxide of the majority elements
present in the different groups of superalloys, i.e., nickel, cobalt, and iron. Similarly,
the refractory alloys (see Sect. 9.2) can rely on the formation of a layer of native
SiO2, amorphous at sufficiently high temperatures, for resisting better to high-
temperature aggressive environments. For the lack of grain boundaries and associ-
ated diffusion paths, and for the adaptive capacity to the changes of the underlying
substrate alloy, the silica layer represents a particularly effective protection barrier.
However, the aluminum, chromium, and silicon oxides are also those featuring the
comparatively lower kinetic constants (Fig. 8.12). Therefore, their preferential
formation requires to overcome kinetics constraints, through a selective oxidation,
in order to overtake faster growing, not protective, competitor oxides. This process
can be enhanced through the design of suitable alloy and protective coating
8.2 High-Temperature Corrosion 469

Fig. 8.15 Schematic representation of the evolution of the surface oxide layer of a Ni-Cr-Al alloy
in subsequent time steps from (a–f). Detailed description of each phase is given in main text.
(Redrawn from Giggins and Pettit 1971)

compositions, taking into account the compositional changes in the outer region of
the base alloy or coating, in the oxide scale and in the local atmosphere. The case of
alumina- and chromia-forming nickel-based alloys has been thoroughly considered
in Giggins and Pettit (1971), a study that provides an excellent insight into the main
oxidation reactions occurring in Ni-Al-Cr systems. The outcome is interesting for
interpreting the oxidation behavior of both nickel-based superalloys and overlay
coatings. With reference to Fig. 8.15, it turns out that, irrespective of the actual alloy
composition, the early stages of oxidation occur by a direct reaction of the alloy with
the oxygen present in the working atmosphere, with a very limited contribution from
diffusive phenomena. The main oxide phases that form under these conditions are Ni
(Cr,Al)2O4 and NiO (Fig. 8.15a). Depending on the actual composition of the base
alloys, these two phases may be “doped” to different extent by outward diffusing
chromium and aluminum atoms. Moreover, since nickel oxide and the spinel are not
preventing oxygen inward diffusion, the formation of a mixed layer, made of
chromia and alumina precipitates, can be observed (Fig. 8.15b), favored by the
oxygen pressure reduction and relative increase in the concentrations of these two
elements, determined by the outward diffusion of nickel. After this initial transient,
the oxidation kinetics enters a steady-state condition featuring a parabolic oxidation,
with products having different characteristics, depending on the alloy composition.
470 8 Corrosion

Fig. 8.16 Isothermal diagram at 1000  C at 0.1 atm O2 from showing the classification of Ni-Cr-Al
alloys as concerns their oxidation behavior. Group I: external scale of NiO; subscale made of a
mixture of Cr2O3 and Al2O3. Group II: the initial oxidation transient would feature the formation of
a non-protective scale made of a mixture NiO and the spinel Ni(Cr,Al)2O4. Group III: Cr2O3 and
Al2O3 are the prevailing phases forming on the alloy surface. (Redrawn from Giggins and Pettit
1971)

In case of nickel alloys with low concentrations of chromium and aluminum (Group
I), the external oxide layer is made of NiO, whereas the internal oxidation leads to
the formation of a discontinuous mixed Cr2O3-Al2O3 subscale (Fig. 8.15c). In Group
II, the subscale Cr2O3 becomes continuous, although, it would not prevent internal
oxidation of aluminum, since the scale contamination by the alloy elements would
still allow the inward diffusion of oxygen (Fig. 8.15e). Eventually, if aluminum
concentration is sufficiently high (Group III), the Al2O3 subscale becomes continu-
ous, and the oxidation kinetic is generally quite slow (Fig. 8.15f). This alloy
classification is summarized in Fig. 8.16, showing an isothermal (1000  C) section
of the ternary Ni-Cr-Al phase diagram, in which data points referring to the alloys of
the three groups are indicated (Giggins and Pettit 1971).
Although the results discussed so far refer to oxidation experiments carried out in
a low pure oxygen pressure, still, they provide reliable indications for the develop-
ment of coating systems, alloy materials, and relevant oxidation protection strate-
gies. The coexistence of aluminum and chromium reduces, with respect to binary
Ni-Al or Ni-Cr alloys, the concentration that each one of these elements should have
in order to produce a continuous oxide scale. This aspect has been exploited in the
8.2 High-Temperature Corrosion 471

Table 8.4 Standard free Sulfides ΔG0 [kJ/mole] Oxides ΔG0 [kJ/mole]
energy of formation of some
1/3Al2S3
219 1/3Al2O3
429
sulfides and relevant oxides at
1000  C (Reed 2006) 1/3Cr2S3
135 1/3Cr2O3
269
CoS
80 CoO
136
FeS
86 FeO
176
MnS
190 MnO
294
NiS
88 NiO
127

development of MCrAlY overlay coatings (see Sects. 7.3.1 and 7.3.2) and has
important consequences in keeping the concentration of chromium at sufficiently
low levels, thus reducing the risk of forming the TCP σ phase (see Sect. 6.4.2).
An optimum oxidation protection is achieved when the external selective oxida-
tion produces a stable, compact and adherent scale. In real working atmospheres,
containing other gaseous species than oxygen, the attempt is still to obtain a
protective surface layer mainly made of stable oxides. In this respect, the thermo-
dynamic drive may help, considering the higher stability of most common oxides
than the relevant compounds from which they may form. As an example, in
Table 8.4, the standard free energies of some sulfides and of the corresponding
oxides are listed.
The classification of superalloys and oxidation protective coatings in terms of
alumina and chromia formers is coherent with the compatibility of their parabolic
oxidation constants with those of the two oxides. However, two aspects should be
considered. The equivalence of the kinetic constants refers to steady-state condi-
tions, which are not always the same all over the lifetime of a real component. In fact,
several high-temperature oxidation regimes can be rather envisaged for an alloy
exposed to real operating conditions (see also Fig. 8.13). Moreover, reactions
involving multicomponent alloys result in the formation of surface scales containing
also other compounds, e.g., nitrides, than just the oxide phases described so far (Liu
et al. 2010; Pei et al. 2017). Therefore, chromia and alumina should rather be
regarded as the majority phases in real surface scales, whose real composition can
be better predicted on the basis of more accurate models, including the effect also of
the constitutional defect structures present in the growing scale (Sato et al. 2011).
An effective protection by the surface oxide scale depends also on its degree of
adhesion to the substrate. The surface scale has to resist to the shear stresses
determined by the volume changes induced by internal, to the scale, oxidation
reactions. Another shear stress contribution may come from the mechanical
mismatch associated with temperature gradients across the alloy-scale interface.
Additions of reactive elements, like yttrium, zirconium, just to mention the most
used ones, have introduced remarkable improvements in the adhesion strength of the
oxide scale to the alloy, as discussed already with reference to Ni-based superalloy
composition selection (see Sect. 6.4.2) and overlay coatings (see Sect. 7.3.1). These
additions are either pure elements or oxide particles, e.g., Y2O3. This latter condition
is rapidly reached anyway, owing to the elevated affinity to oxygen of these
elements, which are thereby easily oxidized by both the oxygen present in the
472 8 Corrosion

working atmosphere or scavenged from the alloy matrix. In case of elemental

additions, concentrations not exceeding 1–2% are recommended, in order to limit
any excessive contribution to the oxidation process. Several mechanisms have been
proposed to explain the beneficial effects of reactive elements on the adhesion
strength of the oxide scale (Hou and Stringer 1995; Huntz 1987; Stott and Wood
1987). The fine oxide particles act as heterogeneous nucleation sites for the scale
forming on the alloy surface. This results in a finer grain size of the surface oxide
layer that will exhibit a higher intrinsic plasticity and fracture toughness. Conse-
quently, internal and interfacial thermomechanical stresses will be better accommo-
dated. Owing to their elevated reactivity with oxygen, but also with sulfur, these
elements tend to act as purifying getters of the alloy. Since the active elements are
usually much slower diffusers than the other scale-forming elements, like aluminum
and chromium, the particles of oxides, mostly, and sulfides are localized at the scale-
alloy interface. This has a twofold beneficial effect. The first regards the mechanical
gripping action that these particles would deploy, behaving like a sort of roots of the
outer scale (pegging effect). Moreover, the interfaces of the oxide particles with the
alloy matrix act as vacancy sinks (see a similar effect also in ODS alloys in nuclear
plant applications in Sect. 9.3), thus keeping the Kirkendall voids, produced by
vacancy coalescence during the oxidation process, smaller. These mechanisms
contribute to different extent to enhance the oxidation resistance of superalloys,
ODS superalloys, containing in their matrix a fine oxide particle dispersion (see Sect.
9.3), and overlay coatings (see Sect. 7.3.1).

8.2.3 Surface Nanocrystallization

In Sect. 7.3.1, the main coating systems used for the protection of superalloys against
high-temperature oxidation have been introduced. The base idea is to have a
reservoir of those metallic elements promoting the formation of a protective oxide
scale and, at the same time, to provide effective adhesion mechanisms of the oxide
scale growing on the coating surface. Diffusion coatings, based on aluminide phases,
e.g., NiAl, FeAl, and CoAl, have been for quite sometimes the best choice. In several
respects, overlay coatings have introduced important improvements in the oxidation
protection, thanks to a better design of their composition. Indeed, the coexistence of
chromium and aluminum affords a significant reduction of the minimum concentra-
tion, particularly of aluminum, necessary to the formation of a protective outer oxide
scale. An indirect consequence of this situation is that lower concentration gradients
are present across the coating-alloy interface, with obvious advantages for the
durability of the coating, whose degradation mechanisms are also dependent on
the interdiffusion into the base alloy. A further reduction in the interdiffusion
kinetics may come from the modification of the microstructure of the alloy surface
without significantly changing its composition. This can be achieved through surface
nanocrystallization (SNC), an approach that is attracting increasing interests, within
the broader field of the nanotechnological research (Lu and Lu 1999).
8.2 High-Temperature Corrosion 473

Notwithstanding the elevated concentration of grain boundaries and, consequently,

the higher diffusivities, several studies have demonstrated the possibility of improv-
ing the oxidation resistance of fine-grained alloys, with respect to their macro-
grained counterparts. The earlier studies on this matter regarded Ni-Cr alloys and
AISI 304 stainless steels (Baer and Merz 1980; Liu et al. 2000; Merz 1979). The
better reported oxidation resistance was ascribed to a more efficient oxidation of
chromium, to form a chromia-rich surface scale, due to an increased outward flux of
this element determined by the increased contribution of grain boundary diffusion.
Similar results have been reported as concerns the oxidation of nickel-based super-
alloys, both chromia and alumina formers, in experiments carried out in air up to
1000  C (Lou et al. 1996). The K38G nickel-based superalloy (Ni-16.3Cr-8.4Co-
3.81Ti-4.01Al-2.7W-1.8Mo-1.75Ta-0.76Nb-0.16C-0.01B wt%) was studied in the
as-cast bulk condition and as a sputter coating, with a thickness of 40 μm and an
average grain size of 0.1 μm (Lou et al. 1995, 1996). In this latter, lower oxidation
rates, than in bulk microcrystalline specimens, were observed under isothermal
conditions at temperatures of 800  C, 900  C, and 1000  C. The K38G superalloy,
owing to the comparatively high Cr concentration, is regarded as a chromia-former,
as indeed confirmed by high-temperature oxidation experiments on bulk samples.
However, particularly at higher temperatures, the scale exhibited a tendency to spall-
off. A dense, well-adherent alumina scale was observed instead on the nanograined
K38G coatings, coherently with the reported slower oxidation kinetics. The transi-
tion from chromia to alumina former, induced by the reduction in the average grain
size of the alloy specimens, is determined by the relevant reduction in the critical
concentration of aluminum necessary to the formation of a protective alumina scale.
As predicted by the theory of Wagner for the isothermal oxidation of binary alloys
(Kofstad 1988), the outward flux of solute atoms capable of sustaining external
oxidation leading to the formation of an external scale depends on the critical solute
concentration, NB (B ¼ Al in the present alloy system), that can be expressed as:
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
πg PO V mM
NB ¼ ð8:13Þ
2DB V mOX

where:
• PO is the oxygen permeability in the alloy matrix, i.e., the product of oxygen
concentration times its diffusivity.
• DB is the solute diffusivity.
• g is the critical volume fraction of oxide required for the transition from internal
to external oxidation. The value is often adopted as about 0.3.
• VmM and VmOX are the molar volumes of the alloy and oxide, respectively.
The diffusivity of the solute atoms increases upon reducing the average grain size
of the alloy since new diffusion paths, i.e., the grain boundaries in the nanostructured
alloy, become available. Therefore, an alumina scale can even form on a substrate
alloy having a sufficiently refined grain structure, although the same alloy would
474 8 Corrosion

rather behave like a chromia-former in a coarse-grained condition (Wang 1997).

This aspect will be considered further on, in connection to the oxidation of γ-TiAl
alloys.
It is now interesting to analyze the factors that are responsible for the improved
oxidation resistance of nanostructured alloy systems. The oxide scale growing on a
nanostructured surface would feature a nanometric microstructure too. This is so,
considering the heterogeneous nucleation action induced by the alloy grain bound-
aries. The refined microstructure of the oxide scale will better relieve internal stresses
and those induced by thermal cycling. Stress relief may more easily occur thanks to
the plastic flow of the oxide scale, assisted by the sliding along grain boundaries. In
this way, the spallation limit stress of the metal-scale interface is not reached, and the
surface oxide continues to protect the underlying alloy. Thermal stresses in the oxide
scale are proportional to the mismatch in thermal expansion coefficients of the scale
and the substrate material through the elastic modulus of the oxide. Dealing with
thermodynamic and mechanical properties of nanocrystalline materials, it has been
shown that they typically display a higher coefficient of thermal expansion and a
lower elastic modulus (Fecht 1990; Klam et al. 1987). These two conditions
contribute to the reduction of the stress in the surface oxide layer. The oxide scales
observed on oxidized nanostructured alloys or coatings are generally characterized
by a compact microstructure with limited porosity, also at the interface with the
alloy. This is another consequence of a substantial levelling up of the diffusion
coefficients of the involved atomic species. The important fraction of diffusive flux
occurring along grain boundaries, of the alloy only initially, and, afterwards, also of
the newly formed oxide scale, guarantees a fast diffusion to occupy the lattice sites,
thus keeping the vacancy concentration comparatively low and reducing the clus-
tering tendency to form larger pores.
The formation of a nanocrystalline layer on metallic alloys can be obtained with
different approaches that can be classified into three main categories: hybrid surface
nanocrystallization, deposition of surface coatings, and surface self-
nanocrystallization (Gupta et al. 2012; Lu and Lu 2004).
Hybrid surface nanocrystallization is achieved in two steps, conducted in
sequence or simultaneously. A possible approach might involve a surface mechan-
ical attrition treatment (SMAT) for the production of a nanostructured layer that can
subsequently be transformed through a metallurgical, chemical, or thermal process
(Lu and Lu 2004).
For the deposition of nanocrystalline coatings, several PVD techniques have been
used, also with nickel-based superalloys (Wang et al. 2014, 2015). By tuning the
deposition parameters, selected features of the nano-grained coatings can be
attained, as concerns average grain size, grain size distribution, adhesion strength
to the alloy substrate, residual stresses, and possible composition fluctuations. To
overcome the reported limited hot corrosion resistance in sulfate environment of
nickel-based superalloys (see Sect. 8.2.5), a two-layer structure was developed
(Yang et al. 2016). On top of the sputter-deposited nanocrystalline coating, having
the same composition as the base superalloy N5 (Ni-7.5Co-7.0Cr-1.5Mo-5.0W-
6.5ta-6.2Al-3.0Re), a second NiCrAlY coating was deposited, using an arc-ion
8.2 High-Temperature Corrosion 475

Fig. 8.17 Shot peening. Repeated multidirectional plastic deformation and consequent formation
of shear bands with a high density of dislocation arrays within the grain results in the formation of a
nanostructured surface layer, whose composition is substantially the same as the base alloy.
(Redrawn from Lu and Lu 1999)

plating PVD technique. The combined protective action of the two layers turned out
to be effective over a broad range of oxidation conditions, and, as expected,
improved also the sulfate hot corrosion behavior.
The main idea behind the self-nanocrystallization techniques is to produce nano-
crystalline surface layers by direct modification of the alloy microstructure, without
significant compositional changes. There are two main types of surface self-
nanocrystallization processes: mechanically induced surface self-nanocrystallization
(MISS-NC) and thermally induced surface self-nanocrystallization (TISS-NC). An
interesting example of the first process is shot peening, a SMAT widely used to
improve the mechanical properties of an alloy through its surface hardening (Sheng
2011). Nanocrystallization can be induced by shot peening and other SMATs if the
applied loads are sufficient to determine a severe plastic deformation field
(Fig. 8.17), accompanied by the formation of arrays of dislocations. Following
complex reactions of dislocation annihilation and recombination, the formation of
a domain structure is obtained, featuring small-angle grain boundaries separating
individual nanometric grains (Xia et al. 2015).
Mechanical surface treatments are not only meant to modify the properties of the
alloy surface layer but also its surface finish, like can be done using sand blasting.
Still using the superalloy N5, the effects of different roughness values, obtained by
sand blasting (using 100-mesh Al2O3 particles at a pressure of 0.2 MPa for 2 min),
mechanical polishing (1 μm diamond paste), and grinding (using a 2000 grit SiC
paper), have been investigated (Yang et al. 2016).
476 8 Corrosion

Fig. 8.18 Simplified Ellingham diagram, showing the position of Ti-O oxides with respect to
Al2O3. (Redrawn from Okafor and Reddy 1999)

The topographies obtained on the alloy surface and transferred also to the coating
deposited onto it influence the morphology and the spalling mode of the oxide scale.
The higher surface roughness of the sand-blasted alloy determines a layer-by-layer
spalling off of the scale, which would otherwise display a bulk fracture mode, in case
of smoother surfaces, like those obtained with polishing and grinding.
The TISS-NC processes are considered also in connection to the oxidation
protection of γ-TiAl alloys (see Sect. 4.3.2). As shown by the simplified Ellingham
diagram in Fig. 8.18, the formation of an alumina-rich scale would be favored in
view of their more negative energy of formation as compared to titanium oxides, in
particular the stable polymorph, TiO2 (Okafor and Reddy 1999). However, also in
TiAl alloys, notwithstanding the thermodynamic drive and the elevated concentra-
tion of aluminum, the kinetic constraint prevails, and titanium oxide tends to form
preferentially, as it can actually be inferred from the values of the parabolic constants
for oxidation shown in Fig. 8.12. Owing to their columnar morphology, the titania
grains cannot provide an effective protection, and any increase in the alumina
concentration in the surface oxide scale is very much appreciated. A proposed
strategy for obtaining this result is based on a TISS-NC approach. The surface of a
TiAl has been modified using rapid solidification, following the flash melting of the
alloy surface obtained with a plasma torch, generally used for the deposition of
TBCs (see Sect. 7.2) (Tomasi et al. 2002). This surface treatment results in the
formation of a nanostructured surface layer, with a thickness of a few micrometers
(Fig. 8.19a), which introduces remarkable improvements in the quality of the oxide
scale that forms on the alloy surface during isothermal oxidation treatments carried
out in air at 800  C and 900  C. The relevant thermogravimetric curves (Fig. 8.19b)
show the weight increase of alloy samples, having the same composition, but two
8.2 High-Temperature Corrosion 477

Fig. 8.19 (a) Cross section of the plasma-treated TiAl, showing a surface layer with a nanometric
grain size; (b) thermogravimetric curves showing the oxidation kinetics in air at 800  C and 900  C
of specimens of the TiAl alloy, tested in the original surface condition and after plasma surface
treatment. (Images property of the Authors, unpublished data)

Fig. 8.20 Cross-sectional micrographs showing the oxide scale grown in a TiAl-based alloy. (a)
Oxidized surface in its original condition; (b) surface after a plasma TISS-NC treatment. (Images
property of the Authors, unpublished data)

different surface conditions. The oxidation kinetics of the alloy in the pristine
condition turns out to be faster than the kinetics of the alloy with a nanocrystalline
surface layer. Similar results have been obtained using an alternative TISS-NC
approach based on laser ablation to produce the nanostructured surface layer
(Mandelli et al. 2007).
The composition of the nanostructured layer obtained on the surface of the TiAl
alloy is substantially unchanged with respect to the base alloy. The protective action
of the surface nanocrystallization treatment is based on the balance of the diffusiv-
ities of aluminum and titanium, so to favor from the beginning the formation of an
alumina-rich oxide scale, compact and well adherent. This is actually the case, when
comparing the oxide scale grown on the untreated (Fig. 8.20a) and plasma-treated
(Fig. 8.20b) TiAl-based alloy.
478 8 Corrosion

The approach discussed herewith is one of the several developed so far for the
protection against high- temperature corrosion attack of TiAl-based alloys. As
emerges from review publications on this theme, research efforts have developed
different protection strategies, which can be selected in view of the specific appli-
cations and oxidation mechanisms (Leyens et al. 2006, Pflumm et al. 2015). Con-
sidering the growing interest in these alloys for their established position as strategic
materials for gas turbines applications, it is likely that novel techniques will be
developed soon.

8.2.4 Gas Phase Corrosion: Sulfur-Induced Corrosive

Processes

Combustion in gas turbine jet engines produces exhaust gases containing substances,
like SOx, NOx, COx, HOx, HxSOy, and HNOx (see also Fig. 2.21), that activate
corrosion mechanisms that are definitely more complex than a simple oxidation.
This, for instance, is the case of sulfidation and other reactions involving sulfur and
its compounds. Sulfur is usually present in fossil fuels, also refined and purified, like
those used to power aircraft engines, in which this element is usually below the
0.3 wt%. Sulfidation reactions, relevant to gas turbine engines, can either occur in
mixed gaseous environments, with sufficiently high activities of oxygen, to keep the
oxides present on the alloy surface stable, or under a salt deposit (i.e., salt-induced
attack) that may form from the condensation of the sulfates present in the exhaust gas
stream. Reducing environments, with partial oxygen pressures below the dissocia-
tion pressures of the oxide containing chemical species, would be a further option,
although not that interesting for the oxygen-rich atmospheres of gas turbine engines.
Thermodynamic aspects can be dealt with relevant Ellingham diagrams (see
Fig. 8.9). However, considering that very often sulfidation environments would
feature both oxygen and sulfur partial pressures, metal-sulfur-oxygen stability dia-
grams are very useful in predicting the stable phases that actually form on the alloy
surface, depending on the local relative pressures. The formation of metal sulfides
may occur as a result of a solid-state diffusion through the oxide scale, sustained by
the gradient of sulfur existing between metal/oxide and oxide/atmosphere interfaces
and determined by the reduction in the partial pressure of sulfur due to the presence
of the oxide scale. The resulting parabolic kinetic law of sulfidation features rate
constants definitely higher than the corresponding oxides. Moreover, the reaction
products are generally not protective. Therefore, this corrosive process is generally
regarded as more aggressive and severe to the metal alloys than pure oxidation.
Sulfidation rate constants for various metals under specific temperature and partial
pressure of S2 are listed in Table 8.5. In this table, the ratios between the sulfidation
and the oxidation kinetic constants are also indicated. They provide a direct evidence
of how faster sulfidation typically is with respect to oxidation, owing to the elevated
reactivity of metals versus sulfur. For instance, the sulfidation rate of chromium is
8.2 High-Temperature Corrosion 479

Table 8.5 Sulfidation rate data for some pure metals. K(S) is the sulfidation rate constant, K(O) is the
oxidation rate constant, and EA is the activation energy for the growth of the sulfides (Young 2015)
Metal T ( C) pS2 (atm) EA (kJ/mol) K(S) (g2/cm4s)a K(S)/K(O)b
Co 650–800 1 142 3  10
6 450
500–700 0.013 78 – –
800–1000 2  10
3 210 – –
Cr 800–1000 1 192 3  10
8 >104
700–900 1 72 1  10
7 –
Fe 650–750 1 92 – 102
700–850 0.26 84 1  10
5 –
650–900 0.13 70 – –
600–1000 1  10
4 61 – –
1  10
8 12 – –
527–697 2  10
6 26 – –
Mn 800–1000 10
8–0
4 82–96 – 1
700–1000 1 130 4  10
9 –
550–711 5  10
9– 72 – –
4  10
3
Mo 850–1050 50% H2S 199 8  10
11 (900  C) –
700–950 1.3  10
3 147 7  10
12 –
Nb 800–1200 1 58 1  10
13 –
Ni 420–640 1 92 8  10
7 (600  C) 1  105 (600  C)
Ta 800–1000 1 64 7  10
10 –
W 380,980 50% H2S 135 3  10
10 –
a
Rate is obtained at 800  C, pS2 ¼ 1 atm unless differently specified
b
Ratio between sulfidation and oxidation rate constants at 800  C, unless differently specified

about four orders of magnitude larger than the corresponding oxidation rate. Since
chromium is a key constituent of several alloys that have to withstand high-
temperature working conditions, this is certainly an issue to be considered as
concerns the corrosion resistance of structural components and protective coatings
(Young 2015). The presence of cracks in the oxide scale or even detached parts may
produce a local change in the relative pressures of sulfur and oxygen. The oxidation
occurring at the metal/oxide interface, determines an increase in local pressure of
sulfur, with a consequent rise in the sulfidation rate. Since sulfide formation involves
also elements like chromium and aluminum, alloy surface depletion takes place, thus
reducing the possibility for the scale self-healing. This is a critical condition, since it
may lead to a fast sulfidation kinetics and propagation of the attack to the alloy.
480 8 Corrosion

8.2.5 Hot Corrosion

The condensation of salt deposits, in particular, sodium sulfate, on hot parts of gas
turbine engines, triggers important corrosion phenomena, usually named as hot
corrosion, to distinguish them from electrochemical, low-temperature corrosion
(see Sect. 8.1). The main components of the salt deposit come from combustion
exhaust gases, mainly, but not exclusively, sulfur compounds, and the outer atmo-
sphere. In marine environment, the air entering the engine contains sodium chloride,
so that reactions like the one given by Eq. 8.14 may very well occur, considering the
high air to fuel ratio and the elevated stability temperature of sodium sulfate.

1
SO2 þ H2 O þ 2NaCl þ O2 ! Na2 SO4 þ 2HCl ð8:14Þ
2

High-pressure turbine vanes and blades are the engine components more liable to
salt condensation, when their temperature falls below the dew point of sodium
sulfate, whose melting point is 884  C. Sodium sulfate is not the only substance
interesting for this phenomenology, and other sources of contaminants can be
identified. Seawater, from which the aerosol carrying sodium chloride into the
engine comes, contains also other impurity elements, like Mg, S, and K, just to
mention those more interesting for corrosive phenomena. As concerns the fuel,
sulfur is not the only impurity that may enter the hot corrosion reactions. Depending
on the provenance of the raw material (crude oil) and the refining process, concen-
trations of minor elements can be found in the fuel, and some of them are particularly
important in the present context (Molière and Sire 1993). Although the main
contribution to the formation of sulfate comes from the outer atmosphere, sodium
is also one of the impurity minority elements present in oil. Comparatively low
concentrations of sulfur are tolerated in aviation fuels, typically below 0.3 wt%,
whereas up to 4 wt% can be found in heavy oils. Elemental sulfur reacts with oxygen
to form the metastable SO compound, which is oxidized to form SO2 and, to partial
extent, SO3. Thermodynamic (especially over the temperature range 1300–2300  C)
and kinetic (below 1000  C) constraints limit to a few percent the conversion of SO2
into SO3. This is an important aspect, considering the strongly acidic character of
SO3 that may induce aggressive corrosive attacks, particularly in the 650–750  C
range (vide infra).
Another alkaline metal, i.e., potassium, may contribute to the formation of the
sulfate deposit (K2SO4). Its origin is still seawater, oil, and also water contamination
coming from the oil extraction process and ballast operation in the tankers, consid-
ering that very low limits for alkali metals altogether are admissible for aviation
fuels, i.e., below 1 ppm (Chevron Product Company 2007). Alkali metal, and also
earth-metal e.g., Mg, sulfates reduce the melting temperature of pure Na2SO4, with a
consequent speeding up of the corrosive reactions also at temperatures lower than
884  C.
8.2 High-Temperature Corrosion 481

Fig. 8.21 Qualitative plot of corrosion rate vs temperature featuring the two maxima,
corresponding to Type I and Type II hot corrosion. In the graphs, the main mechanisms and relevant
agents are indicated and discussed in the main text

Other hot corrosion-relevant elements are vanadium, nickel, and iron, structurally
embedded in porphyrin molecules, present in crude oil, with concentration and
character depending on the provenance regions (Ali et al. 1993). Vanadium may
contribute to the formation of corrosive deposits and ashes, with an impact that
depends on its concentration, ranging from a few ppm up to 1200 ppm. V2O5, whose
melting temperature is 690  C, is produced during combustion and may contribute to
the formation of vanadate by reacting with the sulfate components (see Eq. 8.20).
Hot corrosion phenomena can be classified as type I and type II (see Fig. 8.21).
Type I high-temperature hot corrosion (HTHC) has a maximum around 850  C, a
temperature that is compatible with the presence of a molten salt deposit on the
corroding alloy surface. Type II corrosion is observed at lower temperatures, i.e.,
over the 650–750  C range, where another maximum in the corrosion rate curve
(Fig. 8.21) is present. For this reason, type II corrosion is also called low-temperature
hot corrosion (LTHC). Temperatures at which type II process occurs are well below
the melting point of pure sodium sulfate, and this suggests that in this corrosion
regime, the deposit would be generally solid, unless the presence of “contaminants”
is reducing its melting temperature, so to have localized liquid regions in the deposit.
Type II corrosion, as discussed below, is ruled in several respects by the partial
pressure of SO3, which depends on the composition of the gaseous atmosphere and
the contribution from reactions involving the oxides of the refractory metals
contained in the alloy.
482 8 Corrosion

Before going into the details of the models developed for the interpretation of the
hot corrosion phenomenology, it is important to note that, owing to the compara-
tively little separation between the two temperature ranges of HTHC and LTHC, it is
not so uncommon to observe both hot corrosion types in different regions of the
same component, of course featuring different local conditions (Nicholls 2003).
Several models have been proposed to interpret the phenomenology of hot
corrosion and relevant experimental observations, obtained both from diagnostics
of components after exposure to real service conditions and from laboratory tests,
used to model real situations (see Sect. 8.2.6). The approaches based on the
sulfidation and the fluxing model are presented herewith.
The possibility of forming sulfides out of gas phase reaction has already been
discussed in Sect. 8.2.4. The presence of a Na2SO4 deposit on the alloy surface may
create sulfidation conditions, essentially through the oxidation of pyrolytic carbon,
present in the exhaust gases and due to an incomplete fuel combustion. The
oxidation of metals, present in the alloy, may also contribute to the oxidation
process. In general terms, the two reactions can be written as:

