Nickel-Base Superalloys; Physical Metallurgy of Recycling

J.J. deBARBADILLO

Because of stringent performance specifications, alloys included in the category of nickel-base super-
alloys are some of the most metallurgically complex and expensive in the field of metallurgy. The fact
that these complex alloys can be manufactured commercially to make reliable components reflects the
skill which alloy producers apply to control composition to meet rigid specification of trace elements
as well as alloy constituents. Nevertheless, a substantial amount of scrap is used in the manufacture
of these alloys, and highly specialized operations have been developed to process this scrap. This paper
provides an overview of the current use of scrap in the industry and procedures for handling, melting,
and refining scrap. The effects of residual elements commonly encountered in superalloy scrap on
structure and properties are reviewed. Finally, a brief assessment of trends in generation and use of
scrap is presented.

I. INTRODUCTION substantial amount of "home scrap",* and often some pur-

chased "prompt industrial scrap",* or "obsolete scrap".*
THE aircraft gas turbine engine with its stringent
specifications for materials performance and reliability is *Home Scrap--Scrap generated by an alloy producer during the con-
version of raw materials to primary product.
the best known application for superalloys. Indeed, the Prompt Industrial Scrap--Scrap generated by a fabricator or manu-
American Society for Metals definition of a superalloy: "An facturer during conversion of a primary product to a finished product at a
alloy developed for very high temperature service where location removed from the melting facility.
relatively high stresses are encountered and where oxidation ObsoleteScrap-- Scrap that is generated when used equipment is over-
resistance is frequently required", aptly describes materials hauled or dismantled.
for these applications. ~ However, as this class of alloys ap- There are a number of reasons v~hy recycling is practiced so
proaches its technical maturity, important applications uti- extensively by the superalloy industry, perhaps the most
lizing other characteristics of these remarkable materials important being that the industry itself generates such large
have developed (resistance to aqueous corrosion, hydrogen amounts of scrap. For example, throughout the industry the
embrittlement, etc.). For this paper, a broader definition of weight of semi-finished products shipped is less than one
these alloys will be adopted to include all of the nickel- third of the weight of alloy melted. Some complex machined
chromium and nickel-iron-chromium alloys produced by the components may have a final product yield of less than
traditional manufacturers of superalloys. This broader defi- 10 pct. This scrap has a high intrinsic value because most of
nition is convenient because of the extensive interchange of it is not contaminated, contains valuable and recoverable
scrap and commonality of physical effects of residual ele- alloying elements, and can be readily identified. Even obso-
ments among these materials. lete superalloy scrap is extensively recycled because it is
Whatever the application, superalloys are specified for used in relatively select and identifiable applications where
their outstanding physical characteristics which make it pos- salvage is extensively practiced. In addition, this is a highly
sible to exploit extremes of environment. Since component disciplined, high technology industry where source sepa-
properties and reliability frequently take precedence over ration and alloy identification is practical. Because the in-
price, costly alloying ingredients and sophisticated process- dustry produces many complex proprietary alloys, scrap
ing techniques are routinely used by this industry. Base generally has more value to the generator than it has on the
prices for these alloys may range from about $10 to well over open market. Finally, concerns about supply of scarce or
$200 per kilogram, yet it is not uncommon for finished imported raw materials, reduction of energy consumption,
components to sell for ten times the base alloy price. and even difficulty of disposing of certain low-grade mate-
In view of this, it may be surprising that there is a signifi- rials have all stimulated recycling in recent years.
cant amount of scrap recycled within the superalloy indus-
try. While direct recycling for some applications such as
those for rotating parts in gas turbine engines is effectively II. QUANTITY OF SCRAP GENERATED
precluded, the melt charge for most superalloys includes a AND R E C Y C L E D
Unlike the trade in ferrous scrap, which is closely moni-
J.J. deBARBADILLO is Department Manager, Processing Research tored, very little quantitative information on the generation
and Development, Inco Alloy Products Company, Research Center, and use of superalloy scrap is available. An estimate of
Suffern, NY 10901. 10 pct scrap recycled is frequently quoted in the literature.
This paper is based on a presentation made at a symposium on
"Metallurgical Implications of Recycling" held at the Louisville meeting
This figure, which was derived from United States Bureau
of The Metallurgical Society of AIME, October 14, 1981, under the spon- of Mines data, 2 actually refers to the percentage of domestic
sorship of the TMS Non-Ferrous Metals Committee. demand for chromium which is supplied by purchased

ISSN 0360-2133/83/0311-0329500.75/0
METALLURGICAL TRANSACTIONS A 9 1983 AMERICAN SOCIETY FOR METALS AND VOLUME 14A, MARCH 1983--329
THE METALLURGICAL SOCIETY OF AIME
scrap. This gives a somewhat misleading picture of the for this study was greatly aided by efforts then underway
actual degree of recycling which is practiced by the super- at several companies to develop material balances for
alloy industry. Recognizing the potential value of critical their operations.
metals in superalloy scrap which is not recycled and the The original study considered six classes of alloys, two of
lack of accurate information on scrap recycling practices, which were not superalloys and are hence excluded from
the United States Bureau of Mines in 1978 initiated a de- this paper. The four pertinent classes and the range of com-
tailed survey of scrap recycling practices in the United position of the major elements are shown in Table I. Note
States. The following information is excerpted from reports that although cobalt-base alloys are included in this com-
of this survey.3'4'5 pilation, they constitute less than 10 pct of the total alloy
Although the primary goal of this study was to define the production. The principal types of scrap which are generated
amount of scrap generated and recycled, a direct accounting are shown in Table II. Some of these materials are generally
of the scrap flow was, for several reasons, deemed imprac- considered wastes since they are largely discarded and some
tical. Reliable industry-wide sources of statistical data on are downgraded by recycling to other industries.
scrap do not exist. Much superalloy scrap is recycled within
In addition to those shown in Table II, small quantities of
companies or by exclusive arrangements with scrap pro-
other forms of scrap are generated by specialized industries
cessors and refiners; the alloy classes and industry cate-
gories for existing data bases (raw material purchased, alloy or processes. Physical losses, although not listed, were also
produced, parts sold) overlap extensively. Consequently, considered in developing the overall materials balance.
These include loss of parts in service, service wastage
the approach taken was to construct a materials flow model
for the entire industry which would account for the origin (corrosion and erosion), and a variety of losses which could
be termed "inventory losses".
and destination of all metal units. All of the derived quan-
tifies were based on estimates of alloy produced, data which Although the types of scrap described above are generated
were considered to be the most reliable. Generation of data in various amounts by most alloy producers, the ability to

Table I. Rangeof Compositionof Alloy Classes Used in the Study (Curwick3)

Definition of Alloys 1'2
Quantity of Alloy
Major Alloying Elements Produced in 1976,
Alloy Class Ni Cr Co Fe Thousand Metric Tons (Mg)
Investment Cast
Nickel-base 50 to 75 5 to 20 0 to 20 0 to 20 7.9
Cobalt-base 0 to 30 20 to 30 40 to 65 0 to 5 2.6
Hardfacing
Nickel-base 65 to 85 5 to 20 0 to 2 0 to 5 1.1
Cobalt-base 0 to 5 25 to 35 50 to 65 0 to 5 2.6
Wrought Nickel-Base 45 to 80 < 1 to 35 0 to 20 0 to 20 40.83
Wrought Nickel-Iron 20 to 45 10 to 30 0 to 20 40 to 55 20.9
~Adaptedfrom Reference3.
2Individualalloysmay contain 1 pct or moreof additionalelementsincluding W, Mo, Cb, Ta, Hf, Ti, A1, Re and in case of hard facingalloysC, Si, and B.
3Includesa small amountof wrought cobalt-base.

