-
THE FACULTY OF ENGINEERING AND THE BUILT ENVIRONMENT The
Department of Civil Engineering
The Influence of Curing on Restrained Shrinkage Cracking of
Bonded Overlays
Prepared By: Nick Bester
Supervisor: A/Prof. Hans Beushausen
Date of Submission: 30th January 2015
A dissertation submitted to the Department of Civil Engineering,
University of Cape Town in partial fulfilment of the requirements
for the degree of Master of Science in Civil Engineering.
Concrete Materials and Structural Integrity Research Unit.
The financial assistance of the National Research Foundation
(DAAD-NRF) towards this research is hereby acknowledged. Opinions
expressed and conclusions arrived at, are those of the author and
are not necessarily to be attributed to the DAAD-NRF.
-
i
Plagiarism Declaration 1. I know that plagiarism is wrong.
Plagiarism is to use anothers work and to pretend that it is
ones own.
2. I have used the Harvard Convention for citation and
referencing. Each significant contribution to and quotation in this
report form the work or works of other people has been attributed
and has been cited and referenced.
3. This dissertation is my own work.
4. I have not allowed and I will not allow anyone to copy my
work with the intension of passing it as his or her own work.
Student Number: BSTNIC005 Surname: Bester
Date: 30/01/2015 Signature:________
Plagiarism Declaration
-
ii
Acknowledgements The author would like to thank and acknowledge
with gratitude the following persons, institutions, and companies
who made significant contributions towards the completion of this
dissertation.
The supervisor, Associate Professor Hans Beushausen, for his
invaluable guidance, encouragement, and continual academic and
technical assistance throughout the duration of this
dissertation.
The Directors of the Concrete Materials and Structural Integrity
Research Unit (CoMSIRU), Professor Mark Alexander and Professor
Pilate Moyo, for providing useful suggestions and constructive
criticism.
Mr Steve Crosswell, on behalf of Portland Pretoria Cement Ltd
(PPC), for their donation of cement used in this research.
Mr Kevin Kimbrey, on behalf of Sika South Africa (Pty) Ltd, for
their donation of products used in this research.
Ms Rhuwayda Lalla, on behalf of Lafarge South Africa (Pty) Ltd,
for their donation of readymix concrete used in this research.
The administration staff of the Department of Civil Engineering,
especially Elly Yelverton, the Research Administrative Officer of
CoMSIRU, for their assistance with administrative matters relating
to this research.
Mr Nooredien Hassen, the Civil Engineering Concrete Laboratory
Manager, for his assistance with the laboratory-related
experimentation carried out.
Mr Charles Nicholas, the Civil Engineering Workshop Principal
Technical Officer, for manufacturing the moulds and various
experimentation-related items that were required for the
experimentation carried out.
The Civil Engineering Concrete Laboratory staff, Mr Charlie May,
Mr Leonard Adams, Mr Elvino Witbooi, and Mr Hector Mafungwa, for
assisting in the laboratory with experimental work when needed.
The Postgraduate Students of CoMSIRU, especially Mr Gavin
Golden, for an invaluable friendship and continual assistance with
the experimentation conducted in this dissertation.
Family and friends for their continual encouragement, utmost
support and unconditional love.
The author would also like to thanks and acknowledge, with
gratitude, the financial support throughout the duration of the
dissertation (2013-2014) received from the following institutions
and companies.
The University of Cape Town (UCT) The Concrete Institute (TCI)
The National Research Foundation
(DAAD-NRF) Sika South Africa (Pty) Ltd Pretoria Portland Cement
Ltd (PPC)
AfriSam South Africa (Pty) Ltd Haw & Inglis Civil
Engineering (Pty) Ltd The Tertiary Education Support
Programme (TESP) of ESKOM The Water Research Commission
(WRC)
Acknowledgements
-
iii
Summary As more and more concrete structures are being
constructed worldwide, the amount of infrastructure requiring
repair and rehabilitation is increasing globally. This has resulted
in an increasing need for effective concrete repair techniques.
Bonded mortar or concrete overlays are a common type of concrete
repair technique that involves casting of a new repair material, in
the form of an overlay, in the place of the damaged or deteriorated
portion of concrete. Bonded overlays are however prone to premature
failure, due to differential volume changes between the newly cast
overlay and the existing concrete substrate, which is primarily
manifested as debonding or cracking.
The differential volume changes experienced by the overlay arise
due to the inherent shrinkage of cementitious materials. The
restraint provided by the substrate to the shrinkage of the overlay
results in the induction of tensile stresses within the overlay.
These stresses are proportional to the shrinkage strain and elastic
modulus of the overlay but are partially reduced due to tensile
relaxation. If the tensile stresses induced are greater than the
tensile strength of the overlay then cracking of the overlay,
termed restrained shrinkage cracking, will begin to develop. The
time-dependent material properties of the overlay that govern the
restrained shrinkage cracking are therefore the shrinkage, tensile
strength, elastic modulus, and tensile relaxation of the
overlay.
Despite the relatively sound understanding of the mechanisms of
restrained shrinkage cracking and the numerous measures that can be
taken to mitigate it, such as the use of a reduced paste content,
increased aggregate content, admixtures, fibres, etc., restrained
shrinkage cracking still often occurs. Another measure thought to
increase an overlays resistance to restrained shrinkage cracking is
adequate curing. However, the influence of curing on restrained
shrinkage cracking has not yet been thoroughly investigated and so
its influence is not fully known and understood.
In this study, the influence of curing on restrained shrinkage
cracking of bonded overlays was investigated. The influence of a
total of seven curing regimes which included air curing, sealed
curing (plastic sheets) for 2 and 7 days, water curing (wet cloths)
for 2 and 7 days, a protective coating and a curing compound was
investigated for four mix types which included three
laboratory-made mixes with water-binder (w/b) ratios of 0.45, 0.60
and 0.80, and one commercial repair mortar. The influence of the
seven curing regimes on the material properties governing
restrained shrinkage cracking (shrinkage, tensile strength, elastic
modulus, and tensile relaxation) and on the age at cracking and
crack area of ring tests and composite overlay-substrate specimens
was investigated for the four mix types. Analytical modelling using
a simple analytical model was carried out using the results of the
individual material properties in order to predict the age at
cracking. The results of the analytical modelling, in conjunction
with the tests conducted on the material properties governing
restrained shrinkage cracking, were used to explain the age at
cracking results observed from the ring test and composite
overlay-substrate specimens.