NaSO4 þ R1 ! Na2 O þ 3CO þ S

and ð8:15Þ
NaSO4 þ R2 ! Na2 O þ 3MO þ S

where:
• R1 and R2 are representing the reducing agents, i.e., pyrolytic carbon and an alloy
metal.
• M is a metal originally present in the alloy, particularly in the surface region
involved with the corrosive process.
Sulfur liberated by the reactions (Eq. 8.15) may contribute to the formation of
sulfides. In this regard, the salt deposit behaves like the oxide scale described in Sect.
8.2.4, since it reduces the partial pressure of oxygen and favors the formation of the
sulfide phases. This oxidation stage of the process can be regarded as the initiation of
the corrosive attack. It is followed by the propagation, featuring the oxidation of the
sulfides that start to form at the alloy/deposit interface:

NaSO4 þ MS ! MO þ Na2 O þ 2S ð8:16Þ

Since this oxidation produces more sulfur that may then react to form more
sulfides, the reaction can be regarded as autocatalytic. Table 8.6 gives the standard
free energy of formation of some metal sulfides with the relative formation reactions,
showing that, according to the free energies of metal sulfide formation, sulfur is a
strong oxidizing agent.
The reactions Eqs. 8.15 and 8.16 describe one of the key aspects of the hot
corrosion process, i.e., sulfidation (see Sect. 8.2.4). However, the proposed model is
not so flexible to describe other reactions that may take place during a corrosive
8.2 High-Temperature Corrosion 483

Table 8.6 Standard free energy (ΔG0) values normalized to ½ S2(g) of metal sulfide formation
reactions (Young 2015)
ΔG0 ¼ A + BT [J/mol]
Reaction T [K] A B
Fe + 12S2 ¼ FeS 411.15–1190
148.53 52.80
9
8Co + 12S2 ¼ 18Co9S8 <1053.15
165.77 83.28
4
3Co + 12S2 ¼ 13Co4S3 1053.15–1153.15
129.49 50.04
Co9S8 + 12S2 ¼ 9CoS 733.15–1108.15 Not linear with temperature
2Ni + 2S2 ¼ 2Ni3S2
3 1 1 733.15–808.15
165.56 78.03
2Ni + 2S2 ¼ 2Ni3S2-x
3 1 1 808.15–923.15
140.08 34.94
Ni3S2 + 2S2 ¼ Ni6S5
1 673.15–798.15
102.84 40.71
Ni3S2+x + 12S2 ¼ 3NiS 833.15–108.15
159.54 118.74
Ni6S5 + 12S2 ¼ 6NiS 673.15–833.15
111.92 62.93
Cr + 12S2 ¼ CrS 1173.15–1313.15
204.06 56.80
3CrS + 12S2 ¼ Cr3S4 973.15–1173.15
226.17 169.00
2Cr3S4 + 12S2 ¼ 3Cr2S3 873.15–1173.15
42.719 52.00
Mn + 12S2 ¼ MnS <1517.15
263.38 60.00
3Mo + 2S2 ¼ 3Mo2S3
2 1 1 973.15–1373.15
179.43 75.90
Mo2S3 + 12S2 ¼ MoS2 973.15–1373.15
179.72 82.20

attack. For instance, the sulfate deposit may form not on the alloy surface directly but
on the already existing scale of native oxides (see Fig. 8.22).
From a thermodynamic point of view, the stability field of various substances in a
molten salt deposit can be described by means of relevant phase stability diagrams,
calculated starting from the standard free energy of the involved phases. For the
sodium-sulfur-oxygen system, the stability diagram is given in Fig. 8.23, with
different stability fields depending on the relative concentration of the basic
(Na2O) and acidic (SO3) components in which sodium sulfate may decompose
according to Eq. 8.17:

Na2 SO4 ! Na2 O þ SO3 ðgÞ ð8:17Þ

Phase stability diagrams are evaluated at a fixed temperature, and the boundaries
between different fields are obtained by considering unit activity for the
corresponding condensed phases. The thermodynamic state of the molten salt at a
fixed temperature is therefore determined.
With a similar approach, the phase stability diagram including also the metallic
species undergoing a corrosive attack can be evaluated, as shown by Fig. 8.24 for the
Ni-O-S system at 900  C. Depending on the basic or acidic character of the salt
deposit, different reactions may occur. Particularly interesting for the hot corrosion
phenomenology is to consider the dissolution reactions (fluxing) that may involve
oxides, either originally present on the alloy surface before the deposition of the salt
484 8 Corrosion

Fig. 8.22 Hot corrosion by dissolution and reprecipitation of the oxide scale under a molten salt
layer based on the fluxing model, developed by Otsuka and Rapp. In this example, the system
exhibits a decreasing Na2O activity moving from the oxide/deposit interface outward. The pressure
of SO3 is decreasing in the opposite directions. This condition is typical of basic dissolution of the
metal oxide (basic fluxing). (Redrawn from Young 2015)

or resulting from the ongoing oxidation taking place under or within the deposit.
With reference to the diagram in Fig. 8.24, it turns out that in a Na2O-rich deposit,
i.e., a basic salt, the fluxing reaction that may take place is:

NiO þ Na2 ! Na2 NiO2 ð8:18Þ

whereas, in an acidic deposit (relatively higher SO3 pressure), the favored reaction
is:

NiO þ SO3 ! NiSO4 ð8:19Þ

The two reactions (Eqs. 8.18 and 8.19) can be taken as reference for basic and
acidic fluxing phenomena, respectively. Similar diagrams as the one in Fig. 8.24 are
8.2 High-Temperature Corrosion 485

Fig. 8.23 The Na-S-O

phase stability diagram at
900  C. (Redrawn from
Young 2015)

Fig. 8.24 Phase stability

diagram for the Ni-O-S
system at 900  C in the
presence of Na2SO4.
(Redrawn from Young
2015)
486 8 Corrosion

Fig. 8.25 Oxide solubility

curve in molten sodium
sulfate (Na2SO4) at 1200 K
(927  C) and pO2 of 1 atm.
(Redrawn from Rapp and
Zhang 1994)

available for single metals or, if necessary, for a combination of several of them, in
order to take into account also their relative interactions. Since these diagrams are
equilibrium ones, they cannot predict the effects of kinetic constraints that may limit
and modify the actual stability fields. Moreover, the composition of the deposit is
generally different at different depths in its thickness and may change for the release
of the reaction products and variations in the working atmosphere.
Considering the reactions for the fluxing model (Eqs. 8.18 and 8.19), it turns out
that the basic fluxing tends to increase the oxide ion concentration in the deposit.
A concentration gradient of the reacting species, like shown in Fig. 8.22, may
result in the solutioning of the oxide at the scale/salt interface and its re-precipitation
inside the salt and at the salt/atmosphere interface. This picture suggests that the hot
corrosion process is sustained by the presence of a negative solubility gradient, as
proposed in Rapp and Goto (1981). Referring to the curves in Fig. 8.25, it can be
noticed that the solubility of several oxides in molten Na2SO4 exhibits relatively
larger values in the two lateral branches, corresponding to acidic and basic salt
conditions. In between, a solubility minimum is present. Referring to these curves, it
is possible to predict if hot corrosion can proceed or if a sort of buffer condition is
reached at some stage and the corrosive attack is consequently interrupted.
As suggested by Rapp (2002), and shown by the general examples in Fig. 8.26,
the solubility values for the fluxing oxide species at the oxide scale/salt (point I in
Fig. 8.26) and salt deposit/atmosphere (point II in Fig. 8.26) interfaces can be
marked on the relevant solubility curve. In cases A and C, a basic and acidic fluxing
may occur, with the dissolution of the oxide scale present on the alloy surface and its
re-precipitation in the salt deposit and particularly at the free salt surface, which is a
preferential nucleation site. In all cases, the re-precipitated oxide is non-protective,
being discontinuous and porous. Cases B represent in principle situations in which
8.2 High-Temperature Corrosion 487

Fig. 8.26 Hot corrosion

general cases of a pure
metal. I and II are the
solubility at the oxide/salt
and salt/atmosphere
interfaces, respectively.
Case A and C basic and
acidic fluxing are sustained
by the negative solubility
gradients. In cases B, in
principle, hot corrosion
cannot occur, since no
solubility gradient is
present. However, any
minimum change toward the
solubility minimum would
activate corrosion.
(Redrawn from Rapp 2002)

no solubility gradient is present within the salt deposit. From a practical point of
view, this may correspond to a compositional homogeneity of the molten salt
determined by thermal convective remixing. However, the equilibrium is very
unstable indeed, and as soon as a local minimum in the solubility is reached, fluxing
will start. Incidentally, it is worth noting that both chromia and alumina are close to
the minima in their solubility curves under the operating conditions of a gas turbine
engine. Therefore, they are not so prone to hot corrosion fluxing, a further positive
aspect of these oxides that, as discussed in the previous sections (Sects. 8.2.1 and
7.3.1), justifies their role in the protection strategy of gas turbine components.
The solubility curves can also be used to follow the evolution of the salt
composition as the oxide dissolution proceeds. Basic fluxing consumes the sodium
sulfate deposit and replaces it with sodium-metal oxide (see, for instance, Eq. 8.18
for the formation of Na2NiO2), until the reaction subsides. From this point of view,
acidic fluxing may be more deleterious for the integrity of the component. The
intermediate stage that is implicit in the reaction Eq. 8.19 is the decomposition of the
NiO in its component ions: Ni2+ and O2
. The dissociation of the oxide may
continue as long as the activity of the acidic agent, i.e., SO3, renders the deposit
aggressive to the alloy. Also in view of developing protective strategies, it is
important to identify the main possible sources of SO3 that may determine the acidic
fluxing. In the first place, part of SO3 may come from the exhaust, in which the
oxidation of sulfur present as an impurity in the fuel may take place, as described
earlier on. The same may occur to vanadium, to form V2O5. The reaction of
vanadium oxide with the salt deposit determines an increase of its acidity, owing
to the increase in the partial pressure of SO3:

V2 O5 þ Na2 SO4 ! 2NaVO3 þ SO3 ð8:20Þ

488 8 Corrosion

More critical, since it may result in the catastrophic corrosion of the alloy, is the
acidic fluxing sustained by the reaction of sodium sulfate with the oxide of refractory
metals present in the alloy, like molybdenum and tungsten. The formation of sodium
molybdate is induced by the following reaction, involving MoO3:

Na2 SO4 þ MoO3 ! Na2 MoO4 þ SO3 ð8:21Þ

A similar reaction, involving WO3, would result in the formation of sodium

tungstate, Na2WO4. This reaction (Eq. 8.21) is sustained by an alloying element,
and therefore it can continue as long as this element is available. Therefore, the
acidic corrosive attack may propagate to significant regions of the alloy, even
considering that the acidic character of the salt would determine the dissociation of
possible oxide scale originally present on the alloy surface, including stable oxides,
as alumina, according to dissolution reactions, like:

Al2 O3 þ 3MoO3 ! 2Al3þ þ 3MoO2

4 ð8:22Þ

The particularly aggressive character of this type of acidic fluxing, called alloy-
induced acidic fluxing and normally observed over a temperature range of
900–1000  C (i.e., Type I hot corrosion), is due to the combination of the two acidic
agents: SO3 and MoO3. They are both present in the salt deposit, and, since the acidic
reaction involves an alloy component, e.g., molybdenum, the negative solubility
gradient criterion (see Fig. 8.26) is generally satisfied. Interestingly, it has been
demonstrated that by increasing the SO3 pressure in the gas atmosphere, the rate of
alloy-induced corrosion is reduced, owing to a lower equilibrium activity of MoO3 in
the deposit (Mishra 1986).
The fluxing mechanisms, both basic and acidic, involve the solutioning or
dissociation of native oxides present already on the alloy surface before the salt
deposit would form, or growing during the hot corrosion process. There is a third
category of oxides that may be destabilized by hot corrosion, that is to say thermal
barrier coatings (TBCs, see Sect. 7.3.2). The zirconia-based coatings, used to protect
the internal wall of combustors and the high-pressure stage vanes and blades of the
turbine from an excessive temperature rise, may suffer from fluxing effects due to
sulfate deposits. This may damage the TBCs reducing their protection efficiency in
case their thickness is reduced. The critical aspect is the heterogeneity that typically
characterizes the corrosive attack and consequent thermal gradients induced in the
alloy component (Chen et al. 2003). Moreover, it should be considered that a
selective fluxing of the zirconia stabilizing oxides, e.g., Y2O3, may take place, for
the different solubility of the diverse oxides (Zhang and Rapp 1987). The result is the
destabilization of the zirconia matrix, with consequent phase change and volume
expansion. The stresses building up within the coating may determine the local
spalling off of the TBC and the direct exposure of the alloy to the flux of hot gases.
The critical consequences that this failure mechanism may have on the stability of
the component require a careful consideration.
8.2 High-Temperature Corrosion 489

Fig. 8.27 Stability diagram

for the Na2SO4-NiSO4
system. Below the
shadowed curve, a Na2SO4-
NiSO4 solid solution is
stable. Above it, a liquid
solution Na2SO4-NiSO4 is
stable, still containing some
solid fractions both of the
solid solution (Na2SO4-
NiSO4) and of NiO. The
dotted line shows the
pressure values of SO3
above which NiSO4 is more
stable than NiO, which
would therefore tend to
disappear. (Redrawn from
Luthra and Shores 1980)

The formation reaction of nickel sulfate (Eq. 8.19) is interesting also for intro-
ducing Type II hot corrosion (LTHC), occurring at lower temperatures than Type I
(HTHC), below the melting temperature of pure Na2SO4. In fact, the formation of
metal sulfates contributes to the reduction of the incipient melting temperature of the
salt deposit. The crucial aspect is that in this context the formation of these com-
pounds is mainly due to a sufficiently high pressure of SO3, which is the main agent
of this corrosion attack. In the stability diagram for the Na2SO4-NiSO4 system
(Fig. 8.27), the existence of molten phases for sufficiently high pressures of SO3,
depending on the temperature, is displayed. Still referring to the SO3 pressure, the
(dotted) line above which NiSO4 is more stable than NiO is also shown.
Similar reactions as Eq. 8.19, involving cobalt oxides, may lead to the formation
of cobalt sulfate, starting from its main oxide phases (Luthra 1982):

CoO þ SO3 ! CoSO4

and ð8:23Þ
1
Co3 O4 þ 3SO3 ! 3CoSO4 þ O2
2

The reduction of the liquidus temperature of the sulfate solution is fully compat-
ible with the LTHC regime, considering that the eutectics for the binary systems are
Na2SO4 – CoSO4 (TE ¼ 671  C) and Na2SO4 – NiSO4 (TE ¼ 565  C). Other
substances may contribute to the reduction of the liquidus temperature of the deposit,
490 8 Corrosion

Fig. 8.28 (a) Macroscopic view of the damaged turbine blade (concave surface); (b) macroscopic
view of the damaged turbine blade (convex surface) with arrows indicating a crack propagating
along the blade section. (Eliaz et al. 2002, reproduced with permission of the Publisher)

like MgSO4 and other contaminant either of the fuel or from the environment,
including the unwanted dust that can be sucked into the engine from the runaways,
during taxing and take-off operations. As concerns fuel contaminants, the formation
of sodium vanadate (Eq. 8.20) may contribute to reducing the incipient melting
temperature of the Na2SO4 – NaVO3 solution, in principle down to the minimum of
500  C at the eutectics (Singh et al. 2007).
Type II corrosion is typically a nonuniform corrosive attack with a characteristic
pitting morphology produced by the mentioned low-melting sulfate solutions that
form in those regions of the deposit where particular compositional conditions are
realized. Hot corrosion attacks determine different kinds of damages, which may
lead to part failure. Two case studies are discussed in Eliaz et al. (2002) and illustrate
common ways in which hot corrosion may affect the integrity of structural compo-
nents. The first, refers to MAR-M200 (see Sect. 6.4) blades from the first stage of a
PT6 gas turbine engine, presenting diffuse material loss and localized deformation.
Cross-sectional observations revealed the presence of a porous oxide surface scale,
showing in the subscale region the typical product of corrosion reactions, e.g., metal
sulfides. In this case, the reduction of the load-bearing section caused the detected
plastic deformation of the component, which was thereafter replaced. Another case
history is that of a low-cycle low-stress fatigue crack (arrowed in Fig. 8.28b)
observed in a IN713 C (see Sect. 6.4) blade (Fig. 8.28a) that suffered from the
detachment of a large fragment from its trailing edge. As revealed by microanalytical
investigations, traces of chlorine, sulfur, and potassium were present on the exposed
surface of the crack, confirming that the initial macroscopic defects had been created
by localized hot corrosion. If not replaced on time, the crack initiated by stress
concentration after the spalling off of the initial defect, might have propagated to the
whole section of the blade, thus resulting in a catastrophic failure.
8.2 High-Temperature Corrosion 491

These examples and the mechanisms underlying Type I and II hot corrosion
suggest possible protection and remediation strategies that have been actually
adopted to counteract the very detrimental effects of this process (Eliaz et al. 2002).
Nickel and cobalt base superalloys are the first choice for the hottest and more
demanding, in terms of in- service conditions, components of gas turbine engines.
For these alloys, hot corrosion is definitely an important aspect to consider. How-
ever, as concerns the improvement of the corrosion resistance of the superalloys, the
requirements are often conflicting with those for better mechanical properties, both
static and time-dependent, and high-temperature structural stability. Refractory
elements, like tungsten and molybdenum, are excellent solid solution strengtheners
and, through the reduction of the alloy diffusivity (see Fig. 6.24), have also benefi-
cial effects on creep resistance. Nonetheless, under particular environmental condi-
tions, they may determine an important hot corrosion phenomenon (alloy-induced
hot corrosion) that is particularly critical since sustained by alloy elements, in the
first place, and therefore potentially severely destructive for the component. Chro-
mium and aluminum provide the best protection against high-temperature corrosion,
in view of their elevated stability and capacity to form a protective scale on the alloy
surface. These and similar principles have been followed to rank different alloys as
concerns their corrosion resistance, so to provide helpful material selection guide-
lines (Meier 1989; Pettit 2011; Sidhu et al. 2006).
Better compositions to resist hot corrosion have been developed to comply more
effectively with specific application requirements. Cobalt-based superalloys have
demonstrated an excellent corrosion resistance, which has been ascribed to some
extent to the higher eutectic temperature of the Co-Co3S4 mixture, higher than that of
the corresponding mixture forming in nickel-based superalloys (i.e., Ni-NixSy).
Therefore, the novel generation of the low-density cobalt-based superalloys (see
Sect. 6.2.2) has this further specific characteristic that justifies the growing interests
in these materials. Also in the development of single-crystal nickel-based superal-
loys, the possibility of improving hot corrosion resistance has been considered as an
important design parameter, to improve a strategic material property (Zhang et al.
1993a, b).
Hot corrosion protection has also become an issue for the TiAl-based materials,
considering their introduction as material for gas turbine components (see Sects.
4.3.2 and 4.6), and similar protection strategies are being considered, with the
necessary adaptations to specific compositional systems (Tang et al. 1999; Zhao
et al. 2010). The protective coatings developed for enhancing the oxidation resis-
tance of high- temperature aerospace alloys (see Sect. 7.3.1) would indirectly
improve also their resistance against hot corrosion attack. In fact, the oxide scales
that form on the coating surface, owing to the optimized composition and micro-
structure, are relatively more stable and better adherent to the substrate, so to be more
resistant also in the salt deposit aggressive environment. In case the corrosive
damage would produce the partial removal of the oxide scale, this can reform,
owing to the larger reservoir of scale-forming elements in the coatings. In this
respect, MCrAlX overlay coatings (see Sect. 7.3.1) have proved to be particularly
successful, since their composition can be tuned to better respond specific external
492 8 Corrosion

conditions. For prevailing LTHC, high chromium and low aluminum are preferable
compositions, whereas HTHC can be better tackled with lower chromium and higher
aluminum coatings (Sudiro et al. 2011). As concerns the aluminide diffusion coat-
ings, platinum additions, beneficial to the oxidation resistance, turn out to be
important also for better hot corrosion performances (Yang et al. 2017).
Optimal operating conditions, in terms of air and fuel purity, and engine main-
tenance procedures are additional effective tools for containing the negative effects
of hot corrosion on the engine lifetime. Air filters, quite common in land-based large
gas turbine plants, have been developed also for aerospace applications with partic-
ular regard for helicopter engines (Scimone 2003; Wilcox et al. 2012).
Sulfur, vanadium, and alkali metals, i.e., Na and K, are the main fuel impurities
that may initiate hot corrosion once they react to form phases that may condensate
and form a liquid or a solid deposit. Therefore, there are maximum acceptable values
for the above impurities. Typical limiting concentrations are (Na + K)  0.2–0.6 ppm,
V  0.5 ppm, and S  1%. These limiting concentration may be overcome if the
more exposed components are adequately protected, using appropriate coatings or
suitable additives. These additives are meant to reduce the corrosion rate by pro-
moting competitive reactions. For instance, an excess of vanadium may be tolerated
if a suitable concentration of MgO is added, which reacts preferentially with V2O5 to
form Mg3V2O8. This vanadate remains solid even above the HTHC temperature
interval (Stringer 1987).
To complete the picture on the diverse approaches used to protect the structural
components from hot corrosion attack, one possibility is based on the water wash of
the engine to remove all those contaminating substances that may initiate the hot
corrosion attack. Standard washing protocols have been established and delivered in
specific manuals (Need et al. 2015). The frequency and other parameters of the
procedure are determined on the basis of the specific risk of corrosive attack both
during operation, or for engines exposed to aggressive atmospheric conditions, like
marine environments. In this case, washing is very effective in reducing the corro-
sion load also of the compressor, particularly as concerns the parts exposed to the air
intake flux, since in these cases the prevailing corrosion agents are ingested atmo-
spheric pollutants. These are removed by introducing water and detergent substances
into the intake air flux of an operating engine. The application of protective plugs to
the open ends of an engine, i.e., air intake and exhaust duct, is a common practice to
reduce the access of contaminants into the engine during prolonged idle periods,
particularly for aircrafts exposed to critical and harsh environmental conditions
(Carter 2005; Hoyland and Antoine 2005).

8.2.6 Hot Corrosion Testing

The experimental investigation on the hot corrosion behavior of alloys is not a

simple matter, mainly for the complexity of the involved sub-systems, alloy, deposit,
atmosphere, and relevant compositional changes, as the corrosion reactions proceed.
8.2 High-Temperature Corrosion 493

Fig. 8.29 Schematic of the exposure test

However, there are established methods for hot corrosion testing that provide
extremely useful indications for predicting material behavior under service condi-
tions. The results of real part inspections are very important for the definition of
testing conditions that may provide response as concerns maintenance procedures,
safe lifetime estimation, and part replacement. The three most commonly used
testing methods are:
• Exposure test
• Crucible test
• Burner rig test
The exposure test is usually performed in an electrically heated furnace. The test
implies the contact between the specimens and a sulfur-containing atmosphere.
Exposure test is also known as bead test. The molten salt deposition is reproduced
by evaporation and condensation of sodium sulfate or other sulfur-containing salts.
Condensation is obtained through a suitable temperature profile inside the furnace
chamber and using a gas carrier, to simulate also the effect of the different atmo-
spheres (see Fig. 8.29).
A sample holder may contain up to 10–20 specimens, whose initial weight and
exposed surface are measured before starting the test. In isothermal conditions, the
samples are extracted at regular intervals, and their weight is measured, thus giving
information about the degradation process they have been subjected to. The crucible
test (Fig. 8.30) requires the immersion for fixed time intervals (exposure time) of
specimens with defined cross-section geometry (circular for cylindrical samples,
rectangular for sheet coupons, etc.) into a crucible filled with a molten Na2SO4-based
salt mixture, kept at constant temperature. The aim of the crucible test is both to
measure the metal loss after the corrosive attack and to evaluate the depth reached by
the corrosive attack. For this aspect, suitable material characterization methods are
494 8 Corrosion

Fig. 8.30 Schematic of the crucible test. (Redrawn from Baboian et al. 2005)

used. The test is quite simple and definitely useful for comparative studies, although
it exhibits evident limitations for an effective study of the hot corrosion mechanism.
This is mainly due to the following aspects:
• The composition of the salt bath changes during the test, owing to the reactions
with the sample.
• The effect of gas atmosphere cannot be accounted for, since the partial pressures
of the gaseous components are different at different depths in the bath.
In this regard, the burner rig test is definitely much more effective and reliable for
the assessment of the hot corrosion resistance of gas turbine alloys (Sumner et al.
2013).
A salt solution is injected into a flame produced in a burner chamber. The exhaust
flux, containing the salt droplets, impinges onto the surface of the samples, placed on
a rotating carousel (sample stage in Fig. 8.31). The samples can also be moved in and
out of the hot gas flux, so to test thermal cycling effects. For a full understanding of
the reactions occurring during any of the tests, to evaluate the volume of corroded
material and the depth beneath the alloy surface reached by the corrosion attack, a set
of microscopy and crystallo-chemical investigations are absolutely necessary for a
complete characterization of the corrosion products. Surface analyses, like X-ray
diffraction (XRD) and X-ray fluorescence (XRF), are important to characterize the
phase composition and crystallographic phases present in the scale and in the
subscale interdiffusion layer, if the thickness of the surface layer of salt deposit
and reaction products is not too large. The surface topography can be surveyed with
scanning electron microscopy (SEM) observations. Cross-sectional analyses are
fundamental to understand the extension of the corrosion damage inside the alloy,
the internal corrosion products, and their composition. Metallographic sample prep-
aration is necessary, and it is advisable to protect the surface layer to avoid damage
during handling and polishing operations. SEM observations can be accompanied by
8.2 High-Temperature Corrosion 495

Fig. 8.31 Schematic of a burner rig

Fig. 8.32 Examples of thermogravimetric curves obtained from hot corrosion tests. (a): Weight
gain of the Ni-8Cr-6Al-3Mo alloy hot corroded under different thicknesses of molten Na2SO4
deposit; (b) hot corrosion tests of the IN792 superalloy protected by a CoCrAlY overlay coating.
The effect of different compositions of the atmosphere, as indicated in the graph, is considered. The
tests are conducted under LTHC regime, with solid salt deposit, with the exception of the dotted
curve referring to a liquid deposit obtained by mixing Na2SO4 win MgSO4. (Redrawn from Pettit
2011)

energy dispersive X-ray spectroscopy (EDXS) analyses, which can be used for
evaluating the chemical composition of the different regions in the corrosion altered
alloy (X-ray mapping). To complete the picture, microhardness profile can be
acquired across this same altered layer as a further indication of the changes brought
about by the corrosion attack. Scientific journals and monographs on material
characterization techniques provide a complete information on the possibilities of
the different approaches, as concerns also the sample preparation procedures of each
technique (Brandon and Kaplan 2008). Thermogravimetric curves, like those
presented in the previous section, can be used also for hot corrosion tests data, by
496 8 Corrosion

plotting the specific weight change as a function of the time of exposure to the hot
corrosion conditions (Fig. 8.32).
An important aspect to consider in thermogravimetric tests in hot corrosion
studies, involving salt deposit, concerns a reliable and reproducible procedure of
removing the residuals of the deposit before weighting the tested samples at the end
of each test period. This is usually made with an accurate washing of the specimens.
A possible procedure requires washing in boiling water for 30 min, mostly for the
removal of the remaining salts, followed by ultrasonic cleaning, effective for elim-
inating the corrosion product not adherent to the alloy substrate (Yuan and Wang
2010).

