Laboratory Investigations of Controlled Low - Strength Material PDF

Copyright
by
Lianxiang Du
2001
Laboratory Investigations of Controlled Low-Strength Material
by
Lianxiang Du, B.S., M.S.C.E.
Dissertation
Presented to the Faculty of the Graduate School of
The University of Texas at Austin
in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
The University of Texas at Austin

August, 2001
UMI Number: 3031045
Copyright 2001 by
Du, Lianxiang
All rights reserved.
________________________________________________________
UMI Microform 3031045
Copyright 2001 Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
____________________________________________________________
Bell & Howell Information and Learning Company

300 North Zeeb Road
P.O. Box 1346
Ann Arbor, MI 48106-1346
The Dissertation Committee for Lianxiang Du
Certifies that this is the approved version of the following dissertation:
Committee:
Kevin J. Folliard, Supervisor
David W. Fowler
David Trejo
Ellen Rathje
Alan F. Rauch
Dedication
This dissertation is dedicated to my parents, my wife, Maohui Zhou, and other
people who encouraged and supported me during my study in China and in US.
This dissertation is especially dedicated to my dad, who always stayed late with
me until I finished my assignments when I was a kid.

Acknowledgements
This report would not have been possible without Dr. Kevin J. Folliard.
He spent countless time advising, discussing, and encouraging me. He opened
my eyes to the world of concrete technology. He was so patient that he spent
many hours with my English skills. I owe him so much.
Special gratitude goes to Dr. David Trejo at Texas A&M University, who
co-sponsored the research work presented in this dissertation. His spiritual
inspirations will benefit my whole life. Longhorns and Aggies can be not only
rivals, but also friends.
I also appreciate very much the cooperation and support of my other
committee members, Dr. David Fowler, Dr. Ellen Rathje, and Dr. Alan Rauch.
Their guidance and technical support were very valuable to me. Special thanks
also goes to Dr. Dov Leshchinsky at University of Delaware. He was always
there providing answers to issues in geotechnical engineering. The involving of
Lilma dos Santos Ribeiro and Jessica D. Reath in part of the research are
appreciated. I would also like to thank members of the Construction Materials
Research Group, specifically Mike Rung, Kerry Rothenberg, David Whitney,
v
Sherian Williams, and many undergraduate students who helped me with the
laboratory work.
vi
Publication No._____________
Lianxiang Du, Ph.D.
The University of Texas at Austin, 2001
Supervisor: Kevin J. Folliard
This report presents the results of experimental studies on the properties of
controlled low-strength material (CLSM) and the corresponding test methods for
evaluating the characteristics of this material. A wide range of CLSM mixtures
were included in this study.
Fresh properties of CLSM were evaluated. The use of bottom ash,
foundry sand, and high-carbon fly ash was found to increase water demand to
achieve the desired flow in CLSM. Bottom ash tended to increase bleeding,
whereas foundry sand decreased bleeding tendencies. Two methods were used to
evaluate the setting and hardening of fresh CLSM mixtures, and their mechanisms
were evaluated. New methods were developed to measure the segregation and
subsidence of fresh CLSM mixtures.
The hardened properties of CLSM, including compressive strength,
excavatability, California Bearing Ratio, resilient modulus, water permeability,
vii
triaxial shear strength, and drying shrinkage were studied. Predictive models
were proposed for the strength development of CLSM mixtures. Modifications to
ASTM D 4832 were proposed to make it a more appropriate method for
measuring of the compressive strength of CLSM. The effects of alternative
capping materials, including neoprene pads, gypsum and sulfur, were studied. In
addition, factors affecting the strength gain of CLSM mixtures, including curing
temperature and humidity, were examined.
The resistance of CLSM mixtures to freezing and thawing was evaluated
using modified version of ASTM D 560. Higher air contents and compressive
strengths were found to improve the durability of CLSM mixtures exposed to
freezing-and-thawing cycles.
Although CLSM was found less corrosive than soil, the potential
formation of galvanic cells may be a concern to pipeline engineers. Factors
influencing the corrosion behaviors of ductile-iron samples were identified. For
coupled and uncoupled corrosion conditions, different contributing factors and
mechanisms were identified.
viii
Table of Contents
LIST OF TABLES XVI
LIST OF FIGURES XVIII
CHAPTER 1: INTRODUCTION 1
1.1 GENERAL ........................................................................................................ 1
1.2 RESEARCH OBJECTIVES .............................................................................. 3
1.3 OUTLINE OF THE DISSERTATION ............................................................. 4
CHAPTER 2: LITERATURE REVIEW 6
2.1 INTRODUCTION............................................................................................. 6
2.2 MATERIALS .................................................................................................... 9

2.2.1 Bonding Materials ................................................................................. 9
2.2.1.1 Portland Cement ........................................................................ 9
2.2.1.2 Fly ash ....................................................................................... 9
2.2.2 Aggregate ............................................................................................ 11
2.2.2.1 Concrete Aggregate ................................................................. 11
2.2.2.2 Foundry Sand ........................................................................... 12
2.2.2.3 Excess Fines from Aggregate Crushing Operation................. 13
2.2.2.4 Bottom Ash.............................................................................. 14
2.2.2.5 Other Aggregate Materials ...................................................... 15
2.2.3 Admixtures .......................................................................................... 15
2.2.3.1 Mineral Admixtures................................................................. 15
2.2.3.2 Chemical Admixtures .............................................................. 17
2.2.4 Water ................................................................................................... 17
ix
2.3 MIXTURE PROPORTIONS........................................................................... 17
2.4 BATCHING, MIXING, AND TRANSPORTATING .................................... 20
2.5 CLSM PROPERTIES AND TEST METHODS ............................................. 21

2.5.1 Fresh CLSM Properties ....................................................................... 22
2.5.1.1 Flowability............................................................................... 22
2.5.1.2 Segregation and Bleeding........................................................ 25
2.5.1.3 Hardening Time ....................................................................... 26
2.5.1.4 Subsidence ............................................................................... 27
2.5.2 Hardened CLSM Properties ................................................................ 27
2.5.2.1 Compressive Strength.............................................................. 27
2.5.2.2 Permeability............................................................................. 29
2.5.2.3 Shear Strength ......................................................................... 31
2.5.2.4 California Bearing Ratio (CBR) and Resilient Modulus......... 31
2.5.2.5 Drying Shrinkage ..................................................................... 33
2.5.2.6 Thermal Conductivity.............................................................. 34
2.5.2.7 Excavatability.......................................................................... 35
2.5.2.8 Freezing-and- Thawing Resistance .......................................... 36
2.5.2.9 Erosion..................................................................................... 37
2.5.2.10 Leaching and Environmental Impact..................................... 38
2.5.2.11 Corrosion............................................................................... 39
2.5.2.12 Utility Degradation................................................................ 41
2.6 ECONOMICS.................................................................................................. 41
2.7 CURRENT PRACTICE .................................................................................. 41

2.7.1 Case Histories ...................................................................................... 41
2.7.1.1 Introduction ............................................................................. 41
2.7.1.2 Backfills................................................................................... 44
2.7.1.3 Bridge Replacement ................................................................ 47
2.7.1.4 Structural Fill ........................................................................... 48
x
2.7.1.5 Utility Bedding ........................................................................ 51
2.7.1.6 Erosion Control ....................................................................... 57
2.7.1.7 Void Fill................................................................................... 57
2.7.1.8 Environmental Protection........................................................ 59
2.8 RESEARCH NEEDS ...................................................................................... 59

2.8.1 Introduction ......................................................................................... 59
2.8.2 Research Needs ................................................................................... 60
2.8.2.1 Standard Test Methods............................................................ 60
2.8.2.2 Effective Use of By-Product Materials in CLSM ................... 61
2.8.2.3 Excessive Long-Term Strength Gain ...................................... 61
2.8.2.4 Durability Aspects of CLSM ................................................... 62
2.8.2.5 Environmental Impact ............................................................. 62
2.8.2.6 Applicability of Soil Test Methods ......................................... 63
CHAPTER 3: MATERIALS AND MIXTURE PROPORTIONS 64
3.1 INTRODUCTION........................................................................................... 64
3.2 FINE AGGREGATE ....................................................................................... 64

3.2.1 Aggregate Type ................................................................................... 64
3.2.2 Aggregate Gradation ........................................................................... 65
3.2.3 Absorption Capacity and Specific Gravity of Fine Aggregate............ 66
3.2.4 Chemical Analysis of Foundry Sand ................................................... 68
3.3 MINERAL ADMIXTURES ............................................................................ 68

3.3.1 Fly ash ................................................................................................. 68
3.3.2 Chemical Analysis of Foundry Sand ................................................... 69
3.4 CHEMICAL ADMIXTURES ......................................................................... 70

3.4.1 Air-Entraining Agent ........................................................................... 70
3.4.2 Accelerating Agent .............................................................................. 71
xi
3.5 CEMENT......................................................................................................... 71
3.6 MIXING WATER ........................................................................................... 71
3.7 CLSM MIXTURE PROPORTIONS............................................................... 72

3.7.1 Mixture Proportioning ......................................................................... 72
3.7.2 Mixture Proportions ............................................................................. 74
CHAPTER 4: TEST METHODS AND PROCEDURES 82
4.1 INTRODUCTION........................................................................................... 82
4.2 FRESH CLSM TEST METHODS .................................................................. 85

4.2.1 Mixing Procedure and Measurement of Flow, Air Content, and
Unit Weight ......................................................................................... 85
4.2.2 Setting and Hardening ......................................................................... 87
4.2.3 Segregation.......................................................................................... 89
4.2.4 Subsidence ........................................................................................... 90
4.3 HARDENED PROPERTIES OF CLSM......................................................... 91

4.3.1 Unconfined Compressive Strength...................................................... 91
4.3.1.1 Specimen Preparation.............................................................. 91
4.3.1.2 Compression Machine ............................................................. 93
4.3.1.3 Effects of Loading Rate on Compressive strength.................. 95
4.3.1.4 Cylinder Curing and Conditions .............................................. 96
4.3.1.5 Effects of Curing Temperature and Humidity on
Compressive Strength................................................................. 98
4.3.1.6 Effects of Drainage Conditions on Compressive Strength...... 99
4.3.1.7 Alternative Capping Materials for Compression Testing...... 101
4.3.2 Triaxial Shear Strength...................................................................... 103
4.3.3 Water Permeability of CLSM............................................................ 104
4.3.4 Drying Shrinkage of CLSM mixtures ............................................... 105
4.3.5 California Bearing Ratio .................................................................... 106
4.3.6 Resilient Modulus.............................................................................. 106
xii
4.3.7 Excavation Study by CLSM Mixtures .............................................. 107
4.4 DURABILITY OF CLSM............................................................................. 109

4.4.1 Freezing-and- Thawing Resistance .................................................... 109
4.4.2 Wetting-and-Drying Resistance of CLSM ........................................ 110
4.5 CORROSION OF DUCTILE IRON IN CLSM ............................................ 110

4.5.1 Research Approach............................................................................ 110
4.5.2 Mass Loss Test .................................................................................. 111
4.5.2.1 Uncoupled Condition............................................................. 111
4.5.2.2 Coupled Condition (Galvanic) .............................................. 113
4.5.3 pH of Solutions Extruded from CLSM ............................................. 113
4.5.4 The Resistivity of CLSM................................................................... 115
4.5.5 Chloride Diffusion............................................................................. 116
4.6 LEACHING AND ENVIRONMENTAL IMPACT..................................... 116
CHAPTER 5: RESULTS AND DISCUSSIONS 120
5.1 INTRODUCTION......................................................................................... 120
5.2 CLSM MIXTURE PROPORTIONS AND FRESH PROPERTIES ............. 120

5.2.1 Water Demand/Flow Behavior .......................................................... 120
5.2.2 Air Contents of Fresh CLSM Mixtures............................................. 124
5.2.3 Unit Weight of Fresh CLSM Mixtures.............................................. 125
5.2.4 Setting/Hardening.............................................................................. 125
5.2.5 Bleeding of Fresh CLSM Mixtures ................................................... 133
5.2.6 Segregation of Fresh CLSM Mixtures .............................................. 134
5.2.7 Subsidence ......................................................................................... 136
5.3 HARDENED PROPERTIES OF CLSM....................................................... 137

5.3.1 Unconfined Compressive Strength.................................................... 137
5.3.1.1 Compressive Strength Development ..................................... 137
5.3.1.2 Effects of Cylinder Curing on Compressive Strength........... 146
xiii
5.3.1.3 Effects of Drainage Conditions on Compressive Strength.... 154
5.3.1.4 Specimen Preparation before Testing.................................... 157
5.3.1.5 Alternative Capping Materials for Compression Testing...... 159
5.3.1.6 Effects of Load Rate on Compressive Strength .................... 168
5.3.1.7 Effect of Specimen Size ........................................................ 172
5.3.1.8 Effects of Different Testing Machine .................................... 173
5.3.1.9 Effects of Curing Temperature and Humidity....................... 174
5.3.2 Triaxial Shear Strength...................................................................... 182
5.3.3 Water Permeability............................................................................ 184
5.3.4 Drying Shrinkage ............................................................................... 186
5.3.5 California Bearing Ratio .................................................................... 187
5.3.6 Resilient Modulus.............................................................................. 188
5.3.7 Excavation Study............................................................................... 189
5.4 DURABILITY OF CLSM............................................................................. 201

5.4.1 Freezing-and- Thawing Study............................................................ 201
5.4.2 Wetting-and-Drying Study................................................................ 204
5.5 CORROSION OF DUCTILE IRON IN CLSM ............................................ 206

5.5.1 Introduction ....................................................................................... 206
5.5.2 pH values of Extruded Solutions form CLSM Mixtures................... 207
5.5.3 Resistivity of CLSM Mixture ............................................................ 207
5.5.4 Chloride Concentration and Profiles ................................................. 209
5.5.5 Mass Loss of Ductile-Iron Specimens ............................................... 211
5.5.5.1 Mass Loss of Uncoupled Coupons ........................................ 211
5.5.5.2 Mass Loss of Coupled Coupons ............................................ 219
xiv
5.6 LEACHING AND ENVIRONMENTAL IMPACT..................................... 224
CHAPTER 6: CONCLUSIONS AND FUTURE WORK 228
6.1 CONCLUSIONS ........................................................................................... 228
6.2 FUTURE WORK .......................................................................................... 233
APPENDIX A: REFERCED STANDARD TEST METHODS 236
APPENDIX B: SETTING/HARDENING TEST RESULTS 240
REFERENCES 244
VITA 255
xv
List of Tables
Table 2.1: Four basic types of CLSM .................................................................. 18

Table 2.2: Typical mixture proportions for low fly ash content CLSM............... 19
Table 2.3: Typical Mixture Proportions for High Fly Ash Content CLSM ......... 20
Table 2.4: Currently used or potential test methods for CLSM ........................... 22
Table 2.4: Currently used or potential test methods for CLSM (cont’d) .............. 23
Table 3.1: Aggregate gradations used in laboratory test program........................ 65
Table 3.2: Materials included in laboratory test program ................................... 67
Table 3.3: Chemical characteristics of foundry sand ........................................... 68
Table 3.4: Chemical composition of fly ashes ..................................................... 70
Table 3.5: Che mical analysis of ASTM Type I portland cement (weight percent) .
............................................................................................................................... 71
Table 3.6: Non-air entrained CLSM mixture proportions .................................... 75
Table 3.7: Air-entrained CLSM mixture proportions ........................................... 76
Table 3.8: Selected CLSM mixtures included in Stage One ................................ 77
Table 3.9: CLSM mixture proportions for load-rate study.................................... 79
Table 3.10: CLSM mixtures for cylinder curing/specimen conditioning study... 79
Table 3.11: CLSM mixtures for excavation study ............................................... 79
Table 3.12: CLSM mixtures used for freezing and thawing study....................... 80
Table 3.13: Mixture proportions for alternative capping materials study............ 80
Table 3.14: CLSM mixture proportions for drainage-condition study................. 80
Table 3.15: CLSM mixture proportions for cylinder storage-condition study..... 81
Table 3.16: Mixture proportions for effects of curing temperatures and humidity81
Table 3.17: CLSM mixture proportions for triaxial shear strength and water
permeability............................................................................................... 81
Table 4.1: Summary of initial test program for this study ................................... 83
Table 4.2: Overview of all properties and tests studied ....................................... 84
Table 4.3: Four curing conditions used in the study (using G-series mixtures)... 98
Table 5.1: CLSM mixture proportions and fresh properties .............................. 121
Table 5.2: Analysis of variance (ANOVA) for water demand ........................... 124
Table 5.3: Segregation data for selected CLSM mixtures.................................. 135
Table 5.4: Subsidence results of six selected CLSM mixtures .......................... 136
Table 5.5: Unconfined compressive strength of original 38 CLSM mixtures .... 139
Table 5.6: Comparison of capping materials and methods ................................ 140
Table 5.7: Comparison between predicted compressive strengths (using equations
shown in this section) and actual strengths from previous study (using
different constituent materials) ................................................................ 145
Table 5.8: Effects of curing methods on compressive strength (Mixture B-1) ... 147
Table 5.9: Compressive strength at 28 days using different curing methods...... 148
xvi
Table 5.10: Compressive strength at 91 days using different curing methods.... 149
Table 5.11: Compressive strength at 7 days under different drainage conditions .....
............................................................................................................................. 156
Table 5.12: Compressive strength at 28 days under different drainage conditions ..
............................................................................................................................. 157
Table 5.13: Influence of specimen air-drying on compressive strength (Mixture B-
2).............................................................................................................. 158
Table 5.14: Compressive strength results using different capping materials ..... 160
Table 5.15: Coefficients of variation for compressive strengths using different
capping materials ..................................................................................... 161
Table 5.16: Comparison of “a” values calculated by two methods.................... 165
Table 5.17: Effects of loading rates (wide range) on compressive strength ...... 169
Table 5.18: Effects of load rates (narrow range) on compressive strength........ 170
Table 5.19: Effects of loading rates (wide range) on deformation at peak load 170
Table 5.20: Effect of narrow range load rates (narrow range) on deformation at
peak load.................................................................................................. 171
Table 5.21: Effects of cylinder size on unconfined compressive strength......... 173
Table 5.22: Effects of different testing machines on compressive strength
(coefficient of variation is included in parentheses)................................ 174
Table 5.23: Compressive strength of specimens cured at different temperatures at
different ages ........................................................................................... 177
Table 5.24: Results of triaxial compression tests ............................................... 182
Table 5.25: Water permeability (hydraulic conductivity) of selected CLSM
mixtures ................................................................................................... 184
Table 5.26: Test results of water permeability after freezing-and-thawing cycles ...
............................................................................................................................. 185
Table 5.27: Drying shrinkage of selected CLSM mixtures ................................ 187
Table 5. 28: CBR values for six selected CLSM mixtures................................. 187
Table 5.29: Resilient modulus of six selected CLSM mixtures ......................... 188
Table 5.30: Results of excavation study............................................................. 190
Table 5.31: Comparison between field penetrometer values (field-cured samples)
and compressive strength values (field-cured cylinders) ........................ 191
Table 5.32: Results of excavatability study........................................................ 194
Table 5.33: Compressive and splitting tensile strengths at 7 and 28 days ......... 198
Table 5.34: Effects of temperature and drying conditions on splitting tensile
strength of CLSM.................................................................................... 200
Table 5.35: Freezing-and-thawing results for selected CLSM mixtures (after 7
days of curing) ......................................................................................... 202
days of curing) ......................................................................................... 202
Table 5.37: Freezing-and-thawing test results of D-series mixtures .................. 203
xvii
Table 5.38: Wetting-and-drying results for selected CLSM mixtures (after 7 days
of curing) ................................................................................................. 205
Table 5.39: Wetting and drying results for selected CLSM mixtures (after 28
days of curing) ......................................................................................... 206
Table 5.40: pH and resistivity values of CLSM mixtures.................................. 208
Table 5.41: Percent mass loss for samples embedded in CLSM only................ 213
Table 5.42: Statistical comparison of repeated CLSM mixtures (uncoupled
corrosion)................................................................................................. 216
Table 5.43: Statistical results of mass-loss test for uncoupled samples (CLSM
mixtures containing fly ash) .................................................................... 218
mixtures containing no fly ash) ............................................................... 219
Table 5.45: Percent mass loss for coupled samples embedded in sand .............. 221
Table 5.46: Statistical comparison of repeated CLSM mixtures on corrosion of
coupled coupons embedded in sand ........................................................ 222
Table 5.47: Statistical results of mass-loss test for coupled samples (CLSM
mixtures containing fly ash) ...................................................................... 223
mixtures without fly ash) ......................................................................... 224
Table 5.49: Analysis of heavy metal concentration of extract from raw materials
(mg/L) ...................................................................................................... 226
Table 5. 50: TCLP test results for Class C fly ash, Class F fly ash, and bottom ash
(mg/L) ...................................................................................................... 227
Table B-1: Needle penetration (NP) and soil penetrometer (SP) results............ 240
Table B-1: Needle penetrAtion (NP) and soil penetrometer (SP) results (cont’d)....
............................................................................................................................. 241
Table B-1: Needle penetration (NP) and soil penetrometer (SP) results (cont’d)242
Table B-2: Needle penetration (NP), soil penetrometer (SP) and vane shear tester
(VS) Results............................................................................................. 243
xviii
List of Figures
Figure 2.1: Applications of CLSM. Case A: backfill; Case B: utility bedding;

Case C: structural fill. ................................................................................ 44
Figure 2.2: Class D: impermissible beddings. load factor = 1.1. (From Soil
Engineering, 4th Edition, M. G. Spangler and R. L. Handy, Harper & Row,
Publisher, New York, 1982) ...................................................................... 52
Figure 2.3: Class C ordinary beddings. load factor = 1.5 (From Soil Engineering,
4th Edition, M. G. Spangler and R. L. Handy, Harper & Row, Publisher,
New York, 1982) ....................................................................................... 53
Figure 2.4: Class B first class beddings. load factor for ditch conduits =1.9
(From Soil Engineering, 4th Edition, M. G. Spangler and R. L. Handy,
Harper & Row, Publisher, New York, 1982) ............................................ 54
Figure 2.5: Class A: concrete-cradle bedding. load factor 2.25 to 3.4. (From Soil
Engineering, 4th Edition, M. G. Spangler and R. L. Handy, Harper & Row,
Publisher, New York, 1982) ...................................................................... 55
Figure 3.1: Gradations of three fine aggregates and ASTM C 33 limits.............. 66
Figure 4.1: A typical soil pocket penetrometer ..................................................... 88
Figure 4.2: Pocket vane shear tester used in this program .................................. 89
Figure 4.3: Plastic mold were pre-cut and taped before casting specimens ......... 93
Figure 4.4: A typical set-up of compression testing using geotechnical unconfined
compression machine. ............................................................................... 94
Figure 4.5: Drainage condition simulation......................................................... 101
Figure 4.6: Method of capping cylinders with gypsum compound .................... 103
Figure 4.7: Specimen set-up for testing of permeability of CLSM mixture after
freezing-and-thawing cycles.................................................................... 105
Figure 4.8: Corrosion test set-up for comparing corrosion performance of coupons
in CLSM and sand (uncoupled)............................................................... 112
Figure 4.9: Corrosion test set-up for comparing corrosion performance of
galvanically coupled coupons in CLSM and sand. ................................. 113
Figure 4.10: Set-up for resistance measurement ................................................. 114
Figure 4.11: Set-up for chloride ion diffusion.................................................... 115
Figure 4.12: Flow chart to study toxicity (heavy metals) of CLSM constituent
materials .................................................................................................. 119
Figure 5.1: Effects of variables on water demand of non-air-entrained CLSM
mixtures ................................................................................................... 123
Figure 5.2: Correlation between ASTM C 403 and soil penetrometer values ... 127
Figure 5.3: Stand for the pocket soil penetrometer ............................................. 128
xix
Figure 5.4: Comparison between needle penetration (NP) and soil penetrometer
reading (SP) values and the back calculated resistance from soil
penetrometer (BSP). ................................................................................ 130
Figure 5.5: Setting time needle and the wedge at the tip of the needle (following
ASTM C 403).......................................................................................... 132
Figure 5.6: Comparison between soil penetrometer and vane shear values ........ 133
Figure 5.7: Effects of variables on bleeding of non-air-entrained CLSM mixtures.
“+” means increasing the bleeding, “-“ means decreasing, and “X” means
the effect is not significant compared with variations. ........................... 134
Figure 5.8: Correlation between bleeding (weight percent) and subsidence....... 137
Figure 5.9: Load-deformation response of Mixture #12 at 3, 7, and 28 days .... 140
Figure 5. 10: Comparison of measured and predicted strengths of air-entrained
CLSM mixtures at the ages of 7, 28, and 91 days. .................................. 142
Figure 5.11: Comparison of measured and predicted strengths of non-air-entrained
CLSM mixtures ....................................................................................... 144
Figure 5.12: Effects of curing conditions on compressive strength for Mixtures G-
7 and G-8 ................................................................................................. 150
Figure 5.13: Effects of curing conditions on compressive strength for Mixtures G-
9 and G-10 ............................................................................................... 151
Figure 5.14: Curing condition effects on compressive strength of Mixture G-3
and G-6 .................................................................................................... 152
and G-5 .................................................................................................... 153
Figure 5.16: Effects of capping methods on coefficient of variation (7-day
compressive strength) .............................................................................. 162
Figure 5.17: Effects of capping methods on 163
compressive strength) .............................................................................. 163
Figure 5.19: Load-deflection curves for mixture E-1 (compression testing at 28
days) ........................................................................................................ 167
Figure 5.20: Compressive strength of mixture H-1 for different curing conditions .
............................................................................................................................. 178
............................................................................................................................. 178
............................................................................................................................. 179
............................................................................................................................. 179
............................................................................................................................. 180
............................................................................................................................. 180
xx
Figure 5.26: Stress-strain curves of mixture I-1 and I-4 at 28 days (triaxial
compression testing with an effective confining pressure of 69 kPa)..... 183
Figure 5.27: Penetration resistance (ASTM C 403) vs. compressive strength at
different ages (all C-series mixtures) ...................................................... 192
Figure 5.28: Cracking of CLSM mixture C-2 during DCP penetration............. 195
Figure 5.29: Correlation between stiffness (from GeoGauge) and DCP index.. 196
Figure 5.30: Correlation between DCP Index and Removibility Modulus (RE)
calculated using 28-day compressive strength (lab-cured cylinders) ...... 197
calculated using 240-day compressive strength (field-cured cylinders) . 197
Figure 5.32: A cylinder from mixture E-1, before and after being tested for
splitting tensile strength........................................................................... 199
Figure 5.33: Specimens after freezing-and-thawing cycles (CLSM mixtures
containing foundry sand) ......................................................................... 204
Figure 5.34: Chloride profiles for select CLSM mixtures after 22 weeks of
exposure................................................................................................... 210
xxi
Chapter 1: Introduction
1.1 GENERAL
Controlled low-strength material (CLSM) is a relatively new technology
whose use has grown in recent years. CLSM, often referred to as flowable fill, is
a highly flowable material typically comprised of water, cement, fine aggregates,
and often times, fly ash. Other by-product materials, such as foundry sand and
bottom ash, and chemical admixtures, including air-entraining agents, foaming
agents, and accelerators have also been used successfully in CLSM.
CLSM is typically specified and used in lieu of compacted fill in various
applications, especially for backfill, utility bedding, void fill, and bridge
approaches. Backfill includes applications such as backfilling walls (e.g.,
retaining walls) or trenches. Utility bedding applications involve the use of
CLSM as a bedding material for pipe, electrical, and other types of utilities and
conduits. Void-filling applications include the filling of sewers, tunnel shafts,
basements, or other underground structures. CLSM is also used in bridge
approaches, either as a subbase for the bridge approach slab or as backfill against
wingwalls or other elements.
There are various inherent advantages of using CLSM instead of
compacted fill in these applications. These benefits include reduced labor and
equipment costs (due to self-leveling properties and no need for compaction),
faster construction, and the ability to place material in confined spaces. The
relatively low strength of CLSM is advantageous because it allows for future
1
excavation, if required. Another advantage of CLSM is that it often contains by-
product materials, such as fly ash and foundry sand, thereby reducing the demand
on landfills, where these materials may otherwise be deposited.
Despite these benefits and advantages over compacted fill, the use of
CLSM is not currently as widespread as its potential might predict. One reason
for this is that CLSM is somewhat a hybrid material; that is, it is a cementitious
material that behaves more like a compacted fill. As such, much of the
information and discussions on its uses and benefits have fallen outside the
traditional specialties of concrete materials and geotechnical engineering.
Although there is considerable literature available on the topic, it is often not
given the level of attention it deserves by either group.
Many states have developed specifications (in some cases, provisional)
that govern the use of CLSM. However, these specifications differ from state to
state, and moreover, a variety of different test methods are currently being used to
define the same intended properties. This lack of conformity, both on
specifications and testing methods, has also hindered the proliferation of CLSM
applications.
There are also technical challenges that have served as obstacles to
widespread CLSM use. For instance, it is often observed in the field that
excessive long-term strength gain makes it difficult to excavate CLSM at later
ages. This can be a significant problem that translates into added cost and labor.
Other technical issues deserving attention are the compatibility of CLSM with
different types of utilities and pipes, the potential leaching of constituent materials
2
and elements, and the durability of CLSM subjected to freezing-and-thawing
cycles.
1.2 RESEARCH OBJECTIVES
This dissertation is directed towards identifying deficiencies in the state-
of-the-practice, modifying or developing new test methods, and understanding
factors affecting behaviors of CLSM mixtures. Specifically, the objectives of the
study are to:
1. Summarize the state-of-the-art and state-of-the-practice of CLSM
technology,
2. Identify and understand factors affecting the water demand of CLSM
mixtures to reach a target flow,
3. Evaluate bleeding and segregation of fresh CLSM mixtures, identify
factors influencing the amounts of bleed water, and develop test
method for measuring segregation,
4. Understand the setting and hardening behavior of fresh CLSM
mixtures and recommend test methods for laboratory and field testing,
5. Develop a suitable compression test method to measure strength of
CLSM mixtures, including providing guidance on loading rate,
cylinder curing conditions, and appropriate capping materials,
6. Develop predictive models for the short- and long-term compressive
strength of CLSM,
3
7. Study the effects of curing temperature, humidity, and other
parameters on the strength gain of CLSM mixtures,
8. Study the suitability of certain standard test methods on CLSM,
including triaxial shear test, water permeability, California Bearing
Ratio (CBR), and resilient modulus,
9. Evaluate drying shrinkage of CLSM mixtures,
10. Evaluate methods to predict future excavation of CLSM mixtures,
11. Study durability of CLSM mixtures, including freezing-and-thawing
resistance and wetting-and-drying resistance,
12. Evaluate corrosion behaviors of ductile-iron specimens in CLSM,
including coupled and uncoupled cases, and identify potential
constituents affecting corrosion rate,
13. Propose a procedure to evaluate the leaching and environmental
impact of CLSM mixtures.
1.3 OUTLINE OF THE DISSERTATION
The dissertation is comprised of six chapters. A brief summary of each
chapter is provided below.
Chapter 1 explains the overall objectives of this study.
Chapter 2 summarizes the state-of-the-art and current practice related to
CLSM. Typical applications of CLSM are also provided with case histories.
Chapter 3 describes the CLSM mixture proportions used in the study,
which included 38 mixtures in the Stage One study and 67 mixtures in Stage Two.
4
Three types of aggregates (concrete sand, foundry sand, and bottom ash), three
types of fly ash (Class C, Class F, and High carbon), and American Society for
Testing and Materials (ASTM) Type I cement were included in the study.
Relevant physical and chemical properties of these materials were also measured.
Chapter 4 provides the detailed test program in this research. The
program included the measurements of bleeding and segregation, flow, unit
weight, air content, compressive strength, CBR, resilient modulus, traixial shear
strength, water permeability, drying shrinkage, freezing-and-thawing resistance,
wetting-and-drying resistance, corrosion of ductile-iron samples, and
environmental impact. Factors affecting the testing and development of
compressive strength are included in the test program.
Chapter 5 provides the results and discussions. Models predicting strength
development are provided. Methods predicting setting and hardening are
compared. Effects of curing temperature and humidity are discussed. Factors
affecting the testing of compressive strength of CLSM are studied and
recommendations to modify current ASTM standard test methods are provided.
Methods predicting excavatability are compared. Factors potentially affecting
corrosion behaviors are identified. A systematic procedure was developed to
assess the potential toxicity of raw materials used in CLSM.
Chapter 6 summarizes the main conclusions of the study and identifies
technical issues that deserve attention in future research.
5
Chapter 2: Literature Review
2.1 INTRODUCTION
Controlled low-strength material (CLSM) is, as defined by American
Concrete Institute (ACI) Committee 229, a self-compacted, cementitious material
used primarily as a backfill in lieu of compacted fill.1 Several terms are currently
used to describe this material, including flowable fill, unshrinkable fill, controlled
density fill, flowable mortar, plastic soil-cement, soil-cement slurry, K-Krete and
other various names.
Controlled low-strength materials are defined by “Cement and Concrete
Terminology (ACI 116R)” as materials that result in a compressive strength of 8.3
MPa or less.1 However, most current CLSM applications require unconfined
compressive strength of 2.1 MPa or less. Some researchers consider the range of
0.3 to 1.1 MPa as a good index of sufficient strength and easy future excavation.
For applications where future excavations are expected, the excavatability of
CLSM is critical and this may determine the success of CLSM in practices such
as utility bedding.
Soil-cement has been a widely used material in geotechnical-engineering
practices for a long time. CLSM is relatively new and is different from
conventional soil-cement. Compaction is usually required for soil-cement, which
is not the case for CLSM. One of its earliest applications was carried out in 1964
by the US Bureau of Reclamation as the bedding of the 515-km long pipelines in
the Canadian River Aqueduct Project, which runs from north of Amarillo to south
6
of Lubbock, Texas.2 The material used in that project was called plastic soil-
cement. The soil used consisted of local blow sand deposits. A new construction
procedure was introduced, and the cost of this project was estimated to be 40
percent less than using conventional backfilling techniques. The productivity was
increased from about 120 to 305 m of pipe placed per shift.
In the early 1970s, Detroit Edison Company, in cooperation with Kuhlman
Corp., a ready-mix concrete producer in Toledo, Ohio, investigated an alternative
to compacted granular fill utilizing fly ash and concrete batching techniques. This
new backfill material, called flowable fly ash, was used in several applications in
the late 1970s.3-5 This material was composed principally of fly ash and typically
4 to 5 percent cement, along with an appropriate amount of water. In the Belle
River project, it was estimated that more than $1 million was saved by using this
new material.3 One impressive and exciting feature of this material was that it
remained cohesive when being placed. Another characteristic was the steep angle
of repose when it was placed either above or under water.
Eventually, a company known as K-Krete Inc. was formed. In 1977, four
patents for K-Krete were issued to “Brewer et al.”.6 A typical K-Krete mixture
consisted of 1305 to 1661 kg of sand, 166 to 297 kg of fly ash, 24 to 119 kg of
cement, and up to 0.35 to 0.40 m3 of water per m3 of product. The four patents
included mixture design, backfill technique, pipe bedding, and dike construction.
These patents were sold to Contech, Inc., in Minneapolis, Minn., who ceded the
patent rights later to the National Ready Mix Concrete Association (NRMCA)
with the stipulation that those rights may not be used in a proprietary manner.6,7
7
Since then, ready-mixed concrete producers and contractors have used similar
materials to K-Krete without violating patent restrictions.
Following K-Krete’s emergence as a replacement material for
conventional compacted fill, similar materials have been developed and used
throughout the United States and Canada. However, the lack of a centralized
source for obtaining and disseminating information within the marketplace
appeared to cause confusion and reluctance on the part of the engineering
community to use these materials. In response to the proposal of Brewer, ACI
Committee 229 was established in 1984 under the title “Controlled Low-Strength
Materials (CLSM)”. After years of efforts, in 1994, the committee published a
report called “Controlled Low Strength Materials (CLSM),”1 which has been
referenced widely. In 1999, the revised edition was published with the same title.
In 1998, the American Society for Testing and Materials (ASTM)
published a book titled “The Design and Application of Controlled Low-Strength
Materials (flowable fill).”8 The articles in this book represented the state of art
and practice of CLSM in the field and in the research laboratory at that time.
Different types of waste materials were included in CLSM mixtures to recycle
waste and reduce the cost. Currently, there are five ASTM testing standards
available for CLSM. Specific information and description on these and other test
methods relevant to CLSM are provided later in this dissertation. For
convenience, these tests are also listed in Appendix A.
Many universities and state Departments of Transportation (DOTs) have
conducted comprehensive studies on CLSM.9-19 Some of these are still in
8
progress. For instance, The University of Louisville and Purdue University have
conducted extensive research on CLSM. State DOTs have used CLSM in
constructing of small-span bridges and culverts.20-22 The Iowa DOT used it as
backfill for bridge approaches and bridges replacement.23 Certain companies,
such as W. R. Grace Construction Products and Master Builders Technologies,
have developed powerful air-entraining agents to produce CLSM with high air
content. The air content of the air-modified CLSM can be as high as 15 to 35%,
and the corresponding density is significantly decreased.24
A comprehensive investigation of CLSM, funded by the National
Cooperative Highway Research Program (NCHRP) under Project 24-12 and 24-
12(1) (“Controlled Low-Strength Material for Backfill, Utility Bedding, Void Fill,
and Bridge Approaches”) has culminated in two reports. One report summarized
the state-of-the-art and current practice related to CLSM in 1999 and the other
report summarized the findings of a comprehensive laboratory investigation of
CLSM in 2001.25,26 Much of the laboratory study described in Chapter 4 of this
dissertation was part of these NCHRP-funded efforts.
2.2 MATERIALS
To compete with conventional compacted fill, the cost of CLSM plays a
key role. The materials used in CLSM are thus of crucial importance for this
technology. One significant benefit of CLSM is its ability to use a wide range of
local materials, including by-product materials. Portland cement, fly ash,
9
aggregates (e.g., foundry sand), chemical admixtures, and other by-product
materials can be included in the production of CLSM.
2.2.1 Bonding Materials
2.2.1.1 Portland Cement
Although there is no restriction placed on the types of portland cement
used in CLSM, ASTM Type I is most commonly used. The prevailing criteria are
the local availability and cost of cement, and Type II or Type I/II cements may be
more common in certain regions of the United States. Because of the
comparatively low cement contents in CLSM, common concrete durability
problems, such as alkali-aggregate reaction and sulfate attack, appear quite
unlikely. Type III portland cement has been successfully used in CLSM to
achieve higher early strength and to reduce subsidence.25
2.2.1.2 Fly Ash
Class C fly ash can be used as a binder because of its cementitious
property. The ash typically contains 15 to 35 percent analytical CaO. With
similar proportions, CLSM mixtures with Class C fly ash demonstrate higher
compressive strength than those with Class F fly ash. One example is Flash Fill,
which contains predominantly Class C fly ash, concrete sand and water.27 No
portland cement is added for binding. Its quick-setting, self-leveling, self-
compacting characteristics make trench road repairs faster, easier, and more
economical. These characteristics make it a promising material. Other uses of

10
flash fill include building foundations, fill around pipes, gas lines, and manholes,
and replacement of weak subgrade beneath footings. Flash fill can be hand-
excavated without power-assisted tools or machinery. However, the quick-setting
property of flash fill may cause potential construction difficulties.
2.2.2 Aggregate
Unlike conventional soil-cement, CLSM does not use local soil as an
ingredient. Because of the low cement content of CLSM, the cost of aggregate
materials actually dominates the total cost. It is common that CLSM producers
use whatever suitable aggregate materials that are locally available in the
mixtures, such as concrete sand, foundry sand, crushed glass, by-product fines,
and other aggregates.
2.2.2.1 Concrete Aggregate
Although a wide range of aggregate materials may be used successfully in
CLSM, conventional concrete sand (satisfying ASTM C 33) is most commonly
utilized, especially for CLSM produced at ready-mix concrete plants (the
dominant source of CLSM). Sand that barely meets the ASTM C 33 requirements
can be used in CLSM production as long as specified flowability is satisfied.
CLSM have mostly evolved using only sand as aggregate. CLSM in the
Pacific Northwest has developed differently in that many mixtures use gravels up
to 25-mm top size.28 The reasons for the use of gravel center around economy
and performance. Concrete technology demonstrates that if the largest top-size
11
aggregate is used, the lowest void content in the combined aggregates will be
achieved. Reduced voids result in a lower paste requirement, which
correspondingly reduces the cost on cementitious materials. Gravel can be a
viable material as aggregate in CLSM proportions. Economics are likely to
determine whether gravel is used or not. Performance of CLSM mixtures with
gravel may be expected to be similar to those with sand only. Subsidence can be
reduced as a result of low mixing water requirement.
2.2.2.2 Foundry Sand
Foundry sand, a by-product of the metal casting industry, has been studied
and used successfully in CLSM, and its use has increased in recent years.10,30-32
Foundry sand is a viable candidate for use in CLSM because of its lower cost,
increasing availability, and satisfactory performance.
It is estimated that a typical foundry generates approximately one ton of
waste sand for every ton of metal castings produced and shipped.33 The most
commonly used waste foundry sand in CLSM is “green sand,” a term applied
when the original uniform sand is treated with a bonding agent (usually clay) to
optimize the efficiency of the sand in the molding process. After molding is
completed, the sand is discarded, often in landfills, and in 1994 typical costs were
$22/tonne to $44/tonne.8
The Federal Highway Administration (FHWA) issued a report, “User
Guidelines for Waste and By-Product Materials in Pavement Construction,”
which covers in detail the use of foundry sand (and fly ash) in CLSM and
12
provides guidelines for its proper usage.34 The Environmental Protection Agency
(EPA) has also recognized foundry sand, along with fly ash, as suitable materials
for CLSM.35
There has been a perception that using foundry sand in CLSM may have a
detrimental environmental impact because heavy metals may leach from the
foundry sand. Ferrous foundry sands are more commonly used in CLSM because
there are concerns about the contents of heavy metal in non-ferrous foundry sand.
The EPA does not recommend using non-ferrous foundry sand in CLSM because
of the concerns over the potential leaching of phenols and heavy metals, such as
cadmium, lead, copper, nickel, and zinc.35 Bhat et al. reported the results of their
environmental tests.10 Bio-assay tests were performed as part of this study to
screen out potentially hazardous waste foundry sands. The test was performed
both on expressed CLSM pore solution and raw foundry sand. No firm
conclusions were presented in that report.
2.2.2.3 Excess Fines from Aggregate Crushing Operation
With the development of performance-enhancing admixtures (air-
entraining foam), high-fines limestone screenings have been used as aggregate for
CLSM.36,37 It was estimated that more than 81.6 million tonnes of quarry by-
products were generated in 1993.36 It is difficult to find viable uses for by-
product fines, which account for 15- to 20-percent of aggregate produced, because
most construction and highway material specifications limit the aggregate finer
than 0.075 mm to 5 or 7 percent or less specified by ASTM C 33. A study
13
showed that 14 to 30 percent of air worked well for CLSM mixtures containing
high-fines aggregate.37 Screenings that contain up to 21 percent finer than 0.075
mm particles can be used as aggregate to produce CLSM mixtures meeting the
National Ready Mixed Concrete Association (NRMCA) performance
recommendations.
Another source of high-fines aggregate is recycled concrete. Current
practice usually involves using recycled concrete as coarse aggregate, leaving an
abundance of fines (passing 300-ìm sieve), which may be well suited for use in
CLSM.
2.2.2.4 Bottom Ash
Bottom ash and fly ash are both by-product materials of coal combustion.
Bottom ash is formed by large noncombustible particles that cannot be carried by
the hot gases. These particles descend on hoppers or conveyors, at the bottom of
the furnace, in a solid or partially molten condition. The particles gradually cool
down to form bottom ash. Bottom ash particles are typically porous and angular
in shape. As a by-product material, bottom ash is commonly disposed of in
ponds. In this process, bottom ash is passed through a crusher to reduce the size
of large particles, and it is transported hydraulically through pipelines to the pond
site. The typical range of particle sizes falls between 75 microns and 25
millimeters. Bottom ash can be used in compacted fill when combined with fly
ash.38 Researchers have successfully used bottom ash in CLSM.39
14
2.2.2.5 Other Aggregate Materials
In addition to the aggregate materials previously described, there are other
materials used in CLSM as aggregates. Colored glass that cannot be recycled by
local bottle manufacturers has been crushed to pass a 12.5-mm (½ in) sieve and
successfully used in CLSM as an aggregate.40 A special process was utilized so
that the crushed glass could be handled with bare hands. Phosphogypsum is a by-
product of the production of phosphoric acid and has been shown to be a viable
aggregate for CLSM.41 Toxicity characteristic leaching procedure (TCLP) testing
showed that the toxic contents of the individual materials and final design pass
mixtures were well below the EPA leachate standards. Ground granulated blast
furnace slag (GGBFS) and bentonite have also been used in CLSM.59
2.2.3 Admixtures
2.2.3.1 Mineral Admixtures
Fly ash is a by-product of coal combustion and has been used in a wide
range of construction applications, including CLSM. Fly ashes are broadly
classified into two classes based on the calcium content and pozzolanic properties
(ASTM C 618). Fly ash normally produced from lignite or subituminous coals is
classified as Class C fly ash, which has cementitious properties in addition to
pozzolanic properties. Fly ash normally produced from anthracite or bituminous
coal is classified as Class F fly ash. It has only pozzolanic properties and has no
self-cementitious characteristics because of the absence of sufficient amount of
calcium oxide.
15
Fly ash is used mostly in portland cement concrete, but its use in CLSM
has increased considerably in recent years. Even so, approximately 70 to 75
percent of fly ash generated annually is still disposed in landfills.42 A majority of
this unused fly ash does not meet the specifications for use in portland cement
concrete (the dominant application), sometimes because of the high percentages
of unburned carbon, as measured by the loss on ignition (LOI) test. Values of
LOI of fly ash used in concrete are typically limited to six percent for Class C and
Class F fly ash according to ASTM C 618, whereas certain bituminous fly ashes
may have LOI values in excess of 15 to 20 percent. Higher unburned-carbon
contents increase water demand in concrete and can significantly raise chemical
admixture demand (especially air-entraining agents and superplasticizers).
Despite the limitations placed on fly ash for conventional concrete, it has been
demonstrated that CLSM can be successfully produced using a wide variety of fly
ash types and sources, including high-carbon fly ash that is not permitted in
concrete.
Because ASTM C 618 Class F and Class C fly ash are commonly
available commercially, they are common mineral admixtures used in CLSM.
There are numerous benefits of using fly ash in CLSM, including improved
flowability, reduced segregation and bleeding, and in many cases decreased
material cost.
16
2.2.3.2 Chemical Admixtures
Air-entraining agents are the most commonly used chemical admixtures in
CLSM, and set accelerators have been used to a lesser extent to increase the speed
of construction (e.g., earlier opening of traffic) and to minimize subsidence.
CLSM is often produced with relatively high air contents (e.g. 20-35 percent),
either through the use of potent, liquid air-entraining agents or with foaming
agents applied through a foaming gun.24,43,44 The advantages of such CLSM
include low density, improved insulation properties, reduced segregation and
bleeding, decreased water and/or cement content, improved frost resistance, and
lower material cost. Also, high air contents may be used to limit long-term
strength gain to assure future excavatability.
2.2.4 Water
There are no special requirements for water to be used in CLSM. As a
general rule, any water that is suitable for concrete will work well for CLSM.
2.3 MIXTURE PROPORTIONS
Currently, there is no standard mixture proportioning method that has been
widely adopted for CLSM. A group of studies have focused on CLSM mixture
proportions,29-30,45-46 but most CLSM producers specify mixtures based on past
experience with locally available materials. Because a variety of by-product and
non-standard materials are typically used in CLSM, it is difficult to find a
standard approach to mixture proportioning.
17
In general, there are four typical types of CLSM used in current practice,
as shown in Table 2.1.
Table 2.1: Four basic types of CLSM
Type Binder Flow-driving factors Aggregate
I (K-Krete) Cement, fly ash Water, fly ash Filler*
II (high fly ash) Cement, fly ash Water, fly ash Fly ash
III (high air) Cement High air, water Filler
IV (flash fill) Fly ash Water, fly ash Filler