Table II. Typesof Snperalloy Scrap Generated by the Aerospace Industry

Origin Type of Scrap Classification I Character Typical Contaminants
Home Solids P, MA Solid--many shapes <1 pct oxide
Turnings P, MA Chips 5 pct oil
Grindings P, MA Dust 20 pct grinding wheel debris
P, MA Swarf 30 pct oil
Slag cleanings MA Grit 20 pct slag
Furnace dust MA Dry dust Fully oxidized; <40 pct recoverable metal
Mill scale MA Flaky grit Fully oxidized; 5 pct oil or water
Prompt Industrial Solids P Punchings, parts < 1 pct oxide
Turnings P, MA Chips 5 pct oil
Grindings P, MA Sludge Wet or oily with grinding wheel debris
ECM and pickle sludge MA Wet sludges >80 pct water, salts, or lime
Obsolete Solids P, MA Many shapes May be oxidized, coated, or contaminated
~P - - pedigree
MA-- mixed alloy (does not conformto a standard alloy specification)

330--VOLUME14A,MARCH1983 METALLURGICALTRANSACTIONSA
reuse them and hence their ultimate disposition varies con-
siderably throughout the industry. This will be discussed in
succeeding sections. A general summary showing the dispo-
sition of various types of superalloy scrap is shown in
Table III. There is considerable overlap in disposal methods
from one company to another so that material considered
waste by one company may be recycled directly to the alloy
melt charge by another.
The primary tool used for developing the survey results
is the production model based on a mass balance approach.
A separate materials flow diagram was generated for each
of the alloy classes defined in Table I. A detailed discussion
of the procedures, assumptions, and sources of data is
presented in Reference 3. A simplified composite flow dia-
PRII
gram is shown in Figure 1 to illustrate general materials
flow and overall quantifies of scrap generated in the year
surveyed (1976). The quantity of scrap in the various catego-
ries and its disposition are shown in Table IV. The informa-
tion in this table highlights some of the observations from
Table III as it shows that most solid scrap was remelted ~3CFINI
although some was downgraded or exported. Turnings,
grindings, and mixed scrap (skulls, slag cleanings, oxides)
were largely downgraded. A note of caution is necessary;
substantial progress in recycling of some lower grade OBSOLETE. . . . . . . . . . . .
materials has been made since 1976. For example, some SCRAP
companies now recycle all of their grindings; however,
Fig. 1--Materials flow diagram for the 4 alloy classes (Curwick4).
industry-wide the proportion of low quality material re-
cycled is still small.
In particular, obsolete solid scrap was not efficiently re-
cycled as less than half of this material was remelted by
superaUoy producers. The significance of the downgrading Table IV. Quantitative Summary of Types of Scrap
of scrap is in the role it could play in offsetting the amount Generated and Disposition in 1976~ (Curwick 3)
of primary raw materials required by the superalloy in-
dustry. In terms of contained elements, in 1976 downgraded Quantity, Thousand Metric Tons (Mg)
scrap contained approximately 10 pct of the tungsten, 7 pct Solids Turnings Grindings Mixed Waste Total
of the columbium and tantalum, 15 pct of the nickel, and Remelted 77.9 5.1 0 0 0 83.0
30 pct of the cobalt consumed as primary metals by the Exported 9.0 0 0 0 0 9.0
United States. While most of the nickel and much of the Downgraded 9.9 13.2 9.2 1.4 0 33.7
cobalt was recovered for reuse by the steel and cast iron Lost 2.5 0 0 0 5.5 8.0
industries, most of the other contained metals were lost. 1DataderivedfromReference4
Hence, one conclusion of the Bureau of Mines Report was
that further improvements in recycling efficiency of super-
alloy scrap is highly desirable.

Table III. Disposition of Principal Types of Superalloy Scrap

Remeltedby Same Industry

Direct reuse in new alloy melt Pedigree solid scrap (all types)
Preliminary refining or processing prior to reuse (master melt) Turnings, mixed solid scrap, skulls, slag cleanings
Downgraded to Lower Value Use
Direct use by alloy steel stainless steel or cast iron melters Pedigree and mixed solid scrap turnings, skulls, grindings
Partial refining for use in secondary alloy product Turnings, grindings, skulls, slag cleanings, furnace dusts, mill scale
Refined by primary metal producer Furnace dusts, sludges, mill scale
Lost Material
Physical service loss Components
Inventory loss Components and scrap
Service wastage Corrosion and erosion products
Deliberate disposal Sludges, furnace dusts
Exports/Imports (remelted or downgraded)