References

Ali MF et al (1993) Nickel and Vanadyl Porphyrins in Saudi Arabian Crude Oils. Energy & Fuels 7:
179–184
ASM International (1992) Handbook of Case Histories in Failure Analysis – Vol. 1. ASM Inter-
national, Materials Park, Ohio
ASM International (2003) ASM Handbook Vol. 13A – Corrosion: Fundamentals, Testing and
Protection. ASM International
Baboian et al. (2005) Corrosion Tests and Standards: Application and Interpretation. ASTM
International
Baer D R, Merz M D (1980) Differences in Oxides on Large- and Small-Grained 304 Stainless
Steel. Metallurgical Transactions A 11 (12): 1973–1980
Bagnoli F, Bernabei M (2008) Fatigue analysis of a P180 Aircraft Main Landing Gear Wheel
Flange. Engineering Failure Analysis 15: 654–665
Benavides S (2014) Corrosion Control in the Aerospace Industry. CRC Press
Brandon D, Kaplan W (2008) Microstructural Characterization of Materials. John Wiley and Sons
Ltd
Buckley et al (1981) Early Development of Ceramic Fiber Insulation for Space Shuttle. Ceramic
Bulletin 60: 1196–1200
Carter T J (2005) Common Failures in Gas Turbine Blades. Engineering Failure Analysis 12 (2):
237–247
Chen Z et al (2003) Effect of Al2O3 Overlay on Hot-Corrosion Behavior of Yttria-stabilized
Zirconia Coating in Molten Sulfate-vanadate Salt. Thin Solid Films 443 (1): 46–52
Chevron Product Company (2007) Aviation Fuels – Technical Review. Chevron Product Com-
pany, CA, USA
DuQuesnay D L et al (2002) Fatigue Crack Growth from Corrosion Damage in 7075-T6511
Aluminum Alloy under Aircraft Loading. International Journal of Fatigue 25 (5): 371–377
Eliaz N et al (2002) Hot Corrosion in Gas Turbine Components. Engineering Failure Analysis 9 (1):
31–43
Fecht H J (1990) Thermodynamic Properties and Stability of Grain Boundaries in Metals based on
the Universal Equation of State at Negative Pressure. Acta Metallurgica et Materialia 38 (10):
1927–1932
Findlay J S, Harrison N D (2002) Why Aircraft Fail. Materials Today 5 (11): 18–25
Fontana M G (1987) Corrosion Engineering. McGraw-Hill, Singapore
Giggins C S, Pettit F S (1971) Oxidation of Ni-Cr-Al Alloys Between 1000  C and 1200  C. Journal
of the Electrochemical Society 118 (11): 1782–1790
References 497

Gupta R K et al (2012) Oxidation Resistance of Nanocrystalline Alloys. Available via DIALOG

https://research.monash.edu/en/publications/oxidation-resistance-of-nanocrystalline-alloys.
Accessed 23 Dec 2018
Hou P Y, Stringer J (1995) The Effect of Reactive Element Additions on the Selective Oxidation,
Growth and Adhesion of Chromia Scales. Materials Science and Engineering A 202 (1–2): 1–10
Housecroft C E, Constable E C (2006) Chemistry. Pearson
Hovland G, Antoine M (2005) Scheduling of Gas Turbine Compressor Washing. Intelligent
Automation and Soft Computing 11: 1–12
Huntz A M (1987) Influence of Active Elements on the Oxidation Mechanism of M-Cr-Al Alloys.
Materials Science and Engineering 87: 251–260
Jones R H (1992) Stress-Corrosion Cracking. ASM International
Klam H J et al (1987) The Thermal Expansion of Grain Boundaries. Acta Metallurgica 35 (8):
2101–2104
Kofstad P (1988) High Temperature Corrosion. Elsevier Applied Science, London
Korb et al (1981) The Shuttle Orbiter Thermal Protection System. Ceramic Bulletin 60: 1188–1193
Lee B et al (2011) Investigation on Fretting Fatigue in Aircraft Engine Compressor Blade.
Engineering Failure Analysis 18: 1900–1908
Leyens C et al (2006) Recent Progress in the Coating Protection of Gamma Titanium Aluminides.
JOM 58 (1): 17–21
Liu C T et al (2010) Oxidation Behavior of a Single-Crystal Ni-base superalloy between 900 and
1000  C in Air. Journal of Alloys and Compounds 491 (1–2): 522–526
Liu et al (2000) Surface Nanocrystallization of 316L Stainless Steel Induced by Ultrasonic Shot
Peening. Materials Science and Engineering A 11 (4): 91–95
Lou H et al (1995) Rehealing Ability of Oxide Scales Formed on Microcrystalline K38G Coatings.
Oxidation of Metals 43 (3–4): 317–328
Lou H et al (1996) Oxidation Behavior of Sputtered Microcrystalline Coating of Superalloy K17F
at High Temperature. Materials Science and Engineering A 207 (1): 121–128
Lu K, Lu J (1999) Surface Nanocrystallization (SNC) of Metallic Materials-Presentation of the
Concept behind a new Approach. Journal of Materials Science and Technology 15 (3): 193–197
Lu K, Lu J (2004) Nanostructured Surface Layer in Metallic Materials induced by Surface
Mechanical Attrition Treatment. Materials Science and Engineering A 375–377: 38–45
Luthra K L (1982) Low Temperature Hot Corrosion of Cobalt-Base Alloys: Part I. Morphology of
the Reaction Product. Metallurgical Transactions A 13 (10): 1853–1864
Luthra K L (1982) Low Temperature Hot Corrosion of Cobalt-Base Alloys: Part II. Reaction
Mechanism. Metallurgical Transactions A 13 (10): 1853–1864
Luthra K L, Shores D A (1980) Mechanism of Na2SO4 Induced Corrosion at 600–900 C. Journal of
Electrochemical Society 127 (10): 2202–2210
Mandelli M et al (2007) Surface Treatments and Oxidation Resistance of a γ-α2 TiAl Alloy (Ti-30.5
Al-9W-0.4Si). La Metallurgia Italiana March issue: 21–28
McCafferty E (2010) Introduction to Corrosion Science. Springer Verlag, New York
Meier G H (1989) A review of Advances in High-Temperature Corrosion. Materials Science and
Engineering A 120–121: 1–11
Merz M D (1979) The Oxidation Resistance of Fine-Grained Sputter-Deposited 304 Stainless Steel.
Metallurgical Transactions A 10 (1): 71–77
Mills T et al (2014) Corrosion and the Threat to Aircraft Structural Integrity. In: Benavides S
(ed) Corrosion Control in the Aerospace Industry. Woodhead Publishing Limited, p 35–66
Mishra A K (1986) Mechanism of Na2SO4-1nduced Corrosion of Molybdenum Containing Nickel-
Base Superalloys at High Temperatures II. Journal of the Electrochemical Society 133:
1038–1042
Molière M, Sire J (1993) Heavy Duty Gas Turbine Experience with Ash-forming Fuels. Journal de
Physique IV Colloque 03 (C9): 719–730
Mouritz A (2012) Introduction to Aerospace Materials. Woodhead Publishing Limited
498 8 Corrosion

Need R et al (2015) Reliability Evaluation of a Gas Turbine Water Wash System – A Case Study.
Journal of Natural Gas Science and Engineering 24: 457–463
Nicholls J R (2003) Advances in Coating Design for High-Performance Gas Turbines. MRS
Bulletin 28 (9): 659–670
Okafor I C I, Reddy R G (1999) The Oxidation Behavior of High-Temperature Aluminides. JOM
51 (6): 35–40
Pedeferri P (2010) Corrosione e Protezione dei Materiali Metallici Vol. 1 and 2. Polipress
Pei H et al (2017) Long-Term Oxidation Behavior and Mechanism of DD6 Ni-based Single Crystal
Superalloy at 1050  C and 1100  C in Air. Journal of Alloys and Compounds 704: 218–226
Pettit F (2011) Hot Corrosion of Metals and Alloys. Oxidation of Metals 76 (1–2): 1–21
Pflumm R et al (2015) Oxidation Protection of γ-TiAl-based Alloys – A Review. Intermetallics 56:
1–14
Popov B (2015) Corrosion Engineering: Principles and Solved Problems. Elsevier
Rapp A (2002) Hot Corrosion of Materials: A Fluxing Mechanism? Corrosion Science 44 (2):
209–221
Rapp R A, Goto K S (1981) The Hot Corrosion of Metals by Molten Salts. Proceedings of the
Electrochemical Society 1980: 159–177
Rapp R, Zhang Y (1994) Hot Corrosion of Materials: Fundamentals Studies. Journal of the
Minerals 46 (12): 47–55
Reed R C (2006) The Superalloys – Fundamentals and Applications. Cambridge University Press,
Cambridge (UK)
Sato A et al (2011) Oxidation of Nickel-based Single-Crustal Superalloy for Industrial Gas Turbine
Applications. Acta Materialia 59 (1): 225–240
Scimone M J (2003) Aircraft Engine Air Filter and Method, US Patent 6.595.743, B2
Sheng G (2011) Investigation of Surface Self-Nanocrystallization in 0Cr18Ni9Ti Induced by
Surface Mechanical Attrition Treatment. International Scholarly Research Network Nanotech-
nology. Available via DIALOG https://www.hindawi.com/journals/isrn/2011/583215/.
Accessed 12 Nov 2018
Sidhu T S et al (2006) Hot Corrosion and Performance of Nickel-based Coatings. Current Science
90 (1): 41–47
Singh H et al (2007) An overview of Na2SO4 and /or V2O5 induced Hot Corrosion of Fe- and
Ni-based Superalloys. Reviews on Advanced Materials Science 16 (1): 27–50
Stott F H, Wood G C (1987) Growth and Adhesion of Oxide Scales on Al2O3 forming Alloys and
Coatings. Materials Science and Engineering 87: 267–274
Stringer J (1987) High-Temperature Corrosion of Superalloys. Materials Science and Technology
3 (7): 482–493
Sudiro T et al (2011) High Temperature Corrosion of CoNiCrAlY-Si Alloys in an Air-Na2SO4-
NaCl Gas Atmosphere. Materials Transactions 52 (3): 433–438
Sumner J et al (2013) High Temperature Oxidation and Corrosion of Gas Turbine Materials in
Burner Rig Exposure. Materials Science and Technology 29 (7): 813–821
Tang Z et al (1999) Hot-Corrosion Behavior of TiAl-Base Intermetallics in Molten Salts. Oxidation
of Metals 51 (3–4): 235–250
Tomasi A et al (2002) Oxidation Protection of γ-TiAl Alloys by Surface Heat Treatments. In:
Proceedings of the International Symposium on High Temperature Corrosion in Energy Related
Systems, Angra dos Reis, RJ, September 2002
Wallace W et al (1985) AGARD Corrosion Handbook Volume 1 – Aircraft Corrosion: Causes and
Case Histories. AGARD-Advisory Group for Aerospace Research and Development
Wang F (1997) The Effect of Nanocrystallization on the Selective Oxidation and Adhesion of
Al2O3 Scales. Oxidation of Metals 48 (3–4): 215–224
Wang M J et al (2015) Comparative Study of Oxidation and Interdiffusion Behavior of AIP
NiCrALY and Sputtered Nanocrystalline Coatings on a Nickel-based Single-Crystal Superal-
loy. Corrosion Science 98: 530–540
References 499

Wang X et al (2014) Influence of Sputtered Nanocrystalline Coating on Oxidation and Hot

Corrosion of a Nickel-based Superalloy M951. Journal of Materials Science & Technology
30 (9): 867–877
Wanhill R et al (2011) Stress Corrosion Cracking (SCC) in Aerospace Vehicles. In: Raja V S, Shoji
T (eds) Stress Corrosion Cracking: Theory and Practice, Whoodhead Publishing Limited, p
608–650
Wilcox M et al (2012) Technology Review of Modern Gas Turbine Inlet Filtration Systems.
International Journal of Rotating Machinery 2012 (12)
Xia Z X et al (2015) Improve Oxidation Resistance at High Temperature by Nanocrystalline
Surface Layer. Scientific Reports 5. Available via DIALOG https://www.nature.com/articles/
srep13027. Accessed 11 Dec 2018
Yang L et al (2016) A Duplex Coating for High Temperature Applications on Single-Crystal
Superalloys. Corrosion Science 102: 72–83
Yang L et al (2016) Effects of surface Finish of Single Crystal Superalloy Substrate on Cyclic
Thermal Oxidation of Its Nanocrystalline Coating. Corrosion Science 111: 313–324
Yang Y F et al (2017) Hot Corrosion Behaviour of Single-Phase Platinum-modified Aluminide
Coatings: Effect of Pt Content and Pre-oxidation. Corrosion Sience 127: 82–90
Young D J (2015) High Temperature Oxidation and Corrosion of Metals. Elsevier
Yuan L, Wang H M (2010) Hot Corrosion Behaviors of a Cr13Ni5Si2-based Metal Silicide Alloy in
Na2SO4+ 25 wt.% K2SO4 and Na2SO4+25 wt.% NaCl molten Salts. Intermetallics 18 (3):
324–329
Zhang J S et al (1993a) Design and Development of Hot Corrosion-Resistant Nickel-Base Single-
Crystal Superalloys by the d-Electrons Alloy Design Theory: Part I – Characterization of the
Phase Stability. Metallurgical Transactions A 24: 2443–2450
Zhang J S et al (1993b) Design and Development of Hot Corrosion-Resistant Nickel-Base Single-
Crystal Superalloys by the d-Electrons Alloy Design Theory: Part II – Effects of Refractory
Metals Ti, Ta and Nb on Microstructures and Properties. Metallurgical Transactions A 24:
2451–2464
Zhang Y S, Rapp R A (1987) Solubilities of CeO2, HfO2 and Y2O3 in Fused Na2SO4 -30 mol%
NaVO3 and CeO2 in Pure Na2SO4 at 900  C. Corrosion 43 (7): 348–352
Zhao W et al (2010) The Effect of High Temperature Corrosion on Mechanical Behavior of a
Gamma TiAl Alloy. International Journal of Modern Physics B 24 (15–16): 2970–2975

Further Reading

Craig B, Anderson D (1995) Handbook of Corrosion Data. ASM International

De Luccia J et al (1987) AGARD Corrosion Handbook Volume 2 – Aircraft Corrosion Control
Documents: A Descriptive Catalogue. AGARD-Advisory Group for Aerospace Research and
Development
DeHoff R T (2006) Thermodynamics in Materials Science 2nd ed. CRC Press
Embury J (1990) High-Temperature Oxidation and Sulphidation Processes. Pergamon Press
Khanna A S (2002) Introduction to High Temperature Oxidation and Corrosion. ASM International
Materials Park, OH (USA)
Lai G Y (2007) High-Temperature Corrosion and Materials Applications. ASM International
Lee B, Karkkainen R L (2018) Failure investigation of hydrogen induced cracking of stabilator skin
in jet aircraft. Engineering Failure Analysis 92:182-194
Rangone D V (1995) Thermodynamics of Materials – Volumes 1 and 2. Wiley & Sons
Roberge P (2008) Corrosion Engineering: Principles and Practice. McGraw-Hill
Wanhill R, Windisch M (2017) Corrosion and Stress Corrosion Testing of Aerospace Vehicles
Structural Alloys. Springer
Chapter 9
Other Interesting Alloys for Aerospace
and Related Applications

9.1 Introduction

This last chapter is dedicated to four classes of materials that have been considered
and are still investigated in view of their potential applications in the aerospace field,
as concerns in particular hot parts of aero-engines and functional applications, like
actuators and control systems.
New materials for aircraft structures, and relevant production techniques, have
continuously been developed over more than one hundred years now. This trend
seems to be substantially unaltered in the future perspective, considering that
innovation has always been a pillar in aerospace field. Initially, in view of the limited
power of the engines, weight reduction of all parts of an aircraft was the most
effective approach to enhance its aerodynamic efficiency. For achieving this, since
the early attempts by Orville and Wilbur Wright (Chap. 1), wood, varnished fabric,
and steel, this latter for the load-bearing structures, were the main materials used for
the Flyer I and successive aircraft structures. Following the same principle, for the
engine, the Wright brothers used a cast aluminum alloy (King et al. 2009). Mass
reduction and better specific structural properties of materials continued to be
fundamental selection parameters even after the pioneering times of aviation. How-
ever, more issues came up later, playing a decisive role not only in the development
of new materials but also in the optimization of the properties of the already existing
ones and in the implementation and validation of novel process technologies. In
recent times, considering the planetary dimension achieved by airborne transporta-
tion, the environmental impact has become a crucial aspect (see Sect. 2.6). The same
can be said for fuel saving and, thereby, higher propulsion efficiency, just to mention
a couple of issues that are nowadays central in the development of the field, whereas
they were out of the agenda until comparatively recent times (see Fig. 9.1). The
ongoing related research efforts are proceeding in quite many directions. In this final
chapter, some of the systems that are under investigation for answering novel and
even more demanding technological requirements are presented. The first three

© Springer Nature Switzerland AG 2020 501

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8_9
502 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.1 Effect of different technological approaches on the improvement in aircraft fuel consump-
tion. (Redrawn from Cantor et al. 2001)

examples refer to alloys having as common ground the search for higher engine
operating temperatures: refractory element, oxide dispersion strengthened (ODS)
and intermetallic alloys (Sects. 9.2, 9.3, and 9.4). Actually, intermetallic compounds
would comprise also Nitinol alloys, based on the NiTi-ordered phase, a fundamental
group of shape memory alloys (SMAs, Sect. 9.5). These alloys are rather interesting
for their functional properties and relevant possibility to manufacture smart compo-
nents. They might allow important improvement in the overall aircraft performances,
coherently with the mentioned guidelines for the present development strategies of
aerospace structures.

9.2 Refractory Alloys

9.2.1 General Aspects

Increasing the operating temperatures of gas turbine engines has been a constant
challenge for machine designers and materials scientists over the years. Together
with the improvement of engine thrust performances, other aspects are nowadays
regarded as crucial, like the need for cleaner emissions and a general saving of fossil
fuels, in order to reduce the associated greenhouse effects (see Sect. 2.6). The
increase of the operating temperatures, particularly the raise of the turbine entrance
temperature (TET, see Sect. 2.2 and Fig. 2.20), is an obvious approach to achieve
9.2 Refractory Alloys 503

Table 9.1 Melting temperature and density of bcc refractory metals

Symbol Element Melting temperature [
C] Density [g/cm3]
Cr Chromium 1863 7.2
Nb Niobium 2469 8.6
Mo Molybdenum 2623 10.2
Ta Tantalum 2996 16.6
W Tungsten 3410 19.3

this goal. Structural materials featuring adequate thermal stability are necessary. In
this regard, nickel-based superalloys, once the metal temperature overcomes
1100
C, start creeping at intolerable rates. Nonetheless, this temperature limit can
definitely be regarded as absolutely remarkable, considering that it is just 250
C
below the incipient melting point of state-of-the-art superalloys. Moreover, the gas
temperature in the hottest parts of the engine, i.e., the combustion chamber, can
reach peaks as high as 2000
C (see Fig. 2.12). It is thanks to a combination of
efficient design and operation of the cooling system present in the hollow vanes and
blades, associated with the protective action by TBCs (see Sect. 7.3.2), that the
components can resist these extreme operating conditions. Refractory metals are
characterized by melting temperatures that make them potentially interesting as base
materials for components capable to break the limits of superalloys, as concerns
structural applications at high temperatures (Perepezko 2009). The melting points of
some interesting metals are given in Table 9.1, all featuring values of their melting
temperature above that of pure nickel, i.e., 1455
C. Although this is coherent with
their excellent intrinsic creep resistance, the early investigations on the potentials of
refractory metals and their alloys highlighted several restrictions and critical aspects
(Wadsworth et al. 1988). Some descend from the bcc structure that all of the metals
listed in Table 9.1 have, with a ductile to brittle transition (DBTT) above room
temperature, and consequent limited room-temperature ductility. This renders their
processing and forming rather critical even in view of the elevated strength that these
metals would usually exhibit. Joining and welding are also rather challenging, since
the resulting embrittlement has consequences not only on the production but also on
the in-service alloy performances. Eventually, the exposure to high-temperature
oxidation conditions has demonstrated the insufficient intrinsic resistance of these
alloys.
Notwithstanding this complex picture, mixing together positive and critical
aspects, several applications have importantly benefitted from using refractory
metal alloys. These materials are today the top priority choice for several parts
working under particularly demanding conditions, like dies for precision metal
forging, used also in the production of aircraft fan blades and rotors (Starck Fabri-
cated Products 2013). The alloy systems that have actually been employed in several
fields, including aerospace, are those based on niobium, molybdenum, tungsten, and
tantalum. More recently, feasibility studies have been conducted to evaluate alloys
based on chromium, listed among other refractory metals in Table 9.1. An overview
504 9 Other Interesting Alloys for Aerospace and Related Applications

Table 9.2 Code name and compositions of the commercial refractory metal alloys, subdivided
according to the main element on which they are based (Lipetzky 2002)
Base element Alloy name Composition (wt%)
Mo TZC Mo-1.2Ti-0.25Zr-0.15C
TZM Mo-0.5Ti-0.1Zr-0.02C
Mo-Re Mo-50Re
Mo-13Re Mo-13Re
Nb Nb-1Zr Nb-1.0Zr-0.005C
Nb1 Nb-30W-1Zr-0.06C
B88 Nb-28W-2Hf-0.067C
FS85 Nb-28Ta-10W-1Zr-0.004C
C129Y Nb-l0W-10Hf-0.2Y-0.015C
C-103 Nb-10Hf-1.0Ti-0.7Zr-0.015C
Ta T-111 Ta-8W-2Hf
T-222 Ta-10W-2.5Hf-0.01C
ASTAR 811C Ta-8W—1.0Hf-1.0Re-0.025C
W W-Re W-25Re
3D W-3Re

on the present knowledge on chromium-based alloys will be presented in Sect. 9.2.3,

together with other alloy systems, including the alloys of the platinum group metals
(PGMs).

9.2.2 Refractory Metal Alloys

The refractory alloy systems that have attracted the main interests so far are based on
Mo, Nb, Ta, and W. Table 9.2 gives a list of alloys based on these refractory
elements, developed over the years and successfully employed in several high-
temperature applications.
As concerns niobium, in the 1960s, several high-strength alloys were developed
(Wojcik 1991). The typical production cycle for Nb-based alloys starts with an
electron beam furnace for melting the pure metal, followed by vacuum arc remelting
(VAR, see Sect. 5.3.6), during which alloying elements are added. For the final
refinement, hot extrusion or forging processes close to the recrystallization temper-
ature range, i.e., 1100–1380
C, are employed. Due to an inadequate fracture
toughness observed in some Nb-based alloys, medium strength, higher ductility,
and better weldable materials, like nickel-based superalloys, are still preferred.
However, the C103 alloy, which is one of the most used niobium-based alloys
ever, is still an excellent choice for aerospace applications requiring exposure to
temperatures ranging from 1200 to 1550
C. This alloy exhibits an excellent
weldability in addition to a good workability, so that also complex forms, like rocket
thrust cones, high-temperature valves, and connections, can be fabricated. In partic-
ular, the C103 alloy was used for the descent nozzle of the Apollo lunar module and
9.2 Refractory Alloys 505

thrusters for extra-atmospheric spacecraft positioning. However, the C103 alloy, as

other niobium alloys and like, in general, all alloys based on refractory metals,
exhibits a very limited oxidation resistance that hinders their use in oxidizing
environments. In case of Nb-alloys, the oxide phase that tends to form on the surface
of these materials, Nb2O5, is not protective, since it does not form a continuous,
adherent, and compact layer. Protective coatings, based on silicide phases, have been
developed to provide a sufficient level of protection to the refractory metal alloys
(Bewlay et al. 2003a, b). These coatings are typically obtained using a halide-
activated pack-cementation process (Kung and Rapp 1989), similar to that consid-
ered for nickel aluminide diffusion coatings (see Sect. 7.2.3). The growth of the
coating requires two main steps: first, deposition of silicon on the alloy surface, and
second, diffusion to form the silicide phases. Actually, not only silicon is deposited
onto the surface to be protected but also a series of other alloying elements, like
titanium, chromium, iron, etc., that are introduced to improve the coating perfor-
mances and durability. For this reason, setups for co-deposition are commonly used,
selecting the halide species in order to achieve different values of partial pressures of
the metal precursors. A number of compositions and phases have been investigated,
guided by the knowledge of the relevant phase diagrams (Harper and Rapp 1994;
Vilasi et al. 1998a, b). Coatings with general compositions like M7Si6 and M6Si5
(M being a combination of several elements like Co, Cr, Fe, Nb, Ni, and Ti) have
been developed (Knittel et al. 2013). These coatings provide an adequate protection
to Nb-based alloys and composites, even because their composition can be tuned and
modified to better match the substrate features and comply with the requirements of
the real servicing conditions (Vilasi et al. 1993; Vilasi et al. 1998a, b). Boron is often
present in the coating formulations, having a general stabilization role. Boron
additions result in lower diffusivities, thus rendering the coating less permeable to
gas infiltration but also less prone to interdiffusion with the substrate, a positive
factor for extending the coating lifetime. In case of surface cracking, the borides
present in the outer surface layer act as a reservoir of this element, which contributes
to the formation of a low-temperature melting borosilicate glass. At the operating
temperatures, this glass can flow and mend the damaged regions of the coating, with
a self-healing action (Portebois et al. 2014). The excellent surface stability and
oxidation resistance achieved thanks to these different approaches, coupled with a
relative ease of deformation, render niobium-based alloys promising candidates for
hot components of future gas turbine engines, as indirectly demonstrated by the
industrial development projects that have been carried out on this matter (Bewlay
et al. 2003a, b; Jehanno et al. 2005; Sankar et al. 2015).
Similar issues, as concerns oxidation protection, apply also to molybdenum-
based alloys. Molybdenum has a higher density than niobium (Table 9.1), and its
alloys are used for ultrahigh-temperature applications, under protective atmosphere,
so that gas-phase corrosion cannot take place (Dimiduk and Perepezko 2003). In
these conditions, the structural stability and the good mechanical and creep resis-
tance of molybdenum alloys can be fully exploited up to temperatures in excess of
1200
C. The oxidation behavior of molybdenum-based alloys is somehow similar
to that of the chromia-forming ones. In fact, the main oxide phase that forms on the
506 9 Other Interesting Alloys for Aerospace and Related Applications

exposed surfaces of molybdenum alloys, i.e., MoO3, is volatile above about 700
C.
Therefore, this phase cannot provide any protection to the underlying alloy above
this temperature but can rather promote gas-phase hot corrosion (see Sect. 8.2.5;
Jones et al. 1958). The similarity with chromia-forming alloys is that also Cr2O3
transforms into the CrO3 volatile phase above 950
C approximately (see Eq. 6.13).
Successful strategies to overcome this limit are either based on bulk compositional
changes or on the alloy protection using suitable coatings. As for niobium alloys,
silicon additions provide an essential constituent for the formation of a protective
silicon oxide-based surface scale. Boron is also added in order to enhance the silicon
storage capabilities of the alloy, beyond the solid solubility limit of 3 at% for silicon
and far below 1 at% for boron itself (Heilmeier et al. 2009), through the formation of
molybdenum and boron-silicide precipitates. After an initial oxidation transient,
corresponding to a weight reduction for the formation of the volatile molybdenum
oxide phase, MoO3, a steady-state oxidation regime is entered, during which a
borosilicate surface layer starts to form. This barrier protects the alloy and slows
down the overall oxidation kinetics (Parthasarathy et al. 2002). To reach this
condition, at least temperatures as high as 1000
C are necessary. In fact, at
intermediate temperatures, in between 700 and 1000
C, the formation of a porous
scale is still observed, featuring the coexistence of MoO3 with the borosilicate
component. At intermediate temperatures, below 1000
C though, a continuous
layer of SiO2-B2O3 glassy oxide forms on the alloy surface with an underlying
MoO2-rich layer. This stable molybdenum oxide phase may form thanks to the
reduction in the oxygen residual partial pressure attained beneath the outer borosil-
icate scale. The residual formation of a limited amount of MoO3, if any, can be
inferred from the presence of gas bubbles, trapped into the outer glassy scale. A truly
parabolic oxidation kinetics can be observed at temperatures of 1300
C or so (Azim
et al. 2017; Burk et al. 2010). The viscosity of the protective outer scale can be
modulated by changing the concentration of boron. Moreover, as reported for Nb-Si-
B alloys, the borosilicate glassy phase may contribute to heal cracks forming in the
oxide surface layer (Akinc et al. 1999; Sakidja et al. 2005). From the structural point
of view, the silicate and borosilicate precipitates introduce also an important
strengthening effect, which, in association with the intrinsic creep resistance, renders
molybdenum alloys of potential interests for structural applications, particularly
under important mechanical loads. Nonetheless, the limited room-temperature duc-
tility is still a main issue to consider when dealing with molybdenum and
other refractory metal alloys. In this respect, important steps forward have been
made both in terms of composition design and alloy processing (El-Genk and
Tournier 2005; Wadsworth et al. 1988). The challenge is to render the alloys
sufficiently deformable, without reducing their mechanical strength. Alloying
molybdenum with rhenium, which enhances the dislocation mobility, is a strategy
that has been abandoned, since, in addition to an excessive softening of the alloy, the
material cost was intolerably high, owing to the high price of the raw materials. An
alternative route was initiated starting from the experimental evidence that the reason
for the limited ductility of Mo alloys strongly depends on their sensitivity to the
impurities, e.g., oxygen and nitrogen, segregating along the alloy grain boundaries
9.2 Refractory Alloys 507

(Brosse et al. 1981; Medvedeva et al. 2007; Trinkle and Woodward 2005;
Wadsworth et al. 1988). Several strategies can be implemented to modify the
size distribution of grain boundaries and solute segregation. One possibility is
promoting the precipitation of secondary phases, in the form of carbides and oxides
(Cockeram 2009; Majumdar et el. 2009; Miller et al. 2003; Mueller et al. 2000).
They can act as nucleation sites for new alloy grains and hinder the recrystallization
and coarsening of the already existing ones. A possible route to obtain these
structures is mechanical alloying, a powder metallurgy approach described in Sect.
9.3, concerning oxide dispersion strengthened alloys. An innovative liquid-based
process is reported herewith, most promising as concerns the production of Mo
alloys featuring extremely large deformation to rupture (Liu et al. 2013). The key
factor that determines the exceptional mechanical behavior of these alloys is their
extremely refined microstructure, with molybdenum alloy grains below 1 μm and a
uniform nanometric oxide dispersion. The process is based on a liquid-liquid
mixture of the precursors of the two components, i.e., the molybdenum matrix and
the lanthanum oxide dispersion. Lanthanum oxide (La2O3) is dissolved into an
aqueous solution of La(NO3)3 (lanthanum nitrate), subsequently mixed with another
solution, containing ammonium bi-molybdate, which promotes the precipitation of a
NH4La(Mo2O7)2 powder. The grains of this powder act as heterogeneous nucleation
sites for the (NH4)2Mo2O7, so that a core-shell structure is obtained. Subsequent
thermal treatments and hydrogen reduction result in the formation of molybdenum
grains, still core-shelled, with an inner part made of La2O3. The compaction and
densification of this powder results in a molybdenum matrix reinforced by a disper-
sion of finely distributed lanthanum oxide particles. The mechanical properties of
this nanostructured (NS) alloy have been measured and compared with those of
another molybdenum alloy, having the same composition but obtained using a
conventional powder metallurgy route, typical of the oxide dispersion strengthened
(ODS) alloys (Cockeram 2006; Sturm et al. 2007). Some of these results are
summarized in Table 9.3, where the data for a commercially pure (CP) Mo alloy
are shown for a further comparison and better illustration of the excellent properties
of the newly developed alloy.
The excellent values of the tensile ductility and mechanical strength can mainly
be ascribed to the refined microstructure of this alloy and to its effect on the
dislocation dynamics, as demonstrated by the thorough electron microscopy

Table 9.3 Comparison between the mechanical properties of two ODS Mo alloys, one of which
having a nanostructured (NS) grain. Commercially pure (CP) molybdenum is quoted as refer-
ence (Liu et al. 2013)
Yield
Average grain strength UTS Fracture toughness DBTT
Alloy size [μm] [MPa] [MPa] E% [MPa√m] [
C]
CP-Mo 97  8 385 534 13.0 24.2  2.3 25
ODS Mo 1.4  0.7 710  37 746  49 14.1  6.5 52.8 60
NS-Mo 0.5  0.3 813  3 865  5 37.5  3.0 121.5  3.5 100
508 9 Other Interesting Alloys for Aerospace and Related Applications

characterization carried out on the mechanically tested samples. Interestingly, this

approach may in principle be applied to other refractory alloys, like tantalum- and
tungsten-based ones, that also suffer from a limited ductility. These alloys (see
Table 9.2) have been developed with a particular focus related to aerospace appli-
cations, for high-density components (vide infra), for which these systems are
particularly suited. Tantalum, different from other refractory metals, has a compar-
atively good oxidation resistance and, like niobium, exhibits an acceptable
deformability at room temperature (Schmidt and Ogden 1963). In fact, it can be
processed using liquid nitrogen, as protective environment, so to avoid oxidation
losses and, at the same time, not induce unwanted embrittlement, since its absolutely
cryogenic DBTT is below 269
C (Buckman 2000). As concerns tungsten, the
main aspects of its metallurgy, including extraction, processing, and corrosion
resistance can be found in Lassner and Schubert (1999).
Although the main feature of the refractory element alloys discussed so far is their
high-temperature stability, tantalum and tungsten alloys have found their main
aerospace applications for another specific property, i.e., their elevated density.
This is the case of the so-called balancing weights, introduced into aircraft or
helicopter structures to reduce their vibrations, thus improving the comfort of the
passengers and the crew. These weights are produced with different shapes and
geometries, so that they can be mounted on the control surfaces of the ailerons,
elevators, and rudder sections of aircraft, rotor blades of helicopters, and cockpit
instrumentation as well. Since for these components a limited volume is paramount,
heavier alloys, i.e., tungsten-based, are the first choice (Starck Fabricated Products
2013). In addition to aircraft balancing weights, other applications of tungsten- and
tantalum-based alloys in aerospace are rim components of flywheels or rotating
members in governors and airborne antenna bases. As concerns the high-temperature
applications, the explorative use of tungsten wires to reinforce nickel-based super-
alloy turbine blades can be mentioned as one of the few involving this class of alloys
(McLean 1983). Niobium and molybdenum alloys have had instead rather important
high-temperature application in aerospace. Niobium alloys are the standard choice
for ablative systems of the exhaust cones of missiles and rockets, including the
recommended protection by suitable silicide coatings (Wojcik 1991). A creep
resistance better than the benchmark CMSX4 nickel-based single-crystal superalloy
has been recorded for the Mo-9Si-8B alloy that in this respect can be regarded as an
excellent candidate for sustaining the search for more performing engines, as other
alloys of the Mo-Si-B system mentioned in this section (Heilmeier et al. 2009;
Muktinutalapati 2011). All refractory metals, together with rhenium that differs
from the other bcc metals for its hexagonal close-packed (hcp) structure, are used
as alloying elements in nickel-based superalloys (see Sect. 6.4), as solid solution
strengtheners and creep resistance enhancers, especially for their influence on the
alloy diffusivities. Indeed, as emerges from the above overview, the interest for
refractory metals and relevant alloys has been initially driven by their potential at
temperatures exceeding those accessible to advanced nickel-based superalloys.
Notwithstanding the research efforts, successful in many respects to overcome the
main critical aspects of refractory element alloys, the feeling is that other materials
9.2 Refractory Alloys 509

systems in the end will take the lead. The reason is that the search for more efficient
and better performing engines and, in general, aerospace structures cannot override
the environmental aspects, for which weight saving is a central prerequisite. Tougher
ceramics and carbon-based composites are nowadays regarded as most attractive
systems in this new development strategy (Lu et al. 2013; Richerson 2004). They
might be preferred to metallic systems considering the remarkable weight saving and
the temperature raise that they would afford. From this point of view, the evolution
of structural materials for aero-engines, to be used in the hottest spots, shows some
analogy and reminds some aspects of the development research dedicated to the
space shuttle thermal protection system (TPS), for which the possibility of using
niobium-based shield was initially considered (Buckley et al. 1981). It is well known
that the final choice was completely different, i.e., based on the lighter, cheaper, and
more performing TPS, made of ceramic felt tiles and carbon-carbon composites for
the hottest parts (Korb et al. 1981). However, the research invested in the develop-
ment of refractory alloys for aerospace has demonstrated the feasibility of using
these alloys also for other, different applications (Briant 2000). Molybdenum was
employed for microelectronic device, in flat-panel displays, using field emission tips,
rather than liquid crystals. The same occurred for tantalum-based alloys, employed
for the manufacture of electronic components, like capacitors. Moreover, the high
density of Ta alloys suggested a number of military applications, like penetrators or
gun barrel liner, capable to resist to extremely high flame temperatures, in excess of
3000
C, reached during the shot firing. To complete the picture, regarding the field
of space exploration, tantalum alloys listed in Table 9.2 were all developed for space
nuclear power systems (Buckman 2000). This is an issue that will be considered
further in Sect. 9.3 dealing with ODS alloys. Tantalum alloys were selected for the
best combination of mechanical strength, in particular creep, also under the effect of
radiation-induced deformation, and corrosion resistance, and for their compatibility
with liquid alkali metals and molten plutonium. Moreover, tantalum alloys exhibit
good fabricability and weldability, so that they are the best candidates for
nuclear applications, notwithstanding the comparatively high density.