The results of the experimental testing showed that curing
influences all of the material properties governing restrained
shrinkage cracking. Prolonged and/or more effective curing that
minimised moisture loss and promoted hydration was shown to either
delay or reduce the rate of
Summary
-
iv shrinkage respectively (dependent on the curing method),
increase the tensile strength and elastic modulus, and decrease the
tensile relaxation. In general, prolonged and/or more effective
curing was shown to have a positive influence on restrained
shrinkage cracking by increasing the age and net age at cracking.
Wet cloth curing was shown to outperform plastic sheet curing for
all mix types. An increased curing duration was observed to
increase the age at cracking, however only provided an increase in
crack resistance for higher w/b ratio mixes as the net age at
cracking was only observed to increase for higher w/b ratio mixes.
A protective coating and curing compound were observed to
significantly increase the resistance to cracking by increasing the
age and net age at cracking, with the latter outperforming the
former. The optimal curing regime, in terms of the age at cracking,
was observed to be 7-day wet cloth curing for lower w/b ratio mixes
and a curing compound for higher w/b ratio mixes. The optimal
curing regime, in terms of increasing crack resistance, was
observed to be a curing compound as it resulted in the greatest net
age at cracking. Crack areas were shown not to be directly
correlated to the curing regime but rather to the net age at
cracking and the free shrinkage strain once cracking had
occurred.
Additional findings not directly related to the influence of
curing on restrained shrinkage cracking showed that an increasing
w/b ratio was observed to increase the crack resistance under
restrained shrinkage conditions; the age at cracking results of
ring test specimens were observed to be correlated to composite
overlay-substrate specimens, with the age at cracking results of
the composite overlay-substrate specimens being, on average, 64 15%
greater than the age at cracking of the ring tests; and that a
degree of restraint of 55% 1% was observed for all composite
overlay-substrate specimens which confirmed the typical degree of
restraint of 60% measured in literature.
Summary
-
v
Table of Contents Plagiarism Declaration i Acknowledgements ii
Summary iii Table of Contents v List of Figures viii List of Tables
xii Abbreviations xiv List of Variables xv 1. Introduction 1
1.1 Background to Research Problem 1 1.2 Problem to be
Investigated 2 1.3 Hypothesis 2 1.4 Research Aim and Objectives 2
1.5 Key Research Questions 3 1.6 Scope and Limitations 3 1.7 Thesis
Outline 4
2. Literature Review 6 2.1 Introduction 6 2.2 Fundamentals of
Bonded Overlays 6
2.2.1 Definition 6 2.2.2 Applications 6 2.2.3 Design and
Construction Considerations 7
2.3 Failure of Bonded Overlays 12 2.3.1 Overview 12 2.3.2
Debonding 13 2.3.3 Restrained Shrinkage Cracking 14
2.4 Restrained Shrinkage Cracking of Bonded Overlays 16 2.4.1
Overview 16 2.4.2 Shrinkage 17 2.4.3 Tensile Strength 26 2.4.4
Elastic Modulus 27 2.4.5 Tensile Relaxation 28 2.4.6 Factors
Influencing Restrained Drying Shrinkage Cracking 29
2.5 Curing of Concrete 41 2.5.1 Fundamentals of Curing 41 2.5.2
The Need for Curing 44 2.5.3 Curing Methods 46 2.5.4 Specifying
Curing 59 2.5.5 Influence of Curing on Properties of Concrete
61
2.6 Summary 63
Table of Contents
-
vi 3.Methodology 65 3.1 Introduction 65 3.2 Overview and
Approach 65 3.3 Mortar Mixes 67
3.3.1 Laboratory-Made Mortars 67 3.3.2 Commercial Repair Mortar
69
3.4 Curing and Environmental Conditions 70 3.4.1 Curing Methods
70 3.4.2 Curing Duration 74 3.4.3 Environmental Conditions 75
3.5 Experimental Tests 77 3.5.1 Compressive Strength 77 3.5.2
Free Shrinkage 77 3.5.3 Tensile Strength 78 3.5.4 Elastic Modulus
80 3.5.5 Tensile Relaxation 81 3.5.6 Ring Tests 82 3.5.7 Composite
Overlay-Substrate Specimens 85 3.5.8 Experimental Testing Summary
90
3.6 Analytical Modelling 91 3.6.1 Modelling Method 91 3.6.2
Modelling Assumptions 92
3.7 Summary 93 4. Results and Discussions 94
4.1 Introduction 94 4.2 Compressive Strength 94 4.3 Free
Shrinkage 96
4.3.1 Influence of Curing Regime 96 4.3.2 Influence of Mix Type
99
4.4 Tensile Strength 101 4.5 Elastic Modulus 102 4.6 Tensile
Relaxation 104 4.7 Ring Tests 106
4.7.1 Age at Cracking 106 4.7.2 Crack Area 109
4.8 Composite Overlay-Substrate Specimens 112 4.8.1 Age at
Cracking 112 4.8.2 Restrained Shrinkage 115
4.9 Analytical Modelling 117 4.10 Comparison of Age and Net Age
at Cracking 119
5. Conclusions and Recommendations 122 5.1 Introduction 122 5.2
Influence of the Curing Method on Restrained Shrinkage Cracking 122
5.3 Influence of the Curing Duration on Restrained Shrinkage
Cracking 123
Table of Contents
-
vii
5.4 Additional Findings 124 5.5 Conclusions 125 5.6
Recommendations for Future Research 126
6. References 127 Appendix A: Detailed Experimentation Results
133
A1: Compressive Strength Results 133 A2: Free Shrinkage Results
136 A3: Tensile Strength Results 142 A4: Elastic Modulus Results
142 A5: Tensile Relaxation Results 143 A6: Ring Test Results 144
A7: Composite Overlay-Substrate Specimen Results 146
A7.1: Overlay Mortar Results 146 A7.2: Substrate Concrete
Results 147
Appendix B: Detailed Analytical Modelling Results 149 B1:
Regression Analysis Results 149 B2: Material Property Data 149
Appendix C: Environmental Recordings 155 Appendix D: Comparison
of Ring Test Setups 156 Appendix E: Assessment of Ethics in
Research Projects (EBE Faculty) 158
Table of Contents
-
viii
List of Figures Figure 1.1: Schematic overview of thesis
outline. 5 Figure 2.1: Overview of design and construction
considerations of bonded concrete overlays (Adapted
from Beushausen & Alexander, 2009). 7 Figure 2.2: Failure
modes of bonded overlays (adapted from Crlsward, 2006). 12 Figure
2.3: Schematic overview of the time-dependent material properties
governing restrained
shrinkage cracking of bonded overlays (adapted from Troxell et
al., 1968). 15 Figure 2.4: Illustration of a simplified
representation of (a) drying conditions, (b) free shrinkage
strains, (c) restrained stresses, and (d) cracking due to
restrained shrinkage of bonded overlays (adapted from Pigeon &
Bissonnette, 1999). 15
Figure 2.5: (a) Schematic representation of chemical and
autogenous shrinkage volume changes of fresh and hardened cement
paste (not drawn to scale) (adapted from Kosmatka et al., 2003),
and (b) relationship between autogenous shrinkage and chemical
shrinkage of cement paste at early ages (adapted from Hammer,
1999). 19
Figure 2.6: Schematic representation of (a) shrinkage-water loss
relationships, and (b) shrinkage and internal relative humidity
relationship during drying (adapted from Mindess et al., 2003).
