Sustainable Civil Infrastructures

Wissem Frikha
Serge Varaksin
Antonio Viana da Fonseca Editors

Soil Testing, Soil

Stability
and Ground
Improvement
Proceedings of the 1st GeoMEast
International Congress and Exhibition,
Egypt 2017 on Sustainable
Civil Infrastructures
Sustainable Civil Infrastructures

Editor-in-chief
Hany Farouk Shehata, Cairo, Egypt

Advisory Board
Dar-Hao Chen, Texas, USA
Khalid M. El-Zahaby, Giza, Egypt
About this Series

Sustainable Infrastructure impacts our well-being and day-to-day lives. The

infrastructures we are building today will shape our lives tomorrow. The complex
and diverse nature of the impacts due to weather extremes on transportation and
civil infrastructures can be seen in our roadways, bridges, and buildings. Extreme
summer temperatures, droughts, flash floods, and rising numbers of freeze-thaw
cycles pose challenges for civil infrastructure and can endanger public safety. We
constantly hear how civil infrastructures need constant attention, preservation, and
upgrading. Such improvements and developments would obviously benefit from
our desired book series that provide sustainable engineering materials and designs.
The economic impact is huge and much research has been conducted worldwide.
The future holds many opportunities, not only for researchers in a given country,
but also for the worldwide field engineers who apply and implement these
technologies. We believe that no approach can succeed if it does not unite the
efforts of various engineering disciplines from all over the world under one
umbrella to offer a beacon of modern solutions to the global infrastructure. Experts
from the various engineering disciplines around the globe will participate in this
series, including: Geotechnical, Geological, Geoscience, Petroleum, Structural,
Transportation, Bridge, Infrastructure, Energy, Architectural, Chemical and
Materials, and other related Engineering disciplines.

More information about this series at http://www.springer.com/series/15140

Wissem Frikha Serge Varaksin
•

Antonio Viana da Fonseca

Editors

Soil Testing, Soil Stability

and Ground Improvement
Proceedings of the 1st GeoMEast International
Congress and Exhibition, Egypt 2017
on Sustainable Civil Infrastructures

123
Editors
Wissem Frikha Antonio Viana da Fonseca
National Engineering School of Tunis Faculdade de Engenharia
(ENIT) Universidade de Porto
Tunis Porto
Tunisia Portugal

Serge Varaksin
APAGEO
Paris
France

ISSN 2366-3405 ISSN 2366-3413 (electronic)

Sustainable Civil Infrastructures
ISBN 978-3-319-61901-9 ISBN 978-3-319-61902-6 (eBook)
DOI 10.1007/978-3-319-61902-6
Library of Congress Control Number: 2017946470

© Springer International Publishing AG 2018

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.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface

Toward building sustainable and longer civil infrastructures, the engineering

community around the globe continues undertaking research and development to
improve existing design, modeling, and analytical capability. Such initiatives are
also the core mission of the Soil-Structure Interaction Group in Egypt (SSIGE) to
contribute to the ongoing research toward sustainable infrastructure. This confer-
ence series “GeoMEast International Congress and Exhibition” is one of these
initiatives.
Ancient peoples built their structures to withstand the test of time. If we think in
the same way, our current projects will be a heritage for future generations. In this
context, an urgent need has quickly motivated the SSIGE and its friends around the
globe to start a new congress series that can bring together researchers and prac-
titioners to pursue “Sustainable Civil Infrastructures.” The GeoMEast 2017 is a
unique forum in the Middle East and Africa that transfers from the innovation in
research into the practical wisdom to serve directly the practitioners of the industry.
More than eight hundred abstracts were received for the first edition of this
conference series “GeoMEast 2017” in response to the Call for Papers. The
abstracts were reviewed by the Organizing and Scientific Committees. All papers
were reviewed following the same procedure and at the same technical standards of
practice of the TRB, ASCE, ICE, ISSMGE, IGS, IAEG, DFI, ISAP, ISCP, ITA,
ISHMII, PDCA, IUGS, ICC, and other professional organizations who have sup-
ported the technical program of the GeoMEast 2017. All papers received a mini-
mum of two full reviews coordinated by various track chairs and supervised by the
volumes editors through the Editorial Manager of the SUCI “Sustainable Civil
Infrastructure” book series. As a result, 15 volumes have been formed of the final
+320 accepted papers. The authors of the accepted papers have addressed all the
comments of the reviewers to the satisfaction of the track chairs, the volumes
editors, and the proceedings editor. It is hoped that readers of this proceedings
of the GeoMEast 2017 will be stimulated and inspired by the wide range of papers
written by a distinguished group of national and international authors.

v
vi Preface

Publication of this quality of technical papers would not have been possible
without the dedication and professionalism of the anonymous papers reviewers. The
names of these reviewers appear in the acknowledgment that follows. For any
additional reviewers whose names were inadvertently missed, we offer our sincere
apologies.
We are thankful to Dr. Hany Farouk Shehata, Dr. Nabil Khelifi, Dr. Khalid M.
ElZahaby, Dr. Mohamed F. Shehata, and to all the distinguished volumes editors
of the proceedings of the GeoMEast 2017. Appreciation is extended to the authors
and track chairs for their significant contributions. Thanks are also extended to
Springer for their coordination and enthusiastic support to this conference. The
editors acknowledge the assistance of Ms. Janet Sterritt-Brunner, Mr. Arulmurugan
Venkatasalam in the final production of the 15 edited volumes “Proceedings of
GeoMEast 2017”.
Contents

Study of Slope Stability (Open Pit Mining, Algeria) . . . . . . . . . . . . . . . . . 1

Mohamed Fredj, Abdallah Hafsaoui, Youcef Khedri, Riadh Boukarm,
Radouane Nakache, Abderrazak Saadoun, and Kamel Menacer
Prediction of Drained Settlement and Ultimate Bearing Capacity
for Stone Columns Supported Foundation . . . . . . . . . . . . . . . . . . . . . . . . 12
Kok Shien Ng
Considerations on the Stiffness of Sensitive Soft Soils . . . . . . . . . . . . . . . 20
Lamia Touiti and William Van Impe
A Novel Field Device for the Measurement of Soil Collapsibility . . . . . . 38
Mehdi Mokhberi and Seyed Ayuob Rafieean
Moisture Variation in Expansive Subgrade Through Field
Instrumentation and Geophysical Testing . . . . . . . . . . . . . . . . . . . . . . . . . 45
Asif Ahmed, MD Sahadat Hossain, Mohammad Sadik Khan,
Kelli Greenwood, and Aya Shishani
Geotechnical Properties of Sediments by In Situ Tests . . . . . . . . . . . . . . 59
Sara Rios, António Viana da Fonseca, Nuno Cristelo, and Claver Pinheiro
Comparison of the CO2 Record of Different Slope
Stabilization Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Dennis Gross, Susanne Kytzia, and Armin Roduner
Bearing Capacity of Strip Foundation on Soft Soil Reinforced
with Stone Columns Using Method of Slices . . . . . . . . . . . . . . . . . . . . . . . 79
M. Khalifa, M. Etezad, A. Hanna, and M. Sabry
The Behavior of a Foundation Laterally Loaded at the Top
over Highly Porous and Collapsible Soil . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Roberto Kassouf, David de Carvalho, Paulo José Rocha de Albuquerque,
and Nelson L. Fonte Jr.

vii
viii Contents

Early Applications of DMT in Arabian Gulf Area – Three

Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Sharif Emad
Some Laboratory and Numerical Studies on the Behaviour
of Stone Columns Installed in Mumbai Marine Clay . . . . . . . . . . . . . . . . 125
Vinay Bhushan Chauhan, Yashwant A. Kolekar,
and Satyanarayana M. Dasaka
Performance of Recycled Plastic Pin (RPP) for Slope Stabilization . . . . . 136
M.S. Khan, M. Sahadat Hossain, M.A. Khan, and Mohammad Faysal
Evaluation of Liquefaction Potential of New Caledonian Nickel Ores. . . . 149
Samar Daoud, Imen Said, Samir Ennour, and Mounir Bouassida
Effect of Variability of Soil Parameters in the Behavior
of Shallow Foundations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Tahar Messafer
Improvement of Collapsible Soil Conditions for Industrial Floors . . . . . 177
Nelson L. Fonte Jr., David de Carvalho, and Roberto Kassouf
Ground Improvement of Tank Foundations in the Middle East . . . . . . . 194
Babak Hamidi and Serge Varaksin
Default k-Values for Estimating Resilient Modulus
of Coarse-Grained Nigerian Subgrade Soils . . . . . . . . . . . . . . . . . . . . . . . 210
Abdulfatai Adinoyi Murana
Finite Element Analysis of Rock Slope Stability Using Shear
Strength Reduction Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
Greg You, Maged Al Mandalawi, Ahmed Soliman, Kim Dowling,
and Peter Dahlhaus
Study of the Behavior of Floating Stone Columns in Soft Clay
Formations Using Numerical Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
Ayman L. Fayed, Tamer M. Sorour, and Hany F. Shehata
Auto‐Controlled Ménard Pressuremeter: A Novel Tool
for Optimal Use of the Pressuremeter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Wissem Frikha and Serge Varaksin
Assessment of Relationship Between Static and Dynamic Load
Using Regression Analysis and Artificial Neural Network Model . . . . . . 269
Ahmed H. Abulkareem
Scanning Electron Microscope Analysis of Fly Ash, Quarry
Dust Stabilized Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
P. Indiramma and CH. Sudharani
Contents ix

Analysis of Cylindrical Cavity Expansion in Modified Cam Clay

with Ko Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Vincenzo Silvestri and Claudette Tabib
Modeling and Interpretation Wave Fields in Hierarchical
Heterogeneous Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Olga A. Hachay, Andrey Y. Khachay, and Oleg Y. Khachay
Strength and Stiffness Studies of Cement Stabilized Granular
Lateritic Soil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Dipti Ranjan Biswal, Umesh Chandra Sahoo, and Suresh Ranjan Dash
An Experimental Study on Partial Replacement of Clayey Soil
with an Industrial Effluent: Stabilization of Soil Subgrade . . . . . . . . . . . 337
Godavarthi V.L.N. Murthy, Atkuri Venkata Krishna,
and Vedula V.N. Prabhakara Rao
Ground Improvement of Titanium Dioxide Waste Spoils
and Compressible Organics with In-Situ Mixing with Portland
Cement and Surcharging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
James J. Serpico
Research and Application on the Prediction Method of Pearl Model
of High Filling Subgrade Settlement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
Xiangxing Kong
Research on the Prediction Method of Gompertz Model
for High Subgrade Settlement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
Xiangxing Kong
Study on Methods for Predicting the Settlements
of Soft Clay Roadbed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
Baotong Shi and Xiangxing Kong
Geotechnical Investigations on Highway BR 135 for Earthworks,
Asphalt Paving and Special Art Works in Subsection
Municipalities Manga-Itacarambi, North of Minas Gerais. . . . . . . . . . . . 378
Rideci Farias, Haroldo Paranhos, José Jailson Nogueira,
Marco Aurélio Diana Costa, and Roberto de Pimentel Sousa Junior
Geotechnical Investigations and Proposal for Stabilization
and Recovery Slope on Highway BR-010 Stretch in Aparecida
do Rio Negro to Goiatins, Tocantins, Brazil . . . . . . . . . . . . . . . . . . . . . . . 393
Rideci Farias, Haroldo Paranhos, Jovino Rachid Araújo,
Leonardo Ramanho Sales, and Roberto Pimentel
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
 Study of Slope Stability
(Open Pit Mining, Algeria)

Mohamed Fredj1,2(&), Abdallah Hafsaoui1,3, Youcef Khedri1,4,

Riadh Boukarm2, Radouane Nakache3, Abderrazak Saadoun1,2,
and Kamel Menacer1,3
1
Laboratory: Mineral Resources and Planning,
University of Annaba, Annaba, Algeria
2
Mining and Geology Department, Faculty of Technology,
University of Bejaia, Béjaïa, Algeria
[email protected]
3
Mining Department, Faculty of Earth Science, University of Annaba,
BP 12, Annaba, Algeria
4
Mechanical Department, Faculty of Technology,
University of Annaba, Annaba, Algeria

Abstract. The objective of our work is the study of the influence of the
dynamic load due to the use of the explosive in the stability of the slope of an
open pit mine, for this purpose we used the method of limit equilibrium to
calculate the value of the safety factor through the geotechnical software SLIDE.
On the one hand the calculation was carried out under static loading and on the
other hand the dynamic loading is taken into account, which allowed us to
conclude that the dynamic load influences the safety coefficient. Finally, a
proposal to stabilize the slope at through a reprofiling of the bench is presented.

1 Introduction

As the mining method is turning from open pit mining to sag by mining, the vertical
height of open-pit mine slope continues to increase, the slope deformation and failure
mode has close relationship with regional geological structure characteristics and the
rock mass structure feature, the stability is influenced largely by rock mass, joints and
fissures, and blasting vibration (Qiao and Li 2004). Particularly, open-pit mine’s
productive blasting vibration and rainfall has an important effect on the mine slope
stability. Productive blasting vibration has an indirect dangerous damage on the high
and steep slope. It is mainly caused by blasting seismic wave after the blasting,
especially for the joint fissures development. The high and steep rock slope containing
a fault or fracture zone is more likely as this (Li et al. 2006). The rock instabilities occur
when a number of factors come together, and for one reason or another, the state of
precarious stability prevailing until then passes an unstable situation resulting in rocks
characterized by displacement their types (sliding, flowing, falling), (Chalhoub 2006).
The stability calculation can be performed in two distinct circumstances; before or
after the out break of the movement. In the first case, the slope is apparently stable;
the purpose of stability calculation is to define a sliding surface that would have the

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_1
2 M. Fredj et al.

best chance of appearing. In other words, the stability calculation allows both to assess
the safety margin is on this side of the break, set the site the most threatened areas of
instability and examine the influence some work (earthworks, buildings, blasting,
earthquakes). On the safety margin which has been defined for the virgin slope. This
calculation step, therefore, seems very important because it allows choosing the nec-
essary parameters for the work to ensure the stability of the whole (book and website).
Unlike the first case, and when the movement has already appeared on the slope, the
stability calculation is performed to assess the safety margin between the current state
of the site of the equilibrium state. In this second case, the parameter values necessary
to introduce in the calculation are normally given by the investigations already carried
out on site, are real values such as: The geometry of the surface of the slide, the
Geotechnical characteristics of the massive and the sliding surface, etc. In this case, the
stability calculation also of great interest because it can identify the causes of the
emergence of the movement and define the work of consolidating, devices necessary to
limit the risk (Faure 2000).
According to Schroeder (2010), the different forms of rock slope instabilities
depend on factors “internal” own the massive relate to nature, to the morphology of
rock masses and the characteristics of discontinuities that effect, and factors “external”
can be natural or related to human activities.

2 Geological Setting

Jebel Onk is located in the North East of Algeria, in the eastern end of the mounts of
Nemenmcha; last membered of the Saharan Atlas. It is the natural geographic boundary
between the high plateaus Constantine and the Saharan area (Cieslinski et al. 1985,
1987); (Prian and Cortiel 1993). The study of the site area HAS year Approximately
250 hectares and belongs to the Sami mining basin Metlaoui than mine phosphate (SW
Tunisia) (Mokadem et al. 2014). It contains approximately half the phosphate reserves
of Algeria (Estimated at 2 trillion tons of reserves) (Amiour Dass et al. 2013), (Fig. 1).
According to the report made by EREM Cieslinski et al. 1985 to 1987, the deposit
Kef-Essnoun is located south of the massif of Jebel Onk 4 km from the town of Bir
El-Ater, the administrative headquarters of the region (Daira) is 2 km to the west of
Djemi-Djema deposit between Jebel Moufe and Jebel Tarfaye. The coordinates of this
deposit are x = y = 951 500 and 168.0 to 170.0.
The pit is excavated as benches with slope angles of 75° to 85°, 30 m in height, and
10 m in width, (Fig. 2). The depth of the base of mine is 70 m. The thickness of the
phosphate layer is about 50 m; the barren covering Consists of a series of Y Persian
dolomite limestone with flint, locally Overcome by lutetium limestone, Miocene sands,
and quaternary alluvium.
 Study of Slope Stability (Open Pit Mining, Algeria) 3

Fig. 1. Location map of the study area.

Fig. 2. Quarry of Kef Essnoun before sliding.

3 Study of Instability

Following the great landslide, which took place on September 8, 2007 in the North East
side of the quarry, a large mass of rock broke off from the solid, almost completely
filling the pit (estimated volume of 6,000,000 m3), (Fig. 3).
4 M. Fredj et al.

Fig. 3. Geological section of the studied slope (Kef-Essnoun) after sliding.

According to the geological section and the position of the sliding surface at the
lamination joint (marl interface phosphate), the slip is classified as a plan year. The
expert reports on this shift has resulted in the fact that the probable causes of this hazard
are attributed to some geological factors and operating (Benaissa 2003).

3.1 Geological Factors

– Presence of marl interface phosphates;

– The presence of a near-vertical recovery of the layer of marl, which allows the mass
detached, is following a true inclined plane;
– The highly tectonized character of the massif.

3.2 Operating Factors

– Very straightened Front, almost vertically from the figure to a height of 70 m;

– Reduced dimensions of the berm;
– Presence at the foot of the front layer of marl, which played a key role in the sliding
of the whole;
– The cumulative effect of shooting with explosives on the platform.
After this shift, operation has been shifted to the west side and southwest.
However, cracking was observed in February 2013, about 250 m upstream of the
platform, suggesting a beginning of a new potential slip, (Fig. 4). The observed cracks
 Study of Slope Stability (Open Pit Mining, Algeria) 5

are located along the outcrops of marl located stratigraphically below the level of the
exploited phosphates. The most likely cause of this cracking is related much more to the
vibrations generated by the rock blasting, fire with explosives that a landslide primer.

Fig. 4. Changes in cracks.

4 Methodology

In this article, we will examine the current stability of the northern flank of Kef
Essnoun, (Fig. 5), and provide a potential risk of instability represented by a safety
factor determined by the limit equilibrium methods (SLIDE). In our case study, the
parameters used were identified in the available data submitted by the company. They
come from previous tests at a laboratory in accordance with standard AFNOR Euro-
code 07 (XP P 94-010), Table 1.
To achieve the modeling according to selected software, sections were cut along the
northern flank of Kef-Essnoun to have profiles, with the thicknesses of the different
facies.
6 M. Fredj et al.

Fig. 5. Study site profile.

Table 1. The physical and mechanical properties of the different layers of along the north side
of Kef-Essnoun mining.
Properties Symbols Rock formations
units Limestone Phosphate Marl Danio
Ypresian-Lutetian limestone-montien
Compressive Rc MPa 60 25 8 40
strength
Density qinsat 27 21 23 27
unsaturated kn/m3
Saturated qsat kn/m3 27,46 24,81 24 27,46
density
Cohesion C kn/m2 5400 2300 160 3600
Internal friction u (°) 37 37 16 37
angle
Dilatant w (°) 7 7 0 7
Young’s E MPa 27000 24000 1000 27000
modulus
Poisson’s ratio m 0,250 0,28 0,25 0,25

5 Analysis of Stability

According to observations carried out on the site, it was deduced that the slippage
affects the layer of marl. Given the lack of data regarding this facies of our study site,
we opted the results of analysis of the old sliding whose geological conditions are
 Study of Slope Stability (Open Pit Mining, Algeria) 7

comparable to those of current place of study to determine the most likely Geotechnical
characteristics. This is to trace the value of cohesion and internal friction angle of marl,
compatible for the 2007 landslide, with a critical safety factor (SF = 1).
According to the analysis results showing the variation of SF in terms of cohesion
and internal friction angle of the facies of marl is found that a value of a critical safety
factor (SF = 1): C = 45 kPa and u = 14°.
The next step is to study the stability with or without the influence of vibration
induced by blasting is simulated by integrating the parameter ground acceleration
induced earthquake, the value obtained for horizontal and vertical acceleration esti-
mated respectively at: Ah/g = 0.05 and Av/g = 0. 0125.
Where; Ah/g - horizontal acceleration of gravity; Av/g - vertical acceleration of
gravity.

5.1 Phase Calculation

Static Load
We chose three calculation methods: Spencer, Bishop and Janbu, (Fig. 6).
SF of the values calculated according to the three methods mentioned previously
shown in the following Table 2. According to the results of the safety factor, the limit
equilibrium method (SLIDE), and static load (effect explosives), we find that the
northern flank of Kef-Essnoun is stable, (Fig. 6).

Fig. 6. The sliding surface and the safety factor value (static load).
8 M. Fredj et al.

Table 2. Value of safety factor with zero acceleration.

Methods Bishop simplified Janbu simplified Spencer
Value of safety factor (SF) 1.281 1.226 1.302

Dynamic Load
SF of the values calculated according to the three methods mentioned previously
shown in the following Table 3. For the same Geotechnical characteristics (specific
gravity, cohesion, friction) and dynamic load (effect explosives): Ah/g = 0.05 and
Av/g = 0. 0125, we find that the northern flank of Kef Essnoun is unstable and the
break line is localized more exactly at the phosphate marl interface, (Fig. 7).

Table 3. Value of safety factor with acceleration.

Methods Bishop simplified Janbu simplified Spencer
Value of safety factor (SF) 0.736 0.700 0.700

Fig. 7. The sliding surface and the safety factor value (dynamic load).

5.2 Proposal of a Method of Reinforcement

According to the analysis conducted earlier results, adequate reinforcement method to
stabilize our study site is profiling. It consists of:
– The reduction of the crest to a level thus making it possible to eliminate the effect of
the straightening of the layers, (Fig. 8).
– Creation of bench a height of 15 m to increase stability with the creation and
widening of the platforms, (Fig. 9).
 Study of Slope Stability (Open Pit Mining, Algeria) 9

– Profiling bleachers and platforms, (Fig. 10).

– The reinforcement method chosen made it possible to increase the safety factor and
thus to ensure the stability of the site, (Fig. 11).
According to the result, it is found that profiling makes it possible to increase the
stability with Safty Factor of the order of 1.66.

Fig. 8. Reduction of the crest.

Fig. 9. Creation of stands and expanding platforms.

10 M. Fredj et al.

Fig. 10. Reprofiling: bleachers and platforms.

Fig. 11. The sliding surface and the safety factor value with the dynamic load after reprofiling.
 Study of Slope Stability (Open Pit Mining, Algeria) 11

6 Conclusions

The analysis of the results allows us to say that in addition to the geological and
geotechnical factors that can influence the stability of a slope, the dynamic loading due
to the explosive must be taken into account. The control of this parameter can be
achieved by reducing the height of the bench, the widening of the working platform
and the modification of the operating method of exploitation.

Acknowledgments. The author would like to thank the staff of the mining laboratory of the
Annaba University (Algeria). Also, a special thanks to A. Hafsaoui (Annaba University -
Algeria) and to Boukarm R (Bejaia University - Algeria) for her objective comments and
corrections.

References
Dass Amiour, M., Mezghache, H., Elouadi, B.: The use of three physico-chemical methods in the
study of the organic matter associated with the sedimentary phosphorites in Djebel Onk
Basin, Algeria. Arab. J. Geosci. 6, 309 (2013). doi:10.1007/s12517-011-0381-9
Benaissa, A.: Glissements de terrain: Calcul de stabilité, 2éme édition. Edition office des
publications universitaires, Alger (2003)
Chalhoub, M.: Apports des méthodes d’homogénéisation numériques a la classification des
massifs rocheux fracturées. Thèse de doctorat, Ecole Nationale Supérieure des Mines de Paris,
spécialité géologie de l’ingénieur, pp. 39–45 (2006)
Cieslinski, S., et al.: Travaux de prospection et d’évaluation des phosphates de la région de Bir El
Ater, EREM, pp. 3–10, 14–26, 39–44, 62, 63, 73–84, inédit (1985, 1987)
Faure, R.M.: L’évolution des méthodes de calcul en stabilité des pentes partie I méthodes à la
rupture, Revue Française de Géotechnique (2000)
Li, W., Wang, J., Dai, L.: Safety criterion of blasting shock for the high deep slope of open-pit in
mountain area. Ind. Miner. Process. 1, 20–22 (2006)
Mokadem, N., Hamed, Y., Ben Saad, A., Gargouri, I.: Atmospheric pollution in North Africa
(ecosystems-atmosphere interactions): a case study in the mining basin of El Guettar-M’Dilla
(southwestern Tunisia). Arab. J. Geosci. 7(5), 2071–2079 (2014)
Prian, G.P., Cortiel, P.: Etude de développement du gisement de phosphate de Djebel Onk
(Algérie). Rapport d’expertise géologique, B.R.G.M. France, pp. 11–29, 133–149, 169–173,
inédit. (1993)
Qiao, L., Li, Y.: Engineering geological model of high-steep slope damage in open cut mines.
J. Univ. Sci. Technol. Beijing 26(5), 461–464 (2004)
Schroeder, C.: Etude de stabilité des parois rocheuses. Revue de société Belge de géologie de
l’ingénieur et de mécanique des roches (SBGIMR), pp. 15–18, 21–29, 34 (2010)
 Prediction of Drained Settlement and Ultimate
Bearing Capacity for Stone Columns
Supported Foundation

Kok Shien Ng(&)

Faculty of Civil Engineering, Universiti Teknologi MARA,

Pulau Pinang Branch, 13500 Permatang Pauh, Penang, Malaysia
[email protected]

Abstract. This paper presents a validation of simplified design approach against

a field study and a laboratory test. The design approach has been developed to
predict drained settlement and the ultimate bearing capacity of stone columns
supported foundation adopting a hyperbolic relationship to simulate the strain
hardening behavior. The result of a plain footing load test is required to give two
curve fitting parameters and then adapted them by introducing a reduction factor
which take into account the length ratio and the area replacement ratio to predict
the stress-strain characteristic of a stone column supported foundation. Good
agreements are found when the predictions using the simplified approach are
compared with the laboratory testing and field case study.

1 Introduction

Stone columns has been used intensively all around the world to improve the subsoil
for supporting many civil infrastructure including road and railway embankment,
building, factory, large storage tank, etc. that can tolerate some settlements. Design of
stone columns supported foundation is often carried out in two steps. First, it has to be
checked for the ultimate bearing capacity. Second, serviceability check is performed
particularly on the long term drained settlement which is usually more critical.
Stone columns are used in groups and can be installed either as end bearing or
floating type. Most of the stone column design is catered for end bearing columns
(Balaam and Booker 1981; Priebe 1995; Han and Ye 2001; Castro and Sagaseta 2009)
and very little for floating columns (Rao and Ranjan 1985; Lawton et al. 1994; Shahu
and Reddy 2014; Ng and Tan 2014; Bouassida and Carter 2014). Many model tests
(either laboratory or numerical) and field tests have been performed to study different
aspects of stone column behavior such as settlement, consolidation rate, ultimate
bearing capacity, failure modes, load sharing, group performance, etc. (Babu et al.
2013; Najjar 2013; Bouassida and Hazzar 2015). Based on limit analysis, Bouassida
et al. (2009) proposed an analytical solution to predict the bearing capacity of a rigid
foundation on a soil reinforced by a group of floating columns. Considering the elastic
strain in the surrounding soil and plastic strain in column, Castro (2016) developed a
closed-form solution to estimate the settlement of small column groups. A mechanical
model has been developed by Das and Deb (2014) to study settlement as well as

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_2
 Prediction of Drained Settlement and Ultimate Bearing Capacity 13

bending moment and shear force of the raft foundation resting on a stone column
improved ground.
Load-deformation behavior of a plain footing is a complex phenomenon where
three types of deformation mode have been observed, i.e. strain hardening, strain
softening and limiting capacity. With the inclusion of stone columns, the complexity is
further increased due to multiple failure modes that can happen in the improved ground
system such as wedge shearing, bulging, and punching (Tan and Ng 2013). Prediction
of the nonlinear behavior of the stone columns reinforced foundations is difficult except
by numerical approach. However, this modern computational technique requires
advanced constitutive model to simulate the soil and the column behavior where the
soil/column properties are sometimes difficult and expensive to obtain.
In previous study, a simplified design approach was developed to predict the long
term drained settlement of a stone columns supported foundation. It was first compared
against a series of finite element analysis where a good match was obtained for the
load-deformation curve (Ng et al. 2014). In this study, the soundness of the design
approach was further tested against a laboratory model test and a field test.

2 Simplified Design Approach

The general hyperbolic equation has been used to simulate this nonlinear stress-strain
response of soils (Kondner 1963; Duncan and Chang 1970). Jeon and Kulhawy (2001)
further adapted the hyperbolic equation to displacement and the equation is given
below:

D
q¼ ð1Þ
a þ bD

where q is the applied bearing pressure, D is the footing displacement, and a and b are
curve fitting parameters. The curve fitting parameters can be obtained by plotting the
actual load-deformation curve in a linearized hyperbolic form as shown in Fig. 1 where
a is the intercept of D/q versus D plot while b is the slope of D/q versus D plot. The
parameter a is the inverse of initial tangent modulus of the ground while parameter b is
related to the inverse of ultimate bearing capacity, qult of the footing system. The
ultimate bearing capacity and the actual failure stress are assumed to be the same since
the footing drained performance usually exhibit strain hardening behavior.
Footing supported by stone columns displays stiffer response under load as well as
having a higher ultimate bearing capacity compared to a plain footing which also means
smaller values for a and b parameters in the Eq. (1). The term ‘plain footing’ is used to
refer to the un-reinforced shallow foundation. The determination of a and b parameters
for improved ground is the essence of this new design approach. These two curve fitting
parameters acquired from the load-deformation curve for a plain footing in drained
manner is needed in the first place. Then, reduction factor, M applies to these parameters
to obtain the hyperbolic curve for the reinforced stone column foundation, give as:
14 K.S. Ng

Fig. 1. (a) Hyperbolic stress-strain curve; (b) Transformation of hyperbolic relation.

a
a1 ¼
M
ð2Þ
b
b1 ¼
M
which they are named as modified a and b parameters for stone columns reinforced
foundation. Modified parameters are dependent on two factors: area ratio, a and column
length, L. Area ratio is defined as the ratio of stone columns cross sectional area to the
cross sectional area of the footing. Reduction factor, M is proposed as:

M ¼ 1 þ LLopt ð2a  0:15Þ L  Lopt

ð3Þ
M ¼ 1 þ Ld ð2a  0:15Þ Ld

where d is the thickness of soft soil below footing, Lopt is the optimum length of stone
column and L/Lopt and L/d are denoted as length ratio. Columns longer than a certain
length which give no further improvement in settlement performance is termed optimum
length, Lopt. The M value was first approximate from a series of parametric studies
conducted during the preliminary study by means of axisymmetric finite element
method using 15-node triangular element with elastic-perfectly plastic material for both
columns and soils where M is taken to be equal to the basic settlement improvement
factor, n0 = 2.575a + 0.931 (n = the ratio of settlement without improvement over
settlement with improvement) for cases with column length longer than the optimum
length However, further improvement was made to account for nonlinear behavior of
soil, thus the M value becomes slightly smaller i.e. M = 2a + 0.85 for end bearing
columns or columns longer than optimum length. This value is influenced by the area
replacement ratio together with the length ratio and those are the important factors in
determining the stone column performance. A previous study by Tan et al. (2014)
suggested that the optimum length can be computed as:
 Prediction of Drained Settlement and Ultimate Bearing Capacity 15

Lopt ðmÞ ¼ ð2a þ 0:8ÞD þ 0:5  t ð4Þ

where D equal to the diameter or breath, B of a footing, and t equal to the thickness of
granular mat above the stone columns. The load-deformation curve for the stone col-
umn reinforced foundation can be drawn using Eq. (1) when the values of a1, and b1
are obtained. The design procedure to predict the load-deformation curve for a small
column group is summarized in Fig. 2.

Obtain plain footing load-displacement curve

Find parameters a & b from hyperbolic relation

Apply reduction factor, M to parameters a & b

Use Eq. (1) to predict load-displacement curve for

stone column

Fig. 2. Calculation procedures in hyperbolic method

The simplified design approach differs from the conventional method because it
does not require the arbitrary selection of soil parameters, but is capable of simulating
the yielding response of the improved ground that is strain hardening. Besides, the
effect of footing embedded depth, ground water table and soil layering are mooted in
this design approach since the only requirement is the stress-strain characteristic of a
plain footing under similar testing condition.

3 Validation with Model Test and Field Load Test

3.1 Model Test (Wood et al. 2000)

In a model test by Muir-Wood et al., floating columns with the length, L of 1.7D
(>Lopt  1.33D) and a = 0.24 were used to support a rigid circular footing of 0.1 m
diameter. The sand columns were formed with replacement technique and placed in a
square grid extending just beyond the edge of the footing. The relative density of the
sand poured into the columns was around 0.5. The undrained shear strength of the
surrounding soil is Cu = 10 kN/m2. The displacement controlled test was conducted
slowly in a drained manner until 30 mm settlement was reached.
Figure 3 shows the hyperbolic relation of the plain footing without stone columns
as used in the model test. The parameter a and b are obtained as 0.01 and 0.13
respectively. Using Eq. (3), the reduction factor M was computed as 1.33. The
16 K.S. Ng

Fig. 3. Hyperbolic relation.

Fig. 4. Load deformation results.

prediction is well comparable to the model test as depicted in Fig. 4 despite slight
overestimation of settlements, especially in the beginning of loading stage. There are
probably two reasons for this overestimation. First, the footings are not only supported
underneath, but with extra “buttressing” columns outside the edge of the footing which
give extra constraint effect. Second, the footing was contained in a tank with 300 mm
diameter, thus the boundary is only 1.5 times the footing diameter measured from the
centerline and this boundary effect may have influenced the footing performance.

3.2 Field Load Test (Stuedlein et al. 2011)

A field load test (denoted as G4) on square footings of 2.7 m width supported with 5
numbers of floating stone columns. The column length is 3.05 m which is shorter than
the optimum length, Lopt of 4.34 m. The diameter of columns is 0.76 m and the area
 Prediction of Drained Settlement and Ultimate Bearing Capacity 17

replacement ratio, a is 0.3. The stone columns in the test are vibro-compacted. The first
soil layer is 0.7 m thick hard, desiccated clay crust, followed by soft to medium stiff
upper clay layer to a depth of 3.7 m, overlying sandy silt/silty sand layer with total
thickness of 0.7 m. 4.5 m deep onwards, lie the fourth and deepest layer which is a
slightly silty clay. Laboratory soil tests showed the upper 4.5 m was over-consolidated.
A plain footing test (G3) acted as control test. Square footings in both tests were placed
on the ground surface, direct contact with stone columns. The duration of load tests was
approximately 500 to 600 min (8.5 to 10 h).
Using the proposed method, the reduction factor, M was calculated to be 1.32. The
result of field load test was compared with the result obtained from the proposed
method where a good agreement was obtained as shown in Fig. 5. The slight over
prediction may probably due to the short period of test which the drained condition had

Fig. 5. Field test (a) Hyperbolic relation, and (b) Load-deformation results.
18 K.S. Ng

not been fully achieved. In addition, the ultimate bearing capacity for the column group
was predicted to be 602 kN/m2.

4 Conclusion

This paper presents a validation of a simplified design approach against a model test
and a field load test where the results proved the method to be sufficiently accurate in
estimating the drained settlement of stone columns supported foundation. The
load-deformation response of the foundation can be characterized with hyperbolic
function as shown in the above validation exercise. Albeit simple and easy, the method
is indeed a rational method which has taken into account many intrinsic behaviors of
stone columns such as the load sharing mechanism, nonlinearity and length ratio.

References
Babu, M.D., Nayak, S., Shivashankar, R.: A critical review of construction, analysis and
behaviour of stone columns. Geotech. Geol. Eng. 31(1), 1–22 (2013)
Balaam, N.P., Booker, J.R.: Analysis of rigid rafts supported by granular piles. Int. J. Numer.
Anal. Methods Geomech. 5(4), 379–403 (1981)
Bouassida, M., Jellali, B., Porbaha, A.: Limit analysis of rigid foundations on floating columns.
Int. J. Geomech. 9(3), 89–101 (2009)
Bouassida, M., Carter, J.P.: Optimization of design of column-reinforced foundations. Int.
J. Geomech. 14(6), 04014031 (2014)
Bouassida, M., Hazzar, L.: Performance of Soft Clays Reinforced by Floating Columns. In:
Ground Improvement Case-Histories: Embankments with Special Reference to Consolidation
and Other Physical Methods, p. 433 (2015)
Castro, J., Sagaseta, C.: Consolidation around stone columns. Influence of column deformation.
Int. J. Numer. Anal. Meth. Geomech. 33(7), 851–877 (2009)
Castro, J.: An analytical solution for the settlement of stone columns beneath rigid footings. Acta
Geotech. 11(2), 309–324 (2016)
Das, A.K., Deb, K.: Modeling of uniformly loaded circular raft resting on stone
column-improved ground. Soils Found. 54(6), 1212–1224 (2014)
Duncan, J.M., Chang, C.Y.: Nonlinear analysis of stress and strain in soils. J. Soil Mech. Found.
Div. 96(5), 1629–1653 (1970)
Han, J., Ye, S.L.: Simplified method for consolidation rate of stone column reinforced
foundations. J. Geotech. Geoenviron. Eng. 127, 597–603 (2001)
Jeon, S.S., Kulhawy, F.H.: Evaluation of axial compression behavior of micropiles. In:
Foundations and Ground Improvement, pp. 460–471. ASCE (2001)
Kondner, R.L.: Hyperbolic stress-strain response: cohesive soils. J. Soil Mech. Found. Div. 89
(1), 115–143 (1963). ASCE
Lawton, E.C., Fox, N.S.: Settlement of structures supported on marginal or inadequate soils
stiffened with short aggregate piers. In: Geotechnical Special Publication 40, pp. 962–974
(1994)
Najjar, S.S.: A state-of-the-art review of stone/sand-column reinforced clay systems. Geotech.
Geol. Eng. 31(2), 355–386 (2013)
 Prediction of Drained Settlement and Ultimate Bearing Capacity 19

Ng, K.S., Tan, S.A.: Design and analyses of floating stone columns. Soils and Foundation 54(3),
478–487 (2014)
Ng, K.S., Tan, S.A., Sun, J.: A new approach to the estimation of settlement and ultimate bearing
capacity of stone columns supported shallow foundation. In: The 8th European Conference on
Numerical Methods in Geotechnical Engineering (NUMGE 2014), 18–20 June 2014, Delft
(2014)
Priebe, H.J.: The Design of Vibro Replacement, pp. 31–37. Ground Engineering (1995)
Rao, B.G., Ranjan, G.: Settlement analysis of skirted granular piles. J. Geotech. Eng. 111, 1264–
1283 (1985)
Shahu, J.T., Reddy, Y.R.: Estimating long-term settlement of floating stone column groups. Can.
Geotech. J. 51, 770–781 (2014)
Stuedlein, A.W., Holtz, R.D.: Analysis of footing load tests on aggregate pier reinforced clay.
J. Geotech. Geoenviron. Eng. 138(9), 1091–1103 (2011)
Tan, S.A., Ng, K.S.: Stone columns foundation analysis with concentric ring approach. In:
Proceedings of the Third International Symposium on Computational Geomechanics
(ComGeo III), Krakow, Poland (2013)
Tan, S.A., Ng, K.S., Sun, J.: Column groups analyses for stone column reinforced foundation.
Geotechnical Special Publication 233, pp. 597–608 (2014)
Wood, D.M., Hu, W., Nash, D.F.T.: Group effects in stone column foundations model tests.
Geotechnique 50(6), 689–698 (2000)
 Considerations on the Stiffness of Sensitive
Soft Soils

Lamia Touiti1(&) and William Van Impe2

1
LGC, Civil Engineering Laboratory - Ecole Nationale d’Ingénieurs de Tunis,
Université de Tunis El Manar, Tunis, Tunisia
[email protected]
2
ISSMGE, Ghent University Belgium, Gent, Belgium

Abstract. The present paper comments on the evaluation of the initial stiffness
of sensitive soft soils as an outcome of several testing methods: Resonant
Column (RC), Bender Element (BE) tests, Seismic cone tests and Empirical
relationships (correlations). The RC, BE and common geotechnical laboratory
tests were carried out on soft clays adjacent to the Tunis Lake in Tunisia. The
database was completed making use of existing studies carried out on other
types of Swedish soft clays from the literature related to seismic cone tests.
A number of empirical correlations for determining the initial value of the shear
modulus of soils in the literature was adopted in order to re-analyze the available
data set. The authors suggested an adapted empirical lower bound correlation
aiming at evaluating the analyzed soft soils’ small strain shear modulus (Gmax),
probably even useful for less undisturbed sampling. The proposed equation aims
at giving an indication of the effect of, for example, the mean effective stress (p′)
and the corresponding void ratio (e) on the initial soil stiffness more clearly.

1 Introduction

The small strain shear modulus (Gmax) of soil is an important parameter for a large
variety of geotechnical design problems. Such modulus is typically associated with
shear strain levels of about 10−4% and below. The critical role of soil stiffness at
small-strains in the design and analysis of geotechnical structures is widely accepted.
For example, Gmax is a key parameter in small-strain static and dynamic analyses
(certainly covering foundation engineering problems such as those related to earth-
quake, wind power plants, machine foundations or rail/road traffic…) (Schnaid 2005;
Stokoe et al. 2005). Gmax can be equally important for small-strain cyclic situations
(wind or wave loading).
Gmax is reasonably well connected also to other soil properties, such as density and
sample disturbance (Nash 1999). Its value depends on a number of parameters,
including void ratio, confining stress, soil structure, degree of saturation, stress history
and time factors. Table 1 shows a list of parameters affecting Gmax at different levels of
importance, originally listed by Vucetic and Dobry (1991) and updated by Benz
(2007). As shown in this table, the most important parameters affecting the soft soils
small-strain stiffness are confining pressure, void ratio, geologic age, cementation,
overconsolidation ratio, plasticity index as well as the frequency of a loading pattern.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_3
 Considerations on the Stiffness of Sensitive Soft Soils 21

Table 1. Parameters affecting the small-strain stiffness of soils (Theenathayarl (2011), modified
from the original table presented by Vucetic and Dobry (1991))
Increasing factor Gmax
Confining pressure (r’m) Increases with (r’m)
Void ratio (e) Decreases with e
Geologic age (tg) Increases with tg
Cementation (c) Increases with c
OCR Increases with OCR
Plasticity Index (PI) Increases with PI if OCR > 1
Stays constant if OCR = 1
Cyclic strain (cc) –
Frequency of loading (f) Increases with f

Fig. 1. Field and laboratory methods for determining shear modulus (Shneider et al. 1999).
22 L. Touiti and W. Van Impe

It is common practice to qualify a reliable value of Gmax of the soil (at induced strain
levels less than 0.0001%) from the shear wave velocity:

G0 ¼ V2s  q; kPa ð1Þ

Where: G0 = Initial shear modulus; q = mass density; Vs = shear waves velocity

for a linear, elastic and isotropic medium.
Most testing techniques are consisting of a combination of standard geotechnical
tests with the geophysical modulus used both in laboratory and field conditions. Such
hybrid (field/lab) method is shown in Fig. 1 (Schneider et al. 1999). Field techniques,
besides of the seismic cone (SCPTU) and seismic flat dilatometer, include cross hole
and down hole tests in the typical geophysical version and the SASW method. Lab-
oratory tests in turn, are described as the resonant column (RC), torsional shear and
triaxial tests with local strain measurement and the measurement of the shear wave
velocity based on bender element (BE). Such configurations diminish the disadvan-
tages of each group of tests and considerably enhance the optimization of data col-
lection (Wolski et al. 2006).
The focus of this paper is to evaluate the Gmax using RC, BE and seismic cone test
results verifying some empirical formulas, from literature and adding own test results.
The analyzed data correspond to both laboratory tests conducted on Tunis soft soil (RC
and bender BE tests) and collected data related to Swedish soft soils (laboratory tests
and seismic cone as field tests). Having analyzed the results, the authors suggest an
adapted correlation which fits well the sensitive soils.

2 Materials and Methods

2.1 Tunis Soft Soils

The soil for laboratory tests, sampled as “undisturbed” standard tube samples, was
collected from the region close to the Tunis Lake in Tunisia, more specifically, from the
test sites located in and near the Ghana avenue (south lake). The soil samples can be
described as gray-black soft silty clay and do belong to a natural cohesive soil for-
mation, of recent Quaternary origin, deposited in a marine environment. The deposit
qualifies as a normally consolidated soft soil with illite as the dominating clay and
quartz in the silt particles (Kaâniche et al. 2000; Touiti et al. 2009).
All laboratory tests were carried out at the Geotechnical Institute for Experimental
Models (ISMGEO Laboratory) in Italy and at the Geotechnical laboratory of the Ghent
University (Belgium). The test results summarized in Table 2 do show a clay and silt
content of about 40%, an organic content in the range of 2 to 5%, a water content from
84 to 110% (slightly above the liquid limit), an undrained shear strength derived from
TXCIU tests from 15 to 28 kPa and Gmax, determined using RC and BE tests, in the
range of 3 to 8 MPa. The sensitivity of the Tunis soft soils samples and the one at the
Ska-Edeby site was estimated on the basis of the liquidity index (LI) and the undrained
shear strength (Cu), as established by Leroueil et al. (1983):
 Considerations on the Stiffness of Sensitive Soft Soils 23

Table 2. Identification and soil index properties of selected Tunis soft soils
Specimen Depth Soil Water Bulk density Void Liquid Plasticity Liquidity Organic CaCO3 Cu St
name (m) type content ch/cd index limit LL index PI index Matter CC [%] kPa
W [%] [KNm−3] e [–] [%] [%] LI [–] OM [%]
RC1(SC1) 5.0–5.8 MH 98 14.8/7.5 2.7 96 41 0.85 4.5 34 28 20
RC2(SC2) 7.8 CH 92 14.2/7.4 2.1 85 49 1.10 2.84 39 28 24
BE1(SC3) 3.2–4.0 CH 111 13.8/6.5 2.6 85 58 1.13 – – 15 22
BE2(SC3) 3.2–4.0 CH 91 14.8/7.7 2.4 80 58 0.79 – – 20 19
BE3(SC3) 3.2–4.0 CH 84 15.1/8.2 2.3 81 58 0.67 – – 25 18
BE4(SC4) 5.0–5.7 MH 89 14.1/7.5 2.3 79 30 1.13 2.2 30 18 23

1
Cur ðkPaÞ ¼ ð2Þ
ðIL  0:21Þ2

St ¼ CuðIL  0:21Þ2 ð3Þ

The sensitivity of Tunis soft soils ranged from 18 to 24.

2.1.1 Resonant Column Tests

The RC tests were performed using the ISMGEO laboratory Resonant Column
Apparatus (4–100 Hz) which is an example of Hardin-Drnevich resonant column using
“fixed-free” configuration (Fig. 2). This equipment is commonly used to study the
dynamic deformation characteristics of soil.

Fig. 2. Resonant column apparatus used for the tests - ISMGEO laboratory – Italy

The procedure for testing the soils with RC method followed that outlined in
ASTM D 4015-07. In fact, the specimen is rigidly fixed at the base while the torsional
oscillation is applied to the free end by a drive head. The frequency of vibration is
gradually increased until reaching the first-mode fundamental frequency of the sample,
at which measurements of the resonance frequency and amplitude of vibration are made.
Knowing the geometry and the end constraints of the sample, the measured resonance
frequency is then used to calculate the shear wave propagation velocity using the wave
propagation equation and the theory of elasticity. The shear modulus is subsequently
obtained from the derived shear wave velocity and the density of the sample.
24 L. Touiti and W. Van Impe

2.1.2 Bender Element Tests

The triaxial tests were performed on 6 undisturbed Tunis soft soil specimen in three stages:
saturation (back pressure method), consolidation and shearing (strain controlled mode).
The bender elements are small electro-mechanical transducers which either bend as
an applied voltage is changed or generate the voltage as they bend. Bender elements
were placed into soil sample during the triaxial tests (Fig. 3). The measurements of the
shear wave velocity were performed at the end of each of the saturation and consoli-
dation stage during triaxial tests (Fig. 3). In order to rule out the incorrect measure-
ments, additional measurements of the travel time at the same stress condition but with
different frequency of input signal were made.

Fig. 3. (a) Tunis soft soil samples wrapped in filter paper and rubber skin. (b) Bender elements
placed into Tunis soft soil sample (c) Triaxial apparatus equipped with bender elements used for
the tests

2.2 Swedish Soft Soils

A geotechnical soft soil database including stiffness properties measured by seismic
cone tests was elaborated. The data base was completed on the basis of several inves-
tigations and studies performed on Swedish soft soils (Larsson and Mulabdic 1991).
The analyzed Swedish test sites were selected among the well documented test
fields previously used by the Swedish Geotechnical Institute (SGI): Skå-Edeby,
Norrköping and Bäckebol sites. The soft soils at those sites have a thickness of 5 to
10 m predominantly marine post-glacial clay with the presence of shells and thin layers
or seams of silt.
The natural water content is generally higher than the liquid limit. The sensitivity of
the clays varies between 8 and 24. The undrained shear strength as determined by field
vane tests reaches a minimum of 7.5 to 9 kPa at 2 m below the ground surface for the
various sites and seems to increase with depth.
The geotechnical parameters of the various sites are summarized in the Table 3.

Table 3. Geotechnical parameters of selected Swedish soft soils

Site Z (m) W (%) WL (%) IP (%) P′ (kPa) e Cu (kPa) OC (%) St Vs m/s OCR
Norrköping 3–10 70–118 62–84 _ 33–50 1.7–3.3 9–15 _ 10–20 66–98 1.2–1.6
Skå-Edeby site 3–9 64–87 55–83 30–56 17–38 1.7–2.4 8–15 1–5 6–16 – 1.1–1.2
Bäckebol site 2–8 78–100 69–86 36–51 22–37 2.0–2.7 11–15 <1% 12–24 56–85 1.2–1.9
 Considerations on the Stiffness of Sensitive Soft Soils 25

2.2.1 Norrköping Site

According to Larson and Mulabdic (1991), the investigated soil profile consists of
about 8 meters of gray varved clay (2–10 m) with thin layers or seams of silt which
occur at about 5 m and do appear even more regular from 7 m depth on. The natural
water content is higher than the liquid limit and varies from 120% at the upper layers to
about 70% in the silty bottom layers. The soil is slightly over consolidated; the over
consolidation ratio ranges from 1.2 to 1.6. The sensitivity of the clay measured with the
field vane tests varies between 10 and 20. The small strain shear modulus deduced from
the seismic cone tests ranges from 5 to 15 MPa.

2.2.2 Skå-Edeby Site

The test field is located on Svartsjölandet. A large number of tests has been made in
that test field at Ska-Edeby (Kallstenius and Bergau 1961; Kallstenius 1963; oysterman
and Lindskog 1963; Push 1970; Holtz and Linkskog 1972; Holtz and Broms 1972;
Holtz and Holm 1972; BomanandBroms 1975; Wiezel 1975; Massarsch et al. 1975;
Torstensson 1976; Holm and Holtz 1977; Larsson 1986; Larsson and Eskilson 1989;
Larsson and Mulabdic 1991).
According to the above-mentioned studies, the soil profile (3–9 m) consists of 3 m
of post-glacial clay, highly plastic and slightly organic, followed by 6 m of varved
glacial clay. For both layers, the water content is higher than the liquid limit, the soil is
slightly overconsolidated (OCR = 1.1 to 1.2), the undrained shear strength (field vane
testing) increases with depth from 8 kPa to 14 kPa, the sensitivity varies from 6 to 16
and is common of 15 and the small strain shear modulus (seismic cone tests) varies
with depth from 3 MPa to 14 MPa showing a uniform and continuous increase.

2.2.3 Bäckebol Site

The test field is located near the river GötaÄlv and concerns mainly the upper 10 m of
the soil site. According to Larrsson and Mulabdic (1991), the clay consists of post
glacial marine clay, with illite as the dominating clay mineral and quartz and feldspar in
the silt particles, the clay content being about 60%, the organic content less than 1%,
the water content reaching some 70 to 90% which is slightly above the liquid limit.
The undrained shear strength varies between 11 to 15 kPa, its sensitivity shows
values ranging from 12 to 24 and the small strain shear modulus derived from the
seismic cone tests varies from 5 to 11 MPa.

3 Interpretation and Discussion of RC and BE Test Results

and Collected Data

3.1 Tunis Soft Soils

The results obtained in the laboratory from BE and RC tests are presented in Figs. 4
and 5. The values of the shear wave velocity and the small strain shear modulus varied
from 40 to 80 m/s and 3 to 8 MPa respectively at an applied mean effective stress
ranging from 10 to 50 kPa.
26 L. Touiti and W. Van Impe

Fig. 4. Shear modulus and damping ratio for Tunis soft soil

Fig. 5. Small strain shear modulus and shear wave velocity vs. mean effective stress for Tunis
soft soils
 Considerations on the Stiffness of Sensitive Soft Soils 27

The relationship between the estimated shear modulus, the mean effective stress
and the void ratio for tested soft soils (Fig. 6) confirm the positive impact of the mean
effective stress and the negative impact of voids on the Gmax.

Fig. 6. Normalized Gmax/p′ deduced from authors’ bender element and resonant column tests
versus void ratio for Tunis soft soils

3.2 Swedish Soft Soils

The mean effective stress and the void ratio were calculated and deduced based on the
collected data related to Swedish soft soils from the test fields in Norrköping, Skå-
Edeby and Bäckebol sites.
The small strain shear modulus versus the mean effective stress and the void ratio
for the different Swedish sites was elaborated. Swedish soft soil stiffness properties
resulting from seismic cone tests were collected and re-interpreted as well and sum-
marized in Table 4.

Table 4. The stiffness parameters of selected Swedish soft soils

Gmax (MPa) e P′(kPa)
3–15 1.8–3.4 17–50
28 L. Touiti and W. Van Impe

3.2.1 Norrköping Site

The small strain shear modulus versus the mean effective stress and the small strain
shear modulus versus the void ratio are illustrated in Figs. 7 and 8.

Fig. 7. Small strain shear modulus versus mean effective stress for sensitive soils in Norrköping

The singular points (A and B) at depths of 5 and 6 m (37.7 and 39.98 kPa), are
indicating a change of the curving slope and a greater stiffness of the soil. This is
probably due to the presence of 3 different layers: the first layer from 3 to 5 m
corresponding to soft clay, the second one, from 5 to 7 m, related to the irregular
varved clay with thin silt layers and the third layer appearing from 7 to 10 m and
corresponding to the regular varved clay.

3.2.2 Skå-Edeby Site

The small strain shear modulus versus the mean effective stress (Fig. 9) illustrates a
uniform and linear relation for each soft clay layer while the standardized small strain
shear modulus versus the void ratio suggests a non-linear but continuous relationship
(Fig. 10), except for 1 isolated point (C) at 6 m depth. This can possibly be explained
by the presence of a thin seam with quite deviating geotechnical characteristics.
 Considerations on the Stiffness of Sensitive Soft Soils 29

Fig. 8. Normalized small strain shear modulus versus mean effective stress for sensitive soils in
Norrköping

Fig. 9. Small strain shear modulus versus mean effective stress for sensitive soils in Skå-Edeby
30 L. Touiti and W. Van Impe

Fig. 10. Normalized small strain shear modulus versus mean effective stress for sensitive soils
in Skå-Edeby

3.2.3 Bäckebol Site

Figures 11 and 12 do show a very straightforward relationship between the small strain
shear modulus, the mean effective stress and the void ratio, with the exception of 2
isolated points (D and E) at 5 and 6 m, respectively. This can possibly be explained by
the presence of a thin seam of other geotechnical characteristics.

Fig. 11. Small strain shear modulus versus mean effective stress for sensitive soils in Bäckebol
 Considerations on the Stiffness of Sensitive Soft Soils 31

Fig. 12. Normalized small strain shear modulus versus mean effective stress for sensitive soils
in Bäckebol

3.3 Empirical Correlations for Soft Soils

Many empirical correlations for the evaluation of Gmax have been proposed in the
literature. The authors would suggest a new empirical correlation for such type of soft
sensitive soils, besides of the standard proposals gathered in Table 5.

Table 5. Selected empirical formulas for calculating Gmax for soft soils
Form of the function Authors Kind of soil
0
ðp paÞ0;5 Hardin (1978) Normally consolidated cohesive
G0 ¼ 625: ð0;3 þ 0;7e2 Þ
soils
0 0;5
G0 ¼ 625: ðpðepaÞ
1;3 Þ
Jamilkowski et al. (1991) NC cohesive soils

G0 ¼ 14; 24:p00;23 :e1;26 Markowska-Lech et al. Cohesive soils

(2007)
G0 ¼ p0;85 :e1;26 Gabryś and Szymańsk (2015) NC cohesive soils
32 L. Touiti and W. Van Impe

As mentioned above, Gmax depends on a number of parameters and the void ratio is
recognized as the second meaningful parameter affecting soil stiffness in combination
with the effective stress level.
The collected data and the tests results indicate the possibility of estimating the
small strain shear modulus in this type of soft sensitive soils from the mean effective
stress and the void ratio. The authors proposed a new correlation which could be
helpful in estimating the shear modulus at very small strain without the need for the
shear wave velocity measurement:

Gmax ðMPaÞ ¼ 0; 53:p0 :e0;93 ð4Þ

Table 6. Laboratory fitting parameters of test soils for authors’ equation

Form of the Coefficient Standard error of Correlation Statistics
function the coefficient coefficient Reduced Adj.
a b c a b c R2 Chi-Sqr R-square
Gmax = a. p′b. ec 0,53 1,00 −0,93 0,15 0,06 0,14 0,96 0,530 0,956

Table 6 shows the values of the coefficients as well as the basic fitting parameters
for the proposed function (Eq. (4)). The proposed correlation explains at least 96% of
the Gmax variation (R2 = 0.96) (when ignoring the isolated points). It indeed very
closely matches the data set. A graphical illustration of the proposed equation is
illustrated in Fig. 13.

Fig. 13. Correlation of the experimental shear modulus with the mean effective stress and void
ratio for sensitive soils.
 Considerations on the Stiffness of Sensitive Soft Soils 33

To examine the reliability of the proposed empirical equation, some standard

empirical equations (Table 5) were also used to verify their suitability to the data set of
each site (Figs. 14, 15, 16 and 17). As can be seen from those Figures, the authors’
proposed expression fits the results of the tests for Tunis, Norrköping, Skå-Edeby and
Bäckebol soft soils well.

Fig. 14. Correlation of the experimental values of normalized shear modulus with the void ratio
for the Norrköping sensitive soil.
34 L. Touiti and W. Van Impe

Fig. 15. The experimental and the estimated values of the normalized shear modulus with the
void ratio for Skå-Edeby sensitive soil

Fig. 16. The experimental and the estimated values of the normalized shear modulus with the
void ratio for Bäckebol soft soil.
 Considerations on the Stiffness of Sensitive Soft Soils 35

Fig. 17. The authors’ experimental and the estimated values of the normalized shear modulus
with the void ratio for Tunis soft soil.

Fig. 18. The normalized small strain shear strength versus the void ratio by indicating the
sensitivity range of each investigated soft soil.
36 L. Touiti and W. Van Impe

The sensitivity is obviously a meaningful parameter affecting the small strain shear
modulus (Fig. 18). Consequently, soft soils with the same void ratio and mean effective
stress but different sensitivity values would have different stiffness.
The proposed equation fits the data related to Tunis, Norrköping and Bäckebol soft
soils a lot more than those related to Skå-Edeby soft soil as it is less sensitive (Figs. 14,
15, 16 and 17).
Obviously, we are aware that the proposed lower bound correlation requires further
verification by performing further research taking into account other sensitive soils and
more diverse sites.

4 Conclusion

The small strain shear modulus (Gmax) of soft soils from Tunis (Tunisia) and various
sites in Sweden was measured by means of resonant column, Bender element and cone
seismic tests. Based on these results, some conclusions can be drawn:
– In all the analyzed cases, the values of Gmax do increase with an increase of the
mean effective stress.
– The results clearly indicate that the Gmax increases with a decreasing void ratio for
most of the specimens, which is in agreement with results in literature.
– The authors proposed a suitable correlation pattern (R2 = 0.96), enabling to cover
up to 96% of the Gmax variation.
– The soft soil sensitivity (St) is also a meaningful parameter when describing the
small strain shear modulus of soft soils, however not fully and well enough
understood.

Acknowledgements. The first author would like to acknowledge that all laboratory tests were
conducted at the Geotechnical Institute for Experimental Models (ISMGEO Laboratory) in Italy
and thank all the staff for their assistance. We would also like to thank Prof. Larsson for
permitting us to use his experimental data about the Swedish soils.

References
Schnaid, F.: Geo-characterization and properties of natural soils by in situ tests. In: Proceedings
of 16th International Conference on Soil Mechanics and Geotechnical Engineering, Osaka,
pp. 3–45 (2005)
Stokoe II, K.H., Rathje, E.M., Axtell, P.J.: Development of an in-situ method to measure the non
linear shear modulus of soil. In: Proceedings of the 16th International Conference on Soil
Mechanics and Geotechnical Engineering, Osaka, pp. 751–754 (2005)
Nash, D.F.T., Lings, M.L., Pennington, D.S.: The dependence of anisotropic go shear moduli on
void ratio and stress state for reconstituted Gault clay. In: Lancellotta&LoPresti, J. (eds.)
Prefailure Deformation Characteristic of Geomaterial, pp. 229–238. Rotterdam (1999)
Benz, T.: Small-Strain Stiffness of Soils and its Numerical Consequences. Ph.D. thesis,
University of Stuttgart (2007)
 Considerations on the Stiffness of Sensitive Soft Soils 37

Schneider, J.A., Hoyos Jr., L., Mayne, P., Macari, E.J., Rix, G.J.: Field and laboratory
measurements of dynamic shear modulus of Piedmont residual soils. ASCE Geotech. Spec.
Publ. 92, 12–25 (1999)
Wolski, W., Lipinski, M.J.: Site characterisation for geotechnical and environmental purposes.
In: Proceedings of the XIIIth Danube-European Conference on Geotechnical Engineering,
Ljubljana, pp. 129–150 (2006)
Kaâniche, A., Inoubli, M.H., Zargouni, F.: Développement d’un SIGG et réalisation d’un atlas
géotechnique électronique. Bulletin Anglai deGéologie et d’environnement. Springer,
Heidelberg (2000)
Touiti, L., Bouassida, M., Van Impe, W.: Discussion on the Tunis soft clay sensitivity. Geotechn.
Geol. Eng. J. (2009). doi:10.1007/s10706-009-9263-2
Markowska-Lech, K., Lech, M., Szymanski, A.: Estimation of shear modulus from seismic tests
on Pliocene clays. In: Proceedings of International Conference on Numerical Models in
Geomechanics, London, pp. 153–157 (2007)
Larsson, R., Mulabdic, M.: Shear moduli in scandinavian clays. Swedish Geotechnical Institute,
report n°. 40, Linkoping, pp. 40–94 (1991)
 A Novel Field Device for the Measurement
of Soil Collapsibility

Mehdi Mokhberi(&) and Seyed Ayuob Rafieean

Islamic Azad University, Tehran, Iran

Abstract. Soil collapsibility or soil swelling is typically measured in the labo-

ratory by using a single or double consolidation test. Such a test requires time and
is costly: its stages include preparing the samples, transferring them to the lab-
oratory, and conducting related tests. In most cases, geotechnical engineers have
to carry out experiments during the stage of site investigation to evaluate the soil
in terms of collapsibility. However, soil collapsibility might be easily estimated
by a simple desert test. The device designed, developed, and used by the authors
of the present study has the ability to determine the soil’s collapsibility and offers
its primary percentage. The experiments that were conducted on the downstream
alluvium of Seevand Dam, with high collapsibility, to assess the performance of
the device indicated that the device is able to appropriately evaluate the col-
lapsibility of soil. Furthermore, comparing the results of these experiments with
those of experiments using the consolidation method confirm the accuracy of the
findings present study, with a high percentage. Given the probability of dis-
turbing soils in laboratories, it might be contended that this device obtains results
that are more valid than laboratory results.

Keywords: Collapsibility
Field test
Unsaturated soil
Double

consolidation
Seevand Dam

1 Introduction

Collapsible soils are widespread in Iran and the world. These types of soils are usually
known for their low natural humidity percentage and also low unit weight. Structures
placed on collapsible soils, in the case of soil saturation, might settle unexpectedly. In
the planning of most engineering structures, it is essential to pay special attention to
soil collapsibility as it might destroy the foundation of a structure, destroy a dam, or
produce road subsidence.
Furthermore, the existence of other technical buildings in these areas, and also
given the development of cities and the necessities of the development of big cities,
building residential areas, water and waste water pipes, etc. on these types of soil is of
great importance (Derbyshire 2001). In the process of the development of soil
mechanics science, which formally started in 1936, it was in the late 1960’s and during
the seventh and eighth international soil mechanics conferences that the investigation
of the scientific principles of unsaturated soil was taken into consideration. In 1959,
one of the first criteria of collapsibility based on dry density was offered by Clevenger
(1958). After that, in 1962, Gibbs and Bara (1962) suggested using the unit weight of

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_4
 A Novel Field Device for the Measurement of Soil Collapsibility 39

dry density and soil smoothness limit as criteria for separating collapsible and
non-collapsible oils. The trend of offering criteria for distinguishing collapsible and
non-collapsible soils continued during the 1970’s, 1980’s, and 1990’s, and different
benchmarks were introduced by Denisov (1963), Feda (1966) and Fookes and Best
(1969). In a Russian conference on soil mechanics and foundation engineering, the
instability of a collapsible soil’s skeleton was investigated and discussed. Jennings and
Knight (1957) suggested the double consolidation test to assess the behavior of soil
under conditions of saturation and loading with different levels of stress. Since the late
1980’s, considerable attempts have been made to offer mathematical and computer
models. In these models, the attention was on modeling the behavior of collapsible
soils and the distribution process of these soils’ porosity (Handy 1973; Houston et al.
1988). Since then, many studies have been carried out to estimate collapsibility,
numeric modeling, failure model offering, etc. Similarly, different studies have also
been conducted in Iran about the behavior of soils. For example, Habibagahi and
Mokhberi (1998) investigated the effect of humidity on collapsibility estimation by
modifying the relation offered by Duncan and Chang (1970) and also involving
important parameters. They could offer a new model for soil failure. Tarantino et al.
(2005) attempted to model the behavior of collapsible soils through laboratory and
computational methods, especially the finite element method and using remolded
specimens of loose soil and computer programs continue to be used. They have offered
related research studies related to conducted experiments in their book.
The common methods of determining soil collapsibility require taking disturbed
and non-disturbed soil samples. Such methods are not only time consuming and costly,
but also change the behavior of the soil. Additionally, recreation of collapsible soils in
the laboratory needs experience and special accuracy. The present research study is, in
fact, an attempt to provide a small portable device to determine the collapsibility of
soil. The soil collapsibility field identification device offered in the present study is used
to evaluate the soil’s collapsibility in non-laboratory and field contexts. The device is
able to evaluate soils in terms of their potentials.

2 Materials and Method

2.1 Development of Device
The offered device consists of two sections, one for collapsibility and the other for
loading. Figure 1 presents the whole of the device. In brief, the main parts of the device
are as follows:
1. Stabilizer panel: This is a square rigid metal panel with sides of 30 cm and
thickness of 1 cm, made from iron. There is a hole whose diameter is equal to the
outer diameter of the cylinder in the limiting center of the cylinder within which the
sampling cylinder is inserted.
2. Sampling cylinder: This is a hollow cylinder made from steel and is completely
resistant to the materials of the soil. Its length is 13 cm.
3. Porous stone: This is placed into the sampling cylinder and above the soil sample.
40 M. Mokhberi and S.A. Rafieean

Fig. 1. Schematic design and image of the offered device to assess the soil collapsibility

4. Loading panel: This is made from steel and its diameter is equal to that of the
porous stone. It is completely resistant to the soil materials and is an important point
in the design of the device.
5. Moving loading bar: This is a solid metal bar of steel and is completely polished so
that it can easily move as the soil settles.
6. Device Weights: The weight is imposed through the metal weights placed on the
moving loading bar.
7. Sample Thickness Change Measuring Gauge: The gauge used in this device has
0.01 mm accuracy. There is a magnet in the foot of this gauge that might be
connected to the cylinder limiting panel.
The information obtained from this device is used to assess the soil’s collapsibility and
also to calculate the indices that are about the relation between effective the stress and
the strain or vacuum symptom. Additionally, the information resulting from this device
might be used to increase or estimate speed, determine unsteady settlement and total
settlement of a structure or collapsible soil piles.

2.2 Method
First, the device was placed and leveled in the intended area. Then, the user gradually
puts pressure steadily on the two handles inserted on the cylinder stabilizer panel so
that the cylinder penetrates into the soil. After that, filter paper and the porous stone
were placed on the cylinder. Next, the primary load is done for 5 kPa. Then, within a
5-minute time period after the primary loading, further loading steps are gradually done
at intervals of one hour in natural humidity conditions. The loading process continues
 A Novel Field Device for the Measurement of Soil Collapsibility 41

until the settlement stops. The loading steps should be 10, 25, 50, 100, 200 kPa. Before
each step, any change of form is observed. The amount of stress on the sample before
its saturation is determined based on conditions. One hour after reaching the intended
stress, any changes of the soil form in natural humidity is measured. After that, the
sample is drowned by pouring water into the cylinder. Water steadily enters into the
porous stone through the three holes on the panel. In this way, the whole of the sample
is steadily saturated downwards in line with natural patterns. After drowning the
sample, the sample form changes are read and observed at time intervals of approxi-
mately 0.1, 0.25, 0.5, 1, 2, 4, 8, 15, 30, and 60 min. It should, however, be mentioned
that in soils with high penetration, collapsibility might happen quickly and makes
reading the time difficult. Adding water to the sample should be in a way that it is
saturated downwards and air is not imprisoned in the soil. After adding water into the
cylinder and drowning the sample, the necessary time for putting on a load would
continue for one day or until the primary consolidation (based on D2435 standard) is
finished. The degree of sample form change is read in the above-mentioned time
intervals. Finally, the information obtained from the device is used to determine the
collapsibility percentage of non-saturated soils after being drowned.

2.3 Evaluation of Test Results

2.3.1 Results of Test Device by Using the Device
In this section, the results obtained from the tests using the developed device and those
of an experimental consolidation device are presented. These tests were done on three
points of soil in the downstream area of Seevand Dam located 100 km to the north of
Shiraz, which is known to be collapsible soil. The area soil is of CL type. Table 1
presents the soil consolidation properties in this area.

Table 1. The consolidation properties of soil of the studied area

Station e0 ef x0 ð%Þ Cs Cc Gs
P1 0.95 0.23 5 0.06 0.31 2.56
P2 0.88 0.19 6 0.05 0.3 2.56
P3 0.99 0.25 5 0.06 0.33 2.56

To find the soil collapsibility using the device, the soil changes were recorded and
measured after placing the device and putting the load on the soil. Figures 2a to c and
Table 2 show the soil changes before saturation, during saturation, and after saturation.

2.3.2 Laboratory Consolidation Device Results

The single consolidation tests were carried out based on standard (ASTM D 5333-03)
to determine the soil’s collapsibility. In this experiment, a sample of soil is placed in the
oeodometer device and vertical pressure is increased until reaching the probable
pressures of real earth. At this loading level, water is injected into the device, the
sample is saturated, and resulting changes are recorded.
42 M. Mokhberi and S.A. Rafieean

Fig. 2. Comparison of soil collapsibility from offered device (left) and single consolidation
tests (right)

Table 2. Comparison of soil collapsibility from offered device and single consolidation test
Station Collapse potential from oedometer Collapse potential from proposed
test (%) device (%)
P1 6.9 7.8
P2 6.1 7.2
P3 5.8 7.1
 A Novel Field Device for the Measurement of Soil Collapsibility 43

The consolidation graphs of three different natural soils have been presented below.
It should, however, be pointed out that in the consolidation tests, stresses of 25, 50, and
100 kPa were loaded in natural humidity status and then the samples were consolidated
with stresses of 100, 200, 400, and 800 kPa. After that, the loading process was done to
the weight of one kilogram on one cubic centimeter. Figures 2d to f and Table 2 show
the related results.

2.3.3 Comparison of Soil Collapsibility Determined by the Two Methods

The soil collapsibility was determined by using (Lutenegger and Saber 1988) relation (1).

De
I=  100 ð1Þ
1þe

As Fig. 2 and Table 2 show, the collapsibility of the reconstructed sample is

approximately 6.3 in the laboratory and 7.4 in the field, which implies the correct
performance of the device.

3 Conclusions

The comparison of the results obtained from the developed device and those from an
experimental consolidation device shows that the developed device has appropriately
assessed the collapsibility of the area soil. Additionally, due to its description of the soil
collapsibility, it has a better and more complete capability than other comparable
devices. Therefore, it could be used in similar projects of the district as a suitable
criterion to document the soil collapsibility. In fact, the soil collapsibility field device is
a device that has removed the technical problems of existing related devices and could
simulate the real and natural conditions of soil and as a result, the obtained results are
completely coincident with the natural patterns of the soil. This device, in essence,
increases the accuracy and output of the work considerably.

References
Clevenger, W.A.T.: Experiences with loess as foundation material. Trans. Am. Soc. Civ. Eng.
123(1), 151–169 (1958)
Denisov, N.Y.: About the nature of high sensitivity of Quick clays. Osnov. Fudam. Mekh. Grunt.
5, 5–8 (1963)
Derbyshire, E.: Geological hazards in loess terrain, with particular reference to the loess regions
of China. Earth Sci. Rev. 54(1), 231–260 (2001)
Duncan, J.M., Chang, C.-Y.: Nonlinear analysis of stress and strain in soils. J. Soil Mech. Found.
Div. (1970)
Feda, J.: Structural stability of subsident loess soil from Praha-Device. Eng. Geol. 1(3), 201–219
(1966)
Fookes, P.G., Best, R.: Consolidation characteristics of some kate Pleistocene periglacial
metastable soils of East Kent. Q. J. Eng. Geol. Hydrogeol. 2(2), 103–128 (1969)
44 M. Mokhberi and S.A. Rafieean

Gibbs, H., Bara, J.: Predicting surface subsidence from basic soil tests. In: Field Testing of Soils.
ASTM International (1962)
Habibagahi, G., Mokhberi, M.: A hyperbolic model for volume change behavior of collapsible
soils. Canad. Geotech. J. 35(2), 264–272 (1998)
Handy, R.L.: Collapsible loess in Iowa. Soil Sci. Soc. Am. J. 37(2), 281–284 (1973)
Houston, S.L., Houston, W.N., Spadola, D.J.: Prediction of field collapse of soils due to wetting.
J. Geotech. Eng. 114(1), 40–58 (1988)
Jennings, J., Knight, K.: The additional settlement of foundations due to a collapse of structure of
sandy subsoils on wetting. Proceedings (1957)
Lutenegger, A.J., Saber, R.T.: Determination of collapse potential of soils (1988)
Tarantino, A., Romero, E., Cui, Y.J.: Advanced experimental unsaturated soil mechanics. In:
Proceedings of the International Symposium on Advanced Experimental Unsaturated Soil
Mechanics, Trento, Italy, 27–29 June 2005. CRC Press (2005)
 Moisture Variation in Expansive Subgrade
Through Field Instrumentation
and Geophysical Testing

Asif Ahmed1(&), MD Sahadat Hossain1, Mohammad Sadik Khan2,

Kelli Greenwood3, and Aya Shishani3
1
Civil Engineering, University of Texas at Arlington, Texas, USA
[email protected], [email protected]
2
Civil and Environmental Engineering, Jackson State University,
Mississippi, USA
[email protected]
3
Department of Civil Engineering, University of Texas at Arlington,
Texas, USA
{kelli.greenwood,aya.shishani}@mavs.uta.edu

Abstract. Seasonal climatic variations in subgrade soil affect pavement

responses and can reduce pavement serviceability. In addition to causing
shrinkage and swelling in expansive subgrade, variations in moisture suction
may alter the material properties of soil, which ultimately affect pavement
performance. The current study monitored the seasonal variation of subgrade
moisture content, rainfall, and pavement deformation of a section over State
Highway 342, in Dallas, Texas. Moisture sensors were installed at different
depths up to 4.5 m. The soil was found to be highly plastic clay (CH) in the
selected site. In addition to acquiring real-time moisture data from sensors,
geophysical testing was also conducted on the slope of the pavement. Electrical
Resistivity Imaging (ERI) was carried out at the slope of the instrumented
pavement section to observe the moisture flow at the edge of the pavement.
Based on the field monitoring data, moisture variation ranged from 5% to 14%,
with higher moisture contents correlating with rainfall events. While moisture
sensors provided point information, resistivity imaging yielded a continuous
portrayal of subsurface moisture flow. Furthermore, rainfall-associated defor-
mation of the pavement was monitored. Based on the monitored data, it was
observed that pavement deformation varied with rainfall. A total deformation of
38 mm was recorded over the monitoring period.

1 Introduction

The engineering properties of expansive soils are highly dependent on moisture content
changes in the active zone. Such soils undergo volumetric deformation that gradually
damages essential infrastructure such as foundation slabs, bridges, roadways, and
residential homes. Present in both humid and arid/semi-arid environments, expansive
soils cover nearly a quarter of the area of the United States (Nelson and Miller 1992).
Annually, expansive soils alone incur more financial losses to US property owners than

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_5
46 A. Ahmed et al.

earthquakes, flood, hurricanes, and tornadoes combined (Jones and Jefferson 2012). In
a typical year, the associated financial losses can be as high as 15 billion dollars (Jones
and Jefferson 2012). Their low stiffness, light loading, and extended presence over the
country render pavements especially susceptible to deterioration caused by expansive
subgrades. Continuous volumetric deformation of the problematic soils increases the
pavement’s roughness, resulting in reduced serviceability. Consequently, a better
understanding of the causes of pavement distress and the behavior of expansive soils is
necessary for acceptable pavement performance and design (Sebesta 2002).
Climatic factors, especially moisture variations, induce substantial changes in the
structural characteristics of pavement systems. Fluctuations in moisture content can
have major impacts on the long-term performance of pavements. Principal sources of
moisture variation are rainfall, intrusion from cracks, freeze-thaw cycles, leakage, and
evapotranspiration. Moisture variation affects the hydraulic conductivity, shear
strength, chemical diffusivity, specific heat, and thermal conductivity of a soil
Lu 2015). Furthermore, the success or failure of a pavement system is dependent on the
support provided by subgrade layers. An increase in moisture content has been shown
to decrease the resilient modulus, which quantifies the support that the subgrade can
offer (Mehrota 2011). Moreover, variation in moisture content may cause swelling and
shrinkage in highly plastic subgrade soils. Surficial cracks are generated due to this
behavior and require departments of transportation to increase expenditures related to
pavement maintenance (Zapata and Houston 2008). Because expansive subgrade is
seldom identified as the source of pavement failure, maintenance routines typically
only consist of roadway surface treatment. Consequently, surface roughness can
reappear a short period of two years and six months later. For 18 out of 52 Texas
Department of Transportation (TxDOT) districts, such problems are commonplace
(Wanyan et al. 2010).
To consider the possible effects of moisture in the design procedures, many
researchers have conducted field-based analyses to relate subgrade characteristics with
environmental factors. Bayomy and Salem (2004) instrumented five different sites in
Ohio. After monitoring for a period of five years (1999–2003), the researchers reported
the presence of seasonal moisture variation. Manosuthkij (2008) monitored four
instrumented sites in Texas and found that when the mean moisture content, or the
difference between the maximum and minimum moisture content of each month, was
greater than 20%, edge cracking was likely to occur. In Ohio, Heydinger (2003)
obtained both seasonal variation and temporal variation due to precipitation at two
instrumented sites.
In addition to installing moisture sensors that monitor moisture variation, geo-
physical investigation can also be significant when observing the moisture flow in
subgrade. ERI is widely used in hydrogeological, environmental, and geotechnical
research due to its potential to reveal the subsurface image (Aizebeokhali 2010). It is a
convenient method for determining spatial variations of moisture in subsoil (Kibria and
Hossain 2012). In the geotechnical and geomorphological fields, ERI is gaining pop-
ularity for the user-friendly data acquisition system inherent to the method. The tech-
nology is used in geotechnical applications to determine unknown foundation depths,
investigate slope and foundation failures, detect cracks and the existence of sinkholes,
record moisture variation in pavement base materials, and track soil movement with
 Moisture Variation in Expansive Subgrade through Field Instrumentation 47

accurate results. ERI can measure both the horizontal and vertical variability of soil
arrangement over several meters of a perspective area (Tabbagh et al. 2000).
The resistivity profile provides a comprehensive interpretation of the subsurface
condition. The most interesting and important advantage of using ERI is that the
resulting resistivity profile offers a clear picture of the moisture distribution within the
test area. Accordingly, numerous studies have utilized ERI to study moisture distri-
bution in soil. Benderitter and Schott (1999) observed the water content variation in an
unsaturated soil and the effect of rainfall by observing the short-term variation of
resistivity. Clarke (2006) measured subgrade resistivity over nine months to determine
the active zone, edge moisture variation distance, and long-term equilibrium moisture
beneath the covered area. The author also monitored the seasonal fluctuations of
moisture at the edge. The study found that the most extreme changes occurred at the
surface and diminished with depth. Kean et al. (1987) studied the moisture migration in
the vadose zone using resistivity tomography. Finally, Zhou et al. (2001) performed
spatial and temporal monitoring of soil water content using electrical resistivity
tomography.
During the current study, both dielectric sensors and two-dimensional electrical
resistivity imaging techniques were utilized to study the moisture variations in
expansive subgrade and pavement slope. The objective of the study was to determine
the variation of subgrade moisture due to both seasonal variation and precipitation
events. The study was conducted in a state highway designated as SH 342 in Ellis
County, Texas, USA. Sensors were installed up to 4.5 m depth beneath the driving
lanes. ERI was conducted at the slope of the pavement during different seasons. A rain
gauge was also installed to record the precipitation events.

