REINFORCEMENT OF CEMENT-BASED MATRICES WITH GRAPHITE
NANOMATERIALS
By

Muhammad Maqbool Sadiq

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Civil Engineering – Doctor of Philosophy
2013

ABSTRACT
REINFORCEMENT OF CEMENT-BASED MATRICES WITH GRAPHITE
NANOMATERIALS
By

Muhammad Maqbool Sadiq

Cement-based materials offer a desirable balance of compressive strength, moisture resistance,
durability, economy and energy-efficiency; their tensile strength, fracture energy and durability
in aggressive environments, however, could benefit from further improvements. An option for
realizing some of these improvements involves introduction of discrete fibers into concrete.
When compared with today’s micro-scale (steel, polypropylene, glass, etc.) fibers, graphite
nanomaterials (carbon nanotube, nanofiber and graphite nanoplatelet) offer superior geometric,
mechanical and physical characteristics. Graphite nanomaterials would realize their
reinforcement potential as far as they are thoroughly dispersed within cement-based matrices,
and effectively bond to cement hydrates.
The research reported herein developed non-covalent and covalent surface modification
techniques to improve the dispersion and interfacial interactions of graphite nanomaterials in
cement-based matrices with a dense and well graded micro-structure. The most successful
approach involved polymer wrapping of nanomaterials for increasing the density of hydrophilic
groups on the nanomaterial surface without causing any damage to the their structure.
The nanomaterials were characterized using various spectrometry techniques, and SEM
(Scanning Electron Microscopy). The graphite nanomaterials were dispersed via selected
sonication procedures in the mixing water of the cement-based matrix; conventional mixing and
sample preparation techniques were then employed to prepare the cement-based nanocomposite

samples, which were subjected to steam curing. Comprehensive engineering and durability
characteristics of cement-based nanocomposites were determined and their chemical
composition, microstructure and failure mechanisms were also assessed through various
spectrometry, thermogravimetry, electron microscopy and elemental analyses.
Both functionalized and non-functionalized nanomaterials as well as different micro-scale fibers
were used for comparison purposes at different volume fractions. Replicated mixes and tests
were considered to provide the basis for statistically reliable inferences.
Theoretical studies were conducted in order to develop insight into the reinforcement
mechanisms of properly functionalized graphite nanomaterials. The results suggested that
modified graphite nanomaterials improve the mechanical performance of cement-based matrices
primarily through control of microcrack size and propagation, relying on their close spacing
within matrix and dissipation of substantial energy by debonding and frictional pullout over their
enormous surface areas. The gains in barrier qualities of cement-based materials with
introduction of modified graphite nanomaterials could be attributed to the increased tortuosity of
diffusion paths in the presence of closely spaced nanomaterials.
Experimental investigations were designed and implemented towards identification of the
optimum (nano- and micro-scale) reinforcement systems for high-performance concrete through
RSA (Response Surface Analysis). A comprehensive experimental data base was developed on
the mechanical, physical and durability characteristics as well as the structure and composition of
high-performance cementitious nanocomposites reinforced with modified graphite nanomaterials
and/ or different micro-fibers.

DEDICATION
To my Ammi, Abbu and Brothers
for their endless love, support and encouragement…
Also, to My Wife and the Four Fairies in my life who have made my life so beautiful and are a
constant source of love, motivation and inspiration…

iv

ACKNOWLEDGMENTS
All Praises to Allah Almighty for giving me the ability and serenity to conduct this research,
strength to persevere through hard times and for showering His countless blessings on me. I
would also like to express my humble gratitude to my great country Pakistan for providing me
with all the wonderful opportunities in my life, especially to Pakistan Army for giving me this
chance to pursue this advance degree.
It is with immense gratitude that I acknowledge my Academic Advisor and my Committee
Chair, Professor Parviz Soroushian for his tugs and pushes as well as support and guidance
during the course of my research. I would also like to thank Professor Rigoberto Burgueño,
Professor Lawrence T. Drzal and Professor Venkatesh Kodur for joining my PhD committee,
and for their valuable advice throughout my education at MSU. Many thanks and appreciation
for Dr. Anagi Balachandra for all her help during the initial chemistry part of this research work.
I would especially like to thank Professor and Chairperson Neeraj Buch, Professor and Associate
Dean of Graduate Studies and Faculty Development M. Koochesfahani and Director Office of
International Students and Scholars Peter Briggs for supporting me through some very hard
times. And, I would also like to extend my thanks to Laura Taylor, Mary Mroz, and Margaret
Conner for all the help they provided.
I would also like to thank my fellow graduate students for their advice and support through out
the duration of my graduate studies. Last but not the least I would like to thank Michigan State
University as a whole for providing such a wonderful atmosphere of learning and development.
Hats off to you my Alma Mater – Go Green, Go White.

v

TABLE OF CONTENTS
LIST OF TABLES.……………………………………………………………………………..xiii
LIST OF FIGURES……………………………………………………………………………...xx
Chapter 1 ......................................................................................................................................... 1
1
Introduction ..................................................................................................................... 1
1.1
General ........................................................................................................................ 1
1.2
Graphite Nanomaterials .............................................................................................. 3
1.2.1 Carbon Nanotubes ................................................................................................... 3
1.2.2 Carbon Nanofibers .................................................................................................. 6
1.2.3 Graphite Nanoplatelets............................................................................................ 7
1.3
Merits of Graphite Nanomaterials .............................................................................. 7
1.4
The Need and Routes for Functionalization of Graphite Nanomaterials .................. 10
1.5
Fresh Cement Past and Hydrates: Morphological Relationships with Graphite
Nanomaterials ....................................................................................................................... 15
1.6
Research Hypothesis ................................................................................................. 17
1.7
Research Objectives .................................................................................................. 17
1.8
Scope ......................................................................................................................... 20
Chapter 2…………………………………………………………………………………………38
2
Literature Review, Materials and Methods ................................................................... 38
2.1
General ...................................................................................................................... 38
2.2
Experimental Studies ................................................................................................ 39
2.2.1 Carbon Nanotubes ................................................................................................. 40
2.3
Knowledge Gaps ....................................................................................................... 46
2.4
Graphite Nanomaterials and Microfibers.................................................................. 49
2.5
Cementitious Matrix and Processing Techniques ..................................................... 49
2.5.1 Cementitious Matrix ............................................................................................. 49
2.5.2 Dispersion and Mixing Procedure ........................................................................ 57
2.6
Experimental Methods .............................................................................................. 60
Chapter 3…………………………………………………………………………………………94
3
Reinforcement of High-Performance Cementitious Matrices with Relatively Low
Volume Fractions of Graphite Nanomaterials .............................................................................. 94
3.1
General ...................................................................................................................... 94
3.2
Introduction ............................................................................................................... 94
3.3
Materials and Methods .............................................................................................. 96
3.3.1 Graphite Nanomaterials and Microfibers.............................................................. 96
3.3.2 Cementitious Matrices, Mixing and Curing Procedure ........................................ 97
3.3.3 Experimental Methods .......................................................................................... 97
3.4
Characterization
of
Non-Functionalized
and
Functionalized
Graphite
Nanomaterials.. ................................................................................................................. 98
3.4.1 FTIR Spectrometry ............................................................................................... 98
3.4.2 Scanning Electron Microscopy ............................................................................. 99
vi

3.4.3 Raman Spectrometry............................................................................................. 99
3.4.4 X-Ray Photoelectron Spectroscopy (XPS) ......................................................... 100
3.5
Engineering Properties, Structure and Failure Modes of Cementitious
Nanocomposites .................................................................................................................. 101
3.5.1 Engineering Properties ........................................................................................ 101
3.5.2 Transmission and Scanning Electron Microscopic Investigations ..................... 103
3.5.3 X-Ray Diffraction (XRD) Analysis .................................................................... 104
3.5.4 Thermogravimetric Analysis (TGA)................................................................... 105
3.5.5 Energy-Dispersive Spectroscopy (EDS) ............................................................. 107
3.6
Summary ................................................................................................................. 108
Chapter 4 ..................................................................................................................................... 137
4
Effect of Refined Functionalization of Nanotubes on their Reinforcement Effeciency at
Different Volume Fraction in Cementitious Matrices ................................................................ 137
4.1
General .................................................................................................................... 137
4.2
Materials and Methods ............................................................................................ 138
4.2.1 Materials ............................................................................................................. 138
4.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 138
4.2.3 Experimental Methods ........................................................................................ 139
4.3
Further Functionalization and Characterization of Carbon Nanotubes .................. 139
4.3.1 Polymer Physisorption ........................................................................................ 139
4.3.2 Covalent Tethering of Polymers onto Nanotube Walls ...................................... 141
4.3.3 Further Functionalization of Carboxylated Nanotubes Starting from Carboxyl
Groups. ............................................................................................................................ 143
4.3.4 Microwave Irradiation ........................................................................................ 144
4.4
Experimental Results .............................................................................................. 145
4.4.1 Reinforcement Efficiency of Carbon Nanotubes with Different Functionalization
Conditions ....................................................................................................................... 145
4.4.2 Scanning Electron Microscope Investigations .................................................... 147
4.4.3 X-Ray Diffraction Analysis (XRD) .................................................................... 148
4.4.4 Thermogravimetric Analysis (TGA)................................................................... 148
4.5
Summary ................................................................................................................. 149
Chapter 5………………………………………………………………………………………..170
5
Reinforcing Effects of Multiwalled Carbon Nanotubes and Fibers in High-Performance
Cementitious Materials ............................................................................................................... 170
5.1
General .................................................................................................................... 170
5.2
Materials and Methods ............................................................................................ 170
5.2.1 Graphite Nanomaterials and Microfibers............................................................ 171
5.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 171
5.2.3 Experimental Methods ........................................................................................ 171
5.3
Experimental Results and Discussion ..................................................................... 172
5.3.1 Multiwalled Carbon Nanotube Aspect Ratio and Volume Fraction ................... 172
5.3.2 Effects of Hybrid (nano- and micro-scale) Reinforcement Systems .................. 177
5.4
Summary ................................................................................................................. 180

vii

Chapter 6………………………………………………………………………………………..193
6
Reinforcing Effects of Multiwalled Carbon Nanotubes at Different Volume Fractions
in Cementitious Pastes ................................................................................................................ 193
6.1
General .................................................................................................................... 193
6.2
Materials and Methods ............................................................................................ 194
6.2.1 Graphite Nanomaterials ...................................................................................... 194
6.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 194
6.2.3 Experimental Methods ........................................................................................ 194
6.3
Experimental Results and Discussion ..................................................................... 195
6.3.1 Effects of MWNT D Volume Fraction and Superplasticizer Type .................... 195
6.3.2 Effect of Dispersing Multiwalled Carbon Nanotubes in Different Fractions of
Mixing Water .................................................................................................................. 197
6.4
Summary ................................................................................................................. 199
Chapter 7 ..................................................................................................................................... 205
7
Nano- and/or Micro-Scale Reinforcements of High-Performance Cementitious
Mortar…. .................................................................................................................................... 205
7.1
General .................................................................................................................... 205
7.2
Materials and Methods ............................................................................................ 206
7.2.1 Materials, Graphite Nanomaterials and Microfibers .......................................... 206
7.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 207
7.2.3 Experimental Methods ........................................................................................ 207
7.3
Experimental Results and Discussion ..................................................................... 207
7.3.1 Carbon Nanotubes and/ or Carbon Microfibers at Different Volume Fractions and
with Different Superplasticizers ..................................................................................... 208
7.3.2 Enhanced Hybrid Reinforcement of High-Performance Cementitious Mortar .. 212
7.4
Summary ................................................................................................................. 214
Chapter 8 ..................................................................................................................................... 229
8
Damage-Sensitive Electrical Conductivity of High-Performacne Mortars Reinforced
with Multiwalled Carbon Nanotubes .......................................................................................... 229
8.1
General .................................................................................................................... 229
8.2
Materials and Methods ............................................................................................ 230
8.2.1 Materials and Graphite Nanomaterials ............................................................... 230
8.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 231
8.2.3 Experimental Methods ........................................................................................ 231
8.3
Experimental Results and Discussion ..................................................................... 232
8.3.1 Effects of Modified Multiwalled Carbon Nanotubes on the Electrical
Conductivity of High-Performance Cementitious Mortars............................................. 232
8.3.2 Damage-Sensitivity of the Electrical Conductivity of High-Performance
Cementitious Mortars Incorporating Modified Carbon Nanotubes ................................ 233
8.3.3 Scanning Electron Microscope Observations ..................................................... 233
8.4
Summary ................................................................................................................. 234

viii

Chapter 9………………………………………………………………………………………..240
9
Evaluation of Hybrid (Nano- and Micro-scale) Reinforcemet Effects on the
Engineering of High-Performance Concrete .............................................................................. 240
9.1
General .................................................................................................................... 240
9.2
Materials and Methods ............................................................................................ 242
9.2.1 Graphite Nanomaterials and Microfibers............................................................ 242
9.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 242
9.2.3 Experimental Methods ........................................................................................ 243
9.3
Experimental Results and Discussion ..................................................................... 243
9.3.1 Carbon Nanotubes and/ or Carbon Microfibers at Different Volume Fractions and
with Different Superplasticizers ..................................................................................... 243
9.4
Summary ................................................................................................................. 249
Chapter 10………………………………………………………………………………………260
10
Evaluation of carbon Nano and/ or Micro-scale Reinforcements on the Engineering
Properties of High Strength Concrete and Comparison with High-Performance Concrete ....... 260
10.1 General .................................................................................................................... 260
10.2 Materials and Methods ............................................................................................ 261
10.2.1 Graphite Nanomaterials and Microfibers............................................................ 261
10.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 261
10.2.3 Experimental Methods ........................................................................................ 262
10.3 Experimental Results and Discussion ..................................................................... 262
10.3.1 Flexural Performance .......................................................................................... 262
10.3.2 Compressive Strength ......................................................................................... 264
10.3.3 Impact Resistance ............................................................................................... 265
10.3.4 Abrasion Resistance ............................................................................................ 265
10.4 Summary ................................................................................................................. 266
Chapter 11………………………………………………………………………………………269
11
Evaluation of Reinforcement Efficiency of Carbon Nano- and Micro-fibers in HighPerformance Cementitious Paste ................................................................................................ 269
11.1 General .................................................................................................................... 269
11.2 Materials and Methods ............................................................................................ 271
11.2.1 Graphite Nanomaterials and Microfibers............................................................ 272
11.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 272
11.2.3 Experimental Methods ........................................................................................ 272
11.3 Experimental Results and Discussion ..................................................................... 273
11.3.1 Effects of Carbon Nanofiber Surface Functionalization on Engineering
Properties.. ...................................................................................................................... 273
11.3.2 Effects of Hybrid (Nano- and Micro-scale) Reinforcement Systems ................. 276
11.4 Summary ................................................................................................................. 278
Chapter 12………………………………………………………………………………………290
12
Evaluation of Reinforcement Efficiency of Carbon Nano- and Micro-fibers in HighPerformance Cementitious Mortar .............................................................................................. 290
12.1 General .................................................................................................................... 290

ix

12.2 Materials and Methods ............................................................................................ 292
12.2.1 Graphite Nanomaterials and Microfibers............................................................ 292
12.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 292
12.2.3 Experimental Methods ........................................................................................ 293
12.3 Experimental Results and Discussion ..................................................................... 293
12.3.1 Effects of Carbon Nano- and/or Micro-fibers Used at Different Volume Fraction
and with Different Superplasticizers ............................................................................... 293
12.3.2 Energy-Dispersive Spectroscopy (EDS) ............................................................. 297
12.3.3 Response Surface Analysis of High-Performance Cementitious Mortar ........... 297
12.4 Summary ................................................................................................................. 298
Chapter 13………………………………………………………………………………………306
13
Evaluation of Reinforcement Efficiency of Carbon Nano- and Micro-fibers in HighPerformance Cementitious Concrete .......................................................................................... 306
13.1 General .................................................................................................................... 306
13.2 Materials and Methods ............................................................................................ 308
13.2.1 Graphite Nanomaterials and Microfibers............................................................ 308
13.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 308
13.2.3 Experimental Methods ........................................................................................ 309
13.3 Experimental Results and Discussion ..................................................................... 309
13.3.1 Flexural Performance .......................................................................................... 309
13.3.2 Compressive Strength ......................................................................................... 312
13.3.3 Impact Resistance ............................................................................................... 312
13.3.4 Abrasion Resistance ............................................................................................ 313
13.3.5 Response Surface Analysis ................................................................................. 313
13.3.6 Scanning Electron Microscope Observations ..................................................... 315
13.4 Summary ................................................................................................................. 316
Chapter 14………………………………………………………………………………………324
14
Graphite Nanoplatelets in High-Performance Matrices .............................................. 324
14.1 General .................................................................................................................... 324
14.2 Materials and Methods ............................................................................................ 326
14.2.1 Graphite Nanomaterials and Microfibers............................................................ 326
14.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 326
14.2.3 Experimental Methods ........................................................................................ 326
14.3 Experimental Results and Discussion ..................................................................... 327
14.3.1 Effects of Graphite Platelets and Carbon Microfibers on Engineering Properties of
DSP Paste ........................................................................................................................ 327
14.3.2 Effects of Graphite Platelets and Carbon Microfibers on Engineering Properties of
DSP Mortar ..................................................................................................................... 328
14.3.3 Effects of Graphite Platelets and Different Microfibers on Engineering Properties
of DSP Concrete ............................................................................................................. 331
14.4 Summary ................................................................................................................. 336

x

Chapter 15………………………………………………………………………………………343
15
Durability Characteristics of High-Performance Cementitious Matric Reinforced with
Nano and/or Micro Reinforcements............................................................................................ 343
15.1 General .................................................................................................................... 343
15.2 Materials and Methods ............................................................................................ 344
15.2.1 Graphite Nanomaterials and Microfibers............................................................ 344
15.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 345
15.2.3 Experimental Methods ........................................................................................ 345
15.3 Test Results and Discussion of DSP Mortar Matrix ............................................... 345
15.3.1 DSP Mortar with Different Nano and/ or Micro Reinforcements ...................... 346
15.3.2 DSP Mortar with Refined Nano and/ or Micro Reinforcements ........................ 349
15.4 Test Results and Discussion of DSP Concrete Matrix............................................ 352
15.4.1 Exposure to Elevated Temperature ..................................................................... 352
15.4.2 Freeze-Thaw Cycles............................................................................................ 353
15.4.3 Wet-Dry Cycles .................................................................................................. 354
15.4.4 Hot Water Immersion ......................................................................................... 355
15.5 Summary ................................................................................................................. 356
Chapter 16……………………………………………………………………………………....374
16
Effects of Modified Graphite Nanomaterials on the Moisture Sorption Characteristics
of High-Performance Cementitious Materials ............................................................................ 374
16.1 General .................................................................................................................... 374
16.2 Materials and Methods ............................................................................................ 375
16.2.1 Graphite Nanomaterials and Microfibers............................................................ 375
16.2.2 Cementitious Matrices, Mixing and Curing Procedure ...................................... 376
16.2.3 Experimental Methods ........................................................................................ 376
16.3 Experimental Results and Discussion ..................................................................... 377
16.3.1 Rate of Moisture Absorption of DSP Paste Matrix ............................................ 377
16.3.2 Rate of Moisture Absorption of DSP Mortar Matrix .......................................... 379
16.3.3 Rate of Moisture Absorption of DSP Concrete Matrix ...................................... 380
16.4 Comparison of Graphite Nanomaterials with other Methods of Reducing
Permeability of Concrete .................................................................................................... 382
16.5 Summary ................................................................................................................. 384
Chapter 17………………………………………………………………………………………396
17
Plant Production and Placement of Nanocomposites – Challenges and Results ........ 396
17.1 General .................................................................................................................... 396
17.2 Materials and Methods ............................................................................................ 398
17.2.1 Graphite Nanomaterials and Microfibers............................................................ 398
17.2.2 Cementitious Matrices, Dispersion, Mixing and Curing Procedure ................... 398
17.2.3 Challenges Associated with Plant Production .................................................... 400
17.2.4 Experimental Methods ........................................................................................ 402
17.3 Experimental Results and Discussion ..................................................................... 402
17.3.1 Flexural Performance .......................................................................................... 402
17.3.2 Compressive Strength ......................................................................................... 404
17.3.3 Impact Resistance ............................................................................................... 405

xi

17.3.4 Abrasion Resistance ............................................................................................ 406
17.3.5 Rate of Moisture Absorption by Capillary Rise ................................................. 407
17.3.6 General Observations .......................................................................................... 409
17.4 Summary ................................................................................................................. 409
Chapter 18………………………………………………………………………………………420
18
Theoratical Studies and Performance to Cost Analysis of DSP Matrices Reinforced
with Graphite Nanomaterials ...................................................................................................... 420
18.1 General .................................................................................................................... 420
18.2 Merits of Graphite Nanomaterials .......................................................................... 422
18.2.1 Average Spacing and Specific Surface Area of Nanomaterials in Cementitious
Matrices........................................................................................................................... 422
18.3 Reinforcement Efficiency of Multiwalled Carbon Nanotubes ............................... 424
18.3.1 Optimum Volume Fraction of MWNTs ............................................................. 424
18.4 Analysis of the Debonding Process of Functionalized Carbon Nanotubes, and Their
Contribution to Tensile Strength ......................................................................................... 426
18.4.1 DSP Paste ............................................................................................................ 426
18.5 Analysis of the Frictional Pullout of Carbon Nanotubes, and their Contribution to
Fracture Energy ................................................................................................................... 429
18.6 Elastic Modulus and Tensile Strength Based on the Composite Materials
Approach… ......................................................................................................................... 431
18.7 Theoretical Predictions versus Experimental Results for High-Performance
Cementitious Paste, Mortar and Concrete .......................................................................... 433
18.7.1 High-Performance Paste ..................................................................................... 433
18.7.2 High Performance Mortar and Concrete ............................................................. 434
18.8 Synergistic Role of Nano- and Micro-scale Reinforcement ................................... 435
18.9 Diffusion Resistance of Cement-Based Nanocomposites....................................... 438
18.10 Electric Conductivity and Its Damage Sensitivity .................................................. 440
18.11 Cost versus Performance of Cementitious Nanocomposites .................................. 442
18.12 Targeted Applications ............................................................................................. 444
18.13 Summary ................................................................................................................. 447
Chapter 19……………………………………………………………………………………....469
19
Conclusions ................................................................................................................. 469
19.1 General .................................................................................................................... 469
19.2 Key Findings ........................................................................................................... 472
19.3 Recommendations for Future Research .................................................................. 493
19.4 Impact of the Research Effort ................................................................................. 495
REFERENCES ........................................................................................................................... 496

xii

LIST OF TABLES
Table 1.1- Applications, performance requirements, and competing fibers. ................................ 23
Table 1.2- Typical properties of graphite nanomaterials and carbon fiber. .................................. 24
Table 1.3- Key properties governing the reinforcement efficiency of multi-walled carbon
nanotubes versus conventional steel, polypropylene and glass fibers. ......................................... 24
Table 2.1- Summary review of experimental studies reported on the use of nanomaterials in
cement-based matrices. ................................................................................................................. 63
Table 2.2- Strength test results after 28 days of curing (Li, Wang et al. 2005). ........................... 81
Table 2.3- Physical attributes of all graphite nano and micro-scale reinforcement systems. ....... 81
Table 2.4- The cementitious matrices selected for evaluation of the reinforcement efficiency of
graphite nanomaterials. ................................................................................................................. 82
Table 2.5- Chemical composition of Type I Portland cement. ..................................................... 82
Table 3.1- Elemental analysis of fluorinated MWNTs and ethyl carboxylated MWNTs. ......... 109
Table 3.2- Mix designations of DSP pastes reinforced with non-functionalized and functionalized
graphite nano and micro reinforcements at 0.08 vol.%. ............................................................. 109
Table 3.3- Mix designation of high-performance DSP mortar with functionalized graphite
nanomaterials at 0.03 vol.%. ....................................................................................................... 110
Table 5.1- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DPS paste matrix reinforced with 0%, 0.24% and 0.48% volume fractions of nonfunctionalized and acid-functionalized MWNT D & E. ............................................................. 181
Table 5.2- Mean values of compressive strength test results for DSP paste matrix reinforced with
0%, 0.24% and 0.48% volume fractions of non-functionalized and acid-functionalized carbon
nanotubes. ................................................................................................................................... 181
Table 5.3- Mean values of impact resistance test results for DSP paste reinforced with 0%,
0.24% and 0.48% volume fractions of non-functionalized and acid-functionalized carbon
nanotubes MWNT D & E. .......................................................................................................... 182
Table 5.4- Mean weight loss test results after abrasion for DSP paste reinforced with 0%, 0.24%
and 0.48% volume fractions of non-functionalized and acid-functionalized carbon nanotubes
MWNT D & E. ........................................................................................................................... 182

xiii

Table 5.5- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DSP paste reinforced different vol. fractions of micro- and/or nano-scale reinforcement..... 183
Table 5.6- Compressive strength test results for the DSP paste matrix reinforced with different
volume fractions of nano and micro-scale reinforcements. ........................................................ 183
Table 5.7- Impact resistance test results of DSP pastes with nano- and/or micro-scale
reinforcement. ............................................................................................................................. 183
Table 5.8- Abrasion weight losses of DSP pastes with nano- and/or micro-scale reinforcement.
..................................................................................................................................................... 184
Table 6.1- Mean values of the flexural properties of DSP pastes with functionalized and nonfunctionalized MWNT D at 0.24 vol.% and 0.48vol.%, prepared using different carboxylic-based
superplasticizers. ......................................................................................................................... 201
Table 6.2- Mean values and standard errors of the compressive strength test results for DSP
pastes with functionalized and non-functionalized MWNT D at 0.24 vol.% and 0.48 vol.%,
prepared using different poly carboxylate-based superplasticizers. ........................................... 201
Table 6.3- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DSP paste with and without 0.24 vol.% of polymer-wrapped and non-functionalized MWNT
D dispersed in different fractions of mixing water. .................................................................... 202
Table 6.4- Mean values and standard errors of the compressive strength test results for DSP paste
with and without 0.24 vol.% polymer-wrapped and non-functionalized MWNT D dispersed in
different fractions of the mixing water. ...................................................................................... 202
Table 7.1- Mean values of the flexural attributes high-performance cementitious mortars with
nano- and/or micro-scale reinforcement systems. ...................................................................... 216
Table 7.2- Mean values and standard errors of the compressive strength of the high-performance
cementitious mortar matrices with nano- and/or micro-scale reinforcement systems. ............... 217
Table 7.3- Mean values of the impact resistance test results for high-performance cementitious
mortars reinforced with nano- and/or micro-scale reinforcement systems. ................................ 218
Table 7.4- Mean values of the abrasion resistance test results for high-performance cementitious
mortars reinforced with nano- and/or micro-scale reinforcement systems. ................................ 219
Table 7.5- Mean values of flexural strength, maximum deflection and energy absorption capacity
of high-performance cementitious mortars with nano- and/or micro-scale reinforcement systems.
..................................................................................................................................................... 220
Table 7.6- Mean values and standard errors of the compressive strength test results for high –
performance cementitious mortars with nano- and/or micro-scale reinforcement systems. ...... 220

xiv

Table 7.7- Mean values of the impact resistance test results for high-performance cementitious
mortars with nano- and/or micro-scale reinforcement. ............................................................... 221
Table 8.1- Mean values of the electrical conductivity test results for high-performance
cementitious mortars with different volume fractions of modified multiwalled carbon nanotubes
prepared using two different superplasticizers. .......................................................................... 235
Table 8.2- Mean values of the electrical conductivity of high-performance cementitious mortar
specimens with different volume fractions of functionalized multiwalled carbon nanotubes after
application of different compressive stress levels. ..................................................................... 235
Table 9.1- Mean values of flexural attributes of high-performance concrete with and without
different nano- and/ or micro-scale reinforcement systems. ....................................................... 250
Table 9.2- Mean values and standard errors of the compressive strength test results for highperformance concrete with and without nano- and/ or micro-scale reinforcement systems....... 250
Table 9.3- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement. .............................................................................................. 251
Table 9.4- Mean abrasion weight losses of high-performance concretes with and without nanoand/ or micro-scale reinforcement systems................................................................................. 251
Table 9.5- Outcomes of Ridge analysis of the flexural attributes and impact resistance test data
for high performance concrete with nano- and/or micro-scale reinforcement. .......................... 252
Table 10.1- Starting material selections and mix proportions (weight ratios) of high-performance
and high-strength concrete matrix. ............................................................................................. 267
Table 10.2- Mean values of flexural attributes of high-strength concrete with and without nanoand/ or micro-scale reinforcement systems and comparison with high-performance concrete. . 267
Table 10.3- Compressive strength values of high-strength concrete with and without nano- and/
or micro-scale reinforcement systems and comparison with high-performance concrete. ......... 267
Table 10.4- Impact resistance test results for high-strength concretes with and without nano- and/
or micro-scale reinforcement systems and comparison with high-performance concrete. ......... 268
Table 10.5- Mean abrasion weight losses of high-strength concretes with and without nano- and/
or micro-scale reinforcement systems and comparison with high-performance concrete. ......... 268
Table 11.1- Mean values of the flexural strength, maximum deflection and energy absorption
capacity test results for high-performance cementitious pastes with 0% and 0.24% (by volume of
cement plus silica fume) of non-functionalized and acid-functionalized CNFs. ........................ 279

xv

Table 11.2- Mean values of the flexural strength, maximum deflection and energy absorption
capacity test results for high-performance cementitious paste with carbon nanofiber and/or
microfiber reinforcement systems............................................................................................... 279
Table 11.3- Mean values and standard errors of the compressive strength test results for highperformance cementitious paste reinforced with CNFs and/or microfibers. .............................. 279
Table 11.4- Mean values of the impact resistance test results for high-performance cementitious
pastes with carbon nanofiber and/or microfiber reinforcement systems. ................................... 280
Table 11.5- Mean values and standard errors of the abrasion test results for high-performance
cementitious pastes with carbon nanofiber and/or microfiber reinforcement systems. .............. 280
Table 12.1- Mean values of the flexural attributes of high-performance (DSP) cementitious
mortar with nano- and/or micro-scale reinforcement systems. ................................................... 300
Table 12.2- Mean values and standard errors of the compressive strength test results for DSP
mortar with nano- and/or micro-scale reinforcement systems. ................................................... 301
Table 12.3- Impact resistance test results for high-performance (DSP) cementitious mortar with
nano- and/or micro-scale reinforcement systems. ...................................................................... 302
Table 12.4 - Mean abrasion weight losses of high-performance DSP cementitious mortars with
nano- and/or micro-scale reinforcement. .................................................................................... 302
Table 13.1- Mean values of flexural attributes of plain high-performance concrete and those
reinforced with different nano- and/ or micro-scale reinforcement systems. ............................. 317
Table 13.2- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems. ..... 317
Table 13.3- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement. .............................................................................................. 318
Table 13.4- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems. ............................................................................................ 318
Table 13.5- Outcomes of Ridge analysis of the flexural attributes and impact resistance test data
for high performance concrete with nano- and/or micro-scale reinforcement. .......................... 319
Table 14.1- Mean values of the flexural properties of DSP pastes with graphite nanoplatelet
and/or carbon microfiber reinforcement. .................................................................................... 337
Table 14.2- Mean values and standard errors of the compressive strength test results for DSP
pastes with graphite nanoplatelet and/or carbon microfiber reinforcement................................ 337

xvi

Table 14.3- Mean values and standard errors of the impact resistance test results for DSP pastes
with graphite nanoplatelet and/or carbon microfiber reinforcement. ......................................... 337
Table 14.4- Mean values and standard errors of the abrasion weight loss test results for DSP
pastes with graphite nanoplatelet and/or carbon microfiber reinforcement................................ 338
Table 14.5- Mean values of flexural strength, maximum deflection and energy absorption
capacity of DSP mortar with nano- and/or micro-scale reinforcement systems. ........................ 338
Table 14.6- Mean values and standard errors of compressive strength test results for DSP mortar
with nano- and/or micro-scale reinforcement systems. .............................................................. 338
Table 14.7- Mean values of impact resistance test results for DSP mortar reinforced with nanoand/or micro-scale reinforcement systems.................................................................................. 339
Table 14.8- Mean values of abrasion resistance test results for DSP mortar reinforced with nanoand/or micro-scale reinforcement systems.................................................................................. 339
Table 14.9- Mean values of flexural attributes of plain DSP concrete and those reinforced with
different nano- and/ or micro-scale reinforcement systems. ....................................................... 340
Table 14.10- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems. ..... 340
Table 14.11- Impact resistance test results for DSP concrete with and without nano- and/or
micro-scale reinforcement. ......................................................................................................... 341
Table 14.12- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems. ............................................................................................ 341
Table 15.1- Mean values of flexural attributes of DSP mortar reinforced with micro- and/or
nano-scale reinforcement prior to and after exposure to elevated temperature (ET). ................ 357
Table 15.2 - Mean values of the flexural attributes of DSP mortar samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 cycles of freeze-thaw (FT)............ 357
Table 15.3- Mean values of the flexural attributes of DSP mortars with micro- and/or nano-scale
reinforcement prior to and after exposure to 300 wet-dry (WD) cycles. .................................... 358
Table 15.4- Mean values of the flexural attributes of DSP mortars with micro- and/or nano-scale
reinforcement prior to and after 56 days of hot water (HW) immersion. ................................... 358
Table 15.5- Mean values of flexural attributes of DSP mortar reinforced with preferred hybrid
reinforcement systems prior to and after exposure to elevated temperature (ET). ..................... 359

xvii

Table 15.6- Mean values of the flexural attributes of DSP mortar specimens with preferred
hybrid reinforcement systems prior to and after 56 days of immersion in hot water (HW). ...... 359
Table 15.7- Mean values of flexural attributes of DSP concrete reinforced with micro- and/or
nano-scale reinforcement prior to and after exposure to elevated temperature (ET). ................ 360
Table 15.8- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 cycles of freeze-thaw (FT)............ 360
Table 15.9- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 wet-dry cycles (WD). ................... 361
Table 15.10- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 56 days in hot water (HW). ................. 361
Table 17.1- Starting material selections and mix proportions of high-strength concrete matrix for
plant production. ......................................................................................................................... 411
Table 17.2- Mean values of flexural attributes of plain high-performance concrete and those
reinforced with different nano- and/ or micro-scale reinforcement systems. ............................. 411
Table 17.3- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems. ..... 412
Table 17.4- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement. .............................................................................................. 412
Table 17.5- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems. ............................................................................................ 413
Table 18.1- Typical improvements in properties of high-performance cement-based materials
(concrete/mortar/paste) with 0.16 vol.% graphite nanomaterials, 0.24 vol.% carbon microfibers,
or hybrid reinforcement comprising 0.16 vol.% functionalized carbon nanotube and 0.24 vol.%
carbon microfibers. ..................................................................................................................... 448
Table 18.2- Mean values of the electrical conductivity test results for high-performance
cementitious mortars with different volume fractions of modified multiwalled carbon nanotubes
prepared using two different superplasticizers. .......................................................................... 448
Table 18.3- Mean values of the electrical conductivity of high-performance cementitious mortar
specimens with different volume fractions of functionalized multiwalled carbon nanotubes after
application of different compressive stress levels. ..................................................................... 449
Table 18.4- Performance gains of high-performance concrete per vol.% of reinforcement,
normalized with respect to the reinforcement cost. .................................................................... 449

xviii

Table 18.5- Typical performance gains of cement-based materials with introduction of the
specified volume fractions of functionalized carbon nanotube, and steel, polypropylene and glass
fibers. .......................................................................................................................................... 450

xix

LIST OF FIGURES
Figure 1.1- Arrangement of graphene sheets in carbon nanotube and nanofiber, graphite
nanoplatelet, and carbon micro-fiber. ........................................................................................... 25
Figure 1.2- Carbon nanotube ........................................................................................................ 25
Figure 1.3- Multi-walled carbon nanotubes. ................................................................................. 26
Figure 1.4- Roll-up of a graphene sheet to form different nanotube types. .................................. 26
Figure 1.5- Schematic depiction and electron microscope images of carbon nanofiber. ............. 27
Figure 1.6- SEM and TEM images of exfoliated graphite nanoplatelets (Li, Sham et al. 2007). 28
Figure 1.7- Reinforcement mechanisms of discrete reinforcement systems in composites. ........ 28
Figure 1.8- Loss of mechanical properties of carbon nanotubes due to agglomeration (Ajayan et
al, 2003). ....................................................................................................................................... 29
Figure 1.9- Schematics of the interactions of pristine and functionalized nanotubes with cement
hydration products. ....................................................................................................................... 30
Figure 2.1- Modulus-of-rupture (a) and compression resistance (b) test results for cement mortars
with and without MWNTs (Musso, Tulliani et al. 2009). ............................................................ 83
Figure 2.2- Schematic representation of the experimental setup based on continuous feeding of
cement particles using a screw feeder (Nasibulin, Shandakov et al. 2009). ................................. 84
Figure 2.3- TEM and SEM micrographs of different nano- and micro-scale reinforcement
systems. ......................................................................................................................................... 84
Figure 2.4- DSP cementitious paste, mortar and concrete.(Guerrini 2000).................................. 89
Figure 2.5- Steric and electrostatic repulsion of separated nanotubes. ......................................... 90
Figure 2.6- Dispersion and stability conditions of carbon nanotubes in aqueous media. ............. 90
Figure 2.7- Dispersion equipment: (a) Magnetic stirrer (b) Sonic horn. ...................................... 91
Figure 2.8- Experimental setups. .................................................................................................. 91
Figure 2.9- Durability test systems. .............................................................................................. 92
Figure 3.1 - FTIR spectra of pristine and acid-functionalized carbon nanotubes and nanofibers.
..................................................................................................................................................... 111

xx

Figure 3.2- FTIR spectra for MWNTs A prior to and after fluorination, and after ethyl
carboxylation upon fluorination. ................................................................................................. 114
Figure 3.3 - SEM images of MWNTs B prior to and after fluorination. .................................... 115
Figure 3.4- Raman spectra for MWNT A and B prior to and after fluorination......................... 116
Figure 3.5- XPS spectrum. .......................................................................................................... 117
Figure 3.6-Typical flexural load-deflection behavior of DSP paste with 0% and 0.08% volume
fractions of various non-functionalized and acid-functionalized graphite nanomaterials. ......... 119
Figure 3.7- Mean and standard error of various engineering properties of DSP paste reinforced
with non-functionalized and functionalized graphite nano and micro reinforcements at 0.08
vol.%. .......................................................................................................................................... 120
Figure 3.8- Effects of functionalized graphite nanomaterials at 0.03% on the engineering
properties of high-performance DSP mortar (means & standard errors). ................................... 126
Figure 3.9- Visual appearances of failed specimens after different tests. .................................. 131
Figure 3.10- TEM image of functionalized MWNT B, and SEM images of the fractured surface
of a cementitious matrix with 0.04 vol.% of functionalized MWNT C. .................................... 132
Figure 3.11- Bridging of micro-cracks by acid-functionalized graphite nanomaterials introduced
into cementitious matrices at 0.08% volume fraction. ................................................................ 133
Figure 3.12- Arrest and deflection of a micro-crack by a bundle of CNFs within the cementitious
matrix. ......................................................................................................................................... 134
Figure 3.13- SEM images of DSP paste with 0.08 vol.% of carbon micro-fiber exhibiting
showing pulled-out micro-fibers on a crack surface. .................................................................. 134
Figure 3.14- XRD graphs of the high-performance cementitious paste matrix with 0% and 0.04
vol.% of acid-functionalized MWNT A. .................................................................................... 135
Figure 3.15- Results of TGA for conventional mortar and for the DSP paste with 0% and 0.08
vol.% acid-functionalized MWNT A, MWNT B, carbon nano-fiber, fluorinated MWNT A. ... 136
Figure 4.1- Schematic depiction of the random and disordered adsorption of polymers upon
carbon nanotube walls................................................................................................................. 151
Figure 4.2- FTIR spectra for acid-functionalized carbon nanotubes (CNT-COOH), Gum Arabic
(GA), and acid-functionalized nanotubes modified with Gum Arabic (GA-CNTCOOH). ........ 151

xxi

Figure 4.3- FTIR spectra of acid-functionalized carbon nanotubes (CNT-COOH), PAA, and
acid-functionalized nanotubes modified with PAA (PAA-CNTCOOH).................................... 152
Figure 4.4- FTIR spectra of acid-functionalize carbon nanotubes (CNT-COOH), Glenium®, and
acid-functionalized nanotubes modified with Glenium® (Glenium-CNTCOOH)..................... 152
Figure 4.5- Schematic depiction of the tethering (covalent attachment) of polymer chains via
functional groups onto nanotube walls. ...................................................................................... 153
Figure 4.6- Covalent tethering of PAA onto carboxylated carbon nanotube. ............................ 153
Figure 4.7- Tethering of Glenium® onto carboxylated nanotube. ............................................. 153
Figure 4.8- FTIR spectra of carboxylated nanotubes (CNT-COOH), PAA and PAA-tethered
carboxylated nanotubes (PAA-CNTCOOH). ............................................................................. 154
Figure 4.9- FTIR spectra of carboxylated nanotubes (CNT-COOH), Glenium® and Glenium®tethered carboxylated nanotubes (Glenium-CNTCOOH). ......................................................... 154
Figure 4.10- Schematic depiction of covalent tethering of PEI to carboxylated nanotubes. ...... 155
Figure 4.11- FTIR spectra of carboxylated nanotubes (CNT-COOH), PEI, and PEI-tethered
carboxylated nanotubes (PEI-CNTCOOH). ............................................................................... 155
Figure 4.12- Mean values and standard errors of the flexural attributes and compressive strength
of DSP paste reinforced with 0%, 0.16% and 0.24% volume fractions of acid-functionalized and
non-functionalized MWNT D with different further treatments. ............................................... 156
Figure 4.13- Mean values and standard errors of the flexural attributes and compressive strength
of cementitious (DSP) pastes reinforced with 0%, 0.16% and 0.24% volume fractions of acidfunctionalized and non-functionalized MWNT E with different further treatments. ................. 159
Figure 4.14- Mean values and standard deviations of the flexural strength and energy absorption
capacity test results for cementitious (DSP) pastes reinforced with 0% and 0.16% volume
fractions of non-functionalized MWNT F with different polymer treatments. .......................... 162
Figure 4.15- Flexural strength and energy absorption capacity of DSP cement pastes reinforced
with 0% and 0.16% volume fractions of microwave treated MWNTs F and G. ........................ 164
Figure 4.16- Scanning electron microscope images of the cementitious DSP paste with different
volume fractions of surface-modified carbon nanotubes. ........................................................... 166
Figure 4.17- XRD spectrographs of DSP pastes with 0% and 0.16% volume fractions of acidfunctionalized MWNT D and MWNT E, that are further modified with Gum Arabic. ............ 168
Figure 4.18- Results of thermogravimetric analysis showing weight loss versus temperature. . 169

xxii

Figure 5.1- Flexural strength and energy absorption capacity of DSP cement pastes with 0.24 and
0.48 vol.% non-functionalized and acid-functionalized MWNT D & E. ................................... 185
Figure 5.2- Excessive clumping of carbon nanotubes in DSP paste at 0.48% volume fraction. 187
Figure 5.3- Typical SEM images of DSP paste. ......................................................................... 187
Figure 5.4- The results of thermogravimetric analysis of DSP cement paste with 0% and 0.24%
volume fractions of functionalized multi-walled carbon nanotubes MWNT D & E. ................. 190
Figure 5.5- Typical SEM mages of DSP paste reinforced with 0.16 vol.% of MWNT D-PAA and
0.24 vol.% of CMF: (a) dispersed MWNTs; and (b) failure mechanism of micro-scale fibers. 191
Figure 5.6- Results of thermogravimetric analysis (weight loss vs. temperature) for the highperformance cementitious matrix (DSP paste) with different volume fractions of micro and nano
reinforcements............................................................................................................................. 192
Figure 6.1- SEM mages of DSP paste reinforced with 0.24 vol.% of MWNT D casted with
ADVA® Cast 575 superplasticizer. ............................................................................................ 203
Figure 7.1- SEM image of a cement-based material incorporating well dispersed PAAphysisorbed multiwalled carbon nanotubes (CNT D-PAA). ...................................................... 222
Figure 7.2- Transmission electron microscope image of PAA physisorbed multiwalled carbon
nanotube Type D. ........................................................................................................................ 222
Figure 7.3- Complementary roles of nano- and micro-scale reinforcement in damage control by
suppression of micro-crack inception and control of microcrack growth. ................................. 223
Figure 7.4- Typical SEM images of the fractured surfaces of high-performance cementitious
mortars with nano- and micro-scale reinforcement systems....................................................... 226
Figure 8.1- Percolated network. .................................................................................................. 236
Figure 8.2- Electrical conductivity test setup and specimens. .................................................... 236
Figure 8.3- Typical SEM mages of high-performance cementitious mortar reinforced with
modified multiwalled carbon nanotube. ..................................................................................... 238
Figure 9.1- SEM images of fractured surfaces of high performance concrete. .......................... 256
Figure 11.1- Representative flexural load-deflection curves of high-performance cementitious
pastes with 0 and 0.24 vol.% of non-functionalized and oxidized nanofibers. .......................... 281
Figure 11.2- Flexural attributes of high-performance cementitious pastes. ............................... 281

xxiii

Figure 11.3- Clumping of non-functionalized CNFs in the high-performance cementitious paste
at 0.24% by volume of cement plus silica fume. ........................................................................ 282
Figure 11.4- Mean values and standard errors of the compressive strength test results for highperformance cementitious paste reinforced with 0 and 0.24 vol.% of non-functionalized and acidfunctionalized CNFs. .................................................................................................................. 283
Figure 11.5- Failed impact test specimens. ................................................................................. 283
Figure 11.6- Mean values and standard errors of the impact resistance test results for highperformance cementitious paste reinforced with 0 and 0.24 vol.% of non-functionalized and acidfunctionalized CNFs. .................................................................................................................. 284
Figure 11.7- Abrasion test specimens after test. ......................................................................... 284
Figure 11.8- Mean values of abrasion weight loss for high-performance cementitious pastes
reinforced with 0% and 0.24% of non-functionalized and acid-functionalized CNFs. .............. 285
Figure 11.9- Typical SEM images of high-performance pastes reinforced with 0.24 vol.% of
CNFs. .......................................................................................................................................... 285
Figure 11.10- Typical SEM mages of DSP cementitious paste reinforced with carbon microfibers
and/or oxidized CNFs. ................................................................................................................ 288
Figure 12.1- Typical SEM images of high-performance cementitious mortar reinforced with
different contents of polymer-wrapped or acid- functionalized CNFs. ...................................... 303
Figure 12.2- Typical SEM images of DSP cementitious materials reinforced with CNFs and/or
CMFs........................................................................................................................................... 305
Figure 13.1- SEM images of fractured surfaces of high performance concrete. ........................ 321
Figure 14.1- SEM images of failed specimens of DSP paste reinforced with GP, providing
indications of the graphite nanoplatelet delamination and their interactions with microcracks. 342
Figure 15.1- Tortuous diffusion paths forced by nano-scale reinforcement. .............................. 362
Figure 15.2- DSP mortar specimens with different reinforcement conditions after exposure to
elevated temperature. .................................................................................................................. 362
Figure 15.3- Effects of exposure to freeze-thaw on the dynamic modulus of DSP mortars with
micro- and/or nano-scale reinforcement after 300 cycles. .......................................................... 364
Figure 15.4- Effects of exposure to wet-dry cycles on the dynamic modulus of DSP mortars with
micro- and/or nano-scale reinforcement. .................................................................................... 365

xxiv

Figure 15.5- Effects of immersion in hot water on the dynamic modulus of DSP mortars with
micro- and/or nano-scale reinforcement. .................................................................................... 366
Figure 15.6- High-performance (DSP) mortar specimens with different reinforcement conditions
after exposure to elevated temperature. ...................................................................................... 367
Figure 15.7- Effects of immersion in hot water on the dynamic modulus of DSP mortars with
preferred hybrid reinforcement systems after 56 days. ............................................................... 368
Figure 15.8- DSP concrete specimens with different reinforcement conditions after exposure to
elevated temperature. .................................................................................................................. 369
Figure 15.9- Effects of exposure to freeze-thaw on the dynamic modulus of DSP concrete with
micro- and/or nano-scale reinforcement after 300 cycles. .......................................................... 371
Figure 15.10- Effects of exposure to wet-dry on the dynamic modulus of DSP concrete with
micro- and/or nano-scale reinforcement. .................................................................................... 372
Figure 15.11- Effects of immersion in hot water on the dynamic modulus of DSP concrete with
micro- and/or nano-scale reinforcement. .................................................................................... 373
Figure 17.1- Mean ROMS in HSC matrix reinforced with and without different volume fraction
of graphite nanomaterials and microfibers produced in laboratory and concrete mixing plant. 414
Figure 17.2- Average TWA in HSC matrix reinforced with and without different volume fraction
of graphite nanomaterials and microfibers produced in laboratory and concrete mixing plant. 415
Figure 17.3- Photos of Plant Production and laying of slab overlay. ......................................... 416
Figure 18.1- Evolution of DSP paste into DSP mortar and RPC cementitious products (Guerrini
2000). .......................................................................................................................................... 451
Figure 18.2- Average nanotube spacing versus nanotube diameter at different nanotube volume
fractions ranging from 0.1% to 1% (K=1.38 for 3-D random orientation of nanotubes). .......... 452
Figure 18.3- The specific surface area of nanotubes versus nanotube diameter for different
nanotube volume fractions. ......................................................................................................... 453
Figure 18.4- The crack suppression effect of closely spaced nanotubes. ................................... 454
Figure 18.5- Schematic depiction of the relationship between first-crack stress and nanotube
spacing. ....................................................................................................................................... 454
Figure 18.6- The trends in measured values of flexural strength at different volume fractions of
MWNTs. ..................................................................................................................................... 455

xxv

Figure 18.7- Typical SEM mages of the fractured surfaces of DSP paste reinforced with 0.48%
volume fraction of acid-functionalized and non- functionalized MWNT E. .............................. 459
Figure 18.8- The debonding model. ............................................................................................ 460
Figure 18.9- Interfacial shear stress distribution for critical length (lc) calculation. .................. 460
Figure 18.10- Pullout load versus displacement for various ratios of adhesion shear to interfacial
friction resistance. ....................................................................................................................... 461
Figure 18.11- Gains in engineering properties of high-performance concrete brought about by
0.16 vol.% functionalized carbon nanotube and/or 0.24 vol.% carbon (micro)fiber.................. 462
Figure 18.12- Three-dimensional assembly of carbon nanotubes within a cement-based matrix.
..................................................................................................................................................... 463
Figure 18.13- Tortuous diffusion paths forced by nano-scale reinforcement. ............................ 463
Figure 18.14- Diffusivity factor versus the reinforcement diameter at 0.24% volume fraction for
a diameter range covering multi-walled carbon nanotubes and conventional micro-scale fibers.
..................................................................................................................................................... 464
Figure 18.15- Relative decrease in rate of moisture sorption for different high-performance
matrices incorporating 0.24 vol.% of different nano and micro-scale reinforcements. .............. 465
Figure 18.16- Percolated network. .............................................................................................. 466
Figure 18.17- Percolation of conductive inclusions within a matrix. ......................................... 466
Figure 18.18- Electrical conductivity values plotted for different volume fractions of PAAphysisorbed MWNT D using different β values. ........................................................................ 467
Figure 18.19- Typical SEM image of high-performance cementitious mortar reinforced with
modified MWNTs showing uniform dispersion and percolation of nanotubes within the matrix.
..................................................................................................................................................... 468
Figure 18.20- Self-sensing smart concrete (Baoguo, Xun et al. 2009). ...................................... 468

xxvi

CHAPTER 1

1

INTRODUCTION

1.1 General
This research focuses on enhancement of the cement- and concrete-based civilian and military
infrastructure with functionalized graphite nanomaterials (carbon nanotube and nanofiber, and
graphite nanoplatelet). Concrete-based materials offer suitable engineering properties
(compressive strength, moisture resistance and durability) at low cost, combined with energysaving and ecological benefits. They constitute the most widely used class of manufactured
materials worldwide. The energy consumption for production of concrete is only 3% that of steel
(per unit weight), and cement-based materials are far superior to competing construction
materials in terms of ecological impact (pollution, deforestation, use of exhaustible raw
materials, etc.) (Mehta 1993). Cement-based materials, however, lack adequate toughness,
impact resistance and durability in aggressive environments. These are properties targeted to be
improved through reinforcement with discrete reinforcement systems, including micro-scale
fibers and functionalized graphite nanomaterials.
Introduction of functionalized graphite nanomaterials promises major gains in the structural
performance, durability (weathering resistance and fatigue life), and safety (under fire, explosion
and earthquake) of cement- and concrete-based infrastructure systems. Enhancement of cementbased materials with functionalized graphite nanomaterials promises qualitative improvements
over the successful and growing practice of reinforcing cement-based materials with micro-scale

1

steel, polypropylene, glass, cellulose, carbon and other fibers (Li 2002). The worldwide
consumption rate of these fibers in cement- and concrete-based infrastructure exceeds 1.5 million
tons/yr, representing close to 1 billion tons/year global consumption of fiber cement and fiber
concrete products. Existing fiber cement and fiber concrete applications, especially those with
greater demands for structural performance, durability and safety, represent viable entry points
for market introduction of functionalized graphite nanomaterials. These applications are
introduced in Table 1.1 together with the relevant performance considerations and the competing
fibers.
The targeted applications are of value to both the military and civilian infrastructure systems.
Applications of graphite nanomaterials towards development of explosion-resistant infrastructure
represent a priority topic actively pursued by the U.S. Army (Survivability Branch).
Graphite nanomaterials (e.g., carbon nanotubes), when compared with today’s (micro-scale)
fibers, offer distinctly high mechanical and physical attributes as well as enormous surface areas
and very close spacing within matrix even at low volume fractions (below 0.1%). The high
reinforcement efficiency of graphite nanomaterials in cement-based matrices could be realized
only if steps are taken to ensure thorough dispersion of nanomaterials within the matrix and
effective interfacial interactions between nanomaterials and cement hydrates. The high surface
area of nanomaterials generates strong tendencies towards agglomeration (via van der Waals
adhesion forces). The largely inert surfaces of graphite nanomaterials also offer limited
potentials for chemical bonding to cement hydrates.

2

1.2 Graphite Nanomaterials
The graphite nanomaterials considered in this investigation (Fig. 1.1a) include carbon nanotube,
carbon nanofiber, and graphite nanoplatelet. When compared with the near-perfect structure of,
say, carbon nanotube, the imperfect arrangement of graphene sheets in conventional (microscale) carbon fibers (Fig. 1.1b) generates substantial defects which greatly compromise
engineering properties.
1.2.1

Carbon Nanotubes

CNTs (Carbon nanotubes) are cylinder-shaped macromolecules with nano-scale diameter and
micro-scale (or larger) length. The walls of nanotubes are made up of a hexagonal lattice of
carbon atoms analogous to the atomic planes of graphite. They are capped at their ends by one
half of a fullerene-like (semi-spherical) molecule (Fig. 1.2). In the most general case, a CNT is
composed of a concentric arrangement of many cylinders (Fig. 1.3) bound together by weak van
der Waals interactions (Breuer and Sundararaj 2004). Such MWNTs (Multi-walled Carbon
Nanotubes) have diameters ranging from few up to 100 nm (nanometer), with lengths ranging
from few micrometers to several centimeters. SWNTs (Single-walled Carbon Nanotubes)
possess the simplest geometry, and have been observed with diameters ranging from 0.4 to 3 nm.
The formation of a SWCNT can be visualized through the rolling of a graphene sheet. Based on
the orientation of the tube axis with respect to the hexagonal lattice, the structure of a nanotube
can be completely specified through its chiral vector (Fig. 1.4), which is denoted by the chiral
indices (n, m). The diameter and chirality of carbon nanotubes influence their properties.
Historically the multi-walled nanotubes were the first to be discovered (1991), followed by their
single-walled counterparts (1993). Different methods are now well established for the production

3

of SWNTs, including catalytic decomposition of a gaseous carbon feedstock, laser ablation, and
electric arc-discharge (Balasubramanian and Burghard 2005).
Although carbon nanotubes (CNTs) are similar in composition to graphite, CNTs are highly
isotropic; it is this topology that distinguishes nanotube from other carbon structures, and gives
them their unique properties. Since carbon-carbon covalent bonds are among the strongest bonds
in nature, a structure based on perfect arrangement of these bonds oriented along the axis of the
nanotubes produces a very strong material with an extremely high strength-to-weight ratio. The
mechanical properties of various nanotube types have been extensively investigated. Their
tensile strength (~100-600 GPa (Giga Pascal) is about two orders of magnitude higher than that
3

of current high-strength carbon fibers, and their density is around 1.3 g/cm (Grams per
3

Centimeter Cube), lower than the density of commercial carbon fibers (1.8-1.9 g/cm ). The
compressive strength of carbon nanotubes is approximately two orders of magnitude higher than
the compressive strength of any known fiber. Carbon nanotubes are also one of the stiffest
structures ever made; compared to carbon fibers, which typically have Young’s moduli of about
380 GPa, moduli of nanotubes range between 1 and 5 TPa (Tera Pascal). The most striking
effect, however, is the combination of high ductility with high stiffness and strength, a property
that is absent in carbon fibers. Carbon nanotubes can accommodate large deformations without
rupture; the ultimate tensile strain of carbon nanotubes ranges from 10% to 30%, compared to
0.1% to 2% for carbon fibers. They can be flattened, twister and buckled as they deform
reversibly. The high aspect ratio of carbon nanotubes (as high as 10,000) favors their use in
composites to impart conductivity and reinforcement effects (Breuer and Sundararaj 2004). The
strength and modulus data for carbon nanotubes are occasionally calculated using the crosssectional area of wall(s) and not that of the nanotube (with an individual wall thickness of 0.34
4

nm). When multi-walled carbon nanotubes are evaluated in the context of composites, an
important consideration is that direct stress transfer occurs between the matrix and the outer wall
of the nanotube. The weak van der Waals interactions between nanotube walls limit the
contributions of inner walls. This would reduce the effective strength and modulus of multiwalled carbon nanotubes which can be mobilized to render reinforcing effects within composites.
A conservative approach is to neglect the contributions of inner tubes to the strength and
modulus of multi-walled carbon nanotubes when they are used in composites.
Carbon nanotubes have a unique electronic character which depends upon their chirality, and
ranges from high-conductivity metallic behavior to semiconducting with a large band gap.
Metallic carbon nanotubes exhibit ballistic transport with zero resistance along the tube, and they
can have conductivities several times that of copper. Because of their very low energy
2

dissipation, nanotubes carry tremendous current densities, higher than 100 MA/cm (MilliAmperes per Centimeter Square) for multi-walled carbon nanotubes, compared to tens of
2

kA/cm for superconducting wires. The thermal conductivity of carbon nanotubes is highly
anisotropic; it is diamond-like over the length of the tube, and insulting in the transverse
direction. The room-temperature thermal conductivity (along the length) of carbon nanotubes is
measured at 200 W/(m.K) (Watts per Meter Kelvin), compared to the theoretical prediction of
6,000 W/(m.K) (Breuer and Sundararaj 2004).
Carbon nanotubes have important surface characteristics, which affect their dispersion within a
matrix. The strong van der Waals interactions between the large SSA (Specific Surface Area) of
nanotubes encourage their agglomeration. These agglomerates may break by shearing along their
axes, causing failure between nanotubes at stresses much below the intrinsic capabilities of

5

individual nanotubes. The nanotube surface characteristics can be tailored through
functionalization.
1.2.2

Carbon Nanofibers

CNFs (Carbon Nanofibers) consist of stacked conical graphite planes with a hollow core (Fig.
o

1.5a); the graphite basal planes are oriented at about 25

with respect to the nanofiber

longitudinal axis (Tibbetts, Lake et al. 2007). TEM (Transmission Electron Microscope) and
SEM (Scanning Electron Microscope) images of CNFs are shown in Figs. 1.5b and 1.5c,
respectively.
CNFs generally have larger diameters (50-100 nm) when compared with multi-walled carbon
nanotubes. Their inner diameter ranges from 30 to 90 nanometer, and their length from 50 to 100
µm (micrometer) (with aspect ratios ranging from 100 to 500). The tensile strength and elastic
modulus of CNFs are about 3 GPa and 240 GPa, respectively, and their thermal conductivity is
quite high with a value of 1,950 W/(m.K). CNFs, when compared with carbon nanotubes, have a
higher defect concentration, which lowers their mechanical properties below those of carbon
nanotubes. On the other hand, CNFs are now produced at relatively large quantities, and their
current cost is much lower than that of carbon nanotubes (Breuer and Sundararaj 2004). One of
the most outstanding features of CNFs is the presence of a large number of edges on their
surfaces, which present sites readily available for functionalization. CNFs are generally vaporgrown; this process tends to deposit a layer of (less graphitic) carbon upon nanofibers which
could cement the nanofibers together (Tibbetts, Lake et al. 2007).

6

1.2.3

Graphite Nanoplatelets

GP (Graphite Nanoplatelets) (Fig. 1.6) are obtained by exfoliation of natural graphite; their
thickness can be controlled in a range from 1 to 25 nm (with about 0.5 nm thickness of each
grapheme layer), and their diameter from the sub-micron level to tens of microns. Graphite, in its
basal plane, offers elastic modulus of 1,060 GPa, thermal conductivity of 250 W/m.K, and
-5

electrical resistivity of 4x10

Ω.cm (Ohm Centimeter) (Kalaitzidou, Fukushima et al. 2007;

Kalaitzidou, Fukushima et al. 2007; Li, Sham et al. 2007). Stress concentrations at the rough
edges of graphite nanoplatelets lower their tensile strength below that of carbon nanotubes.

1.3 Merits of Graphite Nanomaterials
Typical properties of three graphite nanomaterials (multi-walled carbon nanotube, carbon
nanofiber and graphite nanoplatelet) are presented in Table 1.2, and compared with typical
properties of carbon fibers. Multi-walled carbon nanotubes are observed to be distinguished from
carbon fibers by their high aspect ratio, specific surface area, elastic modulus, tensile strength,
thermal conductivity, electrical conductivity and strain capacity. The advantages of graphite
nanomaterials over carbon fibers include high aspect ratio, specific surface area, and thermal and
electrical conductivity. The high surface area and elastic modulus of graphite nanoplatelets as
well as their distinct geometry distinguish them from carbon fibers. The costs of graphite
nanomaterials have experienced tremendous drops in recent years; their performance-to-cost
ratios are starting to match and surpass those of carbon fibers.
The geometric features of graphite nanomaterials provide them with distinct capabilities for: (i)
control of defect size and propagation within composites; (ii) energy dissipation via frictional
pullout; and (iii) adding to the tortuosity of crack propagation and diffusion paths. These unique

7

features of graphite nanomaterials are compared in the following with those of conventional
(micro-scale) carbon fibers.
Close spacing of discrete reinforcement systems in composites offer important advantages. With
proper selection of materials and processing conditions, the defect size in composites would be
limited to the spacing between the discrete reinforcement systems. For uniformly dispersed
discrete reinforcement (with different orientation conditions) within a matrix, the average
spacing (S) of the reinforcement can be expressed in terms of the reinforcement diameter (d) and
volume fraction (Vf) as S=K.d / Vf

0.5

. The coefficient K in this equation is a constant which

ranges from 0.8 to 1.38 depending on the reinforcement orientation condition (1-, 2- or 3-D) and
the assumptions made in the calculation. At 0.1% volume fraction, for the typical diameters
introduced in Table 1, the average reinforcement spacing within composites, calculated using the
above equation (with K=1), are 632, 3160 and 221,200 nm For multi-walled carbon nanotube,
carbon nanofiber and carbon fiber, respectively. The average spacing of carbon fiber is thus
about 350 times that of multi-walled carbon nanotube at similar volume fraction. This provides
nanotubes with superior capabilities for control of defect size and propagation within
composites.
A primary purpose of using discrete reinforcement systems in brittle matrices (including
cementitious materials) is to enhance their toughness characteristics. One key mechanism
contributing to fracture toughness is energy dissipation through frictional pullout (Fig. 1.7a) of
the discrete reinforcement, which is proportional to N1.(L/4).(π.d) where N1 is the number of
2

reinforcement crossing a unit area of the crack (N1=0.53Vf/d ), and L and d are the length and
diameter of the reinforcement. At a reinforcement volume fraction of 0.1%, with the typical

8

geometric attributes of Table 1.1, the frictional pullout energy dissipation (per unit crack area) of
carbon nanotube, carbon nanofiber and carbon fiber would be proportional to 0.42, 0.21 and
0.059, respectively. Hence, the (geometric) potential of multi-walled carbon nanotubes for
toughening of composites via frictional pullout is about an order of magnitude greater than that
of conventional (micro-scale) carbon fibers.
Another measure of the reinforcement efficiency of discrete reinforcement systems in
composites is the tortousity of a (crack or diffusion) path within the matrix and interfaces
(avoiding the reinforcement), as shown in Fig. 1.7b. This tortuosity can enhance the fracture
toughness and diffusion resistance of composites in the presence of discrete reinforcement
systems. The tortuosity factor (t) is defined as the ratio of the actual distance that a crack or
penetrant must travel (for avoiding the reinforcement) to the shortest distance that it would have
traveled in the absence of the reinforcement, and can be expressed (for reinforcement with
platelet geometry, length L and thickness T) as t=1+Vf.L/(2T). This equation can be strictly
applied only to carbon nanoplatelets; we assume that it would also be applicable to carbon
nanotube, nanofiber and microfiber. The tortuosity factor for carbon nanotube, carbon nanofiber,
graphite nanoplatelet and carbon microfiber at 0.1% volume fraction would thus be (for the
typical geometric attributes of Table 1.2) 1.5, 1.25, 1.15 and 1.07, respectively. While this
approximate approach could underestimate the tortuosity factor of graphite nanoplatelet, it still
shows that graphite nanomaterials, when compared with carbon fiber, are more effective in
enhancing the diffusion resistance and fracture toughness of composites.
Existing markets for fiber reinforced cement composites are dominated by steel, polypropylene
and glass fibers. Table 1.3 compares the key properties (governing reinforcement efficiency) of
multi-walled carbon nanotube with those of steel, polypropylene and glass fibers. This
9

comparison further highlights the distinctly high reinforcement efficiency of graphite
nanomaterials in cement-based materials.

1.4 The Need and Routes for Functionalization of Graphite Nanomaterials
Successful use of graphite nanomaterials in cement-based matrices requires: (i) thorough
dispersion of individual nanomaterials within the aqueous fresh mix; and (ii) adequate bonding
of cement hydration products to the nanomaterial surfaces for effective stress transfer across
interfaces. The surfaces of graphite nanomaterials are, disregarding the presence of minor
defects, hydrophobic and atomically smooth (i.e., without sites for primary bond development).
The hydrophobic surfaces of graphite nanomaterials would, in an aqueous solution (e.g., fresh
cement mix), have the energetic preference for bonding to each other (i.e., agglomerating) versus
exposing their surfaces to the aqueous environment. Agglomeration of nanotubes undermines all
key features governing their reinforcement efficiency in composites. For example, the
mechanical properties of nanotube agglomerates are dominated by the secondary (weak)
intertube interactions and not the mechanical properties of individual nanotubes. Fig. 1.8 shows
the sharp drop in elastic modulus of nanotube agglomerates with increasing diameter (number)
of agglomerated nanotubes. Agglomeration also compromises the unique geometric advantages
of nanotubes (high surface area and aspect ratio, close spacing, and large tortuosity factor). The
hydrophobic surfaces of carbon nanotubes can be made hydrophilic through introduction of
selected functional groups to facilitate their thorough dispersion in aqueous media.
Besides thorough dispersion in aqueous media, effective bonding to cement hydration products is
another criterion governing the reinforcement efficiency of graphite nanomaterials in cementbased materials. The atomically smooth surfaces of graphite nanomaterials do not offer any sites
for development of strong (primary) bonds to cement hydration products (Fig. 1.9a). The absence
10

of such bonds prevents effective mechanical and physical interactions (e.g., for stress and heat
transfer) between nanotubes and the cement-based matrix. Therefore, the distinctly high
mechanical and physical attributes of nanotubes cannot be mobilized toward enhancement of the
cementitious matrix. In order to develop primary bonds between cement hydration products and
graphite nanomaterials, one can introduce functional groups via primary bonds (to carbon atoms)
upon nanotubes, which can also develop primary bonds with cement hydration products (Fig.
1.9b). Functional groups, when present at adequate concentrations on nanotube surfaces, would
provide enhanced interfacial stiffness, strength, thermal conductivity, etc. for effective
mechanical and physical interactions between carbon nanotubes and cement hydration products.
These interfacial interactions enable mobilization of the distinct mechanical and physical
attributes of graphite nanomaterials towards enhancement of the performance characteristics of
cement-based materials.
Hydrated cement comprises primarily of calcium silicate hydrate and calcium hydroxide.

An

effective way of interacting with these products of cement hydration is through introduction of
COOH functional groups upon graphite nanomaterials.

These functional groups promote

formation of strong covalent bonds between carbon nanotubes and cement hydration products,
and therefore increase the load transfer efficiency at the interface (and lower the interfacial
electrical and thermal resistance). Introduction of COOH functional groups also makes the
nanotube surfaces more hydrophilic, and thus benefits their dispersion in aqueous media.
Refluxing of carbon nanotubes in a H2SO4/HNO3 mixture yields functionalized nanotubes with
main functionalities comprising -COOH, -C=O, and –OH groups (Fig. 1.10) approximately in
4:2:1 proportions. The concentration of these groups in nanotubes treated by different oxidants
varies in the range of 2 x 10

20

20

to 10 x 10

sites/gram. On a molar basis, the concentration of
11

functional groups is about 4% to 6%. The carboxyl and other functional groups introduced upon
carbon nanotubes represent useful sites for covalent bonding to cement hydration products (and
also to other molecules for introduction of a wide range of functional moieties for diverse fields
of application) (Balasubramanian and Burghard 2005).
The acid-functionalization process introduces the functional groups at defects which are either
pre-existing or are generated in the course of acid-functionalization. Typically around 1-3% of
the carbon atoms in a nanotube are located at defect sites formed during its synthesis. Acidfunctionalization also introduces defects (holes) the concentration of which depends upon the
severity of the acid exposure conditions. Carboxyl and other oxygen-containing functional
groups bond to nanotubes at such defect sites. The defects associated with the functional groups
introduced via functionalization tend to compromise the mechanical and physical characteristics
of nanotubes; such negative effects, however, can be minimized with less vigorous conditions
such as refluxing with nitric acid (Balasubramanian and Burghard 2005).
Dispersion of graphite nanomaterials in the aqueous fresh cement mix environment benefits from
the hydrophilic nature of functionalized nanomaterials and also from the reduction in of van der
Waals interactions in the presence of carboxyl groups, which strongly facilitates the separation of
nanotube agglomerates into individual nanotubes (Balasubramanian and Burghard 2005).
The association of acid-functionalization (oxidation) with defects implies some damage to the
mechanical and physical properties of acid-functionalized graphite nanomaterials. “Addition
functionalization” offers the potential for less association of functional groups with defects.
Addition reactions involve direct coupling of functional groups onto the conjugated carbon
framework of nanotubes. The required reactive species are in general made available through
thermally activated reactions. A variety of addition reactions are possible, and the most

12

important one are schematically shown in Fig. 1.11a. In principle, addition reactions can be
initiated exclusively on the intact sidewall (in the absence of defects), or in parallel at defect sites
from where the reaction could proceed further. An exception is fluorination, for which direct
addition to defect-free sidewalls is viable (Balasubramanian and Burghard 2005). Slightly
elevated temperatures are needed to overcome the activation barrier to fluorination. The fluorine
atoms in fluorinated graphite nanomaterials can be replaced through nucleophilic substitution
reactions with relative ease, thus opening a flexible approach for providing the sidewall with
various types of functional groups, as illustrated in Fig. 1.11b.
Analogous with functionalization with carboxyl groups, the direct covalent attachment of
functional moieties followed by substitution reactions can be used to significantly facilitate
dispersion of graphite nanomaterials in aqueous media (Balasubramanian and Burghard 2005).
The enormous specific surface area of nanotubes together with their high aspect ratio and relative
flexibility generates strong tendencies toward agglomeration and entanglement. Aggressive
covalent treatment (e.g., acid-functionalization) of carbon nanotubes could compromise their
desirable performance attributes. Non-covalent treatment could be used to further facilitate the
dispersion of nanotubes at increased concentrations in aqueous media without damaging the
nanotubes (except for altering their electrical properties). Non-covalent surface treatment by
surfactants is a versatile route to enhance the dispersion of nanotubes in aqueous media.
Surfactants (surface acting agents) are amphilic compounds, containing both hydrophobic groups
(tails) and hydrophilic groups (heads). The association of hydrophobic tails with nanotubes and
the hydrophilic heads with water (schematically depicted in Fig. 1.12) is the fundamental
mechanism facilitating dispersion of nanotubes in aqueous media with surfactants.

13

In a typical approach to dispersion of nanotubes with surfactants, the nanotubes are mechanically
exfoliation, and the surfactant is adsorbed on their surfaces; further ultrasonication then helps the
surfactant to debundle nanomaterials by steric or electrostatic repulsions. Ultrasonic treatment of
carbon nanotube agglomerate (Fig. 1.13i) generates high local shear, particularly at the nanotube
bundle end (Fig. 1.13ii). Once spaces or gaps at the bundle ends are formed, they are propagated
by surfactant adsorption (Fig. 1.13iii), ultimately separating the individual nanotubes from the
bundle (Fig. 1.13iv). Generally, ionic surfactants are preferred for dispersion of carbon
nanotubes in aqueous media.
Carbon nanotubes can be dispersed in water when coated by adsorbed surfactants, preferentially
with those that have relatively high HLB (Hydrophile-Lydrophyle Balance). The nature of
surfactant, its concentration, and type of interaction play crucial roles in the dispersion process.
Knowing the surface charge of carbon nanotubes in different media is essential for understanding
the interaction (adsorption) mechanism with ionic surfactants, and to predict colloidal stability of
nanotube dispersions. The adsorption mechanism of ionic surfactants, which is promoted by
electrostatic interactions with the nanotube surface, is heavily controlled by the purification
process and functionalization of nanotubes, which in turn determine their surface charge. Fig.
1.14 shows examples of the structure of adsorbed surfactants on carbon nanotubes.
Experience with the use of functionalized carbon nanotubes in biominerals and ceramics indicate
that introduction of a coating of matrix on nanotubes (by precipitation of biominerals or
adsorption of nanoparticles) yields important gains in performance characteristics of inorganic
matrix nanocomposites (Wang, Ye et al. 2007). An example involved reinforcement of calcium
phosphate cement with carboxyl-functionalized carbon nanotube (for development of a new bone
repair material). When calcium phosphate cement was exposed to simulated body fluid prior to

14

introduction into calcium phosphate cement, a uniform layer of ydroxyapatite precipitated on the
functionalized nanotubes (Fig. 1.15). The resulting (biomineralized) carbon nanotubes proved to
be a more effective reinforcement in calcium phosphate cement when compared with
functionalized (but not biomineralized) nanotubes (Wang, Ye et al. 2007). Similar improvements
in the reinforcement efficiency of carbon nanotubes in ceramic matrices have been detected
when the acid-functionalized nanotubes are coated with ceramic nano-particles (Fig. 1.16) prior
to their use for processing ceramic matrix nanocomposites (Gao, Jiang et al. 2006).