21
Figure 2.7: Illustration of mechanisms of drying shrinkage; (a)
capillary tension, (b) disjoining pressure, (c) surface tension
(adapted from Mindess et al., 2003), and (d) movement of interlayer
water (adapted from Wittmann, 1982, as cited in Kovler &
Zhutovsky, 2006). 23
Figure 2.8: Basic characteristics of shrinkage when concrete is
(a) dried from age t0 until age t and after which it is
re-saturated, and (b), dried from age t0 until age t after which it
is subjected to cycles of drying and wetting (adapted from
Alexander & Beushausen, 2009). 24
Figure 2.9: The effect of relative humidity on drying shrinkage
and carbonation shrinkage (adapted from Neville, 2004; Alexander
& Beushausen, 2009). 26
Figure 2.10: (a) Typical stress-strain behaviour of individual
phases of concrete (adapted from Mehta et al., 2006), and (b)
different forms of elastic modulus (adapted from Alexander &
Beushausen, 2009). 28
Figure 2.11: Characteristics of tensile relaxation. (a)
Strain-time graph, and (b) stress-time graph as a result of tensile
relaxation (adapted from Alexander & Beushasuen, 2009). 29
Figure 2.12: Influence of w/b ratio on (a) the shrinkage of
cement pastes (adapted from Haller, 1940, as cited in Alexander
& Beushausen, 2009), and (b) the tensile strength of mortar of
various compositions (adapted from Graf et al., 1960, as cited in
Reinhardt, 2013). 30
Figure 2.13: Influence of w/b ratio on (a) 28-day elastic
modulus, and (b) 28-day tensile relaxation (data from Kizito,
2013). Note the scaling used to illustrate the trends. 31
Figure 2.14: Influence of w/b ratio on the (a) age at cracking,
and (b) crack area of ring tests and bonded overlays (data from
Chilwesa, 2012). 31
Figure 2.15: Influence of water content on (a) shrinkage for a
variety of mix proportions (adapted from Hobbs, 1974, as cited in
Alexander & Beushausen, 2009), and (b) influence tensile
relaxation (data from Dittmer, 2013). Note the scaling used to
illustrate the trends. 32
Figure 2.16: Influence of water content on (a) 28-day elastic
modulus, and (b) age at cracking of ring test specimens (data from
Dittmer, 2013). Note the scaling used to illustrate the trends.
33
Figure 2.17: Influence of aggregate content and size on (a)
56-day free shrinkage (50% RH), and (b) 28-day tensile strength
(data from Dittmer, 2013). Note the scaling used to illustrate the
trends. 35
Figure 2.18: Influence of aggregate content and size on (a)
elastic modulus, and (b) tensile relaxation (data from Dittmer,
2013). Note the scaling used to illustrate the trends. 35
List of Figures and Tables
-
ix Figure 2.19: Influence of aggregate content and size on (a)
age at cracking of ring test specimens, and
(b) crack area of ring test specimens (data from Dittmer, 2013).
36 Figure 2.20: Influence of shrinkage-reducing admixtures on (a)
drying shrinkage (Adapted from Shah et
al., 1998), and (b) age at cracking of ring test specimens (data
from Shah et al., 1998). 37 Figure 2.21: Influence of (a) steel
fibre volume on age of visible cracking (adapted from Shah et
al.,
2006), and (b) fibre type and volume on average crack width
(adapted from Grzybowski & Shah, 1990). 38
Figure 2.22: Influence of (a) specimen size on the drying
shrinkage of concrete (adapted from Hansen & Mattock, 1966),
and (b) overlay thickness on the age at cracking (data from Weiss
& Shah, 2002). 39
Figure 2.23: Influence of overlay thickness on crack area
(adapted from Laurence et al., 2000). 39 Figure 2.24: Relation
between shrinkage and time for concretes stored in different
relative humidities
(adapted from Troxell et al., 1958). 40 Figure 2.25: (a)
Unhydrated particles of Portland cement (magnification of 2000),
and (b) partially
hydrated Portland cement (magnification of 4000), as observed
through a scanning electron microscope (Sroos, 1994, as cited in
ACI 308, 2001) 42
Figure 2.26: Influence of curing on (a) compressive strength of
150 300 mm cylinders, and (b) water permeability of concrete
(adapted from Kosmatka et al., 2003). 43
Figure 2.27: Example of the variation of the internal relative
humidity from the surface of concrete, at 7 days and 28 days, after
exposure to a severe summer environment (adapted from Carrier,
1983). 44
Figure 2.28: Nomograph for estimating the rate of evaporation of
water from a concrete surface (adapted from ACI 308, 2001). 46
Figure 2.29: Categorisation of methods of curing concrete
(adapted from Kovler & Jensen, 2007). 47 Figure 2.30: (a) Water
ponding of finished concrete (Online Civil Enigineering, 2010) and
(b) immersion
of finished concrete in a water curing bath. 48 Figure 2.31: (a)
Continuous fogging of a recently cast concrete slab (Infrastructure
Research Institute,
2000), and (b) close-up photo of a fog nozzle atomising water
into fog-like mist (Kosmatka et al., 2003). 49
Figure 2.32: (a) Water spraying of already set concrete
(Specifier, 2013), and (b) concrete covered with a fabric covering
(hessian) to prevent drying of the concrete during intermittent
water sprinkling (New River Bridge, 2011). 50
Figure 2.33: Saturated/wet hessian coverings curing recently
cast concrete (New River Bridge, 2011). 51 Figure 2.34: (a) White
plastic sheets curing concrete (Kosmatka et al., 2003), and (b)
concrete surface
temperature variations under clear, black and white plastic
sheets due to solar radiation (adapted from Wojokowski, 1999).