9.2.3 Other Refractory Alloys: Chromium and the Platinum

Group Metals

Two more groups of refractory alloys are considered in this section. The first one is
based on chromium, another bcc metal (see Table 9.1). This element can be regarded
as “refractory” in the present context since its melting temperature falls above that of
nickel and, consequently, above the incipient melting temperature of nickel-based
superalloys, of which chromium is a most important alloying element (see Sect. 6.4)
(Gu et al. 2004). Chromium exhibits specific properties interesting for structural
applications at high temperatures. First of all, a low density 20% lower than nickel
and significantly lower than the density of tantalum and tungsten (see Table 9.1).
510 9 Other Interesting Alloys for Aerospace and Related Applications

Moreover, chromium features an elevated solubility for other elements, so that an

effective solid solution strengthening is potentially achievable. This latter feature
may also be accompanied by precipitation hardening, thanks to the intervention of
strengthening intermetallic precipitates, like Cr silicides. Alloying with silicon is
interesting also as concerns the oxidation behavior of these alloys, in analogy with
the observations reported for niobium and molybdenum alloys (Sect. 9.2.2). In
particular, chromium alloys can rely on the formation of a surface protective oxide
scale (Cr2O3), stable up to 950
C approximately. At higher temperatures, the
presence of a sufficient amount of silicon guarantees an outward diffusion flux of
this element, resulting in the formation of a protective, extremely stable layer of
amorphous silica (SiO2) (Soleimani-Dorcheh et al. 2014). In this way, the alloys can
be operated even up to very high temperatures without the need of applying any
protective coating, unless particularly severe conditions are faced (Gu et al. 2004).
The thermal conductivity of chromium alloys is higher than that of the majority of
nickel-based superalloys, so that thermal gradients and associated stresses remain
below critical values, with obvious advantages for the integrity and lifetime of the
structural components. Eventually, chromium is also less expensive than all the other
refractory elements in Table 9.1, and its availability on the international market is not
a critical issue. However, two main aspects have still to be improved for a wide and
reliable use of this metal and its alloys. First of all, chromium alloys show a tendency
to become brittle when exposed to air for the combined effect of the contamination
with nitrogen and oxygen, similarly to what is reported for molybdenum and
tungsten. In the second place, coherently with the previous datum, chromium alloys
show a low-impact ductility, especially at room temperature and even with a
particular sensitivity of the DBTT on the level of impurities (Gu et al. 2005). This
aspect can be remediated by introducing suitable alloying elements, like cobalt,
iridium, iron, and manganese, which scavenge and clean up the alloy matrix from the
undesired contaminants. Alloy design developments have introduced remarkable
improvements in the mechanical properties, so that chromium alloys can now be
regarded as reliable replacements for conventional superalloys (Gao et al. 2008).
From the discussion on refractory metal alloys, it turns out that a critical aspect,
common to all these materials, although to different extents, is their limited intrinsic
resistance to high-temperature corrosion. This aspect has been tackled with valuable
solutions, which have rendered these alloys useable also in environmentally aggres-
sive conditions. In Sect. 7.3.1, the importance of platinum additions to nickel
aluminide diffusion coatings (see Sect. 7.2.3) for improving their oxidation resis-
tance has been considered. Platinum and other elements of the same group, like
rhodium and iridium, indeed named platinum group metals (PGMs), do show high
melting temperatures, elevated oxidation resistance, and an adequate corrosion
resistance also when exposed to molten salts (Alven 2004; Fischer et al. 1999).
The critical issue is that the global availability of these metals is extremely limited
and consequently their prices are very high. Just as a reference value, the worldwide
extraction of iridium is estimated in 22 metric tons per year. Nonetheless, there have
been ongoing researches aiming at assessing the mechanical properties and corrosion
resistance of PGM alloys (Potgieter and Maledi 2014; Yamabe-Mitarai and
Murakami 2014). The focus has mainly been the high-temperature behavior,
9.3 ODS: Oxide Dispersion Strengthened Alloys 511

considering the fields of interest for future applications of these materials. In addition
to the excellent mechanical properties and high-temperature stability, an aspect that
contributes to render these refractory alloys particularly attractive is the similarity
encountered in the microstructural features of two-phase platinum and iridium alloys
with those of γ-γ0 nickel-based superalloys, including the dislocation structures and
dynamics (Liebscher and Glatzel 2014; Vorberg et al. 2005). This aspect is definitely
useful for the validation of material properties important for any specific application
and related thermal treatment parameters, in view of their structural applications
(Cornish et al. 2006; Vorberg et al. 2004; Wenderoth et al. 2005; Yamabe-Mitarai
et al. 2004).

9.3 ODS: Oxide Dispersion Strengthened Alloys

The leading idea that was followed in the early stages of the development of oxide
dispersion strengthened (ODS) superalloys was to extend the temperature range of
the reinforcement beyond the stability limit of the γ0 precipitates, i.e., the γ0 solvus.
Closely related to this aspect was the search for a strengthening mechanism acting on
the dislocations, once their prevailing interaction with γ0 precipitates is shifting from
cutting to climbing, as a consequence of the coarsening of the precipitates due to the
exposure to servicing conditions (Ostwald ripening, see Sect. 6.5.2). Nanometric
oxide particles were identified soon as potentially effective strengthening agents,
complying with the above requirements. In the earliest unsuccessful attempts to
obtain a fine dispersion of the particles, they were insufflated into the molten alloy,
resulting, though, in uncontrolled segregations. Therefore, this approach was aban-
doned soon. Mechanical alloying (MA) instead turned out to be the right choice.
MA, that is indeed often in the relevant alloy code name (see Table 9.4), was an
already established powder metallurgy approach, developed earlier on, originally
intended to mix up hard metal powders, like WC-Co, and to reinforce aluminum
alloys, the so-called SAP alloys (Benjamin 1990), from sinter-aluminum-pulver
(German), after the definition introduced in 1946 by the Austrian chemist Roland
Irmann. In collaboration with his co-workers, he did use high-energy ball-milling
and hot-pressing sintering to prepare aluminum samples for spectrographic mea-
surements (Pickens 1991). The unexpectedly high mechanical strength obtained in
the powder compacts was ascribed to the interplay between the oxide strengthening
and the effective sintering induced by mechanical attrition. The oxide dispersion
came, mainly, from the fragments of the aluminum oxide layer initially present on
the surface of the powder particles. The intense deformation imparted during milling
to the grains of powder determines also the formation of fresh metal-metal contact
surfaces, promoting intense intergranular bonding and diffusional phenomena dur-
ing sintering. In this regard, these experiments indicated the main effects that a
mechanical milling process, using high-energy tools, may determine in a ductile
metallic powder. In Table 9.4, the compositions of nickel-based superalloys and
aluminum alloys reinforced by a dispersion of oxide particles are listed (see
Table 9.5 for ferritic matrix alloys). The oxides used in these materials are selected
512 9 Other Interesting Alloys for Aerospace and Related Applications

Table 9.4 Compositions of some mechanically alloyed ODS nickel-based superalloys and alumi-
num alloys. Please note that oxygen concentrations quoted for these latter are those referring to
Al2O3 reinforcing particles. The composition of the main ferritic ODS alloys can be found in
Table 9.5 (Elzey and Artz 1988)
Alloy Ni Fe Al Cr Mg Li C Ti Y2O3 Others
Alloy 51 Bal. – 8.5 9.5 – – 0.05 – 1.1 B ¼ 0.01
Mo ¼ 3.4
W ¼ 6.6
Zr ¼ 0.15
Alloy 92 Bal. – 6.5 8.0 – – 0.05 1.0 1.0 B ¼ 0.01
Mo ¼ 1.5
Ta ¼ 3.0
W ¼ 6.0
Zr ¼ 0.15
IncoMAP AL-9052 – – Bal. – 4.0 – 1.15 – – O ¼ 0.5
IncoMAP AL-905XL – – Bal. – 4.0 1.3 1.15 – – O ¼ 0.5
Inconel MA 6000 Bal. – 4.5 15.0 – – 0.05 2.5 1.1 B ¼ 0.01
Mo ¼ 2.0
Ta ¼ 2.0
W ¼ 4.0
Zr ¼ 0.15
Inconel MA 760 Bal. – 6.0 20.0 – – 0.05 – 0.95 B ¼ 0.01
Mo ¼ 2.0
W ¼ 3.5
Zr ¼ 0.15
Inconel MA754 Bal. 1.0 0.3 20.0 – – 0.05 0.5 0.6 –
MA 758 Bal. – 0.3 30.0 – – 0.05 – 0.6 W ¼ 0.5
PM1000 Bal. 3.0 0.3 20.0 – – 0.05 0.5 0.6 –
PM3030 Bal. – 6.0 17.0 – – 0.05 – 0.9 Mo ¼ 2.0
Ta ¼ 2.0
W ¼ 3.5
Zr ¼ 0.15

Table 9.5 Compositions of some mechanically alloyed ODS ferritic/martensitic steels (Donachie
and Donachie 2002)
Alloy Fe Cr C Ti Mo Y2O3 TiO2
DT2906 Bal. 13.0 0.05 2.9 1.5 – 1.8
DT2203Y05 Bal. 13.0 0.05 2.2 1.5 0.5 –
DT Bal. 13.0 0.05 2.9 1.5 – 1.8
DY Bal. 13.0 0.05 2.2 – 0.5 0.9
MA957 Bal. 14.0 0.01 1.0 0.3 0.25 –
9.3 ODS: Oxide Dispersion Strengthened Alloys 513

among the phases not soluble in the metallic matrix, with incoherent interfaces, and
that at the same time may guarantee a good thermomechanical compatibility over the
whole range of working conditions. In Al-alloys, in addition to oxide particles,
aluminum carbides, e.g., Al4C3, that form in situ from the reaction between graphite
and aluminum powders guarantee a further strengthening action (Gilman and Nix
1981).The main features of the strengthening dispersoid particles, i.e., incoherence,
no solubility in the metallic matrix, and nanometric grain size, are paramount as
concerns the resulting microstructure and relevant mechanical properties of the
alloys. The application of MA process to ODS superalloys dates back to 1966,
with the pioneering works conducted at INCO by John Benjamin, using high-energy
ball-milling to mix and refine the microstructure of the different ingredients before
proceeding with the compaction and sintering steps. Nickel- and iron-based super-
alloys were the first investigated systems within a research activity that lead to the
use of the MA754 ODS nickel-based superalloy for gas turbine vanes in the F404
engine in 1977 (Froes 1990; Muktinutalapati 2011).
The production of ODS alloys starts with the preparation of a master batch mix,
comprising elemental metal, pre-alloyed alloy powders and the oxide particles.
Several oxide phases have been considered, including, in the early times, ThO2,
subsequently abandoned for its radioactivity, intolerable particularly during the
manipulation involved with processing and part manufacturing. For high-
temperature alloys, with a nickel- and iron-based matrix, the standard choice is
Y2O3, an oxide that guarantees excellent stability particularly at the highest temper-
atures, good thermomechanical compatibility, and incoherency with the metallic
matrix. For MA superalloys, in general no process control agents (PCAs), like
graphite, stearic acid, or other surfactants, are used, in order to avoid the introduction
of possible contaminations. In other systems, like MA aluminum alloys, PCAs,
including ethanol, are used instead, for reducing cold welding among particles,
and the sticking the particles to the milling media, both effects that reduce the
efficiency of the process drastically. Suitable PCAs, like graphite, are introduced
also for inducing mechanochemical reactions with the metallic powder particles,
thus promoting the in situ formation of carbide strengthening dispersions (Benjamin
and Bomford 1977; Cao et al. 2013). Eventually, if not removed, before proceeding
with the further processing steps, PCAs can be left in the milled powder on purpose,
helping the subsequent compaction and sintering stages, by reducing the inner
friction among the grains of powder and between the powder and the walls of the
compaction die.
After milling the following main processing steps are necessary to obtain the final
products:
• Preliminary consolidation of the powder up to approximately 80% of the full
density
• Hot isostatic pressing (HIP), usually in austenitic steel vacuum sealed cans to full
density
• Hot forging or extrusion for pre-forming the final product and inducing a crys-
tallographic texture in the alloy grain
• Directional recrystallization (DRX), used for ODS superalloys only
514 9 Other Interesting Alloys for Aerospace and Related Applications

After HIPing, the alloy still presents a metallic component with a micrometric
grain size, as attained by the mechanical grinding, with a nanometric oxide disper-
sion, having a stabilizing effect against recrystallization of the metal grains along
transverse to the texture, due to forging, direction. DRX induces instead a preferen-
tial grain growth precisely in the texture direction. This process is favored by the
direction of movement of the ODS product with respect to a localized thermal
gradient, realized with an induction coil heat source. Before entering the hot zone,
the alloy can be kept in a cooling bath or in an insulated region of the furnace,
sufficiently far away from the heat source. In this way, any spontaneous recrystal-
lization is limited, and the dominant grain growth occurs when the alloy piece goes
through the hot zone. An optimized microstructure is obtained when the advance-
ment speed of the alloy in the thermal gradient does not exceed the natural grain
growth velocity. This parameter depends on the alloy composition but also on the
degree of deformation texture induced by the thermomechanical treatments that
precede DRX. The microstructure induced by the DRX can be qualified and ranked
according to the grain aspect ratio (GAR), defined by the ratio between the longitu-
dinal and transverse average grain size. In general, the larger GAR, the better are the
resulting mechanical properties of the alloy, especially as concerns fatigue and creep
resistance with load applied along the longitudinal direction.
As concerns creep, the ODS alloys exhibit quite a peculiar trend of the creep rate
( ε_ ) vs applied stress (σ) curves, over the temperature and stress range where
dislocation creep is the prevailing mechanism, and a power-law creep can be
assumed to model the experimental data. The creep rate of an ODS material can
be expressed as:
σ  σ n
th
ε_ ¼ D ð9:1Þ
E

where
• D is a diffusion factor.
• E is the average Young’s modulus of the alloy.
• n is the creep exponent.
• σ th is the so-called threshold stress.
• σ is the applied stress.
The elevated values of n, as high as 20–30, and the presence of a stress below
which no detectable creep is observed can be explained in terms of an attraction
exerted by the oxide particles on the climbing dislocations.
The relaxation of the dislocation stress fields and, thereby, of their energy reduc-
tion explains the tendency for a dislocation line to remain stuck onto the surface of
the incoherent oxide particles. Once the applied stress overcomes the threshold
stress, a fast rise of creep rate is observed, since at this stage the applied load is by
far larger than any value the alloy can possibly stand. The threshold stress, being
influenced by the whole alloy response to diffusion-assisted phenomena, is also
dependent on the GAR of the alloy, increasing linearly with this microstructural
parameter (Howson et al. 1980). Thanks to the available information, it can be
9.3 ODS: Oxide Dispersion Strengthened Alloys 515

Fig. 9.2 Larson-Miller creep curves for different ODS alloys, compared to single-crystal superal-
loys (for alloy compositions see Tables 6.14, 6.16, and 9.4). (Redrawn from Harada 2003)

concluded that the threshold stress marks an important stress limit, to be considered
in the part design, to extend its lifetime. Still concerning the creep behavior, ODS
superalloys can actually reach comparable performances as single-crystal nickel-
based superalloys (Fig. 9.2), although these latter have always been, and still are, the
priority choice for blades and vanes in state-of-the-art GTEs. Several reasons can
explain this situation. The early steps of development of the ODS superalloys
occurred at the same time as the investment cast technology, applied to the produc-
tion of directionally solidified and single-crystal components. The better process
control and consequently better and more reproducible properties of the products
obtained from investment casting turned out to be particularly important in deter-
mining the success of DS and SC components with respect to the mechanically
alloyed ones. Indeed, the microstructural defects typically present in ODS superal-
loys, like recrystallization defects, i.e., fine post-milling grains surviving the DRX
process or clusters of oxide particles, are responsible for crack initiation, particularly
under low-cycle fatigue (LCF) and creep-fatigue conditions (Elzey and Artz 1988).
Therefore, ODS superalloys were used as a possible alternative to DS and SC
superalloys for high-pressure gas turbine vanes only, components that may definitely
benefit from the stability and creep resistance of ODS alloys, although less critical
than turbine blades, for the lack of applied centrifugal forces. However, also this
application of ODS is now largely abandoned, in view of the weight reduction
afforded by the investment cast hollow vanes, associated with the high-reliability
standards achieved by the EB-PVD thermal barrier coatings and consequent reduc-
tion of the peak metal temperatures.
516 9 Other Interesting Alloys for Aerospace and Related Applications

Although a substantial loss of interest for ODS superalloys is there, as concerns

aircraft gas turbine applications, in more recent years, ODS alloys have attracted a
renewed interest in view of their use for nuclear power plant applications, with
regard to both space and terrestrial reactors. Nuclear energy has been used in the past
for several space applications, before solar energy took over, whenever feasible.
USA and USSR used nuclear reactors for satellites, capable to provide the energy
required for the relevant operations. The SNAP 10A was actually the only nuclear
reactor installed on a US satellite, back in 1965, i.e., the very early years of the
“space rush.” More numerous are the reactors launched by the USSR: 33 altogether,
the majority being the BES-5 and just 2 TOPAZ reactors. Incidentally, the relicts of
these satellites continue to orbit around the Earth, if not on some occasion moved to
more delocalized trajectories, the so-called disposal orbits, not to interfere and create
potential risks and damages to newer satellite deployment. Notwithstanding these
cares, a number of disposed reactors have released radioactive matter, especially
NaK used as coolant liquid, which is now contaminating Earth orbits with droplets
and fragments with estimated diameters ranging from tenths of millimeters up to
several centimeters and an average velocity of 7–13 km/s (Reese and Vick 1983;
Wiedemann et al. 2005). The total amount of released nuclear waste is unknown and
has raised the concern of the scientific community, and the UN has issued two
recommendations on the peaceful use of nuclear power in the outer space (United
Nations General Assembly 1992). Indirectly, this situation promoted the use of solar
panels, having the advantage of being not polluting and to be deployable using
comparatively simple devices, like smart actuators relying on shape memory alloys
(see Sect. 9.5). Although the use of solar panels has been a strong push to abandon
the nuclear technology for space missions, on the other hand, there are still some that
cannot rely on energy harvesting via photovoltaic solar panels only. A few important
examples of these actual and future space mission are: outer solar system missions,
like the Rosetta ESA mission, and, considering closer to Earth scenarios, the Mars
and Moon lander missions and in the future, as concerns Mars, also manned mis-
sions. These latter will require the transportation to the Red Planet of several tons,
probably tens of tons of equipment and components. The duration of the tests
requires also overnight measurements, and test might be conducted in no ideal
conditions, like Martian nights and dust storms. From this point of view, nuclear
power plants are essential energy providers, also considering their undisputable
additional advantages, like the possibility of generating energy independently from
the orbit and distance from the sun. In principle, high power density can be reached;
different forms of energy and not just electrical can be obtained selecting the more
appropriate reactor design and heat conversion systems (El-Genk 2009). Other
missions that would be possible thanks to the deployment of nuclear reactors for
spacecraft energy production are related to robotic exploration, like removal of
abandoned spacecrafts and orbital stations, but also to the active mission of
deflecting Near-Earth objects (NEO), to protect the Earth from meteorite deep
impacts.
Nuclear reactors may also provide power for electrical or combined thermal and
electrical thrusters on interplanetary spacecrafts, as demonstrated by the NASA
9.3 ODS: Oxide Dispersion Strengthened Alloys 517

Fig. 9.3 Specific energy of space nuclear power systems versus reactor exit temperature. (Redrawn
from El-Genk and Tournier 2005)

concept studies on the nuclear thermal propulsion (NTP) and nuclear electric
propulsion (NEP) technologies. Irrespective of the technology used to convert
energy into electricity, some heat must be released at lower temperatures for obvious
thermodynamic reasons. Differently from terrestrial applications, heat rejection in
space environment is a rather critical operation and can only be achieved by
irradiation from suitable surfaces or by ejecting heated matter directly from the
spacecraft. The radiator solution has the advantage not to contribute to space
pollution, affords longer operational times if needed, and, moreover, provides
solutions that might be transferred to terrestrial plants.
Although lower working temperatures would be less critical for the materials,
which might be chosen among established alloys, still, to improve the thermody-
namic efficiency, both peak and radiation temperatures have to be increased
(El-Genk and Tournier 2005). As shown by Fig. 9.3, different classes of materials
are potentially suitable for the different reactor types. In Fig. 9.4, the temperature
ranges faced by different materials, with relevant allowable domains, considering
real working conditions and material limitations, are indicated. They depend on the
DBTT, the irradiation embrittlement, and the embrittlement determined by unwanted
contamination during processing or installation procedures.
The elevated standards in terms of performances, safety, and reliability are central
for a successful development and deployment of nuclear space reactors. Although
518 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.4 Recommended operation temperature range for structural materials in space nuclear
power systems. (Redrawn from El-Genk and Tournier 2005)

fully different in size and layout, still similar material requirements are there for
generation-IV terrestrial nuclear reactors. These new reactors represent the evolution
of the present technology and should introduce significant improvements as con-
cerns safety, operational reliability, sustainability, and useful life, as specified in the
initiative statements launched in year 2000 by the US Department of Energy
(Mansur et al. 2004). The structural materials for the space nuclear reactors
(Figs. 9.3 and 9.4) are being considered also for the novel terrestrial plants. Two
broad categories can be identified. The first one includes materials already available
and with consolidated and widespread applications in the nuclear field, whose
performances have to be reconsidered and verified for the more demanding operating
conditions foreseen in the new plants and depending on higher operating tempera-
tures and neutron doses, coupled with corrosive phenomena, determined by the
interaction of the structural materials with the coolants and fuel. In addition to
these, there is a second category, including other materials that would potentially
provide adequate responses to the Gen-IV reactor technology, although still requir-
ing research and investigations to assess and qualify their performances under the
operating conditions they would face. Ferritic martensitic (F/M) steels, initially
9.3 ODS: Oxide Dispersion Strengthened Alloys 519

developed for fossil-fuel power plants, are still important structural alloys for
Gen-IV reactors (Klueh 1993; Klueh and Nelson 2007). The compositions that
should provide the better results, having tuned the right temper conditions, are
those with 9–12 wt% Cr (see also Table 5.16). The relatively high chromium
concentration is interesting since it guarantees excellent corrosion and oxidation
resistance. Moreover, these steels have demonstrated a reduced activation
(RA) tendency, i.e., a rapid decay of the radioactivity induced by the exposure to a
neutron flux. This aspect is crucial for the decommissioning stage and disposal or
storage of the reactor parts, with relevant effects on the life cycle and life cycle cost
assessments, essential parameters to take the right and more convenient decision in
economic terms. Another aspect that renders the F/M steels particularly interesting is
their dimensional stability with respect of swelling and void formation, much better
than in other potential candidates, like austenitic steels and superalloys. Irradiation
swelling is a further deformation mechanism, active when alloys are used under
intense radiation and nuclear particles, especially neutron, fluxes (Hoffelner 2011). It
can be implemented in material-specific Ashby deformation maps, like those shown
in Fig. 9.5 for the AISI 316 stainless steel, with the occurrence of an irradiation
deformation field (Mutry and Charit 2008).
In addition to the mechanical instability and dimensional changes due to irradi-
ation swelling, the austenitic steels also suffer from microstructural evolution,
involving recrystallization and precipitate ripening, which may deeply change the
overall mechanical strength both under stationary and time-dependent stress and
irradiation conditions. Moreover, the corrosion resistance for the exposure to typical
coolant liquids used in nuclear plants, like lead and lead alloys, is insufficient to
comply with the required useful lifetime and safety requirements. These results have
confirmed a preferential choice for F/M steels and indicate a promising strategy to
enhance their properties through the development of ODS F/M steels (Byun et al.
2013; Chen 2013). The dispersion of fine oxide particles introduces vacancy sinks at
their interfaces with the metallic matrix, in which diffusional flux determines an
accumulation of these defects (vacancies). This vacancy sequestration mechanism
reduces their concentration in the alloy, with a consequent slowing down of the
diffusive phenomena. Moreover, as expected, the oxide particles hinder the dislo-
cation movements in the alloy, particularly in the higher-temperature range, where
dislocation creep, determined by the diffusive climb of dislocations, is prevailing
(Yvon and Carré 2009). The combination of these mechanisms makes F/M ODS
steels extremely stable over a wide range of operational conditions, as demonstrated
in specific studies that have been conducted to assess the performances of the alloys
listed in Table 9.5 (Chen and Hoffelner 2009; Miller et al. 2005, 2006).
The ODS approach has also been used for refractory metals, to provide additional
strengthening mechanisms to these alloys (Patra et al. 2015). However, the limited
intrinsic ductility of refractory metals and their alloys (Sect. 9.2) renders the control
of the defects introduced by the mechanical alloying technology even more critical
than in standard alloys.
520 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.5 Deformation maps for AISI 316 stainless steel. (a) Standard conditions and (b) under
irradiation conditions. Note the additional deformation mechanism associated with this second
regime. (Redrawn from Mutry and Charit 2008)

9.4 Intermetallic Compounds and Ordered Alloys

In previous chapters, several intermetallic phases in alloy systems used for aircraft
structures and engine parts have been cited. In hardenable Al-alloys, equilibrium
intermetallic phases, like Al2Cu, in 2xxx or Mg2Si in 6xxx alloys, would form in the
overaged conditions (see Sect. 3.2). These phases should be avoided, since their
presence is in general indicative of an alloy temper featuring poor mechanical
9.4 Intermetallic Compounds and Ordered Alloys 521

Table 9.6 Main advanced aluminide alloys, based on the Ni3Al phase (Sikka et al. 2000)
Alloy Al Cr Fe Zr Mo B Ni
IC50 21.73 – – 1.15 – – 77.12
IC-74M 23.51 – – – – 0.04 76.45
IC-218 16.66 15.29 – 1.57 – 0.10 66.39
IC2181Zr 17.00 15.82 – 0.39 – 0.04 66.75
IC-221 16.73 15.35 – 3.35 – 0.04 64.54
IC357 18.36 13.53 21.64 0.77 2.51 0.04 43.15
IC-396M 15.96 15.36 – 1.60 5.98 0.02 61.08

properties. The γ0 ordered phase, based on the L12-ordered Ni3Al compound, is

fundamental as concerns the mechanical properties, both static and time-dependent,
of iron- and nickel-based superalloys (see Sects. 6.3 and 6.4). The dependence of the
yield strength of Ni3Al on the temperature (see Fig. 6.21), a consequence of the
dislocation structure and dynamics in the ordered crystallographic structure of this
alloy, has driven the research for single-phase intermetallic alloys, based on the
abovementioned binary compound. An alloy fully made of the strengthening phase
of the nickel-based superalloys was expected to have remarkable mechanical per-
formances. Alloy design studies provided indications on how to improve mechanical
strength of the binary intermetallic and, at the same time, increasing its ductility and
fracture toughness up to values suitable for both processing and structural applica-
tions (Jiao 2016; Pettifor 1988a, b). Boron “doping” (concentration in the 0.1 wt%
range) was employed and turned out to be an effective approach to strengthen grain
boundary bonding, with a consequent beneficial effect on room-temperature ductil-
ity. Advanced aluminides (Table 9.6) were the alloys developed in this context (Aoki
and Izumi 1979; Liu et al. 1985; Sikka et al. 2000). However, they never found their
way to aerospace applications, since the overall mechanical behavior was definitely
worse than state-of-the-art nickel-based superalloys: the lack of the γ-γ0 interfaces
and the optimized microstructure that can be achieved in multiphase alloys (see Sect.
6.4), using appropriate thermal treatments, are the main reasons for this outcome.
In view of the high oxidation and carburization resistance, as it is for wear and
cavitation-erosion resistance of these alloys, several terrestrial applications have
been considered instead: furnace rolls, radiant burner tubes for steel industry, forging
dies, and corrosion resistant fixtures for chemical industries, center posts for
pit-carburization furnaces (Stoloff 2000). A similar development has regarded also
the iron aluminide alloys, based on the FeAl and Fe3Al compounds. Their excessive
brittleness was mitigated using ternary additions, like Mo, Nb, Ti, and Zr also in
association with second phase particles, like TiB2 (Liu and Kumar 1993; McKamey
et al. 1991). Boron doping turned out to be detrimental for Fe3Al, whereas it
provided strengthening and toughening effects in FeAl, depending on the actual
composition of the alloy (Baker et al. 1998). Several alloys have been developed and
used as steel replacement in components working in aggressive environments, like
coal gasification plants (Saunders et al. 1997). With reference to nickel aluminides,
another intermetallic has been extensively investigated: NiAl. NiAl, together with
522 9 Other Interesting Alloys for Aerospace and Related Applications