METALLURGICAL TRANSACTIONS A VOLUME 14A, MARCH 1983--331

 IlL SCRAP PROCESSING AND TRADE Obviously, sophisticated and costly procedures are not
justified for downgraded scrap, and very low grade scrap
The scrap metal recycling industry fulfills an essential and waste receive rather limited treatment. However, as
role in our economy and in the superalloy field has a interest in recycling these low grade materials within the
major share of the responsibility for producing critical superalloy industry grows, the need for improved sam-
alloys at minimum cost. The special raw material needs of piing, analysis, and processing will develop. Efficient
the superalloy producers have spawned a scrap processing handling of these materials will present some challenging
industry with technology consistent with the quality re- research problems.
quired. Cremisio and Wasserman provided a detailed
description of procedures used for identifying, sorting,
cleaning, and packaging scrap suitable for use by superalloy
melters .6The participants in this business must work closely IV. USE OF SCRAP IN ALLOY MELTING
with and fully understand the materials requirements and To achieve the combination of properties required in to-
processing capabilities of the alloy melters. Cremisio and day's superalloys, advanced melting and refining techniques
Wasserman identified three important considerations for are used in conjunction with careful selection and control of
scrap processors: raw materials and processing procedures. General dis-
cussions of practices and equipment used in this industry
1. "Careful identification of the alloy type including veri-
are contained in References 7 to 12. Numerous articles on
fication of compliance with trace element specifications."
specific processes are contained in the proceedings of the
2. "Absolute cleanliness and freedom from volatile sub-
stances and inclusion-forming ingredients." annual Vacuum Metallurgy and Electric Furnace Confer-
3. "Size, form, and physical distribution of the prepared ences, the biennial International Conference on Electroslag
Remelting, the quadrennial International Supperalloy Sym-
scrap to enhance optimum packing factors for the type of
posium, and various technical journals.
metal involved."
Although virtually every commercially available melting
To meet these requirements, scrap processors have had to process is applied to superalloys, the processes and the
employ instrumental chemical analysis including emission alloys they produce can, for convenience, be divided into
and atomic absorption spectroscopy to analyze impurity ele- the broad categories of air-melting and vacuum-melting.
ments to the parts per million level and portable X-ray The former process is used primarily for wrought alloys
fluorescent analyzers which facilitate hand sorting of in- used for plate, sheet, bar tube, and forging stock or for
dividual pieces of scrap, as well as new cleaning and producing master alloys for subsequent remelting by the
packaging techniques. vacuum processes. The vacuum process is used for complex
Scrap dealers generally identify superalloy scrap accord- alloys which contain reactive elements or which must meet
ing to three broad categories: vacuum-melting grade, air- the highest quality requirements. These alloys may be used
melting grade, and refinery grade. Vacuum-melting quality for investment casting, production of powders for sub-
material must be clean, certified free of tramp elements not sequent consolidation, or for conventional wrought shapes.
tolerated in superalloys, and identified according to specific The refining capabilities and hence the ability to use scrap
alloy. This premium material is normally sold for a fixed of these two melting systems are very different.
percentage (75 to 90 pct) of the current market price of all The air-melting approach as it is most commonly applied
the contained elements. Recovery of major elements may is comprised of two principal units, an electric arc furnace
approach 100 pct. Air-melting grade scrap may contain for melting and an argon-oxygen decarburization vessel
some oxide scale and some impurities which can be re- (AOD) for refining. Scrap and primary metals are melted as
moved in air melting and will be identified according to rapidly as possible in the electric furnace. A limited amount
alloy class. This material sells for a percentage ( - 5 0 pct) of of refining is possible in this furnace through oxygen lancing
the current market price of the contained metals. Recovery and slag control; however, most alloy producers prefer to
of elements such as nickel and cobalt exceeds 90 pct, but exploit the efficient melting capabilities of the furnace and
easily oxidizable elements such as aluminum, titanium, col- do their refining in the AOD vessel. The molten metal is
umbium, and tantalum are largely lost to the slag. The third transferred, sometimes with slag, to the AOD vessel where
form of scrap, refinery grade, is more loosely defined. This argon and oxygen are blown through tuyeres in the bottom
material may contain large quantities of oxygen, sulfur, of the vessel. Bath turbulence during the oxygen blow pro-
carbon, or other contaminants, rarely conforms to an alloy vides excellent mixing and rapid reaction kinetics. Oxi-
classification, and is often in a physical form incompatible dation of impurities and degassing are readily achieved. The
with alloy melting. Much of this scrap requires special and turbulent mixing can be used in conjunction with special
often expensive processing before it can be used at all. For additives to obtain further refining.
example, particulate materials may be agglomerated or pel- AOD technology is still evolving and producers continue
letized with water or binders, or mechanically briquetted. to advance their abilities to exploit its advantages. This
Turnings are deoiled thermally or with solvents and then technology has made it possible to produce a better quality
crushed to compact them. Wet sludges must be dried and product from lower grade scrap materials by effectively
pelletized. The cost of this processing is considered when removing many oxidizable impurity elements commonly
the value of a particular scrap material is defined. The value present in scrap. For example, in the process of oxygen
of these materials is based on a fraction of the recoverable blowing, carbon can be reduced to less than 0.01 pct. Mn,
elements, often only nickel, and may range from 50 pct of Si, and the very reactive elements, such as A1, Ti, Zr, etc.,
market price to essentially negative (sludges where the sup- are oxidized and partitioned to the slag. Some elements such
plier must pay drying and transportion costs). as Bi, Pb, Te, and Se are volatilized and can be reduced to

332--VOLUME 14A, MARCH 1983 METALLURGICAL TRANSACTIONS A

levels of less than 1 ppm. 13 Sulfur removal occurs by vola- As

oI
tilization and combination with a basic slag. Dissolved gases
Sn
such as H2 and N2 are removed by the purging action of
the argon.
Although the AOD process removes many impurities,
some elements cannot be effectively controlled. For ex- Sb
ample, Cu, Sn, and Fe, which are common impurities in
scrap, are not removed. Reduction of phosphorus content
depends on the alloy and operating practice. The type of
scrap which can be processed efficiently is limited by the "S
very poor recovery of valuable elements such as Cb, Ta,
o
and V. Ultimately the direct use of AOD processed materials
is limited by tolerance for oxygen. Oxygen is controlled by v
Q_

addition of strong deoxidizers such as Mg, Zr, or Ca, but the Z .01
O
inclusions which are always formed cannot be tolerated for F--
many superalloy applications. Se
I-
In principle, a plant operating an AOD melt shop should Z
uJ
be able to accept virtually any combination of scrap to
Z
produce its alloys, providing the level of noncontrollable 0
L2
impurities such as Cu is kept under control. In fact, such
plants do consume a great deal of scrap; however, good plant
practice places additional restrictions on the amount and
type of scrap accepted. In-house scrap, which noted earlier .001
is always preferred, may provide 50 pct of raw material To
needs. Some of this scrap will be of relatively low grade
(partially oxidized or particulate) or have residual impurity
content such that no additional low grade scrap is acceptable Pb
despite its low price. In addition, the margin for com-
position adjustment requires that some primary metal be 0 20 40 60 80 I00 120 140
used, some of which also contain significant quantities of TIME (rain.)
impurities which must be removed. As a result, despite its Fig. 2--Evaporation of elements from 80 pct nickel-20 pct chromium
refining capabilities and the advent of computer aided selec- melts under vacuum at 1566 ~ (Turillonl4).
tion of charge materials, the air melt process for superalloys
cannot directly consume all of the obsolete scrap which is
potentially available for recycling. link between scrap processor and alloy producer has arisen.
Vacuum processing is used extensively for the production This is the position of the master melt producer whose role
of superalloys for rotating parts in gas turbine engines as is to prepare a certified master alloy for melting and final
well as many other parts requiring a high degree of service refining by the alloy producer. The master melt producer
integrity. This classification includes a wide range of may use air-melting or a combination of air- and vacuum-
processes including vacuum-induction melting (VIM), melting and often has the capacity to consume a consid-
vacuum-arc remelting (VAR), electroflux remelting (EFR), erable amount of scrap. This industry has evolved to the
electron beam melting (EBM), and others. Normally several point where much of the scrap generated by vacuum melters
processes are used in tandem and the products include in- is now directly recycled and refined through the master melt
vestment castings, powders, and ingots cast under the producers. In some cases a single company may perform
protection of slag or vacuum. The alloys frequently con- both functions while in others separate companies are in-
tain reactive metals such as A1, Ti, Ta, and Hf in concen- volved. This system has relieved the producers of vacuum-
trations of 1 pct or more. The melting and refining units melted alloys from some of their dependence on primary
may be as large as those used for air melting, but they have metals and has greatly increased the efficiency of scrap
a limited ability physically to handle scrap and are unable to recycling. Some relatively low grade materials such as turn-
remove effectively some common impurities. Normally, ings, dry grindings, ladle skulls, and slag cleanings are
scrap would be limited to large pieces of solid scrap or processed in this way. However, it must be emphasized that
master alloy ingots of carefully controlled composition in- due to the number of complex alloys produced and the very
corporated in the initial melt charge. The great utility of the stringent purity specifications, the extent of recycling is
vacuum melting approach lies in its ability for producing currently lower than for the air-melt alloy producers.
alloys with extremely low and predictable levels of impurity The form of scrap is of great importance with solid scrap
elements. The selective removal of impurities by vacuum being the most desirable. Very fine materials are difficult to
evaporation at 1566 ~ is illustrated in Figure 2.14 Note that handle and melt and have higher losses. Gross contamina-
Pb and Bi are readily removed but Sn is virtually unaffected. tion with oxides or carbon presents such severe problems to
However, these processes are not well suited to remove large the production of superalloys that such materials are either
quantities of impurities, and consequently the charge raw air melted or downgraded. Another problem is control of the
material must be very carefully monitored and controlled. major alloying elements. Superalloys frequently contain six
As the vacuum-melting industry developed and became and sometimes ten or more major alloying additions. It is
ever more specialized, a new business which forms a key impractical to control scrap recycling to the extent that scrap