52
Figure 2.35: Spray application of a curing compound with (a)
hand-operated (Lambert Chemicals, 2014), and (b) power-driven
(Gomaco, 2014) spray equipment to cure concrete. 54
Figure 2.36: A (a) pumice stone particle used as a
water-saturated light-weight aggregate, and (b) dry, collapsed and
saturate, swollen super-absorbent polymers used to internally cure
concrete (Kovler & Jensen, 2007). 57
Figure 2.37: (a) Influence of sealed curing on direct tensile
strength of concrete (adapted from Heilmann et al., 1969, as cited
in Reinhardt, 2013), and (b) influence of plastic sheet curing on
direct tensile strength of mortar (data from Chilwesa, 2012).
62
Figure 2.38: Influence of plastic sheet curing on elastic
modulus (compression) of mortar (data from Chilwesa, 2012). 63
Figure 3.1: Schematic flowchart of the methodology. 66
List of Figures and Tables
-
x Figure 3.2: Grading curve of fine aggregate (Klipheuwel sand)
used. 68 Figure 3.3: Photographs of (a) a ring test specimen, and
(b) parameter experimentation specimens
cured with plastic sheets. 71 Figure 3.4: Photographs of (a) a
ring test specimen, and (b) parameter experimentation specimens
cured with wet cloths (covered plastic sheets). 72 Figure 3.5:
Photographs of (a) a ring test specimen, and (b) parameter
experimentation specimens
cured with the protective coating. 73 Figure 3.6: Photographs of
(a) a ring test specimen, and (b) parameter experimentation
specimens
cured with the curing compound. 74 Figure 3.7: Schematic
representation of the curing regimes considered. 75 Figure 3.8:
Photographs of (a) the large laboratory room in which the specimens
were placed, and (b)
the environmental data logger used to monitor the environmental
conditions. 76 Figure 3.9: Free shrinkage strain prismatic specimen
geometry. 77 Figure 3.10: Photographs of the (a) free shrinkage
specimens placed next to their corresponding overlay
of the composite overlay-substrate specimens in the curing
environment, and (b) the strain extensometer used to take shrinkage
strain readings. 78
Figure 3.11: (a) Tensile strength notched dog-bone specimen
geometry, and (b) notch detail. 79 Figure 3.12: Photographs of the
(a) general UTM setup for tensile strength testing, and (b) a
failed
tensile strength specimen showing failure path through the
notched cross-section. 79 Figure 3.13: (a) Elastic modulus specimen
geometry, and (b) a photograph of the elastic modulus
specimens drying in the appropriate environment. 80 Figure 3.14:
Photographs of the (a) general UTM setup for elastic modulus tests,
and (b) measurement
of longitudinal strains of the elastic modulus specimens under
load. 81 Figure 3.15: Tensile relaxation dog-bone specimen
geometry. 81 Figure 3.16: Photographs of (a) a tensile relaxation
specimen sealed with paraffin wax, and (b) the
general UTM setup for tensile relaxation tests. 82 Figure 3.17:
(a) Top view illustration (with dimensions labelled), and (b)
section A-A of full ring test
setup. Not drawn to scale. 83 Figure 3.18: (a) A typical
photograph of a cracked ring test specimen showing crack
measuring
locations, and (b) a screenshot of the crack-length algorithm
used to calculate the crack length. 85
Figure 3.19: 3D illustration of composite overlay-substrate
specimens. 87 Figure 3.20: Photographs of the substrate surface (a)
after casting before surface preparation, and (b)
after 24 hours and after surface preparation. 87 Figure 3.21: 3D
illustration showing the location of strain targets used to monitor
the substrate slab
shrinkage. 88 Figure 3.22: Photographs showing (a) the sides of
the overlays sealed with paraffin wax, and (b) the
curing regimes of a set of overlays and the free shrinkage
specimens placed next to their overlay counterparts. 89
Figure 3.23: 3D illustration showing the location of strain
targets used to monitor the overlay shrinkage. 90 Figure 4.1:
28-Day compressive strength development of the various overlay
mortar mixes. 95 Figure 4.2: (a-d) 28-Day free shrinkage strain
development, and (e-h) 56-day free shrinkage strain
development from the completion of curing of the w/b 0.60 mix
for the various curing regimes (see Appendix A2 for full-size
figures). 97
Figure 4.3: 56-day free shrinkage strain development for the
various curing regimes (see Appendix A2 for full-size figures).