2 Project Background and Site Description

The effects of moisture on soil depend on the soil properties. Based on accessibility,
distance from the flooding zone, and TxDOT recommendations, a section on State
Highway (SH) 342, a two-lane hot mix asphalt road, was selected for investigation.
Two 3.3 m (11 ft.) wide lanes with a 3.3 m (11 ft.) wide shoulder on each side of the
road were present at the site. Details of the chosen section can be seen in Fig. 1. The
project site was located near Lancaster, Texas, on the border between Ellis and Dallas
counties. While the pavement was fairly level, edge cracking and other pavement
distresses were observed. Adjacent to the pavement, the ground was covered with
grass. Additionally, dense trees could be observed on both sides approximately 6 m
(20 ft.) from the pavement. To the west, a rail line ran parallel to the roadway. Further
northeast, the road overpassed Bear Creek, while a residential area was located a short
distance to the southeast. No water body was directly observed nearby.
48 A. Ahmed et al.

Fig. 1. State Highway (SH 342) in Dallas, Texas, USA

3 Field Instrumentation

A comprehensive field instrumentation plan was implemented at SH 342 to monitor the

moisture variations in the subgrade soil. Decagon 5TM moisture sensors, an
ECRN-100 high resolution tipping-bucket rain gauge, an EM50 data logger, and a
horizontal digitilt inclinometer probe were installed at the site. The moisture sensors
and the 85 mm horizontal inclinometer casings were installed from the centerline to the
edge of the northbound lane, as seen in Fig. 2.
Continuous monitoring of moisture content, rainfall, and vertical deformation were
obtained from field instrumentation. Data loggers were programed to take hourly read-
ings of moisture content, and rainfall data. Furthermore, the pavement site was visited
monthly to obtain inclinometer readings. Fifteen moisture sensors and six 3 m (10 ft.)
 Moisture Variation in Expansive Subgrade through Field Instrumentation 49

long horizontal inclinometer castings were installed. The depth of the sensors ranged
from 1.2 m (4 ft.) to 4.5 m (15 ft.) below the pavement.

(a) (b)
Fig. 2. (a) Instrumentation Plan, (b) Instrumentation at the site SH 342

4 Soil Properties

The geotechnical properties of soil samples collected from SH 342 during field
instrumentation were obtained through laboratory testing. Upon investigation of the
particle size distribution, the samples contained very fine subgrade soil. All samples
were composed of over 85% clay. The liquid limits ranged between 50% and 64%,
while plasticity ranged between 28% and 42%. As specified by the Unified Soil
Classification System (USCS), sieve analysis and Atterberg limits results indicated that
the soil samples were high-plastic clay (CH). Specific gravity ranged between 2.68 and
2.72, with an average of 2.70. Optimum moisture content was determined to be 22%,
and dry density at the optimum moisture content was found to be 18.9 kN/m3.

5 Moisture Variation in Subgrade by Sensor

Subgrade moisture sensor data was recorded along the center and edge of pavement at
SH 342 from March 2014 to April 2016. Moisture content variation was determined
through analysis of the recorded moisture data. The following discussion presents
interpretable results obtained from the center borehole, which were representative of
data obtained from remaining sensors at the edge.
To obtain moisture data, sensors were installed in the center borehole at depths of
1.2 m (4 ft.), 2.4 m (8 ft.), 3.6 m (12 ft.), and 4.5 m (15 ft.). Moisture variation in the
center borehole is shown in Fig. 3. While rainfall was immediately detected by the
shallower sensors, the deepest sensor (TM 15, located at 4.5 m [15 ft.] depth) observed
no detectable moisture content changes in response to rainfall. TM 15 recorded an
50 A. Ahmed et al.

30 140

25 120
Moisure Content (%)

100

Rainfall (mm)
20
80
15
60
10
40
5 20

0 0
07-Mar-14 09-Jul-14 10-Nov-14 14-Mar-15 16-Jul-15 17-Nov-15 20-Mar-16

Rainfall 1.2 m 2.4 m 3.6 m 4.5 m

Fig. 3. Moisture variation across center borehole (BH-5)

average soil moisture content of 19%, which corresponded to the point of saturation at
the depth of 4.5 m (15 ft.). Moisture contents stabilized to equilibrium moisture con-
tents at all depths. Respectively, equilibrium moisture contents for TM 4, TM 8, and TM
12 were 11%, 6%, and 15%. Moisture readings immediately following rainfall events
rose 4% to 15% in amplitude, with increases limited to the temporary saturation point at
the respective depths. Shallow sensors experienced greater increases in moisture con-
tent. After draining, moisture contents returned to equilibrium. Temporary changes in
moisture content are important to study in the case of expansive clay soils. Brief fluc-
tuations in moisture can significantly affect the volumetric deformation of the soil.
The monthly variation in total rainfall and maximum moisture content recorded by
the sensors is shown in Fig. 4. At shallow depths, the rises and falls in maximum
moisture content mirrored the behavior of rainfall. That is, when precipitation

25 160
140
20
Moisture Content (%)

120
Total Rainfall (mm)

15 100
80
10 60
40
5
20
0 0
Mar-14 Jun-14 Sep-14 Dec-14 Mar-15 Jun-15 Sep-15 Dec-15 Mar-16

Rainfall 1.2 m 2.4 m 3.6 m 4.5 m

Fig. 4. Maximum moisture variation across center borehole (BH-5)

 Moisture Variation in Expansive Subgrade through Field Instrumentation 51

experienced a decline during the dry season, maximum moisture content also dropped.
For example, the dry periods from May to September 2014 and May to August 2015
saw sensors at 2.4 m (8 ft.) experience a net decrease of up to 15% in recorded
moisture content. Conversely, rising levels of rainfall during the wet season produced
increasing magnitudes of moisture content. For instance, the accumulation of nearly
91 mm (3.6 in.) of rainfall between March and May 2014 returned a net increase of
15% in moisture content. Due to the expansive nature of clayey subgrade, the pavement
at SH 342 could be expected to deform seasonally in response to the rises and falls in
moisture content.
While subgrade moisture content varies in response to rainfall events, moisture
content may also vary due to seasonal effects. Seasonal contributions to moisture
variation can depend on the soil compaction, soil type, ground water table, and initial
moisture content. Consequently, some locations observe seasonal variations while
others do not. After five years of monitoring in Ohio, Heydinger (2003) reported that a
seasonal effect on moisture variation existed at the study site. In Texas, Manosuthikij
(2008) found a seasonal contribution to moisture variation at sites in Fort Worth and
Houston, whereas the study site in San Antonio did not experience a seasonal trend. To
ascertain if a seasonal pattern existed at SH 342, electronic resistivity imaging
(ERI) was performed across the grass slope adjacent to the pavement. This method of
geophysical testing provides a continuous picture of moisture content distribution.
Comparison of ERI results obtained from a monthly basis could help conclude whether
moisture variation at the site underwent seasonal changes.

6 Theory of ERI

ERI is an active geophysical method which measures the electric potential differences
at specific locations after injecting a controlled electric current at other locations. By
controlling the current injected in an entirely homogeneous half-space, a resistivity
value can be calculated for the subsurface by measuring the resulting electric potential
difference. Figure 5 illustrates the concept of subsurface electric current flow and the
influence of subsurface heterogeneities.
The ERI method follows Ohm’s law in that resulting potential differences are
measured by transferring artificially generated currents to the soil. The electrical
resistivity depends on several factors of the subsurface condition. Main factors
influencing resistivity include the size, shape, arrangement, mineralogy, voids distri-
bution, porosity, connectivity between the particles, degree of saturation, concentration
of soluble minerals, and temperature of the subsurface soil.
For a simple soil body, the resistivity q (ohm-m) is defined as:

q ¼ RðA=LÞ ð1Þ
52 A. Ahmed et al.

(a) (b)
Fig. 5. (a) Variations in subsurface electric current density (Aizebeokhai 2010) (b) Relation
between resistance and resistivity (Tabbagh et al. 2000)

Where, R is the electrical resistance, L is the length of the cylinder (m) and A is the
cross-sectional area (m2). The electrical resistance of a cylindrical body is defined by
Ohm’s law:

R ¼ V=I ð2Þ

Where, V is the potential difference measure in Volt, and I is the current in Ampere.

7 Geophysical Testing by Resistivity

While moisture sensors returned distinct moisture content recordings, further infor-
mation regarding moisture variation was desired. Thus, geophysical testing in the form
of resistivity imaging was utilized. Geophysical testing provides a continuous picture of
the moisture variation beneath the pavement. Since sensors were not installed beneath
the grass slope adjacent to the pavement, geophysical testing allowed for determination
of the moisture variation beneath the slope beside the pavement.
Instrumentation required for ERI included a super-sting R8/IP resistivity meter
manufactured by Advanced Geosciences Institute (AGI) and a switch box with a
12-volt battery for power supply. RI testing was conducted with 28 electrodes placed at
0.9 m (3 ft.) intervals, resulting in a test line of 24.7 m (81 ft.). Improved resolution in
the horizontal and vertical directions was obtained by employing a dipole-dipole array
(Manzur et al. 2016). The schematic of the layout of testing and site photos are
presented in Fig. 6. The collected data were analyzed in the Earth Imager 2D software
(AGI 2004), which uses a forward modeling technique to calculate apparent resistivity
values from the field data. The software uses non-linear least-squares optimization
technique to yield the final output as 2D resistivity image of the subsurface.
 Moisture Variation in Expansive Subgrade through Field Instrumentation 53

(a)

(b)
Fig. 6. (a) Field setup for resistivity (b) schematic at slope of pavement

8 Moisture Variation in Subgrade by Resistivity

ERI provided a continuous picture of the moisture content distribution at different

levels and locations of the subgrade soil. An inverse relationship between resistivity
and saturation was used to interpret the qualitative data presented in Fig. 7. High
readings of resistivity corresponded to lower moisture contents, while low readings of
resistivity translated to higher moisture contents. The scale from red to blue in Fig. 7
indicated levels of high and low resistivity values, respectively. ERI performed during
the dry season (from May to October) returned high resistivity readings and produced
the image depicting moisture content distribution shown in Fig. 7(a). Conversely, ERI
testing during the wet season (from November to April) produced Fig. 7(b) with low
resistivity values. ERI conducted immediately following a rainfall event revealed the
54 A. Ahmed et al.

(a)

(b)

(c)

Fig. 7. Resistivity variation in (a) dry period, (b) wet period, and (c) moisture intrusion through
edge after rainfall

Oct-15 Dec-15 Feb-16 Mar-16

Apr-16 May-16 Jun-16 Jul-16

Resistivity (Ohm-m)
0 5 10 15 20 25
0

1
Low resistivity in
wet period

2
Depth (m)

3
High resistivity in
dry period
4

Change of
5
resistivity is lower
after 3 m (10 ft.)
6

Fig. 8. Variation of resistivity across depth in central section

 Moisture Variation in Expansive Subgrade through Field Instrumentation 55

presence of cracks or holes in the pavement. An apparent edge crack appeared in May
2016, as seen from the region of low resistivity developing in Fig. 7(c). Low resistivity
values where higher values were previously recorded can indicate the presence of
cracks, potentially serving to determine which areas of pavement need the most
maintenance.
Results of resistivity variation across depth produced the plot shown in Fig. 8.
Resistivity values ranged between 7 to 13 ohm-m between the months of November
and April, translating to high readings of moisture content. Thus, the period was
identified as the wet season. In contrast, the period between May and October returned
resistivity values as high as 23 ohm-m, corresponding to the dry season. Geophysical
testing revealed the presence of a seasonal trend. Given the relationship between
resistivity and moisture content, a seasonal effect on resistivity suggested that a sea-
sonal effect on moisture content variation also existed. Below a depth of 3 m (10 ft.),
resistivity did not undergo significant fluctuations. This finding corroborated findings
obtained from other studies conducted in the same area, which identified 3–3.66 m
(10–12 ft.) as the depth of the active zone (Khan et al. 2015).

9 Deformation Analysis

Both temporal and seasonal changes in subgrade moisture content induce volumetric
deformation in expansive clay soils. Pavement performance can be negatively affected
as a response to subgrade deformation. To monitor the pavement deformation at SH
342, monthly inclinometer results were obtained over a period of two years and are
presented in Fig. 9.
The inclinometer profile showed that swelling and shrinking of the expansive
subgrade caused up to 38 mm (1.5 in.) of vertical movement across the pavement.
A cyclical trend in deformation was readily observed by plotting pavement deformation

70
60 Wet Period’14 Dry
50 Wet Period’14 Period’15
Deformation (mm)

40
30
20
10
0
3 5 7 9 11 13 15
-10
Swelling’16 Baseline Swelling’15
-20 (April’14)
-30
Distance Along Pavement (m)

Fig. 9. Variation of deformation along distance in central section

56 A. Ahmed et al.

(a)

Rainfall Edge Center

50 120
Deformation (mm)

40 100

Rainfall (mm)
30 80
20 60
10 40
0 20
-10 0
Jan-14 Apr-14 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Mar-16
Time (Months)

(b)

Fig. 10. (a) Seasonal trend in deformation, (b) movement in response of rainfall events

of edge over time (Fig. 10a). Seasonal effects were further examined by including
precipitation in a plot of deformation over time (Fig. 10b). Changes in pavement
elevation were seen to correlate with rainfall events. For example, the center of the
pavement experienced 38 mm (1.5 in.) of swelling as the result of 114 mm (4.5 in.) of
rainfall during the wet season from March to May 2014. After the summer of 2014,
during which precipitation levels declined, the pavement dropped by almost 25 mm
(1 in.). Another pattern of swelling and shrinkage was observed in October 2015, when
the pavement dropped 38 mm (1.5 in.) in response to a decrease in rainfall. However,
pavement elevation rose again as precipitation levels increased upon the arrival of the
wet season. Pavement deformation mirrored rainfall patterns which followed a seasonal
trend. Thus, pavement deformation was also concluded to adhere to a seasonal trend.

10 Conclusions

The current study was conducted to determine the variation in subgrade moisture
content resulting from seasonal effects and rainfall events. A two-lane road identified as
SH 342 in Dallas, Texas, USA was instrumented with moisture sensors to record
moisture content and a rain gauge to record precipitation. Geophysical testing using
 Moisture Variation in Expansive Subgrade through Field Instrumentation 57

electrical resistivity imaging was conducted to determine the presence of a seasonal

effect on subgrade moisture variation. Pavement deformation was recorded using a
horizontal inclinometer installed at the site. The observed results are summarized
below.
• Laboratory testing indicated that subgrade soil at SH 342 was made of highly plastic
clay.
• Moisture sensors installed in the pavement captured temporary net increases of 4%
to 15% moisture content following rainfall events.
• Electrical resistivity imaging performed across the grass slope adjacent to the
pavement captured the presence of a seasonal effect on moisture variation. The
detection of water intrusion also revealed the development of cracks in the
pavement.
• Subgrade shrinkage and swelling was observed to follow a seasonal trend. Wet
periods (from November to April) caused a rise in pavement elevation, while dry
periods (from May to October) resulted in a drop. Up to 38 mm (1.5 in.) of
deformation was recorded.
The observation of moisture variation beneath pavement can be linked to defor-
mation behavior of the pavement built on expansive subgrade. A complete deformation
model in response to the moisture variation can be effective to predict the pavement
performance. It will ultimately lead to better pavement preservation technique.

Acknowledgments. The project was funded by Texas Department of Transportation (TxDOT)

and is greatly acknowledged.

References
Advanced Geosciences, Inc.: Instruction Manual for Earth Imager 2D Version 1.7.4. Resistivity
and IP Inversion Software. Advanced Geosciences, Inc., Austin (2004)
Aizebeokhai, A.P.: 2D and 3D geoelectrical resistivity imaging: theory and field design. Sci. Res.
Essays 5(23), 3592–3605 (2010)
Bayomy, F., Salem, H.: Monitoring and modeling subgrade soil moisture for pavement design
and rehabilitation in Idaho: Phase III: Data collection and analysis (Final report). University
of Idaho: National Institute for Advanced Transportation Technology (2004)
Benderitter, Y., Schott, J.J.: Short time variation of the resistivity in an unsaturated soil: the
relationship with rainfall. Eur. J. Environ. Eng. Geophys. 4(1), 37–49 (1999)
Clarke, C.R.: Monitoring long-term subgrade moisture changes with electrical resistivity
tomography. In: Unsaturated Soils, pp. 258–268 (2006)
Heydinger, A.G.: Evaluation of seasonal effects on subgrade soils. Transp. Res. Rec.
J. Transp. Res. Board 1821(1), 47–55 (2003)
Khan, M.S., Hossain, S., Kibria, G.: Slope stabilization using recycled plastic pins. J. Perform.
Constructed Facil. 30(3), 04015054 (2015)
Kean, W.F., Waller, M.J., Layson, H.R.: Monitoring moisture migration in the vadose zone with
resistivity. Ground Water 27(5), 561–562 (1987)
Kibria, G., Hossain, M.S.: Investigation of geotechnical parameters affecting electrical resistivity
of compacted clays. J. Geotech. Geoenviron. Eng. 138(12), 1520–1529 (2012)
58 A. Ahmed et al.

Jones, L.D., Jefferson, I.: Expansive soils. In ICE Manual of Geotechnical Engineering, Volume 1:
Geotechnical Engineering Principles, Problematic Soils and Site Investigation, pp. 413–441.
ICE Publishing, London (2012)
Lu, Y.: Temperature effect on unsaturated hydraulic properties of two fine-grained soils and its
influence on moisture movement under an airfield test facility (MS thesis). Arizona State
University (2015)
Manosuthikij, T.: Studies on volume change movements in high PI clays for better design of low
volume pavements (Doctoral dissertation). University of Texas at Arlington (2008)
Manzur, S.R., Hossain, M.S., Kemler, V., Khan, M.S.: Monitoring extent of moisture variations
due to leachate recirculation in an ELR/bioreactor landfill using resistivity imaging. Waste
Manage., 38–48 (2016)
Mehrotra, A.: Evaluating the influence of moisture variation on resilient modulus for unsaturated
pavement subgrades (MS Thesis). Louisiana State University (2011)
Nelson, J., Miller, J.D.: Expansive Soils: Problems and Practice in Foundation and Pavement
Engineering. Wiley, New York (1992)
Sebesta, S.: Investigation of maintenance base repairs over expansive soils: year 1 report
(Publication FHWA/TX-03/0-4395-1). FHWA, Texas Department of Transportation (2002)
Tabbagh, A., Camerlynck, C., Cosenza, P.: Numerical modeling for investigating the physical
meaning of the relationship between relative dielectric permittivity and water content of soils.
Water Resour. Res. 36 (2000). doi:10.1029/2000WR900181. ISSN: 0043–1397
Wanyan, Y., Abdallah, I., Nazarian, S., Puppala, A.J.: Expert system for design of low-volume
roads over expansive soils. Transp. Res. Rec. J. Transp. Res. Board 2154(1), 81–90 (2010)
Zapata, C.E., Houston, W.N.: Calibration and validation of enhanced integration climatic model
for pavement design (Report No. 602). Transportation Research Board, National Research
Council, Washington, DC, USA (2008)
Zhou, Q.Y., Shimada, J., Sato, A.: Three-dimensional spatial and temporal monitoring of soil
water content using electrical resistivity tomography. Water Resour. Res. 37(2), 273–285
(2001)
 Geotechnical Properties of Sediments
by In Situ Tests

Sara Rios1(&), António Viana da Fonseca1, Nuno Cristelo2,

and Claver Pinheiro1
1
CONSTRUCT-GEO, Department of Civil Engineering,
Faculty of Engineering, University of Porto, Porto, Portugal
{sara.rios,viana}@fe.up.pt, [email protected]
2
Department of Engineering, CQVR,
University of Trás-os-Montes e Alto Douro, Vila Real, Portugal
[email protected]

Abstract. River sediments are mainly composed by intermediate materials,

between sand and clay, for which partial drainage conditions apply. In these
cases, the interpretation of CPTU tests may be wrong since existing correlations
are based on fully drained or fully undrained conditions. This paper presents
results from CPTU tests performed in a river area to evaluate whether partial
drainage conditions were observed. The results, presented in terms of the nor-
malized velocity, show that great part of the analyzed profiles are in this con-
dition. For this reason, the angle of shearing resistance was presented as a
conservative estimate of the soil strength in these areas.

1 Introduction

Harbors management involves the execution of dredging operations to overcome

excessive deposition of sediments in the harbor’s basin and channel accesses, ensuring
proper navigability conditions. Sediment disposal management depends on the nature
of the dredged materials, which generally comprises fine materials such as sand
(particles dimensions in the range of 0.2 to 0.06 mm), silt (between 0.06 and
0.002 mm) and clay (below 0.002 mm). For that reason, a proper in situ geotechnical
characterization of such sediments is necessary before a dredging operation. While the
geotechnical behavior or sands and clays is relatively well understood with very well
established assumptions (e.g., usually drained response and use of effective stress
parameters for sand and, undrained response and undrained shear strength for clays),
intermediate materials such as silts and various mixtures with sand and clay, that occur
naturally and are frequently found worldwide, violate such assumptions. Behavior of
intermediate soils (including sands with fines) can be transitional between sands and
clays or can exhibit unique behavior that differs from both sands and clays. As a result,
a given intermediate soil may simultaneously exhibit properties like a sand, while other
properties may be more similar to a clay.
In particular, the extent to which penetrometer testing occurs under drained,
undrained or intermediate conditions depends on the soil consolidation characteristics
relative to the penetration rate and size of the penetrometer (Finnie and Randolph 1994).
© Springer International Publishing AG 2018
W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_6
60 S. Rios et al.

For that reason, a reliable evaluation of the mass permeability is a preliminary require-
ment for interpretation of piezocone (CPTU) tests (Randolph and Hope 2004; Schneider
et al. 2007; Kim et al. 2008; DeJong and Randolph 2012; Mahmoodzadeh and Randolph
2014). As reported by Schnaid et al. (2004), for soils with permeability in the range of
10−5 to 10−8 m/s, the simplest accepted approach of a broad distinction between drained
(gravel and sand) and undrained (clay) conditions cannot be applied to the interpretation
of in situ tests without a great deal of uncertainty. Implications are that the cone resistance
varies with penetration rate depending on whether the soil is contractive or dilatant as
defined by the state parameter framework proposed by Jefferies and Been (Jefferies and
Been 2006). As reported by DeJong and Krage (2014), if the soil has a large positive state
(state > 0.20; signifying very loose conditions, i.e., contractive behaviour), in drained
cone penetration, no excess pore pressure would develop, but the drained tip resistance
would be relatively low since the soil is contractive. In undrained cone penetration large
positive excess pore pressures would be generated, which reduces the effective stress in
the soil, and hence reduces the cone tip resistance relative to the drained tip resistance
(qun < qdr). If the state is highly negative (state < −0.20; dense conditions; i.e., dilatant
behaviour) in drained cone penetration, this would again result in no development of
excess pore pressures, but the magnitude of the drained tip resistance would be much
higher than for the loose condition. In undrained cone penetration negative excess pore
pressures would develop, increasing the effective stress, and making the undrained tip
resistance higher than the drained tip resistance (qun > qdr).
This means that for the standard penetration rate, a partial drainage in a loose soil
would result in lower estimation of the soil strength relative to a fully drained condi-
tion, and a higher estimation of the soil strength relative to a fully undrained condition.
Consequently, for the soil layers identified by the Robertson (2009) unified approach as
“sands”, the soil resistance generally evaluated by the angle of shearing resistance will
be underestimated if partial draining conditions occur due to the presence of fines. On
the contrary, for the soil layers identified as silts and clays, where the undrained
strength would generally be used, an overestimation of this parameter will occur in
partial draining conditions.
Discussion of this topic will be made in this paper for the specific case of CPTU
tests in river sediments of a harbor site.

2 Test Site and Procedures

The study area is near the city center of Vila do Conde, in north Portugal, located in the
left bank of Ave river where the new shipyards of Vila do Conde were installed. This
area is very close to the river mouth as indicated in Fig. 1, which means that it suffers
from the tidal influence. In fact, the tides in this region of the Atlantic coast are very
wide frequently achieving a difference of 4 m of water height between the low and high
tide. This had an important effect not only in the tests execution but also in the water
table evaluation. Since the tests were performed on shore in an area that was only dry
(and accessible) on the low tide the hour to perform the tests had to be judiciously
chosen and there was no time for unexpected problems. Moreover, the water table
depth, very useful in the interpretation of pore pressure measurements in CPTU tests,
 Geotechnical Properties of Sediments by In Situ Tests 61

Fig. 1. Site area satellite view (Google maps: https://www.google.pt/maps/@41.3403005,-8.

7439096,1462m/data=!3m1!1e3)

was not constant throughout the tests. It is possible that it can also vary from layer to
layer depending on the soil permeability, as less permeable layers take more time to
reduce the pore pressures when the tide falls.
This area was chosen because it is a preferential area of sediment deposition, not
only from Ave river but also from another small stream that converges to Ave river in
this specific point. Unfortunately, it is not possible to see it in Fig. 1 because part of
this stream is piped. On the other hand, this area was also a deposition place of dredged
sediments from Ave river in the past due to the current need of assuring river
navigability.
At this site three points were selected (Fig. 2), providing that they were all
accessible in the low tide. In these points cone penetrometer tests (CPTU) were per-
formed. The test procedure included first the execution of the light dynamic probing,
followed by CPTU test until a maximum tip resistance (qc) of 20 MPa was obtained
since the aim was only to study the loosest layers of sediments. At the maximum depth,
dissipation tests were performed but without success as the pore pressure remained
constant. It can be argue that dissipation tests could have been performed at the higher
excess pore pressure but since the water table was quite difficult to evaluate, it was even
more difficult to know the best depth to perform dissipation tests. After removing the
cone penetrometer, the water table depth was measured in the hole left by the cone,
which was used in the interpretation of CPTU tests although it does not adjusts to the
data as it will be shown below. A simple auger was also used to extract some undis-
turbed samples for future analysis that are out of the scope of this paper. Notwith-
standing, it is worth mentioning that by visual inspection the soil is mainly composed
by sand with fines and in some parts the sand becomes darker with more cohesion as an
indication that the amount of clay/silt particles increases.
62 S. Rios et al.

First point
Second point

Third point

Fig. 2. Zoom of the site area with identification of the three test points

For the CPTU tests, a standard cone was used with a cone area of 10 cm2, a
penetrometer diameter of 3.57 cm and a cone area ratio (a) of 0.58. The tests were
performed at the standard rate of 2 cm/s.

3 CPTU Interpretation Methodology

The tests were interpreted by the unified approach proposed by Robertson (2009) from
the basic CPTU parameters (qc, fs and u2). However, the main expressions used to
obtain some of the parameters will be identified here since there are several correlations
proposed by different authors for the same parameter.
The soil unit weight (c) was evaluated by value obtained through expression (1)
from Robertson and Cabal (2010):
 
c qt
= 0:27
log ðRf Þ þ 0:36
log þ 1:236 ð1Þ
cw pa

where,
qt is the corrected cone resistance as defined by Robertson (2009): qt = qc + (1 − a) * u0
Rf is the friction ratio as defined by Robertson (2009): Rf = fs/qc
pa is the atmospheric pressure
cw is the water unit weight
 Geotechnical Properties of Sediments by In Situ Tests 63

The constrained modulus (M) was evaluated by one of the expressions proposed by
Robertson (2009) as follows,

M ¼ aM ðqt  rv0 Þ ð2Þ

where
aM ¼ Qtn when Qtn  14
aM ¼ 14 when Qtn [ 14
The permeability coefficient (k) was obtained by expression (3) from Robertson
(2010a),

k ¼ 10ð0:9523:04
IcÞ if 1:0 \Ic\3:27

ð3Þ
k ¼ 10ð0:9523:04
IcÞ if 3:27\Ic\4:0

Since it was not possible to evaluate the horizontal consolidation coefficient (ch)
from dissipation tests, it was estimated by the consolidation theory according to Eq. (4)
taking into account the constrained modulus (M) and the permeability coefficient (k)
obtained as described above:

k
M
ch ¼ ð4Þ
cw

This enabled the calculation of the normalized velocity (V) as proposed by

Randolph and Hope (2004) and DeJong et al. (2013), by expression (5) and Fig. 3.

v
d
V¼ ð5Þ
ch

where,
v is the cone penetration rate (in this case is the standard v = 2 cm/s)
d is the penetrometer diameter (is this case d = 3.57 cm)
Note that according to Fig. 3, a standard rate of 2 cm/s induces partial drainage
conditions from fine sands through most silts.
In terms of strength parameters, the angle of shear resistance, can be obtained by
the following expression proposed by Robertson (2010b) in function of the angle of
0
shearing strength at critical state (/cv ) assumed 30º,
64 S. Rios et al.

Fig. 3. Field decision chart for 10 cm2 cone presenting relation between coefficient of
consolidation, penetration velocity and normalized velocity (DeJong et al. 2013)

0 0

/ ¼ /cv þ 14:44 log Qtn;cs  22:31 ð6Þ

where,

Qtn;cs ¼ KC Q ð7Þ

KC ¼ 1:0 if; Ic \1:64

ð8Þ
KC ¼ 5:581 IC3  0:403 IC4  21:63 IC2 þ 33:75 IC  17:88 if; Ic [ 1:64

4 Tests Results

Figures 4, 5 and 6 present the obtained results for the three points indicated in Fig. 2.
The water levels present (indicated as WT) are the ones obtained after removal of the
penetrometer. On point 1 the soil is mainly composed by sand with fines and a few
layers of silt. There is also clay between 3 and 4 m of depth. This point is the most
affected by the dredged operations on the river, as it is close to the breakwater that
separates the Ave river from the shipyards bay, so it is likely that the surface sand
layers came from the river. On point 2 the surface layers are more silty while sand
 Geotechnical Properties of Sediments by In Situ Tests 65

Fig. 4. CPTU test results for the first point

Fig. 5. CPTU test results for the second point

layers appear with depth. This point is the closest to the water in the bay and for that
reason the water table is higher here. It is interesting to notice that both in points 1 and
2 the same clay layer appears between 3 and 4 m of depth. The third point, located near
old wood piles (possibly used in a previous small pier), shows an intercalation of sand
and silt layers.
66 S. Rios et al.

Fig. 6. CPTU test results for the third point

When the normalized velocity is computed it is observed that a significant part of

the tests profiles has normalized velocities between 0.1 and 10, which are associated
with partial drainage conditions according to Schnaid et al. (2004), Schneider et al.
(2008), DeJong et al. (2013). This means that with exception of some layers of finer
soils for which undrained conditions could be acceptable without major errors, the soil
(mainly in point 1 and 3) has an intermediate behavior. Since the estimation of the
angle of shearing resistance (assuming fully drained conditions) corresponds to a
conservative estimation of the soil strength according to the explanation presented in

50

40

30

20 CPTU 1
CPTU 2
10
CPTU 3
0
0 1 2 3 4 5 6
depth (m)

Fig. 7. Angle of shearing resistance profiles: (a) first point; (b) second point; (c) third point
 Geotechnical Properties of Sediments by In Situ Tests 67

the introduction section of this paper, these profiles were plotted in Fig. 7. It is clear
that in these areas of intermediate soil behavior the angle of shearing resistance is
between 35° and 40°, which is a reasonable value for loose well graded sandy soils
(Hough 1957).

5 Conclusions

This paper presents results from CPTU tests performed in a river area influenced by the
tides and dredging operations. The main purpose was to evaluate the suitability of
CPTU tests to evaluate the permeability and strength properties of river sediments
composed mainly by intermediate materials between clay and sand. The normalized
velocity proposed by Randolph and Hope (2004) showed that this is actually the case
for the soil presented in this paper, since normalized values between 1 and 10 were
found in a great part of the analyzed profiles, indicating partial draining conditions.
Although laboratory tests are needed to verify the soil hydraulic and mechanical
properties obtained in situ, it is possible that the obtained angle of shearing resistance
between 35° and 40° is a safe estimation of the effective strength of this soil.

Acknowledgments. The authors would like to acknowledge the MCTES/FCT (Portuguese

Science and Technology Foundation of Portuguese Ministry of Science and Technology) for
their financial support through the SFRH/BPD/85863/2012 scholarship, which is co-funded by
the European Social Fund by POCH program, the project CONSTRUCT
(POCI-01-0145-FEDER-007457) funded by COMPETE 2020, and CNPQ (the Brazilian council
for scientific and technological development) for its financial support in 201465/2015-9 schol-
arship of the “Science without borders” program.

References
DeJong, J.T., Randolph, M.F.: Influence of partial consolidation during cone penetration on
estimated soil behavior type and pore pressure dissipation measurements. ASCE J. Geotech.
Geoenviron. Eng. 138(7), 777–788 (2012). doi:10.1061/(ASCE)GT.1943-5606.0000646
DeJong, J.T., Jaeger, R.A., Boulanger, R.W., Randolph, M.F., Wahl, D.A.J.: Variable
penetration rate cone testing for characterization of intermediate soils. In: Coutinho, R.,
Mayne, P. (eds.) Proceedings of the 4th International Conference on Site Characterization
(ISC-4), Porto Galinhas, Brasil, pp. 25–42 (2013)
Finnie, I.M.S., Randolph, M.F.: Punch-through and liquefaction induced failure of shallow
foundations on calcareous sediments. In: Proceedings of 17th International Conference on the
Behavior of Offshore Structures, Massachusetts Institute of Technology, Boston, vol. 1,
pp. 217–230 (1994)
Hough, B.K.: Basic Soil Engineering. Ronald Press, New York (1957)
Jefferies, M., Been, K.: Soil Liquefaction: A Critical State Approach. CRC Press, Boca Raton
(2006)
Kim, K., Prezzi, M., Salgado, R., Lee, W.: Effect of penetration rate on cone penetration
resistance in saturated clayey soils. ASCE J. Geotech. Geoenviron. Eng. 134(8), 1142–1153
(2008)
68 S. Rios et al.

Krage, DeJong: Variable penetration rate cone testing for liquefaction evaluation of sand with
fines. U.S. Geological survey final technical report, Davis, California, USA (2014)
Mahmoodzadeh, H., Randolph, M.F.: Penetrometer testing: effect of partial consolidation on
subsequent dissipation response. ASCE J. Geotech. Geoenviron. Eng. 138(7), 777–788
(2014). doi:10.1061/(ASCE)GT.1943-5606.0000646
Randolph, M.F., Hope, S.: Effect of cone velocity on cone resistance and excess pore pressures.
In: Proceedings of International Symposium on Engineering Practice and Performance of Soft
Deposits, Yodogawa Kogisha, Osaka, pp. 147–152 (2004)
Robertson, P.K.: Estimating In-Situ soil permeability from CPT and CPTu. In: 2nd International
Symposium on Cone Penetration Testing, Huntington Beach, USA, May 2010 (2010a)
Robertson, P.K.: Estimating In-Situ state parameter and friction angle in sandy soils from CPT.
In: 2nd International Symposium on Cone Penetration Testing, Huntington Beach, USA, May
2010 (2010b)
Robertson, P.K.: Interpretation of cone penetration tests – a unified approach. Can. Geotech. J. 46,
1337–1355 (2009). doi:10.1139/T09-065
Robertson, P.K., Cabal, K.L.: Estimating soil unit weight from CPT. In: 2nd International
Symposium on Cone Penetration Testing, Huntington Beach, USA, May 2010 (2010)
Schnaid, F., Lehane, B.M., Fahey, M.: In situ test characterization of unusual geomaterials. In:
Viana da Fonseca, A., Mayne, P. (eds.) Proceedings of the 2nd International Conference on
Site Characterization (ISC-2), Porto, Portugal, vol. 1, pp. 49–75 (2004)
Schneider, J.A., Lehane, B.M., Schnaid, F.: Velocity effects on piezocone tests in normally and
overconsolidated clays. Int. J. Phys. Modell. Geotech. 7(2), 23–34 (2007)
Schneider, J.A., Randolph, M.F., Mayne, P.W., Ramsey, N.R.: Analysis of factors influencing
soil classification using normalized piezocone tip resistance and pore pressure parameters.
ASCE J. Geotech. Geoenviron. Eng. 134(11), 1567–1586 (2008)
 Comparison of the CO2 Record of Different
Slope Stabilization Methods

Dennis Gross1(&), Susanne Kytzia2, and Armin Roduner1

1
GEOBRUGG AG, Romanshorn, Switzerland
2
University of Applied Sciences, Rapperswil, Switzerland

Abstract. In times of increasing concern for the environment, the CO2-foot-

print method is widely accepted to evaluate the use of greenhouse emitting
production processes for a given product. This allows for comparisons between
products and makes environmentally friendly choices of a product more trans-
parent to decision makers.
In the field of slope stabilization, the high tensile steel wire mesh TECCO® in
combination with soil or rock nailing is the state of the art in many countries.
Compared to shotcrete, load transfer capacity is equal or higher. The open mesh
leaves enough space for plants to grow through the system.
Recent CO2-footprint evaluation taking into account production and transport
of the material of an equivalent slope stabilization measure shows that the
TECCO® system has a very low CO2-footprint. The methods used for assessing
the impact were, the level of carbon dioxide emissions from burning of fossil
fuels as well as all other emissions which contribute to climate change. These
other emissions have been recorded and weighted according to their specific
contribution to give an overall index, “Global Warming Potential” (GWP).
Compared to shotcrete, the CO2-footprint of the mesh solution is 4–5 times
lower! One reason for the good result is the high CO2-emitting level of concrete
in general. Furthermore, less material weight and also transport costs for the
same stabilizing effect of the product on the slope also account for a better CO2
balance of the mesh solution.

1 Introduction

The flexible TECCO® slope stabilization system is a proven protection system using a
high-tensile steel wire mesh in combination with soil or rock nails for stabilizing slopes
endangered by slides and break-outs in loose and solid rock. Special spike plates actively
brace the system against the subsoil, positively influencing the deformation behavior of
the protection system. This is an open, transparent system with no possibility of
hydrostatic water pressure building up behind the mesh covering (Tables 1, 2 and 3).
The TECCO® system offers an alternative approach for slope stabilization com-
pared to anchored, reinforced shotcrete. Due to the fact that the TECCO® system
requires less construction materials, it can be expected that this approach accounts less
for climate change than regular methods do. This was clearly confirmed in a study done
by the Institute of Civil and Environmental Engineering at the University of Applied
Sciences Rapperswil (HSR), Switzerland.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_7
70 D. Gross et al.

Table 1. Used materials in flexible facing solutions.

TECCO – system Material Weight (kg)
Pinning Nail = 4.0 m Steel 1817.3
(WT 32 gross) Nail = 6.0 m Steel 4088.9
Injection for nails Cement, 23400.0
w/c = 0.4
Wire mesh TECCO mesh High-strength wire 1963.6
Corrosion protection: galvanization Zinc 169.6
Corrosion protection: galvanization Aluminium 8.9
Gripping plate Sheet steel 495.0
Corrosion protection for gripping plate Zinc coating 123.8
Press claws Steel
Corrosion protection Zinc coating
Stranded rope (as side guys, top/bottom) High-strength wire 100.0
Corrosion protection for stranded rope Zinc coating 0.5
Spiral rope anchor = 4.0 m (top/bottom) High-strength wire 42.4
Corrosion protection for spiral rope anchor Zinc coating 0.2
Injection for spiral rope anchors Cement 896.0
Planting TECMAT erosion protection mat Polypropylene 714.0
Total material weight 34 t
Materials used for a slope stabilization structure constructed using flexible TECCO® wire netting
for a slope 100 m long.

Table 2. Used materials in hard facing solutions.

Shotcrete placement Material Weight (kg)
Pinning (WT 28 gross) Nails = 4.0 m Steel 1081.9
Nails = 5.0 m Nails Steel 6713.7
Injection for Nägel Cement, 40350.0
w/c = 0.4
Shotcrete Total shotcrete {spraying loss *25%) 564360.0
Clips Steel 6324.8
Top plate Steel 589.2
Corrosion protection for top plate Zinc coating
Drainage Drainage pipes PVC 64.0
Total material weight 619 t
Materials used for a slope stabilization structure constructed using shotcrete placement for a slope
100 m long.
 Comparison of the CO2 Record of Different Slope Stabilization Methods 71

Table 3. Standard processes.

Material Description Unit C02 - GWP
fossil
Steel
Reinforcing steel 63% converter steel, low-alloyed, [kg C02/kg] 0.54 1.25
37% electric steel, unalloyed and
low-alloy
Low-alloyed This process produces primary steel [kg C02/kg] 0.72 1.7
converter steel
Cement
Portland cement, 90% clinker, 5% base additives, 5% [kg C02/kg] 0.764 0.829
strength class Z 42.5, plaster
at plant
Shotcrete
Concrete, exacting 375 kg cement, 150 kg water, [kg C02/kg] 0.121 0.135
1880 kg gravel
Transport
Truck, 28 t [kg C02/tkm] 0.174 0.223
Transoceanic freight [kg C02/tkm] 0.0001 0.011
Coatings
Zinc Zinc for zinc coating [kg C02/kg] 0.483 2.46
Aluminium Aluminium for corrosion protection [kg C02/kg] 1.82 11.7
Others
Drainage pipes for the Polyvinyl chloride pipes for draining [kg C02/kg] 0.02 2.16
shot-crete the slope
Erosion protection Liner to go underneath the Tecco [kg C02/kg] 0.002 2
mat mesh made of polypropylene
List of the standard processes used as well as their environmental impacts.

2 The TECCO® High-Tensile Slope Stabilisation System

The use of wire meshes and wire rope nets for flexible slope stabilization has stood the
test in many cases and frequently offers an alternative to solid concrete or shotcrete
constructions. The open structure of meshes and nets enables the entire surface to be
grassed over. Most often used for slope stabilization are wire meshes with a tensile
strength of ca. 50 kN/m, respective wires from ca. 500 N/mm2. Taking into account an
economic nail spacing however, these are often unable to absorb the occurring forces
and transfer them precisely to the nails.
The development of a wire mesh of high-tensile steel with a wire tensile strength of
at least 1770 N/mm2 offers an interesting possibility for efficient slope stabilization
which can be dimensioned using adapted soil and rock static dimensioning models.
72 D. Gross et al.

In principle the TECCO® slope stabilization system comprises the following

elements:
• High-tensile TECCO® steel wire mesh
• TECCO® system spike plates
• Standard soil or rock nails available on the market, e.g. type GEWI, TITAN or IBO.
TECCO® high-tensile steel wire mesh used as standard for slope stabilization
consists of a 3 mm thick high-tensile steel wire coated with aluminum-zinc for cor-
rosion protection (GEOBRUGG SUPERCOATING®). The 83 mm  143 mm
diamond-shaped mesh forms are produced by a simple twisting process. The TECCO®
steel wire mesh has a tensile strength of at least 150 kN/m. The three-dimensional
structure of the mesh positively influences the mesh-subsoil interaction. Together with
a resulting favorable friction this also offers a good basis for grassing over using hydro
or spray seeding.
In form, size and bending strength, the diamond shaped system spike plates were
adapted to the TECCO® mesh through numerous bending, punching and shearing tests
to achieve optimal supporting behavior of the stabilization system. Close contact and
where possible by lightly pressing-in the spike plates, the mesh can be optimally braced
against the subsoil to be stabilized.

3 Method

3.1 Basic Principle

The basic principle for this study is the method of Life Cycle Assessment (LCA) which
analyses the overall impact to the environment of a structure from the production of the
raw materials to the disposal of the structure as described in ISO 12040:2006 and ISO
14044:2006. In contrast to LCA, our analysis focuses on emissions relevant for climate
change, emissions characterized by their «Global Warming Potential» (according to
Guinee et al. 2001). To indicate this focus we refer to our study as «CO2 - Footprint».

3.2 Example Slope

The climate relevant emissions were estimated for two comparable slope stabilization
systems and an example slope with a height of 8.5 m and a length of 100 m: One with
the TECCO® system and one with a conventional shotcrete cover. The analysis of the
life cycle is limited to the production of the construction materials and their transport to
the construction site since no significant differences in climate relevant emissions from
construction, operation, maintenance, removal and disposal can be expected (Figs. 1
and 2).
 Comparison of the CO2 Record of Different Slope Stabilization Methods 73

Fig. 1. TECCO® mesh and system spike plates

Fig. 2. Example of application in Queensland, Australia

3.3 Used Materials

The materials listed in the tables below were used in these two structures. These
quantities correspond to a stabilization structure consisting of a slope 100 m long in
each case as represented in the diagrams shown in Fig. 3.

Fig. 3. Slope stabilization with flexible TECCO® mesh (left) and shotcrete (right)
74 D. Gross et al.

3.4 System Definition

The service life analysis will be limited to the manufacture of the building materials and
their transport to the construction site. Construction, operation, maintenance, removal
and disposal will not be taken into consideration (see Fig. 4).

Fig. 4. Life cycle of a product and CO2 footprint (schematic diagram)

 Comparison of the CO2 Record of Different Slope Stabilization Methods 75

This definition is based on the following considerations:

• Studies published on life cycle assessment of structures are based on the assumption
that the creation of the structure will only make a relatively small contribution
towards the pollution of the environment. Kasser (1998) and Geiger and Fleischer
(1997) come to the conclusion that in the case of residential buildings, less than 1%
of the cumulative energy demand of a building can be attributed to construction.
The construction of the structures considered here is not expected to result in
(relatively) higher energy expenditure.
• It has been assumed that in general, there will be no expenditure on operation and
maintenance.
• When the two structures are removed, it is mainly steel and concrete which will
need to be dealt with. If these two waste materials are collected separately, then
their processing and disposal will only result in very low energy consumption and
thus only to a low level of emissions which might affect climate change.
In addition, the following processes involved in the manufacture of building
materials were not taken into consideration:
• The galvanizing of the TECCO® mesh: here, only the manufacture of the zinc or
aluminum coating has been taken into consideration. However, the energy con-
sumption associated with galvanizing itself has not been taken into consideration.
• The refinement of the steel for the TECCO® mesh.
This simplification is necessary as these processes cannot be assessed using stan-
dard data.
However, it can be assumed that the energy consumption associated with galva-
nization and refinement is negligibly low compared to the energy consumption asso-
ciated with the steel manufacturing process.

3.5 Data Used as a Basis for the Life Cycle Inventory Analysis
The following table provides information on the processes used and the associated
environmental impacts.
The emissions from the manufacture of materials which might affect climate change
(incl. the recovery of materials, the manufacture of fabricated materials and the supply
of energy) as well as from transport processes have been taken from the Ecoinvent
database from the Swiss Federal Institute of Technology (ETH) domain (see Ecoinvent
center 2007).
Within the context of a conservative estimate of emissions (consideration of the
“worst case scenario”), in case of doubt, processes which produce the highest level of
environmental pollution will be chosen in each case. This includes above all:
• The selection of distances and means of transport. These have been based on an
average distance of 100 km for suppliers from Switzerland, 500 km for suppliers
from Germany and 700 km for suppliers from the rest of Central Europe. A (28 t)
truck has been used as the means of transport for all journeys from European
76 D. Gross et al.

destinations. Transport from China has been based on a distance of 8000 km for
transport by ship and 500 km for transport by truck.
• The selection of steel grades: as no information was available on the proportion of
recycling and alloying, the process of manufacturing ‘low-alloy oxygen steel’
which creates a comparatively high level of emissions has been taken as a basis.

3.6 Method for Assessing Impact

The emissions which might affect climate change have been shown by the level of
carbon dioxide emissions from the burning of fossil fuels (in kg of “CO2 - fossil”) on
the one hand. On the other hand, all the emissions which contribute towards climate
change have been recorded and weighted according to their specific contribution -
relating to carbon dioxide as a reference variable - and added together to give an overall
index. This overall index is known as the “Global Warming Potential” (“GWP” for
short); the unit in which it is measured is kg CO2 equivalent. This method of assess-
ment has been used as part of the Life Cycle Assessment in many studies and is
internationally recognized (see Guinee et al. 2001 and Frischknecht et al. 2003).

4 Results

The comparison of the total climate relevant emissions during the life cycle of the two
structures clearly shows that the TECCO® system accounts significantly less for the
greenhouse effect than the anchored shotcrete cover (see Fig. 5). This is the case for both

Comparison CO2 emssions of different slope stabilization systems

160'000

140'000

120'000 Greening / Drainage

Coatings
100'000 Steel
Grout & Concrete
kg

80'000 Transport

60'000

40'000

20'000

0
TECCO Shotcrete TECCO Shotcrete

CO2 - Equivalent CO2 - Fossil

Fig. 5. Climate relevant emissions during the life cycle of two comparable slope stabilization
structures built with two different methods.
 Comparison of the CO2 Record of Different Slope Stabilization Methods 77

«CO2 - fossil» and the «Global Warming Potential (GWP)». The environmental impact
of the TECCO® system is approx. 4–5 times smaller with both evaluation methods.
The difference can be explained with the differences in the used quantities of
materials. In the structure executed with anchors and shotcrete, approx. 14’700 kg of
steel, 40’300 kg of cement and 564’000 kg of concrete are required. For the same slope
with the TECCO® system, approx. 8’100 kg of steel and 23’400 kg of cement are
used. These differences have a big effect on the environmental impact due to both the
production of the materials and also their transport to site.
For the TECCO® system it can be concluded that:
• The anchorage accounts more for the greenhouse effect than the mesh and the
greening.
• The used cement has the biggest contribution to the total environmental impact,
followed by the steel (in relation 3 to 1).
• The effect of the transport is relatively small with 5% of the total impact.
For the system with the anchored shotcrete cover:
• The shotcrete cover accounts more to the greenhouse effect with 70% of the total
environmental impact than the anchorage with approx. 30%.
• The used cement has the biggest contribution to the total environmental impact,
followed by the steel (in relation 12 to 1).
• The effect of the transport is with approx. 9% more significant than the impact of the
steel.

5 Conclusion

On the base of these results, it can be concluded that the TECCO® system accounts less
for the greenhouse effect than a comparable slope stabilization system executed with
anchored shotcrete. The difference between the two systems is significant as the here
presented study shows. The environmental impact of the slope stabilization with a
TECCO® system is approx. 4–5 times smaller than with shotcrete.
Since the difference between the two systems is quite big, a detailed analysis
considering all the processes during the life cycle and a more detailed examination of
the steel production and the transports would not lead to fundamentally new findings.
This can be stated for the assumptions of life span, maintenance and repair. The result
is robust here as well.

References
Ecoinvent Center, Ecoinvent Database V2.0 (2007). http://www.ecoinvent.ch. Editor.
Dubendorf, CH
Frischknecht, R., Jungbluth, N., Althaus, H.-J., Doka, G., Dones, R., Hischier, R., Hellweg, S.,
Humbert, S., Margni, M., Nemecek, T., Spielmann, M.: Implementation of life cycle impact
assessment methods. Final report ecoinvent 2000 No. 3. Swiss Centre for Life Cycle
Inventories, Dübendorf (2003)
78 D. Gross et al.

Geiger, B., Fleischer, T.: Stoffliche und energetische Lebenszyklusanalysen von Wohngebäuden.
In: Gesamtheitliche Betrachtung von Energiesystemen. VDI-1328. Düsseldorf (1997)
Guinee, J.B., Gorrée, M., Heijungs, R., Huppes, G., Kleijn, R., de Koning, A., van Oers, L.,
Wegener Sleeswijk, A., Suh, S., de Haes, H.A., de Bruijn, H., van Duin, R., Huijbregts, M.A.
J., Lindeijer, E., Roorda, A.A.H., Weidema, B.P.: Life cycle assessment; An operational
guide to the ISO standards; Characterisation and Normalisation Factors. Leiden (2001)
Kasser, U.: Gebäude gesamtenergetisch beurteilt. Sonderdruck aus: Schweizer Architekt, Nr.
13 (1998)
 Bearing Capacity of Strip Foundation on Soft
Soil Reinforced with Stone Columns
Using Method of Slices

M. Khalifa1(&), M. Etezad2, A. Hanna1, and M. Sabry3

1
Department of Building, Civil and Environmental Engineering,
Concordia University, 1455 De Maisonneuve Blvd. W. Montréal,
Québec, Canada
2
Golder Associates Ltd., 6925 Century Avenue,
Suite #100, Mississauga, ON L5N 7K2, Canada
3
Bechtel Corporation, 12011 Sunset Hills Road,
Suit 110, Reston, VA 20190, USA

Abstract. Stone columns are recognized as an environmently-friendly and

cost-effective ground improvement technique. Stone columns are used in soft
soil to increase the bearing capacity, reduce the settlement, increase the rate of
settlement and reduce the liquefaction potential of the ground.
This paper presents an analytical model utilizing the method of slices to
predict the ultimate bearing capacity of the soil reinforced with a group of stone
columns. The soil within the failure zone was divided into slices and the limit
equilibrium technique was adopted to perform the analysis. Shear forces and
passive earth pressure on the boundaries of each slice were determined. By
utilizing a circular failure plane, the minimum inter-slice force coefficients were
determined.
The analysis was carried out using the Morgenstern-Price method to estimate
the failure surface together with the bearing capacity of the reinforced ground.
The failure surface was determined by trial and error to estimate the minimum
factor of safety. The ultimate bearing capacity was defined by increasing the
foundation load until the factor of safety of one was obtained. Results of the
present theory were compared with those available in the literature, where a
good agreement between the two was noted.

1 Introduction

Stone columns technique is a ground improvement method widely used over the past
decades. Stone columns are stiffer and have higher shearing resistance than the native
soft cohesive soil. This results in an increase of the soil bearing capacity and reduce in
the corresponding settlement (Etezad et al. (2015); Mitchell et al. (1985); Muir Wood
et al. (2000); Priebe (1995)). Furthermore, stone columns have higher permeability than
the native surrounding soil, which leads to reduction of the drainage path and
accordingly reduces the time required to complete the consolidation settlement
(McKelvey et al. (2004)).

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_8
80 M. Khalifa et al.

Unit cell concept was one of the first methods used to estimate the bearing capacity
of the ground reinforced with stone columns. The bulging mode of failure was used to
model this process. Accordingly, the bearing capacity of a single stone column was
predicted by estimating the horizontal capacity of the clay soil around the pile. In this
method, the capacity of the group is the total capacities of the individual columns.
Hughes and Withers (1974) used elastic-plastic theory developed by Gibson and
Anderson (1961) to calculate the maximum vertical stress that can be carried by a
single stone column due to bulging failure. Similar approach has also been used by
Balaam and Booker (1981).
Hu (1995) performed laboratory tests on a group of end bearing and floating stone
columns. Shear and punching mode of failure were reported in this study. He concluded
that the collapse pattern for soil mass reinforced by stone columns changes from
general shear failure mechanism to punching shear based on the stone column’s length.
The group interaction reported by Hu (1995) was confirmed by the numerical analysis
carried out by Lee and Pande (1998); Muir Wood et al. (2000).
Priebe (1995) reported two methods to estimate the bearing capacity of a footing on
the group of stone columns based on the general shear failure and the equivalent
homogeneous composite. In the first approach, the weighted average values of the
friction angle and the cohesion were calculated along the failure plane, and the bearing
capacity was estimated using theories for homogeneous soils. In the second approach, a
larger footing width was assumed, and the bearing capacity was determined using the
new footing width and the shear properties of the unreinforced soil. (Lee et al. 1998)
introduced the concept of composite properties of the reinforced soil. They used finite
element technique to estimate the bearing capacity and settlement of the reinforced soil.
Bouassida et al. (2009) presented design charts to determine the ultimate bearing
capacity of a group of floating stone columns. In their study, the friction between the
footing and the soil was neglected as well as the distribution of stone columns.
An analytical model was developed by Etezad et al. (2015) using limit equilibrium
method and equivalent soil properties under the footing to calculate the bearing
capacity of a rigid footing placed on the ground reinforced with stone columns. The
method utilize the general shear failure and slip surface, which was deduced from the
results of the numerical model of Hanna et al. (2013).
Many of the theories developed to predict the bearing capacity utilize the theory for
homogeneous soil as a simplified assumption. In the present paper, the method of slices
was adopted to estimate the bearing capacity of strip footing on a compacted inho-
mogeneous soil. Terashi et al. (1991) conducted a series of centrifuge tests and
full-scale tests to estimate the bearing capacity of the improved ground by compacted
sand piles. The experimental results agreed well with the bearing capacity calculated
using a circular type of slip surface. Morgenstern-Price method was utilized in this
study to calculate the bearing capacity of a footing rests on clay soil reinforced with
stone columns.
 Bearing Capacity of Strip Foundation 81

2 Bearing Capacity Calculations Based on Slip Circle

A 2-D model made of soft clay reinforced with stone columns was developed. The ground
was loaded with a uniform pressure to simulate the case of strip footing. The failure zone
was divided into 50 slides and the forces acting on each slide are shown in Fig. 1. The
limit equilibrium method by Morgenstern et al. (1965) method was used in the computer
program SIDE V-6.020, developed by Rocscience (2012), to calculate the minimum factor
of safety of the circular slip surface (Fig. 2). The analysis was also check with the
Bishop’s Simplified method for comparison (Bishop 1955). The footing was loaded to the
ultimate capacity in the form of uniform pressure applied on the ground surface.

Slip circle
center

α
i
Footing Width
[B] Uniform Load;
[qu]

Vi+1 wi S S
Ei+1 Ei
Vi b
Ti
Ni
Stone columns

Fig. 1. Forces acting on a slide in the slip circle

2.1 Validation
The results obtained from the present analysis was compared with the laboratory and
numerical results available in the literature (Hanna et al. (2013); Hu (1995); McKelvey
et al. (2004)). Table 1 presents these comparison, where a good agreement can be
noted. This further validate the methods of Morgenstern-Price and the Bishop’s Sim-
plified method of slices as viable techniques to estimate the bearing capacity of clay
soil reinforced by a group of stone columns. However, Bishop’s Simplified method
generally overestimates the factor of safety as compared to Morgenstern-Price method
(Turnbull et al. 1967). Nevertheless, both approaches underestimate the bearing
capacity of the reinforced soil.
82 M. Khalifa et al.

Fig. 2. An example of the bearing capacity calculation using slip circle method
Table 1. Comparison between the bearing capacity of the reinforced soil estimated by the method of slices and the available experimental and numerical
results.
No. Clay properties Stone properties Area Footing Surcharge qu, qu, B. S. % Difference qu, Morgenstern % Difference
Cu /c cc [Degree] cs replacement width, B load, q Measured method from Price method from
[kPa] [Degree] [kN/m3] [kN/m3] ratio, As% [m] [kPa] [kPa] [kPa] measured [kPa] measured
1a 32 0 14 34 17.3 24 0.09 0 272 222 20% 218.8 18%
2a 20.5 0 9.9 34 20.3 40 0.05 0 160 165 1% 161 3%
3b 10.5 0 13.1 30 15.47 30 0.1 0 75 73 4% 71.7 3%
4b 11.5 0 13.1 30 15.47 30 0.1 0 79 80 1% 78.5 1%
5c 5 12 13 45 21 35 2.5 2.6 352 270 24% 268.2 23%
6c 5 13 13 40 19 35 2.5 2.6 280 250 12% 246.5 11%
7c 5 15 14 45 21 35 2.5 2.8 420 380 9% 380.2 10%
8c 15 15 14 45 21 35 2.5 2.8 660 644 2% 646 2%
9c 10 13 13 40 19 30 2.62 2.6 366 270 27% 267 26%
10c 15 13 13 40 19 30 2.62 2.6 458 350 24% 346.5 24%
a experimantal work of McKelvey et al. (2004)
b experimental work of Hu (1995)
c numerical work of Hanna et al. (2013)
Bearing Capacity of Strip Foundation
83
84 M. Khalifa et al.

2.2 Parametric Study

In this study, the results are presented in the form of improvement ratio (IR), which is
defined as the ratio of the capacity of improved soil to the capacity of the unimproved
soil. In this analysis, the effect of the stone columns spacing to diameter ratio (S/D),
and the number of stone column rows (N) on the bearing capacity at a given
replacement ratio (As %) was performed. Each row of the stone columns was assumed
as a trench with a width (b) and the spacing between trenches was considered the same
as the spacing between columns as shown in Fig. 3. The trench width was calculated
using Eq. 1. In this case, the replacement ratio (As) was calculated using Eq. 2. The
ratio of S/D was examined in the range of 1 to 3.0 as given in Table 2.

p  D2
b¼ ð1Þ
4S

N  b N  p  D2
As ¼ ¼ ð2Þ
B 4SB

Where,
N is the number of stone columns rows, b is the trench width, B is footing width, D
is stone column diameter, and S is the spacing between stone columns.
Moreover, this method was also used to determine the effect of the undrained
strength of the clay soil (Cu), and the replacement ratio (As %). In this analysis, the
replacement ratio ranged from 10% to 30% which widely used in practice (Hanna et al.
(2013); Hu (1995)). The range of the other parameters believed to govern the bearing
capacity are presented in Table 3.
The effect of the column arrangement on the bearing capacity was also investigated.
Figure 4 presents the spacing/ diameter ratio of stone columns versus replacement
ratios, assuming all other parameters are constant. It can be noted that the bearing
capacity increases with the increase of the replacement ratio. Also. It can be noted that
the spacing between stone columns slightly influences on the improvement ratio. For
low replacement ratio (<10%), the improvement ratio is almost constant, which agreed
well with Castro (2014) observation. However, the improvement ratio reduces with the
increase of the columns spacing ratio; for higher replacement ratio (20%, and 30%),
which confirm that for small spacing between columns, the lateral support from the
surrounding soil increases and accordingly, will show significant improvement.
Regarding the effect of the clay shearing resistance, it is noticeable that the
improvement ratio reduces with the increase of shear strength of clay soil (Cu) for the
same spacing/ diameter ratio as shown in Fig. 5. Contrary, the improvement ratio
raised by the increase of the stone columns shearing resistance angle (/) as shown in
Fig. 6.
The influence of stone columns number under the footing has been also investi-
gated. In this analysis, the columns number varied from N = 1 to N = 4. The columns
diameter ranged from 0.3 m to 1.6 m, which covers the maximum and minimum
ranges that may be used in practice. Figure 7 presents the effect of the stone columns
number on the bearing capacity, and it can be noted that there is no remarkable change
 Bearing Capacity of Strip Foundation 85

Fig. 3. Stone columns arrangement

86 M. Khalifa et al.

Table 2. Range of stone columns dimensions used in parametric study

Area replacement No. of Column Column Spacing/Diameter Trench
ratio, As % columns, N diameter, spacing, ratio, S/D width
D [m] S [m] b [m]
10% 1 0.70 0.70 1.00 0.50
2 0.35 0.38 1.09 0.25
2 0.40 0.50 1.25 0.25
2 0.45 0.63 1.40 0.25
2 0.49 0.75 1.53 0.25
2 0.63 1.25 1.98 0.25
2 0.89 2.50 2.80 0.25
20% 1 1.30 1.30 1.00 1.00
2 0.69 0.75 1.09 0.50
2 0.80 1.00 1.25 0.50
2 0.89 1.25 1.40 0.50
2 0.98 1.50 1.53 0.50
2 1.26 2.50 1.98 0.50
2 1.69 4.50 2.66 0.50
3 0.46 0.50 1.09 0.33
3 0.53 0.67 1.25 0.33
3 0.59 0.83 1.40 0.33
3 0.65 1.00 1.53 0.33
3 0.84 1.67 1.98 0.33
3 1.00 2.33 2.34 0.33
30% 1 2.10 2.10 1.00 1.50
2 1.04 1.13 1.09 0.75
2 1.20 1.50 1.25 0.75
2 1.34 1.88 1.40 0.75
2 1.47 2.25 1.53 0.75
2 1.89 3.75 1.98 0.75
3 0.69 0.75 1.09 0.50
3 0.80 1.00 1.25 0.50
3 0.89 1.25 1.40 0.50
3 0.98 1.50 1.53 0.50
3 1.13 2.00 1.77 0.50
4 0.52 0.56 1.09 0.38
4 0.60 0.75 1.25 0.38
4 0.67 0.94 1.40 0.38
4 0.73 1.13 1.53 0.38
4 0.85 1.50 1.77 0.38
 Bearing Capacity of Strip Foundation 87

Fig. 4. Improvement ratio (IR) versus S/D for different replacement area, Cu = 5 kPa, U = 35o:
(a) As = 10%; (b) As = 20%; (c) As = 30%
88 M. Khalifa et al.

Fig. 5. Effect of shear resistance of clay on the improvement ratio for different replacement
ratios, U = 35o: (a) S/D = 1.25; (b) S/D = 1.50; (c) S/D = 2.00
 Bearing Capacity of Strip Foundation 89

Table 3. Range of parameter used in the parametric study

Parameter Range of value
Cohesion of clay soil, Cu [kPa] 5, 10, and 15
Unit weight of clay soil, cc [kN/m3] 18
Angle of shear resistance of stone, /s [Degree] 35, 38, and 40
Unit weight of stone, cs [kN/m3] 20
Stone column diameter, D [m] 0.3–1.60
Stone column diameter to spacing ratio 1.00–2.50
Replacement area, As [%] 10, 20, and 30
Footing width, B [m] 5

Fig. 6. Effect of shear resistance of stone on the improvement ratio for different replacement
ratios, Cu = 10 kPa; S/D = 1.50

Fig. 7. The relation between number of stone column rows versus the improvement ratio, at
S/D = 1.5, Cu = 5 kPa, U = 35o
90 M. Khalifa et al.

in the improvement ratio particularly for low replacement ratio (<10%). However, there
is a slight reduction in the improvement ratio for higher values of replacement ratio
(30%). For a large number of stone columns, a small diameter was used to keep the
replacement ratio constant. By reducing the column’s diameter more load will be
transferred to the clay soil (less strength material), which lead to the reduction in the
improvement ratio (Black et al. (2007); Hanna et al. (2013))

3 Conclusions

Morgenstern-Price method of slices was used to calculate the bearing capacity of a soft
clay reinforced with stone columns. The theory developed compared well with the
available results in the literature. Parametric study was conducted on the parameters
believed to govern the behavior of this system. The following was concluded:
1- Morgenstern-Price method of slices was successfully used to estimate the bearing
capacity of reinforced clay soil.
2- (IR) significantly increases with the increrase of the replacement ratio.
3- (IR) reduces due to the increase of the spacing/ diameter ratio (S/D)
4- Based on the results of the present study, in order to optimize the benefit of the use
of reinforced soft clay with stone columns is to use a ratio of columns spacing to
columns diameter (S/D) equal to 1.5
5- For the same stone columns arrangement and shear resistance (diameter, spacing)
the improvement ratio (IR) increases with a decrease of the shear strength of
surrounding clay. However, the ultimate bearing capacity of the system is signif-
icantly increased due to the increase of the shear strength of the clay soil as well as
the stone. Furthermore, the improvement ratio (IR) increases with the increase of
the stone columns shearing resistance angle.

Acknowledgments. The financial support received from Concordia University is

acknowledged.

References
Balaam, N., Booker, J.R.: Analysis of rigid rafts supported by granular piles. Int. J. Numer. Anal.
Meth. Geomech. 5(4), 379–403 (1981)
Bishop, A.W.: The use of the slip circle in the stability analysis of slopes. Geotechnique 5(1), 7–
17 (1955)
Black, J., Sivakumar, V., McKinley, J.: Performance of clay samples reinforced with vertical
granular columns. Can. Geotech. J. 44(1), 89–95 (2007)
Bouassida, M., Jellali, B., Porbaha, A.: Limit analysis of rigid foundations on floating columns.
Int. J. Geomech. 9(3), 89–101 (2009)
Castro, J.: Numerical modelling of stone columns beneath a rigid footing. Comput. Geotech. 60,
77–87 (2014)
 Bearing Capacity of Strip Foundation 91

Etezad, M., Hanna, A., Ayadat, T.: Bearing capacity of a group of stone columns in soft soil. Int.
J. Geomech. 15(2) (2015)
Gibson, R., Anderson, W.: In situ measurement of soil properties with the pressuremeter. Civil
Eng. Public Works Rev. 56(658), 615–618 (1961)
Hanna, A., Etezad, M., Ayadat, T.: Mode of failure of a group of stone columns in soft soil. Int.
J. Geomech. 13(1), 87–96 (2013)
Hu, W.: Physical modelling of group behaviour of stone column foundation. (Ph.D. dissertation),
University of Glasgow, UK (1995)
Hughes, J., Withers, N.: Reinforcing of soft cohesive soils with stone columns. Ground Eng. 7
(3), 42–49 (1974)
Lee, J.S., Pande, G.N.: Analysis of stone-column reinforced foundations. Int. J. Numer. Anal.
Meth. Geomech. 22(12), 1001–1020 (1998)
McKelvey, D., Sivakumar, V., Bell, A., Graham, J.: A Laboratory Model Study of the
Performance of Vibro Stone Columns in Soft Clay. J. Geotech. Eng. 152, 1–13 (2004)
Mitchell, J.K., Huber, T.R.: Performance of a stone column foundation. J. Geotech. Eng. 111(2),
205–223 (1985)
Morgenstern, N., Price, V.E.: The analysis of the stability of general slip surfaces. Geotechnique
15(1), 79–93 (1965)
Muir Wood, D., Hu, W., Nash, D.: Group effects in stone column foundations: model tests.
Geotechnique 50(6), 689–698 (2000)
Priebe, H.J.: The design of vibro replacement. Ground Eng. 28(10), 31 (1995)
Rocscience: SLIDE 6.0—2D Slope Stability Analysis for Soil and Rock Slopes. Rocscience Inc.
(2012)
Terashi, M., Kitazume, M., Minagawa, S.: Bearing capacity of improved ground by sand
compaction piles. In: Deep Foundation Improvements: Design, Construction, and Testing.
ASTM International (1991)
Turnbull, W. J., Hvorslev, M.J.: Special problems in slope stability. J. Soil Mech. Found. Div.
(1967)
 The Behavior of a Foundation Laterally
Loaded at the Top over Highly Porous
and Collapsible Soil

Roberto Kassouf1(&), David de Carvalho2,

Paulo José Rocha de Albuquerque2, and Nelson L. Fonte Jr.3
1
Metropolitan College of Campinas - DeVry Metrocamp, Campinas, Brazil
[email protected]
2
State University of Campinas, Campinas, Brazil
[email protected], [email protected]
3
Head of Geoponto Engineering, Mogi das Cruzes, São Paulo, Brazil
[email protected]

Abstract. In geotechnical engineering, the problem of laterally top-loaded piles

occurs frequently. This type of foundation is often used in highly porous, col-
lapsible soils, which are common in several regions of Brazil. Because of the
limited information available in the literature, several load tests have been
performed on piles subjected to this load. To analyze the behavior of piles in
these collapsible soils, load tests were performed in steel piles (I), W
250
32.7 section (mm
kg/m), length 12 m, conventional bored piles
(/ = 0.40 m; L = 12 m) and continuous helical auger piles (/ = 0.40 m;
L = 12 m). All of them were tested at the same site. For each type of pile, a first
load test was carried out with the soil in its natural condition of moisture
content, followed by a second load test after the surface soil was flooded for
48 h. The results indicated a significant negative effect of flooding on the topsoil
which consisted of sandy-silty clay with collapsible features down to the depth
of 6 m. Load vs. horizontal displacement curves and soil coefficients of hori-
zontal reaction were obtained. The results allowed the proposal of parameters
for use in the soil under study. Before executing the load tests, laboratory and
in-situ tests were performed to investigate the local subsoil.

Keywords: Laterally load tests  Piles  Porous soil  Collapsible soil

1 Introduction

The deep foundation designer, in addition to the executive-system-related aspects and

the type of material employed, is concerned with the loading system, which can pro-
duce displacements due to axial load, transversal load and bending moments.
Regarding lateral loading, the top of foundations is generally subject to these forces.
This is the case of bridges, viaducts, transmission line towers, wind power generation
towers, or along the piles shaft due earth’s pressure. In countries with seismic activities,
the construction code requires consideration of the horizontal load in foundation
designs, thus minimizing the consequences of a possible earthquake.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_9
 The Behavior of a Foundation Laterally Loaded 93

When dimensioning foundations to resist lateral loads, the project criteria involve
not only the ultimate horizontal loading capacity, but also the maximum pre-established
displacement that can occur. Currently, there are many mathematical methods for
prediction of the horizontal displacement of a pile. The common difficulty about these
methods concerns the adoption of geotechnical parameters to be used in the
calculations.
The main parameter used is the modulus of horizontal reaction (nh), which is
defined as the soil resistance along the foundation divided by its deflection at a point.
Simplified mathematical models have been created for the analysis because the mod-
eling of the transversal action problem is three-dimensional and extremely complex for
routine solutions by project designers. The most commonly known and widespread
theory for evaluation of these actions is the “Theory of Horizontal Reaction of the
Soil”, where the nh factor represents the proportionality between the reaction and
displacement acting on the soil mass. However, this factor is difficult to be theoretically
estimated. Nevertheless, this factor can be “measured” using load tests to get a reliable
value for horizontal resistance of the soil where the construction is to take place.
The reaction of the soil is a function of many factors, such as pile properties, soil
stress vs. strain behavior, depth of the point analyzed, level of foundation displacement,
etc. Because of the difficulty to establish a function covering all of these factors, the
simplified Winkler’s (1875) hypothesis is generally used, where the reaction of the soil
is considered proportional to the displacement of the pile.
For horizontal loading, in the first few meters the surface soil has great influence on
the load vs. lateral displacement behavior of the foundations. To predict the behavior of
horizontally loaded foundations, theoretical approaches are available in the literature;
however, parameters should be determined for the local soil before they can be used.
Surface soils with porosities above 50% cover vast areas in Midwest Brazil.
Because of their large void volumes, these soils undergo great strain under load. In
addition, many of these soils are collapsible, i.e., when the soils are under load and
when a significant increase in the moisture content or soil saturation occurs, the
structure collapses, which results in unacceptable displacement values for the buildings.
Given the lack of available information in the literature concerning horizontally
loaded piles on highly porous collapsible diabase soil, this study was developed in
order to review the performance of three types of piles in this type of soil. The tests
were performed with the soil at their natural moisture content and after flooded on the
surface.
Based on the horizontal loading tests, the effect of soil flooding on the load vs.
horizontal displacement curve and the values for the horizontal reaction coefficient for
both the natural and pre-flooded soil conditions were verified. The results obtained for
the horizontal reaction coefficient were compared to the results for other types of
foundations on similar soils.
The acting stresses and displacements on a pile under bending moments and hor-
izontal loadings were determined using the theory of soil horizontal reaction, which is
based on the model proposed by Winkler (1875). The soil behavior under horizontal
forces is simulated by a set of independent, identical and equally spaced springs. Thus,
the reaction of the soil is considered proportional to the displacement of the point being
analyzed. This supposition simplifies the problem, considering that the ratio between
94 R. Kassouf et al.

the pressure of contact at the base of the foundation and its corresponding consolidation
is the same for any supporting area. Using the model proposed by Winkler, the concept
of the modulus of horizontal reaction, K, was introduced by Terzaghi (1955). It is
defined as the ratio between the reaction of the soil (in units of force applied by the pile
length) and the corresponding displacement (Eq. 1):
p
K ¼ ð1Þ
y

Where: K = the modulus of horizontal reaction (FL−2), p = the applied pressure

−1
(FL ) and y = the horizontal displacement (L).
This notation presents the advantage of being independent of the diameter of the
foundation. Therefore, the Eq. 2 can be rewritten as:

K ¼ kh :D ð2Þ

Where: kh = the horizontal reaction coefficient (FL−3) and D = the diameter of the
foundation (L).
For pure sands, the elasticity modulus increases (approximately) linearly with
depth. Therefore, the soil reaction to the load applied to the pile is assumed to increase
linearly with depth (Eq. 3):
p
K¼ ¼ nh :z ð3Þ
y

Where: nh = the modulus of the horizontal reaction of the soil (FL−3) and z = the
depth (L).
Understanding the variation of K along the foundation is required for analysis of its
behavior based on the theory of soil reaction. Refinements and sophistications in the
reaction modulus function by depth are not justified since the errors in the results of
calculations are minor compared with the ones involved in the estimation of numerical
values of the modulus of soil reaction. Matlock and Reese (1960) agree with this
assessment because the results are satisfactory and can be obtained for most practical
cases as simple forms of variation of the reaction modulus with depth. Additionally, in
practical problems, the uncertainty inherent to the estimation of soil behavior based on
conventional tests is generally compatible with the minor errors that can be introduced
by the depth using a simple form of the soil reaction modulus function.
Alizadeh and Davisson (1970) were the first researchers to present curves obtained
for horizontal loads tests on sandy soils. They presented these curves in the form of nh
at the y-axis and the displacement y0 at the x-axis. To create these curves, the authors
used Eq. (4) by Matlock and Reese for the displacement for the application of only one
horizontal load parallel to the ground surface, i.e.:

3
HT
yo ¼ Dy ð4Þ
E p Ip
 The Behavior of a Foundation Laterally Loaded 95

Where: T is the relative stiffness between the pile and the soil, and for soils with
sandy behavior and normally consolidated clays. It is defined by Eq. 5:
rffiffiffiffiffiffiffiffiffi
5 Ep Ip
T¼ ð5Þ
nh

To calculate nh, Eq. 6 is used:

4; 42H 5=3
nh ¼ 5=3  2=3 ð6Þ
yo E p I p

In the present study, the soil characteristics were analyzed using geotechnical
laboratory tests and in-situ tests, with the aim of predicting soil behavior in terms of
deformability, resistance and collapsibility. To determine the horizontal reaction
coefficient (nh), curves of horizontal reaction coefficients (nh) vs. horizontal displace-
ment at the surface (yo) were calculated based on the load tests performed and adopting
a range of values of horizontal displacement. Alizadeh and Davisson (1970) proposed
the curves (nh) vs. (yo) using values in the 6.35–12.70 mm range. The curves of
horizontal reaction coefficients vs. horizontal displacement were divided into two
separate groups: load tests performed with soil in its natural moisture content and load
tests performed with the soil pre-flooded on the surface.

2 Materials and Methods

The aims of this study were achieved based on geotechnical analysis of the soil under
investigation, execution of piles, conduction of load tests on piles and analysis of the
obtained data.

2.1 Geological and Geotechnical Characteristics

The soil is composed of diabase bodies which can also be found embedded in the
Itararé Formation and in the Crystalline Complex in the form of sills and dykes. Such
materials are pedologically classified as purple latosols, and are mineralogically con-
stituted by quartz, ilmenite, magnetite, kaolinite, gibbsite, iron oxides and hydroxides.
Thicknesses range from 5 to 30 m. (Zuquete 1987).
The load tests were carried out at the Experimental Site of the University of
Campinas (Cavalcante et al. 2007), where the profile of the subsoil is diabase soil. It
has a highly thick 6-m thick sandy-silty clay surface layer followed by a sandy-clayey
silt layer down to the depth of 19 m. The water table is found at the depth of 18 m. The
mean geotechnical profile of the subsoil was obtained from the results of tests on
undeformed samples taken after opening two wells and performing Standard Pene-
tration Tests (SPT) and Static Penetration Tests (CPT). The results of the laboratory
tests and field tests performed can be found in Peixoto (2001), Fig. 1.
96 R. Kassouf et al.

Fig. 1. Static Penetration Tests (CPT) and the results of the laboratory tests and field tests
performed Peixoto (2001).

The characteristics of collapsibility of the subsoil of the Experimental Site were

reviewed by Monacci (1995). The criterion used was the one proposed by Vargas
(1978), which defines soils as collapsible when the coefficient of structural collapse,
“i”, is greater than 2%, considering that “i” is defined by the following equation:

Dec
i¼ ð7Þ
1 þ ei
 The Behavior of a Foundation Laterally Loaded 97

In the equation above, Dec represents the variation in the void ratio due to collapse
of the soil structure, and ei is the index of voids before flooding.
The indexes of collapse at the depths of 0.75 m, 5.00 m and 8.00 m, were deter-
mined from simple oedometric tests, with soil flooded at determined pressures, which
are presented in Table 1. Note that collapsibility in the layer of high porous soil is
reduced as a function of depth.

Table 1. Collapse coefficients according to the applied voltage and the depth - Monacci (1995)
Depth 0,75 m Depth 5,00 m Depth
8,00 m
r (kPa) i r (kPa) i r (kPa) i
(%) (%) (%)
5 4,97 – – – –
9,8 11,09 9,8 2,41 – –
19,4 7,4 19,4 3,76 – –

Gon (2011) performed oedometric tests with stresses of 100 kPa, 200 kPa and
400 kPa. The samples taken at 1 m of depth were observed to be collapsible, with a
very high index for all flooding stresses. For the 2 m and 3 m deep layers, it was
possible to observe the low collapsibility index for the 100 kPa and 200 kPa stress,
which in turn had a high index for the 400 kPa stress. Based on the data obtained for
the 100 kPa stress, only the 1 m, 4 m and 8 m depths were observed to be collapsible.
However, for the 200 kPa stress, only the 2 m and 3 m depths were not collapsible.
Finally, for the flooding stress of 400 kPa, all of the depths were observed to be
collapsible, with the exception of the 8 m depth.

3 Load Tests

3.1 Piles Analyzed

3.1.1 Steel Piles
A 12 m long steel pile I was used; 250
32.7 section W (kg
mm/m); Moment of
Inertia Ix = 4,937 cm4; Young’s Modulus E = 205,000 MPa and Area = 42.1 cm2; in
Gerdau (2015). The pile dimensions are shown in Fig. 2.

3.1.2 Continuous Flight Auger Piles and Bored Piles

These piles were constructed with 0.40 m nominal diameter and 12 m length. The
concrete used in continuous flight auger piles were pumped, cement consumption of
400 kg/m3 and aggregates (sand and gravel); slump ±240 mm. The characteristic
compression resistance of concrete (fck) was 15 MPa, slump ±70 mm, using gravel
and sand as aggregates. The longitudinal frame of both piles were 4 bars with
98 R. Kassouf et al.