1.5 Fresh Cement Past and Hydrates: Morphological Relationships with
Graphite Nanomaterials
Carbon nanotubes (and other graphite nanomaterials) offer exceptional mechanical, physical and
functional attributes. Development of cement-based nanocomposites which make effective use of
the exceptional mechanical, physical and geometric attributes of graphite nanomaterials requires
thorough dispersion and intimate interfacial interactions of the nanomaterials within matrix.
Efforts to meet these requirements encounter particular challenges relating to the morphological
attributes of cement-based matrices. The fresh cementitious paste is a dispersion of micro-scale
particles in an aqueous solution, which complicates the dispersion process of graphite
nanomaterials. After curing, conventional cement-based matrices incorporate large quantities of
micro- to nano-scale pores which mitigate intimate interfacial contacts over the surface area of
graphite nanomaterials.
Fig. 1.17a depicts the relative dimensions of carbon nanotubes, cement particles, and carbon
fibers. The micro-scale dimensions of carbon fibers facilitate their dispersion within a suspension
of micro-scale cement particles (Fig. 1.17b). Graphite nanomaterials with at least one dimension

15

that is three orders of magnitude smaller than cement particles, on the other hand, would be
squeezed into the space between the micro-scale particles in fresh cement paste (Fig. 1.17c). This
phenomenon increases the chances for agglomeration of the nanomaterials.
The potential for thorough dispersion of carbon nanotubes and other graphite nanomaterials
within a cementitious matrix can be improved through partial replacement of micro-scale cement
particles with nano-scale silica fume particles in order to improve generate a smooth particle size
distribution which facilitates uniform dispersion of nanomaterials. Fig. 1.18a depicts the relative
dimensions of carbon nanotubes, silica fume particles and cement particles. Cementitious pastes
with relatively high replacement levels of cement with silica fume are referred to as DSP
(Densified with Cement Particles). The fresh DSP cementitious paste offers a particle size
distribution (in aqueous solution) which favors dispersion of carbon nanotubes, Fig. 1.18b.
During hydration, the cement particles (dispersed in water) react with water and form hydration
products which partially fill the space between them, Fig. 1.19a; the unfilled fraction of space
forms micro-scale capillary pores. The structure of hydrated cement, Fig. 1.19b, comprises nanoscale C-S-H (Calcium Silicate Hydrate) particles (their van der Waals interaction is the primary
binding force within hydrated cement paste), micro-scale C-H (Calcium Hydroxide) crystals,
capillary pores, gel pores within calcium silicate hydrate, and other hydration products. Capillary
pores constitute several percent (depending on the water/cement ratio) of hydrated cement. The
micro-scale calcium hydroxide crystals have a relatively low specific surface area and thus
limited binding capability. The prevalence of micro-scale pores and crystals limit the potential
for effective interactions between hydrated cement and graphite nanomaterials, Fig. 1.19c.
Improving the gradation of particles in fresh cement paste by introduction of (nano-scale) silica
fume particles, Fig. 1.18b, together with the use of superplasticizers (surfactants) to lower the

16

required water content (for fresh mix workability) yield dense hydrated cements, Fig. 1.19d,
where micro-scale calcium hydroxide crystals and (through pozzolanic reactions) converted to
calcium silicate hydrate (comprising nano-scale particles with desirable binding qualities). The
improved physical compatibility of the refined cement hydrates with graphite nanomaterials,
compare Figs. 1.19c & 1.19d, are expected to enhance the reinforcement efficiency of
nanomaterials in modified cement-based matrices. Further gains would be realized by
functionalization of graphite nanomaterials to improve their chemical compatibility with cement
hydrates.

1.6 Research Hypothesis
This research was conducted primarily with two hypotheses in mind:


Properly functionalized graphite nanomaterials when used in an appropriately graded
matrix can bring about substantial increase in all engineering and durability
characteristics of high-performance cement based matrices.



Complementary use of nano- and micro-scale reinforcement can provide cementitious
materials with desired balances of engineering properties.

1.7 Research Objectives
In order to verify the above hypothesis, the main thrust of this research is to develop and
thoroughly characterize high-performance cement-based materials (paste, mortar and concrete)
reinforced with functionalized graphite nanomaterials and suitable micro-scale fibers. A
theoretical framework will also be developed to rationalize the experimental results and to guide
further development of the technology. The objectives to be accomplished in the project are
presented below together with the approach followed for meeting each objective.

17

Objective 1. Select cementitious materials, mix designs and production conditions for
enhancing the reinforcement efficiency of graphite nanomaterials.
Cementitious materials were selected to effectively interact with nanomaterials of distinctly high
specific surface area, close spacing and fine geometry in order to enhance the benefits of
functionalized nanomaterials to mechanical, barrier and durability characteristics. Fundamental
studies using cementitious paste provided the basis to design and evaluate cementitious
nanocomposites incorporating fine and eventually coarse aggregates for introduction into
mainstream concrete construction practices.
Objective 2. Identify surface functionalization conditions which benefit the dispersion and
reinforcing action of graphite nanomaterials in cement-based matrices.
Covalent functionalization techniques as well as non-covalent (physisorption) methods were
evaluated for enhancing the dispersion of graphite nanomaterials in aqueous media, and their
interfacial bonding to cement hydrates. Due consideration was given to achieving desired
compatibility with aqueous media and cement hydrates without excessively damaging the
nanomaterials.
Objective 3. Make complementary use of graphite nanomaterials and micro-scale discrete
reinforcement towards achieving balanced gains in engineering properties.
The potential for positive interactions between nano- and micro-scale reinforcement in
cementitious matrices were evaluated. Optimization experimental programs were implemented in
order to identify hybrid reinforcement systems which bring about balanced gains in key
engineering properties of cementitious materials. The experimental results were used to verify
and refine the theoretical models.

18

Objective 4. Thoroughly characterize selected cement-based materials with nano-scale
and hybrid reinforcement systems.
Selected cementitious (mortar and concrete) materials with optimum nano-scale and hybrid
reinforcement systems were thoroughly characterized through evaluation of their mechanical,
barrier, durability and physical characteristics as well as their structure and failure mechanisms.
Control tests were conducted on cementitious matrices. Statistical analyses were undertaken to
verify the performance gains realized by cementitious mortar and concrete materials through
nano-scale and hybrid reinforcement.
Objective 5. Develop a theoretical framework for the reinforcing action of graphite
nanomaterials in cement-based matrices. Assess competitive performance
and cost of cement-based nanocomposites, and identify priority applications.
The interfacial interactions, geometric characteristics, spacing and percolation conditions of
functionalized (and non-functionalized) graphite nanomaterials in cementitious matrices were
used towards development of theoretical models for prediction of the mechanical performance,
moisture transport characteristics and electrical conductivity of cementitious materials with
nano-scale and hybrid (nano- and micro-scale) reinforcement systems as evident from the
experimental results. The performance characteristics and economics of cementitious mortar and
concrete materials with nano-scale and hybrid reinforcement systems were evaluated against
those of high-performance concrete materials (without and with micro-scale discrete
reinforcement) in order to identify priority applications for cementitious nanocomposites.

19

1.8 Scope
The research undertaken to address the above objectives is organized in a number of chapters, as
described below:


Chapter 1 provides introduction to different nanomaterials, a general background and
considerations for use of nanomaterials in cement-based matrices, and the objectives of
this study.



Chapter 2 covers a review the literature on the use of graphite nanomaterials in cementbased matrices, identifying key knowledge gaps in the field.



Chapter 3 deals with the preliminary results generated on the use of graphite
nanomaterials (both pristine and functionalized) and carbon microfibers at relatively low
volume fractions in high-performance cementitious paste and mortar matrices.



Chapter 4 presents an evaluation of different covalent and non-covalent surface
modification techniques for enhancing the concentration, dispersion and interaction of
carbon nanotubes in high-strength cementitious matrix.



Chapter 5, 6 and 7

present the results of addition of various multiwalled carbon

nanotubes, non-functionalized as well as functionalized, at different volume fraction in
high-performance (densified with small particles - DSP) paste and mortar matrices along
with the use of hybrid reinforcement with different microfibers, and outcomes of
statistical analyses and optimization efforts.


Chapter 8 evaluates the effects of nanotubes on the electrical conductivity of the
cementitious mortar and the damage-sensitivity of electrical conductivity under
compressive loads.

20



Chapter 9 presents the experimental results of addition of non-functionalized and PAAphysisorbed multiwalled carbon nanotubes and their hybrid combination with different
microfibers at different volume fraction on the engineering properties and microstructure
of DSP concrete. Whereas, Chapter 10 compares the results of reinforcement of high
strength concrete with PAA-physisorbed multiwalled carbon nanotubes and its hybrid
combination with carbon microfibers.



Chapter 11, 12 and 13 presents effects on the engineering properties of high-performance
DSP paste, mortar and concrete matrices with the addition of functionalized and nonfunctionalized nanofibers at different volume fraction along with the use of hybrid
reinforcement with different microfibers, and outcomes of statistical analyses and
optimization efforts.



Chapter 14 deals with the effects on the engineering properties of high-performance DSP
paste, mortar and concrete matrices with the addition of functionalized and graphite
nanoplatelets at different volume fraction along with the use of hybrid reinforcement with
different microfibers, and outcomes of statistical analyses and optimization efforts.



Chapter 15 and 16 present the experimental results on the effects of modified
nanomaterials and their hybrid combinations on the durability characteristics of highperformance cementitious materials.



Chapter 17 presents the results of plant production and field project highlighting the
challenges and results of first ever scale up plant production of high-strength concrete
nanocomposites.

21



Chapter 18 develops a theoretical framework for contributions of graphite nanomaterials
(emphasizing carbon nanotubes) to diverse aspects of concrete engineering properties and
competitive performance and cost analyses of cementitious nanocomposites.



Chapter 19 reviews the main findings and conclusions of the research.

22

Table 1.1- Applications, performance requirements, and competing fibers.
Application

Primary Performance Considerations

Competing
Fibers
Explosion-resistant
Impact resistance, energy absorption Steel, Glass
infrastructure systems (e.g., capacity, ductility, fire resistance, weight
military shelters and mission- (thickness) reduction
critical structures), armor, antiblast doors, security products
Tunnel linings, water and Flexural strength, toughness, ductility, Steel, Glass
sewage
network
linings, abrasion resistance, moisture diffusion
hydraulic
structures
wear resistance, durability, fatigue life, shrinkage
linings
crack control, adhesion capacity, fire
resistance
Repair and retrofit of civilian Ductility, tensile and flexural strength, Steel,
and military infrastructure adhesion capacity, shrinkage crack control, Polypropylene,
systems (bridges, building toughness and energy absorption capacity, Glass
structures, shelters, pavements, impact resistance, fatigue life, abrasion
etc.)
resistance, durability
Water, wastewater and nuclear Flexural strength, ductility, moisture Steel,
waste
containment diffusion resistance, durability, fatigue life, Polypropylene,
infrastructure systems (pipes, shrinkage crack control, fire resistance, Glass
canals, tanks, containers)
impact resistance
Thin-sheet products (roofing, Flexural and tensile strength, toughness, Cellulose,
siding)
ductility, impact resistance, moisture PVA,
diffusion
resistance,
durability,
fire polypropylene,
resistance, shrinkage crack control
Glass, Carbon
Highway
and
airfield Fatigue life, durability, flexural strength, Steel,
pavement,
bridge
decks, ductility, abrasion resistance, impact Polypropylene,
parking garage and industrial resistance, shrinkage crack control
Glass
floors
Structural components (beams, Flexural and tensile strength, ductility, Steel, Glass
columns, walls, joints)
toughness, energy absorption capacity,
impact resistance, bond strength to
conventional reinforcement

23

Table 1.2- Typical properties of graphite nanomaterials and carbon fiber.
Graphite
Nanomat.

Elastic
Modulus,
*
GPa

Tensile
Strength,
*
GPa

Thermal
Conductivity,
*
W/m.K

Dia.,
nm

Aspect
Ratio

MWNT

20

20

1,000

Surface
Area,
2
m /g
500

Nanofiber

50

100

500

500

Nanoplatelet
*

Length, μm

3

10

300

100

1,000

~(10-20)

250

CarbonFiber

1,000

7,000

140

16

350

3

Cost,
$/kg

600

***

1,200
240

***

350
3

***

2,800
1,950

**

45

****

10

*****

10

12

Property measured in basal plane or along length

**

Mid-term price target

***

Measurements are ‘true’ values based on wall cross-sectional area

****

Mid-term price target- Pyrograf Product Inc; a subsidiary of ASZ

*****

Mid-term price target- XG Sciences

Table 1.3- Key properties governing the reinforcement efficiency of multi-walled carbon nanotubes versus conventional steel,
polypropylene and glass fibers.
Reinforcement

Aspect
Ratio

Density,
3
g/cm

Nanotube

*

Dia
μm
0.02

1,000

Steel
Polypropylene
AR Glass

540
50
14

65
450
700

2.1
7.8
0.9
2.7

*

Surface
Area,
2
m /g

Elastic
Modulus,
GPa

500

1,200
200
4
72

0.00095
0.089
0.11

*

Tensile
Strength,
GPa
*

350
1.2
0.35
1.7

Strain
Capacity,
%

Tortuosity
Factor @
1 Vol.%

20

Spacing
@1
Vol.%,
μm
0.2

20
15
2.4

5,400
500
140

1.3
3.2
4.5

6

Measurements are ‘true’ values based on wall cross-sectional area (‘apparent’ values based on total cross-sectional area would by
75% of ‘true’ values, assuming a typical inner-to-outer diameter ratio of 0.5 for MWNT)

24

For the interpretation of the references to colors in this and all other figures, the reader is referred
to the electronic version of this dissertation

(a) Graphite nanomaterials

(b) Micro-scale carbon fiber
Figure 1.1- Arrangement of graphene sheets in carbon nanotube and nanofiber, graphite
nanoplatelet, and carbon micro-fiber.

Figure 1.2- Carbon nanotube

25

Figure 1.3- Multi-walled carbon nanotubes.

Figure 1.4- Roll-up of a graphene sheet to form different nanotube types.

26

(a) Schematics

(b) TEM Images
Figure 1.5- Schematic depiction and electron microscope images of carbon nanofiber.

27

(a) SEM image

(b) TEM image

Figure 1.6- SEM and TEM images of exfoliated graphite nanoplatelets (Li, Sham et al. 2007).

(a) Frictional pullout

(b) Tortuosity of Paths

Figure 1.7- Reinforcement mechanisms of discrete reinforcement systems in composites.

28

Figure 1.8- Loss of mechanical properties of carbon nanotubes due to agglomeration (Ajayan et
al, 2003).

29

(a) Lack of primary bonding with pristine nanotube

(b) Primary bonding with functionalized nanotube
Figure 1.9- Schematics of the interactions of pristine and functionalized nanotubes with cement hydration products.

30

Figure 1.10- Schematic depiction of acid-functionalization of carbon nanotube.

Figure 1.11- Fluorination followed by ethylcarboxylation of carbon nanotube.

31

Figure 1.12- Schematic depiction of the dispersion mechanism of surfactants.

Figure 1.13- Mechanisms of nanotube separation from bundle by ultrasonicaiton and surfactant
stabilization (Vaisman, Wagner et al. 2006).

32

(a) Encapsulation in a cylindrical surfactant micelle

(b) Hemimicellar adsorption of surfactant molecules

(c) Random adsorption of surfactant molecules
Figure 1.14- Schematic presentation of the surfactant assembly structures on carbon nanotubes (Vaisman, Wagner et al. 2006).

33

Figure 1.15- TEM image and selected area diffraction pattern of biomineralized nanotube in
simulated body fluid (SBF) for use in calcium phosphate cement (Wang, Ye et al. 2007).

Figure 1.16- TEM image of carbon nanotube coated with ceramic nanoparticles (Gao, Jiang et
al. 2006).

34

(a) Relative dimensions

(b) Carbon fiber in cement paste

(c) Carbon nanotube in cement paste

Figure 1.17- Relative dimensions and dispersion conditions of carbon nanotubes and fibers.

(a) Relative dimensions

(b) CNT in cement (left) & cement/silica fume (right)

Figure 1.18- Improved dispersion of carbon nanotubes in DSP (cement/silica fume) cement
paste.

(a) Hydration of cement particles partially filling the space between them
Figure 1.19- The interactions of cement hydrates with graphite nanomaterials

35

Figure 1.19 (cont’d)

(b) Structure of hydrated cement depicted with carbon nanotube length scale

36

Figure 1.19 (cont’d)

(c) Conventional

(d) Compact

37

CHAPTER 2

2

LITERATURE REVIEW, MATERIALS AND METHODS

2.1 General
Advances in concrete admixtures have played enabling roles in development of highperformance concrete materials with increasing strength levels. These high-performance concrete
materials, however, tend to be brittle with low tensile strength and sensitivity to early-age
microcracking under restrained (drying/autogeneous) shrinkage and thermal stresses. These
drawbacks in development of high-performance and ultra-high-performance concrete materials
can be addressed through discrete (millimeter- and micrometer-scale) fiber reinforcement
Cementitious materials, however, incorporate flaws at different scales; finer flaws with nanoscale size and spacing challenge the effectiveness of conventional reinforcing fibers; these finer
flaws can form and grow within the cementitious matrix in between conventional (coarser) fibers
without encountering the fibers. The use of nano-scale fibers (with nano-scale spacing) in highperformance cementitious materials can overcome this drawback of conventional reinforcing
fibers.
Graphite Nanomaterials, including carbon nanotubes (CNTs), present several distinct advantages
over conventional (micro-scale) fibers as reinforcement in high-performance cementitious
materials. Graphite nanomaterials generally provide significantly higher strength and modulus
than conventional fibers. These nanomaterials also provide distinctly high specific surface areas
and aspect ratios, which benefit their interactions with the cementitious matrix and effectiveness

38

in crack arrest. Provided that the nanomaterials are uniformly dispersed within the cementitious
matrix, due to their nano-scale cross-sectional dimensions, the spacing between nanomaterials
tends to be very small. The inception and propagation of microcracks in the space between nanoscale reinforcement would thus be effectively hindered.
Few attempts have been made to introduce graphite nanomaterials as reinforcement in
cementitious matrices. A comprehensive review of the literature in this field is presented in the
following. This review covers material selections, the approach to dispersion of nanomaterials,
and their mixing into the cementitious matrix; benefits of nanomaterials to performance
characteristics of cementitious materials are also discussed.

2.2 Experimental Studies
The addition of carbon nanotubes (CNTs), carbon nanofibers (CNFs) and exfoliated graphite
nanoplatelets (GPs) to polymer-based matrices is among the highly developed areas of research.
The results indicate that CNTs and CNFs can tremendously benefit the strength and energyabsorbing capacity of polymers, and also improve their electrical and thermal conductivity.
Graphite nanoplatelets are incorporated into polymer matrices as low-cost alternatives to CNTs
for realizing such benefits as vibration damping, and electric and heat conduction.
Applications of graphite nanomaterials in cementitious matrices are at early stages of
development. A summary of notable research on nanocomposites involving CNTs is presented in
Table 2.1. For each investigation, the objectives, the test parameters, the test methods/features,
and major conclusions have been summarized in this table. In addition, the strengths and
drawbacks, if any, of each investigation are also presented. Additional details, including
significant findings of these studies are discussed in the following.

39

2.2.1

Carbon Nanotubes

In one of the earliest uses of carbon nanotubes in cement-based materials, reported by Fu et
al.(Fu and Chung 1998), it was found that surface treatment of carbon by ozone gas (0.3 vol% in
o

air, 160 C, 10 minutes) can benefit the tensile and compressive strengths or cementitious
nanocomposites.
Campillo et al. tested carbon nanotube/cement nanocomposite under compression, and found that
single- and multi-walled carbon nanotubes increased the strength of plain cement paste by 6%
and 30%, respectively.
Kowald (Kowald 2004) found that 0.5% (by weight of cement) multiwalled carbon nanotubes
(MWNTs) produced 12% gain in the compressive strength or the cementitious matrix; the
consistency of the paste (i.e. the ability of freshly mixed cement paste to flow) was, however,
reduced substantially, and pressures up to 125 MPa had to be applied for more than 45 min to
adequately compact the fresh cementitious nanocomposite.
Makar et al.(Makar and Beaudoin 2004; Makar, Margeson et al. 2005) investigated the
reinforcing effect of 2.0 wt.% CNTs in cement using Vickers hardness measurements and
scanning electron microscope observations. The results indicated that carbon nanotubes may
affect the early progress of hydration, producing higher hydration rates.
Li et al.(Li, Wang et al. 2005) used multiwalled carbon nanotubes, treated by sulfuric and nitric
acid, at 0.5% by weight of cement, in a cement mortar (0.45:1:1.5). Experimental results
indicated that the compressive and flexural strength of mortar could be increased by 19% and
25% with the addition of nanotubes. Moreover, the addition of carbon nanotubes reduced the
porosity and produced a finer pore size distribution in the cementitious mortar. It is worth

40

mentioning that carbon fibers at similar volume fractions could increase the flexural of the
cementitious mortar by 22%, but produced loss of compressive strength (Table 2.2).
Ibarra et al.(Saez De Ibarra, Gaitero et al. 2006) used both single walled carbon nanotubes
(SWNTs) and multiwalled carbon nanotubes (MWNTs) in cement paste matrix at 0.05 to 0.20
percent by weight of cement. Dispersion of nanotubes was found to be a key to their successful
use in cement paste. When gum Arabic was used to disperse nanotubes in the aqueous
environment, the addition of nanotubes benefited the Young’s modulus and hardness of cement
paste; this was particularly true in the case of single walled carbon nanotubes. Without gum
Arabic, however, the addition of nanotubes damaged the mechanical performance of cement
paste.
Cwirzen et al.(Cwirzen, Habermehl-Cwirzen et al. 2008) investigated the effects of surface
treatment on the wetability of multi-walled carbon nanotubes (MWNTs) and the mechanical
properties of cement pastes incorporating dispersions of both pristine and acid functionalized
MWNTs. Experimental results indicated that stable and homogenous dispersions of MWNTs in
water can be produced through surface functionalization of nanotubes followed by decoration
with polyacrylic acid polymers. Cement pastes incorporating dispersions of these nanotubes
provided good workability and an increase in the compressive strength of nearly 50% even with
only a small addition of the MWNTs, namely 0.045–0.15% by weight of cement. These positive
results may have resulted from the existence of chemical bonds between the OH groups of the
functionalized MWNTs and the C–S–H phase of the cement matrix, which benefits interfacial
stress transfer. The effects of decoration of MWNTs with polyacrylic acid polymers and gum
Arabic were also investigated. These dispersions appeared to be homogeneous only for
approximately 2 hours, after which a progressive sedimentation occurred. Good workability was

41

found for the cement pastes produced with all of nanotube dispersions; the only significant
difference was the slower hydration of the cement paste incorporating gum Arabic. When the
amount of gum Arabic was as high as 0.8%, hydration of the cement was hindered for over 3
days.
Musso et al.(Musso, Tulliani et al. 2009) compared the flexural and compressive of unreinforced
cement mortar (0.4:1:1.5) versus three different cement-based composites containing as-grown,
annealed and functionalized multiwalled carbon nanotubes at 0.5% by weight of cement. The
phase composition of composites was also evaluated by thermogravimetric analysis (TGA)
coupled with mass spectroscopy (MS), and the mineralogy and microstructure were analyzed
using X-ray diffractometer (XRD) and scanning electron microscope (SEM) techniques. Flexure
test results indicated that the addition of carboxyl functionalized MWNTs lowered the strength
of cement-based materials by a factor of 2.5; addition of pristine MWNTs, on the other hand,
produced close to 34% rise in flexural strength. Annealed MWNTs produced a slight gain of
about 9% in the flexural strength of cement-based materials. Comparable trends were observed
in compressive strength test results. While carboxyl-functionalized nanotubes produced a sharp
drop in compressive strength, addition of pristine nanotubes led to 10-20% rise in compressive
strength, Figure 2.1.
Konsta-Gdoutos et al.(Metaxa, Konsta-Gdoutos et al. 2010) found that small amounts of
effectively dispersed MWNTs (0.025–0.08 wt.% of cement) can significantly increase the
strength and stiffness of the cementitious matrix. In particular, lower dosages (0.025–0.048
wt.%) of relatively long MWNTs produced desired reinforcement effects; higher dosages (close
to 0.08 wt.%) of short MWNTs were required to render similar reinforcement effects. Nanoindentation test results indicated that MWNTs can strongly modify and reinforce the

42

nanostructure of the cementitious matrix. Compared to plain cement matrix, nanocomposites
appeared to have a larger amount of high-stiffness C–S–H accompanied with reduced nanoporosity. Due to their small diameters (20–40 nm), MWNTs appear to specifically reduce the
amount of fine pores. This phenomenon leads to a reduction of capillary stresses, resulting in a
beneficial effect on the early strain capacity of nanocomposites.
Nasibulin et al. (Nasibulin, Shandakov et al. 2009) synthesized “Cement Hybrid Material”
(CHM) in a simple, one-step process by the chemical vapor deposition (CVD), Figure 2.2, where
cement particles were used both as catalyst and as support material. This approach allowed
synthesis of novel hybrid nano-structured materials, wherein carbon nanotubes and nanofibers
were attached to cement particles, facilitating dispersion of carbon nanomaterials in the cement
matrix. Investigations of the physical properties of the paste made of the CHM revealed as high
as 2 times increase in compressive strength and 40 times rise in the electrical conductivity after
28 days of curing in water.
Makar and Chan (Makar and Chan 2009) examined the effect of 1% SWNT on the early stages
of hydration of ordinary Portland cement using isothermal conduction calorimetry, highresolution SEM and TGA (Thermogravimetric Analysis). Introduction of non-functionalized
SWCNT bundles distributed by sonication in isopropanol on cement grains accelerated the
hydration of cement as compared both with control cement-based materials sonicated alone and
with as-delivered cement. SEM images showed the formation of C-S-H on the surface of
nanotube bundles. The images also showed evidence of classical reinforcing behavior including
bridging and pullout at microcracks.
Manzur and Yazdani (Manzur and Yazdani 2010) explored the effects of treated and untreated
CNTs at different dosages on the mechanical properties of cement mortars with different w/c

43

ratios. An increase in mean strength was observed up to 0.5% MWNT addition with both types
of MWNT; addition of 0.3% MWNTs produced the highest mean compressive strengths. The
maximum gains in mean compressive strength were about 29% and 25% at 7 and 28 days,
respectively, relative to the control samples. A smaller-sized MWNT produced higher mortar
compressive strength. Smaller MWNTs were distributed at a much finer scale, and therefore
filled the nano-pore space within the cement matrix more efficiently.
Wille and Loh (Wille and Loh 2010) designed a hybrid fiber-reinforced ultra-high-performance
concrete (UHPC) that incorporated different sizes of fibers ranging from nanometer to millimeter
length scales. MWNTs in the amount of 0.022% (relative to cement weight) were included in the
mix design. The bond behavior of steel fibers pulled out of UHPC significantly increased with
the addition of low concentrations of MWNTs to the concrete mix. MWNTs showed no
significant influence on mechanical properties such as compressive and flexural strength.
Collins et al. (Collins, Lambert et al. 2012) reported results of preliminary investigations into the
dispersion and rheology of cement pastes reinforced with carbon nanotubes, with and without
dispersants/surfactants. Most paste mixtures were composed with water/cement ratio of 0.5.
Mini-slump consistency increased with increasing w/c ratio (as expected). Addition of carbon
nanotubes (at 0.5%, 1%, and 2% by weight of cement), significantly reduced the consistency of
cement paste. Despite ultrasonication, the compressive strength of cement pastes decreased with
increasing nanotube content. The paste with w/c of 0.4 demonstrated the most significant drop in
compressive strength with the addition of nanotubes; at 2% CNT, compressive strength was only
24% of the compressive strength of the control paste made without nanotubes. The consistence
of cement paste reinforced with carbon nanotube greatly improved with the addition of
polycarboxylate-based superplasticizer, with a highly flowable mixtures achieved with w/c ratios

44

as low as 0.35. This cement paste provided a compressive strength that was 25% higher than that
of the control cement paste made without nanotubes.
Chan and Andrawes (Chan and Andrawes 2010) utilized finite element (FE) models to study the
improvements in tensile strength and ductility of cement brought about by the use of carbon
nanotubes (CNTs) as reinforcement. Nano-scale and macro-scale models were used to
investigate the interfacial bond behavior of an individual CNT embedded in cement matrix, and
the flexural response of a CNT/cement composite beam. Experimental results on CNT/cement
composite beams were used to calibrate the numerical models. It was found that CNTs increased
the load-carrying capacity by about 47%. Moreover, toughness of the composite beams was
greater by 25% on average when compared with the plain cement beams. However, the
mechanical test results did not show any enhancement in the ductility of the composite beams.
A comparison between the experimental and numerical behavior of the composite beams
suggested an effective average (nanotube/cement) interfacial shear strength value of 6.5 MPa. It
was verified that the numerical models could successfully predict the behavior of CNT/cement
composites. FE analysis also predicted that increasing the shear strength from 6.5 MPa to 20.0
MPa, would increase the composite’s flexural strength, ductility, and toughness by 141%, 259%,
and 1976%, respectively. These results were significant considering that the enhancement was
provided by CNTs alone, and pointed at the great potential of surface-treating CNTs for
enhancing their interfacial bond strength and thus reinforcement efficiency in cement-based
matrices.

45

2.3 Knowledge Gaps
The above literature review indicates that most developments conducted thus far in the area of
cement-based nanocomposites have been directed towards processing of carbon nanotubes into
cementitious materials, and understanding the mechanisms of action of nanotubes in
cementitious matrices. Lower-cost graphite nanomaterials (carbon nanofiber and particularly
graphite nanoplatelet) have so far received less attention as reinforcement/additive in cementbased materials. The investigations reviewed above and summarized in Table 2.1 represent
exploratory studies which provide limited insight into the mechanisms of action and potential
value of graphite nanomaterials in cement-based matrices. Some key issues pertaining to
enhancement of the contributions of graphite nanomaterials in cementitious matrices are
presented in the following.


The high reinforcement efficiency of graphite nanomaterials in cement-based matrices
could be realized only if steps are taken to ensure thorough dispersion of nanomaterials
within the matrix and effective interfacial interactions between nanomaterials and cement
hydrates. The hydrophobic surfaces of carbon nanotubes can be made hydrophilic
through introduction of selected functional groups, surfactants and/or polyelectrolytes
which facilitate their thorough dispersion in aqueous media. Non-covalent interactions of
surfactants and polyelectrolytes with graphite nanomaterials can beneficially modify the
nanomaterial surfaces without damaging their structure and properties. Scalability of the
nanomaterial dispersion process would benefit from increasing the concentration of
(modified) nanomaterials dispersed in aqueous media.



Besides thorough dispersion in aqueous media, effective bonding to cement hydrates is a
key criterion governing the reinforcement efficiency of graphite nanomaterials in cement46

based materials. In order to develop primary bonds between cement hydration products
and graphite nanomaterials, one can introduce functional groups via primary bonds (to
carbon atoms) upon nanotubes, which can also develop primary bonds with cement
hydration products. Functional groups, when present at adequate concentrations on
nanotube surfaces, would provide enhanced interfacial stiffness, strength, thermal
conductivity, etc. for effective mechanical and physical interactions between carbon
nanotubes and cement hydration products. Non-covalent treatments (involving
surfactants and polyelectrolytes) could further benefit interfacial interactions with cement
hydrates without damaging the graphite nanomaterials.


The fresh cementitious paste is a dispersion of micro-scale particles in an aqueous
solution, which complicates the dispersion of nanomaterials which would only have the
residual space between cement particles for dispersion. After curing, conventional
cement-based matrices incorporate large quantities of micro- to nano-scale pores which
mitigate intimate interfacial contacts over the surface area of graphite nanomaterials. The
potential for thorough dispersion of carbon nanotubes and other graphite nanomaterials
within a cementitious matrix can be improved through partial replacement of micro-scale
cement particles with nano-scale silica fume particles in order to generate a smooth
particle size distribution which benefits the dispersion of nanomaterials and limit the
porosity and the maximum pore size in cement hydrates. Therefore, proper selection of
cementitious matrix with proper particle size gradation/packing would benefit the
reinforcement efficiency of graphite nanomaterials.



Nanomaterials should first be dispersed in water, and then introduced into the mixing
process of cement-based materials. Some nanomaterial modification schemes (e.g.,

47

physisorption of surfactants and polyelectrolytes) can be implemented as they are being
dispersed in water. The efforts needed for dispersion of nanomaterials in water have
implications in terms of added cost and reduced throughput. Any modification of the
conventional mixing process of cement for ensuring thorough dispersion of nanomaterials
within the fresh cement-based matrix would also have implications in terms of cost and
productivity. For a particular type and dosage of nanomaterials, the modification
conditions of nanomaterials and the (tailored) mix proportions should be selected to
minimize the dispersion efforts at different stages of production.


Given the porous structure of cement-based materials and the presence of micro-scale
crystalline constituents among cement hydrates, judicious selection of mix design and
curing conditions is important for reducing the porosity and pore size, and hindering the
formation of larger crystalline constituents during hydration of cement.



Given a particular combination of nanomaterial type, functionalization/modification
condition, dispersion method, and cementitious matrix mix design, mixing method and
curing condition, there is an optimum nanomaterial dosage for achieving balanced
improvements in the structure and engineering properties of the cement-based matrix.



The reliability of cementitious nanocomposite production, in light of the variations in the
qualities, dispersion and interfacial interactions of nanomaterials, need to be thoroughly
assessed. Comprehensive experimental data are also needed on diverse engineering
properties of cementitious nanocomposites.



Theoretical explanations are needed for the effects of nanomaterials on the engineering
properties of cementitious materials to support further development and improvement of
this new class of cementitious materials.
48

2.4 Graphite Nanomaterials and Microfibers
Seven types of multi-walled carbon nanotubes (MWNT A, B, C, D, E, F & G), in both acidfunctionalize and non-functionalized conditions, two types of carbon nanofiber in plain (CNF1
and CNF) and oxidized (CNF-OX) conditions, Two type of graphite nanoplatelets (GP and
GP1), two types of carbon microfiber (TT 143 and TT 150), and polypropylene microfiber (PP)
were used in the project. All graphite nano- and micro-scale reinforcement systems used during
this research are introduced in Table 2.3. TEM and SEM images of these different fiber
reinforcements are presented in Figure 2.3.

2.5 Cementitious Matrix and Processing Techniques
2.5.1

Cementitious Matrix

In high-performance cementitious matrices two lines of research have been followed. The first
concerns compact granular matrix concretes (DSP), with high superplasticizer and silica fume
content, also incorporating ultra-hard aggregate (calcinated bauxite or granite). The other line of
research relates to MDF (Macro Defect Free) polymer pastes. These pastes have very high
tensile strength (150 MPa or more), in particular when mixed with aluminous cements. In spite
of these favorable properties, it is necessary to underline the problems concerning MDF’s
sensitivity to water that, in many application, could be a weak point.
Dense cement-based matrices with a smooth particle size gradation down to nano-scale range,
capable of dispersing nanomaterials and effective interactions with their interfaces for effective
mobilization of their tremendous mechanical qualities, were used in this research. The primary
category of cement-based materials used here is referred to as densified with small particles
(DSP). These cementitious materials comprise micro-scale cement and nano-scale silica fume

49

particles, dispersed and densified with the help of superplasticizer (Figure 2.4). The silica
fume/binder ratio generally ranges from 0.15 to 0.25, and the water/binder ratio from 0.15 to
0.20 (Guerrini 2000). Using this basic concept, and introducing other ingredients (e.g., highquality aggregates and reinforcing fibers), it is possible to produce high-performance
cementitious materials with balanced qualities suiting different fields of application (Guerrini
2000). RPC (Reactive Powder Concrete) is a variation of DSP where the microstructure, particle
packing density and thus mechanical properties are improved by steam or high-pressure steam
curing (optionally accompanied with pressure compaction) in the presence of reactive microscale aggregates. The main thrust of this research is to evaluate the efficiency of different
graphite nanomaterials in progressively coarser high-performance densified with small particles
(DSP) matrices. Therefore, it is appropriate to delve in to details of the criterion to select such
matrices and the influence of different constituent materials.
2.5.1.1 Homogeneity Enhancement
Conventional concrete is a heterogeneous material, in which the aggregates (sand and gravel)
form a skeleton of contiguous granular elements in the cementitious paste (cement, additives and
water). The hardness of the aggregates is greater than that of the paste. For example, the Young’s
modulus for silica is 50-70 GPa, compared with between 18 and 22 MPa for the paste.
Heterogeneity-related problems are substantially reduced with DSP matrices for the following
reasons (i) Elimination of classical coarse aggregates, replaced by fine sand, (600 µm maximum)
and granite gravel (less than 9 mm), if required; (ii) Improved mechanical properties of the paste;
and (iii) Reduction in the aggregate/matrix ratio.


Effect of Aggregate Size. Conventional concrete aggregates form a set of rigid inclusions.
On application of a compression force shear and tensile stresses appearing at the

50

paste/aggregate interface generate cracks in the paste. The sizes of these cracks is related
to the extent of the zone under tensile or shear stress. In the case of a spherical inclusion,
the size of the equatorial crack is directly proportional, to the diameter of the inclusion.
For DSP mortar matrix, with a reduction in the size of the coarsest aggregate by a factor
of about 50 (e.g, 400 µm instead of 20 mm), a major reduction is obtained in the size of
mechanical (external loads); chemical (autogenous shrinkage); and thermo-mechanical
(differential expansion between paste and aggregate under the effects of heat-treatment)
microcracks.


Effect of Enhanced Mechanical Properties of Paste. In all cases, the increase in the
Young’s modulus for the DSP paste, by comparison with that of conventional
cementitious pastes, tends to attenuate the effects associated with disturbance of the
mechanical stress field.



Effect of Sand Content. The effects mentioned above are linked to aggregate size, and
can be described as “mesoeffects”. Reduction of sand content represents a more global
“macro-effect”. In a conventional concrete, the aggregates (sand and gravel) are the
majority components in terms of volume, and form a rigid skeleton of contiguous
granular elements. This means that a major proportion of paste shrinkage is blocked by
the granular skeleton, with results in increased porosity. In the case of a DSP, the volume
of the paste is at least 20% greater than the voids index of non-compacted sand. Thus the
aggregates used in DSP do not form a rigid skeleton, but a set of inclusions trapped in a
continuous matrix. Paste shrinkage is blocked locally round each aggregate particle (with
diminished consequences as a result of the size limitation effect mentioned above),
whereas global shrinkage is not blocked by the rigid skeleton. Each grain can be

51

transported by the paste, and can migrate with respect to its neighbors. This advantage
only concerns structures where global shrinkage is unrestricted by any external source
(eg, form work).
2.5.1.2 Optimization of Granular Mixture
Optimization of the granular mixture can be achieved by the use of packing models. DSP mix
design is based on the following principles:


A mixture comprising a number of granular classes is obtained, with a tight granular
range inside each class;



Classes are separated, selecting a high ratio (exceeding 13) between mean diameter d, for
two consecutive granular classes;



Cement/superplasticizer compatibility is studied, and optimum ratios are determined by
rheological analysis;



Only the least agglomerated fine sand is selected.

2.5.1.3 Selection of Granular Components


Sand. Sand selection parameters to be defined consist, (i) mineral composition; (ii) mean
particle size; (iii) granular range; (iv) particle shape; and (v) mixture ratio by weight.
Insofar as mineral composition is concerned, silica offers the advantages of being a very
hard material, excellent paste/aggregate interfaces and readily available and low cost.
Mean particle size is defined in line with the homogeneity criterion, and the separation of
granular classes criterion that have both already been described. The next lowest granular
class is the cement, for which mean diameter d, can vary between 11 and 15µm. Sand

52

with a mean particle diameter of about 250 µm is therefore selected. The particle size
range is defined indirectly by the desirable maximum and minimum particle sizes.
Maximum particle size is limited to less than 600 µm, Fine sand is obtained by screening
crushed silica sand, where the grains are highly angular, or natural quarry sand, where the
grains are more spherical. Both types of sand can be used for DSP matrices. However
water demand is slightly less for natural sand.


Cement. The selection of cement cannot be dissociated from that of the superplasticizer.
From the point of view of chemical composition, cements with low C3A content give
better results. As for particle size, over-ground cements with a high Blaine fineness are
not satisfactory, due to their high water demand. The best cement in terms of rheological
characteristics and mechanical performance is high-silica-modulus cement. However this
type of cement has the disadvantage of a very slow setting rate, preventing its use for
certain applications. Conventional quick-setting high performance cement offers very
similar mechanical performance, despite a higher water demand. The most efficient
superplasticizers are polycarboxylate-based dispersing agents, but which also exhibit a
retarding characteristic which can present a problem for practical applications. The
conventional superplasticizers (nepthalene-based) selected for their compatibility with the
cement give slightly poorer results. For the low w/c ratios used for DSP concretes, the
optimum superplasticizer ratio is high (solid content of approx. 1.6% of cement content).



Silica Fume. The silica fume used in DSP matrices has three main functions, (i) filling
the voids between the next larger class particles (cement); (ii) enhancement of rheological
characteristics by the lubrication effect resulting from the perfect sphericity of the basic

53

particles; and (iii) production of secondary hydrates by pozzolanic reaction with the lime
resulting from primary hydration.
Therefore, the following parameters are used for silica fume characterization: (i) degree
of particle aggregation; (ii) nature and quantity of impurities; and (iii) basic particle size.
The main quality of a silica fume is the absence of aggregates. This leads to the use of
non-compacted silica fumes. Slurry cannot be used, as the quantity of water contained in
the slurry exceeds the total quantity of water required for the mixture. Particle size is a
secondary factor. Indeed the best results are obtained with silica fume procured from the
zirconia industry, being free from impurities and totally disaggregated. However Blaine
2

2

fineness is lower than that for conventional fumes (14m /g compared with18m /g). On
2

the other hand, an impurity-free fume with a high Blaine fineness value (22m /g)
produced mediocre results, due to the aggregation of the finest particles. Typically the
fume/cement ratio used for DSP is less than 0.25. This ratio corresponds to optimum
filling performance (Collepardi, Corinaldesi et al. 2002) and it is close to the dosage
required for complete consumption of the lime resulting from total hydration of the
cement. However, cement hydration is incomplete in DSP matrices, and the available
quantity of silica fume is more than required by the pozzolanic reaction.


Crushed Silica Sand. Crushed crystalline silica powder is an essential ingredient for DSP
mortar and concretes. Maximum reactivity during heat-treating is obtained for a mean
particle size of between 5 and 39 µm (11). The mean particle size of the crushed silica
used for DSP is 10µm, and is therefore in the same granular class as the cement. The ratio
by weight adopted corresponds to the stoechiometric optimum for conversion of

54

amorphous hydrates into tobermorite characterized by a C/S molar ratio of 516 = 0.83.
This is achieved with a silica/cement ratio of 0.56. This ratio is obtained by adding silica
fume and crushed silica sand as a complement (see Table 2.4).
2.5.1.4 Microstructure Enhancement by Heat Treatment
Heat-treating is performed after the matrix has set, by simply heating at ambient pressure. Heato

treating at 70 - 90°C substantially accelerates the pozzolanic reaction, while modifying the
o

microstructure of the hydrates which have formed. However, for our experimental regimen 70 C
was selected to conserve on the energy requirement for production of high-performance DSP
matrices.
2.5.1.5 Ductility Enhancement by Incorporation of Fibers
The behavior of DSP matrices is purely linear and elastic, corresponding to a fracture energy not
-2

exceeding 30 Jm

Micro-fibers must be added to enhance DSP for ductility. The fibers are

introduced into the mix at a ratio of between 1.5 and 3% by volume. The economic optimum
3

corresponds to a ratio of 2%, or about 155kg/m . However, most of the micro-fibers increase the
toughness but tend to decrease the flexural strength of the matrix, especially at lower volume
fractions. Besides, other draw-backs related to durability and long-term performance.
Keeping all the factors mentioned earlier in mind a comprehensive literature review was
conducted on DSP (and RPC) cement-based materials (Tjiptobroto and Hansen 1991; Sun and
Young 1993; Sun and Young 1993; Tjiptobroto and Hansen 1993; Tjiptobroto and Hansen 1993;
Beaudoin, Feldman et al. 1994; Cheyrezy, Maret et al. 1995; Richard and Cheyrezy 1995;
Shannag, Hansen et al. 1995; Feylessoufi, Villieras et al. 1996; Roux, Andrade et al. 1996;

55

Shannag, Brincker et al. 1996; Young 1996; Zanni, Cheyrezy et al. 1996; Shannag, Brincker et
al. 1997; Hu, Fang et al. 1999; Matte and Moranville 1999; Shannag, Hansen et al. 1999; Zivica
1999; Guerrini 2000; Shannag and Hansen 2000; Moranville-Regourd 2001; Singh, Kumar et al.
2001; Collepardi, Corinaldesi et al. 2002; Morin, Cohen-Tenoudji et al. 2002; Badanoiu,
Georgescu et al. 2003; Bayard and Ple 2003; Chan and Chu 2004; Washer, Fuchs et al. 2004;
Camilleri, Montesin et al. 2006; Childs, Wong et al. 2007; Lee, Wang et al. 2007; Wong, Childs
et al. 2007). Based on this literature review, the cementitious matrices introduced in Table 2.4
were selected for evaluation of the reinforcement effects of graphite nanomaterials in cementbased matrices. The polycarboxylate-based superplasticizers used for dispersion of graphite
nanomaterials were also used in these cement-based matrices. The materials selected for use in
the cement-based matrices (Table 2.4) are introduced in the following.


Portland Cement: Type I Portland cement was acquired from Lafarge - North America.
The chemical composition of this cement is given in Table 2.5.



Silica Fume: Undensified silica fume was acquired from Norchem, Inc. It comprises
nano-scale (200 nm average particle size) amorphous silica particles, with a minimum
silicon dioxide content of 85% and a specific gravity of 2.25. The specific surface area of
2

this silica fume is 15 m /g, and its minimum 7-day pozzolanic activity index is 105%.


Silica Sand: Sand comprising more than 99.5% silicon dioxide was acquired from
Fairmount Minerals, Chardon, Ohio. It was ball milled in our laboratory to average
particle sizes of about 39 μm and 350 μm. Sand particles were used in high-performance
mortar and concrete matrices.

56



Granite Gravel: Granite gravel ranging in size from 1 mm to less than 9 mm, with
average particle size of 3.55 mm was acquired from Fairmount Minerals, Chardon, Ohio.
It was used in high-performance matrices.



Superplasticizer: Glenium® 7700 and ADVA® Cast 575 polycarboxylate-based
superplasticizers (high-range water reducers) were acquired from BASF and W.R. Grace,
respectively. They offer excellent slump retention without compromising early-age
compressive strength development and set time. These admixtures meet the ASTM C 494
requirements for Type F (high-range water-reducing) admixtures.

2.5.2

Dispersion and Mixing Procedure

Thorough dispersion of nanomaterials is a key step toward their effective use as reinforcement in
various matrix systems. The strong tendencies toward physical interactions over the large surface
area of nanomaterials seriously challenge development of (aqueous) dispersions with adequate
nanomaterial concentrations. As-produced carbon nanotubes align parallel to each other and pack
into crystalline ropes, which is due to the strong van der Waals attractions between nanotubes.
The ropes further aggregate into tangled networks. The aggregation of nanotubes is an obstacle
in most applications, diminishing the special mechanical, geometric and physical properties of
individual nanotubes. Hence, effective use of carbon nanotubes in cementitious nanocomposites
requires deagglomeration and dispersing of the nanotubes within the matrix. The surface of
carbon nanotubes is atomically sharp, hydrophobic and inert. Functionalization of the nanotube
surfaces to make them compatible with cementitious matrices would be another key step toward
successful use of graphite nanomaterials as reinforcement in cementitious matrices.
Before mixing with the cementitious paste, graphite nanomaterials need to be well dispersed in
aqueous media. The mechanical action needed for dispersion of CNT/graphite nanomaterials
57

will be applied through high-shear mixing using magnetic stirrers followed by ultrasonication
(involving high-power sonication using a tip or horn). During ultasonication, high-frequency
sound waves are applied to carbon nanotubes, which can agitate the molecules and breaks
intermolecular bonds.

The separated nanotubes can re-aggregate if the dispersion is not

stabilized using steric and/or electrostatic effects (Fig. 2.5). The objective is to achieve a uniform
and stable dispersion of individual graphite nanomaterials (Fig. 2.6).
Introduction of COOH groups (carboxyl groups) makes the surfaces of graphite nanomaterials
hydrophilic, and thus facilitates their dispersion in the aqueous medium of fresh cementitious
paste. In addition, stabilization of the dispersion is achieved through electrostatic effects, which
is because the COOH groups will be negatively charged in aqueous media, and will thus repel
each other. The COOH-functionalized nanotubes, however, yield stable dispersions as far as the
nanotube concentration in the aqueous medium does not exceed a relatively low limit (about 3
gram/liter).
Introduction of surfactants also facilitates dispersion of graphite nanomaterials; this technique is
commonly used with graphite nano-sheets.

Polymer wrapping is yet another approach to

uniform dispersion of carbon nanotubes. This research work is employing all these techniques
for achieving uniform and stable dispersions of individual graphite nanomaterials with increased
concentrations of nanomaterials.
The dispersion and mixing process was constantly modified throughout the project in order to
achieve improved dispersion at increased volume fraction of graphite nanomaterials. The
nanomaterial dispersion procedure that evolved in the course of implementing the project
comprised the following steps:

58



Add the required amount of graphite nanomaterial, superplasticizer, and polyelectrolyte
(if used as modifier of nanomaterials) to the mixing water of cementitious material;



Stir the mixture overnight (12 to 15 hours), Fig 2.7a;



Sonicate, Fig 2.7b, the mixture using a probe as follows: (i) Sonicate for ten minutes at
different amplitudes (30%, 45%, 65% and 75%) with 1-minute breaks between different
amplitudes; (ii) Pulse (1 minute on & 30 seconds off) for 10 minutes at 85% amplitude;
(iii) Turn off the sonic probe for 2 minutes, and repeat the pulsing cycle two more times;
and (iv) Repeat the whole sonic probing cycle one more time.



In cases with microfiber (or hybrid reinforcement systems), microfibers were added to
the mixing water half an hour before mixing with cementitious materials, without using
the above dispersion procedure (for hybrid reinforcement, the nanomaterials were first
dispersed following the above procedure, and then microfibers were added half an hour
before mixing).

Cementitious materials (with and without graphite nanomaterials dispersed in the mixing water)
were prepared following ASTM C 192 and C 305 procedures. The high-performance (DSP)
paste specimens, for example, were prepared as follows: (i) mix cement and silica fume for 3
minutes at speed 1 in a mortar mixer (HOBART, Model A200F – Fig. 2.8a); (ii) add water (with
dispersed graphite nanomaterial, if any, at the concentration required to yield the targeted
volume fraction) plus superplasticizer and micro-fibers (if any), and mix for approximately 2
minutes at each of the speeds 1, 2 and 3; (iii) cast specimens following ASTM C 192 procedures,
using a vibrating table (FMC Syntron Power Plus) at vibration intensity of 10.

59

The specimens were moist-cured inside molds after casting (ASTM C 192) over a 24-hour
o

period. They were subsequently demolded and subjected to 48 hours of steam curing at 70 C.
The specimens were then conditioned at 50% relative humidity and room temperature for seven
days prior to testing.

2.6 Experimental Methods
The test procedures employed to determine the engineering properties of cement-based materials
are described in the following.
Compression tests (ASTM (American Society for Testing and Materials) C 109) were performed
(Fig. 2.8b) on 50 mm cube specimens. Flexure tests (ASTM C 1185) were performed (Fig. 2.8c)
on 12.5x50x150 mm specimens by center-point loading on a span of 125 mm using a deflection
controlled Instron test system, with load and deflection data collected using a data acquisition
system. Impact tests (ASTM D 7136) were performed (Fig. 2.8d) on 12x150x150 mm
specimens. Abrasion tests (ASTM C 944) were conducted (Fig. 2.8e) on the surface of
cylindrical specimens with 100 mm diameter (and 50 mm height). Moisture absorption rate tests
(ASTM C 1585) were performed (Fig. 2.8f) on cylindrical specimens with 100 mm diameter and
100 mm height. The sides and top surfaces of specimens were covered with impermeable
adhesive sheets, and the bottom surface was immersed (1-3 mm depth) in water. The mass gain
of the specimen (due to capillary sorption of water) was measured at frequent time intervals
(with the wet surfaces patted dry prior to weight measurement).
In order to evaluate the durability and stability of the high-performance (DSP) cementitious
materials with different nano- and/or micro-scale reinforcement, accelerated aging tests were
conducted. Freeze-Thaw tests were carried out following ASTM C666 – 03, Standard Test

60

Method for Resistance to Rapid Freezing and Thawing (with both freezing and thawing in water)
and ASTM C1185 and 1186. The automated freeze-thaw machine used for testing the specimens
is made by Humboldt Manufacturing Corporation, Model H 3185 (Fig. 2.9a). Wet-dry tests were
carried out following ASTM C1185 – 03, Standard Test Method for Sampling and Testing NonAsbestos Fiber-Cement Flat Sheets. An automated wet-dry machine (with CN 76000 Omega®
controller) was used for exposing the specimens to wet-dry cycles (Fig. 2.9b). Hot Water
Immersion tests were carried out according to ASTM C1185 – 03, Standard Test Method for
Sampling and Testing Non-Asbestos Fiber-Cement Flat Sheets. A hot water bath (Precision
Scientific Model 186) was used for hot water immersion of specimens (Fig. 2.9c). The Elevated
Temperature tests were carried out in an electric furnace (Applied Test System, Inc.) shown in
Fig. 2.9d, with temperature change controlled using a CN76000 Omega® controller.
o

o

Temperature was ramped up at a rate of 2 C/minute to 300 C, where it was maintained for 3
hours before ramping down to room temperature.
During long-term exposures to different severe environments, the specimen degradation over the
course of exposure was monitored by nondestructive evaluation of the dynamic elastic modulus
using a sonometer (Geotest Instrument Corporation model C2010) shown in Fig.2.9e, in
accordance with ASTM C215, Test Method for Fundamental Transverse, Longitudinal, and
Torsional Frequencies of Concrete Specimens and ASTM C1198, Standard Test Method for
Dynamic Young’s Modulus, Shear Modulus, and Poisson’s Ratio for Advanced Ceramics by
Sonic Resonance.
In addition to above tests on the engineering properties of cement-based nanocomposites, the
following experimental techniques were employed to provide further insight into the chemistry,
crystallinity, microstructure and failure mechanisms of cement nanocomposites reinforced with
61

functionalized graphite nanomaterials: (i) SEM; (ii) XRD (X-Ray Diffraction); (iii) DT/TGA
(Differential Thermal and Thermogravimetry Analyses); and (iv) FTIR (Fourier Transform
Infrared Spectrometry).

62

Table 2.1- Summary review of experimental studies reported on the use of nanomaterials in cement-based matrices.
Author

Study Objectives /
Detail
Carbon Nanotubes
Li et al. The study presents the
[2004]
results of addition of
acid treated MWNTs
and carbon fibers in
cement mortar
Objectives:
 Acidic treated carbon
nanotubes were added
to cement, and both
the
mechanical
behavior
and
microstructure of the
resulting
nanocomposites were
studied.
 Phase
composition
was characterized by
FTIR spectroscopy
 Mineralogy
and
microstructure
were
studied using SEM.
 Porosity and pore size
distribution
were
determined
by
mercury
intrusion
porosimetry (MIP).

Features and Methodology

Observations / Conclusions

Strengths/
Drawbacks

 Mixtures of cement mortar
were prepared with water:
cement:
sand
ratio
of
0.45:1:1.5. Plain mortars and
those with with 0.5% by
weight of cement MWNT and
0.5% by weight of cement
CMF were studied.
 MWNTs had an outer dia of
10-30 nm and lengths ranging
from 0.5-500 µm.
 CMF had diameter of 10-14
µm and length of 6 mm.
 40x40x160 mm samples were
casted using a flat beat mixer,
and compacted by vibration.
 Demolded after 1 day, cured
o
in water at 30 C for 26 days.
The samples were dried in
o
oven at 50 C for 24 hours
before testing.
 Specimens were taken form
samples after mechanical tests
to perform other tests.

 The carbon nanotubes used were surfacetreated using a mixed solution of sulfuric
acid and nitric acid.
 It was found that the addition of carbon
nanotubes to cement can greatly enhance
its flexural and compressive strengths, as
well as failure strain. This enhancement
is due to the improvement of material
microstructure. Interfacial interactions
between surface-modified nanotubes and
cement hydrates (such as C–S–H and
calcium hydroxide) produce high bond
strength, and increase the load-transfer
efficiency from cement matrix to the
reinforcement.
 Moreover, the addition of carbon
nanotubes refines the pore size
distribution and decreases the porosity of
cement
composites.
Therefore,
cementitious nanocomposites were found
to be densely packed.
 The bridging action of carbon nanotubes
provided for load transfer across voids
and cracks under tension.

 A pioneering
study showing
the potential of
CNTs
in
cement-based
materials.
 Preliminary
results
not
covering
all
aspects
of
nanocomposite
performance.
 Dispersion of
nanotubes
in
cementitious
matrix is not
emphasized.
 Particle
size
distribution of
the
cementitious
material is not
emphasized.

63

Table 2.1 (cont’d)
Author

Study Objectives /
Detail
Ibarra et This paper presents
al. [2006] preliminary results
on the mechanical
properties of cement
pastes
reinforced
with
carbon
nanotubes.
Objectives:
 Assess
contributions
of
SWNTs
and
MWNTs
to
compressive and
flexural strengths.
 Elastic
modulus
was obtained by
nanoindentation
and
AFM
observation.

Features and Methodology

Observations / Conclusions

Strengths/Drawbacks

 Cement paste was used as the  Both MWNTs and SWNTs were  This is a very basic study
matrix using type I cement at added to cement paste without any on the use of CNTs in
0.34 w/c ratio.
dispersing agent, and as a pre-mix cement-base matrix.
 Both SWNTs and MWNTs of nanotubes and the mixing water  Not enough emphasis is
were used at 0.05 to 0.20 with Gum Arabic.
placed on dispersion and
wt.%. These nanotubes were  Overall, specimens containing mixing of nanomaterials
dispersed both in plain water nanotubes without gum arabic in cement-based matrix.
lower
mechanical  Compression and flexure
and water containing Gum provided
Arabic.
properties than the plain cement test data are not reported.
 SWNTs had an outer dia. of 2 paste due to the intrinsic
nm and lengths of about 20 hydrophobicity of the nanotubes
and their effects on the hydration
µm.
 MWNTs had an outer dia. of process of cement.
10-20 nm and lengths of 1-50  However, when nanotubes were
pre-dispersed with gum arabic, an
µm.
 1x1x6 cm samples were increase in mechanical properties
casted and compacted using was achieved with nanotubes, with
highest gains achieved with
vibration.
 Demolded after 1 day, cured SWNTs.
o
in water at 21 C for 27 days.
 Specimens were taken form
samples after mechanical tests
for nanoindentation tests and
AFM observations.

64

Table 2.1 (cont’d)
Author

Study Objectives /
Detail
Cwirzen This
study
et
al. investigated
the
(2008)
effects of surface
decoration on the
wettability of multiwalled
carbon
nanotubes
(MWNTs), and the
mechanical
properties of the
cement
paste
incorporating
decorated nanotube
dispersions.
Objectives:
 Determine
the
effects
of
surfactants
and
surface
functionalization
of MWCTs on
their ability to
disperse
in
aqueous solutions
and on the stability
of
these
dispersions.

Features and Methodology

Observations / Conclusions

 Cement paste was used as the
matrix,
using
sulphateresistant CEM I 42.5 N
cement with w/c ratio of 0.25
- 0.40.
 Two types of MWNTs were
used: pure MWNT and
MWNT functionalized with
COOH groups.
 Pure MWNTs had an outer
dia. of 10 nm and length of
about 10 µm.
 Functionalized MWNTs had
an outer dia. ranging from few
to few hundred µm, with
unspecified length.
 Different wt.% of MWNTs
were mixed with different
proportions of poly-acrylic
acid (PAA) and Gum Arabic
(GA).
 The mixtures were stirred and
sonicated for 2 - 15 minutes.

Strengths/Drawbacks

 The results showed that stable and  This study explores the
homogenous dispersions of MWNTs effects of both pristine
in water can be obtained by by and oxidized MWNTs
surface functionalization combined on
compressive
with decoration using polyacrylic strength of cement
acid.
paste. Either GA or
 The
cement
paste
specimens PAA was used for
incorporating
these
dispersions further treatment of
offerred good workability and an MWNTs.
increase in the compressive strength  Only
very
small
of nearly 50% with 0.045% by weight improvements
in
of cement addition of the PAA- mechanical properties
of cement paste were
treated MWNTs
 At 0.15 wt.%, functionalized observed.
MWNTs provided less increase in
compressive strength
 These results indicate the existence of
chemical bonds between the OH
groups of the functionalized MWNTs
and probably the C–S–H phase in
cement hydrates, enabling effective
stress transfer.

65

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Detail
 Determine
the  10x10x60 mm samples were casted  Decoration of MWCTs with
influence of the using vacuum mixer in Teflon polyacrylic acid polymers and
type of surface molds, and were not compacted gum Arabic yielded dispersions
decoration on the using vibration.
which
appeared
to
be
mechanical
homogeneous
only
for
 Demolded after 1 day, cured in
properties of the water at 21oC for 27 days.
approximately 2 hr., after which
MWCTs–cement  Specimens were taken form a progressive sedimentation
nanocomposites.
samples after mechanical tests for occurred.
nanoindentation tests and AFM  Good workability was achieved
with
the
cement
pastes
observations.
produced
using
all
of
dispersions.
 Slower hydration of the cement
incorporating gum Arabic was
observed.
 The mechanical properties of
the cement pastes incorporating
MWNTs
treated
with
polyacrylic polymers were
unchanged.

66

Strengths/Drawbacks

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Detail
Musso et Presents flexure and  Mortar was used with water:
al.
compression
test cement: sand ratio of 0.4: 1: 1.5 in
[2009].
results on cement plain condition, and with 0.5 wt.%
mortar
reinforced MWNTs.
with
as-grown,  As-grown, annealed and carboxyl
annealed
and functionalized MWNTs were used.
functionalized
 Pristine MWNTs had an outer dia.
MWNTs.
of 40-80 nm and lengths of 4001000 µm.
 Annealed MWNTs had an outer dia
of 40-80 nm and lengths of 200-400
µm.
 Carboxyl functionalized MWNTs
had an outer dia of 10-20 nm and
lengths of 0.1-10 µm.
 All MWNTs were dispersed in
acetone using an ultrasonic probe
(vibration by ultrasound). After 4 h,
sonication was stopped and the
acetone was allowed to evaporate.
Cement
nanocomposites
were
prepared
in
rotary
mixer.
Superplasticizer
and
viscosity
modifying agents were added to the
mixture during the stirring stage, to
benefit dispersion and cohesion.

67

Observations / Conclusions

Strengths/Drawbacks

The following observations were  Three
different
made:
MWNTs
were
 Both pristine and annealed evaluated in mortar
MWNTs
induced
an matrix.
improvement in the mechanical  Carboxyl
properties of cement-based functionalized
matrix.
MWNTs produced a
 The mechanical strength of drop in mechanical
cement nanocomposite was properties, which is
strongly affected by the defects contrary
to
other
and chemical properties of reported results.
multiwall carbon nanotubes  Desired gradation of
(MWNTs).
the matrix constituents
 Carboxyl
functionalized is not addressed.
MWNTs produced a significant
loss of performance when
compared with pristine MWNT.
This behavior was due to the
hydrophilic
nature
of
functionalized MWNTs, which
absorbed most of the water
contained in the cement
mixture, thus hindering proper
hydration of cement.

Table 2.1 (cont’d)
Author

Study Objectives /
Detail

Features and Methodology

Observations / Conclusions

 After pouring in a stainless steel  This hypothesis was supported
3
prismatic (40x40x160 mm ) mold, by the low amounts of
vibrator was used to compact the tobermorite gel produced during
hydration per outcomes of
specimens.
 Molded specimens were kept for 24 TGA.
h in a water vapor saturated  A suitable hydration of cement
atmosphere, and then demolded and was realized by only increasing
the water/cement ratio from 0.4
o
cured for 27 days in 20 C water.
to 0.56.
 Compressive and flexural strengths  Even if a significant number of
were measured.
defects in the CNT atomic
 Morphology was characterized by network can produce curved
FE-SEM, TEM.
CNTs
with
deteriorated
mechanical
properties,
the
ultimate strength of the
nanocomposite can still be
enhanced

68

Strengths/Drawbacks

Table 2.1 (cont’d)
Author
KonstaGdoutos
[2009]

Study Objectives /
Detail
Objectives:
 Development
of
high-performance
cement
pastes
reinforced
with
multiwalled carbon
nanotubes
was
investigated.
 Effective
dispersion
of
MWNTs in the
mixing water was
achieved
by
application
of
ultrasonic energy
and the use of a
commercially
available
surfactant.

Features and Methodology

Observations / Conclusions

Strengths/Drawbacks

 Type I cement was used to produce
pastes with 0.3 w/c ratio.
 Two types of commercially
available purified multiwalled
carbon
nanotubes
(MWNTs),
designated as short and long, were
used.
 Short MWNTs had an outer dia. of
20-40 nm and lengths of 10-30 µm.
 Long MWNTs had an outer dia. of
20-40 nm and lengths of 10-100
µm.
 Mix proportions of
sonicated
suspensions with short MWNTs at
0.048, 0.08 and 0.10 wt.% of
cement, and long MWNTs at 0.025,
0.048 and 0.08 wt.% of cement
were used.
 Mixing of materials was performed
following ASTM 305 using a
standard Hobart mixer.
 After mixing, the paste was cast in
20 x 20 x 80 mm molds. After
demolding, specimens were cured
in water saturated with lime until
the testing age.

 It was found that small amounts
of
thoroughly
dispersed
MWNTs (0.025–0.08 wt.% of
cement)
can
significantly
increase the strength and
modulus of the cementitious
matrix.
 In particular, lower amounts of
long MWNTs (0.025–0.048
wt.%)
offered
high
reinforcement efficiency.
 Higher concentrations (close to
0.08 wt.%) of short MWNTs
were required to produce the
same reinforcing effects as long
nanotubes.
 Nano-indentation test results
suggesedt that MWNTs can
strongly modify and reinforce
the nanostructure of the
cementitious matrix.
 Compared to plain cement
matrix, nanocomposites appear
to have a higher amount of
high-stiffness
C–S–H
and
reduced nano-porosity.

 Two types of MWNTs
were evaluated in
cement paste.
 Functionalization
of
nanotubes was not
given
due
consideration.
 Matrix gradation was
not considered.
 Only
small
improvements
were
reported
in
compressive strength.
More
improvement
was reported with
smaller-diameter
MWNTs.

69

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Detail
 Fracture mechanics  Fracture tests were conducted to  Due to their small dia. (20–40
tests were performed assess
the
mechanical nm) MWNTs appear to
to investigate the performance of nanocomposites. specifically reduce the amount
effects of the aspect A 6 mm notch was placed in of fine pores. This phenomenon
ratio
and specimens using a water-cooled leads to reduction of capillary
concentration
of diamond saw. The specimens stresses, resulting in a beneficial
MWNTs on the were then subjected to three-point effect on the early strain
fracture
flexure tests at 3, 7 and 28 days of capacity of nanocomposites.
characteristics
of age.
cementitious
 Specimens were obtained form
nanocomposites.
samples after mechanical tests for
 Determination
of SEM observations.
mechanical
properties
and
porosity
of
the
composites
was
carried out through
nano-indentation.
 Autogenous
shrinkage
experiments
conducted determine
effect of MWNTs on
early strain capacity
of the cementitious
matrix.

70

Strengths/Drawbacks

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Strengths/Drawbacks
Detail
Nasibuli-  Synthesis of a  Two types of Portland cement with  Cement particles were utilized  A single-step process
n
etal. novel
cement different concentrations of iron both as catalyst and as support was introduced for
[2009]
hybrid
material oxide (Fe O ): sulphate-resistant material,
which
enabled production of cement
2 3
(CHM), wherein (SR) (4 wt%) white (0.33 wt%) synthesis of novel hybrid nano- with as-grown CNTs
CNTs and CNFs Portland cements.
structured materials, wherein and CNFs.
are adhered to  For
synthesis
of
graphite CNTs and CNFs attached to  Only cement paste
cement particles, nanomaterials,
acetylene
was cement particles, and provided matrix was evaluated.
providing
for chosen as the main carbon source for desired dispersion of carbon  A tremendous rise in
thorough
due to its low decomposition nanomaterials in cement matrix. electrical conductivity
dispersion
of temperature and cost-effectiveness.  This hybrid material was was observed with
carbon
Carbon monoxide and carbon synthesized by a simple one- CHM.
nanomaterials in dioxide were used as promoting step CVD method, which can be
the matrix.
easily integrated into industrial
additives.
 The synthesis of CHM was carried cement production without a
out in a CVD reactor, which significant increase in cement
consists of a quartz tube (with an cost.
internal diameter of 34mm and a  TEM, SEM, XRD, TGA and
length of 100 cm) inserted in a Raman investigations showed
resistively heated furnace (with a high efficiency of the method
heated length of 60 cm), a powder for low-temperature and highfeeder with an adjustable powder yield synthesis of CNTs and
feeding rate, a copper screw feeder, CNFs.
a powder collector and a water
cooling system to keep the ends of
the tube at room temperature.

71

Table 2.1 (cont’d)
Author

Study Objectives /
Detail

Makar
Objective:
and Chan  Examination of the
[2009]
effect
of
1%
SWCNT content
on the early stages
of hydration of
OPC
using
isothermal
conduction
calorimetry, highresolution
SEM
and TGA analysis.

Features and Methodology

Observations / Conclusions

Strengths/Drawbacks

 For the performance of mechanical  Investigations of the physical
tests, beams with dimensions of properties of cement paste made
3
60×10×10 mm were prepared with CHM revealed as high as 2
using Teflon moulds and cured in times increase in compressive
strength and 40 times increase
o
water at 20 C for 7, 14 and 28 days. in electrical conductivity after
 The structure of nanocomposites 28 days of curing in water.
was evaluated by observation under
SEM and TEM, and also by XRD,
TGA and Raman.
 Type I cement paste was used with  Unfunctionalized
SWCNT  A fundamental study
0.5 w/c ratio.
bundles
distributed
by focusing on the early
sonication in isopropanol on stages of hydration,
 SWNTs were purified by a solventvaluable
based method which does not OPC grains accelerated the providing
hydration of OPC as compared information on the
functionalize.
 Samples of OPC were sonicated for with both OPC control samples development of cement
2 h in isopropanol both with and sonicated alone and as-delivered hydrate structure in the
presence of nanotubes,
without 1 wt.% SWNT. Isopropanol OPC.
was then removed in each case by  The morphology and location of but limited information
dessication in a vacuum system, the C3A hydration products on the engineering
of
and the processed samples were were altered by the presence of properties
cementitious
stored under vacuum until needed.
SWCNT, but no evidence was
observed for nucleation of C3A nanocomposites.
hydration products on SWCNT
bundles.

72

Table 2.1 (cont’d)
Author

Study Objectives /
Detail

Features and Methodology

Observations / Conclusions

 Calorimetry measurements were  C3S hydration reactions were
then made, and specific hydration instead responsible for the
durations
were
chosen
for observed change in hydration
subsequent SEM examination. behavior.
Times of 0, 60, 120, 135, 180, and  Both the initial C S hydration
3
240 min were chosen for the peak at the end of the induction
samples sonicated with SWCNT, period and the bulk C S
3
and times of 0, 60, 135, 150, 180, hydration peak were increased
and 240 min were chosen for the in magnitude. Formation of
samples
sonicated
without calcium
hydroxide
during
SWCNT.
hydration was enhanced during
 Image analysis was performed to the first 24 h of hydration, with
determine the average surface area the increase in the calcium
covered by the C3A hydration hydroxide directly reflecting the
products formed during the first 60 increased heat of hydration of
min of hydration.
the SWCNT composite as
compared with the sonicated
control.
 SEM images showed the
preferential formation of C–S–
H on the surface of nanotube
bundles in the form of a
gradually thickening coating.
also evidence of classical
reinforcing behavior, including
fiber pullout

73

Strengths/Drawbacks

Table 2.1 (cont’d)
Author
Manzur
and
Yazdani
[2010]

Study Objectives /
Detail
Objectives:
 Evaluation of the
effects of treated
and
untreated
CNTs
on
the
mechanical
properties
of
cement mortar.
 Assessment of the
effect of CNT
dosage on the
mechanical
properties
of
cement mortar.
 Investigations of
the
effect
of
water/cement ratio
on the strength of
CNT–reinforced
cement mortar.

Features and Methodology

Observations / Conclusions

Strengths/Drawbacks

 Mortar was used with 0.4:1:2.75  The compressive strength of a  Preliminary results on
mix proportions. w/c ratio ranged typical mortar mix increased by the use of two different
form 0.485 to 0.65.
addition of the two types of MWNTs in mortar
matrix.
 Type II cement was used with MWNT.
specially graded sand per ASTM  Mixing technique is important  No consideration given
C109. Two types of MWNTs were in enhancing the compressive to functionalization of
of
mortar nanotubes.
used: (i) MWNT1 had an outer dia. strength
>50 nm and lengths ranging from nanocomposites. Proper mixing  Modest increase in
10 to 20 µm; (ii) MWNT2 had an is necessary to distribute the compressive strength
outer dia. of 20-30 nm and lengths MWNTs uniformly within the reported with smallercement-based matrix, which dia. MWNTs.
ranging from 10 to 30 µm.
in
higher-strength  No
 All MWNTs were dispersed using a results
flexure
tests
sonicator for different durations. nanocomposites. This increase performed.
The cement composite samples in strength indicated that
were
properly
were prepared in rotary mixer. nanotubes
anchored within the cement
Superplasticizer was added to
matrix.
mortar mixtures.
 50 mm cubes were prepared for  Higher MWNT contents yield
compression tests, and prismatic 40 reduced strength as additional
3
x 40 x 160 mm specimens for water adheres to MWNT
surfaces;
higher
nanotube
flexure tests.
concentrations also tend to
 Molded specimens were covered
agglomerate, creating weak
with wet cloth for 24 hrs.
zones.
Subsequently, they were demolded,
and placed in lime-saturated water
until tested.

74

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Detail
 Comparison of the  Compressive and flexural Strengths  Workability is a major concern
compressive and were evaluated.
in CNT-reinforced composites,
flexural strengths  Structure was assessed by FE-SEM, and water–cement ratio plays a
of
ordinary TEM. Evaluation of lattice defects vital role. More water is
portland
cement was performed by Raman. FT-IR required to reach proper
mortar
versus and EDX (Energy Dispersive X- workability
in
nanotubeCNT–reinforced
ray) analysis were used to assess reinforced cement composites.
cement mortar.
the presence of reactive groups on The control sample with higher
CNT surfaces. Thermal oxidation of water/cement ratios yielded less
strength,
suggesting
that
CNTs was by (TGA) analysis.
additional water is necessary for
proper hydration and strength
rise results from the reinforcing
action of CNTs.
An increase in mean strength is
observed with up to 0.5%
MWNT addition for both types
of MWNT. Addition of 0.3%
MWNT produced the highest
mean compressive strength.

75

Strengths/Drawbacks

Table 2.1 (cont’d)
Author

Study Objectives /
Detail
Wille and Objectives:
Loh
 Dispersion of low
[2010]
concentrations of
MWNTs in cement
paste, promoting
nanotube–cement
matrix interactions.
 Design of a hybrid
fiber-reinforced
UHPC
that
incorporates
different sizes of
fibers
covering
nanometer
to
micro-scale
size
scales.

Features and Methodology

Observations / Conclusions

Strengths/Drawbacks

 Both UHPC paste and mortar were  By optimizing the dispersion of  Investigation of hybrid
evaluated with 0.22 w/c ratio.
MWNTs in poly(carboxylate reinforcement systems
MWNT
 MWNTs with lengths up to 10 ether)- based SPLs, three initial comprising
micrometer were used at 0.022 goals were accomplished: (i) and microfibers in a
long-term
stable
MWNT high-performance
wt.% of cement.
 MWNTs were mixed in diluted suspensions; (ii) incorporation mortar.
polyelectrolyte or diluted SPL of MWNTs into the mix design  The improvements in
strength
of
solutions.
The
MWNT– without adversely influencing bond
its workability; and (iii) microfibers
with
polyelectrolyte solutions were
subjected to 180 min. of ultrasonic enhancement of the interface addition of MWNT
bath treatment (135 W, 42 kHz) properties between embedded were minor.
followed by 30 min. of high- steel fibers and the concrete  No
gains
in
powered probe sonication (3.175- matrix.
mechanical properties
 Bond strength of steel fibers to were obtained with the
mm tip, 150 W, 22.0 kHz).
 Two different types of high- UHPP and UHPC increased addition of MWNT.
strength steel fibers were used: (i) significantly with the addition
straight fiber with dia. of 0.20 mm of low concentrations of
(1
mg/mL
in
and length of 13 mm; and (ii) MWNTs
twisted fiber with dia. of 0.3 mm dispersion, or 0.022% of cement
weight).
and length of 30 mm.
Materials were mixed using a  MWNTs did not significantly
paddle mixer. Upon mixing, UHPP influence the compressive and
or UHPC was poured into a small flexural strengths of highflow cone to full capacity to performance concrete materials.
measure the spread on fixed plate.