100
List of Figures and Tables
-
xi Figure 4.4: 7- and 28-day direct tensile strength results of
the moderate strength mix (w/b 0.60) with
various curing regimes. 101 Figure 4.5: 7- and 28-day elastic
modulus results of the moderate strength mix (w/b 0.60) cured
with
various curing regimes. 103 Figure 4.6: 7- and 28-day tensile
relaxation results of the moderate strength mix (w/b 0.60) cured
with
various curing regimes. 104 Figure 4.7: Overview of the age at
cracking of ring test specimens made from various mortar
overlay
mixes cured under various curing regimes. 106 Figure 4.8:
Influence of the curing regime on the age at cracking of ring test
specimens. 107 Figure 4.9: Influence of the curing regime on the
net age at cracking of ring test specimens. 107 Figure 4.10:
Influence of the mix type on the age at cracking of ring test
specimens. 108 Figure 4.11: Influence of the curing regime on the
14-day crack area of ring test specimens. 109 Figure 4.12:
Influence of the mix type on the 14-day crack area of ring test
specimens. 110 Figure 4.13: Correlation between 14-day crack area
and (a) free shrinkage strain at the time of crack
measurement, and (b) net age at cracking. 111 Figure 4.14:
Correlation between stress rate and net age at cracking (adapted
from See et al., 2004) 111 Figure 4.15: Overview of the age at
cracking of composite overlay-substrate specimens made from
various mortar overlay mixes cured under various curing regimes.
112 Figure 4.16: Influence of the curing regime on the age at
cracking of overlay-substrate specimens. 113 Figure 4.17: Influence
of the curing regime on the net age at cracking of
overlay-substrate specimens. 113 Figure 4.18: Influence of the mix
type on the age at cracking of overlay-substrate specimens. 114
Figure 4.19: 28-day free and restrained shrinkage strain
development of the moderate strength mix (w/b
0.60) for the various curing regimes. 115 Figure 4.20:
Analytical modelling results of the moderate strength mix (w/b
0.60) for the various curing
regimes. 117 Figure 4.21: Influence of the curing regime on the
(a) age at cracking, and (b) net age at cracking
predicted from analytical modelling. 118 Figure 4.22:
Correlation between ring test and composite overlay-substrate
specimen (a) age at
cracking, and (b) net age at cracking. 119 Figure 4.23:
Comparison of the age at cracking of the moderate strength mix (w/b
0.60). 120 Figure 4.24: Comparison of the net age at cracking of
the moderate strength mix (w/b 0.60). 120 Figure A1: 90-Day
compressive strength development of the various overlay mortar
mixes. 135 Figure A2: 28-day free shrinkage strain development of
the (a) w/b 0.45, and (b) w/b 0.60 laboratory-
made mortars cured using various curing regimes. 136 Figure A3:
28-day free shrinkage strain development of the (a) w/b 0.80
laboratory-made mortar, and
(b) commercial repair mortar cured using various curing regimes.
137 Figure A4: 56-day free shrinkage strain development of the (a)
w/b 0.45, and (b) w/b 0.60 laboratory-
made mortars cured using various curing regimes. 138 Figure A5:
56-day free shrinkage strain development of the (a) w/b 0.80
laboratory-made mortar, and
(b) commercial repair mortar cured using various curing regimes.
139 Figure A6: 56-day free shrinkage strain development from the
completion of curing of the (a) w/b 0.45,
and (b) w/b 0.60 laboratory-made mortars cured using various
curing regimes. 140 Figure A7: 56-day free shrinkage strain
development from the completion of curing of the (a) w/b 0.80
laboratory-made mortar, and (b) commercial repair mortar cured
using various curing regimes. 141
Figure A8: 90-Day compressive strength development of the
concrete substrate. 147 Figure A9: Shrinkage strain measurements of
the concrete substrate. 148
List of Figures and Tables
-
xii Figure A10: Rate of shrinkage of the top of the concrete
substrate slab. 148 Figure C1: Environmental recordings of the
temperature and relative humidity of the environment
within which the various specimens were allowed to cure. 155
Figure D1: (a) Top view illustration (with dimensions and induced
stresses labelled) and (b) section A-
A of full ring test setup. Not drawn to scale. 156
List of Tables Table 2.1: Simplified summary of factors
influencing restrained shrinkage cracking. 41 Table 2.2:
Recommended curing regime for bonded concrete overlays 60 Table
2.3: Recommended curing durations for repair materials (adapted
from ICRI, 1996) 61 Table 3.1: Mix proportions and key mix
properties of the laboratory-made mortars. 69 Table 3.2: Mix
proportions and key mix properties of the commercial repair mortar.
70 Table 3.3: Key mix properties of the substrate concrete. 86
Table 3.4: Summary of the experimental tests conducted. 91 Table
4.1: Mix and curing regime notations used in the results presented.
94 Table 4.2: 28-Day average degree of restraint for the moderate
strength overlays (w/b 0.60) cured
using the respective curing regime. 116 Table 4.3: Comparison of
the age at cracking results of the moderate strength mix (w/b
0.60). 120 Table 4.4: Comparison of the net age at cracking results
of the moderate strength mix (w/b 0.60). 120 Table A1: Compressive
strength results of w/b 0.45 mix specimens. 133 Table A2:
Compressive strength results of w/b 0.60 mix specimens. 134 Table
A3: Compressive strength results of w/b 0.80 mix specimens. 134
Table A4: Compressive strength results of commercial repair mortar
(CRM) specimens. 135 Table A5: 7-Day tensile strength results for
the w/b 0.60 mix specimens using the respective curing
regime. 142 Table A6: 28-Day tensile strength results for the
w/b 0.60 mix specimens cured using the respective
curing regime. 142 Table A7: 7-Day elastic modulus results of
the w/b 0.60 mix specimens cured using the respective
curing regimes. 142 Table A8: 28-Day elastic modulus results of
the w/b 0.60 mix specimens cured using the respective
curing regimes. 143 Table A9: 7-Day 24-hour tensile relaxation
results for the w/b 0.60 mix specimens cured using the
respective curing regime. 143 Table A10: 28-Day 24-hour tensile
relaxation results for the w/b 0.60 mix specimens cured using
the
respective curing regime. 143 Table A11: Age at cracking and net
age at cracking results of ring test specimens. 144 Table A12:
3-day and 14-day crack areas of ring test specimens. 145 Table A13:
Age at cracking and net age at cracking results of the composite