146

9,1
6,1

239,8

220

258
9,1

Fig. 2. Steel pile dimensions

16 mm diameter, CA-50 steel, 6 m long, plus a 12 m long Dywidag steel bar with
32.0 mm diameter positioned in the center of the pile. The stirrups were 6.3 mm every
20 cm. The Young’s modulus of these piles was in average 20 GPa (Miranda Junior
2006).

3.2 Results and Analysis

Based on the results of the load tests, load vs. horizontal displacement curves were
obtained.
The results for steel piles are shown in Fig. 3.
Figures 4 and 5 show the results for bored and continuous flight auger piles.
The load tests were conducted for the three types of piles with slow maintained
loads. The pre-flooding of the soil was conducted for 48 h through an excavation
around pile head with the head of 0.70 m
0.70 m and 0.50 m depth
It is observed that:
Pre-flooding the topsoil caused drastic reduction of soil resistance to horizontal
loading. Taking a 5 mm displacement as reference for the steel pile into the natural
moisture content soil, the supported load is approximately 50 kN, whereas for the
pre-flooded soil for this displacement, the supported load is approximately 3 kN. For
bored piles in the natural moisture content soil for 5 mm, the supported load is about
48 kN, while for the pre-flooded soil for this displacement, the supported load is about
8 kN. For continuous flight auger piles, the supported load is about 45 kN in soil in its
natural moisture content, whereas for pre-flooded soil for this displacement of 5 mm,
the supported load is about 2 kN.
 The Behavior of a Foundation Laterally Loaded 99

Displacement - mm
75.0
70.0
65.0
60.0
55.0
Lateral Load - kN

50.0
45.0
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0 5 10 15 20 25 30 3 40
Slow Horizontal Loading Flooded Slow Horizontal Loading

Fig. 3. Load vs. horizontal displacement curves on Steel Pile (W250
32,7) mm
kg/m;
L = 12 m.

Slow Horizontal Loading flooded Slow Horizontal Loading

60.0
55.0
50.0
45.0
Lateral Load - kN

40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0 5 10 15 20 25 30 35
Displacement - mm

Fig. 4. Load vs. horizontal displacement curves - Bored pile (/ = 0,40 m; L = 12 m)

100 R. Kassouf et al.

Slow Horizontal Loading Flooded Slow Horizontal Loading

60.0
55.0
50.0
45.0
Lateral Loading - kN

40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
0 5 10 15 20 25 30 35 40 45
Displacement - mm

Fig. 5. Load vs. horizontal displacement curves - Continuous flight auger (/ = 0,40 m;
L = 12 m)

Table 2. Values of horizontal loads, displacements and horizontal reaction coefficients. Source:
The authors.
Type Author Local Soil nh (MN/m3) displacement 6 to
pile Name Brazil condition 12 mm
Steel Silva, Unicamp Natural 7.71
Marcella B.M Flooded 0.40
Bored Miranda Jr., Natural 11.55
Gentil Flooded 0.56
CFA Miranda Jr., Natural 9.86
Gentil Flooded 0.24
Caisson Kassouf, Natural 12
Roberto Flooded 6
 The Behavior of a Foundation Laterally Loaded 101

Fig. 6. Graph of nh vs. authors load tests.

4 Conclusions

The load tests have indicated that the soil has low bearing capacity to horizontal
loading, even in its natural condition of moisture content. These results reinforce the
need to perform load tests on these types of soil to get precise parameters for projects.
High variation in moisture content down to a few meters deep into the soil sig-
nificantly reduces the piles ability to support horizontal loadings. Thus, in foundations
subject to such loading in these types of soils, care should be taken to minimize this
possibility, both in the case of leakage in piping and in the case of infiltration by rain
water.
The permeability coefficients of this soil both in its natural condition and in the
compacted condition are in the range between 5
10−4 cm/s and 5
10−7 cm/s,
respectively.
The values of horizontal coefficients nh obtained from the analysis of steel piles,
bored piles and flight auger piles, that are presented in Table 2, differ from the literature
for different types of deep foundations built on soil with similar behavior to the one
from the present study (under natural and pre-flooded conditions) are presented in
Fig. 6.
However, these values are within the range of values nh found for sandy porous
soils of the State São Paulo, Brazil. Miguel (1996) found 7.50 and 8.0 MN/m3 for
bored, Strauss and root piles, respectively. Ferreira et al. (2001) obtained values nh of
7.4 and 11.0 MN/m3 for bored piles in the city of Bauru. Souza et al. (2008) got the
value of 5 MN/m3 for bored piles in Ilha Solteira. For the Experimental Site under
analysis, Miranda Junior (2006) obtained the average values nh of 7.28 MN/m3 for root
piles, 9.86 MN/m3 for continuous flight auger piles and 11.55 MN/m3 for bored piles.
Carvalho et al. (1996) obtained the value of 11.90 MN/m3 for pre-cast concrete
piles.
102 R. Kassouf et al.

Kassouf (2012) had to place close to the same type of soil the value obtained the
value of 12 MN/m3 for a caisson, for the same type of soil in a place close the
Experimental Site.
The results of load tests conducted by Miranda Junior (2006) show that the rein-
forced concrete blocks in the pile head can increase load capacity of both natural
moisture content soil and soil with significant variations of moisture content.
For collapsible soils, the analysis of the possibility of water saturation or large
variations in moisture content of the surface soil up to few meters, during the con-
struction service life, are of fundamental importance to determine the project
parameters.

Acknowledgments. The authors thank the Unicamp-Universidade Estadual de Campinas, the

FAPESP-Fundação de Amparo à Pesquisa do Estado de São Paulo and DeVry Metrocamp-
Metropolitan College of Campinas.

References
Alizadeh, M., Davisson, M.T.: Lateral load tests on piles – arkansas river project. J. Soil Mech.
Found. Div. ASCE 96(SM5), 1583–1604 (1970)
Carvalho, D., Albuquerque, P.J.R., Claro, A.T., Ferreira, C.V.: Análise de estaca carregada
transversalmente no topo, em solo residual de diabásio, Seminário de Engenharia de
Fundações Especiais (SEFE 3), vol. 1, pp. 145–154 (1996)
Gerdau – Perfis Gerdau Açominas aplicados como Estacas Metálicas em Fundações Profundas.
8ª edição (2015)
Gon, F.S.: Caracterização geotécnica através de ensaios de laboratório de um solo de diabásio da
região de Campinas/SP. M.S thesis. Department of Geotechcnics and Transportation,
University of Campinas, Campinas, Brasil (2011)
Kassouf, R.: Análise de Prova de Carga em Tubulão a Céu Aberto Submetido a Esforço
Horizontal em Solo não Saturado de Diabásio da Região de Campinas. Dissertação de
Mestrado. Universidade Estadual de Campinas, SP (2012). 111p.
Matlock, H., Reese, L.C.: Generalized solutions for laterally loaded piles. J. Soil Mech. Found.
Eng. Div. ASCE 86(SM5), 63–91 (1960)
Miguel, M.G.: Execução e análise de provas de carga horizontal em estacas em solo colapsível.
M.S. thesis. Departament of Geotechcnics, University of São Paulo, São Carlos, Brasil (1996)
Miranda, J.G.: Estacas Submetidas a Esforços Horizontais em Solos Colapsíveis do Interior de
São Paulo nas Condições Natural, Melhorada e Inundada. Ph.D. Dissertation, Faculty of
Agricultural Engineering, University of Campinas, Campinas (2006)
Monacci, M.G.: Estudo da Colapsibilidade de um Solo do Campo experimental da Faculdade de
Engenharia Agrícola, Unicamp. Dissertação de Mestrado, Faculdade de Engenharia Agrícola.
Unicamp, p. 130 (1995)
 The Behavior of a Foundation Laterally Loaded 103

Peixoto, A.S.P.: Estudo do ensaio SPT – T e sua aplicação na pratica de engenharia de fundações
– tese de doutorado, Faculdade de engenharia agricola, Unicamp (2001). 468p.
Terzaghi, K.: Evalution of Coefficients of Subgrade Reaction. Géotechnique 5(4), 297–326
(1955)
Vargas, M.: Introdução à Mecânica dos Solos. McGraw -Hill do Brasil, São Paulo (1978)
Zuquete, L.V.: Análise e proposta metodológica sobre cartografia geotécnica para condições
brasileiras, Tese de Doutorado, vol. 3. EESCUSP, São Carlos (1987)
 Early Applications of DMT in Arabian Gulf
Area – Three Case Studies

Sharif Emad(&)

GTC LAB, Geotechnical Engineering, e.construct, Dubai, UAE

[email protected], [email protected]

Abstract. DMT was used for three main applications in famous projects in
Dubai, UAE and in Duqm Port in Sultanate of Oman in the Arabian Gulf. The
1st two applications were conducted in man-made sandy earthfill embankments
formed by dredging from the sea bed and stabilized by deep ground improve-
ment. The recent correlations of S. Marchetti to incorporate the DMT stress
history parameter KD for CPT-based CRR was used to confirm post-compaction
CPT-based liquefaction analysis in the Duqm Site, whereas DMT was used in
Dubai site to confirm the over consolidated nature of upper sandy fill soil that
was previously Vibro-Compacted (8–10 years). On the other hand, the design of
large diameter flexible steel oil storage tanks is based on controlled & strict,
edge settlement requirements. Without DMT, the estimates of soil modulus are
made based on SPT or CPT tests resulting in conservative estimates, and sug-
gesting the need for deep ground densification/improvement. Site specific cor-
relation was developed for a large site between DMT based M and CPT tip
resistance (qc) indicating at least 50% increase over the CPT based M.

1 Introduction

DMT was only recently introduced for practical use in Dubai and the Gulf Area in
General. This paper describes three main early applications of DMT in main projects.
The several man-made islands in Dubai and other places, present a convenient area of
application of DMT and SDMT for both assessment of the deep man made earth fill,
design and control of deep ground improvement procedures and assessment of lique-
faction potential. Further, the construction of large diameter oil storage flexible steel
tanks presents another important area where DMT can provide valuable information to
accurately assess the settlement of tanks.
This paper presents some of the recent applications of DMT in this area.

1.1 Man-Made Islands/Embankments

Construction on artificial islands or embankments by dredging and filling is very
common in the Arabian Gulf region for different purposes.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_10
 Early Applications of DMT in Arabian Gulf Area 105

1.1.1 Background on Man-Made Islands & Embankments Construction

and Stabilization
Several Man-made Islands were constructed along the shore line of Dubai. The first
was Palm Jumeira Island having an approximate diameter of 8 Kms as shown in Fig. 1.

Fig. 1. Typical shapes of man-made islands

The photos in Fig. 2 show part of the Commercial berth site in Duqm Port in
Oman, proposed to accommodate several industrial facilities for the port.

Fig. 2. Commercial Berth – Duqm New Port

The islands/embankments were constructed with silty sands dredged from adjacent
sea bed and laid by hydraulic filling with heights ranging between 12–18 m above
original sea level as shown in Fig. 3.

Fig. 3. Typical photos of sand dredging operations

106 S. Emad

1.1.2 Ground Densification and Verifications

Stabilization and densification of the upper sandy fill is essential to improve its
Engineering performance and mitigate hazards such as liquefaction and high differ-
ential settlements and long term creep movements.
Common Methods used are Vibro Compaction/Replacement and Dynamic
Compaction/RIC for shallow sand depths. Vibro Compaction is most widely and
commonly used method although other methods were also used.
Verification of VC works is typically based on Pre and post-compaction CPT
measurements and zone loading tests in some cases.
The use of Robertson (2009) CPT-based Profiling (SBT) is very effective to
establish the localities of high silt content/high friction ratio which are not likely
compactable.

1.1.3 Actual Performance

Several roads, buildings on shallow and deep foundations exist since more than 8–10
years by today and performing very well as indicated in Fig. 4.

Fig. 4. Typical light weight and heavy structures in PJ

No published data exists about actual performance of the foundation, however the
observed evidence indicates very well and satisfactory performance of the compacted
earth fill.
Testing showed remarkable stiffness and strength increase with aging. Fills tested
after several years of laying are stiffer than young fills.

1.2 Case 1: Confirming CPT-Based Liquefaction Analysis in Port

of Duqm in Oman
The Duqm area lies along the coasts of the Indian Ocean, about 600 km SW of Muscat,
as shown in Fig. 5, the Capital City of the Sultanate of Oman.
 Early Applications of DMT in Arabian Gulf Area 107

Fig. 5. Map of Oman showing location of Duqm

The area is intended to accommodate one of the largest Oil Ports in the Middle East
with all associated facilities as sea port, airport, dry dock, oil refinery, Storage Ter-
minals, infra-structures like roads, rail way, bridges, buildings and residential villas etc.
that are under construction or on the anvil.

1.2.1 Description of DMT/CPT Application

A main component of the port facilities is the Commercial Berth which is made up of
thick reclaimed sandy fill by dredging from the sea bed and hydraulic filling. The berth
is >250 m in width and more than 1500 m in length extending from the coastal line
into the sea. The typical soil profile indicates a thick (typically 12–18 m), slightly silty
to silty fine sand (reclaimed soil) on top of a thin sea bed layer of silt/Clay soil that is
underlain by bedrock of very to extremely weak claystone.
108 S. Emad

The use of shallow foundations to support the many proposed light to moderately
heavy structures within the area of the Commercial Berth requires satisfaction of both
liquefaction mitigation and acceptable settlement performance criteria, including con-
trol of potential differential settlement.
The Berth was investigated with large number of widely spaced boreholes with
SPT testing conducted at 1 to 1.5 m intervals in each borehole as part of technical
feasibility and initial risk assessment and planning of the port. Later, a more detailed
investigation was conducted within the site of a typical project that lies within the
mid-zone of the Commercial Berth and which is proposed for cement silos and concrete
mixing plant. The subsurface conditions are indicated in the generalized subsurface
profile in Fig. 6, which shows the upper sandy fill of about 15 m depth, underlain by
thin sea bed layer of clayey soil followed by bedrock of claystone, as indicated in
Fig. 6.

Fig. 6. Generalized subsurface profile – Duqm site

The SPT tests results obtained within 5 test borings, has indicated large stiffness
variations and presence of weak zones. This is clearly indicated in Fig. 7 of SPT vs.
elevation.
 Early Applications of DMT in Arabian Gulf Area 109

Fig. 7. SPT vs. Elevation – Duqm site

The obtained SPT results, suggested that deep ground improvement is necessary.
Vibro compaction was found to be technically feasible, time and cost effective method
to achieve the necessary targets. Pre compaction – CPT tests were conducted and used
to assess the liquefaction potential and plan the VC works. The results showed the site
is liquefiable and weak, and therefore VC works were conducted. 1st round VC: The
typical layout of triangular grid of VC points is shown in Fig. 8.
Verification included post-compaction CPTU tests at three locations as indicated in
the above layout. At each location, 2 pair of post-compaction CPT tests were con-
ducted, at 1/3rd spacing from the poker point and at centre of spacing of a typical
triangular grid. Typical post-compaction CPT results and CPT – based liquefaction
analysis results are shown in Fig. 9.
110 S. Emad

Fig. 8. Layout of VC points & post-compaction CPT tests locations – Duqm site

Fig. 9. Typical post-compaction– CPT result & liquefaction analysis Duqm site
 Early Applications of DMT in Arabian Gulf Area 111

Note: Liquefaction analysis is conducted according to NCEER (1997) recommenda-

tions with specialist software, which is commonly specified in most similar projects in
GCC areas (e.g. Dubai). Design Earthquake effects as Magnitude (M) and Maximum
ground acceleration (amax) ranges of 5.5–6 and 0.10–0.15 g, respectively were con-
sidered as adopted by the Geotechnical Investigation Study, with amplification factor of
1.5 applied through the reclaimed soil above rock level.
Later, 3 post-compaction DMT tests were conducted, each at the centre of the test
areas of post-compaction CPTs, to assess KD effects on the CPT-Based liquefaction
potential, and also for settlement performance of circular raft foundations proposed for
the cement silos. Typical DMT results are shown in Fig. 10. Definitions of parameters
measured and interpreted by DMT test are given in “Report of the ISSMGE, Technical
Committee 16, 2001”. The obtained results were repeatable and reliable.

Fig. 10. Post-compaction Compaction DMT 1 (typical result) – Duqm site

Definitions of the different parameters incorporated with DMT test are shown in the
below table extracted from The Flat Dilatometer Test (DMT) in Soil Investigation,
Report of the ISSMGE, Technical Committee 16 on “Ground Property Characterization
from In-Situ Testing”, 2001.
112 S. Emad

Incorporating DMT Results to Support CPT-Based Liquefaction Analysis

In his recent paper, Marchetti (2016) described a method to incorporate the stress
history parameter (KD) of DMT into the Liquefaction Correlations. The final recom-
mended relationship that shows the new correlation is shown in Fig. 11.

Fig. 11. Chart for estimating CRR in clean sand based on Qcn & KD (Marchetti 2016)
 Early Applications of DMT in Arabian Gulf Area 113

The relationship is based on Qcn, KD relationship as Qcn = 25
KD given earlier

by Robertson (2012), therefore, CRR is evaluated from measured CPT (Qcn) and also
from Qcn = 25
KD, and the average CRR is then evaluated as: Average CRR =
[(CRR from Qcn)
(CRR from KD)]0.5. CRR is the cyclic soil resistance. The CRR
values provided by Fig. 11 can be regarded as the sum of two parts. One part, the bulk
of it, is the CRR estimated by the CPT consensus curve. The other part is the KD
correction. An advantage of Fig. 11 is that it shows in one graph the whole picture,
facilitating the recognition of trends. For instance the graph indicates that for high
values of KD and Qcn, corresponding to seasoned sands, the differences between the
curves are higher. This later observation is of importance as the likely improvement
effect of KD on the value of soil cyclic resistance CRR can be quickly noticed.
Therefore, for low KD of <2, the effects may be ignored, whereas for higher KD, then
the effects are regarded as significant. This is also illustrated in Fig. 12 as given by
Marchetti (2015).

Fig. 12. CRR expressed as a function of Qcn & KD. Qualitative framework (Marchetti 2015)

An excel sheet was prepared to conduct the analysis for the above. The 3 DMT tests
conducted in the middle of each test triangle has indicated low KD as indicated in the
above tests results (from slightly <1 to 1 for tests 1 & 3 and around 2 to 3 for test 2).
The conclusion was that KD results did not impose any improvement on CRR, and the
site was still classified as liquefiable and further round of VC was specified. The DMT
results, however has also indicated the spatial variation as indicated by different ranges
of KD, which is in close agreement with the post-compaction CPT findings.
Following re-compaction works conducted by the contractor, other round of
post-compaction CPT tests were only conducted that indicated satisfactory perfor-
mance, as illustrated in Fig. 13. 2nd round of VC - Typical Post-compaction CPT result:
Liquefaction analysis was found satisfactory and the site was accepted, and the
construction commenced.
114 S. Emad

Fig. 13. Typical post-compaction compaction (2nd round) CPT – Duqm site.

1.3 Case 2: Confirming OC Nature of Previously Vibro-Compacted,

Aged Sandy Fill in PJ in Dubai
Dubai is part of United Arab Emirates. It lies on the Arabian Gulf as shown in Fig. 14.
Dubai area is distributed physiographically into main land and coastal areas in a ratio of
9:1. The main land is mostly occupied by Aeolian/desert sand dunes (80% of the area)
with hard encrustations of local/in land sabkha’s in areas with near surface water table
and thin sheets of Aeolian sand overlying the local fans of gravels. The sand dune near
coastal belt is enriched with carbonate source material from the sea-shells and car-
bonate rocks.
Dubai is Famous of its tall towers and several man made islands as indicated in
Fig. 15.
Subsurface geological and geotechnical conditions in the island are described as
follows. Results of Typical Test Borings in the Island would show the typical profile as
illustrated in Fig. 16, composed of:
An upper layer of man hydraulically laid sandy fill of 12–18 m in depth sitting on a
thin, weak sea bed layer of very silty SAND to sandy SILT (Carbonate content exceeds
70% and is >90% for many samples), followed by Cap rock of Calcarenite and
Sandstones.
 Early Applications of DMT in Arabian Gulf Area 115

Fig. 14. Map of UAE indicating Dubai on the map

Fig. 15. Sattelite image of Dubai and coast line

116 S. Emad

Fig. 16. Typical subsurface conditions within PJ Island

1.3.1 Application of DMT in Palm Jumeira

The site of a prestigious new project lies on the crescent Palm Jumeira. The site area is
>150,000 sq. m with 4 towers of 160–190 m height range. The architectural design of
the project as schematically indicated in Fig. 17, will make it a new landmark in Dubai.

Fig. 17. View of the project architecture

 Early Applications of DMT in Arabian Gulf Area 117

The subsurface ground conditions are typical of the Palm Jumeira in general as
indicated in the generalized subsurface profile presented in Fig. 18.

Fig. 18. Generalized subsurface profile along the site – Palm Jumeira Dubai

The upper silty SAND earthfill was previously deep compacted with Vibro
Compaction since more than 8 years and hence the CPT tests results indicated dense
conditions. Seismic studies included in-situ tests conducted to measure the shear wave
and other dynamic properties of the site.
The typical MASW (multi-channel analysis of surface waves) results in the vicinity
of the above CPT test result as shown in Fig. 19, indicate the shear wave velocity of
the upper sand generally exceeds 400 to 500 m/s particularly within the upper 10 m.

Fig. 19. Typical MASW profile – PJ Dubai

118 S. Emad

This was explained to be a result of the over-consolidated nature of the soil due to
aging effects as reflected by the relatively high CPT results. However, a more specific
justification was required to address the OCR which is difficult to assess by CPT tests
results for sandy soil. Accordingly, few DMT tests were conducted at selected locations
to establish the constrained modulus (M) profile and establish the DMT (M)/CPT
(qc) ratio that would indicate the OC nature of the sand. Figure 20 presents the result of
typical DMT (M) and CPT (qc) in Mpa as obtained.

Fig. 20. DMT (M) – blue curve - and CPT (qc) – red curve - in Mpa vs. Depth

The M/qc ratio is presented in the below graph and the 1 m rolling average is also
indicated.
M/qc ratio indicated in Fig. 21 shows that it is greater than 8 for the upper 8 m and
exceeding 25 within the upper OC crust. This is a clear confirmation of the OC nature
of the earthfill and provides the necessary justification of the MASW results. It further
provided general good matching with CPT and MASW results.

1.4 Case 3: Accurate Settlement Prediction of Large Diameter Oil

Storage Tanks
DMT was used in a Tank Farm Project in the Port of Al-Fujeira in UAE for accurate
assessment of the proposed large diameter tanks settlement. The oil storage tanks are
typically constructed within the ports close to coastal lines where relatively poor soil
conditions exist. Therefore, accurate assessment of foundation settlement is very
important to decide if ground improvement is required or not. The settlement estimates
depend on the assessed soil modulus (E) or constrained modulus (M). For silty sandy
soils that are generally encountered in that area, the estimates are generally based on
 Early Applications of DMT in Arabian Gulf Area 119

Fig. 21. M/qc ratio vs. depth

SPT and CPT (qc – tip resistance) tests results, which would generally result in low
range values of E. The most commonly used relations are:
E = N60 (Mpa) – (Ciria Report 43, 1995)
M = 4 qc for (qc < 10 MPa)
M = 2qc + 20 Mpa for (10 < qc < 50 MPa)
M = 120 Mpa for (qc > 50 MPa) – (Lunne et al. 1997)
The Eurocode 7, mentions (E = 2.5qc MPa), which is also consistent with the
above estimate of M for NC conditions of sand. For OC sands, the following relation is
typically recommended: M = 5 qc for (qc < 50 MPa), and M = 250 MPa for (qc >
50 MPa) (Lunne et al. 1997).
The relation between M, Shear Modulus G and Young’s Modulus E according to
Elasticity Theory, based on Poisson’s ratio m are given below:

M=G ¼ 2ð1  vÞ=ð1  2vÞ; and G=E ¼ 2ð1 þ vÞ

The effect of over consolidation (stress history) is a main factor that is generally
underestimated with CPT based M or E estimates.

1.4.1 Effect of Shear Strain/Loading Level

The foundation settlements shall be assessed for shear strain in the range of 0.1% (This
is established based on observations made for well designed actual foundations).
Therefore, with absence of reliable procedure to estimate OCR of sands from the results
of CPT or SPT tests, then, the above is only illustration of the amount of conservatism
or risk which the interpreter has to take to estimate M or E for his calculations!
120 S. Emad

For large diameter tanks (>50 m), the use of conservative estimates of Es would
result in settlements exceeding the maximum permissible limit (API 650) and hence
deep ground improvement and other expensive solutions would be needed. Figure 22
shows illustration of the sensitivity of settlement of the tank to the modulus E of the
upper overburden soil layer.

Fig. 22. E = 35 Mpa – settlement at tank edge = 60 mm

Typical settlement analysis of 68 m Diameter tank for range of E modulus of the

upper overburden silty Sand for 300 kPa uniform Hydrotest pressure.
By adopting DMT based modulus of 60 Mpa, then the edge settlement was found
to be 45 mm only.
Note: Axi-Symmetric FE analysis for settlement was conducted with Sigma/W soft-
ware (www.geo-slope.com).
Local experience has indicated that actual tank settlements under hydrotest loading
test are much less than the theoretical estimates of tank settlements based on the above
SPT and CPT methods. The actual measured settlements are typically <50% of the
theoretical estimates.
The cost of deep ground improvement with Vibro-Compaction or Stone Columns is
several hundred thousands of dollars per each large diameter tank, in addition to several
months of delay of the project.
It is also important to point out that the hydrotest conducted (API 650) at the
completion of construction of each tank is a preload that causes most of the permanent
settlement component to take place and subsequent unloading and reloading of the tank
during its operational life time would follow the more stiff reloading/unloading mod-
ulus with in-significant movements as experienced with so many existing tanks of
various diameters.
 Early Applications of DMT in Arabian Gulf Area 121

Therefore, the accurate and reliable assessment of the soil modulus is very sig-
nificant for confidence of the estimated foundation settlement. DMT was therefore
introduced to provide reliable and direct measurement of in-situ modulus of the soil and
provide site – specific correlation to the other penetration tests as SPT and mostly CPT
that are widely used.
Case Study: The proposed storage tank farm includes tanks having diameters of 68 m
and 36 m for the small ones. All the tanks were designed for an average max. hydrotest
load of up to 300 kPa. The main design criteria is that the shell plate settlement under
max. load not to exceed 50 mm (preferable). However, settlements up to 100 mm may
be tolerated by the owner, provided the settlements are uniform (differential settlements
between edge points at 10 m arc length not to exceed 13 mm) and most of the set-
tlements to be of short term nature (to take place during the hydrotest).
The investigation for this Tank farm project consisted of a number of geotechnical
boreholes drilled up to 40 m depth, CPT and DMT tests in addition to geophysical
survey with MASW (Multi Channel Analysis of Surface Waves), as given in Fig. 23.

Fig. 23. Layout of 3 MASW lines covering all tanks locations

The typical geological cross section indicated the site is made up of thin upper
rockfill layer of 1–4 m thickness that was constructed to provide a level platform with
thickness increasing towards the sea side, underlain by medium dense to dense fine
silty sand that is followed by bedrock of very weak to weak Sandstone and con-
glomerates. GWT exists at 2.5 m depth approximately. Generalized subsurface profile
showing the main soil and rock layers is shown in Fig. 24. SPT results of all boreholes
indicated Medium dense to Dense soil with few localized weak lenses/pockets.
The DMT & CPT results, after 3 m of pre-drilling though rockfill cover, indicated OC
sand followed by Medium dense silty Sand, as shown in Fig. 25.
122 S. Emad

Fig. 24. Subsurface profile indicating general homogeneity – Fujeira site.

Fig. 25. Summary of all the 7 DMT tests & typical CPT test result with SBT profiling

The main parameter controlling the settlement estimates of the sand is obviously
the modulus of deformation E. Based on SPT and CPT results, the average Es was
estimated to be 35–40 Mpa, resulting in settlements exceeding 50 mm at the large
diameter tank edges. Based on direct measurement of DMT, M average was slightly
 Early Applications of DMT in Arabian Gulf Area 123

Fig. 26. M/qc ratio obtained (vertical axis) vs. depth (horizontal axis) in m

around 100 Mpa, and E is >60 Mpa. The typical DMT (M)/ CPT (qc) ratio is shown
graphically in Fig. 26.
It shows that M/qc may safely be taken as 6 instead of 4 as adopted for NC sand,
which means that for average qc = 15 Mpa, then M = 90 Mpa and E will be >60 Mpa.
This is at least 50% greater than the typical CPT based estimates adopting NC sand
conditions.
The final recommendation suggested that no deep ground improvement is neces-
sary for settlement limitation. Vibro Compaction was only suggested for the main tanks
to improve the stiffness within potential weak localities, as indicated by the CPT and
some of the DMT results. The project construction phase has commenced and the next
step will be the observation of the results of actual hydrotests that are planned for each
of the tanks, which will provide the final confirmation on the foundation ground
behavior. As stated previously, previous experience has indicated that the actual results
of settlements are less than the theoretical predictions. This will increase the confidence
in the methods adopted for the soil modulus estimates.

2 Conclusion

The use of DMT in Dubai and the Arabian Gulf Area is relatively recent. The early
applications presented in this paper indicate that DMT would be of great value to
identify the effects of previous stress history and over-consolidation of either man made
and stabilized embankments or natural sandy soils, resulting in better understanding of
actual foundation soil behavior and avoiding expensive foundation solutions that may
not be necessary. Further, the use of DMT to support the confidence on assessment of
the liquefaction potential by CPT or SPT-Based methods is of main significance,
124 S. Emad

particularly for aged sands which are vibro-compacted before some time. With future
accumulated experience, a more rich data base would be established that would
increase the level of confidence of the use of DMT for in-situ modulus estimates and to
provide a site specific calibration to other commonly used penetration tests as CPT and
SPT. The documentation of results of full scale hydrotests on large diameter storage
tanks provides great opportunity for that purpose. Further, applications for design and
quality control of deep ground improvement projects and liquefaction potential
assessment form other wide area of future use of DMT in Dubai and the Gulf Area.

References
API 650: Welded Steel Tanks For Oil Storage, 11th edn., June 2007
Clayton C.R.I.: The Standard Penetration Test (SPT): methods and use. CIRIA Report 143,
London (1995)
Marchetti, S.: Some 2015 updates to the TC16 DMT report 2001. In: Proceedings of 3rd
International Conference on the Flat Dilatometer DMT 2015, Rome, Italy, pp. 43–65 (2015)
Marchetti, S.: Incorporating the stress history parameter KD of DMT into the liquefaction
correlations in clean uncemented sands. J. Geotech. Geoenviron. Eng. 142(2), 04015072
(2016)
NCEER: Proceedings of the NCEER workshop on evaluation of liquefaction resistance of soils.
Technical report NCEER-97-0022, 31 December 1997
Robertson, P.K.: Interpretation of cone penetration tests – a unified approach. Can. Geotech. J. 46
(11), 1337–1355 (2009). doi:10.1139/T09-065
Robertson, P.K.: The James K. Mitchell lecture: interpretation of in-situ tests – some insights. In:
Proceedings of 4th International Conference on Geotechnical and Geophysical Site
Characterization, Porto de Galinhas, vol. 1, pp. 3–24 (2012)
Lunne, T., Robertson, P.K., Powell, J.J.M.: Cone penetration tests in geotechnical practice (1997)
The Flat Dilatometer Test (DMT) in Soil Investigation, Report of the ISSMGE, Technical
Committee 16 on “Ground Property Characterization from In-Situ Testing” (2001)
 Some Laboratory and Numerical Studies
on the Behaviour of Stone Columns Installed
in Mumbai Marine Clay

Vinay Bhushan Chauhan1(&), Yashwant A. Kolekar2,

and Satyanarayana M. Dasaka1
1
Civil Engineering Department, Indian Institute of Technology Bombay,
Mumbai 400076, Maharashtra, India
[email protected],
[email protected]
2
Civil Engineering Department, Government College of Engineering,
Pune 411005, Maharashtra, India
[email protected]

Abstract. Mumbai marine clay is problematic in nature for substructures and it

needs to be strengthening before making it available for any construction
activity. Out of many available geotechnical solutions, stone columns are quite
handy in significantly improving the bearing capacity of marine clay, which also
supplements by dissipating excess pore water pressure and reducing the set-
tlements of clay beds under external loads. For the investigation on behavior of
stone columns, marine clay was collected from the Uran site near Mumbai,
India. Slurry consolidated clay beds were prepared and gravity loading mech-
anism was adopted to consolidate the clay beds in the laboratory. Laboratory
tests were performed to ascertain the uniformity and repeatability of the clay
beds by slurry consolidation. Augers of 50, 75 and 100 mm diameter were used
to bore the holes in the clay bed and aggregates were used for stone column
preparation by replacement method. The static displacement controlled tests
were performed on the stone column reinforced clay bed. In the present study,
the loading is applied on stone column alone, as it leads to the ultimate axial
capacity of stone column improved ground. Clay bed reinforced with stone
column exhibits improved load-settlement response, compared to unreinforced
clay beds. The improvement was very significant at higher settlements. The
failure load of reinforced clay bed is around 6 times more than the unreinforced
clay bed. In order to further understand the behaviour of stone column rein-
forced ground, numerical modeling of stone column reinforced clay bed is
conducted using FLAC3D. Using the validated numerical model of the stone
column, the effect of variation of the diameter of stone column on its perfor-
mance was studied. From the numerical analysis and physical tests conducted in
laboratory, it is observed that the stone columns with smaller diameter when
subjected to vertical loading carried higher bearing pressures when compared to
that of the larger diameter stone columns, which may be due to the greater
confining and larger bulge formed at a depth of 2–3 times the diameter of stone
columns.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_11
126 V.B. Chauhan et al.

1 Introduction

The forced intrusions in the form of aggregates/stones for the densification of the soft
soil strata are commonly referred as granular columns or stone columns. Stone column
construction involves partial replacement of weak subsurface soils by the vertical col-
umn of compacted granular material that usually penetrates the weak strata to reinforce
the surrounding soft soil and accomplish the ground improvement, so that it can safely
withstand the superimposed loads (Barksdale and Bachus 1983). These stone columns
are quite handy in significantly improving the bearing capacity of marine clay, which
also supplements by dissipating excess pore water pressure and reducing the settle-
ments. The vertical axial load applied at the top of stone column also induces lateral
movements in the top portion of the stone column, thereby displacing the surrounding
soil leading to formation of a bulge up to a depth of 2 to 3 times the diameter of stone
column below the ground. This bulge in turn, helps the stone column to develop
additional load carrying capacity by generating passive pressure conditions in the sur-
rounding soil, which provides additional confinement for the stone column through the
development of hoop stresses. To ascertain the behavior of stone column, laboratory
tests and numerical analysis using FLAC3D is performed. Also, a parametric study using
physical modeling tests on the effect of variation of diameter of stone column installed in
the consolidated clay bed on bearing capacity is studied in the present study.

2 Geomaterials and Experimental Program

The marine clay used in for the present study is obtained from the sea coast of
Dronagiri, near Uran, Navi Mumbai, India. Marine clay, thus procured, is placed in
specially prepared large bins of size 1.5 m wide, 3.5 m long and 2.5 m deep, and is
covered with tarpaulin sheets throughout the year to prevent from adverse weather
conditions. The particle size distribution curves of marine clay and aggregate (used for
stone column) are shown in Fig. 1(a). The pre-consolidation pressure of marine clay is
determined according to the BIS: 2720 Part 15 (1986) and is found to be 18 kPa, as
shown in Fig. 1(b). Physical properties of marine clay used for preparation of clay bed
are presented in Table 1.

Fig. 1. (a) Grain size distribution of marine clay and aggregates (b) pre-consolidation pressure
of marine clay
 Behaviour of Stone Columns Installed in Mumbai Marine Clay 127

Table 1. Physical properties of marine clay

Properties Values
Natural moisture content, w (%) 84
Field density, c (kN/m3) 14.96
In-situ vane shear strength, Su (kPa) 8 to 9
Specific gravity of soil solids (GS) 2.74
Liquid limit (in-situ conditions) (%) 101
Liquid limit (oven dried) (%) 51
Plastic limit (%) 40
Plasticity index (%) 61
Classification of soil OH
Pre-consolidation pressure, pc (kPa) 18
Free swell index (%) 31

For preparing the clay bed by consolidation process, amount of water equal to 1.5
times liquid limit of the soil is added, so as to have homogeneous sedimentation
(Sridharan and Prakash 2003; Murugesan and Rajagopal 2007). The soaked clay is
made into slurry with the assistance of mechanical mixer. Murugesan and Rajagopal
(2007), Gneil and Bouazza (2009) and Shahu and Reddy (2011) consolidated the clay
slurry to form clay beds for conducting studies on stone column reinforced clay beds.
Murugesan and Rajagopal (2007), Shahu and Reddy (2011) adopted dead weight
consolidation. To understand the behavior and performance of stone columns in the
laboratory, unit cell concept is undertaken (Barksdale and Bachus 1983; Bae et al.
2002; Ambily and Gandhi 2007; Murugesan and Rajagopal 2007; El-Garhy et al. 2011;
Shahu and Reddy 2011). This unit cell is an imaginary cylindrical volume of soil with
concentric stone column and the boundary of the unit cell is the extent beyond which
the effect of stone column loading cannot be felt when subjected to critical loading.
Unit cell (shown in Fig. 2) is adopted in the present study, which consisted of a
detachable collar at the top and cylindrical tank in the bottom part. The detachable
collar placed above the unit cell has dimensions of internal diameter 350 mm, height of
250 mm and thickness 12 mm. To accomplish consolidation of large clay beds
simultaneously, new simplified and economical gravity loading based lever arm set-up,
as shown in Fig. 3, is devised in the present study.
Pressure equal to pre-consolidation pressure is applied to the clay slurry leading to
removal of excess pore water to achieve vane shear strength in the range of 8 to 9 kPa
and bring the consistency of the clay bed to that at the site and thereby simulating field
conditions in the laboratory. However, Barksdale and Bachus (1983) reported that
stone columns perform better in clays having undrained shear strength in the range of
15 to 50 kPa. So, the consolidation pressure is increased from 18 kPa to 36 kPa, which
resulted in undrained shear strength in the range of 20 to 22 kPa, which is in the
specified range to provide better performance of stone columns. On the completion of
consolidation after 30 to 35 days, the detachable collar is removed and the clay pro-
truding above the unit cell is trimmed and leveled.
128 V.B. Chauhan et al.

Fig. 2. Schematic layout of unit cell and collar with dimensions

Fig. 3. Specially designed loading frame for the consolidation of clay

 Behaviour of Stone Columns Installed in Mumbai Marine Clay 129

3 Parametric Study

Bearing capacity of stone column is also a function of H/D (height/depth) ratio of stone
column. Effect of diameter and H/D ratio of the stone column on the behaviour of stone
column reinforced ground has been studied by various researchers. Murugesan and
Rajagopal (2007, 2008, 2009, 2010) and Ambily and Gandhi (2004, 2007) adopted
H/D ratio of unit cell greater than 1.0 and above, whereas, Malarvizhi and Ilamparuthi
(2007) considered the H/D ratio of unit cell less than 1.0. Lovisa and Sivakugan (2015)
observed that for doubly drained specimen with the H/D ratio up to 3, the wall friction
does not play any role and same average degree of consolidation versus time charts can
be considered. Bae et al. (2002) conducted study of the stone columns having H/D
ratios as such as 5, 6, and 7. Malarvizhi and Ilamparuthi (2007) carried out similar
studies and observed that with increase in H/D ratio of stone columns the settlement
was reduced. So, H/D ratio for the present study is chosen 5, 6.67 and 10.
For preparation of stone column, replacement method is adopted to form fully
penetrating stone columns in the consolidated clay bed using an auger (assisted by a
centering plate) of diameters of 50 mm, 75 mm and 100 mm to prepare bore holes of
corresponding size (Fig. 4). The aggregates, used for construction of stone columns in
the marine clay bed, are procured locally. Aggregates passing through 12.5 mm and
retained on 2 mm sieves are used for 100 mm and 75 mm diameter stone columns;
whereas, for 50 mm diameter stone columns aggregates passing through 9.52 mm and
retained on 2 mm sieves are used, following the recommendations of Nayak (1983).
Aggregates are placed in each layer with the assistance of a hollow flexible pipe
(aggregate placer) attached to the bottom of a funnel so that they reach the bottom and
are not lost midway by getting stuck to the surrounding sticky marine clay. The
aggregates are then compacted in 10 layers with a steel rod and unit weight of the stone
column is maintained at 16 kN/m3. A layer of sand of 2 mm is placed over the surface
of completed stone column. Physical properties of aggregate and sand used in prepa-
ration of stone column is shown in Table 2.
Performance of stone column reinforced clay beds are undertaken in the geotech-
nical laboratory using 1-g small scale models. For static load tests on unreinforced clay
bed, a loading plate of 100 mm diameter is centrally placed over prepared consolidated

Fig. 4. Hole formed by auger and hole filled up with aggregates

130 V.B. Chauhan et al.

Table 2. Physical properties of aggregate and sand

Properties Aggregate Sand
D10 (mm) 4.5 0.65
D30 (mm) 7.0 0.80
D60 (mm) 8.5 1.2
Coefficient of uniformity, Cu 1.28 1.85
Coefficient of curvature, Cc 1.89 0.82
Classification as per USCS Poorly graded gravel (GP) Poorly graded sand (SP)
Internal friction angle (Degrees) 41º –
Unit weight (kN/m3) 16 –

clay bed and loading is applied at a constant displacement rate of 1.2 mm/min up to
50 mm displacement. Pressure-settlement response of clay beds consolidated at 18 kPa
and 36 kPa is studied. For static load tests on reinforced clay bed, loading plates of
12 mm thick and diameter equals to that of the stone column of diameter 100 mm,
75 mm and 50 mm are studied in the present study. Loading procedure is similar to
that of unreinforced clay bed. In the present studies, results from three tests carried out
on the consolidated clay bed in unit cell subjected to 36 kPa consolidation pressure are
presented.

4 Numerical Analysis

Clay beds with stone column having dimensions similar to laboratory test are analysed
using finite difference analysis tool FLAC3D. Elastic modulus of aggregate, clay and
steel are taken as 45 MPa, 4 MPa and 180 GPa. Poisson’s ratio of aggregate, clay and
steel are taken as 0.3, 0.45 and 0.29. Figure 5 shows the numerical grid generated to
simulate the unreinforced and reinforced clay beds placed inside steel tank. Clay bed
and aggregate are modelled as an elasto-plastic material following Mohr-Coulomb
failure criterion and tank is modelled as elastic material. The steel container consisting
of clay bed and stone column is modelled numerically, and the model is fixed at bottom
to represent the physical test conditions. Very fine grid is chosen for the simulation of
unreinforced and reinforced clay bed model as shown in Fig. 5. The interface (shown
in Fig. 6) between stone column and clay is modelled as linear spring-slider system
with interface shear strength defined by the Mohr-Coulomb failure criterion. The
normal stiffness (kn) and shear stiffness (ks) of the interface are taken as per guidelines
of Itasca (2011). Interface parameters, kn and ks, have been assigned a value of
6
108 kN/m2/m at the periphery of stone column. In the present analysis, diameter of
stone column is varied as 50, 75 and 100 mm and displacement at a constant incre-
mental displacement is applied up to 50 mm to observe the load settlement behaviour
of the same and results are validated with physical test results.
 Behaviour of Stone Columns Installed in Mumbai Marine Clay 131

Fig. 5. Numerical grid for unreinforced and reinforced clay bed

Fig. 6. Interface between stone column and clay bed in FLAC3D

5 Results and Discussion

5.1 Effect of Consolidation Pressure
With doubling of the consolidation pressure from 18 kPa to 36 kPa, there is 1.48 times
increase in the bearing pressure at settlement of 50 mm for the unreinforced clay beds
as shown in Fig. 7(a).
132 V.B. Chauhan et al.

(a)

(b)
Fig. 7. Pressure-settlement response of (a) unreinforced clay beds consolidated at 18 kPa and
36 kPa (b) reinforced clay bed with different diameter of stone columns

5.2 Influence of Diameter of Stone Column

The diameter of stone column has significant effects the behaviour and performance of
stone column reinforced clay bed. There is marked improvement in the bearing pres-
sure of the stone column reinforced clay bed over the unreinforced clay bed. The
increase in bearing pressure of clay bed with 100, 75, and 50 mm diameter stone
columns is 4, 5.5 and 5.8 time that of unreinforced clay bed, respectively, as shown in
 Behaviour of Stone Columns Installed in Mumbai Marine Clay 133

Fig. 7(b), at 50 mm settlement. It is observed that the experimental results for the clay
bed and the clay bed reinforced with different diameter stone columns are in tandem
with the numerical results. However, the numerical results are slightly on the higher
side. From the numerical analysis, bulging contours of stone column in lateral direc-
tions (x and z) are presented for 100 mm diameter stone column in Fig. 8. It can be
noted from these figures that the bulging of stone column extends up to a depth of 2D
of stone column, where D is the diameter of the stone column. Similarly, lateral strain
up to 15% is observed in the stone column, which resulted in creating passive pressure
in clay bed, and led to the increase in load bearing capacity of stone column.

Fig. 8. Displacement contours of stone column of 100 mm diameter

5.3 Improvement Factor ðIf Þ

The performance improvement of the stone column reinforced clay bed is represented
using a non-dimensional parameter – bearing pressure improvement factor (If), which is
the ratio of footing pressure (qc) for a reinforced clay bed at a given settlement to the
footing pressure (qo) for an unreinforced clay bed, at the same settlement, as defined
below:

If ¼ qc =qo ð1Þ

Improvement factor for the stone column reinforced clay bed is presented in Table 3.
Briefly discuss the results shown in Table 3.
134 V.B. Chauhan et al.

Table 3. Improvement factor for the stone column reinforced clay bed
Test Bearing pressure at Improvement Bearing pressure at Improvement
description 25 mm settlement factor (If)25 50 mm settlement factor (If)50
(kPa) (kPa)
Unreinforced 32 – 40 –
clay bed
Reinforced 201 6.3 233 5.8
clay bed
50 mm
Reinforced 164 5.1 223 5.6
clay bed
75 mm
Reinforced 115 3.6 164 4.1
clay bed
100 mm

6 Conclusions

In the present study, the behaviour of stone columns installed in very soft cohesive soil
(marine clay) and subjected to static loads is studied using small scale models tested at
1-g conditions. Slurry consolidation method is adopted to prepare clay beds, simulating
the natural conditions. The slurry consolidation is carried out in the unit cell itself.
Stone column reinforced clay beds are subjected to static load to ascertain their per-
formance by varying their diameter. In the present study, stone columns of diameters
50 mm, 75 mm and 100 mm are used to understand the behavior of stone column
reinforced clay beds subjected to static loading. A validated numerical model is also
developed in FLAC3D to have better insight into the stone column behavior. The
following conclusions are drawn from the present study.
1. With doubling of the consolidation pressure there is 1.48 times increase in the
bearing pressure at settlement of 50 mm for the unreinforced clay beds.
2. With the increase in the diameter of stone column there is decrease in the bearing
pressure of stone column which is attributed to the confining stresses, which is more
in smaller diameter stone columns.
3. Clay bed reinforced with stone column exhibits improved load-settlement response,
compared to unreinforced clay beds. The improvement was very significant at
higher settlements. The failure load of reinforced clay bed is around 6 times more
than the unreinforced clay bed, which is in tandem with the observations of other
researchers. It is noted that only top portion of the stone column, i.e. depth ranging
from 2 to 3 times diameter of stone column, undergoes bulging, and influencing the
response of stone column to external loads.
4. The numerical simulation results show bulging at a depth of 2D which develops
passive pressure conditions at the interface.
 Behaviour of Stone Columns Installed in Mumbai Marine Clay 135

References
Ambily, A.P., Gandhi, S.R.: Experimental and theoretical evaluation of stone column in soft
clay. In: Proceedings of ICGGE-2004, IIT Bombay, India, pp. 201–206 (2004)
Ambily, A.P., Gandhi, S.R.: Behaviour of stone columns based on experimental and FEM
analysis. J. Geotech. Geoenviron. Eng. 133(4), 405–415 (2007)
Bae, W.S., Shin, W.S., An, B.C.: Behaviours of stone column improved with stone columns. In:
Chung, S., Kashiwagi, S., Hong, S.W. (eds.) Proceedings of the 12th International Offshore
and Polar Engineering Conference, Kitakyushu, Japan, pp. 675–678, 26–31 May 2002
Barksdale, R.D., Bachus, R.C.: Design and construction of stone columns. Report
No. FHWA/RD-83/026, Federal Highway Administration, Washington, D.C., USA (1983)
BIS: 2720 (Part-15): Methods of Test for Soils-Determination of Consolidation Properties.
Bureau of Indian Standards, New Delhi (1985)
El-Garhy, B., Maraie, M., Youssef, A.: Behaviour of model footings resting on soft clay
reinforced by floating granular piles: experimental study. Int. J. Geotech. Eng. 5(4), 415–424
(2011)
FLAC 3D (5.0), Itasca (2011)
Gniel, J., Bouazza, A.: Improvement of soft soils using geogrid encased stone columns. Geotext.
Geomembr. 27(3), 167–175 (2009)
Lovisa, J., Sivakugan, N.: Tall oedometer: method to account for wall friction. Int. J. Geomech.
15, 1–9 (2015). 04014045
Malarvizhi, S.N., Ilamparuthi, K.: Comparative study on the behaviour of encased stone column
and conventional stone column. Soils Found. 47(5), 873–886 (2007)
Murugesan, S., Rajagopal, K.: Model tests on geosynthetic encased stone columns. Geosyn-
thetics Int. 14(6), 346–354 (2007)
Murugesan, S., Rajagopal, K.: Performance of encased stone columns and design guidelines for
construction on soft clay soils. In: Li, G., Chen, Y., Tang, X. (eds.) Proceedings of the 4th
Asian Regional Congress on Geosynthetics, Shanghai, China, pp. 729–734 (2008)
Murugesan, S., Rajagopal, K.: Shear load tests on stone columns with and without geosynthetic
encasement. Geotech. Test. J. ASTM 32(1), 35–44 (2009)
Murugesan, S., Rajagopal, K.: Studies on the behaviour of single and group of geosynthetic
encased stone columns. J. Geotech. Geoenviron. Eng. ASCE 136(1), 129–139 (2010)
Nayak, N.V.: Recent advances in ground improvements by stone columns. In: Proceedings of
Indian Geotechnical Conference, Madras, vol. 1, p. V-19 (1983)
Shahu, J.T., Reddy, Y.R.: Clayey soil reinforced with stone column group: model tests and
analyses. J. Geotech. Geoenviron. Eng. 137(12), 1265–1274 (2011)
Sridharan, A., Prakash, K.: Self-weight consolidation: compressibility behaviour of segregated
and homogeneous fine grained sediments. Mar. Georesour. Geotechnol. 21(2), 73–80 (2003)
 Performance of Recycled Plastic Pin
(RPP) for Slope Stabilization

M.S. Khan1(&), M. Sahadat Hossain2, M.A. Khan2,

and Mohammad Faysal2
1
Department of Civil Engineering, Jackson State University,
1400 John R Lynch Street, Jackson, MS 39217, USA
[email protected]
2
Department of Civil Engineering, The University of Texas at Arlington,
416 Yates Street NH 119, Arlington, TX 76019, USA
[email protected], {md.khan22,md.faysal}@mavs.uta.edu

Abstract. Surficial failures of highway slopes in clayey soils are quite common
throughout the United States. These failures commonly occur for the slope
constructed with expansive clay, especially after prolonged rainfall. These
failures are also predominant in North Texas area and cause significant main-
tenance problems for the Texas Department of Transportation (TxDOT). As an
alternative to the conventional slope stabilization technique, a green and cost
effective slope stabilization method using the Recycled Plastic Pin (RPP) had
been utilized and tested for its performance. RPPs are driven into the slope face
to provide additional resistance along the slip surface, which increases the factor
of safety against shallow slope failure. Current study summarizes the long term
performance of a highway slope on expansive clay reinforced with RPP. The
slope is located over US 287 near the St. Paul overpass in Midlothian, Texas.
Surficial movement had taken place over the slope, resulting in cracks over the
shoulder and near the bridge abutment. Three 50 ft. slope sections were stabi-
lized using RPPs in March 2011. In addition, two 50 ft. unreinforced control
sections were utilized between the reinforced sections to evaluate the perfor-
mance of slope sections stabilized with RPP. After installation of the RPPs, the
performance of the slope was monitored by using instrumented RPPs, incli-
nometers and topographic survey. The performance monitoring results indicate
that, maximum deformation of the reinforced slope section is less than 3.75 cm
(1.5 in.). However, more than 38 cm (15 in.) of vertical settlement was observed
at the control sections during the last 5 years monitoring period. Also, few
slope-sections just opposite side of reinforced slope at the same highway failed
during the monitoring period. Based on the last 5 years monitoring data, it was
summarized that RPP can be successfully utilized for slope stabilization.

1 Introduction

Shallow slope failure refers to surficial slope instabilities along highway cuts, fill slopes
and embankments. Moderate-to-steep slopes and embankments underlain by expansive
clay soils are susceptible to shallow landslides during intense and prolonged rainfall
events. In many cases, the failure surface is parallel to the slope face. Shallow slope

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_12
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 137

failures generally do not constitute a hazard to human life or cause major damage.
However, they can be a hazard to infrastructure, by causing damage to guardrails,
shoulders, road surfaces, drainage facilities, utility poles, and/or the slope landscaping
(Titi and Helwany 2007). In some cases, shallow slope failures can affect regular traffic
movement if debris flows onto highway pavements.
Typically, failure occurs because of an increase in pore water pressure and reduction
in soil strength due to the progressive wetting of the near-surface soil. This condition is
further exacerbated by moisture variations due to seasonal climatic changes, which
cause cyclic shrinkage and swelling of the upper soils. Cyclic shrinkage and swelling
leads to the change in void ratio of soil and lower the shear strength. Reduction of shear
strength results in sloughing and shallow slope failures which are predominant in the
North Texas. It poses a significant maintenance problem to the Texas Department of
Transportation (TxDOT) (Wright 2005). The depth of the shallow slope failure varies
with soil type and slope geometry, but generally ranges between 0.9 m (3 ft.) and 1.8 m
(6 ft.). (Loehr et al. 2007). Shallow failures often cause significant hazards to guard rails,
shoulders, and roadways, which, if not properly maintained, may require extensive and
expensive repairs.
Reinforcement is an effective remediation technique for relatively shallow slope
failure conditions. In situ reinforcement methods for stabilizing slopes and embank-
ments include soil nails, drilled piers, micro piles, and RPPs (Thompson et al. 2006).
Among the available methods, the RPPs have been recognized as a cost effective
solution for slope stabilization (Loehr and Bowders 2007; Khan et al. 2015). RPPs
were first being utilized in the state of Missouri and Iowa, as a sustainable option to
stabilize highway slopes and few field studies were conducted. The study summarized
that no further failure was noticed after the slopes were stabilized using RPPs and the
performance of the stabilized zone was promising (Loehr and Bowders 2007).
Typically, RPPs are fabricated from recycled plastics and waste materials (polymers,
sawdust, and fly ash) (Chen et al. 2007). It is a lightweight material and less susceptible
to chemical and biological degradation than other reinforcement materials. The use of
RPPs can reduce the waste volume entering landfills and provide additional demand for
recycled plastic (Sommers et al. 2000). A typical RPP is composed of High Density
Polyethylene, HDPE (55%–70%); Low Density Polyethylene, LDPE (5%–10%);
Polystyrene, PS (2%–10%); Polypropylene, PP (2%–7%); Polyethylene-terephthalate,
PET (1%–5%); and varying amounts of additives, i.e., sawdust, fly ash (0%–5%)
(McLaren 1995). The use of glass and wood fiber additives significantly improves the
modulus of elasticity for plastic lumber (Breslin et al. 1998). RPPs driven into the slope
face may provide an additional resistance along the slip plane that restricts the pro-
gression of the slip surface and increases the factor of safety. However, limited field
studies have been conducted on the potential use of RPPs in slope stabilization.
The current study presents the long term performance of RPP for a highway slope
stabilization. Surficial movement and a cracked shoulder were observed on a highway
slope located over highway US 287 in Texas. During March 2011, two 15.25 m (50 ft.)
sections on the southbound side of US 287 were reinforced using RPPs. Another
cracked zone formed over the shoulder of US 287 during the following year, and a new
15.25 m (50 ft.) section was reinforced considered on the southbound slope for the
stabilization, using RPPs. In addition, two 15.25 m (50 ft.) control sections were kept
138 M.S. Khan et al.

between the reinforced sections to compare the performance. The reinforced and
control sections of the slope were monitored after installation to evaluate the perfor-
mance of the slope, using instrumented RPPs, inclinometers and surveying equipment.

2 Project Background

The slope is located near Highway US 287, near the St. Paul overpass in Midlothian,
Texas. It is a 3(H): 1(V) fill slope with 30 ft. height. Surficial movement over the slope
and shoulder cracks were observed on the shoulder near the crest of the slope in
September, 2010. The cracks along the pavement shoulder might have occurred due to
the surficial movement of the slope. The site location and the photo of the cracked
shoulder are presented in Fig. 1.

(a) (b)
Fig. 1. (a) Site location (b) Cracks along the shoulder of Highway US 287

The geology of the location is characterized by the Eagle Ford formation. The
Eagle Ford, in the project vicinity, is composed of residual soils, consisting of clay and
weathered shale (shaly clay), underlain by unweathered shale. The weathered shale is
claylike and contains gypsum in-fills, jointed and fractured with iron pyrites. The
unweathered shale is typically gray to dark gray and commonly includes shell debris,
silty fine sand particles, bentonite and pyrite. The Eagle Ford formation consists of
sedimentary rock in the process of degrading into a soil mass. This formation also
contains smectite clay minerals and sulfates. The smectite clay minerals are highly
expansive in nature and may be a contributing factor to the expansiveness of the soil.
A previous study by Kibria and Hossain (2012) indicated that the dominant mineral of
the soil is montmorillonite mineral, which has high shrink/swell.
A field exploration program was undertaken, which indicated that the slope was
constructed with high plastic clayey (CH) soil. The liquid limits (LL) and plasticity
indices (PI) of the samples ranged between 48 to 79 and 25 to 51, respectively. In
addition, an increase in moisture below 5 ft. was observed, which ranged up to 20 ft.
Based on the back analysis with PLAXIS 2D the factor of safety was 1.05 with fully
soften shear strength (Khan et al. 2015).
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 139

3 Slope Stabilization Using RPP

The definition of the factor of safety of a slope is the ratio of resisting moment (MR) to
the driving moment (MD), as presented in Eq. 1. RPPs installed at the slope provided
an additional resisting moment (DMR) along the slip surface, thereby increasing the
resistance and factor of safety, as presented in Eq. 2. The schematic diagram of RPPs
as slope reinforcement is presented in Fig. 2.

FS ¼ MR =MD ð1Þ

FS ¼ ðMR þ DMR Þ=MD ð2Þ

Where,
MR = Resisting Moment along Slip Surface
MD = Driving Moment along Slip Surface
DMR = Additional Resisting Moment from Plastic Pin.

Fig. 2. Schematic diagram of slope reinforcing with RPP

Commercially available fiber reinforced RPP with square Section (4 in.  4 in.)
and 10 ft. and 8 ft. in length was utilized as reinforcement. RPP reinforcement with
different spacing, (i.e., 3 ft., 5 ft. and 6 ft. c/c), was utilized in the Reinforced Sec-
tion 1. Besides, a 4 ft. c/c spacing of RPP was utilized in Reinforced Sections 2 and 3.
The layout of RPP is presented in Fig. 3(a). The calculated factor of safety from the
FEM model (PLAXIS 2D) was 1.43, 1.48 and 1.54 for Reinforced Sections 1, 2 and 3,
respectively (Khan et al. 2014).
RPPs were installed in Reinforced Section 1 and Reinforced Section 2 on March,
140 M.S. Khan et al.

(a)

(b)
Fig. 3. (a) Layout of RPP at US 287 slope (b) Installation photo

2011, as presented in Fig. 3(b). Reinforced Section 3 was stabilized during March,
2012. A crawler-type drilling rig, having a mast-mounted vibrator hammer (model:
Klemm 802 drill rig along with KD 1011 percussion head drifter), was utilized to
install the RPPs. The crawler-type rig was suitable for the installation process over the
slopes, as no additional anchorage was required to maintain the stability of the
equipment, reducing labor, cost and time involved in the installation process. Based on
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 141

the study, the average installation time for a 10 ft. long RPP was 4 min, and a total of
100 to 120 RPPs could be installed in one day.

4 Instrumentation and Performance Monitoring

Regular topographic survey and inclinometers were used to monitor the performance of
RPP stabilized slope. To evaluate the performance of the reinforced slopes and to
monitor the horizontal displacement, inclinometers were installed after the completion
of field installation in Reinforced Section 1 and Reinforced Section 2.

4.1 Topographic Survey

To compare the performance between the reinforced section and control section,
topographic surveys were performed on monthly basis. The purpose of this survey was
to monitor the settlement of the crest of the slope.

4.2 Inclinometers
A total of 3 inclinometers, (designated as Inclinometer 1, 2 and 3) were installed in
Reinforced Section 1, Control Section and Reinforced Section 2 to monitor the hori-
zontal movement of the slope. The depth of each inclinometer casing was 30 ft. and it
was installed perpendicular to the slope surface, 20 ft. below the crest. The layouts of
the inclinometers are presented in Fig. 4. The inclinometers were monitored on a
bi-weekly basis.

Fig. 4. Layout of inclinometer

142 M.S. Khan et al.

5 Performance Evaluation
5.1 Topographic Survey
The total settlement over the crest of the southbound slope was measured during each
survey, in between, the total settlement plot for September 2016 is presented in Fig. 5.
The total settlement plot shows the control sections had significantly greater settlement
at the crest when compared to the reinforced sections. The maximum settlements were
38 cm (15 in.) and 24 cm (9.5 in.) in Control Section 1 and Control Section 2,
respectively, while Reinforced Section 1 had the lowest settlement of 7 cm (2.75 in.),
followed by the Reinforced Section 3 and Reinforced Section 2, with settlements of
16 cm (6.25 in.) and 14 cm (5.5 in.), respectively. It should be noted that Reinforced
Section 1 had the lowest spacing of RPPs (0.9 m, 3 ft. c/c) at the crest of the slope.
Reinforced Section 2 and Reinforced Section 3 had 1.21 m (4 ft.) c/c spacing at the
crest, which was higher and had higher deformation than Reinforced Section 1.
Maximum incremental settlement on the reinforced section was less than 3.75 cm
(1.5 in.) after five years of installation of RPPs.

6.6.12 10.6.12 12.13.12 7.1.13

5.15.14 12.24.14 6.18.15 12.15.2015
6.15.2016 8.15.2016 9.15.2016
Distance along Roadway (m)
0 10 20 30 40 50 60 70 80
0.0
5.0
Settlement (cm)

10.0
15.0
20.0
25.0
30.0
35.0
40.0 R - 01 C - 01 R - 02 C - 02 R - 03

[R = Reinforced Section; C = Control Section]

Fig. 5. Settlement at the crest of slope (Southbound side)

5.2 Inclinometer
The inclinometers were monitored on a monthly basis. After each monitoring, the
obtained data from the inclined plane of the inclinometer was normalized to the vertical
plane and the horizontal movements of Inclinometer 1 and Inclinometer 3 are shown in
Figs. 6 and 7 respectively. Inclinometer 2 malfunctioned; consequently, that data is not
presented here.
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 143

Fig. 6. Inclinometer - 1: cum displacement A-A with time

Fig. 7. Inclinometer - 3: cum displacement A-A with time

The comparisons of movement between Inclinometer 1 and Inclinometer 3 at

different depths (1.37 m and 3.2 m) are presented in Fig. 8. It was observed that
Inclinometer 3 had higher horizontal displacement (4.3 cm) than Inclinometer 1
(2.8 cm) at 1.37 m depth. A similar trend was observed at 3.2 m depth, where the
observed displacement was 3.9 cm and 1.6 cm for Inclinometer 1 and Inclinometer 3,
respectively. It should be noted that the maximum horizontal displacement took place
at the surface of the slope and consequently reducing with deeper depth.
144 M.S. Khan et al.

Fig. 8. Comparison of inclinometer - 1 & 3

Khan et al. 2014 conducted a numerical study on the slope stabilization using
recycled plastic pin. During the study, the change in horizontal displacement of RPP at
the crest of the slope is investigated using finite element method (FEM), with the
change in RPP spacing. The FEM results indicated that, with the increment in RPP
spacing from 0.2 m (2 ft.) to 2.4 m (8 ft.), the horizontal displacement of RPP
increases from 2.5 cm (1 in.) to 12 cm (4.7 in.). In addition, the observed horizontal
displacement from FEM analysis at the crest is in good agreement with the observed
displacement at the field with 0.9 m (3 ft.) and 1.2 m (4 ft.) spacing.

5.3 Performance of the Northbound Slope

The northbound slope of highway US 287 was inspected visually on a monthly basis
for last five years. A number of shallow slope failures were recorded on the control
(northbound of US 287) slope and the failure locations are shown in Fig. 9. Pho-
tographs of the failures of Location 1 to Location 6 are also shown in Fig. 10. First time
failures (Location 1 and Location 2) on the control slope were observed during
September 2013, after a rainfall event. Soil was backfilled and compacted within the
failure zone during the maintenance period. But the slope failed (Location 3) again
during August, 2014 after a period of a heavy rainfall. Soil was backfilled again within
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 145

Fig. 9. Schematic diagram of slope failure within northbound control slope.

the failure location which was a temporary solution to prevent the movement of the
slope. A global failure (Location 4) of the slope was observed on June, 2015 after a
massive rainfall of 125 mm during that month. A soil nail wall was constructed to
control the movement of the pavement and the whole slope was reconstructed with
some geogrid reinforcement on October, 2015. Even after the reconstruction, the slope
failed (Location 5) after a month of heavy rainfall. Most recent failure was recorded on
August, 2016 and levelled as Location 6.
During the failures of the northbound slope, some sharp increment of horizontal
displacement was observed at Inclinometer 1. The sharp increments of the horizontal
displacement might indicate that the southbound slope had the tendency to slip due to
the rainfall which provided additional load over the recycled plastic pin and may cause
the additional movement. However, the RPP provided resistance and no sign of failure
was observed at the reinforced section during the time of failure of the northbound
slope. Moreover, no failure was also reported at the control section of the southbound
slope. It should be noted that the control sections are located between the reinforced
sections. During the failure of the northbound slope, the control section should tend to
slip which might get resistance from the reinforced section. As a result, no failure was
observed at the control section of the southbound slope.
146 M.S. Khan et al.

Location – 01 (September, 2013) Location – 02 (September, 2013)

Location – 03 (August, 2014) Location – 04 (June, 2015)

Location – 05 (November, 2015) Location – 06 (August, 2016)

Fig. 10. Failure locations (Location 01–06) at the northbound of the US 287 slope

6 Summary

A highway slope located on the southbound side of highway US 287 near the St. Paul
overpass in Midlothian, Texas was stabilized using RPPs. Three 15.25 m (50 ft.)
sections were selected and reinforced, using RPPs, after a crack, caused by slope
movement, was observed on the shoulder. Additionally, two 15.25 m (50 ft.) control
sections were utilized between the reinforced sections to compare the performances
between the unreinforced and reinforced slopes. The field performance of the slope was
monitored using topographic survey and inclinometers. Performance monitoring results
for last 5 (five) years are summarized below.
 Performance of Recycled Plastic Pin (RPP) for Slope Stabilization 147

• The unreinforced control sections of the southbound slope had significant settle-
ment at the crest of the slope, as much as 38 cm whereas the incremental settlement
on the reinforced section was less than 3.75 cm.
• Maximum cumulative displacement for inclinometer 1, at a depth below 1.37 and
3.20 m were 2.80 cm and 1.60 cm respectively.
• Maximum cumulative displacement for inclinometer 3, at a depth below 1.37 and
3.20 m were 4.30 cm and 3.90 cm respectively.
• Both inclinometers showed the same trend that, cumulative displacement decreases
with the depth from the slope surface.
• Having the same geographic and climatic condition, the southbound control slope
failed several times at different locations.
Based on the current study, it can be concluded that the overall performance of the
reinforced sections was better than that of the control sections. The northbound control
slope failed several times during the monitoring period whereas no visual depression or
failure was observed within the southbound slope. Since the shallow failure is a major
issue for slopes constructed with high plastic clayey soil; RPPs could provide a cost
effective and sustainable solution to mitigate this problem.

References
Breslin, V.T., Senturk, U., Berndt, C.C.: Long-term engineering properties of recycled plastic
lumber in pier construction. Resour. Conserv. Recycl. 23(1998), 243–258 (1998)
Chen, C.W., Salim, H., Bowders, J., Loehr, E., Owen, J.: Creep behavior of recycled plastic
lumber in slope stabilization applications. J. Mater. Civ. Eng. 19(2), 130–138 (2007)
Khan, M.S., Hossain, S., Kibria, G.: Stabilisation using recycled plastic pins. J. Perform.
Constructed Facil. 229–234 (2017). doi:10.1201/9781315206202-11. Online publication date:
14 Jun 2017
Khan, M.S., Hossain, M., Lozano, N.: Numerical study of slope stabilization using recycled
plastic pin. In: Geo-Congress 2014 Technical Papers: Geo-characterization and Modeling for
Sustainability, pp. 3092–3101. ASCE, February 2014
Kibria, G., Hossain, M.S.: Investigation of geotechnical parameters affecting electrical resistivity
of compacted clays. J. Geotech. Geoenviron. Eng. (2012). doi:10.1061/(ASCE)GT.1943-
5606.0000722
Loehr, J.E., Bowders, J.J.: Slope Stabilization using Recycled Plastic Pins – Phase III. Final
report: RI98-007D, Missouri Department of Transportation, Jefferson City, Missouri (2007)
Loehr, J.E., Fennessey, T.W., Bowders, J.J.: Stabilization of surficial slides using recycled plastic
reinforcement. Transp. Res. Rec. J. Transp. Res. Board 2, 79–87 (2007). No. 1989
McLaren, M.G.: Recycled plastic lumber and shapes design and specification. In: Proceedings
13th Structures Congress, ASCE, vol. 1, pp. 819–833 (1995)
Sommers, L., Loehr, J.E., Bowders, J.J.: Construction methods for slope stabilization with
recycled plastic pins. In: Proceedings of the Mid-continent Transportation Symposium, Iowa
State University, Ames, Iowa, 15–16 May 2000
Thompson, M.J., White, D.J.: Design of reinforcement with small-diameter piles. In: Proceedings
of Advances in Earth Structures: Research to Practice, GSP no. 151, pp. 1–12 (2006).
Reston/VA
148 M.S. Khan et al.

Titi, H., Helwany, S.: Investigation of Vertical Members to Resist Surficial Slope Instabilities
(No. WHRP 07-03). Wisconsin Department of Transportation, Madison, WI (2007)
Wright, S.G.: Evaluation of Soil Shear Strengths for Slope and Retaining Wall Stability Analyses
with Emphasis on High Plasticity Clays. Federal Highway Administration, Washington, D.C.
(2005). FHWA/TX-06/5-1874-01-1
 Evaluation of Liquefaction Potential of New
Caledonian Nickel Ores

Samar Daoud1(&), Imen Said1, Samir Ennour2,

and Mounir Bouassida1
1
ENIT, Université de Tunis El Manar – Ecole Nationale d’Ingénieurs de Tunis,
LR14ES03-Ingénierie Géotechnique, Tunis, Tunisia
[email protected], [email protected],
[email protected]
2
Mining Engineering, MECATER, Rue Ibn Zohr, Immeuble Alizé,
2015 Le Kram, Tunisia
[email protected]

Abstract. Although Nickel ores are generally characterized as plastic to very

plastic silts they do liquefy under specific conditions. In fact, it was reported that
New Caledonian Nickel ores (NCNo) can undergo liquefaction during their
maritime transportation under wave motions. The purpose of this research is to
develop a new empirical method to evaluate the liquefaction potential of these
soils. First, the study consists on interpreting physical and mechanical charac-
terization tests carried on Nickel ore samples originating from different sites of
New Caledonia. This analysis permitted to conclude on the heterogeneity of
NCNo and enabled the identification of three major types namely, laterites,
earthy and grainy saprolites. The classification of these groups was based on the
grain size distribution and the Atterberg limits tests. Afterwards, a review of the
existing liquefaction potential evaluation methods is elaborated. The application
of these methods to the case of NCNo has not allowed concluding on the
liquefaction susceptibility of this material. Finally, a new method specific to
each NCNo type was introduced based on the physical and mechanical test
results.

1 Introduction

New Caledonian Nickel ores (NCNo) are obtained from open-cast mines and shipped
to smelters located in China, Japan, Australia, Korea, etc. Several case histories of ore
carriers capsize have been reported due to liquefaction of the transported material.
According to The International Maritime Solid Bulk Cargoes Code (IMSBC 2008),
these kind of cargoes are classified as Group A materials where several laboratory tests
namely the Flow Table, the Penetration test and the Proctor/Fagerberg test are rec-
ommended to predict the risk of liquefaction before transport.
With reference to Koromila et al. (2013), cargoes that are at risk of liquefaction are
those containing at least some fine particles and moisture and further mined or stored in
exposed areas which allow the soaking up of large amounts of water. Such cargoes at
the time of loading are typically in granular state and look like dry. However, whilst at

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_13
150 S. Daoud et al.

sea, they are subject to agitation due to the engine vibration; ship’s rolling as well as
swell impact. The oscillatory ship movement leads to resettling of the cargo particles
and compaction of the inter-granular spaces. This compaction raises the water pressure
sharply, forcing the particles apart, potentially leading them to lose direct contact. The
cargo loses its shear strength and thus conditions are created for the material to behave
like a liquid, i.e. to liquefy (IMSBC 2008). As a result, this latter can lead to a “free
surface effect” that may upset the carrier’s balance and causes it to capsize in heavy
swells.
The existing guidelines for identifying potentially liquefiable soils are generally
divided into two major types. The first one covers the laboratory tests enabling to
obtain accurate results especially when carried out on undisturbed samples. In fact,
many experimental tests carried out in laboratories are enough efficient since they are
able to reproduce similar solicitations and field conditions to evaluate the susceptibility
of a type of soil to liquefaction. The most famous tests are the direct shear test, the
cyclic triaxial shear test, shaking table test… The second method is the empirical
approaches developed based on various combinations of characterization test results,
namely the Atterberg limits, the grain size characteristics, and the natural water content.
These methods developed by researchers like Tsuchida (1970), Seed and Idriss (1982),
Andrews and Martin (2000), are deduced from field observations of liquefied and
non-liquefied soils after famous earthquakes. By means of these approaches, it is
possible to determine the likelihood of liquefaction triggering in a particular soil under
a given earthquake. Thus, the choice of this alternative is justified by its simplicity and
cost-effectiveness.
Unfortunately, the efficiency and accuracy of these methods are still questioned
since they do not provide reliable results covering all types of soil and solicitations.
Indeed, although researchers had made efforts on identifying the susceptible liquefiable
soil types based on mechanical and physical characterization tests, there is still no
qualitative and quantitative parameter in the Geotechnical field that could be used as an
evaluation tool of liquefaction susceptibility.
The final aim of this study is the elaboration of a method to evaluate the lique-
faction potential of NCNo prior to their shipping in order to avoid further incidents.
First, the study consists of interpreting physical and mechanical characterization tests
carried on Nickel ore samples originating from different sites of New Caledonia. This
analysis enabled the classification of the tested specimens into different groups and the
identification of three major types of NCNo. Afterwards, an evaluation of the lique-
faction potential of NCNo by means of the existing approaches is elaborated. Finally, a
new method to assess the liquefaction susceptibility adapted for each type of NCNo is
introduced.

2 Data Base Discussion and Interpretation

The first step in assessing the risk of liquefaction is to identify potentially liquefiable
soils according to their physical characteristics (grain size distribution, water content,
Atterberg limits…) and mechanical properties.
 Evaluation of Liquefaction Potential 151

As part of Rheolat project, a series of physical and mechanical tests were conducted
on numerous Nickel ore samples originating from different sites and mines of New
Caledonia. Grain size distribution, Atterberg Limits and Cyclic triaxial shear tests were
carried out by CERMES laboratories. Flow Table Tests were elaborated by SLN*.
In this section, the different test results are presented, discussed and interpreted in
order to classify the tested NCNo samples into groups and in final aim to evaluate their
liquefaction potential.

2.1 Physical Characterization of NCNo

Grain Size Distribution Test Results

Particle size analysis were conducted on 24 samples of NCNo to allow determination of
the soil nature corresponding to standard XP CEN TS/ISO 17892-4.
The obtained curves by combining dry sieving and sedimentometry results, point
out the heterogeneity of NCNo. The representation of grain size distributions on the
same graph in Fig. 1 enables distinguishing three major grading ranges. It is concluded
that each of these latters corresponds to a type of NCNo.

Fig. 1. Grading ranges of the three identified NCNo groups

Group 1: Laterites which are silty sands having most particles smaller than 80 µm.
Group 2: Earthy saprolites that are sandy silts containing a small fraction of coarse
particles (>2 mm) not exceeding 30%.
Group 3: Grainy saprolites which are grainy sands having a high fraction of coarse
particles (>30%).

Atterberg Limits Test Results

The Atterberg limits test is also carried out by CERMES laboratories on the same
samples cited above corresponding to standard NF P 94-051. It is used in order to
characterize the plasticity of a soil since it can influence the soil liquefaction properties.
152 S. Daoud et al.

The interpretation of the Atterberg limit test by representing the results on the
Casagrande graph confirms the repartition of the tested samples into three distingable
groups as shown in Fig. 2. It is also concluded that all NCNo are highly plastic silts.

Fig. 2. Representation of Nickel ore tested samples on the Casagrande Diagram

2.2 Mechanical Characterization of NCNo

First, a series of consolidated undrained tests with measure of pore water pressure
(CU + u) were carried out by CERMES laboratories on samples representative of the
three different NCNo groups. Second, Flow Table tests have been elaborated by SLN
laboratories on NCNo. The results of these mechanical testes are presented and
interpreted in this paragraph. The critical water content is defined as the value above
which the sample gets liquefied. At each test, the ratio of this value on the Liquid limit
of the tested specimen is determined in order to be used in the development of the new
approach (Table 1).
Cyclic Triaxial Shear Test Results CTT
The Cyclic Triaxial Tests (CTT) results allows to differentiate between a liquefiable
and a non-liquefiable material when reaching one of the two following conditions;
either the excess pore water pressure reaches the value of the consolidation stress
Δu = r′c or the peak-to-peak axial strain within a cycle is equal to 5%. In this para-
graph, the results of the saturation degree variation as well a comparison of the three
samples responses to cyclic shearing are discussed in order to evaluate the NCNo
susceptibility to liquefaction.
Figure 3 illustrates the effect of the degree of saturation (SR) variation on the liq-
uefaction properties in the case of laterites samples. The obtained shapes of the curves
are in accordance with those found by Pecker (1984) when analyzing the effect of the
 Evaluation of Liquefaction Potential 153

Table 1. CTT results on NCNo samples

Samples <80 µm (%) >2 mm (%) IP (%) Wc/LL
Laterites
L1 100 – 15 0.86
L2 92 – 16 0.7
L6 87.5 – 6 0.63
L7 86.6 – 15 0.65
Mean value 88.47 – 12.56 0.72
Earthy saprolites
ES1 78 – 43 0.71
ES2 55.6 – 48 0.56
ES4 93.8 – 51 0.57
ES5 33.8 – 61 0.58
Mean value 66.85 – 45 0.67
Grainy saprolites
GS1 40 40 9 0.7
GS2 37 48 11 0.67
GS3 47 34 19 0.55
GS4 36 42 24 0.46
GS9 42 33 16 0.64
Mean value 39 40.62 17.7 0.64

Fig. 3. Impact of the degree of saturation variation on the liquefaction properties in the case of
Laterite samples
154 S. Daoud et al.

degree of saturation on the cyclic shear resistance CSR of a soil. It is mentioned that the
required number of cycles to trigger liquefaction when the sample is partially saturated
is higher than that when the sample is fully saturated. Therefore, the decrease of the
degree of saturation improves the cyclic shear resistance of soils. This result is in
accordance with the statement of Sherif et al. (1977) who proved that liquefaction
resistance for soils increases with decreasing degree of saturation.
CTT were conducted on the three samples under the same Rcyc = 0.2. It is noted that
sample representing group 2 needs a higher number of cycles to liquefy in comparison
with the two other groups. This difference may be attributed to the high plasticity of
earthy saprolites (PI = 43) which is the origin of the liquefaction delay. Actually, as
reported by Alan F. Rauch, plastic fines usually create sufficient adhesion between the
grains to limit the ability of larger particles to move into a denser arrangement and
therefore to prevent liquefaction.
In contrast, sample from group 1 was the first to get liquefied compared to the two
other groups. Hence, laterites are assumed to be less resistant to liquefaction than
earthy and grained saprolites. This difference could be explained by the high per-
centage of fines in this sample (fraction < 80 µm = 100%). In fact, Finn et al. (1988),
Lade and Yamamuro (1997), Zlatovic and Ishihara (1995) reported that liquefaction
resistance decrease as the fine content increase. Moreover, according to Liang et al.
(2000), it is proven that the coarse particles take an important part in soil to resist
liquefaction and this is the case of grainy saprolites. Indeed, soils with a higher per-
centage of gravels tend to mobilize higher strength during shearing, and dissipate
excess pore pressures more rapidly than sands.