CoAl and FeAl, is the main phase in diffusion coatings (see Sect. 7.3.1), the
oxidation protective coatings used for protecting nickel- and cobalt-based superal-
loys and steels, respectively, against oxidation and other high-temperature corrosion
phenomena. NiAl is also investigated as possible replacement for nickel-based
superalloys, considering the interesting properties that this intermetallic phase
exhibits, including a low density, i.e., 5.9–6.3 g/cm3, depending on the actual
alloy composition, that is, approximately two thirds the density of nickel-based
superalloys, typically in the 8.6 g/cm3 range (Darolia 1991; Yu et al. 1993). The
thermal conductivity of NiAl, in the 100 W/mK range, is from four to eight times
higher than the thermal conductivity of nickel-based superalloys, an important
prerequisite to keep the internal thermal stresses below critical values. Coherently
with its mentioned use in diffusion coatings, NiAl has an excellent oxidation
resistance. Although based on the simple cubic structure, featuring a lower number
of active slip systems than the austenitic structure of the superalloys, still this
crystallographic structure can guarantee a sufficient level of plastic deformation,
far better than the complex low-symmetry structures of silicides and other exotic
intermetallics, discussed further on in this section (Miracle 1993). Eventually, the
significantly higher incipient melting temperature (1638
C for the stoichiometric
NiAl composition) corresponds to a higher material stability, particularly as con-
cerns creep deformation and other diffusion-assisted phenomena. Several processing
routes have been adopted over the years to produce NiAl-based alloys and to
manufacture components, including prototypes of jet engine parts. Following the
approach used for single-crystal nickel-based superalloys, single-crystal and
directionally solidified NiAl materials have been developed (Cui et al. 2005).
Directional solidification is particularly interesting for eutectic systems, like NiAl-
Mo, NiAl-Cr, NiAl-Cr(Mo), and NiAl-Cr(Fe), although the fracture toughness of
these alloys turned out to be not always adequate for structural applications (Bei and
George 2005; Wang et al. 2016, 2017; Yang et al. 1997). The microstructure of these
alloys resembles that of composite materials, with a matrix phase, i.e., NiAl,
reinforced and toughened by “long fibers” or lamellae of chromium and/or molyb-
denum. For this reason, eutectic alloys are often named in situ composites, to
highlight the compositional and microstructural features of these materials, obtained
directly from the melt. Eutectic alloys considered for high-temperature applications
involve also other intermetallic compounds, like the Nb-Nb3Al, Cr-Nb, NiAl-
NiAlNb, Nb-Nb5Si3, Nb-Nb5Si3-Nb3Si, MoSi2-Mo5Si3 (Bewlay et al. 1995; Cao
et al. 1994; Mason and Van Atkin 1993; Mendiratta and Dimiduk 1993; Reuss and
Vehoff 1990; Reviere et al. 1992; Whittenberger et al. 1992a; Yu et al. 2017).
Another group of these materials comprises composites in which a metallic matrix
is reinforced by refractory metal carbides, in the form of long fibers: Co-TaC
(COTAC), NiTaC (NITAC), and Cr-Cr2C3 (McLean 1983). Because of the excep-
tional mechanical properties, these alloys had attracted a great deal of interest in the
early days of their development. However, although part of this interest still sur-
vives, the critical aspects associated with material processing, like an intrinsic low
growth rate, still pose serious limitations to a widespread application of these
materials in structural applications at high temperatures.
9.4 Intermetallic Compounds and Ordered Alloys 523

A further development of NiAl alloys concerns alloys strengthened by the

precipitation of secondary phases. The early investigations on these systems date
back to the study of the two phase, β-NiAl-β0 -NiAl2Ti, alloy. The β0 -NiAl2Ti Heusler
phase displays a particularly good creep resistance that can be improved further
when combining the two intermetallic phases in a single alloy and optimizing the
relative fraction of the two (Polvani et al. 1976; Strutt et al. 1976). Microscopy
observations confirmed that the increased mechanical resistance of the two-phase
alloys, reaching the level of a nickel-based superalloy, i.e., MAR-M200, was
strongly dependent on the interaction of the dislocations with the semi-coherent
interfaces between β and β0 phases. Being based on cubic structure with bcc-like
symmetries, these alloys were regarded as the bcc counterparts of the conventional
nickel-based austenitic superalloys so that they were named the “bcc nickel-based
superalloys.” The concept of two-phase intermetallic alloy was extended also to the
FeAl-Fe3Al alloys, which coherently with the previous denomination can be defined
as “bcc iron-based superalloys” (Morris and Gunther 1997). In the above studies, the
two-phase alloys were polycrystalline and obtained by conventional casting, featur-
ing quite a low room-temperature ductility to fracture, below 2%. From this point of
view, a significant improvement was obtained by setting the composition in the
three-phase field, in which the β and β0 phases would coexist with the γ0 phase (Yang
et al. 1991, 1992). The three-phase materials, developed with different compositions,
introducing additional alloying elements, turned out to have very interesting
mechanical properties, including better deformability, also in this case comparable
to a standard nickel-based superalloy, i.e., Udimet 720 (Yang et al. 1992). An
important step forward was the development of multiphase intermetallics, based
on NiAl precipitation hardened by another Heusler phase: Ni2HfAl. The formation
of Ni2HfAl is promoted by 1 at% of hafnium approximately in the formulation, on
some occasion added together with zirconium (Locci et al. 1996). Silicon was also
introduced, with generally positive effects on the resulting properties of the alloy,
thanks to the formation of hardening precipitates (G-Ni16Hf6Si7) (Chen et al. 2000;
Whittenberger 1999). These studies highlighted an important processing issue of
these alloys, having either a single-crystal or directionally solidified microstructure.
Indeed, the presence of silicon oxide coming from the erosion of the mold was
observed. The prolonged contact between the molten metal and the mold inner wall
and the very high temperatures involved may easily result in the contamination of the
alloy during the solidification process. To improve this aspect, specific casting
technologies have been developed for a better control of the alloy final composition
and microstructure (Scheppe et al. 2002). Alternatively, a powder metallurgy
approach, using plasma-spheroidized and subsequently HIPed powders, has been
developed, with excellent performances of the consolidated products (Kaplanskii
2018). Another approach that was successfully applied to the strengthening of the
NiAl intermetallic was through the dispersion of secondary ceramic phases using the
mechanical alloying process, discussed already with regard to ODS alloys (Sect.
9.3). Two main powder metallurgy techniques have been used for processing NiAl-
based alloys: a standard mechanical alloying process, similar to that used for
producing an ODS NiAl alloy, and cryomilling, i.e., high-energy ball-milling of a
524 9 Other Interesting Alloys for Aerospace and Related Applications

NiAl alloy in liquid nitrogen (Whittenberger et al. 1990a, b). In the first approach, as
for ODS superalloys, a limited concentration, usually not exceeding 5 wt%, of a fine-
grained thermodynamically stable phase is distributed throughout the intermetallic
matrix, so to provide an obstacle to the dislocation movement, particularly at high
temperatures. The very first attempts to obtain the oxide dispersion strengthening of
the NiAl alloy involved several different oxides, like Al2O3, ThO2, and Y2O3,
although the thoriated material only resulted to be present as a nanometric dispersion
within the intermetallic alloy, whereas the particles of the other oxides clustered
together in rather coarse agglomerates (Koch and Whittenberger 1996). The alter-
native approach, based on cryomilling, was quite effective in attaining a good
dispersion and grain refinement, also as concerns the metallic component, by
avoiding cold welding. An important output of the mentioned research was not
only an effective metallic grain comminution and fine dispersion of the ceramic
phase, Y2O3 in this case, but also the formation of AlN particles (Whittenberger et al.
1990a, b; Witkin and Lavernia 2006). This occurred for the reaction of liquid
nitrogen with the metallic powder, sustained by the mechanical grinding.
Cryomilling of NiAl and the formation of AlN were investigated further in order
to optimize the properties of the products, particularly in their consolidated
condition. Several processing aspects were considered, like the interplay between
in situ-formed AlN and the ceramic dispersion added to the NiAl powder on purpose
and the possibility of obtaining more stable nitride phases, with the introduction of
suitable reacting metals, e.g., zirconium (Whittenberger et al. 1999). In view of the
positive strengthening effects of the AlN dispersion, alternative routes were
attempted, with the idea of bypassing cryomilling. Roasting treatments of NiAl
powders in a nitrogen atmosphere resulted in the formation of the nitride phase,
although the mechanical properties of the consolidated samples did not reach the
level of cryomilled materials. However, the mentioned treatment produced a general
improvement in the powder quality, as regards sinterability, most probably for the
elimination of the surface contamination present in the grains of metallic powder
(Whittenberger et al. 2000). Nowadays, it is acknowledged that the powder metal-
lurgy route is to be regarded as most promising for the technological exploitation of
NiAl-based alloys. Interestingly, also for NiAl alloys, as it happened in the early
stages of the development of ODS superalloys (Sect. 9.3), a molten metal approach
was adopted. Rapid solidification was in fact used to incorporate HfC, TiB2, and TiC
in NiAl (Jha et al. 1989). The HfC-hardened material provided mechanical proper-
ties, also at high temperatures, definitely better than the un-reinforced matrix (Ray
et al. 1989; Whittenberger et al. 1990b, 1992b). Creep curves exhibited a very
similar trend as for ODS alloys, i.e., an increase of the stress exponent, up to values
as high as 20, and the presence of a threshold stress, below which no detectable creep
rate is recorded. Nonetheless, the excessive costs of the rapid solidification plants,
the comparatively limited productivity, and the difficulty of scaling up make this
technology not so attractive, particularly if compared to mechanical alloying.
Beyond nickel aluminides, intermetallic compounds continue to attract the inter-
est in the search for materials capable to be used at temperatures above the present
limits of nickel-based superalloys, i.e., a metal temperature above 1100
C. The
9.4 Intermetallic Compounds and Ordered Alloys 525

melting temperature might definitely be an effective selection parameter, considering

that higher-melting temperatures correspond to lower diffusivities and therefore
better creep performances (Fairbank et al. 2000). Moreover, the ordered structure
of intermetallics enhances this aspect further, in view of the lower diffusivities in the
ordered rather than disordered state (Schoijet and Girifalco 1967; Schoijet and
Girifalco 1968; Wang and Akbar 1993). These factors have addressed the research
for high-temperature ordered intermetallics toward PGM and silicide alloys, both
systems being considered already in Sect. 9.1 (Zhao and Westbrook 2003). The
counterpart to the ordered structure of intermetallic compounds is their extremely
limited ductility, particularly at room temperature, and fracture toughness. In fact,
the strong covalent character of the chemical bonding and the limited number of slip
systems together with the complex, low-symmetry crystallographic structures render
these alloys in many respects similar to ceramics. The approach successfully applied
to nickel aluminide intermetallic alloys, based on the development of multiphase
systems, is being followed also for these higher-temperature systems, with excellent
results (Mitra 2006; Petrovic and Vasudevan 1999; Perepezko et al. 2014). This
concerns not only the structural properties but also the oxidation and high-
temperature corrosion resistance (see Chap. 8).
Still referring to multiphase systems, intermetallic matrix composites have been
investigated (Ward-Close et al. 1996). These materials feature a matrix made of one
or more intermetallic phases, like nickel, iron, titanium aluminides, niobium, and
molybdenum silicides, to be selected on the basis of the target applications. Since the
main task for intermetallic matrix composites is increasing the fracture toughness of
the matrix, ductile metallic reinforcements are used, like in case of NiAl-Mo and
NiAl-Cr eutectic alloys. Using directional solidification, lamellar NiAl-Mo struc-
tures with excellent fracture toughness have been produced. Alternatively, the
intermetallic matrix can be reinforced with high-strength ceramic particles or fibers,
which may have a beneficial effect not only on static strength but also on creep
resistance at higher temperatures. High-strength NiAl particle composites have been
obtained using a combustion synthesis, a processing technology particularly suitable
for intermetallics that usually have relatively large negative enthalpies of mixing that
can be exploited to sustain the composite formation reaction. In this example the
reinforcement is provided by NbB2 and NbxC, although a very broad range of matrix
reinforcement can be found in the literature, involving NiAl and several other
intermetallic systems (Bochenek and Batista 2015; Mitra 2018). The specific fea-
tures of the intermetallic phases have required design strategies and processing
techniques not necessarily coincident with those applied to metal matrix and ceramic
matrix composites. In the first ones, the ductile and tough metal matrix is reinforced
by strong and stiff ceramics, like Al2O3, SiC, TiB2, etc., introduced as fibers or
particles. In ceramic matrix composites, the brittle matrix is toughened by the
presence of weak interfaces, which become preferential pathways for crack propa-
gation, whose energy is thus absorbed by the fracture surface increase.
Nonetheless, the “multiphase” approach will help overcoming the present obsta-
cles to the aerospace applications of potentially interesting intermetallic compounds,
526 9 Other Interesting Alloys for Aerospace and Related Applications

thus extending the list that comprises already very important examples, like nickel,
cobalt, and iron aluminides in overlay coatings (see Sect. 7.3.1) and titanium
aluminides, especially γ-TiAl-based alloys, in low-pressure stage gas turbine blades
(see Sect. 4.6) (Lasalmonie 2006).

9.5 Shape Memory Alloys

Another family of intermetallic compounds, interesting for the aerospace research, is

that of shape memory alloys (SMAs), of which the B2 compound NiTi is the main
and most notorious representative. The potential of shape memory alloys in aero-
space applications stems from the phenomenology of these materials. SMAs are
capable to react directly, in terms of strain and shape changes, to temperature,
magnetic field, and stress variations. The shape changes and the relevant developed
stresses can be used to operate switches, actuate devices, produce displacements, etc.
This aspect allows to reduce the number of components to be used and also the
design complexity of thermomechanical devices. An immediate consequence of
these features is the weight reduction of the systems involving SMAs parts, an
aspect that is always interesting in aerospace components and structures. The
possibility of tuning the operating temperature range, the recoverable deformation
and stress, adjustable through the selection of a suitable alloy, and relevant sequence
of thermomechanical treatments are all additional qualifying properties of SMAs. In
this section, a few important aspects of the SMAs phenomenology interesting for the
aerospace field are recalled. Then, the main applications of SMAs to aerospace
structures and devices are illustrated with the indication of the main and most
promising developments routes.

9.5.1 Main Phenomenological Aspects and Relevant

Mechanisms

The main aspects of the shape memory phenomenology derive from a few structural
and microstructural aspects, which are related to the martensitic transformation that
occurs in these alloys. Different from the martensitic transformation in steels that is
notoriously irreversible and as such should not be treated as a genuine phase
transition, the martensitic transformation occurring in SMAs is reversible. On the
other hand, the steel-related naming is adopted for the involved SMA phases. The
low-temperature phase is indeed called martensite, the high temperature one aus-
tenite. The denominations for the transition temperatures are coherent with this
convention: upon cooling, austenite will transform into martensite over a tempera-
ture range going from martensite start (Ms) down to martensite finish (Mf). The
reverse transformation occurs from As to Af, austenite start and finish, respectively,
9.5 Shape Memory Alloys 527

Fig. 9.6 Schematic of the displacive martensitic transformation occurring in SMAs. Upon cooling
(T< Ms) the initially fully austenitic alloy (a) starts to transform into martensite (b). The transfor-
mation proceeds for the coherent displacement of large number of atoms for small distances. During
the transformation the two phases coexist across a common interface plane (b–c). At T< Mf, all the
austenite is transformed into martensite (d). The reverse path from (a–d) may occur, because the
transformation is reversible. (Redrawn from Duerig et al. 1990)

on heating martensite until it fully transforms back to austenite. In the NiTi alloys
that are by far the most widely used SMAs, austenite has a cubic CsCl B2 structure,
isomorphous to the β-NiAl phase (see Sect. 7.3.1). Martensite in binary NiTi alloys
is generally monoclinic. The presence of dislocations and other defects introduced
by plastic deformation may result in the formation of an intermediate rhombohedral
phase (R-transformation). Eventually, ternary additions, e.g., copper in concentra-
tions exceeding 10 wt%, determine the formation of a martensitic phase based on the
orthorhombic structure (Duerig et al. 1990). All the involved transformations are
displacive. The displacive character depends on the fact that the transformation from
one phase to the other does not involve any compositional change and, therefore, no
diffusive mass transport is needed. Moreover, in analogy with the martensitic
transformation in steels, the crystalline lattice of the austenite can be obtained by a
distortion of martensite and vice versa (Bain strain). Therefore, the transformation
occurs through a coordinated displacement of large number of atoms (military
transformation), moving for very short distances, smaller than the average
interatomic distance. The transition is a first-order one, i.e., a latent heat exchange
is involved, and the coexistence of the two phases is observed as the transformation
proceeds. The two-dimensional scheme in Fig. 9.6 depicts the essential aspects
described so far.
In real alloys, the transformation takes place over a finite temperature range and
with a hysteresis on cooling with respect to heating. This is due to the mechanical
528 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.7 The two mechanisms intervening for accommodating the shape change due to a martens-
itic transformation: (a) slip and (b) twinning. (Redrawn from Duerig et al. 1990)

constraint faced when the new phase is forming from the parent one. Therefore, both
austenite and martensite are stabilized below and above the equilibrium transition
temperature, respectively.
The crystallographic structure of austenite can be obtained through the lattice
strain of martensite. The transformed volume of the alloy would generally have a
different shape from the initial one (see Fig. 9.6a, d). This requires the accommo-
dation of the new shape within the volume initially occupied by the parent phase.
Slip and twining are the two mechanisms intervening in this process. In general, slip
is observed in those systems exhibiting differences in the molar volume between
austenite and martensite larger than 1%, so that shape accommodation involves
comparatively high stresses. When martensite and austenite have similar molar
volumes, twinning is the main shape accommodating mechanism. The fundamental
difference between slip and twinning is that slip involves breaking of atomic bonds,
so that the structure is irreversibly deformed. In twinned structures, the shape
accommodation is attained through the orientation of the so-called twin variants,
just two in the simple bidimensional model in Fig. 9.7b, more numerous and with
complex orientation relationships in case of real 3D systems. The twin variants have
highly mobile interfaces (twin boundaries) as a consequence of their low surface
energy. In case an external stress is applied, all variants tend to align parallel to the
stress direction, a process that is assisted by the movement of the twin boundaries. At
the same time, all variants that are not favorably oriented tend to disappear. Mar-
tensite de-twinning is a particularly important deformation mechanism for shape
memory alloys, occurring at nearly constant stress, in agreement with the claimed
mobility of the twin boundaries, and reaching strain values up to 8% approximately.
Once de-twinning is complete, the elastic deformation of de-twinned martensite will
start, up to the yield stress of the alloy. These structural and microstructural aspects
are the basis for two important phenomena exhibited by SMAs: the shape memory
effect (SME) and the superelastic behavior.
As concerns the SME, the main idea is that the initial shape that the SMA part is
having in its austenitic phase will be recovered once the part is deformed, so that the
9.5 Shape Memory Alloys 529

Fig. 9.8 Schematic of the base principles of the shape memory effect (SME). The austenitic alloy is
cooled below Mf, so that it turns to martensite, with the usual twinned structure. The deformation of
martensite involves the rearrangement of the twin variants. Upon heating up the deformed mar-
tensite, the reverse transition to austenite is accompanied by the recovery on the original form.
(Redrawn from Duerig et al. 1990)

original shape is changed, in the martensitic field. With reference to the sequence of
events depicted in Fig. 9.8, the austenitic alloy is cooled down in the martensitic
field, and its shape is changed with a deformation based on the reorientation of the
twin variants. Once the alloy is heated up back to austenite, the transition to the high-
temperature phase is accompanied by the return to the initial shape of the alloy.
Two aspects are essential to the shape memory effect. First, the shape change
imparted to the alloy in the martensitic phase that can be actually recovered is only
due to the rearrangement of the twin structure. Any fraction of deformation involv-
ing dislocation movement and slip will not be recovered, and it is thereby called
alloy “amnesia.” Second, the springback effect that allows the return to the initial
austenite shape, when the reverse martensite-austenite transition occurs, is an
implicit consequence of the lower crystal symmetry of the martensitic, with respect
to the austenitic phase. This means that, while there are several ways of obtaining
martensite from austenite, there is just one to obtain austenite starting from
530 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.9 Stress-strain curve

of a superelastic SMA
(Cu-39.8%Zn) starting from
the austenite field. (Redrawn
from Duerig et al. 1990)

martensite. In the simple bidimensional scheme in Fig. 9.8, this corresponds to the
fact that from the square symmetry of austenite, two martensite variants can be
obtained. On the contrary, starting from these latter, just one austenite variant is
obtained.
The Bain strain relationships existing between martensite and austenite can be
exploited to induce the transformation by deforming the austenite structure to
martensite also above Ms. This is the so-called stress-induced martensite (SIM)
and on this mechanism superelasticity (SE) is based. SE is the capability of SMAs
to fully recover even relatively large deformations, i.e., 8% or so. Actually, the
applied stress inducing the austenite to martensite transformation is by all means
acting as a temperature reduction. Therefore, SE is regarded as a thermoelastic
process, since there is a complete equivalence in the temperature and stress response
of the alloy. Once the stress overcomes the onset value for starting the transformation
(σMs), the first martensitic nuclei start to form, aligned along favorable variant
directions. The transformation is completed once all austenite is transformed under
the effect of the applied stress to martensite. At still higher stresses, the elastic
deformation of the SIM is observed up to its yield stress (σyM). At any stress below
σyM, upon unloading the alloy, a complete recovery of the deformation is attained for
the reverse martensite to austenite transformation (Fig. 9.9).
The stress needed to initiate the martensitic transformation (σMs) depends on the
temperature, as it is to be expected for the equivalence of stress and temperature
involved with thermoelasticity. In particular, the higher is the temperature of the
alloy above Ms, the larger is the stress required for its transformation from austenite
to martensite. At temperatures above Md (see Sect. 5.4.3.1), the stress necessary to
induce the martensitic transformation would be above the yield stress of austenite.
Therefore, no transformation to martensite is obtained but just the plastic deforma-
tion of austenite. An interesting approach to the combined exploitation of both SME
and SE has been proposed. The leading idea is not to use a thermally but a
9.5 Shape Memory Alloys 531

mechanically, via SE, induced martensite in SME applications (Casati et al. 2014).
The relevant thermomechanical cycle for devices using the SME is illustrated in
Fig. 9.10a. In practice, the stress and strain available to any potential applications
based on the SME are those provided by the de-twinning of martensite and relevant
values of stresses, particularly interesting in case of constrained recovery of the
shape. Figure 9.10b shows how the stress field that may be provided by the SMA
active part can be enhanced if the working stress field is raised to the value of the
upper stress of the superelastic cycle. This renders particularly attractive SMA
actuators based on high-performance SME, for the increased operating
performances.
SME and SE are intrinsic properties of the SMAs. It is clear from the description
of SME that after the first transformation loop, illustrated in Fig. 9.8, once the alloy
has returned to the initial austenite shape, a further loop requires necessarily an
external stress to deform the alloy in the martensite field. It is possible to “train” the
alloy and induce another effect, the so-called two-way SME (TWSME). It consists in
the repeated change of the alloy shape by just varying the temperature across the
martensitic transition interval. Thanks to the TWSME, in the same alloy, two
different shapes can be recovered by just changing the temperature above Af and
below Mf. The “training” of the alloy, which can be conducted in different ways
(Auricchio et al. 2003), generates a matrix of structural defects and associated stress
fields, introducing into the alloy an anisotropic texture condition. Therefore, also the
twinned martensite that forms upon cooling austenite will produce a macroscopic
shape change for the arrangement of the variants induced by the textured micro-
structure. As compared to SME that can also be called one-way SME (OWSME) for
obvious reasons, the TWSME, although very interesting in several respects, exhibits
some important limitations, to be taken into the right consideration for application
purposes. In the first place, the strain excursion required to oscillate between the two
shapes tends to degrade very rapidly, so that TWSME is not recommended in case a
large number (more than 100) of cycles are required. Secondly, the stress levels
achievable via TWSME are by far lower than those involved with OWSME.
Nonetheless, there are several applications, particularly those involving controlled
displacements that may benefit from TWSME.

9.5.2 Aerospace Applications of Shape Memory Alloys

The early observation of the SME was made in Au-Cd alloys, although the potential
and consequent interest for the shape memory effect roused after the research work
on NiTi alloys (Buehler et al. 1963; Kaufman and Mayo 1997; Ölander 1932). The
unexpected results were obtained within a study on alloys for nuclear applications,
carried out at the Naval Ordnance Laboratory, after which the new family of alloys
was named (Nitinol). The first Nitinol alloy that was commercialized was registered
as Tinel®. Its first application was on the US Navy’s Grumman F-14 “Tomcat” jet
fighter in 1971, as a coupling system for hydraulic pipes, code-named Cryofit, after
532 9 Other Interesting Alloys for Aerospace and Related Applications

Fig. 9.10 (a) Thermomechanical loop for the standard SME and for (b) the high-performance
SME. (Redrawn from Casati et al. 2014)
9.5 Shape Memory Alloys 533

the cryogenic temperatures involved in its installation (Melton 1999). This cylindri-
cal coupling, machined from a bar of Nitinol, had in the austenitic phase a diameter
smaller than the external diameter of the two pipes to be connected. The working
principle is the OWSME, and therefore an important parameter is the temperature at
which the joint is meant to operate. At this temperature the alloy has to be fully
austenitic. The positioning of the Cryofit across the joint between the two pipes
requires its deformation in the martensite field. This is obtained using liquid nitrogen
and an appropriate tool, like a tapered mandrel. The installation is accomplished by
heating up the SMA coupling, back to the austenitic field. This reverse transition
may also occur spontaneously, just leaving the Cryofit warming up at ambient
temperature. To enhance the contact stress and to avoid unwanted slippages, radial
teeth are present on the inner surface of the Cryofit. Therefore, the installation of the
Cryofit can be completed quite quickly and using very basic tools only, mainly for
handling the cryogenically cooled part and to deform it in a controlled way, i.e.,
without introducing unwanted plastic deformations that might depress the recovery
capabilities of the SMA. The space required to conduct the whole procedure is
relatively limited, which renders the approach very interesting in case of a high
density of piping connections. A connection based on SMAs exhibits also the
so-called “live joint” behavior, i.e., the capability to accommodate even large
deformations without changing the tightening stress. This behavior descends from
the superelastic effect of the austenite phase, i.e., the constant stress plateau above
σMs (Fig. 9.9). After the early applications of Cryofit in the military field, for which it
was certified soon after the successful tests, the use was subsequently extended to
commercial aircrafts too, and since then Cryofit has become a standard for the
aerospace pipe connections. These fittings are applicable to connect pipelines with
maximum internal pressures of 40 MPa and represent an interesting solution not only
for new installation but also for repair operations.
The potentials of SMAs disclosed more aerospace applications, particularly once
the main issues concerning the reproducibility of the mechanical behavior of these
alloys and their critical dependence on the actual composition have been better
understood and managed (Lewis 1997). The weight saving and, most importantly,
the performance reliability introduced by SMA actuators are particularly attractive
for reducing the weight of satellites and extraterrestrial mission spacecrafts. These
reasons have guided the development of several devices that have become an
absolute must for the release and deployment of payloads from space carrier
vehicles. The Frangibolt, developed and commercialized by the TiNi Aerospace
Inc., is a certified deployment system, which has replaced, in most space missions,
explosive devices (Johnson 1992). Its working principle is still based on the
OWSME. A NiTi cylinder is coaxially mounted to a notched bolt, which is holding
the payload in place during the take-off and preliminary steps of the flight. In this
condition the SMA is martensitic. A resistive heating system is integrated into the
cylinder, and it is switched on to activate the payload release. The SMA cylinder
becomes austenitic and tries to recover the relevant extended shape that, for this
application, is the shape to be “remembered.” The stress developed in the process
breaks the notched bolt and thus releases the payload. The typical weight of each
534 9 Other Interesting Alloys for Aerospace and Related Applications

active element is 100 g approximately, and the force developed is 45–50 kN, to be
compared with a force of 20–25 kN of the load-holding notched bolt. The main
advantage of the Frangibolt, with respect to an explosive deployment device, is the
possibility of pretesting each component, whose correct operation is paramount and
absolutely critical for the full success of the space mission. The deployment dynam-
ics of the Frangibolt involves a much more limited shock energy than the explosive
approach. Moreover, being essentially based on an electrical current regulation, it is
definitely more controllable than an explosion, and this assures a better control of the
correct orbital positioning of the released object, without the need of subsequent
adjustments.
The Pinpuller is another release device using a Nitinol wire, in which a pin is
unlocked and actuates the deployment. This device was used for the first time in a
space mission on the Mars Global Surveyor in 1996 (Bokaie et al. 1998). In the rest
condition, the pin, on which acts a compressed drive spring, is kept in the extended
outward position by a retractable detent system. The remaining blockage of the pin is
guaranteed by a set of balls kept in outward position by a suitable ball keeper and a
latch, connected to the NiTi wire, martensitic in the rest position of the Pinpuller.
When the wire is resistively heated and goes back to the austenitic phase, it contracts,
for the SME, and, through the latch, drives the balls inward, thus unlocking the pin
that is retracted under the effect of the drive spring. Typical operating parameters for
NiTi wire that is the active part in Pinpullers are pull force ranging from 20 to 450 N,
with a corresponding stroke range from 6.3 to 16.0 mm. Another important and
interesting feature of the Pinpuller is its extremely short actuation time, estimated in
a few milliseconds.
Another operation very important for space orbital stations and satellites is the
deployment and unfolding of solar panels and antennas that are launched stowed and
fully folded for obvious space restriction in the hold. The conventional approach,
before the introduction of SMA systems, was again based on nuts explosively
fractures and hinges connected to spring dampers. All these techniques have been
now replaced by a deployment system made of a NiTi strips (Carpenter and Lyons
2002). The strips, folded in the rest condition (martensite), are heated up to the
austenite unfolded condition by a nickel-chrome resistive heater strip, stuck onto the
inner surface of the SMA element. After deployment the SMA part goes back to
martensite, but the deployed shape is of course retained. In principle, as seen when
introducing the OWSME, the “smart” strip might be folded back again by an
external force and the deployment repeated. In real applications, this is not really
the case, since, once deployed, the solar panels and satellite antennae remain in
position for the rest of their lifetime.
An active device, comprising also a reloading mechanism for the SMA wire, was
developed to move a glass screen back and forth over an element of the solar panel
installed on the Mars Pathfinder micro-rover (Stone 1996). This device was meant to
monitor and evaluate the effects of the accumulation of Martian dust on the solar
panel efficiency under different coverage conditions (Jenkins and Landis 1995). This
is a paramount aspect to be considered in view of future expeditions and possible
human settlements on the Red Planet.
9.5 Shape Memory Alloys 535