overlay-substrate
specimens. 146 Table A14: Compressive strength results of the
concrete substrate (SUB). 147 Table B1: Regression coefficients
obtained from regression analysis. 149 Table B2: Material property
data for the air cured (AC) w/b 0.60 mix. 149 Table B3: Material
property data for the 2-day plastic sheets cured (PS (2D)) w/b 0.60
mix. 150 Table B4: Material property data for the 2-day wet cloth
cured (WC (2D)) w/b 0.60 mix. 151 Table B5: Material property data
for the 7-day plastic sheets cured (PS (7D)) w/b 0.60 mix. 151
Table B6: Material property data for the 7-day wet cloth cured (WC
(7D)) w/b 0.60 mix. 152
List of Figures and Tables
-
xiii Table B7: Material property data for the w/b 0.60 mix cured
with a protective coating (PC). 153 Table B8: Material property
data for the w/b 0.60 mix cured with a curing compound (CC). 154
Table D1: Comparison of ASTM C1581-09a (2009) ring test apparatus
dimensions to those used in
the experimentation. 156 Table D2: Comparison of ASTM C1581-09a
(2009) ring tests induced stresses to those used in the
experimentation. 157
List of Figures and Tables
-
xiv
Abbreviations 2D/7D 2-Day/7-Day Curing Duration AASHTO American
Association of State Highway and Transport Officials AC Air Curing
ACI American Concrete Institute ASTM American Society for Testing
and Materials C-S-H Calcium Silicate Hydrate CC Curing Compound
CCAA Cement Concrete & Aggregates Australia CoV Coefficient of
Variation CRA Crack-Reducing Admixture CRM Commercial Repair Mortar
CSF Condensed Silica Fume FA Fly Ash GGBS Ground Granulated
Blastfurnace Slag GGCS Ground Granulated Corex Slag ICRI
International Concrete Repair Institute ITZ Interfacial Transition
Zone LDPE Low-Density Polyethylene LWA Light-Weight Aggregate
MATLAB Matrix Laboratory NWA Normal-Weight Aggregate OPC Ordinary
Portland Cement OPI Oxygen Permeability Index PS Plastic Sheets PC
Protective Coating PVA Polyvinyl Acetate RH Relative Humidity SANS
South African National Standard SAP Super-Absorbent Polymer SRA
Shrinkage Reducing Admixture SSD Saturated Surface Dry UTM
Universal Testing Machine WC Wet Cloth WSI Water Sorptivity Index
w/b Water-Binder Ratio w/c Water-Cement Ratio
Abbreviations
-
xv
List of Variables , Free shrinkage strain regression
coefficients (unitless) , Tensile strength regression coefficients
(unitless) , Elastic modulus regression coefficients (unitless) ,
Tensile relaxation regression coefficients (unitless) Inner
diameter of mortar annulus (mm) Outer diameter of mortar annulus
(mm) Mean elastic modulus in the interval from 1 to (GPa) Tensile
elastic modulus (GPa) Tensile strength (MPa) , Tensile strength at
(MPa) Height of steel ring (mm) Disjoining pressure (kPa) Capillary
pressure (kPa) Surface free energy (J/g) Arbitrary point in time
(any unit of time) Time i (days) Initial point in time (any unit of
time) Thickness of inner steel ring (mm) Thickness of mortar
annulus (mm) Thickness of outer steel ring (mm) Thickness of wooden
board (mm) Degree of restraint (unitless) Mean tensile relaxation
in the interval from 1 to (%) Free shrinkage strain ( 106) , Change
in free shrinkage strain from 1 to ( 106) Tensile stress (MPa) ,
Tensile stress at (MPa) , Exterior circumferential stress (MPa) ,
Interior circumferential stress (MPa) , Exterior radial stress
(MPa) , Interior radial stress (MPa) Poissons ratio (unitless)
Tensile relaxation factor (unitless) Mean tensile relaxation factor
in the interval from 1 to (unitless)
List of Variables
-
1
1. Introduction 1.1 Background to Research Problem As more and
more concrete structures are being constructed worldwide, the
amount of concrete infrastructure requiring repair and
rehabilitation is increasing. This is due to the fact that over
time concrete structures deteriorate, often prematurely, due to
numerous factors, resulting in structures not reaching their
intended service lives. In order to restore safety and/or
serviceability to the deteriorated concrete structures they must be
repaired or rehabilitated. The costs associated with these concrete
repairs may be significant and can have a large influence on the
feasibility of maintaining structures, particularly large
infrastructure (Beushausen & Alexander, 2009). In the United
States of America alone, the American Society of Civil Engineering
reported that in 2006 the annual cost of concrete repair and
rehabilitation was estimated to be $18 to $21 billion (Emmons,
2006; Raupach, 2006). While the costs associated with repair of
concrete structures can be substantial, the costs of poorly design
or executed concrete repairs may be even higher (Beushausen &
Alexander, 2009). This has resulted in an increasing need for
effective concrete repair techniques.
Bonded mortar or concrete overlays, henceforth referred to as
bonded overlays, are a common type of concrete repair technique
that involves the casting of a new repair material, in the form of
an overlay, in the place of the damaged or deteriorated portion of
concrete. Bonded overlays are however prone to premature failure
due to differential volume changes between the newly cast overlay
and the existing substrate. Failure of bonded overlays is primarily
manifested as debonding or cracking (Beushausen, 2005; Bissonnette
et al., 2011). Debonding can be relatively easily prevented by
ensuring a high bond strength between the overlay and substrate by
thorough substrate preparation and cleaning, overlay compaction,
and curing. Failure due to restrained shrinkage cracking on the
other hand has proven significantly more difficult to prevent
(Beushausen & Alexander, 2009; Bissonnette et al., 2011).
The volume changes experienced by the overlay that lead to its
failure arise due to the inherent shrinkage of cementitious
materials. The restraint provided by the substrate to the shrinkage
of the overlay results in the development of tensile stresses
within the overlay. These stresses are proportional to the
shrinkage strain and elastic modulus of the overlay but are
partially reduced due to tensile relaxation. If the tensile
stresses induced are greater than the tensile strength of the
overlay then cracking of the overlay, termed restrained shrinkage
cracking, will begin to develop. Not only is cracking aesthetically
unpleasing, but it also hinders the primary function of bonded
overlays - to improve durability (Banthia & Gupta, 2009).