* Note: Filler includes concrete sand, bottom ash, foundry sand, and high -fine aggregate, depending on the mixture
proportions.
Type I CLSM was the earliest type that appeared in the construction field.
The mixture proportions tend to vary with different applications, especially the
cement content, which is the main factor affecting the strength of CLSM. Type II
CLSM mixtures usually have cement contents of 4 to 5 percent of the total solid
materials. Because of its excellent flowability, Type II CLSM is used widely in
backfilling spaces where access for compaction is limited, if not impossible.
Type II CLSM could achieve compressive strength values as high as 690 kPa at
28 days. To produce Type III CLSM, potent chemical admixtures are required to
entrain large volume of air voids. Type III CLSM is efficient in reducing fill load
because a density as low as 673 kg/m3 can be reached by using foaming agents.47
Type IV CLSM is patented and little is known about its proportions, but the
18
mixtures are typically high in Class C fly ash content. Type I and III CLSM
mixtures are the most commonly used mixtures by state DOTs.25
Although the unconfined compressive strength of CLSM can be as high as
8 MPa, the vast majority of applications are designed for future excavatability and
have strengths of less than 1 to 1.4 MPa. For these mixtures, low cement contents
are used (e.g., 30 to 60 kg/m3 ), with or without fly ash. Certain mixtures use very
high fly ash contents (Type III), where fly ash effectively serves both as a binder
and as an aggregate. Tables 2.2 and 2.3 show what the FHWA considers to be
typical mixture proportions for low and high fly ash contents. Mixtures that do
not contain fly ash are often produced with high air contents (20 to 35%), where
the high air contents provide certain similar benefits as fly ash, including
improved flow, reduced segregation and bleeding, and decreased cement content.
It also minimizes the subsidence of backfill layers because of little bleeding.
Table 2.2: Typical mixture proportions for low fly ash content CLSM 42
Material Range (kg/m3) Typical mixture proportions (kg/m3)
Fly ash (6-14%)* 119 - 297 178
Cement 30 - 119 59
Sand 1483 - 1780 1542
Water 198 - 494 297
Total: 2076
* Higher calcium fly ash is used in lower amounts than low calcium fly ash
19
Table 2.3: Typical Mixture Proportions for High Fly Ash Content CLSM 42
Material Range (kg/m3) Typical mixture proportions (kg/m3)
Fly ash (95%) 949 - 1542 1234
Cement 47 – 74 62
Water 222 - 371 247
Total: 1543
ACI Committee 229 opposes the use of plastic fines such as clay because
they can produce deleterious results, such as increased shrinkage.1
Brewer categorized CLSM mixtures according to their specific
applications.48 CLSM mixture used in backfill is referred to as CLSM-CDF,
pavement base as CLSM-CPB, structural fill as CLSM-CSF, thermal fill as
CLSM-CTF, anti-corrosion fill as CLSM-ACF , and permeability fill as CLSM-
CPF. The term CDF often appears in literature. It is a term created by Brewer to
represent Controlled Density Fill, which indicates a uniform density from the
trench’s bottom to its top. The density of the CDF material can be adjusted by
changing its components.48
2.4 BATCHING, MIXING, AND TRANSPORTING
CLSM is typically batched, mixed, and transported in similar fashions as
concrete. Most CLSM is batched at ready-mix plants and mixed in truck mixers.
Because of the amount of fines included and the fact that CLSM does not usually
contain coarse aggregate (which would assist in breaking up cement clumps and
dispersing the mixture), it is important to mix CLSM thoroughly. There is no

20
standard mixing procedure for CLSM, and mixing time varies with CLSM types.
Different mixing procedures have been described in the literature.
The high fluidity of CLSM may create difficulties in transporting full or
near-full loads in ready mixed trucks. To address this potential problem, certain
producers hold back part of the mixing water and/or liquid air-entraining agents
for on-the-job-site additions rather than at the plant, thereby reducing the mixture
volume in the truck on route to the site. The high fluidity of CLSM affects
several aspects of placement (e.g., floating of pipes) and construction.
2.5 CLSM PROPERTIES AND TEST METHODS
Table 2.4 summarizes test methods and critical CLSM properties that
affect its performance in backfill, utility bedding, void fill, bridge approach, and
other applications.25 Currently, there are five ASTM test methods specifically
designed for CLSM. The table lists various test methods that are currently being
used to assess CLSM or tests that may have the potential to serve as a suitable
methods. CLSM is a hybrid material (falling between soil and concrete), and as
such, soil and concrete test methods may be adopted for use. Focus should be
placed on the most important CLSM properties, their effects on performance, and
current or potential test methods that are most applicable to CLSM.
21
2.5.1 Fresh CLSM Properties
2.5.1.1 Flowability
One of the most important attributes of CLSM is its capacity to easily flow
into confined areas and obtain required strength and/or density without the need
for conventional placing and compacting equipment.1 The self-leveling properties
of CLSM significantly reduce labor and increase construction speed and safety.
Because the enhanced flow properties of CLSM are critical to successful
placement and performance, flowability is measured regularly as an important
quality-control parameter.
Table 2.4: Currently used or potential test methods for CLSM

CLSM Test Methods
Property
AASHTO T 276, “Developing Early Age Compression Test Values and Projecting Later
Age Strengths"
ASTM D 4832, “Preparation and Testing of Controlled Low-Strength Material (CLSM) Test
Cylinders”
ASTM D 4429, “CBR (California Bearing Ratio) of Soils in Place”
Mechanical AASHTO T 193, “Standard Method of Test for The California Bearing Ratio”
properties AASHTO T 24 “Obtaining and Testing Drilled Cores and Sawed Beams of Concrete”
AASHTO T 222, “Nonrepetitive Static Plate Load Tests of Soils and Flexible Pavement
Components for Use in Evaluation and Design of Airport and Highway Pavements”
ASTM D 6024, “Test Method for Ball Drop on Controlled Low Strength Material to
Determine Suitability for Load Application”
AASHTO T 22, “Compressive Strength of Cylindrical Concrete Specimens”
AASHTO T 106, “Compressive Strength of Hydraulic Cement Mortar”
AASHTO T 292, “Standard Method of Test for Resilient Modulus of Subgrade Soils and
Untreated Base/Subbase Materials”
ASTM D 6103, “Standard Test Method for Flow Consistency of Controlled Low-Strength
Flow Material”
ASTM C 939, “Flow of Grout for Preplaced-Aggregate Concrete (Flow Cone Method)”
CRD-C611, “Method of Test for Flow of Grout Mixtures”
Unit weight ASTM D 6023, “Standard Test Method for Unit Weight, Yield, and Air Content
yield and (Gravimetric) of Controlled Low Strength Material”
air content AASHTO T 121, “Unit Weight, Yield, and Air Content (Gravimetric) of Concrete”
22
Table 2.4: Currently used or potential test methods for CLSM (continued)
CLSM Test methods
property
Hardening AASHTO T 197, “Time of Setting of Concrete Mixtures by Penetration Resistance”
time
Sampling ASTM D 5971, “Standard Practice for Sampling Freshly Mixed Controlled Low-
Strength Material”
AASHTO T 141, “Sampling Fresh Concrete”
Segregation ASTM C 940, “Standard Test Method for Expansion and Bleeding of Freshly Mixed
and bleeding Grouts for Pre- Placed Aggregate Concrete in the Laboratory”
Drying AASHTO T 160, “Shrinkage of Portland Cement Concrete”
shrinkage
Consolidation ASTM D 2435, “Standard Test Method for One-Dimensional Consolidation Properties
of Soil”
ASTM D 5084, “Measurement of Hydraulic Conductivity of Saturated Porous Materials
using a Flexible Wall Permeameter”
Permeability AASHTO T 277, “Rapid Determination of the Chloride Permeability of Concrete”
ASTM D 4525, “Standard Test Method for Permeability of Rocks by Flowing Air”
Freezing-and- AASHTO T 161, “Standard Test Method for Resistance of Concrete to Freezing and
thawing Thawing”
resistance ASTM D 560, “Standard Test Methods for Freezing and Thawing of Compacted Soil-
Cement Mixtures”
ASTM D 5334, “Standard Test Method for Determination of Thermal Conductivity of
Soil and Soft Rock by Thermal Needle Probe Procedure”
Thermal ASTM D 5335, “Standard Test Method for Linear Coefficient of Thermal Expansion of
properties Rock Using Bonded Electrical Resistance Strain Gages”
ASTM D 4612, “Standard Practice for Calculating Thermal Diffusivity of Rocks”
ASTM G 1, “Standard Practice for Preparing, Cleaning, and Evaluating Corrosion Test
Specimens”
Corrosion of
ASTM G 51, “Standard Test Method for Measuring pH of Soil for Use in Corrosion
metals in Testing”
CLSM Modified ASTM G 59, “Standard Practice for Conducting Potentiodynamic Polarization
Resistance Measurements”
Modified ASTM G 109, “Standard Test Method for Determining the Effects of Chemical
Admixtures on the Corrosion of Embedded Steel in Concrete Exposed to Chloride
Environments”
ASTM G 57, “Standard Test Method for Field Measurement of Soil Resistivity Using
the Wenner Four-Electrode Method”
Leaching ASTM D 4874, “Standard Test Method for Leaching Solid Material in a Column
Apparatus”
EPA SW-846, Method 1311, "Toxicity Characteristic Leaching Procedure (TCLP)"
Utility ASTM D 543, “Standard Practices for Evaluating the Resistance of Plastics to Chemical
degradation Reagents”
ASTM D 6103, “Standard Test Method for Flow Consistency of
Controlled Low Strength Material,” has become the most common test to measure
CLSM flow since its adoption by ASTM. The test method uses a 75-mm × 150-
mm cylinder, open at both ends, which is filled with CLSM and lifted between 2
23
to 4 seconds allowing the CLSM to slump and to increase in diameter. The final
diameter of “pancake” is typically used to differentiate between various degrees
of flowability. A final diameter of 200 mm or higher is typical for a highly
flowable mixture. Some researchers have found that this test method is not as
sensitive to consistency as the conventional slump test.49 Water and fine material
may run out of the cylinder, leaving a sand cylinder standing. When the slump
cone was used, the sand mix tended to flow with the water and fines.
ASTM C 143, “Standard Test Method for Slump of Hydraulic Cement
Concrete”, has been used to a lesser extent to measure the flowability of CLSM.
The Washington Aggregate and Concrete Association defines flowability ranges
associated with the slump cone as follows:50
Low flowability: less than 150 mm

Normal flowability: 150 to 200 mm
High flowability: greater than 200 mm
ASTM C 939, “Standard Test Method for Flow of Grout for Preplaced-
Aggregate Concrete,” measures the efflux time of CLSM as it passes through a
flow cone. Several state DOTs have specified this test method for CLSM, and the
Florida and Indiana Departments of Transportation require an efflux time of 30 ±
5 seconds.1
Similar to the workability of concrete, flowability is not a fundamental
property of CLSM. To be meaningful, it must be related to the type of
construction and requirement of flowability. There are many factors that
influence the flowability of CLSM, including CLSM components, aggregate
24
gradation and shape, air content, water content, fly ash type and quantity, and
others. Specific methods of performing ASTM D 6103 affect the value of
flowability, and the lifting force has been found to be particularly important.51
To achieve the desired flow, trial mixtures are recommended. Bhat
proposed the concept of a “flow curves” to study possible factors affecting the
flow of CLSM.10 He found that when proper contents of fly ash were used,
adequate flow was obtained. But when too much fly ash was used, more water
was required. The study was limited mainly to CLSM using foundry sand as
aggregate. Gray used rheology to study the flow behavior and regarded the grout
as a Herschel-Bulkley (HB) or Bingham type fluid.52 Unfortunately, this
approach may be too complicated for field application. Anti-washout admixture
(AWA) and high-range water-reducing admixture (HRWRA) were found to help
underwater placement and reduce cross-contamination.53 Janardhanam et al. have
found that superplasticizers could significantly reduce the water requirement of
flowable fly ash, but the high cost of this admixture usually limits its use in
CLSM. 54
2.5.1.2 Segregation and Bleeding
Because of the high-water content requirement for the high fluidity of
CLSM mixtures, there is the potential for excessive segregation and bleeding,
especially for Type I and II CLSM mixtures. Because all components of CLSM
are in suspension when being placed, it is natural that segregation and bleeding
tend to occur when mixtures consolidate. When water is the main flow-enhancing
25
ingredient and there is insufficient surface area (fines), segregation and bleeding
are likely to occur. Generally, the use of fly ash and air-entraining agents reduce
the water demand, thus minimizing the potential for segregation and excessive
bleeding. Using ASTM Test Method C 940, Hoopes found that the 30% air-
modified mix had 0% bleed water, while the non-air-modified CLSM mixture
yielded 2.4% bleed water at the top of the sample.24 Little has been reported on
segregation and bleeding of Type IV CLSM mixtures.
CLSM mixtures, even with high air content, are generally pumpable with
careful proportioning.
2.5.1.3 Hardening Time
ACI Committee 229 defines hardening time as the approximate period of
time required for CLSM to develop from the plastic state to a hardened state, with
sufficient strength to support the weight of an average person. 1 Hardening time
can be as short as one hour, but it generally takes three to five hours under normal
conditions.55 There are many factors affecting the hardening of CLSM. These
factors include binder type and content, aggregate type, drainage conditions,
proportioning of CLSM, temperature, humidity, and depth of fill.
Hardening time, as defined above, is approximate and likely originated
from the concept of “Standing-On Time” in concrete technology. The most
common and easiest approach is to use penetration resistance as an index of
hardening. ASTM C 403, “Standard Test Method for Time of Setting of Concrete
Mixtures by Penetration Resistance,” has been frequently utilized. The soil
26
pocket penetrometer has also been used to study the hardening of CLSM.
Terzhaghi’s bearing-capacity equation of a circular footing is the theoretical basis
of this method.10 Resistance values from the soil pocket penetrometer generally
correlate well with those of ASTM C 403. Penetration values are often specified
to schedule construction practices and the time to opening of traffic.
2.5.1.4 Subsidence
Subsidence or volume reduction of CLSM results from the loss of water
(through bleeding and absorption into surrounding environment) and/or entrapped
air. CLSM mixtures with significant bleed water may have considerable
subsidence, up to approximately 1 to 2 percent of the fill depth. 56 Subsidence
generally occurs during CLSM placement, up until the fresh material sets.
Accelerating admixtures or high early-strength cement may decrease subsidence
by reducing bleed water. Currently, little information is available on subsidence.
2.5.2 Hardened CLSM Properties
2.5.2.1 Compressive Strength
The strength components of CLSM can be considered in terms of
particulate and non-particulate components.10 The non-particulate component of
strength results from the cementitious (and pozzolanic) reaction of cement and fly
ash with water. The particulate component of strength is composed of pure
friction and dilatancy, similar to granular soil. For unconfined compressive
strength, the non-particulate component is dominant because granular soil has no

27
shear strength without confining pressure. Water cement ratio is considered as
the single most important parameter that determines the unconfined compressive
strength of CLSM mixtures.10 Some researchers have deemed cement content as a
more convenient index of strength development, possibly because of the
insensitivity of strength to water cement ratio variations at high levels (>3).20 The
type and amount of fly ash (if used) also have important effects on the strength
development of CLSM mixtures. Many mixture proportions have been studied in
laboratory testing.57,58 Because of the variety of component materials in CLSM, it
is difficult to develop a general model to predict the strength development of
CLSM mixtures.
Compressive strength is one of the most important CLSM properties and it
is generally measured by most practitioners. CLSM mixtures with compressive
strengths of 0.35 to 0.69 MPa (50 to 100 psi) have similar bearing capacities as
well-compacted granular soil. CLSM strength values are often specified for
future excavation (e.g., maximum allowed values of 0.35 to 1 MPa), pavement
construction, and opening of traffic.
ASTM D 4832, “Preparation and Testing of Controlled Low-Strength
Material (CLSM) Test Cylinders,” was adopted in 1995. It has gained acceptance
with state agencies and commercial testing laboratories. The relatively low
strength of CLSM may cause potential problems, such as damaging the test
cylinder when stripping of CLSM cylinders from plastic molds. Because large-
capacity concrete compression machines usually have poor accuracy in the
required low load ranges, the results often show significant variations.26 The
28
capacities of many load frames used to test CLSM specimens in research
laboratories are typically in the range of 1,300 kN to 2,220 kN. To fail a 150-mm
x 300-mm CLSM specimen with a compressive strength of 1 MPa, a load of only
about 18 kN is required. This load is only about one percent of the load frame
capacity in a typical concrete compression machine. According to one survey,
most state DOTs used concrete machines in the testing of CLSM compression
specimens.25 Researchers using a loading rate of 0.24 MPa/sec (35 psi/sec) failed
CLSM specimens in a few seconds and could not satisfy the minimum
requirement of two-minute failure time specified in the ASTM standard.10
2.5.2.2 Permeability
In certain applications, the permeability of CLSM to both liquids and
gases is essential for construction and field performance. Drainage
characteristics, durability and leaching potential are influenced by permeability.
Because of the relatively low permeability of CLSM, certain specifications limit
the use of CLSM as utility bedding for gas pipelines. Difficulties have been
reported in detecting gas-leaks in pipelines buried in CLSM mixtures.25 Typical
water permeability values of CLSM are in the range of 10-4 to 10-5 cm/sec and
may achieve 10-7 cm/sec in mixtures with high strength or high fine contents.1
CLSM mixtures with 30-percent and 21-percent air contents have yielded
permeability values of 1.7 x 10-2 and 1.2 x 10-3 cm/sec, respectively. 24
ASTM D 5084, “Standard Test Method for Measurement of Hydraulic
Conductivity of Saturated Porous Materials Using a Flexible Wall Permeameter,”
29
is often adopted for CLSM mixtures. In experimental work of Bhat et al, a
backpressure of around 550 kPa was applied and maintained.10 In his study, the
“B” value was checked. The “B” value is the ratio of pore water pressure change
to confining pressure change under undrained conditions:
∆u
B= ( 2.1)
∆σcon
Where ∆u = change of pore water pressure due to change of
confining pressure
∆σcon = change in confining pressure
A “B” value of 0.97 or higher was obtained and a constant head
permeability test was conducted with a hydraulic gradient of approximately 10 in
the study.10
When soil specimens are saturated, water is assumed to fill up the voids in
the soil skeleton and water becomes connected. Although the bulk modulus water
changes with temperature and confining pressure, it is much higher than that of
soil. If a soil sample is saturated and under undrained conditions, the change in
confining pressure will be carried mainly by the water rather than by the soil, as
demonstrated by high “B” values. For a CLSM samples, the skeleton has a
relatively higher bulk modulus, resulting from the bonding strength of the
hydration products. It can carry a portion of the change of confining pressure.
Only under high pressure can water enter certain small sized voids in the
hydration products. In Bhat’s study, a backpressure around 550 kPa was used and
it was higher than the strength of certain CLSM mixtures.10 The confining
pressure was not provided in that study. The pore structure of CLSM specimens
30
can be damaged by such a high pressure. A hydraulic gradient of 10 can seldom
be reached in field conditions. Precautions must be taken when evaluating
permeability values from different sources. CLSM mixtures may best be treated
as rock and the change of “B” value should be checked with the change in
confining pressure, rather than using the “B” value itself.
2.5.2.3 Shear Strength
The main objective of using CLSM is as a viable alternative to
conventional compacted fill. As such, it is natural to characterize the engineering
properties of CLSM using geotechnical parameters familiar to field engineers.
Both the triaxial shear test and direct shear test are commonly performed in soils
laboratories, using ASTM D 3080 for the direct shear test and USACE EM 1110-
2-1906 for the consolidated-drained triaxial shear test. The applicability of these
methods to CLSM deserves significant attention.
There have been only a few published studies focusing on the triaxial
shear strength testing of CLSM.10,11,24,43,45,57,59 The reported internal frictional
angles from triaxial compression tests range from 20 to 40 degrees. There are
even fewer published results of direct shear testing studies.16,24,60 The possible
reason is the difficulty in preparing direct shear samples.
2.5.2.4 California Bearing Ratio (CBR) and Resilient Modulus
As first proposed and used in the 1920s by O. J. Porter, an engineer of the
California State Highways Department, CBR is a penetration-type test developed
31
to evaluate the strength of subbase and subgrade materials. Through years of
development, AASHTO T 193, “The California Bearing Capacity”, is now a
standard method. To evaluate the potential of using CLSM as a subbase or
subgrade material, researchers have frequently evaluated CLSM mixtures by
measuring their CBR values.11,18,60-63
The resilient modulus is defined as the initial tangent modulus of the
stress-strain curve in a triaxial test. After conditioning, a deviator stress, ∆σ1 ,
approximating the working stress, is repeated applied until certain criteria are met.
This test was initially used to evaluate pavement subgrades. The resilient
modulus of CLSM can be evaluated if it is to be used as a subgrade material.
Currently, AASHTO method T 274, “Resilient Modulus of Unbound Granular
Base/Subbase Materials and Subgrade Soils (SHRP Protocol P46)”, is commonly
used. There have been very few studies on the resilient modulus of CLSM
mixtures.
2.5.2.4 Settlement and Consolidation
CLSM may reduce its volume as it releases its free water and entrapped
air through consolidation of the mixture. The consolidation of soil consists of
primary and secondary consolidation. Primary consolidation is mainly caused by
seeping of water. Secondary consolidation is caused by rolling, slipping, sliding,
to some extent by crushing at the particle contact points, and elastic distortions.
Similarly, the initial settlement of CLSM mixture within several hours after
placement maybe regarded as primary consolidation. The further reduction of
32
CLSM volume due to applied load is similar to secondary consolidation of soil.
Because of the relatively strong and rigid skeleton of CLSM after hardening,
consolidation is expected to be very small and corresponding settlement of
backfill will be negligible.
ASTM D 2435, “Standard Test Method for One-Dimensional Properties of
Soil”, can be used to measure the consolidation of CLSM mixtures. The test can
estimate both the rate and total amount of settlement of CLSM used in various
applications. For Type III CLSM mixture, the coefficient of volume
compressibility is in the range of compacted dense gravel fill.24 For non-air-
modified CLSM mixtures, the coefficients are expected to be significantly
smaller.
2.5.2.5 Drying Shrinkage
High water-cement ratio and high water content are two factors known to
cause excessive drying shrinkage in concrete. Although CLSM has higher water
cement ratios and higher water contents than concrete, limited research on drying
shrinkage has shown that CLSM may not exhibit much more shrinkage than
concrete. Typical reported linear-shrinkage values of CLSM are in the range of
0.02 to 0.05 percent, which is similar to concrete with low drying shrinkage.1 The
shrinkage and expansion of CLSM tend to continue varying throughout testing.
Gandham et al found that the maximum shrinkage and expansion values of CLSM
were generally less than the acceptable limit established for concrete.41 Lucht
also showed that shrinkage of CLSM was minimal when a shrinkage-ring method
33
was used.13 A shrinkage ring is often used to measure the cracking of concrete
cast around a steel ring. This approach represents 100% restraint and can be used
to assess different materials and mixtures.
Most published studies have used the conventional concrete method to
measure shrinkage of CLSM specimens.41 This method specifies embedding gage
studs at both ends of the specimens and measuring the length change. Careful
handling of the shrinkage prisms during form removal and subsequent
measurements are required. CLSM specimens could be damaged when using this
approach because of the lower strengths of CLSM. Thus, this approach may not
be appropriate for CLSM. Lucht have also used the shrinkage ring method, which
is not an adopted standard.13
2.5.2.6 Thermal Conductivity
Moisture content and dry density are two major factors affecting thermal
conductivity of CLSM. Less important factors include mineral composition,
particle shape and size, gradation characteristics, organic content and specific
gravity. Low density, air-modified CLSM (Type III) is particularly suitable for
pipe backfill because of its enhanced insulating properties. The choice of thermal
conductivity characteristics depends on the specific application.
Methods to measure thermal conductivity of CLSM include ASTM D
5334, “Determination of Thermal Conductivity of Soil and Soft Rock by Thermal
Needle Probe Procedure”, and ASTM C 177, “Steady-State Heat Flux
Measurements and Thermal Transmission Properties by Means of the Guarded-
34
Hot-Plate Apparatus”. Institute of Electrical and Electronics Engineers (IEEE)
Standard 442-1981 has also been used to measure thermal resistivity of CLSM.27
2.5.2.7 Excavatability
To compete with conventional compacted fill in many applications, CLSM
must be excavatable at later ages. Researchers often use unconfined compressive
strength of CLSM as the index of excavatability.6 In general, CLSM with a
compressive strength of 350 kPa or less can be excavated manually and CLSM
with the range of 690 to 1400 kPa can be removed by mechanical equipment,
2
such as backhoes. Engineers in Hamilton County, Ohio, specify a Removability
Modulus to estimate future excavation of CLSM. It is a function of both the
thirty-day unconfined compressive strength and the dry unit weight. If the
calculated value of the Removability Modulus is less than 1.0, the specific
mixture is deemed to be removable in the future.
To predict excavatability, it is important to predict the long-term strength
development of CLSM. When foundry sand was used in CLSM, the unconfined
compressive strength increased up to 30% from 28 days to 91 days. 10 When the
cement content was high, the strength of CLSM more than doubled from the age
of four to thirteen weeks.64 . Fly ash types and contents also affect the ultimate
strength of CLSM.
CLSM mixtures with high quantities of coarse aggregate may become
very difficult to remove manually even when its strength is low. On the contrary,
Type III CLSM with strengths of 2.1 MPa can be removed with a backhoe.1
35
There is no existing laboratory test method to estimate the actual
excavatability/digibility of CLSM in the field. Actual field excavation is needed
to assess the ease of CLSM removal.
2.5.2.8 Freezing-and-Thawing Resistance
Freezing-and-thawing resistance is an important durability parameter for
construction materials. There have been several laboratory and field studies on
the freezing and thawing resistance of CLSM.4,11,17,57,65 Type II CLSM broke into
pieces about the size of a hand when it was placed in a zone of total water
saturation and subjected to severe winter freezing and temperatures below –18
°C.4 Without water saturation, Type II CLSM appears to perform well under
freeze-and-thaw conditions in the field.4 Although CLSM was considered a
suitable backfill material, Gress proposed that the top 5 to 150 mm of CLSM
mixture fill be replaced by a frost heave-compatible material for a uniform
heaving of pavement and trench after 24 hours of CLSM placement.65 Type III
CLSM and Type II CLSM with a high cement content (10%) are recommended
for construction applications where severe freezing is expected. According to
Hoopes, Type I CLSM specimens failed in only five cycles during the testing
process of ASTM D 560.24 Type III CLSM mixtures with 21% and 30% air
contents performed satisfactorily when they were evaluated with ASTM D 560.
Conventional concrete test methods for freezing-and-thawing resistance
have been found to be severe for CLSM. The low strength and the high
permeability of CLSM make it susceptible to damage under testing conditions for
36
concrete. The high permeability allows water to enter CLSM samples easily and
ice lenses to form inside the specimens. As a result, the material is damaged
internally. This phenomenon is very similar to the frost heave of soil. The
internal hydraulic pressure generated during water freezing also damages the
skeletal structure of CLSM. Nantung have modified existing concrete test
methods by exposing CLSM specimens to less severe environments to simulate
field conditions.17 ASTM D 560 is a method to measure the ability of soil-cement
mixtures to resist freezing-and-thawing cycles. Because of the similarity of
CLSM and soil-cement, the method is beginning to gain acceptance in the
research community.45,65 This method is much less severe than those for concrete
and requires only 12 freezing-and-thawing cycles.
2.5.2.9 Erosion
There is no standard procedure for testing the erodibility of CLSM. Dolen
have tested erodibility according to the guidelines of Corps of Engineers
publication REMR-GT-15.59 Tap water is injected through a formed 3.2-mm
diameter hole, lengthwise through the center of the cylindrical test sample. The
water has a pressure of approximately 275 kPa and the test is continued for 28
days. Then, the total mass of eroded material is determined. Specimens with
sufficiently high strength have demonstrated an erodibility where there was a
mass loss of approximately 0.03%, based on eroded material retained on the 75
µm (No. 200) sieve after 28 days.
37
Krell found that flowable fly ash resisted erosion in both its plastic and
hardened states.4 A tank was constructed with water being pumped continuously
across a sample. This test was developed to compare materials and determine
how they affect water quality. CLSM was shown to be superior to the other
backfill materials. This property eliminates the need for silt curtains, containment
by weirs, and other environmental protection systems when CLSM is used as a
backfill under water.
Type I CLSM was also used in Burlington, Iowa to prevent erosion. 6 A
CLSM mixture with unconfined compressive strength of 3,448 kPa (500 psi) was
used to fill the voids of riprap in a drainage ditch. The voids in front of and
behind the concrete floodwall along the Mississippi River were filled with CLSM
having a compressive strength of 690 kPa (100 psi). Similar material was also
used as an erosion control mat along the Iowa River.
2.5.2.10 Leaching and Environmental Impact
With more and more waste materials or by-product materials, such as
foundry sand and fly ash, introduced into CLSM mixtures, some environmental
concerns have been raised. Heavy-metal ions may leach into the environment
more easily from CLSM than from concrete because of the higher permeability of
CLSM. There are two approaches to assess this problem. One approach is to test
the leachate of source materials and the other is to observe the long-term field
performance of actual CLSM applications. Because CLSM is a relatively new
technology, research has mainly focused on the first approach.
38
The Extraction Procedure (EP) toxicity method, Toxicity Characteristic
Leaching Procedure (TCLP) method, American Foundry Society (AFS) method,
and ASTM method are well known laboratory test procedures to determine the
leaching characteristics of waste materials. Heavy metals include arsenic, barium,
cadmium, chromium, lead, mercury, selenium, and silver. Bhat used “Bio-assay”
testing on foundry sand to predict the effects of foundry sand components,
including organics.10
Bhat also measured the pH values and toxicity of expressed pore solutions
from both source materials and CLSM mixtures. Only one of the eleven mixtures
evaluated was found to be toxic.10 Naik et al. and Gandham et al. studied the pore
solution leachate characteristics of different mixtures using the TCLP test (EPA
SW-846, Method 1311).39,41 They found that the toxic contents of the mixtures
were well below the EPA leachate standards.
2.5.2.11 Corrosion
Although corrosion of metal pipes buried in CLSM has not yet surfaced as
a serious problem in field applications, corrosion is a long-term process that may
appear as a problem in the future. Folliard et al. have discussed the potential
mechanisms of corrosion likely to occur in CLSM.26
Most information on the corrosion of metals in CLSM mixtures has been
obtained from laboratory testing results. Abelleira et al. found that CLSM
significantly improved corrosion performance of buried iron. Metal coupons
were buried in CLSM mixtures and were then submerged under corrosive water
39
containing sulfate,, chloride and bicarbonate ions.60 Ramme et al. tested the
resistivity of CLSM mixtures and found that the corrosion potential of CLSM (fly
ash slurry) was significantly less than that of typical soils used for trench
backfill.61 Krell considered that the alkalinity of fly ash and cement could inhibit
the rusting of iron and steel.4 The pH of extruded pore solution from CLSM
mixtures were also tested.
Several ASTM test methods may be utilized to evaluate the corrosion
potential of steel contacting with CLSM mixture. They include ASTM G 1,
“Standard Test Method for Preparing, Cleaning, and Evaluation Corrosion Test
Specimens”, ASTM G 51, “Standard Test Method for Measuring pH of Soil for
Use in Corrosion Testing”, and modified ASTM G 59, “Standard Practice for
Conducting Potentiodynamic Polarization Resistance Measurements”.
Although the corrosion potential of metals in CLSM is significantly
decreased with respect to soil, there may be a concern with galvanic corrosion.
Galvanic cells may form when a pipe contacts both soil and CLSM, two different
environments. If CLSM is only used as utility bedding and the upper portion of
the backfill is conventional compacted fill, a large galvanic cell will be formed
with a small anode and a large cathode. Even though these pipes are buried in
CLSM, a galvanic cell will form between sections in CLSM and sections in the
surrounding soils.
40
2.5.2.12 Utility Degradation
Degradation of non-metallic utilities in CLSM has not been reported as a
problem in field applications. Some plastics can degrade in high pH solutions.
Because CLSM is basic in nature, the long-term durability of various utilities in
CLSM needs to be investigated.56
2.6 ECONOMICS
A recent survey showed that most state DOTs considered the cost of
CLSM as the largest barrier for widespread use.25 Prices as high as $130 per
cubic meter were reported. When filling the same trench, CLSM was slightly
more expensive than conventional granular backfill.66 At the same time, because
of the flowability of CLSM, it is possible to reduce the trench dimensions and
reduce the cost below that of conventional backfill. Bhat et. al stated that savings
in cost could be achieved when waste foundry sand was used as fine aggregate in
CLSM mixtures.10
2.7 CURRENT PRACTICE
2.7.1 Case Histories
2.7.1.1 Introduction
In general, an embankment refers to a volume of earthen material that is
placed and compacted for the purpose of raising the grade of a roadway (or
railway) above the level of the existing surrounding ground surface. A fill refers
41
to a volume of earthen material that is placed and compacted for the purpose of
filling in a hole or depression.
To categorize the typical applications of CLSM, the following words are
defined. These definitions are used in the discussion to follow in this dissertation.
Bridge Replacement
Bridge replacement refers to the use of CLSM as construction material in
the cases of converting deteriorated bridges. Those bridges are not demolished
but changed into culverts or other structures by using CLSM to fully fill the
original space beneath the bridge.
Backfill
Backfill refers to the use of CLSM in filling the openings formed during
the installation or repair of utility. The material is only utilized to cover the utility
components and transfer loads above. No structural support is provided to
utilities in this case.
Structural Fill
Structural fill refers to the use of CLSM in construction or repair, where
CLSM is to act as a load-bearing and/or load-transferring unit. Applications
include foundation support, pavement bases, bridge approaches, soil retaining
structures, and other components.
Insulation and Isolation Fill
42
Insulation and isolation fill refers to the use of CLSM as insulation and/or
isolation material to take the advantage of its appropriate thermal and acoustic
properties.
Utility Bedding
Utility bedding refers to the use of CLSM as bedding material in utility
installation or repair to provide uniform support. Strictly speaking, utility
bedding belongs to the category of structural fill, but is listed separately here to
indicate its importance.
Erosion Control
Erosion control refers to the use of CLSM in anti-erosion applications,
such as filling the voids of riprap.
Void Filling
Void filling refers to the use of CLSM in the practice of filling voids from
settlement or abandoned space. The material is not intended to bear or transfer
significant loads.
Figure 2.1 illustrates the typical applications of CLSM.
43
2.7.1.2 Backfills
One common application of CLSM is to replace conventional compacted
backfill. The high flowability of CLSM eliminates or decreases the dependence
on accessibility of compaction of the equipment, crew productivity, and degree of
compaction verification. Granular or site-excavated backfill, even when
compacted properly in the required layer thickness may not achieve the
uniformity of CLSM. Folliard et al. found that the production rates when placing
CLSM could be substantially higher than conventional compacted backfill.25
CLSM
A B C
Figure 2.1: Applications of CLSM. Case A: backfill; Case B: utility