METALLURGICAL TRANSACTIONS A VOLUME 14A, MARCH 1983--333

is used only in the same alloy. All superalloy melters have grain boundary segregation, nonmetallic inclusions, phase
developed an internal sequence of "segregation" so that stability, and solidification structure. Each of these will be
scrap of simpler alloys is redirected for use in more complex discussed in general terms and then illustrated by specific
alloys. For example, scrap from alloy 713 (columbium but examples. It should be emphasized that the following sec-
no cobalt) and IN-100 (cobalt but no columbium) can be tions present only a very brief overview of a subject which
used to make IN-738 which contains both elements. Cross has been a main focus of superalloy research for more than
blending obviously approaches a limit beyond which the 30 years. A general review of the effect of impurities and
scrap must be downgraded or refined before it can be re- trace elements in superalloys published by Holt and Wallace
used. Finally, the alloy producers must contend with the host in 1976 contains 145 references .24 The pace of research on
of intentional and unintentional trace elements which have the subject has, if anything, accelerated since that time. The
such a profound effect on the processing and properties of proceedings of a 1980 British Conference on "Residuals,
these alloys. The following section deals with the effect Additives and Materials Properties" contains a number of
of these elements. papers dealing with superalloys.25
Grain boundary segregation has been introduced in the
preceding section. Many elements have been shown to seg-
V. EFFECT OF RESIDUAL ELEMENTS regate and since such small amounts can be so detrimental
to a host of properties, a great deal of work has been devoted
As a class, nickel-base superalloys have predominantly to the problem. The agents which cause this phenomenon
single-phase face-centered-cubic crystal structure on solidi- may derive from impurities in the scrap or primary metals,
fication and they undergo no solid-state transformations. from the furnace refractories, or occasionally from an excess
These materials may be strengthened by solid solution of or synergism between desired alloying elements. Specific
alloying elements or through precipitation of alloy carbides examples of grain boundary segregation problems will be
or intermetallic compounds. Many complex alloys utilize presented in a subsequent section.
all three strengthening mechanisms and may have primary The mechanical properties of nickel-base superalloys,
carbide or intermetallic compounds present in the as- whether wrought, cast, or powder processed, are often very
cast condition. sensitive to the size and number of nonmetallic inclusions.
There are a number of excellent reviews which describe Oxide, nitride, and carbide inclusions are all considered
the microstructure of these alloys and its control through detrimental; however, oxides are most frequently encoun-
alloying and thermal processing. 14-2~ Although the basic tered because excess oxygen is always available. The inclu-
metallurgical features are common to many alloy systems, sions which are of primary concern are known as exogenous
nickel-base superalloys have some unique characteristics inclusions and tend to be large (sometimes millimeters in
which make them especially sensitive to the presence of diameter). Exogenous inclusions are usually derived from
impurities. These include the formation of low melting point the furnace or ladle refractories or agglomerated oxidized
compounds with many elements, a limited solid solubility alloy constituents. They should be distinguished from the
which decreases sharply with temperature, and a marked smaller, less harmful, endogenous inclusions which form
tendency for grain boundary segregation and wetting. The during alloy solidification.
effect of sulfur was first recognized by Fleitman in 187921 Superalloys are especially prone to this inclusion problem
and the mechanism of embrittlement defined by Merica in because they frequently contain large quantities of reactive
1925.22 The solubility of sulfur in liquid nickel is greater elements. These elements are capable of reducing many
than 50 pct. The solubility of sulfur in solid nickel is not furnace refractories, thereby introducing other unwanted
known with certainty but is believed to be less than elements into the alloy. In lower grade forms of scrap the
0.001 pet.23 The eutectic between nickel and the precipi- reactive elements may be already oxidized. Recovery of
tating Ni3S2 phase melts at 635 ~ or half of the homologous these elements during melting is very low because they
temperature of nickel. Hence, it is hardly surprising that a combine readily with the slag. However, even in vacuum
small quantity of sulfur can cause severe hot shortness in melting of clean scrap, residual oxygen in the scrap and
nickel alloys. Fortunately, the effect of small amounts of melting system can produce a significant amount of "oxide
sulfur, a ubiquitous impurity in nickel, can be controlled by dirt". Hafnium and tantalum are especially troublesome be-
making a small addition of magnesium, a powerful sulfur cause their oxides are dense and do not readily separate from
scavenger. Some other impurities cannot be so readily neu- the liquid. Exogenous inclusions affect both mechanical
tralized. In addition, most nickel-base alloys are specifically properties and processing of superalloys. Large inclusions
designed to have high strength at elevated temperatures. act as sites for crack initiation. Degradation of fracture
Some advanced nickel-base alloys may contain more than toughness in rotating components of gas turbine engines
50 vol pet of hard phases. Even solid solution strengthened due to exogenous inclusions is a major concern for engine
alloys may exhibit high flow stress and considerable re- builders. A great deal of effort is currently being devoted to
sistance to the processes of grain boundary migration and nondestructive inspection and to techniques such as filtering
recrystallization. Finally, the incipient melting point of of pouring streams. Processing problems attributed to ex-
many nickel-base alloys is low because of the high alloy ogenous inclusions include blockage of pouring nozzles and
content. These factors combine to make wrought and cast misruns and surface defects in complex castings.
nickel-base alloys more sensitive to the presence of em- The solidification structure of nickel-base superalloys can
brittling conditions than other alloy systems. also be affected by impurity content. This is most important
There are at least four ways in which residual elements for investment castings and powder metallurgy products,
derived from scrap can influence superalloy properties: but adverse solidification structures may be encountered in