3 Methods to Evaluate the Liquefaction Potential

In this section, the widely used empirical approaches to assess the liquefaction sus-
ceptibility of soils are presented. Afterwards, an attempt to develop a new empirical
method adapted to NCNo is introduced.

3.1 Existing Empirical Approaches

Many empirical methods have been developed in order to predict the liquefaction
susceptibility of soils. These approaches have been developed to assess the soil
potential liquefaction using different combinations of soil characteristics.
To begin with, Tsuchida (1970) proposed particle size distribution boundary
curves, using the results of sieve analyses on soils that did or did not liquefy during past
earthquakes. Thus, representing the grain size curve of any soil on this graph, allows
deducing its liquefaction potential.
Besides, Seed and Idriss (1982) stated that cohesive soils which are susceptible to
liquefaction must fulfill the following criteria (combined):
• Percent finer than 0.005 mm < 15%
• Liquid Limit (LL) < 35%
• Water Content > 0.9  Liquid Limit
 Evaluation of Liquefaction Potential 155

If soils with these characteristics, plot below the A-line on the Plasticity chart, the
best means of determining their cyclic loading characteristics is by additional test.
Otherwise, clayey soils may be considered non-vulnerable to liquefaction.
In addition, the widely used Chinese criteria as presented by Koester (1992) on
Fig. 4 showed that based on the soil representation on the Casagrande graph potentially
liquefiable and non-liquefiable soils can be distinguished.

Fig. 4. Chinese Criteria Adapted to ASTM Definitions of Soil Properties (Perlea et al. 1999)

Furthermore, Andrews and Martin (2000) developed an empirical approach based

on LL and percentage of fraction passing 2 µm presented in the following table.
Last but not least, Seed et al. (2003) recommended an assessment chart shown in
Fig. 8 which is divided into three zones; Zone A where soils are considered potentially
susceptible to liquefaction if Wc > 80% LL; Zone B where soils are considered
potentially liquefiable with detailed laboratory testing recommended if Wc > 85% LL
and Zone C where soils are considered generally not susceptible to classic cyclic
liquefaction, although they should be checked for potential sensitivity.

3.2 Development of a New Empirical Approach for NCNo Case

Based on the existing widely used empirical approaches to evaluate the liquefaction
potential of soils, a new method is proposed in this research which is more adapted to
the case of NCNo. Most of the existing criteria are based on field observations where
some soils got liquefied and some not under certain earthquakes. In this study, since it
is not the case of a seismic solicitation, the results of CTT reproducing the transport
conditions, are exploited to distinguish liquefiable soils from non-liquefiable ones and
156 S. Daoud et al.

therefore a criterion for liquefaction is deduced based on the soil’s physical and
mechanical characterization.

4 Results

4.1 Evaluation of NCNo Liquefaction Potential Based on Existing

Empirical Methods
The liquefaction susceptibility of NCNo samples was evaluated by means of the
existing empirical approaches widely used in para-seismic analysis.
First, the grading ranges of each NCNo type was added to Tsuchida (1970) chart. It
is seen from Figs. 5, 6 and 7 that all NCNo samples curves pass through both the
‘Potentially Liquefiable’ and ‘Most Liquefiable Soil’ boundaries. Therefore, according
to this criterion and based on the representative grading ranges of each NCNo type, it
can be concluded that the NCNo are susceptible to liquefaction under certain cir-
cumstances. Further, it should be mentioned when comparing the three obtained graphs
that the most liquefiable type of NCNo is the Group 1 since its grading ranges falls
exactly between the boundaries of potentially liquefiable soils. However, for Group 2
and 3 of NCNo only a part of their grain size distribution curves are plotted inside the
boundaries defined by Tsuchida (1970).

Fig. 5. Representation of the grain size distributions of Laterites NCNo on the Tsuchida graph

However, conforming to Seed and Idriss (1982) approach, NCNo are not prone to
liquefaction since all their Liquid limits are above 35% and for all tested samples the
water content does not exceed 0.9  Liquid Limit (Cf. Table 1).
Moreover, according to the widely used Chinese criteria as presented by Koester
(1992) NCNo are classified as non-liquefiable soils since all Liquid limits are above
33.5% and all the water content does not exceed 0.87  Liquid Limit (Cf. Table 1).
 Evaluation of Liquefaction Potential 157

Fig. 6. Representation of the grain size distributions of Earthy Saprolites NCNo on the Tsuchida
graph

Fig. 7. Representation of the grain size distributions of Grainy Saprolites NCNo on the
Tsuchida graph

In addition, following Andrews and Martin (2000) method, the three types of
lateritic soils require further studies since their LL values are higher than 32% and their
Minus 2 µm fraction is generally less than 10% (Cf. Table 2).

Table 2. Liquefaction susceptibility criteria by Andrews and Martin (2000)

LL < 32 LL  32
Minus 2 µm Susceptible to liquefaction Further studies required [Consider
fraction < 10% plastic non-clay sized grains]
Minus 2 µm Further studies required [Consider Not susceptible to liquefaction
fraction  10% plastic non-clay sized grains]
158 S. Daoud et al.

Besides, the representation of NCNo on the graph recommended by Seed et al.

(2003) showed that all the tested samples are plotted in zone C and thus are considered
non susceptible to classic cyclic liquefaction, although they should be checked for
potential sensitivity (Fig. 8).

Fig. 8. Representation of the tested lateritic soils on Seed et al. (2003) recommended chart

Although these methods are well known and widely used, their usability is in doubt
since they do not allow asserting on the liquefaction potential of NCNo. In fact while
some of these methods consider that the tested Nickel ore samples are non-liquefiable,
others claim they are rather susceptible to liquefaction or even liquefiable soils. This
disagreement may be due to first, the fact that most of these empirical approaches were
developed primarily for sands and not particularly for this type of soils (highly plastic
silts). Second, these methods are based on observations of soil behavior under seismic
loading which is doubtless a different solicitation comparing to the one due to swell
effect.
For these reasons, in the following paragraph an adapted method to NCNo based on
their physical and mechanical characterization is proposed in order to assess their
liquefaction potential during maritime transport.
 Evaluation of Liquefaction Potential 159

4.2 New Proposed Method for Evaluating the Liquefaction Potential

of NCNo
As mentioned before, the existing empirical approaches used in para-seismic analysis
are not adopted to the New Caledonian Nickel ores. Therefore a new method to
evaluate the liquefaction potential of this type of soil when exposed to the swell
solicitations is developed. This approach is mainly based on the Seed and Idriss (1982)
method and Seed et al. (2003) criteria adding appropriate modifications to fit with the
tested samples according to Cyclic Triaxial Test (CTT) results (Cf. Table 1). Moreover,
taking into account the heterogeneity of NCNo, a criterion for each ore type is elab-
orated considering its specificity compared to the others (Cf. Table 3). These
approaches are applicable for fully saturated samples.

Table 3. New approach for liquefaction assessment of New Caledonian Nickel ore under the
swell effect
New approach for liquefaction assessment of New Caledonian Nickel
ore under the swell effect
Laterites Earthy saprolites Grainy saprolites
• Infra 80 µm > 80% • Infra 80 µm  50% • Supra 2 mm > 30%
• IP  10% • MBV  5 • IP  12%
• Wc > 0,6 LL • IP  30% • Wc > 0,5 LL
• Wc > 0,55 LL

5 Conclusions

First, the interpretation of physical characterization tests carried on NCNo allowed

identifying three major types of this material:
• Group 1: Laterites which are very fine soils having most particles smaller than
80 µm. Based on their characterization tests, they are classified as highly plastic
silty sands.
• Group 2: Earthy saprolites which are sandy silts containning a small fraction of
coarse particles (>2 mm) not exceeding 30%. According to the United Soil Clas-
sification System (LCPC-USCS), this soil is a high plastic silt having a high clay
content (MBV > 5).
• Group 3: Grainy saprolites are grainy sands having a high fraction of coarse par-
ticles (>30%).
Second, the discussion of CTT results conducted on the three types of NCNo
proved that these latters behave differently under the same solicitations. In fact, it is
mentioned that earthy saprolites may be the less liquefiable type of NCNo. This high
resistance to cyclic shearing may be attributed to the high plasticity of this ore (PI =
43) which is believed to contribute to the liquefaction delay. On the other hand,
laterites appears to be the most susceptible type of NCNo to liquefaction since they
acquires the less number of cycles to reach the criteria. This difference was explained
160 S. Daoud et al.

by the high percentage of fines in this sample (fraction < 80 µm = 100%). For the
grainy saprolites, it is proven that the coarse particles take an important part in soil to
resist liquefaction.
Afterwards, the evaluation of the liquefaction potential of NCNo by means of the
existing methods hadn’t provided reliable results to conclude. In fact, according to
Tsuchida (1970) chart, all three types of NCNo appear to be susceptible to liquefaction.
However, conforming to Seed and Idriss (1982) approach as well as the Chinese
criteria as presented by Koester (1992), NCNo are not prone to liquefaction since they
are characterized by high vales of Liquid limits and for all tested samples the water
content does not exceed 0.87  Liquid Limit.
Finally, a new method adapted to NCNo was developed to assess the susceptibility
of each type of ore to liquefaction. This approach has been inspired from the existing
criteria and based on the physical and mechanical test results.

Acknowledgments. This research has been developed with the financial support of CNRT
(Centre National de Recherche Technique “Nickel et son Environnement”) and SEM (Syndicat
des Producteurs-Exportateurs et Exportateurs de Minerai de Nickel de NC) in the scope of
Rheolat I and II projects 2011–2016. We thank them both for providing insight and expertise that
greatly assisted this research. We would also like to show our gratitude to the CERMES-Navier
laboratories who has conducted the physical characterization and the Cyclic Triaxial tests.

References
Andrews, D.C.A., Martin, G.R.: Criteria for liquefaction of silty soils. In: Proceedings of 12th
WCEE, Auckland, New Zealand (2000)
Finn, W.D.L.: Dynamic analysis in geotechnical engineering. In: Proceedings, Earthquake
Engineering and Soil Dynamics II – Recent Advances in Ground Motion Evaluations.
Geotechnical Special Publication, ASCE, vol. 20, pp. 523–591 (1988)
IMSBC: Adoption of The International Maritime Solid Bulk Cargoes Code (2008), Annex 3,
Resolution MSC.268 (85) (2008)
Koester, J.P.: The influence of test procedure on correlation of Atterberg limits with liquefaction
in fine-grained soils. Geotech. Test. J. 15(4), 352–360 (1992). ASTM
Koromila, I.A., et al.: Experimental investigation of cargo liquefaction and impact on the stability
of a bulk – carrier. In: Proceedings of the 13th International Ship Stability Workshop, pp. 1–7
(2013)
Lade, P.V., Yamamuro, J.A.: Effects of non-plastic fines on static liquefaction of sands. Can.
Geotech. J. 34(6), 918–928 (1997)
Liang, R.W., et al.: Effect of clay particle content on liquefaction of soil. In: 12th World
Conference on Earthquake Engineering, Auckland, New Zealand (2000)
Pecker, A.: Dynamique des sols «Collection géotechnique», 259 p. Presses de l’École nationale
des ponts et chaussées, Paris (1984)
Perlea, V.G., Koester, J.P., Prakash, S.: “How Liquefiable are Cohesive Soils?” In: Proceedings
of the Second International Conference on Earthquake Geotechnical Engineering, Lisbon,
Portugal, vol. 2, pp. 611–618 (1999)
Seed, H.B., Idriss, I.M.: Ground motions and soil liquefaction during earthquakes, p. 134.
Earthquake Engineering Research Institute, Berkeley (1982)
 Evaluation of Liquefaction Potential 161

Seed, R.B., et al.: Recent advances in soil liquefaction engineering: a unified and consistent
framework. EERC-2003–06. Earthquake Engineering Research Institute, Berkeley (2003)
Sherif, M.A., et al.: Saturation effect on initial soil liquefaction. J. Geotech. Eng. Div. 103,
914–917 (1977). American Society of Civil Engineers
Tsuchida, H.: Prediction and countermeasure against liquefaction in sand deposits. Abstract of
the Seminar of the Port and Harbour Research Institute, Ministry of Transport, Yokosuka,
Japan, pp. 3.1–3.33 (1970)
Zlatovic, S., Ishihara, K.: On the influence of non-plastic fines on residual strength. In:
Proceedings of IS-Tokyo 1995, First International Conference on Earthquake Geotechnical
Engineering, vol. 1, pp. 239–244 (1995)
 Effect of Variability of Soil Parameters
in the Behavior of Shallow Foundations

Tahar Messafer(&)

Civil Engineering Department, University M’Hamed Bougara Boumerdes,

Boumerdès, Algeria
[email protected]

Abstract. This work is divided into two folds. The first stage was to carry out a
statistical study of geotechnical soil parameters obtained for a housing project.
This comprised analysis of the mean, standard deviation, coefficient of variance,
histograms, cumulative densities and normal distribution laws. In addition
correlation using linear regression analysis was carried out between pairs of soil
parameters obtained from laboratory tests. Then autocorrelation functions were
developed for pressuremeter modulus.
The second stage was to develop a probabilistic approach to design shallow
foundations. This assumed that strengths parameters such as cohesion and angle
of shearing strength are variable. The results were compared to traditional
techniques based on Terzaghi methods such as DIN and DTU and also to semi
probabilistic approach such as Eurocode.

Keywords: Statistical analysis
Correlation
Laboratory tests
In situ test

Foundation analysis
Probabilistic approach

1 Introduction

The design of civil engineering structures requires a good knowledge of the subgrade.
The first stage of a project is a geotechnical investigation as it allows the engineer to
select representative values of soil characteristics necessary for the design. However, it
is impossible to define in any point of a site the soil properties because the determination
of representative parameter values is generally carried out on the basis of a few samples
taken almost at random and in situ tests executed following a more or less wide mesh.
This is why the development of methods of statistical analysis and probability for
the characterization of physical and mechanical properties of soils should solve the
problem of variability of soil parameters. Then the use of these statistical methods in
the probabilistic foundation design might be more advantageous compared to tradi-
tional methods used at present.

2 Site Presentation

The project is located in a mountainous region (Medea) in Algeria and includes the
completion of 500 and 200 housing units as part of a 2000 housing program. The inves-
tigation in situ comprised the completion of 11 boreholes of 15 ml, 6 boreholes of 20 ml,
© Springer International Publishing AG 2018
W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_14
 Effect of variability of soil parameters 163

5 pressuremeter tests of 10 ml, 60 dynamic penetration tests and 2 piezometers 20 ml each

deep.
Analysis of the results indicated the presence of two main layers which are:
Layer 1: silty clay with a thickness of about 4 m
Layer 2: gray marl

3 Statistical Analysis of Geotechnical Parameters

Data analysis will focus on the results of laboratory and in situ tests for model one layer
(Site) and model two layers (clay and marl). The results compare values of the min,
max, average, standard deviation and coefficient of variance (CV) for the 2 models.
Several geotechnical parameters were analyzed such as the density, water content,
degree of saturation, grading analysis, Atterberg limits, compressibility parameters,
compression index, swelling coefficient, cohesion and angle of friction. As well as the
results of pressuremeter tests.
Only limits results obtained for models one layer and two layers are provided on
Table 1 and these show that the two layers are both plastic.

Table 1. Summary of models Atterberg limits single layer, two layers

Paramètres Model Nbre sampl Min Max Average Stand CV
retained values values values déviation (%)
WL (%) Site 36 45 59 54,24 4,13 7,62
Clay 26 46 59 54,08 4,21 7,79
Marl 10 45 59 54,10 4,18 7,72
IP (%) Site 36 22 29 27,00 2,09 7,76
Clay 26 23 29 26,96 2,16 8,02
Marl 10 22 29 25.50 2,11 8.27

4 Analysis of Variability

Histograms and cumulative distribution of geotechnical parameters have been devel-

oped. Results for plasticity index (Ip) are shown on Fig. 1. A graphical representation
of classes is given in histograms shown in Fig. 2.
From the results it can be seen that the Ip histogram for clay is identical to the site
histogram. Hence to calculate the frequency or density of the site, clay data can be
taken. Adopting Fig. 1, the cumulative distribution of Ip is shown in Fig. 3.
164 T. Messafer

Fig. 1. Distribution de Ip with depth for model 1(a) and model 2(b and c)

Fig. 2. Histograms of Ip

Figures 4, 5 and 6 show Gauss normal laws f(z) and the corresponding probabilities
for Ip for the whole site and layers of Clay and Marl respectively.
 Effect of variability of soil parameters 165

Fig. 3. Cumulative distribution of Ip

Fig. 4. Normal distribution de Ip pour le Site

Fig. 5. Normal distribution of Ip for clay

166 T. Messafer

Fig. 6. Normal distribution of Marl

5 Correlations and Lineair Regressions

Figure 7 shows results of water content and Atterberg Limits correlations. The result gave
good correlations between WL and IP for the whole site, clay and marl layers reflecting
similarity of site materials origins. These results are similar to correlations obtained by
several authors, for example by Magnan [1]: WL ¼ 1:04 Ip þ 27:02 R = 0.90

Fig. 7. Relations between water content and Atterberg limits

6 Spatial Auto-correlation

Analysis of spatial variability of physical and mechanical properties of the site was
performed for the pressuremeter survey. The analysis was carried out in the vertical
direction as it contained enough regularly spaced data. The purpose of this analysis is to
determine the auto-correlation function which describes the representation given in the
 Effect of variability of soil parameters 167

vertical direction and also to determine the remote auto correlation (or fluctuation scale)
which determines the degree of dispersion of data.
The auto-correlation is a measure of the dependence of two or more points in a data
set. From measurements yi given xi equidistant from a property, the estimated value of the
auto-correlation between two values distant by x can be defined as follows, Salloum [2]:
Pns
i¼1 ðyi
yi Þðyi þ s 
yi þ s Þ
qs ¼ Pn s ¼ 0; 1; 2; . . .; n  1 ð1Þ
i¼1 ðyi 
yi Þ 2

Where n is the number of measurements of the soil property and s is the shift of the
data set.
yi and
yi þ s are values of the measured trends at xi and xi þ s respectively.
Among the auto-correlation function models, exponential functions of the fol-
jsj
lowing form: C:e a can be used (Imanzadeh [3]):
s represents the distance between two points of the ground where it is desired to
determine the correlation and a is the auto correlation distance.
A simple but approximate method for calculating the vertical scale of fluctuation is
given by Vanmarcke [4] as follows:

1 Xn1
dv  0:8d
with d
¼ d
i¼1 i
ð2Þ
n1

Where dv is the vertical fluctuation scale and d
is the average value of distances limited
by the intersections of the trend function with the function n(z) of soil property.

6.1 Drift Average Values (Linear Regression)

Observation of measured values distributions of soil parameters as a function of depth
leads to visually distinguish two layers: a 4 m thick followed by a layer of 6 m as
shown in Fig. 8. This represents measured values of pressuremeter module (EM) in a
pressuremeter test.

Fig. 8. Spatial variability of pressuremeter module (EM)

168 T. Messafer

Fig. 9. Autocorrelation diagrams and EM autocorrelation functions in clay and marl

Fig. 10. Capacity-Solicitation model

6.2 Functions and Auto-correlation Distances

Figures 9 and 10 show the auto-correlation charts and autocorrelation functions of EM
in the vertical direction for each layer separately (Marl and clay) and for the whole site.
j xj
For each auto-correlation diagram, an exponential function of the form, y ¼ C:e a ,
has been adjusted on the first three or four values of the coefficient of auto-correlation,
distance of auto correlation and fluctuation scale h = 2a.
Table 2 shows the results of the function of autocorrelation adjusted for pres-
suremeter modulus data in the clay and the marl.
The auto-correlation function in the vertical direction of EM was determined for
clay and marl. The best results were obtained with the exponential function (Figs. 9
and 10) for which the value of R2 is the highest.
The R2 value obtained was found greater than 0.903, and the vertical autocorre-
lation distance was found around 0.52 m in clay and 0.59 m in the marl. This means
that the pressuremeter modulus vary in the clay layer in an identical manner to the marl.
 Effect of variability of soil parameters 169

Table 2. Autocorrelation function, fluctuation scale h wide field and the coefficient of
determination R2 in clay and marl and the whole site
Paraméter Nature Auto-corrélation Auto-corrélation Fluctuation R2
of soil function qðsÞ distance a scale h
EM Clay y = 3,747e−1,32x 0.266 0.52 1
Marl y = 3,383e−1,21x 0.295 0.59 1
Site y = 5,188e−1,38x 0.72 m 1.44 m 0,903

The vertical distance of autocorrelation obtained for EM varies between 0.5 and
1.44. This indicates that these values are not dispersed.

7 Fondation Calculation

A probabilistic approach enables the study of the risk of failure by taking into account
the variability of geotechnical parameters and also the variability of the pressure acting
on the foundations. In what follows, traditional calculation results (Terzaghi, DTU,
DIN) and Eurocode will be compared to probabilistic results, considering a normal
distribution law.

7.1 Calculation of Bearing Capacity from Laboratory Tests

In the case of a strip footing, the bearing capacity under a vertical centered load is given
by the following general relation (superposition method of Terzaghi):

1
ql ¼ c BNc ðuÞ þ cNc ðuÞ þ ðq þ c2 DÞNq ðuÞ ð3Þ
2 1

With:
ql and q: bearing capacity and vertical lateral pressure respectively
c1 ; c2 : Volumetric weight of the soil under the base of the foundation and laterally
respectively
c: Cohesion of soil under the base of the foundation
B and D: Width of the foundation and ancrage depth of the foundation, respectively
Nc ðuÞ,Nc ðuÞ and Nq ðuÞ : Bearing Factors of the foundation
Table 3 contains the formulas of these three factors according to the French,
German and Eurocode conventional concepts.

Table 3. Bearing capacity factors

Concept Bearing factors
Nc Nc Nq
   
Conventionnel French 1,85ðNq - 1Þtanu Nq  1 cotu ep tanu
tan2 4p + 2u
(DTU13.12)
Conventionnel German 2ðNq  1Þtanu
Eurocode 7-1 2ðNq  1Þtanu
170 T. Messafer

7.2 Allowable Bearing Capacity of a Foundation (According to Terzaghi)

ql  c1 :D
qa ¼ c1 :D þ ð4Þ
F

7.3 French Conventional Concept (DTU13.12)

qreel  qad ð5Þ

qreel , qad : real stress applied to the foundation and allowable pressure respectively
 
c B2 Nc þ cD Nq 1 þ cNc
qad ¼ cD þ ð6Þ
Fs

7.4 Conventional German Concept DIN 1054

Vb
 Fs With Vb ¼ qL : A ð7Þ
V

A: area of the foundation and Fs: global safety factor

Vb and V: Limit load and external load applied on the foundation

7.5 Eurocode 7.1 (Semi-probabilistic Calculation for Safety)

Partial safety factors are applied to the soil parameters (c, c, u). The bearing capacity
calculation is calculated as follows:
 
Rd ¼ QL u0d ; c0d ; cd ð8Þ

c0 tan u0k
With, c0d ¼ c k0 ; tan u0d ¼ cu0 cd ¼ ck
c

The variation of c specific weight is very low, for this reason we take c ¼ 1.
cu0 ; cc0 : Partial safety factors applied to u′ and c′ respectively
Indices k and d mean characteristic value and design value respectively

7.6 Probabilistic Method

The few published work on this subject (Mounji et al. [5], Chew et al. [6]) has shown the
importance of taking account of soil variability. In this work, the capacity (Q) and solic-
itation (S) are considered random variables each having a mean and standard deviation.
The difference between the two random variables ðZ ¼ QSÞ is called margin of
safety and is also a random variable. A limit state is reached when Z ¼ 0 and failure
occurs when Z \ 0, and the probability of failure can be defined as (Boutahir et al. [7]):
 Effect of variability of soil parameters 171

Pf ¼ Pf ½ðQ  SÞ  0 ¼ Pf ðQ  SÞ ð9Þ

Where R and S follow Gauss normal distributions, the safety margin (Z) also follows a
normal distribution.

1 2
f ðZ Þ ¼ pffiffiffiffiffiffi : eðZlZ Þ =2sz
2
ð10Þ
sz 2p

Whose mean and standard deviation are respectively:

qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
lZ ¼ lR  lS and sz ¼ s2R  s2S

The hatched area of Fig. 10 being the failure probability Pf calculated with the integral:
Z 0
Pf ¼ fZ ðZ Þ:dZ ð11Þ
1

Bearing capacity ql , based on random variables u, c and c

The function ql ðu; c; cÞ can be developed into a Taylor series. If we take the terms of
the lowest order non null, we can calculate the mathematical expectation
y and the
variance S2y of a function y = f ðx1 ; x2 . . .xn Þ, (Hahn and Shapiro [8]), with xi random
variables defined by their average li and their variance S2i :

1 Xn d2 f 2

y ¼ f ðl1 ; l2 ; . . .:; ln Þ þ i¼1 d2
ðSi Þ ð12Þ
2 xi

Xn  
df 2  2 
S2y ¼ Si
i¼1
dxi

The coefficient of variation of y will be:

Sy
Vy ¼

y

In the case of qP (u, c and c), expressions (12) and (13) become:
 
B 1 d2Nc d2Nq B d2Nc
lq ¼ q
p ¼ cNc kc þ ctNq kq þ c Nc kc ckc 2 þ ctkq 2 þ c kc S2u ð13Þ
2 2 du du 2 du2
   2
dNc dNq B dNc 2 2 B
S2q ¼ ckc þ ctkq þ ckc Su þ Nc2 k2c S2c þ tNq kq þ Nc kc S2c ð14Þ
du du 2 du 2
172 T. Messafer

7.7 Calculation Results

7.7.1 Traditional Method
The geotechnical investigation conducted on the site helped fix a founding level of 2 m
depth below ground level. Widths calculations at this depth for a strip footing and for
c1 ¼ c2 ¼ 20 kN=m3 and solicitation S ¼ 300 kN=ml are shown in Table 4.

Table 4. Widths B with Terzaghi method for Fs = 2 and 3

u’[°]/c’[kPa] 13/40 15/61 15/49 14/51 15/44 15/53 16/53 14/58 15/51
Fs = 2 1,09 0,7 0,83 0,86 0,90 0,79 0,73 0,78 0,81
Fs = 3 1,60 1,04 1,23 1,27 1,32 1,17 1,09 1,16 1,20

7.7.2 DTU, DIN and Eurocode Methods

The results are given on Table 5.

Table 5. Widths B for DTU, DIN and Eurocode 7.1 methods

u’[°]/c’[kPa] DIN 1054 DTU 13.12 Eurocode 7
Fs = 2 Fs = 3 Fs = 2 Fs = 3
13/40 1,09 1,60 1,01 1,40 1,58
15/49 0,83 1,23 0,79 1,11 1,21
15/44 0,90 1,32 0,85 1,18 1,30
16/53 0,73 1,09 0,70 0,99 1,07
15/51 0,81 1,20 0,77 1,09 1,18

The results show that the widths obtained from DIN approach are slightly higher than
those of the DTU. However the results of Eurocode are much closer to the results of
DIN for FS ¼ 3.

7.7.3 Probabilistic Method

The foundation is a strip footing width of B and parameters shown on Table 6.

Table 6. Mechanical properties of the soil

µ s V s2
u’ (°) 13 15 15 14 15 15 16 14 15 15 0,87 0,06 0,75
c1 kN.m-3 20,9 20,9 20,6 20,8 20,6 20,1 20,6 20,5 20,0 20,6 0,32 0,02 0,10
c’ kN.m-2 40 61 49 51 44 53 53 58 51 51,1 6,43 0,13 41,36

It is considered that the solicitation (S) and the capacity (Q) are random and follow
a normal probability law and thus the safety margin (Z = Q–S) also follows a normal
distribution.
The results of long-term load bearing capacities are summarized on Table 7.
Values lq and sq used for the calculation of risk are determined from Table 8.
 Effect of variability of soil parameters 173

Table 7. Probability of a value being smaller than z for a standard normal distribution
Bearing capacity
B = 0,5 m B = 0,75 m B=1m B = 1, 5 m
ql Ql Z F(z) ql Ql Z F(z) ql Ql Z F(z) ql Ql Z F(z)
200 100 −2,08 0,05 200 150 −2,07 0,05 200 200 −2,06 0,05 200 300 −2,05 0,05
300 150 −1,57 0,12 300 225 −1,57 0,12 300 300 −1,57 0,12 300 450 −1,56 0,12
400 200 −1,06 0,23 400 300 −1,06 0,23 400 400 −1,07 0,23 400 600 −1,08 0,22
500 250 −0,55 0,34 500 375 −0,56 0,34 500 500 −0,57 0,34 500 750 −0,60 0,33
538,96 269,48 −0,35 0,38 544,11 408,08 −0,34 0,38 549,25 549,25 −0,33 0,38 559,55 839,32 −0,31 0,38
600 300 −0,04 0,40 600 450 −0,06 0,40 600 600 −0,08 0,40 600 900 −0,12 0,40
659,10 329,55 0,26 0,39 665,92 499,44 0,27 0,38 672,74 672,74 0,28 0,38 686,39 1029,59 0,30 0,38
690,12 345,06 0,42 0,37 696,08 522,06 0,42 0,36 702,03 702,03 0,43 0,36 713,94 1070,91 0,43 0,36
714,00 357 0,54 0,34 720,82 540,61 0,55 0,34 727,64 727,64 0,55 0,34 741,29 1111,94 0,56 0,34
730,83 365,42 0,63 0,33 737,46 553,09 0,63 0,33 744,08 744,08 0,63 0,33 757,33 1136,00 0,64 0,33
753,90 376,95 0,75 0,30 760,57 570,43 0,75 0,30 767,24 767,24 0,75 0,30 780,57 1170,86 0,75 0,30
760,39 380,19 0,78 0,29 766,25 574,69 0,78 0,30 772,12 772,12 0,77 0,30 783,86 1175,79 0,77 0,30
810,96 405,48 1,04 0,23 818,84 614,13 1,04 0,23 826,72 826,72 1,04 0,23 842,48 1263,71 1,05 0,23
848,32 424,16 1,23 0,19 855,24 641,43 1,22 0,19 862,16 862,16 1,22 0,19 876,01 1314,02 1,21 0,19
µ 607,61 303,81 0,00 611,81 458,85 0,00 616,00 616,00 0,00 624,39 936,58 0,00
s 196,16 98,08 1,00 198,98 149,23 1,00 201,81 201,81 1,00 207,52 311,29 1,00

Table 8. Calculation of lq et de sq for différent B

Sol Clay
B = 0,5 m B = 0,75 m B = 1 m B = 1,5 m
c Nc 658,6 658,6 658,6 658,6
c t Nq 91,4 91,4 91,4 91,4
c B=2 Nc 15,0 22,5 30,0 45,0
ql 765,1 772,6 780,1 795,1
1=2 c d 2 Nc=du2 s2u 3 867,8 3 867,8 3 867,8 3 867,8
1=2 c t d 2 Nq=du2 s2u 1 216,5 1 216,5 1 216,5 1 216,5
1=2c B=2d 2 Nc=du2 s2u 561,9 842,9 1 123,9 1 685,8
Somme 5 646,3 5 927,2 6 208,2 6 770,1
moyenne lq (kPa) 6 411,3 6 699,8 6 988,3 7 565,2
c dN c=du 2 294,4 2 294,4 2 294,4 2 294,4
c t dNq=du 530,9 530,9 530,9 530,9
c B=2 dNc=du 131,2 196,8 262,3 393,5
Somme 2 956,5 3 022,1 3 087,7 3 218,8
(..somme..)2 s2u 6555589,07 6849668,34 7150199,79 7770619,2
Nc2 s2c 6867,97 6867,97 6867,97 6867,97
ðt Nq þ B=2 NcÞ2 s2c 2,67 3,056 3,47 4,38
s2q 6562459,70 6856539,37 7157071,23 7777491,55
Ecart type sq 2561,73 2618,5 2675,27 2788,81
variance Vq 0,4 0,39 0,38 0,37
174 T. Messafer

First Case: Assume that the load is constant of intensity S ¼ 300 kN=m, the proba-
bility of failure for widths B of 0.5 m; 0.75 m; 1.0 m and 1.5 m are summarized on
Table 9. Figure 11 shows the method of calculating the probability of failure for a
width equal to B of 1 m.

Table 9. Risk of failure for constant loads

Widths B (m) 0,5 0,75 1,0 1,5
Risk of failure (%) 59,3 29,8 18,2 10,1

Fig. 11. Risk of failure for B = 1 m

Second case: We assume that the stress (S) and the capacity (Q) are random with a
normal distribution, the load to be transmitted to the ground is characterized by the
following values.

Table 10. Risk of failure for variable loads and capacities

Widths B (m) 0,5 0,75 1,0 1,5
Probability de failure (%) 1,4 0,9 0,7 0,5

Results for probability of risk calculation are shown in Table 10 for different widths
B. Figure 12 shows the distributions of Q and S in the case where the width B is 1 m.
0 1
B lS  lQ C
Pf ¼ 1  F @qffiffiffiffiffiffiffiffiffiffiffiffiffiffiA ¼ 1  F ð þ 2; 46Þ ¼ 0; 69% ð15Þ
s2S þ s2Q
 Effect of variability of soil parameters 175

Fig. 12. Diagram of normal distribution of capacity and solicitation for B = 1 m

The results show that the risk of failure in the case of a variable load is negligible
for widths ranging between 0.5 to 1.5 m. However in the case of a constant load,
probability of failure varies from 60 (%) in the case of a 0.5 m width of foundation to
10 (%) in the case of a width of 1.5 m. In general for widths larger than 1 m the
probability of failure is less than 20 (%). With traditional methods (Terzaghi, DIN and
DTU) to obtain B values between 1 and 1.5 m, an Fs equal to 3 is required.

8 General Conclusions

The geotechnical investigation showed that the soil is composed of two essential layers
namely a clay layer with thickness of 4 m followed by a layer of marl. In order to
compare the results we decompose the soil in 2 models. The first model is the whole
site as one layer (site) and the 2nd is a two-layer soil (clay and marl).
The results of the geotechnical parameters were presented in the form of his-
tograms, cumulative distributions and normal laws. Regression equations were estab-
lished between pairs of parameters and gave results comparable to those in the
literature.
We presented an analysis of the spatial variability of the pressuremeter modulus
EM. We found that if we assume that EM is an exponential function, the autocorre-
lation distance is approximately 0.5 m indicating that EM values are not dispersed.
Foundation calculations were carried out by traditional probabilistic methods. We
showed that in the probabilistic approach bearing capacity is a random parameter
because it is based on parameters c and u which are themselves random. Compared to
traditional methods the probabilistic approach is a powerful calculation tool but
requires to be utilized among practitioners.
176 T. Messafer

References
1. Magnan, J.P, Baki. A., Pouget, P.: Analyse statistique de la variabilité des propriétés
physiques et mécaniques d’un sol, Bulletin de liaison Laboratoire Centrale des Ponts et
Chausse, No. 186 (1993)
2. Salloum, N.: Evaluation de la variabilité spatiale des paramètres géotechniques du sol à partir
de mesures géophysiques: application à la plaine alluviale de Nahr-Beyrouth (Liban), Thèse
de doctorat, Université de Grenoble, Liban (2015)
3. Imanzadeh, S.: Effects of uncertainties and spatial variation of soil and structure properties on
geotechnical design. Cases of continuous spread footings and buried pipes, these de Doctorat,
Université de Bordeaux, France (2013)
4. Vanmarcke, E.H.: Probabilistic modeling of soil profiles, America, society of civil engineers.
J. Geotech. Eng. Div. 103(11), 1227–1246 (1977)
5. Mounji, M., Lahmili, A., Ouadif, L., Baba, K., Bahi, L.: Probabilistic approach for the
selection of the shallow foundation’s safety factor. Int. J. Eng. Technol. (IJET) 8(2), 7 (2016)
6. Chew, Y.M., NG, K.S., NG, S.F.: The effect of soil variability on the ultimate bearing
capacity of shallow foundation. J. Eng. Sci. Technol., 1–13 (2015). Special issue on ACEE
2015 Conference
7. Boutahir, M., Belabed, L., Benyaghia, H.: Analyse fiabiliste de la capacité portante des
fondations superficielles, Séminaire International, innovation et valorisation en génie civil et
materiaux de construction, No 5P-092, Rabat, Maroc (2011)
8. Hahn, G.J., Shapiro, S.S.: Statistical models in engineering. Wiley Interscience Publisher
(1994)
 Improvement of Collapsible Soil Conditions
for Industrial Floors

Nelson L. Fonte Jr.1, David de Carvalho2(&), and Roberto Kassouf3

1
Head of Geoponto Engineering, Mogi das Cruzes, São Paulo, Brazil
[email protected]
2
University of Campinas, Campinas, São Paulo, Brazil
[email protected]
3
Metropolitan College of Campinas, DeVry Metrocamp, Campinas, Brazil
[email protected]

Abstract. Unsaturated surface soils with porosities above 50% cover great
extensions of areas in Midwest Brazil. Because of their large volumes of voids,
these soils undergo great strain under loads. In addition, many of these soils are
collapsible, i.e., when the soils are under load and in case of a significant
increase in the moisture content or saturation of the soil, the structure collapses,
thus producing unacceptable displacement values for the buildings. Because of
these characteristics, problems often occur in industrial floors, pavements and
other types of slabs on ground and shallow foundations.
To solve this problem, compaction of the topsoil is performed before starting
the construction. For this study, the soil characteristics are analyzed using both
geotechnical laboratory tests and field tests in order to predict the soil behavior
in terms of deformability, resistance and collapsibility.
The study was conducted at the experimental site located at the State
University of Campinas - Unicamp, in the municipality of Campinas, State of
São Paulo, Brazil. The geotechnical properties of the subsoil were determined
by collecting undisturbed samples down to 8 m in depth and deformed samples
up to 9 m of depth (impenetrable). Simple recognition surveys were performed:
standard penetration test (SPT) and electric static cone penetration test (CPT).
The edometric tests conducted on undisturbed samples with flooding at dif-
ferent levels of stress revealed collapsible characteristics of the soil. With the
conduction of tri-axial tests (CU), numerical values were obtained for the angle
of friction and the cohesion intercepts for each depth. Paschoalin Filho (2008)
verified a significant reduction in these values with soil saturation.
The influence of soil deformability and of the type of load on industrial floors
is analyzed. The results indicate huge influence of soil deformability in the case
of distributed loads, and a minor influence in case of concentrated loads.
The thickness of the compacted soil layer is very important to the definition of
the modulus of subgrade reaction (k) for the studied soil.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_15
178 N.L. Fonte Jr. et al.

1 Introduction

Collapsible soils are non-saturated soils with a potentially unstable porous structure.
Once under the action of a load, their index of voids is suddenly reduced in case of a
certain increase in moisture content. For such, there must be cementation or suction to
keep the soil structure stable in its natural condition. However, In case the soil is
flooded with water, suction ceases to exist and the cement may dissolve destabilizing
the structure of the soil.
Soil collapse may damage buildings and floors due to the magnitude of differential
settlements in case of rupture of pipes, flooding, water table increase, among others
(Gon 2011).
Excavating the soil down to a certain depth and filling the same material back in a
controlled manner is a way to improve the supporting soil of shallow foundations,
industrial floors and pavements on collapsible soils by increasing resistance, rigidity
and stability against collapse, and reducing both strain and permeability (Ribeiro and
Futai 2010).
This improvement can be verified in the soil analyzed in this work by the increased
rigidity and resistance obtained via edometric and triaxial tests using compacted and
undisturbed samples of soil. The effect of this improvement on the design of industrial
floors is analyzed.
The study was conducted at the campus of the State University of Campinas -
Unicamp, city of Campinas, State of São Paulo, Brazil. Several researches have already
been conducted on the soil of the campus, and laboratory tests were made in
non-deformed and deformed samples down to 9-m depth, as well as field tests: SPT
with torque check (SPT-T), electric (CPTU) and mechanical (CPT) cone tests with
Begeman sleeve, cross-hole test, vertical seismic sounding, and Menard-type pressio-
metric and dilatometric (DMT) tests (Paschoalin 2008).

1.1 Occurrence of Collapsible Soils in Brazil

Collapsible soils cover approximately 50% of the southwest region of Brazil. Giacheti
(1991) presents a map pinpointing collapsible soils with geological/geotechnical
characteristics potentially similar to the soil studied in the city of Campinas (Fig. 1).
Besides this vast region, Ferreira et al. (2007) present records of collapsible soils in
different states of the northeast and north of Brazil: Amazonas, Pará, Tocantins, Piauí,
Paraíba, Rio Grande do Norte, Pernambuco and Bahia. Bandeira et al. (2012) also
mention occurrence of such collapsible soils in the state of Ceará.
Taking into account the large areas of collapsible soils in Brazil, it is necessary to
learn more about the behavior of such soils and take their characteristics into due
account when designing shallow foundations, industrial floors and rigid pavements.
 Improvement of Collapsible Soil Conditions for Industrial Floors 179

Fig. 1. Regions in Brazil where the same profile as that of Campinas may exist (Giacheti 1991
adapted by Cury Filho 2016)

1.2 Compaction as Soil Improvement

The most commonly method used worldwide to construct buildings with the use of
surface foundations on porous, collapsible ground is to remove the surface layer,
compact it and return it to the same place (Ribeiro and Futai 2010).
Souza and Cintra (1994) carried out four circular plate load tests and two tests in
continuous footings measuring 0.70 m
3.02 m, all of which were seated at a depth
of 0.70 m. The study was carried out in porous soils in the countryside of the state of
São Paulo, in the city of Ilha Solteira. Two of the plate tests were performed with soil in
its natural moisture, one of which on a compacted soil layer. Two more plate tests were
performed likewise but in flooded soil.
The authors concluded that compaction reduced collapse settlement by 86%. Set-
tlements in other steps of loading were reduced by 50%. The conclusion was that
compaction was a cost-effective, efficient and easy to use process to enable use of
shallow foundations in this type of soil. After an analysis of the results shown in this
research, the observation was that compacted soil increased the bearing capacity of
foundations by approximately 100% when compared to the same moisture content.
Another conclusion was that compacted soil, when flooded, resisted 36% more than
soil in its natural condition.
In the cases mentioned, the soil was compacted down to a “Z” depth equal to the
smallest dimension of “B” footing; the width of the compacted regions is the width of
“B” footing, with an addition of B/2 on each side, as shown in Fig. 2.
180 N.L. Fonte Jr. et al.

Fig. 2. Use of footings (shallow foundations) In compacted collapsible soil (Cintra et al. 2003)

Considering that stresses propagate at depth in a 1:2 distribution, compaction with

the width of B plus B/2 on each side causes the added stress to occur only on the
compacted soil at depth Z = B and, at this depth, this added stress is only 25% of the
stress applied by the footing, thus reducing the collapse of the natural soil due to
reduced stress (Cintra and Aoki 2013).
Guimarães and Ferreira (1998) conducted several laboratory tests with compacted
samples and also confirmed the efficacy of controlled compaction to improve soils in
the state of Pernambuco as to reduction of collapse.

1.3 Study Site

There are two experimental sites at the campus of the State University of Campinas -
Unicamp, in the city of Campinas, state of São Paulo, where several researches have
been conducted. One is the experimental site of the School of Agricultural Engineering
(FEAGRI). The other is the experimental site of the School of Civil Engineering,
Architecture and Urbanism (FEC). The distance between the sites is approximately
300 m and both are installed in an area with the same type of soil, so the results can be
interpreted as representative of one large region.
The FEC experimental site has a 600-m2 area. Its geodesic coordinates in the
WGS84 geodesic system are: latitude −22º49′01″ and longitude −47º03′41″; the
experimental site of FEAGRI as a 400-m2 area, with the following coordinates: latitude
−22º49′10″ and longitude −47º03′40″. Figure 3 shows the location of the experimental
sites at the Unicamp campus. Paschoalin (2008) described part of the researches
conducted at the experimental site of FEAGRI until 2008: Peixoto (2001), in which
several SPT-T type tests were performed; Fontaine (2004) executed electric cone tests
and pressiometer tests; Carvalho et al. (2000) provided geotechnical characteristics
obtained by means of several field and laboratory tests; Albuquerque (2001), Nogueira
(2004) and Paschoalin (2008) carried out several load tests on piles. Giachetti (1991)
and Albuquerque (1996) conducted several laboratory tests for geotechnical charac-
terization; Monacci (1995) studied the soil as to its collapsibility. Paschoalin (2002)
 Improvement of Collapsible Soil Conditions for Industrial Floors 181

Fig. 3. Location of the experimental sites

studied several characteristics of this soil in its “natural” and compacted condition in
different moisture contents.
Gon (2011) collected samples down to 9-m depth at the FEC experimental site for
tests of characterization, permeability, edometric, triaxial and proctor tests.
Based on these researches, it can be stated that the soil in the region under analysis
comprises primarily one first layer of approximately 6.0 m of highly porous sandy clay;
in some spots a 0.50-m layer of clayey-silty fine and medium sand can be found. There
is a second layer of residual diabase soil composed of clayey sandy silt down to the
depth of 16 m. The water table was not found down to this depth. Figure 4 shows this
profile.
The first layer is formed by mature soil which underwent intense weathering. The
second layer is formed by young residual soil, which has preserved inherited charac-
teristics from the original rock (Albuquerque 2001). Figure 5 shows the geological
profile of the Unicamp region (Barão Geraldo District).
Figures 6, 7, 8, 9 and 10 show results of field tests carried out by Carvalho et al.
(2004) and resubmitted by Cury Filho (2016). Figures 6, 7 and 8 show average,
maximum and minimum data of SPT index, Tmax e Tmin resistance of SPT-T.
Figures 9 and 10 show results of static cone penetration tests (CPT) using electric
cone, obtained by Carvalho et al. (2000). The average, minimum and maximum values
of point resistance of the cone (qc) and lateral friction (fs) are shown.
Monacci (1995) and Gon (2011) carried out edometric compression tests in
non-deformed samples, with flooding at different levels of stress to check soil col-
lapsibility in conformance with Vargas method (1978).
According to Tables 1 and 2, most tests show that the soil under study is
collapsible.
182 N.L. Fonte Jr. et al.

Fig. 4. Geotechnical profile - average values

In order to analyze the influence of soil deformability both in its “natural” condition
and after compaction when designing slabs on soil, the initial elasticity modulus was
obtained via triaxial test.
Figure 11 shows the stress - strain behavior of a non-deformed sample collected at
the depth of 1.0 m, with different stresses of confinement (Gon 2011). This Fig-
ure shows the initial modulus of deformability for stress of confinement of 100 kPa,
resulting in the value of 3.3 MPa.
Figure 12 shows the stress - strain behavior of a sample of compacted soil at its
optimal moisture of the Normal Proctor Test. Through these curves, the data in Table 3
was obtained, which shows the tensile strength and the value obtained for the initial
 Improvement of Collapsible Soil Conditions for Industrial Floors 183

Fig. 5. Geological profile of the region of Unicamp (Dirney 2016 adapted from Zuquete 1987)

Fig. 6. Summary N resistance (Cury Filho 2016)

modulus of elasticity for each confinement stress. This table demonstrates that the
average value of the initial modulus of elasticity for the compacted soil was 22 MPa.
The increase of the modulus of elasticity due to compaction is about 6,7, similar the
difference between a structural steel (E = 205 GPa) and a structural concrete
(E = 30 GPa).
Figures 13 and 14 show the curves obtained by Paschoalin (2002) in the edometric
test both for natural and compacted soil. A large reduction in voids can be seen when
the soil is compacted.
Although compaction is a common method to improve most types of soil around
the world, it is especially efficient for the studied soil.
184 N.L. Fonte Jr. et al.

Fig. 7. Summary of Tmáx resistance (Cury Filho 2016)

Fig. 8. Summary of Tres resistance (Cury Filho 2016)

Three CBR tests were performed with compacted soil in its optimal moisture. The
results are described in Table 4. The average value is 4%.

1.4 Impact on Industrial Floors Design

When designing slabs on ground, whether for industrial floors, rigid pavements or
radier-type foundations, in many methods the soil influences the dimensioning because
of its deformability via consideration of the settlement coefficient k, also named vertical
 Improvement of Collapsible Soil Conditions for Industrial Floors 185

Fig. 9. Summary of qc values (Carvalho et al. 2004 apud Cury Filho 2016)

Fig. 10. Summary of fs values (Carvalho et al. 2004 apud Cury Filho 2016)

Table 1. Colapse coefficient (Monacci 1995)

Depth 0,5 a 0,75 m 4,75 a 5,00 m 7,75 a 8,00 m
r [kPa] i [%]
5,00 4,97 – –
9,80 11,97 2,41 –
19,40 7,40 3,76 –
29,10 9,20 – –
38,70 9,98 7,72 3,10
77,30 23,19 15,51 5,26
186 N.L. Fonte Jr. et al.

Table 2. Colapse coefficient (Gon 2011)

Depth CC 100 kPa CC 200 kPa CC 400 kPa
1 7,80% 27,00% 6,00%
2 1,30% 0,65% 23,00%
3 0,45% 0,04% 19,84%
4 9,90% 7,50% 26,80%
5 0,28% 7,03% 7,24%
6 1,10% 3,40% 9,20%
7 1,40% 5,60% 10,74%
8 5,20% 6,60% 0,09%
*CC > 2 are collapsible soils.

Fig. 11. Stress-strain behavior of natural soil for the 1st meter (Adapted from Gon 2011)

Fig. 12. Stress-strain behavior of compacted soil

 Improvement of Collapsible Soil Conditions for Industrial Floors 187

Table 3. Summary of compacted soil stress-strain behavior – triaxial tests

Sample Confinant stress (kPa) Rupture stress (kPa) Elastic modulus (MPa)
CP1 25 160 13
CP2 50 200 30
CP3 100 260 20
CP4 150 260 23
Average – – 22

Fig. 13. Variation in the void index with addition of pressure for compacted soil (Paschoalin
2002)

Fig. 14. Variation in the void index with addition of pressure for undisturbed soil (Paschoalin
2002)
188 N.L. Fonte Jr. et al.

Table 4. CBR tests

Corpo de Prova CBR (%) Expansão (%)
CP1 – S 3 0,015
CP2 – S 3 0,075
CP3 – S 6 0,20
Média 4 0,097

reaction modulus. In an elastic model, k, is the ratio between the stress applied on a
point on the soil and the corresponding displacement.
One of the ways to get the modulus of vertical reaction of the soil (subgrade) is
through correlation with the CBR index (ACI 360R-06). Figure 15 shows a curve
relating the CBR index of the soil with k (Rodrigues et al. 2006). With this curve, it
was possible to get the value of k of 33 MPa/m for the soil under study, considering the
CBR index of 4%. Using this method is not possible to take into account the thickness
of the compacted soil layer.

Fig. 15. Correlation between CBR
k (Vaz Rodrigues et al. 2006)

Another form to get k is by means of a plate load test. There are standards that set a
form to adopt k. For the specific purpose of designing rigid pavements, there is
ASTM D 1196. In Brazil, there is standard 055-2004 ME of DNIT. Both standards
determine a load test on a plate with 76 cm diameter and adopt as settlement coefficient
the ratio of stress that causes settlement of 1.27 mm by the value of this settlement
(Fig. 16).
 Improvement of Collapsible Soil Conditions for Industrial Floors 189

Fig. 16. Plate load-deformation diagram (ACI 360R-06)

Through the theory of elasticity, one can simulate the load test, if the modulus of
elasticity of the soil is known, and get K for different conditions of the soil profile.
Using the modulus of elasticity of the natural and compacted soil obtained via triaxial
test (En = 3.3 MPa and Ec = 22 MPa), it was possible to get k for different thicknesses
of compacted soil from zero to the 8-m layer. To do so, a soil profile with 8-m thickness
on a non-deformed layer was adopted. Using the method of Janbu et al. (1956, apud
Cintra 1998) k was obtained for different situations, including variation of Poisson’s
coefficient of 0.20, 0.35 and 0.50. The results are shown in Table 5, which shows that k
ranged from 5.6 MPa/m to 48.3 MPa/m.

Table 5. k for different thickness of compacted soil layer

Compacted soil Ens = 3,3 MPa e Ecs = 22 MPa
thickness (m) k (MPa/m) k (MPa/m) k (MPa/m)
p/m = 0,5 p/m = 0,35 p/m = 0,20
0 7,2 6,2 5,6
0,2 9,5 8,1 7,4
0,25 10,4 8,9 8,1
0,3 10,7 9,1 8,4
0,4 12,4 10,6 9,7
0,5 13,8 11,8 10,8
0,8 17,5 15,0 13,7
1 19,4 16,6 15,2
1,5 25 21,4 19,5
2 31 26,5 24,2
8 48,3 41,3 37,7

Two situations of loading were used to assess the influence of variation of k in

designing an industrial floor. The first situation is the case of a concentrated load of
190 N.L. Fonte Jr. et al.

50 kN applied by a tire with calibration pressure of 700 kPa. The second situation is the
case of a distributed load of 40 kN/m2 with aisles without loading.
For the first case, the method of Meyerhof (1962) is used for floors with distributed
reinforcement on the top, considering the load applied in the interior region of the floor
(away from the edges of the plate). The bending moment acting in the concrete slab is
obtained by means of Eq. 1, which, in turn, uses Eq. 2, which uses the modulus of
elasticity of concrete of 26,840 MPa and Poisson’s coefficient of concrete of 0.20,
where h is the thickness of the concrete of the industrial floor. In Eq. 1, ‘a’ is the radius
of the area of contact between the tire and the industrial floor. Once the acting moment
is obtained, the dimensioning is made by limiting the tensile stress in concrete by the
admissible tensile stress in bending. For concrete of fck 30 MPa, an admissible stress of
2.07 MPa is adopted.

P
Mi ¼ ð1Þ
6½1 þ ð2a=lÞ
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4 E
h3
l¼ ð2Þ
12xð1  l2 Þ
K

For the case of distributed load, the method of Packard (1996) is used. This method
uses Eq. 3. C is the admissible distributed load, radm is the admissible tensile stress of
concrete (adopted 2.07 MPa), h is the thickness of concrete in cm and k is the coef-
ficient of settlement in MPa/m.
pffiffiffiffiffiffi
C ¼ 1; 03:radm : h:k ð3Þ

Figure 17 shows the variation in thickness of a floor submitted to concentrated load

with variation in the value of k. It can be noted that, even with such a large variation of
k ranging from 5 to 50 MPa/m, the thickness of concrete varied approximately
1.20 cm, i.e., approximately 10%.

Fig. 17. Influence of k on floor thickness for cases of concentrated load

 Improvement of Collapsible Soil Conditions for Industrial Floors 191

Figure 18 shows the same variation for distributed load, where the large influence
of the value of k can be noted. By increasing 10 times its value, the thickness of the
floor varied from 7 to 70 cm, which is directly proportional to the variation of k.

Fig. 18. Influence of k on floor thickness for cases of distributed load

Considering two different support condition for an industrial floor: one without any
compacted soil layer, the slab directly on the natural soil, and other with 1 m of
compacted soil. Using the results presented in Table 5 (m = 0,35) the k value for the
first condition is 6,2 MPa/m, and 16,6 MPa for the second condition. With these K
values, Fig. 17 shows that for concentrated load the slab thickness in a design would be
14 cm for both support conditions. Figure 18, for distributed load, shows for the first
condition the slab thickness would be about 70 cm, and about 21 cm for the second
support condition. The influence of the compacted soil in this case is huge.

2 Conclusions

Based on the results obtained, it can be stated that, for the soil under study, compaction
improved the performance of the soil as to the following aspects:
– Increased bearing capacity.
– Increased rigidity and consequent reduction in strain due to stress.
– Eliminated the collapsible characteristic of natural soil.
The soil under study is unsaturated, highly porous and collapsible. Its compaction is
easy to be executed and cost-effective.
Compaction of collapsible surface soil is a solution that can make it possible to use
shallow foundations in soils similar to the soil of the region under study, within certain
limits of stress.
The layer of compacted soil significantly influences the value of the modulus of
subgrade reaction for the soil under analysis.
192 N.L. Fonte Jr. et al.

For concentrated loads, the value of k has little influence on the industrial floors
design, thus justifying the use of simpler methods to obtain it, such as the correlation
with the CBR index. In these cases, the recommendation is a minimum layer of
20–25-cm thick compacted soil to even out the support and to get stabilization against
collapse.
For distributed loads, the value of k has large influence on the industrial floors
design, thus justifying the use of more reliable methods to obtain it, such as plate load
tests.
The compacted soil layer affects strongly the slab support deformability for the
studied soil.

Acknowledgements. We wish to thank Vander Ferreira Alves, Iago Leandro dos Santos, Paulo
Gustavo Krejci Nunes and Juliana Silva for their help in developing the tests.

References
Albuquerque, P.J.R.: Analysis of the behavior of instrumented driven piles of small diameters in
diabasic residual soil of Campinas region. M.Sc. dissertation of Agriculture Engineering
College, State University of Campinas (1996)
Albuquerque, P.J.R.: Bored piles, CFA and omega piles: study of the behavior subjected to
compression in diabasic residual soil, through instrumented load test. Ph.D. thesis of
Polytechnic School of São Paulo University (2001)
Bandeira, A.P.N., et al.: Characterization of collapsible soils in municipality of Juazeiro do
Norte, interior of Ceará. Brazilian Congress of Soil Mechanics and Geotechnical Engineering,
COBRAMSEG, ABMS, Porto de Galinhas (2012)
Carvalho, D., et al.: Experimental site for studies of soil mechanics and foundation in Campinas –
SP. In: Seminar of Specials Foundation Engineering, vol. 4, São Paulo (2000)
Carvalho, J.M., Viana da Fonseca, A., Almeida, F., Hermosilha,H.: ISC’2 experimental site
investigation and char acterization: conventional and to mographic P and S waves refraction
seismics vs. electrical resistivity. In: Viana da Fonseca, A., Mayne, P.W. (eds.) Geotechnical
and Geophysical Site Characterization. Millpress, Rotterdam (2004)
Cintra, J.C.A.: Foundation on Collapsible Soils, 1st edn. Graphical Service of Engineering
School of São Carlos, University of São Paulo, São Carlos (1998)
Cintra, J.C.A., Aoki, N.: Foundation Design on Collapsible Soils, 1st edn. Graphical Service of
Engineering School of São Carlos, University of São Paulo, São Carlos (2013)
Cury Filho, D.: Load capacity estimation for steel piles subjected to compression in diabasic soil
of Campinas region. M.Sc. dissertation of Civil Engineering College, State University of
Campinas (2016)
Ferreira, R.S.M.: Research report soil volume variation behavior due to flooding. Civil
Engineering Department, UFP (2007)
Fontaine, E.B.: Utilization of specials field tests (electric cone and pressiometer test) for
estimation of geotechnical properties and pile load capacity in some tropical soils. Ph.D.
thesis of Agriculture Engineering College, State University of Campinas (2004)
Giachetti, H.L.: Experimental study of dynamic parameters of some São Paulo state tropical soils.
Ph.D. thesis of Engineering School of São Carlos, University of São Paulo, São Carlos (1991)
Gon, F.S.: Geotechnical characterization through laboratory tests of a diabasic soil of Campinas
region. M.Sc. dissertation of State University of Campinas (2011)
 Improvement of Collapsible Soil Conditions for Industrial Floors 193

Guimarães, N., Ferreira, S.R.M.: Collapse due to flooding in compacted soils. Brazilian Congress
of Soil Mechanics and Geotechnical Engineering, COBRAMSEG, ABMS, Brasília (1998)
Meyerhof, G.G.: Load-carrying capacity of concrete pavements. In: ASCE Proceedings (1950)
Monacci, M.D.: Study of soil collapsibility in experimental site of the Agriculture Engineering
College – Unicamp. M.Sc. dissertation of the Agriculture Engineering College, State
University of Campinas (1995)
Nogueira, R.C.R.: Behavior of root piles, instrumented, subjected ti axial compression, in
diabasic soil. M.Sc. dissertation of Civil Engineering College, State University of Campinas
(2004)
Packard, R.G.: Slab Thickness Design for Industrial Concrete Floors on Grade. Portland Cement
Association, Concrete Information, Skokie (1996)
Paschoalin, J.A.: Study of different piles subjected to axial tensile in diabasic soil of Campinas
region. Ph.D. thesis of State University of Campinas – Unicamp (2008)
Paschoalin, J.A.: Utilization of diabasic residual soil as foundation element and construction
material of compacted embankments for small dams. M.Sc. dissertation of State University of
Campinas – Unicamp (2002)
Peixoto, A.S.P.: Study of SPT-T test and its application in foundation engineering practice. Ph.D.
thesis of Agriculture Engineering College, State University of Campinas (2001)
Junior, I.R., Futai, M.M.: Study case of collapsible soil improvement with compaction. In:
Brazilian Congress of Soil Mechanics and Geotechnical Engineering, COBRAMSEG,
ABMS, Gramado (2010)
Rodrigues, P.P.F., et al.: Gerdau Manual of Industrial Floors, 1st edn. Ed. Pini, São Paulo (2006)
Souza, A., Cintra, J.C.A.: Shallow foundation on collapsible soil of Ilha Solteira – SP. In: 10th
Brazilian Congress of Soil Mechanics and Foundation Engineering, Foz do Iguaçu, vol.
1 (1994)
Vaz Rodrigues R., Muttoni A., Burdet O.: Large scale tests on bridge slabs cantilevers subjected
to traffic loads. In: Proceedings of the 2nd fib Congress, vol. 1, p. 232, Naples (2006)
 Ground Improvement of Tank Foundations
in the Middle East

Babak Hamidi1(&) and Serge Varaksin2

1
ISSMGE TC-211 Member, Perth, Australia
[email protected]
2
Apageo, ISSMGE TC-211 Co-chairman, Paris, France

Abstract. Although tank construction is well advanced in the Middle East, not
all tanks are located on suitable grounds, and many require the implementation
of specific foundation measures to ensure that design specifications are satisfied.
While ground improvement of tank foundations in the region has a proven track
record that dates back to the 1980s, and can potentially be the most appropriate
solution for many tanks, it is the authors’ experience that occasionally the
projects’ settlement requirements appear to have been influenced by building
specifications that have not been developed for the purpose. Consequently and
unnecessarily, more expensive solutions with longer construction schedules may
have to be adopted that will not bring any additional benefits to the foundation
solution. This paper initially reviews settlement criteria of a number of inter-
nationally recognized standards and then presents a number of case studies of
tank foundation improvement by dynamic compaction and quality control by
Menard pressuremeter tests.

1 Introduction

Industrial scale tanks are usually large and heavily loaded structures that are used to
store fluids and must remain in uninterrupted service due to their commercial value and
social importance. For example, any disruption in the supply of potable water or fuel
due to a tank becoming temporarily non-operational will immediately attract public
attention and can lead to social anxiety or unrest if the matter is not addressed and
resolved rapidly.
With consideration that tanks are relatively thin structures, unforeseen and unac-
counted ground deformations can impede the performance of the tanks, and ultimately
lead to tank failure. Hence, implementation of specific foundation techniques has
become common practice when in-situ ground conditions are not able to provide
foundations with the required bearing or deformation limitations.
Ground improvement in general and dynamic compaction, in particular, have been
used extensively for tank foundation improvement since the early 1970s. Dynamic
compaction was first used for the improvement of a Petro France tank in Villeneuve La
Garenne in 1972. The same technique was also used for the first time respectively for oil
tanks in Africa (Congo) in 1973, in Asia (Japan) in 1975, in North America (Los Angles)
in 1977, and in South America (La Plata, Argentina) in 1980. However, it was only not
until 1988 that dynamic compaction and replacement were first used in the Middle East

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_16
 Ground Improvement of Tank Foundations in the Middle East 195

for the treatment of Iraqi Sate Company for Oil Project’s tank farm in Yanbu, Saudi
Arabia. Among other structures and infrastructures, this project included 10 large crude
oil storage tanks, each with diameters and heights that were respectively 110.8 m and
20 m. Two smaller fuel storage tanks were also treated in the same manner.
Today, hundreds of tank foundations have been treated using ground improvement
throughout the world, many of which have been documented through conference
proceedings or journal publications. Some interesting publications include Cognon
et al. (1983); Hendy and Muir (1997); Ihm and Masse (2001); Hamidi and Jullienne
(2007); Buschmeier et al. (2012); Debats et al. (2013) and Hamidi et al. (2016).

2 Tank Settlements

Unlike buildings that are basically hollow structures, tanks are heavily loaded
thin-shelled structures that exert large loads to large foundations. Hence, tank owners
are generally and rightfully very sensitive to foundation movements as any disturbance
in the performance of tanks can have significant commercial or social impacts. How-
ever, it is the authors’ experience that when decision-making engineers have a civil or
building background or are not sufficiently familiar with existing research and current
tank codes of practices, they tend to stipulate settlement criteria requirements that are
unrealistically and unnecessarily onerous and more appropriate and applicable to
buildings rather than tanks. The authors have even been involved in discussions where
the settlement requirement was zero, which is physically meaningless.
In reality, the impact of ground movement on tanks has been well researched since
years ago (Marr et al. 1982; D’Orazio and Duncan 1987), and contrary to most
structures whose global performances due to full loadings are only assessable after
completion of construction and during utilization phase, tanks are typically subjected to
a full-scale hydro test loading before becoming operational.
Internationally recognized standards such as American Petroleum Institute (API),
American Concrete Institute (ACI), and Engineering Equipment and Materials Users
Association (EEMUA) provide guidelines and minimum requirements for tank foun-
dation deformations. Based on internal knowledge and know-how, other organizations
such as Mobil (1990) have also developed their own specifications.

2.1 Types of Tank Settlement

Tank settlements can be envisaged to be uniform, rigid body tilting or planar tilting,
and out-of-plane settlement or differential settlement. These types of movements can
result in various forms of failure, of which the ones that are most concerning are the
distortion of the shell, rupture of the shell or bottom plate or shell-bottom plate con-
nection (Marr et al. 1982).
Uniform Settlement
The uniform settlement of a tank is shown in Fig. 1. Most standards do not appear to be
overly sensitive to this type of settlement, but occasionally draw attention to its effect
196 B. Hamidi and S. Varaksin

on the piping, which can be resolved by using flexible connections or periodically

repositioning the pipe supports.

Fig. 1. Uniform settlement

API 620 (American Petroleum Institute 2002) that is applicable to the design and
construction of large, welded, low-pressure storage tanks does not stipulate limits on
settlement, but notes that uniformity of support and avoidance of excessive settlement
are much more important for tanks that have formed bottom plates than they are in the
case of flat-bottom, vertical storage tanks.
API 625 (American Petroleum Institute 2010) that is used for refrigerated liquefied
gas storage tanks systems simply notes that the amount of acceptable uniform settle-
ment depends on piping and structural connections between the tank system and
adjacent structures.
API 650 (American Petroleum Institute 2007) that is utilized for welded tanks also
does not limit this type of settlement but states that total settlement must not strain
connecting piping or produce gauging inaccuracies, and that settlement should not
continue to a point at which the tank bottom is below the surrounding ground surface.
This standard further specifies that estimated settlement should be within the acceptable
tolerances for the tank shell and bottom, and adds that if a large settlement is expected,
then the tank bottom elevation should be raised so that the final elevation above grade
will be a minimum of 150 mm after settlement.
API 653 (American Petroleum Institute 2003) that is applicable to tank inspection,
repair, alteration, and reconstruction notes that excessive foundation settlement of
storage tanks can affect the integrity of tank shells and bottoms, and adds that this type
of settlement could affect tank nozzles that have piping attached to them. While this
standard does not limit total settlements, it states that for existing tanks with histories of
successful services, it may be possible to accept greater settlements and distortions of
the foundation from a true plane than new tank construction standards allow.
ACI 376 (American Concrete Institute 2010) that has been developed for the design
and construction of concrete structures for the containment of refrigerated liquefied
gasses permits uniform settlement of concrete shallow foundations provided that the
other provisions of the standard are met, and the connecting piping system accom-
modates the settlement.
 Ground Improvement of Tank Foundations in the Middle East 197

Rigid Body Tilting

Figure 2 shows rigid body tilting or planar tilting. This form of settlement of the shell
reduces freeboard, alters the shape of the fluid surface, and places additional stresses in
the shell (Marr et al. 1982). Marr et al. show that a planar tilt that is 5% of the tank
diameter will increase the hoop stress at the bottom of the shell by 10%. While they
assess that failure of the shell from planar tilt alone seems unlikely, and note that this
type of tilting appears to have no detrimental consequence to the structural integrity of
the bottom plate, Marr et al. propose limits to avoid spilling oil from floating roofs or
exceeding the tolerance of the floating roof by stressing and deforming the shell.

Fig. 2. Rigid body or planar tilting

While API 620 does not stipulate limits to planar tilting of steel tanks, API 625
allows variations from the settlement limits that it specifies provided that they are
accounted for in the design of the tank system and interconnecting components, which
in the authors’ opinion, is a very rational approach that should be considered as good
practice. API comments that while large tanks may be able to accommodate significant
tilting without damage, other components usually require lower values of tilt, and thus
limits its predicted value to Eq. 1:

D
GT\25:4a ð1Þ
H

GT = global (rigid body) tilt settlement, in mm

a = 5, but often restricted to lesser values as a result of the same issues that are of
concern for uniform settlement.
D = tank diameter
H = tank height.
ACI 376 limits differential settlement or planar tilting to 1/500.
Out of Plane Settlement
Dishing of the tank bottom, which is shown in Fig. 3 is the most commonly envisaged
form of out-of-plane settlement. However, non-planar settlements do not necessarily
198 B. Hamidi and S. Varaksin

Fig. 3. Out of plane settlement in the form of tank bottom dishing

Fig. 4. Out of plane settlement in the form of warping

have to be from the center of the tank to the shell and can be localized at the connection
of the shell to the bottom plate in the form of a localized bulge or warping, which is
shown in Fig. 4.
Out-of-plane, non-planar or differential settlement may radially distort or overstress
the shell, which can result in the malfunction of a floating roof. Overstressing can also
rupture the shell. Dish-shaped settlements and localized depressions in the bottom
plates of tanks will overstress the plate and welds. Alternatively, the shell-bottom plate
connection may rupture when the shell bridges over a soft spot, but the bottom plate
settles with it (Marr et al. 1982).
API 620 does specify any limits for non-planar settlement of steel tanks. However,
API 625 limits predicted differential settlements between the edge of the tank and the
center (DS), and around the periphery of the tank (TS) respectively to Eqs. 2 and 3.

DS\R=240 ð2Þ

R = tank radius

TS\1=1000 ð3Þ
 Ground Improvement of Tank Foundations in the Middle East 199

API 653 seems to be more tolerant on differential settlement of existing tanks. Based
on the publication of Marr et al. (1982) this standard uses Eq. 4 to calculate the
maximum permissible out of plane deflection of the shell:

11L2 Y
j Sj  ð4Þ
2EH

S = deflection
L = arc length between two points
Y = yield strength
E = Young’s modulus.
The units of L, S, and Y are the same as are the units of E and Y.
API 653 limits the permissible bulge or depression of the tank bottom to:

BB  30:8R ð5Þ

BB = maximum height of bulge or depth of local depression, in mm

R = radius of inscribed circle in bulged area or local depression, in m.
ACI 376 limits dishing along a radial line from the outer perimeter to the tank center to
10 mm drop of 1/300, and notes that restricting the dishing settlement to 1/300
maintains the bending curvatures within acceptable limits so that insulation materials
are not damaged. Specialists have commented on the 10 mm criterion as being unclear
and too conservative. ACI Committee’s response is that it is agreed that 10 mm cri-
terion should be deleted (ACI Committee 376 2010).
ACI also limits footing settlement around the perimeter of the tank to the lesser of
1/500 and the maximum settlement limit calculated for the uniform (planar) tilting of
the tank.

3 Case Histories of Dynamic Compaction Application

to Tanks in the Middle East
3.1 LNG Tanks in Ras Laffan, Qatar

Introduction and Ground Conditions

LNG Tanks T-4, T5, and T-6 are located within the reclaimed port area of Ras Laffan in
Qatar, and the foundations of all three have been treated using dynamic compaction.
Although there were some differences in the specifications and construction procedures
of these tanks, as the ground improvement methodology of all three were very similar,
only the process implemented during the ground improvement works of Tank T-6
(Hamidi and Jullienne 2007) will be reviewed.
200 B. Hamidi and S. Varaksin

Tank T-6’s capacity is 140,000 m3, and its inner steel tank diameter and height are
respectively 74.3 m and 35 m. The outer concrete shell has an internal diameter of
76.3 m. The LNG storage level is 34 m, and the maximum hydrostatic test level was
specified to be 19.98 m.
The tank’s site was fairly flat. Based on boreholes that were as deep as 80 m,
cross-hole tests to depths of 45 m, cone penetration tests (CPT) within the soil layers,
Menard Pressuremeter Tests (PMT), and laboratory tests on recovered samples, the
ground at the tank’s location was understood to consist of a superficial well graded fine
to coarse calcareous sand layer that was mixed with angular gravel. This layer’s fines
content was less than 30%. At localized areas, the sand was slightly cemented at the
base of the layer. Weathered limestone, approximately 3.5 m thick, was encountered at
about 4 m below ground level, followed by 3 m of upper limestone, 12 m of lower
limestone, 30 m of calcareous siltstone, and at least 90 m of calcareous sandstone
down to end of boring. Cavities were detected in the siltstone and sandstone layers, and
groundwater was encountered approximately 2 m below ground level.
Acceptance criteria for the project were defined to be:
• Maximum settlement at the tank’s perimeter: 80 mm
• Maximum deviation from any planar tilt of the foundation between any two points:
1/700 (25 mm differential settlement over any 17.5 m horizontal distance).
Using the specification’s differential settlement criterion over a distance of 40 m
between the center and edge of the tank measured from the concreting of the joint at the
wall to the mat, it can be calculated that the center of the tank was able to settle an
additional 57 mm or a total of 137 mm.
Hamidi and Jullienne (2007) have tabulated the design values for the thicknesses,
densities, Young moduli, Poisson ratios, cohesion, and internal friction angle of each
layer. Initial calculations suggested that the ground would not be able to satisfy the
specifications, and ground improvement was considered as a viable solution of interest.
Foundation Solution: Dynamic Compaction with a Shear Ring Trench
The ground improvement works for satisfying the design basis were awarded to a
specialist contractor who had proposed the application of dynamic compaction. As
shown in Fig. 5, in this scheme a shear ring trench was constructed by excavating a

Fig. 5. Schematic cross section of tank and ground treatment scheme

 Ground Improvement of Tank Foundations in the Middle East 201

3.5 m deep trench around the tank and simultaneously backfilling it with crushed stone.
The trench had a trapezoidal cross-section, with the top and bottom bases being
respectively 12 m and 5 m.
The crushed stone used was up to 200 mm in diameter, the Los Angles abrasion of
aggregate coefficient was less than 35, and fines content was less than 5%.
Similar to dynamic replacement that is typically performed by backfilling the
impact craters with crushed stone and applying additional blows to enlarge, deepen and
compact the granular inclusions, in this project the prints were initially excavated and
backfilled with crushed stone to provide larger granular columns with better
mechanical properties than compacted sand. The trench was then dynamically com-
pacted to increase local bearing capacity, resistance against local shear failure under the
tank’s shell and to reduce edge settlements.
Testing and Verification
Analysis of settlements was undertaken using finite element analysis software with
consideration of the tank’s construction methodology; i.e. the construction of the
perimeter ring beam and tank wall to approximately 38 m height, construction of the
central portion of the mat, construction of the joint between the ring beam and the mat,
construction of the steel cupola, construction of the inner tank, construction of the outer
tank wall haunch and roof, hydro testing, and operation of the tank.
It was calculated that the maximum settlement of the tank would occur in its center
during the hydro test (57.5 mm), but the maximum differential settlement (0.83/700)
would take place after construction of the ring beam and tank wall and prior to the
placement of concrete at the joint.
Field tests were performed during calibration of dynamic compaction parameters
and after treatment. PMTs were carried out at 25 locations, of which 3 were during the
calibration phase and the remaining were after the treatment. The tests were performed
at every one meter before reaching bedrock. 14 plate load tests were also carried out on
circular plates with diameters of 1 m, of which 2 were performed during the calibration
phase.
Testing and interpretation of the results indicated that the average moduli of
deformation in the dynamic replacement columns within the tank, compacted sand, and
ring trench were respectively 75.2 MPa, 36.9 MPa, and 70.1 MPa.
While maximum vertical settlement at the end of construction was calculated to be
31.7 mm under the tank wall, surveying of 18 plates installed on the wall indicated that
measured total settlement during the hydro test were from 4 to 10 mm, and differential
settlement between any two adjacent points was in the range of null to 3 mm.