bedding; Case C: structural fill.
The following are case studies where CLSM has been used as backfill:
Case history 1: Because of severe settlement problems with soil backfill,
the city of Peoria, Illinois, tested CLSM for backfilling utility trenches in 1988.1
As a result, the city changed its backfilling procedure to require the use of CLSM
on all street openings.
44
Case history 2: At one location in Iowa, a laid-up limestone block
67
retaining wall bulged out, indicating the initial stage of failure. A new “H”-pile
wall with wood planks was built in front of the old wall and flowable mortar was
used to fill the gap between the two walls.
Case history 3: In a Maine DOT project, “Evaluation of Flowable Fill for
Use in Highway and Bridge Applications”, it was found that unshrinkable backfill
(25 kg cement per cubic meters, conventional concrete aggregates, 160 to 200 mm
slump, usually air entrained, maximum 28 days compressive strength of 0.4 MPa)
satisfied the technical requirements for utility cut restorations and this material
was cost effective compared to properly compacted and inspected granular
backfill.68
Case history 4: In 1988, during reconstruction of the New England
Thruway (I-95) between Pelham Parkway and Bartow Avenue, Albany, New
York state, lightweight CLSM (672kg/m3 , minimum 28 days compressive
strength of 0.69 MPa) was specified for backfilling certain utilities, which was
placed above a soft organic silt containing a strata of highly organic peat and
organic clay in a random pattern.47
Case history 5: During the construction of Edison’s 1350 MW Belle River
plant, 122,570 m3 of flowable fly ash were used in trench fill and 84,267 m3 was
used in the power house area. The project estimates indicated that more than $1
million were saved through the use of the cement-stabilized ash.3
45
Case history 6: In one tilt-up construction project in Denver, it was
estimated that approximately two days of construction time were saved by using
CLSM for floor construction. 69
Case history 7: The Minnesota District #9 Maintenance Department used
CLSM to fill voids under bridge pier footings in the Robert Street Bridge in St.
Paul, Minn. This avoided building cofferdams or replacing the bridge.
Approximately 126 m3 (165 yd 3 ) of K-Krete was placed under two piers at a cost
of $107,000.6
Case history 8: CLSM provided the solution when severe spring flooding
washed all of the fill from above and around a multiplate steel arch pipe in
Hutchinson County, South Dakota. The pipe was repaired in nine days at a cost
of $12,000. Replacing the pipe was originally estimated to cost $40,000. 6
Case history 9: In Des Moines County, Iowa, CLSM was used to fill a
void along a wooden culvert, where access for conventional repairs was not
available. 6
Case history 10: Controlled-density fill was utilized to completely fill an
underslab void for the US Navy at Rough & Ready Island, Stockton, California.
The total cost was less than 20 percent of the amount authorized by the owner for
placing and compacting conventional granular backfill.70
Case history 11: A hollow sidewalk cavity containing locker room
facilities was backfilled with CLSM in downtown Milwaukee during the summer
of 1984.71
46
Case history 12: CLSM was used in backfill excavations to remove oil-
contaminated soil adjacent to foundations of existing structures. CLSM was also
placed underwater at Harbor in Plymouth, Massachusetts.72
2.7.1.3 Bridge Replacement
The Iowa DOT developed a creative process to replace abandoned
bridges. This process involved installing metal culverts, pipes, or reinforced box
culverts under the bridges. Soil was used as forms at both ends of the bridge.
The space between the bottom of the bridge and the top of the culvert was
backfilled with CLSM.23
Case history 13: Iowa DOT engineers developed flowable mortar, which
includes sand, portland cement, fly ash and water, to convert 10 bridges into
culverts. US Route 30 was kept open to normal traffic most of the time. The cost
was one-third less than that of routine repair (1985).23
Case history 14: Six narrow bridges were replaced in a 9-mile length of U.
S. Route 30 between Woodbine and Logan in Harrison County, Iowa. Four of
these bridges were modified by using flowable mortar with concrete pipe culverts.
One was modified by using flowable mortar with a reinforced-concrete box
culvert. One was replaced with a new bridge. The installation of pipe culvert
with CLSM cost only one-third of a reinforced-concrete box culvert and one-
fourth of conventional bridge replacement.67,6
47
2.7.1.4 Structural Fill
Because of the high strength and low-consolidation potential of CLSM
compared with compacted fill, CLSM is used widely as a structural fill. In
structural fill, the compressive strength of CLSM can vary from 0.69 MPa to 8.3
MPa (100 psi to 1200 psi) depending upon the application. The structural
coefficient of CLSM layer is estimated to range from 0.16 to 0.28 for compressive
strengths from 2.8 MPa to 8.3 MPa (400 psi to 1,200 psi).1
There are five primary causes of bridge-approach settlement, which causes
“the bump at the end of the bridge”73 :
1. time-dependent consolidation of the embankment foundation soil under
the embankment,
2. time-dependent consolidation of the approach embankment,
3. poor compaction of abutment backfill caused by restricted access of
standard compaction equipment,
4. erosion of soil at the abutment face, and
5. poor drainage of the embankment and abutment backfill
State DOTs are especially interested in the application of CLSM in bridge
approach construction because use of CLSM may eliminate causes No. 3 and No.
4. Possibly, if CLSM is used for the bridge approach embankment, cause No. 2
can be avoided.
Case history 15: Near Boone, Iowa, 2,145 cubic meters of CLSM were
used to provide proper bearing capacity for the footing of a grain elevator that was
built on load bearing soil.7
48
Case history 16: In June 1979, a small peninsula about 48.8 meters long
and 9.1 meters wide was constructed of flowable fly ash at Edison's Monroe
Power Plant.3
Case history 17: CLSM was used to fill between a spread footing and the
bearing soil during a cast-in-place hospital parking garage construction. It was
also used for updating the manufacturing plant without interrupting the
manufacturing process. The mix was line pumped over 305 m during some
placements.74
Case history 18: In a micro tunneling project in Newark, California,
CLSM was used to stabilize the soil surrounding the sheet-piled shaft that would
be used to launch a micro tunnel boring machine. The use of this fly-ash-based
CLSM greatly improved the stability of the soils and increased the safety of the
shaft during the launch. The use of CLSM provided a saving of more than
$100,000 in the Newark project.75
Case history 19: In July 1995, a contractor placed more than 306 cubic
meters (400 cubic yards) of CLSM as backfill for the abutments of a bridge
located along Colorado State Highway 135 near Crested Butte, Colorado.
Approximately 3060 m3 (4000 yd 3 ) of CLSM was used on a large commercial
distribution center in Loveland, Colorado for foundation wall backfill in 1991.69
Case history 20: In the Muroran Bay Bridge project, a man-made island of
67 m diameter was built by underwater placement of a self-hardening slurry on a
soft seabed. From October 1988, 53,600 m3 of slurry, containing 46.1 x 106 kg of
fly ash, 22.2 x 106 kg of volcanic ash, and 2.6 x 106 kg of cement, was filled into a
49
steel-pile cofferdam during a three-month period. Observations included 1)
reduced lateral earth pressure on the cofferdam walls, resulting in smaller bending
moments and lateral displacements; and 2) higher strength of the island for
subsequent construction work. In spite of the low temperatures in winter, heat
supplied from the hydration of the slurry kept the fill temperature above 10 ºC and
strength development of the mixture was higher than expected.76
Case history 21: Pockets of soft material of an underlying soil layer were
excavated and then backfilled with flowable fly ash (Class F ash). The next
morning, the contractor continued his work on building the six-story apartment
building.4
Case history 22: CLSM mitigated roadway settlement above buried
culvert and pipes in Wyoming.21
Case history 23: The Oklahoma DOT and Oklahoma State University
School of Civil and Environmental Engineering conducted (1998) a research
program to evaluate options to minimize approach embankment settlement.77 The
research program involved construction of three new bridges on U.S. 177 north of
Stillwater, Oklahoma. They found that the use of CLSM as construction material
for approach embankment was a simple and reasonably cost-effective method to
reduce the potential for “the bump at the end of the bridge”.
Case history 24: Flowable fly ash was used to save Detroit's Northeast
78
Raw Water Tunnel, Detroit, Michigan, 1985-6. CLSM was used to support the
tunnel before the grouting was finished.
50
2.7.1.5 Utility Bedding
Inadequate bedding support is known to reduce the service life of pipes or
utilities. Improper bedding preparation can decrease the desired pipe carrying
capacity or cause excessive backfill settlement. There are several methods to test
the inherent strength of pipe sections. These methods include the two-edge
bearing test, the three-edge bearing test, the sand bearing test, and the Minnesota
bearing test. The three-edge bearing method is the simplest to perform. The ratio
of the strength of a pipe under any stated condition of loading, whether in the field
or in the laboratory, to its strength by the three-edge bearing test is called the load
factor or strength ratio for the stated condition and is designated as Lf. There are
four classes of bedding.79
Impermissible bedding (Class D), Figure 2.2, is the case where little or no
effort is taken to shape the foundation to fit the lower part of the conduit exterior
or to refill all spaces under and around the conduit with granular materials, at least
partially compacted.
51
Figure 2.2: Class D: impermissible beddings. load factor = 1.1. (From Soil
Engineering, 4th Edition, M. G. Spangler and R. L. Handy, Harper &
Row, Publisher, New York, 1982)
Ordinary bedding (Class C), shown in Figure 2.3, is the case where the
conduit is bedded with “ordinary” care in an earth foundation, pre-shaped to fit
the lower part of the conduit exterior with reasonable closeness for a width of at
least 50% of the conduit breadth. The remainder of the conduit is surrounded to a
height of at least 0.15 m above its top by granular materials that are shovel-placed
and shovel-tamped to completely fill all spaces under and adjacent to the conduit.
All this work must be done under the general direction of a competent engineer.
52
H > 7.3 m-13 mm per 305 mm of H
Figure 2.3: Class C ordinary beddings. load factor = 1.5 (From Soil
Engineering, 4th Edition, M. G. Spangler and R. L. Handy, Harper
& Row, Publisher, New York, 1982)
First class bedding (Class B), shown in Figure 2.4, is the case where the
conduit is carefully bedded on fine granular materials in an earth foundation that
is carefully pre-shaped by means of a template to fit the lower part of the conduit
exterior for a width of at least 60% of the conduit breadth. The remainder of the
conduit is entirely surrounded to a height of at least 0.3 m above its top by
granular materials that are carefully placed to completely fill all spaces under and
adjacent to the conduit. The granular materials are thoroughly tamped on each
side and under the conduit as far as practicable in layers not exceeding 0.15 m in
thickness.
53
Figure 2.4: Class B first class beddings. load factor for ditch conduits =1.9 (From
Soil Engineering, 4th Edition, M. G. Spangler and R. L. Handy,
Harper & Row, Publisher, New York, 1982)
Concrete-cradle bedding (Class A), Figure 2.5, is the case where the lower
part of the conduit exterior is bedded in plain or reinforced concrete of suitable
thickness under the lowest part of the conduit and extending upward on each side
of the conduit for a distance not greater than one-fourth the outside diameter.
When CLSM is used for utility bedding, there is no or only minimal preparation
needed. Labor-intensive compaction in the haunch zone is eliminated. The
flowability and strength characteristics of CLSM may make this material a cost-
effective alternative to conventional bedding materials.
Case history 25: The US Bureau of Reclamation (USBR) used CLSM
(soil-cement slurry) in 1964 as bedding material for 381- to 2438- mm diameter
concrete pipe along the entire Canadian River Aqueduct Project, which stretches
515 km (322 miles) from Amarillo to Lubbock, Texas. It was estimated that soil-
54
cement slurry reduced bedding costs by 40 percent. Production increased from
122 linear meters to 305 linear meters of pipe placed per shift.2
Figure 2.5: Class A: concrete-cradle bedding. load factor 2.25 to 3.4. (From
Soil Engineering, 4th Edition, M. G. Spangler and R. L. Handy,
Harper & Row, Publisher, New York, 1982)
Case history 26: In the research project FHWA/OH-93/014, Brewer
installed a pipe arch culvert (381 cm × 241 cm, 2745 cm long with only 840 mm
of CLSM top cover) in Paulding County, 1989 and a long span arch culvert (10.5
m long) in Cuyahoga County in 1990. He found that: 1): the minimum wall
thickness required for the structural capacity of flexible culverts, pipe arches and
long span arches, could be reduced when using CLSM-CDF as the backfill
because of a reduction in required handling stiffness; 2) a relationship was
established between the horizontal and vertical stress of CLSM-CDF backfill,
which was accompanied by the recognition of CLSM-CDF's cohesion
55
development; 3) more triaxial shear information is needed for wider CLSM-CDF
strength ranges, (0.35 to 10.4 MPa (50 to 1500 psi)) to help develop more
accurate determination of CLSM-CDF's cohesion development; and, 4) initially,
conventionally placed backfill exerts less vertical and horizontal load on a culvert
than does a CLSM-CDF backfill, but after CLSM-CDF's cohesion buildup the
reverse is true.20
Case history 27: Quick-setting CLSM was used for replacement of
approximately 305 m of 914-mm water main under numerous train tracks in
downtown Seattle.74
Case history 28: In 1995, for a project at Kent County International
Airport, CLSM was used to backfill a 1.5 m diameter rigid concrete pipe in a
narrow 3.1 m trench, where the material was placed up to the pipe springline. The
culvert pipe has been inspected three times since installation and it is performing
well.80
Case history 29: CLSM was specified as bedding material for a 32,000-m
pipe drainage system at the new Denver International Airport (constructed
between 1991 and 1993). The flowable backfill was designed to completely
surround the pipe and extend a minimum of 152 mm above the top of the pipe.69
Case history 30: A roadway over a myriad of petroleum pipes was needed
to accommodate construction traffic for a project at the SOHIO petroleum plant in
Toledo, Ohio. A rectangular area was excavated and backfilled with a monolithic
mass of CLSM to totally envelope the pipes. After a three-day curing period, a
56
100-mm granular layer was placed over the area as a wearing course for the
roadway. 6
2.7.1.6 Erosion Control
Erosion of soil or backfill is known to cause riverbank failure, bridge
abutment movement, bridge approach settlement, and other damage. The use of
CLSM may help eliminate or alleviate these problems.
Case history 31: 1. Two projects in Burlington, Iowa, used CLSM to
prevent erosion. 2. Wapello, Iowa's business district, utilized CLSM as an
erosion control mat to avoid further erosion of the Iowa River bank.6
Case history 32: Riprap was put into the erosion cavity and the spaces
between the rocks were filled with flowable mortar in Iowa counties.7
2.7.1.7 Void Fill
Because of its flowability, CLSM is an excellent material to fill the voids
inside and/or between objects. In practice, CLSM is used widely to fill tunnel
shafts, sewers, basements, and other underground structures.
Case history 33: CLSM was used to fill an abandoned sewer and a
deserted tunnel that passed under the Menomonee River in downtown Milwaukee.
The material was reported to flow 72 meters and 92 meters, respectively. Only
four hours were needed to fill an exploratory shaft 36.6 m deep, 3.7 m in diameter
with a 9.2 m long branch tunnel. A total of 602 cubic meters of CLSM were
cast.1
57
Case history 34: To verify the practicality of the hydraulic backfilling
concept, a small-scale field injection of about 765 m3 of ash-bentonite-water
slurry into an abandoned mine was completed at the Fairfax mine in Preston
County, West Virginia. It demonstrated that the hydraulic backfilling method is a
practical solution to ash disposal with collateral benefits in preventing Acid Mine
Drainage (AMD) and subsidence.52
Case history 35: CLSM was chosen to fill the an abondoned pipeline
beneath a critical segment of Interstate 70 near Officers Gulch, immediately east
of Copper Mountain, Colorado.69
Case history 36: Flowable fly ash was used to fill abandoned gasoline
tanks, tunnels, pits, and sewers.4
Case history 37: 1.CLSM was used to fill the a void under a 30.5-meter
long metal culvert undermined by water passing beneath. 2. At the University of
Iowa at Iowa City, old stream tunnels were filled with CLSM. 3. Many
abandoned fuel tanks have been filled with CLSM. 7
Case history 38: 1. Wisconsin Electric filled two obsolete steam service
tunnels in downtown Milwaukee in December 1983. One tunnel was 1.8 m in
diameter by 88.5 m long (1.8 x 88m) and the other was a 1.5 m high by 1.2 m
wide ellipsoid section. 2. A number of abandoned steam utility facilities in the
Menomonee River Valley of Milwaukee were filled using CLSM. 3. The
abandoned Indian Creek Parkway Sewer was filled with CLSM.71
Case history 39: In LaSalle, Illinois, 306 m3 of CLSM was pumped to fill
the basement of a building. In Seattle, Washington, 19,150 m3 voids over each
58
bus station in a tunnel was filled with CLSM. CLSM has also been used to fill
abandoned underground tanks.1
Case history 40: CLSM was used to fill old basements full of loose
materials including bricks, broken concrete, and timber.4
Case history 41: Iowa DOT filled two 7.6-m3 and one 3.8-m3 abandoned
fuel tanks with approximately 20 m3 of CLSM near Ames in 1987. The project
was completed at a cost of $1,140. The cost of removing the tanks was estimated
at approximately $8,000.6
Case history 42: With CLSM, the City of Des Moines repaired the grade
under the street pavement and filled an undermined culvert for Pope John Paul's
visit.7
2.7.1.8 Environmental Protection
Case history 43: In Seattle, Washington, a CLSM mixture containing a
contaminated soil was designed to encapsulate it during the construction of a new
manufacturing plant.74 The alternative choice of the owner is to haul the
contaminated soil to a landfill over 640 km away,
2.8 RESEARCH NEEDS
2.8.1 Introduction
Although CLSM has gained popularity in recent years, there are several
obstacles for which some engineers are reluctant to specify it for construction
projects. The purpose of this section is to identify these challenges and

59
corresponding research needed for CLSM. One important obstacle is that
engineers are not familiar with this material. Education is extremely important
for increased use and acceptance of CLSM.
2.8.2 Research Needs
The following are some important issues that deserve additional emphasis
in current and future research project.
2.8.2.1 Standard Test Methods
To date, there are only five ASTM test methods and no AASHTO method
on CLSM. The five ASTM test methods are:
• ASTM D 4832-95, “Standard Test Method for Preparation and Testing
of Controlled Low Strength Material (CLSM) Test Cylinders”;
• ASTM D 5971-96, “Standard Practice for Sampling Freshly Mixed
Controlled Low-Strength Material”;
• ASTM D 6023-96, “Standard Test Method for Unit Weight, Yield,
Cement Content, and Air Content (Gravimetric) of Controlled Low Strength
Material (CLSM)”;
• ASTM D 6024-96, “Standard Test Method for Ball Drop on Controlled
Low Strength Material (CLSM) to Determine Suitability for Load Application”;
• ASTM D 6103-97, “Standard Test Method for Flow Consistency of
Controlled Low Strength Material (CLSM)”.
60
Obviously, there is a lack of standard test methods for CLSM. Only a few
CLSM properties are routinely measured by state DOTs and commercial testing
laboratories, and even these properties are measured with different test methods.25
These different methods may yield different results. It is important to identify
suitable methods if they exist, and to create new methods, if necessary, to
characterize the relevant properties of CLSM.
2.8.2.2 Effective Use of By-product Materials in CLSM
One outstanding characteristic of CLSM is its ability to consume various
and large amounts of by-product materials. Different by-product materials are
used in different locations throughout the United States. These materials include
fly ash of different carbon contents, bottom ash, crushed concrete, high fines, and
crushed recycle glasses. Unfortunately, there is no general agreement among
CLSM users about how to best use and specify these materials, or how to address
specific issues that may arise from their inclusion in CLSM.25 Procedures or test
methods to identify suitable by-product materials to be used in current and future
CLSM applications are needed.
2.8.2.3 Excessive Long-Term Strength Gain
A technical and practical concern is the uncertainty of long-term strength
gain for CLSM because the majority of CLSM applications may eventually
require excavation (e.g., to repair or replace utilities). Strength of CLSM is
considered as the most important factor that determines the excavatability or
61
diggibility of CLSM mixtures placed in field. Density and aggregate type also
influence the actual ease of excavation. There is little published data or
information on this aspect.
To study the long-term strength gain of CLSM, it is important to study the
effects of constituent materials and different mixture proportions. For instance,
CLSM containing fly ash demonstrates more long-term strength gain because of
the pozzolanic and/or cementitious reaction of fly ash. Because CLSM is usually
made with locally available materials and the properties of these materials may
vary significantly, it is necessary to perform accelerated or long-term testing to
predict the long-term strength gain of CLSM mixtures.
2.8.2.4 Durability Aspects of CLSM
It is essential to study the durability characteristics of CLSM as a
construction material. There have not yet been significant durability problems
identified in field applications.25 But this may be due to the relatively short
period of time that CLSM has been used. Issues of particular interest include
freezing and thawing, corrosion, and chemical attack.
2.8.2.5 Environmental Impact
Fly ash, foundry sand, and other by-product materials are often used in
CLSM. These materials may contain heavy metals and other potentially harmful
materials that may leach into the surrounding environments. The leaching
potential and consequent results deserve further attention. Research is needed to
62
develop suitable test approaches and to provide specifications to warrant the safe
and effective use of by-product materials in CLSM.
2.8.2.6 Applicability of Soil Test Methods
As CLSM is mainly used to replace conventional compacted granular fill,
it may be desirable to use soil testing methods to determine the properties of
CLSM. Many parameters describing soil can be used to define properties of
CLSM, such as resilient modulus, CBR values, water permeability, cohesion,
internal friction angle, consolidation and settlement, and bedding factors for
pipelines. However, the applicability of these methods should be verified in
practice.
Soil test methods that may be applied to CLSM should be simple methods,
using equipment available at most soil laboratories. Testing should only be
performed when specifically required for a given application, especially when
design parameters are needed as part of the design process.
63
Chapter 3: Materials and Mixture Proportions
3.1 INTRODUCTION
Component materials used in CLSM include fine aggregates, fly ash,
portland cement, and water. Chemical admixtures are sometimes used to obtain
different objectives, such as air-entraining and setting-accelerating. In this
chapter, materials utilized in this study are described. In addition, a description of
the research program is presented.
Thirty-eight mixes were initially cast to study the important properties of
CLSM. Additional follow-up mixtures were cast to further investigate certain
technical issues. Because of the importance of compressive strength in CLSM
specifications, particular emphasis was placed on evaluating parameters affecting
CLSM strength.
3.2 FINE AGGREGATE
3.2.1 Aggregate Type
One advantage of CLSM is its ability to accommodate waste and by-
product materials. Foundry sand and bottom ash are two of the most common by-
product materials used in CLSM, and both were included in this study. A natural
sand, typically used in concrete, was also used throughout the investigation.
64
3.2.2 Aggregate Gradation
The gradations of the three fine aggregates were determined in accordance
with ASTM C 136-96a, “Standard Test Method for Sieve Analysis of Fine and
Coarse Aggregates.” Because of the large-sized particles in the as-received
bottom ash, only materials passing the 6.35 mm sieve were used in this study.
The gradations of the three fine aggregates are shown in Table 3.1 and Figure 3.1.
For comparison, the upper and lower limits set by ASTM C 33, “Standard
Specification for Concrete Aggregates,” are also shown in the graph. Although
the river sand met the specification of ASTM C 33-93, it approached the coarse
limit of the gradation band. The bottom ash was found to be slightly coarser and
the foundry sand slightly finer than ASTM C 33 gradation limits.
Table 3.1: Aggregate gradations used in laboratory test program

Sieve # Concrete sand Bottom ash* Foundry sand
Cumulative %Cumulative % Cumulative % Cumulative % Cumulative %Cumulative %
retained passing retained passing retained passing
#4 (4.75 mm) 2.73 97.27 9.26 90.74 0.38 99.62
#8 (2.36mm) 20.81 79.19 31.70 68.30 4.26 95.74
#16 (1.18mm) 39.45 60.55 35.78 64.22 11.25 88.75
#30 (600µm) 59.11 40.89 49.66 50.34 30.46 69.54
#50 (300µm) 83.08 16.92 74.98 25.02 71.07 28.93
#100 (150µm) 96.02 3.98 87.85 12.15 96.12 3.88
#200 (75µm) 99.09 0.91 95.39 4.61 98.68 1.32
<#200 (<75 µm) 100 0 100 0 100 0
Fineness 3.0 2.89 2.14
modulus
* The bottom ash gradation reflects the particle size after “scalping” off the material retained above the 6.35
mm sieve
65
100
90
80
70
Percent passing (%)
60
50
40 River sand
30 Bottom ash
Foundry sand
20 ASTM C33 (lower limit)
ASTM C33 (upper limit)
10
0
0.1 1 10
Sieve opening size (mm)
Figure 3.1: Gradations of three fine aggregates and ASTM C 33 limits.
3.2.3 Absorption Capacity and Specific Gravity of Fine Aggregates
The absorption capacity and specific gravity of each aggregate were found
in accordance with ASTM C 128-97, “Standard Test Method for Specific Gravity
and Absorption of Fine Aggregate.” The surface-saturated dry (SSD) specific
gravity and absorption capacity of each aggregate were determined. The results
are shown in Table 3.2. The Fineness Modulus (FM) of the river sand was
slightly higher than those used in concrete. FM is an empirical factor used in
66
concrete technology to assess the fineness of an aggregate. The cumulative
percentages retained sieves: No. 100 (150 ìm), No. 50 (300 ìm), No. 30 (600
ìm), No. 16 (1.18 mm), No. 8 (2.36 mm), No. 4 (4.75 mm), and larger –
increasing in the ratio of 2 to 1, are calculated and divided by 100. The value
obtain is called FM. The absorption capacities of foundry sand and bottom ash
were found to be significantly higher than aggregates typically used in concrete.
Loss on ignition (LOI) of fly ashes were also measured.
Table 3.2: Materials included in laboratory test program*
Material Type or class

Portland cement ASTM Type I (S.G.=3.15)
Fly ash ASTM Class F (LOI = 2.9%, S.G.=2.41)
ASTM Class C (LOI = 0.37%, S.G.=2.51)
High-carbon fly ash (LOI = 14.44%, S.G.=2.09)
Fine aggregate ASTM C 33 river sand (S.G.=2.60, Absorption=1.0%, FM = 3.0)
Foundry sand (ferrous) (S.G.=2.36, Absorption=5.6%, FM = 2.14)
Bottom ash** (S.G.= 2.28, Absorption = 8.9%, FM = 2.89)
Chemical admixtures Air-entraining agent (liquid)
Accelerating admixture (non-chloride)
* Different materials were used to measure permeability and triaxial shear strength, ASTM
Type II cement and a non-classified fly ash were used.
** Bottom ash is classified as a fine aggregate because of similar particle sizes.
The aggregates used in this study were based on saturated surface-dry
(SSD) state. The following equation was used to perform moisture correction:
Wbatch=WSSD(1+p)/(1+a) (3.1)
Where, p – moisture content
A – aggregate absorption capacity

67
3.2.4 Chemical Analyses of Foundry Sand
The chemical analysis of foundry sand was performed by a commercial
analytical laboratory and the results are shown in Table 3.3. The Material Safety
Data Sheet (MSDS) from the producer indicated that this material might comprise
0 to 4 % bentonite and 0 to 4 % coal fines.
Table 3.3: Chemical characteristics of foundry sand
Chemical composition, % by weight Foundry sand

Silicon Dioxide, SiO2 85.20
Aluminum Oxide, Al2 O3 3.92
Iron Oxide, Fe2 O3 3.46
Total (SiO2 + Al2 O3 + Fe2 O3 ) 92.58
Calcium Oxide, CaO 0.79
Magnesium Oxide, MgO 0.58
Sodium Oxide,Na 2 O 0.98
Potassium Oxide, K2 O 0.17
Titanium Dioxide, TiO2 0.21
Manganese Dioxide, MnO2 0.11
Phosphorus Pentoxide, P2 O5 0.00
Strontium Oxide, SrO 0.01
Barium Oxide, BaO 0.07
Sulfur Trioxide, SO3 0.20
Loss On Ignition (LOI) 4.30
3.3 MINERAL ADMIXTURES
3.3.1 Fly Ash
The CLSM mixtures in this study included three types of fly ash, ASTM C
618 Class F, Class C and high-carbon fly ashes. Fly ash with less than 70% SiO 2
and Al2 O3 , resulting in 15-35%CaO is classified as Class C fly ash. On the other
68
hand, fly ash with more than 70% SiO 2 and Al2 O3 , and lower CaO contents, is
classified as Class F fly ash. According to ASTM C 618–94a, “Standard
Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use as
a Mineral Admixture in Portland Cement Concrete”, the LOI limit maximum is
six percent for Class C and Class F fly ashes. The LOI of each fly ash was
measured in accordance with ASTM C 311-98b, “Standard Test Method for
Sampling and Testing Fly ash or Natural Pozzolans for Use as a Mineral
Admixture in Portland-Cement Concrete.” Higher LOI values in fly ash typically
reflect higher carbon contents. The high-carbon fly ash used in this study would
not typically be allowed for use in conventional concrete. High-carbon fly ash is
known to cause increased water demands and difficulties in entraining air in
concrete, but high-carbon fly ash has been allowed in CLSM mixtures by state
DOTs according a survey. 25 The LOI results of the fly ashes are shown in Table
3.2.
3.3.2 Chemical Analyses of Fly Ashes
Chemical analyses of the four fly ash used in this study were performed by
a commercial analytical laboratories, and the results are shown in Table 3.4.
Class F fly ash had high proportions of silica and alumina (a total of 85%) where
Class C had only 55% of these oxide. As a result, the Class C fly ash had a much
higher calcium oxide content. Because of the presence of free lime, some Class C
fly ashes rapidly set and harden when being mixed with water, and are considered
to be true hydraulic cements. The high-carbon fly ash had a higher calcium oxide
69
content than the Class F fly ash, but this by itself does not necessarily indicate that
high-carbon fly ash has higher reactivity than the Class F fly ash. The fly ash
used in water permeability and triaxial shear tests was also analyzed.
Table 3.4: Chemical composition of fly ashes

Chemical composition Class F Class C fly High- Fly ash
fly ash ash carbon fly used in
ash triaxial test
Silicon Dioxide, SiO2 55.24 34.40 47.29 56.41
Aluminum Oxide, Al2 O3 29.43 20.20 19.53 26.40
Iron Oxide, Fe 2 O3 5.19 5.76 5.21 3.48
Total (SiO 2 + Al2 O3 + Fe 2 O3 ) 89.86 60.36 72.03 86.29
Calcium Oxide, CaO 1.59 26.73 6.01 1.51
Magnesium Oxide, MgO 0.93 5.15 2.05 0.58
Sodium Oxide,Na2 O 0.24 1.58 2.47 0.28
Potassium Oxide, K2 O 2.18 0.36 0.83 1.57
Titanium Dioxide, TiO 2 1.44 1.32 0.75 1.43
Manganese Dioxide, MnO 2 0.03 0.06 0.04 0.03
Phosphorus Pentoxide, P2 O5 0.28 1.08 0.40 0.10
Strontium Oxide, SrO 0.10 0.39 0.31 0.12
Barium Oxide, BaO 0.07 0.60 0.30 0.06
Sulfur Trioxide, SO 3 0.38 1.98 0.38 0.12
Loss on Ignition (LOI) 2.90 0.37 14.44 7.92
3.4 CHEMICAL ADMIXTURES
3.4.1 Air-Entraining Agent
A liquid air-entraining agent (AEA), designed specifically for use in
CLSM, was used in selected CLSM mixtures. It is a viscous solution of organic
compounds. The average specific gravity of the agent is 1.0 and the average total
solids are 95%. This admixture is capable of entraining stable air contents
ranging from 15-35%.
70
3.4.2 Accelerating Agent
A liquid, non-chloride accelerating admixture, typically used in concrete,
was utilized in this study for a small number of mixtures. This agent belongs to
ASTM C 494 Type C, and the average specific gravity is 1.34. It has a solid
contents of 41.5%.
3.5 CEMENT
ASTM Type I portland cement was used in this study. ASTM Type II
cement was used only in the study of water permeability and triaxail shear tests.
Its chemical composition, provided by the manufacturer, is shown in Table 3.5.
Table 3.5: Chemical analysis of ASTM Type I portland cement (weight

percent)
Chemical Analysis Type I Type II
(%) (%)
SiO2 21.0 20.14
Al2 O3 4.9 4.54
Fe2 O3 2.3 3.73
CaO 64.8 62.11
MgO 1.7 4.08
Available alkalis as Na 2 O 0.3 0.85
Compound Composition Weight percent (%)
C3 S 62.0
C2 S 13.0
C3 A 9.0
C4 AF 7.0
3.6 MIXING WATER
Laboratory tap water was used throughout this study.
71
3.7 CLSM MIXTURE PROPORTIONS
3.7.1 Mixture Proportioning
This section summarizes the mixture proportions used in this study. There
were two distinct stages of mixing for this study. In the first stage, a total of 38
mixtures were cast and tested. In the second stage, additional mixtures were cast
to study certain issues in greater detail.
To determine the mixes for the first-stage study, important CLSM
properties (and corresponding test methods) were prioritized and classified into
the following three groups, depending on the research objectives of this study:
I. Important CLSM properties

- Flow, setting time, unconfined compressive strength, and corrosion
measured for all 38 mixtures
II. Potentially important CLSM properties
- Segregation and bleeding, subsidence, triaxial shear, CBR, resilient
modulus, water permeability, drying shrinkage, excavatability, freezing
and thawing, leaching/environmental impact
- Measured for selected mixtures only
- Only seeking “Order of Magnitude” differences
III. Less important CLSM properties
- Direct shear strength, air/gas permeability, consolidation,
thermal conductivity
- Not included in laboratory study
- Literature-based and existing-practice-based coverage only
The most common types of CLSM materials and mixtures were selected
for the initial laboratory study. For each of these mixtures, the types and amounts
of cement, fly ash, and aggregates were selected prior to mixing, and the water
content of each mixture was then adjusted to achieve a flow of about 200 mm, as
measured by ASTM D 6103-97, “Standard Test Method for Flow Consistency of
Controlled Low-Strength Material (CLSM).” The three main groups of mixtures
and the range of materials chosen for the initial 38 mixtures included:
72
CLSM (with fine aggregates)
Type I portland cement: 1 type, 2 levels - 30 kg/m3 , 60 kg/m3
Fly ash: 3 types, 3 levels - 0 kg/m3 , 180 kg/m3 , 360 kg/m3
Fine aggregate: 2 types, 1 level - 1500 kg/m3
Air content: 3 levels - Entrapped air only, 15-20% air, 25-30% air
(Air-entraining agents will not be used for CLSM containing fly
ash)
CLSM (without fine aggregates)

Type I portland cement: 1 type, 1 level - 60 kg/m3
Fly ash: 3 types, 1 level - 1200 kg/m3
Air content: 1 level - Entrapped air only
CLSM (with set accelerator)

Selected mixtures from the test matrix
After selecting the rationale for the mixture proportions, as previously
defined, and specifying the desired dosage range for the key components, a
statistical program (ECHIP) was used to generate portions for a test matrix of the
non-air-entrained CLSM mixtures. This software uses experimental design
concepts to produce statistically significant results with a minimal number of
trials. In other words, rather than producing CLSM with every possible
combination of material and dosage, which would not be practical, an optimized
test matrix was produced that could be used to predict test results across the entire
spectrum of variables. In addition, the program can be used to statistically
compare the results of one test to another or to determine the effects of individual
or combined variables on test results. The program is also designed to assess the
repeatability of test results by requiring duplication of certain mixtures within the
test matrix. The repeatability of a test method greatly affects its acceptance as a
standard test method.
73
Although an air-entraining dosage 13 times the manufacture’s
recommended dosage was used in the CLSM mixtures containing foundry sand,
less than 3% of air was introduced in these mixtures. This may be attributed to
the presence of coals and organic materials in the foundry sand. Thus, foundry
sand was not included in air-entrained CLSM mixtures. For the air-entrained
mixtures, a full test matrix was used. Seven CLSM mixtures other than the above
mentioned mixtures were chosen for the stage 1 study. As such, a total of 38
mixtures were tested.
In stage 2, based on the results of stage 1, more specific CLSM properties
were studied in further detail. For each aspect studied, specially selected CLSM
mixtures were cast.
3.7.2 Mixture Proportions
To maximize the practicality of the research program, two separate
mixture series were generated using the statistical software, one for non-air-
entrained CLSM (with fly ash), and one for air-entrained CLSM (without fly ash).
The non-air-entrained mixtures are shown in Table 3.6. Mixture proportions for
air-entrained mixes are shown in Table 3.7. To maintain continuity in mixture
numbering, two new mixtures, mixture #20 and #21, were substituted for the
originally proposed foundry sand mixtures.
74
Table 3.6: Non-air entrained CLSM mixture proportions
Mixture* Type I cement Fly ash type Fly ash content Fine aggregate
content type**
(kg/m3 ) (kg/m3 )
1 30 Class C 180 River sand
2 60 Class C 180 River sand
1r 30 Class C 180 River sand
9 60 Class F 360 Foundry sand
15 30 Class C 360 Foundry sand
4 30 Class F 360 River sand
5 60 Class F 180 Bottom ash
3 60 Class C 360 Bottom ash
8 60 High carbon 180 Foundry sand
10 30 High carbon 180 Bottom ash
12 30 Class C 360 Bottom ash
6 30 High carbon 360 River sand
3r 60 Class C 360 Bottom ash
4r 30 Class F 360 River sand
7 30 Class F 180 Foundry sand
11 60 High carbon 360 Bottom ash
14 60 Class F 360 River sand
13 60 Class C 360 Foundry sand
5r 60 Class F 180 Bottom ash
2r 60 Class C 180 River sand
* ECHIP randomizes order of mixtures and provides for duplicates.
** Fine aggregate content was held constant at 1500 kg/m3.
Because of the modifications to the initially planned matrix for the air-
entrained mixtures, the statistical program for designing the air-entrained mixtures
was not used. Mixtures covering all of the selected variables were cast and
evaluated. That is, two cement contents (30 kg/m3 and 60 kg/m3 ), two target air
contents (15-20% and 25-30%), and two aggregate types (concrete sand and
bottom ash) were used in all combinations to create a total of eight mixtures.
From these eight mixtures, three were selected for replicate mixtures, bringing the
total number of mixtures in Table 3.7 from 7 to 11.
75
Table 3.7: Air-entrained CLSM mixture proportions
Mix # Cement content Air content Fine aggregate
(kg/m3 ) type**
23 60 25-30% Bottom Ash
17 30 15-20% Bottom Ash
19 30 25-30% Bottom Ash
18 60 15-20% Concrete Sand
20* 60 15-20% Bottom Ash
21* 30 25-30% Concrete Sand
20* 60 15-20% Bottom Ash
* These mixes were substituted for the originally proposed mixtures because of difficulty in
entraining air in mixtures containing foundry sand. The originally proposed mixtures
containing foundry sand were still cast, but without entrained air. Table 3.8 shows these
mixtures (Mix # 29 and #30).
** Fine aggregate content was held constant at 1500 kg/m3.
The mixtures shown in Table 3.8 were strategically chosen to investigate
specific mixture types that may be of interest. The mixtures represent typical
CLSM paste mixtures (i.e., 5% cement, 95% fly ash) and also include mixtures
containing an accelerating admixture. Lastly, this table includes non-air-entrained
CLSM mixtures containing foundry sand (selected after the challenges
encountered in entraining air in mixtures containing foundry sand).
Mixtures #24 and #25 were CLSM mixtures without fine aggregates. The
effect of acceleration admixtures was studied on mixtures #26, #27 and #28.
Because of the high fine content of foundry sand, it was possible to design CLSM
mixtures without fly ash or high air contents. Two mixtures, #29 and #30, were
proposed to for further investigation.
76
Table 3.8: Selected CLSM mixtures included in Stage One
Type I cement Fly ash
Mixture content Fly ash content Air content Fine aggregate
# (kg/m3 ) type (kg/m3 ) type*
24 60 Class F 1200 Entrapped air only None
25 60 High 1200 Entrapped air only None
carbon
26** 60 None 0 25-30% River sand
27** 60 Class F 1200 Entrapped air only None
28** 60 Class F 180 Entrapped air only River sand
29 60 None 0 Entrapped air only Foundry sand
30 30 None 0 Entrapped air only Foundry sand
* For mixtures 26, 28, 29, and 30, the fine aggregate content was held constant at 1500 kg/m3
** Mixtures contain accelerating admixture
After casting and testing the initially proposed mixtures (as summarized in
Table 3.6 through 3.8), additional mixtures were cast to further investigate or
refine selected test methods or to study selected CLSM properties in more detail.
The mixtures were based in most cases on previously cast mixtures (from the
original 38). However, there were several variations used to evaluate certain
properties or tests.
Because the compressive strength of CLSM is the most important property
commonly measured (and often the only hardened property measured), particular
emphasis was placed on developing a refined test that is more reliable and
reproducible. Issues such as load rate, curing condition, temperature effects, and
capping methods were studied in detail. The development of an improved
compressive strength test method is also critical because of the inclusion of
strength in most specifications, especially as it relates to excavatability.
77
Nine sets of additional mixtures were cast and will be referred to
throughout this dissertation by the designation of Mixture Series A through I, as
shown below.
Mixture Number of
Series Description Mixtures
A Effects of load rate on compressive strength (Table 3.9) 7
B Effects of curing and air-drying on compressive strength (Table 3.10) 2
C Long-term strength gain and excavatability (Table 3.11) 9
D Freezing and thawing resistance (Table 3.12) 11
E Alternative capping materials for compression cylinders (Table 3.13) 8
F Effects of drainage on compression cylinders (Table 3.14) 8
G Effects of storage conditions on compression cylinders (Table 3.15) 10
H Effects of temperature and humidity on compressive strength (Table 3.16) 6
I Permeability and triaxial shear strength (Table 3.17) 6
Tables 3.9 through 3.17 show the mixture proportions for the additional
investigations. To be consistent with the other mixtures in this study, the
aggregate content was held constant at 1500 kg/m3 . These tables also contain
selected information on the fresh CLSM properties. Later in Chapter 5, the
findings of these investigations are provided. These specialty studies were
designed to assist in refining and optimizing the test methods.
78
Table 3.9: CLSM mixture proportions for load-rate study
Mixture Cement Fly ash Fly ash Fine Water Flow Air Unit
3
content type content aggregate type (kg/m ) (mm) content weight
3 3
(kg/m ) (kg/m ) (%) (kg/m3 )
A-1 60 None None River sand 156 175 25.0 1739
A-2 60 Class F 360 Foundry sand 520 200 1.9 1755
A-3 60 Class F 1200 None 486 213 1.6 1620
A-4 30 Class C 180 River sand 265 330 1.7 2161
A-5 30 Class C 180 River sand 213 216 Not 2226
measured
A-6 60 Class F 1200 None 501 216 Not 1635
measured
A-7 60 None None River sand 156 165 24.5 1740
Table 3.10: CLSM mixtures for cylinder curing/specimen conditioning study

Mixture Cement content Fly ash Fly ash content Fine aggregate Water Flow
(kg/m3 ) Type (kg/m3 ) type (kg/m3 ) (mm)
B-1 30 Class C 180 Concrete Sand 203 250
B-2 30 Class C 180 Concrete Sand 189 200
Table 3.11: CLSM mixtures for excavation study

Mixture Cement River sand Fly ash Fly ash type Water Flow Air content Unit
(kg/m3 ) content content (kg/m3 ) (mm) (%) weight
3 3
(kg/ m ) (kg/m ) (kg/m3 )
C-1 60 None 1195 Class F 485 200 Entrapped 1637
C-2 0 2000 275 Class C 252 229 Entrapped 2148
C-3 30 1500 None None 112 178 28% 1642
C-4 15 1500 180 Class F 177 200 Entrapped 2192
C-5 30 1500 180 Class F 175 200 Entrapped 2158
C-6 15 1500 180 High-carbon 224 216 Entrapped 2095
C-7 30 1500 180 High-carbon 224 216 Entrapped 2115
C-8 15 1500 180 Class C 170 206 Entrapped 2190
C-9 45 1500 None None 103 178 25.5 1652
79
Table 3.12: CLSM mixtures used for freezing and thawing study
Mixture Cement Sand Fly ash Water Flow Air content Unit weight
(kg/m3 ) (1500 kg/m3 ) (180kg/m3 ) (kg/m3 ) (mm) (%) (kg/m3 )
D-1 30 River None 119 180 27% 1630
D-2 30 River Class F 205 200 Entrapped 2196
D-3 30 River High-carbon 256 229 Entrapped 2078
D-4 30 River Class C 200 216 Entrapped 1980
D-7 30 Foundry Class F 425 238 Entrapped 1835
D-6 30 Foundry High-carbon 481 229 Entrapped 1757
D-5 30 Foundry Class C 399 200 Entrapped 1800
D-8 30 Bottom ash Class F 357 200 Entrapped 1870
D-9 30 Bottom ash High-carbon 407 200 Entrapped 1733
D-10 30 Bottom ash Class C 282 200 Entrapped 1896
D-11 45 River None 96 152 30% 1569
Table 3.13: Mixture proportions for alternative capping materials study

Cement Fly ash Fly ash Concret Water Flow Total Air Unit
Mixture content type content e sand (kg/m (mm) bleeding content weight
(kg/m3) (kg/m3) (kg/m3) 3
) (%) (%) (kg/m3)
E-1 60 Class F 1140 None 480 200 2.30 1.4 1630
E-2 None Class C 206 1500 155 200 0.00 1.2 1626
E-3 30 None None 1500 109 175 0.00 29 1651
E-4 15 Class F 180 1500 184 200 2.07 1.7 1693
E-5 30 Class F 180 1500 176 200 0.87 2.5 2180
E-6 15 High- 180 1500 202 200 2.50 1.8 2122
carbon
E-7 30 High- 180 1500 229 225 3.04 - 2136
carbon
E-8 15 Class C 180 1500 109 200 0.85 1.75 2235
Table 3.14: CLSM mixture proportions for drainage-condition study

Cement Fly ash Fly ash River Water Flow Total Air Fresh unit
Mixture content type content sand (kg/m3) (mm) bleeding content weight
(kg/m3) (kg/m3 (kg/m3) (Weight %) (%) (kg/m3)
)
F-1 60 Class F 1140 None 480 200 2.30 1.4 1630
F-2 None Class C 180 2000 250 200 0.00 1.2 1626
F-3 30 None None 1500 109 175 0.00 29 1651
F-4 15 Class F 180 1500 184 200 2.07 1.7 1693
F-5 30 Class F 180 1500 176 200 0.87 2.5 2180
F-6 15 High- 180 1500 202 200 2.50 1.8 2122
carbon
F-7 30 High- 180 1500 229 225 3.04 Not 2136
carbon measure
d
F-8 15 Class C 180 1500 109 200 0.85 1.75 2235
80
Table 3.15: CLSM mixture proportions for cylinder storage-condition study
Mixture Cement Sand Fly ash Fly ash Aggregate Water Flow Air content
(kg/m3 ) (kg/m3 ) (kg/m3 ) type type (kg/m3 ) (mm) (%)
G-1 60 - 1140 Class F - 485 200 Entrapped
G-2 0 2000 275 Class C River sand 252 200 Entrapped
G-3 30 1500 - - River sand 112 187 29
G-4 15 1500 180 Class F River sand 177 200 Entrapped
G-5 30 1500 180 Class F River sand 175 200 Entrapped
G-6 45 1500 - - River sand 103 190 30
G-7 30 1500 180 Class F Foundry sand 349 216 Entrapped
G-8 30 1500 180 Class C Foundry sand 352 190 Entrapped
G-9 30 1500 180 Class F Bottom ash 424 140 Entrapped
G-10 30 1500 180 Class C Bottom ash 367 152 Entrapped
Table 3.16: Mixture proportions for effects of curing temperatures and

humidity
Cement Fly ash Fly ash Concrete Water Flow Air Unit
Mixture content type content sand (kg/m3 ) (mm) content weight
(kg/m3 ) (kg/m3 ) (kg/m3 ) (%) (kg/m3 )
H-1 60 Class F 1200 none 492 220 2.35 1631
H-2 15 Class F 240 1500 197 240 1.15 2191
H-3 15 Class C 240 1500 175 240 1.35 2212
H-4 30 Class C 180 1500 181 200 1.2 2163
H-5 30 Class F 180 1500 188 220 1.35 2210
H-6 60 none 0 1500 123 190 25.5 1603
Table 3.17: CLSM mixture proportions for triaxial shear strength and water
permeability
Mixture Cement Fly ash Sand Water Flow Air Fresh unit Dry unit Moisture
(kg/m3 ) (kg/m3 (kg/m3 (kg/m3 ) (mm) content weight weight content
) ) (%) (kg/m3 ) (kg/m3 ) (%)
I-1 30 180 1500 283 200 1.30 2036 1746 16.55