334--VOLUME 14A, MARCH 1983 METALLURGICAL TRANSACTIONS A

wrought products as well. Both macroscopic and micro- because of preferential partitioning of elements to the differ-
scopic segregation of alloying elements occur during solidi- ent phases. Additional elements such as Zr, B, and Mg can
fication of nickel-base alloys. Segregation of this type may influence the morphology of grain boundary carbide pre-
lead to local concentration of impurities and thereby con- cipitation, while strong carbide formers such as Hf, Ta, and
tribute to a number of secondary processing defects. 26'27 In Cb alter the identity and size of carbides. 16 All of the dis-
addition, impurity elements can act directly to affect solidi- solved elements affect the electron vacancy number and
fication structure. For example, Antony and Radavich re- hence the tendency to form TCP phases. Elements such as
lated the percentage of microporosity and the creep strength manganese, iron, and especially silicon which have rela-
of an experimental investment casting alloy to the boron and tively minor effects on other properties can lead to serious
zirconium content of the alloy.2s The degree of porosity was phase instability problems in superalloys. 15 Strict control of
related to the effect of the elements on the liquidus-solidus alloy composition is clearly a prerequisite to the micro-
range of the alloy. Although the total amount and change in structural reproducibility and ultimately to component in-
porosity was small (<0.5 pct), the effect on stress rupture tegrity. To a large extent, however, this is incompatible with
life was very large. Subsequent work has indicated that the the desire to use the maximum amount of alloy scrap with
effect of residual elements on microporosity may be quite the attendant sampling and analysis uncertainties.
general. 29 During the past decade new technologies for A review of the effect of residual elements would not be
production of advanced engine parts have evolved. These complete without a mention of fusion welding. Although
include hot isostatic pressing (HIP) of powders produced most complex cast nickel-base turbine blade materials are
by gas atomization, melt spinning, and recently rapid very difficult to weld, other nickel-base alloys in commer-
solidification technology. The powder shape, surface cial use today are routinely welded by processes essentially
composition, and precipitation morphology are all strongly the same as those used for stainless steels.35 Weldability of
influenced by impurity content. 3~ Advanced casting superalloys may be affected by all four of the mechanisms
technologies include directional solidification, single mentioned in the preceding sections. Hence, the presence of
crystal technology, ESR cast-to-shape process, and others. nonmetallic inclusions leads to slag defects, blow holes, arc
Grain shape and orientation are sensitive to composition, sputter, and fusion line cracking. Residual elements may
and hence it is very likely that control of residual element alter the weld pool shape, grain structure, and alloy distribu-
content will prove important to reproducible development tion in the fusion zone. Impurities can alter the micro-
of properties. structure and segregate to grain boundaries in both the weld
The final way that impurities can influence superalloy metal and the heat-affected zone. These processes contrib-
properties is through modification of precipitation reactions. ute to cracking during the welding process and degradation
Control of these complex reactions is vital to the devel- of mechanical and physical properties of the fabricated com-
opment of their unique properties, and consequently most ponent. Such behavior is not unique to nickel-base super-
superalloy compositions are delicately balanced. Precipi- alloys, but these alloys are considered more difficult to weld
tating phases include a variety of different carbides and and less tolerant of impurities than many other alloys sys-
intermetallic compounds. The amount, composition, and tems. Any generalization concerning the effect of impurities
morphology and stability of the phases during service expo- is dangerous because there are so many welding processes
sure is very sensitive to the thermal and mechanical history, and alloys. One attempt was made by G.R. Pease 36 who
base composition, impurity content, and even the solidi- listed a number of common elements according to their
fication procedure. A particular alloy may show a wide effect on inert-gas, shielded-arc welding processes. This list
range of strength and ductility depending on the precipitate which was subsequently modified by Linnert 37 is shown in
morphology. A number of reviews of precipitation in super- Table V. This list should be considered only to be an indi-
alloys have been published) 5'19'2~As the applications for cator of effects which might be encountered at levels of
superaUoys became established, it was recognized that some elements commonly encountered in commercial alloys. It is
alloys were subject to microstructural instability during not quantitative and does not consider the possibility of
elevated temperature service. Massive precipitation of inter- over-treatment in the case of elements such as magnesium.
metallic compounds which was observed in a variety of Finally, one might arrive at a different ranking if electron
alloys was sometimes associated with loss of ductility and beam, laser, or resistance weldments were assessed.
other properties. Concentrated research on the problem led
to the identification of a series of phases termed topo-
logically closed packed (TCP) which may occur in alloys
with compositions which put them within range having Table V. Effect of Some Commonly Present
critical values of the electron vacancy number (Nv). One of Elements on the Weldability of Nickel
these phases, known as sigma, occurs in many other alloy and Nickel Alloys (Linnert37) 1
systems, but its effect on properties is best documented in
Beneficial No Particular Variable Harmful
superalloys. The formation of these TCP phases was inten- Effects Influence Influence Effects
sively studied and formed the theme of the First Interna-
tional Superalloy Symposium. 33 Sims presents an excellent Cb Cu AI Pb
review of the formation, structure, and effect of these Mg Cr Ti S
Mn Fe C P
phases) 4 Because the compositional balance of superalloys Co Mo Zr
is so delicate, surprisingly small amounts of some impurities Si B
can have significant effects on reactions. This is especially
~Listproposedby Pease36and modifiedby Linnert3v
so in alloys having high volume fractions of precipitates

METALLURGICALTRANSACTIONSA VOLUME14A,MARCH1983--335
 VI. GRAIN BOUNDARY SEGREGATION
COLD RED SHORT I HOT [ INCIPIENT
MALLEABILITY ZONE MALLEABILITY MELTING
The preceding section presented a general overview of ZONE ZONE ZONE
problems which might be encountered in using recycled
scrap for the manufacture of nickel-base superalloys. Those
readers with a broad familiarity with metals processing will
recognize that impurity effects on processing, phase sta-
bility, and solidification structures are common to the pro-
duction of many alloys. Grain boundary problems are also
encountered in other alloys, but nowhere is the problem
so acute as with superalloys. This is because superalloys
(by definition) were developed for service at elevated tem-
peratures where grain boundaries provide the predominant
fracture path and because nickel alloys are more sensitive to
embrittlement by trace amounts of impurities. It thus seems
appropriate to consider the relationships between grain
boundary segregation and tramp elements in greater detail. I I I
500 I000 1500
The important effect of sulfur on the elevated temperature TEMPERATURE (*C)
mechanical properties of nickel-base alloys was mentioned
earlier. The embrittling effect of sulfur in commercial alloys Fig. 3 - - Effect of temperature, minor constituents, and other factors on the
malleability of nickel (Bieber39).
was alleviated by making small additions of magnesium,
calcium, or rare earth elements. However, in commercial
practice reactive metal treatment sometimes caused more
severe embrittlement. It was also recognized that elements material specifications such as SAE Aerospace Materials
which could not be controlled by reactive metal additions Specification (AMS) 2280 (Table VI). This specification
were also deleterious. The general principles of solute segre- covers only cast alloys for military aircraft applications;
gation to grain boundaries and their effect on properties however, it has been widely applied for guidance in writing
were already well established38 when Bieber and Decker raw materials specifications.4~Detection of elements in the
published their review of the effect of minor constituents on tenths of ppm range has stimulated several innovations in
the malleability of nickel alloys .39However, this latter paper sampling and chemical analysis. Note that the elements
provided the first comprehensive overview of the problem identified in the AMS 2280 specification are primarily those
which was to serve as a foundation for developing melt falling in classes 1 and 2 of the Bieber and Decker system
practices and raw material specifications for commercial (see Figure 4).
nickel alloys. The elements in class 3, including B, Ca, Mg, Zr, and the
The effect of temperature, minor constituents, and other rare earth metals, are especially interesting because they can
factors on the malleability of nickel was explained by Bieber produce both beneficial or harmful effects depending on the
and Decker with the schematic diagram shown in Figure 3. amount added, the method of addition, the purity of the base
Most malleability problems are experienced within the alloy, melting process, and thermal-mechanical processing
"red-short" zone 500 ~ to 900 ~ indicated on the diagram. procedures. There has been a considerable amount of specu-
High purity nickel is seen to have ductility which increases lation as to the mechanism of the beneficial effects of these
uniformly with temperature (A) while commercially pure elements. Clearly, Mg, Ca, and rare earth metals combine
nickel has a distinct minimum of ductility in the red-short with sulfur and oxygen and thereby neutralize the harmful
zone (B). Material with a critical quantity of a known em- effects of these elements. In a similar manner zirconium is
brittling agent (e.g., 0.01 pct sulfur) may have almost no generally believed to tie up residual nitrogen. It has also
ductility at all in the red-short zone (C). The addition of a been suggested from thermodynamic data that rare earth
malleabilizer such as magnesium to either the commercially elements may combine with lead and other "white metals"
pure or embrittled alloy restores the ductility to a level in nickel alloys.41 Other beneficial effects of malleabilizers
approaching that of high purity alloys (D). may be more subtle such as preferential segregation to grain
As a next step, Bieber and Decker reviewed the elements boundary sites. It has been speculated that boron and zir-
of the periodic table and classified them into five categories conium neutralize harmful effects of lead by exclusion or
according to their effect on the malleability of nickel. This alteration of the grain boundary energy rather than by com-
classification is shown in Figure 4. Although at first glance pound formation.42
this analysis might seem academic, the quantities of im- The loss of ductility caused by an excess of group 3
purities such as bismuth and tellurium are so small that they elements can be dramatic and unpredictable. This effect is
are of practical concern for selection of raw materials. This illustrated in Figure 5 where a nickel-base alloy containing
concern was especially valid for the use of scrap from 0.02 pct S was treated with different amounts of calcium 43
sources where some embrittling elements are widely em- When no calcium was added, hot shortness was noted in
ployed (soldered electronic devices, coated or brazed engine upset forgings. Grain boundary films of NiaS2 were identi-
parts, etc.). fied. Good forgeability was seen when the calcium content
The recognition of the broad range of elements which can ranged from 0.017 pct to 0.025 pct. However, severe hot
embrittle the nickel alloys and the wide number of potential tearing occurred at 0.035 pct calcium. This behavior was
sources of impurities led to the emergence of stringent raw associated with the appearance of filmy low-melting-point