3.2 STP Tanks in Palm Jumeira, Dubai

Introduction and Ground Conditions

A sewerage treatment plant (STP) tank has been built on each of the two outer crescent
islands of Palm Jumeira in Dubai, UAE. Due to the similarity of the ground
improvement technique applied at these two sites, only the first plant (Lot A-A) will be
reviewed in this paper; however, Hamidi et al. (2011) and Hamidi and Varaksin (2015)
202 B. Hamidi and S. Varaksin

have described the ground improvement works of both tanks (Lots A-A and G-G) in
detail.
Palm Jumeira is a reclaimed group of man-made islands that consists of a trunk, a
crown with 17 fronds, three surrounding crescent islands that form a breakwater and
two smaller islands that are in the shape of the logo of project’s developer. The main
body of the development consists of calcareous sand that was dredged from the Persian
Gulf. When possible, sand was discharged from the bottom of the hulls of trailing
suction hopper dredgers, but when the water was shallow, the dredgers sprayed the
sand and water mixture onto the reclamation by rainbow discharge.
Due to the low strength and high compressibility of the soil, ground improvement
by vibro compaction was carried out almost throughout the reclamation. Additionally,
piles were also installed to support the heavily loaded structures. The two STP tanks are
the only heavily loaded structures of the development that are only found on ground
that has been improved by dynamic compaction and dynamic surcharging.
The STP lots are located on the tips of the outer crescents. Each plant includes one
reinforced concrete tank with a diameter of 35.1 m, which was envisaged to be sub-
jected to a total uniform load of 120 kPa.
As part of the general geotechnical investigation of the reclamation, two SPT
(standard penetration test) boreholes and two CPTs (cone penetration test) were per-
formed near the location of Tank A-A. The boreholes indicated that the upper crust of the
soil was generally very dense with SPT blow counts, N, reaching 28. The deeper layers of
soil were less dense, with minimum N in the upper 8 m of soil being as low as 5. The soil
then appeared to become denser with a minimum N of 18 and exceeding 50 at the depth
of 13 m. The fines content of the soil in these two boreholes were from 2 to 10% in the
upper 13 m of soil but increased to 22% at the depth of 14.5 m. Ground level was at
+4 m RL (reduced level) and groundwater was at the depth of about 3 m. CPT cone
resistance, qc, in the upper 2 m of sand was as high as 25 MPa. The soil then became
loose with qc dropping to as low as 3 to 4 MPa to the depth of about 13 m where refusal
was achieved. Although these tests suggested that the soil was clean sand, fines content
as high as 30% was observed in a number of boreholes that were not very far from the
project.
At later stages, four additional SPT boreholes were drilled in Lot A-A’s tank
location. These boreholes indicated that the upper 3 m of sand was indeed very dense,
but the soil then became very loose to medium dense at water level. N values at depths
of 3 to 8 m ranged from 4 to 14, and then shifted to values in between 11 to 20 to
depths of approximately 12 to 13 m. N then exceeded 50. The average fines content of
the 38 samples that were extracted from the four boreholes generally ranged from 16 to
21% and up to 30%, which was significantly more than the average 5% that was
indicated by the preliminary geotechnical investigation.
Two PMTs were also carried out in Lot A-A. Testing commenced below sea level.
Limit pressure, PLM, was from less than 100 to about 700 kPa. Pressuremeter Moduli,
EM, varied from less than 1 MPa to approximately 6 MPa.
Preliminary studies suggested that mat foundations could not be used for the tanks
without the implementation of specific foundation measures.
 Ground Improvement of Tank Foundations in the Middle East 203

Foundation Solution: Dynamic Compaction and Dynamic Surcharging

Piling and ground improvement were both deemed as possible solutions; however, the
former solution was discarded due to the high costs associated with it.
Several ground improvement techniques were considered. Although vibro com-
paction was the commonly practiced method of ground improvement on Palm Jumeira,
the possible presence of silty sand made the applicability of this technique dubious.
Stone columns were feasible but appeared to be very costly. Dynamic compaction was
considered as both economical and reliable, and the ground improvement works were
awarded to a specialist contractor who had proposed this technique.
Soil improvement acceptance criteria for a tank foundation level at level +2.5 m RL
were developed as:
• Allowable bearing capacity: 160 kPa with a safety factor of 3
• Differential settlement: 1/750 for a uniformly distributed load of 120 kPa.
A study on the available rigs in the region at the time indicated that the maximum
pounder lift capacity was limited to 15 tons; thus, in addition to classical dynamic
compaction, the specialist contractor decided to implement dynamic surcharging to
improve the achievable results at depth.
Dynamic surcharging is the combined effect of static loading and high energy
impacts. Its purpose was to create acceleration in the soil under the static loading in
such a way to generate a shearing process around the surcharge fill to reduce the load
spreading zone of the high strength superficial layer. Furthermore, vibrations and
increase of the pore pressure under the tank were to reduce the friction in between the
granular soil particles and ultimately result in the collapse and densification of the
foundation soil under the influence of dynamic surcharging.
In this technique a surcharge is initially placed over the treatment area, then
dynamic compaction is performed. Although granular materials settle under static
loads, as dynamic shear modulus has been found to decrease significantly with
increasing values of shear strain amplitude (Silver and Seed 1971), it can be expected
that introducing vibration will increase the amount of settlement under the surcharge.
Furthermore, the rate of consolidation of fine soils is most when the pore water pressure
is high, and it is possible to increase the rate of consolidation back to previously high
values by inducing pore water pressure through vibration.
Four settlement monitoring plates were installed on the ground surface. One plate
was installed in the center of the tank and the others were installed at 120o angles on a
ring whose radius was 17.5 m.
4 m of surcharge was then placed at the tank location in the shape of a truncated
cone whose top and bottom diameters were respectively 32.2 m and 44.2 m. It can be
calculated that the surcharge volume was approximately 4,700 m3. Assuming that the
in-situ unit weight of surcharge was 17 kN/m3, it can be further estimated that its total
weight was 80 MN. As tank foundation level was at +2.5 m RL, an additional over-
burden weight of 25 MN can be added to total surcharge weight. 4 m of surcharge and
1.5 m of overburden generates 93.5 kPa of pressure within the tank’s layout, which
was equivalent to 78% of the tank’s total load; however, the total surcharge load of
105 MN approximately equated to 90% of the tank’s total load.
204 B. Hamidi and S. Varaksin

Fig. 6. Ground settlement in tank A-A during static and dynamic surcharging (Hamidi and
Varaksin 2015)

Tank A-A settlements during placement of surcharge, static loading and dynamic
surcharging are shown in Fig. 6. It can be seen that the plates settled almost prorate
with surcharge height. Once placement of surcharge was completed, the preload was
left in place for five additional days. By then the settlement rate considerably
decreased, and it can be extrapolated that the ground could have additionally settled up
to 5 mm in the long term.
To increase the depth of treatment influence, 26 prints were excavated to the depth
of approximately 1 m on a periphery ring with a diameter of 49.2 m before applying
dynamic surcharging by dropping a 15-ton pounder from 20 m a total of 30 times in
five cycles at the location of each print.
It can be observed in Fig. 6 that dynamic surcharging increased the settlements of
Tank A-A by 1.6 to 5.2 times the values of the static settlements. The maximum effect
was on the tank’s periphery where impact wave amplitudes were greatest. Also,
although the maximum differential settlement of the outer monitoring plates was only
7 mm at the end of static surcharging, maximum differential settlement during dynamic
surcharging increased by 4 times to 40 mm, which suggests the possibility of large
differential settlements under seismic and vibratory loads in untreated areas of the same
ground.
The surcharge was removed and the ground was excavated within a diameter of
41.2 m to working platform level at +2.8 m RL upon completion of dynamic sur-
charging. The excavation sides sloped outwards in such a way that the top of exca-
vation had a diameter of 46 m.
Dynamic compaction was performed on prints located in the center of the tank and
on 4 concentric rings around the central print. Each print was additionally excavated by
approximately 1 m to facilitate pounder penetration and to increase the depth of
influence. 150 m3 of crushed rock and cobbles were also added to the total of 58 prints
in each tank. This amounts to about 2.6 m3 of added rock per print or an equivalent of
approximately 0.13 m of rock thickness per every meter of ground within the treatment
zone. This amount of stone was insufficient to efficiently increase the ground strength
as was intended and described in Ras Laffan Tank T-6 but was rather used to increase
 Ground Improvement of Tank Foundations in the Middle East 205

soil permeability. Pounder drop height during the first and second phases of com-
paction was 20 m. A lesser impact energy was used during the ironing phase.
Pounder penetrations and top of crater diameters were measured during the first two
phases of dynamic compaction for each print. The average values of these two in phase 1
were respectively 1.7 m and 5 m. These average figures respectively reduced to 0.4 m
and 2.3 m in the second phase. It is noted that the pounder’s dimension was 1.7 m.
The ground level dropped to +2.25 m RL at the end of dynamic compaction. With
consideration of the added stone volume, it can be calculated that the ground had
settled 0.68 m in addition to the settlements induced by dynamic surcharging.
Although the magnitude of this settlement is significantly greater than what was
measured during dynamic surcharging, it is noted that the purpose of dynamic sur-
charging was to reduce the settlements of the deep layer that were less impacted by the
energy of the 15-ton pounder.
Testing and Verification
Upon completion of dynamic compaction and leveling of the site, 4 PMTs were carried
out in Tank A-A. PLM and EM before and after ground improvement are shown in
Fig. 7. It can be observed that most of the improvement has occurred to depths of
approximately 8 to 9 m; however, due to the combination of dynamic surcharging and
pre-excavated dynamic compaction, PLM increased by 380% and 70% respectively at
depths of 5 m and 10 m. Furthermore, minimum PLM after improvement was observed
to be greater than 600 kPa, which demonstrates that the young hydraulic fill was no
longer subject to creep due to self-weight (Menard 1975).
Bearing capacity can be calculated using the equation proposed by Menard (1975).
In Tank A-A, the geometric mean of the average of the 4 post ground improvement
PMT was 1,684 kPa. Conservatively assuming that the deeper layers also have the
same value and that the foundation was on ground level, the allowable bearing capacity
can be calculated to be 449 kPa, which is considerably greater than the required
160 kPa.
Tank settlements were calculated by taking the tank-soil interaction into account
and using finite element analysis and three-dimensional modeling. The modeled tank
was supported by a 0.5 m thick concrete raft, which was underlain by a very dense
upper layer and a lesser dense bottom layer. The ground in the model was made stiffer
on one side by varying the thicknesses of two soil layers to assess differential settle-
ment effects. In this procedure, the thicknesses of the upper and lower sand layers on
the left side of the tank were respectively assumed to be 7.5 m and 3.5 m. On the right
side, the upper and lower sand layers were each assumed to be 6.5 m. Young’s moduli,
Ey, for the sand layers were calculated from EM (Menard 1975), and assumed to be 57.9
and 22.5 MPa respectively for the upper and lower sand layers.
Finite element analysis suggested that maximum settlement at the center of the tank
and minimum settlement at the shell were respectively 21.35 mm and 10.91 mm. Thus,
the differential settlement over the radius length of 17.55 m was 10.44 mm or less than
1/1,681, which is substantially less than the allowed value of 1/750. Differential set-
tlement from one side to the other side of the tank was calculated to be 3.13 mm or less
than 1/11,200.
206 B. Hamidi and S. Varaksin

Fig. 7. PLM and EM values before and after ground improvement (Hamidi and Varaksin 2015)

3.3 Fuel Tanks in Sohar, Oman

Introduction and Ground Conditions

Sohar Aluminum Factory is located in Sohar, Oman, and includes a power plant that
contains 4 gas turbines, 2 steam turbines, heavy recovery generators, associated
buildings, and tanks. Phase one of the power plant includes 2 fuel tanks, and a third
tank has been envisaged for the second phase of the plant. The diameters of these steel
tanks were approximately 31.25 m, and the required bearing capacity was 200 kPa.
Initial geotechnical studies indicated that the ground at the location of the tanks was
composed of loose to dense silty sand from the natural ground level at +5.35 m CD
(Chart Datum) to −13 m CD, where very dense silty sand or sandstone was encoun-
tered. Groundwater level was identified at +1.2 m CD.
Supplementary PMTs that were later carried out indicated that the upper crust of the
ground surface was very dense with PLM of 5 MPa. The limit pressure then varied in
the range of approximately 1.5 to 2.1 MPa.
It was also assessed that the in-situ ground conditions did not satisfy the designer’s
requirements, and implementation of specific foundation measures was investigated.
Foundation Solution: Dynamic Compaction
Acceptance criterion that was specified required the allowable bearing capacity of the
tank foundations to be 200 kPa with a safety factor of 3. Tank base levels were defined
at +4.5 m CD. It is noted that it is the authors’ experience that due to their foundation
sizes, tanks practically do not fail in bearing and that local shear failure beneath the
shell, disruption of tank serviceability or structural damage due to ground deformations
are more probable problems that need to be considered.
 Ground Improvement of Tank Foundations in the Middle East 207

Prior to commencement of treatment the working platform level was cut down to
elevation +4.5 m CD.
A 14.5-ton pounder was used to dynamically compact the ground. Pounder drop
height in the first and second phases of ground improvement was 20 m but was reduced
to 10 m for the ironing phase. 10 and 8 blows were applied per print respectively in
phase 1 and phase 2, but only a single blow was applied to each ironing print.
Testing and Verification
Pressuremeter test results before and after dynamic compaction are shown in Fig. 8.
The reference level for depth is at +4.5 m CD. It can be observed that even though the
strength and deformation parameters at the tank locations was on the denser side of
materials, nevertheless PLM and EM respectively improved by approximately 1.6 and
1.9 times in the 4 m of soil beneath the hard upper crust.
Calculation of allowable bearing capacity using the method of Menard (1975)
indicates that the project’s criterion has been well satisfied. Also, it can be calculated
that settlements within the treated layers of the ground will have reduced to approxi-
mately half of the untreated values.

Fig. 8. PLM and EM values before and after ground improvement

4 Conclusion

Tanks undergo a combination of uniform settlements, planar tilting and non-planar

differential settlements that can undermine the performance and serviceability of tanks,
result in damages or ultimately lead to failure. While they are relatively thin, heavily
loaded and sensitive structures, international specifications appear to be relatively
tolerant to uniform settlements but restrict planar tilting and non-planar deformations
more stringently.
208 B. Hamidi and S. Varaksin

Ground improvement techniques have been used successfully in numerous tank

projects throughout the world in general and in the Middle East in particular. Several
successful cases of ground improvement by application of dynamic compaction and
verification by Menard pressuremeter test for various types of tanks in the region were
presented in this paper.

Acknowledgments. The authors would like to express their appreciation to Menard for pro-
viding the information used in this paper.

References
ACI Committee 376: Agenda for Webinars on November 30, December 2 and 16, 2010, p. 23
(2010)
American Concrete Institute: Code Requirements for Design and Construction of Concrete
Structures for the Containment of Refrigerated Liquefied Gases, p. 170. ACI, Farmington
Hills (2010)
American Petroleum Institute: API Standard 653: Tank Inspection, Repair, Alteration, and
Reconstruction, 3rd edn., Washington, D.C., p. 112 (2003)
American Petroleum Institute: API Standard 625: Tank Systems for Refrigerated Liquefied Gas
Storage, 1st edn., Washington, D.C., p. 72 (2010)
American Petroleum Institute: API Standard 650: Welded Tanks for Oil Storage, 11th edn.,
Washington, D.C., p. 445 (2011)
Buschmeier, B., Masse, F., Swift, S., Walker, M.: Full scale instrumented load test for support of
oil tanks on deep soft clay deposits in Louisiana using controlled modulus columns. In:
International Symposium on Ground Improvement (IS-GI) Brussels 2012, Brussels, pp. 359–
372 (2012)
Cognon, J.M., Liausu, P., Vialard, R.: Combination of the drains and surcharge method with
dynamic compaction. In: 8th European Conference on Soil Mechanics and Foundation
Engineering, Helsinki, pp. 219–224. A.A. Balkema, Rotterdam, Netherlands (1983)
Debats, J.M., Scharff, G., Balderas, J., Melentijevic, S.: Ground improvement efficiency and
back-analysis of settlements. Ground Improv. 166(3), 138–154 (2013)
D’Orazio, T.B., Duncan, J.M.: Differential settlement in steel tanks. J. Geotech. Eng. ASCE
113(9), 967–983 (1987)
Hamidi, B., Jullienne, D.: The construction and performance of foundation using dynamic
compaction and dynamic replacement for an LNG tank in Qatar. In: 16th South East Asian
Geotechnical Conference, pp. 523–527 (2007)
Hamidi, B., Masse, F., Racinais, J., Varaksin, S.: The boundary between deep foundations and
ground improvement. Geotech. Eng. 169(GE2), 201–213 (2016)
Hamidi, B., Varaksin, S., Nikraz, H.: Application of dynamic surcharging for construction of
tanks on reclaimed ground. In: International Conference on Advances in Geotechnical
Engineering (ICAGE), Perth, pp. 873–878 (2011)
Hamidi, B., Varaksin, S.: Dynamic compaction and dynamic surcharging at Dubai’s Palm
Jumeira STP. In: Ground Improvement: Case Histories and New Directions, pp. 309–335.
Elsevier, Oxford (2015)
 Ground Improvement of Tank Foundations in the Middle East 209

Hendy, M.S., Muir, I.C.: Experience of dynamic replacement on a 40 m deep reclamation in

Hong Kong. In: Third International Conference on Ground Improvement Geosystems:
Ground Improvement Geosystems - Densification and Reinforcement, pp. 76–80. Thomas
Telford, London (1997)
Ihm, C.W., Masse, F.: Successful application of Menard vacuum consolidation method to
Nakdong river soft clay in Kimhae, South Korea. In: 15th International Conference on Soil
Mechanics and Foundation Engineering (ICSMGE), Istanbul (2001)
Marr, W.A., Ramos, J.A., Lambe, T.W.: Criteria for settlement of tanks. J. Geotech. Geoenviron.
Eng. 108(GT8), 1017–1039 (1982). ASCE
Menard, L.: The Menard Pressuremeter, Sols Soils No. 26 (1975)
Mobil: Chapter 6 - Foundation Design. Mobil Engineering Guide: EGS 262-90. Mobil (1990)
Silver, M.L., Seed, H.B.: Deformation characteristics of sands under cyclic loading. J. Soil Mech.
Found. Div. 9(SM8), 1081–1098 (1971). ASCE
 Default k-Values for Estimating Resilient
Modulus of Coarse-Grained Nigerian
Subgrade Soils

Abdulfatai Adinoyi Murana(&)

Department of Civil Engineering, Ahmadu Bello University,

Zaria, Kaduna, Nigeria
[email protected]

Abstract. In this paper subgrade materials from different locations in Nigeria

were characterized for use in the Mechanistic-Empirical Pavement Analysis and
Design. The engineering properties of the coarse-grained Nigerian subgrade
soils materials were obtained in the laboratory. Seven selected resilient modulus
constitutive equation for estimating the resilient modulus of coarse-grained
subgrade soils were used to estimate the default values using the repeated load
triaxial test result conducted on coarse-grained Nigerian subgrade soils. These
default resilient modulus parameters developed can be used to estimate the
resilient modulus of the compacted subgrade soils with reasonable accuracy and
utilized as level 3 resilient modulus input for Mechanistic-Empirical Pavement
Analysis and Design.

Keywords: Coarse-grained soils
Resilient modulus
Mechanistic-Empirical

Flexible pavement
Subgrade soils

1 Introduction

A conventional flexible pavement consists of a prepared subgrade or foundation and

layers of sub-base, base and surface courses (AASHTO 1993).
For the roadbed soils, the seasonal variation of resilient moduli is considered and used
directly to determine the design or effective roadbed soil resilient modulus. However,
seasonal variation of the resilient moduli for pavement materials is not used or considered
in the design process, even though the resilient modulus of pavement materials can vary
substantially throughout the year (Von Quintus and Killingsworth 1997).
The design and evaluation of pavement structures on base and subgrade soils requires
a significant amount of supporting data such as traffic loading characteristics, base,
subbase and subgrade material properties, environmental conditions and construction
procedures. Characterization of pavement materials is a key requirement for the pave-
ment design process. The characterization task involves obtaining material properties
that identify the material response to external stimuli of traffic loading and environmental
conditions. Characterization of subgrade materials using resilient modulus involves
obtaining material properties (index properties, physical and compaction properties) that
identify the material response to external stimuli of traffic loading and environmental

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_17
 Default k-Values for Estimating Resilient Modulus 211

conditions. In its 2002 design guide, the AASHTO advocated the use of the resilient
modulus parameter for describing granular material behaviour. Laboratory techniques
provide a means for directly measuring the resilient modulus parameter, wherein the
process usually involves elaborate and extensive testing at various stress levels and
physical conditions to completely map the range of the resilient modulus parameter for
any material under consideration. To be able to adopt the Mechanistic-Empirical pave-
ment design method for use in Nigeria, calibration of the subgrade material models to
reflect Nigerian conditions need to be carried out.
A more widely used recent test method on which pavement designs are based is the
resilient modulus value. It is defined as the ratio between repeated deviator stress and
resilient strain. The laboratory testing procedures for determining the resilient modulus
values is time consuming and needs expensive equipment and highly trained personnel
(Vogrig et al. 2003).
The resilient modulus is a fundamental engineering material property that describes
the non-linear stress-strain behaviour of pavement materials under repeated loading. It
is defined as the ratio of the maximum cyclic stress to the recoverable resilient (elastic)
strain in a repeated dynamic loading (Mohammad et al. 2007).
It is a measure or estimate of the elastic modulus of the material at a given stress or
temperature. Mathematically it is expressed as the ratio of applied deviator stress to
recoverable strain (George 2004).
In Level 1 design/analysis, the MEPDG requires input of the regression constants
of the stress-dependent constitutive equation for resilient modulus of a particular
unbound material (subgrade soil or base aggregate). This ensures a more accurate
assessment of the modulus during the analysis over the design period including sea-
sonal variation and varying stress conditions. Constitutive equation coefficients
(k-values) are usually obtained from the regression analysis of resilient modulus test
data for an actual soil/aggregate sample (Hossain 2010).
Some agencies consider the cost, time, complication, and sampling resolution
required for meaningful resilient modulus testing to be too cumbersome for its appli-
cation in less critical projects. Regardless of project size, it is often difficult to predict
and consequently reproduce the in-situ conditions, usually with respect to the state of
stress, further complicating the use of resilient modulus testing. Because of this, cor-
relations are desired for estimating resilient modulus, especially for use (or verification
of default values) associated with MEPDG Level 2 design/analysis. A common method
to predict a resilient modulus value is to use the stress-dependent constitutive equation
with the k-values estimated from soil index properties through further regression
equations. MEPDG Level 3 design/analysis also requires a specific resilient modulus
value as input (Hossain 2010).

2 Resilient Modulus Constitutive Models

Mathematical models are generally used to express the resilient modulus of subgrade
soils such as the bulk stress model and the deviatoric stress model. These models were
utilized to correlate resilient modulus with stresses and fundamental soil properties.
212 A.A. Murana

A valid resilient modulus model should represent and address most factors that affect
the resilient modulus of subgrade soils (Titi et al. 2006).
Several other models were reported in the literature, which use both stresses (either
confining and deviatoric stresses or bulk or octahedral stresses) that are functions of
confining and deviatoric stresses. The most general form of a three-parameter model is
as shown in Eq. 1 (Ooi et al. 2006; NCHRP 2008):

MR ¼ k1 Pa ½f ðcÞk2 ½gðsÞk3 ð1Þ

where f(c) is a function of confinement; g(s) is a function of shear and k1, k2, and k3 are
constants.
The effects of confinement in these models can be expressed in terms of the minor
principal stress (r3), bulk stress (h), or octahedral stress (roct = h/3), while the
parameter options for modelling the effects of shear include the deviatoric stress or
octahedral shear stress (soct). The three-parameter models represented by the Eq. 1 are
more versatile and apply to all soils (NCHRP 2008).
Uzan (1985) studied and discussed different existing models for estimating resilient
modulus. The Uzan equation was developed as a combination of the bulk and deviator
stress models in an effort to improve the predicted response of Mr test results by
including both axial and shear effects. The model defined the resilient modulus as
shown in Eq. 2 (Uzan 1985; NCHRP 2008):
 k2  k3
h rd
MR ¼ k1 Pa ð2Þ
Pa Pa

where k1, k2, and k3 are material constants; h = bulk stress; rd = deviatoric stress; Pa is
the atmospheric pressure
An equation similar to Uzan’s model using the octahedral shear stress instead of the
deviator stress was developed by Witczak and Uzan as shown in Eq. 3 (Witczak and
Uzan 1988; NCHRP 2008):
 k2  k3
h soct
MR ¼ k1 Pa ð3Þ
Pa Pa

where h = bulk stress (r1 + r2 + r3); r1 = major principal stress

r2 = intermediate principal stress = r3 for MR test on cylindrical specimen.
r3 = minor principal stress/confining pressure; soct = Octahedral shear stress.

1 1=2
soct ¼ ðr1  r2 Þ2 ðr2  r3 Þ2 ðr3  r1 Þ2
3

An equation similar to Uzan’s model using the confining pressure instead of the
bulk stress was recommended by Pezo as shown in Eq. 4 (Pezo 1993; NCHRP 2008):
 Default k-Values for Estimating Resilient Modulus 213

 k2  k3
r3 rd
MR ¼ k1 Pa ð4Þ
Pa Pa

An equation similar to Pezo’s model using the confining pressure and deviator
stress in a three-parameter formulation was recommended by Ni et al. as shown in
Eq. 5 (Ni et al. 2002; NCHRP 2008):
   
r3 k2 rd k3
M R ¼ k1 P a 1 þ 1þ ð5Þ
Pa Pa

Ooi et al. slightly modified the equation recommended by Ni et al. using the bulk
stress, octahedral shear stress and deviator stress in a three-parameter formulation into
two models as shown in Eqs. 6 and 7 (Ooi et al. 2004; NCHRP 2008):
 k2  
h rd k3
M R ¼ k1 P a 1þ 1þ ð6Þ
Pa Pa
   
h k2 soct k3
M R ¼ k1 P a 1 þ 1þ ð7Þ
Pa Pa

An equation similar to Ooi et al.’s model using the octahedral shear stress and bulk
stress was recommended by the NCHRP project 1-28 A as shown in Eq. 8 (NCHRP
2008):
 k2  
h soct k3
MR ¼ k1 Pa 1þ ð8Þ
Pa Pa

3 Evaluation of the Resilient Modulus Model Parameters

for Coarse-Grained Soils

The resilient modulus of coarse-grained soils obtained in the laboratory were statisti-
cally analysed. These values were used in evaluating the Mr parameters of the
coarse-grained soils using the seven resilient modulus equations presented in literature.
Figures 1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 3c, 4a, 4b, 4c, 5a, 5b, 5c, 6a, 6b, 6c, 7a, 7b and 7c
presents the histogram of the resilient modulus parameters (ki) values of coarse-grained
soils obtained from the resilient modulus equations evaluated.
Figure 1a, 1b and 1c shows the histogram of resilient modulus parameters (ki)
values obtained from Uzan model.
Figures 2a, 2b and 2c present the histogram of the resilient modulus model
214 A.A. Murana

16

14

12

10
Frequency

0
-2000 0 2000 4000 6000 8000
k1 from Uzan model

Fig. 1a. Histograms of Uzan resilient modulus model parameters k1

4
Frequency

0
-0.03 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04
k2 from Uzan model

Fig. 1b. Histograms of Uzan resilient modulus model parameters k2

 Default k-Values for Estimating Resilient Modulus 215

12

10

8
Frequency

0
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
k3 from Uzan model

Fig. 1c. Histograms of Uzan’s resilient modulus model parameters k3

parameters ki obtained using the Witczak and Uzan resilient modulus model.

10

8
Frequency

0
-2000 0 2000 4000 6000 8000 10000
k1 from Witczak and Uzan model

Fig. 2a. Histograms of Witczak and Uzan resilient modulus model parameters k1
216 A.A. Murana

4
Frequency

0
-0.03 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04
k2 from Witczak and Uzan model

Fig. 2b. Histograms of Witczak and Uzan resilient modulus model parameters k2

12

10

8
Frequency

0
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
k3 from Witczak and Uzan model

Fig. 2c. Histograms of Witczak and Uzan resilient modulus model parameters k3
 Default k-Values for Estimating Resilient Modulus 217

Figures 3a, 3b and 3c present the histogram of the resilient modulus model
parameters ki obtained using the Pezo resilient modulus model.

6
Frequency

0
-2200 -400 1400 3200 5000 6800
k1 from Pezo model

Fig. 3a. Histograms of Pezo’s resilient modulus model parameters k1

3.0

2.5

2.0
Frequency

1.5

1.0

0.5

0.0
-0.03 -0.02 -0.01 0.00 0.01 0.02 0.03
k2 from Pezo model

Fig. 3b. Histograms of Pezo’s resilient modulus model parameters k2

218 A.A. Murana

12

10

8
Frequency

0
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
k3 from Pezo model

Fig. 3c. Histograms of Pezo’s resilient modulus model parameters k3

Figures 4a, 4b and 4c present the histogram of the resilient modulus model
parameters ki obtained using the Ni et al. resilient modulus model.

12

10

8
Frequency

0
-500 0 500 1000 1500 2000 2500 3000
k1 from Ni et al model

Fig. 4a. Histograms of Ni et al.’s resilient modulus model parameters k1

 Default k-Values for Estimating Resilient Modulus 219

10

8
Frequency

0
-0.20 -0.15 -0.10 -0.05 0.00
k2 from Ni et al model

Fig. 4b. Histograms of Ni et al.’s resilient modulus model parameters k2

6
Frequency

0
0 1 2 3 4
k3 from Ni et al model

Fig. 4c. Histograms of Ni et al.’s resilient modulus model parameters k3

220 A.A. Murana

Figures 5a, 5b and 5c present the histogram of the resilient modulus model
parameters ki obtained using the Ooi et al. A resilient modulus model.

10

8
Frequency

0
-500 0 500 1000 1500 2000 2500 3000
k1 from Ooi et al model A

Fig. 5a. Histograms of Ooi et al. A resilient modulus model parameters k1

6
Frequency

0
-0.12 -0.08 -0.04 0.00
k2 from Ooi et al model A

Fig. 5b. Histograms of Ooi et al. A resilient modulus model parameters k2

 Default k-Values for Estimating Resilient Modulus 221

10

8
Frequency

0
0 1 2 3 4
k3 from Ooi et al model A

Fig. 5c. Histograms of Ooi et al. A resilient modulus model parameters k3

Figures 6a, 6b and 6c present the histogram of the resilient modulus model
parameters ki obtained using the Ooi et al. B resilient modulus model.

6
Frequency

0
-500 0 500 1000 1500 2000 2500 3000
k1 from Ooi et al model B

Fig. 6a. Histograms of Ooi et al. B resilient modulus model parameters k1

222 A.A. Murana

6
Frequency

0
-0.12 -0.08 -0.04 0.00
k2 from Ooi et al model B

Fig. 6b. Histograms of Ooi et al. B resilient modulus model parameters k2

6
Frequency

0
0 2 4 6
k3 from Ooi et al model B

Fig. 6c. Histograms of Ooi et al. B resilient modulus model parameters k3

 Default k-Values for Estimating Resilient Modulus 223

Figures 7a, 7b and 7c present the histogram of the resilient modulus model
parameters ki obtained using the NCHRP’s resilient modulus model.

6
Frequency

0
-500 0 500 1000 1500 2000 2500 3000
k1 from NCHRP and MPEDG model

Fig. 7a. Histograms of NCHRP resilient modulus model parameters k1

10

8
Frequency

0
-0.08 -0.06 -0.04 -0.02 0.00
k2 from NCHRP and MPEDG model

Fig. 7b. Histograms of NCHRP resilient modulus model parameters k2

224 A.A. Murana

6
Frequency

0
0 2 4 6
k3 from NCHRP and MPEDG model

Fig. 7c. Histograms of NCHRP resilient modulus model parameters k3

Table 1. Statistical data of ki obtained from the test results of the coarse-grained soils using
resilient modulus models.
S/No. Model Resilient modulus model A-1-b A-2-4 A-2-7
parameters
1 Uzan 1985 k1 1128.2647 1567.4997 1493.3910
k2 7.98E-17 5.31E-17 −4.1E-17
k3 0.2695 0.3094 0.2140
2 Witczak and Uzan k1 1381.7457 2011.3930 1769.3172
1988 k2 8.88E-17 5.45E-17 −4.2E-17
k3 0.2695 0.3094 0.2140
3 Pezo 1993 k1 1128.2647 1567.4997 1493.3910
k2 8.67E-17 3.2E-17 −3.4E-17
k3 0.2695 0.3094 0.2140
4 Ooi et al. 2004 k1 601.3008 797.2932 907.2052
k2 −0.0339 −0.0389 −0.0269
k3 0.9587 1.1006 0.7613
5 Ooi et al. 2004 k1 620.7723 823.0154 929.3132
k2 −0.0339 −0.0389 −0.0269
k3 1.6028 1.8400 1.2727
6 Ni et al. 2002 k1 594.1272 787.8599 899.0481
k2 −0.0485 −0.0557 −0.0385
k3 0.9457 1.0856 0.7509
7 NCHRP 2008 k1 605.4392 802.7457 911.9080
k2 −0.0192 −0.0220 −0.0152
k3 1.5964 1.8326 1.2676
 Default k-Values for Estimating Resilient Modulus 225

From the evaluation, as presented in Figs. 1a, 1b, 1c, 2a, 2b, 2c, 3a, 3b, 3c, 4a, 4b,
4c, 5a, 5b, 5c, 6a, 6b, 6c, 7a, 7b and 7c, the resultant resilient modulus parameters of
coarse-grained soils with the following classifications (A-1-b, A-2-4 and A-2-7) using
the resilient modulus models are as presented in Table 1.
Based on the evaluation of the resilient modulus equations for coarse-grained soils,
it was observed from Table 1 for level 3 analysis that the resilient modulus equation
adopted by NCHRP was the best in determining resilient modulus of coarse-grained
soils.

4 Conclusion

Based on the results of this research, the following conclusions are reached:
1. Resilient modulus constitutive equation adopted by NCHRP and MEPDG was
adopted for estimating resilient modulus of coarse-grained soils.
2. Default values of resilient modulus parameters was determined for coarse-grained
soils as level 3 resilient modulus input.

References
AASHTO: Guide for Design of Pavement Structures. American Association of State Highway
and Transportation Officials, Washington, DC (1993)
George, K.P.: Prediction of Resilient Modulus from Soil Index Properties,
FHWA/MS-DOT-RD-04-172, Department of Civil Engineering, The University of Missis-
sippi, Mississippi (2004). http://www.mdot.state.ms.us/research/pdf/ResMod.pdf
Hossain, S.M.: Characterization of Unbound Pavement Materials From Virginia Sources for Use
in the New Mechanistic-Empirical Pavement Design Procedure. Virginia Transportation
Research Council, Commonwealth of Virginia, Virginia (2010)
Mohammad, L.N., Kevin, G.P., Ananda, H.P., Munir, D.N.: Comparative Evaluation of
Subgrade Resilient Modulus from Non-destructive, In-situ, and Laboratory Methods.
Louisiana Transportation Research Center, Louisiana Department of Transportation and
Development. National Technical Information Service, Springfield (2007)
NCHRP: Synthesis 382: Estimating Stiffness of Subgrade and Unbound Materials for Pavement
Design. National Cooperative Highway Research Program, Transportation Research Board,
Washington, DC (2008)
Ni, B., Hopkins, T.C., Sun, L., Beckham, T.L.: Modeling the resilient modulus of soils. In:
Proceedings of the 6th International Conference on the Bearing Capacity of Roads, Railways,
and Airfields, vol. 2, pp. 1131–1142. Balkema Publishers, Rotterdam, the Netherlands (2002)
Ooi, P.S., Archilla, A.R., Sandefur, K.G.: Resilient modulus models for compacted cohesive
soils. Transportation Research Record No. 1874 Transportation Research Board, National
Research Council, pp. 115–124 (2004)
Ooi, P.S., Sandefur, K.G., Archilla, A.R.: Correlation of resilient modulus of fine-grained soils
with common soil parameters for use in design of flexible pavements. Honolulu: Report
No. HWY-L-2000-06, Hawaii Department of Transportation (2006)
226 A.A. Murana

Pezo, R.F.: A general method of reporting resilient modulus tests of soils: a pavement engineer’s
point of view. In: 72nd Annual Meeting of the Transportation Research Board. Transportation
Research Board, Washington, DC (1993)
Titi, H.H., Mohammed, B.E., Sam, H.: Determination of Typical Resilient Modulus Values for
Selected Soils in Wisconsin. University of Wisconsin, Department of Civil Engineering and
Mechanics. National Technical Information Service 5285 Port Royal Road, Springfield,
Milwaukee (2006)
Uzan, J.: Characterization of granular material. In: Transportation Research Record 1022,
Transportation Research Board, National Research Council, pp. 52–59 (1985)
Vogrig, M., McDonald, A., Vanapalli, S.K., Siekmeier, J., Roberson, R., Garven, E.: A
laboratory technique for estimating the resilient modulus of unsaturated soil specimens from
CBR and unconfined compression tests. In: 56th Canadian Geotechnical Conference, 4th
Joint IAH-CNC/CGS Conference, 2003 NAGS Conference, Canada (2003)
Von Quintus, H., Killingsworth, B.: Design Pamphlet for the Determination of Design Subgrade
in Support of the AASHTO Guide for the Design of Pavement Structures. Office of
Engineering R&D, Federal Highway Administration, National Technical Information
Services, Springfield (1997)
Witczak, M.W., Uzan, J.: The Universal Airport Pavement Design System, Report 1 of 4,
Granular Material Characterization. University of Maryland, College Park (1988)
 Finite Element Analysis of Rock Slope Stability
Using Shear Strength Reduction Method

Greg You1(&), Maged Al Mandalawi1, Ahmed Soliman2,

Kim Dowling1, and Peter Dahlhaus1
1
Faculty of Science and Technology,
Federation University Australia, Ballarat, Australia
[email protected]
2
Glencore, Mount Isa Mines, Zinc Assets, Mount Isa, Australia

Abstract. Finite element analysis incorporating the shear strength reduction

method was applied to study the west slope stability of an open cut mine in
Australia using Mohr–Coulomb and generalized Hoek–Brown criteria. The pit
of the mine had multiphase excavations and reached 180 m in depth. The study
investigated three slope configurations, namely, Stage 1 inter ramp slope 43°,
Stage 2 inter ramp slope 49° and optimized Stage 2 slope 54°. When imple-
menting the generalized Hoek–Brown failure criterion, the equivalent factor of
safety was 1.96, 1.87 and 1.40 under dry slope for the three configurations,
respectively. However, under partly saturated conditions, the optimised slope
would have a factor of safety 1.16. Furthermore, the generalised Hoek–Brown
criterion generated lower factors of safety than the Mohr–Coulomb failure cri-
terion. The difference is related to an overestimation of the shear strength
parameters by the linear Mohr–Coulomb criterion under low confining stresses
compared with the non-linear Hoek–Brown.

1 Introduction

Slopes need be designed and cut with a margin of safety for open pit mines. Economics
could be improved by steepening the slope thereby reducing the amount of waste
excavation; however, excessive steepening of slope could result in failure leading to
loss of life and damage to property (Singh et al. 1989; Singh and Singh 1992). The
factors, which mainly influence the stability of a typical open-pit slope, are the shear
strength parameters of slope forming material, the presence and characteristics of
discontinuities in the slope mass and the groundwater conditions (Singh and Monjezi
2000; Singh et al. 2008).
The shear strength reduction (SSR) method was used for soil slope stability analysis
in 1975 (Zienkiewicz et al. 1975). Duncan (1996) defined the factor of safety of soil
slopes as the ratio of actual shear strength to the minimum shear strength required to
avoid failure, or the factor by which shear strength must be reduced to bring a slope to
failure. This method was applied for rock masses (Shangyi et al. 2003). Yingren and
Shangyi (2004) and Hammah et al. (2007) demonstrated the efficiency of SSR method
for slopes of soil and rock masses. This method was also adopted in several other
studies, such as Yeo and Chan (1993), Dowson et al. (1999), Hammah et al. (2004a, b),

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_18
228 G. You et al.

Zheng et al. (2009) and Gupta et al. (2016). The advantages of the SSR method over
limit equilibrium method (LEM) are: (1) No assumption is required of the interslice
shear force distribution; (2) the critical failure surface in the slope can be found from
the shear strain; and (3) this method is suitable for complex slope conditions in order to
interpret details of displacements, stresses and water pressures. LEM may give lower
estimates of failure volumes than SSR method in numerical modelling (Chiwaye 2010).
The SSR approach includes the search for a stress reduction factor (SRF) value that
brings the slope to fail. The shear strength reduced by a factor of safety F can be
determined using a series of trials to adjust the friction angle (/0 ), and the cohesion (c′)
of slope rock mass. For example of the Mohr-Coulomb criterion (Eq. 1), the process of
shear stress reduction can be expressed in Eq. 1a.

s ¼ c0 þ r0 tan u0 ð1Þ

s c0 r0 tan u0
¼ þ ð1aÞ
F F F
0
Where, the reduced value of cohesion c0 ¼ cF and the reduced value of internal
 0
friction angle u0 ¼ arctan tanu
F :
This paper presents a finite element (FE) analysis of slope stability using SSR
method for three inter-ramp slopes at the Handlebar open pit mine, namely Stage 1
(43°), Stage 2 (49°) and an optimised Stage 2 slope (54°), under both dry and partly
saturated conditions that implements the Mohr–Coulomb and the generalised Hoek–
Brown failure criteria.

2 Site Geology

Handlebar Hill open - pit mine is located at Mt. Isa, north Queensland, Australia
(Fig. 1). In the Mount Isa Valley, the rock formation is represented by Magazine Shale,
Spears-Kennedy Siltstone, Urquhart Shale, Native Bee Siltstone, and Eastern Creek
Volcanics (Figs. 1 and 2). The ground is oxidized and leached of sulphides. The
leaching can extend to great depths aided by faults (Fig. 2). The leached zone in the
rock slope is regarded as an unconfined aquifer (Rosengren and Associates 2007).
Pre-mining groundwater level data indicated that groundwater levels coincided with the
base of the oxidized zone at a depth of about 50 m below ground level.
Mining started in 2008 through multiple stages. Stage 1 was excavated to a depth of
77 m and Stage 2 to 180 m. There were six benches at each stage of varying bench
height from 10 m to 17 m (Fig. 2). The inter-ramp angle (IRA) was 43° and 49° for
Stage 1 and Stage 2, respectively, on the west wall at the 4440 N cross-section.
 Finite Element Analysis of Rock Slope Stability 229

Fig. 1. Location and geological map of the study area (after Conaghan et al. 2003)

Fig. 2. Rock domains of the west slope at N4440 cross-section

230 G. You et al.

3 Data Preparation and Finite Element Analysis

Both the Mohr–Coulomb criterion (Eq. 1) and the generalized Hoek–Brown criterion
(Eq. 2) are used in this analysis.
 a
r03
r01 ¼ r03 þ rci mb þs ð2Þ
rci

Some of the parameters used in the Hoek–Brown criterion (e.g., s, mb, and a) are
available from rock mass data, while other parameters used (mi, ʋ, rci) were derived
using laboratory rock tests (Seville 1981; Tarrant and Lee 1984). From June 2006 to
June 2007, a total of 20,189.6 m of diamond core was drilled and 3,080 m of rock
chips were completed (Rosengren and Associates 2008). The geological strength index
GSI was determined from the diamond core logging and in-situ surface mapping. The
mean GSI value of the overall slope was reduced by 20% in this study (Table 1). For
the weathered Magazine Shale, a GSI value of 24 was used because this represented a
poorer rock mass quality below the base of complete oxidization in the simulations.
The rock mass modulus (Erm) of all domains was estimated according to Hoek and
Diederichs (2006). A disturbance factor of D = 0.7 was used for small scale blasting in
civil engineering slopes that results in modest rock mass damage if controlled blasting
is used (Hoek et al. 2002). The parameters used in the study are detailed in Table 1.

Table 1. Input data used in the slope stability analysis

Parameter Unit Domain
1 2 3 4 5 6 7 8
c kN/m3 27.4 28.3 26.5 26.5 23.5 27.2 26.9 31.1
Erm MPa 3000 10000 4500 4500 1500 8500 2000 12000
u (°) 23 32 13 25 20 35 22 44
c kPa 350 800 50 410 240 990 300 1250
rt MPa 0.1 0.3 0.0 0.1 0.0 0.4 0.0 0.2
rci MPa 36.4 55.5 32.0 32.0 26.3 111.1 32.7 108
ʋ − 0.3 0.2 0.25 0.2 0.25 0.2 0.25 0.2
mb − 0.184 0.42 0.271 0.271 0.148 0.319 0.165 1.036
mi − 4 4 4 4 4 4 4 11
s − 0.0003 0.0026 0.0008 0.0008 0.0002 0.0013 0.0002 0.002
a − 0.509 0.503 0.505 0.505 0.511 0.504 0.510 0.504
D − 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7
80% GSI − 36 48 24 41 32 43 34 46
Dilation (°) 5 8 4 5 7 12 4 12

SSR technique is often used with FEM to solve sophisticated problems such as
estimating stability of slope (Ng et al. 2000). In this study, Phase2 (RocScience 2014)
FE program is used for the slope stability simulations under different scenarios.
The model comprises an area of 750 m in width and 450 m in depth (Fig. 2).
 Finite Element Analysis of Rock Slope Stability 231

The pit bottom is located at an elevation of 3,368 m (175 m deep), and the top of the
west slope is at an elevation of 3,543 m (ground–surface level). The ramp crosses the
west slope section at an elevation of 3,463 m (77 m deep).
Phase2 is a powerful FE program, which can be used for a wide range of engi-
neering projects including slope stability, groundwater seepage and probabilistic
analysis. It can simulate and analyze a complex multi-stage model. Progressive failures
and explicit modelling of discontinuities can be simulated to gain further insight into
the rock mass behavior of the slope. When setting up the project, the multi-stage model
is available to simulate the stresses resulting from different excavations. The program
was designed to calculate the last excavation stage for the critical SRF, or factor of
safety. Therefore, the strength reduction factor analysis has to be run many times for
the two stages of excavations. The FE simulations were carried out to further improve
the overall slope stability analysis, namely to optimize the Stage 2 IRA.
Figure 3 shows maximum shear strains simulated for IRA 43° and 49° using the
Mohr–Coulomb criterion and the generalized Hoek–Brown criterion. At the upper IRA
43° (Stage 1, excavation depth 77 m), there is a localized high shear strain zone in

Fig. 3. Maximum shear strain contours, (a) Stage 1, IRA 43°, (b) Stage 2, IRA 49° and (c) IRA
49° with enhanced slope cohesion strength, dry slope conditions. Right: Generalized Hoek–
Brown criterion, left: Mohr–Coulomb criterion
232 G. You et al.

Domains 3 and 5, which is structurally controlled leached Magazine shale and Spears
siltstone (Fig. 3a, left). At the lower IRA 49° (Stage 2, excavation depth 180 m), the
shear strain concentration at toe is prominent (Fig. 3b, c). In comparison, there is less
deformation in the slope using the Mohr–Coulomb criterion than using the generalized
Hoek–Brown failure criterion (Fig. 3). These results indicate that an increase in
maximum shear stress occurs at the final excavation of stage 2.
In the third case of the study, the Stage 2 IRA was increased 5° to optimize the
open–pit design to be as steep as possible. Figure 4 shows the maximum shear strain
contours for the IRA 54° under dry and partly saturated conditions. As the effective
normal stress decreases under partly saturated slope conditions, larger maximum shear
strains develop. There is a clear increase in shear strain in Domains 3 and 5 and at the
toe of the slope (Fig. 4b, c).

a b

c d

Fig. 4. Maximum shear strain contours for optimized slope after two excavation stages, for
lower IRA of 54°, (a, b) dry slope, (c, d) partly saturated slope

4 Results and Analyses

The critical SRF or factor of safety of the FE analyses is tabulated in Table 2 for IRAs
43°, 49° and optimized IRA 54° under dry and partly saturated conditions using both
Mohr–Coulomb and generalized Hoek–Brown failure criteria. As can be expected,
rising slope angle leads to lower critical SRF. Furthermore, the critical SRF is lower in
partly saturated slope than the dry one. The modelling predicted a higher factor of
safety using the Mohr–Coulomb criterion than using the generalized Hoek–Brown
criterion (Table 2). This may attribute to the linear failure envelope in the Mohr–
Coulomb criterion (Fig. 5). Figure 5 is plotted based on the shear strength analysis of
Urquhart Shale rock using RocLab 1.0 (RocScience 2014). Rock mass strength in the
 Finite Element Analysis of Rock Slope Stability 233

Table 2. Critical SRF from FE analyses for different slopes

IRA Dry slope Partly saturated slope
Mohr–Coulomb Generalized Hoek–Brown Mohr–Coulomb Generalized Hoek–Brown
43° 2.76 1.96 2.70 1.71
49° 2.00 1.87 1.86 1.59
54° 1.98 1.40 1.85 1.16

(a) (b)

Fig. 5. Comparison between the Mohr–Coulomb criterion and the generalized Hoek–Brown
failure criterion, (a) in term of major and minor stresses, (b) in term of shear and normal stresses

area is a non-linear stress function. The linear Mohr–Coulomb failure criterion may not
agree well with the rock mass failure envelope where the low confining stresses in an
open-pit slope make the non-linearity of rock behaviors more possible (Shen 2013). On
the other hand, the Mohr–Coulomb strength parameters may overestimate the shear
strength of a rock mass at a high stress level because it was derived using a straight line
fitted over the Hoek–Brown curve (Lin et al. 2014). Hammah et al. (2004a, b) stated
that the generalized Hoek–Brown criterion is the most suitable strength model for
predicting the failure of rock masses, especially in low normal stress ranges.
If the Stage 2 slope increases 5° to IRA 54°, the critical SRF is 1.16 under partly
saturated condition using the generalized Hoek–Brown criterion, which is less than 1.3.
This will have an influence on the stability of the west slope if the shear strength failure
234 G. You et al.

mechanisms of the rock masses are considered to dominate the slope stability in wet
conditions. The results for slopes under the other conditions show a stable slope with
SRF > 1.3 using both failure criteria. For the long term operations and after heavy
rain-fall, bench scale instabilities can be expected. Given that a GSI reduction of 20%
was assumed and the Erm was estimated as a worst case scenario for the rock mass
strength parameters, the slope is expected to be stable. The groundwater has influence
on the slope stability, in particular, in the case of steeper slope with an IRA of 54°.

5 Conclusions

Slope stability analysis of the west slope of Handlebar Hill open-pit mine was con-
ducted using non-directional strength of rock mass properties through finite element
analyses implemented the shear strength reduction technique. The generalized Hoek–
Brown criterion did not generate any SRF greater than the results predicted by the
Mohr–Coulomb failure criterion. The difference is related to an overestimation of the
shear strength parameters by linear fit under low confining stresses compared with the
non-linear Hoek–Brown. The stability of the west slope would be overestimated when
directly using the shear strength parameters for the analysis. Consequently, the dis-
turbance factor value (D), Erm and the GSI must be significantly calibrated and con-
sidered as main inputs to model the slope.
Presuming that a minimum factor of safety 1.3 is adopted for long term excava-
tions, the Stage 2 IRA could be optimized and increased to 54°, in which SRF was 1.4.
However, this IRA would have a SRF of 1.16 in case of partly saturated slope con-
ditions. Inter-ramp stability can be controlled by both geological structure and rock
mass strength, however, the combined failure mechanism is not the scope of this study.

Acknowledgments. The authors would like to express their appreciation to Glencore zinc for
providing permission to carry out this research and publish this paper.

References
Chiwaye, H.: A Comparison of the Limit Equilibrium and Numerical Modelling Approaches to
Risk Analysis for Open-Pit Mine Slopes. University of the Witwatersrand, Johannesburg (2010)
Conaghan, L.E., Hannan, W.K., Tolman, J.: Mount Isa Cu and Pb-Zn-Ag deposits, NW
Queensland, Australia. In: Advances in Regolith: Proceedings of the Cooperative Research
Centre for Landscape Environments and Mineral Exploration (CRC LEME). Regional
Regolith Symposia, Queensland, Mt. Isa (2003)
Dawson, E.M., Roth, W.H., Drescher, A.: Slope stability analysis by strength reduction.
Géotechnique 49(6), 835–840 (1999)
Duncan, J.M.: State of the art: limit equilibrium and finite-element analysis of slopes. J. Geotech.
Eng. 122(7), 577–596 (1996)
Gupta, V., Bhasin, R.K., Kaynia, A.M., Kumar, V., Saini, A.S., Tandon, R.S., Pabst, T.: Finite
element analysis of failed slope by shear strength reduction technique: a case study for
Surabhi Resort Landslide, Mussoorie township, Garhwal Himalaya. Geomatics Nat. Hazards
Risk 7(5), 1677–1690 (2016). doi:10.1080/19475705.2015.1102778
 Finite Element Analysis of Rock Slope Stability 235

Hammah, R.E., Curran, J.H., Yacoub, T. E., Corkum, B.: Stability analysis of rock slopes using the
finite element method. In: Proceedings of the ISRM Regional Symposium EUROCK (2004a)
Hammah, R.E., Curran J.H., Corkum, B., Yacoub T.E.: Stability analysis of rock slopes using the
finite element method. In: Euoreka 2004 and 53rd Geomechanics Colloquium. Schubert
(2004b)
Hammah, R.E., Yacoub, T.E., Corkum, B., Wibowo, F., Curran, J.H.: Analysis of blocky rock
slopes with finite element shear strength reduction analysis. In: Proceedings of the 1st
Canada-U.S. Rock Mechanics Symposium, Vancouver, Canada, pp. 329–334 (2007)
Hoek, E., Diederichs, M.S.: Empirical estimation of rock mass modulus. Int. J. Rock Mech. Min.
Soc. 43, 203–215 (2006). Canada
Hoek, E., Carranza-Torres, C., Corkum, B.: Hoek-Brown failure criterion-2002 edition. In:
Proceedings of 5th North American Rock Mechanics Symposium, pp. 267–273. University of
Toronto Press, Toronto (2002)
Lin, H., Zhong, W., Xiong, W., Tang, W.: Slope stability analysis using limit equilibrium method
in nonlinear criterion. Sci. World J. (2014). Article ID 206062, 7 pages. Hindawi Publishing
Corporation
Yeo, K., Chen, A.: Finite element slope stability analysis by shear strength reduction technique.
Jpn. Geotech. Soc. 33(2), 194–196 (1993)
Ng, C.W.W., Zhang, L.M., Ho, K.K.S., Choy, C.K.: Influence of laterally loaded sleeved piles on
slope stability. In: GeoEng 2000, Melbourne, Australia (2000)
Phase2 V9.0/RocScience: A Two-Dimensional Finite Element Analysis of Soil and Rock for
Underground and Excavated Slopes, User’s Manual, RocScience Inc., Toronto (2014)
RocLab 1.0/RocScience: Program for Determining Rock Mass Strength Parameters, User’s
Manual, RocScience Inc., Toronto (2014)
Rosengren, K., and Associates: Proposed Handlebar Hill Open Cut geotechnical review
(Technical Report No. 27010). Xstrata, Mt. Isa (2007)
Rosengren, K., and Associates: Handlebar Hill Open Cut Pit. Internal Technical report, in Xstrata
report in 2011. Xstrata, Mt. Isa (2008)
Seville, R.: Review of Hilton Mine Rock Property Data. Mount Isa Mines Limited (1981)
Shangyi, Z., Yingren, Z., Weidong, D.: Stability analysis on jointed rock slope by strength
reduction FEM. Chin. J. Rock Mech. Eng. 2, 020 (2003)
Shen, J.: Analytical and numerical analyses for rock slope stability using the generalized
Hoek-Brown criterion. Doctoral Thesis, Adelaide University, Australia (2013)
Singh, T.N., Ulabhaje, A., Singh, D.P.: Planning of Slope Stability in an Opencast Mine- A
Model Approach, National Seminar on Surface Mining, Dhanbad edited by the Ind. Min. and
Eng. Jr., pp. 5.2.1–5.2.10 (1989)
Singh, T.N., Singh, D.P.: Slope stability study in an opencast mine over previously worked seam.
In: International Symposium on Rock Slope, New Delhi, pp. 467–477 (1992)
Singh, T.N., Gulati, A., Dontha, L., Bhardwaj, V.: Evaluating cut slope failure by numerical
analysis – a case study. Nat. Hazards 47, 263–279 (2008)
Singh, T.N., Monjezi, M.: Slope stability study in jointed rock mass - a numerical approach. Min.
Eng. J. 1(10), 12–13 (2000)
Tarrant, G., Lee, M.: Mount Isa rock properties (Technical No. RES MIN 60). Mount Isa Mines
Limited (1984)
Yingren, Z., Shangyi, Z.: Application of strength reduction FEM in soil and rock slopes. Chin.
J. Rock Mech. Eng. 23(19), 3381–3388 (2004)
Zheng, Y., Tang, X., Zhao, S., Deng, C., Lei, W.: Strength reduction and step-loading finite element
approaches in geotechnical engineering. J. Rock Mech. Geotech. Eng. 1(1), 21–30 (2009)
Zienkiewicz, O.C., Humpheson, C., Lewis, R.W.: Associated and nonassociated. Visco-plasticity
and plasticity in soil mechanics. Geotechnique 25(4), 671–689 (1975)
 Study of the Behavior of Floating Stone
Columns in Soft Clay Formations
Using Numerical Modeling

Ayman L. Fayed1, Tamer M. Sorour1, and Hany F. Shehata2(&)

1
Structural Engineering Department, Ain-Shams University,
Abbasseya, Cairo, Egypt
{ayman_fayed,tamer.sorour}@eng.asu.edu.eg
2
Housing and Building National Research Center (HBRC), Dokki, Giza, Egypt
[email protected]

Abstract. Stone columns have been increasingly acknowledged as one of the

viable foundation techniques in soft ground conditions. The concept behind this
option is to enhance soft soils by insertion of special elements stiffer than the
original soil. The stiffness of the improved ground depends on the geometry of
the stone columns in addition to the relative stiffness between the original soft
soil and the installed elements. Back analysis of case histories is always a
successful and reliable approach in studying the in situ behavior of different
ground improvement methods. In this paper, a well instrumented case study of
floating stone columns installed in the Bothkennar soft clay formation is studied
using numerical modeling utilizing the finite elements method. Different soil
constitutive models are investigated to define the most appropriate one in
simulating the actual Bothkennar soft clay behavior. Results of the comparison
between the monitored field performance and the estimated behavior from the
numerical model using different constitutive models are presented and dis-
cussed, while recommendations for the most suitable stress strain model of the
studied clay are concluded.

1 Introduction

Vibro replacement is a popular form of ground improvement commonly used to

enhance the settlement and bearing capacity characteristics of soft soils. Vertical col-
umns of compact stones are formed in the ground using either the top or bottom feed
systems. Vibro replacement process typically involves replacing 10–35% of the in situ
soil with crushed stones. The high stiffness properties of the crushed stones reduce the
overall and differential settlements of the structures founded on the treated soil. The
consolidation time is also reduced due to the high permeability of the crushed stones.
Stone columns can be used to provide support for a variety of loading scenarios ranging
from small footings (i.e. pads/strip footings) to large area loadings (i.e. embankments
and large dimensions slabs).
Design of foundations on soft soils is usually governed by settlement criteria rather
than bearing capacity due to the high compressibility characteristics of soft soils. The
majority of analytical design methods developed to date contain many simplifying
© Springer International Publishing AG 2018
W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_19
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 237

assumptions, such as the unit cell concept, which assumes an infinite grid of columns
supporting an infinitely wide load area. Therefore, they do not account for the loss of
lateral confinement associated with groups of columns supporting small footings. In
addition, the reduction in vertical stresses with depth beneath small footings is much
sharper than that beneath wide area loadings and, therefore, offers the possibility of
partial depth treatment. While some correction factors exist to account for the loss of
lateral confinement, current design methods do not consider the loss of lateral con-
finement for small groups of floating stone columns. While the bearing capacity of
small groups of stone columns has been well researched, a lack of information exists
regarding the settlement performance of small groups of columns. This is highlighted
by McCabe et al. (2009) whose settlement database of over 20 case studies comprised
only three cases related to small groups of columns.

2 Numerical Modelling: Procedures and Discussions

Many numerical studies conducted to study the behavior of the stone columns are
axisymmetric analyses of large groups of columns. The shortage of information regarding
the settlement performance of small groups of stone columns was identified by Black
(2006) who conducted some high quality laboratory research; however, it is difficult to
extrapolate the findings due to scale effects associated with laboratory tests and also as
some of the area ratios considered are at the high end of typical values used in practice.
This research aims of studying the behavior of floating small groups of columns
using finite element modelling by PLAXIS 3D 2016 for full-scale trial models in
Bothkennar soft clay. The effect of the used constitutive laws on the performance of
stone columns and the associated ground deformations are also investigated for both
treated and untreated soft clay profiles. Lastly, some recommendations to properly
simulate the small group of stone columns by numerical modelling in Bothkennar soil
profile are provided.

2.1 Constitutive Models

Three common constitutive models that have already been implemented in most of the
geotechnical finite element codes are used in this research; Mohr-Coulomb Model
(MCM), Hardening Soil Model (HSM) and Soft Soil Model (SSM). A brief description
of these models is presented below.

2.1.1 Mohr Coulomb Model (MCM)

Mohr-Coulomb model as shown in Fig. 1a is an elastic-perfectly plastic model that is
often used to model soil behavior and serves as a first-order model. In general stress
state, the model’s stress-strain behaves linearly in the elastic range, with two defining
parameters from Hooke’s law (Young’s modulus, E and Poisson’s ratio, m). There are
two parameters define the failure criteria (the friction angle, / and cohesion, c) and also
a parameter to describe the flow rule (dilatancy angle, w which comes from the use of
non-associated flow rule which is used to model a realistic irreversible change in
238 A.L. Fayed et al.

(a) Elastic-perfectly plastic assumption of Mohr-Coulomb constitutive

model.

(b) The Mohr-Coulomb yield surface in principal stress space (c = 0)

Fig. 1. Mohr-Coulomb constitutive model.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 239

volume due to shearing). Mohr-Coulomb model is a simple and applicable to

three-dimensional stress space model, as shown in Fig. 1b, with only two strength
parameters to describe the plastic behavior.

2.1.2 Hardening Soil Model (HSM)

The Hardening Soil model (Brinkgreve and Vermeer 1997; Schanz 1998) is a true
second order model for soils in general, for any type of application (Brinkgreve 2005).
The model involves friction hardening to model the plastic shear strain in deviatoric
loading, and cap hardening to model the plastic volumetric strain in primary com-
pression. Distinction can be made between two main types of hardening, namely shear
hardening and compression hardening. Shear hardening is used to model irreversible
strains due to primary deviatoric loading. Compression hardening is used to model
irreversible plastic strains due to primary compression in oedometer loading and

Fig. 2. Total yield contour of the hardening-soil constitutive model in principal stress space for
cohessionless soil.
240 A.L. Fayed et al.

isotropic loading. Both types of hardening are contained in this model. Yield contour of
the model in three-dimensional space is shown in Fig. 2. Failure is defined by means of
Mohr-Coulomb failure criterion. Some basic characteristics of the model are Stress
dependent stiffness according to a power law (m), plastic straining due to primary
deviatoric loading (Eref ref
50 ), plastic straining due to primary compression (Eoed ), elastic
unloading/reloading input parameters (Eref ur , mur) and failure criterion according to the
Mohr-Coulomb model (c, u and w).

2.1.3 Soft Soil Model (SSM)

The soft soils could be considered like near-normally consolidated clays, clayey silts
and peat. A special feature of such materials is their high degree of compressibility.
This is best demonstrated by oedometer test data as reported for instance by Janbu in
his Rankine lecture (1985). Considering tangent stiffness moduli at a reference
oedometer pressure of 100 kPa, he reported for normally consolidated clays Eoed = 1 to
4 MPa, depending on the particular type of clay considered. The differences between
these values and stiffnesses for NC-sands are considerable as here the values are in the
range of 10 to 50 MPa, at least for non-cemented laboratory samples. Hence, in
oedometer testing normally consolidated clays behave ten times softer than normally

Fig. 3. Total yield contour of the soft-soil constitutive model in principal stress space.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 241

consolidated sands. This illustrates the extreme compressibility of soft soils. The Soft
Soil model is a Cam-Clay type model especially meant for primary compression of near
normally consolidated clay-type soils. The total yield contour in principal stress space
is shown in Fig. 3. The parameters of the Soft Soil model include compression (Cc) and
swelling (Cs) indicies, which are typical for soft soils, as well as the Mohr-Coulomb
model failure parameters.

2.2 Subsurface Conditions of the Case Studies Site

The site selected for the field trials at Bothkennar was a facility for large or full scale
soft clay researches, with the site having been used extensively for research into in-situ
testing and also ‘undisturbed’ sampling techniques. A comprehensive ‘state-of-the-art’
ground investigation and geotechnical data existed for the site. A wealth of researches
have been previously undertaken, including Hight et al. (1992), Paul et al. (1992) and
Nash et al. (1992a, b) and was detailed in the Institution of Civil Engineers
Géotechnique Symposium (Vol. 42, No. 2: 1992). The detailed geotechnical profile of
the site under study is shown in Fig. 4.
The Bothkennar clay profile shown in Fig. 4, is a result of a well-documented and
well-established geotechnical investigation through many research projects. Hight et al.
(1992) and Nash et al. (1992a) have provided rigorous investigation works, which had
been carried out to provide data characterizing the clay for the benefit of future research
projects. The clay properties used in the soil model are presented in Table 1 and are
divided into crust, upper Carse clay and lower Carse clay.

Fig. 4. Key geotechnical profile for the Bothkennar soft clay research site (after Nash et al.
(1992a, b))
242 A.L. Fayed et al.

A critical state friction angle (u′) of 34° is used for the Carse clay. A slightly
cohesion value of 3 kPa was used for the crust layers. Eref ref
50 is assumed equal to Eoed
according to Elshazly et al. (2009).
In choosing the friction angle of the stone backfill, the value of u′ = 45◦ shown in
Table 1, in which the stone column properties have been included, should be readily
achievable, as noticed by McCabe et al. (2009). Default values of E50 = 70 MPa and
Eur = 210 MPa were adopted. The exponent, m, for the Hardening Soil model was
chosen as m = 0.3. These stiffness parameters (i.e. E50, Eur and m) are in accordance
with the values adopted by Gäb et al. (2008).

Table 1. Summary of the geotechnical material parameters adopted in the analysis of the
Bothkennar test site
Soil parameter Crust Upper carse Lower carse Stone
clay clay columns
Depth (m) 0.0–1.2:1.5 1.2:1.5–2.5 2.5–14.5 As needed
Bulk unit weight, c (kN/m3) 18.0 16.5 16.5 19.0
Over-consolidation ratio (OCR) 1.0 1.0 1.5 –
Pre-overburden pressure (kPa) 15 15 0 –
Coefficient of lateral earth pressure, K0 1.5 1.0 0.75 0.3
Effective cohesion, c′ (kPa) 3 1 1 1
Angle of internal friction, /′ (degrees) 34 34 34 45
Dilatancy angle, w (degrees) 0 0 0 15
Initial voids ratio, e0 1.0 1.2 2.0 –
Compression index, CC (SSM) 0.07 0.25 1.12 –
Swelling index, CS (SSM) 0.01 0.03 0.16 –
Eref
50 , kPa (HSM)
1068 506 231 70000
Eref
ur , kPa (HSM) 5382 3036 1164 210000
E, kPa (MCM) 4485 2530 970.3 70000
m (HSM) 1.0 1.0 1.0 0.3
Reference pressure, pref (kPa) 13 20 30 100
Vertical coefficient of permeability, 6.9  10−5 6.9  10−5 6.9  10−5 1.2
kvert (m/day)
Horizontal coefficient of permeability, 1.0  10−4 1.0  10−4 1.0  10−4 1.2
khorz (m/day)

Case Study No (1)

Instrumented field load test on unreinforced rigid pad footings at the Bothkennar test
site, documented by Jardine et al. (1995), is numerically modelled during the scope of
this paper. The footing was loaded to failure using kentledge in less than five days, with
pauses in loading occurring overnight and whenever the rate of settlement exceeds
8 mm/hr. The foundation depth of the tested pad is 0.80 m below ground level, while
the square pad side length is 2.20 m.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 243

2.2.1 Finite Element Modelling

The load test is simulated using the software PLAXIS 3D 2016 as an undrained loading
(with effective stress material parameters) due to the short duration of the load test.
However, the crust in the adopted soil profile is modelled as a drained material as the
lower part of the crust contains a significant proportion of shelly fragments (Nash et al.
1992b) and the upper part of the crust is above the groundwater level. A fine mesh has
been generated to avoid any effects from the meshing size on the results. Wide
boundary conditions have been chosen in order not to affect the settlement response.
Figure 5 shows the discretization of the problem.

Fig. 5. Finite element model in case study no (1).

2.2.2 Results and Discussions

The relations between the load and settlement for both the field measurements and the
results of the numerical model using different soil constitutive models are presented in
Fig. 6. It can be recognized that the numerical model, in general, using the different
constitutive models; (MCM, HSM and SSM), captures the trend of settlement versus
applied pressure as compared to the monitored correlation from the field loading test.
For the HSM, the settlement predicted value at 140.0 kPa, which is the maximum
tested stress and represents almost the ultimate bearing pressure as can be understood
from the field stress-settlement relation, is almost the same as the field measurement,
244 A.L. Fayed et al.

Fig. 6. Comparison between numerical model results using different soil constitutive models
and the actual load–displacement behavior for a pad footing.

however there is always a slight over-prediction of the settlement values at any other
stress levels. MCM results are generally in a good agreement with the measured
settlement values at most stress levels, except at the ultimate stress, 140 kPa, where it
slightly underestimates the settlement value. The SSM yields acceptable estimate of the
settlement values at low stress levels up to about 30% of the ultimate stress value.
Above that level and at higher stress values, the SSM, in general, underestimates the

Fig. 7. Displacement field beneath the loaded pad at stress level of 140 kPa, using the HSM.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 245

settlement values of the loaded footings. For this case study, it can be concluded that
the three investigated constitutive models can reasonable simulate the actual
Bothkennar clay behavior, with the preference to the HSM at high stress levels. Fig-
ure 7 presents the total deformations field at an applied stress value of 140 kPa that is
almost the ultimate pressure sustained by the loaded pad utilizing the HSM.
Case Study No (2)
The second case study numerically analyzed in this research is fully detailed by
Colin J. Serridge (2013) in his PhD thesis comprising a full-scale field testing of 8
models resting on Bothkennar clay. Eight models were constructed to represent the
behavior of the rectangular and square footings in which seven models were resting on
a treated soil with small groups of stone columns and one model was loaded without
soil improvement. Table 2 summarizes the eight trial footings configurations, while
Table 3 provides the loading stages of each model. The first and second loading stages
were consolidated in 5 months each.

Table 2. Summary of the trial tests configurations used in case study no (2)
Trial Dimensions Foundation Ground treatment
footing (L  B) m depth (m) Number of Stone columns Stone columns lengths
no. stone spacing (m) (below footings) m
columns
1 6.0  0.75 0.50 4 1.50 5.70
2 6.0  0.75 0.50 3 2.00 5.70
3 3.0  0.75 0.50 2 1.50 3.70
4 3.0  0.75 0.50 2 1.50 5.70
5 3.0  0.75 0.50 2 1.50 7.70
6 3.0  0.75 1.20 2 1.50 5.70
7 1.5  1.5 0.50 2 1.20 5.70
8 3.0  0.75 0.50 Untreated soil – –

Table 3. Summary of the loading increments of the trial footings for case study no (2)
Trial footing Foundation depth 1st loading increment 2nd loading increment
no. (m) kN/m2 kN/m2
1 0.50 35.5 72.0
2 0.50 32.9 67.1
3 0.50 33.1 67.8
4 0.50 34.9 71.7
5 0.50 32.1 67.0
6 1.20 34.2 69.6
7 0.50 32.7 67.0
8 0.50 34.3 71.6
246 A.L. Fayed et al.

2.2.3 Finite Element Modelling

The full-scale models configurations, shown in Table 2, and with the loading stages,
summarized in Table 3, have been numerically modeled using PLAXIS 3D 2016. The
models represent rectangular as well as square footings configurations. In addition, the
foundation depths vary from 0.5 m to 1.2 m in order to check whether the crust layer
would affect the results or not. The effect of the number of stone columns and the
columns spacing (S) have also been investigated through these full-scale field tests.
One model (Footing no (8)) has been loaded without adding stone columns, to examine
the load-settlement behavior with time for the untreated soil.

2.2.4 Effect of Constitutive Laws on the Load-Settlement-Time

Relationship
Figure 8 and Table 4 present the load-settlement-time relations pertaining to Footing
no (8) for both the field and numerical models. It is noticed that, the MCM can’t mimic
the behavior of the Bothkennar clay with time especially at the high stress levels and
results in underestimated settlement values. On the other hand, the SSM provides a
better estimation of the load-settlement relation especially as the duration of loading
increases. The error in predicting the settlement, as shown in Table 4, ranges between
approximately (−10%) and (−18%) compared to the error values ranging between
approximately (−25%) & (−40%) for the MCM. The HSM predicted settlement values
are, in general, more close to the field measured deformations with error values ranging
between (−5%) & (+14%), as can be noticed from Table 4. Accordingly, it can be
concluded that the HSM can represent the load-settlement behavior of the footings
founded on the untreated Bothkennar clay in a better manner than both the MCM and
the SSM. However, the HSM overestimates the settlement values by (+10% to +14%)
compared to the field results of stress levels above 50% of the approximate ultimate
level.

Fig. 8. Measured and predicted load-settlement relations with time for footing no (8) without
ground improvement.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 247

Table 4. Measured and predicted settlement at different stress levels and durations for case
study no (2).
Applied Time Measured Numerical model results
stress (day) settlement MCM SSM HSM
(kPa) (mm) Estimated Error Estimated Error Estimated Error
settlement (%) settlement (%) settlement (%)
value alue (mm) Value
(mm) (mm)
34.3 0 −7.80 −7.05 −09.62 −7.25 −07.05 −7.40 −05.13
40 −7.90 −7.15 −09.49 −7.30 −07.59 −7.55 −04.43
120 −8.00 −7.25 −09.38 −7.40 −07.50 −7.80 −02.50
160 −8.10 −7.30 −09.88 −7.50 −07.41 −7.90 −02.47
71.6 160 −15.80 −9.50 −39.87 −13.50 −14.56 −17.50 +10.76
220 −20.40 −13.20 −35.29 −17.15 −15.93 −22.50 +10.29
280 −22.85 −15.25 −33.26 −19.90 −12.91 −25.95 +13.57
320 −25.90 −15.60 −39.77 −21.30 −17.76 −27.30 +05.41
107.6 321 −29.10 −20.00 −31.27 −25.50 −12.37 −32.50 +11.68
124.8 322 −47.40 −35.25 −25.63 −42.30 −10.76 −54.15 +14.24

2.2.5 Effect of the Stone Columns on the Load-Settlement-Time Relations

The first footing with stone columns (Footing no (1)) was numerically modelled rep-
resenting a small group of 4 stone columns with spacing of 1.5 m and length of 5.70 m
with a foundation depth of 0.5 m. The stone columns are modelled using the Mohr
Coulomb criterion, while the soil layers were simulated using the three types of soil
constitutive models; MCM, HSM and SSM. Figure 9 presents the discretization of the
model. Figure 10 shows the relation between the stresses and estimated average set-
tlement with time along with the field measurements. From the results, it can be
concluded that both the MCM and SSM constitutive models can’t mimic the softy soil
behavior properly and results in underestimated predicted values for the settlement

(a) Meshing and nodes. (b) Cross-section to show the stone columns

Fig. 9. Finite element models of footing no (1) for the second case study.
248 A.L. Fayed et al.

Fig. 10. Measured and predicted load-settlement relations with time for trial footing no (1) of
the second case study (footing with four stone columns).

with time. However, all the three soil models could be used at a low stress levels, in
which the stress is lower than 30% of the ultimate bearing pressure. The same as the
footing without improvement (Footing 8), the HSM could represent the soil layers of
the Bothkennar soil profile, however it results in higher predicted values of settlement
in the range of (+12% to +17%) above the field measurements. Figure 11 shows the
deformed shape of the stone columns under an applied stress of 72.0 kPa and using the
HSM, in which it can be noticed that bulging partially occurrs with some tilting of the
edge columns.
In the light of the previous results, the trial model (Footing 5) was modelled, which
represents a different group of only 2 stone columns with the same spacing of 1.5 m

Fig. 11. Deformed shape of the stone columns, for a loading stress of 72.0 kPa, using PLAXIS
3D 2016.
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 249

and foundation depth of 0.5 m but with longer stone columns lengths of 7.7 m. Only
the HSM has been adopted to represent the soil layers behavior. Figure 12 shows the
relation between the stress and average settlement for both the numerical model and
field measurements. It could be noticed that the HSM can represent the loading process,
in general, but results in an overestimation of the settlement values by about (+13% to
+18%) than the field measurements.

Fig. 12. Measured and predicted load-settlement relations with time for trial footing no (5) of
the second case study (footing with two stone columns) utilizing the HSM.

Footing no. (6), as detailed in Table 2, has been also simulated using the same
conditions, which represents a different foundation depth of 1.2 m to rest directly on
the upper carse clay, and with using the stone column’s spacing of 1.5 m and length of
5.7 m. The first construction stage was to excavate the overall crust layer, which is
1.2 m in height, in order to lay the footing directly on the upper carse clay. The second

Fig. 13. Measured and predicted load-settlement relations with time for trial footing no (6) of
the second case study (footing with two stone columns).
250 A.L. Fayed et al.

construction stage was to construct the stone columns and with the configuration given
in Table 2. Due to the time duration gap between the two stages, significant heave has
been occurred. Figure 13 shows the results of both the numerical model and the field
measurements. In this case, and due to the heave resulted from the stages of con-
struction, lower values of the settlement, ranging between (−5% to−9%), have been
predicted using the HSM. However, this range of error is considered still small and
would not greatly affect the results. Hence, the HSM could be used to reasonably
simulate the behavior of the Bothkennar soil profile in predicting the settlement values.

3 Conclusions

This paper presents results of the study of effect of the adopted constitutive models for
soft clays behavior simulation on the accuracy of numerical models in predicting the
settlement values associated with loading of footings supported on stone columns. The
finite element code Plaxis 3D 2016 is used in the performed analyses in which the three
common constitutive models; MCM, SSM and HSM, have been used to investigate
their suitability in modelling the behavior of the studied soft clay. Two case studies
were analyzed in which the first case was for a short term loading of a footing founded
on the Bothkennar clay without ground improvement. The second case study included
modelling of several trial footings with and without ground improvement and with
different configurations of the stone columns for the ground treated cases. The per-
formed analyses showed that the three investigated constitutive models can capture the
stress-settlement behavior of the Bothkennar soft clay with the preference of the HSM
over other models especially for the long term loading conditions. Using the HSM
resulted in predicting the settlement values associated with loading footings supported
on stone columns in the Bothkennar soft clay with an approximate deviation in the
accuracy ranging between (−9% to +20%) compared with the actual field measured
values. Both the MCM and SSM can be used in simulating the behavior of the studied
clay formation but with less accuracy ranging between approximately (−25% to −40%)
and (−10% to −18%), simultaneously.

References
Black, J.A.: The settlement performance of a footing supported on soft clay reinforced with
vibrated stone columns. Ph.D. thesis, Queen’s University of Belfast (2006)
Brinkgreve, R.B.J., Vermeer, P.A.: Plaxis Finite Element Code for Soil and Rock
Analysis-Version 7. Balkema, Rotterdam (1997)
Brinkgreve, R.B.J.: Selection of soil models and parameters for geotechnical engineering
application. In: Yamamuro, J.A., Kaliakin, V.N., (eds.) Geotechnical Special Publication
No. 128, ACSE, pp. 69–98 (2005)
Serridge, C.J.: An evaluation of partial depth dry bottom-feed vibro stone columns to support
shallow footings in deep soft clay deposits. Ph.D. thesis, Anglia Ruskin University, UK
(2013)
 Study of the Behavior of Floating Stone Columns in Soft Clay Formations 251

Elshazly, H.A., Hafez, D.H., Mossaad, M.E.: Reliability of conventional settlement evaluation
for circular foundations on stone columns. Geotech. Geol. Eng. 26(3), 323–334 (2009)
Gäb, M., Schweiger, H.F., Kamrat-Pietraszewska, D., Karstunen, M.: Numerical analysis of a
floating stone column foundation using different constitutive models. In: Proceedings of the
2nd International Workshop on Geotechnics of Soft Soils, Glasgow, pp. 137–142 (2008)
Hight, D.W., Bond, A.J., Legge, J.D.: Characterisation of the Bothkennar clay: an overview.
Geotechnique 42(2), 303–347 (1992)
Institution of Civil Engineers: Bothkennar soft clay test site: characterization and lessons learned.
Geotechnique 42(2), 161–378 (1992)
Jardine, R.J., Lehane, B.M., Smith, P.R., Gildea, P.A.: Vertical loading experiments on rigid pad
foundations at Bothkennar. Geotechnique 45(4), 573–597 (1995)
McCabe, B.A., Nimmons, G.J., Egan, D.: A review of field performance of stone columns in soft
soils. In: Proceedings of ICE Geotechnical Engineering (2009)
Nash, D.F.T., Sills, G.C., Davison, L.R.: Onedimensional consolidation testing of soft clay from
Bothkennar. Geotechnique 42(2), 241–256 (1992a)
Nash, D.F.T., Powell, J.J.M., Lloyd, I.M.: Initial investigations of the soft clay site at
Bothkennar. Geotechnique 42(2), 163–181 (1992b)
Paul, M.A., Peacock, J.D., Wood, B.F.: The engineering geology of the Carse clay of the national
soft clay research site. Bothkennar. Géotechnique 42(2), 183–198 (1992)
Schanz, T.: Zur modellierung des mechanishen verhaltens von reibungsmaterialen. Habilitation.
Stuttgard University (1998)
 Auto‐Controlled Ménard Pressuremeter:
A Novel Tool for Optimal Use
of the Pressuremeter

Wissem Frikha1,2(&) and Serge Varaksin2

1
Université Tunis El Manar, Ecole Nationale D’ingénieurs
De Tunis LR14ES03, Ingénierie Géotechnique, Tunis, Tunisia
[email protected]
2
APAGEO, ZA de Gomberville-rue Salvador Allende,
78114 Magny les Hameaux, France
[email protected]

Abstract. The pressuremeter test is an in situ controlled load-deformation test

that is performed on the wall of a borehole using a radially expanded cylindrical
probe. From the test readings (volume variation based on controlled pressure), a
plane deformation stress‐strain curve can be obtained for the soil.
Since the initial prototype, the pressuremeter has consistently been improved
in its design, and the latest version of the pressuremeter, which is called the
“auto-controlled Pressuremeter” has been developed to address the issues of
repeatability and accumulation of approximations in a test. This apparatus is
fully automatic and autonomous, and manages all steps of the test as instructed
by the operator. The auto‐controlled Pressuremeter simplifies the work proce-
dure for the operator, reinforces the reliability of the results, and reduces the
time of set up.
The present paper describes the auto‐controlled pressuremeter, and compares
it with the manual pressuremeter. The first part of the paper compares the
deficiencies of the manual and auto‐controlled Pressuremeter. In the second part,
test results of the two types of pressuremeter that have been performed in similar
conditions are presented and commented.

1 Introduction

The pressuremeter test is an in situ controlled load-deformation test that is performed

on the wall of a borehole using a radially expanded cylindrical probe. From the test
readings (volume variation based on controlled pressure), a plane deformation stress‐
strain curve can be obtained for the soil.
The first pressuremeter prototype was developed by Louis Ménard in January 1955.
Since then, the pressuremeter has consistently been improved in its design.
The first pressuremeter prototype, which is called Type A, consists of a hand pump
for injecting constant increments of water and a large probe with a diameter of
140 mm. After this first model, over the years, several prototypes have been designed
by selecting better materials for pressure application, minimizing approximations, and

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_20
 Auto-Controlled Ménard Pressuremeter 253

improving the sensor and recording system; i.e. Types B, C, D, E, F, G, etc. (Cassan
2005).
Type G pressuremeter refers to devices with nested cell probes. Old notations GA,
GC used in 1965–1966, correspond to pressure-volume controllers interchangeable
with type G. For these devices, the guard cells are inflated with compressed air. The
pressuremeter GB, has a totally different design, characterized by two guard cells
inflated with water and by two volumeters.
From 1984, a pressuremeter identical to the GA was put into service, except that
there is no differential pressure gauge. The latter is calculated by the operator from the
pressures measured by gauges from the measuring cell and the guard cells. At present,
it is the most used pressuremeter. Over three thousand pressuremeters have been built
and used in all the word (over hundred countries).
Recently, there has been technological developments such as pressure gauges that
connect to the circuit and allow working with automatic data acquisition (Geospad®).
This system contains electrical recording components connected to pressure and vol-
ume sensors.
The last generation of pressuremeter follows the electronic and automatic techno-
logical developments. It is called the “auto-controlled pressuremeter” and has been
developed to address the issues of repeatability and accumulation of inaccuracies in the
test. According to ISO 22476‐4 standard, this apparatus is fully automatic, autono-
mous, and manages all steps of the test as preselected by the operator. The auto‐
controlled pressuremeter simplifies the work procedure for the operator, enhances the
reliability of the results, and reduces the time of preparation.
The first part of this paper describes the manual and the auto‐controlled pres-
suremeters. The second part of the present paper presents the process of correction of
pressure loss provided by the auto-controlled pressuremeter. Finally, a comparison
between tests that have been performed (in situ and in artificial ground) using the two
types of pressuremeter, in similar conditions, are presented and commented.