I-2 60 180 1535 327 210 1.25 2077 1748 18.79
I-3 120 180 1485 335 220 0.65 2087 1759 18.61
I-4 60 - 1471 212 190 22.5 1607 1415 13.59
I-5 60 - 1181 172 210 27.0 1569 1381 13.65
I-6 60 120 - 420 200 0.5 1529 1133 35.00
81
Chapter 4: Test Methods and Procedures
4.1 INTRODUCTION
The originally proposed testing program is summarized in Table 4.1.
Table 4.2 provides a summary of all the tests ultimately performed in this study,
which includes the originally proposed tests, in addition to supplemental studies
that were later added to the program (using the mixtures in Tables 3.9 to 3.17).
This section provides information on the specific test methods used in this study,
including various modifications to selected methods. Test methods are grouped
into three categories based on the properties that they are intended to measure:
fresh CLSM properties, hardened CLSM properties, and durability properties.
Some properties were studied in more detail than others. As previously
mentioned, significant emphasis was placed on the unconfined compressive
strength of CLSM.
82
Table 4.1: Summary of initial test program for this study
CLSM Test Specimen Test ages
property method(s) details
IMPORTANT PROPERTIES - TESTING FOR ALL MIXTURES
Flow ASTM D 6103 75 mm x 150 mm After mixing
Setting/Hardening Accordingly (function of early
ASTM C 403 150 mm x 150 mm
time stiffness)
Compressive 75 mm x 150 mm & 3 cylinders each @ 3, 7, 28,
ASTM D 4832
strength 150 mm x 300 mm and 91 days
CLSM -sand ASTM G 1 75 mm x 150 mm 182 days
comparative study
ASTM G 1 &
Galvanic cells 100 mm x 200 mm 182 days and monthly
ASTM G 109
50 mm diameter x 100
pH ASTM G 51 7, 28, and 182 days
mm height
100 mm x 150 mm x
Resistivity ASTM G 57 28 and 182 days
50 mm prism
POTENTIALLY IMPORTANT PROPERTIES - TESTING FOR LIMITED MIXTURES
Segregation and No standard, see test Every 15 min. for first hour,
150 mm x 150 mm
bleeding procedure then hourly
Subsidence No standard 610 mm x 100 mm After hardening & 48 hours
Triaxial shear USACE EM 1110- 73 mm diameter, 150 3 samples each at 7, 28, and
strength 2-1906 mm height 91 days
California Bearing 152 mm diameter, 178
AASHTO T 193 1 sample at 28 days
Ratio (CBR) mm height
73 mm diameter, 150
Resilient modulus AASHTO T 292 1 sample at 28 days
mm height
73 mm diameter, 150 2 samples each at 28 and 91
Water permeability ASTM D 5084
mm height days
87.5 mm × 26.3 mm × 1 sample each daily for 7 days
Drying shrinkage No standard
1000 mm angle & weekly
450 mm x 450mm x
Excavatability No standard 91 days (and later ages)
300 mm
Chloride diffusion ASTM C 1152 150 mm x 300 mm 30 and 91 days
Freezing and 102 mm diameter, 125
ASTM D 560 28 days
thawing mm height
EPA SW-846,
Leaching/environ-
Method 1311
mental impact
(TCLP)
LESS IMPORTANT PROPERTIES - NO TESTING
136 mm diameter, 20-
Direct shear strength ASTM D 3080 None
24 mm height
Thermal ASTM D 5334 76 mm x 203 mm None
conductivity
Air/gas permeability ASTM D 4525 73 mm x 150 mm None
Consolidation ASTM D 2435 63 mm x 200 mm None
83
Table 4.2: Overview of all properties and tests studied
Initial study (see Table
CLSM property Additional studies
4.1)
Flow ASTM D 6103
Soil pocket penetrometer
Setting/Hardening time ASTM C 403
Pocket torvane
Sample size effect
Small vs large machine
Loading rate
Effect of drying sample for 4
to 8 hours before testing
Compressive strength ASTM D 4832 Alternative capping materials
Effects of drainage
Effects of curing methods
Effects of curing temperature
and relative humidity
CLSM-sand comparative study ASTM G 1
ASTM G 1 & ASTM G
Galvanic cells
109
pH ASTM G 51
Resistivity ASTM G 57
Segregation and Bleeding ASTM C 940
Subsidence No Standard
ASACE EM 1110-2-
Triaxial shear strength
1906
California Bearing Ratio (CBR) AASHTO T 193
Resilient modulus AASHTO T 292
Water permeability ASTM D 5084
Drying shrinkage No standard
Excavatability No standard Splitting tensile strength
Chloride diffusion ASTM C 1152
Freezing and thawing ASTM D 560 Effects on permeability
Direct shear strength None
Thermal conductivity None
Air/gas permeability None
Consolidation None
Chemical and toxicity
Leaching/environmental impact None analyses of constituent
materials
84
4.2 FRESH CLSM TEST METHODS
4.2.1 Mixing Procedure and Measurement of Flow, Air Content, and Unit
Weight
For each of the 38 initial mixtures studied trial mixing was performed to
determine the approximate water demand needed for the target 200 to 250 mm
flow. To measure flow in trial mixtures, as well as actual test mixtures, ASTM D
6103-97 was followed. This method uses a 75-mm x 150-mm plastic cylinder,
with the bottom removed. The cylinder is raised, and the flow diameter is
measured (sometimes referred to as the “pancake” diameter).
After determining the target water content, the actual mixtures were cast
and test samples were prepared. For all of the 38 mixtures, the constituent
materials were batched using a Sartoris 38-kg capacity scale (0.5-gram accuracy).
For the smaller mix volumes, a 0.028-m3 drum mixer was used. For the larger
mix volumes (needed for measuring additional properties on the six selected
mixtures), a high capacity (0.056-0.070-m3 ) laboratory mixer was used. The
mixing procedures were different for air-entrained and non-air-entrained
mixtures, as discussed below.
For non-air-entrained mixtures, the dry materials (e.g., sand, fly ash, and
cement) were first mixed with approximately half of the expected mixing water
(based on trial mixing) for three minutes, followed by a two-minute rest period.
After the rest period, the remainder of the batched water was added, followed by
three or more additional minutes of mixing. Immediately after mixing, flow
measurements were taken. In most cases, as a result of the trial mixing, the target
flow of 200 to 250 mm was obtained. If the flow was less than desired, small
85
amounts of water were added, followed by an additional minute of mixing, to
obtain the target flow. Rarely was the flow greater than the desired range. Care
was taken to avoid excessive bleeding or segregation.
For air-entrained CLSM mixtures, a relatively dry consistency (i.e., zero
slump) was first obtained in the mixer, followed by the addition of the AEA. This
process was necessary because of the potency of the AEA. If the AEA was added
to an already fluid mixture, the flow would far exceed the desired range, and the
mixture would often suffer from excessive bleeding.
Immediately after mixing, the flow was measured according to ASTM D
6103, and the air content was measured using slightly modified ASTM C 231
(pressure method). If the flow and air content met the desired ranges, test samples
were then cast. If the air content was found to be lower than desired, additional
AEA was added (within the specified dosage of the admixture). It can often be
challenging to obtain both the desired flow and air content. When CLSM is
relatively dry, adding AEA will increase both air content and flow. Conversely,
adding water to a mixture that already has the desired air content but has a lower
than desired flow may actually decrease the air content (the flow may actually
increase or decrease, depending on the effects on air content). The experience
gained in trial mixing allowed converging on the target air content and flow in the
test mixtures much more efficiently. After obtaining the desired flow and air
content, the unit weight was measured using ASTM C 231. Upon successfully
meeting the target air content and flow, various test samples were cast and tested.
86
4.2.2 Setting and Hardening
The setting time of CLSM was measured using two methods, needle
penetration (ASTM C 403) and soil penetrometer (or “pocket” penetrometer). A
typical soil penetrometer is shown in Figure 4.1. Fresh CLSM was placed in two
standard containers (150 mm x 150 mm), one container for each test method.
Before each measurement, the bleed water was removed and weighed. The
cumulative weight of the bleed water removed throughout the test was used to
calculate the sample bleeding (bleed water mass divided by total sample mass).
The air temperature during testing was maintained at 21 °C.
It should be noted that the depth of penetration for the needle penetrometer
was approximately 25 mm, compared with only 6.4 mm for the soil penetrometer.
The differences in penetration depth, as well as contact area, may have a
significant impact on the measured values of penetration resistance. In particular,
the relatively high bleeding of CLSM (compared to concrete) may lead to
difficulties in early measurements, especially when using the smaller needle
penetrometer. Modifications were made to the needle penetrometer to minimize
the influence of bleed water and obtain a more accurate estimate of early
stiffness/strength gain.
87
Figure 4.1: A typical soil pocket penetrometer
To estimate the early-age shear strength of CLSM, a pocket vane shear
(VS) tester, as shown in Figure 4.2, was used. This device is often used by
geotechnical engineers to estimate the undrained shear strength of cohesive soils
in the field. It was utilized in this study to assess the setting/hardening of fresh
CLSM mixtures. A typical pocket VS tester is shown in Figure 4.2
88
Figure 4.2: Pocket vane shear tester used in this program
4.2.3 Segregation
The segregation of six CLSM mixtures was measured quantitatively. A
specially designed mold, which consisted of three separate cylindrical sections,
was used for this purpose. Each cylindrical section had a diameter of 100 mm and
a height of 75 mm. The sections were connected together vertically to produce a
sample cylinder with a diameter of 100 mm and a height of approximately 225
mm. After the samples had set, steel separating plates were inserted at the
junctions between the cylinder sections, thus yielding three separate samples
(upper, middle, and lower). Each sample was then wet sieved, using the #4, #8,
#16, #30, #50, #100 and #200 sieves. Each portion retained on these sieves was
89
then dried in the oven at 110 ºC for 24 hours and weighed. Material passing the
#200 sieve was not collected. Using the resultant gradation from each of the three
sections, a “pseudo” FM was calculated, using the same mathematical approach
as typically used for FM. As is the case with the normal treatment of the FM, two
different gradations can yield the same FM. As such, the overall gradation was
also considered when analyzing the results, as described later.
4.2.4 Subsidence
The term “subsidence” used here is defined as the settlement of the top
surface detected within several hours of CLSM mixture placement. To measure
subsidence, three 100-mm x 200-mm cylindrical plastic molds were vertically
aligned and firmly bonded together using high-strength tape. The bottoms of the
upper two molds were removed, allowing for the casting of a 600 mm high CLSM
specimen for subsidence measurement. Extra tall specimens were cast to ensure
that sufficient subsidence would occur and that the reduction in the column height
could be more readily measured. The choice of specimen height was rather
arbitrary. Because the maximum pressure increases with the height of fresh
CLSM mixture and higher pressure will result in denser material, it is anticipated
that the measured subsidence (in percentage) will increase with specimen height.
A 600-mm height was chosen to represent a typical layer thickness. After casting
the specimen, the top edge of the specimen was leveled off and periodic
measurements were made of the subsided surface, using a specially designed
probe, which was lowered from the top of the mold onto the center of the
90
subsided surface. Bleed water usually evaporated before the first measurement.
Measurements were conducted until subsidence ceased, usually within two hours
of casting the specimens. Because the friction from the mold wall restrains the
settlement of CLSM, the results from laboratory tests may under estimate the real
settlement in the field.
4.3 HARDENED PROPERTIES OF CLSM
4.3.1 Unconfined Compressive Strength
Because of the importance of the compressive strength of CLSM in
specifications, design, and construction, considerable emphasis was placed on
developing a test method with improved accuracy and reliability. This section
describes the basic procedures followed to test the unconfined compressive
strength of the initial 38 mixtures, including methods of preparing test cylinders,
curing, capping, and testing. After describing this approach, information is
provided on the various modifications and improvements investigated using the
mixtures previously shown in Tables 3.9 through 3.16. The results of the initial
compression study, as well as the findings of the various follow-up studies, can be
used to develop and recommend an improved unconfined compression test for
CLSM.
4.3.1.1 Specimen Preparation
Because of the relatively low strength of CLSM (compared to concrete), it
is very important to carefully handle test cylinders, especially when stripping the
91
cylinders from the molds. Although steel split molds are a good option for
casting CLSM specimens, because of the large number of specimens cast in this
program, a lower cost method was used. Before casting CLSM cylinders for
compression testing, the plastic cylinder molds were cut lengthwise in half till the
bottom. Electrical tape was then used to bind the mold back to its original shape.
This approach only has negligible impact on the shape and size of CLSM
specimens. Figure 4.3 shows a typical plastic mold in this study. After mixing
the CLSM, the cylinders were filled and tapped lightly on the sides to remove
large entrapped air voids. Plastic lids were then placed firmly on the cylinders,
and the specimens were moved immediately to a moist-curing or “fog” room,
which was maintained at 100% relative humidity (RH) and 23 ºC. After curing in
the molds for seven days, the cylinders were then stripped by simply removing the
electrical tape and removing the CLSM specimens from the cylinders.
Conventional stripping tools were not used because of possible damage to the
specimens. The cylinders were then kept outside of their molds in the fog room
until testing. Some CLSM mixtures tend to leach and soften slightly upon long-
term fog-room exposure. An investigation on this issue is described later in this
chapter, in which the effects of cylinder storage were studied in detail.
Moist curing was selected for this study so that test results can be
compared from one laboratory to another, even though CLSM is rarely, if ever,
moist-cured in the field. The same argument can also be made for concrete
testing. That is, concrete is rarely moist cured for more than seven days (if at all)
92
in field applications, but standard curing in a fog room provides a benchmark for
specification and construction acceptance.
Figure 4.3: Plastic mold were pre-cut and taped before casting specimens
4.3.1.2 Compression Machine
Although there is an existing ASTM method for measuring the unconfined
compressive strength of CLSM (ASTM D 4832), some modifications were made
to the method for this project, as described later. Most of the compression tests
were performed on a relatively low-load capacity and machine with displacement
control (100 kN Instron machine), but some testing was also performed on a
larger capacity machine with load control (1780 kN Tinius Olson machine) to
evaluate the effects of machine capacity. When using the smaller machine,
displacement control was used. Additional testing was also performed to examine
93
the effects of cylinder size (75 x 100 mm, 100 x 200 mm, and 150 x 300 mm),
and, for this study, a constant apparent strain rate, 2,493 microstrain/min, was
used. For this, the crosshead displacement was set at 0.38 mm/min for the 150-
mm-high specimen, 0.51 mm/min for the 200-mm-high specimen, and 0.76
mm/min for the 300-mm-high specimen. The objective was to produce failure in
about the same amount of time for each cylinder size for a given mixture. A
floating, spherical head was used to minimize eccentricities while loading. The
load-displacement curves were automatically logged to data files on a desktop
computer. A typical compression machine suitable for the testing is shown in
Figure 4.4.
Figure 4.4: A typical set-up of compression testing using geotechnical

unconfined compression machine.
94
For the larger capacity compression machine, load-controlled testing was
employed, as is the case for concrete testing. The typical load rates used for
concrete, 138 to 345 kPa/sec, would fail most CLSM specimens in a matter of
seconds. Thus, a lower load rate of 6.9 kPa/sec was selected. This lower load
rate was possible on the machine for this study but may not be available for many
standard concrete compression machines.
Only the peak load was obtained from the large capacity machine, and no
deformation data was obtained, which again is typical of concrete testing. Sulfur
capping was used for almost all of the cylinders, except for some weaker mixtures
at early ages, where it was not possible to use sulfur caps. For these mixtures,
neoprene pads were used for testing at seven days. As previously mentioned,
several variations were investigated for the unconfined compression test,
including cylinder size, machine capacity, capping method, load rate, and curing.
The results of these studies are described later.
4.3.1.3 Effects of Loading Rate on Compressive Strength
ASTM D 4832-95 provides little guidance regarding load rate. The
method states only to “Apply the load at a constant rate such that the cylinder will
fail in not less than 2 min.” Because of the vagueness in defining the load rate,
additional testing was performed to investigate the effects of loading rate on
compressive strength.
Table 3.9 shows the mixture proportions used for the load rate study.
These mixtures (a total of seven) were selected based on previous testing of the 38
95
mixtures in this project and were chosen to provide a wide range of materials and
proportions. Using these seven mixtures, the effects of displacement rate (cross-
head displacement of small load-frame) on compressive strength and deformation
at peak load were studied.
From each of the mixtures in Table 3.9 (A1-A7), standard cylinders (75
mm x 150 mm) were cast and moist-cured until the age of testing. The effects of
loading rate were first assessed for mixtures A-1 to A-4 with a wide range of load
rates (0.13 mm/min, 0.38 mm/min, and 0.89 mm/min). After performing these
tests, a narrower range of load rates (0.25 mm/min, 0.38 mm/min, 0.51 mm/min,
and 0.64 mm/min) were evaluated using mixtures A5-A7. Later in this chapter,
the terms “wide range” and “narrow range” refer to the aforementioned
investigations of load rates. The aim was to determine a suitable load rate range
that produces repeatable compressive strength values and can be performed in a
relatively short time. The latter concern was due to the fact that several mixtures
from early research took a relatively long time (i.e., greater than 10 to 15 minutes)
to fail in compression under displacement control, which would not be ideal for a
testing laboratory that must test many cylinders daily.
4.3.1.4 Cylinder Curing and Conditioning
Another possible source of error and confusion in ASTM D 4832 involves
the curing conditions and the treatment of cylinders before testing. According to
ASTM D 4832, curing of CLSM cylinders is performed with the specimens in the
molds (in the fog room) until the time of testing. This is different than the
96
normal concrete approach to stripping the cylinders from their molds after about
the first day of curing and then curing them in the fog room. ASTM D 4832 also
specifies a drying time of 4 to 8 hours for test cylinders after the moist-curing
period and before they are tested in compression. Concrete cylinders, on the other
hand, are specified to remain moist until the time of testing, with no required
drying time. Research was conducted to investigate the effects of cylinder storage
(i.e., in or out of molds) and specimen conditioning or drying prior to testing.
The mixtures shown in Table 3.10 were used to study the effects of drying
time (0.5, 2, 4, and 8 hours) and cylinder curing (in mold, out of mold, and in
lime-saturated water) and on the compressive strength of CLSM. The curing
regimes studied included:
• Leaving CLSM in the molds and curing in the fog room at 23 o C

• Stripping the cylinders at 3 days and curing in the fog room at 23 o C.
• Stripping the cylinders at 7 days and curing in lime-saturated water at 23 o C
After curing according to the above regimes, the test cylinders were then
capped with sulfur-capping compound and loaded at a standard loading rate of
0.38 mm/min. The cylinders were kept moist until the time of testing, i.e., no
specimen drying was allowed.
Because of the interesting outcome of these tests, as described later,
additional mixtures were subsequently cast to study these issues in more detail.
To study variations in curing regimes, the mixtures in Table 3.15 were cast and
test cylinders were prepared. This study aimed to identify possible differences in
compressive strength when four different curing conditions were utilized, as
97
summarized in Table 4.3. Note that for curing condition D, the cylinders were
placed outside the laboratory and were exposed to the high summer temperature
and dry atmosphere of Austin, Texas. All cylinders were capped with sulfur
capping compound and tested at a loading rate of 0.38 mm/min.
Table 4.3: Four curing conditions used in the study (using G-series mixtures)
Curing Curing regime
condition
A (normal) Keep sample in mold with cap on, for seven days in fog room. Then
strip cylinder and keep cylinders in fog room until time of testing
B (mold) Keep sample in mold, with cap on, for seven days in fog room. Then
remove cap and keep cylinder in mold in fog room until time of testing
C (cap) Keep sample in mold, with cap on, in the fog room until time of testing
D (out) Keep sample in mold, with cap off, outdoors until time of testing
4.3.1.5 Effects of Curing Temperature and Humidity on Compressive Strength
As already addressed, the temperature that CLSM is exposed to during its
strength-gain process may be very important, especially when mixtures containing
certain fly ashes are used. Because CLSM is used in many different environments
in practice, the same mixture proportions could exhibit different strength values.
This study is intended to identify factors affecting strength gain of CLSM
mixtures. This study was a follow-up to earlier testing that suggested that
temperature plays a major role in CLSM strength development.
Three curing temperatures (10 ºC, 21 ºC, and 38 ºC) and six CLSM
mixtures (H-1 to H-6 in Table 3.16) were selected to study the strength gain of
CLSM across a range of practical construction conditions. CLSM was cast into
standard cylinder molds (75 mm x 150 mm) and moved to the appropriate
98
temperature-controlled chamber until the date of testing. The cylinders were
stored in two different manners. Half the cylinders from each mixture were
stripped after three days and returned to the same chamber until the time of testing
(without control over relative humidity in the chamber). This condition is
designated later in this report as “dry” curing. Temperature and humidity were
monitored throughout the test. The other half of the specimens from a given
mixture were kept inside the molds with the caps firmly placed on top until the
day of testing (designated as “wet” curing). These cylinders were placed directly
next to the cylinders that had already been stripped.
Cylinders were tested for compressive strength at 7, 28, and 91 days.
After compression testing, the moisture contents of the specimens were measured
to assess the effects of curing conditions on the moisture content (or evaporable
water content) and strength of CLSM.
4.3.1.6 Effects of Drainage Conditions on Compressive Strength
Unlike conventional concrete, CLSM is rarely, if ever cured in the field. It
is used as a geotechnical material and placed without concern for curing. During
the strength-gaining process of CLSM, much of the CLSM is continuously
exposed to air and in contact with the surrounding soil and/or structures.
Different environments may significantly affect the final strength of CLSM as the
water cement ratio may be affected by the seepage of water into surrounding
materials or the loss of water through evaporation of bleed water.
99
The effects of seepage and evaporation were investigated in a study using
the mixtures detailed in Table 3.14 (F-1 to F-8). This study also investigated the
effects of temperature on strength gain, using the fog room as a control and
ambient conditions (hot Texas summer weather) as a test condition. As described
later in this report, the findings of this temperature effects study were quite
interesting, and subsequent testing was performed using controlled-temperature
environments to further elucidate the influence of temperature on CLSM strength,
especially for mixtures containing high volumes of fly ash.
To simulate the field conditions, CLSM mixtures were cast in plastic
molds buried in loose sand. Figure 4.5 shows the test set-up. To simulate the
condition of no water loss, CLSM mixtures were cast in plastic molds without
holes and tight lids were put on (condition “Cap”) immediately after casting. To
simulate conditions that only surface water evaporation is possible, mixtures were
cast in plastic molds without holes and lids (condition “No cap”). To simulate
moderate water seepage, mixtures were cast into molds with seven uniformly
distributed 3.6 mm diameter holes on the bottom (condition “Bottom holes”). No
lid was put on the cylinders. To simulate severe drainage condition, mixtures
were cast into molds with holes not only on the bottom but also on the side
(condition “Side holes”). There were 36 holes on the wall and seven holes on the
bottom per mold. All holes were of 3.6 mm diameter. No lid was put on the
cylinders. To reduce local variations, CLSM specimens from a given mixture
were randomly placed throughout the test box
100
Loose
Mold sand
450 mm 300 mm
Compacted
450 mm sand
450 mm
Figure 4.5: Drainage condition simulation
4.3.1.7 Alternative Capping Materials for Compression Testing
In preliminary laboratory trials and throughout the whole laboratory
program, sulfur capping was found to be an effective method of obtaining
repeatable compressive strength data. However, for early age samples and/or for
particularly low strength cylinders, it may not be possible to cap cylinders with
sulfur because of the risk of damage to the sample. In these instances, neoprene
pads were used. In addition, some limited comparative studies were performed
early in the laboratory program in which sulfur capping was compared to
neoprene pads. However, because only limited testing was performed, it was
decided to significantly expand the scope of the original work to investigate not
only sulfur and neoprene, but also high-strength gypsum capping compound, for a
range of CLSM mixtures. Other motivations for studying alternatives to sulfur
101
are the potential health concerns over the fumes generated from sulfur capping
stations and the length of time needed to cap cylinders with sulfur.
The use of neoprene pads has gained popularity in recent years for
personnel testing concrete. In addition to avoiding the fumes associated with
sulfur, neoprene pads are easier and faster to use than having to cap with sulfur. It
is well known that higher strength concrete requires higher neoprene durometer
values, and vice-versa. Thus, for CLSM, it is expected that softer neoprene pads
(much softer than those used for concrete) may be needed. Another possible
alternative to sulfur capping compound is high strength gypsum, which is
sometimes used for concrete. Because of the high flowability of the gypsum
paste, it could be helpful in capping CLSM cylinders that are prone to suffering
edge damage (e.g., weak cylinders). The set-up of gypsum capping is shown in
Figure 4.6. In addition, gypsum is non-hazardous, without the odor and health
concerns associated with sulfur.
To address the capping-related issues, a comprehensive investigation of
alternative capping materials was launched. Included in this study were sulfur
caps, gypsum (or hydrostone) caps, and neoprene pads with durometer values of
20, 40, 50, 60, and 70. Eight mixtures (E-series) were used in this study, as
previously described in Table 3.13.
102
Figure 4.6: Method of capping cylinders with gypsum compound
4.3.2 Triaxial Shear Stre ngth
The conventional soil triaxial test (USACE EM 1110-2-1906) was
followed for testing six CLSM mixtures, as I-series shown in Table 3.17. The
samples were cast in Shelby tubes of approximately 70 mm diameter and were
stripped after seven days. The testing was performed under consolidated and
drained conditions. The pore water pressure was maintained at 34.5 kPa, and the
confining pressures were 69.0 kPa, 103.5 kPa, and 172.5 kPa, respectively.
Theses values resulted in effective confining stress of 34.5, 69.0, and 138.0 kPa,
respectively. The loading rate was 0.38 mm/min, the same loading rate used for
most of the unconfined compression tests. The tests were terminated when the
residual strength was reached or the stress-strain curve became essentially flat.
The approach of stress paths was used to interpret the data. By curve fitting, the
effective internal friction angle, φ´, and effective cohesion intercept, c´, were
found.
103
4.3.3 Water Permeability of CLSM
The water permeability of six CLSM mixtures (I-series) was measured
using ASTM D 5084 -90, “Standard Test Method for Measurement of Hydraulic
Conductivity of Saturated Porous Materials Using a Flexible Wall Permeameter”.
The cylinders were cured in the fog room for 28 days. A back pressure of 69 kPa
was applied and maintained until no additional water entered the sample
(approximately 30 minutes). This was assumed to represent a saturated condition.
Because the samples were moist-cured prior to testing, the samples were already
essentially saturated prior to sample conditioning. CLSM specimens were
relatively incompressible, thus, the requirement that B ≥ 0.95 was lifted. Rather,
the change of the “B” value with respect to the change of confining pressure was
monitored similar to the testing of rock. Constant-head test was used.
Eleven mixtures, designated as D-series in Table 3.12, were utilized to
study the freezing and thawing effects on permeability. The CLSM cylinders
were first moisture cured for 28 days and then were exposed to 12 freezing-and-
thawing cycles. Because CLSM specimens without air entrainment usually break
down in the testing process, polyvinyl hear-shrink plastic film was used to hold
the specimens together. The specimens were 100 mm × 125 mm. Porous stones
were wrapped on both ends of the specimens. Figure 4.7 shows the specimen set-
up. The falling-head method was used for these mixtures.
104
Figure 4.7: Specimen set-up for testing of permeability of CLSM mixture
after freezing-and-thawing cycles
4.3.4 Drying Shrinkage of CLSM Mixtures
No standard test methods exist for the measurement of the drying
shrinkage of CLSM. A method commonly used (but not a standard method) in
Germany for self-leveling floor screeds was modified and used in this study.
CLSM was cast into an 87.5 mm × 26.3 mm × 1000 mm steel “C” channel. This
angle had one fixed end plate with an anchor and one movable end plate with an
anchor. Prior to casting, wax paper was placed on the inside of the angle to
reduce friction. CLSM was then placed in the fabricated forms. The amount of
shrinkage was measured by an linear voltage displacement transducer (LVDT) of
105
an accuracy of 0.01 mm, which measured the displacement of the movable end
plate. Shrinkage measurements were taken daily for the first week and once a
week thereafter.
4.3.5 California Bearing Ratio
No significant modifications were made to AASHTO T 193-99, “Standard
Method of Test for The California Bearing Ratio,” for testing CLSM in this study.
The only difference was that the material (CLSM) was cast or poured into the
molds, without the need for compaction, as is needed for testing soils. After
seven days of curing, the collar was removed, and the surface was trimmed level
using a straight edge. The specimens were tested at the age of 28 days. Six
selected CLSM mixtures were tested.
4.3.6 Resilient Modulus
The resilient modulus of six selected CLSM mixtures was determined in
accordance with AASHTO T 292–91, “Standard Method of Test for Resilient
Modulus of Subgrade Soils and Untreated Base/Subbase Materials”. In trial
testing, it was found that the deviator stresses listed in Table 4 of AASHTO T
292-91 were not high enough to introduce sufficient deformations. Thus, the
selection of deviator stresses was based on experience performed at Texas A&M
University. Load conditioning of 41 kPa was used for the 1000 repetitions.
106
4.3.7 Excavation Study of CLSM Mixtures
The excavatability of CLSM was assessed for six of the original 38 CLSM
mixtures included in this study to gain an “order of magnitude feel” for the
relative ease of excavating various CLSM mixtures. CLSM was cast into 450-
mm x 450-mm x 300-mm plywood boxes and allowed to harden. Attempts were
made to correlate “walkability” with soil penetrometer values as the CLSM
gained strength in the first few hours. Walkability is the time at which an average
person can walk on the material. Long-term excavatability was assessed at an age
of approximately nine months using typical hand tools, including a shovel and a
pick for six selected mixtures. The compressive strength of lab-cured cylinders
was also measured. In addition, a relatively new instrument, the Humboldt
GeoGauge® (stiffness gauge), was used at the time of excavation to attempt to
correlate excavatability with the stiffness of CLSM, as measured by the
GeoGauge®.
After the initial excavation study, a more comprehensive study on long-
term strength gain and excavatability was conducted. Nine CLSM mixtures (C-1
to C-9 in Table 3.11) were included in the study. A field penetrometer (field
version of ASTM C 403) was used to evaluate the strength gain of CLSM
mixtures in the boxes. The dynamic cone penetrometer (DCP) was also utilized
to estimate the excavatability of CLSM mixtures. The DCP is a modified and
simplified version of the penetrometer used by the Country Roads Board,
Victoria, Australia. It is used by geotechnical engineers to obtain an index of in-
situ CBR and to estimate the strength of soil as a function of depth. The testing
107
consists of dropping a hammer (8 kg in mass) from a height of 575 mm, which
forces a steel rod with a conical head into the CLSM or soil. The penetration
depth per blow is then recorded. The corresponding DCP index value was used to
estimate a soil strength value. This value is often correlated with CBR.
Another approach to estimate the excavatibility of CLSM is the
calculation of its removability modulus. In its CLSM specification, Hamilton
County, Ohio, defines a Removability Modulus (RE), as reproduced below in
Equation 1.a.
W 1 .5 × 104 × C 0 .5 (4.1)
RE =
106
Where, W – In situ unit weight (pcf)
C – Thirty-day unconfined compressive strength (psi)
When SI units are used, as required by AASHTO, the equation is rewritten
as Equation 1.b below:
W 1. 5 × 0.619 × C 0 .5 (4.2)
RE =
106
Where, W – In situ unit weight (kg/m3 )
C – 30-day unconfined compressive strength (kPa)
Engineers in Hamilton County, Ohio and the city of Cincinnati have found
this methodology to be effective in limiting long-term strength gain and ensuring
future excavatability. The same approach was used to calculate RE values for the
C-series mixtures and comparing the results to other direct or indirect indices of
CLSM strength gain. If a CLSM mixture demonstrates a RE value higher than or
equal to 1.0, it is deemed not excavatible in the future.
108
In preliminary investigations, it was found that the splitting tensile
strength of CLSM to be a very simple property to measure (without the need to
cap the cylinders), and the test results may be helpful in predicting excavatability.
In addition, the stress conditions of CLSM specimens under splitting tensile
testing may be quite similar to the stress conditions associated with digging the
material with a shovel or backhoe. Eight mixtures (E-series mixtures shown in
Table 3.13) were tested for tensile strength, and comparisons were made with
other CLSM properties, such as compressive strength.
4.4 DURABILITY OF CLSM
4.4.1 Freezing-and-Thawing Resistance
ASTM D 560, “Standard Test Methods for Freezing and Thawing
Compacted Soil-Cement Mixtures,” a method designed to measure the freeze-
thaw resistance of soil-cement mixtures, was used with one minor modification.
The modification was that the application of wire scratch brush on thawed
specimens was not required because of the low strength of CLSM. One freeze-
thaw cycle involved cycling between -18 °C (a freezer) and 23 °C (the fog room).
Samples were exposed to 12 cycles unless they suffered severe damage at an
earlier time. Mass loss was used as the index of damage. For the initial study,
three cylinders (100 mm x 125 mm) were moist-cured for seven days and three
other cylinders were moist-cured for 28 days prior to exposure to freeze-thaw
cycles. To study the effects of freeze-thaw damage on permeability the
specimens were moist-cured for 28 days prior to freezing-thawing exposure. It
109
should be noted that tests typically used for concrete, such as ASTM C 666, were
found to be too severe in preliminary trials, and the modified soil cement method
was found to be a more suitable approach.
4.4.2 Wetting-and-Drying Resistance of CLSM
A preliminary study on the wetting-and-drying resistance of CLSM
mixtures was performed. After 7 and 28 days of moist curing, the wetting-and
drying cycles were initiated. The specimens were soaked in water at 23 ºC for 24
hours and then dried for 24 hours at room temperature (relative humidity
approximately 50%). The mass of CLSM specimens was recorded before and
after soaking. The testing was stopped after 12 cycles. Six selected mixtures (#4,
#6, #23, #24, #22r, and #26) were investigated.
4.5 CORROSION OF DUCTILE IRON IN CLSM
4.5.1 Research Approach
Corrosion of underground utilities can cause premature failures and can
result in the need for costly repairs. Concerns have been expressed regarding the
corrosion performance of ductile iron pipe in soil and CLSM. Limited work has
been performed on corrosion of ductile iron pipe embedded in CLSM and work is
needed to either substantiate or dismiss these concerns. In general, several factors
can influence the corrosion performance of utilities embedded in both soils and
CLSM. These factors can include84 :
• the presence of aggressive chemicals or ions
110
• Acidity or alkalinity (pH of the solution at the pipe interface)
• electrical conductivity or resistivity of the soil or CLSM
• porosity (aeration) or drainage conditions near the utility
• moisture
As pitting depths depend on sizes of test specimens, it is difficult, if not
impossible, to simulate field performance of iron pipelines. Thus, in the testing
program, uniform corrosion will be assumed such that mass loss can be measured.
Because CLSM is quite different from native soils from an electrochemical
perspective, it is possible that a galvanic cell is formed by a pipe contacting
CLSM and soil at different parts of one cross-section or at different sections along
the pipe. For the first scenario, when a small portion of pipe contacts native soil
(forming an anode of a small area) and the large part contacts CLSM (forming a
cathode of a large area), the worst case of galvanic corrosion is set up. Mass loss
testing was performed in both uncoupled and coupled conditions to assess the
inherent corrodibility of metals placed solely in CLSM and to determine the
impact of galvanic cells on the corrosion process.
4.5.2 Mass Loss Test
4.5.2.1 Uncoupled Condition
To evaluate the actual corrosion performance of ductile iron pipe (and
other metallic materials) embedded in CLSM, samples can be embedded in
CLSM and evaluated for mass loss in laboratory conditions. Analyses of the
111
corrosion activity of coupons embedded in CLSM mixtures can be compared with
the corrosion activity of coupons embedded in a standard ASTM Ottawa sand.
For comparative testing, Ottawa sand was selected as a control material due to its
minimal variance in gradation and composition. However, any soil type could be
used in such a test. The experimental test set-up is shown in Figure 4.8. All of
the initial 38 CLSM mixtures were evaluated in this study. Specimens were cast
and moist cured for 28 days, then immersed in a 3.5% sodium chloride (NaCl)
solution to accelerate corrosion. Mass loss measurements were taken after 18
months of exposure. Preparing, cleaning, and weighing of specimens were
performed in accordance with ASTM G 1-90, “Standard Practice for Preparing,
Cleaning, and Evaluating Corrosion Test Specimens.” Mass loss provides a
measurement of average corrosion rate during the exposure period and thus can be
used as a parameter for evaluating the corrosion susceptibility of coupons
embedded in different materials and exposed to different environments.
75 x 150mm
cylinder
3.5% NaCl
solution
Pipe
coupon 100 mm
CLSM
or sand

coupons in CLSM and sand (uncoupled).
112
4.5.2.2 Coupled Condition (Galvanic)
Figure 4.9 shows the laboratory test set-up for evaluating samples that
may experience galvanic coupling. Samples were moist cured for 28 days before
being exposed to a 3.5% NaCl solution and being electrically connected by a 10-
Ω resistor. Two coupons with essentially equal areas were used in this set-up.
All of the 38 original CLSM mixtures were evaluated in this study. Preparing,
cleaning, and weighing of specimens was performed in accordance with ASTM G
1–90.
10 Ω
resistor
100 x 200 mm
cylinder
3.5% NaCl solution
Coupons CLSM Sand 100 mm

galvanically coupled coupons in CLSM and sand.
4.5.3 pH of Solutions Extruded from CLSM
CLSM sample solutions were extruded from 50 mm × 100 mm cylinder
specimens. The pH values were measured with a pocket pH-meter according to
113
ASTM G 51-95, “Standard Test Method for Measuring pH of Soil for Corrosion
Testing.” The pH values were recorded when the data were stabilized, usually
two or three minutes after inserting the meter into the solution.
4.5.4 The Resistivity of CLSM
This test followed the California DOT Test 532, “Method for Estimating
the Time to Corrosion of Reinforced Concrete Structures.” The set-up for
measuring the resistivity of CLSM is shown in Figure 4.10. The set-up is
composed of plastic plates for the frame and steel plates as electrodes for the
current. The box size measures 100 mm ×150 mm × 50 mm. The resistance
values of 38 CLSM mixtures were measured at an age of 182 days.
Steel Screws
Plates
Figure 4.10: Set-up for resistance measurement
114
4.5.5 Chloride Diffusion
For underground iron pipes, the rate of corrosion can be accelerated by the
presence of aggressive ions. In predicting the susceptibility of pipes placed in
engineered backfill materials, it may be beneficial to establish the quantity of
aggressive ions at the pipe location. In concrete applications, chloride contents as
a function of depth can be determined using specific ion probes to determine how
chlorides migrate through CLSM.
Five CLSM mixtures (#4, #6, #22r, #24, and #26) were evaluated for
chloride ingress rate, as shown in Figure 4.11. From each mixture, two CLSM
cylinders (150 mm x 300 mm) were ponded for 22 weeks with a 3.5% NaCl
solution. Samples were then tested for chloride contents as a function of cylinder
depth.
Chloride
Solution
150 x 300 mm
CLSM cylinder
Figure 4.11: Set-up for chloride ion diffusion
115
4.6 LEACHING AND ENVIRONMENTAL IMPACT
Recent concerns raised by the Environmental Protection Agency (EPA)
regarding the use of coal combustion products in construction have rekindled the
interest in studying the environmental implications of using CLSM containing
coal combustion products and other waste materials. Although the EPA did not
ban or curtail the use of coal combustion products, as originally expected, the
topic will likely resurface in the future. Because the use of by-product and waste
materials in CLSM is so common, some research was performed to assess the
potential toxicity of materials used in this study and to develop a protocol for
assessing materials being considered for use in construction.
The laboratory testing included a full chemical analysis of the by-product
and recycled materials included in this project, including fly ash (class F, Class C,
and high-carbon), bottom ash, and foundry sand. The chemical analysis was
conducted using inductively coupled plasma emission spectroscopy (ICP), gas
chromatography-mass spectrometry (GC-MS), total organic carbon (TOC), and
atomic absorption. Information such as the calcium oxide (CaO) content of fly
ashes can be very helpful in assessing the potential reactivity and effects on long-
term strength gain.
The assessment of the toxicity of the by-product materials used in this
study focused on the identification and quantification of heavy metals. Generally,
heavy metals are more of a concern than organics in fly ash, so the limited
laboratory study only tested for heavy metals. Various techniques are available
116
for the measurement of organics in constituent materials and CLSM mixtures, but
they were not included in this study.
For each of the by-products included in this study (three fly ashes, one
bottom ash, and one foundry sand), the total heavy metal concentration was first
determined in accordance with EPA Method 610, where nitric acid and hydrogen
peroxide were used to digest the materials. The eight elements analyzed included
arsenic, barium, cadmium, chromium, lead, mercury, selenium, and silver.
Because this testing determines the total amount of heavy metals, and not the
leachable amount, the extraction values are often assumed to be 20 times that of
the TCLP limits. If any of the by-product materials yielded values in excess of
the toxicity limits (20 times the TCLP limit), TCLP was then conducted to assess
the type and amount of heavy metals that are actually leachable from the
materials. Method 40CFR 261.24 was used to extract the samples for TCLP
testing, and the same eight heavy metals as previously listed were measured. The
concentrations of the extracts were compared to the TCLP limits which follow:
Element TCLP Limits (ppm)

Arsenic 5.0
Barium 100.0
Cadmium 1.0
Chromium 5.0
Lead 5.0
Mercury 0.2
Selenium 1.0
Silver 5.0
Testing the constituent materials directly, rather than testing the materials
encapsulated in CLSM, represents the worst-case scenario for the purpose of
117
toxicity screening. If the constituent materials test below the TCLP toxicity limits
under this worst-case scenario, the materials can be classified as non-toxic, and no
additional tests need to be performed. None of the materials tested in this study
exceeded the TCLP limits, and thus tests on actual CLSM specimens were not
deemed necessary. However, if any of the constituent materials had exceeded the
TCLP limits when tested by themselves, CLSM mixtures would have been cast
for subsequent leaching tests. The overall approach to the toxicity and leaching
tests performed in this study is shown in Figure 4.12.
118
STEP 1:
Chemical analysis of all raw materials
STEP 2:
Heavy metal test of all raw materials
using EPA 610
Yes
Less than
TCLP limits ?
No
STEP 3:
TCLP test (Method 40CFR 261.24)
on potentially toxic raw materials
from Step 2
Yes Raw materials deemed non-toxic.

Less than Accepted to be used in CLSM
TCLP limits ? mixtures.
No
STEP 4:
TCLP test (Method 40 CFR 261.24)
on CLSM mixture containing
potentially toxic raw materials from
Step 3
Yes
Less than
TCLP limits ?
No
Raw materials deemed toxic.