336--VOLUME 14A, MARCH 1983 METALLURGICAL TRANSACTIONS A

 IA IIA Ilia IV B VB VIB

D o e s not alloy with molten nickel at 1 atm or less,

Harmful to malleability, even with minute additions.
Beneficial to malleability with minute additions, but harmful with larger additions still less than 0.5 percent.
Relatively neutral with minute additions, but harmful with larger additions Still less than 0.5 percent.
Soluble above 0,5 percent; can be used as major alloying element.

Fig. 4--Periodic table showing behavior of elements in commercially pure nickel (Bieber39).

Table VI. Essential Features of AMS 2280 none in alloys melted from scrap. The cause was ultimately
Specification for the Control of Trace Elements traced to minor differences in the residual nitrogen content
in Nickel Alloy Castings after Cremisio4~ of the two alloys.
Element Limit (ppm) Element Limit (ppm) The embrittling effects of sulfur and the group 3 metals
such as calcium are almost certainly directly caused by the
Lead 5.0 Gold 50.0 ~ formation of intermetallic compounds which may be liquid
Bismuth 0.5 Indium 50.0 ~ at mechanical working or service temperatures. The low
Selenium 3.0 Mercury 50.0 ~
Tellurium 0.5 Potassium 50.0 ~ melting point sulfide phase Ni3Sz has been identified at grain
Thallium 5.0 Silver 50.0 ~ boundaries in alloys containing as little as I0 ppm sulfur.~
Antimony 50.01 Sodium 50.0 ~ Many other embrittling elements, including lead and bis-
Arsenic 50.0 ~ Uranium 50.0 ~ muth, do not form compounds with nickel and dramatic
Cadmium 50.0 ~ Zinc 50.0 ~ embrittlement is often observed at impurity contents of a
Gallium 50.01 Thorium 50.01 few parts per million. Many quantitative studies have been
Germanium 50.01 Tin 50.05 published in recent years showing the effects of tramp ele-
5400 ppm maximum of the sum of all elements in antimony-tin series ments on properties of nickel alloys.24 Hot workability is
adversely affected, but concentrations of tramp elements
great enough to cause hot working problems are seldom
nickel-calcium intermetallic compounds at grain boundaries encountered in commercial practice. However, essentially
when uncombined calcium was present in the alloy. Similar all elevated temperature mechanical properties are affected
effects have been identified for magnesium 43 and lanthanum at very low tramp element concentrations. This includes hot
and zirconium.44 tensile ductility, stress rupture life and ductility, thermal and
In the commercial production of nickel-base alloys, the mechanical fatigue, and fracture toughness. Data drawn
precise amount of reactive element which should be added from the work of Thomas and Gibbons 47 showing the effect
to neutralize harmful impurities completely without produc- of Pb, Sn, Se, Te, and Ti on stress rupture life of
ing further embrittlement is very difficult to predict. In the NIMONIC* alloy 105 are typical (Figure 6).
case of calcium, the critical amount which would be needed
*Trademark of the Inco family of companies.
to combine with sulfur would be a function of both oxygen
and sulfur contents of the alloy. The amount of free nitrogen The mechanism for embrittlement by the white metals has
may also be a factor as some calcium can be associated with not been fully established. Many of these metals melt at very
nitfides 45 The effect of nitrogen was illustrated by the ex- low temperatures, and since they do not form compounds
perience of one alloy manufacturer which encountered hot with nickel it has been suggested that liquid films may coat
shortness in alloys melted from high purity raw materials but the grain boundaries as is clearly the case for the preceding

METALLURGICAL TRANSACTIONS A VOLUME 14A, MARCH 1983--337

group of elements. However, optical and electron metal- cast superalloys.4s A concentration profile for bismuth is
lography has not identified such grain boundary phases. shown in Figure 7. The bismuth concentration in the bound-
Auger electron spectroscopy (AES) has provided direct evi- ary is about 5 at. pct in an alloy with a bulk bismuth concen-
dence that bismuth and tellurium can preferentially segre- tration of 22 ppm. Lead was not segregated to the boundary
gate to grain boundaries in embrittled samples of investment in this study, but other work indicates that lead can also
338--VOLUME 14A, MARCH 1983 METALLURGICAL TRANSACTIONS A
 120 I ' I 1 I i I i I
.....-- -
t ._.....'I may be a reduction in grain boundary interfacial energy
....-.- -
......-- - leading to boundary decohesion as discussed by A s a r o . 49
" . f "
TIN

100
VII. TRENDS AND CHALLENGES
FOR RECYCLING
80
Over the past ten years there has been great interest in the
subject of recycling, and considerable progress has been
made in the technology of recycling superalloy scrap. Scrap
LEAD
60 which was once routinely downgraded is now recycled di-
TELLURIUM rectly or indirectly. Procedures for identifying and sorting
scrap by processors and selecting charge materials by
40
Smooth Bor Specimens
melters have been upgraded. Operating experience with
Tested at 8150C, 350 MPo
both air- and vacuum-melting has made it possible re-
producibly to meet trace element specifications. Material
;'0
which was once considered waste is now treated for re-
covery of some or all of the valuable elements. Although
most of this material is downgraded, some is now recycled
0 I I t I I I 1 I t I I within the superalloy industry. There is no component of
0 20 40 60 80 I00 120 scrap or waste on which some research laboratory is not
CONCENTRATION, ppm working to develop a treatment method. A few nickel
Fig. 6--Stress rupture properties of Nimonic alloy 105 as a function of
alloy producers have virtually achieved their goal of zero
impurity content (Thomas47). metal loss by recycling all of their scrap and waste, al-
though industry-wide a gap still remains. The progress
is encouraging.
The motivation for these trends has been well publicized:
I I I I I (1) high and unstable prices of many metals, (2) concern
over availability of some strategic metals, (3) reduction of
total energy costs by using scrap, (4) difficulty in disposing
of wastes because of environmental laws, (5) changes in
BULK COMPOSITION
accounting procedures which alter the internal value of
2?. p p m BISMUTH
scrap, (6) greater competition from foreign superalloy pro-
ducers, and (7) the perception that conservation is essential
for a stable economy in a world with finite resources.
Recycling has advanced both in theory and practice to the
point where Nutting's vision of design of parts for complete
i- recycling seems achievable? ~ However, the superalloy in-
t,.)
I1 dustry has not been static in other areas such as in the
development of new alloys, processing methods, and fabri-
cated components. Freche and Ault have recently reviewed
Z this technology? ~ In general, but particularly in the gas tur-
0
F- bine industry, there has been a progressive movement
rr
k-
toward stronger and tougher alloys, longer service life, more
Z
t.=.l
severe chemical environments, higher operating tempera-
Z
tures, and lighter-weight structures. The alloy producers,
0 component manufacturers, and system fabricators have
made rapid strides to meet these challenges, and in most
instances the solutions have had the effect of making re-
cycling more difficult.
Some of these advances are really just extensions of his-
O torical trends within the industry while others, which will be
0 I0 20 30 40 50 60 discussed in more detail, are new and may have unforeseen
DISTANCE FROM FRACTURE (ANGSTROMS) future consequences. Some historical trends which affect
recycling are ever changing and more complex alloy types,
Fig. 7 - - B i s m u t h sputter-depth profiles for cast nickel (Walsh~),
more stringent trace element specifications, and the evo-
lution of more specialized melt practices. Each of these is
important; however, the effect on the industry is predictable
segregate. 47 The AES studies described above show that and has been discussed previously. What is less predictable
significant loss of ductility can occur at concentrations of is the development of new alloy production techniques, new
embrittling elements lower than that required to produce a fabrication concepts, and composite materials. These tech-
continuous grain boundary film or even a monolayer of the nologies have been under development for ten years or more
embrittling agent. Hence, the mechanism for embrittlement and are now being adopted for production. Their impact on