In addition to the above applications, since the early 1990s, several studies aimed
at applying SMAs to active aircraft structures. The DARPA “Smart Wing” project
considered the feasibility of using SMA actuators, located within the wings, to
modify their profile (Jardine et al. 1997; Kudva 2004). A similar approach, relying
on actuators inside the components, had been proposed earlier on for changing the
geometry of the aircraft wings (Strelec et al. 2003). A similar approach has been
adopted for the intake engine and exhaust ducts, demonstrated in a realistic envi-
ronment on a full-scale F-15 smart structure within the DARPA-funded SAMPSON
project (Pitt et al. 2001, 2002). This idea was developed further, with an approach
still based on the OWSME, developed to modulate the gap of the nozzle of a bypass
fan gas turbine engine, in order to reduce the jet noise during take-off and the shock-
cell noise under cruising flight conditions. At the same time, the control of the fluxes
emerging from the rear part of the engine should also optimize specific fuel con-
sumption, particularly during altitude cruise flight (Calkins et al. 2006).
SMA-functionalized chevrons have been developed and dislocated along the trailing
edge of the thrust reverser. Three Nitinol elements, including resistive heaters, are
embedded into the chevron body, made of composite material (typically carbon
fiber-reinforced polymer matrix). The correct selection of the composite stiffness is
paramount, since a springback action is required to push back and reload the active
SMA elements, once they have completed their deployment action. These variable
geometry chevrons (VGCs) can be operated in two different ways. In the powered
mode, the tip of each chevron is actively modulated by the heaters connected to each
one of the SMA elements, in order to optimize noise reduction, without affecting too
much the propulsion efficiency of the engine (Mabe et al. 2007). Alternatively, the
profile modifications of the chevrons are induced autonomously (autonomous,
unpowered mode) by the temperature changes induced by the take-off and altitude
cruise flight conditions. Specific NiTi alloy compositions, also with ternary addi-
tions, were investigated and characterized to optimize the part performances, over
the complete temperature range (Hart et al. 2010; Noebe et al. 2009; Van Humbeeck
2012). Noise reduction is attained, through the chevrons, by promoting an optimal
mixing of the free gas stream with the fractions coming from the engine fan and from
the hot engine core, i.e., combustor and turbine (Mabe et al. 2005; Saiyed et al.
2000). The VGCs have been successfully tested for the first time on a Boeing
777 aircraft using GE-115B engines, showing a significant noise reduction under
different flight conditions, in association with the expected optimization of the
engine performances as concerns fuel consumption. These results were very much
appreciated, also in view of the growing concern on the impact of the noise from
aircraft jet engines and relevant legislation (Brueckner and Girvin 2008). The VGC
program has demonstrated the potentials of SMAs to support the development of
aircraft structures and systems capable of self-adaptation and optimization to all
flight conditions (Calkins et al. 2008). These aspects are fundamental for morphing
(from metamorphose), i.e., changing the shape of aero-structures (Barbarino et al.
2011). In fact, conventional morphing technologies usually involve complex sys-
tems for actuating the relevant shape changes. Hydraulic pumps, compressors, and
536 9 Other Interesting Alloys for Aerospace and Related Applications

electric motors operating mechanical actuators are not only complex but also rather
expensive and heavy. Moreover, the route to a complete acceptance of these
solutions, notwithstanding the undisputable benefits they bring, is rather complex,
and delays are regularly faced until a full certification is issued. SMAs but also other
functional materials, like piezoelectrics, electro- and magnetostrictive materials, and
electroactive polymers, may enable unique morphing capabilities (Loewy 1997). In
fact, different from traditional approaches, “smart” components do not require
sensors and servomechanisms to be actuated, in the first place, since they are
sensitive and directly responsive to a suitable external input, like a temperature
and/or stress change. Moreover, these materials are not only self-activated but,
owing to their peculiar thermomechanical properties, can directly perform the
required actuation (Calkins and Habe 2010). In this regard, SMAs may represent a
major breakthrough in the field of morphing of aerospace structures, as demonstrated
by several research studies. The reconfigurable rotor blade (RRB) program was
aimed at improving the performances of a rotorcraft, a particularly sensitive aspect
for VSTOL (vertical and/or short take-off and landing) aircrafts. The attempted
approach was to develop an SMA actuator, embedded in the blades, capable of
changing their twist angle during cruise and hover flight conditions (Caldwell et al.
2007; Jacot et al. 2002; Mabe et al. 2006; Prahlad and Chopra 2001; Ruggieri et al.
2008). A prototype was wind tunnel-tested and provided extremely positive results,
with no recorded losses in performances all through the duration of the test that, for
its extreme conditions, is now regarded as a new benchmark for SMA actuators and
active rotor blades (Calkins and Habe 2010). An outcome of the RRB program was
the development of a deployable rotor blade aerodynamic device to reduce the noise
due to the blade-vortex interaction (BVI). This is quite a well-known issue, and
whereas noise reduction is strictly required, a successful solution is based on
applying small tabs on the rotor blades that reduce the noise associated with the
BVI. The counterpart of this solution is the efficiency reduction due to the drag
introduced by the tabs. The improvement, whose feasibility has been demonstrated
by wind tunnel tests, is the use of deployable tabs, operated by SMA actuators,
which can fit inside the blades. The tabs are extended when noise reduction and quiet
operations are desirable or, like in some urban areas, mandatory. In normal condi-
tions, they can be retracted and stowed within the blade profile, thus not influencing
negatively the aerodynamics (Calkins et al. 2006; Kennedy et al. 2000; Singh and
Chopra 2002). Still concerning rotorcraft morphing to reduce noise emission, SMA
may provide interesting solutions for its control, also acting on the blade tips (Loewy
1997; Testa et al. 2005). This latter study is based on a theoretical model and
numerical investigation. The results show the beneficial effect of a SMA actuator,
used to maximize the separation between the tip vortex and the rotor plane, a
condition to reduce the noise produced by VBI. The effect can be enhanced if
used in association with a magneto-rheological fluid that selectively reduces the
bending stiffness spanwise in order to increase the actuation efficiency. Noise
reduction, and also an improvement in the specific fuel consumption, can be
obtained by varying the area of the fan entrance nozzle. A larger diameter at take-
off and approach reduces the jet velocity and thereby the noise produced by the
References 537

engine. Variable area fan nozzles (VAFNs) employ SMA flexure actuators. Test on
full-scale prototypes has demonstrated also for this application the successful design
that can be implemented thanks to SMAs (Mabe et al. 2008).
The perspectives of using SMA actuators and parts in adaptive aircraft compo-
nents appear promising and, in some cases, the only feasible way for implementing
novel solutions. Design and modeling studies have demonstrated already the poten-
tial of SMA in wing morphing (Icardi and Ferrero 2009; Sofla et al. 2010). Consid-
ering the increasing interest for more efficient, less polluting, and noisy aircrafts, it is
likely that SMA-based applications for which the feasibility has been assessed
already will be certified soon and applied both on the existing fleet, whenever
possible in replacement of conventional components, and on the newly designed
aircrafts (Hartl and Lagoudas 2007).

References

Akinc M et al (1999) Boron-doped Molybdenum Silicides for Structural Applications. Materials

Science and Engineering A 261 (1–2): 16–23
Alven D A (2004) Refractory and Precious Metal-Based Superalloys. JOM 56 (9): 27
Aoki K, Izumi O (1979) Improvement in Room Temperature Ductility of the L12 Type Intermetallic
Compound Ni3Al by Boron addition. Journal of the Japan Institute of Metals 43 (12): 119–1196
Auricchio F et al (2003) Modelling of SMA Materials: Training and Two Way Memory Effects.
Computers and Structures 81: 2301–2317
Azim M A et al (2017) Characterization of Oxidation Kinetics of Mo-Si-B Based Materials.
Oxidation of Metals 87 (1–2): 89–108
Baker I et al (1998) The Room Temperature Strengthening Effect of Boron as a Function of
Aluminum Concentration in FeAl. Intermetallics 6 (3): 177–183
Barbarino S et al (2011) A Review of Morphing Aircraft. Journal of Intelligent Material Systems
and Structures 22: 823–877
Bei H, George E P (2005) Microstructure and Mechanical Properties of a Directionally Solidified
NiAl-Mo Eutectic Alloy. Acta Materialia 53 (1): 69–77
Benjamin J S (1990) Mechanical Alloying – A Perspective. Metal Powder Report 45 (2): 122–127
Benjamin J S, Bomford MJ (1977) Dispersion Strengthened Aluminum Made by Mechanical
Alloying. Metallurgical Transactions A 8 (8): 1301–1305
Bewlay B P et al (1995) Solidification Processing of High Temperature Intermetallic Eutectic-
Based Alloys. Materials Science and Engineering A 192: 534–543
Bewlay B P et al (2003a) A Review of Very-High-Temperature Nb-Silicide-Based Composites.
Metallurgical and Materials Transactions A 34 (10): 2043–2052
Bewlay B P et al (2003b) Ultrahigh-Temperature Nb-Silicide-Based Composites. MRS Bulletin
28 (09): 646–653
Bochenek K, Batista M (2015) Advances in Processing of NiAl Intermetallic Alloys and Compos-
ites for High Temperature Aerospace Applications. Progress in Aerospace Sciences 79:
136–146.
Bokaie M D et al (1998) Release Device for Retaining Pins. US Patent 5,771,742
Briant C L (2000) New Applications for Refractory Metals. JOM 51 (3): 36
Brosse J B et al (1981) Intrinsic Intergranular Brittleness of Molybdenum. Scripta Metallurgica
15 (6): 619–623
Brueckner J, Girvin R (2008) Airport Noise Regulation, Airline Service Quality, an Social Welfare.
Transport Research Part B: Methodological 42 (1): 19–37
538 9 Other Interesting Alloys for Aerospace and Related Applications

Buckley J D et al (1981) Early Development of Ceramic Fiber Insulation for Space Shuttle. Ceramic
Bulletin 60: 1196–1200
Buckman R W (2000) New Applications for Tantalum and Tantalum Alloys. JOM 52 (3): 40–41
Buehler W J et al (1963) Effect of Low-Temperature Phase Changes on the Mechanical Properties
of Alloys near Composition TiNi. Journal of Applied Physics 34 (5): 1475–1477
Burk S et al (2010) Effect of Zr Addition on the High-Temperature Oxidation Behaviour of Mo-Si-
B Alloys. Oxidation of Metals 73 (1): 163–181
Byun T S et al (2013) Irradiation Dose Temperature Dependence of Fracture Toughness in High
Dose HT9 Steel from the Fuel Duct of FFTF. Journal of Nuclear Materials 432 (1–3): 1–8
Caldwell N et al (2007) Heat Transfer Model for Blade Twist Actuator System. Journal of
Thermophysics and Heat Transfer 21 (2): 350–360
Calkins F T et al (2006) Variable Geometry Chevrons for Jet Noise Reduction. Paper presented at
the 12th AIAA/CEAS Aeroacoustics Conference (27th AIAA Aeroacoustics Conference), Cam-
bridge, Massachusetts, 8–10 May 2006
Calkins F T et al (2008) Overview of Boeing’s Shape Memory Alloy Based Morphing
Aerostructures. Paper presented at the ASME 2008 Conference on Smart Materials, Adaptive
Structures and Intelligent Systems, Ellicott City, Maryland, USA, 28–30 October
Calkins F T, Habe J H (2010) Shape Memory Alloy Based Morphing Aerostructures. Journal of
Mechanical Design 132 (11): 111012
Cantor B et al (2001) Aerospace Materials. CRC Press
Cao H et al (1994) Mechanical Properties of an In Situ Synthesized Nb/Nb3Al Layered Composite.
Materials Science and Engineering A 185 (1–2): 87–95
Cao H et al (2013) Preparation of Dispersion Strengthened Aluminum Alloy by High Energy Ball
Milling. Advanced Materials Research 602–604: 598–601
Carpenter B, Lyons J (2002) Lightweight Flexible Solar Array Validation Report. Available via
DIALOG. https://eo1.gsfc.nasa.gov/new/validationReport/Technology/Documents/Reports/
LFSA.pdf. Accessed 17 Sept 2018
Casati R et al (2014) Thermal Cycling of Stress-Induced Martensite for High-Performance Shape
Memory Effect. Scripta Materialia 80 (13–16)
Chen J, Hoffelner W (2009) Irradiation Creep of Oxide Dispersion Strengthened (ODS) Steels for
Advanced Nuclear Applications. Journal of Nuclear Materials 392 (2): 360–363
Chen Y (2013) Irradiation Effects of HT-9 Martensitic Steel. Nuclear Engineering and Technology
45 (3): 311–322
Chen Y X et al (2000) Microstructure and Phase Stability Studies on Heusler Phase Ni2AlHf and
G-Phase Ni16Hf6Si7 in Directionally Solidified NiAl-Cr(Mo) Eutectic Alloyed with Hf. Journal
of Materials Research 15 (6): 1261–1270
Cockeram B V (2006) The Mechanical Properties and Fracture Mechanisms of Wrought Low
Carbon Arc Cast (LCAC), Molybdenum-0.5pct Titanium-0.1pct Zirconium (TZM), and Oxide
Dispersion Strengthened (ODS) Molybdenum Flat Products. Materials Science and Engineering
A 418 (1–2): 120–136
Cockeram B V (2009) The Fracture Toughness and Toughening Mechanism of Commercially
Available Unalloyed Molybdenum and Oxide Dispersion Strengthened Molybdenum with an
Equiaxed, Large Grain Structure. Metallurgical and Materials Transactions A 40: 2843–2860
Cornish L A et al (2006) New Pt-based Alloys for High Temperature Application in Aggressive
Environments: The Next Stage. International Platinum Conference “Platinum Surges Ahead”,
The South African Institute of Mining and Metallurgy: 57–66
Cui C Y et al (2005) High Tensile Elongation of a Directionally Solidified NiAl Multiphase Alloy at
High Temperatures. Materials Science and Engineering A 396 (1–2): 194–201
Darolia R (1991) NiAl Alloys for High-Temperature Structural Applications. JOM 43 (3): 44–49
Dimiduk D M, Perepezko J H (2003) Mo-Si-B Alloys: Developing a Revolutionary Turbine-Engine
Material. MRS Bulletin 28 (9): 639–645
Donachie M J, Donachie S J (2002) Superalloys – A Technical Guide. ASM International
Duerig T W et al (1990) Engineering Aspects of Shape Memory Alloys. Butterworth-Heinemann
References 539

El-Genk M S (2009) Deployment History and Design Considerations for Space Reactor Power
Systems. Acta Astronautica 64 (9–10): 833.849
El-Genk M S, Tournier J M (2005) A Review of Refractory Metal Alloys and Mechanically
Alloyed-Oxide Dispersion Strengthened Steels for Space Nuclear Power Systems. Journal of
Nuclear Materials 340 (1): 93–112
Elzey D M, Artz E (1988) Oxide Dispersion Strengthened Superalloys: The Role of Grain Structure
and Dispersion during High Temperature Low Cycle Fatigue. In: Reichman S et al
(ed) Superalloys, The Metallurgical Society, p 595–604
Fairbank G B et al (2000) Ultra-High Temperature Intermetallics for The Third Millennium.
Intermetallics 8 (9–11): 1091–1100
Fischer B et al (1999) High Temperature Mechanical Properties of the Platinum Group Metals.
Platinum Metals Review 43 (l): 18–28
Froes F H (1990) The Structural Applications of Mechanical Alloying. In: Froes F H, deBarbdillo J
J (eds) Proceedings of an ASM International Conference, Myrtle Beach, South Carolina,
27–29 March 1990. ASM International, Materials Park, Ohio
Gao M C et al (2008) The First-Principles Design of Ductile Refractory Alloys. JOM 60 (7): 61–65
Gilman P S, Nix W D (1981) The Structure and Properties of Aluminum Alloys Produced by
Mechanical Alloying: Powder Processing and Resultant Powder Structures. Metallurgical and
Materials Transactions A 12 (5): 813–824
Gu Y F et al (2004) Chromium and Chromium-Based Alloys: Problems and possibilities for High
Temperature Service. JOM 56 (9): 28–33
Gu Y F et al (2005) Microstructural Evolution and Mechanical Properties of Cr-Ru Alloys.
Metallurgical and Materials Transactions A 36 (3): 577–582
H C Starck Fabricated Products (2013) Defense and Aerospace Applications of Molybdenum,
Tungsten and Tantalum Products. Available via DIALOG. https://www.azom.com/article.aspx?
ArticleID¼9073. Accessed 15 Nov 2018
Harada H (2003) High Temperature Materials for Gas Turbines. The Present and Future. In:
Proceedings of the International Gas Turbine Congress, Tokyo, 2–7 November 2003
Harper M A, Rapp R A (1994) Codeposited Chromium and Silicon Diffusion Coatings for Fe-Base
Alloys via Pack Cementation. Oxidation of Metals 42 (3–4): 303–333
Hart D J et al (2010) Jet Engine Chevron Application: II Experimentally Validated Numerical
Analysis. Smart Materials and Structures 19 (1): 015021
Hart D J et al (2010) Use of a Ni60Ti Shape Memory Alloy for Active Jet Engine Chevron
Application: I Thermomechanical Characterization. Smart Materials and Structures 19 (1):
015020
Hartl D J, Lagoudas D C (2007) Aerospace Applications of Shape Memory Alloys. Proceedings of
the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering 221, p
535–552
Heilmeier M et al (2009) Metallic Materials for Structural Applications Beyond Nickel-based
Superalloys. JOM 61 (7): 61–67
Hoffelner W (2011) Design Related Aspects in Advanced Nuclear Fission Plants. Journal of
Nuclear Materials 409 (2): 112–116
Howson T E et al (1980) Creep and Stress Rupture of Oxide Dispersion Strengthened Mechanically
Alloyed Inconel Alloy MA754. Metallurgical Transactions A 11 (9): 1599–10607
Icardi U, Ferrero L (2009) Preliminary Study of an Adaptive Wing with Shape Memory Alloy
Torsion Actuators. Materials & Design 30 (10): 4200–4210
Jacot A D et al (2002) Shape Memory Alloy Device and Control Method. US Patent 6,499,952 B1
Jardine A P et al (1997) Smart Wing Shape Memory Alloy Actuator Design and Performance. In:
Goldstein D H, Chipman R A (eds) Proceedings of SPIE – The International Society for Optical
Engineering. Polarization: Measurement, Analysis, and Remote Sensing, San Diego, CA,
United States, 30 July – 1 August 1997
Jehanno P et al (2005) Assessment of a Powder Metallurgical Processing Route for Refractory
Metal Silicide Alloys. Metallurgical and Materials Transactions A 36 (3): 515–523
540 9 Other Interesting Alloys for Aerospace and Related Applications

Jenkins P P, Landis G A (1995) A Rotating Arm using Shape Memory Alloy. In NASA (ed) The
29th Aerospace Mechanisms Symposium, p 167–171
Jha SC et al (1989) Dispersoid in Rapidly Solidified B2 Nickel Aluminides. Scripta Metallurgica
23: 805–810
Jiao Z B et al (2016) Strategies for Improving Ductility of Ordered Intermetallics. Progress in
Natural Science: Materials International 26 (1): 1–12
Johnson A D (1992) Non-Explosive Separation Device. US Patent 5,119,555
Jones E S et al (1958) The Oxidation of Molybdenum. Corrosion 14 (1): 20–26
Kaplanskii Y Y (2018) Microstructure and Thermomechanical Behavior of Heusler Phase Ni2AlHf-
strengthened NiAl-Cr(Co) Alloy produced by HIP of Plasma-spheroidized Powder. Materials
Science and Engineering A 729: 398–410
Kaufman G, Mayo I (1997) The Story of Nitinol: The Serendipitous Discovery of the Memory
Metal and its Applications. The Chemical Educator 2 (2): 1–21
Kennedy D K et al (2000) Development of an SMA Actuator for In-Flight Rotor Blade Tracking.
Journal of Intelligent Material Systems and Structures 15 (4): 235–248
King D et al (2009) Advanced Aerospace Materials: Past, Present and Future. Aviation and the
Environment 3: 22–27
Klueh R L (1993) Tensile Behavior of Neutron-Irradiated Martensitic Steels: A Review. Nuclear
Technology 102 (3): 376–385
Klueh RL, Nelson A T (2007) Ferritic/martensitic steels for next generation reactors. Journal of
Nuclear Materials 371:37–52
Knittel S et al (2013) Development of Silicide Coatings to Ensure the Protection of Nb and Silicide
Composites against High Temperature Oxidation. Surface and Coatings Technology 235:
401–406
Koch C C, Whittenberger J D (1996) Mechanical Milling/alloying of intermetallics. Intermetallics
4 (5): 339–355
Korb L J et al (1981) The Shuttle Orbiter Thermal Protection System. Ceramic Bulletin 60:
1188–1193
Kudva J (2004) Overview of the DARPA Smart Wing Project. Journal of Intelligent Material
Systems and Structures 15: 261–267
Kung S C, Rapp R A (1989) Analyses of the Gaseous Species in Halide-Activated Cementation
Coating Packs. Oxidation of Metals 32 (1–2): 89–109
Lasalmonie A (2006) Intermetallics: Why is it so Difficult to Introduce Them in Gas Turbine
Engines? Intermetallics 14 (10–11): 1123–1129
Lassner E, Schubert W D (1999) Tungsten – Properties, Chemistry, Technology of the Element,
Alloys, and Chemical Compounds. Kluwer Academic/Plenum Publisher, New York
Lewis P (1997) Aircraft Industry Accepts Shape Memory Alloy Technology. Aircraft Engineering
and Aerospace Technology 69 (1): 31–34
Liebscher C H, Glatzel U (2014) Configuration of Superdislocations in the γ’-Pt3Al Phase of a
Pt-based Superalloy. Intermetallics 48: 71–78
Lipetzky P (2002) Refractory Metals: A Primer. JOM 54 (3): 47–49
Liu C T et al (1985) Effect of Boron on Grain-Boundaries in Ni3Al. Acta Metallurgica 33 (2):
213–229
Liu C T, Kumar K S (1993) Ordered Intermetallic Alloys, Part I: Nickel and Iron Aluminides. JOM
45 (3): 38–44
Liu G et al (2013) Nanostructured High-Strength Molybdenum Alloys with Unprecedented Tensile
Ductility. Nature Materials 12 (4): 344–350
Locci I E et al (1996) Microstructure and Phase Stability of Single Crystal NiAl alloyed with Hf and
Zr. Journal of Materials Research 11 (12): 3024–3038
Loewy R G (1997) Recent Developments in Smart Structures with Aeronautical Applications.
Smart Materials and Structures 6 (5): R11–R42
Lu Z L et al (2013) Fabricating Hollow Turbine Blades using Short Carbon Fiber-reinforced SiC
Composite. International Journal of Advanced Manufacturing Technology 69 (1–4): 417–425
References 541

Mabe J H et al (2005) Design and Control of a Morphing Chevron for Takeoff and Cruise Noise
Reduction. In: Proceedings of the 26th Annual AIAA Aeroacoustics Conference, Monterey, CA,
23–25 May 2005
Mabe J H et al (2006) Boeing’s Variable Geometry Chevron, Morphing Aerostructure for Jet Noise
Reduction. Paper presented at the 47th AIAA/ASME/AHS/ASC Structures, Structural Dynam-
ics, and Materials Conference, Newport, Rhode Island, 1–4 May 2006
Mabe J H et al (2007) Full-Scale Flight Tests of Aircraft Morphing Structures using SMA
Actuators. In: Proceedings Volume 6525, Active and Passive Smart Structures and Integrated
Systems 2007. SPIE Smart Structures and Materials + Nondestructive Evaluation and Health
Monitoring, San Diego, CA, United States, 2007
Mabe J H et al (2008) Variable Area Jet Nozzle Using Shape Memory Alloy Actuators in an
Antagonistic Design. In: Proceedings Volume 6930, Industrial and Commercial Applications of
Smart Structures Technologies 2008. SPIE Smart Structures and Materials + Nondestructive
Evaluation and Health Monitoring, San Diego, CA, United States, 2008
Majumdar S et al (2009) Densification and Grain Growth during Isothermal Sintering of Mo and
Mechanically Alloyed Mo-TZM. Acta Materialia 57 (14): 415–4168
Mansur L F et al (2004) Materials Needs for Fusion, Generation-IV Fission Reactors and Spallation
Neutron Sources – Similarities and Differences. Journal of Nuclear Materials 329–333 Part A:
166–172
Mason D P, Van Atkin D C (1993) The Effect of Microstructural Scale on Hardness of MoSi2-
Mo5Si3 Eutectics. Scripta Metallurgica et Materialia 28 (2): 185–189
McKamey C G et al (1991) A Review of Recent Developments in Fe3Al-based Alloys. Journal of
Materials Research 6 (8): 1779–1805
McLean M (1983) Directionally Solidified Materials for High Temperature Service. The Metal
Society, London, UK
Medvedeva N I et al (2007) Solid Solution Softening and Hardening in the Group-V and Group-IV
bcc Transition Metals Alloys: First Principles Calculations and Atomistic Modeling. Physical
Review B 76 (21): 212104
Melton K N (1999) General Applications of Shape Memory Alloys and Smart Materials. In:
Otsuka K, Wayman C M (eds) Shape Memory Materials. Cambridge University Press, p
220–239
Mendiratta M G, Dimiduk D M (1993) Strength and Toughness of a Nb/Nb5Si3 Composite.
Metallurgical and Materials Transactions A 24 (2): 501–504
Miller M K et al (2003) Improvement in the Ductility of Molybdenum Alloys due to Grain
Boundary Segregation. Scripta Metallurgica 46 (4): 299–303
Miller M K et al (2005) Stability of Ferritic MA/ODS Alloys at High Temperatures. Intermetallics
13 (3–4): 387–392
Miller M K et al (2006) Characterization of Precipitates in MA/ODS Ferritic Alloys. Journal of
Nuclear Materials 351 (1–3): 261–268
Miracle D B (1993) The Physical and Mechanical Properties of NiAl. Acta Metallurgica et
Materialia 41 (39): 649–684
Mitra R (2006) Mechanical Behaviour and Oxidation Resistance of Structural Silicides. Interna-
tional Materials Reviews 51 (1): 13–64
Mitra R (2018) Intermetallic Matrix Composites. Woodhead Publishing, Duxford, UK
Morris D, Gunther S (1997) Room and High Temperature Mechanical Behavior of a Fe3Al-based
Alloy with α-α” Microstructure. Acta Materialia 45 (2): 811–822
Mueller A J et al (2000) Evaluation of Oxide Dispersion Strengthened (ODS) Molybdenum and
Molybdenum-Rhenium Alloys. International Journal of Refractory Metals and Hard Materials
18 (4–5): 205–211
Muktinutalapati N R (2011) Materials for Gas Turbines – An Overview. In: Benini E (ed) Advances
in Gas Turbine Technology, IntecOpen, p 293–314
Mutry K L, Charit I (2008) Structural Materials for Gen-IV Nuclear Reactors: Challenges and
Opportunities. Journal of Nuclear Materials 383 (1–2): 189–195
542 9 Other Interesting Alloys for Aerospace and Related Applications

Noebe R D et al (2009) High Work Output Ni-Ti-Pt High Temperature Shape Memory Alloys and
associated Processing Methods.
Ölander A (1932) An Electrochemical Investigation of Solid Cadmium-Gold Alloys. Journal of the
American Chemical Society 54 (10): 3819–3833
Parthasarathy T A et al (2002) Oxidation Mechanisms in Mo-reinforced Mo5SiB2(T2)– Mo3Si
Alloys. Acta Materialia 50 (7): 1857–1868
Patra A et al (2015) Tungsten Based ODS Alloys-A Comprehensive Survey. Lap Lambert Aca-
demic Publishing
Perepezko J H (2009) The Hotter the Engine, the Better. Science 356 (5956): 1068–1069
Perepezko J H et al (2014) Structural Intermetallics, Alloy Design, Processing, and Applications.
Advanced Materials and Processes 172 (9): 22–26
Petrovic J J, Vasudevan A K (1999) Key Development in High Temperature Structural Silicides.
Materials Science and Engineering A 261: 1–5
Pettifor D G (1988a) Structure Maps for Pseudobinary and Ternary Phases. Materials Science and
Technology 4 (8): 675–691
Pettifor D G (1988b) Theoretical Predictions of Structure and Related Properties of Intermetallics.
Materials Science and Technology 8 (4): 345–349
Pickens J R (1991) High-Strength Aluminum P/M Alloys. In: ASM International (ed) ASM
Handbook Volume 2 – Properties and Selection: Nonferrous Alloys and Special-Purpose
Materials. ASM International
Pitt D M et al (2001) SAMPSON Smart Inlet SMA Powered Adaptive Lip Design and Static Test.
In: 19th AIAA Applied Aerodynamics Conference, Fluid Dynamics and Co-located Confer-
ences, Anaheim, CA, USA
Pitt D M et al (2002) SAMPSON Smart Inlet Design Overview and Wind Tunnel Test: II. In:
Proceedings Volume 4698, Smart Structures and Materials 2002: Industrial and Commercial
Applications of Smart Structures Technologies. SPIE’s 9th Annual International Symposium on
Smart Structures and Materials, San Diego, CA, United States, 2002
Pitt D M et al (2002) SAMPSON Smart Inlet Design Overview and Wind Tunnel Test: I. In:
Proceedings Volume 4698, Smart Structures and Materials 2002: Industrial and Commercial
Applications of Smart Structures Technologies. SPIE’s 9th Annual International Symposium on
Smart Structures and Materials, San Diego, CA, United States, 2002
Polvani R S et al (1976) High Temperature Creep in a Semi-Coherent NiAl-Ni2AITi Alloy.
Metallurgical Transactions A 7: 33–40
Portebois L et al (2014) Effect of Boron Addition on the Oxidation Resistance of Silicide Protective
Coatings: A Focus on Boron Location in as-coated and oxidized coated Niobium Alloys.
Surface and Coatings Technology 253: 292–299
Potgieter J H, Maledi N B (2014) High Temperature Corrosion Resistance of Pt-Based Superalloys
in 0.2% SO2-N2 Gas. The Open Materials Science Journal 8: 18–26
Prahlad H, Chopra I (2001) Design of a Variable Twist Tilt-rotor Blade Using Shape Memory Alloy
(SMA) Actuators. In: Proceedings Volume 4327, Smart Structures and Materials 2001: Smart
Structures and Integrated Systems. SPIE’s 8th Annual International Symposium on Smart
Structures and Materials. Newport Beach, CA, United States, p 46–59
Ray R et al (1989) Carbide-dispersion-strengthened B2 NiAl. Materials Science and Engineering A
119: 103–111
Reese R T, Vick C P (1983) Soviet Nuclear Powered Satellites. British Interplanetary Society
Journal 36: 457–462
Reuss S, Vehoff H (1990) Temperature Dependence of the Fracture Toughness of Single Phase and
Two Phase Intermetallics. Scripta Metallurigica et Materialia 24 (6): 1021–1026
Reviere R D et al (1992) Processing Microstructure and Low-Temperature Properties of
Directionally Solidified NiAl/NiAlNb Alloys. Materials Letters 14 (2–3) 149–155
Richerson D W (2004) Ceramic Components for Gas Turbine Engines: Why Has It Taken So Long?
Ceramics Engineering and Science Proceedings 25 (3): 3–32
References 543