Rejected to be used in CLSM
mixtures.
Figure 4.12: Flow chart to study toxicity (heavy metals) of CLSM

constituent materials
119
Chapter 5: Results and Discussions
5.1 INTRODUCTION
In this section, results from the testing program described in Chapter 4 are
provided. Specifically, the results of tests on the fresh and hardened properties,
durability characteristics, and environmental leaching aspects of CLSM mixtures
are presented. Detailed discussions on the test results are included, along with
some discussions on related mechanisms. Focus is placed on the applicability of
test methods for field applications and on better understanding factors affecting the
short- and long-term behavior of CLSM, with specific focus on factors influencing
strength.
5.2 CLSM MIXTURE PROPORTIONS AND FRESH PROPERTIES
5.2.1 Water Demand/Flow Behavior
An important aspect of this study was the assessment of the fresh or plastic
properties of CLSM, both in terms of evaluating candidate test methods and
determining the relationship between constituent materials, mixture proportions,
and fresh CLSM properties. Table 5.1 summarizes some of the important
parameters, including water demand (to obtain the target flow), air content, flow,
unit weight, and bleeding (%) for the initial 38 mixtures. Note that mixtures ending
in “r”, such as mixture 2r, denote mixtures that were repeated for statistical
120
Table 5.1: CLSM mixture proportions and fresh properties
Mixture Cement Fly ash type Fly ash Fine aggregate Water flow Total Air Fresh unit
content content type demand bleeding content weight
(kg/m3) (kg/m3) (kg/m3) (cm) (%) (%) (kg/m3)
1 30 Class C 180 Concrete sand 211 20 NA 0.9 1965
2 60 Class C 180 Concrete sand 206 20 2.45 0.95 2108
1r 30 Class C 180 Concrete sand 206 21 2.08 0.9 1974
15 30 Class C 360 Foundry sand 486 20 0.13 2.75 1741
3 60 Class C 360 Bottom ash 577 18 4.32 1.65 1754
8 60 High carbon 180 Foundry sand 532 24 1.04 3.3 1647
10 30 High carbon 180 Bottom ash 628 14 4.81 2.0 1681
9 60 Class F 360 Foundry sand 520 20 0.54 2.5 1684
5 60 Class F 180 Bottom ash 600 18 5.84 2.5 1739
12 30 Class C 360 Bottom ash 572 22 3.64 2.7 1774
4 30 Class F 360 Concrete sand 220 20 0.39 2.2 2199
7 30 Class F 180 Foundry sand 501 20 0.57 2.1 1817
3r 60 Class C 360 Bottom ash 541 20 2.58 2.1 1997
4r 30 Class F 360 Concrete sand 220 22 2.92 1.8 2211
24 60 Class F 1200 None 486 24 2.25 2.8 1635
23 60 None 0 Bottom ash 454 14 1.30 28.5 1382
18 60 None 0 Concrete sand 200 22 0.70 16.5 1836
2r 60 Class C 180 Concrete sand 206 25 0.21 0.5 2291
29 60 Foundry sand 0 None 373 23 0.28 2.6 1812
30 30 Foundry sand 0 None 414 20 0.40 2.0 1789
17 30 None 0 Bottom ash 582 13 4.35 20.0 1447
11 60 High carbon 360 Bottom ash 573 23 6.42 1.7 1743
6 30 High carbon 360 Concrete sand 315 20 2.26 1.3 2103
22r 60 None 0 Concrete sand 136 18 0.43 25.5 1802
5r 60 Class F 180 Bottom ash 600 16 7.20 1.4 1887
26 60 None 0 Concrete sand 136 17 0 25.5 1802
16r 30 None 0 Concrete sand 295 19 2.35 15.5 1874
13 60 Class C 360 Foundry sand 499 20 0 1.8 1902
25 60 High carbon 1200 None 853 24 7.38 1.3 1322
19 30 None 0 Bottom ash 492 13 1.08 25.0 1385
20 60 None 0 Bottom ash 525 13 3.41 18.5 1485
27 60 Class F 1200 None 486 23 1.28 0.7 1638
20r 60 None 0 Bottom ash 525 13 1.44 15.5 1511
121
purposes. The results of the mixtures are presented in the order in which they were
mixed and are not sorted by mixture type. The fresh properties of the A- to I-series
mixtures can be found in Table 3.9 through 3.17.
The water demand to obtain the target flow was found to be one of the most
important parameters affecting CLSM. Certain constituent materials, such as high
carbon fly ash, bottom ash, and foundry sand, were found to particularly increase
the water demand of CLSM. By increasing the water content of CLSM, most of
the important properties, such as unconfined compressive strength, were impacted.
ASTM D 6103 was found to be a suitable method for measuring the flow of
CLSM. For mixtures with low density (high air content), it was found that the 200
mm flow was sometimes difficult to obtain, and for these mixtures, a flow of 175
mm was found to be satisfactory. The bottom ash used in this study lacked
sufficient fines for good workability and was prone to bleeding. However, when
used with high contents of fly ash, the mixtures were more workable and exhibited
reduced bleeding.
The effects of material source and quantity on the water demand of CLSM
were analyzed using a statistical program (ECHIP) for the non-air-entrained
mixtures. Figure 5.1 shows the results. This graph illustrates the statistically
significant variables that affect water demand. In this graph, the effect is the
difference between the specified variable level and the mean water demand of
446.3 kg/m3 . The reference variable levels are concrete (river) sand, Class C fly
ash, 180 kg/m3 fly ash content, and 30 kg/m3 cement. The effect of reference level
is the opposite of the sum of remaining levels. The plus sign (+) in the key after the
122
variable description indicates that the difference was positive and dash (-) indicates
that the difference was negative. The negative sign means that the variable level
will reduce that amount of water from the mean. The effect of a reference level is
the opposite of the combination effects of all other variable levels. From the figure
it can be seen that the fine aggregate source was the most significant factor
affecting the water demand of mixtures. The use high-carbon fly ash also increased
the water demand. There is no significant difference between the use of the Class
C and Class F fly ash. The effect of high content of fly ash (360 kg/m3 ) was
negligible.
10
Bottom ash (+)

Foundry sand (+)
High carbon (+)
60kg cement*high carbon (-)
60kg cement*class F (-)
60kg cement (-)
Class F (-)
60kg cement*360kg fly ash (-)
360kg fly ash
0
0 20 40 60 80 100 120 140 160
Effect on water demand (kg/m3 )
Figure 5.1: Effects of variables on water demand of non-air-entrained

CLSM mixtures
123
Table 5.2 shows the results from the analysis of variance (ANOVA)
calculations on water demand. ANOVA is a statistical method for comparing two
or more populations or treatment. The Mean Squares column shows the summary
statistic for the source and the degree of freedom (DF) column shows the number
of model terms summarized by the mean square. The P column gives the
significance. Values less than 0.05 are usually considered statistically significant
and thus affecting the water demand. Thus, this table identifies the significant
variables as fly ash type, fine aggregate source, and the interactions between
cement content and fly ash type.
Table 5.2: Analysis of variance (ANOVA) for water demand
Source of Variation Mean Square Degree of freedom (DF) Significance, P

Cement content 894.919 1 0.0595
Fly ash type 3986.39 2 0.0003
Fly ash content 1.6386 1 0.9293
Sand type 214854 2 0
Cement content * fly ash type 2082.62 2 0.0035
Cement content * fly ash content 681.909 1 0.0932
Error 198.063 10
Replicate error 132.1 5
5.2.2 Air Contents of Fresh CLSM Mixtures
The air contents of CLSM mixtures were measured by the approach
presented in Chapter 4. This approach was found to be easy and fast. The results
are shown in Table 5.1. The range of entrapped air contents was similar to those of
conventional concrete. It was extremely difficult to entrain air in mixtures
containing foundry sand. This is most likely attributed to the presence of clay, coal
fines and other organic materials in the foundry sand. It is believed that the
124
foundry sand prevented the encapsulation or formation of entrapped tiny air voids,
thus making the AEA none-effective. Air-entraining agent dosages were increased
up to 1300% of the manufacturer’s recommended dosage, but it was still not
possible to entrain considerable air.
5.2.3 Unit Weight of Fresh CLSM Mixtures
The results of unit weights of fresh CLSM mixtures are shown in Table
5.1. The unit weights of CLSM mixtures in this study ranged from 1322 to 2291
kg/m3 . The unit weights of CLSM mixtures with entrapped air were only slightly
lower than ordinary Portland cement concrete because of the higher water contents
in CLSM mixtures. Mixtures without fine aggregates only exhibited lower
densities as a result of the higher water demand and the lower specific gravity of fly
ash.
5.2.4 Setting/Hardening
The setting/hardening behavior of CLSM is important for many
applications, especially those where early strengths are required to satisfy
construction demands (i.e., timing between lifts or early opening to traffic). Test
methods are needed to more easily assess the setting of CLSM, both in the
laboratory and in the field. This section discusses some of the preliminary findings
regarding the setting time of CLSM using the following methods:
• Needle penetrometer
125
• Soil pocket penetrometer
• Pocket vane shear tester
Because CLSM exhibits substantially lower strength than normal concrete
mixtures, it is not possible to assign initial and final CLSM setting times as is
typically associated with concrete. In this study, however, penetration resistance
(ASTM C 403) was correlated with soil penetrometer values for all mixtures, and
hardening time related to “walkability time” was investigated for select mixtures.
Appendix B shows the raw data from the soil penetrometer (SP) and needle
penetrometer (NP) tests for the Stage One 38 mixtures, as well as the eight
mixtures from the I-series. The walkability time was assessed by preparing large
“CLSM boxes” that were walked on at various ages. It was found that the soil
penetrometer values were in the range of 4.3 kPa to 7.4 kPa (average of 6.1 kPa)
when CLSM mixtures were able to support the weight of an average person with an
approximate 6 mm indentation.
When measuring the needle penetration of CLSM, it should be noted that a
certain minimum strength of CLSM is required to obtain meaningful test results.
Thus, it is often not feasible to compare the setting time of CLSM mixtures to each
other at predefined time increments, but rather, the timing of measurements was a
function of constituent materials and mixture proportions. Also, because a needle
penetrometer penetrates deeper into mixtures than a soil penetrometer, it is less
subject to bleed water effects. Despite these differences (25 mm versus 6.4 mm) in
penetration depth, there was a limited correlation between soil penetrometer and
126
needle penetrometer values for the 38 mixtures, as shown in Figure 5.2 with a R2
(the square of the correlation coefficient) of approximately 0.75 for all the CLSM
penetration data combined.
0.8
0.7
0.6 R2 =0.75
0.5
0.4
0.3
0.2
0.1
0
0 2 4 6 8 10 12 14
ASTM C 403 penetration (MPa)
Figure 5.2: Correlation between ASTM C 403 and soil penetrometer values
One potential problem with using the soil pocket penetrometer is the effects
of bleeding on the measurements. The accumulation of significant bleed water
(and fines) is common in CLSM construction and can make surface measurements
or assessments challenging. It is difficulty to locate the “true” top surface of fresh
CLSM and being able to quantify the penetration resistance of the top layer of
CLSM. To address this potential problem, a stand was designed, as shown in
Figure 5.3, to minimize the effects of bleeding on field measurements. When using
127
the pokcet penetrometer, the stand was first placed on the mixture surface. The top
of the penetrometer was then inserted into the hole on the top the stand, where its
dead weight pushed the tip of the penetrometer through the soft fines and bleed
water and came to a rest on the dense surface that marked the “true” top. The 6.4
mm penetration depth was then marked on the tip, based on the measurement taken
from the top surface of the stand. Finally, the needle was pushed down and the
reading was recorded. This modified stand was used for the E-series mixtures.
Figure 5.3: Stand for the pocket soil penetrometer
The soil pocket penetrometer gives an estimation of the unconfined
compressive strength of cohesive soil by penetration resistance, while the needle
penetrometer gives the bearing pressure. This test is based on Terzaghi’s bearing
capacity equation for circular footings, and the soil friction angle is assumed to be
zero. Thus, the bearing capacity/resistance is only attributed to internal cohesion.
The unconfined compressive strength is two times the cohesion, according to the
theoretical total stress failure envelope. For the pocket soil penetrometer, the
penetration is required to be 6.4 mm and the soil underneath is assumed to have
failed at this depth. According to Bhat, Bucchi suggested this method for the rapid
determination of soil strength components.10 Using Terzaghi’s equation, which
assumes that only cohesion contributes to the bearing strength, the following
formula can be derived:
128
1.3 × 5.7q u
qf = 1.3 Nc Cu = = 3.71 qu (5. 1)
2
Where qf = ultimate bearing capacity
Cu= undrained cohesion
qu = unconfined compressive strength
The pocket penetrometer is made to give readings directly in qu. By
multiplying the reading from the soil penetrometer by 3.71, the penetration
resistance (qf) can be back calculated. This is used in Figure 5.4 to compare the
penetration resistance measured with the pocket penetrometer and needle
penetration tests.
129
3.00
SP
BSP
2.50
Series3
back calculated soil penetrometer resistance
a
Soil penetrometer reading (SP) and
2.00
(BSP) (MPa)
1.50
1.00
0.50
0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00
Needle penetration (MPa)
Figure 5.4: Comparison between needle penetration (NP) and soil

penetrometer reading (SP) values and the back calculated
resistance from soil penetrometer (BSP).
From Figure 5.4, it was observed that at the early stage of setting, the soil
penetrometer may yield higher resistance values than those measured by needle
penetration. This may be because the soil penetrometer is more sensitive at low
values than the needle penetrometer that measures resistance up to 890 N. As
hydration proceeds and CLSM gains stiffness, the needle penetrometer clearly
exhibited higher values than those of the soil penetrometer. When using the needle
penetrometer for a penetration depth of 25.4 mm, it was observed that that the load
per unit area increased to a peak and then dropped off considerably. This
phenomenon was not observed for the soil penetrometer, which only requires a
130
penetration depth of 6.4 mm. Because CLSM mixtures are often plastic for several
hours after mixing, the soil penetrometer may not generate a bearing failure.
Figure 5.5 shows a needle (ASTM C 403) after the penetration testing.
Note the small wedge at the tip of the needle. This confirmed that a general failure
has been reached and is in accordance with Terzaghi’s assumption on ultimate
bearing capacity. Cracks caused by the needle penetration were observed on
specimen surfaces. Clearly, the penetration of 25.4 mm caused large plastic
deformations, thus affecting the bearing capacity. This also offers a possible
explanation as to why penetration values sometimes decreases, even though the
mixture is still gaining strength.
The pocket VS tester was used to test the shear strength of CLSM mixtures
at early ages. There was some limited correlation between the VS values and the
soil penetrometer values, as shown in Figure 5.6 (R2 =0.45). The VS values
represent an estimate of cohesion and unconfined compressive strength based on
some assumptions on failure plane, whereas soil penetrometer values are measured
indirectly.
131
Wedge
Figure 5.5: Setting time needle and the wedge at the tip of the needle
(following ASTM C 403).
132
0.50
0.40
R2 = 0.45
0.30
0.20
0.10
0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
Soil penetrometer (MPa)
Figure 5.6: Comparison between soil penetrometer and vane shear values
5.2.5 Bleeding of Fresh CLSM Mixtures
Bleeding was found to be common for CLSM mixtures. Even for CLSM
mixtures with high air content, a very small amount of bleeding water was
observed on the mixture surface. Flash fill (mixture C-2) showed little bleeding
because of its fast setting nature. The bottom ash used in this study lacked
sufficient fines for good workability and was prone to bleeding. However, when
used with high contents of fly ash, the mixtures were workable with reduced
bleeding.
Factors affecting the bleeding of CLSM mixtures with moderate air
contents were also analyzed using the statistical software program, ECHIP. The
results from the analysis are shown in Figure 5.7. The mean bleeding is 2.52%.
The foundry sand used in this study reduced the bleeding significantly, while the
133
bottom ash significantly increased the bleeding. The fly ash type had a minimal
effect on the bleeding of CLSM mixtures.
10
Bottom ash (+)
Foundry sand (-)

XXXXXX High carbon FA
XXXX 60 kg cement * high carbon FA
XXXX 360 kg fly ash
XXX Class F FA
XX 60 kg cement
XX 60 kg cement * Class F
XX 60 kg cement * 360 kg FA
0
0 1 2 3 4
Bleeding (% of weight)
Figure 5.7: Effects of variables on bleeding of non-air-entrained CLSM

mixtures. “+” means increasing the bleeding, “-“ means
decreasing, and “X” means the effect is not significant compared
with variations.
5.2.6 Segregation of Fresh CLSM Mixtures
Using the method for assessing segregation described earlier in Chapter 4,
the results from the test shown in Table 5.3 were obtained. These data represent
the gradation of aggregates sieved from different depths within a CLSM cylinder.
A “pseudo-fineness modulus” was calculated using the same approach used to
calculate the FM for sand used in concrete. Based on the data shown in Table 5.3,
134
very little segregation occurred in any of the mixtures. However, the test method
developed to assess segregation showed promise as a useful tool to quantify
segregation. This method will be applied in the proposed field-testing program,
with the intention of correlating the laboratory results with actual field
performance.
Table 5.3: Segregation data for selected CLSM mixtures

Cumulative Cumulative Cumulative Cumulative Cumulative Cumulative
retained retained retained retained retained retained
(%) (%) (%) (%) (%) (%)
Sieve # Mix # 4 Mix # 23
4 4.55 3.30 6.83 10.82 21.10 16.63
8 23.70 24.01 27.60 33.41 41.05 35.13
16 46.97 45.68 50.08 52.88 54.88 50.98
30 70.29 67.40 71.41 66.83 68.26 63.89
50 86.90 82.68 87.69 79.49 82.27 78.57
100 95.94 92.11 96.09 92.31 92.55 91.00
200 100 100 100 100 100 100
Pseudo-
Fineness 3.28 3.15 3.40 3.36 3.6 3.36
modulus
Sieve # Mix # 6 Mix # 22r
4 5.32 3.43 3.70 4.44 3.47 3.53
8 14.72 23.27 24.26 24.38 25.94 24.75
16 38.70 45.23 43.94 45.56 45.93 44.76
30 60.30 63.81 65.90 65.34 68.39 66.08
50 83.78 84.69 85.22 88.03 88.59 89.72
100 95.30 95.97 95.31 98.09 98.04 98.29
200 100.00 100.00 100.00 100 100 100
Pseudo-
Fineness 2.98 3.16 3.18 3.26 3.30 3.27
modulus
Sieve # Mix # 26
4 3.75 5.44 3.74
8 27.06 29.06 25.87
16 47.03 47.67 46.72
30 67.41 72.09 69.52
50 88.79 90.12 89.65
100 98.23 98.07 98.26
200 100 100 100
Pseudo-
Fineness 3.32 3.42 3.34
modulus
*Note: Because mixture 24 did not contain aggregates, segregation was not measured
135
5.2.7 Subsidence
The results of subsidence testing for six CLSM mixtures are shown in Table
5.4. All of the CLSM mixtures exhibited appreciable subsidence, with the
exception of mixture #23, which had relatively poor flowability. With the
exception of mixture #23, there was a limited correlation between subsidence and
bleeding, as shown in Figure 5.8. The use of an accelerating admixture reduced
bleeding and slightly reduced subsidence, as shown by comparing mixtures #26
and #22r (identical mixtures but mixture #26 contained an accelerator).
Table 5.4: Subsidence results of six selected CLSM mixtures
Mix No. 4 Mix No. 24

Time (hrs) Subsidence (mm) Time (hrs) Subsidence (mm)
1.25 2.95 3 5.3
2.3 3.43 5.5 6.8
3.57 3.41 6 6.8
4.57 3.5
5.17 3.41
Mix No. 23 Mix No. 6
3 0.3 1 9.4
4 0.3 2.17 10.5
3.33 12.65
4.33 14.85
6.5 15.85
Mix No. 22r Mix No. 26
3.17 1.22 3.67 1.5
5.17 2.15 5.67 1.5
6.33 2.15
Note: The total specimen height was 600 mm
136
3
2.5
1.5
0.5
0
0 0.5 1 1.5 2 2.5
Bleeding ( weight %)
Figure 5.8: Correlation between bleeding (weight percent) and subsidence
5.3 HARDENED PROPERTIES OF CLSM
5.3.1 Unconfined Compressive Strength
5.3.1.1 Compressive Strength Development
A great deal of emphasis was placed in Stage One on assessing the
unconfined compressive strength of CLSM. This section first summarizes the
findings from the initial mixtures (38 in all), in which several aspects of
compression testing were examined, including the effects of cylinder size, capping
material, and load rate. Based on these original mixtures, some useful predictive
models were developed to predict the strength gain (short- and long-term) of
CLSM. The original study led to several follow-up studies, each of which focused
137
in more detail on issues involving testing parameters. Detailed investigations on
load rate, curing and conditioning of cylinders, effects of drainage on strength, and
the use of alternative capping materials are included in following sections. The
findings of the initial, broad study and the later, detailed studies could be used to
refine and improve existing methods of measuring the unconfined compressive
strength of CLSM.
The unconfined compressive strengths of the originally proposed mixtures
at 3, 7, 28, and 91 days are shown in Table 5.5. The results were found to be
repeatable, with quite low values of coefficient of variation. The data shown in this
table were for 75 mm × 150 mm small, sulfur-capped cylinders tested on a smaller
capacity machine, as previously described.
An interesting observation that illustrates the uniqueness of CLSM is that
most mixtures show a drastic change in the load-deflection curve as the curing time
is increased. Figure 5.9 illustrates this behavior for Mixture #10, which was typical
of most CLSM mixtures. At early ages, CLSM acts more like a soil, with more
ductile behavior. But as time progresses, CLSM begins to act more like concrete,
with higher strength and lower ductility.
As stated earlier, various modifications to the unconfined compression test
were studied in the initial investigation, some of which were later addressed in
more detailed research. Table 5.6 shows a comparison between sulfur-capped and
neoprene-capped cylinders for the selected six CLSM mixtures. In general, sulfur-
capped cylinders yielded higher strengths than cylinders using neoprene pads. The
138
Table 5.5: Unconfined compressive strength of original 38 CLSM mixtures
3-day f’c C.O.V. 7-day f’c C.O.V. 28-day C.O.V. 91-day C.O.V.
Mixture
(MPa) (%) (MPa) (%) f’c (MPa) (%) f’c (MPa) (%)
1 0.12 8.17 0.21 6.81 1.09 4.91 1.87 2.8
2 0.29 2.42 1.76 10.11 3.69 4.03 6.26 13.5
1r 0.13 14.36 0.24 1.18 1.35 7.69 2.34 0.3
15 0.07 8.22 0.11 5.66 0.18 3.97 0.25 1.4
3 0.33 10.50 0.57 2.42 1.36 6.83 2.02 2.9
8 0.09 9.75 0.11 8.27 0.25 4.69 0.33 10.6
10 0.12 2.27 0.16 15.12 0.22 7.93 0.26 1.3
9 0.09 9.66 0.13 12.40 0.22 4.99 0.25 2.9
5 0.14 13.70 0.18 8.15 0.46 16.11 0.57 5.3
12 0.30 16.17 0.27 6.16 0.57 4.70 0.86 3.4
4 0.34 4.94 0.48 6.18 0.79 11.04 1.08 13.6
7 0.09 3.22 0.11 3.40 0.12 9.70 0.16 5.8
3r 0.46 13.08 0.58 4.36 1.49 5.78 1.97 8.3
4r 0.41 13.61 0.57 5.80 0.94 3.92 1.03 6.9
24 0.34 4.77 0.22 1.40 0.44 0.13 0.58 4.6
23* - - 0.04 6.42 0.14 9.52 0.18 7.6
18* - - 0.33 6.75 0.70 1.08 0.79 4.0
14 0.58 6.61 1.07 13.46 2.15 8.14 3.49 16.8
2r 0.42 9.30 1.58 2.86 4.90 2.44 6.87 0.3
29 0.18 0.57 0.31 0.22 0.63 1.9 0.98 6.1
30 0.09 7.08 0.14 3.10 0.26 8.55 0.28 2.8
17* - - 0.01 31.79 0.07 18.90 0.13 16.8
11 0.33 1.66 0.42 4.29 0.75 3.48 0.94 4.7
6 0.40 10.20 0.47 0.74 0.83 4.88 1.09 4.7
16* - - 0.06 11.90 0.13 12.03 0.16 8.5
21* - - 0.09 10.59 0.16 11.84 0.18 11.7
22* - - 0.43 8.97 0.73 4.33 1.01 4.8
22r 0.32 4.19 0.50 9.68 0.96 17.53 0.93 7.0
5r 0.17 12.50 0.28 10.33 0.55 10.08 0.78 16.3
26 0.43 7.08 0.76 8.23 1.14 15.75 1.53 2.6
16r* - - 0.07 9.74 0.15 23.57 0.17 8.3
13 0.28 0.45 0.35 3.33 0.74 3.2 1.12 4.4
25 0.17 4.44 0.30 5.72 0.40 30.90 0.50 9.0
19* - - 0.02 0.71 0.06 45.02 0.06 13.2
20* - - 0.04 36.42 0.21 1.01 0.29 26.0
27 0.22 4.61 0.29 1.72 0.36 3.83 0.55 6.6
20r* - - 0.04 49.90 0.15 32.77 0.24 10.4
28 0.28 3.02 0.47 0.86 0.70 1.95 0.94 0.2
* Mixtures were too weak to be tested at three days.
139
3000
28 days
2500
2000
7 days
Load (N)
1500
1000
3 days
500
0
0 1 2 3 4 5
Deformation (mm)
Figure 5.9: Load-deformation response of Mixture #12 at 3, 7, and 28 days
Table 5.6: Comparison of capping materials and methods

Mixture 3 days 7 days 28 days
Sulfur Neoprene Sulfur Neoprene Sulfur Neoprene
(C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.) (C.O.V.)
MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%)
4 0.34 (4.9) 0.25(15.1) 0.48 (6.2) 0.34 (6.0) 0.79(11.0) 0.55(14.4)
6 0.45 (6.8) 0.25(14.7) 0.47 (0.7) 0.36 (6.3) 0.81 (4.9) 0.61 (3.2)
23 - - 0.04 (6.42) 0.03 (12.92) 0.14 (9.52) 0.12 (3.14)
24 0.34 (4.8) 0.15 (0.3) 0.22 (1.4) 0.19 (20.5) 0.44 (0.1) 0.30(18.8)
22r 0.32 (4.2) 0.22 (8.1) 0.50 (9.7) 0.37 (10.6) 0.93(17.5) 0.54 (4.4)
26 0.43 (7.1) 0.29 (9.1) 0.76 (8.2) 0.42 (11.2) 1.11(15.8) 0.73 (7.2)
neoprene pads used in this study had a durometer value of 50 (an index of
hardness), which is a typical durometer value for conventional concrete cylinder
testing. A more comprehensive study on capping materials, including neoprene
140
pads with varying durometer values, was subsequently performed, as described
later in this chapter.
Efforts were made in this study to develop predictive models for the
compressive strength of CLSM. Various models and statistical approaches were
considered. No single model was found to work well for the entire range of
materials and mixture proportions; however, predictive models for subsets of the
mixtures were found to be quite accurate. For instance, separate models were
developed for air-entrained (both for “high” and “moderate” air contents), and non-
air-entrained CLSM.
In developing the predictive model for air-entrained CLSM (up to 91 days),
the mixtures containing bottom ash as the fine aggregate, because of the
aforementioned problems with bleeding, were not included in the model. By trial-
and-error, the formula predicting the compressive strength of air-entrained CLSM
mixtures (not including bottom ash mixtures) is shown below:
f c' = a ⋅ e b ( w / c ) ( 5 .2 )
a = 0 . 3074 ⋅ ln( t ) + 0 . 22
b = 0 .0086 ⋅ ln( t ) − 0 . 272
Where, f c' is the compressive strength in MPa, w/c is the water-cement ratio, and t
is the age in days.
The measured and predicted compressive strengths for air-entrained
mixtures are plotted in Figure 5.10. There was very good correlation, with a R2
value of 0.97. This formula was also found to be effective in predicting long-term
141
strength gain (i.e., beyond 91 days). For example, cylinders from mixture 22r were
tested for compressive strength after 256 days, with an average strength of 1.0
MPa, compared to the predicted value of 1.1 MPa.
1.4
1.2
R2 =0.97
1
Predicted strength (MPa)
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Measured strength (MPa)
Figure 5. 10: Comparison of measured and predicted strengths of air-

entrained CLSM mixtures at the ages of 7, 28, and 91 days.
A predictive formula was also developed for non-air-entrained CLSM
mixtures. An initial approach, similar to that used by Bhat10 , was used in which the
water-cement ratio was the only variable used to predict the compressive strength.
This approach yielded an R2 value of 0.8. To improve this accuracy, a formula was
developed that included the water-cement ratio, aggregate type, fly ash type, and
the fly ash content as strength-predicting variables. A critical aspect to this
approach was to assign numerical values to the non-numerical variables used in the
142
formula. Through a trial and error process, the following constants (k) were
selected for the materials used in this investigation. Concrete or river sand (kriver
sand) was assigned a value of 1.0, foundry sand (kfoundry sand) a value of 0.2, bottom
ash (kbottom ash ) a value of 1.0, Class C fly ash (kC ash ) a value of 2.2, Class F fly ash
(kF ash ) a value of 1.0, and high-carbon fly ash (kHC ash ) a value of 0.75. The
equation for predicting the compressive strength, S(t), is shown below, and a
comparison between predicted and actual compressive strengths is shown in Figure
5.11.
S(t) = b 0 (t) ⋅ (k agg.type ) b 1 (t) ⋅ (k fly ash type ) b 2 (t) ⋅ (w/c) b 3 (t) ⋅ (k fly ash content ) b 4 (t) (5. 2)
where,
b0 (t)= -0.00071•t2 + 0.131•t -0.755
b1 (t)= -0.000118•t2 +0.0134•t+0.417
b2 (t)= -0.0000842• t2 +0.0154• t+0.0941
b3 (t)= -0.0034•t-1.03
b4 (t)= 0.745-0.0178•t when t 30 days
b4 (t)=0.217 when t > 30 days
Where, S(t) is the compressive strength in MPa and t is the age in days.
For field applications, the empirical equations presented in this report can
be used as a guide for developing mixture proportions to ensure future
excavatability of CLSM or to ensure that a minimum strength is obtained to meet a
design specification. It should be noted that the derived equations have limitations.
As is the case for experimental studies, the validity of the equations is only for the
143
specific materials used in this study. It is recommended that those interested in
predicting CLSM strength perform a series of mixtures using local materials to
generate a database that can serve as the basis for predictive models.
6 R2 =0.97
5
0
0 1 2 3 4 5 6 7
Predicted strength (MPa)
Figure 5.11: Comparison of measured and predicted strengths of

non-air-entrained CLSM mixtures
Although the predictive equations presented in this report were developed
for the specific materials used in this study, a reasonable correlation was obtained
between the equations and test data from a different series of mixtures using
completely different materials sources. Table 5.7 shows the mixture proportions
for this study, along with a comparison between the measured and predicted 7- and
144
28-day compressive strength values. Although there was a reasonable correlation
between predicted and actual strengths, this correlation could have been improved
by using the specific material types and behaviors in developing materials-specific
parameters and constants (k values).
Table 5.7: Comparison between predicted compressive strengths (using

equations shown in this section) and actual strengths from previous
study (using different constituent materials)
Mixture* Cement Fly ash Concrete Water Air Flow Fresh unit
(kg/m3 ) (kg/m3 ) sand (kg/m3 ) content (mm) weight
(kg/m3 ) (%) (kg/m3 )
I-1 30 180 1500 283 1.30 200 2036
I-2 60 180 1500 327 1.25 210 2077
I-3 120 180 1500 335 0.65 220 2087
Mixture* 7 days (MPa) 28 days (MPa)
Measured Predicted Measured Predicted
I-1 0.202 0.314 0.368 0.704
I-2 0.572 0.536 1.062 1.277
I-3 1.832 1.081 2.728 2.696
* Mixture codes are different from those used in present study. These mixtures from a previous study contained high carbon
fly ash in combination with three dosages of Type I portland cement
Some interesting information on how constituent materials affect CLSM
properties was obtained from this study. Most of the significant effects were
related to the influence of water demand on compressive strength. Materials that
tended to increase water demand (such as foundry sand and high-carbon fly ash)
generally yielded lower compressive strengths than materials with lower inherent
water demands. The chemical reactivity of fly ashes was found to be critical, as the
strength of CLSM containing Class C fly ash was higher than similar mixtures
containing Class F or high-carbon fly ash. Class C fly ashes, by definition, have
higher calcium (CaO) contents than Class F fly ashes (and the high-carbon fly ash
145
used in this study) and are more effective in increasing strength, even as early as
seven days.
5.3.1.2 Effects of Cylinder Curing on Compressive Strength
This section describes the findings of investigations on the effect of cylinder
storage conditions on the unconfined compressive strength of CLSM. An initial,
smaller investigation on the effects of cylinder curing (i.e., in mold, out of mold, in
lime water) is described first (using the B-series mixtures), followed by a more
comprehensive study on cylinder curing (using the G-series mixtures). The overall
objective of these studies was to determine the most efficient and accurate
method(s) of curing CLSM test cylinders.
Table 5.8 shows the effects of curing method on mixture B-1. The findings
show that the compressive strength of CLSM can vary widely, depending on
whether the cylinders are cured in or out of their molds, or whether the cylinders
are cured out of their molds in lime-saturated water. This initial study, using only
one mixture, illustrated the importance of curing parameters, especially when
considering long-term strength values (i.e., at 91 days). The highest 91-day
strength was achieved when the cylinders were left continuously in their molds,
suggesting that cylinders stored out of their molds or in lime water may be subject
to substantial leaching. The concern over cylinder storage conditions, in particular
the potential effects of specimen leaching and loss of mass, led to further
investigations using a range of materials and mixture proportions, as described later
in this section.
146
Table 5.8: Effects of curing methods on compressive strength (Mixture B-1)
Curing 7-day strength C.O.V. 28-day strength C.O.V. 91-day strength C.O.V.
method* (kPa) (%) (kPa) (%) (kPa) (%)
Cylinders kept
in molds, 313.6 2.9 1360.9 2.0 3811.5 9.4
stored in fog
room
Cylinders were
stripped at 3 362.2 1.2 1394.7 3.8 2396.5 4.4
days and cured
in fog room
Cylinders were
stripped at 7 352.6 4.6 1157.2 4.8 2670.8 3.8
days and cured
in lime-
saturated water
* Refers to curing regimes previously described in Chapter 4. All curing was performed at 23 °C.
To further study the effects of curing regime on compressive strength, a
more comprehensive investigation was performed using ten CLSM mixtures (G-
series mixtures). A range of constituent materials and mixture proportions was
used to better understand the potential effects on strength values for different curing
methods. The curing conditions investigated included the following:
• Curing Condition A (“normal”) – Cylinder kept in mold (with cap on) for
seven days in fog room. Cylinder then stripped and kept in fog room until time
of testing. This is referred to as “normal” because most cylinders tested in this
project followed this procedure.
• Curing Condition B (“mold”) – Cylinder kept in mold (with cap on) for seven
days in fog room. Cap is then removed and cylinder is kept in fog room until
time of testing.
• Curing Condition C (“cap”) – Cylinder kept in mold (with cap on) until time of
testing.
• Curing Condition D (“out”) – Cylinder kept in mold (with cap off) outdoors
until time of testing.
147
After subjecting cylinders from each of the G-series mixtures to the curing
regimes just described, all cylinders were capped with sulfur capping compound
and tested at a loading rate of 0.38 mm/min. The compressive strengths measured
at 28 and 91 days are shown in Tables 5.9 and 5.10, respectively. The remainder of
this section discusses these results and also describes some of the nuances observed
when testing different materials and mixture proportions. Some interesting
observations were made that illustrate that the strength of CLSM is significantly
affected by variations in curing regime, and further, that these variations are a
function of specific materials and mixture proportions. Some of the more
interesting results are illustrated in Figures 5.12 to 5.15, as described later in this
section.
Table 5.9: Compressive strength at 28 days using different curing methods
Mixture Curing condition A Curing condition B Curing condition C Curing condition D

(normal) (mold) (cap) (outside)
Average C.O.V Average C.O.V Average C.O.V Average C.O.V
(kPa) (%) (kPa (%) (kPa) (%) (kPa) (%)
G-1 559.1 3.2 365.1 2.3 344.3 6.0 536.0 5.9
G-2 246.8 16.5 267.9 4.7 269.2 14.3 380.1 5.2
G-3 89.5 24.1 93.6 24.6 59.4 6.5 too weak -
G-4 247.8 11.6 167.0 16.1 164.9 4.8 - -
G-5 893.5 1.7 877.0 3.3 991.3 4.4 1259.0 4.8
G-6 369.5 7.8 326.1 10.2 306.6 7.3 295.5 8.2
G-7 150.8 11.5 145.4 4.8 161.5 3.7 280.0 8.3
G-8 170.3 9.2 137.3 6.4 160.9 16.1 600.1 8.3
G-9 317.9 12.9 328.6 13.3 279.0 18.7 496.0 14.5
G-10 486.3 3.6 412.2 9.6 411.6 6.1 986.9 4.4
In general, there were substantial differences between the strength of CLSM

cylinders stored outdoors (in hot Austin, TX weather) and cylinders cured in the
fog room. However, there was not a consistent trend for all the mixtures studied,
illustrating that temperature and curing conditions are sensitive to constituent
148
material type and proportions. Some mixtures lost strength when stored outdoors,
whereas others showed significant increases in strength when stored outdoors.
Table 5.10: Compressive strength at 91 days using different curing methods
Mixture Curing condition A Curing condition B Curing condition C Curing condition D

(normal) (mold) (cap) (outside)
Average C.O.V Average C.O.V Average C.O.V Average C.O.V
(%) (%) (%) (%)
G-1 754.6 7.7 500.7 5.0 479.4 6.0 808.8 2.3
G-2 346.1 7.8 308.6 11.1 266.5 15.3 422.7 4.5
G-3 101.9 7.3 97.4 2.0 96.3 15.5 49.6 6.1
G-4 305.8 11.4 272.9 7.9 239.0 8.1 408.5 7.1
G-5 1248.8 19.2 1342.3 14.2 1244.3 7.9 - -
G-6 378.0 13.9 367.7 3.7 366.6 7.1 378.0 13.9
G-7 175.8 10.7 179.3 4.1 199.6 7.6 271.9 8.3
G-8 218.0 2.0 210.5 8.8 201.5 12.6 943.0 2.7
G-9 408.8 6.3 347.3 14.0 353.4 24.1 501.1 13.7
G-10 785.9 17.8 688.0 6.2 617.7 5.4 1080.1 6.3
The largest increase in strength was observed for mixtures G-2, G-8, and G-
10, which contained Class C fly ash, where the high temperatures apparently
helped to activate the fly ash. Mixtures G-2, a typical “flash fill” mixture with 275
kg/m3 of Class C fly ash and no portland cement, showed a 40 percent increase in
strength when stored outside rather than in the fog room.
A similar trend is observed when CLSM containing Class C fly ash is
compared directly to Class F fly ash for mixtures containing foundry sand (Figure
5.12) and bottom ash (Figure 5.13). Mixtures containing Class F or Class C fly
ashes exhibited higher strengths for cylinders cured outdoors in a hot climate, but
the differences were more profound for Class C fly ash. In fact, the difference
149
between fog-room cured and outdoor-cured cylinders were as high as 250 percent
for the mixtures containing Class C fly ash. Thus, using laboratory-cured cylinders
to assess the field performance of CLSM containing fly ash (especially high-
calcium fly ash) in hot environments must be done with caution. The effects of
temperature on CLSM hydration are especially important when large amounts of
fly ash are used, and when the fly ash to cement ratio is high. Another impact of
higher temperature and outdoor curing (with the caps removed from the cylinders)
is the reduction in water content caused by evaporation, which lowers the effective
water-binder ratio.
Figure 5.12: Effects of curing conditions on compressive strength for Mixtures

G-7 and G-8
For high air content mixtures, G-3 and G-6, the effects of curing methods
varied with cement content (and strength levels), as shown in Figure 5.14. Mixture
G-3 contained 30 kg/m3 of cement and exhibited relatively low strengths. In fact,
150
mixture G-3 suffered such a reduction in strength when stored outdoors (compared
to fog-room curing) that cylinders could not be tested at an age of 28 days. At 91
days, cylinders could be tested, but the resultant strength was significantly lower
than fog-room cured cylinders. Mixture G-6, which contained 45 kg/m3 of cement,
exhibited less difference in strength (comparing outdoor-curing to fog-room curing)
than the lower cement content mixture (G-3). For mixture G-6, there was a
reduction in strength for outdoor-stored cylinders compared to fog-room cured
cylinders at 28 days, but this difference became negligible at 91 days. These
findings suggest that high-air content CLSM mixtures (without fly ash) benefit
more from moist-curing than they do from high-temperature exposures.
Figure 5.13: Effects of curing conditions on compressive strength for

Mixtures G-9 and G-10
151
and G-6
The influence of cement content on CLSM strength is also illustrated in
Figure 5.15 for mixtures containing low (15 kg/m3 ) and intermediate (30 kg/m3 )
cement contents and Class F fly ash. Mixture G-4, the low cement mixture, yielded
quite low strength values, and curing in the mold showed negative effects on the
strength when compared to stripped samples, especially after 28 days. Mixture G-
5, with the intermediate cement content, the strengths of outside cured specimens
showed 1.4 times those of specimens cured in the fog room. This further confirms
that both material type and mixture proportions affect the compressive strength of
CLSM, and that different combinations of these parameters will be affected by
methods of curing and testing. Thus, the development and recommendation of a
152
standard test method is essential in generating reproducible test results that can be
applied to specifications, design criteria, and construction guidelines.

and G-5
The findings obtained from the G-series mixtures regarding curing
temperature and cylinder storage led to a final investigation on the effects of curing
temperature and humidity on compressive strength, as discussed before. This
additional investigation was deemed important to better quantify the effects of
temperature on CLSM hydration and strength gain, especially when CLSM
incorporates fly ashes of varying reactivities.
153
5.3.1.3 Effects of Drainage Conditions on Compressive Strength
As described in the Chapter 4 of this dissertation, a comprehensive study
was performed to assess the effects of drainage conditions on CLSM strength. A
specially designed “curing box” was constructed to simulate the placement of
CLSM in native soil, and various drainage conditions were assessed. This section
presents the results of compression testing for cylinders stored in the following
manners:
• “Normal” – standard fog room curing (the same curing regime used throughout
most of this study)
• “No Cap” – cylinders were placed in the curing box with the cylinder caps
removed to simulate surface evaporation.
• “Cap” – cylinders were placed in the curing box with the cylinder caps left
firmly on top to simulate a situation without water loss.
• “Bottom holes” - cylinders were placed in the curing box with the cylinder caps
removed and with holes drilled in the bottom of the cylinders to simulate both
surface evaporation and bottom drainage.
• “Side holes” - cylinders were placed in curing box with the cylinder caps
removed and with holes drilled in the bottom and sides of the cylinders to
severe drainage (from top, bottom, and sides).
Additional information on this test set-up, including curing conditions, was

provided in Chapter 4. It should be noted that the curing boxes were kept in a hot
environment (in a non-air-conditioned room in the laboratory in mid-summer),
whereas the “normal” curing was performed at 23 °C in the fog room. Thus, the
effects of both temperature and drainage could be observed. The “cap” curing, as
previously described, was a useful reference in that these cylinders followed the
154
same time-temperature history as the other cylinders stored in the curing box, but
the cylinders were not subjected to drainage or evaporation.
The compressive strength test results at 7 and 28 days are shown in Table
5.11 and Table 5.12, respectively. Note that mixture F-2, which was a Class C fly
ash mixture without cement, was not tested at 28 days, due to the rapid setting time
of this mixture, which made it difficult to cast sufficient specimens.
In general, there were not substantial differences in compressive strengths
for the four curing methods that utilized the curing box, suggesting that the
influence of drainage and evaporation may not be critically important, at least
based on this laboratory simulation study. If differences were observed for a given
mixture, the difference was more apparent at 7 days than at 28 days. Once again,
the effects of temperature were evident as cylinders cured in the plywood box set-
up (ambient hot weather) tended to exhibit higher strengths than companion
cylinders cured in the fog room. This was particularly evident for mixtures
containing fly ash. The only fly ash mixture that did not show higher strengths in
the curing box, compared to the fog room, was mixture F-2, which was a so-called
“flash fill” mixture. This mixture set extremely fast, with little or no water
available for seepage or evaporation. Also, because the mixture set so fast and
hydrated so rapidly within the first few hours, the effects of long-term temperature
differences were probably not as important.
Another interesting observation, which confirmed some earlier findings,
was that the effects of curing temperature on strength may be quite complex
because it is not just a function of constituent material type but also of material
155
quantity. For instance, when low cement contents (15 kg/m3 ) were used in
combination with fly ash, such as in mixtures F-4, F-6, and F-8, there were only
small differences in strength as curing temperature was increased. However when
the cement content was increased to 30 kg/m3 (with the fly ash content remaining
constant), the temperature effects were significantly more pronounced, such as in
the case of mixtures F-5 and F-7. Thus, one must consider the combined effects of
material type and quantity when assessing the long-term strength gain of CLSM,
especially when using fly ash.
Table 5.11: Compressive strength at 7 days under different drainage

conditions
Mixture Normal No cap Cap Bottom holes Side holes
Average C.O.V. Average C.O.V. Average C.O.V. Average C.O.V. Average C.O.V.
(%) (%) (%) (%) (%)
F-1 308.8 1.5 496.1 6.2 377.3 22.4 469.1 10.1 559.9 1.9
F-2 105.3 5.8 101.8 - 110.0 0.4 110.1 3.3 116.7 -
F-3 84.1 5.0 86.6 22.4 81.7 19.8 97.6 2.6 99.4 15.2
F-4 127.8 7.1 107.2 16.2 105.8 16.7 93.2 11.6 116.8 16.0
F-5 303.8 1.9 435.7 4.0 463.6 5.1 485.6 7.7 523.2 1.7
F-6 132.8 4.1 170.3 4.0 141.7 4.3 196.4 3.9 192.8 1.1
F-7 252.4 10.4 360.7 6.1 388.5 1.9 411.7 7.6 407.8 8.0
F-8 162.5 8.0 187.3 4.0 201.6 6.7 192.2 11.6 212.6 7.5
156
Table 5.12: Compressive strength at 28 days under different drainage
conditions
Mixture Normal No cap Cap Bottom holes Side holes
Average C.O.V. Average C.O.V. Average C.O.V. Average C.O.V. Average C.O.V.
(%) (%) (%) (%) (%)
F-1 532.3 16.3 615.1 14.4 511.5 29.7 554.2 19.4 658.7 9.0
F-2* - - - - - - - - - -
F-3 144.1 14.2 138.4 14.2 132.0 17.7 153.6 25.6 157.3 30.0
F-4 291.9 5.7 309.0 7.5 253.1 16.3 250.7 6.8 283.6 10.2
F-5 842.1 9.0 1497.2 3.1 1329.3 3.6 1576.2 5.0 1586.0 3.6
F-6 255.1 2.7 334.3 18.6 297.4 14.1 394.8 8.6 412.3 9.2
F-7 779.5 14.1 1215.4 8.1 1412.6 2.2 1217.2 4.0 1278.3 1.7
F-8 493.1 6.9 499.8 5.7 462.5 2.4 528.7 3.8 560.3 4.0
* Cylinders from mixture F-2 were not tested at 28 days.
In summary, this investigation confirmed that temperature plays a key role
in many CLSM mixtures and suggests that drainage and evaporation may not be as
critical as temperature-induced effects. Additional work is needed to investigate
CLSM placed in actual applications to gain insight into real drainage conditions, as
opposed to the simulated conditions used in this study.
5.3.1.4 Specimen Preparation Before Testing
This section describes the findings on the effect of cylinder conditioning on
the unconfined compressive strength of CLSM. A preliminary study on cylinder
conditioning (i.e., drying specimens after curing but before testing) is described.
As previously mentioned, ASTM D 4832 specifies that cylinders be allowed
to air-dry for 4 to 8 hours before capping. It was observed that this long waiting
period may be inconvenient and impractical in a testing laboratory and may also
lead to variations in test results. To investigate this, cylinders from mixture B-2
157
were moist-cured for various ages, removed from the fog room, then subjected to
the following conditions:
• Cylinders air-dried for 0 to 0.5 hour before being sulfur-capped and

tested
• Cylinders air-dried for 2 hours before being sulfur-capped and tested
The results of testing after 7, 28, and 91 days of curing are shown are shown
in Table 5.13. Although there were some variations in strength results at each age
when the specimens were allowed to dry for different periods of time, the
variations were not appreciable in most cases. This suggests that it is not necessary
to air dry specimens for 4 to 8 hours before testing, as is specified in the ASTM
standard. This important finding might be incorporated into a proposed
compression test method, in which cylinders are specified to be tested without
undergoing a mandatory drying period. This recommendation is consistent with
the testing of conventional concrete. Also, by eliminating the drying period, it is
expected that more accurate results may be obtained, and laboratory productivity
will increase (i.e., more cylinders can be tested in a single day).
Table 5.13: Influence of specimen air-drying on compressive strength