76

Table 2.1 (cont’d)
Author

Study Objectives /
Detail

Collins et Objectives:
al. [2011]  Evaluation of the
dispersion
and
rheology of CNT–
OPC
paste
mixtures, with and
without
dispersants/
surfactants.
 Initial screening of
aqueous solutions
containing CNTs
and
several
generically
different types of
admixtures
that
were designed to
disperse OPC.

Features and Methodology
 Specimens for single fiber pullout,
compression and flexure tests were
cast and slightly compacted on a
vibration table (30 s).
After casting, the specimens were
covered with plastic sheets and
stored at room temperature for 24 h;
they were then removed from
molds, and stored in a water tank at
20°C for 25 days.
 Type I cement paste was used as the
matrix, with 0.3 w/c ratio.
 Three types of commercially
available purified multiwalled
carbon nanotubes (MWNTs) were
used.
 MWNT1020 had an outer dia. of
10-20 nm and lengths of 5-15 or 1-2
µm.
 MWNT1030 had an outer dia. of
10-30 nm and lengths of 5-15 or 1-2
µm.
 MWNT60100 had an outer dia. of
60-100 nm and lengths of 5-15 or 12 µm.

77

Observations / Conclusions

Strengths/Drawbacks

The experimental work yielded  Paste matrix was used
the following conclusions:
to
evaluate
three
types
of
 In the case of CNTs air- different
entrained aqueous solutions, CNTs.
however,
sedimentation  Not
dispersed
occurred within days following properly.
mixing. Mechanical agitation by  Only
increase
in
ultrasonication greatly reduced compressive strength
sedimentation when observed 9 was reported.
days following mixing.
 Ultrasonication
of
polycarboxylate
and
lignosulfonate
admixtures
provided good dispersion of
CNTs in aqueous solutions,
even 9 days

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Detail
 SEM observation  Different
types
of  Air entraining admixture and
of
fractured dispersants/surfactants were chosen lignosulfonate, when included
surfaces
of because of their compatibility when in OPC paste mixtures, resulted
selected mixtures utilized in cement paste.
in reduced consistency and
for evaluating the  Sedimentation in aqueous solution, strength.
distribution
of mini slump and compressive  CNT–OPC paste consistency
CNTs in cement- strength tests as well as SEM was greatly improved in the
based matrix.
investigations was undertaken.
case
of
polycarboxylate
admixture addition, with highly
 Mixing was performed in a
flowable mixtures achieved
planetary mixer.
 Paste cylinders, measuring 50 mm when w/c was as low as 0.35.
diameter by 100 mm height, were At w/c of 0.35, the compressive
tested in compression at 28 days. strength of OPC–CNT–PC was
Following mixing of paste, the 25% higher than companion
molds were filled and compacted on OPC mixtures, indicating the
a vibrating table for 30 s. improved dispersion of CNT
Specimens were then covered with (visible when observed by SEM
polyethylene film to ensure analysis), which has probably
minimal air circulation over the been a key contributing factor
sample. Following elapse of 24 h, toward strength improvement.
the samples were removed from The active non-polar groups
polycarboxylate
molds, and machine-finished at within
molecule disperse CNTs, while
both ends to ensure smoothness.
polar groups disperse cement.

78

Strengths/Drawbacks

Table 2.1 (cont’d)
Author

Study Objectives /
Detail
Chan and Objectives:
Andrawe  Finite
element
s [2010]
(FE) analysis of
improvements in
tensile strength and
ductility
of
cement-based
materials
with
carbon nanotube
reinforcement.
 Use of nano-scale
and
macro-scale
models
to
investigate
the
interfacial
bond
behavior of an
individual
CNT
embedded
in
cement matrix, and
the
flexural
response
of
a
CNT/cement
composite beam,
respectively.

Features and Methodology
 At the first stage of this study,
small-scale were conducted to
investigate the efficacy of using
uniformly dispersed, randomly
oriented CNTs as reinforcement in
cementitious composites.
 Three-point flexure tests were
performed on 13 mm x 26 mm x
165 mm beams.
 The CNTs used in this work were
industrial-grade MWNTs with OD
of 20 nm and length of 5 µm at a
concentration of 0.25% by the total
weight of cement paste.
 20% PVP by weight of CNTs was
added to the mixture to help with
their dispersion.
 The CNT/water mixture was
divided into six smaller portions;
each portion was sonicated for 4 h
first,
which
intensified
the
ultrasonic energy input. After that,
the six portions were mixed
together, and were sonicated for a
day and a half to produce a uniform
mixture.

79

Observations / Conclusions

Strengths/Drawbacks

 It was found that CNTs  One of few numerical
increased flexural strength by investigations
of
about
47%.
Moreover, CNT/cement
toughness increased by 25% nanocomposites.
with CNTs.
 Reported increase in
 Deformation capacity, however, both strength and
did not benefit significantly toughness.
with introduction of CNTs.
 Suggests
that
 A comparison between the tremendous reinforcing
experimental and numerical effects can be realized
behavior of the composite by surface treatment of
beams suggested an effective CNTs.
average shear strength value of
6.5
MPa
at
the
CNT/cementitious
matrix
interface.
 Numerical
models
could
successfully
predict
the
behavior
of
CNT/cement
nanocomposites.

Table 2.1 (cont’d)
Author

Study Objectives /
Features and Methodology
Observations / Conclusions
Detail
 Use
of  After sonication, the mixture was  As predicted by FE analysis,
experimental
added to cement, and the paste was increasing the interfacial shear
results on CNT/ mixed using an electric mixer for 5 strength from 6.5 MPa to 20.0
cement composite min. The final product was placed MPa, would increase the
beams to calibrate in wooden molds in layers, which composite’s flexural strength,
models.
were compacted by a tamping rod. ductility, and toughness by
The specimens were then cured in a 141%, 259% and 1976%,
moist room for 24 h before removal respectively. These findings
from molds. The demolded pointed at the great potential of
specimens were kept in moist room surface-treated
CNTs
as
for 13 days.
reinforcement in cement-based
materials.

80

Strengths/Drawbacks

Table 2.2- Strength test results after 28 days of curing (Li, Wang et al. 2005).
Mix
PCC
PCCF
PCNT

Compressive Strength (MPa)
Flexural Strength (MPa)
52.27 (+/-) 1.4%
6.69 (+/-) 1.5%
47.51 (+/-) 3.1%
8.14 (+/-) 2.6%
62.13 (+/-) 2.3%
8.37 (+/-) 2.1%

Table 2.3- Physical attributes of all graphite nano and micro-scale reinforcement systems.
Reinforcement

Diameter

Length

Manufacturer

MWNT A

20-40 nm

5-15 µm

NanoAmor, Inc.

MWNT B

60-100 nm

5-15 µm

NanoAmor, Inc.

MWNT C

5-15 nm

5-20 µm

NanoLab, Inc.

MWNT D

8-15 nm

10-50 µm-oxidized

Cheap Tubes, Inc.

100 µm- pristine
MWNT E

20-30 nm

10-30 µm-oxidized

Cheap Tubes, Inc.

50 µm- pristine
MWNT F

8-15 nm

Up to 200 µm

Cheap Tubes, Inc.

MWNT G

20-40 nm

Up to 200 µm

Cheap Tubes, Inc.

CNF1

200-500 nm

10-40 µm

Pyrograf Products, Inc.

CNF-OX

60-150 nm

30-100 µm

Pyrograf Products, Inc.

CNF

60-150 nm

30-50 µm

Pyrograf Products, Inc.

GP1

5-10 nm

15 µm

XG Sciences, Inc.

GP

3-90 nm

500 µm (100-150 µm after XG Sciences, Inc.
sonication

CMF (TT143)

6-7 µm

6 mm

Toho Tenax America,
Inc.

(TT 6-7 µm

3 mm

Toho Tenax America,
Inc.

19 mm

New Nycon, Inc.

CMF1
150)

PP Microfiber

50 µm
3

All MWNTs have ~2.1 g/cm true density, and >95% purity.
3

All CNFs have ~1.95 g/cm true density, and >95% purity.

81

Table 2.4- The cementitious matrices selected for evaluation of the reinforcement efficiency of
graphite nanomaterials.
Mix Proportions

DSP Paste

DSP Mortar

DSP Concrete

Silica Fume/Binder

0.20

0.20

0.20

Water/Binder

0.185 (0.24)*

0.185 (0.24)*

0.185 (0.26)*

Superplasticizer/Binder

Adjusted for diff Adjusted for diff Adjusted for diff
reinforcements
reinforcements
reinforcements
0.36
0.36

Silica Sand (0 – 0.18 mm) / Binder

Silica Sand (0.18 -0.5 mm) / Binder -

0.86

0.86

Granite Gravel/ Binder

-

0.50

-

* including the water content of superplasticizer

Table 2.5- Chemical composition of Type I Portland cement.
Compound
C3S

Content (%)
55
19

C2S

10

C3A

7

C4AF
MgO

2.8
2.9

SO3
Ignition Loss
Free CaO

1.0
1.0

82

(a)

(b)
Figure 2.1- Modulus-of-rupture (a) and compression resistance (b) test results for cement
mortars with and without MWNTs (Musso, Tulliani et al. 2009).
83

Figure 2.2- Schematic representation of the experimental setup based on continuous feeding of
cement particles using a screw feeder (Nasibulin, Shandakov et al. 2009).

(a) MWNT D with OD 8-15
Figure 2.3- TEM and SEM micrographs of different nano- and micro-scale reinforcement
systems.

84

Figure 2.3 (cont’d)

(b) MWNT E with OD 20-40 nm

(c) MWNT F with OD 8-15 nm and length up to 200 µm

85

Figure 2.3 (cont’d)

(d) MWNT G with OD 20-40 nm and length up to 200 µm

(e) Oxidized Nanofibers (NF-OX) with 60-150 nm outer diameter

86

Figure 2.3 (cont’d)

(f) Plain Nanofibers (NF) with 60-150 nm outer diameter

(g) Graphite platelet

87

Figure 2.3 (cont’d)

(h) TT 143 microfibers with length of 6 mm

(i) PP microfibers with length 19 mm

88

Figure 2.4- DSP cementitious paste, mortar and concrete.(Guerrini 2000)

89

Figure 2.5- Steric and electrostatic repulsion of separated nanotubes.

Figure 2.6- Dispersion and stability conditions of carbon nanotubes in aqueous media.

90

(a)

(b)

Figure 2.7- Dispersion equipment: (a) Magnetic stirrer (b) Sonic horn.

(a) Mortar Mixer

(b) Compression

(d) Impact

(c) Flexure

(e) Abrasion
Figure 2.8- Experimental setups.

91

Figure 2.8 (cont’d)

(f) Sorption rate

(a) Freeze-Thaw

(b) Wet-Dry

(c) Hot Water Immersion

(d) Elevated Temperature

Figure 2.9- Durability test systems.
92

Figure 2.9 (cont’d)

(e) Sonometer

93

CHAPTER 3

3

REINFORCEMENT OF HIGH-PERFORMANCE CEMENTITIOUS
MATRICES WITH RELATIVELY LOW VOLUME FRACTIONS OF
GRAPHITE NANOMATERIALS

3.1 General
Graphite nanomaterials would realize their reinforcement potential as far as they are thoroughly
dispersed within cement-based matrices, and also develop primary bonds with cement hydrates.
Thorough dispersion of graphite nanomaterials in the aqueous medium of fresh cementitious
matrix encounters challenges associated with the hydrophobic nature of nanomaterial surfaces as
well as the strong tendency towards agglomeration due to the attractive van der Waals forces
over the enormous surface area of graphite nanomaterials. Effective interfacial interactions with
cement hydrates are further challenged by the largely inert nature of nanomaterial surfaces. The
research reported herein developed surface functionalization techniques to improve the
dispersion and interfacial interactions of graphite nanomaterials in cement-based matrices
without compromising their highly desired structure and engineering properties. The covalent
functionalization techniques considered here included acid-oxidation under mild conditions, and
fluorination/carboxylation. The reinforcement efficiencies of different graphite nanomaterials
were evaluated at relatively low volume fractions in a high performance cementitious matrix.

3.2 Introduction
Cement- and concrete-based materials offer suitable engineering properties (compressive
strength, moisture resistance and durability) at low cost, combined with energy-saving and
94

ecological benefits. They are the most widely used class of manufactured materials worldwide.
Cement-based materials offer important advantages over competing construction materials in
terms of ecological impact (pollution, deforestation, use of exhaustible raw materials, etc.)
(Mehta 1993). Cement-based materials, however, lack adequate toughness and impact resistance;
further improvement of their durability in aggressive environments would also yield major
benefits in terms of the life-cycle cost of infrastructure systems. The main thrust of this research
is to make effective use of the unique geometric and mechanical attributes of graphite
nanomaterials towards achieving these needed improvements in concrete properties.
Enhancement of cement-based materials with graphite nanomaterials draws upon the experience
gained with reinforcement of cement-based materials with micro-scale fibers. Graphite
nanomaterials offer unique reinforcement qualities in cementitious materials, which can benefit
applications requiring high levels of structural performance, durability and safety.
Graphite nanomaterials, when compared with today’s (micro-scale) fibers, offer distinctly high
mechanical and physical attributes, tremendous specific surface areas, and close spacing within
matrix even at relatively low volume fractions of about 0.1%. These advantages of graphite
nanomaterials can be realized as far as they are thoroughly dispersed within the cementitious
matrix, and effectively interact with cement hydrates. The high surface area of nanomaterials
generates strong tendencies towards agglomeration via van der Waals interactions. The largely
inert surfaces of graphite nanomaterials also offer limited potentials for strong bonding to cement
hydrates. This paper presents an investigation into modifications of nanomaterials for enhancing
their dispersion and interfacial interactions, and thus reinforcing effects in cementitious material.

95

3.3 Materials and Methods
3.3.1

Graphite Nanomaterials and Microfibers

Different types of acid-functionalized and non-functionalized multiwalled carbon nanotubes as
well as carbon nanofibers, graphite nanoplatelets and chopped (micro-scale) carbon fibers were
evaluated for use in cement-based nanocomposites. These graphite nanomaterials and carbon
microfibers have been introduced in detail in Chapter 2, Section 2.4and Table 2.3. Multiwalled
carbon nanotubes (MWNT A, MWNT B and MWNT C), carbon nanofibers (CNF1, CNF, CNFOX), graphite nanoplatelets (GP1, GP) and carbon microfibers (CMF) were used in highperformance cement-based paste and mortar matrices, Fig 2.3.
Acid Oxidation (Fig. 1.11) involves refluxing graphite nanomaterials in sulfuric/nitric acid (3/1
volume ratio). The process can introduce a relatively high concentration of carboxyl (-COOH)
groups on the surface of graphite nanomaterials; it also generates other (e.g. –OH) groups. After
functionalization, carboxylated nanotubes would typically have 2-7 wt.% COOH (measured by
titration).
Direct fluorination of carbon nanotubes adds fluorine to the sidewalls of nanotubes to
dramatically enhance their chemical reactivity (accompanied by some changes in physical
properties). Direct fluorination is carried out by heating carbon nanotubes in the presence of
hydrogen and fluorine (Fig. 1.13). The reaction causes in-situ generation of HF which, in turn,
acts as catalyst for fluorination. After fluorination, –bonds get more activated, and other
electron-rich molecules can be covalently added to fluorinated surfaces. For effective
interactions with cement hydrates, fluorinated graphite nanomaterials were reacted with succinic
o

acid peroxide (1:10 weight ratio of fluorinated nanomaterial to succinic acid peroxide) at 90 C

96

over 24 hours. The intermolecular elimination of HF in the reaction results in the formation of
ethylcarboxylate-functionalized graphite nanomaterials.
3.3.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cement-based matrices with a smooth particle size gradation covering nano- to microscale range promise to effectively mobilize the tremendous mechanical qualities of graphite
nanomaterials within cementitious nanocomposites. The cementitious matrices selected have
been introduced in detail in Chapter 2, Section 2.5. Both DSP (Densified with Small Particles)
paste and mortar matrices were use to evaluate the reinforcement efficiency of both graphite
nanomaterials and carbon microfibers at relatively low volume fractions.
Cementitious materials with and without functionalized and non-functionalized graphite
nanomaterials dispersed in the mixing water via sonication , described in detail in Chapter 2,
Section 2.5.2 ) and/or microfibers, were prepared following ASTM C 192 and C 305 procedures
as described in the same section.
3.3.3

Experimental Methods

The test procedures employed to determine the engineering properties of cement-based
nanocomposites are described in this section and have been discussed in detail in Chapter 2,
Section 2.6. Compressive Strength tests (ASTM C 109), flexure tests (ASTM C 1185), Impact
tests (ASTM D 7136), abrasion tests (ASTM C 944) and moisture absorption rate tests (ASTM
C 1585) were performed to estimate the engineering and durability properties of highperformance cement-based nanocomposites.
The following experimental methods were also employed to gain further insight into the
chemistry,

crystallinity,

microstructure

and

97

failure

mechanisms

of

cement-based

nanocomposites: (i) SEM; (ii) XRD; (iii) DT/TGA; (iv) XPS; (v) Raman Spectrometry; and (vi)
FTIR. Graphite nanomaterials were also evaluated prior to and after functionalization.

3.4 Characterization of Non-Functionalized and Functionalized Graphite
Nanomaterials
3.4.1

FTIR Spectrometry

ATR FTIR spectra for three types of multiwalled carbon nanotubes and the carbon nanofiber
considered in this experimental program were obtained after acid oxidation (Fig. 3.1a).

Since
-

surfaces of carbon nanotubes and nanofibers were oxidized with nitric acid, the peak at 1648 cm
1

is associated with C=O stretching of carboxylic acids. This indicates that treatment with the

strong oxidizing acid led to attachment of oxygen-containing groups (such as carboxylic and
-1

hydroxyl) to the carbon nanotube and nanofiber surfaces. The peak near 3400 cm

is associated

with the -OH stretching mode of the -COOH group.
Since ATR FTIR offers relatively low detection limit and sensitivity, a transmission FTIR device
equipped with more sensitive MCT (Mercury Cadmium Telluride) detector was used to further
investigate the relatively low concentration of functional groups introduced after acid oxidation
upon carbon nanotubes and nanofibers. Fig. 3.1b and c present the transmission FTIR spectra of
a multiwalled carbon nanotube prior to and after acid oxidation. Like graphite, pristine nanotubes
have featureless FTIR spectra with extremely low infrared absorption intensities. After acid-1

oxidation, two new bands appear at 1715 and 1160 cm , which are attributed to the C=O and CO stretching vibrations, respectively, of the carboxylic group (-COOH). Peaks at 1635 and 1560
-1

cm

-1

are due to C=C stretching. The intense, broad line centered at 3422 cm

98

is associated with

the –OH stretching mode of the –COOH group. The increase in relative intensities of bands at
-1

1068 and 3420 cm

suggests a rise in –OH groups on nanotube surfaces after acid-oxidation.

Similar trends are observed in the case of acid-functionalized carbon nanofiber (Fig. 3.1d).
The FTIR spectra for a multiwalled carbon nanotube prior to and after fluorination and ethyl
carboxylation are presented in Fig. 3.2. After fluorination, the appearance of C=C and C-F bonds
confirms the success of fluorination. No peaks at these wave numbers were observed in pristine
multiwalled carbon nanotubes.

After ethyl carboxylation, a new peak was observed at ~1700

-1

cm confirming carboxylation.
3.4.2

Scanning Electron Microscopy

Comparison of the SEM images of multiwalled carbon nanotubes prior to and after fluorination
(Fig. 3.3) did not reveal appearance of additional defects on nanotube surfaces after fluorination.
The surfaces appear even smoother after fluorination (pointing at “Teflonization”).
3.4.3

Raman Spectrometry

The Raman spectra of two types of multiwalled carbon nanotubes (MWNT A & B) prior to and
after fluorination were used for diagnosis of the nanotube type. The D mode (the disorder band
-1

located between1330-1360 cm , when excited with a visible laser) is associated with multiwalled nanotubes. However, when observed with single-walled nanotubes, one assumes that it is
-

due to the presence of defects. The G mode or (TM- Tangential Mode), located around 1580 cm
1

, corresponds to the stretching mode in the graphite plane. The results presented in Fig. 3.4

suggest that, after fluorination, MWNT A shows downshifts in the positions of both D (~ 1300
-1

-1

cm ) and G (near 1600 cm ) peaks due to the fluorine groups attached to the nanotube
99

-1

sidewalls. However, MWNT B shows a 4 cm

downshift of only the G peak due to a smaller

bulk amount of fluorine as compared to the smaller-diameter MWNT A.
3.4.4

X-Ray Photoelectron Spectroscopy (XPS)

Table 3.1 presents the elemental analyses obtained for MWNT A (20-40 nm diameter) and
MWNT B (60-100 nm diameter) in pristine condition and after fluorination (at different
temperatures) and subsequent ethyl carboxylation, confirming the introduction of fluorine and
oxygen atoms upon fluorination and subsequent carboxylation, respectively.
Introduction of fluorine and oxygen atoms with electron negativity onto the MWNT surfaces can
significantly change the chemical environment of carbon atoms.

Figs 3.5a and 3.5b show the

C1s high resolution spectra (red curves) obtained for fluorinated (at two different temperatures)
and subsequently carboxylated MWNTs (with two different diameters). Unlike pristine
nanotubes, an asymmetric peak is observed for the C1s high resolution XPS spectrum.

This

indicates that carbon atoms are in different environments, which is an indication of successful
functionalization. The main peak centered at 284.12 eV (Electron Volts) was assigned to carbon
atoms bound to hydrogen atoms. The peak at 284.32 eV originated from the sp3–hybridized
carbon atoms. Peaks at 285.57, 288.65 and 290.95 eV are assigned to carbon atoms bound to
oxygen atoms and fluorine atoms, respectively. Carbon atoms bound to oxygen and fluorine
could lead to higher binding energies because of their high electron negativity.

100

3.5 Engineering Properties, Structure and Failure Modes of Cementitious
Nanocomposites
3.5.1

Engineering Properties

Typical flexural load-deflection behavior of the high-performance cementitious (DSP) paste
without any reinforcement and with 0.08 vol.% (volume fractions of dry cementitious materials)
of different acid-functionalized graphite nanomaterials (and micro-scale carbon fibers) are
presented in Fig. 3.6. With proper functionalization, graphite nanomaterials are observed to yield
(at only 0.08 vol.%) important gains in the flexural strength and energy absorption capacity (area
underneath the load-deflection curve) of cement-based matrix.
Fig. 3.7 with mix designation given in Table 3.2, summarizes the effects of different graphite
nanomaterial types and functionalization conditions, at 0.08 vol.%, on the flexural performance,
abrasion resistance, moisture sorption rate and compressive strength of the high-performance
cementitious (DSP) paste. Analysis of variance of test results confirmed that functionalized
graphite nanomaterials at 0.08 vol.% yield statistically significant gains (at 0.05 significance
level) in the flexural strength and energy absorption capacity, abrasion resistance and moisture
sorption resistance of the high-performance cementitious paste. The corresponding effects of
graphite

nanomaterials

on

compressive

strength

were

not

statistically

significant.

Functionalization of graphite nanomaterials via acid-oxidation and, to a generally smaller extent,
fluorination/carboxylation, led to improved reinforcement efficiency of graphite nanomaterials in
cementitious paste. Analysis of variance, followed by pair-wise comparisons, of test results
indicated that the acid-oxidation effects on the reinforcement efficiency of graphite
nanomaterials at 0.08 vol.% were statistically significant at 0.05 significance level as far as

101

flexural strength and energy absorption capacity, maximum deflection and abrasion resistance
are concerned. In the case of resistance to moisture sorption, the acid-oxidation effect was
statistically significant at 0.02 significance level. The maximum improvement in all engineering
properties was achieved with oxidized MWNT B.
Analysis of variance of test results followed by pair-wise comparisons indicated that the microscale carbon fiber is inferior to all nano-scale reinforcement systems considered here (at 0.05
significance level) as far as the flexural strength and energy absorption capacity, and moisture
sorption resistance of the cementitious paste are concerned. The large variations in abrasion test
results lowered the statistical confidence in superiority of graphite nanomaterials over the microscale carbon fiber in improving the abrasion resistance of cementitious paste.
Graphite nanoplatelets were introduced in non-functionalized form into the high-performance
(DSP) cementitious paste at 0.08 vol.% with the primary objective of lowering the moisture
sorption rate by forcing a tortuous diffusion paths. The mean moisture sorption rate was lowered
by more than 50% with 0.08 vol.% of non-functionalized graphite nanoplatelets. This effect was
statistically significant at 0.05 level of significance.
Fig. 3.8 with mix designations in Table 3.3, summarizes the effects of introducing a very low
volume fraction (0.03 vol.%) of acid-functionalized multi-walled carbon nanotube and oxidized
carbon nanofiber on the flexural performance, abrasion resistance, moisture sorption rate and
compressive strength of the high-performance (DSP) mortar. The gains in flexural energy
absorption capacity, maximum deflection and abrasion resistance were statistically significant at
0.05 level of significance for both MWNT A and B. The gains in flexural strength and
compressive strength were not statistically significant for this very low nanomaterial volume
fraction.

102

The introduction of acid-functionalized multiwalled carbon nanotube MWNT B led to a more
complex failure mode (involving more extensive cracking) of the high-performance cementitious
(DSP) paste under impact loading (Fig. 3.9a). The failure modes in flexure (Fig. 3.9b) and
abrasion (Fig. 3.9c) did not undergo obvious changes with introduction of ≤0.08 vol.% of
graphite nanomaterials.
3.5.2

Transmission and Scanning Electron Microscopic Investigations

The failed surfaces of flexure and compression test specimens were evaluated under a highprecision scanning electron microscope (JOEL JM-6300F). All samples were coated with
Osmium (Using Osmium Coater Neoc-AN, Meiwa Shoji) prior to SEM observations.
A transmission electron microscope (TEM) image of the multiwalled carbon nanotube MWNT B
after acid-oxidation followed by nucleation of cement hydrates on their surfaces is shown in Fig.
3.10a. One can distinguish the nucleated cement hydrates formed upon the functionalized
nanotube surfaces. SEM images of the cementitious matrix with 0.04% volume fraction of
multiwalled carbon nanotubes subjected to acid-oxidation followed by nucleation of cement
hydrates (Fig. 3.10b) suggest that the carbon nanotubes exposed at the failed surface of flexure
test specimens are thoroughly coated with cement hydration products. Debonding seems to have
taken place away from the actual interface within the matrix, which points at the adequately
strong bond developed between the functionalized nanotubes and the cementitious matrix. The
SEM images of Fig. 3.10b provide indications for the uniform dispersion of individual acidfunctionalized carbon nanotubes within the cementitious matrix.
SEM images of cementitious pastes reinforced with 0.08% volume fraction of acidfunctionalized graphite nanomaterials provided strong evidence for the bridging and pull-out

103

actions of carbon nanotubes and nanofibers at microcracks (Fig. 3.11), which are key to their
contributions to the flexural strength and energy absorption capacity of cementitious matrices.
While the functionalized multiwalled carbon nanotubes and nanofibers were largely dispersed
within the cementitious paste, some bundling of nanomaterials occurred at the higher volume
fraction of 0.08 vol.%. Even the bundled graphite nanomaterials exhibited positive effects
through arrest and deflection of microcracks (Fig. 3.12).
The cementitious paste matrix with 0.08 vol.% of acid-functionalized microfibers exhibited
thorough dispersion of fibers, with relatively large fiber spacing. The prevalent failure mode
involved pullout of microfibers (Fig. 3.13).
3.5.3

X-Ray Diffraction (XRD) Analysis

XRD Analysis was carried out on two cementitious pastes with 0% and 0.04 vol.% of acidfunctionalized MWNT A using Rotaflex Rigaku 200B Analyzer. The spectrum was analyzed
using JADE to identify various elements according to their crystalline structure.
The XRD graphs of the two cementitious materials (Fig. 3.14) exhibit no significant differences,
indicating that there were no major changes in the crystalline structure of the matrix with
introduction of a low volume fraction (0.04 vol.%) of acid-functionalized multi-walled carbon
nanotube. The two major compounds identified in these two cementitious materials were calcium
silicate (Ca3SiO5) and calcium silicate oxide (Ca3(SiO4)O). As compared to the XRD graph of a
conventional cement mortar, the typical calcium hydroxide peaks were not observed; this points
at substantial conversion of calcium hydroxide to calcium silicate hydrate via pozzolanic
reactions involving silica fume.

104

3.5.4

Thermogravimetric Analysis (TGA)

Thermogravimetric analyses of the high-performance cementitious paste matrix, plain or with
acid-functionalized graphite nanomaterials were performed using the TA Instrument Hi-Res
TGA 2950 Thermogravimetric Analyzer. A sample of conventional mortar having 1: 1.22: 0.45,
cement: sand: water ratio was also tested in order to identify their differences. A finely ground
powder (passing 100 mesh) of the material was used in thermogravimetric analysis. The samples
o

o

were heated at a constant rate of 10 C per minute up to 1000 C, with weight loss recorded as a
function of temperature. Tests were conducted 21 days after casting of the cementitious samples.
The results of thermogravimetric analysis are summarized in Fig. 3.15; some interpretations of
these results are presented below.


o

o

The peak in weight loss between 450 C and 500 C is assumed to be due to C-H
dehydration. This peak is quite strong for conventional mortar, and weak for all highperformance (DSP) matrices, which indicates the presence of more calcium hydroxide in
conventional mortar than in DSP matrices. This has been confirmed by scanning electron
microscopic analyses where C-H crystals are found only rarely in DSP matrices. This
finding could be anticipated because the relatively high dosage of silica fume used in
DSP matrices converts C-H to C-S-H through pozzolanic reactions.



o

o

The initial loss of mass from 100 C to 122 C is assumed to be due to the loss of
o

o

physically held water, and from 600 C to around 797 C due to high-temperature
decomposition of CaCO3 (Calcium Carbonate). The physically held water content of DSP
paste is greater than that of conventional mortar, which could be attributed to the higher

105

surface area of the hydration products present in DSP paste. The young age of the
cementitious materials tested implies that the CaCO3 content is relatively small. Over
time, the higher C-H content of the conventional mortar (which is prone to carbonation)
could increase its CaCO3 content when compared with DSP paste.


o

o

The mass loss from 819 C to 899 C, observed only for DSP samples which contain silica
fume, is characteristic of the calcium silicate hydrate (C-S-H(I)) with CaO/SiO2<1.5,
which forms primarily through pozzolanic reaction. This effect is caused by the
reorganization of the lattice network of anhydrous calcium silicate.



For the purpose of evaluating the chemically bound water in C-S-H, with a certain degree
o

o

of approximation, only the mass loss between 100 C to 400 C was considered. This
approximation helps avoid the errors caused by the possible presence of CaCO3 and the
o

mass loss associated with its decomposition at higher temperatures (around 800 C), and
o

by the presence of free water below 100 C. The maximum weight loss of 9.49% occurs
for DSP Paste containing 0.08 vol.% of carbon nanofiber, followed by 7.88% for the case
with acid-functionalized MWNT B. The acid-functionalized and fluorinated MWNT A
had weight losses of 6.57% and 6.29%, respectively, as compared to 6.87% for plain DSP
paste. In conventional mortar, the total weight loss was 3.93%. These results indicate that
more water is chemically bound with C-S-H in the high-performance cementitious
material (DSP paste) when compared with conventional mortar. The effect of a low

106

volume fraction (0.08%) of functionalized nanotubes on the C-S-H content of DSP paste
is observed to be relatively small.


Total weight loss is greater for the high-performance cementitious material (DSP paste),
at 18.0% for plain DSP paste, and 17.5% with fluorinated MWNT A, 17.4% with carbon
nanofiber, 17.3% with MWNT B and 16% with acid-functionalized MWNT A as
compared to 10.2% for the conventional mortar. This could result from a greater content
of hydration products in DSP paste when compared with the conventional mortar (which
partly results from the presence of sand in conventional mortar).

The findings of thermogravimetric analysis confirm the outcomes of XRD analysis, which
suggests that the introduction of acid-functionalized graphite nanomaterials does not alter the
chemistry and crystalline structure of cement hydration products.
3.5.5

Energy-Dispersive Spectroscopy (EDS)

Elemental analyses were performed on 2.5x12x12 mm samples of cementitious materials with
and without acid-functionalized graphite nanomaterials. The specimens were coated with carbon
(using Carbon String Evaporator, Ernest F. Fullam) to make them more conductive for the
purpose of EDS analysis. Elemental (EDS) analyses of cementitious matrices with 0 and 0.08
vol.% of acid‐functionalized carbon nanotube or nanofiber indicated that cementitious matrices
with graphite nanomaterials exhibit elemental compositions similar to those of the plain
cementitious matrix, except for the higher carbon content in systems incorporating graphite
nanomaterials. This is in line with the findings of x-ray diffraction and thermogravimetry
analyses, indicating that introduction of functionalized graphite nanomaterials did not alter the
chemistry of cement hydration products.

107

3.6 Summary
Graphite nanomaterials (multi-walled carbon nanotube, carbon nanofiber and graphite
nanoplatelet) were evaluated as reinforcement systems in cement-based materials. While the
excellent mechanical and physical characteristics as well as the distinct geometric attributes of
graphite nanomaterials promise superior reinforcement efficiency in cement-based matrices,
there are challenges associated with their dispersion and interfacial interactions within cementbased materials. Uniform dispersion of graphite nanomaterials in an aqueous medium is
complicated by the predominantly hydrophobic nature of nanomaterial surfaces and the strong
tendency towards agglomeration driven by van der Waals attractions between the enormous
surface areas of graphite nanomaterials. The largely inert surfaces of nanomaterials cannot
effectively bond to cement hydrates, which lower the interfacial stress transfer needed to
mobilize the enormous mechanical qualities of graphite nanomaterials. Complementary covalent
functionalization techniques were developed to address these challenges and improve the
reinforcement efficiency of graphite nanomaterials in cement-based matrices. The methods
developed for covalent functionalization included acid-oxidation under mild conditions, and
fluorination/ethyl carboxylation, both of which introduced carboxyl groups via primary bonds
upon graphite nanomaterials. Preparation of cement nanocomposites involved careful dispersion
of functionalized graphite nanomaterials via sonication in the mixing water, followed by
conventional mixing, molding and steam curing of the nanocomposites. The nanomaterial
volume fractions considered included 0.04% and 0.08 vol.% (in cementitious pastes) and 0.03
vol.% in mortar. Replicated mixes and tests were considered to provide the basis for statistically
reliable inferences.

108

Table 3.1- Elemental analysis of fluorinated MWNTs and ethyl carboxylated MWNTs.

Sample

C1s

F1s

O1s

MWNT A

97.4
70.4

1.0
25.8

1.6
3.7

150 C F-MWNT A-(CH2)2COOH

74.2

19.7

6.1 (notable ethyl
carboxylation)

250 C F-MWNT A

69.8

29.1

1.1

250 C F-MWNT A-(CH2)2COOH

72.7

24.8

2.6 (notable ethyl
carboxylation)

150 C F-MWNT B

70.2

25.9

3.9

150 C F-MWNT B-(CH2)2COOH

74.7

20.8

4.4 (notable ethyl
carboxylation)

250 C F-MWNT B

63.5

33.7

2.8

250 C F-MWNT B-(CH2)2COOH

65.4

31.2

3.4 (minor ethyl
carboxylation)

o

150 C F-MWNT A
o
o
o
o
o
o
o

Table 3.2- Mix designations of DSP pastes reinforced with non-functionalized and functionalized
graphite nano and micro reinforcements at 0.08 vol.%.
Reinforcement Condition
Plain
Non-functionalized MWNT A, 0.08 vol.%
Acid-functionalized MWNT A, 0.08 vol.%

Mix
Designation
1
2
3

Fluorinated MWNT A, 0.08 vol.%

4

Non-functionalized MWNT B, 0.08 vol.%

5

Acid-functionalized MWNT B, 0.08 vol.%

6

Fluorinated MWNT B, 0.08 vol.%

7

Acid-functionalized CNF1, 0.08 vol.%

8

GP1, 0.08 vol.%

9

Acid-functionalized CMF, 0.08 vol.%

10

vol.% - volume percent with respect to dry cementitious materials

109

Table 3.3- Mix designation of high-performance DSP mortar with functionalized graphite
nanomaterials at 0.03 vol.%.
Reinforcement Condition
Plain
Acid-functionalized MWNT A, 0.03 vol.%
Acid-functionalized MWNT B, 0.03 vol.%

Mix
Designation
1
2
3

Acid-functionalized Carbon Nanofiber, 0.03 vol.%

4

vol.% - volume percent with respect to dry cementitious materials

110

(a) FTIR spectra for three types of carbon nanotubes and one type of carbon nanofiber after
acid oxidation
Figure 3.1 - FTIR spectra of pristine and acid-functionalized carbon nanotubes and nanofibers.

111

Figure 3.1 (cont’d)

(b) MWNT A prior to and after acid-oxidation

112

Figure 3.1 (cont’d)

(c) MWNT B prior to and after acid-oxidation

113

Figure 3.1 (cont’d)

(d) Acid-oxidized carbon nanofiber

(a)
Figure 3.2- FTIR spectra for MWNTs A prior to and after fluorination, and after ethyl
carboxylation upon fluorination.
114

Figure 3.2 (cont’d)

(b)

Figure 3.3 - SEM images of MWNTs B prior to and after fluorination.

115

 
(a)

(b)
Figure 3.4- Raman spectra for MWNT A and B prior to and after fluorination.

116

Figure 3.5- XPS spectrum.
117

Figure 3.5 (cont’d)

118

Figure 3.6-Typical flexural load-deflection behavior of DSP paste with 0% and 0.08% volume
fractions of various non-functionalized and acid-functionalized graphite nanomaterials.

119

M
W W
N N
T T
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Flexural Strength (MPa)

16

15

14

13

12

11

10

9

8

7

6

Reinforcement Type

(a)

Figure 3.7- Mean and standard error of various engineering properties of DSP paste reinforced
with non-functionalized and functionalized graphite nano and micro reinforcements at 0.08
vol.%.

120

500

400

300

200

100

M
W W
N N
T T
Energy Absorption Capacity (N.mm)
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Figure 3.7 (cont’d)

Reinforcement Type

(b)

121

M
W W
N N
T T
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Maximum Deflection (mm)

Figure 3.7 (cont’d)

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

Reinforcement Type

(c)

122

M
W W
N N
T T
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Compressive Strength (MPa)

Figure 3.7 (cont’d)

200

150

100

50

Reinforcement Type

(d)

123

M
W W
N N
T T
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Loss of Mass (grams)

Figure 3.7 (cont’d)

0.5

0.4

0.3

0.2

0.1

0.0

Reinforcement Type

(e)

124

0.00008

0.00007

0.00006

0.00005

0.00004

0.00003

0.00002

M
W W
Moisture Sorption Rate (mm/sec^0.5)
N N
T T
M A-N A-F
W o lN nF 0.0
T
M
W M A- n-0 8 v
N W O .0 ol
T N x- 8 .%
M B-N T B 0.0 vol
8 .
M WN on -Fl- vo %
ic T Fn 0. l.
r
N ofi B-O -0. 08 %
an b x 08 vo
e N ofib r-O 0.0 vol l.
an e x 8 . %
op r-O -0. vo
la x- 08 l.%
te 0. vo
le 08 l.
t-0 v %
.0 ol
8 .%
vo
l.
N %
on
e

M

Figure 3.7 (cont’d)

Reinforcement Type

(f)

125

Flexural Strength (MPa)

25

20

15

M

N
on
e

W
N
T

A
-O
M
xW
0.
N
03
T
B
vo
-O
N
l.%
an
xof
0.
03
ib
er
vo
-O
l.%
x0.
03
vo
l.%

10

Reinforcement Type
(a) Flexural strength
Figure 3.8- Effects of functionalized graphite nanomaterials at 0.03% on the engineering
properties of high-performance DSP mortar (means & standard errors).

126

(b) Energy absorption capacity

127

N
on
e

A
-O
M
xW
0.
N
03
T
B
vo
-O
N
l.%
an
xof
0.
03
ib
er
vo
-O
l.%
x0.
03
vo
l.%

T

N

W

M

Energy Absorption Capacity (N.mm)

Figure 3.8 (cont’d)

400

350

300

250

200

Reinforcement Type

Figure 3.8 (cont’d)

Maximum Deflection (mm)

0.9
0.8
0.7
0.6
0.5
0.4
0.3

N
on
e

W
N
T
M

M

W
N
T

A

-O

x-

0.

03
B
vo
-O
N
l.%
an
xof
0.
03
ib
er
vo
-O
l.%
x0.
03
vo
l.%

0.2

Reinforcement Type
(c) Maximum Deflection

128

(d) Compressive strength

129

N
on
e

A
-O
M
xW
0.
N
03
T
B
vo
-O
N
l.%
an
xof
0.
03
ib
er
vo
-O
l.%
x0.
03
vo
l.%

T

N

W

M

Compressive Strength (MPa)

Figure 3.8 (cont’d)

200

150

100

50

Reinforcement Type

Figure 3.8 (cont’d)

Loss of Mass (grams)

1.2
1.0
0.8
0.6
0.4
0.2

M

W

N

T

N
on
e

A

-O
M
xW
0.
N
03
T
B
vo
-O
N
l.%
an
xof
0.
03
ib
er
vo
-O
l.%
x0.
03
vo
l.%

0.0

Reinforcement Type
(e) Abrasion weight loss

130

0.00008
0.00007
0.00006
0.00005
0.00004
0.00003
0.00002

03

03
N
an
of
ib

er

-O

x-

0.

0.
xB
-O

N
on
e

l.%
vo

vo
03
0.
xM

W
N
T

A
-O
W
N
T
M

vo
l.%

0.00001
l.%

Moisture Sorption Rate (mm/sec^0.5)

Figure 3.8 (cont’d)

Reinforcement Type
(f) Sorption rate

Plain matrix

0.04 vol.% MWNT B
(a) Impact

Figure 3.9- Visual appearances of failed specimens after different tests.

131

Figure 3.9 (cont’d)

(b) Flexure

(c) Abrasion

(a) TEM image

(b) SEM images
Figure 3.10- TEM image of functionalized MWNT B, and SEM images of the fractured surface of
a cementitious matrix with 0.04 vol.% of functionalized MWNT C.
132

(a)

(b)
Figure 3.11- Bridging of micro-cracks by acid-functionalized graphite nanomaterials introduced
into cementitious matrices at 0.08% volume fraction.
133

Figure 3.12- Arrest and deflection of a micro-crack by a bundle of CNFs within the cementitious
matrix.

Figure 3.13- SEM images of DSP paste with 0.08 vol.% of carbon micro-fiber exhibiting
showing pulled-out micro-fibers on a crack surface.
134

1200

Intensity (CPS)

1000

MWNT A, 0.04 vol.%

800
DSP Paste Plain
600
400
200
0
20

30

40

50

60

70

Two-Theta (deg)
Figure 3.14- XRD graphs of the high-performance cementitious paste matrix with 0% and 0.04
vol.% of acid-functionalized MWNT A.

135

1000

900

800

700

600

500

400

300

200

100

0
0

Loss of Weight (mg)

-0.002
-0.004
-0.006

Conventional Mortar

-0.008

DSP Paste Plain
0.08% vol.% MWNT A, Acid
Oxidized
0.08% vol.% MWNT B, Acid
Oxidized
0.08% vol.% MWNT A, Fluorinated

-0.01
-0.012
-0.014

0.08% vol.% CNF1, Acid Oxidized

-0.016

Temperature(C)
Figure 3.15- Results of TGA for conventional mortar and for the DSP paste with 0% and 0.08 vol.% acid-functionalized MWNT A,
MWNT B, carbon nano-fiber, fluorinated MWNT A.

136

CHAPTER 4

4

EFFECT OF REFINED FUNCTIONALIZATION OF NANOTUBES ON
THEIR REINFORCEMENT EFFECIENCY AT DIFFERENT VOLUME
FRACTIONS IN CEMENTITIOUS MATRICES

4.1 General
Carbon nanotubes offer distinct engineering and geometric attributes for use as reinforcement in
cementitious materials. The tendency of nanotubes towards clustering, via van der Waals
interactions over their large specific surface area, and the hydrophobic nature of their surfaces
challenge efforts to make effective use of the distinct qualities of nanotubes in cementitious
materials. Functionalization/modification of the nanotube surfaces can enhance their wetting
attributes for improved dispersion in aqueous media, and their potential for bonding with cement
hydrates. Such surface functionalization/modification of nanotubes should not damage their
structure and properties. The objective of modifying nanotube surfaces for increasing the density
of hydrophilic groups on carbon nanotubes without damaging their structure can be achieved by
wrapping nanotubes using selected polymer chains, which does not require introduction of
defects or shortening of nanotube (unlike commonly used nanotube functionalization techniques,
including acid-oxidation). Polymer-wrapping of nanotubes can be realized through: (i)
physisorption of (water-soluble) polymer molecules upon COOH-functionalized nanotubes
(O'Connell 2001; Banerjee, Kahn et al. 2003); or (ii) covalent tethering of polymers to nanotube
walls (Xie, Xu et al. 2007). Other functionalization strategies have also been explored (Hawley
2002; Bensebaa, Zavaliche et al. 2004; Kubota, Sano et al. 2005).

137

4.2 Materials and Methods
4.2.1

Materials

Gum Arabic from acacia tree (GA, Reagent grade) and poly(acrylic acid) (PAA, average Mw
~100,000, 35 wt.% in H2O), N,N′-dicyclohexyl carbodiimide (DCC, 99% purity), N,Ndimethylformamide (DMF, ACS reagent, ≥99.8%), ethanol (ACS reagent, ≥99.5%), N-(3dimethylaminopropyl)-N′-ethylcarbodiimide (EDC, ≥97.0% purity) and N-hydroxysuccinimide
(NHS, 98% purity, 1-Hydroxy-2,5-pyrrolidinedione) were purchased from Sigma-Aldrich.
Glenium® 7700 (40 wt.% solution in water) was purchased from BASF (Master Builders).
Polyethyleneimine (PEI, branched, Mw ~70,000, 30% w/v aqueous solution) was purchased
from Alfar Aesar, Ward Hill, MA. Two types of both carboxylic acid-functionalized and nonfunctionalized multiwalled carbon nanotubes (MWNT D and E) were used, introduced in detail
in Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. Two longer non-functionalized multiwalled
carbon nanotubes (MWNT F and G) were also used. All multiwalled carbon nanotubes had a
purity of >95%. Deionized (DI) water was used for all solution preparations.
4.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cement-based matrix selected has been introduced in detail in Chapter 2, Section 2.5.1 and
Table 2.4. DSP (Densified with Small Particles) paste was use to evaluate the reinforcement
efficiency of different multiwalled carbon nanotubes with different surface treatments.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication, described in detail in Chapter 2, Section 2.5.2) were prepared and
cured following ASTM procedures described in the same section.

138

4.2.3

Experimental Methods

The test procedures employed for determining the engineering properties of cement-based
materials were Compressive Strength test (ASTM C 109) and Flexural Strength test (ASTM C
1185). The following experimental methods were also employed to gain further insight into the
chemistry,

crystallinity,

microstructure

and

failure

mechanisms

of

cement-based

nanocomposites: (i) SEM; (ii) XRD; (iii) DT/TGA; and (vi) FTIR. The details of these
experimental tests and methods have been covered in Chapter 2, Section 2.6.

4.3 Further Functionalization and Characterization of Carbon Nanotubes
4.3.1

Polymer Physisorption

Fig. 4.1 schematically depicts the adsorption of polymer/surfactants upon the nanotube sidewalls.
The following three polymers were chosen to evaluate the potential of polymer physisorption in
enhancing the dispersion and reinforcing effects of nanotubes in cement-based matrices: (i) Gum
Arabic (GA); (ii) poly(acrylic acid) (PAA); and (iii) Glenium® (which is also used as a
superplasticizer in the project).
The above mentioned polymers were introduced into DI water (1 wt.%), and the mixture was
stirred to dissolve the polymers. Subsequently, a dispersion of either non-functionalized or acidfunctionalized nanotubes was introduced (1 wt.%, with the dispersion mixed thoroughly
overnight). The volume of the mix was adjusted to the targeted level (for use as the mixing water
of the cementitious material) by addition of water. The dispersion was then sonicated, first at
different amplitudes (30%, 45% , 65% and 75%) for 10 minutes each (with one-minute break
intervals in between), and then subjected to three ten-minute intervals of pulsing (1 minute on 30
seconds off) at 85% amplitude with two-minute breaks in between. Finally, the whole sonication

139

process was repeated one more time. The resulting nanotube dispersion was immediately mixed
with the cementitious paste; it was also characterized through Fourier Transform Infrared
Spectrometry (FTIR).
Fig. 4.2 shows the FTIR spectra of acid-functionalized nanotubes, Gum Arabic, and the acid-

functionalized nanotubes modified with Gum Arabic. The appearance of new peaks at 1600 cm
1

-1

, 1720 cm

-1

and 3300 cm

indicate that GA is physisorbed upon nanotubes.
-1

appearance of a new broad peak around 2900 cm

Further

due to alkane stretching confirms the

presence of a polymer chain.
The FTIR spectra for acid-functionalized nanotubes, PAA and PAA-modified functionalized
nanotubes are shown in Fig. 4.3. The appearance of new peaks for PAA-modified nanotubes at
-1

1720 cm

indicates that PAA is physisorbed upon acid-functionalized nanotubes.
-1

appearance of a new broad peak around 3000 cm

Further

indicates the presence of free OH groups.

Hence, FTIR spectra confirm the presence of polymer chains associated with nanotubes.
Fig. 4.4 shows the FTIR spectra of acid-functionalized nanotubes, Glenium®, and Glenium®modified functionalized nanotubes. The appearance of new peaks for Glenium®-modified
-1

-1

-1

nanotubes at 1545 cm , 1720 cm

and 3200 cm

point at the association of Glenium® with
1

acid-functionalized nanotubes. The peak at 1720 cm- corresponds to the carbonyl group of
-1

Glenium®, and the one at 3200 cm
groups in Glenium®.

corresponds to the OH stretching of the carboxylic acid
-1

Further appearance of a new set of peaks around 2900 cm

is due to

alkane stretching, which confirms the presence of a polymer chain. The appearance of all these

140

peaks confirms that Glenium® is incorporated upon nanotube walls, probably through noncovalent interactions.
4.3.2

Covalent Tethering of Polymers onto Nanotube Walls

While nanotube dispersions can be prepared using surfactants and physisorbed polymers, the
interactions between nanotubes and polymers or surfactants are non-covalent in these cases.
Such interactions tend to weaken upon changes in pH, ionic strength and other attributes of the
medium. In addition, polymer wrapping decreases the effective aspect ratio of the reinforcement.
Covalent attachment of functional groups or polymers onto the nanotube wall can improve the
load transfer efficiency. Oxidation is one of the most commonly used covalent functionalization
methods, in which a mixture of sulfuric acid and nitric acid is commonly used to form carboxylic
acid groups on nanotube surfaces. This approach involves tethering of water-soluble polymers
upon carbon nanotubes (Fig. 4.5) via covalent bonding to the COOH groups introduced at
nanotube surfaces by acid-oxidation. Two water-soluble polymers were chosen to be tethered to
carboxylated nanotubes: (i) poly(acrylic acid) (PAA); and (ii) Glenium®. Both these polymers
have high densities of COOH groups, and have been shown to improve the dispersability of
nanotubes and their bonding to cementitious matrices.
PAA and Glenium® were chosen here as the anchoring polymers because of their watersolubility and high density of COOH groups. The process of tethering these polymers onto
nanotube walls starts with acid-oxidation of nanotubes, resulting in covalent attachment of the
carboxylic acid (COOH) groups onto the surface of nanotubes. This is followed by a coupling
reaction of the COOH groups on nanotube surfaces with the COOH groups of PAA or
Glenium® in the presence of coupling reagents (e.g, dicyclohexylcarbodiimide - DCC), as
schematically depicted in Figs. 4.6 and 4.7. This procedure will covalently tether the polymer

141

chains onto the nanotube surfaces. Thorough dispersion of the resulting modified nanotubes in
water was accomplished through sonication with a sonic probe as described earlier.
The FTIR spectra of carboxylated nanotubes, PAA and PAA-tethered carboxylated nanotubes
are shown in Fig. 4.8. The appearance of new peaks for PAA-CNTCOOH (versus CNT-COOH)
-1

-1

-1

-1

-1

-1

-1

at ~3300 cm , 2900 cm , 2800 cm , 1625 cm , 1573 cm , 1400 cm , 1300 cm and 1200
-1

cm

points at tethering of PAA via COOH onto CNT. The appearance of a new set of peaks
-1

around 2900 cm

-1

and 2800 cm , due to alkane (CH2, CH3 and CH) stretching, confirms the
-1

presence of a polymer chain. The peak at ~3300 cm is due to OH stretching in COOH groups;
this is usually a broad peak, but it is sharp here indicating that the OH groups are not H-bonded.
-1

Peaks at 1625 cm

-1

and 1573 cm , due to carbonyl stretching, occur lower than the

corresponding neat PAA carbonyl peaks, indicating that the COOH groups of PAA are being
-1

-1

reacted. The medium peak at 1400 cm is due to C-OH bending. Peaks at 1300 cm and 1200
-1

cm

are due to C-O stretching. In short, the FTIR spectra shown in Fig. 4.8 confirm that PAA

is covalently bonded to carboxylated nanotubes.
Fig. 4.9 shows the FTIR spectra of carboxylated nanotubes, Glenium®, and Glenium®-tethered
carboxylated nanotubes. Appearance of new peaks in Glenium®-CNTCOOH (versus CNT-1

-1

COOH) at 1545 cm , 1720 cm

-1

and 3200 cm

indicates that Glenium® is tethered via COOH
-1

groups onto nanotubes. Further appearance of a new set of peaks around 2900 cm , attributed to
alkane stretching, confirms the presence of polymer chains in association with nanotubes.

142

4.3.3

Further Functionalization of Carboxylated Nanotubes Starting from Carboxyl
Groups

Amino groups can be attached to carboxylated nanotubes in a way similar to those described
above for attachment of carboxylic acid polymers. In an effort to employ amino groups towards
further modification of carboxylated nanotubes, the water-soluble polymer polyethyleneimine
(PEI), which carries amino groups, was chosen to interact with the carboxyl groups of nanotubes,
as schematically depicted in Fig. 4.10, and also with cement hydrates (through mechanisms
similar to COOH groups). Further functionalization of carboxylated nanotubes with PEI was
accomplished in the presence of N-hydroxysuccinimide (NHS); the COOH groups on
carboxylated nanotubes were activated to form NHS esters. Upon introduction of N,Ndicyclohexyl carbodiimide (EDC), PEI can react with the activated ester groups to form amide
bonds between PEI and nanotubes.
As a first step, EDC (1 M) was mixed together with NHS (1 M) in the required amount of DI
water, and stirred to achieve thorough dissolution. Carboxylated nanotubes were then added to
the mixture (at 0.5 wt.%), followed by further stirring for 30 minutes. Conversion of the
carboxylic acid groups of CNT-COOH into active esters (CNT-NHS) would occur upon
exposure to the solution of EDC (1 M) and NHS (1 M) in water. Subsequently, PEI (0.5 wt.%)
was added to the mix, and it was left for 24 hours at room temperature. During this period, the
amino groups of PEI would covalently bind to CNT-NHS. After 24 hours of stirring, the residue
(amine-anchored nanotube) was filtered, rinsed well with ethanol and DI water, and dried. It was
then mixed with water (10 g/L), and stirred overnight for thorough mixing. Thorough dispersion
of the modified nanotubes in water was accomplished through sonication with a sonic probe.

143

Fig. 4.11 shows the FTIR spectra of carboxylated nanotubes, PEI, and PEI-tethered carboxylated
-1

-1

-1

-1

-1

-1

nanotubes. Peaks at ~3300 cm , 2900 cm , 2800 cm , 1640 cm , 1550 cm , 1460 cm , 1370
-1

-1

-1

cm , 1260 cm and 1060 cm point at the tethering of PEI upon nanotubes. The appearance of
-1

a new set of peaks upon tethering of PEI around 2900 cm

-1

and 2800 cm , attributed to alkane

(CH2, CH3 and CH) stretching, confirms the presence of polymer chains in association with
-1

nanotubes. Peaks at ~3300 cm

(NH stretching), 1550 cm

-1

-1

(NH2 scissoring) and 1060 cm
-1

(C-N stretching) are attributed to the amine groups of PEI. The peak at ~1640 cm is attributed
to carbonyl stretching of the newly formed amide bond between the amine groups of PEI and the
COOH groups of carboxylated nanotubes. This bonding mechanism is further confirmed by the
-1

medium peak at 1460 cm , which is attributed to amide N-H bending. In short, the FTIR spectra
indicate that PEI is covalently bonded to carboxylated nanotubes.
4.3.4

Microwave Irradiation

Carbon nanotubes (0.75 g) were added to 350 ml (milli-liter) of a 1:1 mixture of 70% solution of
nitric acid and 97% solution of sulphuric acid. The microwave power was set at 40% of a total
of 1100 W, and microwave irradiation was performed at atmospheric pressure. The reaction
vessel was subjected to 5 minutes of irradiation followed by a 1-minute break and another 5
minutes of microwave irradiation. For safety reasons, the microwave oven was placed in a fume
hood. The mixture was then diluted with DI water and filtered through 5 µm hydrophilic poly
(tetrafluoroethylene) (PTFE) filter paper under vacuum.

Filtered MWNTs were washed until

the washings were found to be neutral (using pH paper). The functionalized MWNTs were

144

allowed to dry in the filter paper.

Further drying was achieved under vacuum. The extent of

functionalization was evaluated using FTIR spectroscopy.

4.4 Experimental Results
4.4.1

Reinforcement Efficiency of Carbon Nanotubes with Different Functionalization
Conditions

The functionalization conditions of carbon nanotubes were established in a systematic manner.
First, the preferred dosages of different polymers introduced upon carbon nanotubes were
identified and characterized. For many polymers, a 1:1 weight ratio of physisorbed polymer to
carbon nanotube produced desired results. Cementitious nanocomposites were prepared and
characterized with 0.16 vol.% (volume fraction of dry cementitious materials) or 0.24 vol.%
functionalized carbon nanotubes. Flexure and Compression tests were carried out on samples in
order to evaluate the effects of different functionalization conditions. The main findings of this
experimental investigation are summarized below.
The refined functionalization process of carbon nanotubes is sensitive to both the
functionalization condition as well as the diameter of nanotubes. For example, PEI introduction
made positive contributions in the case of functionalized MWNT D (8-15 nm diameter) at 0.24
vol.%, but had negative effects on the flexural strength and energy absorption capacity when
applied to acid-functionalized MWNT E (20-40 nm diameter) (Figs. 4.12 & 4.13). Similar trends
were observed upon treatment of functionalized nanotubes with GA. Another observation is that
a rise in the dosage of the treatment agent proportional with increasing nanotube concentration
may not be optimal in the case of GA and PEI treatments.

145

Test results indicate that PAA is the preferred polymer for further functionalization of nanotubes
via polymer physisorption/tethering. PAA has produced consistent results with both
functionalized and non-functionalized MWNT D. Although some mixes with GA (0.16 vol.% of
acid-functionalized MWNT D) and PEI (0.16 vol.% of non-functionalized MWNT D) gave the
best results, these polymers produced inconsistent overall effects at different nanotube volume
fractions and for different pre-functionalization conditions.
PAA produced positive effects on both the flexural strength and energy absorption capacity of
cementitious nanocomposites at different volume fractions of MWNT D, and with different prefunctionalization conditions. As noted earlier, the exploratory word conducted in the course of
this investigation indicated that further improvements in PAA-functionalized nanotubes can be
realized by not increasing the PAA content in proportions with the nanotube volume fraction.
As was the case with MWNT D, with MWNT E also PAA emerged as the only polymer used for
further functionalization via physisorption/tethering which consistently benefited the
reinforcement efficiency of nanotubes. The increase in flexural strength was evident with
increasing volume fraction of MWNT E. However, the maximum deflection and energy
absorption capacity of cementitious nanocomposites with 0.24 vol.% of non-functionalized
MWNT E were reduced, and the flexural strength was unaffected. This could indicate that these
mixes can be further improved by not increasing the (functionalizing) polymer in proportion with
the nanotube volume fraction.
Cementitious nanocomposites with desired flexural attributes did not exhibit any significant loss
of compressive strength, while those with unfavorable flexural performance exhibited
measurable losses of compressive strength, Fig.s 4.12c and 4.13c.

146

A comparison of the GA, PAA and PEI treatment effects on the MWNT F contributions to
flexural performance at 0.16 vol.% (Fig. 4.14) indicated that only PAA treatment produced
promising results; both GA and PEI treatments caused a marked reduction of flexural strength
and energy absorption capacity of cementitious nanocomposites. A significant drop upon GA
treatment of MWNT F can be attributed to its longer length which could complicate its
dispersion under sonication.
Both MWNT F and G (functionalized through microwave irradiation) at 0.16 vol.% produced an
increase in flexural strength, maximum deflection and energy absorption capacity (Fig. 4.15).
This shows that both MWNT F and G act as effective reinforcement at this volume fraction,
which is partly due to their longer lengths (200 µm). MWNT F, due to its smaller diameter (8-15
nm), showed more promise as reinforcement in DSP cementitious paste. The gains in flexural
strength, maximum deflection and energy absorption capacity with 0.16 vol.% of MWNT F were
24%, 65% and 145%, respectively.
4.4.2

Scanning Electron Microscope Investigations

Observations of failed surfaces of flexure and compression test specimens using a high-precision
scanning electron microscope (JOEL JM-6300F) provided further insight into the mechanisms of
actions of nanotubes with different functionalization conditions in cementitious matrices. All
specimens which provided high flexural strength and energy absorption capacity showed
nanomaterials bridging across (and pulling out at) micro-cracks (Figs. 4.16a & b). Specimens
that provided lower flexural and compressive strengths, on the other hand, exhibited
compromised microstructures with reduced matrix density (Figs. 4.16c & d), which could be
attributed to the presence of excess amounts of functionalizing polymer.

147

4.4.3

X-Ray Diffraction Analysis (XRD)

XRD Analysis of selected specimens was carried out using Rotaflex Rigaku 200B Analyzer. The
spectrum was analyzed using JADE to identify various elements according to their crystalline
structure. Typical XRD spectrographs for DSP Pastes with 0% and 0.16 vol.% of MWNT D and
MWNT E (both acid-functionalized and carrying adsorbed GA) are presented in Fig. 4.17, noting
that higher compressive and flexural strength levels were obtained with MWNT D.

The

spectrum for MWNT D shows no significant difference when compared with that for plain
matrix. The two major compounds identified in these two samples were calcium silicate
(Ca3SiO5) and calcium silicate oxide (Ca3(SiO4)O). However, the mix containing MWNT E
shows peaks corresponding to calcium hydroxide which does not provide the high surface area
and binding qualities of calcium silicate hydrate. This adverse effect of functionalized MWNT E
on the matrix structure could be used to explain the relatively low flexural and compressive
strengths provided by these nanocomposites (see Figs. 4.12 and 4.13). In other words, depending
on the nanotube type and functionalization conditions, introduction of nano-scale reinforcement
with high specific surface area could interfere with the hydration/pozzolanic reactions of
cementitious matrix.
4.4.4

Thermogravimetric Analysis (TGA)

Thermogravimetric analyses of the cementitious matrix (DSP Paste) were performed using the
TA Instrument Hi-Res TGA 2950 Thermogravimetric Analyzer. A finely ground powder
(passing No. 100 mesh) of the material was used in thermogravimetric analysis. The samples
o

o

were heated at a constant rate of 10 C per minute up to 1000 C, with the weight loss recorded as

148

a function of temperature. Tests on selected mixes were conducted 15 days after casting of
cementitious specimens.
The TGA test results for the materials introduced in previous section (on XED) are shown in Fig.
4.18. The lower-strength cementitious nanocomposites with 0.16 vol% MWNT E (acidfunctionalized, with physisorbed Gum Arabic) showed excess weight loss with increasing
temperature. The loss of water is evident, which points at the presence of excess water that was
not chemically bound with cement. In the case of the higher-strength cementitious
nanocomposite reinforced with MWNT D (acid-functionalized, with physisorbed Gum Arabic),
the mass loss trends were similar to those of the plain cementitious matrix.

4.5 Summary
Experimental studies were conducted on the reinforcement efficiency of multiwalled carbon
nanotubes with different aspect ratios and surface modification conditions in high-performance
cementitious materials. The surface modification conditions considered involved (acid)
functionalization of carbon nanotubes and/or anchoring different polymers via chemical bonding
or physical adsorption. Experimental results indicated that a combination of the surface
modification specifics and the nanotube aspect ratio determines their reinforcement efficiency in
high-performance cementitious matrices. The most consistent favorable effects were produced
when poly(acrylic acid) (PAA) was physisorbed on nanotubes. Increased nanotube length and
aspect ratio also benefited their reinforcement efficiency in high-performance cementitious
matrices; excess nanotube lengths, however, complicate their dispersion in the mixing water of
cementitious matrix at required concentrations. The best balance of mechanical properties was
realized when longer multiwalled carbon nanotubes were used with physisorbed PAA, which
were not acid-functionalized prior to introduction of PAA. It seems that the benefits of acid149

functionalization are more than compensated for by their damaging effects involving shortening
of carbon nanotubes. The benefits realized by proper surface modification of nanotubes can be
explained by their contributions towards uniform dispersion in aqueous media, and improved
interfacial interactions with cement hydrates. The extent of surface modification as well as the
concentration of nanotubes in cementitious matrix should be optimized, depending on the
composition and processing conditions of cementitious nanocomposites, in order to avoid
agglomeration of nanotubes and their potential interferences with the hydration process of
cement.

150

Figure 4.1- Schematic depiction of the random and disordered adsorption of polymers upon
carbon nanotube walls.

Figure 4.2- FTIR spectra for acid-functionalized carbon nanotubes (CNT-COOH), Gum Arabic
(GA), and acid-functionalized nanotubes modified with Gum Arabic (GA-CNTCOOH).

151

Figure 4.3- FTIR spectra of acid-functionalized carbon nanotubes (CNT-COOH), PAA, and
acid-functionalized nanotubes modified with PAA (PAA-CNTCOOH).

Figure 4.4- FTIR spectra of acid-functionalize carbon nanotubes (CNT-COOH), Glenium®, and
acid-functionalized nanotubes modified with Glenium® (Glenium-CNTCOOH).

152

Figure 4.5- Schematic depiction of the tethering (covalent attachment) of polymer chains via
functional groups onto nanotube walls.

Figure 4.6- Covalent tethering of PAA onto carboxylated carbon nanotube.

Figure 4.7- Tethering of Glenium® onto carboxylated nanotube.

153

Figure 4.8- FTIR spectra of carboxylated nanotubes (CNT-COOH), PAA and PAA-tethered
carboxylated nanotubes (PAA-CNTCOOH).

Figure 4.9- FTIR spectra of carboxylated nanotubes (CNT-COOH), Glenium® and Glenium®tethered carboxylated nanotubes (Glenium-CNTCOOH).

154

Figure 4.10- Schematic depiction of covalent tethering of PEI to carboxylated nanotubes.

Figure 4.11- FTIR spectra of carboxylated nanotubes (CNT-COOH), PEI, and PEI-tethered
carboxylated nanotubes (PEI-CNTCOOH).

155

18

Plain
MWNT D,
fn
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI

16

Flexure Strength, MPa

14
12
10
8
6
4
2
0

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

Nanotube Type and Further Treatment
(a) Flexural strength
Figure 4.12- Mean values and standard errors of the flexural attributes and compressive strength of DSP paste reinforced with 0%,
0.16% and 0.24% volume fractions of acid-functionalized and non-functionalized MWNT D with different further treatments.

156

Figure 4.12 (cont’d)

Plain

Energy Absorption Capacity, N.mm

400
350
300
250
200
150
100
50
0
Nanotube Type and Further Treatment

(d) Energy absorption capacity

157

MWNT D,
fn
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

Figure 4.12 (cont’d)

Plain

140

Compressive Strength, MPa

120
100
80

60
40
20
0
Nanotube Type and Further Treatment

(c) Compressive strength

158

MWNT D,
fn
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,GA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PAA
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI
MWNT D,
fn,PEI

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

Plain

18
16

Flexure Strength, MPa

14
12
10
8
6
4
2
0
Nanotube Type and Further Treatment

MWNT E,
fn
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

(a) Flexural strength
Figure 4.13- Mean values and standard errors of the flexural attributes and compressive strength of cementitious (DSP) pastes
reinforced with 0%, 0.16% and 0.24% volume fractions of acid-functionalized and non-functionalized MWNT E with different further
treatments.

159

Figure 4.13 (cont’d)

Energy Absorption Capacity, N.mm

300

Plain

250

200

150

100

50

0

Nanotube Type and Further Treatment

(b) Energy absorption capacity

160

MWNT E,
fn
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

Figure 4.13 (cont’d)

Plain

160

Compressive Strength, MPa

140
120
100
80
60
40
20
0
Nanotube Type and Further Treatment

(c) Compressive strength

161

MWNT E,
fn
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,GA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PAA
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI
MWNT E,
fn,PEI

0.16 vol%, acid0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non0.16 vol%, acid0.24 vol%, acid0.16 vol%, non0.24 vol%, non-

12

Flexure Strength, MPa

10
8

Plain

6

MWNT F, 0.16 vol%,
non-fn, GA
MWNT F, 0.16 vol%,
non-fn,PAA

4

MWNT F, 0.16 vol%,
non-fn, PEI

2
0
Type of Further Treatment

(a) Flexural strength
Figure 4.14- Mean values and standard deviations of the flexural strength and energy absorption capacity test results for cementitious
(DSP) pastes reinforced with 0% and 0.16% volume fractions of non-functionalized MWNT F with different polymer treatments.

162

Figure 4.14 (cont’d)

Energy Absorption Capacity, N.mm

250

200
Plain
150

MWNT F, 0.16 vol%,
non-fn, GA
MWNT F, 0.16 vol%,
non-fn,PAA

100

MWNT F, 0.16 vol%,
non-fn, PEI
50

0
Type of Further Treatment

(b) Energy absorption capacity

163

14

Flexural Strength, MPa

12
Plain

10

8

MWNT F-non fn-0.16
vol%, Microwave

6
MWNT G-non fn0.16 vol%,
Microwave

4

2

0
Carbon Nanotubes Types
(a) Flexural strength
Figure 4.15- Flexural strength and energy absorption capacity of DSP cement pastes reinforced with 0% and 0.16% volume fractions
of microwave treated MWNTs F and G.

164

Figure 4.15 (cont’d)

160

Energy Absorption Capacity, N.mm

140
120

Plain

100
MWNT F-non fn-0.16
vol%, Microwave

80
60

MWNT G-non fn0.16 vol%,
Microwave

40
20
0
Carbon Nanotubes Types
(b) Energy Absorption Capacity

165

(a)

(b)
Figure 4.16- Scanning electron microscope images of the cementitious DSP paste with different
volume fractions of surface-modified carbon nanotubes.
166

Figure 4.16 (cont’d)

(c)

(d)

167

1600
Plain DSP Paste

1200
Intensity (CPS)

1400

MWNT D,Acid Fn,Gum
Arabic,0.16 vol.%

1000

MWNT E,Acid Fn, Gum
Arabic,0.16 vol.%

800
600
400
200
0
20

30

40
50
Two-Theta (deg)

60

70

Figure 4.17- XRD spectrographs of DSP pastes with 0% and 0.16% volume fractions of acid-functionalized MWNT D and MWNT E,
that are further modified with Gum Arabic.

168

Figure 4.18- Results of thermogravimetric analysis showing weight loss versus temperature.

169

CHAPTER 5

5

REINFORCING
EFFECTS
OF
NANOTUBES
AND
FIBERS
CEMENTITIOUS MATERIALS

MULTIWALLED
CARBON
IN
HIGH-PERFORMANCE

5.1 General
The effects of surface functionalization of multiwalled carbon nanotubes (MWNTs) through
introduction of carboxyl groups and use of refined dispersion techniques on the reinforcement
efficiency of nanotubes at different volume fractions within a high-performance cementitious
(DSP) matrix were evaluated.

Two types of surface-functionalization and intense

stirring/sonication techniques were employed for improving the reinforcement efficiency of
carbon nanotubes at volume fractions as high as 0.48 vol.% (volume percent of dry cementitious
materials). SEM observations of fractured surfaces pointed at successful dispersion of nanotubes
within the cement-based matrix. Cementitious nanocomposites with increased volume fractions
of functionalized nanotubes provided superior flexural strength and energy absorption capacity.
The effects of the degree of nanotube functionalization as well as the nanotube length and
diameter on their reinforcing effects in the high-performance cement-based matrix were also
investigated. An investigation of hybrid reinforcement systems comprising multiwalled carbon
nanotubes and micro-scale fibers indicated that selected hybrid systems offer a cost-effective
approach for realizing balanced gains in performance characteristics of high-performance
cementitious matrices.

5.2 Materials and Methods
170

5.2.1

Graphite Nanomaterials and Microfibers

Different types of acid-functionalized and non-functionalized multiwalled carbon nanotubes as
well as chopped (micro-scale) carbon fibers were evaluated for use in cement-based
nanocomposites. These graphite nanomaterials and microfibers have been introduced in detail in
Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. Multiwalled carbon nanotubes (MWNT D and
MWNT E) and carbon microfibers (CMF) were used in high-performance cement-based paste
and mortar matrices.
5.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
paste was used to evaluate the reinforcement efficiency of both graphite nanomaterials and
carbon microfibers at relatively low volume fractions.
Cementitious materials with and without functionalized and non-functionalized graphite
nanomaterials dispersed in the mixing water via sonication, described in detail in Chapter 2,
Section 2.5.2, and/or carbon microfibers were prepared and cured following ASTM procedures
described in the same section.
5.2.3

Experimental Methods

The test procedures employed to determine the engineering properties of high-performance
cement-based materials with and without nano- and/ or micro-scale reinforcements include
Compression tests (ASTM C 109), Flexure tests (ASTM C 1185), Impact tests (ASTM D 7136)
and Abrasion tests (ASTM C 944). The following experimental methods were also employed to
gain further insight into the structure and failure mechanisms of cement-based nanocomposites:

171

(i) SEM; (ii) EDS; (iii) XRD; and (iv) DT/TGA. The experimental results were subjected to the
ANOVA (Analysis of Variance) and PWC (Pair-Wise Comparison) statistical analyses.

5.3 Experimental Results and Discussion
5.3.1

Multiwalled Carbon Nanotube Aspect Ratio and Volume Fraction

5.3.1.1 Flexural Performance
The flexural strength of cementitious nanocomposites was higher with the longer/smallerdiameter (higher-aspect-ratio) MWNT D when compared with MWNT E. Non-functionalized
and acid-functionalized MWNT D generated (at 0.24 vol.%) 72% and 43% gains in flexural
strength, and 264% and 161% gains in flexural energy absorption capacity, respectively. In the
case of MWNT D (Figure 5.1 and Table 5.1), the adverse shortening effects of acidfunctionalization overshadowed the contributions associated with the improved interfacial
interactions in the presence of carboxyl functional groups. Non-functionalized and acidfunctionalized MWNT E (at 0.24 vol.%) generated 16% and 20% gains in flexural strength, and
67% and 95% gains in flexural energy absorption capacity, respectively (see Figure 5.1 and
Table 5.1). In the case of MWNT E, the benefits of carboxyl functional groups to interfacial
interactions (upon acid-functionalization) more than compensated for the shortening effects of
acid-functionalization. It should be noted that the acid-functionalized MWNT E had a lower
degree of functionalization (1.2% COOH groups) when compared with functionalized MWNT D
(2.5% COOH groups). This difference (which is caused by the smaller diameter of MWNT D),
together with the observation that the maximum lengths of both nanotubes were reduced to about
half upon functionalization, can be used to explain the overall negative and positive effects of
acid-functionalization on MWNT D and MWNT E, respectively. In spite of these differences,

172

one can conclude that longer and smaller-diameter (i.e., higher-aspect-ratio) multi-walled carbon
nanotubes offer (as predicted theoretically) higher reinforcement efficiencies in cement-based
matrices. Another theoretical prediction verified experimentally is that the flexural energy
absorption capacity of cement-based matrices benefit more than their flexural strength from
introduction of multi-walled carbon nanotubes. The contributions of carbon nanotubes to the
flexural energy absorption capacity and strength of cementitious materials were confirmed
through statistical analysis (ANOVA) of test results at significance levels of 0.001 and 0.050,
respectively. At 0.48 vol.% nanotube volume fraction, drops in flexural strength were observed
(see Figure 5.1 and Table 5.1), with gains in energy absorption capacity experienced in most
conditions. For the cementitious nanocomposite materials and processing conditions considered
here, a nanotube volume fraction of 0.48 vol.% seems to be excessive; the sonication and mixing
procedures employed here may not be capable of effectively dispersing the nanotubes at this
volume fraction. SEM observations of cementitious nanocomposites with 0.48 vol.% nanotube
volume fraction confirmed excess clumping of nanotubes (Figure 5.2). The adverse effects of
0.48% nanotube volume fraction could also result from the excess superplasticizer dosage
required to maintain the fresh mix workability of this cementitious material; at such high dosage,
the superplasticizer molecules covering cement particles (and later cement hydrates) could
mitigate their effective interactions with water for thorough hydration of cement. More extensive
steam curing or extended moist curing periods could help overcome this obstacle and advance
the hydration process of cement. Statistical analysis (of variance) of test results indicated that the
(adverse) effects of 0.48 vol.% MWNT D and MWNT E on flexural strength were not
statistically significant (at 0.05 significance level); the corresponding (positive) effects of
nanotubes on flexural energy absorption, however, were statistically significant.