METALLURGICAL TRANSACTIONS A VOLUME 14A, MARCH 1983-- 339

recycling will not be fully appreciated until the end of the some of these coatings may have to be removed by selective
twentieth century. chemical pickling before recycling. The ultimate conse-
A cursory review of the proceedings of recent superalloy quences of this trend are not known, but broad industrial
conferences shows that great emphasis has been placed on application of coated or clad products would lead to greatly
a concept termed "near-net-shape-processing". Basically increased cost of scrap identification and processing and a
this means that alloy producers attempt to make a product as reduction of scrap value.
close to its final dimensions as possible. The motivation is The use of multialloy fabricated components in the gas
a very powerful one derived from the extraordinarily poor turbine engine continues to expand. Advancements in fric-
product yields once typical of this industry. The processing tion and fusion welding, brazing, and diffusion bonding
advances to meet this challenge take a number of forms: hot have resulted in some very remarkable and complex compo-
isostatic compaction of powdered metals, direct casting of nents. One component, a composite vane assembly used in
billets for precision closed die forging, use of investment an advanced gas turbine engine, is constructed from an
castings of ever increasing complexity, substitution of cen- oxide dispersion strengthened nickel-chromium alloy
trifugal castings for extruded tubes, direct rolling of strip (INCONEL* alloy MA 754) airfoil brazed with a high-
from powder, and many others. These processes impact *Trademark of the Inco family of companies.
recycling in three ways:
silicon nickel-base brazing alloy (AMS 4779) to a cast
(1) By their nature less scrap is generated. However, it cobalt-base alloy (MAR-M509**) sideband. Nickel-
is this home and prompt industrial scrap which is most
desirable for recycling. Ultimately the proportion of **Trademark of the Martin Marietta Corporation.
obsolete scrap in the overall superalloy scrap mix must chromium-molybdenum alloy (INCONEL alloy 625) baffle
increase significantly. and tubes are tack welded to the assembly. For certain
(2) The vacuum melting processes used for producing service conditions the entire assembly may be coated with
these alloys are less able to directly use scrap because of nickel-aluminide. The design and fabrication cost of this
limitations on shape and size of charge materials, need for component is apparently justified by improved engine per-
more precise compositional control, and difficulty in re- formance; but how is this part going to be recycled?
moving some common metallic impurities.
(3) The tolerance for common scrap derived impurities by
components produced by the new techniques remains to be VIII. NEEDS FOR RESEARCH
established. For instance, powder metal parts may be more IN SUPERALLOY RECYCLING
tolerant of white metal impurities (grain boundary dilution)
but less tolerant of oxide impurities (inclusions, particle The progress in recycling technology over the past decade
boundary precipitates). How this will affect recycling will clearly reflects the level of research and development effort
not be known for many years. devoted to the task by industry. Much of this effort will
continue for the immediate future. By the end of the decade
Even before these technologies had become commercially
the proportion of scrap in the basic melt charge will have
established, advancements in alloy processing technology
increased substantially and only a few contaminated, low
produced new materials with such promising properties that
metal content sludges will be discarded. The technical basis
interest in them was very keen. The more conventional
for this is already available and only implementation re-
powder technology spawned oxide dispersion strengthening
by the mechanical alloying processs2 and rapidly solidified mains. The real challenge will be to devise new procedures
metal powders;53 investment casting led to grain oriented to cope with the increasing quantity and complexity of obso-
lete scrap. As noted previously, new and more sophisticated
castings, single crystals, and directionally aligned eutectics.
identification and sorting methods may be needed along
While it is too early to tell whether any of these processes
will ever be used on such a wide scale as to constitute a with new techniques for disassembly of components. Com-
significant fraction of superalloy production, one can envi- panies making master melt alloys for the new melting and
sion some major impact on recycling practices. processing techniques may melt and refine large quantities
of obsolete scrap. Much more thought will have to be de-
Another major trend is toward the greater use of clad-
voted to the development of material and performance speci-
dings, coatings, brazes, and multialloy components. Al-
though these concepts are not new, it has recently been fications which provide systems reliability and yet allow for
flexibility for efficient use of raw materials: 5 In spite of this
reemphasized and is one of the major strategies identified by
progress in recycling scrap, there may always be some scrap
the United States government to increase America's ability
which is too complex or contaminated to reuse directly in
to respond to raw materials shortages :4 Claddings and coat-
superalloys and yet too valuable to downgrade. Complex
ings have been used routinely in specialized applications for
separation techniques employing operations from the extrac-
many years. However, production cost, coating integrity,
tive metallurgy field have been proposed to treat this type of
and other problems have precluded more widespread adop-
material and recover pure separated metals. A recent report
tion. But it is likely that a heavy emphasis on research and
published by the United States Bureau of Mines reviews
development will lead to new processing techniques and
work in this field: 6
much broader application of these technologies. The impli-
cations for recycling are significant. Many clad products are REFERENCES
difficult to recycle because the metals are incompatible with
1. Metals Handbook, 8th edition, T. Lyman, ed., ASM, Metals Park,
existing alloy production techniques. Coated products are OH, 1961, vol. 1, p. 37.
less of a problem because the coating is usually thin and thus 2. J.L. Morning: Mineral Facts and Problems, U.S. Bureau of Mines,
only a small fraction of the metal volume. Nevertheless, Washington, DC, 1976, p. 247.