Ruggieri R T et al (2008). Development of a 1/4-Scale NiTinol Actuator for Reconfigurable

Structures. In: Proceedings Volume 6930, Industrial and Commercial Applications of Smart
Structures Technologies 2008. SPIE Smart Structures and Materials + Nondestructive Evalua-
tion and Health Monitoring, San Diego, CA, United States, 2008
Saiyed N et al (2000) Acoustics and Thrust of Separate-Flow Exhaust Nozzles with Mixing Devices
for High-Bypass-Ratio Engines. Available via DIALOG. https://ntrs.nasa.gov/archive/nasa/
casi.ntrs.nasa.gov/20000083968.pdf. Accessed 24 Feb 2019
Sakidja R et al (2005) Synthesis of Oxidation Resistant Silicide Coatings on Mo-Si-B Alloys.
Scripta Materialia 53 (6) 723–728
Sankar M et al (2015) Microstructure, Oxidation Resistance and Tensile Properties of Silicide
Coated Nb-Alloy C-103. Materials Science and Engineering A 645 (1): 339–346
Saunders S R et al (1997) Behaviour of Fecralloy and Iron Aluminides Alloys in Coal Gasification
Atmospheres Containing HCl. Materials Science Forum 251–254: 583–590
Scheppe F et al (2002) Nickel Aluminides: A Step Toward Industrial Applications. Materials
Science and Engineering A 329–331: 596–601
Schmidt F F, Ogden H R (1963) The Engineering Properties of Tantalum and Tantalum Alloys.
Available via DIALOG. https://apps.dtic.mil/dtic/tr/fulltext/u2/426344.pdf. Accessed 8 Nov
2018
Schoijet M, Girifalco L A (1967) Theory of Diffusion in Ordered Alloys of the β-Brass Type. Solid
State Communications 29 (3): 481–495
Schoijet M, Girifalco L A (1968) Diffusion in order-disorder Alloys. The Face Centered Cubic AB3
Alloy. Journal of Physics and Chemistry of Solids 29 (6): 911–922
Sikka V K et al (2000) Advances in Processing of Ni3Al-based Intermetallics and Applications.
Intermetallics 8 (9–11): 1329–1337
Singh K, Chopra I (2002) Design of an Improved Shape Memory Alloy Actuator for Rotor Blade
Tracking. In: Proceedings Volume 4701, Smart Structures and Materials 2002: Smart Structures
and Integrated Systems. SPIE’s 9th Annual International Symposium on Smart Structures and
Materials, San Diego, CA, United States, 2002
Sofla A Y N et al (2010) Shape Morphing of Aircraft Wing: Status and Challenges. Materials &
Design 31 (4): 1284–1292
Soleimani-Dorcheh A et al (2014) On Ultra-High Temperature Oxidation of Cr-Cr3Si Alloys: Effect
of Germanium. Materials and Corrosion 65 (12): 1143–1150
Stoloff N S (2000) Emerging Applications of Intermetallics. Intermetallics 8 (9–11): 1313–1320
Stone H W (1996) Mars Pathfinder Microrover A Small, Low-Cost, Low-Power Spacecraft. In:
Proceedings of the 1996 AIAA Forum on Advanced Developments in Space Robotics.
Strelec J K et al (2003) Design and Implementation of a Shape Memory Alloy Actuated
Reconfigurable Wing. Journal of Intelligent Material Systems and Structures 14 (4–5): 257–273
Strutt P R et al (1976) Creep Behavior of the Heusler Type Structure Alloy NiAl2Ti. Metallurgical
Transactions A 7: 23–31
Sturm D et al (2007) The Influence of Silicon on the Strength and Fracture Toughness of
Molybdenum. Materials Science and Engineering A 463 (1–2): 107–114
Testa C et al (2005) Feasibility Study on Rotorcraft Blade Morphing in Hovering. Proceedings of
SPIE – The International Society for Optical Engineering 5764: 171–182
Trinkle D R, Woodward C (2005) The Chemistry of Deformation: How Solutes Soften Pure Metals.
Science 310 (5754): 1665–1667
United Nations General Assembly (1992) Principles Relevant to the Use of Nuclear Power Sources
in Outer Space – Resolution 47/68. Available via DIALOG. https://www.un.org/ga/documents/
gares47/list47.htm. Accessed 27 Sept 2016
Van Humbeeck J (2012) Shape Memory Alloys with High Transformation Temperatures. Materials
Research Bulletin 47 (10): 2966–2968
Vilasi M et al (1993) Crystal Structure of Tri-Niobium Tri-Iron Chromium Hexasilicide
Nb¼3Fe¼3Cr¼lSi6: an Intergrowth of Zr4Co4Ge7 and Nb2Cr4Si5 Blocks. Journal of Alloys
and Compounds 194 (1): 127–132
544 9 Other Interesting Alloys for Aerospace and Related Applications

Vilasi M et al (1998a) New Silicides for New Niobium Protective Coatings. Journal of Alloys and
Compounds 264 (1–2): 244–251
Vilasi M et al (1998b) Phase Equilibria in the Nb-Fe-Cr-Si System. Journal of Alloys and
Compounds 269 (1–2): 187–192
Vorberg S et al (2004) Pt-Al-Cr-Ni Superalloys: Heat Treatment and Microstructure. JOM 56 (9):
40–43
Vorberg S et al (2005) A TEM Investigation of the γ-γ’ Phase Boundary in Pt-based Superalloys.
JOM 57 (3): 49–51
Wadsworth J et al (1988) Recent Advances in Aerospace Refractory-Metal Alloys. International
Materials Reviews 33 (1): 131–150
Wang C C, Akbar S A (1993) Diffusion in Ordered Alloys and Intermetallic Compounds. Acta
Metallurgica et Materialia 41 (10): 2807–2813
Wang L et al (2016) Microstructure and Mechanical Properties of NiAl-based Hypereutectic Alloy
obtained by Liquid Metal Cooling and Zone Melted Liquid Metal Cooling Directional Solid-
ification Techniques. Journal of Materials Research 31 (5): 646–654
Wang L et al (2017) Microstructure Evolution and Room Temperature Fracture Toughness of
as-cast and Directionally Solidified Novel NiAl-Cr(Fe) Alloy. Intermetallics 84: 11–19
Ward-Close C M et al (1996) Intermetallic-Matrix Composites – A Review. Intermetallics 4 (3):
217–229
Wenderoth M et al (2005) On the Development and Investigation of Quaternary Pt-Based Super-
alloys with Ni Additions. Metallurgical and Materials Transactions A 36 (3): 567–575
Whittenberger J D (1999) 1300 K Creep Behavior of [001] oriented Ni-49Al-1Hf (at%) Single
Crystals. Materials Science and Engineering A 286 (1): 165–183
Whittenberger J D et al (1990a) 1300 K Compressive Properties of Several Dispersion Strengthened
NiAl Materials. Journal of Materials Science 25: 2771–2776
Whittenberger J D et al (1990b) Preliminary Investigation of a NiAl Composite prepared by
Cryomilling. Journal of Materials Research 5 (2): 271–277
Whittenberger J D et al (1992a) Compressive Strength of Directionally Solidified NiAl-NiAlNb
Intermetallics at 1200 and 1300 K. Scripta Metallurgica et Materialia 26 (6): 987–992
Whittenberger J D et al (1992b) Influence of Grain Size on the Creep Behavior of HfC-dispersed
NiAl. Materials Science and Engineering A 151: 137–146
Whittenberger J D et al (1999) Microstructure and 1000-1400 K Mechanical Properties of
Cryomilled NiAl-0.7Zr. Journal of Materials Research 14 (6): 2418–2429
Whittenberger J D et al (2000) Elevated Temperature Compressive Strength Properties of Oxide
Dispersion Strengthened NiAl after Cryomilling and Roasting in Nitrogen. Materials Science
and Engineering A 291 (1–2): 173–185
Wiedemann C et al (2005). Size Distribution of NaK Droplets released during RORSAT Reactor
Core Ejection. Advances in Space Research 35 (7): 1290–1295
Witkin D B, Lavernia E J (2006) Synthesis and Mechanical Behavior of Nanostructured Materials
via Cryomilling. Progress in Materials Science 51 (1): 1–60
Wojcik C C (1991) High Temperature Niobium Alloys. In: Stephens J J, Ahmad I (eds) High
Temperature Niobium Alloys. The Minerals, Metals and Materials Society, Warrendale, PA,
USA, p 1–12
Yamabe-Mitarai Y et al (2004) Platinum-Group-Metals-Based Intermetallics as High-Temperature
Structural Materials. JOM 56 (9): 34–39
Yamabe-Mitarai Y, Murakami H (2014) Mechanical Properties at 2223 K and Oxidation Behavior
of Ir Alloys. Intermetallics 48: 86–92
Yang J M et al (1997) Microstructure and Mechanical Behavior of in-situ Directional Solidified
NiAl/Cr(Mo) Eutectic Composite. Acta Materialia 45 (1): 295–305
Yang R (1992) Equilibria and Microstructural Evolution in the β/β’/γ’ Region of the Ni-Al-Ti
System: Modeling and Experiment. Acta Metallurgica et Materialia 40 (7): 1553–1562
Yang R et al (1991) A Microstructural Study of a Ni2AlTi-Ni(Al,Ti)-Ni3(Al,Ti) Three-Phase Alloy.
Journal of Materials Research 6 (2): 343–354
References 545

Yang R et al (1992) Three-Phase β/β’/γ’ Ni-Al-Ti-(Cr,Fe) Alloys for High Temperature Use.
Materials Science and Engineering A 152 (1–2): 227–236
Yu J L et al (2017) Mechanical Properties and Fracture Behavior of an Nb-Silicide in situ
Composite. Intermetallics 90: 135–139
Yu K O et al (1993) Investment Casting of NiAl Single-Crystal Alloys. JOM 45 (5): 49–51
Yvon P, Carré F (2009) Structural Materials Challenges for Advanced Reactor Systems. Journal of
Nuclear Materials 385 (2): 217–222
Zhao J-C, Westbrook J H (2003) Ultrahigh-Temperature Materials for Jet Engines. MRS Bulletin
28 (9): 622–630

Further Reading

Lagoudas D C (2007) Shape Memory Alloys – Modelling and Engineering Applications. Springer
Lecce L, Concilio A (2015) Shape Memory Alloy Engineering: for Aerospace, Structural and
Biomedical Applications. Butterworth-Heinemann
Leo D J (2007) Engineering Analysis of Smart Material Systems. John Wiley & Sons Inc.
Liu C T et al (1992) Ordered Intermetallics – Physical Metallurgy and Mechanical Behaviour.
Springer Netherlands
Otsuka K, Ren X (2005) Physical Metallurgy of Ti-Ni-Based Shape Memory Alloys. Progress in
Materials Science 50 (5): 511–678
Sauthoff G (1995) Intermetallics. VCH, Weinheim
Schwartz M (2002) Encyclopedia of Smart Materials – 2 Volume Set. John Wiley & Sons
Stoloff N S, Sikka V K (1996) Physical Metallurgy and Processing of Intermetallic Compounds.
Chapman & Hall, New York
Appendices

Appendix 1: Wrought Aluminum Alloys and Relevant

Compositions

International
designation Si Fe Cu Mn Mg Cr Zn Ti Others
2001 0.20 0.20 5.2–6.0 0.15–0.50 0.20–0.45 0.10 0.10 0.20 Ni ¼ 0.05
Pb ¼ 0.003
Zr ¼ 0.05
2011 0.40 0.70 5.0–6.0 – – – 0.30 – Bi ¼ 0.20–0.60
Pb ¼ 0.20–0.60
2014 0.50–1.20 0.70 3.9–5.0 0.40–1.20 0.20–0.80 0.10 0.25 0.15 –
2017 0.20–0.80 0.70 3.5–4.5 0.40–1.00 0.40–0.80 0.10 0.25 0.15 –
2020 – – 4.5 0.50 – – – – Li ¼ 1.2
2021 0.20 0.30 5.8–6.8 0.20–0.40 0.02 – 0.10 0.02– Sn ¼ 0.03–0.08
0.10 V ¼ 0.05–0.15
Zr ¼ 0.10–0.25
Cd ¼ 0.05–0.20
2023 0.10 0.15 3.6–4.5 0.30 1.0–1.6 0.10 – 0.05 Sc ¼ 0.01–0.06
Zr ¼ 0.05–0.15
2024 0.50 0.50 3.8–4.9 0.30–0.90 1.2–1.8 0.10 0.25 0.15 –
2024A 0.15 0.20 3.7–4.5 0.15–0.80 1.2–1.5 0.10 0.25 0.15 –
2027 0.12 0.15 3.9–4.9 0.50–1.2 1.0–1.5 – 0.20 0.08 Zr ¼ 0.05–0.15
2036 0.50 0.50 2.2–3.0 0.10–0.40 0.30–0.60 0.10 0.25 0.15 –
2050 0.08 0.10 3.2–3.9 0.20–0.50 0.20–0.60 0.05 0.25 0.10 Ag ¼ 0.20–0.70
Ga ¼ 0.05
Li¼0.70–1.30
V ¼ 0.05
Zr ¼ 0.06–0.14
2055 0.07 0.10 3.2–4.2 0.10–0.50 0.20–0.60 – 0.30– 0.10 Ag ¼ 0.20–0.70
0.70 Li ¼ 1.0–1.3
Zr ¼ 0.05–0.15
2060 0.07 0.07 3.4–4.5 0.10–0.50 0.6–1.1 – 0.30– 0.10 Ag ¼ 0.05–0.50
0.50 Li ¼ 0.60–0.90
Zr ¼ 0.05–0.15

(continued)

© Springer Nature Switzerland AG 2020 547

S. Gialanella, A. Malandruccolo, Aerospace Alloys, Topics in Mining, Metallurgy
and Materials Engineering, https://doi.org/10.1007/978-3-030-24440-8
548 Appendices

International
designation Si Fe Cu Mn Mg Cr Zn Ti Others
2065 0.10 0.10 3.8–4.7 0.15–0.50 0.25–0.80 – 0.30 0.10 Ag ¼ 0.15–0.50
Li ¼ 0.8–1.5
Zr ¼ 0.05–0.15
2076 0.12 0.15 2.9–3.8 0.10–0.50 0.05–0.40 – 0.10– 0.10 Li ¼ 1.0–1.4
0.50 Zr ¼ 0.05–0.15
2090 0.10 0.12 2.4–3.0 0.05 0.25 0.05 0.10 0.15 Li ¼ 1.9–2.6
Zr ¼ 0.08–0.15
2091 0.20 0.30 1.8–2.5 0.10 1.1–1.9 0.10 0.25 0.10 Li ¼ 1.7–2.3
Zr ¼ 0.08–0.15
2094 0.12 0.15 4.4–5.2 0.25 0.25–0.80 – 0.25 0.10 Li ¼ 0.70–1.4
Zr ¼ 0.04–0.18
2097 0.12 0.15 2.5–3.1 0.10–0.60 0.35 – 0.35 0.15 Li ¼ 1.2–1.8
Zr ¼ 0.08–0.16
2098 0.12 0.15 3.2–3.8 0.35 0.25–0.80 – 0.35 0.10 Ag ¼ 0.25–0.60
Li ¼ 0.80–1.3
Zr ¼ 0.04–0.18
2099 0.05 0.07 2.4–3.0 0.10–0.50 0.10–0.50 – 0.40– 0.10 Be ¼ 0.0001
1.0 Li ¼ 1.6–2.0
Zr ¼ 0.05–0.12
2124 0.20 0.30 3.8–4.9 0.30–0.90 1.2–1.8 0.10 0.25 0.15 –
2195 – – 4.0 – 0.40 – – – Ag ¼ 0.40
Li ¼ 1.0
Zr ¼ 0.11
2196 – – 2.9 – 0.50 – – – Ag ¼ 0.40
Li ¼ 1.75
Zr ¼ 0.11
2198 – – 3.2 0.50 0.50 – 0.35 – Ag ¼ 0.40
Li ¼ 1.0
Zr ¼ 0.11
2199 0.05 0.07 2.6 0.30 0.30 – 0.55 – Li ¼ 1.8
Zr ¼ 0.09
2218 0.90 1.00 3.5–4.5 0.20 1.2–1.8 0.10 0.25 – Ni ¼ 1.7–2.3
2219 0.20 0.30 5.8–6.8 0.20–0.40 0.02 – 0.10 0.02– V ¼ 0.05–0.15
0.10 Zr ¼ 0.10–0.25
2296 – – 2.45 0.28 0.60 – 0.25 – Ag ¼ 0.43
Li ¼ 1.6
Zr ¼ 0.11
2297 – – 2.8 0.30 0.25 – 0.50 – Li ¼ 1.4
Zr ¼ 0.11
2397 – – 2.8 0.30 0.25 – 0.10 – Li ¼ 1.4
Zr ¼ 0.11
2519 0.25 0.30 5.3–6.4 0.10–0.50 0.05–0.40 – 0.10 0.02– V ¼ 0.05–0.15
0.10 Zr ¼ 0.10–0.25
2618 0.10–0.25 0.9– 1.9–2.7 – 1.3–1.8 – 0.10 0.04– Ni ¼ 0.9–1.2
1.3 0.10
3002 0.08 0.10 0.15 0.05–0.25 0.05–0.20 – 0.05 0.03 V ¼ 0.05
3021 0.50 0.70 0.20– 0.05–0.80 0.10 0.10 0.10 0.10 –
0.60
3030 0.15 0.35 0.10 0.10–0.70 0.05 0.05 0.05– 0.05– –
0.50 0.35
3102 0.40 0.70 0.10 0.05–0.40 – 0.30 0.10 –
3130 0.15 0.20 0.05 0.10–0.40 0.05 – 0.05– 0.05 –
0.30
4004 9.0–10.5 0.80 0.25 0.10 1.0–2.0 – 0.20 – –

(continued)
Appendices 549

International
designation Si Fe Cu Mn Mg Cr Zn Ti Others
4007 1.0–1.7 0.40– 0.20 0.8–1.5 0.20 0.05– 0.10 0.10 Ni ¼ 0.15–0.70
1.0 0.25 Co ¼ 0.05
4043 4.5–6.0 0.80 0.30 0.05 0.05 – 0.10 0.20 –
4047 11.0–13.0 0.80 0.30 0.15 0.10 – 0.20 – –
4104 9.0–10.5 0.80 0.25 0.10 1.0–2.0 – 0.20 – Bi ¼ 0.02–0.20
4115 1.8–2.2 0.70 0.10– 0.6–1.2 0.10–0.50 – 0.20 – –
0.50
4143 4.7–6.0 0.80 0.30 0.05 0.15–0.30 – 0.10 0.20 –
4147 11.0–13.0 0.80 0.25 0.10 0.10–0.50 – 0.20 – –
4343 6.8–8.2 0.8 0.25 0.10 – – – – Ni ¼ 0.20
4643 3.6–4.6 0.80 0.10 0.05 0.10–0.30 – 0.10 0.15 –
5005 0.30 0.70 0.20 0.20 0.50–1.1 0.10 0.25 – –
5017 0.40 0.70 0.18– 0.6–0.8 1.9–2.2 – – 0.09 –
0.28
5022 0.25 0.40 0.20– 0.20 3.5–4.9 0.10 0.25 0.10 –
0.50
5024 0.25 0.40 0.20 0.20 3.9–5.1 0.10 0.25 0.20 Sc ¼ 0.10–0.40
Zr ¼ 0.05–0.20
5025 0.25 0.25 0.10 0.20 4.5–6.0 0.20 0.25 0.05– Be ¼ 0.0009
0.20 Sc ¼ 0.05–0.55
Zr ¼ 0.10–0.25
5050 0.40 0.70 0.20 0.10 1.1–1.8 0.10 0.25 – –
5051 0.40 0.70 0.25 0.20 1.7–2.2 0.10 0.25 0.10 –
5205 0.15 0.70 0.03– 0.10 0.6–1.0 0.10 0.05 – –
0.10
5251 0.40 0.50 0.15 0.10–0.50 1.7–2.4 0.15 0.15 0.15 –
5854 0.40 0.40 0.10 0.50 2.8–3.6 0.30 0.20 0.15 Li ¼ 0.0009
Pb ¼ 0.01
Ca ¼ 0.0009
Na ¼ 0.0009
Mn+Cr ¼ 0.10–0.60
6056 0.70–1.3 0.50 0.50– 0.40–1.0 0.6–1.2 0.25 0.10– Zr+Ti ¼ 0.20
1.1 0.70
6063 0.20–0.60 0.35 0.10 0.10 0.45–0.90 0.10 0.10 0.10 –
6156Cl –
7010 0.12 0.15 1.5–2.0 0.10 2.1–2.6 0.05 5.7– 0.06 Ni ¼ 0.05
6.7 Zr ¼ 0.10–0.16
7040 0.10 0.13 1.5–2.3 0.04 1.7–2.4 0.04 5.7– 0.06 Zr ¼ 0.05–0.12
6.7
7050 0.12 0.15 2.0–2.6 0.10 1.9–2.6 0.04 5.7– 0.06 Zr ¼ 0.08–0.15
6.7
7056 0.10 0.12 1.2–1.9 0.20 1.5–2.3 – 8.5– 0.08 Zr ¼ 0.05–0.15
9.7
7075 0.40 0.50 1.2–2.0 0.30 2.1–2.9 0.18– 5.1– 0.20 Zr+Ti ¼ 0.25 when
0.28 6.1 required for extruded
or forged products
7150 0.12 0.15 1.9–2.5 0.10 2.0–2.7 0.04 5.9– 0.06 Zr ¼ 0.08–0.15
6.9
7349 0.12 0.15 1.4–2.1 0.20 1.8–2.7 0.10– 7.5– – Zr+Ti ¼ 0.25
0.22 8.7
7449 0.12 0.15 1.4–2.1 0.20 1.8–2.7 – 7.5– – Zr+Ti ¼ 0.25
8.7
7475 0.10 0.12 1.2–1.9 0.06 1.9–2.6 0.18– 5.2– 0.06 –
0.25 6.2
8090 0.20 0.30 1.0–1.6 0.10 0.6–1.3 0.10 0.25 0.10 Li ¼ 2.2–2.7
Zr ¼ 0.04–0.16

(continued)
550 Appendices

International
designation Si Fe Cu Mn Mg Cr Zn Ti Others
8091 0.30 0.50 1.6–2.2 0.10 0.50–1.2 0.10 0.25 0.10 Li ¼ 2.4–2.8
Zr ¼ 0.08–0.16
8093 0.10 0.10 1.0–1.6 0.10 0.9–1.6 0.10 0.25 0.10 Li ¼ 1.9–2.6
Zr ¼ 0.04–0.14
Chemical composition of wrought aluminum alloys of the 2XXX, 3XXX, 4XXX, 5XXX, 6XXX, 7XXX, 8XXX groups. Composi-
tions are in wt%. Aluminum content is balance (Lumely 2011; Polmear et al. 2017; Prasad et al. 2014; The Aluminum Association
2015)

Appendix 2: Alloying Elements in Cast and Wrought

Aluminum Alloys

Element Symbol Effects

Chromium Cr Generally added to Al-Mg, Al-Mg-Si, and Al-Mg-Zn alloys. Cr addi-
tions prevent recrystallization during heat treatment. Above 0.35%, it
tends to form coarse intermetallics with other impurities or with Ti and
Mn. The presence of Cr in wrought products leads to the formation of
fine dispersoid phases that inhibit nucleation and grain growth. This
latter aspect is interesting during heat treatments and hot working of
both Al-Mg-Si and Al-Mg-Zn alloys. A critical issue related to the
presence of chromium is the increase in quenching sensitivity of heat-
treatable alloys in case the hardening phases precipitate on pre-existing
Cr-containing particles
Copper Cu Copper is one of the most important alloying elements in Al. Its content
ranges between 2% and 10%. Both cast and wrought alloys can be
subjected to an increase in mechanical properties after solution treat-
ment and subsequent aging. This is possible thanks to the formation of
Cu-containing precipitates (see Sect. 3.2.3.3). Strengthening effect is
maximum for wt% ranging between 4% and 6%, depending on the
alloy and on the influence of the other alloying elements that may have
synergistic hardening effects. In 7XXX series, copper enhances the
aging rate by increasing the degree of supersaturation. In general, Cu
increases the alloy resistance to stress corrosion but decreases the
resistance to general corrosion of Al-Zn-Mg alloys
Iron Fe In Al-Cu-Mg cast alloys, iron increases strength and hardness at
elevated temperatures. The iron-containing alloys are also character-
ized by improved load-bearing capabilities and dimensional stability.
An excess of Fe leads to the formation of the several intermetallic
phases, among which Cu2FeAl7. This decreases the amount of copper
available to precipitate formation, with potentially deleterious effects
on the effectiveness of the heat treatments that may result less effective
in forming the strengthening phases. In Al-4Cu-0.5 Mg wrought alloy,
an Fe content lower than 0.5% lowers the tensile properties after heat
treatment. This phenomenon occurs, unless a sufficient concentration
of Si is present for the formation of the FeSi intermetallic. The strength
properties remain unaltered when a sufficient amount of Si combines
(continued)
Appendices 551

Element Symbol Effects

with Fe to form FeSi, although this phase tends to lower the fracture
toughness of the alloy
Lithium Li Lithium reduces the density of the alloy and increased the yield
strength. Lithium forms ordered δ’ (Al3Li) precipitates that are
responsible for the increase in mechanical properties. Lithium content
may vary depending on the alloy (see Table 3.12). Further details about
the role in aluminum alloys are given in Sect. 3.2.4
Magnesium Mg Magnesium can be added to Al-Cu alloys, resulting in an increase in
strength induced by solution treatment and quenching. Mg shows a
great influence on the aging characteristics of the alloy. Minor addi-
tions (0.5%) can lead to significant variations in properties after aging
treatments. The influence of Mg on corrosion resistance of Al-alloys
depends on the alloy itself and on the thermal treatment. Strengthening
effect of Mg on wrought alloys can be maximized by means of cold-
working processes before aging. This effect is significant in artificially
aged products and not that much with naturally aged and cold-worked
alloys. When present in combination with silicon, magnesium can form
the Mg2Si intermetallic that in the β0 -metastable condition is an
important strengthening element (see Sect. 3.2.3.3)
Manganese Mn Manganese addition leads to an increase in tensile strength, as well as
in yield strength. It prevents recrystallization during heat treatment.
Commercial alloys usually do not contain more than 1% Mn since its
presence can lead to a loss in ductility
Nickel Ni Nickel is used to improve high-temperature strength of both cast and
wrought Al-Cu-Mg alloys. Although 0.5% Ni reduces the tensile
properties of heat-treated wrought Al-4Cu-0.5 Mg-alloys at room
temperature
Silicon Si Silicon combines with magnesium leading to the formation of the β’
(Mg2Si) strengthening phase (see Sect. 3.2.3.3). In 6XXX series alloys,
Mg to Si ratio is approximately kept around 1.73:1 in order to promote
the formation of the Mg2Si intermetallic. The maximum solubility of
Mg2Si in the Al matrix is 1.85%, and it decreases with temperature.
Mg2Si precipitation upon age hardening occurs by formation of GPZs
and very fine precipitates. An excess in Si leads to segregation at the
grain boundaries, causing grain boundary fracture in recrystallized
structures. In Al-Cu-Mg alloys, Si additions reduce cracking tendency
of the material
Silver Ag Silver is added to increase the strength after heat treatment and aging
Ti Titanium Titanium additions are used for grain refinement of castings and ingots.
Presence of boron enhances the effect of titanium
Zinc Zn Zinc forms intermetallics with Mg (MgZn2). An increase in Zn content,
as well as the combined increase in both Mg and Zn contents, decreases
the corrosion resistance of the alloy. In 7XXX alloy series, Zn and Mg
have the control of the aging process
Zirconium Zr Zirconium concentrations in the 0.1–0.3% range are used to inhibit
recrystallization and recovery thanks to the formation of a fine pre-
cipitation of intermetallics within the alloy. Zr additions are used in an
increasing number of alloys, in particular those belonging to the Al-Zn-
Mg family, for increasing recrystallization temperature, thus allowing a
(continued)
552 Appendices

Element Symbol Effects

fine grain size to be retained even at higher treatment temperatures. The
microstructure of wrought products can be controlled by means of Zr
additions. These Zr-containing alloys are less quench sensitive, as
compared to those having Cr additions. As-cast grain size can be
reduced by means of Zr additions. Indeed, some superplastic alloys
employ high Zr contents in order to retain a relatively fine micro-
structure during high-temperature forming processes
Summary of the effects of the most important alloying elements in aluminum alloys. Concentrations
in wt% (Polmear et al. 2017; Prasad and Wanhill 2017)

Appendix 3: Alloying Elements in Magnesium Alloys

Element Symbol Effects

Aluminum Al Aluminum is a common alloying element in Mg-alloys, and it is used
in both wrought and cast alloys. Its main function consists in the
formation of the Mg17Al12 strengthening phase. The maximum solu-
bility of Al in Mg is 12.7% at 473  C. When the Al content exceeds the
6%, the alloy can be heat-treated. However, increasing Al content leads
the alloy to be more prone to stress corrosion cracking. Al is also able
to improve casting properties since it widens the solidification range of
the alloy in which it is contained
Beryllium Be The maximum solubility of beryllium in Mg is 0.001%. It has benefi-
cial effects because it reduces the oxidation of liquid Mg. However, its
use is not recommended in cast alloys because it induces grain
coarsening
Calcium Ca Calcium is used as grain refiner. It enhances the creep as well as the
corrosion resistance. Ca can improve the deformability of alloys
subjected to rolling processes (e.g., sheet production), although its
content must be kept below 0.3% in order to avoid the alloy to become
more prone to crack formation after welding processes. Cast alloys that
contain Ca show a reduced oxidation of the liquid surface of the melt
and are less prone to oxidation during heat treatment
Ceriuma Ce Cerium has positive effects on the elongation, even in small amounts
(0.2%), thus improving the alloy deformability
Cobalt Co Cobalt has detrimental effects on the corrosion resistance. Together
with Cu, Fe, and Ni, it is considered as a harmful impurity
Copper Cu Copper has detrimental effects on the corrosion resistance. A signifi-
cant drop of this property occurs when Cu content exceeds 0.05%.
However, its presence can lead to an improvement of room-
temperature and high-temperature strength
Gadoliniuma Gd Gadolinium improves both the corrosion resistance, the high-
temperature strength, and the creep resistance
Iron Fe Iron is regarded as a detrimental impurity since its presence, like that of
copper, reduces the corrosion resistance of the alloy. In order to obtain
(continued)
Appendices 553