173

5.3.1.2 Compressive Strength
The compressive strength test data (means and standard deviations) are summarized in Table 5.2.
The increase in carbon nanotube volume fraction occasionally resulted in a slight drop in
compressive strength (especially for non-functionalized nanotubes). This slight loss of
compressive strength with introduction of nanotubes (which is also observed with micro-scale
fibers) generally results from the loss of fresh mix workability and thus the reduced compaction
of matrix in the presence of discrete (nano- or micro-scale) reinforcement systems. This effect is
highly dependent on the specifics of mix composition and processing conditions. Another factor
influencing this trend can be the increased amount of superplasticizer used for (partially)
restoring the fresh mix workability at higher nanotube volume fractions. The excess
superplasticizer dosage could coat cement particles and their hydrates, delaying the hydration
process. In order to study this trend, control specimens with higher amounts of superplasticizer
were prepared and tested. The results showed that higher amounts of superplasticizer produced
losses of compressive strength (without adding nanotubes). With the high superplasticizer
dosage, mixes with and without nanotubes provided statistically comparable (at 0.05 significance
level) compressive strengths.
5.3.1.3 Impact Resistance
The impact resistance of the high-performance (DSP) cementitious matrix increased with
introduction of 0.24 vol.% non-functionalized or acid-functionalized MWNT D. However, nonfunctionalized MWNT E produced a drop in impact resistance (see Table 5.3). The two types of
functionalized nanotubes considered here (MWNT D & E), at 0.24 vol.%, produced about 70%
gain in impact resistance. The impact resistance of the cementitious matrix increased further with
increasing the nanotube volume fraction to 0.48 vol.% for both non-functionalized and acid174

functionalized MWNT D. Test results indicated that functionalization of carbon nanotubes
benefits their contributions to the impact resistance of the high-performance cementitious matrix.
The two types of functionalized nanotubes considered here (MWNT D & E), at 0.48 vol.%,
produced about 75% and 58% gains, respectively, in impact resistance, which can be attributed
to arrest and deflection of microcracks by nanotubes, frictional pullout of nanotubes, and the
multiple cracking induced by nanotubes. Statistical analysis (of variance) of impact test results
indicated that the impact resistance of the DSP cementitious matrix experienced statistically
significant gains (at 0.027 significance level) with increasing nanotube volume fractions. Acidfunctionalized nanotubes provided somewhat better reinforcement efficiency (0.010 significance
level) as compared to their non-functionalized counterparts, which could be attributed to
improved interfacial interactions of functionalized nanotubes with the cementitious matrix. In
general, MWNT D produced greater improvements in impact resistance when compared with
MWNT E (which was generally the case with different engineering properties).
5.3.1.4 Abrasion Resistance
The control specimen without nanotube experienced a mean weight loss of 0.28 grams upon
abrasion, compared to only 0.1 gram or less weight loss for the cementitious material with 0.24
vol.% acid-functionalized MWNT D & E. The contributions of functionalized nanotubes to
abrasion resistance may be attributed to the corresponding gains in fracture toughness and the
control of surface (restrained shrinkage) microcracking by nanotubes. However, mixes with 0.48
vol.% acid-functionalized and non-functionalized MWNT D & E provided less abrasion
resistance than those with 0.24 vol.% nanotube, as shown in Table 5.4. As noted earlier, for the
mix designs and processing conditions considered here, 0.48 vol.% seems to exceed the
maximum limit for effective dispersion of nanotubes. Clumping of nanotubes at this excess

175

volume fraction compromises their reinforcement efficiency in the cementitious matrix. The
relatively high dosage of superplasticizer required for restoring the fresh mix workability in the
presence of high nanotube volume fractions further compromises the quality of these
cementitious nanocomposites. Statistical analysis (of variance) of abrasion test results pointed at
a statistically significant improvement in abrasion resistance (at 0.05 significance level) with
0.24 vol.% carbon nanotube. The gains in abrasion resistance would, however, be statistically
insignificant at the higher nanotube volume fraction of 0.48 vol.%).
5.3.1.5 Scanning Electron Microscope Evaluation
The SEM images presented in Figure 5.3 indicate that the refined sonication and mixing steps
developed in the project were successful in dispersion of individual nanotubes within the
cementitious matrix at 0.24 vol.% volume fraction. These images also point at the low porosity
of the high-performance cement-based matrix; the introduction of nanotubes does not seem to
have compromised the high density (low porosity) of this matrix. SEM observations provided
ample evidence for uniform dispersion of both types of nanotubes; evidences of crack
suppression and crack bridging were also found in all specimens examined under SEM.
Functionalized nanotubes, which provided higher levels of reinforcement efficiency, were found
to produce a more uniform dispersion within the cementitious matrix.
5.3.1.6 Thermogravimetric Analysis (TGA)
Results of thermogravimetric analysis for DSP pastes with different volume fractions of MWNT
D and E (Figure 5.4) point at some improvements in the matrix structure resulting from the
addition of nanotubes. Figure 5.4 suggests that the rates of mass loss with increasing temperature
are less steep for cementitious nanocomposites when compared with the plain matrix.

176

5.3.1.7 Energy Dispersive Spectroscopy (EDS)
Elemental analyses were performed on 2.54x12x12 mm samples of DSP Paste with 0 and 0.24
vol.% non-functionalized and acid-functionalized MWNT D & E. Elemental and Quantitative
analysis of these samples indicated that all cementitious materials exhibit similar elemental
compositions, except for the (expected) higher carbon content in materials incorporating carbon
nanotubes. This further confirms the findings of thermogravimetry analyses, indicating that
graphite nanomaterials do not significantly alter the chemistry of cement hydration for the
reinforcement conditions considered here.
5.3.2

Effects of Hybrid (nano- and micro-scale) Reinforcement Systems

Hybrid reinforcement systems comprising multiwalled carbon nanotubes and carbon fibers with
micro-scale diameter were evaluated in DSP paste. Exploratory studies with MWNT D or E
multiwalled carbon nanotube reinforcement and carbon fiber indicated that, for the cementitious
matrix and processing conditions considered here, the nanotube volume fraction should be
lowered below 0.24 vol.% before benefits can be realized from introduction of carbon fibers.
Hence, a nanotube volume fraction of 0.16 vol.% was considered for development of hybrid
reinforcement systems. The nanotubes considered here were either non-functionalized MWNT D
or PAA-physisorbed MWNT D.
5.3.2.1 Flexural Performance
The flexural strength and energy absorption capacity test results for the high-performance (DSP)
cementitious paste with different reinforcement systems are summarized in Table 5.5. When
carbon nanotube and/or microfiber reinforcement were used in DSP paste, the highest values of
flexural strength, maximum deflection and energy absorption capacity were obtained with hybrid

177

reinforcement. The improvements brought about by hybrid reinforcement in flexural strength,
maximum deflection and energy absorption capacity were 61%, 260% and 365%, respectively.
These results point at the synergy of the reinforcement mechanisms at nano and micro scales in a
high-performance cementitious matrix. Statistical analysis (of variance) of test results confirmed
the statistical significance of the gains in flexural performance upon hybrid reinforcement of the
DSP cementitious paste (at significance level of 0.001).
5.3.2.2 Compressive Strength
While there are minor gains in compressive strength with micro-scale and hybrid reinforcement
systems (Table 5.6), the hybrid reinforcement effects on compressive strength were found to be
statistically insignificant (at 0.05 significance level). Past experience with micro-scale fibers
indicates that fibers generally lower the compressive strength of cementitious composites by
lowering the fresh mix workability and introducing fiber clumps which act as defects within
matrix. Hence, the fact that the compressive strength of the matrix is retained with nano-scale
and hybrid reinforcement systems points at the benefits of nano-scale reinforcement to the
structure of cementitious matrix.
5.3.2.3 Impact Resistance
The improvements in impact resistance of DSP paste upon hybrid reinforcement are summarized
in Table 5.7. Statistical analysis of the impact test results indicated that nano- and/or micro-scale
reinforcement of DSP paste led to statistically significant gains in impact resistance (0.038
significance level).

178

5.3.2.4 Abrasion Resistance
The mean values of abrasion weight loss are presented in Table 5.8. The hybrid reinforcement
system reduces the abrasion weight loss of DSP cementitious paste by 60%, which was found to
be statistically significant (at 0.000 significance level).
5.3.2.5 Scanning Electron Microscope Evaluation
Figure 5.5 shows the SEM images indicating the uniform distribution of nanotubes and the
failure mechanism of carbon fibers (involving rupture within the paste and pullout at fracture
surfaces. One may observe in Figure 5.5 the individually dispersed multiwalled carbon
nanotubes within the paste matrix at the fracture surface. The investigation, Figure 5.5b, also
reveals the pullout behavior of carbon microfibers which adds to the flexural strength of the
matrix.
5.3.2.6 Thermogravimetric Analysis (TGA)
Thermogravimetric analysis of DSP paste with microfiber and hybrid (nanotube/microfiber)
reinforcement systems as well as that for a conventional cementitious mortar (Figure 5.6)
indicated that:


o

o

The mass loss between 100 C to 400 C was considered in order to assess, with a certain
degree of approximation, the chemically bound water in calcium silicate hydrate (C-S-H).
This approximation helps avoid the errors caused by the possible presence of CaCO3 and
o

the mass loss associated with its decomposition at higher temperatures (around 800 C),
o

and by the presence of free water below 100 C. The maximum weight loss of 9.65%
occurs for DSP Paste containing hybrid reinforcement, followed by 7.54% for that
179

incorporating carbon microfiber as compared with 6.87% for plain DSP paste. In
conventional mortar, the total weight loss was 3.93%. These results indicate that more
water is chemically bound with C-S-H in the high-performance cementitious material
(DSP paste) when compared with conventional mortar.


Total weight loss is greater for the high-performance cementitious material (DSP paste),
at 20.2% with hybrid reinforcement, 18.5% with carbon microfiber and 17.9% for plain
DSP matrix, versus 10.2% for the conventional mortar. This reflects the higher
concentration of hydration products in DSP paste when compared with conventional
mortar.

5.4 Summary
The experimental results generated for cementitious pastes reinforced with different nonfunctionalized and functionalized multiwalled carbon nanotubes and also their hybrids with
carbon microfibers yield some conclusions that the length, diameter and surface functionalization
(as well as mechanical attributes) of carbon nanotubes all have critical effects on their
reinforcement efficiency in cement-based matrices. Functionalization of nanotubes could
introduce new defects or reduce the length of nanotubes. These adverse effects of
functionalization may overshadow its positive contributions to interfacial bonding and thus the
reinforcement efficiency of nanotubes. Use of polymer-wrapped (PAA physisorbed) multiwalled
carbon nanotubes addresses this issue by enhancing the dispersion/interfacial interactions of
nanotubes within cementitious matrix without introducing defects or shortening the nanotubes.

180

Table 5.1- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DPS paste matrix reinforced with 0%, 0.24% and 0.48% volume fractions of nonfunctionalized and acid-functionalized MWNT D & E.
Reinforcement Condition
Plain
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%
MWNT
vol.%

D, acid-fn, 0.24

Mean Flexure
Strength (MPa)
10.29
14.73

Deflection
(mm)
0.40
0.69

Energy
Absorption(N.mm)
61.84
161.45

D, non-fn, 0.24

15.81

0.78

225.71

E, acid-fn, 0.24

12.36

0.54

120.79

E, non-fn, 0.24

11.90

0.44

103.53

D, acid-fn, 0.48

12.14

0.75

172.99

D, non-fn, 0.48

12.78

0.85

237.32

E, acid-fn, 0.48

9.54

0.60

138.89

E, non-fn, 0.48

8.68

0.44

95.34

Table 5.2- Mean values of compressive strength test results for DSP paste matrix reinforced with
0%, 0.24% and 0.48% volume fractions of non-functionalized and acid-functionalized carbon
nanotubes.
Reinforcement Condition

Mean Compressive
Strength (MPa)

Standard Deviation
(MPa)

Plain

110.92

14.27

MWNT D, acid-fn, 0.24 vol.%

107.96

19.26

MWNT D, non-fn, 0.24 vol.%

111.65

8.48

MWNT E, acid-fn, 0.24 vol.%

103.63

11.69

MWNT E, non-fn, 0.24 vol.%

112.56

16.92

MWNT D, acid-fn, 0.48 vol.%

112.89

19.03

MWNT D, non-fn, 0.48 vol.%

100.14

9.51

MWNT E, acid-fn, 0.48 vol.%

111.39

4.48

MWNT E, non-fn, 0.48 vol.%

100.37

6.62

181

Table 5.3- Mean values of impact resistance test results for DSP paste reinforced with 0%,
0.24% and 0.48% volume fractions of non-functionalized and acid-functionalized carbon
nanotubes MWNT D & E.
Reinforcement Condition

Impact Resistance, θ (mm/mm)

Plain

0.36

MWNT D, acid-fn, 0.24 vol.%

0.57

MWNT D, non-fn, 0.24 vol.%

0.42

MWNT E, acid-fn, 0.24 vol.%

0.52

MWNT E, non-fn, 0.24 vol.%

0.26

MWNT D, acid-fn, 0.48 vol.%

0.63

MWNT D, non-fn, 0.48 vol.%

0.47

MWNT E, acid-fn, 0.48 vol.%

0.57

MWNT E, non-fn, 0.48 vol.%

0.36

Table 5.4- Mean weight loss test results after abrasion for DSP paste reinforced with 0%, 0.24%
and 0.48% volume fractions of non-functionalized and acid-functionalized carbon nanotubes
MWNT D & E.
Reinforcement Condition

Mean Weight Loss
(grams)

Plain

0.28

MWNT D, acid-fn, 0.24 vol.%

0.10

MWNT D, non-fn, 0.24 vol.%

0.15

MWNT E, acid-fn, 0.24 vol.%

0.10

MWNT E, non-fn, 0.24 vol.%

0.25

MWNT D, acid-fn, 0.48 vol.%

0.18

MWNT D, non-fn, 0.48 vol.%

0.23

MWNT E, acid-fn, 0.48 vol.%

0.20

MWNT E, non-fn, 0.48 vol.%

0.28

182

Table 5.5- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DSP paste reinforced different vol. fractions of micro- and/or nano-scale reinforcement.
Reinforcement Condition

Flexural
Strength
(MPa)

Maximum
Deflection (mm)

Energy
Absorption(N.mm)

Plain

10.29

0.40

61.84

CMF (TT 143), 0.24 vol.%

16.06

1.13

251.12

MWNT D (PAA-fn), 0.16
vol.%

13.90

0.49

105.62

CMF (TT 143), 0.24 vol.%
and MWNT D (PAA-fn),
0.16 vol.%

16.64

1.46

287.98

Table 5.6- Compressive strength test results for the DSP paste matrix reinforced with different
volume fractions of nano and micro-scale reinforcements.
Reinforcement Condition

Mean Compressive
Strength (MPa)

Standard
Deviation (MPa)

Plain

110.92

14.27

CMF (TT 143), 0.24 vol.%

116.68

18.01

MWNT D (PAA-fn), 0.16 vol.%

109.64

9.43

CMF (TT 143), 0.24 vol.% and
MWNT D (PAA-fn), 0.16 vol.%

122.76

12.12

Table 5.7- Impact resistance test results of DSP pastes with nano- and/or micro-scale
reinforcement.
Reinforcement Condition

Impact Resistance
(mm/mm)

Plain

0.365

CMF (TT 143), 0.24 vol.%

0.521

MWNT D (PAA-fn), 0.16 vol.%

0.625

CMF (TT 143), 0.24 vol.% and MWNT D (PAA-fn), 0.16 vol.%

0.729

183

Table 5.8- Abrasion weight losses of DSP pastes with nano- and/or micro-scale reinforcement.
Reinforcement Condition

Loss of Mass
(grams)

Plain

0.275

CMF (TT 143), 0.24 vol.%

0.150

MWNT D (PAA-fn), 0.16 vol.%

0.110

CMF (TT 143), 0.24 vol.% and MWNT D (PAA-fn), 0.16 vol.%

0.125

184

20
Plain (48.75ml Glenium
7700)

18

14

MWNT D-acid fn-0.24
vol%

12

MWNT E-non fn-0.24
vol%

10

MWNT E-acid fn-0.24
vol%

8

MWNT D-non fn-0.48
vol%

6

MWNT D-acid fn-0.48
vol%

4

Flexural Strength, MPa

16

MWNT D-non fn-0.24
vol%

MWNT E-non fn-0.48
vo%

2

MWNT E-acid fn-0.48
vol%

0
Carbon Nanotubes Type
(a) Flexural strength
Figure 5.1- Flexural strength and energy absorption capacity of DSP cement pastes with 0.24 and 0.48 vol.% non-functionalized and
acid-functionalized MWNT D & E.

185

Figure 5.1 (cont’d)

350
Plain (48.75ml Glenium
7700)

Energy Absorption Capacity, N.mm

300

MWNT D-non fn-0.24
vol%

250

MWNT D-acid fn-0.24
vol%

200

MWNT E-non fn-0.24
vol%
MWNT E-acid fn-0.24
vol%

150

MWNT D-non fn-0.48
vol%
MWNT D-acid fn-0.48
vol%

100

MWNT E-non fn-0.48
vol%

50

MWNT E-acid fn-0.48
vol%

0
Carbon Nanotubes Type
(b) Energy absorption capacity

186

Figure 5.2- Excessive clumping of carbon nanotubes in DSP paste at 0.48% volume fraction.

(a) Uniform distribution of nanotubes in the matrix
Figure 5.3- Typical SEM images of DSP paste.
187

Figure 5.3 (cont’d)

(b) Micro- crack-bridging

(c) Micro- crack-bridging
188

Figure 5.3 (cont’d)

(d) Dense DSP matrix

189

Figure 5.4- The results of thermogravimetric analysis of DSP cement paste with 0% and 0.24%
volume fractions of functionalized multi-walled carbon nanotubes MWNT D & E.

190

(a)

(b)
Figure 5.5- Typical SEM mages of DSP paste reinforced with 0.16 vol.% of MWNT D-PAA and
0.24 vol.% of CMF: (a) dispersed MWNTs; and (b) failure mechanism of micro-scale fibers.
191

0
90

0

0

80

60

70

0

0
50

0
40

0
30

0
20

0
10

0
0
-0.002

Loss of Weight (mg)

-0.004
-0.006
-0.008
Conventional Mortar
-0.01
CMF(TT143),0.24 vol% & MWNT D
(PAA-fn), 0.16 vol%

-0.012

CMF(TT 143), 0.24 vol%
-0.014
DSP Paste Plain
-0.016
-0.018

Temperature(C)
Figure 5.6- Results of thermogravimetric analysis (weight loss vs. temperature) for the high-performance cementitious matrix (DSP
paste) with different volume fractions of micro and nano reinforcements.

192

CHAPTER 6

6

REINFORCING
EFFECTS
OF
NANOTUBES AT DIFFERENT
CEMENTITIOUS PASTES

MULTIWALLED
CARBON
VOLUME FRACTIONS IN

6.1 General
Multiwalled carbon nanotubes would realize their reinforcement potential within cement-based
materials as far as they are thoroughly dispersed within cementitious matrix, and effectively
bonded to cement hydrates. Superplasticizers, which are commonly used in high-performance
cementitious materials, can facilitate the dispersion of nanotubes. Carboxylate-based
superplasticizers have assumed a prominent position among concrete admixtures; various
proprietary formulations of these superplasticizers are available commercially. Two carboxylatebased superplasticizers were evaluated in the experimental work reported here. Multiwalled
carbon nanotubes were introduced into cementitious pastes at different volume fractions. A highperformance cementitious paste (densified with small particles – DSP) was considered in this
investigation.
It is challenging to disperse nanotubes in water at relatively high concentrations. Nanotubes were
thus dispersed in the whole mixing water of cementitious paste, which can complicate large-scale
production of cementitious nanocomposites. This investigation evaluated the potential for
increasing the concentration nanotube dispersions in water. Cementitious nanocomposites were
thus prepared with nanotubes dispersed in different fractions of mixing water, and the effects on
performance characteristics of cementitious nanocomposites were evaluated. The minimum
193

fraction of mixing water required for effective dispersion of nanotubes was identified through
this experimental program.

6.2 Materials and Methods
6.2.1

Graphite Nanomaterials

Multiwalled carbon nanotubes (MWNT D) were used in non-functionalized, carboxylic acidfunctionalized, and polyacrylic acid-physisorbed conditions. MWNT D has been introduced in
detail in Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3 and was used in high-performance
cement-based paste matrix.
6.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
(Densified with Small Particles) paste was used to evaluate the reinforcement efficiency of
MWNT D at different volume fractions.
Cementitious materials (with and without functionalized and non-functionalized multiwalled
carbon nanotubes dispersed in the mixing water via sonication) were prepared and cured
following ASTM procedures described in the same section.
6.2.3

Experimental Methods

Compression tests (ASTM C 109) and flexure tests (ASTM C 1185) were performed as well as
SEM was also employed to gain further insight into the structure and failure mechanisms of
cement-based nanocomposites. The experimental results were subjected to the ANOVA and
PWC statistical analyses.

194

6.3 Experimental Results and Discussion
6.3.1

Effects of MWNT D Volume Fraction and Superplasticizer Type

6.3.1.1 Flexural Performance
The flexure test data are summarized in Table 6.1. Both functionalized (MWNT D FN) and nonfunctionalized MWNT D (MWNT D NF) produced more pronounced gains in flexural attributes
when used with the ADVA® Cast 575 superplasticizer. ADVA® Cast 575 enabled better
dispersion of non-functionalized and functionalized nanotubes, and seemed to enhance their
interactions with the cementitious matrix when compared with the Glenium® 7700
superplasticizer. The improvements in flexural attributes were more pronounced with nonfunctionalized nanotubes (due to improved dispersion of nanotubes with the ADVA® Cast 575
superplasticizer). ADVA® Cast 575, when compared with Glenium® 7700, was found to be
more effective in improving all flexural attributes as well as the compressive strength of the
high-performance cementitious paste with either MWNT D FN or MWNT D NF at 0.24 and 0.48
vol.%. The gains in flexural strength, maximum deflection and energy absorption capacity of
cementitious nanocomposites incorporating MWNT D FN at 0.24 vol.% (0.48 vol.%) prepared
using ADVA® Cast 575 were 9.23% (59.5%), 127% (137%) and 25.3% (116%), respectively.
The corresponding gains at 0.24 vol.% (0.48 vol.%) for MWNT D NF were 28.5% (60.9%),
128% (146%) and 62.6% (73%), respectively.
Outcomes of statistical analysis of test results, using the analysis of variance and pair-wise
comparison techniques, verified the improved performance of cementitious nanocomposites with
different volume fractions of functionalized or non-functionalized nanotubes realized with
ADVA® Cast 575 versus Glenium® 7700 superplasticizer. Pair-wise comparisons of flexure test

195

data indicated that the gains in flexural strength with 0.24 vol.% acid-functionalized nanotubes
were not statistically significant (0.128 significance level) with the ADVA® Cast 575
superplasticizer. At 0.48 vol.%, however, acid-functionalized nanotubes produced statistically
significant gains (at 0.000 significance level) in the flexural strength of high-performance
cementitious paste. It should be noted that statistical significance levels below 0.05 are generally
considered as indicative of statistically significant effects. In the case of flexural energy
absorption capacity and maximum deflection, the effects of acid-functionalized nanotubes were
found to be statistically significant at both 0.24 and 0.48 vol.% (significance levels ≤0.03) with
the ADVA® Cast 575 superplasticizer. For non-functionalized nanotubes, the gains in flexural
strength, energy absorption capacity and maximum deflection were all statistically significant (at
0.000 significance level). Pair-wise comparisons of experimental results indicated that the
distinctions between the flexural attributes of cementitious nanocomposites incorporating 0.24
vol.% versus 0.48 vol.% nanotubes were not statistically significant.
6.3.1.2 Compressive Strength
The compressive strength test results for high-performance pastes with both acid-functionalized
and non-functionalized MWNT D, at 0.24 vol.% and 0.48 vol.%, prepared using the two
polycarboxylate-based superplasticizers are summarized in Table 6.2. When ADVA® Cast 575
superplasticizer was used, the addition of non-functionalized and functionalized carbon
nanotubes at 0.24 vol.% and especially 0.48 vol.% produced some increase in compressive
strength. The superplasticizer type greatly influenced the effects of nanotubes on compressive
strength. Functionalized MWNT D, at 0.24 vol.% (0.48 vol.%), prepared using the ADVA® Cast
575 superplasticizer increased the compressive strength by 4.62% (11.5%) when compared with
similar nanocomposites prepared with the Glenium® 7700 superplasticizer; these gains were

196

found to be statistically significant at 0.005 (0.000) significance level. The corresponding
improvements in the case of non-functionalized nanotube were 8.88% (31.0%) with 0.087
(0.000) significance levels.
6.3.1.3 Scanning Electron Microscope Evaluation
Fig. 6.1 shows SEM images of cementitious nanocomposites with MWNT D prepared with the
ADVA® Cast 575 superplasticizer. Multiwalled carbon nanotubes were uniformly distributed
within the dense matrix. This uniform dispersion of the reinforcement system is essential for
realizing the observed gains in the engineering properties of cementitious matrices. Figure 6.6c
indicates that carbon nanotubes were pulled out of the matrix during the fracture process. This
pull-out behavior benefits the ductility and energy absorption capacity of cementitious materials.
6.3.2

Effect of Dispersing Multiwalled Carbon Nanotubes in Different Fractions of
Mixing Water

In order to facilitate industrial-scale production of cementitious nanocomposites, one needs to
minimize the fraction of mixing water used for dispersion of the required quantities of
nanomaterials. Comparative studies were conducted involving dispersion of nanomaterials in
different fractions of the mixing water followed by preparation of cementitious nanocomposites.
Polymer-wrapped

multiwalled

carbon

nanotubes

(CNTD-FN)

and

non-functionalized

multiwalled carbon nanotubes (CNTD-NF) were used as reinforcement in the high-performance
(DSP) cementitious paste at 0.24 vol.%. Multiwalled carbon nanotubes were dispersed,
following the procedures described earlier, in 30%, 20% and 10% of the total mixing water of
the cementitious paste. All mixes were prepared using the ADVA® Cast-575 superplasticizer.

197

6.3.2.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity test results for the
high-performance (DSP) cementitious paste reinforced with 0.24 vol.% of CNTD-NF and
CNTD-FN dispersed in different fractions of mixing water are shown in Table 6.3. These results
indicate that the maximum improvements in flexural performance were realized by dispersing
CNTD-FN in the full quantity of mixing water; the corresponding improvements in flexural
strength, energy absorption capacity and maximum deflection were 41.0%, 249% and 160%,
respectively. The test data also suggest that reducing the fraction of mixing water used for
dispersion of both functionalized and non-functionalized nanotubes reduced the reinforcement
efficiency of nanotubes. This trend could be attributed to the reduced thoroughness of nanotube
dispersion resulting from the use of smaller quantities of water during dispersion. The least gains
in the DSP paste flexural performance with nanotube reinforcement were observed when only
10% of mixing water was used for dispersion of CNTD-NF; the corresponding improvements in
flexural strength, energy absorption capacity and maximum deflection were 3.6%, 136% and
20%, respectively. Desired gains in flexural qualities were realized when 30% of the mixing
water was used for dispersion of both CNTD-FN and CNTD-NF.
Analysis of variance followed by pair-wise comparison of test results indicated that reducing the
fraction of mixing water used for dispersion of nanotubes from 100% to 30% did not produce
statistically significant drops in the reinforcement efficiency of the CNTD-FN and CNTD-NF
multiwalled carbon nanotubes. A statistically significant effect was, however, observed for both
flexural strength and energy absorption capacity when the fraction of mixing water used for
dispersion of nanotubes was reduced to 20% or 10% for CNTD-NF. In the case of CNTD-FN, a
statistically significant effect was observed only for flexural strength when 10% of the mixing

198

water was used for dispersion of nanotubes. This implies that, with CNTD-FN, the fraction of
mixing water required for dispersion of nanotube can be reduced to 20%.
6.3.2.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
(DSP) cementitious pastes with 0.24 vol.% of CNTD-FN and CNTD-NF reinforcement
dispersed in different fractions of mixing water are presented in Table 6.4. Analysis of variance
of the compression test results followed by pair-wise comparisons indicated that 0.24 vol.% of
CNTD-FN or CNTD-NF, dispersed in different fractions of mixing water, did not produce a
statistically significant drop in compressive strength (at 0.05 significance level). While with
CNTD-NF the mean value of compressive strength dropped with decreasing fractions of mixing
water used for dispersion of nanotubes, in the case of CNTD-FN, compressive strength did
remained practically unchanged as decreasing fractions of the mixing water used for dispersion
of multiwalled carbon nanotubes.

6.4 Summary
The reinforcement efficiency of functionalized, non-functionalized and polymer-wrapped
multiwalled carbon nanotubes were evaluated in a high-performance cementitious paste. Two
different polycarboxylate-based superplasticizers were used to enhance dispersion of nanotubes
and workability of fresh cementitious nanocomposites. The results showed that different
polycarboxylate-based superplasticizers produce cementitious nanocomposites with significant
differences in performance characteristics. Scanning electron microscope observations of
fractured surfaces of cementitious nanocomposites indicated that multiwalled carbon nanotubes
were uniformly distributed within the dense matrix. This uniform dispersion of the reinforcement

199

system is essential for realizing the observed gains in the engineering properties of cementitious
matrices. Some carbon nanotubes were observed to have pulled out of the cementitious matrix
during the fracture process. This pull-out behavior benefits the ductility and energy absorption
capacity of cementitious materials.
Polymer-wrapped as well as non-functionalized multiwalled carbon nanotubes were added as
reinforcement to the high-performance cementitious paste after dispersion in reduced quantities
of water. While the reinforcing effects of nanotubes decreased with decreasing fraction of the
mixing water used for their dispersion, reasonable results were obtained when nanotubes were
dispersed in 30% of the total mixing water, which benefits scaled-up production of cementitious
nanocomposites. For polymer wrapped nanotubes, the quantity of mixing water could be reduced
to 20% without having a statistically significant drop in the various engineering properties
evaluated.

200

Table 6.1- Mean values of the flexural properties of DSP pastes with functionalized and nonfunctionalized MWNT D at 0.24 vol.% and 0.48vol.%, prepared using different carboxylic-based
superplasticizers.
Reinforcement Condition

Mean Flexure
Strength
(MPa)
10.3
13.9
14.7

0.40
0.71
0.69

Energy
Absorption
(N.mm)
61.8
155
162

15.8

0.86

226

16.1

1.57

202

20.3

1.96

367

12.1

0.75

173

12.8

0.85

237

19.3

1.78

373

20.6

Plain, Glenium 7700
Plain, ADVA CAST 575
MWNT D, acid-fn, 0.24 vol%, Glenium
7700
MWNT D, non-fn, 0.24 vol%, Glenium
7700
MWNT D, acid-fn, 0.24 vol%, ADVA
CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA
CAST 575
MWNT D, acid-fn, 0.48 vol%, Glenium
7700
MWNT D, non-fn, 0.48 vol%, Glenium
7700
MWNT D, acid-fn, 0.48 vol%, ADVA
CAST 575
MWNT D, non-fn, 0.48 vol%, ADVA
CAST 575

Deflection
(mm)

2.09

411

Table 6.2- Mean values and standard errors of the compressive strength test results for DSP
pastes with functionalized and non-functionalized MWNT D at 0.24 vol.% and 0.48 vol.%,
prepared using different poly carboxylate-based superplasticizers.
Reinforcement Condition

Standard
Error (MPa)

Plain, Glenium 7700
Plain, ADVA CAST 575
MWNT D, acid-fn, 0.24 vol%, Glenium 7700

Mean
Compressive
Strength (MPa)
111
120
108

MWNT D, non-fn, 0.24 vol%, Glenium 7700

112

8.48

MWNT D, acid-fn, 0.24 vol%, ADVA CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575
MWNT D, acid-fn, 0.48 vol%, Glenium 7700
MWNT D, non-fn, 0.48 vol%, Glenium 7700
MWNT D, acid-fn, 0.48 vol%, ADVA CAST 575
MWNT D, non-fn, 0.48 vol%, ADVA CAST 575

113
123
113
100
126
133

7.96
9.54
15.7
9.51
7.15
11.9

201

14.3
7.09
16.3

Table 6.3- Mean values of flexural strength, maximum deflection and energy absorption capacity
of DSP paste with and without 0.24 vol.% of polymer-wrapped and non-functionalized MWNT D
dispersed in different fractions of mixing water.
Reinforcement Condition

Plain, ADVA CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575,
30% water
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575,
20% water
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575,
10% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575,
30% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575,
20% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575,
10% water

Mean
Flexure
Strength
(MPa)
13.9
20.3
23.6
18.4

Deflection
Energy
(mm)
Absorption
(N.mm)
0.71
1.96
2.48
1.96

155
367
404
315

15.6

1.81

217

14.4

1.68

186

21.5

2.15

375

19.3

1.91

325

16.6

1.82

265

Table 6.4- Mean values and standard errors of the compressive strength test results for DSP
paste with and without 0.24 vol.% polymer-wrapped and non-functionalized MWNT D dispersed
in different fractions of the mixing water.
Reinforcement Condition

Plain, ADVA CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575, 30% water
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575, 20% water
MWNT D, non-fn, 0.24 vol%, ADVA CAST 575, 10% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575, 30% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575, 20% water
MWNT D, poly-fn, 0.24 vol%, ADVA CAST 575, 10% water

202

Mean
Compressive
Strength
(MPa)
120
123
125
125
121
109
121
121
120

Standard
Error
(MPa)
5.89
9.54
5.00
6.71
11.9
10.3
6.09
9.24
5.79

(a) Dense DSP matrix

(b) Uniform dispersion of nanotubes within matrix
Figure 6.1- SEM mages of DSP paste reinforced with 0.24 vol.% of MWNT D casted with
ADVA® Cast 575 superplasticizer.
203

Figure 6.1 (cont’d)

(c) Pulled-out nanotubes at a fractured surface

204

CHAPTER 7

7

NANO- AND/OR MICRO-SCALE REINFORCEMENTS OF HIGHPERFORMANCE CEMENTITIOUS MORTAR

7.1 General
Multiwalled carbon nanotubes provide major advantages over (micro-scale) carbon fibers in
terms of mechanical, physical and geometric attributes. The geometric features of graphite
nanomaterials (including high aspect ratio and specific surface area, and close spacing within
matrix) benefit: (i) control of defect size and propagation within composites; (ii) energy
dissipation via frictional pullout; and (iii) tortuosity of the diffusion and crack propagation paths.
When effectively dispersed, the spacing of typical multi-walled nanotubes within matrix is three
orders of magnitude smaller than that of typical carbon fibers at similar volume fractions; the
surface area of multiwalled nanotubes available for energy dissipation through frictional pullout
is two orders of magnitude greater than that of carbon fiber.
Successful use of graphite nanomaterials in concrete requires: (i) thorough dispersion of
individual nanomaterials within the aqueous fresh mix; and (ii) adequate bonding of cement
hydrates to nanomaterial surfaces for effective stress transfer. Modification of the MWNT
surfaces for improved compatibility with aqueous media and cement hydrates can effectively
mobilize the distinct qualities of graphite nanomaterials towards reinforcement of cementitious
matrices. As aggressive acid-functionalization (Figure 1.11) for introduction of a high
concentration of functional groups on MWNTs has not produced the highest reinforcement
efficiency in cementitious matrices as shown in Chapter 4. Physisorption (wrapping) of
205

polyacrylic acid (PAA) onto nanotube surfaces (Figure 4.1) was evaluated in the work reported
herein as a means of improving the reinforcement efficiency of nanotubes in cementitious
matrices. Figure 7.1 shows an example SEM image of a fracture surface of a cementitious
nanocomposite produced in this work. This image shows the uniform dispersion of nanotubes
achieved through surface modification of nanotube and proper selection of the nanotube
dispersion methods and the nanocomposite mixing procedures.
Earlier results have shown that functionalized nanotubes, at relatively low (~0.1%) volume
fraction, can bring about balanced gains in diverse engineering properties of high-performance
cementitious matrices (Konsta-Gdoutos, Metaxa et al. 2010; Konsta-Gdoutos, Metaxa et al.
2010). Carbon nanotubes effectively arrest the initiation and propagation of cracks at the nanoscale. The effectiveness of nano-scale reinforcement diminishes as the crack width increases.
Given the effectiveness of micro-scale fibers in arresting larger cracks, nano- and micro-scale
(hybrid) reinforcement systems can render synergistic effects in cementitious matrices. Control
of crack growth by hybrid reinforcement at different scales benefits the mechanical performance
and durability of cementitious materials. The concept of hybridization of fibers has been
explored using micro-scale fibers of different types and dimensions (Lawler, Zampini et al.
2005; Banthia and Sappakittipakorn 2007).

7.2 Materials and Methods
7.2.1

Materials, Graphite Nanomaterials and Microfibers

Multiwalled carbon nanotubes (MWNT) and/or carbon microfibers (CMF) were used as
reinforcement in high-performance (DSP) mortar at different volume fractions. These graphite
nanomaterials and microfibers have been introduced in detail in Chapter 2, Section 2.4, Fig. 2.3

206

and Table 2.3. In the course of dispersing the nanotubes in water PAA was physisorbed upon
multiwalled carbon nanotubes. PAA with average molecular weight (Mw) of about 100,000 at
35 wt.% concentration in water, was purchased from Sigma-Aldrich. Figure 7.2 shows a
transmission electron micrograph PAA-physisorbed nanotube (CNTD-PAA), which exhibited
improved dispersion in aqueous media and interfacial interactions with cement hydrates.
7.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
mortar was used to evaluate the reinforcement efficiency of both multiwalled carbon nanotubes
and carbon microfibers at different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/or microfibers were prepared and cured following ASTM
procedures described in the same section.
7.2.3

Experimental Methods

The test procedures employed to determine the engineering properties of high-performance
mortar included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests
(ASTM D 7136) and Abrasion tests (ASTM C 944). SEM was also employed to gain further
insight into the structure and failure mechanisms of cement-based nanocomposites. Experimental
results were evaluated using ANOVA and PWC techniques. Response Surface Analysis (RSA)
of test results was used to identify the optimum reinforcement condition which yields balanced
gains in various engineering properties of the cementitious matrix.

7.3 Experimental Results and Discussion

207

7.3.1

Carbon Nanotubes and/ or Carbon Microfibers at Different Volume Fractions and
with Different Superplasticizers

7.3.1.1 Flexural Performance
The flexure test results for the high-performance (DSP) cementitious mortar reinforced with
different volume fractions of PAA-physisorbed multiwalled carbon nanotube (CNT D-PAA)
and/or carbon microfiber (CMF) are summarized in Table 7.1. These test results indicate that all
the reinforcement conditions considered here bring about improvements in all aspects of flexural
performance with both superplasticizer types used in the experimental program. The ADVA®
Cast 575 superplasticizer (when compared Glenium® 7700) was particularly effective in
enhancing the flexural attributes of the plain and reinforced high-performance cementitious
mortar, confirming the preliminary findings with high-performance cementitious paste (Sadiq,
Soroushian et al. 2012). The flexural strength, energy absorption capacity and maximum
deflection of plain matrix improved by 12.7%, 39.4% and 27.3%, respectively, when the
ADVA® Cast 575 superplasticizer was used in lieu of Glenium® 7700 superplasticizer.
With the Glenium® 7700 superplasticizer, 0.16 vol.% (volume percent of dry cementitious
materials) modified multiwalled carbon nanotube (CNT D-PAA) was found to be optimum,
producing 37.9%, 106% and 141% gains in flexural strength, energy absorption capacity and
maximum deflection, respectively. At 0.96 vol.% carbon microfiber (CMF), these improvements
were 39.0%, 119% and 156%, respectively (with the Glenium® 7700 superplasticizer).
With the ADVA® Cast 575 superplasticizer, the optimum nanotube (CNT D-PAA) volume
fraction was 0.24%, yielding 35.3%, 58.2% and 86.9% gains in flexural strength, energy
absorption capacity and maximum deflection, respectively, of plain matrix with the same
superplasticizer. The corresponding improvements brought about by 0.96 vol.% CMF were
208

27.8%, 74.9% and 84.8%, respectively. These results indicate that the reinforcing effects,
normalized with respect to volume fractions, are greater with the modified multiwalled carbon
nanotube (CNT D-PAA) when compared with carbon microfiber (CMF).
With both superplasticizers, the most balanced improvements in flexural performance
characteristics of the high-performance cementitious mortar were realized with the hybrid
reinforcement comprising both carbon nanotube and carbon fiber. With the ADVA® Cast 575
(Glenium® 7700) superplasticizer, the gains in flexural strength, energy absorption capacity and
maximum deflection were 13.4% (18.7%), 65.0% (105%) and 53.8% (84.0%), respectively,
noting that these improvements are versus a plain matrix made with the same superplasticizer.
Analysis of variance and pair-wise comparisons of test results indicated that (i) the gains in
flexural attributes of the high-performance cementitious mortars with the nano- and/or microscale reinforcement systems considered here were statistically significant (at 0.05 significance
level) with both superplasticizers; and (ii) the improvements in the flexural performance of
cementitious mortar brought about by the use of ADVA® Cast 575 superplasticizer in lieu of
Glenium® 7700 were also statistically significant.
7.3.1.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for the high-performance
cementitious mortars with nano- and/or micro-scale reinforcement systems and the two
superplasticizers are presented in Table 7.2. The effects of carbon nano- and/or micro-scale
reinforcement systems on the compressive strength of the high-performance cementitious mortar
were relatively small and statistically insignificant (per outcomes of the analysis of variance of
test results), confirming the observations made with cementitious paste presented earlier.

209

7.3.1.3 Impact Resistance
The impact test data are summarized in Table 7.3. With the ADVA® Cast 575 superplasticizer,
PAA-physisorbed carbon nanotubes (CNT D-PAA) at 0.16 and 0.24 vol.% produced 58.2% and
66.7% gains, respectively, in the impact resistance of high-performance cementitious mortar.
Micro-scale carbon fibers (CMF) at 0.24, 0.48 and 0.96 vol.% increased the impact resistance of
the high-performance cementitious mortar by 33.3%, 46.5% and 50.1%, respectively. Hybrid
(nano- and micro-scale) reinforcement produced further gains in impact resistance over those
achieved with micro-scale reinforcement used alone. Analysis of variance and pair-wise
comparisons of the impact test results confirmed the statistical significance of the gains in impact
resistance of the high-performance cementitious mortar with hybrid reinforcement. The ADVA®
Cast 575 superplasticizer was found to be more effective in improving the impact resistance of
the high-performance cementitious mortar in plain form and also with nano- and/or micro-scale
reinforcement systems. Comparisons with preliminary test results produced with highperformance cementitious paste indicated that the high-performance mortar (plain or with
different reinforcement systems) provided higher levels of impact resistance. This finding points
at the benefits rendered by sand particles to the impact resistance of the high-performance
cementitious matrix.
7.3.1.4 Abrasion Resistance
The abrasion test results performed on the high-performance cementitious mortar with carbon
nano- and/or micro-scale reinforcement and different superplasticizers are summarized in Table
7.4. With the ADVA® Cast 575 superplasticizer, PAA-physisorbed carbon nanotubes (CNT DPAA) at different volume fractions (and the hybrid reinforcement system) produced the greatest
improvement (34.9% - 39.0%) gains in the abrasion resistance of the high-performance

210

cementitious mortar. The benefits of all reinforcement systems to the abrasion resistance of the
high-performance cementitious mortar were found to be statistically significant.
7.3.1.5 Scanning Electron Microscope Evaluation
Figures 7.1, 7.3 and 7.4 show SEM images of the high-performance cementitious mortars. The
damaging effects of mechanical loads and aging are generally manifested in cementitious
materials as inception and growth of microcracks, which tend to be concentrated initially at the
aggregate-paste interfaces. The close (nano-scale) spacing and high specific surface of carbon
nanotubes seem to make them effective in suppression of microcrack inception (and initial
growth) (Figs 7.3a, b & c) and control of microcrack propagation by effectively interacting with
the microcrack tips (Figures 7.3a & c). Longer micro-scale fibers, on the other hand, are
particularly effective in bridging across wider cracks, and mitigating their further opening
(Figures 7.3c & f).
The relatively dense structure of the high-performance cementitious mortar is apparent in Figure
7.4a. The uniform dispersion of nanotubes and their bridging/pullout actions across a fine (nanoscale) crack within the cementitious matrix are depicted in Figures 7.1, 7.3b and 7.4b.
Indications of micro-fiber pullout from the matrix are provided in Figures 7.3b, 7.4b and 7.4c.
Scanning electron microscope (SEM) observations of the high-performance cementitious mortar
with hybrid reinforcement (comprising carbon nanotubes and micro-fibers) indicate that the
presence of nanotubes in the matrix around micro-fibers could have benefited the interaction of
matrix with micro-fibers (Figures 7.4b & c), thus benefiting the performance gains associated
with hybrid reinforcement. This strong interaction of matrix with micro-fibers was not observed
when the micro-fiber reinforcement was used alone (Figure 7.4d). Finally, although calcium
hydroxide crystals (C-H) were rarely found in the high-performance cementitious mortar, their

211

deleterious effects were reduced by nanotubes bridging across the weak zones formed by C-H
crystals (Fig. 7.4e). These observations provide some insight into the synergistic action of nanoand micro-scale reinforcement systems in cementitious matrices.
7.3.2

Enhanced Hybrid Reinforcement of High-Performance Cementitious Mortar

Response surface analysis of the test results presented above, with carbon nanotube and carbon
microfiber volume fractions used as input variables, led to selection of the following two
optimum hybrid reinforcement conditions: (i) 0.06 vol.% carbon nanotube (CNT D-PAA) & 0.30
vol.% carbon microfiber (CMF); and (ii) 0.08 vol.% carbon nanotube (CNT D-PAA) & 0.40
vol.% carbon microfiber (CMF). The simple fact that optimum reinforcement conditions
comprise both nano- and micro-scale systems points at the synergistic action of the two in
enhancing the performance characteristics of high-performance cementitious mortars.
The materials, and processing and testing methods introduced earlier were employed for
producing and characterizing high-performance cementitious mortars with the two optimum
reinforcement conditions introduced above (using the ADVA® Cast 575 superplasticizer). The
performance characteristics of these cementitious nanocomposites are reviewed in the following,
and compared against those of the high-performance cementitious mortar in plain form and also
with some preferred nano- and/or micro-scale reinforcement systems evaluated earlier.
7.3.2.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity test results for the
high-performance (DSP) cementitious mortar reinforced with different volume fractions of PAAphysisorbed multiwalled carbon nanotube (CNT D-PAA) and/or carbon microfiber (CMF) are
presented in Table 7.5. These results indicate that all the reinforcement conditions considered

212

here enhance the flexural performance characteristics of the high-performance cementitious
mortar. With a single reinforcement type, maximum improvements in flexural characteristics
were realized with 0.16 vol.% carbon nanotube (CNT D-PAA). This reinforcement condition
yielded 31.6%, 70.9% and 82.6% improvements (versus plain mortar) in flexural strength,
energy absorption capacity and maximum deflection, respectively. Carbon nanotube offered
higher reinforcement efficiency per unit volume when compared with carbon microfiber. The
experimental results presented in Table 6 indicate that the hybrid reinforcement systems provide
particularly high reinforcing effects, confirming the value rendered by the synergistic actions of
nano- and micro-scale reinforcement in high-performance cementitious mortar, confirming the
preliminary results produced with high-performance cementitious paste. The hybrid
reinforcement comprising 0.08 vol.% carbon nanotube and 0.40 vol.% carbon microfiber
produced the maximum improvements in flexural attributes, yielding 16.6%, 66.2% and 70.3%
improvements in flexural strength, energy absorption capacity and maximum deflection,
respectively. Lowering the nanotube volume fraction would reduce the flexural performance
characteristics of the high-performance cementitious mortar with hybrid reinforcement. Analysis
of variance and pair-wise comparison of test results indicated that: (i) carbon nanotube and
hybrid reinforcement systems produce statistically significant gains (at 0.05 significance level) in
all aspects of the flexural performance of the high-performance cementitious mortar. Carbon
microfiber, however, produced statistically significant gains in the flexural energy absorption
capacity (but not the flexural strength and maximum deflection) of the high-performance
cementitious mortar. This finding further confirms the desired reinforcing effects of nano-scale
and hybrid reinforcement systems, when compared with micro-scale fibers, in high-performance
cementitious mortar.

213

7.3.2.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
(DSP) cementitious mortars with carbon nanotube and/or microfiber reinforcement systems are
presented in Table 7.6. The mean values of compressive strength dropped slightly with the
addition of 0.24 vol.% carbon microfiber. The effects of carbon nanotube and/or microfiber on
the compressive strength of high-performance cementitious mortar, however, were not
statistically significant.
7.3.2.3 Impact Resistance
The impact test data are summarized in Table 7.7. The two (optimum) hybrid reinforcement
systems are observed to be effective in enhancing the impact resistance of DSP mortar (by 5065%). The hybrid reinforcement comprising 0.08 vol.% carbon nanotube (CNT D-PAA) and
0.40 vol.% carbon microfiber (CMF) produced the highest improvement in impact resistance of
the high-performance cementitious mortar. The contributions of carbon nanotube and/or carbon
microfiber to impact resistance were found to be statistically significant (at 0.05 significance
level).

7.4 Summary
Effects of reinforcement with multiwalled carbon nanotubes modified through physisorption of a
polyelectrolyte (PAA) and/or carbon microfiber on the flexural and compressive performance,
abrasion resistance, and the structure and failure modes of a high-performance fiber reinforced
cementitious mortar were evaluated. Desired dispersion and interfacial interactions of carbon
nanotubes in cementitious matrix were realized through modification of nanotubes (via PAA
physisorption on their surfaces) and sonication of nanotubes in water resulting in substantial

214

increase in all engineering properties of the high performance matrix. As was seen in highperformance paste matrix the selection of the superplasticizer type can significantly impact the to
reinforcement efficiency of carbon nanotubes in high-performance cementitious mortars. The
results were also substantiated by SEM observations. The use of response surface analysis (RSA)
of experimental results identified optimum hybrid reinforcement condition for high-performance
mortar matrix.

215

Table 7.1- Mean values of the flexural attributes high-performance cementitious mortars with
nano- and/or micro-scale reinforcement systems.
Reinforcement Condition - Superplasticizer
Plain – Gl
Plain - Adva
Carbon microfiber, 0.24 vol% (CMF-0.24) - Gl
Carbon microfiber, 0.24 vol% (CMF-0.24) - Adva
Carbon microfiber, 0.48 vol% (CMF-0.48) - Gl
Carbon microfiber, 0.48 vol% (CMF-0.48) - Adva
Carbon microfiber, 0.96 vol% (CMF-0.96) - Gl
Carbon microfiber, 0.96 vol% (CMF-0.96) - Adva
Carbon nanotube D, PAA-physisorbed, 0.16 vol%
(CNT D-PAA-0.16) - Gl
Carbon nanotube D, PA- physisorbed, 0.16 vol%
(CNT D-PAA-0.16) - Adva
Carbon nanotube D, PAA-physisorbed, 0.24 vol%
(CNTD-PAA-0.24) - Gl
Carbon nanotube D, PAA-physisorbed, 0.24 vol%
(CNTD-PAA-0.24) - Adva
Carbon microfiber, 0.24 vol%, & Carbon nanotube
D, PAA-physisorbed, 0.16 vol% (HYCNT1) - Gl
Carbon microfiber, 0.24 vol%, & Carbon nanotubes
D, PAA physisorbed, 0.16 vol% (HYCNT1) - Adva

216

Flexural
Strength
(MPa)
16.6
18.7
19.3
19.6
22.5
23.0
23.1
23.9
22.9

Deflection
(mm)
0.99
1.38
1.37
1.42
2.31
2.35
2.54
2.55
2.39

Energy
Absorption
(N.mm)
216
275
319
344
424
459
474
481
445

24.6

2.52

470

22.6

2.48

404

25.3

2.58

435

19.7

2.03

398

21.2

2.28

423

Table 7.2- Mean values and standard errors of the compressive strength of the high-performance
cementitious mortar matrices with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Mean
Standard
Compressive
Error
Strength
(MPa)
(MPa)

Plain – Gl

125

17.0

Plain - Adva

129

9.42

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl

119

5.69

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva

121

10.4

Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl

138

8.45

Carbon microfiber, 0.48 vol%-(CMF-0.48) - Adva

137

5.44

Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl

127

17.6

Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva

128

6.57

Carbon nanotubes D, PAA physisorbed, 0.16 vol%-(CNT D-PAA0.16) - Gl

132

6.21

Carbon nanotubes D, PAA physisorbed, 0.16 vol%-(CNT D-PAA0.16) - Adva

135

4.59

Carbon nanotubes D, PAA physisorbed, 0.24 vol%-(CNTD-PAA0.24) - Gl

125

3.18

Carbon nanotubes D, PAA physisorbed, 0.24 vol%-(CNTD-PAA0.24) - Adva

130

3.58

Carbon microfiber, 0.24 vol%, & Carbon nanotubes D, PAA
physisorbed, 0.16 vol%-(HYCNT1) - Gl

126

17.4

Carbon microfiber, 0.24 vol%, & Carbon nanotubes D, PAA
physisorbed, 0.16 vol%-(HYCNT1) - Adva

131

12.8

217

Table 7.3- Mean values of the impact resistance test results for high-performance cementitious
mortars reinforced with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Impact Resistance,
mm/mm

Plain – Gl

0.573

Plain - Adva

0.625

Carbon microfiber, 0.24 vol% (CMF-0.24) - Gl

0.781

Carbon microfiber, 0.24 vol% (CMF-0.24) - Adva

0.833

Carbon microfiber, 0.48 vol% (CMF-0.48) - Gl

0.833

Carbon microfiber, 0.48 vol% (CMF-0.48) - Adva

0.885

Carbon microfiber, 0.96 vol% (CMF-0.96) - Gl

0.938

Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva

0.938

Carbon nanotubes D, PAA physisorbed, 0.16 vol% (CNT D-PAA0.16) - Gl

0.938

Carbon nanotubes D, PAA physisorbed, 0.16 vol% (CNT D-PAA0.16) - Adva

0.989

Carbon nanotubes D, PAA physisorbed, 0.24 vol% (CNTD-PAA0.24) - Gl

0.990

Carbon nanotubes D, PAA-physisorbed, 0.24 vol% (CNTD-PAA0.24) - Adva

1.042

Carbon microfiber, 0.24 vol% & Carbon nanotube D, PAAphysisorbed, 0.16 vol% (HYCNT1) - Gl

0.938

Carbon microfiber, 0.24 vol% & Carbon nanotube D, PAAphysisorbed, 0.16 vol% (HYCNT1) - Adva

0.989

218

Table 7.4- Mean values of the abrasion resistance test results for high-performance cementitious
mortars reinforced with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Loss of
Mass (gr)

Plain – Gl

1.38

Plain - Adva

1.23

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl

0.98

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva

0.93

Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl

0.93

Carbon microfiber, 0.48 vol%-(CMF-0.48) - Adva

0.93

Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl

0.85

Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva

0.83

Carbon nanotubes D, PAA physisorbed, 0.16 vol%-(CNT D-PAA-0.16) - Gl

0.83

Carbon nanotubes D, PAA physisorbed, 0.16 vol%-(CNT D-PAA-0.16) Adva

0.80

Carbon nanotubes D, PAA physisorbed, 0.24 vol%-(CNTD-PAA-0.24) - Gl

0.80

Carbon nanotubes D, PAA physisorbed, 0.24 vol%-(CNTD-PAA-0.24) Adva

0.75

Carbon microfiber, 0.24 vol%, & Carbon nanotubes D, PAA physisorbed,
0.16 vol%-(HYCNT1) - Gl

0.80

Carbon microfiber, 0.24 vol%, & Carbon nanotubes D, PAA physisorbed,
0.16 vol%-(HYCNT1) - Adva

0.80

219

Table 7.5- Mean values of flexural strength, maximum deflection and energy absorption capacity
of high-performance cementitious mortars with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Mean Flexure
Strength
(MPa)

Deflection
(mm)

Energy
Absorption
(N.mm)

Plain - Adva

18.7

1.38

275

Carbon microfiber, 0.24 vol%-(CMF0.24) - Adva

19.6

1.42

344

Carbon nanotubes D, PAA physisorbed,
0.16 vol%-(CNTD-PAA-0.16) - Adva

24.6

2.52

470

Carbon microfiber, 0.24 vol%, & Carbon
nanotubes D, PAA physisorbed, 0.16
vol%-(HYCNT1) - Adva

21.2

2.28

423

Carbon microfiber, 0.30 vol%, & Carbon
nanotubes D, PAA physisorbed, 0.06
vol%-(HYCNT2) - Adva

21.0

2.29

441

Carbon microfiber, 0.40 vol%, & Carbon
nanotubes D, PAA physisorbed, 0.08
vol%-(HYCNT3) - Adva

21.8

2.35

457

Table 7.6- Mean values and standard errors of the compressive strength test results for high –
performance cementitious mortars with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Mean
Compressive
Strength
(MPa)

Standard
Error
(MPa)

Plain - Adva

129

9.42

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva

121

10.4

Carbon nanotubes D, PAA physisorbed, 0.16 vol%(CNTD-PAA-0.16) - Adva

135

4.59

Carbon microfiber, 0.24 vol%, & Carbon nanotubes
D, PAA physisorbed, 0.16 vol%-(HYCNT1) - Adva

131

12.8

Carbon microfiber, 0.30 vol%, & Carbon nanotubes
D, PAA physisorbed, 0.06 vol%-(HYCNT2) - Adva

133

4.83

Carbon microfiber, 0.40 vol%, & Carbon nanotubes
D, PAA physisorbed, 0.08 vol%-(HYCNT3) - Adva

130

7.99

220

Table 7.7- Mean values of the impact resistance test results for high-performance cementitious
mortars with nano- and/or micro-scale reinforcement.
Reinforcement Condition

Impact
Resistance
(mm/mm)

Plain - Adva

0.63

Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva

0.83

Carbon nanotubes D, PAA physisorbed, 0.16 vol%-(CNTD-PAA-0.16) Adva

0.99

Carbon microfiber, 0.24 vol%, & Carbon nanotubes D, PAA physisorbed,
0.16 vol%-(HYCNT1) - Adva

0.99

Carbon microfiber, 0.30 vol%, & Carbon nanotubes D, PAA physisorbed,
0.06 vol%-(HYCNT2) - Adva

0.95

Carbon microfiber, 0.40 vol%, & Carbon nanotubes D, PAA physisorbed,
0.08 vol%-(HYCNT3) - Adva

1.04

221

Figure 7.1- SEM image of a cement-based material incorporating well dispersed PAAphysisorbed multiwalled carbon nanotubes (CNT D-PAA).

Figure 7.2- Transmission electron microscope image of PAA physisorbed multiwalled carbon
nanotube Type D.

222

(a) Schematics

(b)
Figure 7.3- Complementary roles of nano- and micro-scale reinforcement in damage control by
suppression of micro-crack inception and control of microcrack growth.

223

Figure 7.3 (cont’d)

(c) Control of damage inception by nanomaterials

(d)

224

Figure 7.3 (cont’d)

(e) Control of Damage propagation by nanomaterials

(f) Control of damage growth by micro-scale fibers

225

(a)

(b)
Figure 7.4- Typical SEM images of the fractured surfaces of high-performance cementitious
mortars with nano- and micro-scale reinforcement systems.

226

Figure 7.4 (cont’d)

(c)

(d)
227

Figure 7.4 (cont’d)

(e)

228

CHAPTER 8

8

DAMAGE-SENSITIVE ELECTRICAL CONDUCTIVITY OF HIGHPERFORMACNE MORTARS REINFORCED WITH MULTIWALLED
CARBON NANOTUBES

8.1 General
Carbon nanotubes, when properly functionalized and thoroughly dispersed in well-graded
cementitious materials, bring about balanced gains in diverse engineering properties of highperformance concrete. These properties include fracture toughness, impact resistance, strength,
barrier qualities, abrasion resistance, fatigue life, and weathering resistance (Metaxa, KonstaGdoutos et al. 2009; Metaxa, Konsta-Gdoutos et al. 2010; Metaxa, Konsta-Gdoutos et al. 2010).
In addition, multiwalled carbon nanotubes offer distinctly high thermal and electrical
conductivity. They can thus significantly increase the conductivity attributes of cement-based
materials when a percolated (networked) structure of nanomaterials forms within cement-based
matrices (Figure 8.1). The distinctly high aspect ratio of nanomaterials substantially lowers the
“threshold” volume fraction at which the percolation phenomenon occurs. Percolated
nanomaterials provide continuous paths for transport of electricity and heat; given the relatively
high aspect (length-to-diameter) ratio of carbon nanotubes, well-dispersed nanotubes percolate at
relatively low volume fractions. Break-up of the percolated network under damaging effects such
as microcracks and large strains could disrupt the continuous conduction path, and thus lower the
conductivity of cementitious nanocomposites. The damage-sensitivity of electrical conductivity

229

offers a convenient basis for health monitoring of cementitious nanocomposites (and concretebased infrastructure systems incorporating them).
Applications of carbon nanotubes towards development of self-sensing, self-monitoring
materials have generally employed the piezo-resistive qualities of carbon nanotubes in polymer
matrices (Sandlar, Shaffer et al. 1999; Hu, Zhao et al. 2006); limited work has been reported on
the use of carbon nanotubes in cement paste (Li, Wang et al. 2007; Han, Yu et al. 2009; Han, Yu
et al. 2011). The available data point at the potential of carbon nanotubes to yield
nanocomposites with inherent structural health monitoring attributes. The work reported uses
carbon nanotubes in a high-performance cementitious mortar (incorporating fine aggregates),
and relies upon the disturbance of the percolation conditions of carbon nanotubes as the matrix
experiences damage to enable health monitoring via measurement of electrical conductivity.

8.2 Materials and Methods
8.2.1

Materials and Graphite Nanomaterials

Modified multiwalled carbon nanotubes, modified by physisorption of PAA (polyacrylic acid),
were used in a high-performance (DSP) cementitious mortar matrix. These graphite
nanomaterials have been introduced in detail in Chapter 2, Fig. 2.3 and Table 2.3. In the course
of dispersing the nanotubes in water PAA was physisorbed upon multiwalled carbon nanotubes.
PAA with average molecular weight (Mw) of about 100,000 at 35 wt.% concentration in water,
was purchased from Sigma-Aldrich. Figure 7.4 shows a transmission electron micrograph PAAphysisorbed nanotube (CNTD-PAA), which exhibited improved dispersion in aqueous media
and interfacial interactions with cement hydrates.

230

8.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
mortar was used to evaluate the pressure-sensitive properties of both graphite nanomaterials at
different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) were prepared and cured following ASTM procedures described in
the same section.
8.2.3

Experimental Methods

The electrical conductivity of high-performance cementitious mortar specimens (75 x 50 x 12.5
mm) with different fractions of MWNTs modified by PAA physisorption, prepared using two
different superplasticizers, was measured in dry condition using Gamry Electro-Chemical
Spectrometer following the procedures of ASTM D991, D 4496 and D 257. The experimental
setup is shown in Figure 8.2. PELCO® Conductive Silver paint was applied, and copper tapes
were firmly attached to the opposite faces of the specimen. Electrodes were connected to copper
tapes, and electrical conductivity was measured in dry condition in longitudinal direction. In
order to determine the damage-sensitivity of the electrical conductivity of nanocomposites,
conductivity tests were also conducted after subjecting cube (50 x 50 x 50 mm) specimens of the
high-performance cementitious mortars with or without nanotubes to 60% and 80% of their
mean compressive strength. The means (standard errors) of the compressive strength test results
for the high-performance cementitious mortars with 0, 0.16 and 0.24 vol.% (volume percent of
dry cementitious materials) of modified carbon nanotubes, prepared using the Glenium® 7700
superplasticizer, were 125 (17.0), 132 (6.21) and 125 (3.18) MPa, respectively. The
corresponding values for cementitious materials prepared with the ADVA® Cast 575
231

superplasticizer were 129 (9.42), 135 (4.59) and 130 (3.58) MPa, respectively.

For each

specimen, three resistance measurements were made in each (longitudinal and transverse)
direction, and the average was recorded. Electrical conductivity was calculated as inverse of the
measure value of resistance. The specimens were also subjected to SEM in order to gain further
insight into their structure and failure mechanisms. Experimental results were evaluated using
the analysis of variance (ANOVA) and pair-wise comparison techniques.

8.3 Experimental Results and Discussion
8.3.1

Effects of Modified Multiwalled Carbon Nanotubes on the Electrical Conductivity
of High-Performance Cementitious Mortars

The mean values of electrical conductivity for high-performance cementitious mortars with
different volume fractions of modified multiwalled carbon nanotubes (prepared using two
different superplasticizers) are presented in Table 8.1. Depending on the superplasticizer type,
the addition of 0.16 vol.% modified carbon nanotube increased the electrical conductivity of the
high-performance cementitious mortar by two to three orders of magnitude. Increasing the
dosage of modified nanotube from 0.16 to 0.24 vol.% further increased the electrical
conductivity of cementitious mortar by an order of magnitude. The significant rise in electrical
conductivity of cementitious mortars with the addition of 0.16 and 0.24 vol.% modified carbon
nanotubes indicates that a percolated network of nanotubes is formed at either volume fraction.
The experimental results indicate that the selection of superplasticizer can significantly impact
the effects of nanotubes on the electrical conductivity of high-performance cementitious mortars.
This could be attributed to the differences in the thoroughness of nanotube dispersion realized
with different superplasticizers and/or the differences in (nanotube-to-nanotube) interfacial
electrical resistance when different superplasticizers are used.
232

8.3.2

Damage-Sensitivity

of

the

Electrical

Conductivity

of

High-Performance

Cementitious Mortars Incorporating Modified Carbon Nanotubes
The mean values of electrical conductivity for high-performance cementitious mortars with 0.16
and 0.24 vol.% modified multiwalled carbon nanotubes prepared with the ADVA® Cast 575
superplasticizer are presented in Table 8.2. The results clearly indicate establishment of
conductive percolated network of carbon nanotubes at both 0.16 and 0.24 vol.% nanotube
additions. The test results presented in Tables 8.2 and 8.3 were produced using specimens of
different dimensions prepared using different batches of cementitious materials, which can be
used to explain the differences between electrical conductivity measurements for specimens
prepared using the same (ADVA® Cast 575) superplasticizer.
All specimens containing nanotubes, when subjected to increasing compressive stress levels,
experienced a sharp drop in electrical conductivity, which can be attributed to breakdown of the
percolated nanotube network under the damaging effects (e.g., microcracking and large
permanent strains) of compression loading. After application of 60% of compressive strength,
electrical conductivity dropped by one order of magnitude. Application of 80% of compressive
strength produced two orders of magnitude drop in electrical conductivity. The higher
conductivity levels in the presence of 0.24 vol.% carbon nanotubes simplify the conductivity
measurements. The significant damage sensitivity of electrical conductivity in cementitious
materials incorporating carbon nanotubes points at the inherent health monitoring qualities of the
material.
8.3.3

Scanning Electron Microscope Observations

Failed surfaces of compression test specimens were evaluated under a high-precision scanning
electron microscope (JOEL 7500F). All samples were coated with Osmium (using Osmium
233

Coater Neoc-AN, Meiwa Shoji) prior to SEM observations. The scanning electron microscope
images depicted in Figure 8.3 point at a dense cementitious matrix incorporating uniformly
dispersed carbon nanotubes which seem to form a percolated network.

8.4 Summary
Multiwalled carbon nanotubes were modified through physisorption of a polyelectrolyte, and
were dispersed within a high-performance cementitious mortar at relatively low (0.16-0.24
vol.%) volume fractions. The effects of nanotubes on the electrical conductivity of the
cementitious mortar were evaluated, and the damage-sensitivity of electrical conductivity was
evaluated by making conductivity measurements prior to and after application of different
fractions of compressive strength. The results show that functionalized carbon nanotubes form
percolation networks within the matrix thus increasing the conductivity of the matrix. The
pressure sensitive properties of the matrix can be effectively utilized in self-sensing, selfmonitoring and smart structures.

234

Table 8.1- Mean values of the electrical conductivity test results for high-performance
cementitious mortars with different volume fractions of modified multiwalled carbon nanotubes
prepared using two different superplasticizers.
Reinforcement Condition

Electrical
Conductivity, 1/ Ω m
-6

Plain

1.09 x 10

CNT D-PAA, 0.16 vol% - Gl

#

-4

1.19 x 10

-3

CNT D-PAA, 0.16 vol% - Ad*
CNT D-PAA, 0.24 vol% - Gl

4.72 x 10

CNT D-PAA, 0.24 vol% - Ad
#

1.72 x 10

2.07 x 10

Glenium® 7700;

-3
-2

*ADVA® Cast 575

Table 8.2- Mean values of the electrical conductivity of high-performance cementitious mortar
specimens with different volume fractions of functionalized multiwalled carbon nanotubes after
application of different compressive stress levels.
Reinforcement Condition

Electrical
Conductivitylongitudinal
direction,
1/ Ω m
-3

CNT D-PAA, 0.16 vol%

1.92 x 10

CNT D-PAA, 0.16 vol%, 60% ultimate

4.14 x 10

CNT D-PAA, 0.16 vol%, 80% ultimate

1.05 x 10

CNT D-PAA, 0.24 vol%

2.07 x 10

CNT D-PAA, 0.24 vol%, 60% ultimate

3.15 x 10

CNT D-PAA, 0.24 vol%, 80% ultimate

4.23 x 10

-4
-5
-2
-3
-4

235

Electrical
Conductivitytransverse direction,
1/ Ω m
-3

2.25 x 10

-4

4.72 x 10

-4

1.12 x 10

-2

2.80 x 10

-3

4.05 x 10

-4

4.15 x 10

Figure 8.1- Percolated network.

(a) Experimental Setup
Figure 8.2- Electrical conductivity test setup and specimens.

236

Figure 8.2 (cont’d)

(b) Prepared specimens prior to test

(c) Specimen surfaces coated with silver paint

237

(a) Dense matrix

(b) Uniform dispersion of nanotubes within the matrix
Figure 8.3- Typical SEM mages of high-performance cementitious mortar reinforced with
modified multiwalled carbon nanotube.
238

Figure 8.3 (cont’d)

(c) Percolation of nanotubes within the matrix

239

CHAPTER 9

9

EVALUATION OF HYBRID (NANO- AND MICRO-SCALE)
REINFORCEMET EFFECTS ON THE ENGINEERING OF HIGHPERFORMANCE CONCRETE

9.1 General
High-performance cementitious materials are subject of intense development efforts. Given the
brittle nature of high-strength cementitious materials, discrete reinforcement systems (fibers) are
used to render toughening and crack control effects. The reinforcing fibers commonly used in
high-performance cementitious materials have diameters in micrometer to millimeter scales.
Their primary reinforcing effect involves bridging across cracks to control their growth and
dissipate energy via frictional pullout.
Carbon nanotubes (CNTs) present several distinct advantages over conventional fibers as
reinforcement for high-performance cementitious materials. These advantages include: (i) high
strength, modulus and aspect ratio which benefit their crack-arrest and toughening effects; (ii)
high specific surface area and close spacing, which intensify their interactions with cement
hydrates and microcracks at their inception. These advantages can be realized as far as the
nanotubes are uniformly dispersed within the cementitious matrix, and develop strong bonds to
cement hydrates.
The reinforcement efficiency of carbon nanotubes has been evaluated primarily in cementitious
pastes (without the addition of any aggregates), and occasionally in cementitious mortars with
relatively fine aggregates; modest gains in some engineering properties have been reported with
240

the addition of relatively low volume fractions of nanotubes (Makar and Beaudoin 2004;
Cwirzen, Habermehl-Cwirzen et al. 2008; Cwirzen, Habermehl-Cwirzen et al. 2009; Makar and
Chan 2009; Metaxa, Konsta-Gdoutos et al. 2009; Metaxa, Konsta-Gdoutos et al. 2010; Li, Wang
et al. 2005). The work reported herein is part of a comprehensive investigation on the use of
modified carbon nanotubes in high-performance cementitious paste, mortar and concrete. Proper
selections of the nanotube modification and dispersion methods, and the cementitious matrix mix
designs have led to significant gains in performance characteristics of high-performance
cementitious materials with the addition of relatively low nanotube volume fractions. Results of
these background studies have been shown in earlier chapters.
The work reported herein focused on the use of hybrid (nano- and micro-scale) discrete
reinforcement systems in high-performance (DSP) concrete. The experience gained with the use
of nano-scale reinforcement in cementitious materials has largely avoided the use of coarse
aggregates in the matrix. The use of coarse aggregates in this work adds to the complexity of the
structure and failure modes of the cementitious matrix. The presence of coarse aggregates limits
the volume of paste available for dispersion of nano-scale reinforcement. Formation of an
interfacial zone around aggregates, which is more porous, has a higher concentration of coarser
crystals, and is prone to shrinkage microcracking, are some key effects of coarse aggregates in
cementitious matrix. A key objective of this investigation was to assess the contributions of
functionalized multiwalled carbon nanotubes, alone or in combination with micro-scale fibers, to
the performance characteristics of high-performance concrete incorporating coarse aggregate.
Cementitious materials are essentially particulate composites, which are rarely used without
aggregates (particulates). Given the micro- to millimeter-scale dimensions of the sand and gravel
particles used in high-performance (DSP) concrete (Figure 9.1), hybrid reinforcement which

241

renders reinforcing effects at different scales could produce particularly positive effects. In the
past, the concept of using hybrid (micro- and millimeter-scale) reinforcement has been explored
for achieving balanced gains in material properties (Moranville-Regourd 2001; O'Connell, Boul
et al. 2001; Lawler, Zampini et al. 2005; Banthia and Sappakittipakorn 2007). The work reported
herein has extended this background work by making complementary use of reinforcement at
micro- and nano-scales.

9.2 Materials and Methods
9.2.1

Graphite Nanomaterials and Microfibers

PAA (Polyacrylic acid) physisorbed MWNTs and/or carbon microfibers (CMF) and
polypropylene microfibers (PP) were used as reinforcement in high-performance (DSP) concrete
at different volume fractions. These graphite nanomaterials and microfibers have been
introduced in detail in Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. In the course of dispersing
the nanotubes in water PAA was physisorbed upon multiwalled carbon nanotubes.
9.2.2

Cementitious Matrices, Mixing and Curing Procedure

Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5.1.
DSP concrete was used to evaluate the reinforcement efficiency of both graphite nanomaterials
and microfibers at different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or different microfibers were prepared and cured following
ASTM procedures described in the same section.

242

9.2.3

Experimental Methods

The test procedures employed to determine the engineering properties of high-performance
concrete included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests
(ASTM D 7136) and Abrasion tests (ASTM C 944). SEM was also employed to gain further
insight into the structure and failure mechanisms of cement-based nanocomposites. Experimental
results were evaluated using the analysis of variance (ANOVA) and pair-wise comparison
techniques. Response Surface Analysis (RSA) of test results was used to identify the optimum
reinforcement condition which yields balanced gains in various engineering properties of the
cementitious matrix.