2 Manual Ménard Pressuremeter –Type G

The manual Ménard pressuremeter consists of a CU (control unit), a plastic tubing and
a 3-cell probe, which allow performing in situ tests according to the ISO 22476-4 and
ASTM D4719-07 standards. The borehole is drilled to minimize wall disturbance and
to keep the diameter of the hole in accordance with the selected probe size.
The CU is an aluminium box (86 cm
43 cm
26 cm), with a protective cover,
standing on a 65 cm high tripod. It is handle transported, its weight is 24.5 kg (tripod,
3.5 kg) (Fig. 1). The tripod and level allow the vertical setting of the pressuremeter in
all sites.
The CU consists of devices allowing the regulation of the applied pressure and the
reading of the volume change. It includes an 800 cm3 sight tube volumeter for reading
the volume changes of the measuring cell, regulators for both main and differential
pressures, pressure gauges of 0 to 25 bars and 0 to 60 bars for both guard and
measuring cells, and several valves and connectors (Fig. 1).
254 W. Frikha and S. Varaksin

Num DESCRIPTION Num DESCRIPTION

1 Gas circuit stop valve (complete) 15 Pressure regulator (complete)
2 25/100 bar gas selection valve (complete) 16 Fixation for lid type 3
3 Bleed valve for both gas and water (complete) 17 Main pressure regulator
4 Inversion valve 0-10m 18 Sight tube complete (new model)
5 25/100/60 bar water selection valve (complete) 19 Polycarbonate cover plate for sight tube

6 Water circuit stop valve (complete) 20 Quick female socket for extra gauge
7 0-25 bar gauge (vertical outlet) 21 Quick female socket for hose outlet (water and gas)
8 0-60 bar gauge (vertical outlet) 22 Quick female socket for Nitrogen bottle inlet
9 Ø 100 gauge ring 23 Tripod
10 Ø 60 gauge ring 24 Stainless steel tripod axis
11 0-250 bar gauge (vertical outlet) 25 Filter housing complete
12 Ball Fixation for GeoBOX® 26 Rilsan tubing 3x6 Pressuremeter inner black lead
13 Stop watch 1 minute 27 Rilsan tubing 3x6 Pressuremeter inner red lead
14 Protection case for stop watch

Fig. 1. Ménard pressuremeter equipped with central unit computer (GeoBOX®) and central
acquisition tool (Geospad®)
 Auto-Controlled Ménard Pressuremeter 255

The probe is totally protected by a rubber cover (different types according to the
soil stiffness), which is inflated by a gas in the 2 guard cells and by water in the
measuring cell. The different cells’ applied pressures are controlled by the differential
regulator to ensure a cylindrical deformation along the measuring cell and to avoid any
adverse boundary condition. The pressure source is provided through an external
nitrogen gas cylinder.
The pressuremeter can be equipped with system tools that allows instant display of
test results via a central unit computer (GeoBOX®) and central data acquisition tools
(Geospad®). It permits the automatic record of the test data and the specific test
conditions and the visualisation of the evolution of the data during the test. The
pressure of the measuring cell, the differential pressure, the volumes variation, the
number of increments and time are displayed during the performance of the test. The
records are automatically carried out after 0, 15, 30 and 60 s with optimized accuracy:
0.10 cm3 on the volume and 10 kPa on the pressures (in accordance to procedure B of
ISO 22476‐4).
The acquisition tool (Geospad®), integrated into Ménard pressuremeter, is a
waterproof box including 2 pressure sensors of 0 to 100 bars and an ultrasonic sensor
for volume.
To perform the test, a borehole is drilled to minimize wall disturbance and to keep
the cavity diameter in accordance with the probe size. The probe is lowered into the
borehole to the required test depth and the pressure is applied by equal increments.
Pressure and volume is read from the control unit. As soon as the probe is lowered into
the borehole to the required test depth, the operator can start the test by pressure
increments with the control unit.

2.1 Auto-controled Ménard Pressuremeter (Geopac®)

The main differences between the auto-controlled pressuremeter and manual pres-
suremeter is the method of application of the pressure and the volume and their
measurement.
The pressurization of the guard cell is performed by a modulation of nitrogen gas
from an external cylinder using a control pressure regulator and sets of solenoid valves.
Two pressure intervals are considered: 0 to 7 bars with an accuracy range of 0.01 bars
and 7 to 100 bars with an accuracy range of 0.1 bars.
Water pressurizing in the measuring cell is performed using electro-mechanical
tools consisting of a DC electric motor of 24 V and a reducer with a normal torque of
24 N/m, a high ratio gearbox with an epicycloid stage gear train, a ball screw torque, a
transmitter of pushing force, a pressure-transmitting sealing piston, etc. The electronic
control allows the variation of the power associated with the electric motor to command
the speed by using an incremental encoder of small resolution.
The mechanical part of the auto-controlled pressuremeter permits the transforma-
tion of the electrical energy into work pressure applied on the water circuit.
The performance tests, the analysis of operations and the approval of the model
(electrical and mechanical) were performed and certified by an independent national
organisation CETIM (Centre Technique des Industries Mécaniques, France) in 2010.
256 W. Frikha and S. Varaksin

The auto-controlled pressuremeter (GeoPAC®) consists of an adapted piston in a

cylinder driven by a precision micro-motorization and is entirely piloted by a control
unit (GeoBOX®) with a wireless connection (Fig. 2). The pressure sensors chosen for
the device are of a high-precision class, allowing the target pressure to be reached and
regulated quickly without drift. The regulation ensures the simultaneous recording of
data at the imposed times. The precision on the measurement of the volumes, of an
order of magnitude never previously reached in manual pressuremeter test, is
1/100 cm3. The deformation at the interface between the probe and the borehole wall is
therefore measurable with a resoluteness, depending on the type of probe used and its
degree of expansion, relative to the loaded surface, in the order of 1 lm. Even before
starting the soil tests, this precision is enhanced, due to the reliability and repeatability
of the measurements, when the calibrations are carried out in series.

Fig. 2. Auto-controlled pressuremeter (GéoPAC®)

When the test is performed, the operator records the initial parameters and specific
conditions in the control unit, such as borehole number, depth of the test, pressure of
first stage, etc. As soon as the probe is lowered into the borehole to the required test
depth, the control unit can begin the test. All the sequences of the whole process of
standard test are also automated: pressure loss, volume loss, pressure increments and
pressure steps settings. During the entire process, the control unit computer shows the
monitoring of the actual test on its screen (progression, view of the results in real time,
line graph etc.). The operator can stop or modify the test progress from the control unit
at any time.
The auto-controlled pressuremeter can measure volume and pressure respectively
with the precision of 0.05 cm3 and 0.025 bar.
 Auto-Controlled Ménard Pressuremeter 257

2.2 Principle of Pressure Loss Measurement

The pressure loss in a pressuremeter test is related to the tubing dimensions and the
flow rate of the liquid circuit. For the classical Ménard pressuremeter, the regulation is
usually performed after the test. Only auto-controlled pressuremeter is able de correct
instantaneously the measurements by taking into account the regulation of the pressure
loss. The adjustment of the pressure measurements depend on the pressure increment
between two steps. In this phase, pressure should be increased to the next increment in
an optimal time that does not exceed 20 s with adequate differential pressure between
the liquid and gas circuits.
The differential pressure in the probe cannot be relied upon due to the overesti-
mation of the liquid pressure when the pressure loss is not taken into account during the
test. Indeed, at the nearest hydrostatic pressure, for a target differential pressure of
0.1 MPa and with a pressure loss of 0.2 MPa, the real differential pressure in the probe
is then −0.1 MPa. The pressure regulation undergoes a strong disturbance around the
target pressure. This regulation tends to decrease the flow rate when the measured
pressure is near the set point, but the reduction in the flow rate causes a drop in the
measured pressure (Arsonnet et al. 2013).
The control responds to this reduction by applying again the flow, which is
translated by a phenomenon of oscillation going until the hydraulic circuit is excited
(Arsonnet et al. 2013).
In the case of creep in certain levels (in the expansion of probe phase), the
adjustment of flow in the liquid circuit is necessary to maintain the pressure constant.
This flow creates a pressure loss, which adds to the probe pressure and results in the
overestimation of the pressure by the sensor. The main consequence during the trial is
that the differential pressure in the probe is not respected due to the overestimation of
the liquid pressure and the pressurizing level is not made at a lower pressure that varies
according to the instantaneous flow rate injected during this phase. This result is more
evident in a calibration test where the creep is maximal and can cause intense oscil-
lation in tests with long tubing.
To overcome these problems, the first approach for compensating the pressure drop
was to identify the characteristics of the tubing, through laboratory tests, and to
determine the coefficients of pressure loss per unit length (Fig. 3). The results show
different tendencies lines as a function of tubing length. The auto-controlled pres-
suremeter also take into account the singular pressure loss in the entire hydraulic
system, which can be varied significantly with the occasional or permanent sealing
phenomenon: this is related to the in situ extended use of the hydraulic circuits tubing,
their deformation during loading, and their repair by shortening and connector instal-
lation (Arsonnet et al. 2013).
Therefore, the pressuremeter test always requires a minimum volume of liquid to be
injected at the beginning of test and before the probe comes into contact with the wall
of the drilled hole. The auto-controlled pressuremeter consists of an automatic iden-
tification procedure that instantaneously computes the coefficients of the loss pressure
in the hydraulic circuit. This procedure, which is called the “pressure loss calibration
procedure” is carried out automatically without intervention of the operator at the
beginning of the test, based on the control of hydraulic flow.
258 W. Frikha and S. Varaksin

0,30 Calculated for l=25ml

Calculated for l=33ml
Calculated for l=50ml
0,25
Calculated for l=100ml
Measured for l=25ml
0,20 Measured for l=33ml
Measured for l=50ml
Δ P(MPa)

Measured for l=100ml

0,15

0,10

0,05

0,00 ΔP=(LN(LN(l)-2)+0,15)(Q2+300Q)10-6
0 50 100 150 200 250 300 350
3
Flow Q(cm /minute)

Fig. 3. Pressure loss as a function of flow rate for different tubing length (l = 25, 33, 50, 100 m)
(from Arsonnet et al. 2013)

Several “pressure/flow” points are measured at different flow rates after stabiliza-
tion of the hydraulic circuit (Fig. 3). These points are then used to characterize the
pressure losses by calculating the coefficients of their representative function. This
function is used to correct the pressure measurements of the liquid circuit as a function
of the instantaneous flow rate.
This solution, which has been developed exclusively by APAGEO for GéoPAC®
auto-controlled pressuremeter, has many advantages and is of great interest. In fact, it is
integrated into the test and runs seamlessly at start-up without affecting the test progress
itself. The pressure loss covers the entire liquid circuit. The accuracy of pressures level
of soil that represent a certain creep is significantly improved in the case for tests
between the beginning of the creep and the end of the failure, which increase in
repeatability and accuracy. The oscillatory phenomena are completely attenuated and
controlled and the automatic regulation becomes more efficient and accurate.
From the report of CETIM (2010), the flow rate used in the design of GéoPAC® is
270 cm3/min. It is calculated from theses parameters: engine torque, engine reduction
torque, piston travel speed and surface area, etc. Figure 4 shows the changes of
pressure loss DP as a function of tubing length. The values of DP range from 0.07 to
0.26 bars.
The stiffness of auto-controlled pressuremeter is defined as the ratio of the applied
pressure to the corresponding injected volume. The injected volume is important at low
pressures (several tens of cm3); whereas, for high pressure, the injected volume is very
low (less than one cm3). An algorithm has been developed to control the injected
volume at any instant and to compensate the variation in pressure.
The gain (defined as the ratio of the voltage to the pressure) is adapted so that the
process of the control increases with the decrease of the errors in order to command the
 Auto-Controlled Ménard Pressuremeter 259

0,30

0,25

ΔP(MPa)
0,20

0,15

0,10

0,05

0,00
0 20 40 60 80 100 120
length of tubing (m)

Fig. 4. Pressure loss DP as a function of tubing length

rotating the motor even for small deviations. An adaptive position control algorithm is
also involved. A further synchronization is also implemented to maintain a constant
pressure difference between the gas and the liquid pressures.
The inaccuracy obtained by the calibration is very important in manual pres-
suremeters, but is insignificant when using the GéoPAC®. In the auto-controlled
pressuremeter, the difference between the pressure loss recorded for 10 m and 50 m of
tubing is not important, is equal to 0.20 MPa for a volume exceeding 70 cm3, and
corresponds to the value obtained from the equation shown in Fig. 3 that relates flow to
pressure loss (0.187 MPa). The pressure loss is double this value in the manual
pressuremeter (Fig. 5).

800 Manual pressuremeter for l=10ml

Geopac pressuremeter for l=10ml
700 Manual pressuremeter for l=50ml
Geopac pressuremeter for l=50ml
600
ΔV(cm 3)

500

400
ΔP=0,187 MPa

300

200

100

0
0,00 0,05 0,10 0,15 0,20 0,25
Pressure loss Δ P(MPa)

Fig. 5. Volume variation as a function of pressure loss

260 W. Frikha and S. Varaksin

2.3 Test Performed on Artificial Ground

The first set of tests were performed by the auto-controlled and manual pressuremeters
on artificial grounds to eliminate errors due to the heterogeneity of soil and the effect of
boring and installation of the probe. The artificial grounds were high density moulded
sleeves of polyurethane elastomer that were 63 mm
155 mm
500 mm (Fig. 6).
Two types of artificial ground were tested (70 and 90 Shore A) using tubing lengths of
10 and 50 m. According to the indentation hardness of materials (like polymers,
elastomers, and rubbers) by using the durometer (ASTM D2240 scales), the 70 Shore A
is equivalent to 20 bars of resistance for a probe that has a diameter of 60 mm reach a
volume of 600 cm3. The 90 Shore A is equivalent to 30 bars of resistance. Table 1
summarizes the details of all pressuremeter tests that were performed on the artificial
grounds.

Fig. 6. Photograph of an artificial ground that was used for testing

Table 1. Performed test on artificial ground

Tests Ground Length Pressuremeter EM Pf PLM
resistance for of type (MPa) (MPa) (MPa)
600 cm3 of tubing
volume (bars) (m)
Press_AG_20b_10 m 20b 10 Manual type G 11.6 0.89 1.99
Press_AG_20b_50 m 20 50 Manual type G 13 0.65 1.86
Press_AG_30b_10 m 30 10 Manual type G 17.6 1.87 3.2
Press_AG_30b_50 m 30 50 Manual type G 18.9 1.74 3.5
Geopac_AG_20b_10 m 20 10 Auto-controlled 11.6 0.86 2.0
Geopac_AG_20b_50 m 20 50 Auto-controlled 12.1 0.85 2.06
Geopac_AG_30b_10 m 30 10 Auto-controlled 17 1.86 3.08
Geopac_AG_30b_50 m 30 50 Auto-controlled 21 1.75 3.4

Figures 7(a) and (b) show volume variations as a function of corrected limit
pressure obtained from pressuremeter tests performed respectively on artificial grounds
 Auto-Controlled Ménard Pressuremeter 261

of 20 and 30 bars (70 and 90 shore A). Both tests show comparable results and the
pseudo-elastic and plastic behaviours of the artificial grounds. All pressuremeter tests
were performed using at least pressure increments. The volume variation measured
with respect to time was recorded for each pressure increment at 15, 30 and 60 s after
the application of pressure. The first levels (six to seven points) were derived with
pressure increments of 0.05 MPa. For the final points, the pressure increment used was
0.1 MPa. The tests were completed when volume reached 600 cm3.

700 Manual pressuremeter for l=10m

Auto-controlled pressuremeter for l=10m
Manual pressuremeter for l=50m
600
Auto-controlled pressuremeter for l=50m

500
ΔV(cm 3)

400

300

200

100
Artificial ground 20 bars
0
0,00 0,25 0,50 0,75 1,00 1,25 1,50 1,75 2,00
plMPa)

(a)

700 Manual pressuremeter for l=10m

Auto-controlled pressuremeter for l=10m
600 Manual pressuremeter for l=50m
Auto-controlled pressuremeter for l=50m
500
ΔV(cm 3)

400

300

200

100
Artificial ground 30 bars
0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5
pl(MPa)

(b)
Fig. 7. Volume variation as a function of corrected limit pressure from pressuremeter tests
performed in artificial grounds of (a) 20 bars and (b) 30 bars
262 W. Frikha and S. Varaksin

The limit pressure (2 MPa) obtained from either the auto-controlled or manual
pressuremeter for 10 m of tubing corresponds to the real value of the limit pressure of
the 20-bar artificial ground. The limit pressure value that was derived varied for the
manual and auto-controlled pressuremeters respectively from 1.86 to 2.06 bars when
tubing length was 50 m. For the 30-bars artificial ground, the limit pressure derived
from the automatic pressuremeter shows more precision than what was measured by
the manual device. The inaccuracy is not very significant and does not exceed 10%.
The auto-controlled and manual pressuremeters show comparable pressuremeter
moduli for 10 m of tubing lengths for the two types of the artificial grounds. The values
are slightly different when the length of tubing is longer (50 m), but the difference does
not exceed 11%.

3 In Situ Pressuremeter Test

The second set of tests consisted a series of in situ auto-controlled and manual pres-
suremeter tests in Site located at Salvador Allende Road, Magny Les Hameaux, France.
For both devices, the probes were initially calibrated (Fig. 8). The pressure and volume
readings are then adjusted to compensate for the head of water (pressure loss) in the
measuring cell tubing and the inertia of the probe (membrane and probe cover). The
calibrations are required to determine the strength of the probe covers and membrane
assemblies as functions of the expansion of the probe (volume variation) and the
parasitic increase in volume variation in the entire apparatus as a function of the applied
pressure.

Fig. 8. In situ pressuremeter testing

 Auto-Controlled Ménard Pressuremeter 263

The type of membrane and the number of cycles can affect the result. Authors’
experience shows that the strength of the probe (membrane) decreases as a function of
the number of tests that are carried. The recorded volume-pressure curves allowed the
strength of the membranes to be determined for injection volumes of 700 cm3 using
60 mm diameter probes. The pressure increment used was 25 kPa. The calibration
curve obtained at the end of the test was the volume variation, reading at 60 s, as a
function of the applied pressures.
The pre-expansion test of the probe is carried out in situ each time a new probe is
used. The differential valve is adjusted to 0 m depth.
The expansion coefficient “a” is determined using a steel calibration tube that is
usually 1.05 m in length and 65 mm in diameter. The test is carried out in 500 kPa
pressure increments to 5 MPa.
The initial volume of the measuring probe cell is (AFNOR, NF P 94-110):
 
pdi
Vs ¼ ls  Vc ð1Þ
4

Where:
ls = the length of the membrane (210 mm),
di = inside diameter of the calibration tube,
Vc = the ordinate at the origin of the straight line, which gives the best fit with the
second portion of the expansion curve (Fig. 9).

120

100 Vc a

80
Volume(cm3)

60

40

20

0
0 1 2 3 4 5
Pressure (MPa)

Fig. 9. Expansion curve of a pressuremeter probe in a calibration tube

264 W. Frikha and S. Varaksin

The dilation coefficient is then:

DV
a¼ ð2Þ
Dp

The coefficient a must be less than 6 cm3/MPa (0.6 cm3/bar) for a device that is
equipped with 50 m of tubing. The recorded values of the expansion coefficient for the
auto-controlled and manual pressuremeters were respectively 3.668 cm3/bar and
3.112 cm3/bar. The recorded Vc for the auto-controlled and manual pressuremeters
were respectively to 91.795 cm3 and 83.857 cm3.
Table 2 summarizes the details of the second series of auto-controlled and manual
pressuremeter tests that were performed in situ. The tests are carried to the depth of 4 m
every one meter in pre-bored holes. The holes are drilled using a continuous flight
auger with a diameter of 63 mm. The upper layers of soil consisted of sandy to clayey
silt underlain by plastic clay. All pressuremeter tests were performed in at least 9
pressure increments. The auto-controlled test was performed with more increments to
achieve a higher precision.
The diagrams in Fig. 10 show examples of the software output that was used to
determine the various parameters of the pressuremeter (Geovision®): modulus, creep
pressure and limit pressure from two tests carried out at 1 m using the auto-controlled
pressuremeter. Both modulus and pressure derived from the auto-controlled pres-
suremeter, for the sandy silt, were slightly higher than those obtained from the manual
pressuremeter.

Table 2. Detail of performed in situ tests

Tests Depth Soil No of Pressuremeter EM Pf PLM EM/PLM
points type (MPa) (MPa) (MPa)
Press_Situ_1 m 1m Sandy 12 Manual type G 13 0.33 0.785 16.56
Silt
Press_Situ_2 m 2m Sand 9 Manual type G 3.9 0.12 0.291 13.56
clay
Press_Situ_3 m 3m Clay 11 Manual type G 11.2 0.28 0.933 11.98
Press_Situ_4 m 4m Clay 15 Manual type G 34.2 1.24 1.886 18.15
Geopac_ Situ_1 m 1 m Sandy 13 Auto-controlled 16.6 0.43 0.801 20.67
silt
Geopac_ Situ_2 m 2 m Sand 10 Auto-controlled 4.4 0.17 0.334 13.24
clay
Geopac_ Situ_3 m 3 m Clay 11 Auto-controlled 16.5 0.29 0.812 20.29
Geopac_ Situ_4 m 4 m Clay 18 Auto-controlled 27.3 1.44 2.754 9.922

Figure 11 show the results of several different tests. It can be clearly observed that
there are differences between the values that are obtained by the auto-controlled and
manual pressuremeters at the end of the plastic phase or near the limit pressure.
Figure 12 compares the results of the auto-controlled and manual pressuremeters. It
can be observed that all moduli values derived from auto-controlled pressuremeter are
 Auto-Controlled Ménard Pressuremeter 265

Fig. 10. Example of determination of soil proprieties using Geovision® software

266 W. Frikha and S. Varaksin

800 800

700 Menard pressuremeter Menard pressuremeter

700
Geopac pressuremeter Geopac pressuremeter
600 600

500 500
ΔV(cm )

ΔV(cm )
3

3
400 400

300 300

200 200

100 100
Depth D=1m Depth D=2m
0 0
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 0,0 0,1 0,2 0,3 0,4
P(MPa) P(MPa)

800 800
Menard pressuremeter 700 Menard pressuremeter
700 Geopac pressuremeter
Geopac pressuremeter
600 600

500
ΔV(cm )
500 3
ΔV(cm3 )

400 400

300
300
200
200
100
100 Depth D=4 m
Depth D=3 m 0
0 0,0 0,5 1,0 1,5 2,0 2,5
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
P(MPa)
P(MPa)

Fig. 11. Results of the pressuremeter tests (expansion curve)

pM from Manual pressuremeter

0 pf from Manual pressuremeter
0 pM from Auto-controlled pressuremeter
Manual pressuremeter pf from Auto-controlled pressuremeter
1 Auto-controlled pressuremeter
1
Depth (m)
Depth (m)

2 2

3 3

4 4

5 5
1 0,1 p(MPa) 1
EM(MPa) 10

Fig. 12. Modulus, creep pressure, limit pressure versus depth (auto-controlled and manual
pressuremeters)

slightly higher than those obtained from the manual pressuremeter in the first three
meters. The creep pressures derived from the manual pressuremeter are consistently
lower than those obtained from the auto-controlled one. The limit pressures are com-
parable in the first three meters; however, a small difference is noticeable at 4 m depth.
Figure 13 shows the software outputs used for the analyses of data (Geovision®)
with pictures of extracted specimens and equipment used.
 Auto-Controlled Ménard Pressuremeter 267

Fig. 13. Output of software used for analysis of data (Geovision®)

4 Discussion and Conclusion

The main purpose of the present paper is the description of a new apparatus that is able
to automatically perform a pressuremeter test. Such an automatic apparatus enables the
geotechnical engineer to obtain repeatable measurements, and reduces inaccuracies that
might be induced by the operator, uncertainties, and loss of head. Both in situ tests and
tests performed on artificial ground demonstrate the feasibly of the auto-controlled test
and the possibility to obtain good results. The performance tests, the analysis of
operations and the approval of the model (electrical and mechanical) were conducted
by an independent organisation CETIM (Centre Technique des Industries Mécaniques,
France) in 2010. The inaccuracy obtained by the calibration is substantial when the
manual pressuremeter is used; however, this value is insignificant when using the novel
generation pressuremeter.
The solution of self-correction of pressure loss, developed exclusively for
GéoPAC® auto-controlled pressuremeter, has many advantages and is of great interest
especially in deep soils when using long tubing. In fact it is integrated into the test and
runs at start-up without affecting the test progress. The accuracy of pressures of soil that
represent a certain creep is significantly improved. The accuracy of pressures level of
soil that represent a certain creep is significantly improved in the case for tests between
the beginning of the creep and the end of the failure, which increase in repeatability and
accuracy. The oscillatory phenomena are completely attenuated and controlled and the
automatic regulation becomes more efficient and accurate.
The auto-controlled pressuremeter also makes it possible to carry out true cyclic
tests, which can be performed by varying the amplitude or the frequency of the test in
either cases of pressure or volume control. The cyclic(s) loading(s) can be easily
programmed and the record is also possible using the central unit.

Acknowledgments.. The authors gratefully appreciate the help provided by Damien Brechot,
Vincent Brousset, Lionel Daré and Jean Pierre Arsonnet from APAGEO Company during the
carried out experimental work and Hamidi Barak for the review of the present paper.
268 W. Frikha and S. Varaksin

References
AFNOR, NF P 94-110: Sols, Reconnaissances et essais-Essai pressiométrique Ménard. French
Standard, Edited by afnor, Paris-La défense, Publisher Location, July 1991 (1991)
AFNOR, NF P 94-110-1: Sols, Reconnaissances et essais-Essai pressiométrique Ménard, Partie
1: Essai sans cycle. French Standard, Edited by afnor, Paris-La défense, Publisher Location,
January 2000 (2000)
AFNOR, XP P 94-110-2: Sols, Reconnaissances et essais-Essai pressiométrique Ménard, Partie
1: Essai avec cycle. French Standard, Edited by afnor, Paris-La défense, Publisher Location,
December 1999 (1999)
ASTM Standards D2240-15: Standard Test Method for Rubber Property - Durometer Hardness.
Published August 2015. Originally approved in 1964. Last previous edition approved in
August 2015 as D2240 – 15. Developed by Subcommittee: D11.10 (2015)
ASTM Standards D 4719-07: Standard Test Methods for Prebored Pressuremeter Testing in
Soils. Current edition approved Feb. 15, 2007. Published April 2007. Originally approved in
1987. Under the jurisdiction of ASTM Committee D18 on Soil and Rock and the direct
responsibility of Subcommittee D18.02 on Sampling and Related Field Testing for Soil
Evaluations (2007)
Arsonnet, G., Baud, J.-P., Gambin, M., Youssef, W.: Le GéoPAC®, un contrôleur pression
volume automatisé pour les essais pressiométriques de qualité. The Geopac®, an Automated
Control Unit for Quality Ménard PMTs. In: Proceedings of the 18th International Conference
on Soil Mechanics and Geotechnical Engineering, Paris 2013. Parallel session ISP 6 (2013)
Cassan, M.: Les essais pressiométriques et leurs applications en France. Rappels Historiques et
état des connaissances. The pressuremeter test and their applications in France. Historical
Summary and Sate of the art. 50 ans de pressiomètres. Vol. 1 Gambin, Magnan et Mestat (ed.)
Presses de l’ENPC/LCPC, Paris (2005)
Rapport of CETIM: Conception d’un contrôleur pression/volume automatique. CETIM, Centre
Technique des Industries Mécaniques, France, 19 Février 2010 (2010)
ISO 22476-4: Geotechnical investigation and testing—Field testing—Part 4: Ménard pres-
suremeter test. Reconnaissance et essais géotechniques – Essais en place – Partie 4: Essai au
pressiomètre Ménard. International Standard. First edition 2012-12-01. Published in
Switzerland (2012)
 Assessment of Relationship Between Static
and Dynamic Load Using Regression Analysis
and Artificial Neural Network Model

Ahmed H. Abulkareem(&)

Civil Engineering Department, Engineering College,

University of Anbar, Ramadi, Iraq
[email protected]

Abstract. The Light Falling Weight Deflectometer (LFWD) load test has been
developed to directly estimate the in-situ elastic modulus of near surface profiles
as foundation, and subgrade layers is presented in this paper. For this purpose,
field tests were conducted on selected sections from landfill project within
Anbar province. In addition, forty test sections were constructed and tested at
the Civil Engineering Department- University of Anbar. All sections were tested
using the ZFG 3000 model - LFWD in companion with the Plate Load Test
(PLT) that were used as reference measurements. Regression analyses were
conducted to determine the best correlations between the elastic modulus
obtained from LFWD and PLT tests., Evd. ANN model is used to calculate
dynamic deformation modulus, Evd and comparing with the regression statistical
model. The results indicate that ANN model have the capability of predicting
dynamic deformation modulus, Evd with a high degree of accuracy. Good
correlations were obtained, which demonstrated that the LFWD can be reliably
used to predict the modules obtained from plate load test and degree of com-
paction values, and hence can be used to evaluate the stiffness/strength
parameters of shallow subgrade layers.

Keywords: Plate load test
LFWD
Dynamic modulus
ANN
Regression

analysis

1 Introduction

The light falling weight Deflectometer (LFWD) is used in landfill project in Iraq. The
German device (ZFG 3000) from Zorn is a compaction control device used in this
research. It is able to measure the dynamic load bearing capacity of subgrades, subsoils,
embankment layers and backfills.
Extensive application of these apparatus still has not been achieved since the
dynamic modulus in not accepted in the quality assessment and quality control process
of embankments and subgrade layers. Only marginal use of these devices can be
noticed, mainly on areas of low importance (e.g. road shoulders) or trenches where
performing a static plate load test could be complicated (Zoltán 2008).

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_21
270 A.H. Abulkareem

For being able to use these dynamic devices on subgrade and embankment layers
for landfill project, the research for converting the measured dynamic modulus into
static modulus has been initiated.
The main objective was to determine the correlation between static and dynamic
modules. Since direct conversion formulas are not frequently used in practice, intro-
duction of an easy to use table with the required static and dynamic target values has
been aimed. This is achieved by conducting field tests on constructed layers using the
investigated device LFWD along with standard in-situ test device Plate Load Test
(PLT). Otherwise new quality assessment based on dynamic modulus might be able to
substitute the exclusive usage of the slow and complicated static plate load test in the
near future. With the help of these results, new dynamic design methods can be worked
out and applied.

2 Light Falling Weight Deflectometer (LFWD)

Light Falling Weight Deflectometer (LFWD) is a portable falling weight deflectometer

that has been developed in Germany as an alternative in-situ testing device to the plate
load test. Different types of LFWD exist in the market. All types exhibit many simi-
larities in their mechanics of operation although there are many differences in design
and mode of operation, which lead to variations in the measured results. Generally, the
LFWD consist of a loading device that produces a defined load pulse, a loading plate,
and one center geophone sensor (electric deflection data device) to measure the center
surface deflection (Garcia and Thompson 2004).
The ZFG 3000 LFWD from Zorn Instruments was used in this study Fig. 1 is a
compaction control device, according to ASTM E2835-11 and ASTM E2583-07
(2011). Dynamic modulus of deformation, Evd is the most accurate and independent
means for judging deformation (stiffness) and, thus, a material’s level of compaction
(www.ticservicegroup.com.au, 2013).
A 10 kg falling weight is dropped onto a 300 mm diameter plate from a height of
72 cm through guide rod; the vertical displacement of the plate (dc) is recorded by an

Fig. 1. Light Falling Weight Deflectometer (LFWD) (TIC Service Group 2013).
 Assessment of Relationship Between Static and Dynamic Load 271

accelerometer built in a steel case on the top of the plate. The drop weight, drop height
and plate diameter are constants. The plate coefficient (c) and the Poisson’s ratio (µ) are
also set constant, therefore the dynamic subgrade modulus, (Evd) is calculated by a
simplified Boussinesq equation (Zoltán 2008):

Evd ¼ 22:5=dc ð1Þ

To estimate dynamic modulus of deformation, Evd for each test in this study, eight
position (nearly 45° apart) surrounding the location of the static plate load test were
selected and dynamic plate load test were performed. Each test was repeated three
times, in each time three preconsolidation tests (pre-compaction to remove any bedding
errors, and are ignored) were conducted as shown in Fig. 1.

3 Static Plate Load Test (PLT)

The Static Plate Load Test (PLT) has been a useful site investigation tool for many
years and has been used for proof testing of pavement structure layers in many
countries. Currently it is used for both rigid and flexible pavements. The test was
conducted by the procedure recommended by ASTM D1196-93as shown in Fig. 2. The
test consists of a circular plate (450 mm diameter) that is in close contact with the layer
to be tested and measuring deflections under load increments. The load increments
were applied via a hydraulic jack with a suitable load capacity. The load was applied in
increments up to a final value of externally applied stress of 700 kPa. The corre-
sponding settlement was monitored and recorded, by using three suitable dial gauges
(120° apart), for each increment until the settlement has ceased.

Fig. 2. Static Plate Load Test (PLT).

Plate loading tests can be used to estimate the modulus of subgrade reaction (k).
Determination of the modulus of subgrade reaction is made in the field on the selected
subgrade soil at its natural moisture content. This test is conducted by subjecting the
272 A.H. Abulkareem

subgrade to a known stress at a predetermined rate of speed using a loading system, and
recording the resulting settlement. The modulus of subgrade reaction, k, can be cal-
culated using the following relation (Yoder and Witczak 1975):

k ¼ p=d ð2Þ

Where
P = unit load on plate (kN)
d = Settlement of the plate (mm)
The value of Young’s modulus was obtained from the well-known relationship that
correlates the young’s modulus to the modulus of subgrade reaction (Bowels 1997):


Es ¼ kB 1  l2 ð3Þ

Where B in this case represents the diameter of the testing plate and µ is the
Poisson’s ratio and its value can be 0.5 for the assumption of flexible base plate (Yoder
and Witczak 1975).

4 Existing Relationships

The Institute for Transport Sciences in Hungarian (KTI) launched a research program
in 1995 aiming to convert the dynamic modulus obtained by that device (Evd) into the
well-known static plate load test modulus (E2) obtained by conventional measurements
(research report, ÁKMI Kht 1995). After collecting 64 measurement results performed
on different subgrade and subsoil materials, a general conversion formula was sug-
gested as following:

Evd ¼ 0:52
E2 þ 9:1 ð4Þ

Several correlation results between E2 and Evd are available in the international
literature. The most relevant results are summarized in Fig. 3.
Figure 3 shows that the value of the static plate load test modulus clearly exceeds at
least two times that of the Evd modulus. Some of the results show even higher ratios.
Only two publications give a ratio less than two, but both of them are based on
modulus values measured only at few points and within small intervals (Zoltán 2008).
Nazzal (2003) present a correlation study between the PLT and the LFWD on
cement treated soils, lime treated soils, unstabilizsed fine-grained soils and granular
soils. The relationship for modulus (surface modulus for LFWD) thus developed is as
follows:

EPLT ¼ 20:9 þ 0:69ðELFWD Þ ðR2 ¼ 0:94; for 2:5 MPa\ ELFWD \865 MPa ð5Þ

These regression models are illustrated in Fig. 4.

 Assessment of Relationship Between Static and Dynamic Load 273

Fig. 3. Correlation results between E2 and Evd (Zoltán 2008).

Fig. 4. Relationship between EPLT(R2) and ELFWD (Nazzal 2003).

274 A.H. Abulkareem

5 Analysis of Results
5.1 Linear Regression Model
In this study, collecting 40 measurement results for plate load test (PLT) performed on
subgrade crushed limestone material in excess of (38 mm) in diameter by the field
laboratory for civil engineering department at University of Anbar. After the division of
project area as strips and establish the location of each point by mark, it has performed
LFWD testing as shown in Fig. 5.

Fig. 5. Profile of the project area.

In order to characterize the variation of Evd used as an independent value,

descriptive statistics such as; minimum, maximum, mean, mode, median, variance,
standard deviation, skewness and kurtosis etc. were calculated using the SPSS Version
20 (2012) package. Table 1 shows that the independent statistics value. Depending on
the kind of soil and the degree of compaction, it can possible exceeding the limits of the
lower and upper Evd values as (Nazzal 2003).
In this study, simple regression analysis was performed in the analysis. The rela-
tions between the measured Evd, moduli values with conventional static Es, moduli
values were analyzed employing linear, power, logarithmic and exponential functions.
Statistically significant and strong correlations were found to be linear.
The objective of this regression analysis is to determine the parameters in the least
square error models, which is used to predict the Evd moduli from Es moduli, with their
corresponding coefficient of determination, R2. In the linear regression models, the
dependent variable is assumed to be a linear function of one or more independent
 Assessment of Relationship Between Static and Dynamic Load 275

Table 1. Descriptive statistics for Evd as an independent value.

N 40
Mean 49.87
Std. error of mean 4.7394
Median 46.28
Mode 27.61
Std. deviation 29.97452
Variance 898.472
Skewness 3.957
Std. error of skewness 0.374
Kurtosis 20.714
Std. error of kurtosis 0.733
Range 187.7
Minimum 21.3
Maximum 209
Sum 1994.91

variables plus an error introduced to account for all other factors, a typical form of a
regression linear model is as follow:

Yi ¼ bo þ b1 xi1 þ . . .. . .bk xik ð6Þ

Where Yi is the dependent variable, and xi1, …, xik are the independent or
explanatory variables, and bo is the disturbance or error term (SPSS Inc. 2009).
The coefficient of determination, R2, represents the proportion of variation in the
dependant variable that is accounted by the regression model and has values from 0 to
1. If it is equal to one, the entire observed points lie on the suggested least square line,
which means a perfect correlation exists. Significance level is the result of the statistic
test with null hypothesis b1 = 0; it is expressed in percent. The greater the significance
level the more supportive the model to alternative hypothesis (b1 6¼ 0), which indicates
that a relation does exist between the dependent and independent variable. Finally, the
standard error is the square root of the mean square errors (MSE).
After collecting forty measurement results performed on subgrade layer, a general
conversion formula was suggested as follows:

Evd ¼ 7:384 þ 0:527 Es ð7Þ

This formula can be used to convert the measured Evd dynamic moduli values into
conventional static Es moduli values. The coefficient of correlation gives a value of
R2 = 0.801 and standard error = 13.56 for 40 sites, which seems to be acceptable in
geotechnical testing. This regression model is illustrated in Fig. 6. It can be seen that
there are some values of Evd and Es outlier range caused by the low correlation
coefficient (R2).
276 A.H. Abulkareem

0 40 80 120 160 200 240 280 320 360

240 240

200 200

160 160
Evd(N/mm )
2

120 120

80 80

Evd=7.384+0.527Es
2
40 R =0.801 40
No. of Data=40

0 0
0 40 80 120 160 200 240 280 320 360
2
Es(N/mm )

Fig. 6. Relationship between Evd and Es.

5.2 Multiple Regression Model

In this study, 13 test results were selected from 40 sites. These sites have been tested by
PLT and LFWD and have conducted field density test by sand cone method according
to ASTM D1556-00. The modified proctor compaction test according to ASTM
D1557-12 also showed that the maximum dry unit weight and the optimum moisture
content for crushed limestone material in excess of (38 mm) in diameter were
17.6 kN/m3 and 5.6%, respectively. Summary of results for this section is presented in
Table 2.

Table 2. Summary of test results for crushed limestone layer.

Strips Evd MPa Es MPa MC (field) % cdry (field) kN/m3 Dpr, Modified %
6–2 37.08 35.21 7.59 16.99 96.56
8–1 59.63 73.93 13.57 17.27 98.12
9–5 21.3 34.01 16.2 16.25 92.36
10–8 44.2 63.84 14.08 16.54 93.96
9–1 68.8 127.90 15.6 16.42 93.30
10–3 24.53 50.30 15.7 16.23 92.20
10–4 46.33 41.91 13.08 16.72 94.99
10–5 26.55 43.05 13.8 16.13 91.64
10–6 52.63 79.56 14.08 16.89 95.95
10–7 47.53 123.00 13.8 16.74 95.12
9–2 64 101.57 16.11 16.22 92.18
9–6 24.6 59.97 15.32 16.23 92.21
9–7 86.2 143.12 16.2 16.92 96.14
 Assessment of Relationship Between Static and Dynamic Load 277

Multiple regression, a time-honored technique going back to Pearson’s 1908 use of

it, is employed to account for (predict) the variance in an interval dependent, based on
linear combinations of interval, dichotomous, or dummy independent variables. The
general purpose of multiple regression is to learn more about the relationship between
several independent or predictor variables and a dependent or criterion variable. The
typical form of the multiple regression model is as follow (SPSS Inc. 2009):

y ¼ b1 x1 þ b2 x2 þ . . .: þ bn xn þ c ð8Þ

Where b1, b2, …, bn are the regression coefficients, representing the amount the
dependent variable y changes when the corresponding independent changes 1 unit. c is
a constant, where the regression line intercepts the y axis, representing the amount the
dependent y will be when all the independent variables are 0. The standardized versions
of the b coefficients are the beta weights, and the ratio of the beta coefficients is the
ratio of the relative predictive power of the independent variables. The major con-
ceptual limitation of all regression techniques is that one can only ascertain relation-
ships, but never be sure about underlying causal mechanism (SPSS Inc. 2009).

Table 3. Model summaries of multiple regressions for prediction of Evd.

Independent variables Coefficient Std. error t-Value Sig. level
Constant −345.76 193.484 −1.787 0.108
Es, MPa 0.375 0.1 3.733 0.005
MC, % 0.754 1.855 0.406 0.694
Dpr, % 3.75 1.902 1.971 0.08

Multiple regression analysis was carried out to correlate the measured Evd to three
parameters, namely, static modulus (Es), moisture content (MC) and degree of com-
paction (Dpr) (Table 3). Multiple regression model to predict Evd is given:


Evd ¼ 0:375ðEs Þ þ 0:754ðMC%Þ þ 3:75 Dpr % 345:76 ð9Þ

The coefficient of correlation between the measured and predicted values is a good
indicator to check the prediction performance of the model. Figure 7 shows the rela-
tionships between measured and predicted values obtained from the Multiple regres-
sion model for Evd, which good correlation coefficient with R2 = 0.804.

5.3 Target Values for Dynamic Models

Direct conversion between dynamic, static models and degree of compaction is not
frequently used in practice. Generally target values are given for different embankment
and subgrade layers, more often depending on the required degree of compaction of the
tested layer. Evd modulus target values are fixed in Germany, and some other countries.
278 A.H. Abulkareem

0 20 40 60 80 100
100 100

80 80
Evd, Predicted (N/mm )
2

60 60

40 40

20 20
No. of Data=13
2
R =0.804
0 0
0 20 40 60 80 100
2
Evd, Measured (N/mm )

Fig. 7. Relationship of predicted and measured values of Evd for Multiple regression model.

Based on the results above, a table of target values can be introduced. Different Es
and degree of compaction values are given for required Evd values in Table 4. It can be
seen that Evd values reduce in access 10% compared with Es due to kind of soil layers
and degree of compaction. Interpolation between given values is acceptable (Zoltán
2008).

Table 4. Target values for crushed limestone subgrade layers.

Es MPa Evd MPa Dpr % Modified Es/Evd ratio
Crushed limestone subgrade layers
180 100  100 2
150 85  99 2
120 70  98 2
100 60  98 2
80 50  97 2
60 35  95 2
45 30  95 2
20 15  94 1.5

5.4 Artificial Neural Network (ANN) Modeling of Overview

When the materials are natural, there is always an uncertainty sourced from the nature
of the materials. This could be the main reason why soft computing methodologies
such as artificial neural networks, Fuzzy systems, genetic algorithms have been
 Assessment of Relationship Between Static and Dynamic Load 279

developed in recent years and adopted but used for some other application. These
techniques attract more and more attention in several research fields because they
tolerate a wide range of uncertainty (Jin and Jiang 1999). Multilayer feed-forward
neural network model is the most widely used network for its efficient generalization
capabilities. Figure 7 presents typical multi-layer feed-forward neural networks. This
type of neural network consists of an input layer, one or more hidden layer(s) and an
output layer. Layers are fully connected by arrows, and comprise a number of pro-
cessing units, the so-called nodes or neurons. The strength of connections between
neurons is represented by numerical values called weights. Each neuron has an acti-
vation value that is a function of the sum of inputs received from other neurons through
the weighted connections (Demuth et al. 2006). The optimum number of hidden layers
and the number of neurons in each hidden layer is problem specific. Therefore, trial and
error should be carried out to choose an adequate number of hidden layers and the
number of neurons in each hidden layer.
Back propagation is the most successful and widely used in neural network
applications. In this method, the input is propagated from the input layer through the
hidden layers to the output layer. The network input is connected to every neuron in the
first hidden layer while each network output is connected to each neuron in the last
hidden layer. In this case, this would call full connection ANN. The network weights
were originally set to random values and new values of the network parameters
(weights) are computed during the network training phase. The neurons output are
calculated using (Abdul-Razzak and Yousif, 2007):
X
Oi ¼ Fð I
j j
 Wij þ bi Þ ð10Þ

Where
Oi = the output of the neuron i, Ij are the input of j neurons of the previous layer,
Wij = the neuron weights, bi is the bias for the modeling, and
F = the activation function.
The activation function is the portion of the neural network where all the computing
is performed. The activation function maps the input domain (infinite) to an output
domain (finite). The range to which most activation functions map their output is either
in the interval [0, 1] or the interval [−1, 1]. The most common activation functions
belong to five families as follows: (1) linear activation function; (2) step activation
function; (3) ramp activation function; (4) sigmoid activation function; and (5) Gaus-
sian activation function. The network error is then back propagated from the output
layer to the input layer in which the connection weights are adjusted. This process is
repeated until the error is minimized to a preference level. The error incurred during the
learning can be expressed as Mean Squared Error and is calculate using (Abdul Razzak
and Yousif 2007):

1 Xn Xm  2
MSE ¼ t ij  y ij ð11Þ
nm i¼1 j¼1

Where t is the target value, y is the output value.

280 A.H. Abulkareem

5.5 ANN Model for Prediction of Evd

The use of ANN provides an alternative way to estimate dynamic modulus of defor-
mation, Evd. In this work a multi-layered feed-forward neural network with a
back-propagation algorithm was adopted. (MATLAB 7.1 2005) software was used in
neural network analyses having a three-layer feed-forward network (Alvarez and
Babuska 1999). Forty cases of actual measured were extracted from experimental tests
used in this study. The databases is randomly divided into three sets such as; training
(70% of all data), test (15% of all data), and verification (15% of all data). The training
set is used to adjust the connection weights, whereas the testing set is used to check the
performance of the model at various stages of training and to determine when to stop
training to avoid over-fitting. The validation set is used to estimate the performance of
the trained network in the deployed environment (Shahin et al. 2008). A typical
structure of artificial neural networks consists of many processing elements that are
arranged in layers: an input layer, an output layer, and one or more layers in between,
called intermediate or hidden layers (Fig. 8). Each processing element in a specific
layer is interconnected to all the processing elements in the next layer via weighted
connections. The scalar weights determine the strength of the connection between
interconnected neurons. A zero weight refers to no connection between two neurons
and a negative weight refers to a prohibitive relationship (Shahin et al. 2008). The
model has one input parameter and one output parameter. The model has two hidden
layers with nine nodes each (MATLAB software uses for determining the optimal
number of hidden nodes rather than assuming a fixed number of hidden nodes in
advance), and output layer with one node giving dynamic modulus of deformation, Evd.

Fig. 8. Structure of Multi-layered feed-forward network (Michael 2015)

 Assessment of Relationship Between Static and Dynamic Load 281

By increasing the number of hidden neurons they can typically get a better approxi-
mation and do still better by further increasing the number of hidden neurons. In the
analyses network of learning parameter, momentum parameters and networks training
function, which is an activation (transfer) function for all layers, have typical values of
0.01, 0.9, trainLm (training Levenberg-Marquardt function) and tansig (transfer func-
tion) respectively. As in many other network training methods, models and parameters
were used in order to reach the minimum Mean Square Error (MSE) values.

0 40 80 120 160 200 240

240 240

200 200
Evd,Predicted(N/mm )
2

160 160

120 120

80 80

40 40
No. of Data=40
2
R =0.9862
0 0
0 40 80 120 160 200 240
2
Evd,Measured(N/mm )

Fig. 9. Relationship of predicted and measured values of Evd for ANN Model.

In fact, the coefficient of determination between the measured and predicted Evd
values is a good indicator to check the prediction performance of the model. Figure 9
shows the relationships between measured and predicted values obtained from the
models for Evd, with good coefficient of determinations. In this study, variance account
for VAF (Eq. 12) and root mean square error (RMSE) (Eq. 13), indices were also
calculated to control the performance of the prediction capacity of predictive models
developed in the study as employed by (Alvarez and Babuska 1999):
 
var ðy  y0 Þ
VAF ¼ 1   100 ð12Þ
var ð yÞ
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 XN
RMSE ¼ ð y  y0 Þ 2 ð13Þ
N i¼1

Where y and y’ = the measured and predicted values, respectively.

If the VAF is 100 and RMSE is 0, then the model will be excellent. The obtained
values of VAF and RMSE given in Table 5 indicated a high prediction performance.
282 A.H. Abulkareem

Table 5. RMSE, VAF and R2 values used to predict Evd.

Model Predictive model RMSE VAF% R2
LRM Evd = 7.384 + 0.527 Es 9.25 75.09 0.81
ANN 2.57 98.31 0.98

6 Conclusions

The LFWD proved to be a truly light weight FWD and is highly transportable. It is very
easy to operate and changes from the 10 kg to the 20 kg drop weights or loading plates
(200 mm and 300 mm) are quick and easy to do. All tests are saved automatically on a
chip card, simply insert the card and the data can be transferred to the PC immediately.
The basic information provided by the LFWD clearly proved to be very useful for
construction quality control and assurance purposes.
The objective of this study is to evaluate the potential use of non-destructive testing
device to measure the stiffness/strength parameters of highway materials and
embankment soils during and after construction for landfill project. The field testing
program included conducting tests using the investigated devices, in addition to
standard tests, which included the static Plate Load Test (PLT), field density test by the
sand cone method, and modified proctor compaction test). The field testing performed
on subgrade crushed limestone material in excess of (38 mm) in diameter.
The results of the statistical analysis show that a good correlation do exist between
the device under evaluation LFWD and the standard tests PLT, and degree compaction
depending on standard tests. The relations obtained from statistical analysis, were linear
regression to model and multiple regression for another. All regression models had an
adjusted, correlation coefficient (R2) greater than 0.8.
A multi-layered feed-forward neural network with a back-propagation algorithm
was used to demonstrate the feasibility of ANNs to predict the dynamic modulus of
deformation, Evd. Forty cases of actual field measurements were used for model
development and verification. The predicted Evd obtained by utilising ANNs were
compared with the measured Evd. The results indicate that ANN model have the
capability of predicting Evd with a high degree of accuracy. From VAF, RMSE indi-
cators and correlation coefficient (R2) results, it can be seen that the ANN model is
more accurately than regression analysis to predict Evd as in Table 5.
The result of this study suggests that LFWD can be reliably used to predict the
modules obtained from PLT, and degree of compaction values, and hence can be used
to evaluate the stiffness/strength parameters of subgrade layers.
The new dynamic target values could open up the opportunity to perform the
quality control and assess the bearing strengths of the tested layer, not only by static
plate load test, which proved to be time-consuming and labour intensive, but by
dynamic devices too.
The widespread use of mentioned dynamic devices referred to above, may facilitate
for contractors, laboratories and engineers in the highway and railway construction
industry to perform quick and continuous quality control of embankments, subgrade
and subsoil layers and backfills.
 Assessment of Relationship Between Static and Dynamic Load 283

Acknowledgment. The authors express his sincere thanks to the staff of the field laboratory for
civil engineering department at University of Anbar for help in the conduct the field tests.

References
Zoltán, T.: Conversion between static and dynamic load bearing capacity moduli and
introduction of dynamic target values. Periodica Polytech. Civil Eng. 52(2), 97–102 (2008)
Garcia, G., Thompson, M.R.: Subgrade strength/stiffness evaluation. University of Illinois, Dept
of Civil and Environmental Engineering, Technical Note, 1215NCEL, MC-250, Urbana, IL
61801 (2004)
TIC Service Group: Test Equipment for Geotechnics and Pavements (2013). www.
ticservicegroup.com.au
Yoder, E.J., Witczak, M.: Principles of Pavement Design. Wiley, New York (1975)
Bowels, J.E.: Foundation Analysis and Design, 5th edn. The McGraw-Hill Companies,
New York (1997)
Institute for Transport Sciences: Measurement of load bearing capacity with lightweight
deflectometers (Wemex/ ZFG). Research report, ÁKMI Kht (1995)
Nazzal, M.D.: Field evaluation of in-situ test technology for QC/QA during construction of
pavement layers and embankments. M.Sc. thesis in civil engineering, Louisiana State
University and Agricultural and Mechanical College (2003)
SPSS Inc.: PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc. (2009)
Jin, Y., Jiang, J.: Techniques in neural-network based fuzzy system identification and their
application to control of complex systems. In: Leondes, C.T. (ed.) Fuzzy Theory Systems,
pp. 112–128. Techniques and Applications, Academic Press, New York (1999)
Demuth, H.B., Beale, M., Hagan, M.: Neural network toolbox for use with matlab User’s guide.
MathWorks, Incorporated (2006)
Abdul-Razzak, A.A., Yousif, S.T.: Artificial neural network model for predicting nonlinear
response of uniformly loaded fixed plates. Eng. Technol. 25(3), 334–348 (2007)
MATLAB 7.1 Software for technical computing and Model-Based Design. The MathWorks Inc.
(2005)
Alvarez, G.M., Babuska, R.: Fuzzy model for the prediction of unconfined compressive strength
of rock samples. Int. J. Rock Mech. Min. Sci. 36, 339–349 (1999)
Shahin, M.A., Jaksa, M.B., Maier, H.R.: State of the art of artificial neural networks in
geotechnical engineering. Electron. J. Geotech. Eng. 8, 1–26 (2008)
Michael, N.: Neural Networks and deep learning. online book (2015) http://
neuralnetworksanddeeplearning.com/chap4.html
 Scanning Electron Microscope Analysis
of Fly Ash, Quarry Dust Stabilized Soil

P. Indiramma(&) and CH. Sudharani

Department of Civil Engineering, S.V. University College of Engineering,

Tirupati, India
[email protected], [email protected]

Abstract. Soil is a complex material with different minerals present in variant

proportions, since it is formed due to weathering of rocks. The proportion of
mineral governs the behaviour of soils. The soils with Montimorillonite mineral
are more expansive in nature. The structures build with or on these soils are
prone to problems like cracking, foundation failures due to their Swelling and
Shrinking behaviour. Stabilization of expansive soil using admixtures is effec-
tive way of improving the intended soil properties and strength in specific. This
paper focuses on the performance of Fly Ash and Quarry Dust as admixtures to
the selected expansive soil. Laboratory investigations for determining Atterberg
Limits, Compaction Characteristics, Deformation behaviour and Unconfined
Compressive Strength and mineralogical analysis are performed by Scanning
Electron Microscope (SEM) technique. The general behaviour of soil alone and
after admixing with different percentages of Fly Ash and Quarry Dust are
analyzed from laboratory test results and cross checked from Scanning Electron
Microscope studies on the selected expansive soil.

Keywords: Expansive soil
Fly Ash
Quarry Dust
Scanning Electron

Microscope (SEM)
Unconfined Compressive Strength

1 Introduction

Around 20% of total land area in the world are covered by expansive soils. Expansive
soils have low Bearing Capacity, high Swelling and Shrinkage characteristics and high
water susceptibility. Presence of Montimorillonite mineral is responsible for this
behaviour of expansive soils. Structural cracks, premature failure of the pavement
occurs when constructed on or with expansive soils. Stabilization of expansive soils
using different additives such as Fly Ash, Quarry Dust, Lime, Cement and Bitumen are
economical and effective. Fly Ash is an industrial by-product which is generated from
Thermal Power Plants. Fly Ash is classified into class F and class C Fly Ash. Fly Ash is
generally grey in colour, abrasive, mostly alkaline and refractory in nature. Safe dis-
posal of Fly Ash is a major problem and creating severe environmental problems.
Quarry Dust is a mineral solid waste, obtained from crushing of stone aggregates.
Researchers used industrial waste admixtures namely, Quarry Dust, Fly Ash and Steel
Slag etc. as soil stabilizers so that there is a value addition to the industrial wastes and
environmental pollution is also minimized. Many research works are reported in

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_22
 Scanning Electron Microscope Analysis of Fly Ash 285

literature using Fly Ash Phanikumar and Sharma (2004, 2007); Phanikumar (2009) and
Quarry dust Sridharan and Soosan (2005) in their study areas to investigate the vari-
ation of properties of soils to analyze them as beneficial Geotechnical materials.
An attempt has been made by in this investigation to use industrial waste admix-
tures namely, Quarry Dust and Fly Ash as soil stabilizers to the selected expansive soil.
The effect of Fly Ash and Quarry Dust on expansive soils, laboratory tests are con-
ducted for determining Plasticity characteristics, Differential Free Swell Index, Com-
paction Characteristics, and Unconfined Compressive Strength. Soil behaviour mainly
depends on minerals present in soil, so a micro level investigation is carried out using
Scanning Electron Microscope (SEM) technique in order to arrive the mineral com-
position and to understand the mechanism involved in soil stabilization.

2 Materials and Methods Used

2.1 Materials
1. An expansive soil collected from Tirupati Airport site, near Renigunta.
2. Fly ash from Vijayawada Thermal Power Station (VTPS), Vijayawada and
3. Quarry Dust from Tirupati Quarry.

2.2 Admixture Proportions

In the present study two series of tests are conducted. In first series the expansive soil is
mixed with Fly Ash varying from 5% to 25% in increments of 5% by weight of soil
taken. In second series expansive soil is mixed with Quarry Dust varying from 5% to
25% in increments of 5% by weight of soil taken.

2.3 Methods Adopted

The following tests are conducted on selected soil alone and Soil-Fly Ash mixtures and
Soil-Quarry Dust mixtures:
Liquid Limit Test (IS: 2720 (PART V)-1975), Plastic Limit Test (IS: 2720 PART
V-1975), Differential Free Swell Index (DFSI) Test (I.S: 2720 (Part-XL)-1977),
Compaction Test IS: 2720 (PART-7)-1980), Unconfined Compression Test (IS: 2720
(Part-10)-1991) and Scanning Electron Microscope studies (SEM) using FEI Quanta
200 unit (Tables 1, 2 and 3).

3 Analysis of Test Results

The variation of Liquid Limit, Plastic Limit, Plasticity Index, Differential Free Swell
Index, Optimum Moisture Content, Maximum Dry Unit Weight and Strength of soil
admixtures are given in Figs. 1, 2, 3, 4, 5, 6 and 7 in the form of plots and the
quantified values of respective properties are given in Tables 4, 5, 6, 7, 8 and 9.
286 P. Indiramma and CH. Sudharani

Table 1. Properties of tested soil

Sl.No. Property Value
1. Grain Size Distribution
Gravel (%) 2.76
Sand (%) 30.0
Silt and Clay (%) 67.9
2. Atterberg Limits
Liquid Limit (%) 98.0
Plastic Limit (%) 34.0
Plasticity Index (%) 64.0
3. Differential Free Swell Index (%) 377.0
4. Specific Gravity 2.61
5. IS Classification ‘CH’
6. Compaction Characteristics
Maximum Dry Unit Weight (kN/m3) 16.62
Optimum Moisture Content (%) 14.0
7. California Bearing Ratio Value
@ 2.5 mm Penetration 1.83
@ 5.0 mm Penetration 1.71
8. Unconfined Compressive Strength (kN/m2) 118.00

Table 2. Properties of Fly Ash

Sl.No. Property Value
1. Grain Size Distribution
Gravel (%) 0.0
Sand (%) 0.0
Silt (%) 93.0
Clay (%) 7.0
2. Specific Gravity 2.15
3. Compaction Characteristics
Maximum Dry Unit Weight (kN/m3) 12.5
Optimum Moisture Content (%) 21.2
4. Color Grey
5. Free Swell Index (%) 0.0
 Scanning Electron Microscope Analysis of Fly Ash 287

Table 3. Properties of Quarry Dust

Sl. No. Property Value
1 Grain size distribution
Gravel (%) 3.0
Sand (%) 94.0
Silt and Clay (%) 3.0
2 Specific Gravity 2.71
3 Compaction Characteristics
Maximum Dry Unit Weight (kN/m3) 19.1
Optimum Moisture Content (%) 12.6
4 Shear parameters
Angle of Shearing Resistance (deg) 4.1
Cohesion (C) (kN/m2) 0.00

Fig. 1. Variation of Liquid Limit with percentages of admixtures

3.1 Plasticity Characteristics

The variation of Liquid Limit and Plastic Limit with different of percentages of
admixtures as represented in Figs. 1 and 2. It can be observed that as the percentage of
admixtures i.e. Fly ash alone/Quarry dust increases, Liquid Limit and Plastic Limit
decreases. At 25% of admixtures, the maximum percentage of decrease in Liquid Limit
and Plastic Limit can be observed in both the test series. The same trend observed in
variation of Plasticity Index with the addition of soil-admixtures in increasing per-
centages. The same trend can be observed in variation of Plasticity Index with the
addition of soil-admixtures in increasing percentages.
288 P. Indiramma and CH. Sudharani

Fig. 2. Variation of Plastic Limit with percentages of admixtures

Fig. 3. Variation of Plasticity Index with percentages of admixtures

Reduction in Diffused Double Layer thickness and Clay percentage are responsible
for decrease in Liquid Limit with increase the percentage of admixtures. The reduction
in Plastic Limit of soil admixtures is due to filling of voids of flocculated soil thereby
reducing water holding capacity.

3.2 Differential Free Swell Index

The swelling tendency of soils can be directly assessed by conducting Differential Free
Swell Index test. The swelling tendency of selected expansive soil is decreasing with
 Scanning Electron Microscope Analysis of Fly Ash 289

Fig. 4. Variation of Differential Free Swell Index with percentages of admixtures

Fig. 5. Variation of Optimum Moisture Content with percentages of admixtures

addition of admixtures Fly ash or Quarry Dust as shown in Fig. 4. The Differential Free
Swell Index is observed to decrease with increase in percentage of admixtures due to
prevailing of Divalent and Trivalent cations (Ca2+, Al3+, Fe3+ etc.,) increasing floc-
culation of clay particles therby reducing surface area and water affinity of the soil
sample.
290 P. Indiramma and CH. Sudharani

Fig. 6. Variation of Maximum Dry Unit Weight with percentages of admixtures

Fig. 7. Variation of Unconfined Compressive Strength with percentages of admixtures

 Scanning Electron Microscope Analysis of Fly Ash 291

Table 4. Liquid Limit at different percentages of Fly Ash and Quarry Dust
Fly Ash (%) Liquid Limit (%) Quarry Dust (%) Liquid Limit (%)
0 98 0 98
5 83 5 70
10 72 10 65
15 68 15 62
20 62 20 58
25 58 25 54

Table 5. Plastic Limit at different percentages of Fly Ash and Quarry Dust
Fly Ash (%) Plastic Limit (%) Quarry Dust (%) Plastic Limit (%)
0 34 0 34
5 33 5 32
10 31 10 30
15 30 15 28
20 29 20 27
25 27 25 25

Table 6. Plasticity Index at different percentages of Fly Ash and Quarry Dust
Fly Ash (%) Plasticity Index (%) Quarry Dust (%) Plasticity Index (%)
0 64 0 64
5 50 5 38
10 41 10 35
15 38 15 34
20 33 20 31
25 31 25 29

Table 7. Differential Free Swell Index at different percentages of Fly Ash and Quarry Dust
Fly Ash Differential Free Swell Quarry Dust Differential Free Swell
(%) Index (%) (%) Index (%)
0 377 0 377
5 320 5 295
10 295 10 250
15 280 15 217
20 262 20 190
25 250 25 148
292 P. Indiramma and CH. Sudharani

Table 8. Optimum Moisture Content at different percentages of Fly Ash and Quarry Dust
Fly Ash (%) Optimum Moisture Quarry Dust (%) Optimum Moisture
Content (%) Content (%)
0 14 0 14
5 13.5 5 13
10 13 10 12.7
15 12 15 12.4
20 12 20 12
25 11.7 25 11.3

Table 9. Maximum Dry Unit Weight at different percentages of Fly Ash and Quarry Dust
Fly Ash (%) Maximum Dry Unit Quarry Dust (%) Maximum Dry Unit
Weight (kN/m3) Weight (kN/m3)
0 16.62 0 16.62
5 16.69 5 17.00
10 16.91 10 17.20
15 17.12 15 17.29
20 17.18 20 17.50
25 17.23 25 17.63

3.3 Compaction Characterstics

Figures 5 to 6 the Optimum Moisture Content is found to decrease and MDD to
increase with addition of Fly ash or Quarry Dust to the expansive soil. The increasing
attractive pressure of soil particles minimizes dispersion thereby reducing and Opti-
mum Moisture Content increasing Dry Unit Weight

4 Strength Characterstics

Figure 7, depicts the variation of unconfined compressive strength with increasing

percentages of Fly Ash or Quarry Dust to the soil. Maximum Unconfined Compressive
strength is observed at 25% of admixture addition to the soil due to reduction of
cohesion. Soil-Fly Ash strength is more compared to Soil-Quarry Dust strength.

5 Scanning Electron Microscope (SEM) Studies

Scanning Electron Microscope studies are conducted using Quatum 200 unit on soil
alone, soil-Fly Ash and Soil-Quarry Dust admixtures for understanding the mechanism
of behavior of soils with different additives at different percentages. Figures 8, 9 and 10
shows Microscope analysis for soil alone, 10% Fly Ash + 90% Soil and 10% Quarry
Dust + 90% soil respectively. It is observed that structure of the soil has dispersed
 Scanning Electron Microscope Analysis of Fly Ash 293

fabric in natural state as given in Fig. 8, the texture of an soil alone consists of number
of flaky-like particles. Soils with flaky shapes are likely to have low strength.
Flocculated structure is assumed after addition of admixtures. The extent of vari-
ation depends upon the type of admixtures. Figure 9 represents the texture of soil when
Fly Ash is used as a stabilizer. The surfaces seems to be smooth and voids in soils
occupied by Fly Ash. The texture of soil with addition of Quarry Dust is shown in
Fig. 10 where rough surfaces with sharp edges and presence of Quarry Dust particles
on surfaces can be observed. Compared to Fly Ash, Quarry Dust has more voids, so
soil-Fly Ash with Fly Ash mixes gives more strength compared to soil-Quarry Dust
mixtures. These observations strengthen the improvement in Strength, decrease in
Swelling nature and Plasticity Characteristics of Soil-Fly Ash and soil-Quarry Dust
mixtures found from laboratory tests (Table 10).

Fig. 8. Microanalysis report of soil

294 P. Indiramma and CH. Sudharani

6 Conclusions

A series of laboratory tests were conducted to study the effect of Fly Ash and Quarry
Dust on Plasticity Characteristics, Differential Free Swell Index, Compaction Char-
acteristics (MDD, OMC) Compressive Strength on selected expansive soil. Fly Ash
and Quarry Dust were added in different percentages from 0–25% in increments of 5%.
Based on the test results the Plasticity Characteristics indicated by Plasticity Index,
Optimum Moisture Content and Differential Free Swell Index are observed to decrease
whereas Maximum Dry Unit Weight and Strength of soil admixtures increased with
increase in percentage of Fly Ash and Quarry Dust.

Table 10. Unconfined Compressive Strength at different percentages of Fly Ash and Quarry
Dust
Fly Ash (%) Unconfined Compressive Quarry Dust Unconfined Compressive
Strength (kN/m2) (%) Strength (kN/m2)
0 10.6 0 10.6
5 18.1 5 15.0
10 22.5 10 18.6
15 25.0 15 18.9
20 34.6 20 21.2
25 42.2 25 22.1

Fig. 9. Microanalysis report of (Soil + 10% Fly Ash)

 Scanning Electron Microscope Analysis of Fly Ash 295

Fig. 10. Microanalysis report of (Soil + 10% Quarry Dust)

The SEM observations for soil alone depicts flaky structure and addition of Fly Ash
and Quarry Dust the same soil represents Flocculated structure with voids filled by
admixtures resulting in decrement of Plasticity Characteristics, Optimum Moisture
Content and Differential Free Swell Index and increment in Maximum Dry Unit
Weight thereby Strength of Soil-Admixtures compared to soil alone. Fly Ash or Quarry
Dust can be used as effective stabilizing admixtures for expansive soil considered in
this investigation.

References
Sridharan, A., Soosan, T.G.: Utilization of quarry dust to improve the geotechnical properties of
soils in highway construction. Can. Geotech. J. 28(4), 391–400 (2005)
Phanikumar, B.R., Sharma, R.S.: Effect of fly ash on Engg properties of expansive soil.
J. Geotech. Geoenviron. Eng. 130(7), 764–767 (2004)
296 P. Indiramma and CH. Sudharani

Phanikumar, B.R., Sharma, R.S.: Volume change behavior of fly ash-stabilized clays. J. Mater.
Civ. Eng. (ASCE) 19(1) (2007). doi:10.1061/(ASCE)0899-1561(2007)19:1(67)
Phanikumar, B.R.: Effect of lime and fly ash on swell, consolidation and shear strength
characteristics of expansive clays: a comparative study. Geomech. Geoeng. Int. J. 4(2), 175–
181 (2009).
 Analysis of Cylindrical Cavity Expansion
in Modified Cam Clay with Ko Consolidation

Vincenzo Silvestri1(&) and Claudette Tabib2

1
École Polytechnique, Montreal, Canada
[email protected]
2
Montreal, Canada
[email protected]

Abstract. This paper presents explicit expressions for the principal effective
stresses generated around a cylindrical cavity expanded in plane strain and
undrained conditions in Modified Cam Clay. The assumption made in the
present analysis is that Poisson’s ratio v′ remains constant throughout the
shearing process. Theoretical expressions are applied to the simulation of a
cylindrical cavity expansion test in Ko normally consolidated remoulded Boston
Blue Clay modelled as Modified Cam Clay. The results, which are compared to
those obtained by assuming that the shear modulus G0 remains constant, show
that the two approaches are quite similar.

1 Introduction

Chen and Abousleiman (2012) and Silvestri and Abou-Samra (2012) recently obtained
semi-analytical solutions for the plane strain undrained expansion of cylindrical cavi-
ties in Modified Cam Clay. Chen and Abousleiman (2012) employed small strains, but
Silvestri and Abou-Samra (2012) used finite natural strains in both the elastic and
plastic phases of the expansion. The latter authors also considered Almansi strains for
obtaining the limiting radial expansion pressure. In addition, Silvestri and Abou-Samra
(2012) adopted the restrictive assumption that the hardening parameter p0c , which
controls the size of the yield loci, remained constant during shearing. Such simplifying
assumption permitted the determination of explicit expressions for the principal
effective stresses generated in the soil around the cavity, but it led to approximate
responses.
Silvestri and Abou-Samra (2012) also assumed the shear modulus G0 to remain
constant throughout the expansion process, following the approach of Randolph et al.
(1979). From a theoretical point of view, it is preferable to assume a constant shear
modulus G0 , as Zytynski et al. (1978) showed that the use of a constant Poisson’s ratio
m0 would lead to a non-conservative model in the sense that it may not conserve energy
during closed-stress cycles (Yu 2006). However, this effect was not important in the
case treated by Silvestri and Abou-Samra (2012) since there were no unloading-
reloading cycles. As for Chen and Abousleiman (2012), these authors assumed Pois-
son’s ratio m0 to remain constant and obtained semi analytical expressions for the
principal effective stresses.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_23
298 V. Silvestri and C. Tabib

Cao et al. (2001) also obtained an approximate closed-form solution of the

undrained cavity expansion in Modified Cam Clay by combining a large strain theory
in the plastic zone and a small strain theory in the elastic zone. In the present paper, as
the soil is modelled as a non-linear elastic plastic material, elastic and plastic zones are
not treated separately. Thus, the analysis cannot be compared directly with Cao et al.’s
approach.
In the present paper, both the shear modulus G0 and the hardening parameter p0c
vary during undrained shearing, but that Poisson’s ratio m0 remains constant. Indeed,
Gens and Potts (1988) pointed out that a constant shear modulus did not agree well
with experimental measurements and might imply negative values of Poisson’s ratio at
low stresses, which is physical unreasonable (Yu 2006).
As Poisson’s ratio m0 remains constant, such approach allows obtaining explicit
expressions for the principal effective stresses generated in the soil around the
expanding cylindrical cavity. However, both the total radial pressure and the natural
shear strain in the horizontal plane must still be determined numerically. The results
obtained in the present paper are applied first to a well-known benchmark case and are
thereafter compared with those found by assuming that the shear modulus G0 remains
constant during the expansion, but that the hardening parameter varies during shearing.
The results also show that the vertical effective stress, which represents the major
principal stress at the beginning of the expansion, becomes the intermediate principal
stress during the latter stages of the expansion. Similarly, while one of the horizontal
effective stresses (i.e., the radial stress) becomes the major principal stress, the other
horizontal effective stress (i.e., the tangential stress) becomes the minor principal stress
during the expansion. Thus, failure occurs on vertical planes. Such response is different
from the one analysed by Monnet (2007). Indeed, this author found that failure in a
cylindrical cavity expansion could also occur on inclined planes with the vertical and
tangential stresses being the major and minor principal stresses, respectively. Such
situation typically arises in expansion tests carried out at shallow depth.

2 Approach
2.1 Modified Cam Clay Model
The general effective stress invariants used in the Modified Cam Clay model are the
mean effective stress p0 and the deviator stress q, which are defined as:

r0r þ r0h þ r0z

p0 ¼ ð1aÞ
3
and
"
2
2
2 #1=2
r0r  r0h þ r0h  r0z þ r0z  r0r
q¼ ð1bÞ
2
 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 299

Fig. 1. Undrained compression test on normally consolidated clay: (a) p0 : q plane; (b) t : p0
plane (adapted from Wood (2007))

where r0r ; r0h ; r0z ¼ principal radial, tangential, and axial effective stresses. Figure 1a
presents various yield curves of the Modified Cam Clay model. Each curve is described
by the following expression:
300 V. Silvestri and C. Tabib


 1=2
q ¼ p0 M p0c =p0  1 ð2Þ

where M ¼ q=p0 at critical state ¼ 6sin/0 =ð3  sin/0 Þ, with /0 ¼ friction angle and
p0c ¼ hardening parameter. The latter parameter controls the size of the yield locus. An
effective stress path (ESP) followed by a Ko normally consolidated specimen in an
undrained compression test is also reported in Fig. 1a. The ESP is described by the
expression
h

1=K i1=2
q ¼ p0 M 2 þ g2i p0i =p0 M 2 ð3aÞ

where
gi ¼ qi =p0i ¼ 3ð1  Ko Þ=ð1 þ 2Ko Þ;
qi ¼ r0zi  r0ri ¼ r0zi  r0hi ¼ ð1  Ko Þr0zi ;


p0i ¼ r0ri þ r0hi þ r0zi =3 ¼ ð1 þ 2Ko Þr0zi =3;
r0ri ¼ r0hi ; r0zi ¼ initial horizontal and vertical effective stresses, respectively,
Ko ¼ r0ri =r0zi ¼ r0hi =r0zi ¼ in situ coefficient of lateral earth pressure at rest,
K ¼ ðk  jÞ=k;
k ¼ slope of t : ln p0 line in loading (Fig. 1b),
j ¼ slope of t : ln p0 line in unloading (Fig. 1b),
t ¼ 1 þ e ¼ specific volume, and
e ¼ void ratio:
The ESP may also be described by the following equation:
h
1=K i1=2
q ¼ p0 M p0o =p0 1 ð3bÞ

where p0o ¼ value of p0 for q ¼ 0, as shown in Fig. 1a. The ESP of the Ko normally

  
consolidated specimen begins at point C p0i ; qi and ends at point F p0f ; qf on the
critical state line, whose coordinates are p0f ¼ 2K p0o and qf ¼ Mp0f ¼ 2K Mp0o (Wood
pffiffiffi
2007). The undrained shear strength Su is equal to qf / 3. As the expansion occurs
under undrained conditions, the specific volume t remains constant along the ESP, as
shown in Fig. 1b, and each point on the stress path lies on a new yield locus. The test
thus moves across progressive higher yield loci, which are associated with expansion of
the yield locus and with decrease of the mean effective stress p0 for the normally
consolidated specimen.

2.2 Modified Cam Clay Stress-Strain Relationships

Silvestri and Abou-Samra (2012) showed that the incremental elastic-plastic relation-
ships of the Modified Cam Clay model in undrained shearing are given by:
 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 301

"  #
6G0 j r0r  p0
dr0r 0
¼ 2G der þ dp 1 þ 0
 ð4aÞ
tM 2 p0 2p0  p0c
"

0  #
0 0
6G j r h  p
dr0h ¼ 2G0 deh þ dp0 1 þ
 ð4bÞ
tM 2 p0 2p0  p0c

"  #
6G0 j r0z  p0
dr0z 0
¼ 2G dez þ dp 1 þ 0
 ð4cÞ
tM 2 p0 2p0  p0c

where der ; deh ; dez = incremental radial, tangential, and axial natural strains. However,
as dez ¼ 0 in plane strain and der þ deh ¼ 0 in undrained shearing, then Eqs. 4a–4c
become:

"  #
6G0 j r0r  p0
dr0r 0
¼ 2G deh þ dp 1 þ
0  ð5aÞ
tM 2 p0 2p0  p0c
"

0  #
0 0
6G j r  p
dr0h ¼ 2G0 deh þ dp0 1 þ
h  ð5bÞ
tM 2 p0 2p0  p0c
"
 #
6G0 j r0z  p0
dr0z ¼ dp 1 þ 0
 ð5cÞ
tM 2 p0 2p0  p0c

The shear modulus G0 and the hardening parameter p0c are given by (See, for example,
Wood 2007):

3 ð1  2v0 Þ 3 ð1  2v0 Þ tp0

G0 ¼ K ¼ ð6Þ
2 ð 1 þ v0 Þ 2 ð 1 þ v0 Þ j

and

1=K
p0c ¼ p0 p0o =p0 ð7Þ

where v0 ¼ Poisson’s ratio and K ¼ tp0 =j ¼ bulk modulus. According to Eq. 6, the
shear modulus G0 increases as the mean effective stress p0 increases. Substitution of G0
and p0c in Eqs. 5a–5c yields:
302 V. Silvestri and C. Tabib

2
0  3
0 0 0 0
ð1  2v Þ tp 9 ð 1  2v Þ r  p
dr0r ¼ 3 deh þ dp0 41 þ 2  r
 5 ð8aÞ
ð 1 þ v0 Þ j M ð1 þ v0 Þ 2 
p0 =p0 1=K p0
o

2
0  3
0 0 0 0
ð 1  2v Þ tp 9 ð 1  2v Þ r  p
dr0h ¼ 3 deh þ dp0 41 þ 2  h
 5 ð8bÞ
ð 1 þ v0 Þ j M ð1 þ v0 Þ 2 
p0 =p0 1=K p0
o

2
0  3
0 0
9 ð1  2v Þ r  p
dr0z ¼ dp0 41 þ 2  z
 5 ð8cÞ
M ð1 þ v0 Þ 2 
p0 =p0 1=K p0
o

Equation 8c may be rearranged to give:

0 


9 ð1  2m0 Þ rz  p0 dp0
dr0z  dp ¼ d 0
r0z p ¼ 20
0

  0 ð9aÞ
M ð1 þ m Þ 2  p0 =p0 1=K p
o

or


d r0z  p0 9 ð1  2m0 Þ dp0

 ¼ 2   ð9bÞ
r0z  p0 M ð1 þ m0 Þ p0 2  p0 =p0 1=K


o

and, finally,

 dp0
dln r0z  p0 ¼ P 
1=K  ð9cÞ
p0 2  p0o =p0

0
where P ¼ M92 ðð12m Þ
1 þ m0 Þ :

Integration of Eq. 9c gives (See appendix):

"

#PK=2

 2  p0o =p0 1=K p0i 1=K
r0z ¼p þ 0
r0zi p0i
1=K 0 ð10Þ
2  p0o =p0i p

The remaining principal effective stresses r0z and r0h are found by introducing Eq. 10
into the expressions of the mean effective stress p0 and the deviator stress q, as given by
Eqs. 1a and 1b, resulting into:

0 "
1=K
1=K #PK=2
rzi  p0i 2  p0o =p0 p0i
r0r ¼p  0

 1=K p0
þs ð11aÞ
2 2  p0 =p0 o i
 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 303

and

"
1=K
1=K #PK=2
r0  p0i 2  p0o =p0 p0i
r0h 0
¼ p  zi
 1=K p0
s ð11bÞ
2 2  p0o =p0i

0 0 
where the shear stress s in the horizontal plane (r : h) is equal to rr  rh =2 and is
given by:
8 "
0 0 1=K
1=K #P K 9 1=2
1 <4
0

0 2 2  po =p p0i =
s¼ q  3 rzi  pi
2

 ð12Þ
2 :3 0 1=K
2  p0o =pi p0 ;

The natural shear strain c is found by integration of the incremental shear strain
dc ¼ jder  deh j ¼ 2deh , which is determined from Eqs. 5a and 5b, that is, from
0
ds 6jsdp
dc ¼ 0

 ð13aÞ
G t M p0 2p0  p0c
2

or

2 ð1 þ v0 Þ jds 6jsdp0
dc ¼ 
 ð13bÞ
3 ð1  2v0 Þ tp0 tM 2 p0 2p0  p0c

where G0 is given by Eq. 6. Integration of Eq. 13b gives:

2 jð1 þ v0 Þ Zs ds sdp0
0
6j pZ
c¼ 
 ð14Þ
3 tð1  2v0 Þ 0 p0 tM 2 p0 p0 2p0  p0c
i

1=K
where the hardening parameter p0c ¼ p0 p0o =p0i from Eq. 7. It is apparent that the
shear strain c in Eq. 14 must be evaluated numerically due to the complex nature of the
expression of the shear stress s from Eq. 12.
Computations of total radial stress and pore pressure were carried out following the
same approach of Silvestri and Abou-Samra (2012). For completeness, their expres-
sions are briefly repeated herein.
The total radial stress rr acting in the clay around the expanding cylindrical cavity
is given by (Yu 2000; Silvestri and Abou-Samra 2012):
c
Z sdc
rr ¼ c1
þ rri ð15aÞ
0
e
304 V. Silvestri and C. Tabib

which becomes
cZ asdc
rra ¼ c1
þ rri ð15bÞ
0
e

at the wall of the cavity, where rri is the initial total radial or horizontal stress, and the
natural shear strains c and ca are given by:
0
2
r
c ¼ ln½  ð16aÞ
r

and
0
2
a
ca ¼ ln½  ð16bÞ
a

at the wall of the cavity. The pore pressure is determined by subtracting the radial
effective stress r0r given by Eq. 11a from the radial total stress rr given by Eq. 15a. In
Eqs. 16a and 16b, ðr; r 0 Þ and ða; a0 Þ represent respectively, generic radial distances and
cavity radii, before and after the distortion has occurred. In order to obtain the limiting
radial expansion pressure at the wall of the cavity, it is convenient to use the Almansi
tangential strain which is defined as (Baguelin et al. 1978):

1 r 02  r 2
a¼ ð17aÞ
2 r 02

which becomes equal to

1 a02  a2
aa ¼ ð17bÞ
2 a02

at the wall of the cavity. Substitution of Eqs. 17a and 17b into Eqs. 16a and 16b gives:

c ¼ lnð1  2aÞ ð18aÞ

and

ca ¼ lnð1  2aa Þ ð18bÞ

which, when introduced into Eqs. 15a and 15b yield:

Za sda
rr ¼ þ rri ð19aÞ
0
a
 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 305

and
aZa sda
rra ¼ þ rri ð19bÞ
0
a

at the wall of the cavity, as also obtained by Baguelin et al. (1978). As mentioned
above, the Almansi tangential strain was introduced for facilitating the determination of
the limiting radial expansion pressure. Indeed, consideration of the upper limit of the
integral in Eq. 15b indicates that ca must be equal to infinity for the radial pressure to
reach the limiting expansion condition. Such a calculation may involve considerable
computational problems. However, because Eq. 18b shows that ca ¼ 1 is reached for
aa ¼ 0:5, then it becomes relatively easy to carry on the integration process in Eq. 19b
up to aa ¼ 0:5 without experiencing any computational difficulties. This shows the
superiority of the Almansi tangential strain over the natural strain in extending the
integration process to infinity. There is also an additional advantage. In effect, in
several software programs involving either finite elements or finite differences, it is
often assumed that critical state is reached when the cavity radius has doubled in size,
that is, when a0 ¼ 2a in order to avoid numerical difficulties. As this condition cor-
responds to aa ¼ 0:375 from Eq. 17b or to ca ¼ 1:386 from Eq. 18b, it is apparent that
the limiting state of aa ¼ 0:5 is still far away and that by setting aa ¼ 0:375 can only
result in approximate limiting values.