Despite the relatively sound understanding of the mechanisms of
restrained shrinkage cracking and the numerous measures that can be
taken to mitigate it, such as the use of an increased water-binder
(w/b) ratio, reduced paste content, appropriate cement extenders,
increased aggregate content, admixtures, fibres, etc., restrained
shrinkage cracking still often occurs. Another measure thought to
increase an overlays resistance to restrained shrinkage cracking is
curing. However, the
Chapter 1: Introduction
-
2 influence of curing on restrained shrinkage cracking has not
yet been thoroughly investigated, and so its influence is not fully
known and understood.
1.2 Problem to be Investigated Bonded overlays often fail
prematurely due to restrained shrinkage cracking. In an attempt to
remedy this problem, it is suggested that the use of curing may
help to prevent or delay premature failure due to restrained
shrinkage cracking. However, the influence of curing on restrained
shrinkage cracking has not yet been thoroughly investigated and so
its influence is not fully known and understood. The problem to be
investigated, therefore, is to determine the influence of different
curing regimes (method and duration) on restrained shrinkage
cracking of bonded overlays, and the relevant time-dependent
properties of concrete (shrinkage, tensile strength, elastic
modulus, and tensile relaxation), using the latter to provide an
explanation of the influence of curing in terms of the
time-dependent material properties governing restrained shrinkage
cracking.
1.3 Hypothesis This research proposes that different curing
methods have different influences on the age at which restrained
shrinkage cracking is observed and corresponding crack area
measured, and that increasing the curing duration of bonded
overlays increases the age at which restrained shrinkage cracking
occurs. Furthermore, it is proposed that there exists an optimal
curing regime for a particular overlay mix, and that this optimal
curing regime is governed by the interrelation of the
time-dependent material properties of the overlay.
1.4 Research Aim and Objectives The aim of the proposed research
is it to determine the influence of different curing regimes on the
cracking of non-load-bearing bonded overlays due to restrained
shrinkage. The aforementioned aim will be achieved by following the
four objectives of the research:
Experimentally determine the influence of curing on the
time-dependent material properties (shrinkage, tensile strength,
elastic modulus, and tensile relaxation), of specimens of a
moderate strength mortar subjected to various curing regimes.
Experimentally determine the age at cracking and crack area due
to restrained shrinkage on specimens of various mortar mixes
subjected to various curing regimes.
Analytically model, using the experimentally determined
time-dependent material properties, the age at cracking due to
restrained shrinkage cracking of a moderate strength mortar
subjected to various curing regimes.
Critically evaluate the performance of the various mortar mixes
subjected to various curing regimes, using the results of the
influence of curing on the time-dependent material properties and
the associated analytical modelling to provide fundamental
insight.
Chapter 1: Introduction
-
3
1.5 Key Research Questions The key research questions of the
research are listed herein.
What effect do different curing methods have on the governing
time-dependent material properties, the age at cracking, and crack
area, of restrained shrinkage cracking of bonded overlays?
What effect does the curing duration have on the governing
time-dependent material properties, the age at cracking, and crack
area, of restrained shrinkage cracking of bonded overlays?
Are there preferential curing regimes for different overlay
mixes? If so, which curing regime is preferential for each overlay
mix and to what degree can restrained shrinkage cracking be reduced
with the respective curing regimes?
1.6 Scope and Limitations There are numerous factors that
influence the performance of bonded overlays with regard to
restrained shrinkage cracking and so it is necessary to define the
scope of research and to identify limitations as it is not possible
to include all the influencing factors in a limited research
project.
The scope of the research is to determine the influence of a
select number of curing regimes, exposed to a single environmental
condition, on the time-dependent material properties governing
restrained shrinkage cracking (shrinkage, tensile strength, elastic
modulus, and tensile relaxation) and on restrained shrinkage
cracking directly (age at cracking and crack area) of a limited
number of overlay mixes. The results of the material property tests
are to be used in an analytical model which in turn, is used to
explain the trends observed from the results of the direct
restrained shrinkage tests, namely the age at cracking.
The limitations of the research are listed herein.
The number of overlay mixes used in experimentation was limited
to four due to the large number of tests and the extensive amount
of time required for the full range of tests that needed to be
conducted on each mix. Furthermore, these mixes were limited to
laboratory-made, cementitious mortars and a commercial
polymer-modified cementitious mortar. No ordinary,
high-performance, self-compacting or fibre reinforced concretes, or
resin-based overlays were considered.
Only five curing methods and a total of two curing durations of
2 and 7 days, making a total of seven curing regimes, were used due
to similar reasons outlined in the previous point.
Material property tests (shrinkage, tensile strength, elastic
modulus and tensile relaxation) were limited to two test ages per
mix. This was due to similar reasons outlined in the first
point.
Restrained shrinkage cracking tests were limited to the ring
test and composite overlay-substrate specimens.
The analytical modelling was limited to a simple material
property analytical model.
Chapter 1: Introduction
-
4
1.7 Thesis Outline The first chapter (Chapter 1) provides an
introduction to the research by establishing the background to the
research problem, listing the aims and objectives as well as the
key research questions. The scope of the research is also defined
and the limitations are noted. This is followed by a review of
relevant literature (Chapter 2). The literature is reviewed by
first providing a brief overview of the fundamentals of bonded
overlays, including applications and design and construction
considerations. An overview of the failure of bonded overlays is
then presented after which failure due to debonding and restrained
shrinkage cracking is discussed respectively. Following this,
restrained shrinkage cracking of bonded overlays, particularly the
material properties governing it and the factors influencing it, is
reviewed in further detail. An extensive review of curing of
concrete is then provided. General aspects of curing are presented
first followed by a review of the various curing methods available
after which more specific aspects of curing relating to bonded
overlays are then investigated. The experimental approach taken,
curing regimes considered, and details of the tests conducted and
analytical modelling carried out are then discussed in Chapter 3.
The results obtained from the experimentation and analytical
modelling are then presented and discussed in Chapter 4.
Conclusions and recommendations made based on the results presented
are provided in Chapter 5. A list of references used is provided in
Chapter 6. Relevant detailed experimental results, detailed
analytical modelling results, environmental recordings, and a
comparison of ring tests setups are included in Appendix A to D
respectively. A schematic representation of the thesis outline is
provided in Figure 1.1.
Chapter 1: Introduction
-
5
Figure 1.1: Schematic overview of thesis outline.