9.3 Experimental Results and Discussion
9.3.1

Carbon Nanotubes and/ or Carbon Microfibers at Different Volume Fractions and
with Different Superplasticizers

9.3.1.1 Flexural Performance
The flexure test results for high-performance concrete reinforced with multiwalled carbon
nanotubes and/or microfibers are summarized in Table 9.1. These test results indicate that all
reinforcement systems produced gains in the flexural strength, maximum deformation (ductility)
and energy absorption capacity of the high-performance concrete, except for the two microfibers
which lowered the flexural strength of concrete when used alone. The most desired balance of
flexural properties for a single reinforcement system was realized by the addition of 0.16 vol.%
(volume percent of dry cementitious materials) of CNTD-PAA. The corresponding
improvements in flexural strength, energy absorption capacity and maximum deflection with
respect to plain concrete were 23.0%, 151% and 123%, respectively. Test results also indicate

243

that the addition of each of the microfibers to this (relatively brittle) high-strength concrete
produced important gains in two out of the three flexural attributes considered here (maximum
deflection and energy absorption capacity), while causing a relatively small drop in flexural
strength. Polypropylene microfiber produced the maximum rise in energy absorption capacity
(315%) and maximum deflection (550%).
The test presented in earlier chapters for cementitious paste and mortar matrices have pointed at
the positive effects of hybrid (nano- and micro-scale) reinforcement systems in highperformance cementitious materials. The test data presented in Table 9.1 suggest that
polypropylene microfibers of lower modulus and lower cost (when compared with carbon
microfibers) could effectively complement the reinforcing effects of nanomaterials in highperformance concrete. Hybrid reinforcement systems comprising polypropylene microfibers and
multiwalled carbon nanotubes produced balanced gains in the flexural performance
characteristics of high-performance concrete by interacting with and arresting cracks which
developed in matrix at different scales. The hybrid reinforcement comprising 0.16 vol.% PAAphysisorbed multiwalled carbon nanotube and 0.24 vol.% polypropylene microfiber produced
41.8%, 648% and 352% rise in flexural strength, maximum deflection and energy absorption
capacity, respectively. Hybrid reinforcement systems also overcame the adverse effects of
microfibers on flexural strength. The synergistic reinforcement actions of nano- and micro-scale
reinforcement systems point at the merits of multi-scale reinforcement in high-performance
concrete. This effectiveness of hybrid reinforcement was also evident in the case of highmodulus carbon microfibers used together with multiwalled carbon nanotube. This hybrid
reinforcement produced improvements in flexural performance of high-performance concrete
when compared with plain concrete and also that reinforced with carbon microfiber alone.

244

Statistical analysis (of variance) of flexure test results indicated that there are statistically
significant (at 0.05 significance level) improvements in all flexural attributes of highperformance concrete due to the addition of nano-scale and hybrid reinforcement systems. Pairwise comparisons were carried out in order to assess the statistical significance of the effects
associated with the addition of nano- and/or micro-scale reinforcement systems. These results
indicated that nano-scale reinforcement at 0.16 vol.% produced statistically significant
improvements in all flexural attributes of the plain concrete at 0.05 significance level. Both
microfibers produced statistically insignificant drops in flexural strength (0.297 and 0.481
significance levels for carbon and polypropylene microfibers, respectively). Pair-wise analysis
also pointed at statistically insignificant gains in energy absorption capacity (0.502 significance
level) and maximum deflection (0.402 significance level) realized with 0.24 vol.% carbon
microfiber. Combined use of carbon nanotube with any of the two microfibers (low-modulus
polypropylene or high-modulus carbon microfiber) exhibited synergistic effects of nano- and
micro-scale reinforcement, producing balanced gains in different flexural attributes of highperformance concrete at significance levels varying from 0.000 to 0.047.
9.3.1.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
concretes with nano- and/or micro-scale reinforcement systems are presented in Table 9.2. As
earlier results with high-performance paste and mortar indicates, PAA-physisorbed multiwalled
carbon nanotubes as well as polypropylene or carbon microfibers produced relatively small and
statistically insignificant (at 0.05 significance level) effects on the compressive strength of highperformance concrete. The use of hybrid reinforcement systems comprising carbon or
polypropylene microfiber and carbon nanotube restored the compressive strengths of the high-

245

performance concrete by inducing interactions with matrix and its cracks at different scales. The
hybrid reinforcement system comprising PAA-physisorbed multiwalled carbon nanotube and
micro-scale polypropylene fibers produced a small rise in the compressive strength of highperformance concrete.
9.3.1.3 Impact Resistance
The impact test data are summarized in Table 9.3. All nano- and/or micro-scale reinforcement
systems produced improvements in the impact resistance of high-performance concrete. The
maximum rise in impact resistance achieved with a single reinforcement (73%) was for 0.16
vol.% of CNTD-PAA. In the case of micro-scale fibers, polypropylene produced a higher gain
(60%) than carbon in the impact resistance of high-performance concrete.
All hybrid reinforcement systems produced further gains in the impact resistance of highperformance concrete when compared with individual (nano-or micro-scale) reinforcement used
alone. The maximum rise in the impact resistance of high-performance concrete (120%) was
produced by the hybrid reinforcement comprising 0.24 vol.% polypropylene microfiber and 0.16
vol.% multiwalled carbon nanotube; this gain in impact resistance was found to be statistically
significant (at 0.000 significance level). Outcomes of the analysis of variance and pair-wise
comparisons of the impact test data pointed at the statistical significant (at 0.000 to 0.012
significance levels) of nano- and/or micro-scale reinforcement effects on the impact resistance of
high-performance concrete. When compared with the test data presented in earlier chapters on
the impact resistance of high-performance cementitious paste and mortar (with nano-scale
reinforcement), concrete (with nano-scale reinforcement) provides higher levels of impact
resistance; this observation points at the well-known contributions of aggregates (in concrete) to
the toughness of the cementitious matrix.

246

9.3.1.4 Abrasion Resistance
The abrasion test results produced for high-performance concrete with different reinforcement
conditions are summarized in Table 9.4. Multiwalled carbon nanotube at 0.16 vol.% produced
the greatest improvement (40.0%) in the abrasion resistance of high-performance concrete. All
nano- and/or micro-scale reinforcement systems produced marked gains in the abrasion
resistance of the high-performance concrete, which were statistically significant at 0.05
significance level. Outcomes of pair-wise comparisons confirmed that each of the reinforcement
conditions considered here produced statistically significant improvements in the abrasion
resistance of high-performance concrete (with significance levels ranging from 0.006 to 0.000).
9.3.1.5 Response Surface Analysis of Test Results
The test data on high-performance concrete with nano- and/or micro-scale reinforcement systems
was subjected to response surface analysis. The objective of this analysis was to identify
optimum reinforcement conditions of high-performance concrete which maximize the benefits to
specific engineering properties or a combination of such properties.
Response surface analysis (RSA) of the test data was conducted considering the volume fractions
of nano- and micro-scale reinforcement systems as input variables. Flexural strength, energy
absorption capacity, maximum deflection and impact resistance were used as response variables.
The response surface analysis process started with evaluating the effects of input variables on
response variables, and was followed by desirability analyses; two methods of response surface
analysis (canonical and ridge) were used in this investigation.
Response Surface analyses performed separately for the flexural strength, energy absorption
capacity, maximum deflection and impact resistance test data (Figure 9.1) indicated that an

247

optimum reinforcement system comprising 0.15 vol.% CNTD-PAA and 0.23 vol.% PP
microfiber maximizes subject properties. Optimal mean values (95% confidence intervals) for
flexural strength, energy absorption capacity, maximum deflection and impact resistance were
15.6 (15.0 - 16.2) MPa, 282.8 (256.7 - 308.9 N/mm), 3.10 (2.79 - 3.41) mm and 2.50 mm/mm
(2.38 - 2.63) mm/mm, respectively. Outcomes of Ridge analysis, which can be used to further
refine the reinforcement condition, indicated that the optimum hybrid reinforcement identified in
this investigation is reasonable for the performance characteristics considered here.
Desirability analysis of test data was conducted in order to identify reinforcement conditions
which yield a desired balance of all the engineering properties considered here. Desirability
analysis was conducted using the mean values obtained through both canonical and ridge
analyses. These outputs indicate that a desired balance of all response variables can be achieved
with 0.14 (or 0.15) vol.% CNTD-PAA carbon nanotube and 0.19 (or 0.20) vol.% PP microfiber.
9.3.1.6 Scanning Electron Microscope Evaluation
Figure 9.2 shows SEM images of the high-performance concrete samples. Figures 9.2a and 9.2b
point at the relatively dense structure of the high-performance concrete material considered here.
Figure 9.2a depicts an aggregate-paste interfacial crack that has propagated into the cementitious
matrix. Figures 9.2c through 9.2e are indicative of the uniform dispersion of nanotubes within
cementitious matrix, and their bridging/pullout actions across fine cracks developing within
matrix. Figures 9.2f and 9.2g depict pullout of microfibers from the cementitious matrix. The
scanning electron microscope observations presented in Figure 9.2 indicate that nano- and microscale reinforcement systems interact with cracks at different scales, which partly explains their
complementary

reinforcement

effects

in

high-performance

concrete.

The

nano-scale

reinforcement also seems to benefit the interfacial interactions of micro-scale fibers with the

248

cementitious matrix, which can further explain their complementary effects on cementitious
matrix. These observations rationalize the finding that optimum reinforcement systems for highperformance concrete comprise both nano- and micro-scale reinforcement systems.

9.4 Summary
The effects of reinforcement with MWNTs modified by physisorption of a polyelectrolyte (PAA)
and/or polypropylene and carbon microfibers on the engineering properties of a highperformance concrete were evaluated. Proper use of nano-scale reinforcement in highperformance concrete can yield improvements in mechanical properties which surpass those
realized with similar dosages of micro-scale fibers. Nano- and micro-scale reinforcements render
complementary reinforcing effects in high-performance concrete; optimum reinforcement
systems thus incorporate both carbon nanotubes and microfibers.
The synergistic actions of nano- and micro-scale reinforcement systems in high-performance
concrete can be explained by their interactions with cracks developing in the cementitious matrix
at different scales, and also by the positive effects of nanotubes on the interfacial interactions of
micro-scale fibers with the high-performance cementitious matrix. These synergistic actions of
nano- and micro-scale reinforcement were realized with nanotubes that were uniformly dispersed
within the matrix.

249

Table 9.1- Mean values of flexural attributes of high-performance concrete with and without
different nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition  Flexural Strength
(MPa) 

Deflection
(mm) 

Energy
Absorption
(N.mm) 

Plain 

13.4

0.71

92 

CMF-0.24 

12.8

1.01

111 

PP-0.24 

13.0

4.55

383 

CNT D-0.16 

16.5

1.58

231 

CNT D-PAA+PP 

19.7

5.31

416 

CNT D-PAA+CMF 

19.0

1.78

316 

 

Table 9.2- Mean values and standard errors of the compressive strength test results for highperformance concrete with and without nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition

Mean
Compressive
Strength (MPa) 

Standard
Error (MPa) 

Plain 

151

7.75

CMF-0.24 

139

13.5

PP-0.24 

137

18.4

CNT D-0.16 

143

5.55

CNT D-PAA+PP 

153

11.0

CNT D-PAA+CMF 

150

7.53

 

250

Table 9.3- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement.
Reinforcement Condition

Impact
Resistance
(mm/mm) 

Plain 

1.50

CMF-0.24 

1.98

PP-0.24 

2.40

CNT D-0.16

2.60

CNT D-PAA+PP

3.30

CNT D-PAA+CMF

2.65

 

Table 9.4- Mean abrasion weight losses of high-performance concretes with and without nanoand/ or micro-scale reinforcement systems.
Reinforcement Condition

Loss of
Mass
(grams) 

Plain 

1.50

CMF-0.24 

1.15

PP-0.24 

1.05

CNT D-0.16 

0.90

CNT D-PAA+PP

0.92

CNT D-PAA+CMF

0.90

 

251

Table 9.5- Outcomes of Ridge analysis of the flexural attributes and impact resistance test data
for high performance concrete with nano- and/or micro-scale reinforcement.
(a) Flexural Strength
Ridge Analysis for Maximizing Flexural Strength
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response
Upper
Lower
CNT D-PAA
PP
0.000

15.604

15.024

16.183

0.080

0.120

0.100

15.847

15.264

16.429

0.088

0.123

0.200

16.093

15.502

16.684

0.095

0.127

0.300

16.344

15.739

16.949

0.103

0.131

0.400

16.600

15.976

17.225

0.110

0.136

0.500

16.861

16.212

17.511

0.117

0.141

0.600

17.128

16.449

17.807

0.125

0.147

0.700

17.400

16.686

18.115

0.132

0.152

0.800

17.678

16.924

18.433

0.139

0.158

0.900

17.963

17.164

18.762

0.146

0.165

1.000

18.253

17.404

19.103

0.152

0.171

252

(b) Energy Absorption Capacity 
Ridge Analysis for Maximizing Energy Absorption Capacity
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response
Upper
Lower
CNT D-PAA
PP
0.000

282.779

256.693

308.865

0.080

0.120

0.100

295.584

269.367

321.800

0.082

0.132

0.200

308.239

281.635

334.843

0.084

0.143

0.300

320.761

293.521

348.000

0.086

0.155

0.400

333.167

305.059

361.276

0.087

0.167

0.500

345.477

316.288

374.667

0.087

0.179

0.600

357.712

327.254

388.171

0.088

0.191

0.700

369.893

338.003

401.783

0.088

0.203

0.800

382.042

348.582

415.501

0.087

0.215

0.900

394.181

359.037

429.326

0.086

0.228

1.000

406.334

369.406

443.263

0.085

0.240

253

(c) Maximum Deflection 
Ridge Analysis for Maximizing Maximum Deflection
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response
Upper
Lower
CNT D-PAA
PP
0.000

3.101

2.793

3.408

0.080

0.120

0.100

3.296

2.987

3.605

0.082

0.132

0.200

3.492

3.178

3.805

0.083

0.143

0.300

3.687

3.366

4.008

0.085

0.155

0.400

3.882

3.551

4.213

0.086

0.167

0.500

4.076

3.733

4.420

0.088

0.179

0.600

4.271

3.912

4.630

0.089

0.191

0.700

4.465

4.089

4.841

0.091

0.202

0.800

4.659

4.264

5.054

0.092

0.214

0.900

4.853

4.437

5.268

0.093

0.226

1.000

5.046

4.608

5.484

0.094

0.238

254

(d) Impact Resistance 
Ridge Analysis for Maximizing Impact Resistance
Coded Radius Estimated 95.00%

Response

Confidence IntervalUncoded Factor Values

Upper

Lower

CNT D-PAA

PP

0.000

2.500

2.375

2.625

0.080

0.120

0.100

2.567

2.441

2.693

0.086

0.127

0.200

2.633

2.506

2.761

0.092

0.135

0.300

2.700

2.569

2.831

0.099

0.142

0.400

2.767

2.631

2.902

0.105

0.150

0.500

2.833

2.693

2.974

0.111

0.157

0.600

2.900

2.752

3.048

0.117

0.165

0.700

2.967

2.811

3.123

0.124

0.172

0.800

3.034

2.868

3.199

0.130

0.180

0.900

3.100

2.925

3.276

0.136

0.187

1.000

3.167

2.979

3.355

0.142

0.195

255

(a) Crack deflection around sand particle

(b) Dense structure of paste
Figure 9.1- SEM images of fractured surfaces of high performance concrete.
256

Figure 9.1 (cont’d)

(c) Pulled out nanotubes at fractured surface

(d) Pulled out nanotubes at fractured surface

257

Figure 9.1 (cont’d)

 
(e) Distribution of nanotubes within the matrix
 

(f) Pulled out carbon microfibers

258

Figure 9.1 (cont’d)

(g) Carbon nanotubes interacting with carbon microfibers 

259

CHAPTER 10

10 EVALUATION OF CARBON NANO AND/ OR MICRO-SCALE
REINFORCEMENTS ON THE ENGINEERING PROPERTIES OF
HIGH STRENGTH CONCRETE AND COMPARISON WITH HIGHPERFORMANCE CONCRETE
10.1 General
Most of the works involving use of graphite nanomaterials, especially multiwalled carbon
nanotubes, in cementitious matrices have used very fine matrices to evaluate the efficiency of
their reinforcement and reported modest gains in some of the engineering properties (Makar and
Beaudoin 2004; Cwirzen, Habermehl-Cwirzen et al. 2008; Cwirzen, Habermehl-Cwirzen et al.
2009; Makar and Chan 2009; Metaxa, Konsta-Gdoutos et al. 2009; Metaxa, Konsta-Gdoutos et
al. 2010; Li, Wang et al. 2005). A comprehensive approach was taken to evaluate functionalized
graphite nanomaterials in both high-strength and high-performance cementitious matrices of
higher complexity.
The focus of this research is on the use of high-strength concrete, normally cured as the matrix in
which the reinforcement efficiency of optimally functionalized multiwalled carbon nanotubes
and/ or microfibers is evaluated. The larger aggregate size and content in concrete, when
compared with mortar and paste, adds to complexity of behavior and failure modes by
introducing an interfacial zone, generating interactions between aggregates and propagating
microcracks, and the need to disperse nanomaterials in the space between aggregates. The results

260

are compared with same reinforcement conditions used in high-performance (DSP) concrete
matrix.

10.2 Materials and Methods
10.2.1 Graphite Nanomaterials and Microfibers
Polyacrylic acid physisorbed multiwalled carbon nanotubes (CNTD-PAA) and carbon
microfibers (CMF) at different volume fractions were used in a high-strength, normally cured,
concrete matrix. These graphite nanomaterials and microfibers have been introduced in detail in
Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. In the course of dispersing the nanotubes in water
PAA was physisorbed upon multiwalled carbon nanotubes.
10.2.2 Cementitious Matrices, Mixing and Curing Procedure
High-Performance (DSP) concrete matrix has been described in detailed in Chapter 2, Section
2.5.1. For this experimental regimen, a more practical high-strength concrete matrix was selected
with a targeted compressive strength of 15,000 psi (104 MPa). Most of the materials used are the
same as were used for high-performance matrix but with different proportions. Table 10.1 gives
the composition of both high-performance (DSP) and high-strength concrete matrix. The mixes
included a plain without any reinforcement, one with CNT D-PAA, one with CNT D-PAA and
CMF; and with CMF microfibers alone. CNT D-PAA were used at 0.16 vol.% (volume percent
of the dry cementitious materials or binder (cement plus silica fume)) and CMF were used at
0.24 vol.% of binder both individually and when used as hybrid reinforcement.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication, described in detail in Chapter 2, Section 2.5.2) and/ or different
microfibers were prepared and cured following ASTM procedures described in the same section.

261

For high-strength concrete the samples were cured in fresh water for 9 days. At least two batches
were casted with at least four specimens for each test condition in each batch for all
reinforcement conditions and engineering properties.
10.2.3 Experimental Methods
The test procedures employed to determine the engineering properties of this high-strength
concrete matrix included compression tests (ASTM C 109), flexure tests (ASTM C 1185) and
impact tests (ASTM D 7136) and Abrasion tests (ASTM C 944). Experimental results were
evaluated using the analysis of variance (ANOVA) and pair-wise comparison techniques.

10.3 Experimental Results and Discussion
Poly-acrylic acid physisorbed multiwalled carbon nanotubes Type D (CNT D-PAA) were used
as reinforcement at 0.16 vol.% and carbon microfibers(CMF) were used at 0.24 vol.%.
Combination of carbon microfibers with multiwalled carbon nanotubes were also used as hybrid
reinforcement for concrete, with CMF and CNT D-PAA used at 0.24 vol.% and 0.16 vol.%,
respectively.
10.3.1 Flexural Performance
The flexural attributes test results for the both high-strength and high-performance concrete
reinforced with different volume fractions of multiwalled carbon nanotubes and/or carbon
microfibers is summarized in Table 10.2. These test results indicate that all reinforcement
systems produced gains in all flexural attributes considered here, for the high-strength concrete
matrix considered here. Due to relatively lesser brittle nature, the plain high-strength matrix
showed some increase in both maximum (25%) deflection and energy absorption capacity (17%)
but a significant (30.6%) drop in flexural strength with respect to plain DSP concrete matrix.

262

For high-strength concrete, the most desired balance of properties for a single reinforcement was
realized by the addition of 0.16 vol.% of CNT D-PAA. The corresponding improvements in
flexural strength, energy absorption capacity and maximum deflection with respect to plain
concrete were 24.7%, 17% and 53.7%, respectively. Test results also indicate that the addition
of microfibers alone at 0.24 vol.% to this high-strength concrete produced important gains in all
flexural attributes considered here. The most significant gain was realized for maximum
deflection (51.7%) with respect to plain high-strength matrix.
As was the case for high-performance concrete matrix, the test data generated in the past pointed
at the positive effects of hybrid (nano- and micro-scale) reinforcement systems in high-strength
concrete matrix. Carbon microfibers (CMF), when used together with CNT D-PAA as hybrid
reinforcement, produced balanced gains in the flexural performance characteristics of highstrength concrete by interacting with and arresting cracks which developed in matrix at different
scales. Overall, the best balance of flexural attributes was produced by the hybrid reinforcement
comprising CNT D-PAA (0.16 vol.%) and CMF (0.24 vol.%), which produced a 38.7%, 59.6%
and 41.7% rise in flexural strength, maximum deflection and energy absorption capacity,
respectively. The synergistic reinforcement actions of nano- and micro-scale reinforcement
systems point at the merits of multi-scale reinforcement in high-strength concrete. Although the
reinforcement efficiency of CNT D-PAA and/ or CMF in high-strength concrete is lower than in
high-performance DSP concrete matrix. Still they point to the superior reinforcement efficiency
of both nanomaterials and microfibers, especially when used in hybrid.
The experimental results were subjected to statistical analysis of variance (ANOVA). Due to the
large number of observations, ANOVA only gives the general trends in test results. The ANOVA
outcomes indicate that there are statistically significant (at 0.05 significance level) improvements

263

in all flexural attributes of high-strength concrete due to the addition of nano-scale and hybrid
reinforcement systems. Pair-wise comparisons were carried out in order to assess the statistical
significance of the effects associated with the addition of nano- and/or micro-scale reinforcement
systems. These results indicate that nano-scale reinforcement systems at 0.16 vol.% produced
statistically significant improvements in all flexural attributes of the plain matrix at 0.05
significance level when compared with plain concrete. Use of high-modulus carbon microfiber in
conjunction with CNT D-PAA produced balanced gains in the engineering properties of highstrength concrete at significance levels varying from 0.000 to 0.047.
10.3.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-strength
concretes with nano- and/or micro-scale reinforcement systems and compared with highperformance DSP concrete, are presented in Table 10.3. As was the case with DSP paste and
mortar and concrete, PAA physisorbed multiwalled carbon nanotubes as well as its hybrid with
CMF microfibers produced small increase in the mean value of compressive strength. However,
CMF when used alone showed a small drop in compressive strength.
Outcomes of pair-wise comparisons also indicate the effects of nano- and/ or micro-scale
reinforcement systems on compressive strength are not statistically significant (at 0.05
significance level). The use of hybrid reinforcement systems comprising of CNT D-PAA and
CMF restored the compressive strengths of high-strength concrete by inducing interactions with
matrix and its cracks at different scales. This effect was also evident in the results of highperformance DSP concrete.

264

10.3.3 Impact Resistance
The impact test data are summarized in Table 10.4. All nano- and/or micro-scale reinforcement
systems produced improvements in the impact resistance of high-strength concrete. The
maximum rise in impact resistance realized with a single reinforcement (87%) was for 0.16
vol.% of CNT D-PAA. In the case of micro-scale fibers alone, CMF produced a gain 34.8% in
the impact resistance of high-performance (DSP) concrete.
The hybrid nano- and micro-scale reinforcement systems produced further rise in the impact
resistance (104%) of high-strength concrete when compared with individual (nano-or microscale) reinforcement used alone. A significant (0.000 significance level) increase in impact
resistance was observed for the hybrid reinforcement comprising CMF with CNT D-PAA, which
confirms the effectiveness of hybrid reinforcement.
Outcomes of ANOVA and pair-wise comparisons point at the statistical significant (0.000 to
0.012 significance levels) of nano- and/ or micro-scale reinforcement effects on the impact
resistance of high-strength concrete. When compared with high-performance concrete (with
nano-scale reinforcement), high-strength concrete (with nano-scale reinforcement) provides
lower levels of impact resistance; this observation points at the weakening of matrix due to
higher volume fraction of coarse aggregates in the high-strength.
10.3.4 Abrasion Resistance
The abrasion test results produced for high-strength concrete with different volume fractions of
graphite nanomaterials and/ or microfiber reinforcement and its comparison with highperformance concrete are summarized in Table 10.5.

265

Overall high-strength concrete matrix showed lower value of abrasion resistance both plain and
with all reinforcement conditions when compared with high-performance DSP concrete matrix.
As was the case for high-performance concrete, for a single reinforcement, CNT D-PAA at 0.16
vol.% produced the greatest improvement (35%) in the abrasion resistance of high-strength
concrete. All nano- and/or micro-scale reinforcement systems, produced gains in the abrasion
resistance of this high-strength concrete matrix, which were statistically significant at 0.05
significance level. Hybrid reinforcement of CNT D-PAA and CMF produced the maximum gain
of 37% in the abrasion resistance of high-performance concrete matrix. Outcomes of pair-wise
comparisons confirmed that each of the reinforcement conditions considered here produced
statistically significant improvements in the abrasion resistance of high-performance concrete
(with significance levels ranging from 0.006 to 0.000).

10.4 Summary
The experimental results generated for high strength concrete reinforced with PAA physisorbed
multiwalled carbon nanotubes (CNT D-PAA), carbon microfibers (CMF) and their hybrids
casted at optimum volume fractions and with carboxyl-based superplasticizers indicated that
proper use of nano-scale reinforcement can produce improvements in high-strength concrete
mechanical properties which surpass those realized with micro-scale reinforcement. However,
the level of reinforcement efficiency is reduced as we move down from a high-performance to a
high-strength concrete matrix. This is obvious due to normally curing of concrete versus steam
curing of high-performance concrete matrix which makes it denser thus the interaction of the
matrix at with the nano-reinforcement is more for high-performance matrices.

266

Table 10.1- Starting material selections and mix proportions (weight ratios) of high-performance
and high-strength concrete matrix.
Ingredient

DSP Concrete

Silica Fume/Binder
Water/Binder
Superplasticizer/Binder
Silica Sand (0 – 0.18 mm) / Binder
Silica Sand (0.18 -0.6 mm) / Binder
Granite Gravel (1 mm – 9 mm)/
Binder

0.20
0.20
Adjusted for diff
reinforcement
0.36
0.86
0.5

High Strength
Concrete
0.20
0.35
Adjusted for diff
reinforcement
0.36
0.86
1.1

Table 10.2- Mean values of flexural attributes of high-strength concrete with and without nanoand/ or micro-scale reinforcement systems and comparison with high-performance concrete.
Reinforcement Condition

Flexural Strength
(MPa)

Plain DSP
Plain HSC
CMF-0.24-DSP
CMF-0.24-HSC
CNT D-PAA-0.16-DSP
CNT D-PAA-0.16-HSC
CNT D-PAA+CMF-DSP
CNT D-PAA+CMF-HSC

13.4
9.3
12.8
10.9
16.5
11.6
19.0
12.9

Deflection
(mm)
0.71
0.89
1.01
1.35
1.58
1.06
1.78
1.42

Energy
Absorption
(N.mm)
92
108
111
129
231
166
316
153

Table 10.3- Compressive strength values of high-strength concrete with and without nano- and/
or micro-scale reinforcement systems and comparison with high-performance concrete.
Reinforcement Condition
Plain DSP
Plain HSC
CMF-0.24-DSP
CMF-0.24-HSC
CNT D-PAA-0.16-DSP
CNT D-PAA-0.16-HSC
CNT D-PAA+CMF-DSP
CNT D-PAA+CMF-HSC

Mean
Compressive
Strength (MPa)
151
112
139
108
143
118
150
113

267

Standard
Error (MPa)
7.75
7.60
13.5
6.95
5.55
8.26
7.53
5.95

Table 10.4- Impact resistance test results for high-strength concretes with and without nanoand/ or micro-scale reinforcement systems and comparison with high-performance concrete.
Reinforcement Condition
Plain DSP
Plain HSC
CMF-0.24-DSP
CMF-0.24-HSC
CNT D-PAA-0.16-DSP
CNT D-PAA-0.16-HSC
CNT D-PAA+CMF-DSP
CNT D-PAA+CMF-HSC

Impact Resistance
(mm/mm)
1.50
1.15
1.98
1.55
2.60
2.15
2.65
2.35

Table 10.5- Mean abrasion weight losses of high-strength concretes with and without nano- and/
or micro-scale reinforcement systems and comparison with high-performance concrete.
Reinforcement Condition
Plain DSP
Plain HSC
CMF-0.24-DSP
CMF-0.24-HSC
CNT D-PAA-0.16-DSP
CNT D-PAA-0.16-HSC
CNT D-PAA+CMF-DSP
CNT D-PAA+CMF-HSC

268

Loss of Mass
(grams)
1.50
3.75
1.15
3.45
0.90
2.43
0.90
2.35

CHAPTER 11

11 EVALUATION OF REINFORCEMENT EFFICIENCY OF CARBON
NANO
AND
MICRO-FIBERS
IN
HIGH-PERFORMANCE
CEMENTITIOUS PASTE
11.1 General
Cementitious materials are generally characterized as brittle materials with relatively low tensile
strength and strain capacity, which are susceptible to early-age cracking due to restraint of
shrinkage movements. These problems tend to be more pronounced in the case of highperformance cementitious materials with high compressive strengths. Fibers (with micro-scale
diameters) are incorporated into cementitious matrices to overcome these drawbacks. The microscale diameters and relatively low aspect (length-to-diameter) ratios of these fibers, and their
viable volume fractions in cementitious materials, however, yield relatively large fiber-to-fiber
spacing and relatively small specific surface area of fibers. These features limit the effectiveness
of micro-scale fibers in controlling the flaw (microcrack) size and propagation in cementitious
materials. Slender nanomaterials (e.g. CNFs) with nano-scale diameters and relatively high
aspect ratios offer the potential to overcome these drawbacks of micro-scale fibers in
cementitious materials. The relatively close spacing and high specific surface area of
nanomaterials provide them with the potential to effectively control the flaw size and
propagation in cementitious materials. The close spacing of nanomaterials could also enhance
the moisture barrier qualities and thus the durability characteristics of cementitious materials by

269

forcing tortuous diffusion paths of moisture (and aggressive solutions) into cementitious
materials.
Previous research on high-performance cementitious materials reinforced with multiwall carbon
nanotubes (MWNT) has shown that, as far as the uniform dispersion and effective interfacial
interactions of nanotubes within cementitious paste are achieved, important gains in diverse
engineering properties of cementitious materials can be realized at nanotube volume fractions
below 0.1% of the cementitious paste. The work reported herein examines the potential of carbon
nanofibers (CNFs), which are of substantially lower cost when compared with carbon nanotubes,
to bring about balanced gains in engineering properties of high-performance cementitious
materials.
CNFs have a unique hybrid grapheme sheet structure based on conical and tubular elements.
Their diameters range from 50 to 200 nm (which are greater than typical nanotube diameters).
CNFs provide high mechanical properties (e.g., 25 to 200 GPa elastic modulus depending on
their diameter) (Cwirzen and Penttala 2005), specific surface areas and aspect ratios, which are
below those of carbon nanotubes. Their electrical and thermal conductivity as well as thermal
and chemical stability are also relatively high. When compared to carbon nanotubes, CNFs
provide more active sites on their surfaces for bonding to cementitious materials. The desired
combination of performance and cost of CNFs suits applications in the field of nanocomposites.
Past research efforts have primarily focused on development of carbon nanofiber–polymer
nanocomposites (Ping, Beaudoin et al. 1991; Song, Jang et al. 2007; Pfeifer, Moeser et al. 2009).
Limited work has been conducted on the use of CNFs in cementitious matrices. The prerequisites
for successful use of CNFs in cementitious materials include thorough dispersion and effective
interfacial interactions of CNFs within the cementitious matrix (Ahlborn, Harris et al. 2011) .

270

Past research efforts have examined the effects of 0.005%, 0.02%, 0.05%, 0.5% and 2% CNF by
weight of cementitious materials on the microstructure and mechanical properties of a
cementitious matrix with 10% of cement replaced with silica fume (Bentz 2000; Bentz, Jensen et
al. 2000). The effects of surface treatment of CNF with nitric acid on their reinforcement
efficiency (at 0.5% by weight of cementitious materials) were also evaluated. It was found that
the use of silica fume and surface treatment of CNF facilitated their dispersion, and improved
their interfacial interactions with the cementitious paste. The gains in engineering properties
(splitting tensile strength and compressive strength) of cementitious matrix with introduction of
CNFs were, however, insignificant. Other investigations involving microscopic observations
have indicated that CNFs can control cracking in cementitious materials by bridging across fine
cracks, yielding improvements in flexural strength of cementitious materials at weight fractions
of about 0.048% (Metaxa, Konsta-Gdoutos et al. 2010).
In this study, the effects of surface functionalization of carbon nanofibers (CNFs) through
introduction of carboxyl groups and use of refined dispersion techniques on the reinforcement
efficiency of nanotubes at different volume fractions within a high-performance cementitious
(DSP) matrix were evaluated. Scanning electron microscope observations of fractured surfaces
pointed at successful dispersion of nanofibers within the cement-based matrix. An investigation
of hybrid reinforcement systems comprising CNFs and micro-scale fibers (CMFs) indicated that
selected hybrid systems offer a cost-effective approach to realizing balanced gains in
performance characteristics of high-performance cementitious matrices.

11.2 Materials and Methods

271

11.2.1 Graphite Nanomaterials and Microfibers
Different types of acid-functionalized and non-functionalized CNFs as well as chopped CMFs
were evaluated for use in cement-based nanocomposites. These graphite nanomaterials and
microfibers have been introduced in detail in Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3.
Acid-functionalized and non-functionalized carbon nanofibers (CNF-OX and CNF) and carbon
microfibers (CMF) were used in high-performance cement-based paste matrix.
11.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
paste was used to evaluate the reinforcement efficiency of both carbon nanofibers and/or carbon
microfibers at different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon microfibers were prepared and cured following
ASTM procedures described in the same section.
11.2.3 Experimental Methods
The test procedures employed to determine the engineering properties of high-performance paste
included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests (ASTM D
7136) and Abrasion tests (ASTM C 944). SEM and EDS were also employed to gain further
insight into the structure and failure mechanisms of cement-based nanocomposites. The
experimental results were subjected to statistical analysis using the analysis of variance
(ANOVA) and pair-wise comparison methods.

272

11.3 Experimental Results and Discussion
11.3.1 Effects of Carbon Nanofiber Surface Functionalization on Engineering Properties
11.3.1.1 Flexural Performance
Typical flexural load-deflection curves for cementitious (DSP paste) matrices reinforced with
0% and 0.24 vol.% of acid-functionalized and non-functionalized nanofibers are shown in Fig.
11.1. Volume fraction is calculated here with respect to the volume of cement and silica fume.
The mean values of flexural strength, maximum deflection and energy absorption capacity (area
underneath the flexural load-deflection curve) are summarized in Table 11.1. The mean values
and standard deviations of the flexural strength and energy absorption capacity test results are
presented in Figs. 11.2a and 11.2b, respectively.
Acid-oxidized nanofibers produced significant gains in the flexural strength and especially
energy absorption capacity of the cementitious matrix. The improvements in flexural strength
and energy absorption capacity were 52% and 296%, respectively. The relatively large length of
CNFs benefits their contributions to energy absorption capacity via extended frictional pullout at
cracks. The contributions of oxidized CNFs to the flexural energy absorption capacity and
strength of the high-performance cementitious paste were confirmed through statistical analysis
(ANOVA) of test results at 0.001 significance level. Non-oxidized nanofibers produced a drop in
flexural strength, although they contributed to the flexural energy absorption capacity and
maximum deflection of the cementitious matrix. The drop in flexural strength can be attributed
to the potential for clumping of non-functionalized nanofibers in the cementitious matrix. This
observation was confirmed by SEM observations, Fig 11.3. The contributions of CNFs to the

273

flexural performance characteristics of high-performance cementitious materials are attractive,
given their relatively low cost when compared with other nanomaterials.
11.3.1.2 Compressive Strength
The compressive strength test results (means and standard deviations) are summarized in Fig.
11.4. The compressive strength of the high-performance cementitious matrix is observed to be
either preserved or improved with the addition of nanofibers. Analysis of variance of test results
indicated that, at 0.05 significance level, the high compressive strength of the cementitious
matrix was preserved after introduction of nanofibers.
11.3.1.3 Impact Resistance
Examples of failed impact test specimens are shown in Fig. 11.5. The mean values of impact test
results are presented in Fig. 11.6. The impact resistance of cementitious materials is observed to
increases with introduction of 0.24 vol.% of both non-functionalized and acid-functionalized
nanofibers. The test results suggest that functionalization of nanofibers benefits their
contributions to the impact resistance of the cement-based matrix. The functionalized and nonfunctionalized nanofibers, at 0.24 vol.%, yielded about 73.6% and 44.7% gains, respectively, in
impact resistance, which can be attributed to the arrest and deflection of microcracks, pullout of
nanofibers, and multiple cracking phenomena in the presence of nanofibers. Cementitious
materials reinforced with oxidized nanofibers exhibited more pronounced multiple cracking as
compared to those with pristine nanofibers (see Fig. 11.5).
11.3.1.4 Abrasion Resistance
Examples of abrasion test specimens after performance of abrasion tests are shown in Fig. 11.7.
The abrasion test results presented in Fig. 11.8 indicate that cementitious materials with oxidized

274

and pristine nanofibers, 0.24 vol.% showed superior abrasion resistance when compared when
compared with the plain cementitious matrix. The significant increase in abrasion resistance with
oxidized nanofibers can be attributed to their thorough dispersion and effective interfacial
interactions in the high-performance cementitious matrix. Analysis of variance of test results
indicated that the contributions of nanofibers to the abrasion resistance of high-performance
cementitious matrix were statistically significant at 0.05 significance level.
11.3.1.5 Scanning Electron Microscope Evaluation
The SEM images presented in Fig. 11.9 indicate that the refined sonication and mixing steps
developed in the project were successful in dispersion of individual oxidized nanofibers within
the cementitious matrix at 0.24 vol.%. These images also point at the low porosity of the highperformance cementitious matrix used here; the introduction of nanofibers does not seem to have
compromised the high density (low porosity) of this matrix. SEM observations verified the
uniform dispersion of oxidized nanofibers; evidences of crack suppression and crack bridging
were also found in all specimens examined under SEM. Oxidized nanofibers, which provided
higher levels of reinforcement efficiency, were found to be more uniformly dispersed within the
cementitious matrix, as some clumping was found for pristine nanofibers.
11.3.1.6 Energy Dispersive Spectroscopy (EDS)
Elemental and Quantitative analyses of these samples indicated that all cementitious materials
exhibit similar elemental compositions, except for the (expected) higher carbon content in
materials incorporating CNFs.

275

11.3.2 Effects of Hybrid (Nano- and Micro-scale) Reinforcement Systems
Hybrid reinforcement systems comprising oxidized carbon nanofibers (CNF-OX) and carbon
fibers with micro-scale diameter (CMF) were evaluated in DSP paste. Exploratory studies of
carbon nanofiber and carbon microfiber reinforcement indicated that, for the cementitious matrix
and processing conditions considered here, the nanofiber volume fraction (of cement plus silica
fume) should be lowered below 0.24 vol.% before benefits can be realized from introduction of
carbon microfibers. Hence, a nanofiber volume that was 0.16 vol.% that of cement plus silica
fume was considered for development of hybrid reinforcement systems, which also incorporated
carbon microfiber at 0.24 vol.% of cement plus silica fume.
11.3.2.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity test results for the
high-performance (DSP) cementitious paste reinforced with different volume fractions of CNFs
and CMFs are shown in Table 11.2. CNFs and CMFs, and especially their combination, are
observed to yield important gains in the flexural strength, energy absorption capacity and
maximum deflection of DSP cementitious paste. The hybrid (carbon nanofiber and microfiber)
reinforcement system produced 64.9%, 356% and 288% gains in the flexural strength, energy
absorption capacity and maximum deflection, respectively, of DSP paste. These benefits could
be attributed to the multi-scale reinforcing action of the hybrid reinforcement system. ANOVA
and pair-wise comparison of flexure test data confirm, at 0.00 significance level, that the
contributions of nano- and/or micro-scale reinforcement effects to the flexural performance of
this high-performance cementitious paste are statistically significant.

276

11.3.2.2 Compressive Strength
While micro-fiber reinforcement tends to lower the compressive strength of concrete, mainly due
to the introduction of interfacial stress rise and structural defects, nano-scale (and hybrid)
reinforcement systems are observed in Table 11.3 to produce a slight rise in the compressive
strength of the high-performance cementitious paste. This effect, however, is not statistically
significant. The somewhat positive effects of nano-scale reinforcement on compressive strength
could be attributed to: (i) the reduced interfacial stress concentration and structural defects due to
the nano-scale dimensions and spacing of the reinforcement; and (ii) effective arrest of
microcrack propagation by the closely space nanofibers.
11.3.2.3 Impact Resistance
The impact resistance test results presented in Table 11.4 indicate that carbon microfibers, CNFs
and the hybrid reinforcement systems enhance the impact resistance of DSP paste by 42.7%,
71.2% and 92.9%, respectively, which can be attributed to the arrest and deflection of
microcracks by nanofibers, frictional pullout of nanofibers, the multiple cracking induced by
nanofibers, and control of microcrack size by the closely spaced nanofibers. The particularly
significant contribution of the hybrid reinforcement to impact resistance could be attributed to
the multi-scale interaction of the nano- and micro-scale reinforcement within the matrix with
microcracks. The contribution of nano- and/or micro-scale reinforcement to the impact resistance
of the high-performance cementitious paste was found to be statistically significant (at 0.01
significance level).

277

11.3.2.4 Abrasion Resistance
The abrasion test results presented in Table 11.5 indicate that carbon microfibers and especially
hybrid reinforcement, and (to a lesser extent) CNFs enhance the abrasion resistance of the highperformance (DSP) cementitious paste. These benefits were found to be statistically significant
at 0.000 significance level.
11.3.2.5 Scanning Electron Microscope Evaluation
Fig. 11.11 shows SEM images which are indicative of the uniform dispersion of nanofibers
within the cementitious paste (Fig. 11.10a), and a combination of carbon fiber rupture and
pullout at fracture surfaces (Fig. 11.10b & c).

11.4 Summary
Results of experiments conducted on cementitious pastes reinforced with non-functionalized and
functionalized CNFs as well as their hybrids with carbon microfibers yielded that acidoxidization of CNFs benefited their dispersion and reinforcement efficiency in a highperformance cementitious matrix. A hybrid reinforcement comprising CNFs and CMFs at 0.16%
and 0.24% of the volume of cementitious materials produced distinct gains in the engineering
properties of the high-performance cementitious paste, which could not be matched by nano- or
micro-scale reinforcement used alone. This finding points at the synergistic effects offered by
nano- and micro-scale reinforcement systems.

278

Table 11.1- Mean values of the flexural strength, maximum deflection and energy absorption
capacity test results for high-performance cementitious pastes with 0% and 0.24% (by volume of
cement plus silica fume) of non-functionalized and acid-functionalized CNFs.
Reinforcement Condition

Flexural
Strength
(MPa)

Maximum
Deflection
(mm)

Energy
Absorption
(N.mm)

Plain

10.3

0.40

62.0

Acid-functionalized nanofiber (CNF-OX), 0.24 vol.%

15.7

1.18

245

Non-functionalized nanofiber (CNF), 0.24 vol.%

9.09

0.79

102

Table 11.2- Mean values of the flexural strength, maximum deflection and energy absorption
capacity test results for high-performance cementitious paste with carbon nanofiber and/or
microfiber reinforcement systems.
Reinforcement Condition

Flexural
Strength
(MPa)

Maximum
Deflection
(mm)

Energy
Absorption
(N.mm)

Plain

10.3

0.40

61.8

Carbon microfiber, 0.24 vol%

16.1

1.13

251

Carbon nanofiber, 0.16 vol%

14.9

1.05

213

Carbon microfiber, 0.24 vol%, &
Carbon nanofiber, 0.16 vol%

17.0

1.55

282

Table 11.3- Mean values and standard errors of the compressive strength test results for highperformance cementitious paste reinforced with CNFs and/or microfibers.
Reinforcement Condition

Mean
(MPa)

Standard Error
(MPa)

Plain

111

14.3

Carbon microfiber, 0.24 vol%

117

28.0

Carbon nanofiber, 0.16 vol%

113

17.9

Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16
vol%

117

15.3

279

Table 11.4- Mean values of the impact resistance test results for high-performance cementitious
pastes with carbon nanofiber and/or microfiber reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol%
Carbon nanofiber, 0.16 vol%
Carbon microfiber, 0.24 vol% & Carbon nanofiber, 0.16 vol%

Impact Resistance
(mm/mm)
0.365
0.521
0.625
0.704

Table 11.5- Mean values and standard errors of the abrasion test results for high-performance
cementitious pastes with carbon nanofiber and/or microfiber reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol%
Carbon nanofiber, 0.16 vol%
Carbon microfiber, 0.24 vol% & carbon nanofiber, 0.16 vol%

280

Loss of Mass
(grams)
0.275
0.150
0.250
0.125

Figure 11.1- Representative flexural load-deflection curves of high-performance cementitious
pastes with 0 and 0.24 vol.% of non-functionalized and oxidized nanofibers.

18
Flexural Strength, MPa

16
14
DSP Paste
Plain
CNF-0.24 vol.%

12
10
8

CNF-OX-0.24
vol.%

6
4
2
0
Reinforcement Type
(a) Flexural strength

Figure 11.2- Flexural attributes of high-performance cementitious pastes.
281

Figure 11.2 (cont’d)

Energy Absorption Capacity,
(N.mm)

350
300
250

DSP Paste
Plain
CNF-0.24 vol.%

200
150

CNF-OX-0.24
vol.%

100
50
0

Reinforcement Type
(b) Energy absorption capacity

Figure 11.3- Clumping of non-functionalized CNFs in the high-performance cementitious paste
at 0.24% by volume of cement plus silica fume.
282

Compressive Strength, MPa

140
120
100

DSP Paste Plain

80

CNF-OX-0.24
vol.%
CNF-0.24 vol.%

60
40
20
0
Reinforcement Type

Figure 11.4- Mean values and standard errors of the compressive strength test results for highperformance cementitious paste reinforced with 0 and 0.24 vol.% of non-functionalized and
acid-functionalized CNFs.

Figure 11.5- Failed impact test specimens.
283

Impact Strength (mm/ mm)

0.7
0.6
DSP Paste
Plain

0.5
0.4
0.3

CNF-0.24
vol%

0.2

CNF-OX-0.24
vol%

0.1
0
Reinforcement Type

Figure 11.6- Mean values and standard errors of the impact resistance test results for highperformance cementitious paste reinforced with 0 and 0.24 vol.% of non-functionalized and
acid-functionalized CNFs.

Figure 11.7- Abrasion test specimens after test.

284

Total Weight Loss (grams)

0.3
DSP Paste
Plain
0.2

CNF-0.24
vol.%

0.1

CNF-OX0.24 vol.%

0
Reinforcement Type
Figure 11.8- Mean values of abrasion weight loss for high-performance cementitious pastes
reinforced with 0% and 0.24% of non-functionalized and acid-functionalized CNFs.

(a) Uniform dispersion of nanofibers in matrix
Figure 11.9- Typical SEM images of high-performance pastes reinforced with 0.24 vol.% of
CNFs.
285

Figure 11.9 (cont’d)

(b) Micro- crack-bridging action of nanofibers

(c) Micro- crack-bridging action of nanofibers

286

Figure 11.9 (cont’d)

(d) Dense DSP matrix

287

(a) Dispersion of oxidized CNFs within the matrix

(b) Failure mechanism of micro-scale fibers at fracture surfaces
Figure 11.10- Typical SEM mages of DSP cementitious paste reinforced with carbon microfibers
and/or oxidized CNFs.
288

Figure 11.11 (cont’d)

(c) Failure mechanism of micro-scale fibers at fracture surfaces

289

CHAPTER 12

12 EVALUATION OF REINFORCEMENT EFFICIENCY OF CARBON
NANOAND
MICRO-FIBERS
IN
HIGH-PERFORMANCE
CEMENTITIOUS MORTAR
12.1 General
Fibers are incorporated into cementitious matrices to enhance their toughness, tensile strength
and ductility, and crack resistance. Cracks initiate in cementitious materials at fine scales, as a
result of local restraint of the past shrinkage; conventional (micro-scale) fibers are not effective
in controlling the development and size of such fine cracks. The relatively close spacing and high
surface area of nanofibers, on the other hand, could limit the concentration and size of such
“microcracks”, and their propagation in cementitious materials. Thorough dispersion and
effective interfacial interactions of nanofibers are critical to their effectiveness in cementitious
materials. The relatively short lengths of nanofibers, however, limit their ability to bridge across
wider cracks, and provide for substantial energy dissipation via extended frictional pullout.
Longer micro-scale fibers can complement the action of nanofibers at fine scale with crack
control and toughening effects at advanced stages of damage development. Earlier results on use
of nano-scale (primarily multi-walled carbon nanotubes) reinforcement of high-performance
cementitious materials have indicated that major gains in engineering properties of cementitious
materials can be realized with relative low dosages (as small as 0.04 vol.% by weight of cement).
One objective of the work reported herein was to assess the effectiveness of CNFs as relatively
low-cost nano-scale reinforcement in high-performance cementitious mortars. The other

290

objective was to validate the synergic action of a multi-scale reinforcement comprising CNFs
and microfibers as reinforcement in high-performance cementitious mortars.
Carbon nanofibers (CNFs) are made of grapheme sheets formed into cones, which are stacked to
assume a tubular geometry with 50-200 nm outer diameter. The elastic modulus of individual
nanofibers ranges from 25 to 200 GPa, depending on their diameter (Cwirzen and Penttala 2005).
Given their relatively low cost compared to carbon nanotubes, development of carbon nanofiber
reinforced polymer nanocomposites is actively pursued (Ping, Beaudoin et al. 1991; Song, Jang
et al. 2007; Pfeifer, Moeser et al. 2009). Efforts to assess the value of CNFs to inorganic
(including cementitious) matrices have also been initiated; uniform dispersion and effective
interfacial interactions are key to their reinforcement efficiency in cementitious materials
(Ahlborn, Harris et al. 2011). Past experimental evaluations of the reinforcement efficiency of
CNFs in cementitious matrices have yielded mixed results, which could be partly attributed to
the extent of dispersion and interfacial interactions of nanofibers in cementitious materials, and
also to the qualities (e.g., density, pore system characteristics and strength) of the cementitious
matrix. In a series of tests, the introduction of silica fume (with nano-scale particle size) to
cement (with micro-scale particle size) as well as acid treatment of nanofibers were found to
benefit dispersion of CNFs in cementitious matrices; the addition of nanofibers at 0.005% to 2%
by weight of cement, however, did not benefit the mechanical properties of cementitious
materials (Bentz 2000; Bentz, Jensen et al. 2000). Metaxa et al. also investigated the
nanostructure and the mechanical properties of cementitious nanocomposites reinforced with
CNF at low volume fractions. Another experimental work (Metaxa, Konsta-Gdoutos et al. 2010),
on the other hand, showed improvements in the flexural strength of cementitious materials

291

realized by introduction of CNFs at 0.048 wt.%. Scanning electron microscope images showed
that CNFs bridge across fine cracks, and control their propagation.
In this study, the gains in reinforcement efficiency of CNFs, alone or in combination with carbon
microfibers, resulting from oxidation of nanofibers and also the use of refined methods for
dispersion of nanofibers were investigated. A high-performance cementitious matrix (DSP) was
considered here, and different nanofiber volume fractions were evaluated. Two alternative
polycarboxylate-based superplasticizers were used for preparation of cementitious matrices. One
of these superplasticizers, which performed better in cementitious nanocomposites, was used
throughout the investigation.

12.2 Materials and Methods
12.2.1 Graphite Nanomaterials and Microfibers
Different types of functionalized and non-functionalized carbon nanofibers as well as chopped
(micro-scale) carbon fibers were evaluated for use in cement-based nanocomposites. These
graphite nanomaterials and microfibers have been introduced in detail in Chapter 2, Section 2.4,
Fig. 2.3 and Table 2.3. Acid-functionalized and polymer wrapped carbon nanofibers (CNF-OX
and CNF-PAA) and carbon microfibers (CMF) were used in high-performance cement-based
mortar matrix.
12.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
mortar was used to evaluate the reinforcement efficiency of both graphite nanomaterials and/or
carbon microfibers at different volume fractions.

292

Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon microfibers were prepared and cured following
ASTM procedures described in the same section.
12.2.3 Experimental Methods
The test procedures employed to determine the engineering properties of high-performance
mortar included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests
(ASTM D 7136) and Abrasion tests (ASTM C 944). SEM and EDS were also employed to gain
further insight into the structure and failure mechanisms of cement-based nanocomposites. The
experimental results were subjected to statistical analysis using the analysis of variance
(ANOVA) and pair-wise comparison methods. RSA (Response Surface Analysis) was used to
identify reinforcement volume fractions which yield optimum gains in various engineering
properties of the cementitious matrix.

12.3 Experimental Results and Discussion
12.3.1 Effects of Carbon Nano- and/or Micro-fibers Used at Different Volume Fraction
and with Different Superplasticizers
12.3.1.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity (area underneath the
flexural load-deflection curves) were determined for the high-performance (DSP) cementitious
mortar reinforced with different volume fractions of oxidized or PAA-physisorbed CNFs and/or
microfibers, prepared using two different carboxylate-based superplasticizers. The flexure test
data are summarized in Table 12.1. The experimental results indicate that all reinforcement
conditions considered here produced improvements in the flexural attributes of high-performance

293

cementitious mortar. Maximum improvements were realized with PAA-physisorbed carbon
nanofiber (CNF-PAA) and its hybrid combination with carbon microfibers. The maximum gain
in flexural strength versus plain mortar (21.4%) was achieved at 0.16 vol.% CNF-PAA, which
was 4.3% greater than that achieved with a similar volume fraction of CNFs used without
polymer wrapping. It should be noted that the volume fraction of CNFs is calculated here with
respect to the volume of dry cementitious materials (cement and silica fume). The maximum
improvements in energy absorption capacity and maximum deflection versus plain mortar
(62.2% and 52.9%, respectively) were realized with 0.24 vol.% CNF-PAA. The gains in these
flexural attributes were also higher with polymer-wrapped CNFs (CNF-PAA) when compared
with CNFs used without polymer wrapping. Hybrid reinforcement systems comprising both
nano-scale CNF-PAA and micro-scale carbon fibers brought about balanced improvements in
the engineering properties of high-performance cementitious mortars, which surpassed those
achieved with nano- or micro-scale reinforcement used alone. Polymer wrapping of CNFs
consistently improved the reinforcement efficiency of nanofibers in high-performance (DSP)
mortar. ADVA® Cast 575 polycarboxylate-based superplasticizer produced better results when
compared with Glenium® 7700.
Per unit volume fraction, carbon nanofiber was more effective than carbon microfiber in
enhancing the flexural performance characteristics of DSP cementitious mortar. Hybrid (nanoand micro-scale) reinforcement systems comprising 0.16 vol.% CNF-PAA and 0.24 vol.% CMF
produced highly desired, balanced gains in the flexural performance characteristics of highperformance DSP cementitious mortar at a relatively low total reinforcement volume fraction.
Analysis of variance of test results indicated that the reinforcement conditions considered here
produced statistically significant (at 0.05 significance level) improvements in all flexural

294

attributes of the high-performance cementitious mortar. Subsequent pair-wise comparisons of
test results indicated that all nano- and/or micro-scale reinforcement conditions produced
statistically significant gains (at 0.05 significance level) in the flexural performance
characteristics of the high performance (DSP) mortar. Distinct improvements (at 0.00
significance level) in all flexural attributes were realized with the addition of CNF-PAA and/or
CMF reinforcement.
12.3.1.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
cementitious mortars with nano- and/or micro-scale reinforcement systems are presented in
Table 12.2. PAA- physisorbed as well as oxidized or pristine CNFs, when used alone or in
combination with carbon microfibers produced statistically insignificant effects on the
compressive strength of the high-performance cementitious mortar (0.76 significance level).
Pair-wise comparisons indicated that the difference between no pair of compressive strength test
results was statistically significant (at 0.05 significance level).
12.3.1.3 Impact Resistance
The impact test data are summarized in Table 12.3. Nanofiber reinforcement at 0.24 vol.% as
well as hybrid reinforcement of DSP mortar produced more than 50% gain in impact resistance.
At equal volume fractions, nanofibers were more effective than microfibers in enhancing the
impact resistance of the high-performance cementitious mortar with both superplasticizers. The
hybrid (nano- and micro-scale) reinforcement was more effective than micro-scale reinforcement
alone in improving the impact resistance of DSP mortar. The analysis of variance and pair-wise
comparisons of impact test results indicated that the gains in impact resistance with nano- and/or

295

micro-scale reinforcement of DSP mortar were statistically significant (at 0.05 significance
level).
12.3.1.4 Abrasion Resistance
The abrasion test results for DSP mortars with different volume fractions of carbon nanofiber
and/or microfiber reinforcement are summarized in Table 12.4. With ADVA® Cast 575
superplasticizer, carbon nanofiber (0.24 vol.%) and its hybrid combination with carbon
microfiber produced the greatest improvement (34.7%) in the abrasion resistance of DSP mortar.
Nano- and/or micro-scale reinforcement of DSP mortar produced marked improvements in
abrasion resistance, which were statistically significant at 0.05 significance level. Outcomes of
pair-wise comparisons confirmed that each of the reinforcement conditions considered here
produced statistically significant gains in the abrasion resistance of the DSP cementitious mortar.
12.3.1.5 Scanning Electron Microscope Evaluation
The SEM images presented in Fig. 12.1 indicate that the refined sonication and mixing steps
developed in the project were successful in dispersion of individual nanofibers within the
cementitious matrix. These images also point at the low porosity of the high-performance
cement-based matrix; the introduction of nanofibers does not seem to have compromised the
high density (low porosity) of this matrix. SEM observations also provided evidence for the
crack suppression and crack bridging actions of the nanofibers in cementitious matrix. PAAphysisorbed nanofibers, which provided higher levels of reinforcement efficiency, were found to
produce a more uniform dispersion within the cementitious matrix, as some clumping was found
for pristine nanofibers, Fig. 12.2.

296

12.3.2 Energy-Dispersive Spectroscopy (EDS)
Elemental and Quantitative analysis of these samples indicated that all cementitious materials
exhibit similar elemental compositions, except for the (expected) higher carbon content in
materials incorporating carbon reinforcement. These results indicate that graphite nanomaterials
do not significantly alter the chemistry of cement hydration for the reinforcement conditions
considered here.
12.3.3 Response Surface Analysis of High-Performance Cementitious Mortar
The test data on high-performance cementitious mortar with nano- and/or micro-scale
reinforcement systems prepared with two different superplasticizers were subjected to response
surface analysis. The objective of this analysis was to identify optimum reinforcement conditions
which maximize: (i) benefits to specific engineering properties of the high-performance mortar;
and (ii) simultaneous benefits to several engineering properties.
Response surface analysis (RSA) of the test data was conducted considering the volume fractions
of nano- and micro-scale reinforcement systems as input variables. Flexural strength, energy
absorption capacity and maximum deflection, and impact resistance were used as response
variables. The RSA process started with evaluating the effects of input variables on response
variables, and was followed by desirability analyses considering means of response variables,
using both canonical and ridge analysis methods.
Response Surface analyses conducted for each of the flexural strength, energy absorption
capacity, maximum deflection and impact resistance test data indicated that an optimum
reinforcement system comprising 0.09-0.13 vol.% PAA-physisorbed carbon nanofiber and 0.210.23 vol.% carbon microfiber maximized subject properties. Optimal Responses for flexural

297

strength, energy absorption and maximum deflection were calculated at 21.6 MPa, 429 N.mm
and 2.61 mm mean values, respectively, with 95% confidence Interval of 20.4-22.8 MPa, 390467 N.mm and 1.67–3.56 mm, respectively. The optimal response for impact resistance was
calculated at 0.945 mm/mm mean value with 95% Confidence Interval of 0.767-1.12 mm/mm.
All stationary points were found to be saddle points, which do not give a maximum but a general
direction to follow in which the targeted properties can be maximized, Fig. 12.3. Outcomes of
ridge analysis can be used to adjust the reinforcement condition for achieving further gains in
engineering properties. The analyses for flexural attributes and impact resistance show that the
hybrid reinforcement systems considered here occur within the range of optimal reinforcement
conditions.
The test data were also subjected to desirability analyses in order to identify reinforcement
conditions which yield a desired balance of all the engineering properties considered here.
Desirability analyses were conducted using mean values obtained through both canonical and
ridge analyses. These outputs indicate that maximum values of all response variables can be
achieved with CNT-PAA/CMF volume fractions 0.13/0.22 or 0.15/0.26 vol.% (calculated with
respect to the volume of dry cementitious materials). These optimum reinforcement systems do
not differ significantly from those considered in this investigation.

12.4 Summary
Comprehensive experimental evaluations of high-performance cementitious mortar with nanoscale (pristine, polymer-wrapped or oxidized CNFs) and/or micro-scale (carbon microfiber)
reinforcement were conducted using two polycarboxylate-based superplasticizers. As was the
case for carbon nanotubes the experimental results indicated that for similar reinforcement
conditions, the gains in performance characteristics of the cementitious matrix depended upon
298

the specific polycarboxylate-based superplasticizer used. The gains in flexural performance
characteristics per unit volume of reinforcement were more significant with carbon nanofiber
when compared with carbon microfiber.
Similarly, the experimental results also pointed at the merits of hybrid (nano- and micro-scale)
reinforcement towards achieving balanced gains in the flexural performance characteristics and
impact resistance of high-performance cementitious materials. Response surface analysis of
experimental results indicated that optimum reinforcement conditions comprised both nano- and
micro-scale reinforcement systems. The reinforcement conditions considered here produced
minor (and statistically insignificant) changes in the compressive strength of high-performance
cementitious materials.

299

Table 12.1- Mean values of the flexural attributes of high-performance (DSP) cementitious
mortar with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Flexural Maximum
Energy
Strength Deflection Absorption
(MPa)
(mm)
(N.mm)
Plain – Gl
16.6
0.99
216
Plain - Adva
18.7
1.38
275
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl
19.3
1.37
319
Carbon microfiber, 0.24 vol%-(CMF-0.24) 19.6
1.42
344
Adva
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl
22.5
2.31
424
Carbon microfiber, 0.48 vol%-(CMF-0.48) 23.0
2.35
459
Adva
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl
23.1
2.54
474
Carbon microfiber, 0.96 vol%-(CMF-0.96) 23.9
2.55
481
Adva
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) 21.6
1.76
405
Gl
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) 21.9
1.87
430
Adva
Carbon nanofiber, 0.16 vol%-(CNF-PAA-0.16)
22.7
2.02
443
- Adva
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) 20.9
1.87
410
Gl
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) 21.2
1.95
435
Adva
Carbon nanofiber, 0.24 vol%-(CNF-PAA-0.24)
22.2
2.11
446
- Adva
Carbon microfiber, 0.24 vol%, & Carbon
20.4
2.13
375
nanofiber, 0.16 vol%-(CMF;CNF-OX) - Gl
Carbon microfiber, 0.24 vol%, & Carbon
20.9
2.17
423
nanofiber, 0.16 vol%-(CMF;CNF-OX) - Adva
Carbon microfiber, 0.24 vol%, & Carbon
21.4
2.18
425
nanofiber, 0.16 vol%-(CMF;CNF-PAA) - Adva

300

Table 12.2- Mean values and standard errors of the compressive strength test results for DSP
mortar with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Plain – Gl
Plain - Adva
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Adva
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Gl
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Adva
Carbon nanofiber, 0.16 vol%-(CNF-PAA-0.16) - Adva
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Gl
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Adva
Carbon nanofiber, 0.24 vol%-(CNF-PAA-0.24) - Adva
Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16
vol%-(CMF;CNF-OX) - Gl
Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16
vol%-(CMF;CNF-OX) - Adva
Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16
vol%-(CMF;CNF-PAA) - Adva

301

Mean
Compressive
Strength
(MPa)
125
129
119
121
138
137
127
128
129
129
131
122
126
127
132

Standard
Error
(MPa)

131

1.21

133

1.77

17.0
9.42
5.69
10.4
8.45
5.44
17.6
6.57
9.57
7.57
7.72
15.5
5.56
4.75
7.96

Table 12.3- Impact resistance test results for high-performance (DSP) cementitious mortar with
nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Impact Resistance
(mm/mm)
Plain – Gl
0.57
Plain - Adva
0.63
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl
0.78
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva
0.83
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl
0.83
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Adva
0.89
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl
0.94
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva
0.94
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Gl
0.89
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Adva
0.94
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Gl
0.94
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Adva
0.99
Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16 vol%0.94
(CMF;CNF-OX) - Gl
Carbon microfiber, 0.24 vol%, & Carbon nanofiber, 0.16 vol%0.99
(CMF;CNF-OX) - Adva
Table 12.4 - Mean abrasion weight losses of high-performance DSP cementitious mortars with
nano- and/or micro-scale reinforcement.
Reinforcement Condition
Plain – Gl
Plain - Adva
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Gl
Carbon microfiber, 0.24 vol%-(CMF-0.24) - Adva
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Gl
Carbon microfiber, 0.48 vol%-(CMF-0.48) - Adva
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Gl
Carbon microfiber, 0.96 vol%-(CMF-0.96) - Adva
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Gl
Carbon nanofiber, 0.16 vol%-(CNF-OX-0.16) - Adva
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Gl
Carbon nanofiber, 0.24 vol%-(CNF-OX-0.24) - Adva
Carbon microfiber, 0.24 vol%, & Carbon nanofiber,
(CMF;CNF-OX) - Gl
Carbon microfiber, 0.24 vol%, & Carbon nanofiber,
(CMF;CNF-OX) - Adva

302

0.16 vol%-

Loss of
Mass
(grams)
1.38
1.23
0.98
0.93
0.93
0.92
0.85
0.83
0.88
0.85
0.83
0.80
0.88

0.16 vol%-

0.80

(a) Uniform distribution of nanofibers in the matrix

(b) Pulled out nanofibers from matrix
Figure 12.1- Typical SEM images of high-performance cementitious mortar reinforced with
different contents of polymer-wrapped or acid- functionalized CNFs.
303

Figure 12.1 (cont’d)

(c) DSP mortar matrix with embedded sand grain

304

(a)

(b)
Figure 12.2- Typical SEM images of DSP cementitious materials reinforced with CNFs and/or
CMFs.
305

CHAPTER 13

13 EVALUATION OF REINFORCEMENT EFFICIENCY OF CARBON
NANOAND
MICRO-FIBERS
IN
HIGH-PERFORMANCE
CEMENTITIOUS CONCRETE
13.1 General
Advanced technological aspects of cement based materials have recently focused on developing
high-performance cementitious composites, which exhibit high compressive strengths. Such
composites however, exhibit also extremely brittle failure, low tensile capacity and appear
sensitive to early age microcracking as a result of volumetric changes due to high autogenous
shrinkage stresses. These characteristics of cement based materials are serious shortcomings that
not only impose constrains in structural design, but also affect the long term durability of
structures. To overcome the aforementioned disadvantages reinforcement of cementitious
materials is typically provided at the millimeter and/or the micro scale using macro-fibers and
microfibers, respectively. Cementitious matrices however, exhibit flaws at the nano-scale, where
traditional reinforcement is not effective. Graphite Nanomaterials, including carbon nanofibers
(CNFs),

present

several

distinct

advantages

as

a

reinforcing

material

for

high

strength/performance cementitious composites as compared to more traditional fibers. First, they
exhibit significant greater strength and stiffness than conventional fibers, which should improve
overall mechanical behavior. Second, their higher aspect ratio is expected to effectively arrest the
nano-cracks and demand significantly higher energy for crack propagation. Thirdly, provided
that CNFs are uniformly dispersed, and due to their nano-scale diameter, fiber spacing is

306

reduced. This has opened a new field for nano-sized reinforcements that should theoretically
hinder the formation and later propagation of microcracks at the very beginning. Few attempts
have been made to add different graphite nanomaterials as reinforcement in cementitious
matrices.
Most of the works involving use of graphite nanomaterials in cementitious matrices have used
very fine matrices to evaluate the efficiency of their reinforcement and reported modest gains in
some of the engineering properties (Makar and Beaudoin 2004; Cwirzen, Habermehl-Cwirzen et
al. 2008; Cwirzen, Habermehl-Cwirzen et al. 2009; Makar and Chan 2009; Metaxa, KonstaGdoutos et al. 2009; Metaxa, Konsta-Gdoutos et al. 2010; Li, Wang et al. 2005). A
comprehensive approach was taken to evaluate suitably functionalized CNFs in highperformance concrete of higher complexity.
The focus of this research paper is on the use of high-performance (DSP) concrete as the matrix
in which the reinforcement efficiency of suitably functionalized CNFs and/ or different
microfibers is evaluated. The larger aggregate size and content in concrete, when compared with
mortar and paste, adds to complexity of behavior and failure modes by introducing an interfacial
zone, generating interactions between aggregates and propagating microcracks, and the need to
disperse nanomaterials in the space between aggregates.
Evaluation of the reinforcement efficiency of CNFs in a coarser cementitious matrix, highperformance concrete, was undertaken to see the effect of increased particle size on the
interaction of nano-reinforcements with the relatively coarser matrix. This has also highlighted
the filler effect and dimensional stability brought about by use coarser aggregates in a well
graded matrix. Furthermore, micro-scale fibers along with their hybrid combination with CNFs
were evaluated to achieve desired balance of performance and cost efficiency.

307

Cementitious materials are essentially particulate composites, which are rarely used without
aggregates (particulates). Given the micro- to millimeter-scale dimensions of the sand and gravel
particles used in DSP concrete, hybrid reinforcements which complement the reinforcing and
dimensional stabilizing actions of particulates with multi-scale reinforcement mechanisms could
produce particularly positive effects. In the past, the concept of using hybrid (micro- and
millimeter-scale) reinforcement has been explored by the concrete industry for achieving
balanced gains in material properties (Moranville-Regourd 2001; O'Connell, Boul et al. 2001;
Lawler, Zampini et al. 2005; Banthia and Sappakittipakorn 2007).

13.2 Materials and Methods
13.2.1 Graphite Nanomaterials and Microfibers
Different types of acid-functionalized and polyacrylic acid (PAA) physisorbed carbon nanofibers
as well as (micro-scale) carbon and polypropylene fibers were evaluated for use in cement-based
nanocomposites. These graphite nanomaterials and microfibers have been introduced in detail in
Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. Acid-functionalized and PAA physisorbed CNFs
(CNF-OX and CNF-PAA), carbon microfibers (CMF) and polypropylene microfibers (PP) were
used in high-performance cement-based concrete matrix.
13.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
concrete was used to evaluate the reinforcement efficiency of both graphite nanomaterials and/
or carbon or polypropylene microfibers at different volume fractions.

308

Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon; polypropylene microfibers were prepared and cured
following ASTM procedures described in the same section.
13.2.3 Experimental Methods
The test procedures employed to determine the engineering properties of high-performance
concrete included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests
(ASTM D 7136) and Abrasion tests (ASTM C 944). SEM and EDS were also employed to gain
further insight into the structure and failure mechanisms of cement-based nanocomposites. The
experimental results were subjected to statistical analysis using the analysis of variance
(ANOVA) and pair-wise comparison methods. RSA (Response Surface Analysis) was used to
identify reinforcement volume fractions which yield optimum gains in various engineering
properties of the cementitious matrix.

13.3 Experimental Results and Discussion
Acid-oxidized (CNF-OX) and Poly-acrylic acid physisorbed (CNF-PAA) CNFs were used as
reinforcement at 0.16 vol.%; polypropylene (PP)

and carbon microfibers(CMF) were used at

0.24 vol.%. Combinations of polypropylene and carbon microfibers with PAA physisorbed
CNFs were also used as hybrid reinforcement for concrete, with PP/ CMF microfibers and CNFPAA used at 0.24 vol.% and 0.16 vol.%, respectively.
13.3.1 Flexural Performance
The flexural attributes test results for the high-performance concrete reinforced with different
volume fractions of CNFs and/or microfibers are summarized in Table 13.1. Significant gains in
all flexural performance characteristics of concrete are observed with carbon nanofiber

309

reinforcement used alone or in combination with microfibers. Polymer wrapping of nanofibers
improves their reinforcement efficiency in concrete. The most desired balance of properties for a
single reinforcement system was realized with 0.16 vol.% of PAA-physisorbed CNFs (CNFPAA). The corresponding improvements in flexural strength, energy absorption capacity and
maximum deflection versus plain concrete were 20.9%, 134% and 120%, respectively.
The experimental data generated in the past pointed at the positive effects of hybrid (nano- and
micro-scale) reinforcement systems in high-performance cementitious materials [Banthia, 2007].
The test data presented here suggest that micro-scale fibers of lower modulus and lower cost
(when compared with high-modulus carbon microfiber) could effectively complement the
reinforcing effects of polymer-wrapped CNFs in high-performance (DSP) concrete. These lowermodulus microfibers, when used as hybrid reinforcement together with both treated and
untreated carbon microfibers, produced balanced gains in the flexural performance attributes of
the high-performance (DSP) concrete by interacting with and arresting the cracks developing in
matrix at different scales. The hybrid reinforcement of low-modulus polypropylene microfibers
with nanofibers produced particularly pronounced gains in the mechanical properties of DSP
concrete. The best balance of flexural attributes was realized with the hybrid reinforcement
comprising PAA-physisorbed CNFs (0.16 vol.%) and polypropylene microfiber (0.24 vol.%),
which produced 35.8%, 644% and 371% rise in the flexural strength, maximum deflection and
energy absorption capacity of the high-performance concrete, respectively.
The hybrid reinforcement systems also overcame the adverse effects of microfibers on flexural
strength, as reported earlier. The synergistic reinforcing action of nano- and micro-scale
reinforcement was further enhanced by polymer wrapping of CNFs. The effectiveness of hybrid
reinforcement was also evident for high-modulus carbon microfibers when used with different

310

PAA-physisorbed CNFs. This hybrid reinforcement resulted in improvement of all engineering
properties versus plain matrix and also versus a similar hybrid reinforced materials incorporating
untreated CNFs.
The experimental results were subjected to statistical analysis of variance (ANOVA). Due to the
large number of observations, ANOVA only gives the general trends in test results. The ANOVA
outcomes indicate that there are statistically significant (at 0.05 significance level) improvements
in all flexural attributes of high-performance concrete due to the addition of nano-scale and
hybrid reinforcement systems. Pair-wise comparisons were carried out in order to assess the
statistical significance of the effects associated with the addition of nano- and/or micro-scale
reinforcement systems. These results indicate that nano-scale reinforcement systems at 0.16
vol.% produced statistically significant improvements in all flexural attributes of the plain matrix
at 0.05 significance level when compared with plain concrete. Both microfibers produced
statistically insignificant drops in flexural strength (0.297 and 0.481 significance levels for
carbon and polypropylene microfibers, respectively). Pair-wise analysis also pointed at
statistically insignificant gains in energy absorption capacity (0.502 significance level) and
maximum deflection (0.402 significance level) realized with 0.24 vol.% carbon microfiber. Use
of low-modulus polypropylene microfiber as well as high-modulus carbon microfiber in
conjunction with different surface functionalized CNFs reversed any negative trends in the
effects of individual reinforcement, producing balanced gains in the engineering properties of
high-performance (DSP) concrete at significance levels varying form 0.000 to 0.047. The lower
values of significance level were only for maximum deflection when carbon microfibers were
used in conjunction with CNFs.