(Mixture B-2)
Test age 0-0.5 hour air dry 2 hours air dry 4 hours air dry 8 hours air dry
Strength C.O. V. Strength C.O. V. Strength C.O. V. Strength C.O. V.
(kPa) (%) (kPa) (%) (kPa) (%) (kPa) (%)
7 days 310.1 5.8 329.2 11.3 357.5 2.7 363.2 2.0
28 days 1575.1 1.9 1536.5 7.8 1447.2 5.7 1649.8 3.3
91 days 3289.4 2.0 3226.9 4.6 3430.4 3.9 3536.1 4.6
158
5.3.1.5 Alternative Capping Materials for Compression Testing
A comprehensive study was performed on the use of alternative capping
materials for compression testing, as already described in Chapter 4. The materials
included in this investigation were neoprene pads with Shore “A” durometer values
of 20, 40, 50, 60, and 70, as well as sulfur-capping compound and gypsum (or
“hydrostone”). After curing the CLSM cylinders for various ages (7, 28, and 91
days), the cylinders were capped and tested using a similar approach which
followed throughout most of this project. A load rate of 0.38 mm/min was used for
all cylinders, and load and deformation values were recorded by a data acquisition
system.
Initial trials using gypsum led to the development of a more user-friendly
method of using this capping material. When using the gypsum, a water-gypsum
ratio of 0.3 was used, and a split 75-mm x 150-mm plastic mold was used as a
“dam” to retain the fluid gypsum mixture on top of the cylinder end. Sealing
grease was used to prevent the gypsum mixture from leaking through the gap
between the cylinder surface and plastic mold. Because of the self-leveling
property of gypsum mixtures, cylinders with damaged ends could be easily capped.
The long setting time of gypsum mixtures, up to 40 minutes, made the capping
process time consuming, and thus, gypsum was only used for testing at 28 days.
Table 5.14 shows the compressive strength values measured for the
different mixtures and capping materials at 7, 28, and 91 days. Table 5.15 shows
the corresponding coefficients of variation of measured compressive strength using
the various capping methods. The sulfur capping method provided the highest
159
strength values for all the eight mixtures. Also, in general, sulfur capping
generated the lowest variations compared to the other capping methods. Note that
Mixture E-2, a “flash fill” mixture, set so quickly that it was not possible to
produce enough cylinders for testing at all ages. Thus, testing was only performed
at an age of 7 days.
Table 5.14: Compressive strength results using different capping materials

Mixture E-1 E-2 E-3 E-4 E-5 E-6 E-7 E-8
Sulfur 7 day 0.31 0.11 0.08 0.13 0.20 0.34 0.43 0.19
28 day 0.53 0.14 0.29 0.33 1.03 1.19 0.66
91 day 0.85 0.14 0.45 0.48 1.58 1.71 1.24
Gypsum 28 day 0.52 0.11 0.24 0.36 0.95 1.11 0.59
D70 7 day 0.28 0.09 0.06 0.12 0.20 0.28 0.36 0.17
28 day 0.42 0.11 0.24 0.34 0.92 1.05 0.42
91 day 0.10 0.34 0.46 1.12 1.39 0.91
D60 7 day 0.29 0.09 0.05 0.11 0.16 0.33 0.36 0.18
28 day 0.57 0.12 0.26 0.27 0.66 1.02 0.42
91 day 0.10 0.37 0.31 1.29 1.50 0.89
D50 7 day 0.26 0.08 0.05 0.13 0.21 0.34 0.40 0.19
28 day 0.51 0.10 0.25 0.28 0.87 0.96 0.47
91 day 0.10 0.35 0.32 1.23 1.44 1.08
D40 7 day 0.27 0.07 0.04 0.13 0.22 0.33 0.34 0.21
28 day 0.51 0.10 0.22 0.25 0.91 1.01 0.50
91 day 0.79 0.14 0.33 0.45 1.36 1.37 0.80
D20 7 day 0.25 0.09 0.04 0.11 0.20 0.32 0.40 0.17
28 day 0.71 0.09 0.24 0.31 0.90 0.96 0.57
91 day 0.13 0.40 0.36 1.30 1.43 1.05
160
Table 5.15: Coefficients of variation for compressive strengths using different
capping materials
Mixture E-1 E-2 E-3 E-4 E-5 E-6 E-7 E-8
Sulfur 7 day 1.6 5.8 5.0 7.1 15.7 17.6 8.4 6.9
28 day 16.3 14.2 5.7 20.7 13.3 4.0 10.3
91 day 3.6 15.3 10.6 10.5 6.9 5.3 8.1
Gypsum 28 day 18.0 7.7 6.1 5.5 5.3 7.8 14.9
D70 7 day 5.3 12.4 20.3 5.9 3.7 12.0 9.7 9.5
28 day 45.9 19.3 22.0 44.5 16.1 14.9 20.8
91 day 22.8 6.8 24.9 8.3 3.0 3.7
D60 7 day 18.0 6.5 15.5 7.2 17.3 12.0 13.2 22.1
28 day 12.9 2.9 2.5 4.0 23.9 7.6 22.0
91 day 30.4 12.9 16.4 12.5 6.6 21.4
D50 7 day 17.5 5.7 28.7 15.9 7.4 0.7 4.6 30.0
28 day 18.1 15.5 5.0 20.9 15.1 6.4 6.5
91 day 17.0 22.2 12.6 4.2 4.0 11.5
D40 7 day 17.9 27.6 23.3 21.4 10.1 13.0 6.8 15.2
28 day 15.4 14.0 9.5 20.5 19.5 10.8 11.2
91 day 12.7 2.7 4.6 27.8 6.6 14.6 9.4
D20 7 day 16.5 1.9 20.5 15.8 10.3 7.3 9.6 16.6
28 day 7.1 1.1 14.5 10.1 19.2 6.8 5.9
91 day 6.1 7.4 22.2 9.1 6.0 23.5
Figures 5.16 through 5.18 compare the coefficients of variation of strength

values obtained by using different capping methods at 7, 28, and 91 days,
respectively. Neoprene pads resulted in as high as 45% variation in the measured
compressive strength values, with the hardest pads resulting in the largest
variations.
161
compressive strength)
162

163
The process for the qualification of unbonded capping systems described in
ASTM C 1231 was followed in analyzing the test data in this study. Under close
examination, the relationship expressed as equation (3) in the test method was
found to be only approximate. Specifically, it involves a “student t” testing with
the null hypothesis that xp –0.98xs ≥ 0. Therefore, the calculation of standard
deviation of the difference should be based on xp and 0.98 xs. When testing
concrete according to ASTM C 1231, the strength from different capping methods
is assumed to be essentially equal, and the error using the proposed process can
typically be neglected. However, it has been shown that different capping materials
yield different strength values for CLSM and as such, the validity of the specified
ASTM C 1231 analysis process was examined in more detail. The theoretically
correct process should calculate the difference in strength for each pair of cylinders
using the following equation:
d i = x pi − ax si (5. 3)
Where:
di = Difference in strength of a cylinder with unbonded capping compared
to a percent of strength of another cylinder with sulfur capping,
a = Percent. Strength measured with unbonded capping should be larger
than “a” percent of strength measured with sulfur capping at a 95% confidence.
The initial estimate of a0 was selected to be the value obtained through the
ASTM C 1231 process satisfying equation (3) in ASTM standard. Adhering to the
164
process outlined in ASTM C 1231-99 (using the proposed equation 5.4 above, a
value of b0 could be obtained by satisfying:
x p = b0 a0 xs + (tsd ) /( n)1 / 2 (5. 4)
If b0 is equal to or very close to 1.0, the initial estimate could be deemed
correct. Otherwise, an iterative process should be followed, letting a1 =a0 × b0 , and
repeating until converging on an accurate value. Usually, only one or two cycles
were needed for convergence. Table 5.16 illustrates the qualification of unbonded
caps using the unmodified ASTM C 1231 process and the proposed, modified
method. For each durometer hardness and gypsum cement, more than 40 pairs of
specimens were included in the statistical analysis. The differences between the
two processes were found to be within six percent of each other, a relatively small
difference. However, it was believed that it was worthy of consideration, given the
differences in strengths observed using the different capping methods.
Table 5.16: Comparison of “a” values calculated by two methods

Capping system “a” value calculated by “a” value calculated by
ASTM C 1231 process modified ASTM C 1231
Gypsum 0.9203 0.9205
D 20 0.7990 0.8126
D 40 0.7316 0.7914
D 50 0.8032 0.8081
D 60 0.7629 0.7705
D 70 0.7512 0.7379
165
ASTM D 4832 states that the alternative capping systems are acceptable
when the average strength obtained is not less than 80% of the average strength of
companion cylinders capped with sulfur capping compound. According to this
criterion, the use of gypsum and neoprene pads with durometer values of 20 and 50
could be qualified when using the ASTM C 1231 process. By using the modified
method (as shown in Table 5.16), gypsum and neoprene pads with durometer
values of 20, 40, and 50 could be qualified.
As previously described, full load-deformation curves were recorded for
each of the compression tests. Rather than providing all of this data in this
dissertation, some representative results are presented next. Figure 5.19 shows a
typical load deformation curves for compressive cylinders of mixture E-1 at 28
days. Because the deformation-measuring device records the total movement
between loading platens, the compression of neoprene pads is also included the
deformation value. As a result, higher deformation values and longer loading times
were recorded in the testing. Thus, care should be taken in performing tests and
analyzing data when different capping materials are used because both the
measured deflection and testing time are directly influenced. Unlike concrete
cylinders, where deflections can be measured from surface mounted gages, it is
difficult to measure deflections or strains from the cylinder surface because of the
inherent low strength of CLSM. For CLSM, the most relevant data can be obtained
from using rigid caps (gypsum or sulfur) because the measured deflections (based
on relative platen movement) are less affected by deflections in the capping
material.
166
4
D60
3.5
3 D20
D20
2.5 D40
Load (kN)
D70 D50
2
Sulfur D40 D60
1.5
D70
1 Gypsum Gypsum
D50
0.5 Sulfur
0
0 100 200 300 400 500 600 700
Displacement (mm)
Figure 5.19: Load-deflection curves for mixture E-1 (compression

testing at 28 days)
In summary, the main findings from this capping material study include the
following:
1. Sulfur-capped compression cylinders generally produced the highest
strength values with the lowest variations.
2. Gypsum-capped specimens generally yielded higher strength values
than those using neoprene pads. However, gypsum takes a
relatively long time to set and may not be as user-friendly in a
testing laboratory environment.
3. There was not a significant difference between using different
durometer pads, especially when considering the variations in
167
results. However, when applying the acceptance criteria described
in ASTM C 1231, only neoprene pads with durometer values of 50
or less were found to be acceptable for use with CLSM.
5.3.1.6 Effects of Load Rate on Compressive Strength
Because of the general lack of guidance provided in ASTM D 4832
regarding the loading rate, additional emphasis was placed on assessing the effects
of loading rate on the compressive strength of CLSM in this study. This section
describes the results of two studies, the first of which focused on a relatively wide
range of loading rates (0.085 to 0.58 %/min), and the second of which covered a
narrower range of loading rates (0.16 to 0.42 %/min). A deflection-controlled
machine, often used to test soils, was used for this investigation. The results of
these two studies, coupled with the findings from the initial investigation, could be
used to recommend loading rates for a refined method to measure the unconfined
compressive strength of CLSM.
The loading rate is an important parameter when considering compression
testing. The testing of CLSM cylinders should be accurate. If strength values are
found to be influenced by loading rate, then a load rate range must be defined and
required for accurate testing. The loading rate also determines the length of time
needed to test a given cylinder. This required length of time must be sufficient to
ensure accuracy (i.e., not a sudden cylinder failure) and to complete a given test in
a reasonable amount of time. In a laboratory or testing facility, the time required to
test cylinders may be critical, especially if many tests are performed in a single day.
168
The compressive strength results for the two loading rate range studied,
referred to as “wide range” and “narrow range” studies, shown in Tables 5.17 and
5.18. The cylinder measures 75 mm x 150 mm were used in this study. The full
load-deflection curve was also obtained for each cylinder tested, and the
deformation at peak load for each test specimen was recorded, as shown in Tables
5.19 and 5.20. The deformation at peak load was used to determine if changes in
loading rate resulted in changes in modes of failure, such as a change from
relatively ductile to brittle failure. Discussion follows on how loading rates
affected both the compressive strength of CLSM and the deformation at peak load.
Table 5.17: Effects of loading rates (wide range) on compressive strength
Age Mixture 0.085 %/min 0.25 %/min 0.58 %/min

(days) Average C.O.V. Average C.O.V. Average C.O.V.
(kPa) (%) (kPa) (%) (kPa) (%)
A-1 291.5 4.0 344.9 5.5 373.6 7.0
7 A-2 126.0 6.0 108.6 11.4 120.1 2.7
A-3 222.5 5.2 245.7 7.6 269.2 5.7
A-4 128.6 6.1 132.4 6.1 132.8 8.9
A-1 867.8 5.7 914.5 11.7 962.5 4.4
28 A-2 227.1 4.4 238.9 2.3 222.9 5.2
A-3 675.1 2.9 708.1 0.8 682.4 2.0
A-4 587.4 4.4 634.1 3.8 630.3 4.2
A-1 1166.8 6.8 1196.2 2.7 1377.5 0.7
91 A-2 410.9 4.3 429.7 3.1 409.0 2.0
A-3 891.4 12.5 886.7 2.1 987.5 9.6
A-4 1439.7 8.6 1221.5 12.8 1239.5 5.3
169
Table 5.18: Effects of load rates (narrow range) on compressive strength
Age Mixture 0.16 %/min 0.25 %/min 0.33 %/min 0.42 %/min
(days) Average C.O.V. Average C.O.V. Average C.O.V. Average C.O.V.
(kPa) (%) (kPa) (%) (kPa) (%) (kPa) (%)
A-5 533.4 9.1 532.6 4.6 530.8 11.3 565.7 7.3
7 A-6 279.1 4.3 291.3 6.4 288.8 1.9 310.1 2.3
A-7 193.5 8.0 190.5 3.1 188.0 2.2 220.1 8.4
A-5 1050.1 7.1 1090.5 5.0 1048.3 4.1 1066.7 5.0
28 A-6 752.0 9.0 745.3 5.3 760.4 12.1 779.3 10.9
A-7 1039.1 5.4 1022.0 3.5 1085.3 3.6 1078.8 4.0
A-5 1339.5 3.4 1614.4 7.6 1529.3 8.4 1590.2 4.5
91 A-6 829.2 9.0 990.3 5.3 1079.7 12.1 1016.5 10.9
A-7 1871.9 5.4 1784.3 3.5 1863.7 3.6 1983.8 4.0
Table 5.19: Effects of loading rates (wide range) on deformation at peak load
0.085 %/min 0.25 %/min 0.58 %/min
Age Mixture Deformation C.O.V. Deformation C.O.V. Deformation C.O.V.
(days) (mm) (%) (mm) (%) (mm) (%)
A-1 0.711 23.9 0.673 2.7 0.803 20.2
7 A-2 4.252 16.3 3.459 15.9 4.252 2.3
A-3 0.960 23.5 0.892 28.6 0.765 13.4
A-4 1.521 13.4 1.547 19.8 1.402 5.1
A-1 0.572 13.5 0.566 11.8 0.566 6.4
28 A-2 1.466 13.3 1.801 17.2 1.537 11.8
A-3 0.665 2.3 0.861 12.3 0.848 5.1
A-4 0.648 1.4 0.683 20.9 0.691 14.4
A-1 0.716 0.7 0.597 1.7 0.638 1.9
91 A-2 1.920 0.4 2.042 2.7 1.880 3.4
A-3 0.909 1.8 0.925 1.3 0.851 2.6
A-4 0.777 1.7 0.843 4.7 1.031 3.9
170
Table 5.20: Effect of narrow range load rates (narrow range) on deformation
at peak load
0.16 %/min 0.25 %/min 0.33 %/min 0.42 %/min
Age Mixture Deformation C.O.V. Deformation C.O.V. Deformation C.O.V. Deformation C.O.V.(
(days) (mm) (%) (mm) (%) (mm) (%) (mm) %)
A-5 0.577 15.8 0.526 14.5 0.480 34.1 0.434 31.0
7 A-6 0.648 11.6 0.620 12.2 0.627 9.0 0.599 8.1
A-7 1.229 17.9 1.224 31.1 1.166 7.6 1.077 2.9
A-5 0.610 16.0 0.726 20.3 0.650 9.7 0.528 11.6
28 A-6 0.711 7.2 0.640 18.7 0.681 3.0 0.759 5.1
A-7 0.838 12.3 0.861 12.2 0.780 12.8 0.828 23.4
A-5 0.493 4.8 0.732 12.9 0.759 3.2 0.602 6.9
91 A-6 0.866 20.5 0.676 6.7 1.036 17.7 1.214 13.5
A-7 0.775 21.5 0.640 9.6 0.813 20.4 0.808 24.1
The effects of load rate on compressive strength were found to vary,
depending upon the type of mixture and the age of testing. Interestingly, some
mixtures (i.e., mixture A-2) were relatively insensitive to load rate, whereas others
were quite sensitive. There was no consistent trend among the mixtures, which
suggested the compressive strength was either directly or indirectly proportional to
load rate. However, there were some general trends that were insightful, as
described next.
For testing at 7 and 28 days, the range of loading rates from 0.16 to 0.42
%/min was generally found to generate consistent results. The effects of loading
rate on strength were more significant at 91 days than at 7 or 28 days. Mixture A-2
(containing foundry sand) was an interesting case where the compressive strength
varied very little, even across a wide range of loading rates. This is advantageous
from a testing perspective because the deformations at peak load of specimens
containing foundry sand were found to be large, and testing in some cases took
more than 10 minutes per sample when using a loading rate of 0.25 %/min. Higher
171
loading rates could thus be used for these types of mixtures to speed up the testing
without affecting test result accuracy. Because the variations in compressive
strength with loading rate were to some extent mixture or material specific, a load
rate between 0.and 0.42 %/min may be selected for improved compressive strength
testing.
The deformations at peak load provided a useful index of the relative
brittleness or ductility of CLSM. The effects of curing time on CLSM ductility
were profound, where at early ages (7 days), the deformations were somewhat
greater than those at later ages (28 and 91 days), as the specimens were more
ductile and extensible.
In general, when varying the loading rate, the deformations observed at
peak load did not vary as much as the actual compressive strength values. This
suggests that the mode of failure was not greatly affected by loading rate
modifications. Thus, the range of loading rates between 0.16 to 0.42 % per minute
can be recommended for the compression test method, based mainly on selecting a
range that yielded accurate and repeatable strength values in a reasonable amount
of time.
5.3.1.7 Effect of Specimen Size
Table 5.21 shows a comparison between different size cylinders, in which
no discernible size effects were observed. It should be noted that all cylinders were
tested at the same effective strain rate (0.25 %/min), using the same deflection-
172
controlled machine. There is no general trend found for the effects of specimen
size.
Table 5.21: Effects of cylinder size on unconfined compressive strength

Mix 7 days 28 days
75 x 150 mm 100 x 200 mm 75 x 150 mm 100 x 200 mm 150 x 200 mm
MPa (C.O.V.:%) MPa (C.O.V.:%) MPa (C.O.V.:%) MPa (C.O.V.:%) MPa (C.O.V.:%)
4 0.79 (11.0) 1.13 (7.01)
4r 0.58 (6.2) 0.49 (6.3) 0.94 (3.9) 0.84 (6.5)
6 0.47 (0.7) 0.50 (8.8) 0.81 (4.9) 0.87 (2.0) 0.81 (10.9)
23 0.04 (6.4) 0.07 (7.2) 0.14 (9.5) 0.21 (8.3) 0.19 (3.6)
24 0.22 (1.4) 0.30 (11.1) 0.44 (0.1) 0.59 (3.9) 0.52 (1.8)
22r 0.50 (9.7) 0.64 (2.5) 0.93 (17.5) 0.87 (2.5) 0.72 (16.2)
26 0.76 (8.2) 0.70 (7.8) 1.11 (15.8) 1.09 (27.2) 1.08 (2.0)
5.3.1.8 Effects of Different Testing Machine
Most of the research on compression testing performed in this study utilized
a smaller capacity testing machine under deflection control. However, many
laboratories use larger capacity concrete compression machines under load rate
control. To assess the relative difference in strength values, a limited study was
performed to compare the results of a small capacity machine (Instron, 100 kN
capacity) under deflection control with a larger capacity machine (1780 kN Tinius
Olson) under load control. For instance, for specimens from mixture #22 at the age
of 7 days, the small machine yielded a strength of 0.50 MPa and a COV of 1.4%,
and the large machine yielded a compressive strength of 0.24 MPa and a COV of
22.4%. The larger compression machine showed significantly lower strength
values and significantly higher variations than those obtained with the smaller
machine. Thus, caution should be taken when using a large capacity machine
173
(Tinius Olson, 1780 kN capacity) under load control. The results from this
investigation are summarized in Table 5.22. It should be noted that in order to
meet the-time-to-failure limits described in ASTM D 4832 (not less than two
minutes), a load rate of 6.9 kPa/sec was used for the large machine. In general, the
results obtained from the larger compression machines lack the accuracy in the
lower range of load values to assess the strength of CLSM.
Table 5.22: Effects of different testing machines on compressive strength

(coefficient of variation is included in parentheses)
Age 7 days 28 days
Specimen 75 x 150 100 x 200 75 x 150 100 x 200
Size (mm)
Machine Small Large Small Large Small Large Small Large
Strength Strength Strength Strength Strength Strength Strength Strength
MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%) MPa (%)
Mix #4 0.48 (6.2) 0.26 (15.8) 0.79(11.0)
Mix #4r 0.49 (6.3) 0.48 (12.2) 0.94 (3.9) 0.84 (6.5) 0.81 (22.8)
Mix #6 0.47 (0.7) 0.35 (7.3) 0.50 (8.8) 0.43 (9.3) 0.81(4.9) 0.66 (20.6) 0.87 (2.0) 0.84 (2.3)
Mix #23 0.04 (6.4) 0.08 (69.6) 0.07 (7.2) 0.10 (33.8) 0.14(9.5) 0.16 (14.2) 0.21 (8.3) 0.15 (33.2)
Mix #24 0.22 (1.4) 0.24 (28.9) 0.30 (11.1) 0.26 (39.3) 0.44(0.1) 0.35 (12.7) 0.59 (3.9) 0.37 (10.1)
Mix #22r 0.50 (9.7) 0.31 (22.4) 0.64 (2.5) 0.48 (19.6) 0.93(17.5) 0.82 (10.3) 0.87 (2.5) 0.70 (0.5)
Mix #26 0.76 (8.2) 0.35 (7.3) 0.70 (7.8) 0.65 (13.9) 1.11 (15.8) 0.96 (2.1)
5.3.1.9 Effects of Curing Temperature and Humidity
The study summarized in this section was a follow-up to previous testing
that suggested that temperature plays a major role in CLSM strength development.
Three curing temperatures (10 ºC, 21 ºC, and 38 ºC) and six CLSM mixtures (H-1
to H-6 in Table 3.16) were selected to study the strength gain of CLSM across a
range of practical construction conditions. After casting the specimens into plastic
174
molds, the cylinders were immediately transported to the appropriate temperature-
controlled chambers. After three days of storage in the chambers, half the cylinders
from each mixture were stripped or removed from the molds and returned to the
same chamber until the time of testing. This curing regime is referred to as “dry”
in subsequent discussions. The other half of the specimens from a given mixture
were kept inside the molds with the caps firmly on the top of the mold until the day
of testing (designated as “wet” curing). These cylinders were placed directly next
to the cylinders that had already been stripped.
The results of compression testing at 7, 28, and 91 days are shown in Table
5.23. Also included in Table 5.23 are the moisture contents of specimens measured
immediately after compression testing to assess the effects of curing conditions on
the moisture content (or evaporable water content) and strength of CLSM. The test
results at 7, 28, and 91 days are given in Table 5.23 and illustrated in Figures 5.20
through 5.25.
The results summarized in Figures 5.20 to 5.25 confirm the earlier
conclusion that temperature and storage conditions play a major role in the
compressive strength of CLSM cylinders. Once again, it was difficult to discern
clear trends in behavior across all material types and mixture proportions, but
several general observations could be made. These include:
• Mixtures containing fly ash exhibited significant strength gains at 38 °C,

compared to lower temperatures. As expected, the increase in strength was
more pronounced for Class C fly ash, compared to Class F fly ash, mainly
due to the difference in reactivity. For example, compared to cured at 21 °C
, mixture H-5, containing Class C fly ash, showed a 160% increase in
175
strength when cured at 38 ºC, whereas mixture H-2, containing Class F fly
ash, showed a 40% increase in strength for the same conditions. The CaO
content of fly ash is generally the most important variable affecting
compressive strength and should be taken into account when using CLSM,
especially when temperature extremes may be encountered.
176
Table 5.23: Compressive strength of specimens cured at different
temperatures at different ages
Age 7 days 28 days 91 days

Mix Temp* Moisture Compressive C.O.V. Moisture Compressive C.O. V. Moisture Compressive C.O.V.
# content strength (%) content strength content strength
(ºC) (%) (kPa) (%) (kPa) (%) (%) (kPa) (%)
10 dry 36.9 328.6 3.5 17.8 548.5 7.4 2.0 258.5 2.7
10 wet 38.0 210.5 6.0 37.1 240.9 7.8 27.0 323.4 5.5
H-1 21 dry 18.1 367.5 2.8 1.6 607.3 6.5 1.1 480.2 6.5
21 wet 38.0 299.2 23.0 37.8 266.2 11.8 26.8 632.3 3.3
38 dry 19.5 828.5 2.2 1.8 722.3 4.4 1.1 756.7 12.2
38 wet 22.6 440.4 13.5 35.3 802.2 2.4 26.0 917.3 0.4
10 dry 7.4 188.3 10.2 1.2 314.1 7.0 2.7 118.9 13.6
10 wet 10.3 151.9 8.6 9.6 194.2 12.3 5.5 260.5 3.6
H-2 21 dry 1.8 214.1 2.3 0.3 171.4 10.1 6.5 142.7 7.5
21 wet 9.8 172.4 2.1 9.1 233.1 6.5 3.3 345.3 10.2
38 dry 0.3 256.6 12.2 0.2 211.9 4.9 12.2 263.6 3.2
38 wet 8.6 220.9 11.7 8.9 458.2 7.3 0.4 634.6 2.9
10 dry 6.0 1384.3 1.5 1.6 2315.4 4.9 1.1 1395.4 13.0
10 wet 8.7 937.7 16.2 9.1 1141.2 8.9 7.9 1367.9 3.2
H-3 21 dry 1.6 1213.4 5.1 0.6 963.6 9.6 0.5 852.9 8.9
21 wet 8.7 695.7 8.8 8.6 786.2 3.2 7.4 919.8 6.7
38 dry 0.3 1222.5 6.8 0.0 944.5 3.8 0.4 1042.0 13.5
38 wet 6.3 864.0 2.1 6.2 3844.8 7.4 2.3 3880.9 20.4
10 dry 7.4 314.0 2.8 1.4 1486.2 2.8 1.1 895.8 14.8
10 wet 10.0 185.9 9.4 37.7 893.2 9.6 8.1 1670.3 4.9
H-4 21 dry 1.2 669.6 3.0 0.3 628.6 9.8 0.4 458.6 9.9
21 wet 8.8 501.7 4.1 7.7 1570.7 7.3 3.8 3743.6 4.9
38 dry 0.4 2615.0 8.5 0.9 2041.2 3.5 0.3 2060.6 5.6
38 wet 5.9 2098.8 9.2 3.7 12116.8 11.1 1.4 11512.6 7.0
10 dry 8.7 273.1 6.9 1.4 711.4 3.2 0.8 421.9 5.8
10 wet 10.0 232.6 4.5 9.7 544.5 9.6 8.6 1362.6 6.3
H-5 21 dry 1.4 420.8 3.9 0.3 411.2 5.3 0.3 330.7 12.1
21 wet 10.7 316.8 6.7 10.2 815.2 2.9 8.8 1497.7 4.6
38 dry 0.3 1524.7 9.1 0.2 1423.7 4.6 0.1 1339.4 11.9
38 wet 7.4 1472.5 7.3 7.7 2282.0 8.6 3.5 2638.2 12.0
10 dry 6.3 281.9 16.7 1.6 740.5 8.7 1.1 669.3 3.4
10 wet 8.7 210.6 6.9 8.1 470.5 1.2 7 922.2 5.6
H-6 21 dry 1.0 480.4 16.9 0.3 434.5 15.0 0.3 372.9 18.7
21 wet 7.5 371.4 15.8 6.2 744.7 7.2 6.2 929.6 9.4
38 dry 0.4 816.8 11.9 0.2 828.0 28.1 0.2 782.0 4.1
38 wet 6.1 562.5 10.7 4.8 786.3 12.6 0.4 991.3 7.7
*: dry = cylinders stripped after 3 days; wet = cylinders kept in mold until time of testing
177
Figure 5.20: Compressive strength of mixture H-1 for different curing
conditions

conditions
178
conditions

conditions
179
conditions

conditions
180
• Drying (from an age of three to seven days) generally increased the 7-day
strength of CLSM, compared to keeping the cylinders in molds
continuously for 7 days. For 91-day testing, the opposite effect was
generally observed. That is, cylinders continuously left in sealed plastic
molds exhibited higher strengths than companion cylinders that were
allowed to dry after the first three days. The results at an intermediate age
of 28 days were not as well defined, and general trends were not evident.
• The measurement of the moisture content of cylinders at the time of testing
may provide useful information on its performance, especially when
subjected to drying conditions. In some cases, the reduction in moisture
content for cylinders cured outside of their molds at higher temperatures
was accompanied by microcracking and subsequent strength reductions.
• The standard curing methods might severely under-estimate CLSM strength
development in the field, especially in humid environments. On the
contrary, for the case of mixture H-4, the dry specimens at 21 ºC only
developed 25% of compressive strength estimated in the lab. When
choosing a CLSM mixture, its proportions should be carefully studied in
relation to its application environments, such as temperature and loss of
moisture.
• In general, mixtures without fly ash were less sensitive to curing
temperature and drying conditions. The predictive models described earlier
in this chapter were generally effective in predicting CLSM without fly ash,
but significant differences were evident when applying the predictive
models to fly ash mixtures. Thus, particular attention should be paid to
assessing the long-term strength gain of CLSM containing fly ash,
especially when high-calcium ashes are used in warmer climates.
• Field experience is essential in understanding the performance of local
materials in specific applications and climates. The experience gained is
181
particularly important in assessing long-term strength gain and its effects on
excavatability.
• An assessment of the on-site strength of CLSM should take into account
laboratory-obtained test results, but it should also take into account climatic
conditions. An understanding of material reactivity is helpful in
extrapolating laboratory results to field performance.
5.3.2 Triaxial Shear Strength
Using the I-series CLSM mixtures, the triaxial shear strength of several
CLSM mixtures was measured. The results, shown in Table 5.24, confirmed the
observation of other researchers that the strength of CLSM is composed of both
chemical bonding and internal frictional resistance. For the mixtures investigated
in the present study, the behavior was found to be a function of specific material
and mixture proportions, and the effects changed with increased curing time.
By using the approach of stress paths, the strength parameters of CLSM
mixtures were calculated. For mixtures I-1, I-2, and I-3, the internal friction angles
and cohesion both increased with time (between 7 and 28 days). Their friction
angles at 28 days were in the range of a very dense granular soil, and the mixtures
behaved like dense sand, with lower residual strengths than ultimate strengths. For
mixture I-4, the strength development was manifested mainly as an increase in
internal friction angle, whereas for mixture I-5, an increase in cohesion was the
dominant factor. These mixtures (I-4 and I-5) had high air contents and exhibited
behavior similar to loose or uncompacted sands. It was interesting to note that for
mixture I-6, the friction angle decreased and cohesion greatly increased with time.
182
Figure 5.26 illustrates some of the more profound differences in CLSM behavior,
contrasting the stress-strain behavior of mixtures I-1 and I-4 samples at 28 days.
Table 5.24: Results of triaxial compression tests

Mixture 7 days 28 day
Friction angle φ′ Cohesion c′ Friction angle φ′ Cohesion c′
(degree) (kPa) (degree) (kPa)
I-1 36.1 31.8 42.8 40.1
I-2 36.1 96.1 38.7 174.7
I-3 39.4 251.7 47.9 346.2
I-4 22.0 43.9 23.9 43.1
I-5 19.5 89.8 18.5 130.0
I-6 37.3 44.4 33.9 93.4
600
500
I-4
400
I-1
300
200
100
0
0 2 4 6 8 10
Strain (%)
Figure 5.26: Stress-strain curves of mixture I-1 and I-4 at 28 days (triaxial
compression testing with an effective confining pressure of 69
kPa)
183
5.3.3 Water Permeability
The water permeability (or hydraulic conductivity) for six CLSM mixtures
(I-series) are shown in Table 5.25. These permeability values (measured after 28
days of moist-curing) were in the range of silts. The results show that water-
cement ratio affected the coefficient of permeability greatly. Generally, the lower
the w-c ratio, the lower the permeability. Interestingly, the high air content of
mixture I-5 did not increase the permeability significantly, indicating that the
entrained air bubbles were not well connected. The testing performed on these six
mixtures indicates that CLSM can be easily assessed for water permeability using
equipment commonly used to characterize soils. Additional research on water
permeability in which the effects of freezing-and-thawing damage on permeability
were also assessed.
Table 5.25: Water permeability (hydraulic conductivity) of selected CLSM

mixtures
Mixture Permeability Water-cement ratio

(cm/sec)
I-1 2.46×10-4 9.4
I-2 5.33×10-5 5.5
I-3 1.45×10-5 2.8
I-4 4.20×10-4 3.5
I-5 6.75×10-4 2.9
I-6 2.89×10-5 7.0
The water permeability of D-series mixtures was measured after twelve
freezing-and-thawing cycles in accordance with the modified ASTM D 560. The
184
results are shown in Table 5.26 and compared with permeability values of CLSM
specimens that were not exposed to freezing-and-thawing cycles.
Table 5.26: Test results of water permeability after freezing-and-thawing

cycles
Mixture # D-1 D-2 D-3 D-4 D-5 D-6 D-7 D-8 D-9 D-10 D-11
28-day
compressive
0.13 1.02 0.79 1.47 0.11 0.12 0.20 0.38 0.25 0.53 0.34
strength
(MPa)
Permeability
before F-T
7.38 0.35 0.21 13.02 6.60 3.87 1.63 1.60 3.08 3.63 8.06
cycles (× 10-2
mm/sec)
Permeability
after 12 F-T
14.21 1.90 1.61 8.94 0.08 0.94 1.72 0.42 0.73 0.36 5.94
cycles (× 10-2
mm/sec)
Mixtures that would likely suffer freezing-and-thawing damage were
selected to measure possible changes in permeability (before and after testing, as
shown in Table 5.26). However, the effects of freezing-and-thawing damage on
water permeability were somewhat inconclusive, with some mixtures showing
increased permeability and others showing decreased permeability. This was most
likely due to the test set-up, which was designed to keep the samples intact, thus
allowing for subsequent permeability testing. However, by keeping the samples
intact (and confined), it may not have allowed for an accurate estimate of in-situ
permeability. Because the samples were confined, the expansion due to freezing
and thawing may have actually compacted the samples, resulting in an apparent
185
reduction in permeability. More work is needed to elucidate the effects of freezing-
and-thawing damage on permeability.
5.3.4 Drying Shrinkage
As stated in Chapter 4, there are no standard methods to evaluate the drying
shrinkage of CLSM, and only limited emphasis was placed in this project on the
topic. A method developed in Germany for flooring applications was used in this
study, with the results shown in Table 5.27. The temperature during the testing
period was 20 o C, and the relative humidity was approximately 60 to 65%. For
mixtures without air entrainment, most of the shrinkage apparently occurred during
the first day. Much of this may be attributable to early bleeding and subsidence,
which may result in a net volume decrease of the solid mixture. This method is
intended to assess concrete mixtures, as opposed to CLSM, and no attempts were
made to modify or optimize the method for CLSM. It may be possible that CLSM
does not exhibit enough strength (or stiffness) to effectively cause meaningful
movement of the end anchor, thereby limiting its subsequent effectiveness in
measuring shrinkage strains. More research is needed to examine drying shrinkage
of CLSM and to assess and recommend suitable test methods. Because the topic of
drying shrinkage was not identified as a critical issue for this study, it is not
possible to offer firm recommendations on its relevance or on appropriate tests.
186
Table 5.27: Drying shrinkage of selected CLSM mixtures
-6
Shrinkage strain (x 10 )
Time Mix #4 Mix #24 Mix #23 Mix #6 Mix #22r Mix #26
1 day 2260 2830 80 1440 90 10
2 days 2280 2850 80 1450 90 30
3 days 2280 2860 80 1450 90 50
4 days 2280 2860 80 1450 100 70
5 days 2300 2860 80 1460 100 100
6 days 2310 2880 80 1480 110 130
7 days 2330 2880 80 1500 120 160
2 weeks 2390 2930 160 1540 150 200
3 weeks 2410 2960 180 1590 150 190
4 weeks 2410 2980 180 1529 160 210
5 weeks 2410 2980 180 1600 160 220
6 weeks 2420 2960 180 1610 160 220
7 weeks 2410 2960 180 1600
5.3.5 California Bearing Ratio
Based on a prioritization of the testing program in Chapter 4, only limited
testing was performed on assessing CBR values of CLSM. Standard crushed rock
has a CBR value of 100. The results of testing performed on six CLSM mixtures
are shown in Table 5.28. Most of the mixtures exhibited quite high CBR values,
with the exception of some fly ash mixtures and high-air mixtures. Based on the
values obtained, it can be concluded that the specific mixtures investigated would
function as suitable base or subbase material. More importantly, the results and
experience confirm that it is feasible to determine CBR values for CLSM using
equipment commonly used to evaluate soils in typical testing laboratories.
Table 5. 28: CBR values for six selected CLSM mixtures

Mixture 4 6 23 24 22r 26
CBR (%) 216 176 20 62 115 150
187
5.3.6 Resilient Modulus
The resilient moduli of the six selected CLSM mixtures were measured and
the results are shown in Table 5.29. These values were magnitude higher than
typical soils. Based on the values obtained, it can be concluded that the specific
mixtures investigated would function as suitable base or subbase material. More
importantly, the results and experience confirm that it is feasible to determine
resilient modulus values for CLSM using equipment commonly used to evaluate
soils in typical testing laboratories.
Table 5.29: Resilient modulus of six selected CLSM mixtures

Mixture #4 Mixture #6
-3.3517 -2.6284
Regression equation Mr=3.00×1010(Sd ) Regression equation Mr=3.00×106(Sd )
Coefficient of 0.9155 Coefficient of 0.8483
determination R2 determination R2
Deviator stress (kPa) Total Mr (GPa) Deviator stress (kPa) Total Mr (GPa)
276 199.45 69 69.28
345 66.23 138 3.01
414 52.52 207 1.94
276 2.06
Mixture #23 Mixture #24
-2.2978 -2.8847
Regression equation Mr=1.46×105 (Sd ) Regression equation M r=1.00×108(Sd )
34.5 57.61 138 134.32
69 4.79 207 20.63
103.5 3.49 276 11.19
138 2.36 345 8.47
414 4.74
Mixture #22r Mixture #26
-0.4929 -1.4393
Regression equation Mr=3.06×102 (Sd ) Regression equation Mr=6.57×105 (Sd )
207 23.33 414 104.44
276 17.99 483 89.48
345 16.6 690 72.93
414 15.77 828 33.25
483 15.25
552 13.54
188
5.3.7 Excavation Study
This section summarizes the results of tests that are directly or indirectly
related to the excavatability of CLSM. Included are the initial findings from tests
conducted on selected CLSM mixtures (six from the original mixture series) and
the results of subsequent, more comprehensive testing on excavatability and related
indices. The use of the splitting tensile test as a potential index of excavatability
was investigated, and as such, the tensile results are presented in this section.
As previously described in Chapter 4, the excavatability of CLSM was
assessed for six of the original 38 mixtures by casting CLSM into 450-mm x 450-
mm x 300-mm plywood boxes. The early strength or stiffness of CLSM was
assessed using a soil penetrometer, and these values were correlated with the
“walkability” or the time at which an average person can walk on the material with
a slight impression. It was found that the soil penetrometer values in the range of
4.32 kPa to 7.35 kPa correlated with the initial “walkability.” Long-term
excavatability was assessed for the six CLSM mixtures at an age of approximately
nine months using typical hand tools, including a shovel and a pick. Prior to
assessing the excavatability, the “stiffness” of the samples was measured using the
Stiffness Gauge instrument (as described earlier). Compressive strengths of lab-
cured cylinders were also measured at the time of excavation.
As shown in Table 5.30, there was no clear correlation between
compressive strength, excavatability, and stiffness (as measured by the Stiffness
Gauge). For example, the lab-cured compressive strength of mixture #23 was quite
low, but the field-cured excavation box was not excavatable. It has been shown in
189
previous testing that laboratory-cured cylinders may not be accurate indicators of
in-situ strength or stiffness, especially when CLSM is exposed to higher
temperatures in the field (as was the case for these samples). Also, the results
suggest that compressive strength, by itself, may not be a good predictor of
excavatability. Another example of lack of correlation was the fact that mixture
#24 had a higher “stiffness” than mixture #22r, yet it was much easier to excavate.
The findings of this initial study led to more comprehensive research on
excavatability, including the assessment of other test methods and indirect indices,
as described next.
Table 5.30: Results of excavation study

Compressive C.O.V. of Stiffness* C.O.V. of
Mixture # strength strength stiffness Excavatability
(MPa) (%) (MN/m) (%)
24 0.31 4.86 11.62 10.27 Easy
22r 1.01 5.49 9.76 5.40 Difficult
6 0.92 9.66 13.20 6.72 No
4 0.70 5.24 30.72 9.10 No
26 1.61 5.91 34.57 2.09 No
23 0.12 22.31 17.15 7.51 No
* Stiffness was measured using the GeoGauge device
A comprehensive follow-up study to the initial excavatability investigation
was performed. A wide range of CLSM mixtures (C-series) was included in the
investigation, and the following methods or approaches were assessed as possible
indices (direct or indirect):
• Unconfined compressive strength (field-cured cylinders)