340--VOLUME 14A, MARCH 1983 METALLURGICALTRANSACTIONSA

 3. L.R. Curwick, W.A. Petersen, and H.V. Makar: U.S. Bureau of Superalloys, Claitor's Publishing Division, Baton Rouge, LA,
Mines Information Circular 8821, 1980, p. 51. 1976, pp. 137-46.
4. L.R. Curwick, W.A. Petersen, and J.J. deBarbadillo: Superalloys 29. L. Ouichou, E Lavand, and G. Lesoult: SuperaUoys 80, op. cir.,
1980, Proceedings of the Fourth International Symposium on Super- pp. 235-44.
alloys, ASM, Metals Park, OH, 1980, pp. 21-30. 30. P.N. Ross and B.H. Kear: Rapid Solidification Processing,
5. J.J. deBarbadillo: Proceedings, U.S. Department of Commerce Proceedings of the International Conference, Claitor's Publishing
Public Workshop on Critical Materials Needs in the Aerospace Division, Baton Rouge, LA, 1978, pp. 278-85.
Industry, National Bureau of Standards NBSIR 81-2305, July 1981, 31. C. Aubin, J.H. Davidson, and J.P. Trottier: Superalloys-1980,
pp. T18, 1-13. op. cit., pp. 345-54.
6. R.S. Cremisio and L.M. Wasserman: Proceedings of the 1977 32. R.D. Field, A.R. Cox, and H. L. Fraser: ibid, pp. 439-77.
Vacuum Metallurgy Conference, Science Press, Princeton, NJ, 1977, 33. International Symposium on Structural Stability in Superalloys, Seven
pp. 353-88. Springs, PA, 1968, 618 p.
7. R.S. Cremisio: The Superalloys, C.T. Sims and W. C. Hagel, eds., 34. C.T. Sims: The Superalloys, op. cit., pp. 259-84.
John Wiley and Sons, New York, NY, 1972, pp. 373-401. 35. W. Yeniscavich: The Superalloys, op. cit., pp. 509-32.
8. W.E. Jones: Vacuum Metallurgy, R. E Bunshah, ed., Reinhold 36. G.R. Pease: Welding Journal, 1957, vol. 36, pp. 330s-334s.
Publishing Corp., New York, NY, 1958, pp. 189-217. 37. G.E. Linnert: Effects of Minor Elements on the Weldability of High-
9. R.C. Hambleton: The Nimonic Alloys, second edition, W. Betteridge Nickel Alloys, Welding Research Council, New York, NY, 1964,
and J. Heslop, Edward Arnold Ltd., London, 1974, pp. 115-28. pp. 139-44.
10. D.J. Hopkins: ibid, pp. 324-41. 38. D. McLean: Grain Boundaries in Metals, Oxford University Press,
11. Electroslag Remelting and Plasma Arc Melting, National Materials London, 1957, pp. 116-48.
Advisory Board Pub. NMAB-324, National Academy of Sciences, 39. C. Bieber and R.F. Decker: Trans. TMS-AIME, 1961, vol. 221,
Washington, DC, 1976. pp. 629-36.
12. R. Schlatter: Superalloys-Processing, Proceedings of Second Inter- 40. R.S. Cremisio, L.W. Lherbier, C.S. Freer, E.W. Ross, and
national Conference, Metals and Ceramics Information Center, C.H. Lurid: Proceedings of the 1977 Vacuum Metallurgy Confer-
Battelle, Columbus, OH, 1972, pp. A1-A40. ence, Science Press, Princeton, NJ, 1977, pp. 333-52.
13. J.C. Hartman: Electric Furnace Proceedings, Iron and Steel Society 41. R.E. Bailey, R.R. Shiring, and R.J. Anderson: Superalloys--
of AIME, 1978, vol. 35, pp. 286-88. Metallurgy and Manufacture, op. cit., pp. 109-18.
14. P.P. Turillon: Transactions of the Vacuum Melting Conference-1963, 42. R.F. Decker and J. W. Freeman: Trans. TMS-AIME, 1960, vol. 218,
R.F. Bunshah, ed., American Vacuum Society, Boston, MA, 1964, p. 277.
pp. 88-102. 43. J.J. deBarbadillo: Superalloys--Metallurgy and Manufacture,
15. R. E Decker and S. Floreen: Precipitation from Iron-Base Alloys, op. cit., pp. 95-107.
Gordon and Breach Science Publisher, New York, NY, 1965, 44. J.E. Doherty: Can. Met. Quarterly, 1974, vol. 13, p. 229.
pp. 69-128. 45. F.J. Larson and J.J. deBarbadillo: unpublished research, Inco
16. R.F. Decker: Symposium on Steel Strengthening Mechanisms, Research & Development Center, Inc., Suffern, NY, 1976.
Climax Molybdenum Co., Greenwich, CT, 1969, pp. 147-70. 46. A.W. Thompson: Grain Boundaries in Engineering Materials,
17. P.J. Bridges: The Nimonic Alloys, second edition, W. Betteridge and Proceedings of the Fourth Bolton Landing Conference, Claitor's
J. Heslop, eds., Edward Arnold Ltd., London, 1974, pp. 36-61. Publishing Division, Baton Rouge, LA, 1975, pp. 607-18.
18. C.H. White: ibM, pp. 63-97.
19. R.F. Decker and C.T. Sims: The Superalloys, C.T. Sims and 47. G.B. Thomas and T. B. Gibbons: Metals Technology, London, 1979,
W.C. Hagel, eds., John Wiley and Sons, New York, NY, 1972, vol. 6, p. 95-101.
pp. 33-78. 48. J.M. Walsh and N.P. Anderson: Superalloys--Metallurgy and
20. D.R. Muzyka: ibid, pp. 113-44. Manufacture, op. cit., pp. 127-36.
21. T. Fleitmann: Ber. Deutsch. Gesell., 1879, vol. 12, p. 454. 49. R.J. Asaro:Residuals, Additives and Materials Properties, TheRoyal
22. P.D. Merica and R.G. Waltenberg: U.S. National Bureau of Stan- Society, London, 1980, pp. 150-63.
dards Technical Paper No. 28, April 1925. 50. J. Nutting: Metals and Materials, July/August 1977, vol. 11,
23. F.A. Shunk: Constitution of Binary Alloys, second supplement, pp. 30-34.
McGraw-Hill, New York, NY, 1969, pp. 552-53. 51. J.C. Freche and G.M. Ault: Superalloys--Metallurgy and
24. R. 1". Holt and W. Wallace: International Metals Reviews, Review Manufacture, op. cir., pp. 297-320.
No. 203, March 1976, vol. 21, pp. 1-24. 52. J.S. Benjamin: Metall. Trans., 1970, vol. 1, p. 2943.
25. Residuals, Additives and Materials Properties, A. Kelly, 53. A.R. Cox, J.B. Moore, and E.C. VanReuth: Superalloys--
D.W. Pashley, E. D. Hondros, and C. Lea, eds., The Royal Society, Metallurgy and Manufacture, op. cit., pp. 45-54.
London, 1980, p. 341. 54. Proceedings, U.S. Department of Commerce Public Workshop on
26. G.S. Cole and R.S. Cremisio: The Superalloys, op. cir., Critical Materials Needs of the Aerospace Industry, John Wachtman,
pp. 479-508. ed., Doc. NBSIR 81-2305, 1981.
27. T.S. Piwonka and P.N. Atanmo: Proceedings of the 1977 Vacuum 55. J.H. Westbrook: Residuals, Additives and Materials Properties, The
Metallurgy Conference, Science Press, Princeton, NJ, 1977, Royal Society, London, 1980, pp. 25-44.
pp. 507-41. 56. J.J. deBarbadillo, J.K. Pargeter, and M.V. Makar: Process for
28. K.C. Antony and J.F. Radavich: Superalloys: Metallurgy and Recovering Chromium and Other Metals from Superalloy Scrap,
Manufacture, Proceedings of Third International Symposium on U.S. Bureau of Mines Report of Investigations 8570, 1981, 73 p.

METALLURGICALTRANSACTIONSA VOLUME 14A, MARCH 1983--341