311

13.3.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
concretes with nano- and/or micro-scale reinforcement systems are presented in Table 13.2. As
was the case with DSP paste and mortar, PAA physisorbed CNFs as well as both microfibers
produced relatively small and statistically insignificant (at 0.05 significance level) effects on the
compressive strength of high-performance concrete.
Outcomes of pair-wise comparisons also indicate the effects of nano- and micro-scale
reinforcement systems on compressive strength are not statistically significant (at 0.05
significance level). The use of hybrid reinforcement systems comprising carbon or
polypropylene microfiber and carbon nanotubes restored the compressive strengths of DSP
concrete by inducing interactions with matrix and its cracks at different scales. The hybrid
reinforcement system comprising functionalized CNFs and micro-scale polypropylene fiber
produced a small rise in the compressive strength of high-performance concrete.
13.3.3 Impact Resistance
The impact test data are summarized in Table 13.3. All nano- and/or micro-scale reinforcement
systems produced improvements in the impact resistance of high-performance concrete. The
maximum rise in impact resistance realized with a single reinforcement (69%) was for 0.16
vol.% of CNF-PAA. In the case of micro-scale fibers, polypropylene produced a higher gain
(60%) in the impact resistance of high-performance (DSP) concrete.
All hybrid nano- and micro-scale reinforcement systems produced further rise in the mechanical
properties of high-performance concrete when compared with individual (nano-or micro-scale)
reinforcement used alone. A significant (0.000 significance level) increase in impact resistance

312

was observed for the hybrid reinforcement comprising polypropylene microfiber with CNFPAA, which confirms the effectiveness of hybrid reinforcement. The maximum increase in
impact resistance (115%) was brought about by hybrid use of 0.16 vol.% of CNF-PAA and 0.24
vol.% of PP microfiber. Outcomes of ANOVA and pair-wise comparisons point at the statistical
significant (0.000 to 0.012 significance levels) of nano- and/ or micro-scale reinforcement effects
on the impact resistance of high-performance concrete. When compared with high-performance
cement paste and mortar (with nano-scale reinforcement), concrete (with nano-scale
reinforcement) provides higher levels of impact resistance; this observation points at the wellknown contributions of aggregates (in concrete) to the toughness of cementitious matrices.
13.3.4 Abrasion Resistance
The abrasion test results produced for high-performance concrete with different volume fractions
of graphite nanomaterials and/ or microfiber reinforcement are summarized in Table13. 4.
CNF-PAA at 0.16 vol.% produced the greatest improvement (40.0%) in the abrasion resistance
of high-performance (DSP) concrete. All nano- and/or micro-scale reinforcement systems
produced marked gains in the abrasion resistance of this high-performance concrete matrix,
which were statistically significant at 0.05 significance level. Outcomes of pair-wise
comparisons confirmed that each of the reinforcement conditions considered here produced
statistically significant improvements in the abrasion resistance of high-performance concrete
(with significance levels ranging from 0.006 to 0.000).
13.3.5 Response Surface Analysis
The test data on high-performance concrete with nano- and/or micro-scale reinforcement systems
was subjected to response surface analysis. The objective of this analysis was to identify

313

optimum reinforcement conditions which maximize the benefits to specific engineering
properties of high-performance concrete. This section presents optimization of concrete
nanocomposites for enhancing specific engineering properties, and also for simultaneous
enhancement of several engineering properties.
Response surface analysis (RSA) of the test data was conducted considering the volume fractions
of nano- and micro-scale reinforcement systems as input variables. Flexural strength, energy
absorption capacity, maximum deflection and impact resistance were used as response variables.
The RSA process started with evaluating the effects of input variables on response variables, and
was followed with desirability analyses considering means of response variables, using both
canonical and ridge analyses.
Response Surface analyses conducted for each of the flexural strength, energy absorption
capacity, maximum deflection and impact resistance test data indicated that an optimum
reinforcement system comprising 0.12 vol.% CNF-PAA and 0.24 vol.% PP microfiber
maximizes subject properties. Optimal Response for flexural strength as calculated to be 17.20
MPa with 95% Confidence Interval (16.02, 18.18 MPa), Optimal Response for energy absorption
capacity is 482.78 N/mm with 95% Confidence Interval (336.69, 508.86 N.mm), Optimal
Response for maximum deflection is 5.101 mm with 95% Confidence Interval (4.79, 5.41 mm),
and Optimal Response for maximum impact resistance is 2.90 mm/mm with 95% Confidence
Interval (2.75, 3.63 mm/ mm). All stationary points are saddle points which do not give a
maximum but a general direction to follow in which targeted properties can be maximized.
The ridge analysis outcomes can be used to adjust the reinforcement condition for achieving
further gains in engineering properties, Table 13.5. The analyses for flexural attributes and

314

impact resistance show that the current hybrid reinforcement is the appropriate combination to
maximize these properties of the matrix.
Desirability analysis of test data was also conducted in order to identify reinforcement conditions
which yield a desired balance of all the engineering properties considered here. Desirability
analyses were conducted using mean values obtained through both canonical and ridge analyses.
These outputs indicate that maximum values of all response variables can be achieved with 0.11;
0.12 vol.% CNF-PAA and 0.26; 0.24 vol.% PP microfiber. These values are closer to what is
being used as hybrid reinforcement in this high-performance concrete matrix.
13.3.6 Scanning Electron Microscope Observations
Fig. 13.1, shows SEM images of DSP concrete samples. The density of the DSP matrix is
apparent in Fig. 13.1a and 13.1b. Fig 13.1a also highlights the positive effective of coarser
particles in the matrix like sand and gravel. Fig. 13.1c and 13.1d points at the uniform dispersion
of nanotubes and their bridging/pullout actions across a fine (nano-scale) crack within the
cementitious matrix. Figs. 13.1e and 13.1f, present evidence of micro-fiber pullout from the
matrix. SEM observations of cementitious concrete matrices reinforced with hybrid
reinforcement (CNFs and micro-fibers) also indicated that the presence of nanofibers in the
matrix around micro-fibers enhanced the interaction of matrix with micro-fibers, (Fig. 13.1f),
thus benefiting the gains in matrix performance with hybrid reinforcement. This strong
interaction of matrix with micro-fibers was not observed in samples reinforced with only microfiber reinforcement. These observations provide some insight into the synergistic action of nanoand micro-scale reinforcement systems in cementitious matrices. All these observations support
the results of various engineering properties mentioned in earlier.

315

13.4 Summary
The experimental results generated for high performance cementitious concrete reinforced with
both oxidized and (CNF-OX) PAA physisorbed carbon nanofibers (CNF-PAA), polypropylene
(PP) microfibers, carbon microfibers (CMF) and their hybrids casted at various/ optimum
volume fractions and with carboxyl-based superplasticizers yield that proper use of nano-scale
reinforcement can produce improvements in high-performance concrete mechanical properties
which surpass those realized with micro-scale reinforcement. Nano- and micro-scale
reinforcement renders complementary reinforcing effects in high-performance concrete;
optimum reinforcement systems thus incorporate both nano- and micro-scale reinforcement
systems. The experimental results produced using the refined hybrid reinforcement condition,
identified through RSA, point at the effectiveness of response surface analysis in identifying the
desired reinforcement condition for improving the targeted attributes of a matrix.

316

Table 13.1- Mean values of flexural attributes of plain high-performance concrete and those
reinforced with different nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Carbon nanofiber, 0.16 vol.% (CNF-OX-0.16)
Carbon nanofiber, 0.16 vol.% (CNF-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.% (CNF-OX+PP)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.% (CNF-PAA+PP)
Carbon microfiber, 0.24 vol.% and Carbon
nanofiber, 0.16 vol.% (CNF-OX+CMF)
Carbon microfiber, 0.24 vol.% and Carbon
nanofiber, 0.16 vol.% (CNF-PAA+CMF)

Flexural
Strength
(MPa)
13.4
12.8
13.0
15.4
16.2
17.4

Deflection
(mm)
0.71
1.01
4.55
1.52
1.56
5.30

Energy
Absorption
(N.mm)
92
111
383
193
215
400

18.2

5.28

433

17.2

1.55

274

17.5

1.65

269

Table 13.2- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Carbon nanofiber, 0.16 vol.% (CNF-OX-0.16)
Carbon nanofiber, 0.16 vol.% (CNF-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and Carbon
nanofiber, 0.16 vol.% (CNF-OX+PP)
Polypropylene microfiber, 0.24 vol.% and Carbon
nanofiber, 0.16 vol.% (CNF-PAA+PP)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber,
0.16 vol.% (CNF-OX+CMF)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber,
0.16 vol.% (CNF-PAA+CMF)

317

Mean Compressive
Strength (MPa)
151
139
137
140
144
148

Standard
Error (MPa)
7.75
13.5
18.4
8.75
8.23
12.3

148

4.33

150

9.87

151

13.1

Table 13.3- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Carbon nanofiber, 0.16 vol.% (CNF-OX-0.16)
Carbon nanofiber, 0.16 vol.% (CNF-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber, 0.16
vol.% (CNF-OX+PP)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber, 0.16
vol.% (CNF-PAA+PP)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber, 0.16 vol.%
(CNF-OX+CMF)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber, 0.16 vol.%
(CNF-PAA+CMF)

Impact
Resistance
(mm/mm)
1.50
1.98
2.40
2.29
2.32
3.25
3.26
2.35
2.35

Table 13.4- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Carbon nanofiber, 0.16 vol.% (CNF-OX-0.16)
Carbon nanofiber, 0.16 vol.% (CNF-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber, 0.16
vol.% (CNF-OX+PP)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber, 0.16
vol.% (CNF-PAA+PP)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber, 0.16 vol.%
(CNF-OX+CMF)
Carbon microfiber, 0.24 vol.% and Carbon nanofiber, 0.16 vol.%
(CNF-PAA+CMF)

318

Loss of
Mass
(grams)
1.50
1.15
1.05
0.97
0.98
0.99
0.99
1.00
1.01

Table 13.5- Outcomes of Ridge analysis of the flexural attributes and impact resistance test data
for high performance concrete with nano- and/or micro-scale reinforcement.
(a) Flexural Strength
Ridge Analysis for Maximizing Flexural Strength
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response Upper
Lower
CNF-PAA
PP
0.000
15.604
15.024
16.183
0.080
0.120
0.100
15.847
15.264
16.429
0.088
0.123
0.200
16.093
15.502
16.684
0.095
0.127
0.300
16.344
15.739
16.949
0.103
0.131
0.400
16.600
15.976
17.225
0.110
0.136
0.500
16.861
16.212
17.511
0.111
0.141
0.600
17.128
16.449
17.807
0.112
0.147
0.700
17.400
16.686
18.115
0.113
0.152
0.800
17.678
16.924
18.433
0.116
0.208
0.900
17.963
17.164
18.762
0.118
0.225
1.000
18.253
17.404
19.103
0.120
0.240
(b) Energy Absorption Capacity
Ridge Analysis for Maximizing Energy Absorption Capacity
Coded Radius Estimated
Response
0.000
282.779
0.100
295.584
0.200
308.239
0.300
320.761
0.400
333.167
0.500
345.477
0.600
357.712
0.700
369.893
0.800
382.042
0.900
394.181
1.000
406.334

95.00% Confidence IntervalUncoded Factor Values
Upper
Lower
CNF-PAA
PP
256.693
308.865
0.080
0.120
269.367
321.800
0.082
0.132
281.635
334.843
0.084
0.143
293.521
348.000
0.086
0.155
305.059
361.276
0.087
0.167
316.288
374.667
0.087
0.179
327.254
388.171
0.098
0.191
338.003
401.783
0.108
0.203
348.582
415.501
0.117
0.215
359.037
429.326
0.0119
0.228
369.406
443.263
0.120
0.240

319

(c) Maximum Deflection
Ridge Analysis for Maximizing Maximum Deflection
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response Upper
Lower
CNF-PAA
PP
0.000
3.101
2.793
3.408
0.080
0.120
0.100
3.296
2.987
3.605
0.082
0.132
0.200
3.492
3.178
3.805
0.083
0.143
0.300
3.687
3.366
4.008
0.085
0.155
0.400
3.882
3.551
4.213
0.086
0.167
0.500
4.076
3.733
4.420
0.088
0.179
0.600
4.271
3.912
4.630
0.089
0.191
0.700
4.465
4.089
4.841
0.091
0.202
0.800
4.659
4.264
5.054
0.092
0.214
0.900
4.853
4.437
5.268
0.109
0.226
1.000
5.046
4.608
5.484
0.119
0.238
(d) Impact Resistance
Ridge Analysis for Maximizing Impact Resistance
Coded Radius Estimated 95.00% Confidence IntervalUncoded Factor Values
Response Upper
Lower
CNTF-PAA
PP
0.000
2.500
2.375
2.625
0.080
0.120
0.100
2.567
2.441
2.693
0.086
0.127
0.200
2.633
2.506
2.761
0.092
0.135
0.300
2.700
2.569
2.831
0.099
0.142
0.400
2.767
2.631
2.902
0.105
0.150
0.500
2.833
2.693
2.974
0.111
0.157
0.600
2.900
2.752
3.048
0.117
0.165
0.700
2.967
2.811
3.123
0.124
0.172
0.800
3.034
2.868
3.199
0.130
0.180
0.900
3.100
2.925
3.276
0.136
0.227
1.000
3.167
2.979
3.355
0.120
0.240

320

(a) Crack deflection around sand particle

(b) Dense structure of paste
Figure 13.1- SEM images of fractured surfaces of high performance concrete.
321

Figure 13.1 (cont’d)

(c)

(d) Pulled out nanofibers at fractured surface

322

Figure 13.1 (cont’d)

(e) Carbon nano- and micro-fibers in the matrix

(f) CNFs interacting with carbon microfibers

323

CHAPTER 14

14 GRAPHITE
MATRICES

NANOPLATELETS

IN

HIGH

PERFORMANCE

14.1 General
Fibers are incorporated into cementitious matrices to overcome these weaknesses, producing
materials with increased tensile strength, ductility, and toughness and improved durability.
However, cracking in cement-based materials starts at the nano-scale level, where traditional
reinforcement is not effective. The introduction of graphite nanomaterials has opened a new field
for nano-sized reinforcement that should hinder the formation and propagation of nano-cracks
and improve abrasion and impact resistance of the matrix. Previous research by the authors of
this work on reinforcing cementitious materials with multiwall carbon nanotubes (MWNTs) and
carbon nanofibers (CNFs) has shown that all major engineering properties as well as durability
of high-performance cementitious matrices can be significantly increased by adding very low
concentrations of homogeneously dispersed carbon nanotubes and carbon nanofibers (as little as
0.04% by weight of cement). However, higher costs of these nanomaterials sometimes become a
hindrance in their effective use in cement-based matrices. One objective of this study is to
examine whether exfoliated graphite nanoplatelets (GP) can have similar beneficial effects as a
reinforcing material for cementitious matrices.
Graphite nanoplatelets (Fig. 2.3g) are obtained by exfoliation of natural graphite; their thickness
can be controlled in a range from 1 to 25 nm (with about 0.5 nm thickness of each grapheme
layer), and their diameter from the sub-micron level to tens of microns. Graphite, in its basal
324

plane, offers elastic modulus of 1,060 GPa, thermal conductivity of 250 W/m.K, and electrical
-5

resistivity of 4x10

Ω.cm (Kalaitzidou, Fukushima et al. 2007; Kalaitzidou, Fukushima et al.

2007; Li, Sham et al. 2007). Stress concentrations at the rough edges of graphite nanoplatelets
lower their tensile strength below that of carbon nanotubes.
GP appear extremely promising in opening new avenues in the field of nanocomposites.
Research efforts have concentrated on developing GP–polymer composites (Ping, Beaudoin et
al. 1991; Song, Jang et al. 2007; Pfeifer, Moeser et al. 2009). The investigation of graphite
nanomaterials/concrete composites is at a relatively novel stage, and no authentic work regarding
the effectiveness of GP in enhancing the engineering properties of concrete has been published.
To develop high-performance nanoplatelet–cement nanocomposites, a homogeneous dispersion
of the nanoplatelets in cementitious matrices must be achieved (Ahlborn, Harris et al. 2011). In
addition, a strong bond between the nano-reinforcement and the matrix will allow efficient load
transfer.
In this study, the effects of pristine graphite nanoplatelets (GP) and PAA physisorbed graphite
nanoplatelets (GP-PAA) are use of refined dispersion techniques on the reinforcement efficiency
of nanotubes at different volume fractions within a high-performance cementitious (DSP)
matrices were evaluated.

Scanning electron microscope observations of fractured surfaces

pointed at successful dispersion of nanofibers within the cement-based matrix. An investigation
of hybrid reinforcement systems comprising GP and micro-scale fibers carbon (CMF) and
polypropylene (PP) indicated that selected hybrid systems offer a cost-effective approach for
realizing balanced gains in performance characteristics of high-performance cementitious
matrices.

325

14.2 Materials and Methods
14.2.1 Graphite Nanomaterials and Microfibers
Both pristine (GP) and Polyacrylic acid physisorbed graphite nanoplatelets (GP-PAA), carbon
microfibers (CMF) and polypropylene microfibers (PP) at different volume fractions were used
in different DSP matrices. These graphite nanomaterials and microfibers have been introduced in
detail in Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3.
14.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrices selected have been introduced in detail in Chapter 2, Section 2.5.1
and Table 2.4. DSP paste, mortar and concrete were used to evaluate the reinforcement
efficiency of both graphite nanoplatelets and different microfibers at different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon and polypropylene microfibers were prepared and
cured following ASTM procedures described in Chapter 2, Section 2.5.2.
14.2.3 Experimental Methods
The test procedures employed to determine the engineering properties of high-performance
matrices included compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests
(ASTM D 7136) and Abrasion tests (ASTM C 944). SEM and EDS were also employed to gain
further insight into the structure and failure mechanisms of cement-based nanocomposites. The
experimental results were subjected to statistical analysis using the analysis of variance
(ANOVA) and pair-wise comparison methods.

326

14.3 Experimental Results and Discussion
14.3.1 Effects of Graphite Platelets and Carbon Microfibers on Engineering Properties of
DSP Paste
14.3.1.1 Flexural Performance
The flexure test data are summarized in Table 14.1. GPs, CMFs, and especially their
combination (hybrid reinforcement) produce important gains in the flexural strength, energy
absorption capacity and maximum deflection of DSP cementitious pastes. The hybrid
reinforcement system yields 61.7%, 366% and 264% gains in flexural strength, energy
absorption capacity and maximum deflection of cementitious pastes. The ANOVA outcomes
confirm the statistical significance (at significance level, p, of 0.000) of the contributions of
nanoplatelet and microfiber reinforcement to the flexural performance of DSP pastes.
14.3.1.2 Compressive Strength
The compressive strength test results for DSP pastes with graphite nanoplatelet and/or carbon
microfiber reinforcement are summarized in Table 14.2. GPs seem to produce a minor
(statistically insignificant) loss of compressive strength. This could be attributed to their planar
(flaky) geometry and micro-scale planar dimension, which could produce local stress rise. The
ANOVA outputs indicate that reinforcement with graphite nanoplatelet and/or carbon microfiber
has statistically insignificant effects (significant level, p, of 0.482).
14.3.1.3 Impact Resistance
The impact resistance test results presented in Table 14.3 and point at the desirable effects of
carbon microfibers and especially graphite nanoplatelets and hybrid reinforcement systems. The
impact resistance of DSP paste increases by 42.7%, 114% and 158% with carbon microfiber,

327

graphite nanoplatelet and hybrid reinforcement. The ANOVA outcomes also point at the
statistical significance of the graphite nanoplatelet and/or carbon microfiber reinforcement
effects on the impact resistance of DSP paste (at a significance level, p, of 0.001). The high
effectiveness of graphite nanoplatelets in enhancing the impact resistance of cementitious paste
is somewhat surprising given their flaky geometry. Fig. 14.1 shows scanning electron
microscope images of the failed surfaces of impact specimens reinforced with graphite
nanoplatelets. It seems that the layered structure of nanoplatelets have undergone some level of
delamination while their exterior surfaces interacting with microcracks propagating in the
cementitious matrix under impact loads.
14.3.1.4 Abrasion Resistance
The abrasion resistance test results presented in Table 14.4 and indicate that graphite
nanoplatelet and/or carbon microfiber reinforcement of cementitious pastes produce
improvements in the abrasion resistance of plain DSP paste. The abrasion resistance of DSP
paste increases by 45.5%, 58.9% and 56.4% with carbon microfiber, graphite nanoplatelet and
hybrid reinforcement.
The ANOVA outcomes point at the statistical significance of the contributions of graphite
nanoplatelet and/or carbon microfiber reinforcement to the abrasion resistance of DSP
cementitious pastes.
14.3.2 Effects of Graphite Platelets and Carbon Microfibers on Engineering Properties of
DSP Mortar
Different reinforcement systems comprising PAA-physisorbed graphite nanoplatelets (GP-PAA)
and carbon fibers (CMF) with micro-scale diameter were evaluated in DSP mortar. Exploratory

328

studies with different graphite nanomaterial reinforcement indicated that, for the cementitious
matrix and processing conditions considered here, the GP-PAA volume fraction should be
lowered below 0.24 vol.% before benefits can be realized from introduction of carbon
microfibers. Hence, a graphite nanoplatelet volume fraction of 0.16 vol.% was considered for
development of hybrid reinforcement systems.
14.3.2.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity test results for the
high-performance (DSP) cementitious mortar reinforced with different volume fractions of PAAphysisorbed graphite nanoplatelets and/ or microfibers are presented in Table 14.5. These results
indicate that all the reinforcement conditions considered here produce improvements in flexural
strength, maximum deflection and energy absorption capacity of the high-performance (DSP)
cementitious plain mortar matrix. For single reinforcement condition, a highly desired balance of
flexural qualities was produced at 0.16 vol. % of GP-PAA and 0.96 vol.% of CMF. The
improvements in the flexural strength, energy absorption capacity and maximum deflection of
DSP mortar realized with 0.16 vol.% of GP-PAA were 3.6%, 13% and 64%), respectively. The
improvements in flexural strength, maximum deflection and energy absorption capacity with
addition of 0.96 vol.% of carbon microfibers were 39%, 119% and 156%, respectively. The
comparatively lesser improvement in all flexural attributes with use of GP-PAA is expected due
to morphology and physical characteristics of GP-PAA. The experimental results also indicate
that hybrid (nano- and micro-scale) reinforcement systems produce desired effects, which point
at the complementary role of reinforcement systems operating at different length scales.
Due to large number of observations, application of analysis of variance (ANOVA) to test results
only reveals general trends. The analysis of variance of test results was followed by pair-wise

329

comparisons. The key findings of pair-wise comparisons are summarized below. The
comparisons indicate that reinforcement systems using GP-PAA alone or with CMF did not
produce improvement in all flexural attributes of the plain matrix. However CMF, when used
alone, did produce improvements in all flexural attributes at a high significance (p) levels
ranging from 0.042 to 0.000, at different volume fractions.
14.3.2.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for DSP cementitious
mortars with nano- and/or micro-scale reinforcement systems are presented in Table 14.6. As
was the case with DSP paste, carbon nano- and/or micro-scale reinforcement of DSP mortar
produces small and statistically insignificant (0.755 significance level) effects on compressive
strength; pair-wise comparisons, however, indicate GP-PAA (and hybrid reinforcement systems
incorporating GP-PAA/ CMF) tend to have statistically significant (but still small) adverse
effects on the compressive strength of DSP mortar, varying from 0.051 to 0.07.
14.3.2.3 Impact Resistance
The impact resistance test data are summarized in Table 14.8. GP-PAA at 0.16 and 0.24 vol.%
improve the impact resistance of DSP mortar by 99%; 109%. Micro-scale carbon fibers
reinforcement at 0.24, 0.48 and 0.96 vol% increases the impact resistance of DSP mortar by
36%, 46% and 64%, respectively. All hybrid (nano- and micro-scale) reinforcement systems also
produced improvements in impact resistance when compared with their corresponding case with
micro-scale reinforcement alone. ANOVA outcomes and pair wise comparison point at the
statistical significance (p=0.002) of nano- and/or micro-scale reinforcement effects on the impact
resistance of DSP mortar. When compared with DSP paste with nano-scale reinforcement, DSP
mortar with nano-scale reinforcement provides higher levels of impact resistance. DSP mortars
330

reinforced with different volume fractions of GP-PAA generally provided particularly high
levels of impact resistance due to their platy morphology and physical attributes (which was not
the case for flexural performance and compressive strength).
14.3.2.4 Abrasion Resistance
The abrasion test results performed on DSP mortar with different carbon nano- and/or microscale reinforcement are summarized in Table 14.8. At equal volume fractions, GP-PAA and its
hybrid combination with carbon microfiber produced the lesser improvement in the abrasion
resistance of DSP mortar when compared with CMF. However, all nano- and/or micro-scale
reinforcement conditions produced marked improvements in the abrasion resistance of DSP
mortar, which were statistically significant (p = 0.007). Detailed pair-wise comparisons of test
results confirmed that improvements in abrasion resistance with nano- and/or micro-scale
reinforcement (versus plain DSP mortar) were statistically significant at 0.042 to 0.000
significance levels.
14.3.3 Effects of Graphite Platelets and Different Microfibers on Engineering Properties of
DSP Concrete
Different reinforcement systems comprising PAA-physisorbed graphite nanoplatelets (GP-PAA),
carbon fibers (CMF) and polypropylene fibers (PP) with micro-scale diameter were evaluated in
DSP concrete, at different volume fractions.
14.3.3.1 Flexural Performance
The flexural strength, maximum deflection and energy absorption capacity test results for the
high-performance (DSP) concrete reinforced with different volume fractions of PAAphysisorbed graphite nanoplatelets and/or microfibers, prepared using ADVA® Cast 575

331

superplasticizer, are summarized in Table 14.9. These test results indicate that all reinforcement
systems produced gains in all flexural attributes considered here, except for the two microfibers
which lowered the flexural strength of concrete when used alone. The most desired balance of
properties for a single reinforcement was realized by the addition of 0.16 vol.% of GP-PAA. The
corresponding improvements in flexural strength, energy absorption capacity and maximum
deflection with respect to plain concrete were 10.5%, 97% and 76%, respectively. Test results
also indicate that the addition of each of the microfibers to this (relatively brittle) highperformance (DSP) concrete produced important gains in two out of the three flexural attributes
considered here (maximum deflection and energy absorption capacity), while causing a relatively
small drop in the flexural strength. PP microfiber produced the maximum rise in energy
absorption capacity (315%) and maximum deflection (550%).
The test data generated in the past pointed at the positive effects of hybrid (nano- and microscale) reinforcement systems in high-performance cementitious materials. The test data
presented here suggest that micro-scale fibers of lower modulus and lower cost (when compared
with carbon microfiber) could effectively complement the beneficial effects of graphite
nanoplatelets in high-performance (DSP) concrete. These lower-modulus microfibers, when used
together with different graphite nanoplatelets as hybrid reinforcement, produced balanced gains
in the flexural performance characteristics of high-performance (DSP) concrete by interacting
with and arresting cracks which developed in matrix at different scales. The hybrid
reinforcement comprising low-modulus polypropylene microfibers with different graphite
nanomaterials produced particularly pronounced gains in the mechanical properties of DSP
concrete. The best balance of flexural attributes was produced by the hybrid reinforcement
comprising GP-PAA (0.16 vol.%) and PP (0.24 vol.%), which produced a 15%, 573% and

332

329% rise in flexural strength, maximum deflection and energy absorption capacity, respectively.
Hybrid reinforcement systems also overcame the adverse effects of microfibers on flexural
strength. The synergistic reinforcement actions of nano- and micro-scale reinforcement systems
point at the merits of multi-scale reinforcement in high-performance concrete. This effectiveness
of hybrid reinforcement was also evident in the case of high-modulus carbon microfibers used
together with different graphite nanomaterials. This hybrid reinforcement produced
improvements in all the engineering properties of high-performance concrete when compared
with plain concrete and also that reinforced with CMFs alone.
The experimental results were subjected to statistical analysis of variance (ANOVA). Due to the
large number of observations, ANOVA only gives the general trends in test results. Pair-wise
comparisons were carried out in order to assess the statistical significance of the effects
associated with the addition of GP-PAA- and/or micro-scale reinforcement systems. These
results indicate that GP-PAA only produced statistically insignificant gains in maximum
deflection (0.158 significance level) when compared with plain concrete. Both microfibers
produced statistically insignificant drops in flexural Strength (0.297 and 0.481 significance levels
for carbon and polypropylene microfibers, respectively). Pair-wise analysis also pointed at
statistically insignificant gains in energy absorption capacity (0.502 significance level) and
maximum deflection (0.402 significance level) realized with 0.24 vol.% carbon microfiber. Use
of low-modulus PP microfiber as well as high-modulus CMFs in conjunction with GP-PAA
reversed any negative trends in the effects of individual reinforcement, producing balanced gains
in the engineering properties of high-performance (DSP) concrete at significance levels varying
form0.000 to 0.047. The lower values of significance level were only for maximum deflection
when CMFs were used in conjunction with GP-PAA.

333

14.3.3.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for high-performance
(DSP) concretes with GP-PAA- and/or different micro-scale reinforcement systems are presented
in Table 14.10. As was the case with DSP paste and mortar, GP-PAA as well as both microfibers
produced relatively small and statistically insignificant (at 0.05 significance level) effects on the
compressive strength of DSP concrete.
Outcomes of pair-wise comparisons also indicate that, except for GP-PAA (0.003 significance
level), the effects of micro-scale reinforcement systems on compressive strength are not
statistically significant (at 0.05 significance level). The adverse effects of GP-PAA on
compressive strength were also observed in DSP paste and mortar, and can be attributed to the
planar geometry and probably agglomeration of graphite nanoplatelets. The combination of GPPAA and PP or CMF mitigated this adverse effect. This is due to interactions of reinforcements
with matrix and its cracks at different scales.
14.3.3.3 Impact Resistance
The impact test data are summarized in Table 14.11. All GP-PAA- and/or micro-scale
reinforcement systems produced improvements in the impact resistance of plain highperformance concrete matrix. The maximum rise in impact resistance realized with GP-PAA
reinforcement (46%) was for 0.16 vol.% of GP-PAA. In the case of micro-scale fibers, PP
produced a gain of 60% in the impact resistance of high-performance (DSP) concrete at 0.24
vol.%.
All hybrid GP-PAA- and micro-scale reinforcement systems produced further rise in the
mechanical properties of high-performance concrete when compared with individual (nano-or

334

micro-scale) reinforcement used alone. A significant (0.000 significance level) increase in
impact resistance was observed for the hybrid reinforcement comprising PP microfiber with GPPAA, which confirms the effectiveness of hybrid reinforcement. The maximum increase in
impact resistance (106%) was brought about by 0.16 vol.% of GP-PAA and 0.24 vol.% of PP
microfiber. Outcomes of ANOVA and pair-wise comparisons point at statistical significant
(0.000 to 0.012 significance levels) effects on the impact resistance of high-performance
concrete due to addition of GP-PAA- and/ or micro-scale reinforcements. When compared with
high-performance cement paste and mortar (with nano-scale reinforcement), concrete (with
nano-scale reinforcement) provides higher levels of impact resistance; this observation points at
the well-known contributions of aggregates (in concrete) to the toughness of cementitious pastes.
14.3.3.4 Abrasion Resistance
The abrasion test results produced for DSP concrete with different volume fractions of GP-PAA
and/ or different microfiber reinforcement are summarized in Table 14.12.
All GP-PAA- and/or micro-scale reinforcement systems, using in conjunction with the ADVA®
Cast 575 superplasticizer, produced marked gains in the abrasion resistance of DSP concrete,
which were statistically significant at 0.05 significance level. GP-PAA at 0.16 vol.% produced
the greatest improvement (38%) in the abrasion resistance of high-performance (DSP) concrete.
Outcomes of pair-wise comparisons confirmed that each of the reinforcement conditions
considered here produced statistically significant improvements in the abrasion resistance of
high-performance (DSP) concrete (with significance levels ranging from 0.006 to 0.000).

335

14.4 Summary
The experimental results generated for cementitious pastes reinforced with pristine and PAAphysisorbed graphite nanoplatelets and also their hybrids with carbon microfibers/ polypropylene
microfibers in increasingly coarser high-performance matrices indicated that the test results of all
engineering properties indicate that all the reinforcement conditions considered here produce
improvements in flexural attributes, impact and abrasion resistance of the all high-performance
(DSP) cementitious plain matrices.
In all DSP matrices, GP-PAA, CMF, and especially their combination (hybrid reinforcement)
produce important gains in the flexural strength, energy absorption capacity and maximum
deflection of DSP cementitious pastes. The most important contribution of GP-PAA was to the
impact resistance of DSP paste and mortar. This significant increase is due to flaky geometry
which tends to be very effective under sudden loads.
In DSP concrete matrix because of a coarser matrix. GP-PAA effectively reinforced the paste
portion of the matrix. Thus improvements were observed flexural strength, energy absorption
capacity and maximum deflection with respect to plain concrete. Whereas the addition of both
CMF and PP microfibers high-performance (DSP) concrete produced important gains in only
two out of the three flexural attributes considered here (maximum deflection and energy
absorption capacity), while causing a relatively small drop in the flexural strength. The hybrid
reinforcement comprising low-modulus polypropylene microfibers with different graphite
nanomaterials produced particularly pronounced gains in the mechanical properties of DSP
concrete.

336

Table 14.1- Mean values of the flexural properties of DSP pastes with graphite nanoplatelet
and/or carbon microfiber reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol%
(CMF-0.24)
Graphite nanoplatelet, 0.16 vol%
(GP-0.16)
Carbon microfiber, 0.24 vol% &
Graphite nanoplatelet, 0.16 vol%
(CMF; GP)

Mean Flexure
Strength
(MPa)
10.3
16.1

Deflection
(mm)
0.401
1.13

Energy
Absorption
(N.mm)
61.8
251

13.2

0.835

156

16.6

1.46

288

Table 14.2- Mean values and standard errors of the compressive strength test results for DSP
pastes with graphite nanoplatelet and/or carbon microfiber reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Graphite nanoplatelet, 0.16 vol% (GP-0.16)
Carbon microfiber, 0.24 vol% & Graphite
nanoplatelet, 0.16 vol% (CMF; GP)

Mean Compressive
Strength (MPa)
111
117
99.9
106

Standard Error
(MPa)
14.3
28.0
8.77
7.97

Table 14.3- Mean values and standard errors of the impact resistance test results for DSP pastes
with graphite nanoplatelet and/or carbon microfiber reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Graphite nanoplatelet, 0.16 vol% (GP-0.16)
Carbon microfiber, 0.24 vol% & Graphite nanoplatelet,
0.16 vol% (CMF; GP)

337

Impact Resistance (mm/mm)
0.365
0.521
0.781
0.937

Table 14.4- Mean values and standard errors of the abrasion weight loss test results for DSP
pastes with graphite nanoplatelet and/or carbon microfiber reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Graphite nanoplatelet, 0.16 vol% (GP-0.16)
Carbon microfiber, 0.24 vol% & Graphite nanoplatelet, 0.16 vol%
(CMF; GP)

Loss of Mass
(grams)
0.275
0.150
0.113
0.120

Table 14.5- Mean values of flexural strength, maximum deflection and energy absorption
capacity of DSP mortar with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Mean Flexure Deflection
Strength
(mm)
(MPa)
Plain
16.6
0.99
Carbon microfiber, 0.24 vol% (CMF-0.24)
19.3
1.37
Carbon microfiber, 0.48 vol% (CMF-0.48)
22.5
2.31
Carbon microfiber, 0.96 vol% (CMF-0.96)
23.1
2.54
Graphite nanoplatelets, 0.16 vol% (GP-PAA17.2
1.62
0.16)
Graphite nanoplatelets, 0.24 vol% (GP-PAA14.7
1.13
0.24)
Carbon microfiber, 0.24 vol%, & Graphite
18.3
1.68
nanoplatelets, 0.16 vol% (CMF; GP-PAA)

Energy
Absorption
(N.mm)
216
319
424
474
243
201
261

Table 14.6- Mean values and standard errors of compressive strength test results for DSP
mortar with nano- and/or micro-scale reinforcement systems.
Reinforcement Condition

Mean (MPa)

Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Carbon microfiber, 0.48 vol% (CMF-0.48)
Carbon microfiber, 0.96 vol% (CMF-0.96)
Graphite nanoplatelets, 0.16 vol% (GP-PAA-0.16)
Graphite nanoplatelets, 0.24 vol% (GP-PAA-0.24)
Carbon microfiber, 0.24 vol%, & Graphite nanoplatelets,
0.16 vol% (CMF; GP-PAA)

125
119
138
127
119
111
118

338

Std. Error
(MPa)
17.0
5.69
8.45
17.6
14.2
6.10
6.18

Table 14.7- Mean values of impact resistance test results for DSP mortar reinforced with nanoand/or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Carbon microfiber, 0.48 vol% (CMF-0.48)
Carbon microfiber, 0.96 vol% (CMF-0.96)
Graphite nanoplatelets, 0.16 vol% (GP-PAA-0.16)
Graphite nanoplatelets, 0.24 vol% (GP-PAA-0.24)
Carbon microfiber, 0.24 vol%, & Graphite nanoplatelets, 0.16
vol% (CMF; GP-PAA)

Impact Resistance
(mm/mm)
0.57
0.78
0.83
0.94
1.15
1.20
0.99

Table 14.8- Mean values of abrasion resistance test results for DSP mortar reinforced with
nano- and/or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol% (CMF-0.24)
Carbon microfiber, 0.48 vol% (CMF-0.48)
Carbon microfiber, 0.96 vol% (CMF-0.96)
Graphite nanoplatelets, 0.16 vol% (GP-PAA-0.16)
Graphite nanoplatelets, 0.24 vol% (GP-PAA-0.24)
Carbon microfiber, 0.24 vol%, & Graphite nanoplatelets, 0.16
vol% (CMF; GP-PAA)

339

Loss of Mass
(grams)
1.38
0.98
0.93
0.85
0.95
1.01
0.93

Table 14.9- Mean values of flexural attributes of plain DSP concrete and those reinforced with
different nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP0.24)
Graphite Nanoplatelet, 0.16 vol.% (GP-PAA0.16)
Polypropylene microfiber, 0.24 vol.% and
Graphite Nanoplatelet, 0.16 vol.% (PP;GPPAA)
Carbon microfiber, 0.24 vol.% and Graphite
Nanoplatelet, 0.16 vol.% (CMF;GP-PAA)

Flexural
Strength
(MPa)
13.4
12.8
13.0

Deflection
Energy
(mm)
Absorption
(N.mm)
0.71
92
1.01
111
4.55
383

14.8

1.25

181

16.4

4.78

395

16.7

1.37

219

Table 14.10- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Graphite Nanoplatelet, 0.16 vol.% (GP-PAA0.16)
Polypropylene microfiber, 0.24 vol.% and
Graphite Nanoplatelet, 0.16 vol.% (PP;GP-PAA)
Carbon microfiber, 0.24 vol.% and Graphite
Nanoplatelet, 0.16 vol.% (CMF;GP-PAA)

340

Mean
Compressive
Strength (MPa)
151
139
137
131

Standard
Error (MPa)

146

10.4

148

7.85

7.75
13.5
18.4
8.32

Table 14.11- Impact resistance test results for DSP concrete with and without nano- and/or
micro-scale reinforcement.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Graphite Nanoplatelet, 0.16 vol.% (GP-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and Graphite Nanoplatelet, 0.16
vol.% (PP;GP-PAA)
Carbon microfiber, 0.24 vol.% and Graphite Nanoplatelet, 0.16 vol.%
(CMF;GP-PAA)

Impact
Resistance
(mm/mm)
1.50
1.98
2.40
2.19
3.10
2.20

Table 14.12- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems.
Reinforcement Condition
Plain
Carbon microfiber, 0.24 vol.% (CMF-0.24)
Polypropylene microfiber, 0.24 vol.% (PP-0.24)
Graphite Nanoplatelet, 0.16 vol.% (GP-PAA-0.16)
Polypropylene microfiber, 0.24 vol.% and Graphite Nanoplatelet, 0.16
vol.% (PP;GP-PAA)
Carbon microfiber, 0.24 vol.% and Graphite Nanoplatelet, 0.16 vol.%
(CMF;GP-PAA)

341

Loss of
Mass
(grams)
1.50
1.15
1.05
0.93
0.97
0.96

(a)

(b)
Figure 14.1- SEM images of failed specimens of DSP paste reinforced with GP, providing
indications of the graphite nanoplatelet delamination and their interactions with microcracks.
342

CHAPTER 15

15 DURABILITY CHARACTERISTICS OF HIGH-PERFORMANCE
CEMENTITIOUS MATRIX REINFORCED WITH NANO AND/ OR
MICRO REINFORCMENTS
15.1 General
Unlike micro-scale fibers which do not benefit the barrier qualities of concrete, the closely
spaced nanomaterials force tortuous diffusion paths into concrete, which enhance the barrier
qualities (impermeability and diffusion/sorption resistance) of concrete (Fig. 15.1). Nano- and
micro-scale reinforcement controls the growth of defects at different scales; nano-scale
reinforcement could also reduce the rate of diffusion of moisture and aggressive chemicals into
cementitious matrices. These effects can potentially enhance the durability characteristics of
cement-based materials.
Degradation of cementitious materials with hybrid reinforcement systems could be influenced by
degradation of the reinforcement under chemical attack, fiber-matrix physical and chemical
interfacial interactions, and volume instability and cracking of the matrix under various severe
exposures and long-term environmental effects. The chemical attack on nano- and micro-scale
reinforcement can occur by either the highly alkaline (pH>13) nature of pore water, or the
diffusion of aggressive solutions into the cementitious matrix. In conventional fiber cement
composites, the ITZ (Interfacial Transition Zone) tends to be more porous initially, and subject to
continued hydration and densification over time. With dense matrices such as DSP, however,
SEM observations have not indicated the presence of a definite ITZ in matrix. While this
343

observation supports the assumption of a homogenous cement matrix (neglecting any ITZ
effect), aging of these composites seem to enhance fiber-matrix interaction upon normal aging.
This phenomenon causes an increase in the flexural strength of composites reinforced with highmodulus nano- and micro-scale reinforcement. The tailored structure of interface upon aging
would also alter the performance (bending and pullout) of fibers (and nanofibers) at cracks,
thereby refining the mechanical performance of composites (Katz and Bentur 1994; Katz and
Bentur 1995; Katz, Li et al. 1995; Katz and Bentur 1996). Depending on the nature of
accelerated aging, dimensional (and differential dimensional) movements under moisture and
temperature effects are also among the key aging mechanisms of cementitious composites.
Temperature and moisture gradients across the thickness aggravate damage mechanisms caused
by differential dimensional movements.

15.2 Materials and Methods
15.2.1 Graphite Nanomaterials and Microfibers
Polyacrylic acid (PAA) physisorbed carbon nanotubes, graphite nanoplatelets as well as (microscale) carbon and polypropylene fibers were evaluated for use in cement-based nanocomposites.
These graphite nanomaterials and microfibers have been introduced in detail in Chapter 2,
Section 2.4, Fig. 2.3 and Table 2.3. PAA physisorbed CNFs (CNT D-PAA), graphite
nanoplatelets (GP), carbon microfibers (CMF) and polypropylene microfibers (PAA) were used
in high-performance mortar and cement-based concrete matrix.

344

15.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrix selected has been introduced in detail in Chapter 2, Section 2.5. DSP
mortar and concrete was used to evaluate the reinforcement efficiency of both graphite
nanomaterials and/ or carbon or polypropylene microfibers at different volume fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon; polypropylene microfibers were prepared and cured
following ASTM procedures described in Chapter 2, Section 2.5.2.
15.2.3 Experimental Methods
In order to evaluate the durability and stability of DSP matrix with different nano- and/or microscale reinforcement, accelerated aging tests were conducted and introduced in detail in Chapter
2, Section 2.6. Freeze-Thaw tests (ASTM C666), Wet-dry tests (ASTM C1185), Hot Water
Immersion (ASTM C1185) and The Elevated Temperature tests were carried out in an electric
furnace.
During long-term exposures to different severe environments, the specimen degradation over the
course of exposure was monitored by nondestructive evaluation of the dynamic elastic modulus
using a sonometer in accordance with ASTM C215. The flexural attributes of different
specimens were also evaluated after half and full exposure in accordance with ASTM C 1185.
Experimental results were evaluated using the analysis of variance (ANOVA) and pair-wise
comparison techniques.

15.3 Test Results and Discussion of DSP Mortar Matrix

345

15.3.1 DSP Mortar with Different Nano and/ or Micro Reinforcements
15.3.1.1 Exposure to Elevated Temperature
The high-temperature exposure conditions used in our experiments simulated exposure to fire,
and subjected the specimens to high (thermal and vapor pressure) stresses. The specimens
exhibited severe spalling, disintegration and strength loss. Plain DSP mortar specimens as well
as those with graphite nanoplatelets and microfibers together with graphite nanoplatelets did not
survive this severe exposure (Fig. 15.2a-c). However, two samples each, Fig. 15.2d, of DSP
mortar reinforced with hybrid reinforcement (HYCNT1) of CNTD-PAA and CMF as well as
CMF alone survived the exposure to elevated temperature. The chances of survival of a dense
matrix like DSP at elevated temperature are very slim because of its relatively high
impermeability (which increases the internal vapor pressure upon temperature rise) and
brittleness. The fact that half of DSP mortars with microfiber and hybrid (microfiber and
nanotube) reinforcement survives exposure to elevated temperature reflects upon: (i) the gains in
DSP mortar toughness with reinforcement; (ii) increased thermal conductivity in the presence of
graphite micro- and nano-scale reinforcement, which could reduce the temperature gradient
within cementitious specimens; and (iii) combustion of graphite microfibers and nano-scale
reinforcement which could open space within cementitious matrices for release of vapor pressure
at elevated temperature.
The specimens that survived exposure to elevated temperature were saturated by immersion in
water for 24 hours, and then tested in flexure. The results are given in Table 15.1. DSP mortar
with CMF reinforcement experienced 62.3%, 61.8% and 83.0% loss of flexural strength,
maximum deflection and energy absorption capacity after exposure to high temperature. DSP
mortar with hybrid (HYCNT1) reinforcement provided improved thermal stability, experiencing

346

26.4%, 49.3% and 65.3% losses of flexural strength, maximum deflection and energy absorption
capacity, respectively. This finding reflects upon the positive effects of reinforcement at microand nano-scale upon the stability of cementitious materials under severe exposure conditions.
15.3.1.2 Freeze-Thaw Cycles
The relative dynamic elastic moduli of specimens with different reinforcement conditions after
exposure to different freeze-thaw cycles are shown in Fig. 15.3. The flexural performance
characteristics of unaged and aged specimens are summarized in Table 15.2. The dynamic
moduli test results presented in Fig. 15.3 indicate that any adverse effects of freeze-thaw cycles
on dynamic modulus is overshadowed by the positive effects associated with continued
hydration of the cementitious matrix in the presence of moisture. Similar trends are observed as
far as freeze-thaw effects on flexural performance are concerned. The gains in properties are
particularly pronounced in the case of DSP mortar with hybrid reinforcement (HYCNT1)
comprising CNTD-PAA and carbon CMF. This system experienced 14.2%, 44.5% and 36.5%
gains in flexural strength, energy absorption capacity and maximum deflection, respectively,
after freeze-thaw cycles. It is worth mentioning that the high-performance (plain) DSP mortar,
due to its desired combination of impermeability and mechanical performance, exhibits a desired
resistance to freeze-thaw attack. The benefits of nano-scale and hybrid reinforcement systems to
the barrier and mechanical qualities of DSP mortar further enhance its freeze-thaw durability.
Analysis of variance (ANOVA) and pair-wise comparison of test results indicate that, at
significance level of 0.05, only DSP mortar with hybrid reinforcement (HYCNT1) reinforcement
experiences statistically significant gains in all flexural attributes after exposure to freeze-thaw
cycles. Mortars with CMF alone and with HYGP (graphite nanoplatelet/carbon microfiber)

347

reinforcement experience statistically significant gains in flexural energy sorption capacity and
maximum deflection (but not flexural strength) after exposure to freeze-thaw cycles.
15.3.1.3 Wet-Dry Cycles
The relative dynamic elastic moduli of mortars with different reinforcement conditions after
exposure to different wet-dry cycles are shown in Fig. 15.4. The flexural performance
characteristics prior to and after exposure to freeze-thaw cycles are summarized in Table 15.3.
As was the case with freeze-thaw cycles, the elastic moduli and flexural performance
characteristics of DSP mortars (plain or with different reinforcement conditions) improve under
exposure to wet-dry cycles. This finding implies that any damage associated with wet-dry cycles
is overshadowed by the positive effects of continuous hydration of the cementitious material in
the presence of moisture. The DSP mortar with hybrid (HYCNT1) reinforcement benefits the
most from exposure to wet-dry cycles (which was also the case for freeze-thaw cycles). Mortar
with this hybrid reinforcement system experiences 7.6%, 24% and 40% gains in flexural
strength, energy absorption capacity and maximum deflection, respectively, after exposure to
300 wet-dry cycles.
ANOVA and pair-wise comparison of test results indicate that, at 0.05 significance level, only
DSP mortar with hybrid reinforcement (HYCNT1) experiences statistically significant gains in
all flexural attributes after exposure to repeated wet-dry cycles. In the case of mortar with CMF
reinforcement alone, flexural energy absorption capacity and maximum deflection (but not
flexural strength) experience statistically significant gains upon exposure to wet-dry cycles. As
far as the plain and matrix with HYGP is concerned, only the gains in maximum deflection under
wet-dry cycles are statistically significant.

348

15.3.1.4 Hot Water Immersion
The relative dynamic elastic moduli of DSP mortars with different nano- and/or micro-scale
o

reinforcement, after different intervals of immersion in hot water at 60 C, are summarized in
Fig. 15.5. The flexural performance characteristics of these mortars prior to and after immersion
in hot water are presented in Table 15.4. The trends observed here are different to those observed
for tests involving exposure to freeze-thaw and wet-dry cycles. All DSP mortar whether plain or
containing micro and/ or nano reinforcements experienced a drop in flexural strength after 56
days of hot water immersion. The other two flexural attributes were relatively not effected for
matrix containing reinforcements. Both maximum mean deflection and energy absorption
capacity showed increase in mean value. The minimum drop in flexural strength (3%) and
maximum increase in deflection (7%) and energy absorption capacity (8%) was shown by
specimens containing hybrid reinforcement (HYCNT1), containing both CNTD-PAA and CMF
and the maximum drop in flexural strength was shown by plain DSP matrix which was about
29%. At 0.05 level of significance the ANOVA and pair-wise comparison results point at a
statistically significant drop in flexural strength of plain DSP matrix and matrix containing GP
only.
15.3.2 DSP Mortar with Refined Nano and/ or Micro Reinforcements
The preferred hybrid reinforcement systems identified through response surface analysis (RSA)
based on extensive test results conducted earlier were considered here. The preferred hybrid
reinforcement systems identified through RSA, comprising CNTD-PAA and CMF included one
with 0.06 vol.% CNTD-PAA and 0.3 vol.% CMF (HYCNT2) and another with 0.08 vol.%
CNTD-PAA and 0.4 vol.% CMF (HYCNT3).

349

The preferred hybrid reinforcement system

comprising GP and CMF comprised either 0.01 vol.% GP and 0.8 vol.% CMF (HYGP2), or 0.2
vol.% GP and 0.4 vol.% CMF (HYGP3).
To evaluate the durability characteristics, two conditions were selected for evaluation of highperformance (DSP) cementitious mortar with nano- and/or micro-scale reinforcement systems
under extreme exposure/accelerated aging effects: (i) exposure to elevated temperature; and (ii)
extended immersion in hot water. The test procedures have been described earlier. Given the
high qualities of DSP cementitious materials, these conditions were found in our earlier work to
produce some detectable damage to the material.
15.3.2.1 Exposure to Elevated Temperature
Specimens reinforced with the preferred hybrid reinforcement systems comprising CNTD-PAA
and CMF, three specimens with HYCNT3 reinforcement one specimen with HYCNT2
reinforcement survived exposure to elevated temperature (Fig. 15.6b). As pointed earlier, the
chances of survival of a dense matrix like DSP at elevated temperature is very slim because of its
relatively high impermeability (which increases the internal vapor pressure upon temperature
rise) and brittleness. The fact that some DSP mortars with hybrid reinforcement survived
exposure to elevated temperature reflects upon the enhancement of durability by hybrid
reinforcement.

By tailoring these hybrid reinforcements through RSA can give higher

performance characteristics at suitable cost. Specimens that survived exposure to elevated
temperature were saturated by immersion in water for 24 hours, and then tested in flexure. The
results are summarized in Table 15.5. DSP mortar specimens with hybrid reinforcement systems
HYCNT2 (HYCNT3) experienced 42.5% (32.6%), 59.2% (51.5%) and 79.1% (69.0%) losses of
flexural strength, maximum deflection and energy absorption capacity, respectively. This finding
reflects upon the positive effects of hybrid (nano- and micro-scale) reinforcement on the fire

350

resistance of high-performance cementitious (DSP) mortar. When compared to the values of
DSP mortar reinforced with either CMF alone and HYCNT1, the results clearly indicate the
positive effect of CNTD-PAA presence in the matrix. The loss in the flexural attributes is less
with increased volume fraction of carbon nanotubes in the cementitious matrix.
15.3.2.2 Hot Water Immersion
After 56 days of hot water immersion a slight rise in elastic modulus occurs initially, Fig. 15.7,
which reverses after about 14 days. The rise tends to be greater in the case of DSP mortar
specimens with hybrid reinforcement systems comprising carbon nanotube and carbon
microfiber. The test data summarized in Table 15.6 indicate that a drop in mean flexural strength
occurs for all samples after 28 days of immersion in hot water. This drop is relatively small
(4.9% and 9%) for DSP mortar with hybrid reinforcement systems HYCNT3 and HYCNT2,
respectively. It tends to be greater (12% and 19%) for hybrid reinforcement systems (HYGP2
and HYGP3, respectively) comprising GP and carbon microfiber CMF. Other flexural attributes
were less affected by hot water immersion; maximum deflection increased in all samples, and
flexural energy absorption capacity increased in most samples after hot water immersion.
At 0.05 level of significance the ANOVA and pair-wise comparison results point at a statistically
significant drop in flexural strength and energy absorption capacity of both DSP mortars with
hybrid reinforcement comprising GP and CMF. The corresponding effects on DSP mortars with
hybrid reinforcement systems comprising CNTD-PAA and CMF, however, were not statistically
significant. This further augments that hybrid reinforcements involving carbon nanotubes
augment the durability characteristics of DSP cementitious matrix.

351

15.4 Test Results and Discussion of DSP Concrete Matrix
To assess the durability characteristics of DSP concrete matrix, plain matrix as well as matrix
containing CNTD-PAA were used at 0.16 vol.% and/ or the low modulus polypropylene (PP)
microfibers at 0.24 vol.%. Same mixing, curing and accelerated aging conditions were used as
described in Chapter 2, Section 2.5.2.
15.4.1 Exposure to Elevated Temperature
After exposure the specimens exhibited fracture, disintegration and strength loss. Plain DSP
concrete specimens as well as some specimens with PP microfibers showed severe spalling and
broke into pieces, Fig. 15.8c. However, two samples of DSP concrete containing PP microfibers
and all samples containing functionalized carbon nanotubes (CNTD-PAA) and hybrid
reinforcement (PPCNTD), Fig. 15.8b, survived the exposure to elevated temperature. The fact
that almost all DSP concrete samples with microfiber and/or hybrid (microfiber and nanotube)
reinforcement survives exposure to elevated temperature reflects upon in addition to the factors
mentioned earlier to combustion of PP microfibers reinforcement which opens space within
cementitious matrices for release of vapor pressure at elevated temperature.
The specimens that survived exposure to elevated temperature were saturated by immersion in
water for 24 hours, and then tested in flexure. The results are given in Table 15.7. DSP concrete
with polypropylene microfiber reinforcement experienced 39.1%, 74.0% and 71.0% loss of
flexural strength, maximum deflection and energy absorption capacity after exposure to high
temperature. DSP concrete with nano and hybrid (PPCNTD) reinforcement provided improved
thermal stability, experiencing 25.4%; 27.6% and 4.32%; 61.6% loss in flexural strength,
maximum deflection and energy absorption capacity, respectively. Whereas, specimens with

352

nano reinforcement observed on an average 22.8% increase in mean maximum deflection. This
finding reflects upon the positive effects of reinforcement at nano- and micro- (when used in
hybrid) scale upon the stability of cementitious materials under severe exposure conditions.
15.4.2 Freeze-Thaw Cycles
As was the case for DSP mortar, Fig. 15.9 shows no effect on the values of relative dynamic
modulus after repeated exposure to freeze thaw on plain specimen and specimens containing
CNTD-PAA. The values of relative dynamic modulus increase with time due to continuation of
hydration in wet state. However, the relative dynamic modulus for specimens containing PP
microfibers or hybrid (PPCNTD) of CNTD-PAA and PP started showing drop in value of
relative dynamic modulus. Although the value is not yet below 100 but the overall curve starts
showing a negative slope.
Table 15.8 also shows no effect of freeze-thaw cycles and an increase in all flexure attributes of
all plain as well as samples containing CNTD-PAA. This points out to the superior durability of
DSP concrete matrix and its further reinforcement by nanomaterials. The increase in flexure
attributes is the most for matrix reinforced with 0.16 vol.% CNTD-PAA. The increase in flexural
strength, energy absorption capacity and maximum deflection for this matrix is 4.8%, 0.7% and
33.3%, respectively. However, there was a drop observed in the mean values of all flexural
attributes of all the samples containing both PP microfibers and hybrid reinforcement (PP;
CNTD-PAA). The ANOVA results show a little different picture. At 95% confidence or for
values of p less than or equal to 0.05, no DSP concrete matrix whether plain or reinforced with
nano and/or micro reinforcement showed any change in flexural strength. However for matrices
containing PP microfibers and PP; CNTD, the loss in maximum deflection is statistically
significant. But the loss in energy absorption capacity is significant for DSP concrete matrix

353

containing hybrid reinforcement (PPCNTD). The main reason that matrices containing PP
microfibers showed drop in some of the flexural attributes is due to the well-known fact of fibermatrix degradation due to long-term exposure to severe freeze-thaw conditions.
15.4.3 Wet-Dry Cycles
As was the case for freeze-thaw, Fig. 15.10 shows no effect on the values of relative dynamic
modulus after repeated exposure to wet-dry cycles on plain specimen and specimens containing
CNTD-PAA. The values of relative dynamic modulus increase with time due to continuation of
hydration in wet state. However, the relative dynamic modulus for specimens containing PP
microfibers and hybrid reinforcement (PPCNTD) started showing drop in value of relative
dynamic modulus. Although the value is not yet below 100 but the overall curve starts showing a
negative slope.
Similarly, Table 15.9 show no effect of wet-dry cycles and an increase in all flexure attributes of
all plain as well as samples containing CNTD-PAA. The increase in flexure attributes is the most
for matrix reinforced with 0.16 vol.% CNTD-PAA. The increase in flexural strength, energy
absorption capacity and maximum deflection for this matrix is 9.4%, 38.3% and 39.1%,
respectively. However, there was a drop observed in the mean values of flexural strength and
energy absorption capacity of all the samples containing both PP microfibers and hybrid
reinforcement (PPCNTD). The ANOVA results show that at 95% confidence, no DSP concrete
matrix plain or reinforced showed any increase/ decrease in any of the flexural attributes
evaluated.

354

15.4.4 Hot Water Immersion
Fig. 15.11 shows no effect on the values of relative dynamic modulus after hot water immersion
for 56 days on DSP concrete plain specimen and specimens containing CNTD-PAA. The values
of relative dynamic modulus increase with time due to continuation of hydration in wet state,
especially for samples containing CNTD-PAA at 0.16 vol.%. However, the relative dynamic
modulus for specimens containing PP microfibers or with hybrid reinforcement (PPCNTD)
started showing drop in value of relative dynamic modulus. Although the value is not yet below
100 but the overall curve starts showing a negative slope.
Table 15.10 shows no effect of hot water immersion after 56 days and an increase in all flexure
attributes of all plain as well as samples containing CNTD-PAA. The increase in flexural
attributes is more pronounced for matrix reinforced with 0.16 vol.% CNTD-PAA. The increase
in flexural strength, energy absorption capacity and maximum deflection for this matrix is
17.1%, 4.3% and 20.3%,, respectively. However, there was a drop observed in the mean values
of all flexural attributes of all the samples containing both PP microfibers and hybrid
reinforcement (PPCNTD). The ANOVA results show that at 95% confidence level, no DSP
concrete matrix whether plain or reinforced with nano and/or micro reinforcement showed any
change in flexure strength. However for matrices containing PP microfibers and hybrid
reinforcement (PPCNTD), the loss in maximum deflection is statistically significant. But the loss
in energy absorption capacity is significant for DSP concrete matrix containing hybrid
reinforcement only.

355

15.5 Summary
The durability characteristics of high-performance (DSP) cementitious mortars and concrete
were evaluated under different accelerated aging effects. The results indicated indicate that the
DSP mortars provide desirable durability characteristics, which tend to be enhanced through
reinforcement with nano-scale and especially hybrid (nano- and micro-scale) reinforcement
systems. Hybrid reinforcement of DSP cementitious mortar, when used at appropriate volume
fractions, proved very effective in improving key durability characteristics of the matrix through
barrier qualities and synergistic reinforcement at different scales.
For DSP concrete as well the accelerated aging test results showed no effect on the flexural
attributes and dynamic elastic modulus of plain DSP concrete matrix and matrix containing
CNTD-PAA after 300 cycles of freeze-thaw and wet-dry and 56 days immersion in hot water.
However, matrices containing PP microfibers and hybrid (PPCNTD) of CNT-PAA and PP,
showed drop in the mean values of almost all flexural attributes when exposed to above
mentioned accelerated aging test.

356

Table 15.1- Mean values of flexural attributes of DSP mortar reinforced with micro- and/or
nano-scale reinforcement prior to and after exposure to elevated temperature (ET).
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)

CMF (0.24 vol%)
CMF (0.24 vol%), ET
HYCNT1- (CMF 0.24 vol% &
MWNTD-PAA, 0.16 vol%)
HYCNT1-(CMF, 0.24 vol% &
MWNTD-PAA, 0.16 vol%), ET

19.3
7.29
19.7

1.37
0.52
2.03

319
54.1
398

14.5

1.03

138

Table 15.2 - Mean values of the flexural attributes of DSP mortar samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 cycles of freeze-thaw (FT).
Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
16.6
0.99
216
17.0
1.15
272
19.3
1.37
319
19.9
1.77
435
&
19.7
2.03
398

Reinforcement Condition
Plain, unaged
Plain, FT
CMF (0.24 vol%), unaged
CMF(0.24 vol%), FT
HYCNT1-(CMF, 0.24 vol%
MWNTD-PAA, 0.16 vol%), unaged
HYCNT1-(CMF, 0.24 vol% &
MWNTD-PAA, 0.16 vol%), FT
GP (0.16 vol%), unaged
GP (0.16 vol%), FT
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), unaged
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), FT

22.5

2.77

575

17.2
17.8
18.3

1.62
1.83
1.68

244
305
261

19.6

2.07

355

357

Table 15.3- Mean values of the flexural attributes of DSP mortars with micro- and/or nano-scale
reinforcement prior to and after exposure to 300 wet-dry (WD) cycles.
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
16.6
0.99
216
17.1
1.15
275
19.3
1.37
319
20.3
1.91
430
&
19.7
2.03
398

Plain, unaged
Plain, WD
CMF (0.24 vol%), unaged
CMF(0.24 vol%), WD
HYCNT1-(CMF,
0.24
vol%
MWNTD-PAA, 0.16 vol%), unaged
HYCNT1-(CMF,
0.24
vol%
&
MWNTD-PAA, 0.16 vol%), WD
GP (0.16 vol%), unaged
GP (0.16 vol%), WD
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), unaged
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), WD

21.2

2.85

495

17.2
18.1
18.3

1.62
1.88
1.68

244
301
261

19.9

2.04

335

Table 15.4- Mean values of the flexural attributes of DSP mortars with micro- and/or nano-scale
reinforcement prior to and after 56 days of hot water (HW) immersion.
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
16.6
0.99
216
11.8
1.05
220
19.3
1.37
319
17.2
1.40
328
&
19.7
2.10
398

Plain, unaged
Plain, HW
CMF (0.24 vol%), unaged
CMF(0.24 vol%), HW
HYCNT1-(CMF,
0.24
vol%
MWNTD-PAA, 0.16 vol%), unaged
HYCNT1-(CMF,
0.24
vol%
&
MWNTD-PAA, 0.16 vol%), HW
GP (0.16 vol%), unaged
GP (0.16 vol%), HW
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), unaged
HYGP1-(CMF, 0.24 vol% & GP, 0.16
vol%), HW

19.1

2.05

430

17.2
13.8
18.3

1.62
1.65
1.68

244
250
261

15.5

1.69

270

358

Table 15.5- Mean values of flexural attributes of DSP mortar reinforced with preferred hybrid
reinforcement systems prior to and after exposure to elevated temperature (ET).
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
HYCNT2-(CMF, 0.30 vol%, &
21.0
2.29
441
MWNTD-PAA,
0.06
vol%),
unaged
HYCNT2-(CMF, 0.30 vol%, &
12.1
0.93
92
MWNTD-PAA, 0.06 vol%), ET
HYCNT3-(CMF, 0.40 vol%, &
21.8
2.35
457
MWNTD-PAA,
0.08
vol%),
unaged
HYCNT3-(CMF, 0.40 vol%, &
14.7
1.14
142
MWNTD-PAA, 0.08 vol%), ET

Table 15.6- Mean values of the flexural attributes of DSP mortar specimens with preferred
hybrid reinforcement systems prior to and after 56 days of immersion in hot water (HW).
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
HYCNT2-(CMF, 0.30 vol%, &
21.0
2.29
441
MWNTD-PAA, 0.06 vol%), unaged
HYCNT2-(CMF, 0.30 vol%, &
19.2
2.31
450
MWNTD-PAA, 0.06 vol%), HW
HYCNT3-(CMF, 0.40 vol%, &
21.8
2.35
457
MWNTD-PAA, 0.08 vol%), unaged
HYCNT3-(CMF, 0.40 vol%, &
20.7
2.39
473
MWNTD-PAA, 0.08 vol%)- HW
HYGP2-(CMF, 0.80 vol%, & GP,
22.7
2.36
477
0.01 vol%)
HYGP2-(CMF, 0.80 vol%, & GP,
19.9
2.38
403
0.01 vol%)- HW
HYGP3-(CMF, 0.40 vol%, & GP,
19.2
1.83
401
0.20 vol%)
HYGP3-(CMF, 0.40 vol%, & GP,
15.5
1.86
298
0.20 vol%)- HW

359

Table 15.7- Mean values of flexural attributes of DSP concrete reinforced with micro- and/or
nano-scale reinforcement prior to and after exposure to elevated temperature (ET).
Reinforcement Condition
PP(0.24 vol%)
PP(0.24 vol%) -ET
CNTD-PAA(0.16 vol%)
CNTD-PAA(0.16 vol%)-ET
PPCNTD-(PP, 0.24 vol%
CNTD-PAA, 0.16 vol%)
PPCNTD-(PP, 0.24 vol%
CNTD-PAA, 0.16 vol%)-ET

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
9.50
4.54
383
5.79
1.18
111
13.4
1.40
139
9.99
1.72
133
&
13.6
4.77
420
&

9.84

1.83

153

Table 15.8- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 cycles of freeze-thaw (FT).
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
Plain
9.89
1.04
98
Plain -FT
10.19
1.10
90
CNTD-PAA(0.16 vol%)
13.38
1.41
138
CNTD-PAA(0.16 vol%)-FT
14.02
1.42
184
PP(0.24 vol%)
9.50
4.54
383
PP(0.24 vol%) -FT
8.08
3.19
278
PPCNTD-(PP, 0.24 vol% & CNTD13.61
4.77
420
PAA, 0.16 vol%)
PPCNTD-(PP, 0.24 vol% & CNTD12.13
3.41
211
PAA, 0.16 vol%)-FT

360

Table 15.9- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 300 wet-dry cycles (WD).
Reinforcement Condition

Mean Flexural Deflection
Energy
Strength(MPa)
(mm)
Absorption(N.mm)
Plain
9.89
1.04
98
Plain -WD
10.03
1.16
124
CNTD-PAA(0.16 vol%)
13.38
1.41
138
CNTD-PAA(0.16 vol%)-WD
14.64
1.95
192
PP(0.24 vol%)
9.50
4.54
383
PP(0.24 vol%) -WD
8.88
4.84
330
PPCNTD-(PP, 0.24 vol% & CNTD13.61
4.77
420
PAA, 0.16 vol%)
PPCNTD-(PP, 0.24 vol% & CNTD31.42
5.86
355
PAA, 0.16 vol%)-WD

Table 15.10- Mean values of the flexural properties of DSP concrete samples with micro- and/or
nano-scale reinforcement before and after exposure to 56 days in hot water (HW).
Energy
Mean Flexural Deflection
Strength(MPa)
(mm)
Absorption(N.mm)
Plain
9.89
1.04
98
Plain -WD
12.02
1.10
111
CNTD-PAA(0.16 vol%)
13.38
1.41
138
CNTD-PAA(0.16 vol%)-WD
15.67
1.47
166
PP(0.24 vol%)
9.50
4.54
383
PP(0.24 vol%) -WD
8.47
3.04
273
PPCNTD-(PP, 0.24 vol% & CNTD13.61
4.77
420
PAA, 0.16 vol%)
PPCNTD-(PP, 0.24 vol% & CNTD12.71
3.30
218
PAA, 0.16 vol%)-WD
Reinforcement Condition

361

Figure 15.1- Tortuous diffusion paths forced by nano-scale reinforcement.

(a) Failed plain DSP mortar specimens
Figure 15.2- DSP mortar specimens with different reinforcement conditions after exposure to
elevated temperature.

362

Figure 15.2 (cont’d)

(b) Failed DSP mortar with hybrid reinforcement

(c) Survived DSP mortar with hybrid reinforcement

363

Plain

CMF

HYCNT1

GP

HYGP1

Relative Dynamic Elastic Modulus (%)

100.7
100.6
100.5
100.4
100.3
100.2
100.1
100
99.9
99.8

0

8

20

48

96

130

150

198

246

296

Number of Cycles, Freeze-Thaw Test
Figure 15.3- Effects of exposure to freeze-thaw on the dynamic modulus of DSP mortars with micro- and/or nano-scale reinforcement
after 300 cycles.

364

Plain

CMF

HYCNT1

GP

HYGP1

Relative Dynamic Elastic Modulus (%)

100.9
100.8
100.7
100.6
100.5
100.4
100.3
100.2
100.1
100
0

8

20

48

96

130

150

198

246

296

Number of Cycles, Wet-Dry Test
Figure 15.4- Effects of exposure to wet-dry cycles on the dynamic modulus of DSP mortars with micro- and/or nano-scale
reinforcement.

365

Plain

CMF

HYCNT1

GP

HYGP1

Relative Dynamic Elastic Modulus (%)

100.3
100.25
100.2
100.15
100.1
100.05
100
99.95
0

1

2

3

5

7

14

21

28

35

42

49

56

Number of Days, Hot Water Immersion
Figure 15.5- Effects of immersion in hot water on the dynamic modulus of DSP mortars with micro- and/or nano-scale reinforcement.

366

(a) Failed DSP mortar with hybrid reinforcement

(b) Survived DSP mortar with hybrid reinforcement
Figure 15.6- High-performance (DSP) mortar specimens with different reinforcement conditions
after exposure to elevated temperature.
367

Plain

HYGP3

HYCNT3

HYGP2

HYCNT2

Relative Dynamic Elastic Modulus (%)

100.24

100.19

100.14

100.09

100.04

99.99
0

1

2

3

5

7

14

21

28

35

42

49

56

Number of Days in Hot Water
Figure 15.7- Effects of immersion in hot water on the dynamic modulus of DSP mortars with preferred hybrid reinforcement systems
after 56 days.
368

(a) DSP concrete specimens before exposure

(b) Survived samples with nano and/or micro reinforcement
Figure 15.8- DSP concrete specimens with different reinforcement conditions after exposure to
elevated temperature.
369

Figure 15.8 (cont’d)

.
(c) Failed samples without any and with micro reinforcement

370

Plain

PP

CNTD-PAA

PPCNTD

100.4

Dynamic Elastic Modulus (%)

100.35
100.3
100.25
100.2
100.15
100.1
100.05
100
0

8

20

48

96

130

150

198

246

296

Number of Freeze Thaw Cycles
Figure 15.9- Effects of exposure to freeze-thaw on the dynamic modulus of DSP concrete with micro- and/or nano-scale reinforcement
after 300 cycles.

371

Plain

PP

CNTD

48

96

130

CNTD-PP

101

Dynamic Elastic Modulus (%)

100.9
100.8
100.7
100.6
100.5
100.4
100.3
100.2
100.1
100
0

8

20

150

198

246

296

Number of Wet Dry Cycles
Figure 15.10- Effects of exposure to wet-dry on the dynamic modulus of DSP concrete with micro- and/or nano-scale reinforcement.

372

Plain

PP

CNTD-PAA

PPCNTD

100.29

Dynamic Elastic Modulus (%)

100.24

100.19

100.14

100.09

100.04

99.99
0

1

2

3

5

7

14

21

28

35

42

49

Number of Days in Hot Water
Figure 15.11- Effects of immersion in hot water on the dynamic modulus of DSP concrete with micro- and/or nano-scale
reinforcement.

373

56

CHAPTER 16

16 EFFECTS OF MODIFIED GRAPHITE NANOMATERIALS ON THE
MOISTURE
SORPTION
CHARACTERISTICS
OF
HIGHPERFORMANCE CEMENTITIOUS MATERIALS
16.1 General
Realizing balanced gains in diverse engineering properties of concrete is a challenging
undertaking. Lowering the water-to-cementitious ratio of concrete to improve its strength, for
example, compromises the toughness of concrete. Fiber reinforcement of concrete (using
conventional fibers with micro-scale diameter) for achieving improved toughness does not
necessarily benefit the strength of concrete. Another example is latex modification of concrete
for enhancing its barrier attributes, which would not necessarily benefit the strength of concrete.
Among the additives considered so far for use in concrete, nanomaterials provide unique
qualities for bringing about balanced gains in diverse engineering properties of concrete. The
geometric, physical and mechanical properties of nanomaterials offer the potential to realize
balanced improvements in such diverse qualities of concrete as toughness, strength, barrier
attributes, abrasion resistance, thermal/electrical conductivity, and fire resistance. The
investigation reported herein evaluated the effects of graphite nanomaterials on the moisture
sorption characteristics of a high-performance concrete. These effects of nanomaterials were
compared against the corresponding effects of fibers with micro-scale diameter.
The durability of concrete in aggressive environments depends largely upon its transport
qualities (Ping, Beaudoin et al. 1991; Cwirzen and Penttala 2005). Among the primary transport

374

mechanisms (capillary sorption, permeation and diffusion), sorption of moisture via capillary
action is probably the one with the greatest impact on concrete durability (Poon, Lam et al. 1999;
Bentz 2000; Bentz, Jensen et al. 2000; Pfeifer, Moeser et al. 2009). A large fraction of concrete
in service is only partly saturated, which is due to the drying action of sun and wind. The initial
ingress of water and dissolved salts in concrete thus involves, at least partly, capillary sorption
which involves suction of water into empty capillary cavities within hydrated cement paste
(Song, Jang et al. 2007; Ahlborn, Harris et al. 2011).
The resistance of concrete to moisture sorption can be enhanced by the use of supplementary
cementitious materials (e.g., fly ash and silica fume) which tend to block capillary pores by
products of pozzolanic reactions; other measures such as reduction of water-to-cementitious
ratio, use of water-repellents or latex polymers can also lower the moisture sorption of concrete
(Sahmaran and Li 2009). Graphite nanomaterials, some of which have reached industrial-scale
production and offer viable cost structures, provide an alternative means of lowering the
moisture sorption rate of concrete by forcing water molecules to travel tortuous paths within the
hydrated cementitious paste.