• Field penetrometer (field version of ASTM C 403 needle penetrometer)
190
• Dynamic cone penetrometer (DCP)
• CBR (estimated from DCP)
• Stiffness gauge (GeoGauge)
• Removability modulus (RE)
• Splitting tensile strength
The results of field penetrometer and compressive strength testing at

various ages are shown in Table 5.31. In many cases, the mixtures were too strong
to obtain a penetration value. Between 7 and 28 days, all mixtures increased in
compressive strength, but some mixtures (C-1, C-3, and C-9) exhibited reductions
in strength between 28 and 91 days. Some of the excavation box samples also
exhibited reductions in strength or stiffness, perhaps due to the long-term exposure
to the elements. Thus, some samples were not penetrable at earlier ages but
became penetrable at later ages.
Table 5.31: Comparison between field penetrometer values (field-cured

samples) and compressive strength values (field-cured cylinders)
Mix 7 days 28 days 91 day 240 days
# Penetration Strength Penetration Strength Penetration Strength Penetration Strength
(MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa) (MPa)
C-1 5.86 0.44 - 0.62 13.78 0.46 11.9 1.13
C-2 4.27 0.11 - 0.24 - 0.28 - 0.49
C-3 2.27 0.06 2.41 0.13 3.31 0.08 Too soft 0.06
C-4 3.24 0.11 - 0.18 9.92 0.40 14.0 0.45
C-5 - 0.41 - 1.37 - 1.54 - 3.29
C-6 7.65 0.19 - 0.29 - 0.35 - 0.36
C-7 - 0.43 - 1.13 - 1.10 - 1.40
C-8 2.83 0.14 - 0.47 - 0.85 - 1.61
C-9 1.98 0.16 - 0.42 6.06 0.30 6.8 0.20
The dash,“-”, indicates samples that were too strong and penetration values were not btained
Figure 5.27 illustrates the relationship between field penetrometer values
and the compressive strength of field-cured companion cylinders. Although there
191
0.6
0.5
R2 = 0.67
Compressive strength
0.4
(MPa)
0.3
0.2
0.1
0
0 2 4 6 8 10 12 14 16
Needle penetration (ASTM C 403) (MPa)
Figure 5.27: Penetration resistance (ASTM C 403) vs. compressive strength at

different ages (all C-series mixtures)
was a general trend that mixtures with higher compressive strength exhibited higher
penetration values, the correlation was not very strong. A possible explanation for
the lack of fit between the two data sets is that the data include all mixtures from
the C-series, even though some mixture types show different inherent relationships
between strength and penetration resistance. For example, paste mixtures exhibit
relatively high compressive strengths but require substantially less penetration
resistance. Grouping CLSM based on specific materials and mixture proportions
generally yielded a better correlation between strength and penetration resistance.
Thus, it is conceivable that reliable databases linking strength to penetration
resistance can be developed for specific materials and commonly used mixture
192
proportions. However, care should be taken in trying to extrapolate trends
developed for one set of common mixtures (i.e., paste mixtures) to other mixture
types (i.e., high-air mixtures).
Table 5.32 summarizes additional results from the excavatability study,
including DCP values, stiffness values (using GeoGauge), and calculated
Removability Modulus (RE) values. The table also shows the compressive strength
for lab-cured cylinders (at 28 days) and field-cured cylinders, which were cured
adjacent to the excavatability boxes and tested at the time of excavation (240 days).
The densities of the lab-cured and field-cured cylinders were measured before
testing them in compression, and these values were used in RE calculations. The
relative ease of excvation was assessed using a hand shovel.
The DCP index values, which indicate the penetration depth per blow, were
measured for each of the excavation boxes. The minimum value for a recordable
count of blows corresponded to a penetration of at least 25 mm. DCP values were
found to decrease until the specimens ultimately suffered large cracks, as shown in
Figure 5.28. After the large cracks appeared, the DCP values progressively
increased. Thus, the lowest index value was taken for each mixture and used in
Table 5.32 because it represented the most difficult portion to excavate, thus
providing a conservative index.
193
Table 5.32: Results of excavatability study
Mix 28-day Density 240-day RE using RE using DCP CBR Stiffness Relative
# compressive (kg/m3 ) compressive 28-day 240-day index (%) (using ease of
strength strength lab-cured field- (mm) GeoGauge) excavation
(lab-cured) (field-cured) cylinders cured (MN/m) (with
(kPa) (kPa) cylinders shovel)*
C-1 533 1512 1134 0.84 1.22 6.4 37 19.37 3
C-2 807 1946 493 1.51 1.18 6.5 36 40.56 7
C-3 144 1724 63 0.53 0.35 29.0 7 19.03 1
C-4 292 1858 447 0.85 1.05 5.2 46 28.97 6
C-5 1027 2094 3290 1.90 3.39 0.5 100 30.87 9
C-6 332 2252 362 1.20 1.25 5.0 48 18.54 8
C-7 1192 2143 1397 2.12 2.29 1.1 100 11.96 8
C-8 658 2171 1615 1.60 2.51 0.6 100 23.89 10
C-9 417 1660 199 0.85 0.59 10.0 22 23.03 4
*Note: The ease of excavation wa s compared by assigning different values to each mixture, where 1 is easiest and 10 is most
difficult.
The GeoGauge was used to assess the relative “stiffness” of the CLSM
specimens. As CLSM mixtures were quite strong (relative to soil), a thin layer of
wet fine sand was placed on the surface prior to testing, as per the
recommendations of the manufacturer. Three readings were taken for each
mixture. The variations were quite high for the device, with coefficients of
variation as high as 40% for some specimens. A comparison between “stiffness”
values and DCP values is shown in Figure 5.29. There was no clear trend between
stiffness values and DCP, nor was there a clear trend between stiffness values and
actual excavatability (by shovel). In general, the GeoGauge was not found to be an
effective means of assessing the properties of CLSM, both because of poor
reproducibility and inability to predict excavatability.
The correlation between the DCP index and the two sets of calculated RE
values (based on 28-day lab-cured cylinders and 240-day field-cured cylinders) are
shown in Figure 5.30 and 5.31. The RE values calculated using field-cured
194
cylinders, tested at the time of excavation, showed a better correlation with DCP
index than RE values based on 28-day lab-cured cylinders. The RE values based
on 28-day lab-cured cylinders did, however, identify all the mixtures that were not
excavatable (i.e., their RE values were greater than 1). However, using the 28-day
data is quite conservative and may not be suitable for all CLSM mixtures.
Figure 5.28: Cracking of CLSM mixture C-2 during DCP penetration
195
45
40
35
30
25
20
15
10
5
0
0 5 10 15 20 25 30 35
Penetration Index (mm per blow)
Figure 5.29: Correlation between stiffness (from GeoGauge) and DCP index
Using a correlation table provided by the DCP manufacturer, the
corresponding California Bearing Ratio (CBR) values were obtained from the DCP
index. However, according to the manual, the highest possible CBR value (100%)
calculated was reached for some of the CLSM mixtures, making it difficult to use
the value quantitatively. Thus, the DCP index, as previously described, was used
as an index for excavatability, rather than calculated CBR values.
As shown in Figure 5.31, a DCP index of 5 mm per blow correlated well
with an RE value of 1.0. This suggests that the DCP may be an effective, user-
friendly method of assessing excavatability in the field. This approach needs
further investigation in field testing, where excavatability may not be assessed only
196
using hand tools, but also with typical, commercial excavation equipment (i.e.,
backhoe).
35
30
25
y = 5.2798x-2.557
R2 = 0.76
20
15
10
0
0.00 0.50 1.00 1.50 2.00 2.50
RE using 28-day strength
calculated using 28-day compressive strength (lab-cured
cylinders)
40
35
30
25 y = 5.4769x-1.8217
20 R 2 = 0.93
15
10
5
0
0.00 1.00 2.00 3.00 4.00
RE using field-cured cylinders
calculated using 240-day compressive strength (field-cured
cylinders)
197
Another parameter that potentially might be used as an index for
excavatability is the splitting tensile strength of CLSM. Some preliminary trials
found that tensile strength may, in fact, be more suitable than compressive strength
in assessing excavatability. Although splitting tensile tests were not performed on
the C-series mixtures, some tests were performed on other mixture series. The
results are provided in this section because of the potential of applying tensile data
to excavatability predictions.
The splitting tensile strengths of the E-series mixtures were measured, as
shown in Table 5.33. A split specimen of mixture E-1 is shown in Figure 5.32. For
the E-series CLSM mixtures, the splitting tensile strength to compressive strength
ratio ranged from 9 to 17 percent, which is higher than those typically observed for
conventional concrete. Unlike concrete, this ratio did not substantially decrease
with an increase in compressive strength.
Table 5.33: Compressive and splitting tensile strengths at 7 and 28 days

Mixture 7 days 28 days 91 days
Average C.O.V. fst /fc Average C.O.V. fst /fc Average C.O.V. fst /fc
(kPa) (%) (%) (kPa) (%) (%) (kPa) (%) (%)
E-1 57.0 4.0 18.5 62.5 9.8 11.7 161.5 8.2 19.0
E-2 21.1 5.3 17.5 28.4 19.9 14.5 31.2 19.1 11.9
E-3 14.4 8.2 17.1 20.6 17.5 14.3 27.2 11.7 18.9
E-4 25.0 14.5 19.6 29.5 8.3 10.1 57.6 6.7 12.8
E-5 46.9 12.9 23.6 101.1 23.1 9.8 150.7 23.0 9.5
E-6 34.2 19.3 10.0 57.3 14.7 17.3 61.9 8.3 13.0
E-7 50.5 6.9 11.8 164.2 30.3 13.8 188.5 18.5 11.0
E-8 33.2 16.3 17.3 75.4 8.1 11.5 146.1 1.8 11.8
Note: fst is the splitting tensile strength and fc is the compressive strength.
198
Figure 5.32: A cylinder from mixture E-1, before and after being tested for
splitting tensile strength
Additional splitting tensile tests were performed using the H-series mixtures
to assess the effects of drying on tensile strength and the tensile to compressive
strength ratio. This testing was initiated because drying generally has a more
profound effect on tensile strength than compressive strength, at least ni the case of
conventional concrete. This behavior is generally attributed to the effects of
microcracks. The results, shown in Table 5.34, confirmed that drying had a similar
effect on CLSM, significantly lowering the tensile to compressive strength ratio.
The effects of temperature and drying on the tensile strength of CLSM, especially
as it relates to using tensile strength as an index of excavatability, require additional
research. It is expected that further research will help to classify some of these
effects and to determine the potential for using splitting tensile strength (by itself or
in combination with other parameters) as an indicator of excavatability.
199
Table 5.34: Effects of temperature and drying conditions on splitting tensile
strength of CLSM
Mixture H-1 H-2
Condition Average C.O.V. ft /fc Average C.O.V. ft /fc
(kPa) (%) (%) (kPa) (%) (%)
10 ºC, dry 16.7 23.1 6.5 6.0 20.2 5.1
10 ºC, wet 25.6 19.3 7.9
21 ºC, dry 23.8 8.6 4.9 8.8 39.6 6.1
21 ºC, wet 89.0 27.9 14.1
38 ºC, dry 55.0 18.8 7.3 18.9 11.4 7.2
38 ºC, wet 74.5 6.1 8.1 53.4 15.5 8.4
Mixture H-3 H-4
(kPa) (%) (%) (kPa) (%) (%)
10 ºC, dry 100.4 14.1 7.2 65.5 33.0 7.3
10 ºC, wet 166.3 32.1 12.2 95.8 101.4 5.7
21 ºC, dry 49.2 8.5 5.8 32.2 5.0 7.0
21 ºC, wet 114.5 15.5 12.4 455.9 36.0 12.2
38 ºC, dry 87.8 7.5 8.4 158.6 6.3 7.7
38 ºC, wet 525.4 30.9 13.5 1791.7 12.5 15.6
Mixture H-5 H-6
(kPa) (%) (%) (kPa) (%) (%)
10 ºC, dry 28.8 12.3 6.8 56.1 67.5 8.4
10 ºC, wet 133.0 7.5 9.8 84.5 22.8 9.2
21 ºC, dry 25.6 16.6 7.7 64.1 5.3 17.2
21 ºC, wet 127.8 8.6 8.5 142.8 7.3 15.4
38 ºC, dry 106.8 9.1 8.0 114.1 16.6 14.6
38 ºC, wet 198.2 6.0 7.5 125.3 12.5 12.6
200
5.4 DURABILITY OF CLSM
5.4.1 Freezing-and-Thawing Study
This section discusses the results of two studies on the freezing-and-
thawing resistance of CLSM. The first study involved testing six CLSM mixtures
from the original mixture series using a modified version of ASTM D 560, as
described in Chapter 4. This method was originally developed for the assessment
of soil-cement. The second study used the same method to assess a wider range of
CLSM mixtures (D-series), and also measured the effects of freezing-and-thawing
damage on permeability, which is discussed in Section 5.3.3 in this dissertation.
The results of the initial freezing-and-thawing study are shown in Tables
5.35 and 5.36 for samples cured for 7 days and 28 days, respectively. According to
the “soil-cement laboratory handbook”, the maximum allowable weight loss is 14%
after 12 freezing-and-thawing cycles (for a Group A-1 soil).81 Based on this
criterion, most CLSM mixtures with high air content behaved well, especially after
28 days of moist curing. The possible damage of CLSM due to freezing-and-
thawing may be a concern, especially if it leads to a change in volume of the
material. There has been very little field evidence suggesting that this is a problem
with CLSM, but research was needed, as a minimum, to assess the laboratory
performance of CLSM, and to determine if a suitable method could be
recommended. Based on these preliminary findings, ASTM D 560 (with minor
modifications) was found to be an effective and user-friendly method to assess the
freezing and thawing resistance of CLSM.
201
days of curing)
Mixture # 4 24 23 6 22r 26
Initial weight (g) 2241 1639 1441 2085 1861 1868
Initial moisture content (%) 10.86 37.78 30.37 14.79 10.38 9.17
1 cycle (% of initial weight retained) 100.1 101.2 102.1 100.1 100.5 100.4
2 cycles (% of initial weight retained) 100.3 102.3 100.6 99.9 100.4 100.2
7 cycles (% of initial weight retained) 97.7 Damaged Damaged 88.6 100.2 100.5
8 cycles (% of initial weight retained) 92.4 88.7 100.2 100.4
Moisture content at end of test (%) 13.35 15.92 11.16 6.97
Mass loss (%) 27.9 13.7 0.7 -2.5
Table 5.36: Freezing-and-thawing results for selected CLSM mixtures (after

28 days of curing)
Mixture # 4 24 23 6 22r 26
Initial weight (g) 2241 1638.5 1441 2084.5 1861 1867.5
6 cycles (% of initial weight retained) 100.9 Damaged 98.0 84.0 102.6 99.3
7 cycles (% of initial weight retained) 97.3 98.0 82.0 102.7 99.4
8 cycles (% of initial weight retained) - 98.2 81.3 102.6 99.3
Moisture content at end of test (%) 14.16 27.32 16.91 9.87 4.99
Mass loss (%) 17.4 1.1 35.1 -2.7 -2.9
202
The freezing-and-thawing test results of the D-series CLSM mixtures are
shown in Table 5.37. It is interesting to note that D-1 mixture (high-air mixture
with 30 kg/m3 of cement) did not pass the twelve cycles, whereas a similar mixture
with 45 kg/m3 of cement (D-11) did pass the entire 12 cycles, suggesting that both
air-void system and strength contribute to freezing-and-thawing resistance.
Mixture D-10 survived all 12 cycles, most likely due to its higher strength
(contributed from the Class C fly ash). The remaining mixtures (D-2 through D-9)
did not survive all 12 cycles. Mixtures containing foundry sand (D5 to D7) were
damaged after only one or two cycles. Figure 5.33 shows damaged CLSM samples
containing foundry sand. It was observed that specimens demonstrated volume
expansion and cracking due to the formation of ice lenses.
Table 5.37: Freezing-and-thawing test results of D-series mixtures

Mixture # D-1 D-2 D-3 D-4 D-5 D-6 D-7 D-8 D-9 D-10 D-11
Original
1.38 1.77 1.70 1.86 1.53 1.47 1.46 1.50 1.55 1.56 1.33
weight (kg)
Moisture
9.8 10.5 13.6 10.1 30.7 35.5 32.4 27.4 28 25.1 7.3
content (%)
28-day
strength 0.13 1.02 0.79 1.47 0.11 0.12 0.20 0.38 0.25 0.53 0.34
(MPa)
1 cycle (%) 105.2 99.0 100.1 99.9 Damaged 92.8 93.0 98.9 98.1 99.2 103.5
2 cycle (%) 102.3 95.3 97.3 99.8 Damaged Damaged 94.9 94.8 98.1 105.9
3 cycle (%) 84.4 85.4 78.4 99.1 91.5 87.1 98.0 108.1
4 cycle (%) 74.1 75.7 73.2 96.4 88.0 83.6 98.1 110.3
5 cycle (%) 66.4 72.4 71.1 88.4 84.6 79.1 97.7 110.3
6 cycle (%) 59.6 55.4 63.8 61.0 81.4 73.5 97.3 110.3
7 cycle (%) 51.5 54.9 52.1 59.7 76.9 69.2 97.1 108.8
8 cycle (%) 45.8 46.3 54.3 57.9 73.7 62.1 95.8 108.2
9 cycle (%) 37.3 30.1 48.6 43.8 68.1 56.3 94.4 104.6
10 cycle (%) 33.6 Damaged Damaged 40.3 56.1 Damaged 92.2 103.6
11 cycle (%) 25.8 Damaged 49.8 89.4 100.6
12 cycle (%) Damaged Damaged 85.1 97.5
Final
moisture 28.7 21.3
content (%)
Dry weight
11.0 11.9
loss (%)
203
Figure 5.33: Specimens after freezing-and-thawing cycles (CLSM mixtures
containing foundry sand)
5.4.2 Wetting-and-Drying Study
In this section, the effects of wetting-and-drying cycles were discussed.
Specimens from six selected CLSM mixtures were exposed to wetting-and-drying
cycles at the ages of 7 and 28 days respectively. In this study, the samples were
dried at room temperature, approximately 20 °C. Table 5.38 and 5.39 show the
results. In these tables, from 1 to 11 cycles, the weights of dry samples before
being submerged in water were expressed as a percentage of the weights before the
204
wetting-and-drying cycles. After 12 cycles, the weight loss was expressed in terms
of calculated original dry weights.
Table 5.38: Wetting-and-drying results for selected CLSM mixtures (after 7

days of curing)
Mixture # 4 24 23 6 22r 26
Initial weight (g) 2206 1640.5 1416.5 2102.5 1805.1 1868.9
12 cycles (Mass loss, %) 0.6 1.5 11.8 2.0 -0.6 -0.3
Moisture content at end of test (%) 0.099 0.359 0.385 0.117 0.067 0.059
As shown in Table 5.38, CLSM mixtures, with the exception of mixture
#23, cured for 7 and 28 days, performed satisfactorily throughout the wetting-and-
drying cycles. Mixtures 22r and 26 gained weight when exposed to wetting-and-
drying cycles at 7 days. This was probably due to the continuing hydration of
portland cement. The mass loss of mixture #23 was significantly reduced when it
was tested after 28 days of curing when compared to 7 days of curing. This
suggests that strength is an important factor in resisting wetting-and-drying
damage.
205
Table 5.39: Wetting and drying results for selected CLSM mixtures (after 28
days of curing)
Mixture # 4 24 23 6 22r 26
Initial weight (g) 2261.1 1646.9 1472.3 2134 1790.1 1860.5
12 cycles (Mass loss, %) 0.2 3.6 5.8 0.0 0.0 0.0
Moisture content at end of test (%) 7.6 33.2 33.8 11.0 7.1 6.1
5.5 CORROSION OF DUCTILE IRON IN CLSM
5.5.1 Introduction
In this section, the results of corrosion of ductile iron under accelerated
laboratory testing are presented. As mentioned in Chapter 4, the pH and resistivity
values of CLSM mixtures were measured. The chloride penetration was illustrated
in Figure 5.34. Uncoupled and coupled ductile-iron specimens were exposed to the
3.5% NaCl solution after 28 days of curing. The mass loss of steel specimens was
measured after being exposed to the NaCl solution. Because in cases where the
depths of embedment are high, the corrosion of steel is controlled by diffusion of
dissolved oxygen in the water entrapped in the soil, the composition of carbon steel
206
has little effect on corrosion resistance in a given soil. Thus, the composition of the
ductile iron is not included here.
5.5.2 pH Values of Extruded Solutions from CLSM Mixtures
The measured pH values are shown in Table 5.40. The pH values of
extruded solutions were found to vary with time, with no general trend. A
saturated calcium hydroxide solution has a pH value of 12.582 . Thus, CLSM
mixtures whose pore solutions had pH values less than 12.5 likely had a lower
concentration of Na+, K+, and OH- dissolved or available in solution. Factors
influencing the pH value of the pore solution include the amount of calcium
hydroxide provided through the hydration of portland cement or dissolution from
Class C fly ash, alkalis from raw materials, the consumption of calcium hydroxide
by pozzolanic reaction, the leaching of calcium hydroxide and alkalis to the
environments, and other mechanisms. The difference in pH values between
mixture #2 and #2r could possibly be attributed to the leaching of calcium
hydroxide. It is important to measure pH values of in situ solution CLSM mixtures
to evaluate possible corrosion tendency.
5.5.3 Resistivity of CLSM Mixtures
The most corrosive soils are those that contain large concentrations of
soluble salts. Because of the presence of salts, such soils have relatively high
electrical conductivities (or low electrical resistivities). As a result, resistivity is
often used to approximate the aggressiveness of a soil. The resistivity values of
207
CLSM mixtures were measured using the approach specified in Chapter 4. The
182-day results are shown in Table 5.40. The moisture content of a soil also affects
its resistivity. In this aspect, the saturated condition of CLSM mixtures, as required
by the test method in this study, represents the worst-case scenario when no
aggressive ions are present.
Table 5.40: pH and resistivity values of CLSM mixtures

Mix # 7-day pH 28-day pH 182-days pH 182-day resistivity (Ω-cm)
1 12.05 11.61 11.4 2593
2 11.84 11.67 9.8 2855
1r 11.88 11.42 10.33 3484
15 11.43 10.9 8.63 2122
3 11.99 11.91 11.65 2698
8 11.8 11.9 11.18 3222
10 11.84 11.74 11.28 1781
9 11.8 11.63 11.1 1441
5 11.97 11.72 11.24. --
12 - 11.86 11.58 4086
4 11.89 11.92 11.12 --
7 11.76 11.68 10.8 3196
3r 12.06 11.92 11.58 3248
4r 12.01 11.68 11.2 8487
24 12.1 11.95 11.53 1582
23 12.12 11.17 11.4 3536
18 12.18 12.05 11.93 4128
14 12.09 12.12 11.4 12049
2r 12.16 11.89 11.9 3882
29 11.96 12.06 11.28 3143
30 11.95 12.06 11.2 1624
17 11.96 12.05 11.01 2310
11 12.01 11.93 11.3 3468
6 11.66 11.31 8.7 10478
16 12.24 11.85 11.32 1834
2130 12.23 10.37 12.53 7387
22 12.21 12.24 12.3 8120
22r 12.04 12.24 12.3 3824
5r 12.03 11.94 12.25 12573
26 11.827 11.64 9.07 3091
16r 12.28 12.24 9.8 2881
13 12 11.91 10.1 5605
25 12 12.03 12.93 1048
19 12.27 12.14 12.64 7072
20 12.14 11.97 12.89 5291
27 12.24 12.29 13.03 3065
20r 12.02 12.24 12.97 8644
28 12.08 12.03 12.65 6287
208
But during testing, chloride ions were present and, as such, it may be more
appropriate to measure the resistivity values of CLSM specimens exposed to the
same solution as that used for mass loss testing. However, resistivity alone is not a
sufficient index of susceptibility of steel embedded in CLSM to corrosion because
other corrosion characteristics, such as pH, passivity of steel, and chloride
concentration are also important.
5.5.4 Chloride Concentration and Profiles
For underground iron pipes, the rate of corrosion can be accelerated by the
presence of aggressive ions. In predicting the susceptibility of pipes placed in
engineered backfill materials, it may be beneficial to establish the quantity of
aggressive ions at the pipe location. This study evaluated five CLSM mixtures for
chloride ingress rate. CLSM cylinders (150 mm x 300 mm) were ponded for 22
weeks with a 3.5% chloride solution. Samples were then tested for chloride
contents as a function of cylinder depth. Figure 5.34 shows results from the testing
program. These results indicate that chloride ions easily migrate through CLSM,
and corrosive conditions can develop at relatively early ages.
209
Chloride Profiles for CLSM Mixtures Ponded with 3.5% Chloride Solution
0
50
Depth from top of cylinder (mm)
Mix 4
Mix 6
100 Mix 24
Mix 22R
Mix 26
150
200
250
300
0 10 20 30 40 50 60 70
3
Total chloride content (kg/m )
Figure 5.34: Chloride profiles for select CLSM mixtures after 22 weeks of
exposure.
Visual observations during the testing program indicated that the ponded
solution was migrating through the CLSM at relatively high rates. It was observed
that approximately 650 mL was permeating through the sample from mixture #23
within approximately twenty-four hours. Because of the relatively fast migration
of solution through the sample, ponding was stopped and mixture #23 was not
evaluated for chloride profiles.
It was observed that all other samples exhibited wetting of the cylinder
sides as a result of ponding the chloride solution on the top of the samples. It is
likely that permeation (i.e. flow as a function of a pressure head) of the chloride
210
solution into the CLSM samples was the predominate mode of chloride ingress.
Diffusion of chlorides (i.e., the ingress of chlorides due to differences in chloride
concentration) was most likely a minor contributor.
5.5.5 Mass Loss of Ductile-Iron Specimens
5.5.5.1 Mass Loss of Uncoupled Coupons
This section discusses the corrosion of uncoupled coupons under laboratory
conditions. The test set-up was explained in Chapter 4. As is the case for most
laboratory corrosion tests, the results should be interpreted with caution. This is
especially true for evaluating corrosion of metals in soils, where there are inherent
variations in location, seasons, and other environmental issues. Local pollution
may also influence the corrosion characteristics of steel pipelines. Thus, the results
from laboratory testing may, at best, be used as a potential index of corrosion.
Actual field tests should be conducted to evaluate corrosion characteristics of
pipelines. The corresponding parameters characterizing soil/CLSM conditions
should also be evaluated to assess corrosion rates and provide guidelines for future
applications. The drawback of field testing is the length of time required for
corrosion initiation. It is important to find a balance between accurate, but time-
consuming, field tests and accelerated, but preliminary, laboratory tests. Thus, the
results presented next should be interpreted cautiously.
The ductile-iron specimens were cleaned and weighed in accordance with
ASTM G 51, “Standard Practice for Preparing, Cleaning, and Evaluating Corrosion
Test Specimens.” The results of measured percent mass losses of samples
211
embedded in CLSM only are shown in Table 5.41. Significant variations in mass
loss existed between different mixtures, repeated mixtures, and samples within one
mixture.
Several factors might have contributed to the variations in test results. First,
the variations were possibly due to the inherent nature of CLSM. Because of the
lack of coarse aggregates, it is difficult to break cement lumps and disperse the
material uniformly. This phenomenon is not obvious when ordinary CLSM
specimens, such as 75-mm x 150-mm cylinders for compression, were tested. For
corrosion coupons, due to their small size and corresponding small surface areas,
slight changes of local CLSM conditions may significantly affect the corrosion
results. Local areas containing higher cement and fly ash contents may show
higher local pH values and suffer less corrosion. Also, the properties of the
transition zone between CLSM and coupons may affect the corrosion behavior.
Unlike the case of submerging coupons in liquid corrosive solution directly, a
transition zone exists and may play an important role. For example, the existence
of entrapped air may affect the corrosion behavior by increasing the available
chloride ions and dissolved oxygen. Another potential factor leading to variations
is the actual measurement itself. Because of the very small values of percentage of
mass loss, a very small error may significantly influence the results. Although steel
specimen surfaces are prepared to meet certain requirements, variations are still
expected. The different surface conditions of steel specimens could also affect the
corrosion rate significantly.
212
Table 5.41: Percent mass loss for samples embedded in CLSM only
Sample ID Specimen No. Average
1 2 3
1 0.05 0.05 0.05
1R 0.04 0.05 0.05 0.04
2 0.02 0.02 0.05 0.03
2R 0.03 0.01 0.02 0.02
3 0.01 0.01 0.02 0.01
3R 0.01 0.02 0.06 0.03
4 0.03 0.02 0.02 0.02
4R 0.01 0.01 0.01 0.01
5 0.03 0.05 0.05 0.04
5R 0.04 0.03 0.02 0.03
6 0.01 0.02 0.03 0.02
7 0.03 0.02 0.02 0.02
8 0.02 0.00 0.01
9 0.03 0.02 0.02 0.03
11 0.05 0.01 0.01 0.02
12 0.03 0.04 0.04
13 0.02 0.02 0.13 0.06
14 0.02 0.04 0.02 0.02
15 0.02 0.03 0.02 0.02
16 0.19 0.22 0.15 0.19
16R 0.10 0.12 0.09 0.10
17 0.28 0.72 0.50
18 0.13 0.37 0.25
19 0.62 0.35 0.48 0.48
29 0.03 0.03 0.02 0.02
30 0.03 0.01 0.31 0.12
22 0.37 1.02 0.17 0.52
22R 0.08 0.09 0.11 0.09
23 2.10 5.80 4.60 4.17
24 0.04 1.30 0.67
25 0.03 0.03 0.94 0.33
26 0.11 0.07 0.09 0.09
27 0.06 0.12 0.09
28
20 0.03 0.17 0.03 0.08
21 5.85 1.95 2.85 3.55
213
Because of the variations within a given mixture, the results from a repeated
mixture may be quite different or close to the original mixture. This is because
means from two different populations are compared, which are random variables.
As such, the statistical approach provided below is utilized:83
Step 1: Formulate hypotheses. If the means of two populations are
denoted as µ1 and µ2 , the null hypothesis for a test on two independent means
would be:
H0 : µ1 = µ2
The alternative should be chosen as:
HA: µ1 ≠ µ2
Step 2: Select the appropriate model. For the case of two independent
samples, the hypotheses of step 1 can be tested using the following test statistic:
X1 − X 2
t= (5. 6)
1 1
s ( + ) 0. 5
n1 n 2
in which X1 and X 2 are the means of the samples from populations 1 and 2,
respectively; n1 and n2 are the samples sizes for the samples from populations 1 and
2, respectively; t is the value of a random variable having a t distribution with
degrees of freedom of (n1 +n2 -2); and s is the square root of the pooled variance
given by
(n1 − 1) s12 + ( n 2 − 1) s 22
s2 = (5. 7)
n1 + n2 − 2
214
where s12 and s22 are the variances of the samples from populations 1 and 2,
respectively. This test statistic assumes that the variances of the two populations
are equal, but unknown.
α ). Values of either 1 or 5 percent

Step 3: Select the level of significance (α
are used most frequently.
Step 4: Compute estimate of test statistic. Samples are drawn from the
two populations and the sample means and variances are drawn. In this study,
three samples are drawn per population. The statistic is then calculated to test the
null hypothesis, which assumes the two means are equal.
Step 5: Define the region of rejection. The region of rejection is a
function of the degrees of freedom (n1 + n2 –2), the level of significance, and the
statement of the alternative hypothesis. In this case, the region of rejection is:
t< -tα/2 or t > tα/2
Step 6: Select the appropriate hypothesis. If the sample value lies in the
region of rejection, the null hypothesis should be rejected.
Through this approach, the comparisons between repeated mixtures are
shown in Table 5.42. It is estimated that four out of seven repeated mixtures did
not yield the same corrosion mass losses as the original mixtures. There were two
possible causes for the differences between repeated and control mixtures. One
cause was that the mass loss was very sensitive to small changes of CLSM
ingredient materials or/and mixing conditions, which is different for every mixture.
The other was due to the small sample sizes. As already pointed out, the within-
batch variations were found to be significant. Thus, larger sample sizes (> 3) are
215
required to represent the whole population in corrosion test. Further study is
needed to determine which factor is more dominant or whether both are equally
important.
Table 5.42: Statistical comparison of repeated CLSM mixtures (uncoupled

corrosion)
Population Population Estimated Critical Accepted
# #r X1 X2 S2 statistic (t) value (t), hypothesis
α=0.05
1 1r 0.050 0.047 2.222E-05 1.018 3.182 Null
2 2r 0.030 0.020 2.000E-04 1.591 2.776 Null
3 3r 0.013 0.030 3.667E-04 -1.958 2.776 Null
4 4r 0.023 0.010 1.667E-05 7.348 2.776 Alternative
5 5r 0.043 0.030 1.167E-04 2.777 2.776 Alternative
16 16r 0.187 0.103 7.333E-04 6.924 2.776 Alternative
22 22r 0.520 0.093 9.887E-02 3.053 2.776 Alternative
The mass loss of samples in different CLSM mixtures was found to vary
significantly. Many factors may affect the corrosion behavior of metal in different
CLSM mixtures. These factors include constituent materials, permeability,
chloride ion diffusion coefficients, subsidence, segregation and bleeding, pH of
pore solutions, and resistivity of the CLSM mixtures.
Ions dissolved in water can enter CLSM mixtures by capillary suction,
permeation, and diffusion. In general, samples embedded in CLSM mixtures with
entrained air contents from 15 to 30% showed much higher corrosion rates than
those in mixtures with entrapped air only.
216
Constituent materials of CLSM mixtures may also influence the corrosion
behavior of embedded pipe specimens. Statistical software (SAS) was used to
analyze the significance of potentially important factors. The General Linear
Models (GLM) Procedure was used because of its ability to handle unbalanced
experimental design with repeated measurements. Four types of F-tests were
utilized to evaluate the significance of factors. A p-value was calculated for each
F-statistic and the factor was determined significant if the p-value was less than a
0.05 significance level. This procedure can also cope with cases of missing data.
As mentioned above, the use of fly ash may significantly change the corrosion
behavior of samples in CLSM mixtures. Thus, the analyses were performed
separately for mixtures with and without fly ash. Factors analyzed include cement
content (two levels), fly ash type (three levels), fly ash content (two levels),
aggregate type (three levels), and water-cement ratio (continuous variable). The
results are shown in Tables 5.43 and 5.44.
Results in Table 5.43 indicate that there was no single factor significantly
affecting the corrosion of samples in CLSM mixtures with fly ash. However, the
interaction between cement content and fly ash type did seem to play an important
role. For instance, samples in mixture #11 (60 kg/m3 cement and 360 kg/m3 high-
carbon fly ash) corroded less than those in mixture #1 (30 kg/m3 cement and 180
kg/m3 Class C fly ash). Thus, when choosing CLSM mixture proportions,
precautions should be taken to evaluate combinations of cement content and fly ash
type.
217
For CLSM mixtures with entrained air (without fly ash), air content and
aggregate type were found by one test to significantly affect the corrosion rate of
embedded samples. It is worth noting that air content may influence corrosion
rates. One test in Table 5.44 indicates that the air content could affect corrosion
rate. More study is required to further assess these factors. The use of different
aggregates (concrete sand and bottom ash) may also affect the ingress of chloride
ions by permeability.
mixtures containing fly ash)
Source DF Type I SS Mean Square F Value Pr > F
Cement content 1 0.00021188 0.00021188 0.62 0.4335
Fly ash type 2 0.00109877 0.00054939 1.62 0.2091
Fly ash content 1 0.00045676 0.00045676 1.35 0.2519
Aggregate 2 0.00134288 0.00067144 1.98 0.1497
Water-cement ratio 1 0.00003464 0.00003464 0.10 0.7508
Cement * fly ash type 2 0.00224639 0.00112319 3.31 0.0453
Source DF Type II SS Mean Square F Value Pr > F
Cement content 1 0.00022344 0.00022344 0.66 0.4213
Fly ash type 2 0.00115231 0.00057616 1.70 0.1942
Fly ash content 1 0.00022754 0.00022754 0.67 0.4171
Aggregate 2 0.0097992 0.00048996 1.44 0.2464
Source DF Type III SS Mean Square F Value Pr > F
Cement content 1 0.00001570 0.00001570 0.05 0.8306
Fly ash type 2 0.00141477 0.00070739 2.08 0.1358
Fly ash content 1 0.00022754 0.00022754 0.67 0.4171
Aggregate 2 0.0097992 0.00048996 1.44 0.2464
Source DF Type IV SS Mean Square F Value Pr > F
Cement content 1 0.00001570 0.00001570 0.05 0.8306
Fly ash type 2 0.00141477 0.00070739 2.08 0.1358
Fly ash content 1 0.00022754 0.00022754 0.67 0.4171
Aggregate 2 0.0097992 0.00048996 1.44 0.2464
Note that “*” means interaction.
218
mixtures containing no fly ash)
Cement content 1 0.17285714 0.17285714 0.10 0.7584
Air content 1 20.88580011 20.88580011 11.72 0.0024
Aggregate 1 18.73837547 18.73837547 10.52 0.0037
Cement * aggregate 1 0.71627884 0.71627884 0.40 0.5326
Cement content 1 0.09839648 0.09839648 0.06 0.8164
Air content 6.74734560 6.74734560 3.79 0.0645
Aggregate 1 1.75261123 1.75261123 0.98 0.3321
Cement content 1 0.27679692 0.27679692 0.16 0.6973
Air content 1 6.74734560 6.74734560 3.79 0.0645
Aggregate 1 0.00236750 0.00236750 0.00 0.9713
Cement content 1 0.27679692 0.27679692 0.16 0.6973
Air content 1 6.74734560 6.74734560 3.79 0.0645
Aggregate 1 0.00236750 0.00236750 0.00 0.9713
Comparing mixture 22r and 26, 24 and 27 suggests that the addition of an
accelerating admixture did not change the corrosion performance of pipe coupons
in CLSM. Because only two pairs of mixtures were tested, this finding is only
preliminary.
5.5.5.2 Mass Loss of Coupled Coupons
This section provides results of mass loss of coupled coupons, as shown in
Table 5.45. In this testing, a galvanic cell was formed as a result of pipe coupons
219
being embedded in different environments, which were electrically connected to
complete the circuit for corrosion current. A 10-Ω resister was used to obtain
current flow between coupons.
The procedure listed in Section 5.5.5.1 was used to evaluate the
repeatability of CLSM mixtures with regard to corrosion. The results are shown in
Table 5.46. Five of six mixtures were found to be statistically repeatable. The
repeatability of mass loss of coupled samples was much higher than those of
uncoupled specimens. This was probably due to the different corrosion
mechanisms involved with galvanic coupling. The coupon embedded in CLSM
functioned as a cathode and the coupon in sand as anode. The lower variation was
evident for the within-batch and between-batch variations of results provided here.
A t-test indicated that the means of mass loss for coupled samples
embedded in sand were not significantly different for CLSM mixtures with and
without fly ash content. This was probably because current flow resulting from the
differences in potentials of the samples was governing the corrosion rates.
Similar to Section 5.5.5.1, factors possibly affecting sample corrosion were
analyzed separately for CLSM with and without fly ash. The results are shown in
Tables 5.47 and 5.48.
220
Table 5.45: Percent mass loss for coupled samples embedded in sand
Sample ID Specimen No. Average
1 2 3
1R -- -- 1.68 1.68
2 2.42 1.48 1.80 1.90
2R 2.48 1.52 1.80 1.93
3 3.70 -- 5.70 4.70
3R 1.90 1.25 1.55 1.57
4 2.35 1.45 1.85 1.88
4R 3.50 2.20 1.50 2.40
5 1.50 2.65 2.45 2.20
5R 2.65 2.00 3.58 2.74
6 1.75 2.85 4.65 3.08
7 4.60 5.95 1.50 4.02
8 2.80 -- 3.10 2.95
9 1.42 0.18 -- 0.80
11 2.70 7.60 1.60 3.97
12 1.97 1.78 2.35 2.03
13 1.25 1.75 0.10 1.03
14 2.25 1.84 1.60 1.90
15 4.20 3.60 4.00 3.93
16 2.80 8.30 12.00 7.70
16R 3.00 6.25 1.30 3.52
17 3.90 3.50 4.95 4.12
18 2.80 2.95 1.25 2.33
19 5.60 4.70 -- 5.15
29 2.10 1.60 3.44 2.38
30 1.80 1.44 1.61 1.62
22 2.90 2.70 4.25 3.28
22R 1.20 2.35 2.30 1.95
23 2.80 2.60 -- 2.70
24 11.60 2.80 5.30 6.57
25 4.00 4.65 2.00 3.55
26 3.25 2.55 2.55 2.78
27 6.00 2.65 -- 4.33
28 5.50 5.40 0.40 3.77
20 2.90 4.00 5.80 4.23
21 2.30 1.58 1.30 1.73
221
Table 5.46: Statistical comparison of repeated CLSM mixtures on corrosion
of coupled coupons embedded in sand
Population Population Estimated Critical value Accepted
# #r X1 X2 S2 statistic (t) (t), hypothesis
α=0.05
2 2r 1.90 1.93 0.24 -0.084 2.776 Null
3 3r 4.70 1.57 0.74 3.998 3.182 Alternative
4 4r 1.88 2.40 0.62 -0.806 2.776 Null
5 5r 2.20 2.74 0.50 -0.937 2.776 Null
16 16r 7.70 3.52 13.88 1.375 2.776 Null
22 22r 3.28 1.95 0.57 2.169 2.776 Null
For CLSM mixtures with fly ash, all four statistical tests indicated that
water-cement ratio had a significant effect on corrosion rates in coupled cases.
Three out of four tests showed that cement contents and aggregate type were also
important factors.
For CLSM mixtures without fly ash, no single factor or interaction was
found to significantly influence the corrosion rate of coupled specimens. Unlike
the case of uncoupled corrosion, air content did not play a key role here. Only one
test indicated that cement content affects the corrosion behavior. Further study is
needed to identify the important parameters affecting the corrosion process.
222
mixtures containing fly ash)
Cement content 1 1.83194087 1.83194087 0.11 0.7402
Fly ash type 2 22.93176577 11.46588288 0.70 0.5036
Fly ash content 1 23.10524882 23.10524882 1.41 0.2429
Aggregate 2 62.42531783 31.21265891 1.90 0.1633
Cement content 1 84.9477888 84.9477888 5.17 0.0287
Fly ash type 2 0.6473494 0.3236747 0.02 0.9805
Fly ash content 1 21.3668191 21.3668191 1.30 0.2611
Aggregate 2 162.0849627 81.0424813 4.94 0.0124
Cement content 1 84.9477888 84.9477888 5.17 0.0287
Fly ash type 2 0.6473494 0.3236747 0.02 0.9805
Fly ash content 1 21.3668191 21.3668191 1.30 0.2611
Aggregate 2 162.0849627 81.0424813 4.94 0.0124
Cement content 1 84.9477888 84.9477888 5.17 0.0287
Fly ash type 2 0.6473494 0.3236747 0.02 0.9805
Fly ash content 1 21.3668191 21.3668191 1.30 0.2611
Aggregate 2 162.0849627 81.0424813 4.94 0.0124
A comparison between mass loss of samples embedded in sand for Mixture
#29 and #30 showed that higher cement content dramatically increased the
corrosion rate of the anode in this study. There was not enough information to
study the effect of accelerating agent on corrosion rate in coupled cases.
The investigation of the influence of embedding pipe coupons in different
environments, where one environment is CLSM and the other is sand, indicates that
placing pipe in these different environments while coupled, could lead to increased
corrosion.
223
mixtures without fly ash)