3 Application

Before discussing in detail the application of the various theoretical expressions derived
previously to the benchmark case presented below, it should be mentioned that the
Modified Cam Clay model is known to give reasonable results only for isotropically
normally consolidated clays (See, for example, Wood 2007). If either the initial stress
state or the clay fabric, or both, are anisotropic, or if the soil is overconsolidated, better
models should be resorted to, for example, such as the anisotropic Modified Cam Clay
model (Dafalias 1987; Dafalias et al. 2002, 2006) or the Banerjee model (Banerjee and
Yousif 1986; Banerjee et al. 1988). These two models have the advantage over more
sophisticated and complex models that they can account for both inherent and induced
anisotropy with relatively few model parameters. It is the authors’ intention to apply
one of these models to the problem at hand in the near future.
However, as the principal aim of the present study was to obtain the exact solution
of the principal effective stresses generated around a cylindrical cavity in Modified
Cam Clay under plane strain and undrained conditions, computations were carried out
assuming that the Modified Cam Clay model could also be applied to a Ko normally
consolidated clay.
The theoretical relationships derived previously will be applied to the simulation of
a plane strain undrained expansion of a cylindrical cavity in Ko normally consolidated
remoulded Boston Blue Clay modelled as Modified Cam Clay. The properties of
the clay are the following (Randolph et al. 1979): OCR = 1, Ko ¼ 0:55, t ¼ 2:16;
306 V. Silvestri and C. Tabib

k ¼ 0:15; j ¼ 0:03; K ¼ 0:8; and M ¼ 1:2 for /0 ¼ 30 . The OCR is based on the
vertical effective stress.
The initial stress conditions are represented for illustration purposes by:
r0zi ¼ 300 ½kPa, r0ri ¼ r0hi ¼ 165 ½kPa, ui ¼ 0, p0i ¼ 210 ½kPa, qi ¼ 135 ½kPa,
p0o ¼ 256 ½kPa, p0c ¼ 270 ½kPa, and Poisson’s ratio m0 ¼ 0:2855. The choice of m0 ¼
0:2855 was made so that the value of the initial shear modulus G0i , calculated from
Eq. 6, was equal to that obtained from the data of Randolph et al. (1979), that is,
G0i ¼ 7570 ½kPa. These authors assumed G0 ¼ G0i ¼ constant in the entire expansion
process. The stress parameters at critical state are p0f ¼ 147:6 ½kPa, qf ¼ 177:1 ½kPa.
Figure 2 compares the shear stress-shear strain curve obtained in this study with
that derived by assuming G0 to remain constant. Examination of two curves shows that
they are similar.

Fig. 2. Shear stress-shear strain curves

 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 307

Figure 3 compares the total radial stress rra and the excess of pore pressure u
generated at the wall of the cavity as a function of the Almansi tangential strain aa .
Again, comparison between the two sets of curves indicates that the results are prac-
tically equivalent.

Fig. 3. Comparison of total radial stress and excess pore water pressure generated at wall of
cavity as function of Almansi tangential strain aa

Finally, Fig. 4 shows the distributions of the principal effective stresses and of the
excess pore pressures around the cylindrical cavity at critical state. Once again, the two
series of curves are quite similar.
Although the two sets of results presented in Figs. 2, 3 and 4 are practically
equivalent, the approach followed in the present paper, by assuming Poisson’s ratio m0
to remain constant during shearing, is superior to that in which the initial shearing
modulus G0 remains constant, because the present solution allows finding exact explicit
expressions for the principal effective stresses.
308 V. Silvestri and C. Tabib

Fig. 4. Distributions of principal effective stresses and excess pore water pressures around
cylindrical cavity at critical state

4 Conclusions

On the basis of the results reported in this paper, the following main conclusions are
drawn:
1. The assumption of constant Poisson’s ratio allows obtaining explicit expressions for
the principal effective stresses generated around an expanding cylindrical cavity in
Modified Cam Clay.
2. The theoretical relationships are applied to the simulation of an expanding cylin-
drical cavity in Ko normally consolidated remoulded Boston Blue Clay. The results,
which are compared with those obtained by assuming that the shear modulus G0
remains constant during the expansion, show that the two approaches are quite
similar.

Acknowledgements. The author expresses his gratitude to the Natural Sciences and Engi-
neering Research Council of Canada for the financial support received in this study.
 Analysis of Cylindrical Cavity Expansion in Modified Cam Clay 309

Appendix: Derivation of Eq. 10

The integral in Eq. 9c reads

0

r  p0 pZ 0 dp0
ln 0z ¼ P 
  ðA1Þ
rzi  p0i 0 0
p0 p 2  p =p
0 1=K
i o

0
where P ¼ M92 ðð12m Þ
1 þ m0 Þ :

1=K
By letting y ¼ p0o =p0 , then p0 ¼ p0o yK and dp0 ¼ p0o ðKÞyK1 dy, and the
integral in Eq. A1 becomes

Z ðp0o =p0 Þ1=K Z ðp0o =p0 Þ1=K

dy dy
I ¼ PK ¼ PK ðA2Þ
ðp0o =p0i Þ
1=K
yð 2  yÞ ðp0o =p0i Þ
1=K
ðy  2yÞ
2

Integration of Eq. A2 yields:

y
PK y 
I¼ ln  ðA3Þ
2 y  2 yi

or





PK y yi  2 PK yi y2
I¼ ln ¼ ln ðA4Þ
2 yi y2 2 y yi  2

1=K
1=K
where y ¼ p0o =p0 and yi ¼ p0o =p0i . As a result, Eq. A4 becomes:
("
#
)
1=K
PK p0o =p0 p0i 1=K
2
I¼ ln
 1=K
ðA5Þ
2 p0 =p0
o i 2 p

Combining Eq. A5 with Eq. A1 yields:

("
#
)PK2
1=K

 p0o =p0 p0i 1=K
2
r0z  p ¼ r0zi  p0i
0

1=K 0
ðA6Þ
p0 =p0
o i 2 p

which, when rearranged, gives:

("
1=K #
1=K )PK2

 2  p0o =p0 p0i
r0z 0
¼p þ r0zi  p0i
 ðA7Þ
2  p0o =p0i
1=K p0

Eq. A7 is Eq. 10 in the main text of the paper.

310 V. Silvestri and C. Tabib

References
Baguelin, F., Jézéquel, J.F., Shields, D.H.: The Pressuremeter and Foundation Engineering.
Trans Tech Publications, Clausthal (1978)
Banerjee, P.K., Yousif, N.B.: A plasticity model for the mechanical behaviour of anisotropically
consolidated clay. Int. J. Numer. Anal. Meth. Geomech. 10(5), 521–541 (1986)
Banerjee, P.K., Kumbhojkar, A.S., Yousif, N.B.: Finite element analysis of the stability of
vertical cut using an anisotropical soil model. Canad. Geotechn. J. 25(1), 119–127 (1988)
Cao, L., Teh, C.I., Chang, M.-F.: Undrained cavity expansion in modified Cam clay I: theoretical
analysis. Géotechnique 51(4), 323–334 (2001)
Chen, S.L., Abousleiman, Y.N.: Exact undrained elasto-plastic solution for cylindrical cavity
expansion in modified Cam clay. Géotechnique 62(5), 447–456 (2012)
Dafalias, Y.F.: An anisotropic critical state clay plasticity model. In: Proceedings of the Second
International Conference on Constitutive Laws for Engineering Materials. Theory and
Applications, Tucson, Arizona, vol. 1, pp. 513–521. Elsevier, 5–8 January 1987
Dafalias, Y.F., Manzari, M.T., Akaishi, M.: A simple anisotropic clay plasticity model. Mech.
Res. Commun. 29(4), 241–245 (2002)
Dafalias, Y.F., Manzari, M.T., Papadimitriou, A.G.: SANICLAY: simple anisotropic clay
plasticity model. Int. J. Numer. Anal. Meth. Geomech. 30(12), 1231–1257 (2006)
Gens, A., Potts, D.: Critical state models in computational geomechanics. Eng. Comput. 5, 178–
197 (1988)
Monnet, J.: Numerical validation of an elasto-plastic formulation of the conventional limit
pressure measured with the pressuremeter test in cohesive soil. ASCE J. Geotech. Geoen-
viron. Eng. 133(9), 1119–1127 (2007)
Randolph, M.F., Carter, J.P., Wroth, C.P.: Driven piles in clay-the effect of installation and
subsequent consolidation. Géotechnique 24(4), 361–393 (1979)
Silvestri, V., Abou-Samra, G.: Analytical solution for undrained plane strain expansion of a
cavity in modified Cam clay. Geomech. Eng. 4(1), 19–37 (2012)
Yu, H.S.: Plasticity and Geotechnics. Springer, New York (2006)
Yu, H.S.: Cavity Expansion Methods in Geomechanics. Kluwer Academic Publishers, Dordrecht
(2000)
Wood, D.M.: Soil Behaviour and Critical State Soil Mechanics. Cambridge University Press,
Cambridge (2007)
Zytynski, M., Randolph, M.F., Nova, R., Wroth, C.P.: On modelling the unloading-reloading
behaviour of soils. Int. J. Numer. Anal. Meth. Geomech. 2(1), 87–93 (1978)
 Modeling and Interpretation Wave Fields
in Hierarchical Heterogeneous Medium

Olga A. Hachay1(&), Andrey Y. Khachay2, and Oleg Y. Khachay2

1
Institute of Geophysics UB RAS, Ekaterinburg, Russia
2
Ural Federal University, Ekaterinburg, Russia

Abstract. Geological medium is an open system which is influenced by outer

and inner factors that can lead it to a unstable state. That non stability is as a rule
occurred locally and these zones are named as dynamically active elements,
which are indicators of potential catastrophic sources. These objects differ from
the embedded geological medium by their structural forms, which often are of
hierarchical type. The process of their activisation can be searched, using wave
fields monitoring. For that purpose it is needed to develop new algorithms of
modeling wave field’s propagation through the local objects with hierarchical
structure. Also it is needed to develop new theory of interpretation the wave
field distribution for defining the contours of these local hierarchical objects. It
had been constructed algorithms for 2D modeling of sound diffraction and linear
polarized transversal seismic wave on an intrusion of hierarchic structure,
located in the layer number J of N-layered elastic medium. We used the method
of integral and integral-differential equations for a space frequency presentation
of wave field’s distribution. It is developed an algorithm for constructing the
equation of theoretical inverse problem for 2-D electromagnetic field of E and H
polarization and linear polarized longitudinal elastic wave by excitation of the
N-layered conductive or elastic medium with hierarchic conductive or elastic
inclusion located in the m-th layer. From the theory it is obviously that for such
complicated medium each wave field contains its own information about the
inner structure of the hierarchical inclusion. Therefore it is needed to interpret
the monitoring data for each wave field apart, and not to mix the data base.
These results will be the base for constructing new systems of monitoring
observations of dynamical geological systems. Especially it is needed to prevent
rock shocks in deep mines by their exploitation.

Keywords: Hierarchic medium
Electromagnetic field
Seismic field

Algorithms of modeling
Equation of theoretical inverse problem

1 Introduction

The last decades are characterized by active development of Earth’s sciences. We shall
use the materials, published in the book (Dmitrievsky 2009) by Russian academician
Dmitrievsky A.N., who suggested the conception about the development of oil-gas
geology in Russian Federation. The modern research methods and technologies give
the opportunity to obtain new data about the Earth’s structure and processes, which

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_24
312 O.A. Hachay et al.

occur in its interior. The conception development about the nonlinear geodynamics
practically coincides with research of nonlinear processes in different parts of physics.
In geology soliton and auto wave conceptions are developed, principles of synergetic
and self organization become be used, and in geodynamics the macro quantum
behavior of large mass matter, which are in critical state, in geophysics the auto wave
nature of geophysical fields is researched. A new of direction quantum geodynamics
appeared. In contrary from traditional approaches in geodynamics, which are based on
classical models of continuum, quantum geodynamics allows analyzing the Earth’s
energetic structure, which evolutes in time and penetrates in all natural phenomena and
possesses macro quantum time features. The development of possible methods and
approaches must be based on the considering energy of dynamical processes.
For understanding and analyzing the state of the geological medium it is needed to
compare the key ideas of geophysics, which give the main ideas to research different
dynamical events into the Earth.
“Geophysics of XX century” is the understanding of such features:
– geophysical fields are indicators of the processes, which occur in lithosphere;
geophysical parameters, which are registered distantly can have functional or cor-
relation relation with the matter-structural characteristics of geological medium (on
macro- and micro levels);
– analysis of space-time and energetic distribution of geophysical field can give
information about space-time distribution of geological medium properties;
– registration and analysis of geophysical field in the monitoring regime can give
information about the geodynamical processes in near borehole space, in Earth’s
crust and lithosphere in more deep Earth’s layers.
Practical problems of geophysics of XX century had been vigorous stimulus of
evolution of theoretical and experimental physics of thin layered, porous and crack
media. As a result it had been derived new classes of mathematical models fluid
saturated heterogeneous media, it had been researched anisotropic effects of geological
media, and it had been revealed different physical and physical-chemical effects which
occur on the boundaries “solid skeleton-fluid”. Geophysics for the first time set a
question about possibility of construction physical-geological and mathematical models
of geological objects and processes. Applied geophysics of XX century had realized the
possibility of research one and the same geological objects on micro level (nuclear
geophysics), meso level (electrical, heat, magnetic, acoustic fields) and macro level
(fields of elastic waves and low frequency electromagnetic fields).
Key ideas in XXI century in geophysics: Geophysics of the XXI century is the
understanding of such features: the Earth is a self evolutionary, self conditioned geo
cybernetic system for which the geophysical fields fulfill the role of a driving mech-
anism. Hierarchic and quantum features of geophysical space can be used in new
geophysical principles together with nonlinear effects and effects of re-emission of
geophysical fields that allow us to create a new geophysics. At the same time we have
to consider new aspects in theory and mathematical modeling of geophysical fields and
new systems of data interpretation. That will lead to development new equation classes,
which describe the distribution of elastic and electromagnetic fields in heterogeneous
and hierarchic media, with account of various effects of nonlinearity of geological
 Modeling and Interpretation Wave Fields 313

media and irreversibility of geophysical processes. New theories of inverse problems

solution with account of the hierarchic structure of inclusions imbedded into the lay-
ered medium will be developed.
Now we shall investigate and study some results which are into the key direction of
geophysics of XXI century, noted by Dmitrievsky (Dmitrievsky 2009).
In present day for more adequate understanding of the dynamics of processes,
which occur in the geological medium on deep levels by the action of natural and man
caused factors Academician V.E. Panin using the results, which had been derived
together with his colleagues (Panin 2005), had introduced a new paradigm on the
junction between physics and mechanics of deformed solid body, which is the base of
physical mesomechanics.
1. Identification of mechanisms of plastic flow on different structural deformation
levels, which lead to a fundamental change of the initial inner structure of the solid
body and forming into it dissipative substructures as mesoscopic plastic deforma-
tion support.
2. Fixing a relation between the outer action, changing of the initial inner structure,
forming a hierarchy of mesoscopic self matched structural levels of deformation and
occurring as a result of it mechanical fields.
3. Synergetic approach in the methodology of describing the deformed solid body as a
non equilibrium many leveled medium, which in the points of bifurcation losses its
shift stability on different structural levels and becomes to be destroyed in the
conditions of global losses of its shift stability on macro scale level.
For experimental research of deformation mechanisms of specimens on meso level
it had been developed new methods with use of speckle interferometer and optical
television devices of technical vision, measuring of fractal dimension of deformed solid
body. It turns out, that on the meso level 3-D structural elements (meso volumes) move
as a whole. In that case it is sufficient to consider a representative volume, which
consists on some tens meso volumes, for writing the equations of mesomechanics,
taking into account the inner structure of the deforming solid body.
For realizing the second point of the new paradigm Panin with his colleagues had
written a system of equations, which describes the mechanical field in the deforming
solid body on one level (Panin 2005). It turned out, that it is similarly the Maxwell
system of equations for alternating electromagnetic fields. Similarly electromagnetic
field, where alternating electric and magnetic fields are mutually linked, the common
mechanical field occurs in the deforming solid body, which contains organic mutual
linked the translational and rotating modes (Panin 2005).
That result was very significant for choice of geophysical methods, using for
monitoring of structure and state of rock massif, which is under strong man based
action, which is a component of the geological medium in seismic-tectonically areas.
314 O.A. Hachay et al.

2 Mathematical Modeling and Comparison of Seismic

and Electromagnetic Response for a Block-Layered 2D
Model with Homogeneous Inclusions

In that paper are derived integral equations and integral differential equations of 2D
direct problem for the seismic field in the dynamical variant. We had been provided the
joint analysis of the integral equations for 2D direct problems for electromagnetic and
seismic fields. The received results can be used for definition of the complex criterions
of achievement the research of high complicated medium both with seismic and
electromagnetic methods. For the problem of sound diffraction on the 2D elastic
heterogeneity, located in the j-th layer of the n-layered medium, using the approach
from V.I. Dmitriev and V.D. Kupradze (Hachay et al. 2013) we can derive the integral
differential equation for the distribution of the potential for the vector of elastic dis-
placements inside the heterogeneity.
2 R
ðk1ji
2
k1j Þ R r
2p uðMÞGSj ðM; M 0 ÞdsM þ rjaji u0 ðM 0 Þ
ðr r Þ H
Sc
 rjaji 2pji GSj @u @n dc ¼ uðM Þ by M 2 SC
0 0

rji ðk1ji
2
k1j2 R
C
Þ R
ð1Þ
rðM 0 Þ2p uðMÞGSj ðM; M 0 ÞdsM þ u0 ðM 0 Þ
ðrja rji Þ H
Sc
 rðM 0 Þ2p GSj @u
@n dc ¼ uðM Þ by M 62 SC
0 0
C

Using the second integral-differential presentation we can define the potential of the
elastic displacements in the arbitrary layer, and then we can calculate the distribution of
the vector of elastic displacements in the arbitrary layer. Let us compare the derived
expressions with the solution of the diffraction problem for electromagnetic field in the
frame of the same geometrical model. That case corresponds to the problem of exciting
by a plane wave H -polarization, the solution of which is done in the paper (Hachay
et al. 2008a, b). Let us transform it to the form similarly to (1) and let us compare the
derived equations for the solution of the inner 2-D seismic and electromagnetic
problem. ~k2 ðM 0 Þ ¼ ix~ ~0 ¼ 4p107 Hm ; r
~ðM 0 Þ l
l0 r ~ðM 0 Þ - conductivity in the point M0,
i - the imaginary unit, Hx ðM 0 Þ - the summarized component of magnetic field, Hx0 ðM 0 Þ
- the component of magnetic field in the layered medium without heterogeneity,
~k2 ðM 0 Þ ¼ ix~ ~ji ; ~ki2 ðM 0 Þ ¼ ix~
l0 r ~i ; r
l0 r ~ji - conductivity into the heterogeneity, located
ji
into the j-the layer, r ~i - conductivity of the i-th layer of the n-layered medium,
Gm ðM; M 0 Þ - the Green function of the 2-D problem for the case of H-polarization
(Hachay 2007).
 Modeling and Interpretation Wave Fields 315

2 R
ðk1ji
2
k1j Þ R r
2p uðMÞGSj ðM; M 0 ÞdsM þ rjaji u0 ðM 0 Þ
ðr r Þ H
Sc
 rjaji 2pji GSj @u @n dc ¼ uðM Þ by M 2 SC
0 0
C
ð~kji2 ~kj2 Þ R R ~k 2 ð2Þ
2p Hx ðMÞGm ðM; M 0 ÞdsM þ ~kji2 Hx0 ðM 0 Þ
Sc j

ð~kji2 ~kj2 Þ H
 ~k2 2p Hx @G @n dc ¼ Hx ðM Þ by M 2 SC
m 0 0
j C

The difference in the boundary conditions for the seismic and electromagnetic
problems lead to different types of equations: in the seismic case- to the
integral-differential equation, in the electromagnetic case to the load integral Fredholm
equation of the second type. If for the solutions of the direct electromagnetic and
seismic in dynamical variant problems we can establish the similarity in the explicit
expressions for the components of electromagnetic and seismic fields by definite types
of excitation then with complicating of the medium structure as can we see from the
obtained result by the case of the seismic field linked with longitudinal waves the
similarity vanishes. That means that the seismic information is additional to the elec-
tromagnetic information about the structure and state of the medium.
For the problem of diffraction of a linearly polarized elastic transverse wave on the
2-D heterogeneity located in the j-th layer of the n-layered medium, using the approach
described in the paper (Hachay et al. 2008a, b) for the electromagnetic wave 2-D
problem (case H-polarization), (the geometric model is similar to a that described
higher in the previous problem) we obtain the expressions as follows for the compo-
nents of the displacement vector:
2 R
ðk2ji
2
k2j Þ R l
2p ux ðMÞGSs;j ðM; M 0 ÞdsM þ lja u0x ðM 0 Þ
ji
Sc
ðlja lji Þ H @GSs;j
 l 2p ux ðMÞ @n dc ¼ ux ðM Þ by M 0 2 SC
0
ji

lji ðk2ji
2
k2j 2 R
C
Þ R
ð3Þ
lðM 0 Þ2p ux ðMÞGSs;j ðM; M 0 ÞdsM þ u0x ðM 0 Þ
Sc
ðlja lji Þ H @G
 lðM 0 Þ2p ux ðMÞ @nSs;j dc ¼ ux ðM 0 Þ by M 0 2 SC
C

The expressions (3) content the algorithm of seismic field simulation for distri-
bution of transversal waves in the n-layered medium, which contain a 2D hetero-
geneity. The first expression is a Fredholm load integral equation of the second type the
solution of which gives the distribution of the components of the elastic displacements
vector inside the heterogeneity. The second of them is an integral expression for
calculation of the elastic displacements vector in the arbitrary layer of the n-layered
medium. Comparing the expressions (3) with correspondingly for the electromagnetic
field (H-polarization) (2) we see that there is a similarity of the integral structure of
these expressions. The difference is only for the coefficients of corresponding terms in
the expressions (2) and (3). That we can account by choosing the system of observation
with one or another field. We must also account the difference of the medium response
316 O.A. Hachay et al.

frequency dependence from seismic or electromagnetic excitation. But keeping within

the similarity of the coefficients the seismic field, excited by transversal waves, and the
electromagnetic field will contain the similar information about the structure of the
heterogeneous medium and state, linked with it. Those results are confirmed by the
natural experiments described in the paper (Hachay 2007) Thus, it is showed that for
more complicated, than horizontal-layered structures of the geological medium the
similarity between the electromagnetic and seismic problems for longitudinal waves get
broken. Therefore, these observations with two fields allow getting reciprocally addi-
tional information about the structure and especially about the state of the medium.
These fields will differently reflect the peculiarities of the heterogeneous structures and
response on the changing their state. If we can arrange seismic observations only with
the transversal waves together with the magnetic component of electromagnetic one
(H-polarization) in the 2-D medium, it will be establish the similarity, which can be
used by construction of mutual systems of observation for magneto-telluric soundings
and deep seismic soundings on exchanged waves.

3 Mathematical Modeling of Seismic Response

for a Hierarchic Model

From the point of view of the paradigm of physical mesomechanics, which includes the
synergetic approach to the change of rock massif state of different matter content, that
problem can be solved with use of monitoring methods, which are settled on the
research of hierarchic structured media (Panin 1995). For description of these effects it
is needed to consider the wave process in the hierarchic blocked medium. Let us
consider an algorithm of sound diffraction on 2-D elastic heterogeneity with hierarchic
structure, located in the j-th layer of n-layered medium (Hachay et al. 2008a, b).

ðk1jil
2
k1j
2
Þ RR r
2p ul ðMÞGSp;j ðM; M 0 ÞdsMl þ rjilja u0l1 ðM 0 Þ
ðr r Þ H
Scl
 rjajil 2pjil GSp;j @u
@n dcl ¼ ul ðM Þ by M 2 SCl
l 0 0

rjil ðk1jil
2
k1j
2
Þ
Cl
RR ð4Þ
rðM 0 Þ2p ul ðMÞGSp;j ðM; M 0 ÞdsMl þ u0l1 ðM 0 Þ
ðr r Þ H
Scl
 rðM
ja jil
0 Þ2p GSp;j @u
@n dcl ¼ ul ðM Þ by M 62 SCl
l 0 0
Cl

GSp;j ðM; M 0 Þ source function for the longitudinal wave, which coincides with the
function from the paper (Hachay and Khachay 2013), k1jil 2
¼ x2 ðrjil =kjil Þ – wave
number for the longitudinal wave. In the equations, the index ji indicates the features
within the heterogeneity membership, ja – out of the heterogeneity, l = 1…L − 1 –
number of the hierarchic level, ~ ul ¼ gradul , u0l – potential of the normal seismic field
in the layered medium, when the heterogeneity of the previous rank is absent. If
l = 2…L u0l ¼ ul1 , if l = 1, u0l ¼ u0 , this coincides with the expression from the
paper (Hachay et al. 2015).
 Modeling and Interpretation Wave Fields 317

ðk2jil
2
k2j
2
Þ RR lja
2p uxl ðMÞGSs;j ðM; M 0 ÞdsMl þ ljil u0xðl1Þ ðM 0 Þ
Scl
ðlja ljil Þ H @GSs;j
 l 2p uxl ðMÞ @n dcl ¼ uxl ðM 0 Þ by M 0 2 SCl
ji
Cl
ljil ðk2jil
2
k2j
2
Þ RR ð5Þ
lðM 0 Þ2p uxl ðMÞGSs;j ðM; M 0 ÞdsMl þ u0xðl1Þ ðM 0 Þ
Scl
ðlja ljil Þ H @GSs;j
 lðM 0 Þ2p uxl ðMÞ @n dcl ¼ uxl ðM 0 Þ by M 0 2 SCl
Cl

2
k2jil ¼ x2 ðrjil =ljil Þ - wave number for the transversal elastic wave, GSs;j ðM; M 0 Þ
source function for a transversal elastic wave (Hachay and Khachay 2013). If l = 2…
L u0x ¼ uxðl1Þ , if l = 1, u0xl ¼ u0x , this coincides with the expression from the paper
(Hachay et al. 2015). If by transition on the next hierarchic level the axis of
two-dimensionality does not change and only the geometry of the section of embedded
structures change, then we can write the iteration process of modeling of the seismic
field (case generation only longitudinal wave). The iteration process covers to modeling
of the response of transition from the previous hierarchic level on the next level. Inside
each hierarchic level the integral-differential equation and the integral-differential
representation are calculated as it is written in the papers (Hachay et al. 2008a, b).

4 Inverse Problem for Electromagnetic Field Propagation

Through a Layered Conductive Medium with Hierarchic
Inclusions

In the paper (Hachay 1994) it had been considered a conception of a staged inter-
pretation of the alternating electromagnetic field. On the first stage the parameters of
the normal section or the parameters of the one dimensional non magnetic medium in
which are embedded anomalous conductive or magnetic inclusions are defined. On the
second stage it is developed a procedure of anomalous alternating field fitting by a
system of singular sources, which are embedded into the horizontal layered medium
with geoelectrical parameters, which had been defined on the first stage. On the third
stage it is solved the theoretical inverse problem, that is for the given geoelectrical
parameters for the embedded medium for the set of anomalous parameters, defined on
the second stage, we define the contours of the inclusions. We had derived explicit
integral-differential equations of the theoretical inverse problem for 2-D and 3-D
alternating and 3-D stationary electromagnetic fields in a frame of the models: con-
ductive or magnetic body in the J-th layer of the conductive layered half space. Here,
using the approach, which was written in the papers (Hachay 1989, 1990), we had
derived the equation of the theoretical inverse problem for the 2-D alternating elec-
tromagnetic field (scalar case) for the model: a conductive hierarchic inclusion of the k
rank, located in the J-th layer of the conductive N-layered half space. U þ ðk1Þ ðMk Þ -
þ ðk1Þ þ ðk1Þ
anomaly component Ex or Hx for the inclusion of k-th rank by Mk 2 @Dk , U
318 O.A. Hachay et al.

(M) - sum field Ex or Hx , U iðk1Þ ðMÞ - normal field of the layered section Ex or Hx for
k = 1, for k > 1 - the field Ex or Hx , must be calculated, using the direct problem
solution with use the algorithm (Hachay et al., 2013). Gak ðMk ; Mo Þ - Green function for
the inner area of the heterogeneity of the k-th rank, GðMk ; Mo Þ - Green function of the
N-layered medium (Hachay 1989, 1990), bm ; bi ; bak - complex coefficients for the J-th
layer and for the inner area of the heterogeneity of the k-th rank, which are introduced
in (3).
R þ ðk1Þ
2pU þ ðk1Þ ðM0 Þ ¼ @Dk ððUm ðMk Þ
ðMk ÞÞð@G ðM k ;M0 Þ
 ðbbmi Þ @GðM@nk ;M0 ÞÞ
iðk1Þ ak
þ Um @n
þ ðk1Þ iðk1Þ ð6Þ
@Um ðMk Þ @Um ðMk Þ
 bm ð @n þ @n Þ
 ððb1ak ÞGak ðMk ; M0 Þ  ðb1i ÞGðMk ; M0 ÞÞÞdlk

The hierarchic heterogeneities are approximated by embedded non axial conductive

rak cylinders, located along the axes OX. As a result of solution of the Eq. (1) for the
function rðuÞ, this describes the contour of the sought heterogeneity of the k-th rank.

5 Conclusions

We had developed an iterative algorithm for modeling electromagnetic and seismic

fields in a frame of two models for layered-block 2D models with homogeneous and
hierarchic inclusions. The last algorithms differ from the fractal model approach by a
freer selecting of heterogeneities position of each rank. If the boundaries of the
inclusion of the k rank are fractals (Mandelbrot 1982) the surface and contour integrals
in the integral equations must be changed to repeated fractional integrals of Riman -
Liuvill (Simko et al. 1987). We had derived the equation of the theoretical inverse
problem for the 2D alternating electromagnetic field (scalar case) for the model: a
conductive hierarchic inclusion of the k rank, located in the J-th layer of the conductive
N-layered half space.

References
Dmitrievsky, A.N.: Selected works, V.2, Nauka, Moscow (2009)
Hachay, O.A.: Geophysical monitoring of the state of rock massif with use of paradigm of
physical mesomechanics. Phys. Earth 4, 58–64 (2007)
Hachay, O.A., et al.: Modeling of seismic and electromagnetic field in the hierarchic
heterogeneous media. In: Proceedings of International Conference. IGF UB RAS, Ekater-
inburg (2008a)
Hachay, O.A., et al.: Complex electromagnetic and seismic method of research of the crust and
(2008b)
Hachay, O.A.: Mathematical modeling and interpretation of alternating electromagnetic field in
heterogeneous crust and upper mantle of the earth. Professor Dissertation, IGF UB RAS,
Sverdlovsk (1994)
 Modeling and Interpretation Wave Fields 319

Hachay, O.A.: About interpretation 2-D alternative and 3-D stationary anomalies of
electromagnetic field. Phys. Earth 10, 50–58 (1989)
Hachay, O.A.: About the solution of inverse problem of 3-D alternating electromagnetic fields.
Phys. Earth 2, 55–59 (1990)
Mandelbrot, B.B.: The Fractal Geometry of Nature. Freeman, San Francisco (1982)
Panin, V.E.: Physical mesomechanics and computer construction of materials. Novosibirsk,
Nauka, SB RAS (1995)
Simko, S.G., Kilbas, A.A., Marichev, O.I.: Integrals and derivatives of partial rank and some
their applications. Minsk, Technique (1987)
 Strength and Stiffness Studies of Cement
Stabilized Granular Lateritic Soil

Dipti Ranjan Biswal(&), Umesh Chandra Sahoo,

and Suresh Ranjan Dash

School of Infrastructure, IIT, Bhubaneswar 752050, Odisha, India

{drb11,ucsahoo,srdash}@iitbbs.ac.in

Abstract. A huge network of rural roads is being developed in India under the
most ambitious Prime Minister’s rural connectivity programme, PMGSY
(Pradhan Mantri Gramin Sadak Yojna). Under this programme, thousands of
kilometers of rural roads are being constructed in the country, which require
good quality pavement materials like crushed stone. The scarcity of natural
aggregates has compelled to use marginal materials or locally available soils in
structural layers of these pavements, which would reduced the cost of the
project. Granular lateritic soils are widely available in many parts of India and
presently, this is also used as sub-base material in different rural road projects,
where it satisfies the code specifications. However, granular lateritic soils of
some locations do not satisfy the strength and plasticity requirement of sub-base
layer. But, they can probably be made suitable through stabilization. Though
stabilization of soil by cement or lime is a well known process of improving the
strength and stability of soil, the strength and stiffness parameters of stabilized
lateritic soils in terms of modulus of rupture, resilient modulus, flexural modulus
have very limited reference in literature. Therefore, in this study an attempt has
been made to characterize the cement stabilized lateritic soils for use in sub-base
and base layers in rural road pavements. A comprehensive laboratory testing
programme has been conducted on cement stabilized granular lateritic soil
samples collected from five different places of eastern India to study various
strength parameters such as compressive strength, modulus of rupture and
stiffness properties in terms of flexural modulus of cement stabilized granular
lateritic soil. In this paper, strength and stiffness developments of cement sta-
bilized granular lateritic soil in 7 days and 28 days have been studied and its
suitability as a structural layer in rural roads has been investigated. Suitable
modulus values of cement stabilized granular lateritic soils have been proposed
which can be used as an input parameter for the input in mechanistic design of
roads. Also relationships have been proposed to determine modulus of rupture
and flexural modulus of cement stabilized granular lateritic soil from its com-
pressive strengths.

1 Introduction

A huge network of rural roads is being developed in India under different schemes
launched by the government including the Prime Minister’s rural connectivity pro-
gramme (PMGSY). The requirement of enormous amount of good quality pavement

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_25
 Strength and Stiffness Studies 321

materials like crushed stones (whose source is depleting day by day), is becoming a
concern for the administrators and concerned authorities of the country. In this sce-
nario, investigation of the use of marginal aggregates and locally available granular
materials in structural layers of the pavement is timely justified. Granular lateritic soil
(GLS) is available in many parts of India and presently being used as a sub-base
material in various road projects, where this meets the code specifications. However,
GLS from various sources do not satisfy the strength and plasticity requirement for
sub-base and base layers (Biswal et al. 2016). Hence, these soils need to be modified or
stabilized to make them suitable for sub-base and base layers of pavement. Tradi-
tionally, cement and lime or both are used to stabilize the soils for road construction
purpose (Gidigasu 1976). Indian Roads Congress (2010) provides guidelines are
available for cement and lime stabilization (IRC 2010) of soils. Though stabilization of
soil by cement or lime is a very well known process of improving the strength of soil,
very few studies are available on strength and stiffness properties of cement stabilized
granular lateritic soils (CLS). For the mechanistic design of pavements with stabilized
layers, the input parameters usually required are modulus and flexural strength. But,
studies related to characterization of CLS in terms of modulus, flexural strength etc.
under repeated loading condition is scarce. Therefore, in the present study, strength
properties of CLS such as unconfined compressive strength (UCS), modulus of rupture
(MOR) i.e., flexural strength and stiffness in terms of flexural modulus have been
investigated.

1.1 Studies on Stabilized Lateritic Soil

The objective of stabilization is to improve plasticity characteristics, alteration of grain
size distribution, increase in mechanical strength or stiffness and durability under
adverse conditions which may be done by mixing cement or lime and cement, fly ash or
with any chemical stabilisers. A brief summary on stabilization of lateritic soils by
different researchers is presented in Table 1. Properties of cement or lime stabilized
lateritic soils have been studied by many researchers. However, most of the papers have
explained the improvement of the geotechnical properties e.g., OMC, MDD, liquid
limit, plastic limit, UCS, CBR of stabilized lateritic soil. It needs to be emphasized that,
studies regarding flexural modulus, flexural strength including compressive strength of
stabilized lateritic soil is very scarce The study carried out by Joel and Agbede (2010)
on stabilization of lateritic soil with cement and sand, revealed that 45% sand and 6%
cement is the optimum content for the mix design. Similarly, Portelinha et al. (2012),
Jaritngam et al. (2012), Ravi et al. (2008), Joel and Agbede (2010) investigated the
potential of stabilized lateritic soils of Brazil, Thailand, India and Nigeria respectively.
It may be observed that cement and lime proves to be good modifiers or stabilizers for
lateritic soils in order use as a base or sub-base course material.
Strength of Cement Stabilized Soils
As already mentioned, stabilization of soil by cement and lime is a well known process,
and use of stabilized soil in pavements has also been in practice for decades. However,
the proper characterisation of cement stabilized soil as sub-base and base course
322 D.R. Biswal et al.

Table 1. Some studies on stabilised lateritic soil

Location of Reference Optimum 7 day UCS
laterite content (%)
sample
Nigeria Joel and 45% sand 2702 kPa, 270% improvement
Agbade and 6%
(2010) cement
Brazil Portelinha 3% cement 600 kPa for lime (118% improvement)
et al. (2012) or 3% lime 1000 kPa for cement (233% improvement)
Thailand Jaritngam 3% cement 2220 kPa
et al. (2012)
India Ravi et al. 4% cement 1000 kPa, 440% improvement
(2008) with pond
ash

material is required as the mechanistic design of pavements requires input parameters

like elastic modulus and Poisson’s ratio etc. Unconfined compressive strength is one of
the accepted methods for determining the strength of bound materials. UCS is used for
determining the suitability of the mix to perform satisfactorily as a bound sub-base and
base layer (Paige 1998; Vorobeiff 2004; Austroads 2008; Yeo et al. 2011). Addition of
small percentage of stabiliser may improve the strength and plasticity properties of soil,
but it may lack the required tensile strength and behave as an unbound granular
(Vorobeiff 2004). A criteria developed by Austroads (2013) to differentiate modified
soil from stabilized soil is given in Table 2.

Table 2. Typical properties of modified and bound materials (Austroads 2013)

Classification Testing criteria Test conditions
Modified 0.7 MPa < UCS < 1.5 MPa 28 day moist curing and 105 mm diameter
Bound UCS > 1.5 MPa and 115.5 mm height mould and 4 h
soaking prior to testing

Many research reports and standard specifications have put the requirement of
minimum strength in terms of UCS of stabilized soil for use in sub-base and base layer
of pavement. As per Indian Road Congress (IRC:37-2012) guidelines, the minimum
requirement of 7 day UCS value for cement stabilized soils are 4.5 MPa and
0.75–1.5 MPa for base and sub-base respectively. However, in case of low volume
roads the minimum UCS value for stabilized or bound base is 3 MPa which has been
mentioned in Table 3. Further, wet-dry durability tests need to be conducted on sta-
bilized materials to assess the suitability of these materials for use as sub-base or base
layer. The assessment is done in terms of mass loss after 12 wet-dry cycles. As per
PCA (Portland Cement Association) (1992), the maximum permissible mass loss after
12 wet-dry durability cycles are 14% and 7% for granular materials and cohesive clays
respectively Similarly, the performance of stabilized materials is studied by conducting
 Strength and Stiffness Studies 323

shrinkage tests and load induced fatigue tests. However, durability tests, shrinkage tests
and fatigue tests of cement stabilized lateritic soil are not within the scope of this paper.
Table 3 summarises the strength requirement for stabilized layers for different types of
pavement. Test conditions and sample size for determination of UCS vary as shown
Table 3.

Table 3. Criteria in terms of UCS for suitability of stabilised subbase and base
References UCS (MPa) Curing period Sample size
Medium to Low
high volume volume
roads roads
Portland 2.068 (for
Cement subbase/base)
Association
(1992)
Gass (1993) 2.068 (for
subbase/base)
MEPDG 1.72 (for 1.72 (for 7 days for cement 100 mm
(2004) subbase) and subbase) and 28 days for (d)
115 mm
5.17 (for and 5.17 lime-flyash or (h)
base) (for base) cement-flyash
Austroads 2 1–2 28 days curing and 105 mm
(2008) 4 h soaking prior to (d)
115.5 mm
testing (h)
IRC 37 0.75–1.5 (for 1.7 (for 7 days for cement 50 mm
(2012) subbase) and subbase) and 28 days for (d)
100 mm
4.5–7 (for and 3 (for lime-flyash (h)
base) base)
Syed and 1.38 (for
Scullion subbase/base)
(2001)

The strength criteria for lightly trafficked or low volume roads is low as compared to
high volume roads, as there is less concern on fatigue cracking causing damaging
effects on the life of thin bituminous surfacing. It can be observed that, the minimum 28
day UCS for stabilized materials for use in rural road is 1–2 MPa as per Austroads
(2008). The minimum UCS value of 1.7 MPa and 3 MPa is considered suitable for
sub-base and base course as per IRC:SP-72 (2015).
As the bound material is capable of developing tensile stress, which allows it to
resist and transfer the traffic load to layers below, the MOR needs to be evaluated.
Though bending tensile stress developed in bound layers due to traffic load is below the
tensile strength of bound materials, the MOR is an important tensile strength parameter
which is required for determination of cumulative fatigue damage of stabilized layer.
The default value of MOR of soil cement and cement treated aggregate are taken as
0.69 MPa corresponding to minimum 28 days UCS value of 5.17 MPa (ARA 2004).
324 D.R. Biswal et al.

Stiffness of Cement Stabilized Soils

Elastic modulus of stabilized materials is an important parameter required in elastic
layered analysis for determination of stresses and strains at critical locations of a
pavement. Although elastic modulus of cemented materials can be estimated by
monotonic compression test, third point flexure test is better for determination of
modulus due to the similarity in development of stress-strain gradient generated in a
pavement (Wen et al. 2014; Austroads 2008; Arulrajah et al. 2015; Yeo et al. 2011).
Laboratory determination of stiffness or modulus values of cement stabilized materials
depends on the type of test. The modulus values obtained under monotonic com-
pression testing differs from that obtained from cyclic compression testing (Fall et al.
2008) which also differs with flexural modulus determined from beam testing. The
modulus of cemented materials which influences development of tensile strain is the
flexural modulus and therefore, this may be used as the design modulus. For the elastic
analysis of pavement structure having soil-cement base layer, elastic modulus and
Poisson’s ratio are two important input material parameters. Though, the elastic
modulus of cement stabilized base material depends on so many factors, the design
elastic modulus have been presumed to be 5000 MPa, 3447 MPa and 3500 MPa by
Austroads (2012), ARA (2004), IRC (2012) respectively. As the laboratory determi-
nation of elastic modulus is a tedious process, generally UCS of stabilized materials is
used to determine the modulus value of cemented materials. Equations 1, 2 and 3
presents the models used to determine the design modulus values of soil cement by IRC
(2012), ARA (2004), Austroads (2012) respectively.

Ef ¼ 1000 UCS ð1Þ

Ef ¼ 1200 UCS ð2Þ

Ef ¼ 1000 to 1250 UCS ð3Þ

Where,
E = Flexural Modulus in MPa
UCS = Unconfined Compressive Strength in MPa
As, a range of stabilized materials are being used as bound layer in pavements,
modulus of elasticity and strength cannot be uniquely related, (Williams 1972) which
encourages to study the relationship between UCS and modulus values of stabilized
granular lateritic soils.

2 Materials and Methods

2.1 Materials
The materials used in the present investigation are GLS collected from five different
locations of Eastern India as shown in Fig. 1 which are marked as A, B, C, D and E.
The physical properties of five different GLS are summarised in Table 4 and the
chemical compositions of the soils are summarised in Table 5. The lateritic soils
 Strength and Stiffness Studies 325

Fig. 1. Location map of collected granular lateritic soils in eastern India.

Table 4. Summary of the engineering properties of soil

Sample Region Gravel Sand Silt & womc MDD Soaked UCS LL PI Soil type
no Clay (%) (kN/m3) CBR (%) (kPa) (ISCS)
A Khurdha, 35 60 4 13 21 48 15 SW
Odisha
B Baripada, 95 4 0 9 22 20 760 34 10 GW
Odisha
C Sonepur, 42 37 17 9 22 27 950 68 38 SW
Odisha
D Hussainabad, 33 67 0 10 21 23 410 47 19 SW-SM
Jharkhand
E Bokaro, 24 75 0 8 21 24 360 52 29 SW
Jharkhand
326 D.R. Biswal et al.

Table 5. Chemical Compositions of granular lateritic soil and cement used in this study.
Constituents Lateritic Soil (% by mass) Cement
A B C D E
Fe2O3 21.6 32.49 8.47 7.61 31.82 1.67
Al2O3 18.59 14.26 12.51 14.44 17.73 4.41
SiO2 50.32 44.05 64.06 55.02 48.97 16.33
MgO 0.95 1.02 0.76 0.95 1.14 5.49
K2O 2.1 1.34 1.69 7.69 2.45 1.09
CaO 1.03 0.3 0.82 0.66 67.99
TiO2 2.49 2.97 1.36 1.34 3.62
CuO 0.166
SO2 0.53

selected under this study have a wide range of plasticity characteristics and gradation,
in order to get a practical range of strength and stiffness properties of GLS. The lateritic
soils have been air dried prior to testing or sample preparation. The gradation and
plasticity characteristics of the studied soils suggests that both cement and lime can be
used for stabilization of this soil as per IRC: SP89 (2010), However, cement has been
chosen for the study of the stabilization of lateritic soils in the present study. The
cement used in this study is bagged ordinary portland cement (OPC).

2.2 Methods

Unconfined Compressive Strength

The strength properties have been determined in two stages. In the first stage, UCS tests
have been conducted on all soil samples at various cement content such as 2, 3, 4, 5, 6
and 8%. The air dried soil samples have been mixed with calculated amount of cement
and water and samples of size 38 mm in diameter and 76 mm in height have been
prepared and cured for 7 and 28 days. All the soil samples have been prepared at
optimum moisture content and modified dry density at respective dosed state for the
strength and stiffness study.
Flexure Test
The flexure test is conducted for determination of modulus of rupture. In this study, the
test has been conducted as per ASTM D 1635. Beam samples of size 75 mm

75 mm
290 mm have been prepared for all five GLS mixed with two different
cement content i.e., 2% and 5% to prepare stabilized materials for sub-base and base
course respectively. The beam samples have been cured for a period of 28 days prior to
the flexure test. Three specimens have been tested for each soil samples and the average
value of the result has been used for analysis. The flexure test arrangement is shown in
Fig. 2.
 Strength and Stiffness Studies 327

Fig. 2. Flexural Modulus test of beam sample

Flexural Modulus Test

In the absence of a standard method for determination of flexural modulus of stabilized
beam samples, the protocol developed by Yeo et al. (2011) has been adopted for
determination of flexural modulus. The flexure test arrangement which is shown in
Fig. 2 has also been used for flexural modulus test. Two linear variable differential
transformers (LVDTs) have been used for measuring deflection of the beam centre. The
beam specimen have been subjected to haversine cyclic load of 100 cycles having a
peak load which is equal to 20%, 30% and 40% of failure load of beam samples
determined from flexure test. Modulus value corresponding to 40% of failure load have
been taken for determination of flexural modulus (Wen et al. 2014; Yeo et al. 2008).
The first 50 loading cycles are considered as conditioning stage and the average value
of modulus calculated in last 50 cycles is reported and considered as modulus value of
CLS. Following Equation has been used for determination of flexural modulus.
328 D.R. Biswal et al.

Flexural Modulus,

rt Pl
23Pl3
Ef ¼ ¼ 108d
bd 2
¼ ð4Þ
et 23l2
dd 108bd 3 dd

Where,
rt = Maximum bending stress corresponding to failure load P
et = Maximum bending strain corresponding to failure load P
P = Cyclic load in kN
b = Width of beam in mm
d = Thickness of beam in mm
l = Span from centre to centre of support in mm
dd = Vertical deformation at midpoint in mm

3 Results and Discussions

3.1 Compressive Strength of CLS

The use of stabilized materials or soils for their suitability as a sub-base or base is
generally decided by their UCS values (Yeo et al. 2011). Minimum requirement of
UCS values for stabilized soils to satisfy the strength criteria are summarized in
Table 3. UCS values are also used for differentiating mixtures of soil cement, lime, fly
ash or other soil-chemical mixtures as modified soil, lightly bound or heavily bound
soil. According to NAASRA (1986), stabilized material with 7 day UCS value of
0.8 MPa or higher can be used as a structural layers as per the strength criteria. As per
IRC 37-2012, stabilized materials having UCS value more than 1.5 MPa can be used as
a sub-base material and those having UCS value more than 4.5 MPa can be used as
base material as per strength criteria. However, the durability, shrinkage and fatigue
properties of stabilized granular lateritic soils are not within the scope of this paper. The
7 day and 28 day UCS values of CLS of samples A, B, C, D are shown in Figs. 3 and 4
respectively. As the UCS value of CLS with 3% cement is more than 1.5 MPa, it can
be used as sub-base layer in low and high volume roads. UCS values of LSC of
samples A, C, D and E at 5% cement content satisfies the requirement of base layer i.e.,
UCS > 4.5 MPa. However, granular lateritic soil sample B needs minimum 6% cement
to satisfy the IRC requirement for base layer. Further, as per IRC SP: 72-2015, a
minimum laboratory UCS of 3 MPa is required for base whereas, minimum laboratory
UCS of 1.7 MPa is required for sub-base for rural roads. Hence, CLS with 5% cement
can be used for base layer in rural roads and CLS with 2% cement content can be used
as sub-base layer in rural roads.
The relationship between MOR and UCS of 28 days cured cement stabilized
granular lateritic soil samples is shown in Fig. 5 and presented by Eq. 5. A comparison
of MOR predicted from UCS obtained by Eq. 4 (present study), Eq. 5 (Wen et al.
2014) and Eq. 6 (ARA 2004) have been illustrated in Fig. 6. The MOR of CLS having
2% cement and 5% cement varies in the range of 0.27–0.74 MPa and 0.74–1.69 MPa
 Strength and Stiffness Studies 329

9.00

8.00

7.00

6.00
7 day UCS (MPa)

5.00

4.00 7 day UCS A

7 day UCS B
3.00
7 day UCS C
2.00
7 day UCS D
1.00 7 day UCS E

0.00
0.00 2.00 4.00 6.00 8.00 10.00
Cement (%)

Fig. 3. 7 day UCS values of CLS (specimen size 38 mm diameter and 76 mm height)

10.00
9.00
8.00
28 day UCS (MPa)

7.00
6.00
5.00
28 day UCS A
4.00
28 day UCS B
3.00
28 day UCS C
2.00 28 day UCS D
1.00 28 day UCS E
0.00
0.00 2.00 4.00 6.00 8.00 10.00
Cement (%)

Fig. 4. 28 days UCS values of CLS (size 38 mm diameter and 76 mm height)

330 D.R. Biswal et al.

1.8

1.6 y = 0.198x
R² = 0.795
Flexural strength /MOR (MPa)

1.4

1.2

0.8

0.6

0.4

0.2

0
0 2 4 6 8
UCS (MPa)

Fig. 5. MOR versus UCS

1.2

1 Present study
Flexural Strength (MPa)

NCHRP (2014)
0.8
NCHRP (2004)

0.6

0.4

0.2

0
0 1 2 3 4 5 6
UCS (MPa)

Fig. 6. MOR predicted from UCS using different correlations

 Strength and Stiffness Studies 331

respectively. The default value of MOR of soil cement adopted by IRC 37: 2012 and
MEPDG for fatigue analysis is 0.7 MPa for UCS value of 5.17 MPa is very conser-
vative for CLS as the predicted MOR obtained from Eq. 2 is 1.02 MPa. However,
minimum value of MOR of CLS with 5% cement which is 0.74 MPa can be used as the
design value of MOR for CLS.

MOR ¼ 0:1985
UCS R2 ¼ 0:79 ð5Þ

MOR ¼ 0:14
UCS ð6Þ

MOR ¼ 0:2
UCS ð7Þ

3.2 Flexural Modulus of CLS

Flexural moduli of 28 day cured lateritic soil samples with 2% cement and 5% cement
at stress ratios of 0.2, 0.3 and 0.4 are illustrated in Fig. 7. It may be observed that, the
modulus increases with stress ratio. This behavior of CLS is in line with gravel-cement,
silt-cement and clay-lime (Wen et al. 2014). As already mentioned, flexural modulus at
a stress ratio of 0.4 have been taken as the modulus of that sample as per the study of
Austroads (2008).

2000

1800

1600
Flexural Modulus (MPa)

1400
Soil A
1200
Soil B
1000
Soil C
800
Soil D
600
Soil E
400

200

0
0.1 0.2 0.3 0.4 0.5
Stress ratio

Fig. 7. Flexural modulus at various stress ratios

332 D.R. Biswal et al.

The relationship between flexural modulus and UCS is shown in Fig. 8 and pre-
sented in Eq. 8. However, more samples having wide a range of stiffness or strength
values needs to be studied to develop a good correlation between flexural modulus and
UCS. Similarly, the relationship between flexural modulus and UCS is shown in Fig. 9

2000
1800 y = 195.4x + 159.1
1600 R² = 0.66
Flexural Modulus (MPa)

1400
1200
1000
800
600
400
200
0
0 2 4 6 8
UCS (MPa)

Fig. 8. Flexural modulus versus UCS

2000
y = 794.1x + 353.4
1800 R² = 0.83
Flexural Modulus (MPa)

1600
1400
1200
1000
800
600
400
200
0
0 0.5 1 1.5 2
Flexural Strength (MPa)

Fig. 9. Flexural modulus versus MOR

 Strength and Stiffness Studies 333

and presented in Eq. 9. It may be observed that the flexural modulus and MOR
increases with increase in UCS which is in consistent with the earlier studies (Wen
et al. 2014; Arulrajah et al. 2015; Austroads 2008).
Figure 10 shows a comparison between predicted flexural modulus and measured
flexural modulus of CLS after 28 days curing periods. Figure 11 illustrates a com-
parison of predicted flexural modulus determined from UCS using various models such
as Eq. 8 (present study), Eq. 1 (IRC 2012), Eq. 2 (ARA 2004) and Eq. 10 (Wen et al.
2014). It may be observed from Fig. 11 that the elastic modulus predicted from the
relationship developed in present study is very close to the elastic modulus predicted
from Eq. 10 (Wen et al. 2014). However, the design elastic modulus predicted from
Eq. 1 (IRC 2012) is four times more than the elastic modulus predicted from the
relationship developed in present study. Similarly, the design elastic modulus predicted
from Eq. 2 (ARA 2004) are five times more than the elastic modulus predicted from the
relationship developed in present study. Hence, it is very important to determine the
elastic modulus value of cement stabilized soils by using appropriate laboratory of filed
test which simulated the behaviour of stabilized layer in pavement

Ef ¼ 195:43
UCS þ 159:17 R2 ¼ 0:66 ð8Þ

Ef ¼ 794:1
MOR þ 353:4 R2 ¼ 0:83 ð9Þ

Where,
Ef = flexural modulus in MPa
UCS = Unconfined Compressive strength in MPa
MOR = Modulus of Rupture in MPa

2000
Predicted Flexural Modulus (MPa)

R² = 0.680
1500

1000

500 Regression line

Line of Equality

0
0 500 1000 1500 2000
Measured Flexural Modulus (MPa)

Fig. 10. Predicted verses measured flexural modulus values determined from UCS
334 D.R. Biswal et al.

7000
Present Study
6000
NCHRP (2014)
5000 NCHRP(2004)
Elastic Modulus

IRC (2012)
4000

3000

2000

1000

0
0 1 2 3 4 5 6
UCS (MPa)
Fig. 11. Comparison of design modulus predicted from UCS

EF ¼ 131:08
UCS þ 62382 ð10Þ

E*f = flexural modulus in psi

UCS* = Unconfined Compressive strength in psi

4 Conclusions

A detailed laboratory testing programme have been conducted on cement stabilized

granular lateritic soils collected from five different locations of eastern part of India to
investigate the compressive strength, modulus of rupture and flexural modulus or
design modulus. Cement stabilized granular lateritic soil at 2% cement is found suitable
for sub-base layer from compressive strength criteria in both low. However, Cement
stabilized granular lateritic soil at 5% cement content is found adequate for base layer
of rural roads as per IRC:SP 7-2015. The relationship between modulus of rupture and
UCS observed in this study is consistent with ARA (2004), Austroads (2012). Com-
parison has been made between predicted modulus of rupture and modulus determined
from UCS of stabilized granular lateritic soil. Relationships have been developed to
determine modulus of rupture of cement stabilized granular lateritic soil samples from
UCS of CLS, for practical use. Design value of modulus of rupture, mentioned in
IRC (2012), ARA (2004) is very close to the minimum modulus of rupture of CLS
 Strength and Stiffness Studies 335

determined from this study. A relationship has also been established to determine the
modulus of CLS from UCS. However, further study is required with more number of
samples having a range of stiffness and/or strength values to develop a good correlation
between flexural modulus and UCS. However, further studies of some important
aspects like, shrinkage, durability and fatigue shall be conducted to assess the per-
formance of stabilized granular lateritic soils, which are not within the scope of this
paper.

References
Arulrajah, A., Disfani, M.M., Haghighi, H., Mohammadinia, A., Horpibulsuk, S.: Modulus of
rupture evaluation of cement stabilized recycled glass/recycled concrete aggregate blends.
Constr. Build. Mater. 84, 146–155 (2015)
ASTM D1635: Standard Test Method for Flexural Strength of Soil-Cement Using Simple Beam
Using Third-Point Loading
Austroads: Guide to Pavement Technology Part 2: Pavement Structural Design. Sydney, NSW:
AUSTROADS Incorporated (2012)
Austroads: Review of definitions of modified Granular Materials and Bound Materials.
AP-R434-13, Sydney (2013)
Austroads: The development and evaluation of protocols for the laboratory characterisation of
cemented materials, by R Yeo, AP-T101/08, Sydney, NSW (2008)
Gass, B.G., Ventura, D.F.C., De Beer, M.: Erodibility of cemented materials. National
Department of Transport, South Africa (1993)
Fall, M., Sawangsuriya, A., Benson, C.H., Edil, T.B., Bosscher, P.J.: On the investigations of
resilient modulus of residual tropical gravel lateritic soils from Senegal (West Africa).
Geotech. Geol. Eng. J. 26(1), 109–111 (2008)
Gidigasu, M.D.: Laterite Soil Engineering: Pedogenesis and Engineering Principles. Elsevier,
Amsterdam (1976)
IRC: 37-2012: Guidelines for the design of flexible pavements. The Indian Roads Congress, New
Delhi
IRC: SP 89-2010: Guidelines for soil and granular material stabilization using cement, lime and
flyash. The Indian Roads Congress, New Delhi
IRC: SP 72-2015: Guidelines for soil the design of flexible pavements for low volume roads. The
Indian Roads Congress, New Delhi
Joel, M., Agbede, I.O.: Mechanical-cement stabilization of laterite for use as flexible pavement
material. J. Mater. Civil Eng. 23(2), 146–152 (2010)
ARA: Part 2. Design Inputs; Chapter 2. Material Characterisation. NCHRP, TRB (2004)
NAASRA 1986: Guide to stabilization in road works, National Association of Australian State
Road Authorities, Sydney, NSW
Wen, H., Muhunthan, B., Wang, J., Li, X., Edil, T., Tinjum, J.M.: Characterization of
cementitiously stabilized layers for use in pavement design and analysis. Report No-789. TRB
(2014)
Paige-Green, P.: Recent developments in soil stabilization. In: ARRB Transport Research Ltd
Conference, 19th 1998, Sydney, New South Wales Australia (1998)
Portelinha, F.H.M., Lima, D.C., Fontes, M.P.F., Carvalho, C.A.B.: Modification of a lateritic soil
with lime and cement: an economical alternative for flexible pavement layers. J. Soil Rock
35(1), 51–63 (2012)
336 D.R. Biswal et al.

Portland Cement Association: Soil-Cement Laboratory Handbook (1992)

Ravi, S.U.A., Suresha, S.N., Kashinath, B.: Characterisation of lateritic soil modified with pond
ash and cement. Indian Highways 36(6), 21–27 (2008)
Syed, I.M., Scullion, T.: Performance evaluation of recycled and stabilized bases in Texas.
Transp. Res. Rec. 1757, 14–21 (2001)
Vorobieff, G.: Stabilization practices in Australia. In: Proceedings of New Zealand Institute of
Highway Technology (NZIHT) Stabilization of Road Pavements Seminar, pp. 1–14 (2004)
Williams, R.I.T.: Properties of cement stabilised materials. J. Inst. Highway Eng. 19(2), 5–19
(1972)
Yeo, Y.S., Jitsangiam, P., Nikraz, H.: Mix design of cementitious base course. In: International
Conference on Advances in Geotechnical Engineering, pp. 379–385 (2011)
 An Experimental Study on Partial
Replacement of Clayey Soil with an Industrial
Effluent: Stabilization of Soil Subgrade

Godavarthi V.L.N. Murthy(&), Atkuri Venkata Krishna,

and Vedula V.N. Prabhakara Rao

Department of Civil Engineering, Sri Vasavi Engineering College,

Tadepalligudem, Andhra Pradesh, India

Abstract. Rapid growth of infrastructure viz., construction of highways,

embankments limits the construction on varied soil conditions. The construction
sites at times pass through weak/expansive soils. In order to overcome weak and
problematic sub grade world over, soil stabilization becomes inevitable. In this
connection, mechanical stabilization of clayey soils is explored with partial
replacement by Vitrified Tiles Sludge (VTS), an industrial waste. In this
experimental study, expansive soils samples are collected from Palakol town in
the West Godavari District, Andhra Pradesh, India. The experiments showed
significant volume changes in the soil with a change in water content, corre-
sponding decrease in strength and significant increase in compressibility. The
current experimental investigation studies the variation in the geotechnical
characteristics of expansive soil modified with sludge formed by crushing vit-
rified tiles (VTS) in aqueous medium. Our study reveals that there is a signif-
icant decrease in compressibility characteristics of the clay with an increase in
the proportion of vitrified tile sludge. The composite soil proves to be good
construction material for construction of various civil engineering structures
such as embankment, earthen dams, and runways wherever the decrease in
swelling and plasticity are the real targets.

Keywords: Expansive soil
Atterberg limits
Soil stabilization
Vitrified tiles

sludge

1 Introduction

The study area comprising part of Godavari delta extends over Andhra coast from
Rajahmundry to Narsapur. Geographically, the area lies between, 81.8040° E to
81.6966° E Long. and 16.4330° N to 17.0005° N Lat. and is covered by Survey of India
topographical map Nos. 65 D & Hand 66 A & E. The north-western boundary touches
the Madras- Calcutta grand trunk road passing through Vijayawada, Eluru and Rajah-
mundry. Administratively, the area falls in the jurisdiction of East and West Godavari
districts of Andhra Pradesh. The delta is nearly flat and highly irrigated. The geological
succession of the area comprises alluvial deposits (Recent period), Deccan trap
(Eocene), Gondawana (Carboniferous), Cuddapah-Kurnool and Dharwars (Cambrian)
and granite-gneisses (Archaean). The area receives rainfall both in south-west and

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_26
338 G.V.L.N. Murthy et al.

north-east monsoon periods. The actual rain fall in the East Godavari District is
449.7 mm and the actual rain fall in the West Godavari district is 442.0 mm. The main
soil groups are Black cotton soils with the subgroup Typical Pellusterts and Chro-
musterts with the essential clay mineral as fine montmorillonitic. The area under study
has many tile factories where in untreated effluent comprising sludge made up of
grounded tiles and water is left into the cultivable land rendering them nearly useless.
Vitrified Tiles are the latest and largest growing industry alternate for many tiling
requirements across the globe with far superior properties compared to natural stones
and other manmade tiles. India and China are the largest regions to contribute to the
6900 million square meters of production every year. With an annual growth rate of
20% worldwide and 25% in India, Vitrified tile is the fastest growing segment in the
tile industry. Vitrified tiles own 12% share of the overall tile production across the
world. With the increase in production of vitrified tiles in India, there is growing
concern about the huge generation of tile polishing dust.
The raw material composition of Vitrified tiles is:
– Quartz of 99% Silica,
– Potash Feldspar of 12% to 14% Alkalis,
– Soda Feldspar of 12% to 14% Alkalis,
– Strengthening agent, China clay, body stains for producing in various colors.
VTS is collected from a tile factory situated at Narayanapuram Village, Ungutur
Mandal, West Godavari, Tadepalligudem - 534407, Andhra Pradesh, India. The
physical properties of Vitrified Tile Sludge (VTS) are given in the Tables 1 and 2 below.

Table 1. Chemical properties of Vitrified Tile Sludge (VTS)

Materials VTS (%)
SiO2 63.29
Al2O3 18.29
Fe2O3 4.32
CaO 4.46
MgO 0.72
P2O5 0.16
K2O 2.18
Na2O 0.75
SO3 0.10
CL− 0.005
TiO2 0.61
SrO2 0.02
Mn2O3 0.05
L.O.I 1.61

The current experimental study is concerned with the selection of approximate type
of soil to achieve a very high degree of compaction and to expose the compaction
properties of clay. The results of the study can provide thoughts for applying clay soil
in various applications of soil stabilization process.
 An Experimental Study on Partial Replacement 339

Table 2. Engineering properties of pure vitrified tile sludge

% Medium sand 1.43
% Fine sand 95.7
% Silt and Clay 1.07
Specific gravity 2.46
MDD (g/cc) 1.58
OMC (%) 19.4

1.1 Objectives of the Study

The following are the main objectives for the present study:
(a) Evaluate the limiting VTS content to reduce the plasticity i.e., Atterberg limits
viz., Liquid limit by 20% and Plastic limit by 10% off the native clayey soil
(b) To increase the sub grade soil stiffness of the soil by at least 25%.

2 Literature Review

Al-Rawasa et al. (2005) has tried to stabilize expansive soil with lime, cement, com-
binations of lime and cement, sarooj (Artificial pozzolana), heat treatment and have
found that with the addition of 6% lime, both the swell percent and swell pressure
reduced to zero and heat treatment reduced swelling potential to zero. Al-Rawas et al.
(2002) investigated the effectiveness of using cement by-pass dust, copper slag,
granulated blast furnace slag, and slag-cement in reducing the swelling potential and
plasticity of expansive soils. Bell (1989) through his experiments on Clayey soil
showed that addition of a small percentage of lime enhanced many of the engineering
properties of the soil. Boardman et al. (2001) with the help of two clay minerals
assessed to assess the time-dependent effects of mineral structural chemistry on the
lime–clay reaction. Briptive (2004) emphasized the need of a quick or simple test to
show the amount of lime required to react chemically with a soil to bring about these
physical changes to an optimum degree. Buhler and Cerato (2007) and Modak et al.
(2012) mixed highly expansive soil with lime and class C fly ash to study the plasticity
reduction in highly expansive natural clays from Idabel, Oklahoma. Çokça (2001) has
used, high-calcium and low-calcium class C fly ashes from the Soma and Tuncbilek
thermal power plants, respectively, in Turkey, for stabilization of an expansive soil. An
evaluation of the expansive soil-lime, expansive soil-cement, and expansive soil-fly ash
systems is presented. Cristelo et al. (2013) have studied the effects of sodium-based
alkaline activators and class F fly ash on soil stabilization. Eades (1966) tried to show
through simple test, the amount of lime required to react chemically with a soil to bring
about these physical changes to an optimum degree. Edil et al. (2006) have evaluated
the effectiveness of self-cementing fly ashes derived from combustion of
sub-bituminous coal for stabilization of soft fine-grained soils. Addition of fly ash
resulted in appreciable increases in the California bearing ratio (CBR). Ferguson (1993)
had reported that Ash treatment can effectively reduce the swell potential of fat clay
340 G.V.L.N. Murthy et al.

soils and increases the sub grade support capacity of pavement sub grades. Greeves
(1996) used expansive black cotton soil to assess the influence of ambient temperature
in the progress of lime-soil reactions. Higgins (2005) reported the incorporation of
ground granulated blast furnace slag to combat the soil expansion. Hunter (1988)
studied the expansive reactions between lime and sulfate‐bearing clay soils and
reported that the long‐term pozzolanic chemistry of normal lime‐soil reactions found to
be disrupted while Cation exchange, agglomeration, and carbonation are unaffected.
Jones and Holtz (1973) studied the relationship between the mineralogical composition
and water chemistry and their relationship to geotechnical properties and engineering
structures failures. Kamon and Nontananandh (1991) prepared a product by burning
industrial wastes along with lime to produce a by‐product having cementing charac-
teristics and showed that the resulting by‐product is of promise for use in stabilizing a
loam soil. Khattab et al. (2007) have evaluated the long-term stability characteristics of
lime-treated expansive soils to improve the properties of expansive soils. Matsuo and
Kamon (1981) have investigated a new stabilization method of soft clay soils using
multivalent cations, trivalent cations, e.g. Fe3+ and A13+. The beneficial effects of this
treatment are clarified not only for engineering properties but also for microscopic
mechanisms (ionization process and aggregating phenomenon). McCarthy et al.
(2014), investigated the engineering and durability properties of lime-stabilized soils
treated with low-lime fly ash to limit swelling (heave) due to sulphate, associated with
the ground improvement process. Turkoz et al. (2014) have evaluated the effect of
magnesium chloride (MgCl2) solution on the engineering properties of clay soils. The
results show that dispersive and expansive clay soils can be effectively improved using
an additive MgCl2 solution. Puppala (2003) investigated with class F fly ash,
sulphate-resistant cement, ground granulated blast furnace slag (GGBFS) and lime
mixed with fibers as potential stabilizers in enhancing the strength and volume change
properties of soft, expansive and sulphate-rich soils. Petry and Armstrong (1989) had
developed performance-based testing of chemical stabilizers that simulate field con-
ditions. The situations included a physical erosion test to determine dispersion of clays
that have been treated, two swell test preparation sequences simulating injection of
chemicals into clays, a wet-dry test sample preparation using field gradation specifi-
cations, and a three-dimensional swell test for stability of treated clays when subjected
to wetting. Petry (1997) used Chemical stabilizing agents to improve the behavior of
earth materials. Rao and Subba Rao (1994) have discussed the cause of ground heave
of an inherently non-swelling, kaolinite-rich red soil (from Bangalore, India) due to
prolonged spillage of concentrated (40%, weight/weight solution) caustic soda (sodium
hydroxide) solution into the sub-soil through cracked drains in an industrial estab-
lishment. Treatment of the contaminated ground with 5% ferric chloride solution and
simultaneously measures to minimize caustic soda solution spillage mitigate the soil
against heave from caustic soda attack. Reddy et al. (2015) have employed three types
of additives viz., (a) Cementitious: lime and fly ash (b) Non-cementitious: stone dust,
and (c) Chemical additives: CaCl2 and Na2SiO3 to evaluate the performance of sta-
bilized expansive soils. They investigated the influence of valence of cations (viz.,
monovalent, divalent and trivalent) and mean particle diameter of additive(s) on per-
centage reduction of swelling characteristics. Their results revealed that chemical
additives exhibit superior performance over cementitious and non-cementitious
 An Experimental Study on Partial Replacement 341

additives in reducing the swelling characteristics. Brooks (2009) studied the suitability
of expansive soil as a construction material using rice husk ash (RHA) and flyash. Saha
and Saha (1991) have upgraded expansive soil as a construction material using rice
husk ash (RHA) and flyash, a cost comparison was made for the preparation of the
sub-base of a highway project with and without the admixture stabilizations. When the
RHA content of 12% and a flyash content of 25%, Unconfined Compressive Stress
increased by 97% while CBR improved by 47%. Sivanna (1976) had studied the
problem of damage to structures in expansive soils world wide and have proposed that
sodium carbonate and calcium carbonate is stabilizing a black cotton soil. Sivapullaiah
studied the effect of electrolytes on the shear strength of clayey soils. Abood et al.
(2007) had investigated the effect of adding different chloride compounds including
(NaCl, MgCl2, CaCl2) on the engineering properties of silty clay soil. The increase in
the percentage of each of the chloride compounds increased the maximum dry unit
weight and decreases the optimum moisture content. While the liquid limit, plastic limit
and plasticity index decreased with the increase in salt content, the unconfined com-
pressive strength increased with the increase in salt content. Nalbantoglu (2001) has
shown that Fly ash treatment has tremendous potential as an economical method for the
stabilization of the soil. Significant reduction in the swell potential and an increase in
the hydraulic conductivity values were reported.

3 Need for Study

From the literature review it is observed that: Almost all of the researchers have studied
the effect of (1) Industrial Byproducts (Fly ash, Rice Husk Ash), (2) Chemical additives
(Lime, Caustic Soda, Calcium Carbonate, Sodium Carbonate, Magnesium Chloride,
Mono-valence, Divalent and trivalent compounds).
Effect of treatment on soil properties like specific gravity, liquid limit, plastic limit,
plasticity index, grain size analysis, shear strength of soil and bearing resistance have
been studied. Long-term behavior of soil like consolidation properties, hydraulic
conductivity and resilient modulus has not been studied by most researchers.
• There are many methods of laboratory experimentation to inhibit soil expansion but
very few field methods to forecast soil expansion are available.
• Stress – strain behavior of the modified soil needs to be studied.
• Very minimal or no research has been conducted on industrial waste sludge despite
the fact that they constitute environmental hazards and could have their usefulness
in soil stabilization.

4 Experimental Studies

The methods adopted for the present study of investigation are presented below:
Natural moisture content of the soil samples is determined as per IS: 2720 (Part II)
“Methods of Test for Soils” (Water Content), BIS, New Delhi. The specific Gravity of
the soil is determined as per 38. IS 2720 Part-III/Section 2 (1997), “Methods of Tests
342 G.V.L.N. Murthy et al.

for soils” (Specific Gravity), BIS, New Delhi. The grain size analysis is conducted as
per IS 2720 Part-IV (1995), “Methods of Tests for soils” (Grain Size Analysis), BIS,
New Delhi.
The liquid, plastic limit (Atterberg’s limits) are determined as per IS 2720 Part-V
(1995), “Methods of Tests for soils” (Atterberg Limits), BIS, New Delhi. THE maximum
unit weight and optimum moisture content for a given effort of compaction are obtained
as per IS 2720 Part-VIII (1997), “Methods of Tests for soils” (Proctor’s Compaction
Test), BIS, New Delhi. The bearing resistance of the soil in soaked and un-soaked
conditions is determined as per IS: 2720 (Part-16) (1979), “Methods of Tests for soils”
(California Bearing Ratio), BIS, New Delhi. In CBR (Fig. 1) testing the rate of pene-
tration of the plunger was kept at 1.25 mm/min and Proving ring capacity is 10 KN.
Soaked CBR values at 2.5 mm and 5 mm penetration were calculated for all the samples.

Fig. 1. Liquid limit of clayey soil admixed with different proportions of VTS

While the Unconfined Compressive strength of the soil is determined as per IS

2720 (Part-10) (1991), “Methods of Test for Soils”, (Unconfined Compressive
Strength), BIS, New Delhi. UCS testing is based on the fact that the minor principal
stress is zero and the angle of internal friction of the soil is assumed to be zero. Strain
controlled testing was followed and a strain rate of 1% per minute is used. At the time
of testing loading is continued until the load on the specimen begins to decrease or until
at least 20% of strain has been reached. In this case for the sample of 76 mm height
20% strain amounts to 0.2  76 = 15.2 mm.
Specimens for CBR and UCS testing are prepared at predetermined water content and
unit weight. The free swell Index is determined as per IS: 2720 Part XL (1977), “Methods
of Test for Soils”, Determination of Free Swell Index of Soils, BIS, New Delhi.
 An Experimental Study on Partial Replacement 343

5 Results and Discussion on Results

The native clayey soil is admixed with VTS by varying its proportion and the optimum
content of the VTS is determined (Table 3). The admixed proportions of VTS con-
sidered in this study are 0%, 10%, 20%, 30% and 40%.

Table 3. Summary of the laboratory test results

Sample Palakol
% Sand 6
% Silt 32
% Clay 62
Moisture content, n.m.c (%) 15.38
Specific gravity 2.71
Liquid limit 66.4
Plastic limit 32
OMC, % 22
Max dry unit weight (cd)max (g/cc) 1.56
Soaked CBR for 2.5 mm penetration 1.8%
Unsoaked CBR 3.2%
DFS 150

Table 4. Summary of the test results obtained with different proportions of the VTS with
Clayey soil
VTS (%) 0 10 20 30 40
Liquid limit (%) 36.02 34.67 32.12 29.17 26.14
Plastic limit (%) 33.33 28.57 25 23.07 20
O.M.C (%) 18.2 17 15 14.4 10
Max dry unit weight (g/cc) 1.61 1.62 1.635 1.64 1.66
CBR for 2.5 mm penetration 2.50% 2.60% 5.20% 3.80% 3.72%
UCS 6.68 7.06 8.5 8.24 8.18

The following paragraphs describe the laboratory results (Table 4).

The above values are plotted against the different proportions of VTS admixed with
the Clayey soil. The following are the observations found from these tests:
(a) As the admixed proportion of VTS increases from 10% to 40%, the liquid limit of
the clayey soil is decreasing gradually (Fig. 1). The curve observed as falling
trend. The similar trend is also observed in case of plastic limit of the soil (Fig. 2).
(b) In contrast, the dry unit weight of the soil is gradually increasing with slight
reduction with OMC (Fig. 3).
344 G.V.L.N. Murthy et al.

(c) The soaked CBR values are gradually increased from 2.34% to 5.04% with 0%
and 20% replacement and there onwards the CBR values are found to be
decreasing with further increasing of the VTS content (Fig. 4).

Fig. 2. Plastic limit of the soil obtained with different proportions with VTS

Fig. 3. Dry unit weight of the soil obtained with different proportions with VTS
 An Experimental Study on Partial Replacement 345

Fig. 4. Soaked CBR values of soil obtained with different proportions of VTS

(d) The UCS values are gradually increased from 6.68% to 8.5% with 0% and 20%
replacement and thereafter, the UCS values are found to be decreasing with
further increasing of the VTS content (Fig. 5).

Fig. 5. UCS values of soil obtained with different proportions of VTS

346 G.V.L.N. Murthy et al.

6 Conclusions

The following are the conclusions drawn from the present study:
(a) Addition of varied Vitrified Tile Sludge (VTS) content has improved the prop-
erties of the clay sample.
(b) Replacement of Vitrified Tile Sludge by 20% is found to be optimum.
(c) The flow characteristics, plasticity characteristics and shrinkage characteristics of
the soil sample are gradually decreasing with increase in the percentage of VTS in
the clay sample.
(d) At the optimum VTS content of 20% replacement, the maximum dry density,
soaked CBR value and UCS values are obtained as 1.635 gm/cc, 5.2% and 85,
respectively.
Hence, it is concluded that the VTS as a waste product can also be used for soil
stabilization purpose and the obtained results are based on this single study, further
investigation is required to confirm the findings.

Acknowledgments. We sincerely acknowledge Ms. Ch. Tanuja, Assistant Professor at English

in the Department of Basics Science & Humanities, Sri Vasavi Engineering College, Tadepal-
ligudem, West-Godavari District, Andhra Pradesh, India, for rigorous round of proof reading at
English.