16.2 Materials and Methods
16.2.1 Graphite Nanomaterials and Microfibers
Polyacrylic acid (PAA) physisorbed carbon nanotubes and graphite nanoplatelets as well as
(micro-scale) carbon and polypropylene fibers were evaluated for use in cement-based
nanocomposites. These graphite nanomaterials and microfibers have been introduced in detail in
Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. PAA physisorbed carbon nanotubes (CNT DPAA), graphite nanoplatelets (GP-PAA), carbon microfibers (CMF) and polypropylene

375

microfibers (PP) were used in high-performance paste, mortar and cement-based concrete
matrix.
16.2.2 Cementitious Matrices, Mixing and Curing Procedure
Dense cementitious matrices selected have been introduced in detail in Chapter 2, Section 2.5.
DSP paste, mortar and concrete were used to evaluate the water sorption characteristics of both
graphite nanomaterials and/ or carbon or polypropylene microfibers at different volume
fractions.
Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication) and/ or carbon; polypropylene microfibers were prepared and cured
following ASTM procedures as described in Chapter 2, Section 2.5.2.
16.2.3 Experimental Methods
The moisture sorption characteristics (capillary rise) of high-performance cement-based
materials were assessed using ASTM C1585 procedures (Fig. 2.8f). In this test (Hall 1989),
cylindrical samples with 4 in (102 mm) diameter are pre-conditioned (over 18 days) to a known
moisture content; their bottom surfaces are then exposed to liquid water, and the increase in mass
due to moisture sorption is measured. Measurement of mass during the sorption test is performed
more frequently during the first six hours after contact with water, with subsequent
measurements made daily up to fifteen days. The rate of moisture sorption and the total moisture
sorption are two key outcomes of this test method.
For the test the sides and top surface of the specimens were covered with impermeable adhesive
sheets. The bottom surface was immersed in water (1-3 mm depth), and the mass gain of the
specimen over time (due to capillary sorption of water) was measured frequently (with the tested

376

surface patted dry prior to weight measurement). Absorption, I (mm), is mass change divided by
both the cross-section area of the specimen and the density of water.

The initial rate of

absorption is defined as the slope of the line that is the best fit to I plotted against the square root
of time (minute) from 1 minute to 6 hours or until the plot shows a clear change of slope (Nick
point time). The later-age rate of water absorption (mm/min) is the slope of the line that is the
best fit to I plotted against the square root of time (min½) from 1 to 7 days.
The experimental results were subjected to statistical analysis using the analysis of variance
(ANOVA) and pair-wise comparison techniques in order to assess the statistical significance of
the effects of different reinforcement systems on the moisture sorption characteristics of highperformance cementitious materials. Insight into the structure of the cementitious
nanocomposites was gained through scanning electron microscope observations of the
specimens.

16.3 Experimental Results and Discussion
Poly-acrylic acid physisorbed CNFs (CNT D-PAA) and graphite platelets (GP-PAA) were used
as reinforcement at 0.16 vol.% (volume percent of dry cementitious materials); polypropylene
(PP)

and carbon microfibers(CMF) were used at 0.24 vol.%. Combinations of polypropylene

and carbon microfibers with PAA physisorbed carbon nanotubes and graphite nanoplatelets were
also used as hybrid reinforcement for different high-performance matrices, with PP/ CMF
microfibers and CNT D-PAA/ GP-PAA used at 0.24 vol.% and 0.16 vol.%, respectively.
16.3.1 Rate of Moisture Absorption of DSP Paste Matrix
The mean rate of moisture absorption (ROMS) and total amount of water absorbed over a 15
days period for the high-performance paste reinforced with different volume fractions of graphite

377

nanomaterials and microfibers are shown in Fig. 16.1 and Fig. 16.2, respectively. The proper
proportioning of cement and silica fume particles, contribution of silica fume in forming C-S-H
by reaction with calcium hydroxide (C-H) in the matrix through pozzolanic reaction combined
with steamed curing results in a very dense high-performance paste matrix, Fig. 16.4, which
reduce the transport of moisture by capillary rise due to presence of very fine and discontinuous
pours. The introduction of graphite nanomaterials further reduces the sorption rate and total
moisture sorption of the high-performance cementitious pastes.
The addition of Poly-acrylic acid (PAA) physisorbed graphite nanomaterials at different volume
fractions results in almost impermeable matrix. The rise in diffusion resistance with respect to
plain matrix varies form 63.7% for 0.16 vol.% CNT D-PAA to 74.3% for 0.24 vol.% of GPPAA. Unlike conventional (micro-scale) fibers, graphite nanomaterials significantly enhance the
diffusion resistance of cement-based materials. This could be attributed to the close spacing and
large surface area of graphite nanomaterials (Fig. 16.4), which significantly increase the
tortuosity of diffusion paths within the cement-based matrix. Also the presence of graphite
nanomaterials in the matrix refine the existing pore structure by providing discontinuity in a pore
at nano-scale which decreases the permeability of a cement-based matrix.
Li et.al (Li, Wang et al. 2005) also found that addition of 0.5 weight percent of MWNTs refined
the pore structure by mercury intrusion porosimetry (MIP). Konsta-Gdoutos et.al (KonstaGdoutos, Metaxa et al. 2010) also indicated that when compared to plain cement matrix, the
nanocomposites appear to have a higher amount of high stiffness C–S–H and reduced
nanoporosity.
Due to their small diameter MWNTs appear to specifically reduce the amount of fine pores.
These results are in accordance with the results found here. On the other hand, CMF at 0.24

378

vol% produced a very small improvement in diffusion resistance of about 11.7%. The mean
value of total amount of water absorbed by capillary rise after exposure of one surface to water
for 15 days also showed similar trend which were observed for the rate of moisture absorption.
The experimental results were subjected to ANOVA followed by pair-wise comparison. The
statistical analyses clearly suggested a sharp decrease in the permeability of high-performance
DSP paste matrix with addition of different volume fractions of both graphite nanomaterials, at
0.05 significance level. However, the increase in diffusion resistance of the matrix due to CMF
alone was not found to be statistically significant.
16.3.2 Rate of Moisture Absorption of DSP Mortar Matrix
The mean ROMS and total amount of water absorbed over a 15 days period for the highperformance mortar reinforced with different volume fractions of graphite nanomaterials and/ or
microfibers are shown in Fig. 16.5 and Fig. 16.6, respectively. The proper proportioning of
cement, silica fume and fine silica sand particles, combined with the factors mentioned earlier
results in a very dense matrix, Fig 16.7a and 16.7b, which reduce the transport of moisture by
capillary rise due to presence of fine and discontinuous pours. The resulting mean rate of
moisture absorption by capillary rise in the plain matrix is 0.002967 mm/min which results in a
very dense matrix. When compared to a normal mortar with out silica fume and a w/c ratio of
0.45, the increase in diffusion resistance is about 79.3%.
However, due to the addition of silica sand the increase in rate of moisture sorption due to
capillary rise with respect to plain DSP paste is 63.3%. This increase is primarily due to addition
of silica sand and the porosity is increased due to bigger sand grains (500 microns), Fig. 16.8a.
The ITZ around these grains seems more porous and a likely path for crack to progress under
load, Fig. 16.8b. However, as was shown for DSP paste matrix, the addition of both
379

functionalized graphite nanomaterials as well as their hybrid combination with CMF resulted in
drop of both rate of moisture sorption rate as well as the total water absorbed. The maximum
increase was recorded with use of 0.24 vol.% GP-PAA (66.9%). This increase in diffusion
resistance is due to the reinforcement of graphite nanomaterials of the paste portion of the matrix
as well as the ITZ with sand grains, Fig 16.7c. These results show a lot of promise for use of
such material for repair and rehabilitation work. On the other hand, CMF at 0.24 vol% produced
a very small improvement in diffusion resistance of about 5.6%.
The experimental results were subjected to ANOVA followed by pair-wise comparison. The
statistical analyses clearly suggested a sharp decrease in the permeability of high-performance
DSP mortar matrix with addition of different volume fractions of both graphite nanomaterials
and their hybrid combination with CMF, at 0.05 significance level. However, the increase in
diffusion resistance of the matrix due to CMF alone was not found to be statistically significant.
16.3.3 Rate of Moisture Absorption of DSP Concrete Matrix
The mean ROMS and total amount of water absorbed over a 15 days period for the highperformance concrete reinforced with different volume fractions of graphite nanomaterials and/
or microfibers are shown in Fig. 16.8 and Fig. 16.9, respectively. The proper proportioning of
cement, silica fume, fine silica sand particles and granite gravel, combined with the factors
mentioned earlier results in a very dense matrix, which reduce the transport of moisture by
capillary rise due to presence of fine and discontinuous pours. The resulting mean rate of
moisture absorption by capillary rise in the plain matrix is 0.009919 mm/min which results in a
very dense matrix. When compared to a high strength concrete matrix with silica fume (10,000
psi; 69 MPa) and a w/c ratio of 0.35, the increase in diffusion resistance is about 49.2%. This is
much superior to 0.19 mm/min for normal concrete with a W/C ratio of 0.4 which is suggested

380

by (Neville A.M. 1995)and other research suggested that ordinary Portland cement concrete with
W/C ratio of 0.4–0.5 would have sorptivity of about 0.23 mm/min (Tsivilis, Tsantilas et al. 2003;
Chindaprasirt, Jaturapitakkul et al. 2005). The reason for the lower sorptivity of different DSP
matrices samples may be attributed to significantly lower water to cementitious material (w/cm)
ratio, high silica fume content and absence of classic coarse aggregate. The use of silica fume
results in a denser matrix, by reducing the pore size and thickness of transition zone between
aggregate particles/ microfiber and surrounding cementitious matrix. According to the Mehta and
Monteiro (Mehta 1993), the existence of microcracks in the interfacial transition zone at the
interface with coarse aggregate is the primary reason that concrete is more permeable than the
corresponding hydrated cement paste and mortar. Also in general, everything else being the
same, the larger the aggregate size the higher the local water–cement ratio in the interfacial
transition zone and, consequently, the weaker and more permeable would be the concrete.
However, due to the addition of granite gravel the increase in rate of moisture sorption due to
capillary rise with respect to plain DSP mortar is 70%. This increase is primarily due to addition
of much bigger inclusions which are relatively more porous than the bulk past and the porosity is
increased due to bigger gravel particles (9 mm). The ITZ around these gravel particles is much
more pronounced and porous. Being weaker than the bulk paste these gravel particles are and the
ITZ around them are the likely paths for crack to progress under load. However, as was shown
for DSP paste and mortar matrix, the addition of both PAA physisorbed graphite nanomaterials
as well as their hybrid combination with PP microfibers resulted in drop of both moisture
sorption rate as well as the total water absorbed. The maximum increase was recorded with use
of 0.24 vol.% GP-PAA (63.9%). This increase in diffusion resistance is due to the reinforcement
of graphite nanomaterials of the paste portion of the matrix as well as the ITZ with sand and

381

gravel particles. These results show a lot of promise for use of such material for repair and
rehabilitation work as well as when combined with appropriate microfibers these
nanocomposites can be used in structural members with superior engineering and durability
properties. On the other hand, PP microfibers at 0.24 vol% produced a very small improvement
in diffusion resistance of about 5.7%.
The experimental results were subjected to ANOVA followed by pair-wise comparison. The
statistical analyses clearly suggested a sharp decrease in the permeability of high-performance
DSP concrete matrix with addition of different volume fractions of both graphite nanomaterials
and their hybrid combination with PP microfibers, at 0.05 significance level. However, the
increase in diffusion resistance of the matrix due to PP microfibers alone was not found to be
statistically significant.

16.4 Comparison of Graphite Nanomaterials with other Methods of Reducing
Permeability of Concrete
The durability of concrete structures is intimately related to the rate at which water is able to
penetrate the concrete. This is because concrete is susceptible to degradation through leaching,
sulfate attack, freezing-and-thawing damage, and other mechanisms such as steel corrosion in
reinforced concrete structures that depend on the ingress of water. Because cracks significantly
modify the transport properties of concrete, their presence greatly accelerates the deterioration
process. To solve this serious problem, many ways can be adopted i.e. adding water repellant
agents to the concrete, treating the top with latex modified impermeable concrete and using high
strength matrices with secondary supplementary materials like silica fume and fly ash.

382

However, all these agents have a certain service life till they are effective. A first possibility for
the improvement of the life span of concrete structures can be realized by impregnation of the
covercrete with water repellent agents (Ahlborn, Harris et al. 2011). In this method the properties
of the covercrete are optimized in a rational way. The most important parameter controlling the
effectiveness of impregnating water repellent admixtures as means of protecting concrete from
aggressive ions to cause concrete deterioration is the depth of penetration of the water repellent
agents. However, when concrete structures experience cracking in field conditions, the depth of
cracking can easily reach or exceed the full depth of penetration.
Another possibility to reduce water transportation is to modify the structural concrete properties
by adding a water repellent agent directly to the fresh mixture (Katz and Bentur 1996). Using a
water-soluble silicone based water repellent chemical admixture (Dow Corning IE-6683). This
water repellent chemical admixture is supplied by the manufacturer in liquid form with a solid
content of 40%. This type of admixture does not provide an impenetrable physical barrier, but
rather induces a chemical repulsion/repellency of the concrete to water (Sahmaran and Li 2009).
3

This water repellent admixture is usually used at a medium dosage of 13.6 kg/m (about 0.80%
by weight of total dry materials) recommended by the manufacturer and is directly added to the
fresh mixture during the mixing process. However, these chemical agents have limited life
varying form 2-5 years and are neutralized by the action different weathering agents.
Water based polymer systems are often used for improvement in the properties of plain cement
mortar or concrete. Presently, latexes of a single or combinations of polymers like polyvinyl
acetate, copolymers of vinyl acetate–ethylene, styrene– butadiene, styrene–acrylic, and acrylic
and styrene butadiene rubber emulsions are generally used. One of the limitations of these
polymer systems is that they may re-emulsify in humid alkaline conditions.
383

A effective solution to address this problem is to add graphite nanomaterials which interact with
the matrix at the nano level and increase the durability and reduce transport of the moisture
through the matrix by increasing the tortousity of the diffusion path as well as increasing the
density of the CSH phase of the matrix (Konsta-Gdoutos, Metaxa et al. 2010). When comparing
with the above solution to make concrete impermeable, graphite nanomaterials have inherent
advantage of being present in the matrix as inclusion and there effectiveness is not reduced by
different weathering agents. When comparing from price to performance ratio functionalized
graphite nanoplatelets ($10/ kg) at a dosage rate of 3 to 4 lbs/cubic meter of concrete is a very
viable solution to be used as a repair material. Also when combined with appropriate microfibers
these nanocomposites can be used in structural concrete with superior engineering and durability
properties.

16.5 Summary
The durability results generated for different high performance cementitious matrices reinforced
with different volume fractions of graphite nanomaterials and/ or microfibers indicated proper
gradation of matrix materials results in a dense high-performance matrix with much lower
porosity.
Addition of both PAA physisorbed graphite nanomaterials as well as their hybrid combination
with both microfibers resulted in drop of both moisture sorption rate as well as the total water
absorbed. Due to inherent advantages of being present in the matrix as an inclusion and being
relatively inert to all weathering agents, graphite nanomaterials, especially graphite
nanomaterials, provide a cost-effective solution for durable repair materials. When appropriately
combined with microfibers these nanocomposites can be using in structural concrete with
superior

engineering

and
384

durability

properties.

DSP
Paste
Plain

0.001

ROMS (mm/min)

CNT DPAA-0.16
vol.%
CNT DPAA-0.24
vol.%
GP-PAA0.16
vol.%

0.0005

GP-PAA0.24
vol.%
CMF-0.24
vol.%
0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
Figure 16.1- Mean ROMS by capillary rise in high-performance DSP paste matrix reinforced with and without different volume
fraction of graphite nanomaterials and microfibers.

385

Total Water Absorbed (grams)

15

10

5

C

N

C
M
T
FD
-P 0 . 2
C
4
A
N
vo
A
T
-0
D
.1 l . %
-P
6
A
vo
A
-0
l.%
D
.
SP 2 4
v
G
P- Pa ol.%
PA ste
A
Pl
G
P- -0.
a
PA 16 in
vo
A
-0
.2 l.%
4
vo
l.%

0

Reinforcment Condition
Figure 16.2- Mean total water absorbed after 15 days in high-performance DSP paste matrix
reinforced with and without different volume fraction of graphite nanomaterials and microfibers.

386

Figure 16.3- SEM images of fractured surfaces of high performance paste matrix showing dense
structure of the matrix.

Figure 16.4- Three-dimensional assembly of carbon nanotubes within a cement-based matrix.

387

0.015
DSP Mortar
Plain

ROMS, (mm/min)

CNT D-PAA0.16 vol.%
0.01
CNT D-PAA0.24 vol.%
GP-PAA-0.16
vol.%
0.005
GP-PAA-0.24
vol.%
Plain
Mortar-0.45
w/c

0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
(a)
Figure 16.5- Mean ROMS by capillary rise in high-performance DSP mortar matrix reinforced with and without different volume
fraction of graphite nanomaterials and/ or microfibers.

388

Figure 16.5 (cont’d)

0.003

DSP Mortar
Plain

ROMS, (mm/min)

CNT D-PAA0.16 vol.%
CNT D-PAA0.24 vol.%

0.002

GP-PAA-0.16
vol.%
GP-PAA-0.24
vol.%
CMF-0.24
vol.%

0.001

CMF;CNT D0.24;0.16
vol.%
CMF;GP0.24;0.16
vol.%

0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
(b)
389

Total Water Absorbed (grams)

30

20

10

C

M
F;
C
N
C TD
M
F; -P CM
G AA F
P- - - 0
PA 0.2 .2
C A- 4 ; 0 4
N 0
T .2 .16
C D- 4;0
N PA .1
T
D D A- 6
SP -P 0
.
M AA 16
or -0
G tar .24
P- P
P l
G AA ain
P- PA 0.1
A 6
-0
.2
4

0

Reinforcment Condition
Figure 16.6- Mean total water absorbed after 15 days in high-performance DSP mortar matrix
reinforced with and without different volume fraction of graphite nanomaterials and microfibers.

390

(a) Dense structure of the matrix and inclusion of silica sand particles

(b) Diversion of crack around sand particle
Figure 16.7- SEM images of fractured surfaces of high performance mortar matrix.

391

Figure 16.8 (cont’d)

(c) Close spacing of nanomaterials within the matrix

392

0.02

DSP
Concrete
Plain
CNT D-PAA0.16 vol.%

ROMS, (mm/min)

0.015

CNT D-PAA0.24 vol.%
0.01
GP-PAA-0.16
vol.%
GP-PAA-0.24
vol.%

0.005

Plain
Concrete0.35 w/c

0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
(a)
Figure 16.8- Mean ROMS by capillary rise in DSP concrete matrix reinforced with and without different volume fraction of graphite
nanomaterials and/ or microfibers.

393

Figure 16.8 (cont’d)

0.01

DSP Concrete
Plain

0.009

0.007
ROMS, (mm/min)

0.008

CNT D-PAA0.16 vol.%
CNT D-PAA0.24 vol.%
GP-PAA-0.16
vol.%

0.006
0.005

GP-PAA-0.24
vol.%

0.004

PP-0.24 vol.%
0.003
PPCNT D0.24;0.16
vol.%
PPGP0.24;0.16
vol.%

0.002
0.001
0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
(b)
394

Total Water Absorbed (grams)

90
80
70
60
50
40
30

C
N CM
T
C D- F-0
D NT PA .2
SP D A 4
C -PA -0.1
on A 6
cr -0
PP
G ete .24
P- P
;C
P la
N
T GP AA in
PP D- -P -0
;G PA AA .16
P- A- -0
PA 0.2 .2
A 4;0 4
-0
.2 .16
4;
0.
16

20

Reinforcment Condition
Figure 16.9- Average values of total water absorbed by capillary rise after 15 days in highperformance DSP concrete matrix reinforced with and without different volume fraction of
graphite nanomaterials and microfibers.

395

CHAPTER 17

17 PLANT PRODUCTION AND PLACEMENT OF NANOCOMPOSITES –
CHALLENGES AND RESULTS
17.1 General
To overcome extremely brittle failure, low tensile capacity and sensitivity to early age
microcracking high strength concrete matrices are typically provided with fiber reinforcements at
the millimeter and/or the micro scale using macro-fibers and microfibers, respectively. These
matrices however, exhibit flaws at the nano-scale, where traditional reinforcement is not
effective. Graphite Nanomaterials, including relatively low cost carbon nanofibers (CNF) and
graphite nanoplatelets (GP), present several distinct advantages as a reinforcing material for high
strength/performance cementitious composites as compared to more traditional fibers. First, they
exhibit significant greater strength and stiffness than conventional fibers, which should improve
overall mechanical behavior. Second, their higher aspect ratio is expected to effectively arrest the
nano-cracks and demand significantly higher energy for crack propagation. Thirdly, provided
these nanomaterials are uniformly dispersed, and due to their nano-scale diameter, fiber spacing
is reduced. This has opened a new field for nano-sized reinforcements that should theoretically
hinder the formation and later propagation of microcracks at the very beginning. Few attempts
have been made to add different graphite nanomaterials as reinforcement in cementitious
matrices.
Most of the works involving use of graphite nanomaterials in cementitious matrices have used
very fine matrices to evaluate the efficiency of their reinforcement and reported modest gains in
396

some of the engineering properties (Makar and Beaudoin 2004; Cwirzen, Habermehl-Cwirzen et
al. 2008; Cwirzen, Habermehl-Cwirzen et al. 2009; Makar and Chan 2009; Metaxa, KonstaGdoutos et al. 2009; Metaxa, Konsta-Gdoutos et al. 2010; Li, Wang et al. 2005). A
comprehensive approach was taken to evaluate suitably functionalized CNFs and GPs in highperformance cementitious materials of higher complexity, especially concrete. The results of
functionalized and non-functionalized graphite nanomaterials in different high performance DSP
matrices have presented in detail in earlier chapters. However, all the work done in the field of
cement-based nanocomposites is confined to laboratory evaluation.
The focus of this part of research is to take the laboratory experience and scale up the production
of these cement-based nanocomposites by tailoring their use to practical situation. In this
research a HSC (High-Strength Concrete) matrix was selected using the materials commonly
used in plant production of concrete. The matrix was suitably reinforced with polyacrylic acid
(PAA) physisorbed CNFs/ GPs and polypropylene (PP) microfibers. The larger aggregate size
and content in concrete, when compared with mortar and paste, adds to complexity of behavior
and failure modes by introducing an interfacial zone, generating interactions between aggregates
and propagating microcracks, and the need to disperse nanomaterials in the space between
aggregates.
Evaluation of the reinforcement efficiency of CNFs and GPs in a coarser cementitious matrix,
high-strength concrete, was undertaken to see the effect of increased particle size on the
interaction of nano reinforcements with the relatively coarser matrix during and after plant
production. Furthermore, micro-scale PP fibers along with their hybrid combination with CNFs
and GPs were evaluated to achieve desired balance of performance and cost efficiency. Parallel

397

laboratory specimens were also casted to compare the efficiency of mixing procedures adopted
during concrete plant production.
Cementitious materials are essentially particulate composites, which are rarely used without
aggregates (particulates). Given the micro- to millimeter-scale dimensions of the sand and gravel
particles used in HSC, hybrid reinforcements which complement the reinforcing and dimensional
stabilizing actions of particulates with multi-scale reinforcement mechanisms could produce
particularly positive effects. In the past, the concept of using hybrid (micro- and millimeterscale) reinforcement has been explored by the concrete industry for achieving balanced gains in
material properties (Moranville-Regourd 2001; O'Connell, Boul et al. 2001; Lawler, Zampini et
al. 2005; Banthia and Sappakittipakorn 2007).

17.2 Materials and Methods
17.2.1 Graphite Nanomaterials and Microfibers
Polyacrylic acid (PAA) physisorbed carbon nanofibers and graphite nanoplatelets as well as
(micro-scale) polypropylene fibers were evaluated for use in high-strength cement-based
nanocomposites. These graphite nanomaterials and microfibers have been introduced in detail in
Chapter 2, Section 2.4, Fig. 2.3 and Table 2.3. PAA physisorbed CNFs (CNF-PAA), graphite
nanoplatelets (GP-PAA) and polypropylene microfibers (PP) were used in high-performance
cement-based concrete matrix.
17.2.2 Cementitious Matrices, Dispersion, Mixing and Curing Procedure
A more practical HSC matrix was selected with a targeted compressive strength of 10,000 psi
(69 MPa). Most of the materials used are commonly used in concrete manufacturing at any major
concrete plant. Table 17.1 gives the composition of the high-strength concrete matrix as well as

398

the composition of different mixes selected and casted for evaluation of the merits of graphite
nanomaterials and microfibers in cement-based materials. Four mixes were casted at High Grade
Materials, Inc concrete plant located at Lansing, MI. The mixes included a plain without any
reinforcement, one with CNF-PAA and PP, one with GP-PAA and PP; and with PP microfibers
alone. CNF-PAA and GP-PAA nano-reinforcements were used at 0.16 volume percent (vol.%)
of the dry cementitious materials or binder (cement plus silica fume) and PP microfibers were
used at 0.24 vol.% of binder. The plant mixes were made in 1 cubic yd batches and were laid as
over lay of an existing slab. Specimens were taken from these batches to evaluate various
engineering and durability properties. Commensurate laboratory mixes were also casted using
same materials and reinforcement conditions to evaluate different engineering properties and
compare with the values of plant produced specimens. In the initial laboratory evaluation it was
ascertained that dispersing the graphite nanomaterials in up to 20% of the mixing water does not
effect the dispersion and reinforcement efficiency of the nanomaterials. Therefore, only 20% of
the initial mixing water was used for dispersion of nanomaterials in the laboratory. The
nanomaterials were provided in dispersed condition when mixing was done at the concrete plant.
The materials selected for use in cement-based matrices included Type I Portland cement
(Illinois Cement Company), densified silica fume (Elkem Materials, Inc.) with ~200 nm average
particle size, ~15 m2/g specific surface area and minimum 7-day pozzolanic activity index of
105%, 2NS river sand (High Grade Materials, Inc), gravel ranging in size from 1 mm to less than
9.5 mm, with average particle size of 3.55 mm (High Grade Materials, Inc), Polychem SPC a
polycarboxylate-based Type F ASTM C 494 complete range water reducing admixture procured
from General Resource Technology, Inc.

399

Cementitious materials (with and without functionalized graphite nanomaterials dispersed in the
mixing water via sonication as described in Chapter 2, Section 2.5.2) and/ or polypropylene
microfibers were prepared following ASTM C 192 and C 305 procedures. The specimens were
moist-cured inside molds after casting (ASTM C 192) over a 24-hour period, and then demolded
and subjected to curing in water for 27 days before testing.
17.2.3 Challenges Associated with Plant Production
In order to scale up there are different challenges involved with control on the quality including
the batch weights of mix ingredients, mixing procedure and dispersion of reinforcements within
the matrix. Some of the challenges which were faced during the plant production of these highstrength cement based nanocomposite are listed below:


The increase in the quantity of mix ingredients decreases the control on the quality and
quantity of mix ingredients during plant production. Most of the concrete plants produce
batches of 10 to 15 cubic yards where the rounding off of different ingredient weights
does not cause a major change in the engineering properties, especially for normal
strength concrete. But for high-strength concrete when using low water to cement ratio
and high cement content this rounding can have a major effect, especially when a one
cubic yard batch is prepared. A lot of care must be taken to weigh the matrix materials as
close to the calculated materials as possible and variation between the batches must be
avoided.



Most of the concrete producing plants do not handle supplementary cementitious
materials, especially silica fume, in large quantities. The required amount if required is
usually added directly to the mixing truck. This although allows control of weight of
material to be added but becomes cumbersome for large volume concrete jobs.
400



Sonication and dispersion of graphite nanomaterials in the mixing water at larger scale
present a major challenge. Thus in our laboratory work initial evaluation was done to
evaluate the minimum quantity (20% of mix water) of mix water that can be used without
effecting the dispersion and subsequent reinforcement efficiency of graphite
nanomaterials in the matrix. PAA physisorption also prove effective in dispersion of
nanomaterials in lesser quantity of mix water.



The procedure for mixing the dry ingredients and introduction of fiber reinforcements,
especially graphite nanomaterials, in to the matrix should be carefully considered. As at
most of the concrete plants the mixing is done by the mixing truck, therefore, the predispersed graphite nanomaterials were introduced into the mixing drum with 50% of the
remaining mixing water. The matrix materials including all-purpose water reducer and
remaining water were introduced into the drum slowly with the drum of the mix truck
rotating. The ingredients were mixed for 15 minutes at maximum rotation. Microfibers
were added to the mix when the ingredients have turned in to slurry and the concrete mix
was rotated for another 5 minutes before moving to lay-down site. For laboratory
specimen the procedure followed in plant production was roughly followed.



Due to large capacity of the mixing drum and small volume of the mixing batch a lot of
matrix material was attached to the central mixing shaft. One way to address this issue is
to reverse rotate the drum in a controlled fashion so that any material attached to the
mixing shaft is mixed with the main ingredients.



The specimens from the testing must be taken from the main batch both form the starting,
middle and end part of the pouring to get statistically reliable results.

401

17.2.4 Experimental Methods
The test procedures employed to determine the engineering properties of HSC included
compression tests (ASTM C 109), flexure tests (ASTM C 1185), impact tests (ASTM D 7136),
abrasion tests (ASTM C 944) and moisture absorption rate tests (ASTM C 1585). Experimental
results were evaluated using the analysis of variance (ANOVA) and pair-wise comparison
techniques to reach statistically viable conclusions.

17.3 Experimental Results and Discussion
Hybrid combination of poly-acrylic acid physisorbed carbon nanofibers (CNF-PAA) / graphite
nanoplatelets (GP-PAA) were used as reinforcement at 0.16 vol.% and polypropylene (PP)
microfibers were used at 0.24 vol.% of binder. PP microfibers were also casted alone at 0.24
vol.% of binder to assess the beneficial effects of nano-reinforcements when added to the matrix.
17.3.1 Flexural Performance
The flexural attributes test results for the HSC reinforced with different volume fractions of
graphite nanomaterials and polypropylene microfibers produced at a concrete plant and in
laboratory are summarized in Table 17.2.
A drop in the mean values of all flexural attributes was observed in samples casted from plant
mixed batches when compared with those casted in the laboratory. This is expected due to large
quantity of material mixed in a plant as compared to lab casting, rounding up or down of weight
of matrix materials during bathing and mixing in the truck mixer and improper dispersion of
fibers within the matrix. However, it was observed that all hybrid reinforcements, whether casted
in laboratory or mixed in concrete plant, produced increase in the mean values of all flexural
attributes when compared with respective plain matrices. However, samples containing

402

polypropylene microfibers when used alone showed a small drop in the mean flexural strength
both for laboratory and plant produced specimens.
Both in laboratory and plant production, the most desired balance of properties was realized with
hybrid reinforcement consisting of 0.16 vol.% CNF-PAA and 0.24 vol.% of PP microfibers. The
corresponding improvements in flexural strength, energy absorption capacity and maximum
deflection for laboratory and plant produced specimens versus respective plain concrete were
23.4; 22.4%, 252; 356% and 532; 591%, respectively. This points out that hybrid reinforcement
of brittle high-strength concrete matrices can effectively improve all the flexural attributes, thus
bringing about balanced improved at competitive cost.
The experimental data presented earlier pointed at the positive effects of hybrid (nano- and
micro-scale) reinforcement systems in high-performance concrete materials. The test data
presented here suggest that micro-scale fibers of lower modulus and lower cost could effectively
complement the reinforcing effects of polymer-wrapped CNFs and GPs in high-strength
concrete. The hybrid reinforcement systems also overcame the adverse effects of microfibers on
flexural strength. The synergistic reinforcing action of nano- and micro-scale reinforcement was
further enhanced by polymer wrapping of graphite nanomaterials.
The experimental results were subjected to statistical analysis of variance (ANOVA) followed by
pair-wise comparison. The statistical analyses indicated that statically significant improvement in
all flexural attributes was observed for all samples reinforced with CNF-PAA and PP, whether
casted in laboratory or in a concrete plant, with respect to their plain matrices at 95% confidence
level. Other reinforcement systems considered here only showed improvements in energy
absorption capacity and maximum deflection at this level. The pair-wise comparison also points
out that the drop in mean values of flexural attributes of specimens casted from the plant mix

403

with respect to those casted in a laboratory is not statistically significant. The detailed statistical
analysis points out to the effectiveness of procedures adopted during plant production and paves
the way for effective production of high-strength nanocomposites.
17.3.2 Compressive Strength
The compressive strength test results (mean values and standard errors) for HSC with different
volume fractions of graphite nanomaterials and polypropylene microfibers produced at a
concrete plant and in laboratory are summarized in Table 17.3. As was seen for all flexural
attributes a small drop in the mean values of compressive strength was also observed in samples
casted from plant mixed batches when compared with those casted in the laboratory. This is
expected due to the reasons given earlier. However, it was observed that all hybrid
reinforcements, whether casted in laboratory or mixed in concrete plant, produced increase in the
mean values of compressive strength when compared with respective plain matrices. However,
samples containing polypropylene microfibers when used alone showed a drop in the mean
compressive strength both for both laboratory and plant produced specimens. This drop in
compressive strength and flexural strength is commonly seen in cement-based matrices
reinforced with micro and macro fibers while increasing the post-peak deflection and energy
absorption capacity.
Outcomes of pair-wise comparisons indicate that the small drop in compressive strength
observed between plant produced specimens and laboratory specimens was not found
statistically significant (at 0.05 significance level). The outcome of pair-wise comparisons also
revealed that the increase in compressive strength due to addition of hybrid reinforcement is also
not statistically significant at 0.05 significance level. This shows the positive contribution of
nano-reinforcement in maintaining the desired compressive strength.

404

17.3.3 Impact Resistance
The mean values of impact resistance test results for HSC with different volume fractions of
graphite nanomaterials and polypropylene microfibers produced at a concrete plant and in
laboratory are summarized in Table 17.4. As was seen for other engineering properties a small
drop in the mean values of impact resistance was also observed in samples casted from plant
mixed batches when compared with those casted in the laboratory. This is expected due to the
reasons explained earlier. However, it was observed that all hybrid reinforcements as well as
polypropylene microfibers when used alone, whether casted in laboratory or mixed in concrete
plant, produced increase in the mean values of impact resistance when compared with respective
plain matrices. The maximum rise in impact resistance was realized with hybrid reinforcement
(87.5; 95.5%) containing 0.16 vol.% of GP-PAA and 0.24 vol.% of PP microfibers both for
laboratory and plant produced specimens. In the case of micro-scale fibers, polypropylene
produced a gain (56.6; 65.5%) in the impact resistance of high-strength concrete.
Both hybrid nano- and micro-scale reinforcement systems produced better increase in the
mechanical properties of high-strength concrete when compared with individual micro-scale
reinforcement used alone. A significant (0.000 significance level) increase in impact resistance
was observed for the hybrid reinforcement comprising polypropylene microfiber with CNFPAA, which confirms the effectiveness of hybrid reinforcement. The maximum increase in
impact resistance (115%) was brought about by hybrid use of 0.16 vol.% of CNF-PAA and 0.24
vol.% of PP microfiber. Outcomes of ANOVA and pair-wise comparisons point at the statistical
significant (0.000 to 0.012 significance levels) of nano- and/ or micro-scale reinforcement effects
on the impact resistance of high-performance concrete. When compared with high-performance
cement paste and mortar (with nano-scale reinforcement), concrete (with nano-scale

405

reinforcement) provides higher levels of impact resistance; this observation points at the wellknown contributions of aggregates (in concrete) to the toughness of cementitious matrices.
17.3.4 Abrasion Resistance
The mean values of abrasion resistance for the HSC reinforced with different volume fractions of
graphite nanomaterials and polypropylene microfibers produced at a concrete plant and in
laboratory are summarized in Table 17.5.
As was observed with other engineering properties a drop in the mean values of abrasion
resistance was also observed in samples casted from plant mixed batches when compared with
those casted in the laboratory. Detailed statistical analysis indicated that this drop in mean values
was not statistically significant at 0.05 significance level. However, it was observed that both
hybrid reinforcements, whether casted in laboratory or mixed in concrete plant, produced
increase in the mean values of abrasion resistance when compared with respective plain matrices.
The increase ranged from 8.7% for GP-PAA and PP hybrid mixed in concrete plant to 29.5% for
CNF-PAA and PP hybrid high-strength concrete samples casted in the laboratory. However,
samples containing polypropylene microfibers when used alone showed a small drop in the mean
abrasion resistance both for laboratory and plant produced test specimens. But ANOVA followed
by pair-wise comparison indicated that this drop is statistically insignificant at 95% confidence
level.
As was the case for all flexural attributes and compressive strength for both in laboratory and
plant production, the most significant increase in abrasion resistance, with respect to their
respective plain concrete, was realized in specimens with hybrid reinforcement consisting of 0.16
vol.% CNF-PAA and 0.24 vol.% of PP microfibers whether casted in field or mixed in the
concrete plant and was found to be statistically significant. The results of abrasion resistance
406

further point to the benefits of adding graphite nano-reinforcements to high-strength concrete
matrices.
17.3.5 Rate of Moisture Absorption by Capillary Rise
The mean ROMS (Rate of Moisture Absorption) and TWA (Total Water Absorbed) over a 15
days period for the high-strength concrete reinforced with different volume fractions of graphite
nanomaterials and/ or microfibers are shown in Fig. 17.1 and Fig. 17.2, respectively. As was the
case with other engineering properties the plant specimens showed a decrease in diffusion
resistance of the concrete matrix with respect to the specimens casted in the laboratory.
However, resulting mean rate of moisture absorption by capillary rise in the plain concrete
matrix is 0.0195/ 0.0161 mm/min for plant and laboratory produced specimens. This is much
superior to 0.19 mm/min for normal concrete with a W/C (water to cement) ratio of 0.4 which is
suggested by (Neville A.M. 1995)and other research suggested that ordinary Portland cement
1/2

concrete with W/C ratio of 0.4–0.5 would have sorptivity of about 0.23 mm/min

(Tsivilis,

Tsantilas et al. 2003; Chindaprasirt, Jaturapitakkul et al. 2005). The reason for the lower
sorptivity of different high strength concrete matrices samples may be attributed to lower water
to cementitious material (w/cm) ratio, high silica fume content and absence of classic coarse
aggregate. The use of silica fume results in a denser matrix, by reducing the pore size and
thickness of transition zone between aggregate particles/ microfiber and surrounding
cementitious matrix. According to the Mehta and Monteiro (Mehta 1993), the existence of
microcracks in the interfacial transition zone at the interface with coarse aggregate is the primary
reason that concrete is more permeable than the corresponding hydrated cement paste and
mortar. Also in general, everything else being the same, the larger the aggregate size the higher

407

the local water–cement ratio in the ITZ (Interfacial Transition Zone) and, consequently, the
weaker and more permeable would be the concrete.
However, due to the addition of normal gravel the increase in rate of moisture sorption due to
capillary rise with respect to plain DSP concrete is 70%. This increase is primarily due to
addition of weaker inclusions which are relatively more porous than the bulk past. The ITZ
around these gravel particles is much more pronounced and porous. Being weaker than the bulk
paste these gravel particles are and the ITZ around them are the likely paths for crack to progress
under load. However, as was shown for DSP concrete matrix, the addition of both PAA
physisorbed graphite nanomaterials as hybrid combination with PP microfibers resulted in drop
of both moisture sorption rate as well as the total water absorbed. The maximum increase was
recorded with use of 0.16 vol.% GP-PAA and 0.24 vol.% PP microfibers. The increase is 55;
57% for plant and laboratory produced specimens with respect to their plain matrices. This
increase in diffusion resistance is due to the reinforcement of graphite nanomaterials of the paste
portion of the matrix as well as the ITZ with sand and gravel particles. These results show a lot
of promise for use of such material for repair and rehabilitation work as well as when combined
with appropriate microfibers these nanocomposites can be used in structural members with
superior engineering and durability properties. On the other hand, PP microfibers at 0.24 vol%
produced a very small improvement in diffusion resistance of about 4.5; 10% for plant and
laboratory produced specimens, respectively.
The mean value of total amount of water absorbed by capillary rise after exposure of one surface
to water for 15 days also showed similar trend which were observed for the rate of moisture
absorption for the high-strength concrete matrix.

408

The experimental results were subjected to ANOVA followed by pair-wise comparison. The
statistical analyses clearly suggested a sharp decrease in the permeability of high-strength
concrete matrix with addition of different volume fractions of both graphite nanomaterials as
hybrid combination with PP microfibers, at 0.05 significance level. However, the increase in
diffusion resistance of the matrix due to PP microfibers alone was not found to be statistically
significant.
17.3.6 General Observations
All the batches of cement-based nanocomposite produced in plant were produced with a slump
of 10 to 13 inches. The final product was quite fluid and was easily laid, Fig. 17.3b. The highstrength plain matrix and those reinforced with nano and micro-reinforcements set will in
Michigan weather during the month of March. No surface cracking or other detrimental effects
were recorded, Fig. 17.3g and 17.3h. No finish treatment was applied to the slab overlay. These
segments are being watched for their behavior to Michigan weather.

17.4 Summary
The plant casting and field project highlights the challenges associated with scale up plant
production of HSC nanocomposites and points out that many procedures have to be streamlined
before scaling up the production of concrete nanocomposites from laboratory to concrete plant
production. Issues like dispersion of larger quantities of graphite nanomaterials, their
introduction into the plant batching and mixing process and due diligence to ensure proper
dispersion of both nano and micro-reinforcements must be well thought out and executed. A
small drop in mean values of all engineering properties was observed in plant produced
specimens when compared to those casted in laboratory. The most balanced and significant

409

increase in all engineering properties was observed with use of hybrid reinforcement system
including both nano and micro-scale fiber reinforcements. Nano- and micro-scale reinforcement
renders complementary reinforcing effects in high-strength concrete; optimum reinforcement
systems thus incorporate both nano- and micro-scale reinforcement systems.

410

Table 17.1- Starting material selections and mix proportions of high-strength concrete matrix for
plant production.
3

Design
Composition
Cement Type I
1013
Silica Fume
253
Water
443
Superplasticizer - oz. cwt
300
River Sand (2NS)
786
#67 Aggregate
320
Polypropylene Microfibers
4.5
Carbon Nanofibers
3
Graphite Nanoplatelets
3

Plain

PP

1040
253
440
306
789
325
-

Ingredient (lb/yd )

1030
250
445
305
790
325
4.50
-

CNFPAA+PP
1060
256
450
310
790
330
4.5
3
-

GPPAA+PP
1030
250
446
305
785
320
4.5
3

Table 17.2- Mean values of flexural attributes of plain high-performance concrete and those
reinforced with different nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain-Lab
Plain-Plant
Polypropylene microfiber, 0.24 vol.%-Lab
(PP-0.24-Lab)
Polypropylene microfiber, 0.24 vol.%-Plant
(PP-0.24-Plant)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.%-Lab (CNFPAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.%-Plant (CNFPAA+PP-Plant)
Polypropylene microfiber, 0.24 vol.% and
Graphite nanoplatelet, 0.16 vol.%-Lab (GPPAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and
Graphite nanoplatelet, 0.16 vol.%-Plant
(GP-PAA+PP-Plant)

411

Flexural
Strength
(MPa)
10.7
9.56
10.6

Deflection
Energy
(mm)
Absorption
(N.mm)
1.00
168
0.89
120
6.29
534

9.37

6.05

505

13.2

6.32

591

11.7

6.15

548

11.1

6.22

521

10.2

6.00

493

Table 17.3- Mean values and standard errors of the compressive strength test results for plain
DSP concrete and those reinforced with nano- and/ or micro-scale reinforcement systems.
Reinforcement Condition
Plain-Lab
Plain-Plant
Polypropylene microfiber, 0.24 vol.%-Lab
(PP-0.24-Lab)
Polypropylene microfiber, 0.24 vol.%-Plant
(PP-0.24-Plant)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.%-Lab (CNFPAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and
Carbon nanofiber, 0.16 vol.%-Plant (CNFPAA+PP-Plant)
Polypropylene microfiber, 0.24 vol.% and
Graphite nanoplatelet, 0.16 vol.%-Lab (GPPAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and
Graphite nanoplatelet, 0.16 vol.%-Plant (GPPAA+PP-Plant)

Mean
Compressive
Strength (MPa)
77.6
71.6
69.9

Standard
Error (MPa)

60.5

7.5

94.1

8.1

85.5

8.8

77.8

7.7

72.2

11.1

7.8
7.1
10.7

Table 17.4- Impact resistance test results for high-performance concrete with and without nanoand/or micro-scale reinforcement.
Reinforcement Condition
Plain-Lab
Plain-Plant
Polypropylene microfiber, 0.24 vol.%-Lab (PP-0.24-Lab)
Polypropylene microfiber, 0.24 vol.%-Plant (PP-0.24-Plant)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber,
vol.%-Lab (CNF-PAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber,
vol.%-Plant (CNF-PAA+PP-Plant)
Polypropylene microfiber, 0.24 vol.% and Graphite nanoplatelet,
vol.%-Lab (GP-PAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and Graphite nanoplatelet,
vol.%-Plant (GP-PAA+PP-Plant)

412

0.16

Impact
Resistance
(mm/mm)
1.20
1.10
1.88
1.82
2.10

0.16

2.05

0.16

2.25

0.16

2.15

Table 17.5- Mean abrasion weight losses of DSP concretes with and without nano- and/ or
micro-scale reinforcement systems.
Reinforcement Condition
Plain-Lab
Plain-Plant
Polypropylene microfiber, 0.24 vol.%-Lab (PP-0.24-Lab)
Polypropylene microfiber, 0.24 vol.%-Plant (PP-0.24-Plant)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber,
vol.%-Lab (CNF-PAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and Carbon nanofiber,
vol.%-Plant (CNF-PAA+PP-Plant)
Polypropylene microfiber, 0.24 vol.% and Graphite nanoplatelet,
vol.%-Lab (GP-PAA+PP-Lab)
Polypropylene microfiber, 0.24 vol.% and Graphite nanoplatelet,
vol.%-Plant (GP-PAA+PP-Plant)

413

0.16

Loss of
Mass
(grams)
3.90
4.25
3.94
4.29
2.75

0.16

3.31

0.16

3.38

0.16

3.88

0.025

HSC PlainLab
CNFPAA+PPPlt
CNFPAA+PPLab
GPPAA+PPPlt
GPPAA+PPLab
HSC PlainPlt

ROMS, (mm/min)

0.02

0.015

0.01

PP-Lab

0.005

PP-Plt
0
0

10

20

30

40

50

60

70

80

90 100 110 120 130 140 150

Time(minutes)^0.5
Figure 17.1- Mean ROMS in HSC matrix reinforced with and without different volume fraction of graphite nanomaterials and
microfibers produced in laboratory and concrete mixing plant.

414

C
N
Mean Total Water Absorbed (grams)
C F-P
N
A
F- A
PA +
P
G A+ PP
P L
G -PA P- ab
P- A P
PA + lan
A PP- t
H +P La
S P
b
H C P -P l
SC la an
Pl in- t
a i La
n- b
Pl
PP ant
PP -La
-P b
la
nt

200

150

100

50

Reinforcment Condition
Figure 17.2- Average TWA in HSC matrix reinforced with and without different volume fraction
of graphite nanomaterials and microfibers produced in laboratory and concrete mixing plant.

415

(a) Preparation of old slab for an overlay

(b) Pouring of concrete for slab
Figure 17.3- Photos of Plant Production and laying of slab overlay.
416

Figure 17.3 (cont’d)

(c) Overlay outlook of high-strength concrete after being laid

(d) Overlay outlook of high-strength concrete after being laid
417

Figure 17.3 (cont’d)

(e) Casting of specimens for evaluation

(f) Laboratory and plant specimens being cured in water baths

418

Figure 17.3 (cont’d)

(g) Finished look of HSC nanocomposites after 3 months

(h) Finished look of HSC nanocomposites after 3 months

419

CHAPTER 18

18 THEORATICAL STUDIES AND PERFORMANCE TO COST
ANALYSIS OF DSP MATRICES REINFORCED WITH GRAPHITE
NANOMATERIALS
18.1 General
Various UHPC (Ultra-High-Performance Concrete) materials have been developed over the past
two decades, with properties which approach those of lower-quality ceramics. These UHPC
materials are divided into different categories, including MDF (Macro-Defect-Free), DSP
(Densified with Small Particles), and RPC (Reactive Powder Concrete) (Cheyrezy, Maret et al.
1995; Bonneau, Vernet et al. 2000; Chan and Chu 2004). These emerging concrete materials are
distinguished by a very dense microstructure, high compressive strengths of about 150 [22 ksi]
and as high as 840 MPa [122 ksi], and superior barrier attributes and durability characteristics.
The experimental work in project focused on DSP cementitious pastes, mortars and concretes.
The compositions of DSP matrices considered in this investigation are introduced in Table 2.4,
and schematically depicted in Fig. 18.1. DSP cementitious materials offer distinct advantage in a
variety of engineering applications such as blast shelters, impact-resistant structures, nuclear
structures, skyscrapers, corrosion-proof structures and pavements. In order to improve the
strength, deformability and toughness of DSP, fibers of micro-scale diameter (e.g., relatively
short steel fibers) are used to restrain cracks in the matrix (Hu, Fang et al. 1999; Shannag and
Hansen 2000; Collepardi, Corinaldesi et al. 2002; Badanoiu, Georgescu et al. 2003).

420

Fibers enhance the mechanical characteristics of cement-based materials, especially after the
matrix has cracked, by bridging across cracks and providing some post-cracking ductility; they
benefit the fracture toughness and impact resistance of cement-based materials. Fibers of microscale diameter do not generally enhance the flexural strength of cement-based materials,
especially at common volume fractions which are below 1%; they tend to slightly lower the
compressive strength and occasionally the flexural strength of concrete.
Cementitious materials incorporate flaws at different scales; conventional fibers with micro-scale
diameter cannot interact effectively with finer flaws with nano-scale size and spacing; these finer
flaws can form and grow in the relatively large space between microfibers. The use of discrete
reinforcement with smaller dimensions in nano-scale (and nano-scale spacing) can hinder the
formation and growth of fiber flaws in cementitious materials.
This research was based upon the hypothesis that properly functionalized graphite
nanomaterials, when used in an appropriately graded matrix, can bring about substantial gains
in diverse engineering properties of high-performance cementitious materials.
Properly functionalized and thoroughly dispersed graphite nanomaterials, including multiwalled
carbon nanotubes, are used in dense DSP matrices to hinder the formation and growth of finer
flaws/cracks. These effects of nanomaterials can benefit the flexural strength of highperformance cementitious materials. Nanomaterials can also benefit the toughness and impact
resistance of cementitious materials by bridging across microcracks, energy dissipation via
frictional pullout, microcrack arrest and deflection, and blunting of crack tips. The nano-scale
spacing of graphite nanomaterials can also force tortuous diffusion paths within cementitious
pastes, thus benefiting the barrier and durability characteristics of cement-based materials. The
typically high aspect (length-to-diameter) ratios of nanomaterials also allows for their

421

percolation at relatively low volume fractions. The percolated network of nanomaterials could
raise the (thermal and electrical) conductivity and fire resistance of cementitious materials. In
short, graphite nanomaterials can be distinguished from conventional fibers (with micro-scale
diameter) by their ability to bring about balanced improvements in diverse engineering properties
of cement-based materials.

18.2 Merits of Graphite Nanomaterials
Graphite nanomaterials would play multi-faceted roles towards enhancing the mechanical,
physical and functional attributes of cementitious materials. As far mechanical attributes are
concerned, a comprehensive view of the contributions of graphite nanomaterials can be
developed by considering their spacing, debonding and frictional pullout characteristics, and the
implications for the mechanical performance of nanocomposites. Theoretical treatments of these
topics (presented below) can be used as guides toward judicious use of graphite nanomaterials in
cementitious matrices.
18.2.1 Average Spacing and Specific Surface Area of Nanomaterials in Cementitious
Matrices
For nanomaterials uniformly dispersed (with different orientation conditions) within a matrix, the
average nanomaterial spacing can be expressed as follows:
S=K.d / Vf

0.5

where, S is the average nanomaterial (or fiber) spacing, d is the nanomaterial diameter, Vf is the
nanomaterial volume fraction, and K is a constant which ranges from 0.8 to 1.38 depending on
the nanomaterial orientation(1-, 2- or 3-D) and the assumptions made in the calculation.

422

At 0.1% volume fraction, for the typical diameters introduced in Table 1.2, the average
reinforcement spacing within composites, calculated using the above equation (with K=1), are
632, 3160 and 221,200 nm for multi-walled carbon nanotube, carbon nanofiber and carbon fiber,
respectively. The average spacing of carbon fiber is thus about 350 times that of multi-walled
carbon nanotube at similar volume fractions. This provides nanotubes with superior capabilities
for control of defect size and propagation within cementitious materials.
Fig. 18.2 plots the average spacing of carbon nanotubes versus nanotube diameter for different
nanotube volume fractions (ranging from 0.1% to 1%). For typical carbon nanotube diameters of
about 10 nm or less, the nanotube spacing (at less than 1% volume fraction) fall within the nanoscale range (i.e., less than about 100 nm). This extremely close spacing of nanomaterials can be
used, with proper functionalization of graphite nanomaterials, to refine the structure and thus
properties of the hydrated cement paste.
The specific surface area (As), that is the surface area of reinforcement per unit volume of
composite, can be calculated as follows:
As = 4 Vf / d
where, Vf is the nanomaterial volume fraction and d is the nanomaterial diameter. The graphical
depiction of this expression in Fig. 18.3 highlights the substantial gains in the specific surface
area of nanotubes within matrix realized with decreasing the nanotube diameter to typical values
of about 10 nm (for multiwalled carbon nanotubes).
The close spacing of nanotubes in the cementitious matrix would have important implications in
terms of microcrack suppression and control of diffusion. The crack suppression action of
nanomaterials is schematically depicted in Fig. 18.4. Closer spacing of nanotubes would reduce

423

the flaw (microcrack) size within the matrix (size would be limited to nanotube spacing as far as
interfacial defects are controlled), and nanotubes occurring in close vicinity of microcracks
would effectively control the intensity of stress system near the microcrack tip, thus delaying the
propagation and joining of microcracks which lead to formation of larger cracks within
cementitious nanocomposites. Fig. 18.5 schematically depicts the rise in first-crack stress with
reduction of nanotube spacing.
The high specific surface area of nanomaterials would, upon proper functionalization of
nanomaterials, offer opportunities to refine the structure of hydrated cementitious paste by
providing numerous sites for nucleation of hydration products.
The potentials to control defects (microcracks) and to refine the structure of cementitious paste
with nano-scaled inclusions point at the unique capabilities of nanomaterials to render direct
reinforcing effects and also to improve the structure of the cementitious matrix.

18.3 Reinforcement Efficiency of Multiwalled Carbon Nanotubes
18.3.1 Optimum Volume Fraction of MWNTs
The effects of MWNTs D & E at different volume fractions on the flexural strength of DSP paste
were determined. MWNTs D & E, in functionalized and non-functionalized forms, were
introduced into the cementitious paste at 0.24 vol.% and 0.48 vol.%. Flexural strengths were
measured using 12.5x50x150 mm specimens under center-point loading over a span of 125 mm
in accordance with ASTM C 293 procedures. The trends in measured values of flexural strength
with increasing volume fractions of MWNTs D & E are shown in Fig. 18.6. It is clear from Fig.
18.6 that 0.24 vol.% of various nanotubes produced higher flexural strengths. However, the
longer length (100 micrometer) and higher aspect ratio of non-functionalized MWNT D

424

compensates for its inferior interfacial interactions due to lack of functionalization, producing
superior flexural strength test results when compared with its functionalized counterpart with
half the length (50 micrometer).
MWNT E, which has a larger diameter, provides less improvement in flexural strength at similar
volume fractions. The 0.48 vol.% was found to be excessive for the material compositions and
processing conditions considered here. Flexural strength generally decreases as the nanotube
volume fraction increases from 0.24% to 0.48%. It should be emphasized that volume fraction of
nanotubes is calculated here with respect to the volume of cementitious materials (cement and
silica fume). The nanotube dispersion and mixing techniques considered here cannot ensure
thorough dispersion of nanotubes beyond a certain volume fraction, causing excess nanotube
agglomeration, especially for nanotubes of smaller diameter and longer length. Scanning electron
microscope images of agglomerated MWNT E at higher volume fraction are shown in Fig. 18.7
Agglomeration greatly reduces the reinforcement efficiency of carbon nanotubes in cementbased matrices.
Larger-diameter nanotubes yield a smaller number of nanotubes per unit cross-section area of
cement nanocomposites, which is not favorable to their reinforcement efficiency in cement-based
materials. Furthermore, the comparatively shorter length of both functionalized and nonfunctionalized MWNT E, and agglomeration of nanotubes at higher volume fractions negatively
impact the flexural strength of cementitious nanocomposites. Similar trends have been observed
with flexural energy absorption capacity and ductility.
For the material selections and compositions, and the nanotube dispersion and mixing procedures
considered in this investigation, 0.24 vol.% of nanotubes seems to be the optimum, beyond
which a downtrend in the measured values of engineering properties is generally observed.

425

Similar trends were observed for the other two categories of graphite nanomaterials. Based on
the experience gained with the use of nanomaterials, alone or in combination with microfibers, in
cementitious materials, graphite nanomaterials were used at 0.24 vol.% and 0.16 vol.% in
cementitious materials without and with microfiber reinforcement. The volume fraction is
expressed here with respect to the volume of cementitious materials (cement and silica fume).

18.4 Analysis of the Debonding Process of Functionalized Carbon
Nanotubes, and Their Contribution to Tensile Strength
18.4.1 DSP Paste
SEM observations have indicated that debonding of functionalized carbon nanotubes occurs
away from the interface within the cementitious matrix. This observation suggests that the
fracture energy of the cementitious matrix is lower than that of the interface. There are close to
18

38x10

carbon atoms per square meter of carbon nanotube surface area, and the covalent bond
19

energy is of the order of 5x10-

J (Joules). With f representing the fraction of surface carbon

atoms functionalized (and capable of developing chemical bonds with the cement matrix), the
debonding energy per unit area of nanotubes (Gdb) would be:
18

-19

Gdb = (f).(38x10 ).(5x10

2

) = 19f J/m

The compact cementitious pastes considered in our investigation are relatively brittle (due to the
fine microstructure comprising largely of nanostructured calcium silicate hydrate). Considering
the percent of COOH (carboxyl) groups on MWNT D and E, their fracture energy is estimated at
2

2

about 0.475 J/m and 0.228 J/m , respectively.

426

With debonding occurring by fracture of a brittle cementitious matrix, the nanotube (average)
tensile stress (σ) associated with catastrophic debonding (neglecting any frictional shear
resistance) can be calculated as follows (see Fig. 18.8):
0.5

σ = (8 E.Gdb / d)

0.5

= 12 (f.E/d)

where, E and d are the elastic modulus and diameter of the nanotube, and Gdb is the energy
2

2

required to debond a unit surface area of the nanotube (estimated at 0.475 J/m and 0.228 J/m

for MWNT D and E, respectively, noting that debonding actually occurs within the compact
matrix).
The tensile strength and elastic modulus of multi-walled carbon nanotubes are about 300 GPa
and 1 TPa, respectively. In the context of a composite, where stress is transferred via interfacial
bonds to the outer wall of the multi-walled carbon nanotube, one may conservatively assume that
the transferred load is carried by the outer wall alone (with minimal contributions from the inner
walls which interact via weak van der Waals bonds). With a 0.34 nm thickness of each nanotube
wall, the effective (apparent) strength and elastic modulus of multi-walled nanotube would be
-9

-9

300 (1.36x10 /d) GPa and 1 (1.36x10 /d) TPa, respectively (where d is the nanotube diameter
in meter).
Based on discussions with manufacturers of multi-walled carbon nanotubes and reviews of the
relevant literature, it is ascertained that the ranges of nanotube length and diameter stated earlier
are applicable to the nanotubes in a particular batch, with a greater probability of occurrence of
nanotubes with the shorter lengths and larger diameters within the corresponding range. It has
been noted that if the aspect ratio distribution is not symmetric and/or the aspect ratio is not very
high, the predictions of reinforcement contributions based on the distribution of aspect ratio

427

could differ from those based on the mean aspect ratio by as much as 30% (Jiang, Liu et al.
2007).
Based on the above considerations, diameters of MWNT D and E were taken as 15 nm and 25
nm, respectively. For these diameters, the effective values of strength and elastic modulus would
be 27.2, 16.32 GPa, and 90.6 and 54.4 GPa, respectively. Therefore, the debonding stress can be
calculated as follows:

σMWNTD = 4.74 GPa
σMWNTE = 1.99 GPa
For nanotubes that are uniformly dispersed and randomly oriented within matrix, the number of
nanotubes per unit area, N1, can be calculated as follows:
N1 = 0.53 Vf / d

2

where, Vf is the nanotube volume fraction, and d is the nanotube diameter.
12

Functionalized MWNT D and E at 0.24% volume fraction thus represent 2.82x10
2.04x10

12

-6

/m2 nanotubes per unit area, respectively. With 3.8x10

/m2 and

N force developed in each

nanotube, assuming that all nanotubes crossing a crack are mobilized simultaneously, this force
-6

per nanotube generates a tensile stress of 3.8x10 x N1. Hence, the contribution of MWNT D
and E to tensile strength can be estimated at 10.74 MPa and 7.73 MPa at 0.24 vol.%,
respectively.

428

The above predicted contributions are more than the actual increase observed in experiments.
The SEM observations as well as the findings of other researchers indicate that factors such as
lesser than ideal distribution of nanotubes in the brittle matrix, waviness of nanotubes, their
tendency to cluster, and deficient bonding to the cement-based matrix tend to lower the actual
reinforcement efficiency of carbon nanotubes (Zhang, Picu et al. 2008).
Hence, reduction factors need to be incorporated into theoretical models in order to produce
more realistic predictions of the nanotube reinforcement efficiency. Considering the greater
tendency of higher-aspect-ratio nanotubes towards coiling and clustering, an aspect ratio
reduction factor (F) of 0.45 was introduced into our models in light of experimental observations.
The contribution of MWNTs D and E to tensile strength can now be estimated at:
MWNT D = 0.45 x 10.74 = 4.83 MPa (at 0.24 vol.%)
MWNT E = 0.45 x 7.73 = 2.31 MPa (at 0.24 vol.%)
These, results are within 10-15% of the flexural strength test results, which can be used as
approximate measures for tensile strength.

18.5 Analysis of the Frictional Pullout of Carbon Nanotubes, and their
Contribution to Fracture Energy
With longer nanotubes used in later stages of the research program, which were modified by
polymer physisorption, one may estimate the interfacial friction based on the estimated frictional
shear stress (τfu) of 10 MPa developed with the cementitious matrix (Zhang, Sun et al. 2008).
For a frictional (linear) shear stress transfer, the critical nanotube length lc (defined as the

429

minimum nanotube length required for buildup of a stress in nanotube which is equal to its
strength (as shown in Fig. 18.9) can be calculated as:
lc = σu.d/(2 τfu)
where, σu is the effective tensile strength of nanotubes, which was calculated earlier.
For MWNT D and E, the above equation yields a critical nanotube length of lc = 24 micrometer.
For l < lc, nanotube pullout dominates, and a fraction (but not all) of the nanotube tensile
strength would be mobilized during frictional pullout.
In case of PAA-physisorbed, non-functionalized MWNT D and E, average lengths are 24
micrometer and 16 micrometer, respectively, compared to critical length of 24 micrometer.
Frictional pullout would thus be an important factor, with an average pullout length of l/4. The
average nanotube tensile force at the beginning of frictional pullout for MWNT D and E would
-6

-6

be: τfu .π.d.l/4 = 2.82x10 N and 3.14 x10 N, respectively.
-6

The average frictional energy dissipation per nanotube for MWNT D and E would be: (2.36x10
-11

/2).(l/4)=1.69x10

-11

J and 1.25x10

J, respectively.

The debonding energy over the average pullout length of l/4 for MWNT D and E would be:
-13

Gdb.π.d.l/4 = 1.34x10

-13

J and 6.28x10

J, respectively. The debonding and frictional pullout
-11

energies of MWNT D and E thus add up to 1.70x10

-11

J and 1.31x10

J, respectively.

According to the Dugdale-Barenblatt model, a nanomaterial bridging zone can be assumed,
which may be the major source of energy dissipation. The contribution of nanotube debonding

430

and frictional pullout to the energy dissipation of cement nanocomposites can be estimated at
2

2

47.94 J/m and 26.65 J/m for MWNT D and E, respectively, at 0.24 vol.%. The bulk (about
90%) of the nanotube contribution to fracture energy is attributed to the frictional pullout of
nanotubes from the matrix, and the rest (about 10%) to the debonding process. These values
2

constitute significant contributions to the fracture energy (2 J/m ) of the brittle (DSP)
cementitious paste.
Besides their direct contributions (via debonding and frictional pullout) to the fracture energy of
cement-based materials, carbon nanotubes can further benefit the fracture process through
control of defect size, as explained earlier.
Closely spaced and effectively bonded carbon nanotubes would benefit the fracture process
through yet another mechanism – suppression of flaw propagation under stress. Nanotubes
occurring in the close vicinity of microcracks would control the intensity of stress system near
the microcrack tip, thus delaying the propagation and joining of microcracks which lead to crack
formation within nanocomposites.

18.6 Elastic Modulus and Tensile Strength Based on the Composite
Materials Approach
The composite materials approach yields the following expression for the elastic modulus of
carbon nanotube/cementitious matrix nanocomposites:
Ec = Em.(1-Vf) + Fl.Fo.En.Vf

431

where, Ec, Em and En are the nanocomposite, cementitious matrix and nanotube elastic moduli,
respectively, Fl and Fo are the length and orientation efficiency factors, respectively, and Vf is
the nanotube volume fraction.
For low-porosity (DSP) cementitious pastes and mortars with elastic moduli of 15 and 50 GPa,
respectively (Guerrini 2000), with an effective nanotube elastic modulus of 160 GPa, and length
and orientation efficiency factors of 0.9 and 0.2, respectively, there would be minimal effects of
carbon nanotubes (at volume fractions below 1%) on elastic modulus of cementitious
nanocomposites per composite materials approach.
In the post-cracking range of behavior, the composite materials approach predicts the tensile
strength of cementitious nanocomposites (σcu) as follows:

σcu = Fo.Vf.τfu.l/d
where, τfu is the frictional shear resistance of nanotubes in the low-porosity matrix (~10 MPa),
and Fo is the orientation factor. For MWNT D and E, assuming a post-crack orientation
efficiency factor of 0.12 and 0.135, respectively, the tensile strength of the cementitious
nanocomposite using MWNT D and E can be calculated as 4.60 MPa and 2.07 MPa,
respectively, at 0.24% nanotube volume fraction. With MWNT D at 0.16 vol.%, the tensile
strength of DSP paste was 3.07 MPa.

432

18.7 Theoretical Predictions versus Experimental Results for HighPerformance Cementitious Paste, Mortar and Concrete
18.7.1 High-Performance Paste
The theoretical models provided a basis to predict the gains in mechanical properties of the highperformance (DSP) cementitious matrix with introduction of 0.24 vol.% acid-functionalized
MWNT D and E. These theoretically predicted values were compared with experimental results.
The debonding stress (σ) of carbon nanotubes was expressed earlier as:

σ = (8 E.Gdb / d)0.5
With MWNT D and E, the cumulative contributions of nanotube debonding stresses at a section
to the tensile strength of matrix add up to 4.83 MPa and 2.31 MPa, respectively. Since tension
tests were not performed, we relied on the flexural strength test results to approximately verify
the theoretical predictions. It should, however, be noted that flexural strength depends not only
on tensile strength but also on the ductility and toughness of the material.
Experimental results suggest that the mean flexural strength of the high-performance
cementitious paste increased by 4.4 and 2.1 MPa (from 10.3 MPa at 0 vol.% to 14.7 MPa and
12.4 MPa at 0.24 vol.% of MWNT D and E, respectively). The measured gains in flexural
strength are somewhat less than, but still within the range of, the theoretically predicted gains in
tensile strength. The deviations of experimental results from theoretical predictions could have
been caused by imperfect dispersion and interfacial bonding of nanotubes in cementitious paste.
The composite materials approach, assuming an orientation efficiency factor of 0.12 and 0.135
for MWNT D and E, respectively, predicts gains of 4.60 MPa and 2.07 MPa in tensile strength.
Both these results are within 10% of the mean experimental gains. For MWNT D at 0.16 vol.%,
433

the tensile strength of DSP paste was 3.07 MPa. This value is closer to but still lower than the
actual value of 3.2 MPa for DSP paste reinforced with 0.16 vol.% of MWNT D. This trend
could be explained by the improved dispersion of nanotubes, using the stirring, sonication and
mixing procedure used in the project, at reduced nanotube concentrations.
18.7.2 High Performance Mortar and Concrete
Given its superior reinforcing attributes, MWNT D was evaluated in high-performance
cementitious mortar and concrete. As graphite nanomaterials act within the paste portion of the
matrix, contributions of nanotubes at 0.16 and 0.24 vol.% to the tensile strength of highperformance DSP mortar can be estimated to be proportional to the volume of paste within
cementitious matrix. This approach yields 7.16 MPa and 10.74 MPa contributions of nanotubes
to tensile strength at 0.16 and 0.24 vol.%, respectively. Application of reduction factors yields
the following anticipated contributions of nanotubes to the tensile strength of mortar:
MWNT D = 0.90 x 7.16 = 6.44 MPa (at 0.16 vol.%)
MWNT D = 0.55 x 10.74 = 5.91 MPa (at 0.24 vol.%)
The above values are within 10% to 15% of the experimental results produced in the project. The
value of the reduction factor was reduced to reflect the effects of sand particles on dispersion of
nanotubes and formation of interfacial transition zones. Furthermore, the reduction factors reflect
the magnified dispersion problems at volume fractions beyond 0.16%. At 0.16 vol.%, the
contribution of carbon nanotubes to tensile strength is very close to the theoretical value.
Due to economic considerations, and also for improving the dimensional stability of the
cementitious matrix, coarser aggregates were introduced into the high-performance matrix.
Although the coarser aggregate used here (granite) offers high mechanical attributes, it

434

introduces interfacial transition zones which tend to be weaker than the bulk matrix, and are also
prone to cracking. The transition zones represent preferred routes of crack propagation under
load. The contributions of nanotubes to high-performance concrete result from their reinforcing
action within the matrix and also at the ITZ. Theoretically, at 0.16 vol.%, the contribution of
nanotubes to the tensile strength of high-performance DSP concrete can be estimated at 7.16
MPa. Application of the reduction factors yields:
MWNT D = 0.5 x 7.16 = 3.58 MPa (at 0.16 vol.%)
The reduction factor accounts for the inclusion of coarser aggregate in the high-performance
matrix and formation of microcracks in the interfacial transition zone.

18.8 Synergistic Role of Nano- and Micro-scale Reinforcement
An important finding of the project concern the synergistic role that nano- and conventional
micro-scale reinforcement systems can play in enhancement of concrete properties. The
damaging effects of mechanical loads and aging are generally manifested in concrete as
inception and growth of micro-cracks within the aggregate-paste interfacial transition zones and
the bulk cementitious paste. Closely spaced graphite nanomaterials exhibit a strong presence
within the bulk paste and at interfacial zones in concrete. Their close (nano-scale) spacing and
high specific surface are make them highly effective in suppression of microcrack inception (and
initial growth) (Figures 7.9a&b) and control of microcrack propagation by effectively interacting
with their tips (Figures 7.9a&c). Longer micro-scale fibers, on the other hand, are particularly
effective in bridging across wider cracks, and mitigating their further opening (Figures 7.9a&d).
Unlike micro-scale fibers which do not benefit the barrier qualities of concrete, the closely

435

spaced nanomaterials force tortuous diffusion paths into concrete, which enhance the barrier
qualities (impermeability and diffusion/sorption resistance) of concrete.
The theoretical studies presented earlier and the experimental evidence produced in the project
yield the following conclusions:


For viable lengths of multiwalled carbon nanotubes, and the types and degrees of
functionalization considered in this investigation, debonding of nanotubes occurs at a
smaller interfacial (adhesional) shear stress (τau) when compared with frictional shear
stress (τfu). This implies that the nanotube pull-out load-deflection behavior would
follow the diagram of Fig. 18.10a (in lieu of those in Figs. 18.10b & c). In this type of
behavior, after the maximum interfacial shear stress has been exceeded, debonding will
initiate and gradually develop along the entire length of the reinforcement, without
catastrophic failure. The maximum pull-out load can be calculated as:
Pmax = 2πr. τfu
The pull-out load decreases linearly form the maximum load. It is assumed that the
frictional resistances before and after sliding (i.e. static and dynamic) are the same.
Nanotubes mainly reinforce the matrix by increasing the first-crack load through crackblunting, crack-arrest and crack-bridging. Evidences for these effects were produced in
the experimental phases of the research and SEM observations.



Microfibers are effective in cement-based matrices at relatively high volume fraction of
about 2%, below which their contribution to the flexural attributes of the matrix are
relatively small. Ultra-high-performance (DSP, RPC and DUCTAL®) concrete materials

436

generally incorporate microfiber volume fractions of about 2% and higher. Furthermore,
in a brittle cementitious matrix, much of the fiber contribution is made at a stage where
the matrix has cracked and fibers are bridging across cracks. Carbon nanotubes can
complement this action of microcracks by enhancing pre-crack performance, which
mainly results in an increase in flexural strength and, to some extent, toughness.


By using both nano- and micro-scale reinforcements at much lower volume fractions, a
load-deflection behavior similar to that shown in Fig. 18.10b can be achieved, with
improvements realized in both flexural strength and toughness. This condition promises
to yield superior flexural performance at reduced reinforcement volume fractions.

Figure 18.11 presents examples of experimental results which reflect the gains in performance
characteristics of high-performance concrete realized by the addition of 0.16 vol.%
functionalized carbon nanotube and/or 0.24 vol.% carbon (micro)fiber. These results point at the
superior reinforcement efficiency of functionalized carbon nanotube over carbon (micro)fiber,
and the synergistic action of the two in high-performance concrete at lower volume fractions
than those commonly used in fiber reinforced concrete.
Table 18.1 summarizes the gains in engineering properties of high-performance cementitious
paste, mortar and concrete resulting from reinforcement with 0.16 vol.% functionalized carbon
nanotube (CNT) and/or 0.24 vol.% carbon microfiber (CMF), or 0.16 vol.% of functionalized
graphite nanoplatelet or carbon nanofiber. Nano-scale reinforcement brings about balanced
improvements in engineering properties of cement-based materials, surpassing those offered by
micro-scale fibers. The (optimized) hybrid reinforcement comprising multiwalled carbon
nanotube and carbon (micro) fiber produces particularly large gains in engineering properties of
cement-based materials. Similar trends are observed for broader aspects of engineering
437

properties, including impermeability, fire resistance, and durability under aggressive exposures.
It should be emphasized that micro-scale fibers alone can improve the toughness and impact
resistance of concrete; their contributions to strength, abrasion resistance, impermeability and
durability are, however, limited.
The high reinforcement efficiency of nanomaterials in concrete can be explained by their
effectiveness in micro-crack suppression, control of microcrack propagation, frictional energy
dissipation during pullout, and imposition of tortuous diffusion paths within concrete.

18.9 Diffusion Resistance of Cement-Based Nanocomposites
Moisture sorption is a primary factor governing the weathering resistance and long-term
durability of cement- and concrete-based infrastructure systems. Unlike conventional (microscale) fibers, carbon nanotubes were found to significantly enhance the diffusion resistance of
cement-based materials. This could be attributed to the close spacing and large surface area of
nanotubes (Fig. 18.12), which significantly increase the tortuosity of diffusion paths within the
cement-based matrix (Fig. 18.13).
It has been suggested that the diffusivity of a random 3D fiber (or nanotube) structure is
equivalent to that of a cubical lattice of cylinders. This approach implies (Tomadakis and
Robertson 2005) that diffusivity is proportional to:
2

-1

(0.0375 d /Vf).[-ln Vf - 0.931 + (ln Vf) ]
where, d and Vf are the nanotube (or fiber) diameter and volume fraction, respectively. Fig.
18.14 presents this diffusivity factor (as a function of nanotube/fiber diameter) for 0.24%
nanotube (or fiber) volume fraction. The range of reinforcement diameter in this figure includes

438

those of multi-walled carbon nanotubes (few to several nanometers) and conventional microscale fibers (7 to 10 μm). Multi-walled carbon nanotubes are observed to offer diffusivity factors
which are about five orders of magnitude smaller than those of conventional micro-scale fibers.
In other words, reduction of reinforcement diameter to the nano-scale range makes them quite
effective in lowering the diffusivity of cement-based matrices.
The relative drops in rate of moisture sorption and total water absorbed, Fig 18.15, indicate that
properly functionalize carbon nanotubes and graphite nanoplatelets can reduce these values by
more than 2 to 3 times and 2.7 to 3.9 times, respectively.
Another approach is to define a tortuosity factor (t), that is the ratio of the actual distance that a
penetrant (or crack) must travel (for avoiding the reinforcement) to the shortest distance that
would be traveled in the absence of reinforcement, and can be expressed (for reinforcement with
nanoplatelet geometry, length L, and thickness T) as:
t =1+Vf.L/(2T)
This equation can be strictly applied only to graphite nanoplatelets; we assume that it would also
be applicable to carbon nanotubes, nanofibers and microfibers. The tortuosity factor for carbon
nanotubes, graphite nanoplatelets and carbon microfibers at 0.24% volume fraction are (for the
typical geometric attributes of these reinforcement systems used in the project) 3.08, 4.60 and
1.07, respectively. This rise in tortuosity factor is very close to the relative drops in rate of
moisture sorption and total water absorbed in the presence of these reinforcement systems, Fig
18.15.