Cement content 1 24.52736310 24.52736310 4.38 0.0486
Air 1 6.03657813 6.03657813 1.08 0.3108
Aggregate 1 0.17048459 0.17048459 0.03 0.8631
W-C ratio * cement 1 2.43267052 2.43267052 0.43 0.5169
Cement content 1 5.09210317 5.09210317 0.91 0.3510
Air 1 4.02414115 4.02414115 0.72 0.4060
Aggregate 1 0.45015898 0.45015898 0.08 0.7795
W-C ratio * cement 1 2.43267052 2.43267052 0.43 0.5169
Cement content 1 6.48091742 6.48091742 1.16 0.2941
Air 1 4.02414115 4.02414115 0.72 0.4060
Aggregate 1 2.62623812 2.62623812 0.47 0.5008
W-C ratio * cement 1 2.43267052 2.43267052 0.43 0.5169
Cement content 1 6.48091742 6.48091742 1.16 0.2941
Air 1 4.02414115 4.02414115 0.72 0.4060
Aggregate 1 2.62623812 2.62623812 0.47 0.5008
W-C ratio * cement 1 2.43267052 2.43267052 0.43 0.5169
5.6 LEACHING AND ENVIRONMENTAL IMPACT
The potential leaching of constituent materials, specifically heavy metals,
from CLSM was investigated. Interest in leaching and environmental issues
regarding CLSM has increased mainly because of concerns raised by the
Environmental Protection Agency (EPA) regarding the use of fly ash in
construction.26 Emphasis in this study was placed on characterizing the potential
224
toxicity of the by-product materials used in this study, including bottom ash,
foundry sand, and fly ash (Class C, Class F, and high carbon).
The chemical analysis, heavy metal test, and TCLP test on raw materials
were performed by Wyoming Analytical Laboratories, INC., Golden, Colorado. As
previously discussed in the Test Methods and Procedures section of this
dissertation, a systematic approach (shown earlier in Figure 4.13) was taken to first
characterize the chemical composition of the by-product materials used in the
study, then to quantify the total heavy metals present in these materials, and lastly
to measure the heavy metals that actually are able to leach from these materials.
The chemical compositions of the fly ashes and foundry sand were already reported
in Table 3.3 and Table 3.4. This information is particularly important for fly ashes,
where the chemical composition (especially CaO and SiO 2 content) can be used to
estimate pozzolanic and cementitious reactivity, especially as it relates to long-term
strength gain and excavatability.
Table 5.49 summarizes the total concentration of heavy metals present in
the raw by-product materials used in this study. These results represent the total
concentration of the eight key heavy metals. A “rule of thumb” that some
practitioners use is that the concentration of total heavy metals can be up to 20
times the standard TCLP limits.26 According to this guideline, bottom ash, Class C
fly ash, and Class F fly ash exceeded the “rule of thumb” value for arsenic. Thus,
additional testing was performed (using the TCLP method) to determine the actual
amount of heavy metals that are available to leach from these materials. Because
the foundry sand and high-carbon fly ash did not have significant amounts of total
225
heavy metals, the materials were classified as non-toxic, and no subsequent
leaching tests were performed.
Table 5.49: Analysis of heavy metal concentration of extract from raw

materials (mg/L)
Element TCLP 20 times Bottom Foundry Class C Class F High –carbon

limits TCLP limits ash sand fly ash fly ash fly ash
Arsenic 5.0 100 170* 7.7 280* 160* 58
Barium 100.0 2000 2000 240 1300 320 1200
Cadmium 1.0 20 0.23 0.28 1.55 2.1 0.51
Chromium 5.0 100 10 18 87 96 16
Lead 5.0 100 <0.2 18 <0.2 37 <0.2
Mercury 0.2 4 <0.2 <0.2 <0.2 <0.2 <0.2
Selenium 1.0 20 <0.2 <0.2 <0.2 2.4 <0.2
Silver 5.0 100 <0.2 <0.2 <0.2 <0.2 <0.2
* concentration exceeded the “rule of thumb” value of 20X the TCLP limits
The TCLP results for Class C fly ash, Class F fly ash, and bottom ash (raw
materials) are shown in Table 5.50. The concentration of heavy metals that leached
from each material was well below the EPA-recommended TCLP limits, and as
such, the materials were classified as non-toxic and suitable for use in CLSM. If
any of the by-product materials had exhibited significant leaching of heavy metals
(above the TCLP limits), the last step would have been to assess the actual leaching
of heavy metals from CLSM containing the material(s). This systematic approach
can be followed for any material being considered for use in CLSM. Although all
the materials used in this study were deemed non-toxic, it may be possible that
certain materials considered for a given CLSM application may be more of an
environmental concern.
226
Table 5. 50: TCLP test results for Class C fly ash, Class F fly ash, and bottom
ash (mg/L)
Element TCLP limits Bottom ash Class C fly ash Class F fly ash
Arsenic 5.0 0.12 0.074 0.37
Barium 100.0 3.61 0.30 0.17
Cadmium 1.0 0.001 0.004 0.024
Chromium 5.0 0.01 0.29 0.11
Lead 5.0 <0.01 <0.01 <0.01
Mercury 0.2 <0.01 <0.01 0.11
Selenium 1.0 <0.01 0.37 0.02
Silver 5.0 <0.01 <0.01 <0.01
227
Chapter 6: Conclusions and Future Work
6.1 CONCLUSIONS
This study focused on fresh, hardened, and durability characteristics of
CLSM mixtures. A wide range of CLSM mixtures were included in this research
to address the variability of raw materials incorporated in CLSM. Waste
materials, such as fly ash, bottom ash, and foundry sand, were used. Suitable
existing test methods were examined and modified in this study to compensate for
the lack of CLSM test methods. Existing geotechnical test methods, such as
triaxial shear and resilient modulus, were included in the research to assess their
applicatability to CLSM. Based on this comprehensive study, the following
conclusions can be drawn.
Fresh properties of CLSM
1. Source of aggregates i.e. foundry sand, bottom ash, and river sand is
the most significant factor affecting bleeding and water demand of
CLSM mixtures to obtain a target flow;
2. The foundry sand and bottom ash used in this study led to difficulties
in entraining large volumes of air in CLSM;
3. With slight modifications described in Chapter 4, ASTM C 231 is a
good approach to measure the air contents and unit weights of fresh
CLSM mixtures;
228
4. Both the needle (ASTM C 403) and pocket soil penetrometer can be
utilized to characterize the setting and hardening of CLSM mixtures.
Precautions should be taken to interpret the results from the two
methods because the penetration depths and underlying mechanisms
are different;
5. The initial and final setting concepts, used in concrete technology, may
not be appropriate for CLSM.
Hardened properties of CLSM
6. For CLSM mixtures with high air content (without fly ash), the water-
cement ratio is the main factor affecting compressive strength;
7. For CLSM mixtures containing fly ash, a predictive model was
developed to estimate the development of compressive strength. Not
only the water-cement ratio, but also the fly ash type, fly ash content,
aggregate type influences the CLSM strength;
8. Variations in load rate can affect the measured values of compressive
strength of CLSM specimens. A loading rate between 0.042 to 0.16 %
per minute is recommended for compression tests to achieve adequate
accuracy and to accomplish testing in a reasonable time;
9. Testing machines with smaller, more accurate load cells yield more
repeatable compression test results than conventional concrete
compression machines with larger capacities;
229
10. The air drying of CLSM specimens for 4 to 8 hours, as specified by
ASTM D 4832, was found to be unnecessary in this study. Thus, this
waiting period is not recommended as part of standard compression
testing;
11. Leaching of hydration products from CLSM specimens (i.e., during
wet-curing) may lead to variations in strength. Direct contact between
standing water and CLSM should be avoided, especially at early ages.
Dripping water could damage CLSM specimens;
12. Curing temperature and humidity can significantly affect the strength
development of CLSM mixtures with fly ash. The compressive
strength values can be an order of magnitude different with different
curing conditions;
13. The inclusion of Class C fly ash in CLSM mixtures typically yields
higher compressive strengths. If future excavation is specified, the use
of Class C fly ash should be investigated prior to use. Local
experience serves as an excellent tool for mixture proportioning and
obtaining the desired short- and long-term strength;
14. CLSM curing appears to be significantly influenced by surrounding
temperature and humidity rather than drainage conditions. This
provided very critical information to people specifying compressive
strength of CLSM mixtures based on a performance-based
specification. The curing temperature and humidity are so important,
230
that appropriate conditions should be clearly specified for the mixtures
proposed by the contractors;
15. High-strength gypsum is an appropriate alternative capping material to
sulfur. However, the capping procedure may take more than half an
hour per sample;
16. Neoprene pads with hardness durometer less than 50 were found
suitable for testing the compressive strength of CLSM, provided that
the strength reduction (compared to sulfur capping) less than 20% is
accepted as specified in ASTM D 4832;
17. Removability Modulus can be utilized to predict the ease of future
excavation of CLSM mixtures in the field. This approach was found
to be conservative in this study;
18. The dynamic cone penetrometer (DCP) is a viable method to evaluate
the in-situ strength of CLSM mixtures. The DCP index correlates well
with the Removability Modulus (RE) if the strength of field-cured
samples are utilized to calculate RE;
19. No acceptable correlations between compressive strength and needle
penetration resistance (ASTM C 403) were found;
20. Stiffness (measured by Stiffness Gauge) of CLSM mixtures did not
correlate well with penetration index from DCP;
21. Measuring the California Bearing Ratio (CBR) and Resilient Modulus
of CLSM mixtures was found to be feasible;
231
22. The water permeability of CLSM mixtures were found to be in the
range of silts;
23. When testing water permeability according to ASTM D 5084-90, to
check the saturation of samples, the requirement that B > 0.95 can be
waived. Instead, the change of B values (approaching a constant) with
change of confining pressure should be checked;
24. It is feasible to measure the triaxial properties of CLSM mixtures
using conventional geotechnical equipment;
25. Beyond the first day, the drying shrinkage of CLSM mixtures may be
very small (according to the limited testing performed in this study).
Durability Characteristics of CLSM
26. For CLSM specimens cured in the fog room in this study, the pH
values of extruded solutions change with time and mixture
proportions. The pH values ranged from 8.63 to 13.03 in this study;
27. The resistivity of CLSM mixtures after 182 days ranged from 1781 to
12573 Ω-cm for CLSM specimens cured in the fog room;
28. Mass loss of ductile-iron specimens embedded in CLSM in an
uncoupled conditions demonstrated large variations within mixtures
and between mixtures;
29. For the uncoupled conditions, the corrosion rates of specimens in
CLSM mixtures with fly ash might be affected by the interaction
between cement content and fly ash type;
232
30. For the uncoupled conditions, no single factor was found to
significantly influence the corrosion rates of specimens in CLSM
mixtures without fly ash;
31. The mass loss of ductile-iron specimens embedded in sand and in
CLSM in a coupled condition showed less variation than those in an
uncoupled condition;
32. For the coupled conditions, corrosion rates of specimens in sand with
CLSM containing fly ash were affected by the water-cement ratio and
possibly the aggregate source;
33. For the coupled conditions, no factor was found to significantly
influence the corrosion rates of specimens in sand when CLSM
mixtures did not contain fly ash;
34. To survive twelve freezing-and-thawing cycles, CLSM mixtures
required both high air contents and adequate strengths;
35. CLSM mixtures can typically survive wetting-and-drying cycles, even
if cured only for seven days, provided that the samples are allowed to
air dry;
6.2 FUTURE WORK
Although this research focused on many important issues, additional
research is needed to further advance the state-of-the-art regarding CLSM. Some
issues that deserve further exploration include:
233
1. Evaluate the feasibility of ASTM D 6023-96, “Standard Test Method
for Unit Weight, Yield, Cement Content, and Air Content
(Gravimetric) of Controlled Low-Strength Material (CLSM)” ;
2. Optimize the mixture proportions of CLSM containing bottom ash as
aggregate to obtain desired flowability without excessive bleeding;
3. Identify a suitable method to characterize aggregate properties that
may affect the flowability of CLSM;
4. Study the hydration products formed in CLSM mixtures because their
high water-cement ratios significantly exceed normal ranges for
conventional concrete. Also, the ratio of fly ash to portland cement is
much higher than that for concrete.
5. Research is needed to study the hydration of fly ash, especially Class
C fly ash. Focus should be placed on the effects of temperature and
humidity;
6. More research is needed to evaluate the use of CLSM as a pavement
subbase and base;
7. Additional work is required to evaluate the excavatability of CLSM in
field applications and examine the effects of surrounding environments
and local climate conditions;
8. The mechanisms of corrosion of metals in CLSM need to be studied in
more detail. Potential influencing factors should be identified. If
possible, corrosion prediction criteria similar to those for soil should
be established to assist in design and construction;
234
9. More severe wetting-and-drying tests should be performed on CLSM
to evaluate its durability.
235
Appendix A: Referenced Standard Test Methods
ASTM Standard Test Methods:
ASTM C 33-1999, “Standard Specification for Concrete Aggregate”
ASTM C 39-1996, “Standard Test Method for Compressive Strength of

Cylindrical Concrete Specimens”
ASTM C 128-1997, “Standard Test Method for Specific Gravity and Absorption
of Fine Aggregate”
ASTM C 136-1996a, “Standard Test Method for Sieve Analysis of Fine and
Coarse Aggregates”
ASTM C 143-1998, “Standard Test Method for Slump of Hydraulic-Cement

Concrete”
ASTM C 231-1997, “Standard Test Method for Air Concrete of Freshly Mixed
Concrete by the Pressure Method”
ASTM C 311-1998b, “Standard Test Method for Sampling and Testing Fly Ash
or Natural Pozzolans for Use as a Mineral Admixture in Portland-Cement
Concrete”
ASTM C 403-1997, “Standard Test Method for Time of Setting of Concrete

Mixtures by Penetration Resistance”
ASTM C 494-1998a, “Standard Specification for Chemical Admixtures for

Concrete”
ASTM C 496-1996, “Standard Test Method for Splitting Tensile Strength of

Cylindrical Concrete Specimens”
ASTM C 618-1998, “Standard Specification for Coal Fly Ash and Raw or
Calcined Natural Pozzolan for Use as a Mineral Admixture in Concrete”
236
ASTM C 939-1997, “Flow of Grout for Preplaced-Aggregate Concrete (Flow
Cone Method)”
ASTM C 940-1998a, “Standard Test Method for Expansion and Bleeding of

Freshly Mixed Grouts for Pre- Placed Aggregate Concrete in the Laboratory”
ASTM C 1231-1999, “Standard Practice for Use of Unbonded Caps in

Determination of Compressive Strength of Hardened Concrete Cylinders”
ASTM D 559-1996, “Standard Test Methods for Wetting and Drying Compacted
Soil-Cement Mixtures”
ASTM D 560-1996, “Standard Test Methods for Freezing and Thawing of

Compacted Soil-Cement Mixtures”
ASTM D 2435-1996, “Standard Test Method for One-Dimensional Consolidation

Properties of Soil”
ASTM D 4429-1993, “CBR (California Bearing Ratio) of Soils in Place”
ASTM D 4525-1990, “Standard Test Method for Permeability of Rocks by

Flowing Air”
ASTM D 4612-1986, “Standard Practice for Calculating Thermal Diffusivity of

Rocks”
ASTM D 4832-1995, “Preparation and Testing of Controlled Low-Strength

Material (CLSM) Test Cylinders”
ASTM D 5084-1990, “Measurement of Hydraulic Conductivity of Saturated

Porous Materials using a Flexible Wall Permeameter”
ASTM D 5334-1992, “Standard Test Method for Determination of Thermal

Conductivity of Soil and Soft Rock by Thermal Needle Probe Procedure”
ASTM D 5335-1992, “Standard Test Method for Linear Coefficient of Thermal

Expansion of Rock Using Bonded Electrical Resistance Strain Gages”
ASTM D 5971-1996, “Standard Practice for Sampling Freshly Mixed Controlled

Low-Strength Material”
237
ASTM D 6023-1996, “Standard Test Method for Unit Weight, Yield, and Air
Content (Gravimetric) of Controlled Low Strength Material”
ASTM D 6024-1996, “Test Method for Ball Drop on Controlled Low Strength
Material to Determine Suitability for Load Application”
ASTM D 6103-1997, “Standard Test Method for Flow Consistency of Controlled

Low-Strength Material”
ASTM G 1-1995, “Standard Practice for Preparing, Cleaning, and Evaluating

Corrosion Test Specimens”
ASTM G 51-1999, “Standard Test Method for Measuring pH of Soil for Use in
Corrosion Testing”
ASTM G 57-1995, “Standard Test Method for Field Measurement of Soil

Resistivity Using the Wenner Four-Electrode Method”
ASTM G 59-1997, “Standard Practice for Conducting Potentiodynamic

Polarization Resistance Measurements”
ASTM G 109-1999, “Standard Test Method for Determining the Effects of

Chemical Admixtures on the Corrosion of Embedded Steel in Concrete Exposed
to Chloride Environments”
AASHTO Standard Test Methods:
AASHTO T 22-1997, “Compressive Strength of Cylindrical Concrete

Specimens”
AASHTO T 24-1997 “Obtaining and Testing Drilled Cores and Sawed Beams of
Concrete”
AASHTO T 106-2000, “Compressive Strength of Hydraulic Cement Mortar”
AASHTO T 121-1997, “Unit Weight, Yield, and Air Content (Gravimetric) of

Concrete”
AASHTO T 160-1997, “Shrinkage of Portland Cement Concrete”
238
AASHTO T 161-2000, “Standard Test Method for Resistance of Concrete to
Freezing and Thawing”
AASHTO T 193-1999, “Standard Method of Test for The California Bearing

Ratio”
AASHTO T 197-2000, “Time of Setting of Concrete Mixtures by Penetration

Resistance”
AASHTO T 222-1990, “Nonrepetitive Static Plate Load Tests of Soils and

Flexible Pavement Components for Use in Evaluation and Design of Airport and
Highway Pavements”
AASHTO T 292-1991, “Standard Method of Test for Resilient Modulus of

Subgrade Soils and Untreated Base/Subbase Materials”
AASHTO T 276-1997, “Developing Early Age Compression Test Values and

Projecting Later Age Strengths"
AASHTO T 277-1996, “Rapid Determination of the Chloride Permeability of

Concrete”
Corp of Engineers Test Methods:
CRD-C611-1980, “Method of Test for Flow of Grout Mixtures (flow cone

methods)”
239
Appendix B: Setting/Hardening Test Results
Table B-1: Needle penetration (NP) and soil penetrometer (SP) results
Mix # 1 Mix # 2 Mix # 1r Mix # 9

Time NP SP Time NP SP Time NP SP Time NP SP
(hrs) (MPA) (MPA) (hrs) (MPA) (MPA) (hrs) (MPA) (MPA) (hrs) (MPA) (MPA)
1.5 0.00 2 2 0.00 3.33 0.01
2 0.01 3 3 0.00 4.25 0.11 0.01
2.5 0.01 4 4 0.50 0.02 6.17 0.22 0.02
3 0.03 5 5 1.10 0.04 7 0.25 0.02
3.5 0.07 6.17 0.58 0.02 6 1.76 0.08 8 0.28 0.02
4 1.21 0.08 6.67 0.76 0.01 7 2.10 0.12 22.33 0.74 0.12
4.5 1.43 0.37 7.67 - 0.03 8 2.43 0.15 26.17 0.83 0.12
5 1.43 0.17 8.17 0.72 0.04 9 2.32 0.17
5.5 2.07 0.19 20 4.91 0.07 24 2.89 0.27
6 2.32 0.25 24 3.20
8.25 2.32 0.27
24 4.69 0.46
Mix # 15 Mix # 4 Mix # 5 Mix # 3
2 0.00 0.44 0.05 3 1.24 0.01 4 0.44 0.01
3 0.01 4.25 0.58 0.05 4.67 1.32 0.02 5 0.74 0.04
4 0.17 0.02 5.58 0.69 0.09 5.83 1.49 0.02 6 0.61 0.04
5 0.11 0.03 6.33 0.77 0.08 7.67 1.76 0.03 7.5 1.24 0.04
6 0.28 0.04 7 0.94 0.10 22.33 2.70 0.20 8.5 1.41 0.05
7 0.33 0.05 9 1.49 0.12 24.33 2.78 0.26 11 2.37 0.13
8 0.33 0.06 24 3.86 - 12 2.78 0.15
9 0.33 0.06 24 8.41
10 0.39 0.07
24 0.39 0.09
Mix # 8 Mix # 10 Mix # 12 Mix #7
3 0.06 3 1.38 0.08 3 1.02 0.03 3.5 0.03 0.00
4.22 0.08 0.09 4 1.46 0.07 4.33 0.96 0.03 4.33 0.08 0.01
5.17 0.11 0.01 5.33 1.65 0.05 5.5 1.10 0.07 6.73 0.22 0.02
6.5 0.14 0.02 6.83 1.79 0.11 6.67 1.43 0.09 8 0.25 0.02
7.5 0.22 0.02 8.5 2.45 0.15 8 1.93 0.10 9.33 0.28 0.02
8.5 0.28 0.03 24 3.03 0.30 24 8.06 - 24 0.63 0.1
24.67 0.94 0.13
240
Table B-1: Needle penetrAtion (NP) and soil penetrometer (SP) results
(cont’d)
Mix #3r Mix # 4r Mix # 24 Mix # 17

3.00 0.55 0.01 3.67 0.86 0.02 3.40 0.08 0.01 3.00 0.28 0.02
5.00 1.32 0.03 5.00 0.77 0.03 4.67 0.33 0.01 4.33 0.36 0.05
7.00 1.43 0.05 6.00 1.41 0.04 5.50 0.30 0.01 5.33 0.39 0.05
8.50 1.49 0.07 8.25 1.86 0.08 6.50 0.50 0.02 9.50 0.66 0.08
10.50 2.43 0.11 9.33 1.88 0.11 8.00 0.61 0.02 23.00 0.83 0.16
24.00 8.96 - 26.00 5.41 0.42 24.00 2.87 0.18 26.50 0.91 0.21
Mix # 23 Mix # 18 Mix # 14 Mix # 11
5.00 0.08 0.00 4.50 0.08 0.01 2.50 0.30 0.02 3.00 0.41 0.02
6.50 0.14 0.01 5.67 0.22 0.01 3.50 0.50 0.03 4.17 0.58 0.02
8.50 0.33 0.01 6.67 0.30 0.02 4.50 0.66 0.05 5.25 1.10 0.03
22.00 0.61 0.28 8.00 0.47 0.03 5.50 0.99 0.06 6.17 1.43 0.08
24.00 0.88 0.05 14.00 1.38 0.11 8.50 3.03 0.19 7.67 2.07 0.08
28.00 0.77 0.23 31.00 1.60 0.16 26.00 14.76 - 24.00 5.02 0.37
Mix # 2r Mix # 29 Mix # 30 Mix # 6
3.00 0.17 0.01 2.50 0.25 0.02 3.00 0.22 0.02 3.00 0.00 0.00
4.33 0.41 0.02 3.50 0.44 0.05 4.00 0.36 0.03 4.33 0.06 0.01
5.67 1.02 0.10 4.00 0.61 0.08 5.00 0.44 0.04 6.33 0.55 0.02
7.17 3.39 0.30 4.50 0.72 0.08 6.00 0.47 0.05 8.33 0.58 0.03
8.50 5.24 - 7.00 1.27 0.17 9.67 0.74 0.09 9.83 0.80 0.03
22.67 14.76 24.00 2.98 0.34 24.00 1.19 0.12 12.50 1.16 0.05
24.17 2.95 0.20
241
Table B-1: Needle penetration (NP) and soil penetrometer (SP) results
(cont’d)
Mix # 16 Mix # 21 Mix # 22 Mix # 22r

3.5 0.39 0.03 3 3.66 4.83 0.30 0.01
4.67 0.30 0.03 4.33 0 0.00 5 0.25 0.01 6.5 0.72 0.05
7.33 0.50 0.03 5.33 0.03 0.01 6 0.39 0.01 7.67 0.94 0.10
9.67 0.80 0.04 8.33 0.22 0.01 7 0.58 0.03 9.3 1.49 0.15
10.17 1.05 0.06 11 0.39 0.02 9 0.91 0.03 15 1.93 0.21
24.25 1.10 0.08 21.5 0.61 0.04 11 1.05 0.06 24.56 2.34 0.28
26.83 0.61 0.04 25.33 1.21 0.17
Mix # 5r Mix # 26 Mix # 16r Mix # 13
3 0.83 0.03 4 0.41 0.02 2.5 0 0 4.5 0.83 0.12
4 1.02 0.08 5 0.44 0.03 3.75 0.11 0.01 6 1.05 0.15
5 1.05 0.14 6.17 0.91 0.05 5.25 0.11 0.02 7.6 1.10 0.17
7 1.71 0.15 8.17 1.16 0.11 7 0.25 0.03 9 1.27 0.20
9 2.15 0.17 9.17 1.21 0.14 9 0.41 0.03 10.33 1.43 0.22
24.33 4.41 11.17 1.31 0.17 18.67 0.66 0.08 21 2.41 0.31
24.17 2.07 0.25 22.21 0.66 0.11 24.5 2.90 0.37
Mix # 25 Mix # 19 Mix # 20 Mix # 27
3 3.33 0.47 0.04 3 0.55 0.03 2 0 0.00
4 0.01 4.33 0.47 0.04 5 0.61 0.06 3 0.11 0.01
5.33 0.03 0.01 5.67 0.55 0.04 7.5 0.91 0.14 4 0.25 0.02
7.33 0.19 0.02 7.17 0.77 0.06 9.5 1.08 0.16 5 0.30 0.02
8.83 0.25 0.02 9.33 0.72 0.07 19.67 1.86 0.22 8 0.50 0.03
10.67 0.36 0.03 25 0.97 0.14 24 1.86 0.30 21.67 2.12 0.21
18.33 0.83 0.06 24 2.34 0.23
Mix # 20r Mix # 28
Time NP SP Time NP SP
(hrs) (MPA) (MPA) (hrs) (MPA) (MPA)
3 0.25 0.02 2 0.25 0.03
5 0.41 0.04 3 0.52 0.04
7 0.61 0.05 4 1.05 0.10
10.67 0.69 0.08 5 1.46 0.11
23.5 1.41 0.27 6 1.85 0.11
25.83 1.45 0.31 9.17 - 0.16
11.33 - 0.24
24.33 6.21 0.27
242
Table B-2: Needle penetration (NP), soil penetrometer (SP) and vane shear
tester (VS) Results
Mix E-1 Mix E-2

Time NP SP VS Time NP SP VS
(MPa) (MPa) (MPa) (MPa) (MPa) (MPa)
3.18 0.23 0.07 0.10 0.80 0.14 0.02 0.13
5.22 0.41 0.16 0.18 3.33 0.28 0.12 0.15
7.18 1.45 0.18 0.29 5.97 0.77 0.12 0.15
11.02 2.13 0.35 0.43 9.80 0.83 0.12 0.22
22.40 3.26 0.48 21.18 1.21 0.17 0.25
Mix E-3 Mix E-4
6.22 0.00 0.00 0.04 2.28 0.50 0.02 0.00
8.27 0.14 0.01 0.05 3.87 0.66 0.03 0.05
10.22 0.17 0.01 0.05 5.95 1.08 0.07 0.08
20.71 0.22 0.10 0.11 7.95 1.24 0.17 0.11
24.02 0.33 0.27 0.06 18.28 2.48 0.31 0.16
21.62 2.04 0.27 0.24
Mix E-5 Mix E-6
3.95 0.34 3.25 0.14 0.02
6.59 0.86 0.04 0.04 5.90 0.36 0.01 0.01
9.50 1.71 0.08 0.10 7.73 0.79 0.03 0.02
11.50 1.71 0.11 0.11 10.82 0.83 0.04 0.03
22.08 2.43 0.18 0.16 12.73 1.10 0.07 0.04
26.17 2.70 0.23 0.22 23.35 2.07 0.15 0.08
Mix E-7 Mix E-8
2.98 0 0.02 0.01 2.98 0.72 0.03 0.03
5.45 0.50 0.03 0.04 5.45 1.55 0.12 0.10
7.31 0.75 0.03 0.04 7.31 2.13 0.22 0.10
11.52 1.10 0.08 0.08 11.52 2.18 0.24 0.10
13.49 1.57 0.09 0.09 13.49 2.40 0.24 0.11
22.82 0.73 0.27 0.11 22.82 0.62 0.36 0.21
21:37 2.04 0.27 0.24
243
References
1. American Concrete Institute, Committee 229, Controlled Low-Strength
Materials (CLSM), ACI 229R-94 Report, 1994.
2. Adaska, W. S., “Controlled Low Strength Materials,” Concrete International,
Vol. 19, No. 4, 1997.
3. Funston, J. J., Krell, W. C. and Zimmer, F. V., “Flowable Fly Ash, A New
Cement Stabilized Backfill,” Civil Engineering, ASCE, 1984.
4. Krell, W. C., “Flowable Fly Ash,” Transportation Research Record, No.
1234, 1989.
5. Krell, W. C., “Flowable Fly Ash,” Concrete International, Vol. 15, No. 7,
1989.
6. Larsen, R. L., “Sound Use of CLSMs in the Environment,” Concrete
International, Vol. 15, No. 7, 1993.
7. Larsen, R. L., “Use of Controlled Low Strength Materials in Iowa,” Concrete
International, Vol. 10, No. 7, 1988.
8. ASTM, The Design and Application of Controlled Low-Strength Materials
(Flowable Fill), Howard, A. K. and Hitch, J. L., eds, STP 1331, 1998.
9. Adesanya, S. A., Analysis of Controlled Low Strength Backfill Material for
Utility Trenches, M. E. Thesis, University of Louisville, 1988.
10. Bhat, S. T., Use of Coal Combustion Residues and Foundry Sands in
Flowable Fill, Ph. D. Dissertation, Purdue University, 1996.
244
11. Burns, F., Flowable Fly Ash Backfill for Buried Pipelines, M. S. Thesis,
University of North Carolina at Charlotte, 1990.
12. Hagerty, J. L., Investigation and Testing of Controlled Low Strength Material
for Use in the Restoration of Street Cuts, M. S. Thesis, University of
Louisville, 1989.
13. Lucht, D. A., Thermal Performance of Flowable Fill Mixtures for Horizontal
GSHP System, M. S. Thesis, South Dakota State University, 1995.
14. McElroy, N. H., Testing of Controlled Low Strength Material Used as a
Utility Cut Backfill, M. S. Thesis, University of Louisville, 1993.
15. Mididaddi, V. G., Evaluation of Utility Cut Restoration Using Unshrinkable
Backfill, M. S. Thesis, Arizona State University. 1990.
16. Monson, T., Engineering Properties of Controlled Low-Strength Material
Using Fly Ash/AMD Sludge, M. S. Thesis, West Virginia University, 1997.
17. Nantung, T. E., Design Criteria for Controlled Low Strength Materials, Ph. D.
Dissertation, Purdue University, 1993.
18. Okwuegbu, A., Recycle of Industrial By-Products in Concrete Construction,
Ph. D. Dissertation, Michigan State University, 1995.
19. Wakim, A. B., Strength and Durability Characteristics of Flowable Fill, M. S.
Thesis, University of Tennessee, Knoxville, 1993.
20. Brewer, W. E., “The Design and Construction of Small Span Bridges and
Culverts Using Controlled Low Strength Material (CLSM),” Columbus, OH,
Ohio Department of Transportation, 1992.
245
21. Lundvall, J. F., “Mitigation of Roadway Settlement above Buried Culvert and
Pipes,” FHWA/WA-97/01, Cheyenne, WY, Wyoming Department of
Transportation, 1997.
22. Rojas-Gonzalez, L. F., Knott, D. L. and Newman, F. B., “Demonstration
Applications in the Use of Flowable Backfills for Bridge Abutments,”
Research Projects 90-12, Gai-Consultants, Inc., 1993.
23. AASHTO, “A Look at What Some States Are Doing,” AASHTO Quarterly,
Vol. 64, No. 4, 1985.
24. Hoopes, R. J., “Engineering Properties of Air-Modified Controlled Low
Strength,” The Design and Application of Controlled Low-Strength Materials
(Flowable Fill), ASTM STP 1331, West Conshohocken, PA, 1998.
25. Folliard, K. J.,Trejo, D., Du, L., and Sabol, S. A., “Controlled Low-Strength
Material for Backfill, Utility Bedding, Void Fill, and Bridge Approaches,”
NCHRP 24-12 Interim report, 1999.
26. Folliard, K. J.,Trejo, D., Du, L., and Sabol, S. A., “Controlled Low-Strength
Material for Backfill, Utility Bedding, Void Fill, and Bridge Approaches,”
NCHRP 24-12(1) Interim report, 2001.
27. Ayers, P. H., Charlton, C. B. and Frishette, C. W., “Investigation of Flash
Fill® as a Thermal Backfill Material,” Midwest Power Conference,
Proceedings of the Midwest Power Conf., Vol. 57//V1, 1995.
28. Fox, T. A., “Use of Coarse Aggregate in Controlled Low Strength Materials,”
Transportation Research Record, No. 1234, 1989.
246
29. Bhat, S. T. and Lovell, C. W., “Design of Flowable Fill: Waste Foundry Sand
as a Fine Aggregate,” Transportation Research Record, No. 1546, 1996.
30. Bhat, S. T. and Lovell, C. W., “Mix Design for Flowable Fill,” Transportation
Research Record, No. 1589, 1997.
31. Bhat, S. T. and Lovell, C. W., “Flowable Fill Using Waste Sand: a Substitute
for Compacted or Stabilized Soil,” Testing Soil Mixed with Waste or Recycled
Materials, ASTM STP 1275, West Conshohocken, PA, 1997.
32. Stern, K., “The Use of Spent Foundry Sand in Flowable Fill in Ohio,”
Foundry Management & Technology, Vol. 123, No.9, 1995.
33. Kennedy, D. O. and Line, C. L. 1987. “Environmental and Economical
Aspects of Sand Reclamation System,” EPRI, Vol. 2, Palo Alato, CA
34. Federal Highway Administration (FHWA), Chesner, W. H., “User Guidelines
for Waste and By-Product Materials in Pavement Construction,” Turner-
Fairbank Highway Research Center, 1998.
35. Environmental Protection Agency (EPA), “Back Document for Proposed CPG
III and Draft RMAN III,” EPA Report EPA530-R-98-003, 1998.
36. Crouch, L. K., Gamble, R. Brogdon, J. F. and Tucker, C. J., “Use of High-
Fines Limestone Screenings as Aggregate for Controlled Low-Strength
Material (CLSM),” Testing Soil Mixed with Waste or Recycled Materials,
ASTM STP 1275, West Conshohocken, PA, 1998.
37. Crouch, L. K. and Gamble, R., “Putting More Fines in Flowable Fill,” Rock
Products, Vol. 100, No. 7, 1997.
247
38. Karim, A. K., Salgado, R. and Lovell, C. W., “Compaction of Fly and Bottom
Ash Mixtures,” 51st Purdue Industrial Waste Conference Proceedings, 1996.
39. Naik, T. R., Kraus, R. N., Sturzl, R. F. and Ramme, B. W., “Design and
Testing Controlled Low Strength Materials (CLSM) Using Coal Ash,” Testing
Soil Mixed with Waste or Recycled Materials, ASTM STP 1275, West
Conshohocken, PA, 1998.
40. Ohlheiser, T. R., “Utilization of Recycled Glass as Aggregate in Controlled
Low-Strength Material (CLSM),” Testing Soil Mixed with Waste or Recycled
Materials, ASTM STP 1275, West Conshohocken, PA, 1998.
41. Gandham, S., Seals, R. K. and Foxworthy, P. T., “Phosphogypsum as a
Component of Flowable Fill,” Transportation Research Record, No. 1546,
1996.
42. Federal Highway Administration (FHWA), “Fly Ash Facts for Highway
Engineers,” U. S. Department of Transportation, FHWA-SA-94-081, 1995.
43. McGrath, T. J. and Hoopes, R. J., “Bedding Factors and E Values for Buried
Pipe Installations Backfilled with Air-Modified CLSM,” The Design and
Application of Controlled Low-Strength Materials (Flowable Fill), ASTM
STP 1331, West Conshohocken, PA, 1998.
44. Nmai, C. K., McNeal, F. and Martin, D., “New Foaming Agent for CLSM
Application,” Concrete International, Vol. 19, No. 4, 1997.
45. Janardhanam, R., Burns, F. and Peindl, R. D., “Mix Design for Flowable Fly-
Ash Backfill Material,” Journal of Materials in Civil Engineering, Vol. 4, No.
3,1992.
248
46. Ayers, M. E., Wong, S. Z. and Zaman, M., “Optimization of Flowable Fill
Mix Proportions,” Controlled Low-Strength Materials, ACI SP 150, 1994.
47. Fesenmaier, T., “CLSM for New England Thruway,” Concrete International,
Vol. 10, No. 7,1988.
48. Brewer, W. E., “Durability Factors Affecting CLSM,” Controlled Low-
Strength Materials, ACI SP150, 1994.
49. Mulllarky, J. I., “Long Term Strength Gain of Controlled Low-Strength
Materials,” The Design and Application of Controlled Low-Strength Materials
50. Washington Aggregates and Concrete Association, “Suggested Specifications
for Controlled Density Fill,” Seattle, Washington,1992.
51. Dandria, G. G., Frost, D. J., Ashmawy, A. and Patterson, K. R., “Potential
Factors Affecting Flow consistency Test Method for Controlled Low Strength
Materials,” Transportation Research Record, No. 1589, 1997.
52. Gray, D. D., “Filling Abandoned Mines with Fluidized Bed Combustion Ash
Grout,” The Design and Application of Controlled Low-Strength Materials
53. Hepworth, H. K., Davidson, J. S. and Hooyman, J. L., “Admixture Enhanced
Controlled Low-Strength Material for Direct Underwater Injection with
Minimal Cross-Contamination,” The Design and Application of Controlled
Low-Strength Materials (Flowable Fill), ASTM STP 1331, West
Conshohocken, PA, 1998
249
54. Janardhanam, R., Burns, F. and Peindl, R. D., “Mix Design for Flowable Fly-
Ash Backfill Material,” Journal of Materials in Civil Engineering, Vol. 4, No.
2, 1992.
55. Smith, A., “Controlled Low-Strength Material,” Concrete Construction, May
1991.
56. McLaren, R. J. and Balsamo, N. J, Fly Ash Design Manual for Road and Site
Applications, Volume 2, Slurried Placement, Electric Power Research
Institute, 1986.
57. Bernard, R. D. and Tansley, R. S., “Laboratory Testing Program for
Development of a Lean Mix Backfill Specification,” PB83-122952,
Department of Housing and Urban Development, Washington, D. C. , 1981.
58. Ayers, M. E., Wong, S. Z. and Zaman, W., “Optimization of Flowable Fill
Mix Proportions,” Controlled Low-Strength Materials, ACI SP-150, 1994.
59. Dolen, T. P., “Properties of Low-Strength Concrete for Meeks Cabin Dam
Modification Project,” The Design and Application of Controlled Low-
Strength Materials (Flowable Fill), ASTM STP 1331, West Conshohocken,
PA, American Society for Testing Materials, 1998.
60. Abelleria, A., Berke, N. S. and Pickering, D. G., “Corrosion Activity of Steel
in Cementitious Controlled Low-Strength Materials vs. That in Soil,” The
Design and Application of Controlled Low-Strength Materials (Flowable
Fill), ASTM STP 1331, West Conshohocken, PA, American Society for
Testing Materials, 1998.
250
61. Ramme, B. W., Naik, T. R. and Kolbeck, H. J., “Use of Fly Ash Slurry for
Underground Facility Construction,” Construction and Building Materials,
Vol. 8, No. 1, pp. 63-67,1994.
62. Landwermeyer, J. S. and Rice, E. K., “Comparing Quick Set and Regular
CLSM,” Concrete International, Vol. 19, No.5, 1997.
63. Pons, F., Landwermeyer, J. S. and Kerns, L., “Development of Engineering
Properties for Regular and Quick-Set Flowable Fill,” The Design and
Application of Controlled Low-Strength Materials (Flowable Fill), ASTM
STP 1331, West Conshohocken, PA, American Society for Testing Materials,
1998.
64. Mullarky, J. I., “Long Term Strength Gain of Controlled Low-Strength
Materials,” The Design and Application of Controlled Low-Strength Materials
(Flowable Fill), ASTM STP 1331, West Conshohocken, PA, American
Society for Testing Materials, 1998.
65. Gress, D., “The Effect of Freeze-Thaw and Frost Heaving on Flowable Fill,”
UNH Civil Engineering #1096-1 for New Hampshire Department of
Transportation, University of New Hampshire, 1996.
66. Brewer, W. E., “Economic Considerations When Using Controlled Low
Strength Material (CLSM-CDF) as Backfill,” Transportation Research
Record, No. 1315, 1991.
67. Buss, W. E., “Iowa Flowable Mortar Saves Bridges and Culverts,”
Transportation Research Record, No. 1234, 1989.
251
68. Fougere, K. A., “Evaluation of Flowable Fill for Use in Highway and Bridge
Applications,” Technical Paper 94-6, Maine Department of Transportation,
Maine, 1994.
69. Hook, W. and Clem D. A., “Innovative Uses of Controlled Low Strength
Material (CLSM) in Colorado,” The Design and Application of Controlled
Conshohocken, PA, American Society for Testing Materials, 1998.
70. Mason, T. F., “Use of Controlled Density Fill to fill Underslab Void,” The
71. Naik, T. R., Barun R., Ramme, B. W. and Lolbeck, H. J., “Filling Abandoned
Underground Facilities with CLSM Fly Ash Slurry,”Concrete International,
Vol. 12, No. 7, 1990.
72. Walker, M. P., and Ash, J. P., “Flowable Fill Backfill for Use in Sequential
Excavations in Contaminated Soil,” The Design and Application of Controlled
73. Ardani, A., “Bridge Approach Settlement,” Report No. CDOH-DTP-R-87-06,
Colorado Department of Highways, 1987.
74. Gardner, M. R., “Developing controlled Low-Strength Materials to Meet
Industry and Construction Needs,” The Design and Application of Controlled
252
75. Green, B. H., Staheli, K., Bennett, D. and Walley, D. M., “Fly-Ash-Based
Controlled Low-Strength Material (CLSM) Used for Critical Microtunneling
Application,” The Design and Application of Controlled Low-Strength
Materials (Flowable Fill), ASTM STP 1331, West Conshohocken, PA,
American Society for Testing Materials, 1998.
76. Kawasaki, H., Horiuchi, S., Akatsuka, M. and Sano, S., “Fly-Ash Slurry
Island, I. Theoretical and Experimental Investigation.” Journal of Materials in
Civil Engineering, Vol.4, No. 2. 1992.
77. Snethen, D. R. and Benson, J. M., “Construction of CLSM Approach
Embankment to Minimize the Bump at the End of the Bridge,” The Design
and Application of Controlled Low-Strength Materials (Flowable Fill),
ASTM STP 1331, West Conshohocken, PA, American Society for Testing
Materials, 1998.
78. Swaffar, K. M. and Price, H. R., “Tunnel Saved by Fly Ash,” Civil
Engineering, Vol. 57, No. 9, 1987.
79. Spangler, M. G. and Handy, R. L., Soil Engineering, Fourth Edition, Harper &
Row, New York, 1982.
80. Hegarty, J. R. and Eaton, S. J., “Flowable Fill Promotes Trench Safety and
Supports Drainage Pipe Buried 60 ft (18.3m) under New Runway,” The
253
81. Portland Cement Association, “Soil-Cement Laboratory Handbook.” Skokie,
Ill, 59 p., 1992.
82. Lea’s Chemistry of Cement and Concrete, Hewlett, P. C., Editor, Fourth
Edition, 1998, P. 526.
83. McCuen, R. H., Statistical Methods for Engineers, Prentice-Hall, 1985.
254
Vita
Lianxiang Du was born in Emeishan City, Sichuan Province, China on
August 25, 1971, the son of Wuyin Du and Yueying Yuan. After completing his
work at Emei Second High School, Emeishan, Sichuan, in 1989, he entered
Hunan University in Changsha, Hunan Province, China, where he was awarded
the Zhao Zhiying Award. In 1993, he obtained the degree of Bachelor of Science.
In September of the same year, he entered the graduate school at Southwest
Jiaotong University, Chengdu, Sichuan Province, China. In June 1996, he earned
his master’s degree in science with the thesis titled, “Experimental Study on
Properties of Slurry Infiltrated Fiber Concrete (SIFCON).” In Fall 1997, he
received the prestigious Bloc Fellowship and entered The University of Delaware
to pursue his Ph. D. In Fall 1999, he transferred to The University of Texas at
Austin, where he worked as a teaching and research assistant.
Permanent address: Metal Materials Factory 3-7, Emeishan, Sichuan, China,
614200
This dissertation was typed by the author.
255

Laboratory Investigations of Controlled Low - Strength Material PDF

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Laboratory Investigations of Controlled Low - Strength Material PDF

Загружено:

Авторское право:

Доступные форматы

Copyright

Lianxiang Du, B.S., M.S.C.E.

The University of Texas at Austin

for the Degree of

The University of Texas at Austin

All rights reserved.

Bell & Howell Information and Learning Company

Certifies that this is the approved version of the following dissertation:

Laboratory Investigations of Controlled Low-Strength Material

Kevin J. Folliard, Supervisor

This dissertation is dedicated to my parents, my wife, Maohui Zhou, and other

me until I finished my assignments when I was a kid.

He spent countless time advising, discussing, and encouraging me. He opened

my eyes to the world of concrete technology. He was so patient that he spent

many hours with my English skills. I owe him so much.

co-sponsored the research work presented in this dissertation. His spiritual

rivals, but also friends.

I also appreciate very much the cooperation and support of my other

also goes to Dr. Dov Leshchinsky at University of Delaware. He was always

there providing answers to issues in geotechnical engineering. The involving of

appreciated. I would also like to thank members of the Construction Materials

Research Group, specifically Mike Rung, Kerry Rothenberg, David Whitney,

Lianxiang Du, Ph.D.

The University of Texas at Austin, 2001

Supervisor: Kevin J. Folliard

This report presents the results of experimental studies on the properties of

evaluating the characteristics of this material. A wide range of CLSM mixtures

were included in this study.

Fresh properties of CLSM were evaluated. The use of bottom ash,

subsidence of fresh CLSM mixtures.

The hardened properties of CLSM, including compressive strength,

excavatability, California Bearing Ratio, resilient modulus, water permeability,

were proposed for the strength development of CLSM mixtures. Modifications to

ASTM D 4832 were proposed to make it a more appropriate method for

measuring of the compressive strength of CLSM. The effects of alternative

temperature and humidity, were examined.

The resistance of CLSM mixtures to freezing and thawing was evaluated

strengths were found to improve the durability of CLSM mixtures exposed to

formation of galvanic cells may be a concern to pipeline engineers. Factors

influencing the corrosion behaviors of ductile-iron samples were identified. For

coupled and uncoupled corrosion conditions, different contributing factors and

mechanisms were identified.

LIST OF TABLES XVI

LIST OF FIGURES XVIII

1.1 GENERAL ........................................................................................................ 1

1.2 RESEARCH OBJECTIVES .............................................................................. 3

1.3 OUTLINE OF THE DISSERTATION ............................................................. 4

CHAPTER 2: LITERATURE REVIEW 6

2.2 MATERIALS .................................................................................................... 9

2.4 BATCHING, MIXING, AND TRANSPORTATING .................................... 20

2.5 CLSM PROPERTIES AND TEST METHODS ............................................. 21

2.7 CURRENT PRACTICE .................................................................................. 41

2.8 RESEARCH NEEDS ...................................................................................... 59

CHAPTER 3: MATERIALS AND MIXTURE PROPORTIONS 64

3.2 FINE AGGREGATE ....................................................................................... 64

3.3 MINERAL ADMIXTURES ............................................................................ 68

3.4 CHEMICAL ADMIXTURES ......................................................................... 70

3.6 MIXING WATER ........................................................................................... 71

3.7 CLSM MIXTURE PROPORTIONS............................................................... 72

CHAPTER 4: TEST METHODS AND PROCEDURES 82

4.2 FRESH CLSM TEST METHODS .................................................................. 85

4.3 HARDENED PROPERTIES OF CLSM......................................................... 91

4.4 DURABILITY OF CLSM............................................................................. 109