References
Al-Rawas, A., et al.: A comparative evaluation of various additives used in the stabilization of
expansive soils. Geotech. Test. J. 25(2), 199–209 (2002)
Al-Rawasa, A.A., et al.: Build. Environ. 40(5), 681–687 (2005). Elsevier
Bell, F.G.: Lime stabilization of clay soils. Bull. Int. Assoc. Eng. Geol. 39, 67 (1989)
Boardman, D.I., et al.: Development of stabilization and solidification in lime-clay mixes.
Geotechnique 51, 533–543 (2001)
Britpave: Stabilized soils as sub-base or base for roads and other pavements. The British In-situ
Concrete Paving Association, Camberley BP/08 (2004)
Buhler, R.L., Cerato, A.B.: Stabilization of Oklahoma expansive soils using lime and class c
flyash. In: Problematic Soils and Rocks and In-situ Characterization, GSP 162, pp. 1–10.
ASCE (2007)
Çokça, E.: Use of class C flyash for the stabilization of an expansive soil. J. Geotech.
Geoenviron. Eng. 127(7), 568–573 (2001). ASCE
Cristelo, N. et al.: Effects of alkaline-activated fly ash and portland cement on soft soil
stabilization. Acta Geotech. (2013). doi:10.1007/s11440-012-0200-9
Eades, J.L., Grim, R.E.: A quick test to determine requirements for lime stabilization.Highw.
Res. Board Bull. 139(1966), 61–72 (1966)
Edil, T., et al.: Stabilizing soft fine-grained soils with fly ash. J. Mater. Civ. Eng. 18(2), 283–294
(2006)
Ferguson, G.: Use of self commenting flyash as a soil stabilizing agent. Geotechnical Special
Publication, vol. 36. ASCE, New York, pp. 1–15 (1993)
 An Experimental Study on Partial Replacement 347

Greeves, H.M.: An introduction to lime stabilization. In: Proceedings, Seminar on Lime

Stabilization, Loughborough University, pp. 5–12. Thomas Telford, London (1996)
Higgins, D.D.: Soil Stabilisation with ground granulated blastfurnace slag. Cementitious Slag
Makers Association (CSMA) (2005)
Hunter, D.: Lime-induced heave in sulfate-bearing clay soils. J. Geotech. Eng. 114(2), 150–167
(1988). ASCE
IS: 2720 Part XL: Methods of test for soils, Determination of Free Swelkl Index of Soils, BIS,
New Delhi (1977)
IS: 2720 Part-16: Methods of tests for soils (California Bearing Ratio), BIS, New Delhi (1979)
IS 2720 Part-10: Methods of test for soils, (Unconfined Compressive Strength), BIS, New Delhi
(1991)
IS 2720 Part-IV: Methods of tests for soils (Grain Size Analysis), BIS, New Delhi (1995)
IS 2720 Part-V: Methods of tests for soil (Atterberg Limits), BIS, New Delhi (1995)
IS 2720 Part-III: Methods of tests for soils (Specific Gravity), BIS, New Delhi (1997)
IS 2720 Part-VIII: Methods of tests for soils (Proctor’s Compaction Test), BIS, New Delhi
(1997)
Jones, D.E., Holtz, W.G.: Expansive soils- the hidden disaster. Civil Engg. 43(8) (1973)
Kamon, M., Nontananandh, S.: Combining Industrial Wastes with Lime for Soil Stabilization.
J. Geotech. Eng. 117(1), 1–17 (1991). ASCE
Khattab, S.A.A., et al.: Long term stability characteristics of a lime-treated plastic soil. J. Mater.
Civil Eng. 19(4), 358–366 (2007)
Matsuo, S., Kamon, M.: Soil stabilisation by multivalent cations. In: Proceedings of the 10th
ICSMFE, Stockholm, vol. 3, pp. 735–738 (1981)
McCarthy, M.J., Csetenyi, L.J., Dhir, R.K.: Engineering and durability properties of fly ash
treated lime-stabilised sulphate-bearing soils. Eng. Geol. 174, 139–148 (2014). Elsevier
Turkoz, M., et al.: The effect of magnesium chloride solution on the engineering properties of
clay soil with expansive and dispersive characteristics. Appl. Clay Sci. 101, 1–9 (2014)
Modak, P.R., et al.: Stabilization of black cotton soil using admixtures. Int. J. Eng. Innovative
Technol. (IJEIT) 1(5), 1–3 (2012)
Pappula, J., Wattanasanticharoen, E., Punthutaecha, K.: Experimental evaluations of stabilisation
methods for sulphate-rich expansive soils. Proc. Inst. Civil Eng. Ground Improv. 7(1), 25–35
(2003). ISSN 1755-0750
Petry, T.M., Armstrong J.C.: Stabilization of expansive clay soils, RR-1219, TRB, pp. 103–112
(1989)
Petry, T.M.: Performance based testing of chemical stabilizers. Transp. Res. Rec. Transp. Res.
Board Nat. Acad. 1589, 36–41 (1997)
Rao, S.M., Subba Rao, K.S.: Ground heave from caustic soda solution spillage - a case study.
J. Soils Found. 34(2), 13–18 (1994)
Reddy, N.G., et al.: Evaluating the influence of additives on swelling characteristics of expansive
soil. Int. J. Geosynthetics Ground Eng. 1(1), 7 (2015). Springer. doi:10.1007/s40891-015-
0010-x. ISSN 2199-9260
Brooks, R.M.: Soil stabilization with flyash and rice husk ash. Int. J. Res. Rev. Appl. Sci. 1(3),
209–217 (2009)
Saha, S., Saha, P.: Improvement of soils by use of chemicals. In: IGC - 91, Surat, vol. 1, pp. 429–
432 (1991)
348 G.V.L.N. Murthy et al.

Sivanna, G.S.: Strength and Consolidation Characteristics of Black Cotton Soils with Chemical
Additives- CaCl2 & KOH, Report Prepared by Karnataka Engg. Research Station,
Krishnarajasagar, India, pp. 21–27 (1976)
Abood, T.T., et al.: Stabilization of silty clay soil using chloride compounds. J. Eng. Sci.
Technol. 2(1), 102–109 (2007)
Nalbantoglu, Z., Gucbilmez, E.: Improvement of calcareous expansive soils in semi-arid
environments. J. Arid Environ. 47(4), 453–463 (2001). Elsevier B.V
 Ground Improvement of Titanium Dioxide
Waste Spoils and Compressible Organics
with In-Situ Mixing with Portland Cement
and Surcharging

James J. Serpico(&)

Geotechnical Services, Maser Consulting P.A., 331 Newman Springs Road,

Suite 203, Red Bank, NJ 07701, USA
[email protected]

Abstract. Part of an overall $2.4 billion dollar, multi-use, brownfield rede-

velopment project along a major river and highway system in the Borough of
Sayreville, New Jersey, USA, was a need to conduct ground improvement
activities for approximately 40 acres of the 440-acre project for a large retail
structure and portions of the roadway system. During the course of subsurface
explorations it was determined that a 40-acre portion of the project contained
4 to 14 ft of titanium dioxide overlying highly compressible organics. The
titanium dioxide materials were the result of abandoned plant processing
activities in the production of white paint pigments.
The ground improvement program consisted of mixing the titanium dioxide
byproduct, in-place, with Portland cement. The mixing was conducted using
large excavators mixing set dimensioned cells to required depths. Over 375,000
cubic yards of materials were mixed in-place using this method. Many lessons
were learned regarding design and construction considerations to overcome air
quality, Portland cement delivery, mixing thoroughness, mix strength testing
and quality control.
Once in-place cement mixing was completed a full-scale, 28-acre surcharge
program was implemented to consolidate the underlying compressible organics
to support the proposed development. The surcharge program was accomplished
using prefabricated wick drains (PWDs) and surcharge fills consisting of Pro-
cessed Dredge Materials (PDM). PDM typically consist of river and harbor
dredge spoils mixed with at least 8% cement prior to reuse as fills. The sur-
charge program is being remotely monitored 24/7 through the use of digital
piezometers, settlement plates and in-place inclinometers via a project website.

1 Introduction

The project area was formerly used in the production of white pigments by the National
Lead Company. Beginning in the 1930s and continuing until the late 1970s the site
produced white pigment (titanium dioxide) by processing raw ore materials barged to
the site. Initially a sulfuric acid process was used until waste acid disposal costs moved
the plant to consider a chloride process. The sulfuric acid process was halted in the
mid-1970s. During titanium dioxide processing, process waste was discharged into
© Springer International Publishing AG 2018
W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_27
350 J.J. Serpico

several areas at the site. The largest waste containment area was approximately 40 acres
and was commonly known as the “acid lagoon” by locals and site workers. The pH of
the groundwater in the lagoon area was generally found to range between 2.5 to 3.5
with some areas going as low as a pH of 1.8 prior to any material stabilization work.
The strong acid nature of the groundwater resulted in a natural chemical process that
produced high concentrations of heavy metals due to them leaching out of the soil.
Additionally, over the years, transformers for pumping and electrical needs resulted in
PCB (polychlorinated biphenyl) contamination in the lagoon itself. A site photo taken
during mixing operations in May of 2014 (Fig. 1), presents a project scale and outlines
areas to be discussed further in this case study.

Fig. 1. Site photo taken May 2014 during mixing operations

2 Subsurface Conditions

Test boring access was limited in the lagoon area due to existing water levels at the
time and the PCB contamination. The subsurface conditions in the lagoon area gen-
erally consisted of titanium dioxide waste materials directly overlying highly com-
pressible organic silt and clay deposits. Directly underlying the highly compressible
soils there are relatively free-draining sands and gravels. The titanium dioxide waste
materials ranged in thickness from 4 to 14 ft with an average thickness over the area on
the order of 5 ft thick. The highly compressible organic soils ranged in thickness from
8 to 35 ft with depth increasing as one moved toward the river. Groundwater was
acidic as expected and ranged in depth from 3 to 5 ft below existing lagoon bottom
grades.
 Ground Improvement of Titanium Dioxide Waste Spoils 351

Laboratory testing conducted on the compressible organics indicated that a modi-

fied compression index of 0.25 should be used in settlement evaluation based upon
multiple consolidation tests and that the materials were normally consolidated. Atter-
berg Limit testing resulted in an average Liquid Limit value of 85% and a Plasticity
Index of 45%. Natural moisture contents ranged from 58% to 78%.

3 Ground Improvement Considerations

As part of the site development layout and construction phasing considerations it

became quickly apparent that the lagoon area needed to be addressed from an envi-
ronmental and geotechnical viewpoint early on. After site remediation of high PCB
levels at a few isolated areas in the lagoon, test pits were excavated in portions of the
lagoon that had drained to better characterize and evaluate options for the stabilization
of the titanium dioxide waste materials. Samples were collected for treatability studies
for various mix designs. At the same time various options were considered to address
the large volume of titanium dioxide waste materials anticipated in to be in the lagoon.
Consideration was given to complete over-excavation and replacement with structural
fills, mixing the titanium dioxide waste materials with large coarse aggregate (4″ to 6″
crushed stone), utilizing alternate mixing methods such as deep soil mixing, rotating
head mixing and mixing with large track excavator with Portland cement. After con-
sideration of off-site disposal costs due to contaminated soils and site limits to the
lagoon area only for working due to the PCB contamination, mixing in-place with
Portland cement was selected as the design solution.

3.1 In-Situ Soil Stabilization (ISS)

During contractor bidding, mixing in-place with large track excavators was ultimately
selected to be the most cost effective solution. Treatability studies indicated that the
titanium dioxide waste would need a minimum of 7.5% Portland cement to stabilize.
However, the area of the Large Retail store required full-depth mixing completely
through the compressible organics. Treatability evaluations for full-depth mixing indi-
cated that Portland cement percentages would need to be increased to at least 10% due to
impacts of the organics on resulting mixture strength. The contractor proposed three
(3) different size mixing areas based upon cement delivery volumes; Small Size
(10′  50′), Medium Size (20′  60′) and Large Size (20′  90′). The Small Size cells
were for the area with full depth mixing under the Large Retail store with mix depths up
to 16 ft anticipated in the design. Acceptance criteria for each mixed cell was estab-
lished, by contract, to be evaluated with unconfined compressive strength tests on 2″ and
3″ diameter samples taken at designated areas per cell location. The mix acceptance
strength was set at 65 psi at 28 days. The design mix strength requirement was based
upon limited finite element evaluations of anticipated loadings that the area would be
subject to during construction with the understanding that the majority of the lagoon
area would only receive four (4) ft of ISS which in essence would “float” on any
remaining un-mixed titanium dioxide waste materials and the highly compressible soils.
352 J.J. Serpico

Dry Portland cement was to be placed directly over a cell area by direct pumping from a
tanker truck using pneumatic methods then mixed in.
The ISS mixing started in August 2013 with an aggressive mobilization and within
the first four (4) months of ISS, 783 cells of the estimated 1,200 cells required for the
entire lagoon were mixed. However, acceptance rates for the first three (3) months were
less than 40%. Discussions with the contractor in the field did not yield any appreciable
change in procedures until the middle of November 2013 with introduction of a new Site
Superintendent for the contractor. At this time, mixing was essentially stopped until
March of 2014 while the contractor evaluated why the ISS results were not more
successful. Review of pH levels, individual excavator operator performance, water
added, mix depth compared to cell size and percent cement was conducted. Ultimately,
the contractor increased the Portland cement content to a little over 11% and agreed to
increase mix time duration. Figure 2 presents the number of cells mixed each month and
the percent passing on the first mix. As one can see the changes the contractor made
increased the success rate to what was anticipated within the contractor original bid
estimates which was approximately 80%. Unfortunately the mixing changes occurred
too late and the original contractor was forced, in July 2014 due to all the cell remixes
that were needed, to cease work and terminate the business. While a new contractor was
selected to complete the remaining work, the data was reviewed and it appeared that cell
size had a significant impact. For the Small, Medium and Large cell sizes the success
rate was 46%, 37% and 85%, respectively. The new contractor decided to accept the
findings and used a large cell size with 11% cement and the success rate was generally
greater than 80%. The final ISS cell was mixed in April 2015.

Fig. 2. Mixed cell acceptance rate per month

 Ground Improvement of Titanium Dioxide Waste Spoils 353

3.2 Surcharge Program

Once large areas of the lagoon were stabilized and approved, a site contractor started to
install the necessary elements of the surcharge program. In April 2015 the installation of
wick drains over a 28-acre area commenced. Due to time constraints for delivery of the
large retail store and associated parking, the wick drains were spaced between 4 and 5 ft
on a triangular grid. A total of 65,000 wick drains were installed. The material being
used as structural base fills and surcharge fill materials for the project are Processed
Dredge Materials (PDMs). Processed dredge materials typically consist of river and
harbor dredge spoils mixed with at least 8% cement prior to reuse as fills which make
them highly moisture sensitive and relatively impermeable. To account for the imper-
meable nature of the fills to be placed, the surcharge design included the use of strip
drains to collect wick drain seepage and piping to manholes with 18-inches of clean fine
gravels used as the drainage layer under the PDM materials. During wick drain instal-
lation, vibrating wire instrumentation consisting of (32) piezometers, eight (8) settlement
sensors and one (1) in-place inclinometer (5 sensor locations with depth) were installed.
All the instrumentation is wired to dataloggers which transmit real-time data to the
project website for contractor use. The real time data allows monitoring so the site pore
water pressures will not exceed established threshold limits. Additionally, (32) manual
settlement plates were installed and observed on a bimonthly basis.
Base fills were placed over the entire surcharge area to elevations greater than final
design grades to account for anticipated settlements. The PDM requires very thin lifts
due to the relatively high moisture content of the material as delivered to the site which
can range from 55% to 85% moisture. PDM use as a structural fill requires thin lifts
over large areas which are aerated prior to compaction with a sheepsfoot roller. Based
upon proctor tests (ASTM D-1557) conducted on the PDM materials, the optimum
moisture content ranged from 25 to 35% with maximum dry densities ranging from 65
to 85 lb per cubic foot.
The surcharge fill placement was broken down into two (2) separate phases; Phase 1
was to cover the area associated with the parking, roadway and utility needs for the large
retail store construction, and Phase II would utilize the surcharge fills from Phase I and
prepare the ground for future roadway, parking and culvert construction. In computing
the surcharge fill height for the areas the design had to account for the relatively low
anticipated moist unit weight of the PDM materials placed in thick lifts and any impact
of the ISS materials on surcharge load stress distribution. A final surcharge height of
(19) ft was used in construction which included a 5-foot increase in height to account for
any impacts of the ISS. Figure 3 presents the predicted versus actual settlement at
settlement plate location M07 which is located in the middle of the Phase I surcharge.
The observed settlement is close to the predicted total (primary) settlement and it
appears that the additional surcharge fill height added for the ISS concerns was nec-
essary. Surcharge fills were cleared for removal 4.5 months after the start of surcharge
placement which was estimated in the design of the wick drain spacing. Surcharge fills
are currently being removed from Phase I and are being placed on Phase II.
354 J.J. Serpico

Fig. 3. Settlement observations-M07

4 Conclusions

While this is a single case study for the use of in-place soil mixing for structural
support, there are many lessons to be learned from the field observations and testing.
1. Laboratory treatability studies, while important in establishing a baseline for design
and pricing, should be followed-up with actual treatability evaluations in the field.
Ideally, the contract should have a field treatability evaluation program as part of the
mobilization. Construction documents and contracts should be structured that if the
field observation success rate is not as anticipated, work should be stopped until
further evaluation is done.
2. Placement of the dry Portland cement by pneumatic pumping created an issue with
dust. The fine grained nature of the Portland cement meant that it would become
airborne with very little wind. Air monitoring is recommended with contract pro-
visions for dust remediation or work stoppage should air monitoring indicate
unacceptable levels of dust transport.
3. The time of mixing each cell had a significant impact on the success rate.
4. Cell size had a significant impact on the success rate. Small, relatively narrow cells
with deeper mixing depths had the lowest success rate. This is probably due to the
use of large track excavator physical shape limitations on how it can operate in a
confined, narrow trench. Larger cell sizes resulted in quicker mix times and higher
success rates.
5. The real-time instrumentation was very helpful in confirming design assumptions as
to stress distribution and saturated conditions. Piezometer data allowed for confir-
mation of pore-pressure distribution assumptions and construction rates. Settlement
sensors were good in evaluating relative settlement but generally were not as
accurate as the settlement values obtained from the manual settlement plates.
 Ground Improvement of Titanium Dioxide Waste Spoils 355

We were not able to evaluate if the settlement sensor readings were impacted by the
installation process or heavy construction traffic impacting the sensor. Future sur-
charge programs at the site will focus on this issue.

Acknowledgements. The author appreciates the support and input from the redevelopment
owner Brian O’Neill Jr., of the O’Neill Properties Group, the Civil/Site Project Manager for
Maser Consulting P.A, Daniel Busch, P.E., and the Senior Geologist for Maser Consulting P.A.,
William Bolton.
 Research and Application on the Prediction
Method of Pearl Model of High Filling
Subgrade Settlement

Xiangxing Kong(&)

The First Highway Survey and Design Institute of China Communications

Construction Company Ltd., Xi’an, China
[email protected]

Abstract. According to the development law of high embankment settlement,

the Pearl curve model is established. Combined with specific engineering
example, the Pearl curve well reflects the change process of high embankment
settlement, and the prediction result of the model with the measured settlement
value is very accurate and reliable.

1 Introduction

The construction of expressway in the west, high filling and excavation of roadbed
structure are generally occur [1, 2]. After completion construction of high fill
embankment subgrade settlement often appear, even cause the destruction of pavement
structure engineering accident [3]. In order to ensure the safety of the normal operation
of highway, according to the law of development of high filling subgrade settlement
observation data and combining site, Pearl model is applied to predict the settlement.

2 Mechanism of High Filling Subgrade Settlement

In the early period of construction, just after loading soil is in elastic status. The soil
pore water do not exhaust [4, 5]. Due to the soil of instantaneous shear de-formation,
with the increase of load the settlement has approximately linear increase. In the later
period of construction, with increasing load, the load imposed in high embankment are
also more and more big. The pore water in the foundation soil is gradually discharged,
meanwhile excess pore water pressure gradually decreases. So compaction causes more
de-formation, and soil in embankment will be in elastic-plastic state. With the con-
tinuous development of the plastic zone, settlement rate increases quickly, until the
load no longer increases.
After construction, due to the pore water pressure close to disappear completely, the
consolidation process is not complete, settlement will continue with the passage of
time, but significantly decreased sedimentation rate [6, 7]. If the time of settlement is
infinity, settlement will be at ultimate state, and settlement rate is zero, so the settlement
at that time is really the final settlement. In fact, for the highway time takes for 15 years
plus filling time.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_28
 Research and Application on the Prediction Method of Pearl Model 357

3 The Forecast Model of Pearl Curve

Pearl curve is sometimes referred as logistic curve, which is based on the American
biologists and demographer Raymond Pearl [8]. Because the curve can reflect the
growth process of biological, so Pearl curve is widely applied in animal breeding,
development of population statistics and analysis of product life cycle and so on. Pearl
curve prediction model is expressed as below.

L
yðtÞ ¼ ð1Þ
1 þ aebt

Where: L, a, b are three parameters of the model, in which a > 0, b > 0. Figure 1 is
the schematic diagram of typical Pearl curve.

Fig. 1. The carve of Pearl model

Seen from Fig. 1, the change of Pearl curve occurs in the point by the concave. The
point is the inflection point of the Pearl curve, which upper and lower halves are around
the inflection point symmetry. Also Pearl curve graph is a long strip of S, so it is
sometimes referred as S curve. It describes so regularly, and curve is relatively flat. The
description of Pearl curve actually reflects the process of things happening, develop-
ment, maturity and up to a certain limit. Because of the change of process and Pearl
curve settlement of high embankment reflects accurately the process of things devel-
opment, therefore Pearl curve model can reflect the law of development of high filling
subgrade settlement, and is applied to predict the settlement.

4 Calculation of Parameters

Using 3 parameters of Pearl curve model method to solve [5] has the following two
requirements:
358 X. Kong

(1) the settlement data items in the N time series is a multiple of 3, then the calcu-
lation time sequence can be divided into 3 segments, each segment containing
items;
(2) before the argument the time interval is equal or the length of time equal to, and
which is continuous and interval time series. The time of T are numbered
sequentially starting from 1, t = 1, 2, 3,…, n. According to the requirement, then
the number of items in time series is respectively (i = 1, 2,…, n). The time series
is divided into 3 sections: in the first paragraph t = 1, 2, 3,…; in the second
section, R, t = r + 1, r = 2, r + 3,…; in the third section t = 2r + 1, 2r + 2,
2r + 3, 2R,…, 3r.
Let S1, S2, S3 respectively for the 3 section numerical the sum, that is

Xr
1
S1 ¼ ð2Þ
t¼1
yðtÞ

X
2r
1
S2 ¼ ð3Þ
t¼r þ 1
yðtÞ

X
3r
1
S3 ¼ ð4Þ
t¼2r þ 1
yðtÞ

1 1 aebt
¼ þ ð5Þ
yðtÞ L L
ðs1 s2 Þ
ln ðs 2 s3 Þ
b¼ ð6Þ
r
r
L¼ ð7Þ
ðs1 s2 Þ2
s1  ðs1 s2 Þðs2 s3 Þ

ðs1  s2 Þ2 ð1  eb ÞL
a¼ ð8Þ
ððs1  s2 Þ  ðs2  s3 ÞÞeb ð1  erb Þ

So far 3 parameters are calculated by the formula (1), which can be obtained from
the Pearl prediction model. Based on the principle and the method of corresponding
calculation program is compiled with Matlab.
 Research and Application on the Prediction Method of Pearl Model 359

Fig. 2. Results of Pearl and forecasting and measured in K103+380

Table 1. Results of Pearl and forecasting and measured in K103+380

Time (d) Pearl forecasting (mm) Measured in site (mm) Absolute error (mm)
20 26.3828 23.9 2.5
40 32.9468 32.9 0.0
60 40.6426 46.3 5.7
80 49.4299 52.1 2.6
100 59.1656 60.6 1.5
120 69.5986 65.7 3.9
140 80.3868 80.9 0.5
160 91.1385 94.4 3.3
180 101.4672 106.3 4.9
200 111.0450 111.6 0.6
220 119.6399 114.9 4.7
240 127.1288 124.6 2.5
260 133.4881 132.7 0.8
280 138.7712 137.0 1.8
300 143.0809 142.3 0.8
320 146.5447 145.2 1.3
340 149.2954 147.1 2.2
360 151.4591 148.0 3.5
380 153.1483 150.2 2.9
400 154.4594 151.4 3.1
360 X. Kong

Table 2. Results of Pearl and forecasting and measured in K123+560

Time (d) Pearl forecasting (mm) Measured in site (mm) Absolute error (mm)
20 21.48777 21.755 0.2813
40 29.61882 30.305 0.7223
60 38.71763 43.035 4.5446
80 48.45608 48.545 0.0936
100 58.49492 56.62 1.9736
120 68.52094 61.465 7.4273
140 78.27012 75.905 2.4896
160 87.53908 88.73 1.2536
180 96.18579 100.035 4.0518
200 104.1238 105.07 0.996
220 111.3138 108.205 3.2724
240 117.7523 117.42 0.3498
260 123.4631 125.115 1.7388
280 128.4877 129.2 0.7498
300 132.8785 134.235 1.4279
320 136.6934 136.99 0.3122
340 139.9921 138.795 1.2601
360 142.8328 139.65 3.3503
380 145.2708 141.74 3.7166
400 147.3569 142.88 4.7125

Fig. 3. Results of Pearl and forecasting and measured in K123+560

 Research and Application on the Prediction Method of Pearl Model 361

Table 3. Parameters of Pearl method

L a b
158.7895 6.5942 0.2730

5 Engineering Examples

Combined with the subgrade settlement observation data in Hunan province

Changde-Zhangjiajie expressway, the K103+380, K123+560 are chosen as the case
study. The high fill subgrade filler is typical, where is filled with low liquid limit clay
and shale, an embankment height is about 13 m.
In Fig. 2, the time-scattered settlement is “S” shape. The first 12 points (20–240 days)
data are selected as samples, Matlab6.5 is used to calculate related parameters of Pearl
curve model (as shown in Table 1), then the results of Pearl curve model for settlement
prediction of roadbed (as shown in Table 2) are obtained, as well as the sum squared
error (prediction precision) 168.13 (Fig. 3 and Table 3).

6 Conclusion

In the forecasting of road foundation settlement, different methods can often provide
valuable information.
(1) The settlement of high embankment has its own law of development, and Pearl
curve is a good reflection of S shape relationship between settlement and time of
the whole process of roadbed settlement, mean-while its settlement prediction is
accurate and reliable.
(2) Using Pearl curve model in forecasting, in order to obtain good prediction suf-
ficient measured data is basis.
(3) The aging situation also exists in the Pearl prediction model, and with time go
recent settlement should be add as the original sample, so that the prediction effect
is more better.

References
1. Xu, J., Cheng, K.K.: Road Engineering. Tongji University Press, Shanghai (1995)
2. Highway Traffic Division: The Highway Engineering Quality Problems Prevent Guide.
China Communications Press, Beijing (2002)
3. Gu, X., Qian, H.J., et al.: The Foundation and Foundation. China Architecture Industry
Press, Beijing (1993)
4. Zai, J., Mei, K.: Settlement of the whole process of prediction method. Rock Soil Mech. 21
(4), l 322–325 (2000)
5. Zhao, M., Liu, Y., et al.: The soft soil roadbed settlement development law and its
forecasting. J. Cent. South Univ. 35(1), 157–161 (2004)
362 X. Kong

6. Pi, D.Y., Sun, Y.: Control and decision of multi model adaptive control. J. Algorithm 11(1),
77–80 (1996)
7. Bates, J.M., Granger, C.W.J.: The combination of forecasts. J. Oper. Res. Soc. 20(4), 451–468
(1969). doi:10.2307/3008764. http://www.jstor.org/stable/3008764
8. Tang, X., Wang, J., et al.: A new combination forecasting method of adaptive fuzzy variable
weight. J. Univ. Electron. Sci. Technol. Algorithm 26(3), 289–291 (1997)
 Research on the Prediction Method
of Gompertz Model for High Subgrade
Settlement

Xiangxing Kong(&)

The First Highway Survey and Design Institute of China Communications

Construction Company Ltd., Xi’an, China
[email protected]

Abstract. According to the development law of high embankment settlement,

the Gompertz curve model is established. Combined with specific engineering
example, the Gompertz curve well reflects the change process of high
embankment settlement, and the prediction result of the model with the mea-
sured settlement value is very accurate and reliable.

1 Introduction

The construction of expressway in the west, high filling and excavation of roadbed
structure are generally occur [1, 2]. After completion construction of high fill
embankment subgrade settlement often appear, even cause the destruction of pavement
structure engineering accident [3]. In order to ensure the safety of the normal operation
of highway, according to the law of development of high filling subgrade settlement
observation data and combining site, Gompertz curve model is applied to predict the
settlement of the roadbed.

2 Mechanism of High Filling Subgrade Settlement

In the early period of construction, just after loading soil is in elastic status. The soil pore
water do not exhaust [4, 5]. Due to the soil of instantaneous shear de-formation, with the
increase of load the settlement has approximately linear increase. In the later period of
construction, with increasing load, the load imposed in high embankment are also more
and more big. The pore water in the foundation soil is gradually discharged, meanwhile
excess pore water pressure gradually decreases. So compaction causes more de-formation,
and soil in embankment will be in elastic-plastic state. With the continuous development of
the plastic zone, settlement rate increases quickly, until the load no longer increases.
After construction, due to the pore water pressure close to disappear completely, the
consolidation process is not complete, settlement will continue with the passage of
time, but significantly decreased sedimentation rate. If the time of settlement is infinity,
settlement will be at ultimate state, and settlement rate is zero, so the settlement at that
time is really the final settlement. In fact, time generally is large enough, for the
highway time takes for 15 years plus filling time.

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_29
364 X. Kong

3 The Forecast Model of Gompertz Curve

At present, there are two categories in calculation method of settlement of embank-

ment. First, according to the constitutive model of consolidation theory combined with
various soil, it is settlement calculation of the application of numerical methods, such
as visco-elastoplastic finite element model method [6, 7]. The method of the calculation
of the finite element method is very high, generally only used to calculate the key end
of large engineering. And some constitutive model and engineering practice has a large
gap, and needs a large number of soil tests to determine the soil parameters, so it is
difficult to use in the general engineering design. Second, based on the measured data it
is calculated settlement prediction method of quantity and time relations, such as
hyperbola and exponential curve method. This method is concise and practical, and can
satisfy the engineering precision requirement. Method for forecasting subgrade set-
tlement of this selection is one of the Gompertz model.
Gompertz method is proposed by statisticians and mathematicians B. Gompertz,
and it is also a growth curve, which are widely used in economics, management and
statistics. Gompertz forecasting model is a growth curve model, and can be used to
express and describe the process of growth and development. The practice indicated
prediction model can be used as a reliable, effective method to forecast the settlement
prediction of roadbed.

4 Calculation of Parameters

The three parameters in the model can be obtained using the three estimation. First of
all, make it into a modified exponential form, then the data is divided into three sections
(if not divisible by 3, can increase and decrease the individual data which is a multiple
of 3), each containing a number of segments, i.e., seek and obtain.

yt ¼ eðk þ ab Þ
t
ð1Þ

y0 ¼ k þ abt ð2Þ

y1 ¼ k þ ab1 ð3Þ

y2 ¼ k þ ab2 ð4Þ

y3 ¼ k þ ab3 ð5Þ

yt ¼ k þ abT ð6Þ

X X
n
y ¼
1 t
yt ¼ nk þ abðb0 þ b1 þ b2 þ    þ bn1 Þ ð7Þ
t¼1
 Research on the Prediction Method of Gompertz Model 365
X X
2n
y ¼
2 t
yt ¼ nk þ abn þ 1 ðb0 þ b1 þ b2 þ    þ bn1 Þ ð8Þ
t¼n þ 1

X X
3n
y ¼
3 t
yt ¼ nk þ ab2n þ 1 ðb0 þ b1 þ b2 þ    þ bn1 Þ ð9Þ
t¼2n þ 1

bn  1
ðb0 þ b1 þ b2 þ    þ bn1 Þ ¼ ð10Þ
b1
X bn  1
y
1 t
¼ nk þ ab ð11Þ
b1
X bn  1
y
2 t
¼ nk þ abn þ 1 ð12Þ
b1
X bn  1
y ¼ nK þ ab2n þ 1
3 t
ð13Þ
b1
sPffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
yt  2 yt
b¼ P P ð14Þ
n 3
2 yt  1 yt

b1 X X
a¼ 2
ð y
2 t
 yÞ
1 t
ð15Þ
ðbn  1Þ b

1 X bn  1
k ¼ ð 1 yt  ab Þ ð16Þ
n b1
"P P P #
1 yt 3 yt  ð 2 y t Þ 2
k¼ P 1 P P ð17Þ
n 1 yt þ 3 yt  2 2 yt

By using the method of analogy, Gompertz prediction parameters can be calculated as

below:

b1 X X
a¼ 2
ð ln yt  ln yt Þ ð18Þ
ðbn  1Þ b 2 1

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P P
ln yt  2 ln yt
b¼ P P ð19Þ
n 3
2 ln yt  1 ln yt

"P P P #
1 ln yt 3 ln yt  ð 2 ln yt Þ2
k¼ P 1 P P ð20Þ
n 1 ln yt þ 3 ln yt  2 2 ln yt
366 X. Kong

5 Engineering Examples

Combined with the subgrade settlement observation data in Hunan province

Changde-Zhangjiajie expressway, the K103+380, K123+560 are chosen as the case
study. The high fill subgrade filler is typical, where is filled with low liquid limit clay
and shale, an embankment height is about 13 m (Fig. 1).

Fig. 1. Results of Gompertz and forecasting and measured in K123+560

Fig. 2. Results of Pearl and forecasting and measured in K103+380

 Research on the Prediction Method of Gompertz Model 367

In Fig. 2, the time-scattered settlement is “S” shape. The first 12 points (20–240
days) data are selected as samples, Matlab6.5 is used to calculate related parameters of
Gompertz curve model (as shown in Table 1), then the results of Gompertz curve
model for settlement prediction of roadbed (as shown in Table 2) are obtained, as well
as the sum squared error (prediction precision) 168.13 (Table 3).

Table 1. Results of Gompertz and forecasting and measured in K123+560

Time (d) Gompertz forecasting (mm) Measured in site (mm) Absolute error (mm)
20 23.6187 23.9 0.3
40 32.1777 32.9 0.7
60 41.7554 46.3 4.6
80 52.0064 52.1 0.1
100 62.5736 60.6 2.0
120 73.1273 65.7 7.5
140 83.3896 80.9 2.5
160 93.1464 94.4 1.3
180 102.2482 106.3 4.1
200 110.6040 111.6 1.0
220 118.1724 114.9 3.2
240 124.9498 124.6 0.3
260 130.9612 132.7 1.7
280 136.2502 137.0 0.7
300 140.8721 142.3 1.4
320 144.8878 145.2 0.3
340 148.3601 147.1 1.3
360 151.3503 148.0 3.3
380 153.9166 150.2 3.7
400 156.1125 151.4 4.8

Table 2. Results of Pearl and forecasting and measured in K103+380

Time (d) Pearl forecasting (mm) Measured in site (mm) Absolute error (mm)
20 25.32749 22.944 2.383488
40 31.62893 31.584 0.044928
60 39.0169 44.448 5.4311
80 47.4527 50.016 2.5633
100 56.79898 58.176 1.37702
120 66.81466 63.072 3.742656
140 77.17133 77.664 0.49267
160 87.49296 90.624 3.13104
(continued)
368 X. Kong

Table 2. (continued)
Time (d) Pearl forecasting (mm) Measured in site (mm) Absolute error (mm)
180 97.40851 102.048 4.63949
200 106.6032 107.136 0.5328
220 114.8543 110.304 4.550304
240 122.0436 119.616 2.427648
260 128.1486 127.392 0.756576
280 133.2204 131.52 1.700352
300 137.3577 136.608 0.749664
320 140.6829 139.392 1.290912
340 143.3236 141.216 2.107584
360 145.4007 142.08 3.320736
380 147.0224 144.192 2.830368
400 148.281 145.344 2.937024

Table 3. Parameters of Gompertz method

K a b
5.1264 −2.3314 0.8426

6 Conclusion

In the forecasting of road foundation settlement, different methods can often provide
valuable information.
(1) The settlement of high embankment has its own law of development, Gompertz
model can express S shape relationship between settlement and time of the whole
process of subgrade. The settlement reflects very good, but also can be accurately
applied in prediction of high embankment.
(2) Using the Gompertz model to forecast the settlement of roadbed, in order to obtain
good prediction effect enough measured data is basis. The measured settlement
value should be checked, and delete obvious error.
(3) In order to improve the prediction accuracy, the selection of measured sample
points should be increased, and with the passage of time, the settlement of recent
is added as the original sample, so that the prediction effect is better.

References
1. Highway Traffic Division: The Highway Engineering Quality Problems Prevent Guide. China
Communications Press, Beijing (2002)
2. Xu, J., Cheng, K.K.: Road Engineering. Tongji University Press, Shanghai (1995)
 Research on the Prediction Method of Gompertz Model 369

3. Zai, J.M., Mei, K.: Settlement of the whole process of prediction method. Rock Soil Mech.
21(4), 322–325 (2000)
4. Gu, X., Qian, H.J., et al.: The Foundation and Foundation. China Architecture Industry Press,
Beijing (1993)
5. Pi, D.Y., Sun, Y.: Control and decision of multi model adaptive control. J. Algorithm 11(1),
77–80 (1996)
6. Zhao, M., Liu, Y., et al.: The soft soil roadbed settlement development law and its forecasting.
J. Cent. South Univ. 35(1), 157–161 (2004)
7. Bates, J.M., Granger, C.W.J.: The combination of forecasts. J. Oper. Res. Soc. 20(4), 451–468
(1969). doi:10.2307/3008764. http://www.jstor.org/stable/3008764
 Study on Methods for Predicting
the Settlements of Soft Clay Roadbed

Baotong Shi and Xiangxing Kong(&)

The First Highway Survey and Design Institute of China Communications

Construction Company Ltd., Xi’an, China
{117021654,103121153}@qq.com

Abstract. The soft clay has the characteristics of high compressibility, high
water content, low strength and poor permeability, which will lead to large
settlements under long-term loads, therefore, the prediction of settlements of soft
clay roadbed has been one of the important research topics in Geotechnical
Engineering. In this paper, the commonly-used methods in practice to predict
the settlements of soft clay roadbed have been summarized, and the advantages
and disadvantages have been evaluated.

1 Introduction

The soft clay has the characteristics of high compressibility, high water content, low
strength and poor permeability, which will lead to large settlements under long-term
loads. Therefore, the settlement of soft clay roadbed is one of the prominent problems
during the construction of railway and highway in soft clay area, and many engineering
accidents are caused by the large settlement of subgrade. Thus, the prediction of
settlements of soft clay roadbed has been one of the important research topics in
Geotechnical Engineering.
As the settlement of soft clay usually develops with time, the settlement vs. time
curve is usually used to describe the settlement properties of soft clay [1–3]. The most
common method to predict the long term settlement is fitting the settlement data
measured in the field. Many traditional fitting curves have been used to predict the
settlement of soft clay, such as exponential curve, hyperbolic curve, Poisson curve and
Asaoka method [4, 5]. Furthermore, in recent years, more and more new methods based
on new theories have been introduced to improve the prediction of settlement, such as
Grey forecasting method [6], Logistic method [7], and Gompertz model [8]. In this
paper, the commonly used methods mentioned above to predict the settlement of soft
clay have been evaluated and compared with each other.

2 Mechanism of the Settlement of Soft Clay Roadbed

The total settlement of soft clay roadbed contains three components: immediate set-
tlement, consolidation settlement and secondary consolidation settlement. The devel-
opment of settlement can be divided into the following four stages (as shown in Fig. 1):

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_30
 Study on Methods for Predicting the Settlements 371

Fig. 1. Sketch view of settlement vs. time

(1) at the initial status, the soil is elastic and the change of settlement with load is
linear;
(2) then with the increase of load, the settlement of soft clay and the corresponding
rate increase and the nonlinear is obvious;
(3) although the load keeps constant, the settlement continues to increase due to
consolidation and rheological property of clay, but the settlement rate declines;
(4) when the time is long enough, the settlement of clay will reach a limit state, the
corresponding settlement rate is zero and the settlement arrives at its final value.
The purpose to study the properties of soft clay is to develop theoretical or
empirical models to predict the settlement of soft clay at any time and any load. In
recent years, many fitting curve methods based on measured field settlement data have
been proposed, we will introduce these models detailedly in the next section.

3 Fitting Curve Methods

3.1 Exponential Curve

According to the consolidation theory proposed by Terzaghi, the relation between pore
water pressure with time can be described by exponential expression. For soil with
linear elastic property, the consolidation degree of soil defined by using stress equals to
that defined by using strain. the consolidation degree can be expressed as follows:

U ¼ 1  a exp ðbtÞ ð1Þ

Where a and b are consolidation parameters, of which the expressions relate to

drainage condition and are shown in Table 1.
Then according to the measured settlement data, the average consolidation degree
can be expressed as,
372 B. Shi and X. Kong

Table 1. Theoretical values of a and b

Drainage condition a b
Consolidation with vertical drainage 8 p2 cv
p2 4H 2
8 p2 cv
Drainage wells run through the soil layer p2
8ch
Fn de2 þ 4H 2
Drainage wells do not run through the soil layer 8 8ch
p2 q Fn de2

st  sd
U¼ ð2aÞ
sf  sd

Where st is the measured settlement at time t, sd and sf are immediate consolidation

settlement and final consolidation settlement.
Combined Eq. (1) with Eq. (2a), the following expression can be obtained,

st ¼ ðsf  sd Þ½1  a expðbtÞ þ sd ð2bÞ

In exponential curve fitting method, three data points on the settlement vs. time
curve are selected and the time interval between the points must be the same, that is,

Dt ¼ t3  t2 ¼ t2  t1 ð3Þ

Then put the three points into Eq. (2b), three equations can be obtained and the
expressions of b, sf, and sd are shown as follows,

1 s2  s1
b¼ lnð Þ ð4aÞ
Dt s3  s2

s3 ðs2  s1 Þ  s2 ðs3  s1 Þ
sf ¼ ð4bÞ
ðs2  s1 Þ  ðs3  s2 Þ

st  sf ½1  a expðbtÞ
sd ¼ ð4cÞ
a expðbtÞ

The above equations are the controlling equations in exponential curve.

3.2 Hyperbolic Curve

For soft clay with high compressibility, the stress vs. strain curve is non-linear, so the
consolidation process of soft clay does not necessarily conform exponential curve.
 Study on Methods for Predicting the Settlements 373

When 0.6 < U < 0.9, the relation between U and time factor T can be expressed by
equilateral hyperbola. The theory background of hyperbolic curve is given as follows:
The settlement of soft clay at time t can be calculated by the following equation,
0
0 t  t0
s ¼ s0 þ 0 0 0
ð5Þ
a0 þ b ðt  t0 Þ

Rewriting Eq. (5), then,

0
t  t0 0 0 0
0 ¼ a0 þ b ðt  t0 Þ ð6Þ
s  s0

Where and are the intercept and slope of vs. curve.

3.3 Asaoka Method

The one dimensional consolidation equation described by using volumetric strain can
be expressed as follows,

@ 2 ev @ev
cv ¼ ð7Þ
@z2 @t

On the basis of Eq. (7), Asaoka employed a series to approximate the expression,

ds d2s dns
s þ a1 þ a2 2 þ . . . þ an n ¼ b ð8Þ
dt dt dt

Where s is consolidation settlement, ai(i = 1, 2, 3, …, n) is consolidation coeffi-

cient, b is a constant accounting for boundary condition.
According to Asaoka’ suggestion, the parameters in Eq. (8) can be determined by
using field measured data. Furthermore, Eq. (7) can also be simplified as recursive
relation as follows,

X
n
sj ¼ b0 þ bi sj1 ð9Þ
i¼1

Taking the fixed boundary condition, the solution for the equation is given by,

t
sðtÞ ¼ s1  ðs1  s0 Þ expð Þ ð10Þ
a1

where s0 and are the initial and final settlement of soft clay, respectively, a1 = 5h2/
(12cv).
374 B. Shi and X. Kong

3.4 Poisson Curve

Poisson curve is the logistic curve, which is suitable for incremental or decayed “S”
shaped curve. The expression of the commonly-used Poisson curve is given by,

k
st ¼ ð11Þ
1 þ aebt

where a, b and k are parameters that should be determined by measure data.

The parameters in the expression of Poisson curve can be fitted by using the
measured st curve, the main procedures are shown as follows:
Taking a time interval of and the total number of time series is 3n, then divide the time
series into three sub-series and each sub-series has a number of n data points, that is,
8
< sub  series 1 t ¼ 1; 2; 3; . . .; n
sub  series 2 t ¼ n þ 1; n þ 2; n þ 3; . . .; 2n ð12Þ
:
sub  series 3 t ¼ 2n þ 1; 2n þ 2; 2n þ 3; . . .; 3n

Then assuming that S1, S2 and S3 are the sum of the reciprocal of settlements
corresponding to each time sub-series, the expressions to determine the parameters a, b
and k are given as follows,
8 2 b
>
> a ¼ ½ðS1 SðS 1 S2 Þ ð1e Þk
< b
2 ÞðS2 S3 Þe ð1e
nb Þ

b ¼ lnðS1 S2 Þ=ðS 2 S3 Þ

ð13Þ
>
>
n
:k ¼ n
ðS1 S2 Þ2
S1 ðS
1 S2 ÞðS2 S3 Þ

Combining Eq. (13) with Eq. (12), the settlement of clay at any time st can be
calculated, and the final settlement equals to the ultimate value of st, when the time t
trends to infinity, that is,

s1 ¼ lim st ¼ k ð14Þ
t!1

3.5 Grey Forecasting Method

The Grey forecasting method is always referred to as GM model, and the most common
GM Model has isometric time sequence, that is, GM(1, 1) model, the expression of
which is given by the following differential equation,

dxð1Þ
þ cxð1Þ ¼ u ð15Þ
dt
 Study on Methods for Predicting the Settlements 375

The corresponding time response equation is,

u u
^xð1Þ ðtÞ ¼ ½xð1Þ ð0Þ  ect þ ð16Þ
c c
The discrete response equation is,
u u
^xð1Þ ðkÞ ¼ ½xð1Þ ð0Þ  ecðk1Þ þ ð17Þ
c c

3.6 Logistic Model

Logistic model is a kind of growth model that is widely used in the fields of ecology
and demography. The differential form of logistic model is as follows,

ds s
¼ rsð1  Þ ð18Þ
dt K

According to Eq. (18), Logistic curve contains three parameters that should be
determined. If the measured settlement data can be obtained, there are three methods
used to determine these three parameters: three section method, grey model and non-
linear regression method. The main processes of the three methods can be found in
other literatures.

3.7 Gompertz Model

Gompertz model is also a kind of growth curve, the mathematical expression of which
shows as follows,

Fig. 2. Typical Gompertz curve

376 B. Shi and X. Kong

y ¼ m exp½eðnltÞ  ð19Þ

where m, n and l are constants, t is time series.

The typical Gompertz curve is shown in Fig. 2. From the figure, it is found that the
theoretical curve agrees well with measured settlement curve. Therefore, it is feasible to
use Gompertz model to predict the settlement of soft clay.

4 Discussions

In Sect. 3, seven methods to predict the settlement of soft clay have been reviewed
detailedly. Although all of them have been applied to predict the settlements of some
highway subgrade successfully, each method has its own application scope.
The exponential curve conforms to the one-dimensional consolidation theory
proposed by Terzaghi and the parameter b has definite physical meaning. However, the
method to determine the parameters a and b based on three data points is subjective,
which will influence the accuracy.
According to Terzaghi’s one dimensional consolidation theory, the hyperbolic
method will make the relation between consolidation degree U and time factor T
simpler than the exponential curve. Therefore, the hyperbolic method is more suitable
for engineers in the field. However, the disadvantage of this method is that the con-
solidation parameters can not be reflected.
Compared with the consolidation of Terzaghi, it is found that Asaoka method
predicts the settlement trends once the degree of consolidation exceeds 0.6. Further-
more, the final settlement of the soft clay can also be predicted well.
The Poisson curve can describe well the S-shaped relationship between settlement
and time. The applicability of Poisson curve for other settlement vs. time curves is
questionable and should be given more attentions.
The four methods mentioned above are traditional methods used for predicting the
settlement of soft clay, while the Grey forecasting method, Logistic model and
Gompertz model are new methods suitable for special kind of soft clay or different
boundary conditions.

5 Conclusion

In this paper, seven different methods for predicting the settlement of soft clay roadbed
have been reviewed detailedly based on the presentation of the settlement mechanism
of soft clay. The applicability of each method has been analyzed. Although all the
methods introduced in this paper have been applied into practices, the accuracy of each
method should be studied in further researches.
 Study on Methods for Predicting the Settlements 377

References
1. Zhao, M., Liu, Y., Cao, W.: The developing regularity and forecasting of settlement in soft
clay roadbed. J. Cent. South University (Nat. Sci.) 35(1), 157–161 (2004)
2. Chen, S., Wang, X., Xu, X., Yu, F., Qin, S.: Three-point modified exponential curve method
for predicting subgrade settlements. Rock Soil Mech. 32(11), 3355–3360 (2011)
3. Zai, J., Mei, G.: Forecast method of settlement during the complete process of construction
and operation. Rock Soil Mech. 21(4), 322–325 (2000)
4. Tan, S.A.: Hyperbolic method for settlement in clays with vertical drains. Can. Geotech. J.
31(1), 125–131 (1994)
5. Asaoka, A.: Observational procedure of settlement prediction. Soil Found. 18(4), 87–101
(1978)
6. Lei, X., Bai, S., Meng, Q.: The application of Grey forecasting to analyzing soft foundation
settlements. Rock Soil Mech. 21(2), 145–147 (2000)
7. Xu, H., Shi, B., Li, X.: Logistic growth model and its applicability for predicting settlement
during the whole process. Rock Soil Mech. 26(3), 387–391 (2005)
8. Yu, C., Liu, S.: A study on prediction of embankment settlement with the Gompertz model.
Rock Soil Mech. 26(4), 82–86 (2005)
 Geotechnical Investigations on Highway
BR 135 for Earthworks, Asphalt Paving
and Special Art Works in Subsection
Municipalities Manga-Itacarambi,
North of Minas Gerais

Rideci Farias1(&), Haroldo Paranhos1, José Jailson Nogueira2,

Marco Aurélio Diana Costa3, and Roberto de Pimentel Sousa Junior4
1
UCB/Reforsolo Engenharia/UniCEUB/IesPlan, Brasília, Brazil
[email protected], [email protected]
2
VALEC, Brasília, Brazil
[email protected]
3
UniCEUB, Brasília, Brazil
[email protected]
4
UCB/UNB, Brasília, Brazil
[email protected]

Abstract. This article presents part of the results and discussions of the
geotechnical studies carried out in the area and in materials collected in the BR
135 in the section between the counties of Manga and Itacarambi in the state of
Minas Gerais. The study consisted in the execution of field work and laboratory
work. In the field investigations were performed (auger, SPT and rotary), col-
lecting material (soil and rock), densities “in situ” and identification/tactile-visual
classification of materials. In the laboratory, with soils and rocks collected, were
performed additional tests necessary to the design and implementation of the road
in question, in that stretch. The studies focused along the existing highway axis
and the track side, and side guidance in the predicted contours, concentrated
deposits of soil, deposits concentrated of gravel, sands deposits and deposits of
rocky material. With implementation of Auger investigations, it was found that
the soils tactile-visual characteristics are similar to the predominance of
yellow/red sandy soils. It was found that along the highway there are competent
rocky material and also friable, mixed, which are not local, but come from deposits
in the region and which were used in the primary coating of the road. In the
carrying out field work, was done a broad recognition of the BR 135 insertion
region that recognition was found that the region is covered by a wide layer of
typical sandy soil (Red and Yellow Latosol). In the case of rocky material, it was
found that there is a high occurrence of carbonate rocks in the region. This rock is
exploited and used as building material, especially on pavements and as aggre-
gates for concrete (Gravel 1, 2 and Gravel Gravel powder). Being a characteristic
rock materials in the region, it was used for granulometric tests of sand equiva-
lency, organic matter content, form of content, durability, abrasion “Los Angeles”
and adhesiveness. The check of sands deposits, local people reported that most of
the sand used in the region comes from dredging the San Francisco River in the

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_31
 Geotechnical Investigations on Highway BR 135 379

municipality of Itacarambi. two locations sands were also collected in the

municipality of Mango, also the banks of the São Francisco River.

Keywords: Geotechnical investigation
BR 135
Manga
Itacarambi

1 Introduction

This article aims to present some of the results of the geotechnical studies carried out in
the area and collected materials on the BR 135 in the stretch between the towns of Manga
and Itacarambi in the State of Minas Gerais. The studies consisted in performing field
work and laboratory. On the field were executed investigations (Auger, SPT and rotat-
ing), collecting material (soil and rock), in situ density and identification/classification of
materials. In the lab were carried out further testing required for the design and imple-
mentation of the road between the municipalities Manga and Itacarambi, State of Minas
Gerais.
The studies were concentrated along the existing axis and in the sides of the
highway to be paved in the guideline and sides of the contours to be executed, con-
centrated deposits of soil, concentrated deposits of gravel, sand deposits and deposits of
stonelike material.
The field work was conducted in the months of June, July and November, and the
laboratory work were executed in the months of July, August, November and
December 2010.
The field studies were reconciled with the Office so that the information they were
consistent with the topographic study provided.
With the implementation of the Auger investigation, it was found that the
tactile-visual characteristics soils similar to the predominance of yellow/red sandy soils.

2 Location Studies of Execution Area

The area of execution of study is located between the municipalities of Manga and
Itacarambi in the state of Minas Gerais. In terms of organization, the area was divided
into four sections, and was classified as follows:
Stretch 1: Segment between the municipalities of Manga and São João das Missões;
Stretch 2: Segment between the City of São João das Missões to the start of paved
stretch, passing for the small city called Rancharia;
Stretch 3: Manga municipality contour;
Stretch 4: Outline of the City of São João das Missões;
Figures 1 and 2, from Google Earth, show the insertion area of the enterprise in
question.
380 R. Farias et al.

Fig. 1. Macro view of the studies area

Fig. 2. Macro view of studies running area (Excerpt: Manga - Itacarambi).

 Geotechnical Investigations on Highway BR 135 381

Figures 3 and 4, also from Google Earth, present, respectively, the outlines pro-
vided in the municipalities of Manga, and São João da Missões.

Fig. 3. View contouring Manga/MG. Home (approximate coordinates: Start 14º 44′04.6″ S,
043º 58′ 36.9″ W, and end 14 46′25.8″ S, 043º 58′ 19.1″ W).

Fig. 4. View the outline of St. John of the Missions. Home (approximate coordinates: Home 14º
52′15,875″ S, 044º 04′ 44,751″ W, and end 14 53′20.6″ S, 044º 04′ 39.3″ W).

3 General on the Region of Materials

In carrying out of field work was done a detailed recognition of the region of insertion
of BR 135. In recognition was found that the region is covered by a wide layer of
typical sandy soil (Red Latosol and Yellow Latosol), Figs. 5 and 6. Along the track
382 R. Farias et al.

Fig. 5. Material of the primer coating and typical sandy soil found in the region.

Fig. 6. Start of Manga-São João das missões. Also observe the implementation of the survey
auger.

axis and contours, has the typical profile shown in Figs. 7, 8, 9 and 10. Samples of the
surface layers of these profiles were collected but not used in the implementation of
laboratory tests. For soil concentrated study, we used an area near the municipality of
São João da Missões.
In the case of rocky material, it was found that there is a high occurrence of
carbonate rocks in the region. This material is exploited and used as building material,
especially on pavements and as aggregates for concrete (Gravel 1, 2 and Gravel
powder).
 Geotechnical Investigations on Highway BR 135 383

Fig. 7. Typical via banks. Manga-São João das Missões. (GPS 14º 50′28.6″ S, 44º 03′ 12″ W)

Fig. 8. Excerpt from Manga-São João das Missões. Observe typical rocky material in the
surface layer.

Note: It is verified that along the way there is the existence of competent rocky
material and also friable, mixed, which are not local, but come from deposits in the
region and which were used in the primary coating of the road.
Aiming the possible deployment of a bridge on the outline provided in São João das
384 R. Farias et al.

Fig. 9. Excerpt from Manga-São João das Missões. Detail of the typical rocky material in the
surface layer.

Fig. 10. Typical profiles prevailing in the local collection of soil samples

Missions, mixed investigations were performed (SPT + Rotating), Figs. 11 and 12.
 Geotechnical Investigations on Highway BR 135 385

4 Results of Studies Performed

Fig. 11. Implementation of the SPT survey on the river Itacarambi possible to bridge the
implementation outline of São Joao Mission.

Fig. 12. Execution of the rotary drilling on the river Itacarambi possible to bridge the
implementation outline of São Joao Mission.

As already mentioned, the study consisted in the execution of field work and labora-
tory. In the field site investigations were performed (auger and rotary SPT), collecting
material (soil and rock), densities “in situ” and identification/material classification. In
the laboratory, with soils and rocks collected, the following tests were performed:
(a) moisture content; (b) Liquidity Limits and Plasticity; (c) Granulometri by sieving
and sedimentation, using as a dispersant sodium hexametaphosphate; (d) real density of
386 R. Farias et al.

soil grains; (e) compression; (f) California Support Index (ISC or CBR); (g) content of
organic matter; (h) Sand Equivalency; (i) Abrasion “Los Angeles”; (j) Adhesion;
(k) Durability; (l) index of form; (m) Material Classification; (n) Analysis of SPT;
(o) Analysis of Rotating SPT. The following part summaries are presented the results of
studies performed.

4.1 Soil Collected in Axis of Projected via

Figure 13 shows a summary of the average of the physical and mechanical charac-
teristics of the soil along the Via to be paved. Through this figure, has the follow
results: in granulometric terms the predominant soil is sandy; not liquid and not plastic;
Sand equivalency equal to 33%; Group Index equal to zero; The predominant soil-2-4

Fig. 13. Summary and physical characteristics and soil mechanics for kids stretch (North)
 Geotechnical Investigations on Highway BR 135 387

(gravel or silty or clayey sands) the classification TRB (HRB); optimum moisture
content in the normal energy equal to 8%; ISC (CBR) 18%; Expansion of the com-
pacted material after 96 h of immersion of 0.86%; apparent density of 1,77 g/cm3;
natural moisture of 1.86%.

4.2 Deposits of Soil and Gravel

For areas of lending materials, we studied a concentrated deposit of soil and three
deposits of gravel. The field soil has similar characteristics to that shown in Fig. 13.
The soil collected in this field was also used to study soil-gravel mixture. Two Soil
Gravel mixtures were studied. A mixture of 20% and another 30% Gravel, being
respectively 80% and 70% of soil. Figures 14 and 15 show the results obtained for
these mixtures (soil and gravel). In the case of compression it is found that the normal

Fig. 14. Study summary for soil mix (80%) - Gravel (20%).
388 R. Farias et al.

Fig. 15. Study summary for soil mix (70%) - Gravel (30%).

power sample with 20% dry gravel has optimum moisture content and density,
respectively 2,03 g/cm3 and 8.70%. Already 30% of crushed stone features 2.14 g/cm3
and 7.49% of optimum moisture.
Next, in Fig. 16, is shows one of the gravel deposits studied, and Fig. 17 presents a
summary of this study for this deposit. It is observed that in terms of compacting the
material has medium for optimum moisture content and maximum dry density in
normal power (13.2%; 1.90 g/cm3), intermediate (12.7%; 1.96 g/cm3) and modified
(12.0%; 2.01 g/cm3).
 Geotechnical Investigations on Highway BR 135 389

Fig. 16. View of the gravel deposits studied.

Fig. 17. Study summary for the gravel mine.

390 R. Farias et al.

4.3 Sands Deposits

During the check of sands deposits, local people reported that most of the sand used in
the region comes from the dredging of the river San Francisco in the municipality of
Itacarambi. On the way to the site (Fig. 18) and in conversations, it was found that are
marketed three kinds of sand called by traders “fine sand”, “pebble” and “sand with
pebble”, as shown in Fig. 19. In addition to these in two locations, sands were collected
in Manga municipality, also in the margins of the São Francisco river.

Fig. 18. Sand dredgers area view in Itacarambi/MG on the banks of the São Francisco River
(GPS 15º 06′18.7″ S, 44º 05′ 41.1″ W).

Fig. 19. Sands collected on the banks of the São Francisco River in Itacarambi/MG.
 Geotechnical Investigations on Highway BR 135 391

Fig. 20. Study summary for the dredged coarse sand of the São Francisco River in the
municipality of Itacarambi.

Figure 20 shows the summary of the study to the coarse sand dredged from the São
Francisco River in the municipality of Itacarambi. It is observed that the sand in terms
of the equivalent material has 67% with low organic impurities.

5 Analysis and Conclusions

This article aims to present some of the results of geotechnical studies carried out in the
area and materials collected in the BR 135 in the section between the municipalities of
Manga and Itacarambi in the state of Minas Gerais. With support for the implemen-
tation of field work and laboratory and analyzes, we conclude that:
392 R. Farias et al.

(a) the area is covered by soils with visual tactile characteristics similar to the pre-
dominance of latosols sandy yellow/red;
(b) in terms of granulometric the soil is predominantly sandy in the center of the track
to be paved, presents itself as not liquid and not plastic; Sand equivalency equal to
33%; Group Index equal to zero; The predominant soil A-2-4 (gravel or silty or
clayey sands) the classification TRB (HRB); optimum moisture content in the
normal energy equal to 8%; ISC (CBR) 18%; Expansion of the compacted material
after 96 h of immersion of 0.86%; apparent density of 1,77 g/cm3; natural moisture
of 1.86%;
(c) In the case of rocky material, it was found that there is a high occurrence of
carbonate rocks in the region. This material is exploited and used as building
material, especially on pavements and as aggregates for concrete (Gravel 1, 2 and
Gravel Gravel powder);
(d) by mixing the soil in a field with gravel in a 20% to 30% Gravel, being respectively
80% and 70% soil, it was found that the normal power sample with 20% gravel has
a specific mass dried and optimum moisture, respectively 2,03 g/cm3 and 8.70%
with 22.30% CBR. Already the sample with 30% of crushed stone features
2.14 g/cm3 and 7.49% of optimum moisture and CBR of 30%.

Acknowledgments. The Reforsolo Engenharia Ltda, National Department of Transport

Infrastructure (DNIT), Department of Engineering and Construction of the Brazilian Army
(DEC), Centre of Excellence in Transportation Engineering (CENTRAN.); Catholic University
of Brasilia (UCB), IesPlan and UniCEUB with important contributions that made possible the
realization of this work.

References
DNIT (DNER-ES 301/97): Pavimentação - sub-base estabilizada granulometricamente
DNIT (DNER-ES 303/97): Pavimentação - base estabilizada granulometricamente
MEDRADO, Wallen Alexandre. Caracterização Geotécnica de Solo da Região Norte de Minas
Gerais para Aplicação em Obras Rodoviárias. Dissertação de Mestrado. UFOP (2009)
NORMAS Diversas ABNT-NBR/DNIT para ensaios em laboratório
NORMAS Diversas ABNT-NBR/DNIT para trabalhos de campo
 Geotechnical Investigations and Proposal
for Stabilization and Recovery Slope
on Highway BR-010 Stretch in Aparecida
do Rio Negro to Goiatins, Tocantins, Brazil

Rideci Farias1, Haroldo Paranhos1, Jovino Rachid Araújo2,

Leonardo Ramanho Sales3, and Roberto Pimentel4(&)
1
UCB, Reforsolo Engenharia, UniCEUB, IesPlan, Brasília, Brazil
[email protected], [email protected]
2
CMT Engehharia, Brasília, Brazil
[email protected]
3
UniCEUB, Brasília, Brazil
[email protected]
4
UnB, Brasília, Brazil
[email protected]

Abstract. This article presents the results of geotechnical studies and proposals
to stabilize embankments referring to cuts 16, 23, 25, 30 and 31 of the BR-010
referring to Aparecida do Rio Negro a Goiatins, State of Tocantins. The study
consisted in the execution of a field incursions program for verification of
geotechnical conditions along the indicated stretch, running lab tests, and sta-
bility analysis of embankments in order to indicate solutions to cuts made
necessary for the completion of the highway. In terms of characterization of the
problem, it presents part of the photographic survey conducted in November
2007 with the conditions of the most critical slopes. There is in the pictures that
the side slopes to the highway were in the process of degradation and
(or) progressive destabilization requiring urgent interventions aimed at miti-
gating the ongoing problems. In terms of geometry, the slope had heights up to
approximately 35 m. To determine the parameters necessary to verify the sta-
bility of embankments using three undeformed blocks corresponding soil to cuts
25 and 31. Samples of identifications are presented and the summary of the
results of geotechnical tests performed and certain parameters. In laboratory
tests, it was found that the strength parameters (cohesion and angle of friction)
diverged somewhat from the usual laboratory parameters obtained in soils. This
behavior was due to the fact that the heterogeneity of the samples for the
presence mainly of discontinuities in the collected blocks. Before the run sta-
bility analysis and field surveys, it was found, the need for actions aimed at
solving problems for marginal slope to the highway. Thus, due to the proximity
of the intense rainy season, it was proposed the reconstruction of slope with
surface protection of them aimed at stabilization and protection against erosion,
as well as the execution of superficial and deep drainage. Aiming to further
verify the possible degradation of the slopes by rain, was sought in INMET’s
website (National Institute of Meteorology) precipitation to the region. It was
found that for the time in January, February and March rainfall is considerably

© Springer International Publishing AG 2018

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6_32
394 R. Farias et al.

high, which could further compromise the integrity of the executed cuts. Faced
with this possibility, he indicated consider the emergency actions that could
prevent this degradation. Actions such as the reduction and protection of
embankments shown in the study.

1 Introduction

This article presents the results of geotechnical studies and proposals to stabilize
embankments referring to cuts 16, 23, 25, 30 and 31 of the BR-010 referring to the
Black River Aparecida excerpt Goiatins, State of Tocantins, made in the year 2007.
The study consisted in the execution of a field incursions program for verification of
geotechnical conditions along the indicated section, as well as the execution of labo-
ratory tests, and stability analysis of embankments in order to indicate the solutions
made cuts necessary to completion of the highway.

2 Problem Characteristics

In terms of characterization of the problem proceeded to the assessment and field

analysis accompanied by an extensive survey photographic held in November 2007
with the conditions of the slopes 16, 23, 25, 30 and 31. It can be seen in Figs. 1, 2, 3, 4,
5, 6, 7, 8, 9, 10, 11, 12 and 13 which the side slope to the highway, in some cases, were
in the process of degradation and (or) progressive destabilization requiring urgent
interventions aimed at mitigating the ongoing problems. In terms of geometry, the
slope presented heights up to about 35 m, as evidenced in the analyzed sections and
shown in this article.

Fig. 1. General view of part of slope 16.

 Geotechnical Investigations and Proposal for Stabilization 395

Fig. 2. General view of part of cut 16 with superficial erosion.

Fig. 3. General view of part of the cut 16 with superficial erosion.

Fig. 4. View of the slope of cut 23.

396 R. Farias et al.

Fig. 5. Cut 25, still intact (September 2007).

Fig. 6. Slope of cut 25 starting degradation process of slope.

Fig. 7. Material of lower horizon of cut 25.

 Geotechnical Investigations and Proposal for Stabilization 397

Fig. 8. Traction cracks in the berm slope of cut 25.

Fig. 9. Traction cracks on berm slope of cut 25.

Fig. 10. Saturation from the slope basis in cut 25.

398 R. Farias et al.

Fig. 11. Upwelling water in cut 25 slope.

Fig. 12. Failure stretch in cut 25 slope.

Fig. 13. Cut 31 (slope basis saturation).

 Geotechnical Investigations and Proposal for Stabilization 399

3 Geotechnical Testes Performed

To determine the parameters necessary to verify the stability of slopes, we used three
undisturbed samples of soil corresponding to Sections 25 and 31. Table 1 shows the
sample to the Court 25, but also the summary of the results of the conducted
geotechnical testing and certain parameters. The three undisturbed soil samples were
used for executing the natural moisture of assays, bulk density and direct shear to
define the strength parameters. The tests were performed in accordance with relevant
standards and (or) the usual procedures used in geotechnical laboratory. For direct
shear tests (ASTM D3080/D3080M: 2011), the bodies of the test piece were tested in
the flooded condition in order to verify possible critical mechanical behavior of soil.

Table 1. Summary of geotechnical tests realized and determined parameters for sample of cut 25.
CUT 25 - Stake 1410 (slope right side) - Level 310,394 m
Specific weight natural cnat (kN/m3) 22,17
Natural humidity w (%) 9,38
Effective cohesion (Top) C′ (kPa) 78 (Condição Inundada)
Effective angle of friction (Pico) /′ (graus) 19 (Condição Inundada)
Effective cohesion (Residual) C (kPa) 27 (Condição Inundada)
Effective friction angle (Residual) /′ (graus) 15 (Condição Inundada)
Specific gravity apparent dry (kN/m3) cd (kN/m3) 20,27
Specific weight of solids (or grains) cs (g/cm3) 2,81
Specific weight saturated (kN/m3) csat (kN/m3) 22,86
Void index – 0,36
Porosity n (%) 26,43
Saturation level (%) Sr (%) 73,38

4 Comments About Soil Samples Collected

It appears from Table 1, the example of other tests performed for other samples not
shown in this article, the strength parameters (cohesion and friction angle) diverge
somewhat from the usual parameters obtained for laboratory soil. This behavior was
due to the fact that the heterogeneity of the sample by the presence mainly of dis-
continuities in the collected blocks.
Such discontinuities were also found in the field when the three surveys conducted
in the months of September, November and December 2007, to evaluate the cuts run
along the highway.
These discontinuities contributed directly to the technical difficulties encountered
by the Geotechnical Laboratory when the molding of the samples for the appropriate
tests. Such difficulties caused repetitions of several tests in which there was no con-
vergence results.
400 R. Farias et al.

5 Slope Stability Analysis Performed

Due to the difficulties, both in the collection of the blocks as the molding of the test
bodies, obtaining a few parameters for ordinary soils of the North Central region of
Brazil, it was performed several analyzes of stability of embankments through com-
puter program SLOPE/W, version 2007, marketed by Geo-slope International. These
analyzes understood the use of the parameters obtained in the laboratory, as well as
simulations with more realistic parameters that represented stability with safety factor
close to 1 (one) to the slope of the Court 25, in compliance with consistent conditions
of embankments.
After several tests in order to find more realistic parameters for soil that made up
the slopes of the cuts 25 and 31, corrected to the upper and lower horizon soil obtaining
the values shown in Tables 2 and 3 (horizon) and 6 (lower horizon).

Table 2. Strength parameters fixed for upper horizont soil.

Parametrosde Resistencia para o
Horizonte Superior
Coesão Efetiva C (kPa) 14
Ângulo de atrito efetivo /′ (graus) 26

Table 3. Strength parameters fixed for below horizont of soil.

Parametros de Resistencia para o
Horizonte Superior
Coesão Efetiva C (kPa) 44
Ângulo de atrito efetivo /′ (graus) 26

For the cohesion values, remained the value of 14 kPa found for the upper horizon,
and it is a high cohesion value for characteristic soils of the region, decreased to 78 kPa
to 44 kPa related to the lower horizon. The value of 44 kPa due to the fact that several
stability analyzes with a view to the compatibility of what was happening on the field.
It is surprising that value because the soil had low cohesion when desconfinado, but
when confined cohesion tended to be higher. For the case in question on which the
slope face exposure the soil tended to expand with the consequent breakdown neces-
sitating thus of at least surface protection of the slope.
It should be noted that the field located various types of breaks were observed,
highlighting the planar, circular and small spalling. Within this complexity, it broke up
in most cases the reproduction of circular breaks that are quite common in soils.
Simplification adopted in part of the study to circulate break comes the already men-
tioned complexity in terms of heterogeneity and the difficulties encountered, both in the
selection of collecting the material and the molding of the samples in order to obtain
parameters that could represent the field of similarity. Within the subject matter, for the
 Geotechnical Investigations and Proposal for Stabilization 401

analysis and conclusions as to the positioning in the appropriate interventions, aiming

stable geometries, one should consider - in addition to computational studies with
different characteristics, but also the importance of surveys in situ by professionals who
are able to position solutions address the various problems that may be identified.
In terms of stability analyzes were done for several cuts 16, 23, 25, 30 and 31
represent, in general, we found the slopes along the portions studied. Below, we present
the studies to the slope of the cut 25 because it is one of the most critical checked in the
field. Initially shows the stabilities of the cut 25 for the parameters obtained in the

Fig. 14. Cutting 25 stability with lab parameters (FS = 0.693) – Stake 1410 (natural slope –
right side).

Fig. 15. Cutting 25 stability with lab parameters (FS = 0.629) – Stake 1411 (natural slope –
right side).
402 R. Farias et al.

laboratory just to check, since the obtaining of strength parameters has been compro-
mised, as mentioned above. Subsequently, we present the analysis with more realistic
parameters in compliance with those found soils and conditions prevailing in the field.

5.1 Stability with Parameters Obtained in Lab

Next, in Figs. 14 and 15, the stability analyzes are presented, with the strength
parameters obtained in our laboratory for the cuts 25 on cuttings 1410 and 1411.

6 Considerations About Performed Slope Stability Analysis

Before the run stability analysis and field surveys, there was the need for actions aimed
at solving the problems that the marginal slope to the highway had presented. For this,
it was indicated two types of interventions following:
(a) The first type of proposed intervention was the reduction (retaludamento) of slopes
to stable geometries within the engineering practice, considering the economy,
environment, insertion area of the enterprise, in other factors. In this case, it rec-
ommended to check the availability of areas to cut beyond the highway right of
way, if necessary, and place of disposal of material removed;
(b) The second type of intervention was the recommendation for adoption of struc-
tures that allowed the stabilization of the now existing embankments. Within this
practice there are several solutions such as soil nailing, bolting, root cuttings, etc.
For the case in question, due to the proximity to the most intense rainy season, there
was the need to intervene as soon as possible. Within the time that was available, then
recommended him retaludamento sections with the execution of deep drainage and
surface. For deep drainage usually use the DHP - Drain subhorizontal Deep and surface
drainage is commonly contemplated by the vestment of drains and channels.
For simulations of retaludamento proceeded to several stability analysis in the
studied slope. By way of example, is presented in Figs. 16 and 17 the analysis of
stability for the cut 25 in the most critical Cutting (Cutting 1411). For these simula-
tions, broke up the natural sections with the execution of cuts for defining stable
geometries within the practice of engineering. After several simulations came to
embankments with 1 V: 1.5 H for the cuts 16, 23, 30 and 31. This primarily by the
need to install any type of surface protection on the slopes, as an example geocell in
order to prevent erosive processes.
For cutting 25, broke up the natural slope safety factor equal to 1.007 (Fig. 16),
with the implementation of retaludamento and introducing deep subhorizontal drains to
relegation in N.A. within the massif.
Initially, simulated an approximate cross-sectional area of 358 m2 Stake 1411. In
this situation, there was a need to move laterally out of the situation of the road -
counting from the bottom up - the first edge in 5 m, second at 10 m, the third in 15 m, the
fourth at 18.5 m and fifth in 20 m. Figure 17 shows such a situation where we got the
1,118 Safety factor that represents a low value for the stability of the slope in question.
 Geotechnical Investigations and Proposal for Stabilization 403

Fig. 16. Cutting 25 Stability with fixed parameters – Estaca 1411 (natural slope – right side).

Fig. 17. Cutting 25 stability with proposal remotion of soil – Stake 1411 (natural slope – right
side).

In addition to retaludamento, simulated to the lowering of N.A. inside slope which

represents the introduction of the sub horizontal drains for the removal of excess water.
In this situation it obtained the Safety Factor of 1.488, considered satisfactory for the
type of work in question. Figure 18 shows the situation analyzed.
404 R. Farias et al.

Fig. 18. Cutting 25 Stability – Stake 1411 (natural slopes – right side) – Low NA (introdution of
deep subhorizontal drain).

7 Recommendations for Stabilization and Recovery

of the Studied Slopes

Before the run stability analysis and field surveys, it was found, as already mentioned,
the need for actions aimed at solving the problems that the marginal slope to the
highway had presented. Thus, due to the proximity of the intense rainy season
approaching, it proposes the retaludamento Slashes 16, 23, 25, 30 and 31, with surface
protection of slopes aimed at stabilization and protection against erosion that occurred
and came to occur, as well as the execution of superficial and deep drainage.
Another recommendation made was the draining mattress running on the highway
over the cuts implemented in order to prevent possible upwelling of water that could
compromise the surface layers of the highway, as the sub-base, base and coating, that
the saturation finding the bases of slopes in the run cuts.

8 Pluviometric Analysis of the Area

In order to verify the possible degradation of the slopes by rain, was sought in
INMET’s website (National Institute of Meteorology) precipitation for the region in
previous years, and the period of the executed cuts. It was found in Figs. 19, 20, 21, 22,
23 and 24 in the months of January, February and March there is heavy rainfall, which
could further compromise the integrity of the executed cuts. Faced with this possibility,
it had to consider the urgency of actions that could prevent degradation. Actions such
as the reduction and protection of embankments listed in this article and the complete
work developed.
Geotechnical Investigations and Proposal for Stabilization 405

Fig. 19. Accumulated rainfall of 2002.

Fig. 20. Accumulated rainfall of 2003.

Fig. 21. Accumulated rainfall of 2004.

406 R. Farias et al.

Fig. 22. Accumulated rainfall of 2005.

Fig. 23. Accumulated rainfall of 2006.

Fig. 24. Accumulated rainfall of 2007.

 Geotechnical Investigations and Proposal for Stabilization 407

Acknowledgments. The Reforsolo Engenharia Ltd., CMT Engineering, Catholic University of

Brasilia (UCB), University Center of Brasilia (UniCEUB) and the Institute of Higher Education
Plateau with important contributions that made possible the realization of this work.

References
ABNT/ NBR 7181: Solo – Análise granulométrica
ABNT/NBR 6457: Solo – Amostras de solo – Preparação para ensaios de compactação e ensaios
de caracterização
ASTM D3080/D3080M:2011 - Standard Test Method for Direct Shear Test of Soils Under
Consolidated Drained Conditions
Geo-slope international: Stability modeling with SLOPE/W, an engineering methodology
(2007a)
Geo-slope international: Stress deformation modeling with SIGMA/W, an engineering method-
ology (2007b)
INMET (Instituto Nacional de Meteorologia). http://www.inmet.gov.br/portal/
Author Index

A G
Abulkareem, Ahmed H., 269 Greenwood, Kelli, 45
Ahmed, Asif, 45 Gross, Dennis, 69
Araújo, Jovino Rachid, 393
H
B Hachay, Olga A., 311
Biswal, Dipti Ranjan, 320 Hafsaoui, Abdallah, 1
Bouassida, Mounir, 149 Hamidi, Babak, 194
Boukarm, Riadh, 1 Hanna, A., 79
Hossain, M. Sahadat, 136
C I
Chauhan, Vinay Bhushan, 125 Indiramma, P., 284
Costa, Marco Aurélio Diana, 378
Cristelo, Nuno, 59 K
Kassouf, Roberto, 92, 177
D Khachay, Andrey Y., 311
da Fonseca, António Viana, 59 Khachay, Oleg Y., 311
Dahlhaus, Peter, 227 Khalifa, M., 79
Daoud, Samar, 149 Khan, M.A., 136
Dasaka, Satyanarayana M., 125 Khan, M.S., 136
Dash, Suresh Ranjan, 320 Khan, Mohammad Sadik, 45
de Albuquerque, Paulo José Rocha, 92 Khedri, Youcef, 1
de Carvalho, David, 92, 177 Kolekar, Yashwant A., 125
de Pimentel Sousa Junior, Roberto, 378 Kong, Xiangxing, 356, 363, 370
Dowling, Kim, 227 Krishna, Atkuri Venkata, 337
Kytzia, Susanne, 69
E M
Emad, Sharif, 104 Mandalawi, Maged Al, 227
Ennour, Samir, 149 Menacer, Kamel, 1
Etezad, M., 79 Messafer, Tahar, 162
Mokhberi, Mehdi, 38
F Murana, Abdulfatai Adinoyi, 210
Farias, Rideci, 378, 393 Murthy, Godavarthi V.L.N., 337
Fayed, Ayman L., 236
Faysal, Mohammad, 136 N
Fonte Jr., Nelson L., 92, 177 Nakache, Radouane, 1
Fredj, Mohamed, 1 Ng, Kok Shien, 12
Frikha, Wissem, 252 Nogueira, José Jailson, 378

© Springer International Publishing AG 2018 409

W. Frikha et al. (eds.), Soil Testing, Soil Stability and Ground Improvement,
Sustainable Civil Infrastructures, DOI 10.1007/978-3-319-61902-6
410 Author Index

P Shehata, Hany F., 236

Paranhos, Haroldo, 378, 393 Shi, Baotong, 370
Pimentel, Roberto, 393 Shishani, Aya, 45
Pinheiro, Claver, 59 Silvestri, Vincenzo, 297
Soliman, Ahmed, 227
R Sorour, Tamer M., 236
Rafieean, Seyed Ayuob, 38 Sudharani, CH., 284
Rao, Vedula V.N. Prabhakara, 337
Rios, Sara, 59
Roduner, Armin, 69 T
Tabib, Claudette, 297
S Touiti, Lamia, 20
Saadoun, Abderrazak, 1
Sabry, M., 79 V
Sahadat Hossain, MD, 45 Van Impe, William, 20
Sahoo, Umesh Chandra, 320 Varaksin, Serge, 194, 252
Said, Imen, 149
Sales, Leonardo Ramanho, 393 Y
Serpico, James J., 349 You, Greg, 227