439

18.10 Electric Conductivity and Its Damage Sensitivity
Multiwalled carbon nanotubes offer distinctly high thermal and electrical conductivity. They can
thus increase the conductivity of cement-based materials when a percolated (networked)
structure of nanomaterials forms within cement-based matrices (Fig. 18.16). The distinctly high
aspect ratio of nanomaterials substantially lowers the “threshold” volume fraction at which the
percolation phenomenon occurs. Percolated nanomaterials provide continuous paths for transport
of electricity and heat; given the relatively high aspect (length-to-diameter) ratio of carbon
nanotubes, well-dispersed nanotubes percolate at relatively low volume fractions. Break-up of
the percolated network upon cracking and development of large strains could disrupt the
continuous conduction path, and thus lower the conductivity of cementitious nanocomposites.
The damage-sensitivity of electrical conductivity offers a convenient basis for health monitoring
of cementitious nanocomposites (and concrete-based infrastructure systems incorporating them).
It is a common practice to add conductive fillers (e.g., carbon black) to insulating matrices (e.g.,
polymers) in order to increase their electrical conductivity and thus prevent electrostatic charge
buildup (by enabling continuous discharge). Upon percolation, conductive fillers provide
conduction paths within matrix (Fig. 18.17a), causing a sharp rise in conductivity of the
composite (Fig. 18.17b). In the case of conductive fillers of relatively low aspect (length-todiameter) ratio, the percolation threshold (volume fraction) is relatively high. The percolation
volume fraction can be reduced by choosing conductive inclusions of high aspect ratio (Fig.
18.16) (Farriss, Kelley et al. 1995).
The experimental results generated in the project indicated that lower nanotube dosages of about
0.04 vol.%, especially for nanotubes of shorter lengths (5-15 µm), no increase in electric
conductivity was observed in high-performance DSP paste and mortar matrices. However, 0.16

440

vol.% of modified multiwalled carbon nanotubes with longer lengths (50-100 µm) produced to
two to three orders of magnitude rise in electrical conductivity of high-performance cementitious
mortar (depending upon the superplasticizer type). Increasing the nanotube volume fraction from
0.16 to 0.24% produced another order of magnitude rise in electrical conductivity, Table 18.2.
A first-level approximation of electric conductivity can be based on the classical percolation
theory, according to which the electrical conductivity of a filled composite follows a power law
relationship (Hu, Zhao et al. 2006)

σ α (v-vc)β
where,

σ

is the nanocomposite conductivity, v is the nanowire volume fraction,

vc

is the

percolation threshold, and β is the critical exponent which relates to system dimension.
Theoretical predictions of β range from 1.6 to 2.0 for three-dimensional percolating systems; in
the case of nanocomposites incorporating carbon nanotubes, the experimental values of β (in
polymer matrices) range from 0.7 to 3.1 (Hu, Zhao et al. 2006).
The values of electrical conductivity of cementitious nanocomposites are plotted using different
values of β in Fig. 18.18. The graph shows that at 0.16 vol.% of PAA-physisorbed MWNT D,
the values of β range form 0.9 to 1.2. With increase in dosage of PAA-physisorbed MWNT D to
0.24 vol.%, the value of β drops to a range of 0.6 - 0.85. This reduction in value of β applies to
longer, suitably functionalized carbon nanotubes which effectively form a percolated network
within the insulative cementitious matrix as shown in Fig. 18.19 and explained in Chapter 8.
The damaging effects (microcracking and permanent strains) induced by compressive stresses
lower the electrical conductivity of high-performance cementitious materials incorporating
carbon nanotubes at 0.24 vol.% (Table 18.3). After application of compressive stresses
441

corresponding to 60% and 80% of compressive strength, electrical conductivity is observed to
drop by one and two orders of magnitude, respectively. The damage-sensitivity of the electrical
conductivity of high-performance cementitious materials incorporating carbon nanotubes
provides these materials with inherent health monitoring qualities (Fig 18.20) (Baoguo, Xun et
al. 2009).

18.11 Cost versus Performance of Cementitious Nanocomposites
Large-scale availability of functionalized graphite nanomaterials is a pre-requisite for their
transition to concrete markets. Fortunately, there is an explosive growth in supply of carbon
nanotubes and other graphite nanomaterials. The mid-term price targets for carbon nanotubes has
been reducing consistently, benefiting their competitive market position.
Fiber reinforcement of cement and concrete products has enjoyed commercial success over the
past two decades, with an annual growth rate of 15%. The benefits of functionalized graphite
nanomaterials to cement-based materials surpass those of conventional (e.g., polypropylene, steel
and glass) fibers in terms of both the extent of improvements and the range of properties
impacted at relatively low volume fractions of the reinforcement system. Table 18.5 compares
typical performance gains associated with the introduction (at specified volume fractions) of
functionalized carbon nanotube versus discrete steel, polypropylene and glass fibers (Bentur and
Mindess 1990; Balaguru and Shah 1992). Functionalized carbon nanotube is observed to be
distinguished from other discrete reinforcement systems by the extent and the balance of
improvements it brings about per vol.% addition to cement-based materials.
The breadth of material properties improved by functionalized graphite nanomaterials suggests
that markets for nanomaterials in cement and concrete applications can go beyond those for

442

conventional fiber reinforcement systems. While the understanding of the benefits of
functionalized graphite nanomaterials to cement- and concrete-based products is just emerging,
the trends established in the project indicate that functionalized graphite nanomaterials could
have a strong impact on the $10 billion/year markets (with 10% annual growth rate) of cement
and concrete admixtures (including fibers). According to reports prepared by BCC Associates,
the total global consumption of all types of nanomaterials reached 10.3 million tons or $20.5
billion in 2010, with an AAGR (average annual growth rate) of 9.3% in value terms. Multiwalled
carbon nanotubes are now sold in million-pound plus quantities for modifying polymers.
Nanotubes are finding growing applications in military, consumer and industrial products. Longterm applications of nanotubes can easily exceed several billion dollars, but when, is debatable.
The unique value of functionalized graphite nanomaterials to cement- and concrete-based
materials should be evaluated in light of their current and projected future costs. Over the past
two years, scale-up of multi-walled carbon nanotube production (by Arkema, Bayer Material
Science, Showa Dendo, etc.) has led to a sharp price reduction, down from thousands of dollar
per kg to $150/kg for semi-industrial applications. The run for industrial carbon nanotube
production plants has started in order to develop a sustainable business around
commercialization of nanotubes in different fields of application, with a mid-term price target of
$45/kg.
Typical sales prices (in 2008) of steel, polypropylene and alkali-resistant glass fibers are $2.5, $6
and $5.25 per kg. The specific gravity of steel, polypropylene and AR glass are 7.8, 0.91 and 2.7,
respectively. At 1% by volume of cementitious materials, the purchase price of fibers accounts
for approximately 71, 20 and 52 dollars per ton of concrete. With functionalized carbon
nanotubes at 1% volume fraction (which is for calculation purposes, and far exceeds their

443

optimum volume fraction), the purchase price of nanotubes per ton of concrete would be $300.
The normalized values of performance gains per vol.% of reinforcement were further normalized
with respect to the price (at 1 vol.%) per ton of concrete, and the results are summarized in Table
18.5. Noting that the long-term cost of multi-walled carbon nanotubes is expected to experience
significant drops below the mid-term cost used here, their high reinforcement efficiency makes
them quite cost-competitive against competing fibers even with the conservative mid-term costs
used here. It should be noted the range of properties improved by carbon nanotubes far exceeds
those improved by conventional (micro-scale) fibers.
The global markets for fibers in cement-based materials account for about $2 billion annual
sales. A fraction (about 25%) of this market makes secondary use of fibers for plastic shrinkage
crack control. The targeted markets for functionalized carbon nanotubes as primary
reinforcement in cement-based materials thus account for $1.5 billion annual sales (with growing
market penetration). Since functionalized carbon nanotubes, when compared with conventional
microfibers, benefit a broader range of concrete properties, they could compete with other
admixtures (e.g., water-repelling agents and latex polymers) for a share of the $10 billion
markets for concrete admixtures. Based on performance considerations, the fractions of fiber and
admixture markets within which functionalized carbon nanotubes are competitive can amount to
more than $2 billion annual sales (with an anticipated annual growth rate of about 10%).

18.12 Targeted Applications
Cement- and concrete-based products which could benefit from introduction of functionalized
carbon nanotubes account for nearly 11% of the U.S. construction market value ($110 billion).
Cement consumption (normalized with respect to the construction market value) has, in recent
years, experienced an annual growth rate of about 5% (Klemens 2005). The markets for fiber
444

reinforced cement and concrete, which are the primary targets for use of functionalized graphite
nanomaterials, have experienced an annual growth rate of 15% prior to the economic recession
st

during the first decade of 21 century (Li 2002). The growing market share of fibers in concrete
markets can be attributed to the rapid evolution of cement- and concrete-based materials into
complex composites which increasingly incorporate various mineral and chemical admixtures as
well as reinforcing fibers to meet the growing demands for improved performance (Aitcin 2000).
The worldwide markets for concrete admixtures and fibers are estimated at $10 billion, with an
annual growth rate of 10%, which is twice the growth rate of cement consumption.
Functionalized graphite nanomaterials surpass today’s fiber reinforcement systems in terms of
the range of properties they improve. They would thus compete not only with fiber
reinforcement systems but also with other admixtures in broader fields of application. For
example, functionalized graphite nanomaterials greatly improve the moisture barrier (and
durability) characteristics of cement-based materials, and can compete with water-repelling
admixtures and polymer emulsions in applications involving aggressive exposure conditions.
The decisions concerning the use of functionalized carbon nanotubes in consultation with
designers, contractors and materials suppliers have to rely on technical data and specifications.
Some technical specifications impacting the transition of functionalized graphite nanomaterials
to markets for cement- and concrete-based products include: (i) ASTM C 1116: Standard
Specification for Fiber-Reinforced Concrete; (ii) ASTM C 1399: Standard Test Method for
Obtaining Average Residual Strength of Fiber Reinforced Concrete; (iii) ASTM C 1550:
Standard Test Method for Flexural Toughness of Fiber Reinforced Concrete; (iv) ATM C 1609:
Standard Test Method for Flexural Performance of Fiber Reinforced Concrete; (v) ASTM C
1186: Standard Specification for Flat Non-Asbestos Fiber Cement Sheets; (vi) ASTM C 1288:

445

Standard Specification for Discrete Non-Asbestos Fiber Cement Interior Substrate Sheets; and
(vii) JCI-SF4: Method of Test for Flexural Strength and Flexural Toughness of Fiber Reinforced
Concrete. The trend away from recipe-based towards performance-based specifications within
the concrete industry has created a favorable environment for expedited introduction of new
materials based on their performance merits.
The targeted applications are of value to both the military and civilian infrastructure systems.
Applications of functionalized carbon nanotubes towards development of explosion-resistant
infrastructure represent a priority topic actively pursued by the U.S. Army (Survivability
Branch). The benefits of graphite nanomaterials to concrete-based infrastructure systems would
be of direct relevance to this program. Existing fiber cement and fiber concrete applications,
especially those with greater demands for structural performance, durability and safety, represent
viable entry points for market introduction of functionalized carbon nanotubes. These
applications are introduced in Table 1.1 together with the relevant performance considerations
and the competing fibers.
In the course of implementing this project, a pilot-scale project involving mixing in ready-mix
concrete plant and field placement of high-performance concrete nanocomposites was performed
with the assistance of High Grade Materials, a leading concrete supplier in Mid-Michigan. This
project was quite helpful towards identifying major issues related to scaled-up production of
concrete nanocomposites. Some key issues towards industrial-scale production and use of
concrete nanocomposites relate to scaled-up dispersion of graphite nanomaterials for addition to
industrial concrete mixers, control of variability in concrete nanocomposite production at large
scale, and satisfaction of broad aspects of fresh mix workability requirements.

446

18.13 Summary
Theoretical studies were conducted in conjunction with micro-structural, chemical and
crystalline analyses in order to develop insight into the reinforcement mechanisms of properly
functionalized graphite nanomaterials. The results suggested that functionalized graphite
nanomaterials improve the mechanical performance of cement-based matrices primarily through
dissipation of substantial energy by debonding and frictional pullout phenomena over their
enormous surface areas, and also through effective suppression of micro-crack size and
propagation by closely spaced nanomaterials. The gains in barrier qualities of cement-based
materials with introduction of functionalized graphite nanomaterials could be attributed to the
increased tortuosity of diffusion paths in the presence of closely spaced nanomaterials.
Uniformly dispersed graphite nanomaterials percolate within concrete at relatively low volume
fractions, producing a rise in electrical conductivity. Damage-dependence of the electrical
conductivity of concrete nanocomposites provides a convenient basis for health monitoring of
concrete-based infrastructure systems. The electrical conductivity of concrete nanocomposites
was found to be sensitive to damaging effects imposed on concrete. Concrete nanocomposites
thus offer the potential to offer inherent health monitoring qualities.
Theoretical explanations were also presented for the synergistic role of carbon nanotubes and
conventional (micro-scale) fibers towards achieving balanced gains in diverse engineering
properties of concrete.

447

Table 18.1- Typical improvements in properties of high-performance cement-based materials
(concrete/mortar/paste) with 0.16 vol.% graphite nanomaterials, 0.24 vol.% carbon microfibers,
or hybrid reinforcement comprising 0.16 vol.% functionalized carbon nanotube and 0.24 vol.%
carbon microfibers.
Reinforcement System
Multiwalled Carbon Nanotube
(CNT)
Carbon Nanofiber (CNF)
Graphite Nanoplatelet (GP)
Carbon Microfiber (CMF)
Hybrid (CNT & CMF)

Flexural
Toughness
Strength
23% / 43% / 151% / 165%
53%
/ 260%
15% / 38%
110% /
/40%
145% /215%
10% /13% / 96% / 73% /
28%
90%
1% / 14% / 21% / 48% /
10%
55%
45% / 51% / 215% / 85% /
62%
365%

Abrasion
Resistance
40% / 64% /
45%
32% / 50%
/36%
30% / 36% /
54%
23% / 30% /
19%
36% / 40% /
55%

Impact
Resistance
73% / 58% /
20%
62% / 53% /
19%
59% /100% /
114%
32% / 36% /
20%
84% / 64% /
99%

Table 18.2- Mean values of the electrical conductivity test results for high-performance
cementitious mortars with different volume fractions of modified multiwalled carbon nanotubes
prepared using two different superplasticizers.
Reinforcement Condition

Electrical
Conductivity, 1/ Ω
m

Plain

1.09 x 10-6

CNT D-PAA, 0.16 vol% - Gl#

1.19 x 10-4

CNT D-PAA, 0.16 vol% - Ad*

1.72 x 10-3

CNT D-PAA, 0.24 vol% - Gl

4.72 x 10-3

CNT D-PAA, 0.24 vol% - Ad
#
Glenium® 7700; *ADVA® Cast 575

448

2.07 x 10-2

Table 18.3- Mean values of the electrical conductivity of high-performance cementitious mortar
specimens with different volume fractions of functionalized multiwalled carbon nanotubes after
application of different compressive stress levels.
Reinforcement Condition

Electrical
Conductivitylongitudinal
direction,
1/ Ω m

Electrical
Conductivitytransverse direction,
1/ Ω m

-3

CNT D-PAA, 0.16 vol%
CNT D-PAA, 0.16 vol%, 60% ultimate

4.14 x 10

CNT D-PAA, 0.16 vol%, 80% ultimate

1.05 x 10

CNT D-PAA, 0.24 vol%

2.07 x 10

CNT D-PAA, 0.24 vol%, 60% ultimate

3.15 x 10

CNT D-PAA, 0.24 vol%, 80% ultimate

-3

1.92 x 10

4.23 x 10

2.25 x 10

-4

-4

4.72 x 10

-5

-4

1.12 x 10

-2

-2

2.80 x 10

-3

-3

4.05 x 10

-4

-4

4.15 x 10

Table 18.4- Performance gains of high-performance concrete per vol.% of reinforcement,
normalized with respect to the reinforcement cost.
Reinforcement

Flexural
Strength

Toughness

Abrasion
Resistance

Impact
Resistance

2.1

Moisture
Diffusion
Resistance
2.9

Carbon
Nanotube
Steel Fiber
Polypropylene
Fiber

1.9

8.4

0.18
1.2

5.6
8.5

0.035
0

0.035
0

2.1
2.7

449

2.5

Table 18.5- Typical performance gains of cement-based materials with introduction of the
specified volume fractions of functionalized carbon nanotube, and steel, polypropylene and glass
fibers.
Reinf. Type

Reinf.
Vol.%

Carbon
Nanotube
Steel Fiber
Polypropylene
Fiber
Glass Fiber

0.24

% Improvement in Properties (Normalized Value with respect
to Vol.%)
Flex.
Flex.
Abrasion Mois.Diff.Res.
Impact
Strength
Energy
Res.
Res.
50 (625) 200 (2,500) 50 (625)
70 (875)
60 (750)

2
0.5

25 (12.5)
20 (40)

800 (400)
100 (200)

5 (2.5)
0 (0)

5 (2.5)
0 (0)

300 (150)
250 (500)

5

50 (10)

1000 (200)

0 (0)

0 (0)

260 (52)

 
 
 
 
 
 

450

 
 
 
 
 

 

Figure 18.1- Evolution of DSP paste into DSP mortar and RPC cementitious products (Guerrini 2000).

451

 

1.2E-7

1.0E-7

8.0E-8

6.0E-8

4.0E-8

2.0E-8

0.0E+0
Nanotube Spacing, m

1.0E‐5

Vf = 0.25%
1.0E‐6
Vf = 0.5%
Vf = 0.75%
Vf = 1%

1.0E‐7

1.0E‐8
Nanotube Diameter, m
Figure 18.2- Average nanotube spacing versus nanotube diameter at different nanotube volume fractions ranging from 0.1% to 1%
(K=1.38 for 3-D random orientation of nanotubes).

452

Figure 18.3- The specific surface area of nanotubes versus nanotube diameter for different nanotube volume fractions.

453

Figure 18.4- The crack suppression effect of closely spaced nanotubes.

Figure 18.5- Schematic depiction of the relationship between first-crack stress and nanotube
spacing.

454

MWNT D-Acid Functionalized

Flexural Strength (MPa)

18
16
14
12
10
8
6
-0.10%

0.00%

0.10%

0.20%

0.30%

0.40%

Reinforcement Volume Fraction (%)
(a)
Figure 18.6- The trends in measured values of flexural strength at different volume fractions of MWNTs.

455

0.50%

Figure 18.6 (cont’d)

MWNT D- Non-Functionalized
20

Flexural Strength (MPa)

18
16
14
12
10
8
6
-0.10%

0.00%

0.10%

0.20%

0.30%

Reinforcement Volume Fraction (%)
(b)

456

0.40%

0.50%

Figure 18.6 (cont’d)

MWNT E-Acid Functionalized

Flexural Strength (MPa)

18
16
14

12
10

8
6
-0.10%

0.00%

0.10%

0.20%

0.30%

Reinforcement Volume Fraction (%)
(c)

457

0.40%

0.50%

Figure 18.6 (cont’d)

MWNT E- Non-Functionalized
20

Flexural Strength (MPa)

18
16
14
12
10
8
6
-0.10%

0.00%

0.10%

0.20%

0.30%

Reinforcement Volume Fraction (%)
(d)

458

0.40%

0.50%

(a) Clustering of functionalized MWNT E

(b) Non-functionalized MWNT E cluster
Figure 18.7- Typical SEM mages of the fractured surfaces of DSP paste reinforced with 0.48%
volume fraction of acid-functionalized and non- functionalized MWNT E.
459

Figure 18.8- The debonding model.

Figure 18.9- Interfacial shear stress distribution for critical length (lc) calculation.

460

Figure 18.10- Pullout load versus displacement for various ratios of adhesion shear to interfacial friction resistance.

461

Figure 18.11- Gains in engineering properties of high-performance concrete brought about by 0.16 vol.% functionalized carbon
nanotube and/or 0.24 vol.% carbon (micro)fiber.

462

Figure 18.12- Three-dimensional assembly of carbon nanotubes within a cement-based matrix.

Figure 18.13- Tortuous diffusion paths forced by nano-scale reinforcement.

463

Figure 18.14- Diffusivity factor versus the reinforcement diameter at 0.24% volume fraction for a diameter range covering multiwalled carbon nanotubes and conventional micro-scale fibers.

464

C

C
M
F
D C -0.
C -PA MF 24N A - M
T
- 0. o
C D-P 0.2 24- rta
N
T AA 4-C Pa r
G D -0 o st
P- -P . nc e
PA AA 24 re
t
G A-0 -0. Mo e
P- .2 24 rt
P 4 - ar
G AA -C Pas
P- -0 on t
PA . 2 c e
PP A- 4-M rete
-0 0.2 or
. 2 4- t a
4- P r
C as
on te
cr
et
e

N
T

Relative Decrease in ROMS

4

3

2

1

Reinforcement Type

Figure 18.15- Relative decrease in rate of moisture sorption for different high-performance
matrices incorporating 0.24 vol.% of different nano and micro-scale reinforcements.

465

Figure 18.16- Percolated network.

(a) Percolation of fillers
Figure 18.17- Percolation of conductive inclusions within a matrix.

466

Figure 18.17 (cont’d)

Electrical Conductivity-1/ohm-m

(b) Conductivity vs. volume fraction

0.6
0.5
MWNT D0.16
vol.%

0.4
0.3

MWNT D0.24
vol.%

0.2
0.1
0
0

0.5

1

1.5

2

Beta
Figure 18.18- Electrical conductivity values plotted for different volume fractions of PAAphysisorbed MWNT D using different β values.

467

Figure 18.19- Typical SEM image of high-performance cementitious mortar reinforced with
modified MWNTs showing uniform dispersion and percolation of nanotubes within the matrix.

 

Figure 18.20- Self-sensing smart concrete (Baoguo, Xun et al. 2009).

468

CHAPTER 19

19 CONCLUSIONS
19.1 General
Advances in concrete additives have played enabling roles towards development of high-strength
concrete materials. The rise in concrete compressive strength, however, tends to lower the
toughness and tensile-to-compressive strength ratio of concrete, and make it more prone to earlyage microcracking under restrained (drying / autogeneous) shrinkage and thermal stresses. These
issues have been addresses through introduction of discrete fibers with micrometer- to
millimeter-scale diameters. Concrete, however, incorporates flaws at different scales; finer flaws
with nano-scale size and spacing challenge the effectiveness of conventional reinforcing fibers;
these flaws can form and grow within the cementitious matrix in between fibers without
encountering them. The use of nano-scale reinforcement (with nano-scale spacing) in highperformance cementitious materials can overcome this drawback of conventional reinforcing
fibers.
Graphite Nanomaterials, including carbon nanotubes, present several distinct advantages over
conventional (micro-scale) fibers as reinforcement in high-performance cementitious materials.
Graphite nanomaterials generally provide significantly higher strength and modulus than
conventional fibers. These nanomaterials also provide distinctly high specific surface areas and
aspect ratios, which benefit their interactions with the cementitious matrix and effectiveness in
control of crack formation and propagation. Provided that nanomaterials are uniformly dispersed

469

within the cementitious matrix, due to their nano-scale cross-sectional dimensions and high
aspect ratios, the spacing between nanomaterials tends to be very small. The inception and
propagation of microcracks in the space between nano-scale reinforcement would thus be
effectively hindered.
Experimental studies were conducted on the reinforcement efficiency of multiwalled carbon
nanotubes with different aspect ratios and surface modification conditions in high-performance
cementitious materials. While the focus of the project was on multiwalled carbon nanotubes,
graphite nanoplatelets and carbon nanofibers were also evaluated. The surface modification
conditions considered involved (acid) functionalization of carbon nanotubes and/or anchoring
different polymers via chemical bonding or physical adsorption as well as microwave irradiation.
Comparative evaluations of different surface modification conditions led to the development of
an effective nanotube surface modification procedure which favored the dispersion and
interfacial interactions of nanotubes in cement-based materials. With proper surface
modification, nanotubes of higher aspect ratios could be dispersed within cementitious materials
at larger dosages for achieving pronounced, balanced gains in diverse engineering properties.
This research has laid a solid foundation for development of a new class of high-performance
concrete materials with a unique balance of engineering properties at competitive cost.
Comprehensive experimental studies were conducted to tailor the selection and proportioning of
concrete-making materials for effective mobilization of the distinct qualities of modified
multiwalled carbon nanotubes towards achieving balanced gains in engineering properties.
Smooth gradation of cementitious particles was considered to benefit the dispersion of
nanomaterials; this measure, together with pozzolanic reactions, produced dense cement hydrates
which effectively interacted with the modified nanotube surfaces.

470

Experimental investigations were undertaken in order to identify optimum (nano- and microscale) reinforcement systems for high-performance concrete. A comprehensive experimental data
base was developed on the mechanical, physical and durability characteristics as well as the
structure and failure mechanisms of high-performance cementitious nanocomposites reinforced
with modified graphite nanomaterials. Thorough dispersion and effective interfacial interactions
of graphite nanomaterials within concrete, enabled by proper surface modification of
nanomaterials, led to major gains in the engineering properties of high-performance concrete at
relatively low dosages of nanomaterials. Another unique feature of modified nanomaterials was
their ability to bring about balanced gains in diverse engineering properties of high-performance
concrete. Significant gains in fracture toughness and impact resistance were accompanied with
major rise in strength, barrier qualities, abrasion and weathering resistance, fatigue life and
weathering resistance of high-performance concrete.
Graphite nanomaterials percolate within concrete at relatively low volume fractions, causing a
rise in electrical conductivity. Damage-sensitivity of the electrical conductivity of concrete
nanocomposites provides a convenient basis for health monitoring of concrete-based
infrastructure systems. The significant gains in diverse aspects of concrete properties brought
about by low dosages of nanomaterials, combined with the growth in their industrial-scale
production at reduced costs; enable development of a new class of high-performance concrete
materials with unique balance of qualities at competitive cost. A highly desired synergism was
found between the contributions of nano- and micro-scale reinforcement to different engineering
properties of concrete.

471

19.2 Key Findings
The following key conclusions are drawn based on the work conducted in the project:


Past investigations on the use of graphite nanomaterials in cement-based matrices have
been conducted mostly in cementitious pastes, and the test results do not provide a
consistent view of the reinforcement efficiency of nanomaterials in cement-based
matrices.



The initial feasibility studies conducted in the project indicated that:
o

Surface modification is a critical step towards enhancement of the reinforcement
efficiency of graphite nanomaterials in cement-based matrices by improving the
dispersion and interfacial interactions of nanomaterials within matrix.

o

Modified graphite nanomaterials at relatively low volume fractions (<0.1 vol.%)
offer superior reinforcement efficiency when compared with micro-scale (carbon)
fibers and unmodified nanomaterials.

o

Modified graphite nanomaterials yield uniquely balanced gains in the engineering
properties of cement-based matrices; these properties range from flexural
strength, energy absorption capacity and impact resistance to abrasion resistance
and moisture barrier qualities, which cannot be matched by micro-scale fiber
reinforcement systems.

o

The improvements in mechanical properties of cement-based matrices with
introduction of graphite nanomaterials can be attributed to:


Significant energy dissipation associated with debonding and frictional
pullout over the enormous surface area of nanomaterials; and
472



Effective suppression of micro-crack size and propagation by the closely
spaced nanomaterials.

o

The gains in moisture barrier qualities (against capillary sorption) of cementbased matrices with introduction of modified graphite nanomaterials can be
attributed to the tortuous diffusion paths in the presence of closely spaced
nanomaterials.

o

The introduction of modified (and unmodified) graphite nanomaterials at <0.1
vol.% does not fundamentally alter the structure of cement hydration products.



The surface modification conditions and the aspect ratio of multiwalled carbon
nanotubes determine their reinforcement efficiency in high-performance cementitious
matrices.
o

Among the surface modification conditions considered, physisorption of
poly(acrylic acid) (PAA) on multiwalled carbon nanotubes produced the most
consistent favorable effects.

o

The best balance of mechanical properties was realized when longer multiwalled
carbon nanotubes were modified through PAA physisorption without acidfunctionalization of nanotubes prior to introduction of PAA; it seems that any
benefits of acid-functionalization prior to PAA-physisorption are more than
compensated for by their damaging effects (e.g., shortening of carbon nanotubes).

o

Increased nanotube length and aspect ratio benefited their reinforcement
efficiency in high-performance cementitious matrices; excess nanotube lengths,

473

however, complicate their dispersion in the mixing water of cementitious matrix
at required concentrations.
o

The benefits realized by proper surface modification of nanotubes can be
explained by their contributions towards uniform dispersion in aqueous media,
and improved interfacial interactions with cement hydrates.

o

The extent of surface modification as well as the concentration of nanotubes in
cementitious matrix should be optimized, depending upon the composition and
processing conditions of cementitious nanocomposites, in order to avoid
agglomeration of nanotubes and their potential interferences with the hydration
process of cement.



The experimental results generated for cementitious pastes reinforced with different
unmodified and modified multiwalled carbon nanotubes, and also their hybrids with
carbon microfibers at higher volume fractions, yielded the following conclusions:
o

The length, diameter and surface modification conditions (as well as the
mechanical attributes) of carbon nanotubes all have critical effects on their
reinforcement efficiency in cement-based matrices.


Acid functionalization of nanotubes could introduce new defects, and
reduce the length of nanotubes. These adverse effects of functionalization
may overshadow its positive contributions to the dispersion and interfacial
bonding and thus the reinforcement efficiency of nanotubes.



Use of polymer-wrapped (PAA physisorbed) multiwalled carbon
nanotubes enhances the dispersion and interfacial interactions of

474

nanotubes within cementitious matrix without introducing defects or
shortening the nanotubes.
o

Upon acid-functionalization, the larger-diameter multiwalled carbon nanotubes
considered here provided a lower degree of functionalization (1.2% COOH
groups) when compared with the smaller-diameter nanotubes (2.5% COOH
groups). The finer, higher-aspect-ratio nanotubes, in both functionalized and nonfunctionalized conditions, produced (at 0.24% by volume of dry cementitious
materials) more pronounced gains in the engineering properties of the highperformance cementitious (densified with small particles - DSP) paste considered
in the project. The high specific surface area and the nano-scale spacing of
dispersed nanotubes benefit their microcrack arrest action, which explains their
reinforcement efficiency in the cementitious matrix. Well-dispersed multiwalled
carbon nanotubes with desired interfacial interactions brought about balanced
gains in the flexural strength, energy absorption capacity and ductility, impact and
abrasion resistance of high-performance cementitious paste, while preserving its
high compressive strength.

o

Relatively high nanotube contents (0.48 vol.% by volume of dry cementitious
materials) led, for the cementitious matrix composition and processing conditions
considered here, to clumping of nanotubes which compromised their
reinforcement efficiency.

o

Hybrid (combined nano- and micro-scale) reinforcement of the high-performance
cementitious paste, when used at appropriate volume fractions, provided distinct
gains in diverse engineering properties, which could not be realized by nano- or

475

micro-scale reinforcement alone; this point at the synergistic beneficial effects
offered by nano- and micro-scale reinforcement systems.


The reinforcement efficiencies of functionalized, non-functionalized and polymerwrapped multiwalled carbon nanotubes were evaluated in a high-performance
cementitious paste prepared using two different commercially available polycarboxylatebased superplasticizers. It should be noted that the role of superplasticizers in highperformance cementitious nanocomposites was expanded to also cover that of a
nanomaterial dispersant. Experimental results indicated that:
o

Different

polycarboxylate-based

superplasticizers

produce

cementitious

nanocomposites with significant differences in performance characteristics.
Hence, the specifics of polycarboxylate-based superplasticizer formulation impact
its effectiveness towards dispersion, interfacial interactions and reinforcement
efficiency of multiwalled carbon nanotubes in high-performance cementitious
matrices.
o

Nanotubes were dispersed in the mixing water of cementitious paste as a first step
towards preparation of cementitious nanocomposites. This step employed a
specific ultra-sonication schedule to ensure thorough dispersion of modified
nanotubes in the mixing water incorporating superplasticizer. Subsequent mixing
and casting steps for preparation of cementitious nanocomposites followed the
procedures used commonly with plain high-performance cementitious pastes.

o

The gains in flexural strength, maximum deflection and energy absorption
capacity of high-performance cementitious pastes resulting from reinforcement
with modified (PAA-physisorbed) multiwalled carbon nanotubes at 0.24 vol.%
476

(0.48 vol.%) by volume of dry cementitious materials prepared using the
preferred superplasticizer were 46.0% (48.2%), 176% (194%) and 136% (165%),
respectively. The preferred superplasticizer (ADVA® Cast 575) produced gains
in all flexural attributes with increasing nanotube volume fraction contrary to the
other superplasticizer (Glenium® 7700) which caused a drop in flexural strength
at higher volume fraction of both modified and unmodified multiwalled carbon
nanotubes of higher aspect ratios.
o

Modified and unmodified multiwalled carbon nanotubes, when used at 0.24 vol.%
and especially 0.48 vol.%, produced some increase in compressive strength of the
high-performance cementitious paste when prepared with the preferred
superplasticizer (ADVA® Cast 575). Selection of superplasticizer had a major
influence on the effects of nanotubes on the compressive strength of cementitious
paste.

o

SEM observations of the fractured surfaces of cementitious nanocomposites
indicated that multiwalled carbon nanotubes were uniformly dispersed within the
dense matrix. This uniform dispersion of the reinforcement system is essential for
realizing the observed gains in the engineering properties of the cementitious
paste. Some carbon nanotubes were observed to have pulled out of the
cementitious matrix during the fracture process. This pull-out behavior benefits
the ductility and energy absorption capacity of cementitious materials.

o

Modification by polymer wrapping (PAA-physisorption) of multiwalled carbon
nanotubes significantly benefited their reinforcement efficiency in highperformance cementitious pastes. The improvements in flexural strength, energy

477

absorption capacity and maximum deflection of cementitious paste with addition
of 0.24 vol.% polymer-wrapped nanotubes were 41.0%, 249% and 160%,
respectively.
o

Polymer-wrapped as well as unmodified multiwalled carbon nanotubes were
added as reinforcement to the high-performance cementitious paste after
dispersion in reduced quantities of mixing water. While the reinforcing effects of
nanotubes decreased with decreasing fraction of the mixing water used for their
dispersion, reasonable results were obtained when nanotubes were dispersed in as
low as 30% of the total mixing water, which benefits scaled-up production of
cementitious nanocomposites. For polymer wrapped nanotubes, the quantity of
mixing water could be reduced to as low as 20% for dispersion of nanotubes
without producing a statistically significant drop in the various engineering
properties of cementitious nanocomposites evaluated.



Experimental evaluation of reinforcement with multiwalled carbon nanotubes modified
through physisorption of a polyelectrolyte (PAA) and/or carbon microfiber on the
flexural and compressive performance, abrasion resistance, and the structure and failure
modes of a high-performance fiber reinforced cementitious mortar incorporating silica
sand yielded the following conclusions:
o

Reinforcement of high-performance cementitious mortars with different volume
fractions of modified carbon nanotube and/or carbon microfiber produced
important gains in its flexural strength, ductility and energy absorption capacity,
impact resistance and abrasion resistance. At equal volume fractions, carbon
nanotubes produced more pronounced gains in performance characteristics of

478

high-performance cementitious mortar when compared with carbon microfibers.
Hybrid reinforcement systems comprising both carbon nanotubes and carbon
microfibers produced particularly favorable results.
o

Selection of the superplasticizer type can significantly impact the reinforcement
efficiency of carbon nanotubes (modified via PAA physisorption) in highperformance cementitious mortars, which can be attributed to the corresponding
impacts on the dispersion and interfacial interactions of nanotubes in the
cementitious matrix..

o

Response surface analysis of experimental results identified an optimum
reinforcement condition comprising 0.08% modified carbon nanotube and 0.40%
carbon microfiber (by volume of dry cementitious materials), which produced
balanced gains in the performance characteristics of the high-performance
cementitious mortar considered in the project. This finding confirms the
synergistic actions of nano- and micro-scale reinforcement systems in
cementitious materials.

o

SEM observations of the fractured surfaces of high-performance cementitious
(mortar) nanocomposites confirmed uniform dispersion of modified carbon
nanotubes within the dense matrix. The contributions of nanomaterials towards
restraint of microcracks by bridging action and crack arrest were also evident.
Thermogravimetric analyses pointed at increased production of hydration
products with the addition of carbon nano- and micro-scale reinforcement system.



The effects of nanotubes on the electrical conductivity of the cementitious mortar were
evaluated, and the damage-sensitivity of electrical conductivity was assessed by making
479

conductivity measurements prior to and after application of compressive stress (up to
different fractions of the ultimate strength of the cementitious mortar). Experimental
results indicated that:
o

Dispersion of 0.16 vol.% modified multiwalled carbon nanotubes (by volume of
dry cementitious materials) into high-performance cementitious mortar led to two
to three orders of magnitude rise in electrical conductivity (depending upon the
superplasticizer type). Increasing the nanotube volume fraction from 0.16 to 0.24
vol.% produced another order of magnitude rise in electrical conductivity.

o

The damaging effects (microcracking and permanent strains) induced by
increasing compressive stress levels lowered the electrical conductivity of highperformance cementitious mortars incorporating carbon nanotubes. After
application of 60% and 80% of compressive strength, electrical conductivity
dropped by one and two orders of magnitude, respectively. The higher
conductivity levels in the presence of 0.24 vol.% (versus 0.16 vol.%) carbon
nanotubes (by volume of dry cementitious materials) simplify measurement of
electrical conductivity. Damage-sensitivity of the electrical conductivity of highperformance cementitious materials incorporating carbon nanotubes provides
these materials with inherent health monitoring qualities.

o

SEM images confirmed successful dispersion of carbon nanotubes in a dense,
high-performance cementitious matrix, with nanotubes forming a percolated
network within matrix which benefits their contributions to electrical
conductivity.

480



Experimental results were generated for high-performance concrete (incorporating silica
sand and granite coarse aggregate) reinforced with modified (PAA-physisorbed)
multiwalled carbon nanotubes, polypropylene microfibers, carbon microfibers and their
hybrids

prepared

at

various/optimum

volume

fractions

with

carboxyl-based

superplasticizers. Test results indicated that:
o

Proper use of nano-scale reinforcement can produce improvements in highperformance concrete mechanical properties which surpass those realized with
micro-scale reinforcement at equal volume fractions.

o

Nano- and micro-scale reinforcement renders complementary reinforcing effects
in high-performance concrete; optimum reinforcement systems thus incorporate
both nano- and micro-scale reinforcement.

o

Among the single reinforcement conditions evaluated in high-performance
concrete, PAA-physisorbed multiwalled carbon nanotubes produced particularly
significant improvements in the mechanical properties of high-performance
concrete.

o

PAA-physisorbed multiwalled carbon nanotubes act synergistically with microscale reinforcement towards enhancement of the mechanical properties of highperformance concrete; the synergistic effects of nano- and micro-scale
reinforcement are reflected in the fact that hybrid (nano- and micro-scale)
reinforcement overcomes any drawbacks associated with the use of each (nano- or
micro-scale) reinforcement used alone.

481

o

The experimental results produced using the optimu hybrid reinforcement
condition identified through RSA (response surface analysis) pointed at the
effectiveness of RSA in identifying the desired reinforcement condition for
improving the targeted attributes of high-performance concrete. The refined
hybrid reinforcement produced results which further demonstrated the synergetic
actions of nano- and micro-scale reinforcement systems. For the hybrid
reinforcement comprising polymer-wrapped multiwalled carbon nanotubes at
0.15% and polypropylene microfiber at 0.23% by volume of dry cementitious
materials, pronounced and balanced gains in performance characteristics were
realized when compared with systems where either nano- or micro-scale
reinforcement was used alone.



Experimental results generated for high-strength concrete materials with lower
performance characteristics than those considered throughout the project, reinforced
with PAA physisorbed multiwalled carbon nanotubes, carbon microfibers and their
optimum hybrid system, prepared with carboxyl-based superplasticizer, indicated that:
o

Proper use of nano-scale reinforcement can produce improvements in highstrength concrete mechanical properties which surpass those realized with microscale reinforcement at equal volume fractions. However, the reinforcement
efficiency is reduced with reduction of concrete matrix performance (which
accompanied lowering of the cementitious paste content). The higherperformance concrete (which was steam cured) exhibited a denser microstructure
when compared with the concrete considered at this stage of investigation (which

482

was moist cured). The increased density of cementitious matrix benefits its
interactions with nanomaterials.
o

The hybrid (nano- and micro-scale) reinforcement produces complementary
reinforcing effects in high-strength concrete (as it did in the higher-performance
concrete considered earlier); optimum reinforcement systems for this concrete
would also incorporate both nano- and micro-scale reinforcement.

o

Among the nano- and micro-scale reinforcement systems considered, PAAphysisorbed multiwalled carbon nanotubes produced balanced gains in
engineering properties of high-strength concrete when compared with micro-scale
reinforcement.



Experimental results on cementitious pastes reinforced with non-functionalized and
functionalized carbon nanofibers as well as their hybrids with carbon microfibers
indicated that:
o

Acid-oxidization of CNFs benefited their dispersion and reinforcement efficiency
in the high-performance cementitious matrix. The high specific surface area and
the nano-scale spacing of dispersed nanofibers benefit their control of defect
(microcrack) size and mitigation of microcrack growth, thus improving the
nanofiber contributions to the strength, energy absorption capacity and ductility of
high-performance cementitious materials. Properly functionalized and welldispersed CNFs with desired interfacial interactions brought about balanced gains
in the flexural strength, energy absorption and ductility, impact and abrasion
resistance of high-performance cementitious paste, while preserving its high
compressive strength.
483

o

For the cementitious matrix and processing conditions considered in this
investigation, the nanofiber dosage should be lowered below 0.24% (by volume
of dry cementitious materials) before benefits can be realized from introduction of
carbon microfibers. Hence, a nanofiber volume fraction of 0.16% was considered
for development of hybrid reinforcement systems.

o

A hybrid reinforcement comprising CNFs and microfibers at 0.16% and 0.24% by
volume of dry cementitious materials produced distinct gains in the engineering
properties of the high-performance cementitious paste, which could not be
matched by the nano- or micro-scale reinforcement used alone. This finding
points at the synergistic effects of nano- and micro-scale reinforcement systems.



Results of comprehensive experimental evaluations of high-performance cementitious
mortar with nano-scale (pristine, polymer-wrapped or oxidized carbon nanofibers)
and/or micro-scale (carbon microfiber) reinforcement using two polycarboxylate-based
superplasticizers indicated that:
o

As was the case with carbon nanotubes for similar reinforcement conditions, the
gains in performance characteristics of the cementitious matrix depended upon the
specific polycarboxylate-based superplasticizer used. The gains in flexural
performance characteristics per unit volume of reinforcement were more
significant with CNFs when compared with carbon microfibers.

o

The

high-performance

cementitious

mortar experienced

important

(and

statistically significant) gains in all flexural attributes (strength, energy absorption
capacity and maximum deflection), impact and abrasion resistance with
introduction of different volume fractions of polymer-wrapped CNFs (used alone
484

or in combination with microfibers) versus plain mortar or those with similar
volume fractions of pristine or oxidized CNFs.
o

Hybrid reinforcement systems comprising CNFs and microfibers produced
particularly significant gains in the flexural performance characteristics and
impact resistance of high-performance cementitious mortar. Response surface
analysis of experimental results indicated that optimum reinforcement conditions
comprised both CNFs and microfibers. The reinforcement conditions considered
here produced minor (and statistically insignificant) changes in the compressive
strength of high-performance cementitious mortar.



The experimental results generated for high performance concrete reinforced with both
oxidized and PAA physisorbed carbon nanofibers, polypropylene microfibers, carbon
microfibers and their hybrids, prepared at various/optimum volume fractions and with
carboxyl-based superplasticizers, indicated that:
o

Proper use of CNFs can produce improvements in high-performance concrete
mechanical

properties

which

surpass

those

realized

with

micro-scale

reinforcement.
o

CNFs and micro-scale reinforcement render complementary reinforcing effects in
high-performance concrete; optimum reinforcement systems thus incorporate both
nano- and micro-scale reinforcement.

o

Among the individual reinforcement conditions evaluated in high-performance
concrete, PAA-physisorbed CNFs produced particularly high and balanced
improvements in the mechanical properties of high-performance concrete.

485

o

PAA-physisorbed CNFs act synergistically with micro-scale reinforcement
towards enhancement of the mechanical properties of high-performance concrete;
the synergistic effects of CNFs and micro-scale reinforcement are reflected in the
fact that hybrid (nano- and micro-scale) reinforcement overcomes any drawbacks
associated with the use of each (nano- or micro-scale) reinforcement used alone.

o

The optimized hybrid reinforcement system, identified through response surface
analysis (RSA) of experimental results, point at the effectiveness of RSA in
identifying desired reinforcement conditions for improving the targeted attributes
of high-performance concrete. The refined hybrid reinforcement produced results
which further demonstrated the synergetic actions of CNFs and micro-scale
reinforcement. The hybrid reinforcement system comprising polymer-wrapped
CNFs at 0.12% and polypropylene microfibers at 0.24% (by volume of dry
cementitious materials) produced more pronounced and balanced gains in
performance characteristics versus conditions with either nano- or micro-scale
reinforcement used alone.



Experimental results generated for high-performance cementitious paste, mortar and
concrete reinforced with pristine and PAA-physisorbed graphite nanoplatelets and also
their hybrids with carbon or polypropylene microfibers indicated that:
o

All the reinforcement conditions considered produced improvements in flexural
attributes,

impact

and

abrasion

resistance

of

high-performance

(DSP)

cementitious paste, mortar and concrete.
o

In DSP paste, graphite nanoplatelets, carbon microfibers and especially their
combination (hybrid reinforcement) produced important gains in the flexural
486

strength, energy absorption capacity and maximum deflection. The hybrid
reinforcement system yielded 61.7%, 366% and 264% gains in flexural strength,
energy

absorption

capacity

and

maximum

deflection,

respectively,

of

cementitious paste. Graphite nanoplatelets seem to produce a minor (statistically
insignificant) loss of compressive strength. This could be attributed to their planar
(flaky) geometry and micro-scale planar dimension, which could produce local
stress rise. The most important contribution of graphite nanoplatelets was to the
impact resistance of DSP paste, which increased by 42.7%, 114% and 158% with
introduction of carbon microfiber, graphite nanoplatelet and hybrid reinforcement.
The abrasion resistance of DSP paste increased by 45.5%, 58.9% and 56.4% with
carbon microfiber, graphite nanoplatelet and hybrid reinforcement.
o

Similar trends in all engineering properties were observed in the case of the DSP
mortar matrix. The comparatively smaller and statistically insignificant
improvement in flexural attributes with the use of PAA-physisorbed graphite
nanoplatelets can be attributed to the morphology and physical characteristics of
nanoplatelets. The experimental results also indicate that hybrid (nano- and
micro-scale) reinforcement systems produce more pronounced and balanced
benefits, which point at the complementary role of reinforcement systems
operating at different length scales. In mortar matrix, pair-wise comparisons
indicated that PAA-physisorbed graphite nanoplatelets (and their hybrid with
carbon microfiber) produce statistically significant (but still small) adverse effects
on the compressive strength of DSP mortar. As was the case with DSP paste, the
most significant effect of PAA-physisorbed graphite nanoplatelet was on the

487

impact resistance of DSP mortar. PAA-physisorbed graphite nanoplatelet at 0.16
and 0.24% by volume of dry cementitious materials improved the impact
resistance of DSP mortar by 99% and 109%, respectively. This significant
increase is due to flaky geometry which tends to be more effective under dynamic
loads.
o

In DSP concrete incorporating coarse aggregates, PAA-physisorbed graphite
nanoplatelets effectively reinforced the paste portion of the matrix. The
improvements in flexural strength, energy absorption capacity and maximum
deflection with respect to plain concrete were 10.5%, 97% and 76%, respectively.
The addition of carbon or polypropylene microfibers to high-performance (DSP)
concrete produced important gains in only two out of the three flexural attributes
considered here (maximum deflection and energy absorption capacity), while
causing a relatively small drop in the flexural strength. The hybrid reinforcement
comprising low-modulus polypropylene microfibers with graphite nanoplatelets
produced particularly pronounced gains in the mechanical properties of DSP
concrete. The best balance of flexural attributes was produced by the hybrid
reinforcement comprising PAA graphite nanoplatelet (0.16 vol.%) and
polypropylene microfiber (0.24 vol.%), which produced 15%, 573% and 329%
rise in flexural strength, maximum deflection and energy absorption capacity,
respectively. As was the case with DSP paste and mortar, PAA-physisorbed
graphite nanoplatelets as well as carbon or polypropylene microfibers, produced
relatively small and statistically insignificant effects on the compressive strength
of DSP concrete. The modified graphite nanoplatelet and/or micro-scale

488

reinforcement systems produced improvements in the impact resistance of the
high-performance concrete matrix. However, due to addition of coarse aggregates,
the contribution of PAA-physisorbed graphite nanoplatelet was not that
significant. The modified graphite nanoplatelet at 0.16% by volume of dry
cementitious materials produced the greatest improvement (38%) in the abrasion
resistance of high-performance (DSP) concrete when compared with the two
microfibers.


Experimental results on the durability characteristics of high-performance (DSP)
cementitious mortar and concrete indicated that:
o

DSP mortars provide desirable durability characteristics, which tend to be
enhanced through reinforcement with nano-scale and especially hybrid (nano- and
micro-scale) reinforcement systems. Hybrid reinforcement of DSP cementitious
mortar, when used at appropriate volume fractions, proved very effective in
improving key durability characteristics of the matrix through enhancement of
barrier qualities and synergistic reinforcement at different scales.

o

Plain DSP mortars as well as those with hybrid graphite nanoplatelet/carbon
microfiber reinforcement did not survive exposure to elevated temperatures. DSP
mortars with hybrid PAA-physisorbed multiwalled carbon nanotube and carbon
microfiber reinforcement, however, survived exposure to extreme temperatures.
This was attributed to the high toughness, tensile strength and thermal
conductivity of these nanocomposites, and also to the potential for combustion of
percolated nanotubes which opens escape routes to reduce the steam pressure
developed in nanocomposites at elevated temperatures. Extended immersion in

489

hot water had relatively small effects on the dynamic elastic modulus of DSP
mortar nanocomposites. Flexural strength, however, experienced a drop after
extended hot water immersion, which was not statistically significant with hybrid
reinforcement systems comprising PAA-physisorbed carbon nanotubes and
carbon microfibers.
o

In the case of the high-performance (DSP) concrete also, the accelerated aging
effects considered did not damage the flexural attributes and dynamic elastic
modulus of plain DSP concrete and that with PVA-physisorbed multiwalled
carbon nanotube. These accelerated aging conditions included 300 freeze-thaw or
wet-dry cycles, and 56 days immersion in hot water. However, matrices
containing polypropylene microfibers and hybrid (polypropylene microfiber and
PAA-physisorbed carbon nanotube) reinforcement experienced drops in the mean
values of almost all flexural attributes when exposed to accelerated aging effects.
These drops, however, were not generally statistically significant. In general,
polypropylene microfibers did not perform well under accelerated aging
conditions even in a dense DSP concrete matrix. On the other hand, PAAphysisorbed multiwalled carbon nanotubes benefited the durability characteristics
of high-performance cementitious matrices.



Experimental results on the moisture sorption rates of different high-performance
cementitious matrices reinforced with various volume fractions of graphite nanomaterials
and/or microfibers indicated that:
o

The proper gradation of high-performance cementitious materials yields a dense
structure with high barrier qualities.

490

o

Unlike conventional (micro-scale) fibers, graphite nanomaterials significantly
enhance the diffusion resistance of cement-based materials. This could be
attributed to the close spacing and large surface area of graphite nanomaterials,
which significantly increase the tortuosity of diffusion paths within the cementbased matrix. The presence of graphite nanomaterials in the cementitious matrix
also partially blocks the pore system at nano-scale, which further benefits the
barrier qualities of cementitious materials.

o

Addition of either of the two PAA-physisorbed graphite nanomaterials as well as
their hybrid combinations with both microfibers resulted in drops in the moisture
sorption rate as well as the total water absorbed. The maximum drops (among
different reinforcement systems used alone) for all matrices were recorded with
0.24 vol.% PAA-physisorbed graphite nanoplatelet.

o

Due to the inherent advantages of being present in the matrix as an inclusion and
being relatively inert under weathering effects, graphite nanomaterials, especially
graphite nanomaterials, provide a cost-effective solution for enhancing the barrier
qualities and thus durability characteristics of cementitious materials. When
appropriately combined with microfibers, cementitious nanocomposites offer a
desired balance of mechanical, barrier and durability characteristics.



The experience gained with industrial-scale production and field construction of highperformance concrete nanocomposites indicated that:
o

Scaled-up production of concrete nanocomposites requires streamlining of the
dispersion process of larger quantities of graphite nanomaterials in water, and
their introduction into the plant batching and mixing process.
491

o

Concrete nanocomposites produced under scaled-up conditions provided
statistically similar, but slightly smaller, engineering properties when compared
with concrete nanocomposites produced at laboratory scale.

o

As was the case in laboratory studies, under scaled-up production and field
construction conditions also, optimized hybrid (nano- and micro-scale)
reinforcement systems performed better than either of the nano- and micro-scale
reinforcement used alone. This confirms the synergistic effects of nano- and
micro-scale reinforcement towards enhancement of the high-performance
concrete materials properties.

o

Scaled-up studies confirmed that the dense matrix in high-strength concrete
produced high barrier qualities. Both graphite nanomaterials, when used in
combination with polypropylene microfibers, further improved the barrier
qualities of high-performance concrete by forcing tortuous diffusion paths and
partial blocking of capillary pores. The maximum improvements in barrier
qualities were realized with the hybrid reinforcement comprising graphite
nanoplatelets and polypropylene microfibers. When used alone, however,
polypropylene microfibers did not produce statistically significant gains in the
barrier qualities of high-strength concrete.



A theoretical framework was developed for evaluating the contributions of graphite
nanomaterials (emphasizing carbon nanotubes) to diverse aspects of concrete engineering
properties. The theoretical models were verified and refined using experimental results.
These models enabled explanation of the means by which nanomaterials benefit diverse
aspects of concrete performance, ranging from toughness and strength to barrier and

492

durability characteristics, which far surpass the range of properties improved by
conventional (micro-scale) fibers. Theoretical explanations were also presented for the
synergistic role of carbon nanotubes and conventional (micro-scale) fibers towards
achieving balanced gains in diverse engineering properties of concrete


Percolation of carbon nanotubes can occur at relatively low nanotube volume fractions.
This brings about a sharp rise in conductivity (and also fire resistance) of concrete with
introduction of relatively low nanotube volume fractions. The electrical conductivity of
concrete nanocomposites was found to be sensitive to the damaging effects experienced b
concrete. Concrete nanocomposites thus offer inherent health monitoring qualities.



The trend towards industrial-scale production of multiwalled carbon nanotubes and other
graphite nanomaterials is causing a sharp drop in their projected intermediate- and longterm costs. This trend together with the significant benefits rendered by nanomaterials to
diverse aspects of concrete performance at relatively low dosages make nanomaterials,
and especially hybrid (nano- and micro-scale) reinforcement systems, economically
viable additives for development of the next-generation high-performance concrete
materials.

19.3 Recommendations for Future Research
While this study has advanced the state-of-the-art with respect to the use of suitably modified
graphite nanomaterials and/ or microfibers in different high-performance cementitious matrices,
further research is required to fully realize the reinforcement potential of graphite nanomaterials
in a broad range of cement-based matrices. Some key recommendations for further research in
this area are presented in the following.

493



The modification conditions of nanomaterials, and procedures used for their dispersion in
concrete need to be optimized for enhanced effectiveness at viable cost, and should be
streamlined to suit scaled-up production.



The efficiency of modified nanomaterials in broader categories of concrete materials,
with different mix proportions and admixtures, need to be evaluated.



Distinctions between multiwalled carbon nanotubes, carbon nanofibers and graphite
nanoplatelets should be further clarified, and niche applications for each of these
nanomaterials should be identified.



Optimum hybrid (nano- and micro-scale) reinforcement systems should be identified with
broader categories of micro-scale fibers, and for different concrete mix proportions.



More comprehensive laboratory and field test data should be produced on diverse
engineering properties of concrete nanocomposites.



Further micro-structural investigations should be undertaken in order to better understand
the impact of modified nanomaterials on the structure of cement hydrates, their role in
failure mechanisms and sorption resistance of concrete, and their complementary actions
with micro-scale fibers.



Theoretical models should be refined and expanded for reliable predictions of the nanoscale and hybrid reinforcement systems on diverse aspects of concrete engineering
properties.



Focused laboratory and field investigations should be undertaken to demonstrate the
advantages of nanomaterials in niche applications identified for each nanomaterial type.

494



The research findings need to be reflected in standard specifications and guides to
facilitate transition of modified graphite nanomaterials to selected concrete construction
markets.

19.4 Impact of the Research Effort
The pioneering work conducted in this research has produced convincing evidence for the
technical value and competitive cost of modified graphite nanomaterials as reinforcement in
high-performance concrete materials. The hybrid reinforcement system developed in the project,
comprising modified graphite nanomaterials and micro-scale fibers, produces significant and
uniquely balanced gains in diverse engineering properties of high-performance concrete. Initial
steps were also taken towards scaled-up production and field use of concrete nanocomposites.
The research outcomes point at the value of graphite nanomaterials as a new class of concrete
additive, which are distinguished by their impact on a broad range of concrete material
properties. Graphite nanomaterials can make important contributions towards development of the
next-generation high-performance concrete materials and infrastructure systems offering
balanced improvements in diverse mechanical qualities and structural efficiency, barrier and
durability characteristics, safety attributes under extreme conditions and life-cycle economy.

495

REFERENCES

496

REFERENCES
Navielle, A.M., N. (1995). Properties of concrete. London, Longman Group Limited.
Ahlborn, T. M., D. K. Harris, et al. (2011). "Characterization of strength and durability of ultrahigh-performance concrete under variable curing conditions." Transportation Research
Record(2251): 68-75.
Badanoiu, A., M. Georgescu, et al. (2003). "The study of 'DSP' binding systems by
thermogravimetry and differential thermal analysis." Journal of Thermal Analysis and
Calorimetry 74(1): 65-75.
Balasubramanian, K. and M. Burghard (2005). "Chemically functionalized carbon nanotubes."
Small 1(2): 180-192.
Banerjee, S., M. G. C. Kahn, et al. (2003). "Rational chemical strategies for carbon nanotube
functionalization." Chemistry - A European Journal 9(9): 1898-1908.
Banthia, N. and M. Sappakittipakorn (2007). "Toughness enhancement in steel fiber reinforced
concrete through fiber hybridization." Cement and Concrete Research 37(9): 1366-1372.
Bayard, O. and O. Ple (2003). "Fracture mechanics of reactive powder concrete: Material
modelling and experimental investigations." Engineering Fracture Mechanics 70(7-8):
839-851.
Beaudoin, J. J., R. F. Feldman, et al. (1994). "Pore structure of hardened Portland cement pastes
and is influence on properties." Advanced Cement Based Materials 1(5): 224-236.
Bensebaa, F., F. Zavaliche, et al. (2004). "Microwave synthesis and characterization of Co-ferrite
nanoparticles." Journal of Colloid and Interface Science 277(1): 104-110.
Bentz, D. P. (2000). "Influence of silica fume on diffusivity in cement-based materials. II. Multiscale modeling of concrete diffusivity." Cement and Concrete Research 30(7): 11211129.
Bentz, D. P., O. M. Jensen, et al. (2000). "Influence of silica fume on diffusivity in cement-based
materials. I. Experimental and computer modeling studies on cement pastes." Cement and
Concrete Research 30(6): 953-962.
Breuer, O. and U. Sundararaj (2004). "Big returns from small fibers: A review of
polymer/carbon nanotube composites." Polymer Composites 25(6): 630-645.
Camilleri, J., F. E. Montesin, et al. (2006). "Characterization of Portland cement for use as a
dental restorative material." Dental Materials 22(6): 569-575.

497

Chan, L. Y. and B. Andrawes (2010). "Finite element analysis of carbon nanotube/cement
composite with degraded bond strength." Computational Materials Science 47(4): 9941004.
Chan, Y.-W. and S.-H. Chu (2004). "Effect of silica fume on steel fiber bond characteristics in
reactive powder concrete." Cement and Concrete Research 34(7): 1167-1172.
Cheyrezy, M., V. Maret, et al. (1995). "Microstructural analysis of RPC (reactive powder
concrete)." Cement and Concrete Research 25(7): 1491-1500.
Childs, P., A. C. L. Wong, et al. (2007). "Measurement of the coefficient of thermal expansion of
ultra-high strength cementitious composites using fiber optic sensors." Cement and
Concrete Research 37(5): 789-795.
Chindaprasirt, P., C. Jaturapitakkul, et al. (2005). "Effect of fly ash fineness on compressive
strength and pore size of blended cement paste." Cement and Concrete Composites 27(4):
425-428.
Collepardi, M., V. Corinaldesi, et al. (2002). "Applicazioni e sviluppo dei materiali DSP; DSP
materials applications and development progress." Industria Italiana del Cemento 72(JU):
540-545.
Collins, F., J. Lambert, et al. (2012). "The influences of admixtures on the dispersion,
workability, and strength of carbon nanotube-OPC paste mixtures." Cement and Concrete
Composites 34(2): 201-207.
Cwirzen, A., K. Habermehl-Cwirzen, et al. (2009). "SEM/AFM studies of cementitious binder
modified by MWCNT and nano-sized Fe needles." Materials Characterization 60(7):
735-740.
Cwirzen, A., K. Habermehl-Cwirzen, et al. (2008). "Surface decoration of carbon nanotubes and
mechanical properties of cement/carbon nanotube composites." Advances in Cement
Research 20(2): 65-73.
Cwirzen, A. and V. Penttala (2005). "Aggregate-cement paste transition zone properties affecting
the salt-frost damage of high-performance concretes." Cement and Concrete Research
35(4): 671-679.
Feylessoufi, A., F. Villieras, et al. (1996). "Water environmental and nanostructural network in a
reactive powder concrete." Cement & Concrete Composites 18(1): 23-24.
Fu, X. and D. D. L. Chung (1998). "Submicron-diameter-carbon-filament cement-matrix
composites." Carbon 36(4): 459-462.
Gao, L., L. Jiang, et al. (2006). "Carbon nanotube-ceramic composites." Journal of
Electroceramics 17(1): 51-55.

498

Guerrini, G. L. (2000). "Applications of high-performance fiber-reinforced cement-based
composites." Applied Composite Materials 7(2-3): 195-207.
Hall, C. (1989). "Water sorptivity of mortars and concretes: a review." Magazine of Concrete
Research 41(147): 51-61.
Han, B., X. Yu, et al. (2009). "A self-sensing carbon nanotube/cement composite for traffic
monitoring." Nanotechnology 20: 1-5.
Han, B., X. Yu, et al. (2011). "Effects of CNT concentration level and water/cement ratio on the
piezoresistivity of CNT/cement composites." Journal of Composite Materials 46(1): 1925.
Hawley, M. C. (2002). Microwave processing of polymers and composites - Fundamentals and
applications. Composites Manufacturing and Tooling 2001, Febrary 21, 2001 - Febrary
23, 2001, Orlando, FL, United states, Society of Manufacturing Engineers.
Hu, A., Y. Fang, et al. (1999). "Humidity dependence of apparent dielectric constant for DSP
cement materials at high frequencies." Journal of the American Ceramic Society 82(7):
1741-1747.
Hu, G., C. Zhao, et al. (2006). "Low percolation thresholds of electrical conductivity and
rheology in poly(ethylene terephthalate) through the networks of multi-walled carbon
nanotubes." Polymer 47(1): 480-488.
Kalaitzidou, K., H. Fukushima, et al. (2007). "Mechanical properties and morphological
characterization of exfoliated graphite-polypropylene nanocomposites." Composites Part
A: Applied Science and Manufacturing 38(7): 1675-1682.
Kalaitzidou, K., H. Fukushima, et al. (2007). "Multifunctional polypropylene composites
produced by incorporation of exfoliated graphite nanoplatelets." Carbon 45(7): 14461452.
Katz, A. and A. Bentur (1994). "Mechanical properties and pore structure of carbon fiber
reinforced cementitious composites." Cement and Concrete Research 24(2): 214-220.
Katz, A. and A. Bentur (1995). "Effect of matrix composition on the aging of CFRC." Cement
and Concrete Composites 17(2): 87-97.
Katz, A. and A. Bentur (1996). "Mechanisms and processes leading to changes in time in the
properties of CFRC." Advanced Cement Based Materials 3(1): 1-13.
Katz, A., V. C. Li, et al. (1995). "Bond properties of carbon fibers in cementitious matrix."
Journal of Materials in Civil Engineering 7(2): 125-128.
Konsta-Gdoutos, M. S., Z. S. Metaxa, et al. (2010). "Highly dispersed carbon nanotube
reinforced cement based materials." Cement and Concrete Research 40(7): 1052-1059.

499

Konsta-Gdoutos, M. S., Z. S. Metaxa, et al. (2010). "Multi-scale mechanical and fracture
characteristics and early-age strain capacity of high performance carbon nanotube/cement
nanocomposites." Cement and Concrete Composites 32(2): 110-115.
Kowald, T. (2004). Influence of Surface-Modified Carbon Nanotubes on Ultra-High
Performance Concrete. Proceedings of International Symposium on Ultra High
Performance Concrete. Kassel, Germany: 195–202.
Kubota, K., M. Sano, et al. (2005). "Microwave irradiation for chemical modification of carbon
nanotubes for better dispersion." Japanese Journal of Applied Physics, Part 1: Regular
Papers and Short Notes and Review Papers 44(1 A): 465-468.
Lawler, J. S., D. Zampini, et al. (2005). "Microfiber and macrofiber hybrid fiber-reinforced
concrete." Journal of Materials in Civil Engineering 17(5): 595-604.
Lee, M.-G., Y.-C. Wang, et al. (2007). "A preliminary study of reactive powder concrete as a
new repair material." Construction and Building Materials 21(1): 182-189.
Li, G. Y., P. M. Wang, et al. (2007). "Pressure-sensitive properties and microstructure of carbon
nanotube reinforced cement composites." Cement and Concrete Composites 29(5): 377382.
Li, G. Y., P. M. Wang, et al. ( 2005). "Mechanical behavior and microstructure of cement
composites incorporating surface-treated multi-walled carbon nanotubes." Carbon 43(6):
1239–1245.
Li, J., M. L. Sham, et al. (2007). "Morphology and properties of UV/ozone treated graphite
nanoplatelet/epoxy nanocomposites." Composites Science and Technology 67(2): 296305.
Li, V. C. (2002). "Large volume, high-performance applications of fibers in civil engineering."
Journal of Applied Polymer Science 83(3): 660-686.
Makar, J. M. and J. J. Beaudoin (2004). Carbon nanotubes and their applications in the
construction industry. Nanotechnology in construction, Proceedings of the 1st
international symposium on nanotechnology in construction, Royal Society of Chemistry.
P. J. M. Bartos, J. J. Hughes, P. Trtik and W. Zhu: 331–341.
Makar, J. M. and G. W. Chan (2009). "Growth of cement hydration products on single-walled
carbon nanotubes." Journal of the American Ceramic Society 92(6): 1303-1310.
Makar, J. M., J. Margeson, et al. (2005). Carbon nanotube/cement composites – early results and
potential applications. Proceedings of the 3rd international conference on construction
materials: performance, innovations and structural implications, Vancouver, BC, Canada.

500

Manzur, T. and N. Yazdani (2010). "Strength enhancement of cement mortar with carbon
nanotubes: Early results and potential." Transportation Research Record(2142): 102-108.
Matte, V. and M. Moranville (1999). "Durability of reactive powder composites: Influence of
silica fume on the leaching properties of very low water/binder pastes." Cement &
Concrete Composites 21(1): 1-9.
Mehta, A. (1993). Concrete, Prentice Hall.
Metaxa, Z. S., M. S. Konsta-Gdoutos, et al. (2009). Carbon nanotubes reinforced concrete. ACI
Fall 2009 Convention, November 8, 2009 - November 12, 2009, New Orleans, LA,
United states, American Concrete Institute.
Metaxa, Z. S., M. S. Konsta-Gdoutos, et al. (2010). "Carbon Nanofiber-Reinforced cement-based
materials." Transportation Research Record(2142): 114-118.
Metaxa, Z. S., M. S. Konsta-Gdoutos, et al. (2010). Mechanical properties and nanostructure of
cement-based materials reinforced with CNFs and Polyvinyl Alcohol (PVA) microfibers.
ACI Spring 2010 Convention, March 21, 2010 - March 25, 2010, Chicago, IL, United
states, American Concrete Institute.
Moranville-Regourd, M. (2001). New cementitious systems and composite materials. 7th
Conference and Exhibition of the European Ceramic Society, Sep 9-13 2001, Brugge,
Trans Tech Publications Ltd.
Moranville-Regourd, M. (2001). New cementitious systems and composite materials, Brugge,
Trans Tech Publications Ltd.
Morin, V., F. Cohen-Tenoudji, et al. (2002). "Evolution of the capillary network in a reactive
powder concrete during hydration process." Cement and Concrete Research 32(12):
1907-1914.
Musso, S., J.-M. Tulliani, et al. (2009). "Influence of carbon nanotubes structure on the
mechanical behavior of cement composites." Composites Science and Technology 69(1112): 1985-1990.
Nasibulin, A. G., S. D. Shandakov, et al. (2009). "A novel cement-based hybrid material." New
Journal of Physics 11.
O'Connell, M. J., P. Boul, et al. (2001). "Reversible water-solubilization of single-walled carbon
nanotubes by polymer wrapping." Chem. Phys. Lett. 342(3,4): 265-271.
O'Connell, M. J., P. Boul, et al. (2001). "Reversible water-solubilization of single-walled carbon
nanotubes by polymer wrapping." Chem. Phys. Lett. 342(3): 265-271.

501

Pfeifer, C., B. Moeser, et al. (2009). Durability of ultra-high-performance concrete. 10th ACI
International Conference on Recent Advances in Concrete Technology and Sustainability
Issues, October 14, 2009 - October 16, 2009, Seville, Spain, American Concrete Institute.
Ping, X., J. J. Beaudoin, et al. (1991). "Effect of aggregate size on transition zone properties at
the portland cement paste interface." Cement and Concrete Research 21(6): 999-1005.
Poon, C. S., L. Lam, et al. (1999). "Effects of fly ash and silica fume on interfacial porosity of
concrete." Journal of Materials in Civil Engineering 11(3): 197-205.
Richard, P. and M. Cheyrezy (1995). "Composition of reactive powder concretes." Cement and
Concrete Research 25(7): 1501-1511.
Roux, N., C. Andrade, et al. (1996). "Experimental study of durability of reactive powder
concretes." Journal of Materials in Civil Engineering 8(1): 1-6.
Saez De Ibarra, Y., J. J. Gaitero, et al. (2006). Atomic force microscopy and nanoindentation of
cement pastes with nanotube dispersions, Wiley-VCH Verlag.
Sahmaran, M. and V. C. Li (2009). "Influence of microcracking on water absorption and
sorptivity of ECC." Materials and Structures/Materiaux et Constructions 42(5): 593-603.
Sandlar, J., M. S. P. Shaffer, et al. (1999). "Development of a dispersion process for carbon
nanotubes in an epoxy matrix and the resulting electrical properties." Polymer 40: 59675971.
Shannag, M., W. Hansen, et al. (1999). "Interface Debonding in Fiber Reinforced CementMatrix Composites." Journal of Composite Materials 33(2): 158-176.
Shannag, M. J., R. Brincker, et al. (1996). "Interfacial (fiber-matrix) properties of high-strength
mortar (150 MPa) from fiber pullout." ACI Materials Journal 93(5): 480-486.
Shannag, M. J., R. Brincker, et al. (1997). "Pullout behavior of steel fibers from cement-based
composites." Cement and Concrete Research 27(6): 925-936.
Shannag, M. J. and W. Hansen (2000). "Tensile properties of fiber-reinforced very high strength
DSP mortar." Magazine of Concrete Research 52(2): 101-108.
Shannag, M. J., W. Hansen, et al. (1995). Interfacial debonding and sliding in brittle cementmatrix composites from steel fiber pullout tests, Boston, MA, USA, Materials Research
Society, Pittsburgh, PA, USA.
Singh, B., P. Kumar, et al. (2001). "High performance composites for the new millennium."
Journal of Structural Engineering (Madras) 28(1): 17-26.
Song, H.-W., J.-C. Jang, et al. (2007). "An estimation of the diffusivity of silica fume concrete."
Building and Environment 42(3): 1358-1367.

502

Sun, G.-k. and J. F. Young (1993). "Quantitative determination of residual silica fume in DSP
cement pastes by 29-Si NMR." Cement and Concrete Research 23(2): 480-483.
Sun, G. K. and J. F. Young (1993). "Hydration reactions in autoclaved DSP cements." Advances
in Cement Research 5(20): 163-169.
Tibbetts, G. G., M. L. Lake, et al. (2007). "A review of the fabrication and properties of vaporgrown carbon nanofiber/polymer composites." Composites Science and Technology
67(7-8): 1709-1718.
Tjiptobroto, P. and W. Hansen (1991). "Mechanism for tensile strain hardening in high
performance cement-based fiber reinforced composites." Cement & Concrete Composites
13(4): 265-273.
Tjiptobroto, P. and W. Hansen (1993). "Model for predicting the elastic strain of fiber reinforced
composites containing high volume fractions of discontinuous fibers." ACI Materials
Journal (American Concrete Institute) 90(2): 134-142.
Tjiptobroto, P. and W. Hansen (1993). "Tensile strain hardening and multiple cracking in highperformance cement-based composites containing discontinuous fibers." ACI Materials
Journal (American Concrete Institute) 90(1): 16-25.
Tsivilis, S., J. Tsantilas, et al. (2003). "The permeability of Portland limestone cement concrete."
Cement and Concrete Research 33(9): 1465-1471.
Vaisman, L., H. D. Wagner, et al. (2006). "The role of surfactants in dispersion of carbon
nanotubes." Advances in Colloid and Interface Science 128-130: 37-46.
Wang, X., J. Ye, et al. (2007). "Reinforcement of calcium phosphate cement by bio-mineralized
carbon nanotube." Journal of the American Ceramic Society 90(3): 962-964.
Washer, G., P. Fuchs, et al. (2004). "Ultrasonic Testing of Reactive Powder Concrete." IEEE
Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 51(2): 193-201.
Wille, K. and K. J. Loh (2010). "Nanoengineering ultra-high-performance concrete with
multiwalled carbon nanotubes." Transportation Research Record(2142): 119-126.
Wong, A. C. L., P. A. Childs, et al. (2007). "Simultaneous measurement of shrinkage and
temperature of reactive powder concrete at early-age using fiber Bragg grating sensors."
Cement and Concrete Composites 29(6): 490-497.
Xie, L., F. Xu, et al. (2007). "Single-walled carbon nanotubes functionalized with high bonding
density of polymer layers and enhanced mechanical properties of composites."
Macromolecules 40(9): 3296-3305.

503

Young, J. F. (1996). Recent advances in the development of high performance cement-based
materials, Washington, DC, USA, ASCE, New York, NY, USA.
Zanni, H., M. Cheyrezy, et al. (1996). "Investigation of hydration and pozzolanic reaction in
reactive powder concrete (RPC) using Si NMR." Cement and Concrete Research 26(1):
93-100.
Zivica, V. (1999). "Possibilities of a novel use of silica fume in mineral binding systems."
Construction and Building Materials 13(5): 271-277.

504