Element Symbol Effects

the maximum corrosion resistance, 0.005% is the highest Fe content
for Mg-alloys
Lithium Li The maximum solubility of Li in Mg is 5.5% at 588  C. Lithium
additions reduce the density of the alloy. Furthermore, Li reduces the
strength of the alloy while enhancing its ductility
Manganese Mn Manganese leads to an improvement of the alloy weldability. It has
very limited effects on the ultimate and yield strength of the alloy and
can be used for grain-refining purposes. When added to Mg-Sc alloys,
it improves the creep resistance thanks to the formation of Mn2Sc and
other MnxScy compounds
Neodymiuma Nd Neodymium has a high solubility limit in magnesium. The formation
of stable precipitates, both within the grain and at the grain boundaries,
determines an improvement in the strength of the alloy
Nickel Ni Similar to Fe and Cu, nickel is considered as a detrimental impurity for
what concerns the corrosion resistance. In order to obtain the highest
corrosion resistance, the maximum nickel content is 0.005%. How-
ever, the formation of Mg2Ni intermetallic phase is able to enhance the
room-temperature strength
Silicon Si The main effect of silicon is to increase the fluidity of liquid magne-
sium; therefore, cast alloys usually contain this element. When added
together with REs, it forms stable silicides that improve high-
temperature creep and stress resistance properties. In association with
iron leads to a reduction in the corrosion resistance properties.
Silver Ag Silver improves the mechanical properties of Mg-alloys that have been
subjected to age hardening
Strontium Sr Strontium is usually added together with other higher concentration
alloying elements. Sr does not have significant influence on yield and
ultimate strength of the alloy. However, its presence has been reported
to improve the creep performances of the alloy
Thorium Th Since it is a radioactive element, the use of thorium is limited to a few
alloys. The usual Th content in Mg-alloys ranges between 2% and 3%.
Its addition leads to an improvement in creep resistance properties up
370  C. Th improves high-temperature strength as well. The
weldability of Zn-containing alloys is improved by thorium additions
Yttriuma Y Yttrium has high solubility in Mg (~12.6%). High-strength alloys such
as WE54 and WE43 contain Y in the 4–5% range. Yttrium is added
together with RE elements to improve the high-temperature strength,
creep performances, and corrosion resistance
Zinc Zn Zinc introduces both solid solution and age hardening in Mg-alloys. It
is used in combination with Al in order to increase the strength of the
alloy without reducing its ductility. Moreover, Zn has positive effects
on the high-strength precipitation hardenable alloys containing REs,
thorium and zirconium. Its presence enhances the corrosion resistance,
avoiding the detrimental effects of Fe and Ni impurities
Zirconium Zr The maximum solubility of zirconium in Mg is 0.50% at 654  C, while
the maximum solubility at room temperature is 0.28%. It is a strong
grain refiner, especially in alloys containing Ag, REs, Th and Zn. The
addition of Zr to alloys containing REs and Zn leads to an increase in
(continued)
554 Appendices

Element Symbol Effects

both strength and ductility. The grain-refining action of zirconium
fades away in alloys containing Al, C, Fe, and Si and gaseous inclu-
sions of H2, N2, and O2, since it reacts with these elements. The
development of Mg-Zr alloys for aerospace applications resulted in
alloys with enhanced strength, in particular at elevated temperatures. A
good example of Zr-containing alloy is WE43, which long-term tem-
perature stability is maintained up to 250  C
Effect of the most important alloying elements in magnesium alloys, including impurities. Con-
centrations in wt% (Prasad and Wanhill 2017; Zhang and Zhao 2012)
a
REs-containing Mg-alloys have been extensively investigated in recent years. For REs atoms, it is
not easy to diffuse in the alloy, and this has positive effects on the recrystallization temperature.
Furthermore, they form small and stable precipitates. The addition of REs usually leads to high
aging and solid solution effects, thus increasing high-temperature stress resistance, creep resistance,
and corrosion resistance. In some cases, the addition of specific RE elements leads to an improve-
ment of the ductility as well. This is the case of Ce, Gd, Nd, and also Y, not a true RE element.
Finally, the addition of REs leads to a narrowing of the solidification range of the alloy. This latter
feature is beneficial for the minimization of the micro-shrinkage porosities in castings and for the
reduction of the probability to form cracks during welding processes

Appendix 4: Alloying Elements in Steels

Carbides
Element Symbol γ-stabilization formation Main alloying effects
Aluminum Al No No Al is a strong deoxidizer and can be used
for this purpose during alloy refinement,
either alone or in association with Mn
and Si. Al can form non-soluble inclu-
sions (e.g., Al2O3) in molten steel, both
during primary melting and refining
process. It is a very effective element
when used as grain refiner. Its presence
promotes precipitation hardening in
stainless steels. It is used in nitriding
steels leading to the formation of very
hard aluminum nitrides (1000 HV). Al is
used for increasing corrosion resistance
in low-carbon steels and for increasing
oxidation resistance in heat-resistant
steels
Antimony Sb No – It is considered as an impurity and it is
not intentionally added. Moreover, it is
difficult to remove during melting and
refining processes. Its presence lowers
the surface quality of continuously cast
billets. It causes temper embrittlement in
alloy steels. Sb has a strong tendency to
(continued)
Appendices 555

Carbides
Element Symbol γ-stabilization formation Main alloying effects
segregate, especially at grain boundaries
and during cooling processes
Beryllium Be – – Beryllium improves the corrosion resis-
tance of steels
Boron B No – Boron is usually added in very limited
amounts, ranging from 0.0005% to
0.0035%. B tends to segregate at grain
boundaries, thus improving
hardenability of steels. Boron is a strong
nitride former, and this feature interferes
with the hardening effect of elemental
B. The addition of a specific amount of
Ti to the steel protects B from nitride
formation because Ti combines prefer-
entially with N, thus avoiding boron
nitride formation
Calcium Ca – – Ca is a strong deoxidizer. Concerning
steelmaking processes, the SiCa treat-
ment can be used in order to modify
inclusion morphology and type. In par-
ticular, the process helps removing alu-
mina from the molten bath thanks to the
combination of Ca and S. This process
leads to CaO∙xAl2O3 inclusions
surrounded by liquid CaS. Thanks to this
phenomenon, Al2O3 is removed from the
liquid steel and contributes to the forma-
tion of the liquid slag. SiCa treatment can
reduce sulfur and oxygen content in the
alloy. In killed steels, CaS is present as
spherical particles. Ca, Al, and Si tend to
form low-melting compounds that results
in better machinability of the steel. Some
problems arise when CaS-based inclu-
sions become too large, promoting the
formation of fatigue cracks
Carbon C Yes Yes Carbon is, with iron of course, the base
element in steels. It is inexpensive and
can be easily added in various forms
during steelmaking. During solidifica-
tion, carbon shows a moderate tendency
to macrosegregation, being more prone
to segregating at microstructural defects,
such as grain boundaries and disloca-
tions. Carbon is an interstitial, both in fcc
and in bcc iron, although with different
solubility limits, and has a strengthening
effect, enhancing tensile strength, yield
strength, and hardness. This is possible
(continued)
556 Appendices

Carbides
Element Symbol γ-stabilization formation Main alloying effects
through carbide dispersion strengthening
and solid solution strengthening.
Strengthening effect is proportional to
the C content up to a threshold value
(around 1% for carbon steel). Notch
toughness and weldability of steels
decrease at higher carbon concentrations
Chromium Cr No Yes Chromium is one of the most important
alloying elements in steels, it is a carbide
former, and for increasing Cr/C ratios,
Cr2C, and, the most common, Cr7C3 and
Cr23C6 would form. Cr carbides increase
hardness and wear resistance. Chromium
carbides precipitation at the grain bound-
ary (i.e., sensitization) lowers the corrosion
resistance of stainless steels, in particular
in the heat-altered zone (HAZ) after
welding processes. For this reason, tita-
nium is introduced as preferential carbide-
forming element. Chromium implies an
increase in mechanical properties, in
hardenability, in oxidation and in general
corrosion resistance. Stainless steels usu-
ally contain more than 13% Cr, and this
leads to the formation of a thin, stable, and
adherent Cr2O3-based layer on their sur-
face causing an increase in the corrosion
resistance in many different environments
Cobalt Co Yes No The use of cobalt as alloying element has
been recently limited by its toxicity. As a
consequence of this, Co-free alloys have
been developed. By adding cobalt to
carbon steel, the material undergoes a
reduction in its hardenability. When
combined with Cr, it increases the
hardenability of Cr-Mo steels. Its use is
generally limited to the production of
tool steels, maraging steels, creep-
resistant steels, and high-speed steels
and for steels used in high-temperature
applications. Co is a grain boundary sta-
bilizer at high temperature and contrib-
utes to maintaining high-temperature
strength and temper as well. This latter
feature is the reason for the increase in
useful lifetime of tool steel items. Cobalt
promotes precipitation hardening and
acts as a stabilizer agent for martensite
increasing the Ms temperature and
reducing the concentration of retained
austenite in alloy steels
(continued)
Appendices 557

Carbides
Element Symbol γ-stabilization formation Main alloying effects
Copper Cu Yes No Copper is generally regarded as an
impurity in steels, and its content is kept
below 0.04%. It cannot be oxidized and
removed during refining of steels, and
the best approach for reducing Cu con-
tent in steel is through dilution. A par-
ticular class of low-alloy steels contains
an appreciable amount of Cu for
increasing atmospheric corrosion resis-
tance. This class of steels is named
weathering steels, and thanks to the
presence of Cu, P, Ni, and Cr, they are
able to form a protective layer on the
surface. This family of steels is also
known as COR-TEN. PH stainless steels
use Cu for increasing the strength thanks
to the precipitation of Cu-rich interme-
tallics during aging treatment. Copper is
also added to austenitic stainless steels,
and when exceeding 1% it causes an
improvement in the corrosion resistance
against HCl and H2SO4 and against
stress corrosion phenomena (see Sect.
8.1.6)
Hydrogen H – – Hydrogen is an interstitial atom, it is
soluble in liquid steel, and its presence
can be very detrimental to the material.
Modern steelmaking processes imply
degassing stages for removing this ele-
ment from liquid steel. The presence of
hydrogen can lead to hydrogen embrit-
tlement (see Sect. 8.1.7)
Lead Pb – – Lead is insoluble in iron, and it is used
for increasing machinability of the
material. When added to steel, it tends to
form a particle dispersion throughout the
material. This is particularly useful while
machining, since it results both in a good
lubrication and increased chip-breaking
ability. However, leaded steel market has
experienced a contraction in the last
decades because of the environmental
hazard and health risk associated with
lead. Therefore, modern steelmaking
processes tend to prefer the use of sulfur
for the production of steels with
improved machinability
Manganese Mn Yes Yes Manganese is both a γ-stabilizer element
and a moderate carbide-forming ele-
ment; it dissolves in Fe3C and forms
(continued)
558 Appendices

Carbides
Element Symbol γ-stabilization formation Main alloying effects
alloyed cementite as well. Other typical
Mn carbides are Mn23C6 and Mn15C4.
Manganese is present, as substitutional,
in almost all steel grades in a quantity
0.30%. In steelmaking processes, Mn
is used both as deoxidizer agent (often in
association with Al, Si, or both) and for S
removal, through the formation of MnS,
thus avoiding the deleterious formation
of FeS. Manganese enhances the steel
hardenability for concentrations exceed-
ing 0.8%. Steels with Mn concentrations
larger than 2% are used for special pur-
poses, such as railway track connectors
and rock crushers. These steels belong to
a special class named Hadfield manga-
nese steels, also known as austenitic
manganese steels
Molybdenum Mo No Yes Mo is an α-stabilizer element. It is a
strong carbide former element, Mo2C
and Mo6C being the main phases found
in steels. Mo carbides are important for
steel grades used in wear-resistant
applications (e.g., tool steels). Molybde-
num is added for minimizing temper
embrittlement, but it cannot avoid it
completely. It is an effective substitu-
tional solution hardener and can induce
secondary hardening during tempering
of quenched steels and in austenitic
alloys at high temperatures. It acts also
as grain refiner when added to fine-
grained steels. Mo improves corrosion
resistance; it is used both in ferritic and
austenitic stainless steels, acting as a
passivation-enhancing element in stain-
less steels, thus promoting a fast forma-
tion of the passivating Cr2O3 layer, also
increasing its stability. Mo is also present
in duplex stainless steels. Super-
austenitic stainless steels have a Mo
content  6%; the high Mo content
reduces the susceptibility to pitting cor-
rosion but can lead to the formation of
brittle intermetallic compound
(chi-phase), which is responsible for the
reduction in deformability of the
alloy. Mo is also one of the alloying
elements used in Maraging steels (see
Sect. 5.4.1), in view of its positive effects
on hardening, plasticity and ductility in
the aged condition
(continued)
Appendices 559

Carbides
Element Symbol γ-stabilization formation Main alloying effects
Nickel Ni Yes No Nickel is a strong γ-stabilizer and a non-
carbide-forming element, and thereby it
is a necessary alloying element in aus-
tenitic stainless steels. Due to its position
in the Ellingham diagram (see Sect.
8.2.1), it does not undergo oxidation
during steelmaking processes, meaning
that the amount of Ni is constant over
time during the process. The substitu-
tional solid solution of nickel in iron has
a limited effect on increasing the alloy
yield strength. However, this element
improves the hardenability of steel, in
particular when combined with Cr and
Mo. In this condition the material
undergoes also an increase in fatigue
resistance and in impact toughness. It
reduces the ductile to brittle transition
temperature. For this reason, nickel is
used as alloying element in steels for
cryogenic applications. If combined with
Al, it forms with aging treatments the
intermetallic Ni3Al precipitates, on
which precipitation hardening of
PH steels is based
Niobium Nb No Yes Niobium is a strong carbide-, nitride-,
and carbonitride-forming element. Nio-
bium (named columbium in the US lit-
erature) introduces a grain refinement
effect in steels. It is used in HSLA (high-
strength low-alloy) steels in order to
enhance the strength by maintaining
good toughness through precipitation
strengthening and the formation of
Nb-carbonitrides, and it is also able to
prevent carbide dissolution and
reprecipitation in undesirable locations.
For this reason, it is added to some fer-
ritic stainless steels (e.g., AISI 436 and
437), austenitic stainless steels (e.g.,
AISI 347, 348, and 384), and also to
17-4 PH stainless steels
Nitrogen N Yes No Nitrogen is an important γ-stabilizer
element, which forms, together with
carbon and metals, carbonitrides. Nitro-
gen may also partially replace nickel as
austenite stabilizer in austenitic steels, in
many of which it is used and results in
solution or precipitation strengthening.
(continued)
560 Appendices

Carbides
Element Symbol γ-stabilization formation Main alloying effects
This is the case of AISI types 201, 202,
205, 304N, 304LN, 316N, and 316LN.
Nitrogen is also used in duplex stainless
steels for increasing pitting corrosion
resistance and strength
Oxygen O – – Oxygen is considered as an impurity in
steels. Usually, the oxygen content has
to be lower than 60 ppm. The solubility
of oxygen in iron is very low, and this
leads to undesired oxides formation or
porosities
Phosphorus P No No Phosphorus is generally regarded as an
impurity in steels, and its content is usu-
ally below 0.02%. Upon increasing the P
content, the material experiences a
decrease in toughness and ductility. It has
a strong tendency for segregation at the
grain boundary, and this leads to delete-
rious effects like temper embrittlement.
However, for specific steel grades, a high
P content can have beneficial effects.
Low-alloy structural steels with high P
content show better atmospheric corro-
sion resistance. The addition of P can
cause precipitation hardening and yield
strength improvement in some austenitic
Cr-Ni steels. Re-phosphorized steels are a
special class of alloys with improved
strength and enhanced machinability
Rare earth metals – – Rare earth metals are a group of 17 ele-
ments belonging to the III B group. They
can have strong deoxidizing and
desulfurizing effects. These elements
have also a tendency to segregate at the
grain boundaries. Rare earth metals have
effects on shape modification and con-
trol of inclusions; in particular, La and
Ce are added for the sulfide shape con-
trol. Rare earth metals can also increase
the hardenability of steels
Silicon Si No No During steel refinement it is used as
deoxidant, either alone or in combination
with Al and Mn or both. It does not form
carbides in steel, but its presence may
lead to the formation of SiO2 inclusions.
Silicon retards the formation of Fe3C
during tempering and increases yield
strength, hardenability, and wear resis-
tance in heat-treated steels. It is also
(continued)
Appendices 561

Carbides
Element Symbol γ-stabilization formation Main alloying effects
added to a particular family of steels
called transformer steels. They have sil-
icon content around the 3% and are
characterized by low core losses and
high magnetic permeability. Silicon dis-
solves into martensite, and it is able to
retard martensite decomposition up to
300  C
Sulfur S No – Like P and O, sulfur is generally con-
sidered as an impurity, with concentra-
tions below 0.02%. The tendency of
sulfur for segregating at the grain
boundary is very high, and this causes
the reduction of the high-temperature
ductility in alloy steels. This element is
insoluble in Fe and can form deleterious
low melting compounds, such as FeS.
However, similar to P, sulfur is inten-
tionally added to a special class of steels
that have a S content between 0.08% and
0.35% resulting in an improved
machinability of the alloy
Tantalum Ta – Yes Like niobium, tantalum is a strong car-
bide former. It has a very high affinity to
carbon; in some austenitic stainless
steels (i.e., AISI type 347 and AISI 348),
it is used as carbide stabilizing agent
Tin Sn – – Tin is considered as a detrimental impu-
rity, in view of its effects on steel prop-
erties. One of the deleterious
consequences of its presence is the tem-
per embrittlement and the quality of the
surface of continuously cast steels
Titanium Ti – Yes Titanium is considered as a very strong
deoxidizer due to its high affinity to
oxygen. During melting and refining, the
addition of Ti has to be carried out care-
fully in order to avoid its oxidation and
subsequent migration of oxides into the
slag. Titanium is also an effective former
of nitrides and carbides, and its effects
are similar to those of V and
Nb. However, Ti carbides and nitrides
are more stable if compared with Nb-
and V-based ones. Ti nitrides act as grain
boundary stabilizers. Ti is also used as
carbide stabilizing agent in austenitic
stainless steels, ferritic and PH steels.
Titanium improves intergranular
(continued)
562 Appendices

Carbides
Element Symbol γ-stabilization formation Main alloying effects
corrosion (i.e., sensitization) in austen-
itic stainless steels. This element is also
used in Fe-Ni-Cr alloys (e.g., alloy
A286) for the formation of γ’ precipi-
tates. In maraging steels, Ti forms the
intermetallic hardening H-phase (see
Table 5.6). Furthermore, the addition of
Ti prevents boron nitrides formation, so
that boron can retain its elemental form
Tungsten W No Yes Tungsten is an α-stabilizer and carbide
former element. Carbides (WC and
W2C) are very hard and of same stoi-
chiometry and structure of Mo carbides.
Tungsten induces secondary hardening
during tempering treatment of quenched
steels, limits grain growth at high tem-
perature, and together with Mo, is the
most important alloying element in tool
steels
Vanadium V – Yes Vanadium forms extremely hard car-
bides and nitrides. V carbides have
higher hardness than Cr, Mo, and Fe
carbides. Vanadium is used for grain
refinement, resulting in strength and
toughness increase. Vanadium can also
dissolve into austenite thus increasing its
hardenability. Vanadium addition intro-
duces large secondary hardening effect
after tempering, with a general
improvement of the high-temperature
hardness of the material
Zirconium Zr No Yes Zirconium in steel limits the austenite
phase field. It is a carbide- and nitride-
forming element. Its effects on steel are
similar to those of titanium, although it is
very expansive and, therefore, not so
common in steels
There are at least two possible criteria for the classification of alloying elements in steels. The first is
based on the stabilization of the austenite (γ) phase. The second one is based on their tendency to
form carbides. The carbide formation requires that concentration of the carbon would locally
overcome a threshold value, i.e., its solubility limit in the relevant alloy; otherwise the element
would remain in solution. According to the aforementioned criteria, the information concerning the
most important alloying elements, and also impurities, in steels are given in this appendix.
Concentrations in wt% (Campbell 2008, Kutz 2002, Totten 2007)
Appendices 563

Appendix 5: Alloying Elements in Superalloys

Fe-based
Element Symbol Co-based superalloys superalloys Ni-based superalloys
Aluminum Al Better oxidation Precipitation of the Moderate solid solu-
resistance γ0 phase tion hardening effect
Formation of the β-CoAl Delayed formation Precipitation of the
phase of the hexagonal η γ0 phase
Ni3Ti phase
– – Better oxidation
resistance
Boron B Better creep strength Better creep Moderate solid solu-
and ductility strength and tion hardening effect
ductility
Refinement of grain Retards formation Inhibition of car-
boundary of grain boundary bides coarsening
η Ni3Ti
Workability – Better grain bound-
improvement ary strength
– – Better creep strength
and ductility
Carbon C Carbides formation Carbides Carbides formation
formation
Decrease in ductility Fcc matrix Moderate solid solu-
stabilization tion hardening effect
Refinement of grain – –
boundary
Cobalt Co Matrix element – Increase in γ solidus
temperature
– – Moderate increase in
the γ0 volume frac-
tion in some alloys
– – Increase in γ solvus
temperature
– – Better resistance to
hot corrosion under
salt deposit
Chromium Cr Precipitation of M23C6 Hot corrosion and Hot corrosion and
and M7C3 carbides oxidation resis- oxidation resistance
tance improvement improvement
Hot corrosion and oxi- Solid solution Precipitation of
dation resistance hardening M23C6 and M7C3
improvement carbides
Promotes TCP phase – Moderate solid solu-
formation tion hardening effect
– – Moderate increase in
the γ0 volume
fraction
(continued)
564 Appendices

Fe-based
Element Symbol Co-based superalloys superalloys Ni-based superalloys
– – Promotes TCP phase
formation
Hafnium Hf – – Creep strength and
ductility
improvement
– – Improvement in
grain boundary
strength
– – Formation of the
HfC carbide
– – Promotes eutectic γ/
γ0 formation
Iron Fe Workability Matrix element Decrease in the oxi-
improvement dation resistance
Fcc matrix stabilization – Promotes TCP phase
formation
– – Workability
improvement
Molybdenum Mo Solid solution hardening Solid solution High solid solution
hardening hardening effect
Formation of the Co3Mo Precipitation of the Moderate increase in
intermetallic M6C carbide the γ0 volume
fraction
Promotes TCP phase – Increase in alloy
formation density
Niobium Nb Precipitation of MC and Precipitation of the High solid solution
M6C carbides γ00 phase hardening effect
Formation of the Co2Nb Precipitation of the Large increase in the
intermetallic NbC carbide γ0 volume fraction
Reduction in the surface Precipitation of the Formation of the
stability δ-Ni3Nb phase NbC carbide
– – Precipitation of the
γ00 phase
– – Promotes γ0 phase
formation
– – Precipitation of the
δ-Ni3Nb phase
Nickel Ni Fcc matrix stabilization Fcc matrix Matrix element
stabilization
Decrease in hot corro- Inhibition of TCP –
sion resistance phase formation
Rhenium Re – – Moderate solid solu-
tion hardening effect
– – Retards γ0
coarsening
– – Increase in the γ/γ0
lattice mismatch
(continued)
Appendices 565

Fe-based
Element Symbol Co-based superalloys superalloys Ni-based superalloys
Tantalum Ta Precipitation of MC and Precipitation of the High solid solution
M6C carbides γ00 phase hardening effect
Formation of the Co2Ta Precipitation of the Precipitation of the
intermetallic TaC carbide TaC carbide
Reduction in the surface – Large increase in the
stability γ0 volume fraction
– – Oxidation resistance
improvement
Titanium Ti Precipitation of the TiC Precipitation of the Moderate solid solu-
carbide γ0 phase tion hardening effect
Formation of the Co3Ti Precipitation of the Precipitation of the
intermetallic TiC carbide γ0 phase
Formation of the Ni3Ti – Precipitation of the
intermetallic (with suffi- TiC carbide
cient amount of Ni)
Reduction in the surface – –
stability
Tungsten W Effective solid solution Effective solid Effective solid solu-
hardening solution hardening tion hardening effect
Formation of the Co3W Precipitation of the Moderate increase in
intermetallic M6C carbide the γ0 volume
fraction
Promotes TCP phase – Formation of the
formation M6C carbide
– – Increase in alloy
density
– – Promotes TCP phase
formation
Zirconium Zr Formation of MC Creep strength and Moderate solid solu-
carbide ductility tion hardening effect
improvement
Creep strength and duc- Retards formation Inhibition of carbide
tility improvement of grain boundary coarsening
η Ni3Ti
Reduction in the surface – Better grain bound-
stability ary strength
Refinement of grain – Better creep strength
boundary and ductility
Role of alloying elements in superalloys. Concentrations in wt% (Geddes et al. 2010)
566 Appendices

Appendix 6: Precipitates and Secondary Phases

in Superalloys

Crystal Lattice parameters

Phase Symbol structure [nm] Formula Other information
Delta δ Orthorhombic a0 ¼ 0.3624–0.511 Ni3Nb δ phase is an
(ordered b0 ¼ 0.3624–0.4251 intergranular pre-
Cu3Ti) c0 ¼ 0.7406–0.4556 cipitate that has
detrimental conse-
quences on the
creep response of
the alloy. Observed
in Incoloy 718. It
has different mor-
phologies
depending on the
formation mecha-
nism. An acicular
shape is observed
when formed in the
815–980  C range.
It forms by cellular
reaction at low
aging temperatures.
Intragranular pre-
cipitation occurs at
high aging
temperatures
Eta η hcp (D024) a0 ¼ 0.5093 Ni3Ti The η phase has
c0 ¼ 0.8276 been observed in all
the three families of
superalloys when
high Ti/Al ratios are
present. Its forma-
tion occurs as
intergranular pre-
cipitates or
intragranular acicu-
lar platelets in a
Widmanstätten
pattern
(continued)
Appendices 567

Crystal Lattice parameters

Phase Symbol structure [nm] Formula Other information
Gamma γ0 fcc (ordered 0.3561 for pure Ni3Al Main strengthening
prime L12) Ni3Al to 0.3568 for Ni3(Al, phase in Ni- and Fe-
Ni3(Al0.5Ti0.5) Ti) Ni-based superal-
loys. Its shape
varies from spheri-
cal (e.g., in
Nimonic 80 and
Waspaloy) to
cuboidal in more
recent Ni base
superalloys. Varia-
tions in the Al/Ti
ratio and in the Mo
content influence
the morphology of
the phase. In Ni-Fe-
based alloys is
metastable. When
the mismatch with
the matrix is high,
extended exposure
above 700  C leads
to conversion in the
deleterious η
(Ni3Ti) or δ
(Ni3Nb) phases
Gamma γ00 bct (ordered a0 ¼ 0.3624 Ni3Nb γ00 is the main
double D022) c0 ¼ 0.7406 strengthening phase
prime in Incoloy 718.
Usually, precipi-
tates have the form
of coherent disc-
shaped particles
that form on the
{100} planes with
an average diameter
of 600 Å and a
thickness ranging
between 50 and
90 Å. In case of
overaging of the
alloy, γ00 phase
converts into the
deleterious δ phase
(continued)
568 Appendices

Crystal Lattice parameters

Phase Symbol structure [nm] Formula Other information
Laves – Hexagonal a0 ¼ 0.475–0.495 Co2Ti They are classified
Fe2Mo as TCP phases typ-
c0 ¼ 0.770–0.815 Fe2Nb ical of Co and Fe
base superalloys.
Fe2Ta
Observed in
Fe2Ti Co-based superal-
loys with Ti content
greater than 5 wt%
MC – Cubic a0 ¼ 0.430–0.470 HfC Usually primary
NbC carbide that forms
TaC directly from the
melt. MC carbides
ThC
appear as globular,
TiC irregularly shaped
ZrC particles
M7C3 – Hexagonal a0 ¼ 1.398 Cr7C3 M7C3 carbide has
c0 ¼ 0.4523 generally a blocky
intergranular shape.
It has been
observed in some
Co-based superal-
loys and in other
alloys such as
Nimonic 80A after
exposure at temper-
atures exceeding
1000  C
M6C – fcc a0 ¼ 1.085–1.175 Co3Mo3C It is a randomly
Fe3Mo3C distributed carbide.
Fe3Nb3C The metal is com-
monly molybde-
Fe3W3C
num. It can be
Fe4W2C originated from the
Nb3Co3C melt or from the
Ta3Co3C decomposition of
MC carbide at high
temperatures
M3B2 – Tetragonal a0 ¼ 0.560–0.620 Fe3B2 Borides appear very
c0 ¼ 0.300–0.330 Mo2FeB2 similar to carbides;
Nb3B2 they have been
observed in both
Ta3B2
Fe-Ni- and
V3B2 Ni-based superal-
loys when B con-
tent is 0.03%, or
greater
(continued)
Appendices 569

Crystal Lattice parameters

Phase Symbol structure [nm] Formula Other information
M2SC – hcp a0 ¼ 0.3234–3.386 (Nb, Ti, This phase is
c0 ¼ 11.233–11.834 Zr)2SC observed in alloys
with high levels of
S or low C/S ratios
M23C6 – fcc a0 ¼ 1.050–1.070 Cr23C6 The M23C6 carbide
according to (Co, Fe, usually forms at
composition Mo, grain boundaries in
W)23C6 the form of either
globules, lamellae,
or platelets. The
most common
composition
implies Cr as metal
(particularly in
Co-based superal-
loys); however, Co,
Fe, Mo, and W can
substitute Cr. In Co
superalloys it can
originate from the
decomposition of
M6C carbides at
high temperature
MN – Cubic a0 ¼ 0.4240 NbN MN nitrides can be
TiN observed in alloys
ZrN containing Nb, Ti,
or Zr. They are
characterized by
square to rectangu-
lar shapes and usu-
ally are used by
carbides as nucle-
ation sites. Nitrides
are not soluble
below the melting
temperature
Mu μ Rhombohedral a0 ¼ 0.475 (Fe, It generally appears
Co)7(Mo, as coarse, irregular
W)6 Widmanstätten
c0 ¼ 2.577 Co2W7 platelets. It forms at
high temperatures
in alloys with high
content of Mo or W
(continued)
570 Appendices

Crystal Lattice parameters

Phase Symbol structure [nm] Formula Other information
Sigma σ Tetragonal a0 ¼ 0.880–0.910 (Fe, It is classified as a
c0 ¼ 0.450–0.480 Ni)x(Cr, TCP phase typical
Mo)y of Fe-Ni- and
Co-based superal-
loys. It is less com-
mon in Ni base
superalloys. σ
phase has detri-
mental effects on
both mechanical
properties and cor-
rosion resistance
An overview on the main precipitate and secondary phases that can be found in all the three
superalloy families (ASM International 1991; Geddes et al. 2010)