Introduction
Fundamentals of Bonded Overlays
Chapter 2: Literature Review
Chapter 1: Introduction
Chapter 3: Methodology
Chapter 4: Results and Discussions
Chapter 5: Conclusions and
Recommendations
Restrained Shrinkage Cracking of Bonded Overlays
Curing of Concrete
Summary
Restrained Shrinkage Cracking
Time-Dependent Material Properties
Methodology
Results and Discussions
Conclusions and Recommendations
Explanation of Influence of Curing on Restrained Shrinkage
Cracking
Restrained Shrinkage Cracking
Time-Dependent Material Properties
Experimental Tests
Curing Regimes
Failure of Bonded Overlays
Analytical Modelling
Analytical Modelling
Overview and Approach
Chapter 1: Introduction
-
6
2. Literature Review 2.1 Introduction This chapter covers the
fundamentals of aspects relating to the performance of bonded
overlays, with particular emphasis on those that influence cracking
due to restrained drying shrinkage of non-load-bearing bonded
overlays. A brief overview on the fundamentals of bonded overlays
is provided first in Section 2.2, which includes applications and
design and construction considerations. Failure of bonded overlays,
with particular emphasis on debonding and restrained shrinkage
cracking, is then covered in Section 2.3. This is followed by a
review of the restrained shrinkage cracking of bonded overlays in
Section 2.4, focusing on the material properties governing it and
the factors that influence it. A detailed review of curing of
concrete is then provided in Section 2.5, in which general aspects
of curing are presented first, followed by a review of various
curing methods after which more specific aspects of curing relating
to bonded overlays are then investigated. Lastly, a summary of the
literature review is presented in Section 2.6.
2.2 Fundamentals of Bonded Overlays 2.2.1 Definition A bonded
overlay is a layer of mortar or concrete typically 25 to 150 mm in
thickness which is placed on top of a pre-existing concrete
surface, known as the substrate, to which a bond develops (Tayabji
et al., 2009). The primary use of a bonded overlay is to extend the
service life of a structure by restoring or improving the
structural integrity, serviceability and/or durability. Bonded
overlays used for repairs that are relatively small in surface area
are commonly termed patch repairs (Beushausen & Alexander,
2009).
2.2.2 Applications Over time concrete structures start to
deteriorate which leads to a decrease in the structural
performance, reduction in the durability of the structure and may
render the concrete aesthetically unpleasing. Repair of the
concrete is required in order to reinstate the structure to its
original specifications in order to prevent further deterioration
of the member or structure. Among the possible choices for repair
or rehabilitation, bonded overlays are considered the most
economical option for concrete structures (Bissonnette et al.,
2011). They are therefore widely used for repairing, as well as
strengthening concrete structures. Successful implementation of
bonded overlays can result in a number of advantages, these
include:
Restoration of structural strength of the repaired member
Restoration of structural integrity (strength, abrasion resistance,
etc.) of the repaired member Restoration or improvement of the
appearance of the repaired member Restoration of the durability
performance of the repaired member Improvement of the
serviceability of the repaired member
Applications of bonded overlays, as previously mentioned,
include that of thin surface repairs and/or strengthening of
concrete that has delaminated, spalled and/or cracked. They are
also
Chapter 2: Literature Review
-
7 particularly suited for members with large surface areas
(slabs on grade, pavements and tunnel linings, etc.) where the
concrete can be poured or sprayed (Granju et al., 2004).
Additionally, they are used on precast members which receive an
in-situ topping (Beushausen, 2005). The research conducted in this
dissertation largely relates to the use of non-load-bearing bonded
overlays for concrete repair, and so the use of the term bonded
overlay will henceforth refer to this type of overlay.
2.2.3 Design and Construction Considerations In order to achieve
successful implementation of bonded overlays, there are various
design and construction factors that must be considered to ensure
sufficient bond strength and crack resistance. The former being
largely linked to substrate surface preparation and properties
whilst the latter being linked to overlay material properties
(tensile strength, shrinkage, elastic modulus, and relaxation). An
overview of these factors are shown diagrammatically in Figure 2.1
and are elaborated on in Sections 2.2.3.1 to 2.2.3.4. Further
details of the design and construction of bonded overlays are given
in standards and specifications (ICRI ACI, 2013) and in
recommendations (Bissonnette et al., 2011). However, standards and
specifications for design of bonded overlays generally lack scope
and detail, and so it is often left to the engineer to specify
appropriate materials and construction procedures (Beushausen,
2005). This is largely due to the fact that each repair situation
is unique and so needs to be treated as such.
Figure 2.1: Overview of design and construction considerations
of bonded concrete
overlays (Adapted from Beushausen & Alexander, 2009).
2.2.3.1 Substrate Preparation and Properties Correct substrate
preparation is generally considered to be the most crucial steps in
a concrete patch repair as it significantly influences the bond
strength of the substrate. A poorly prepared substrate will always
limit the performance of an overlay, no matter how good the repair
material or how well it was applied. The substrate should therefore
be prepared by removing deteriorated concrete and creating a good
surface interface. Substrate preparation consists of two distinct
processes that are both equally important substrate surface removal
and substrate surface cleaning (Beushausen & Alexander, 2009).
These two processes, as well as substrate moisture condition and
the use of bonding agents, are discussed further in the following
four subsections.
ENVIRONMENT Overlay material properties
(fresh and hardened properties) Curing procedures
Edge conditioning
Substrate preparation Roughness Cleanliness Bonding agent
Rebar preparation: Cleaning from rust Protective coating
Undercutting of corroded rebars
Substrate Removal of
contaminated Prevention of
Chapter 2: Literature Review
-
8 Substrate Surface Removal Substrate surface removal involves
the bulk removal of poor-quality (including the laitance layer) and
deteriorated substrate concrete, and any previously applied
coatings. This should be carried out until the substrate is sound
and has sufficient surface texture to form a strong bond. Edges of
removed concrete should be straight and normal to free edges (as
shown in Figure 2.1) to prevent feather edges which result in
localised overlay failure (Beushausen & Alexander, 2009).
Additionally, any corroded steel must be undercut and all corrosion
products should be removed.
Mechanical surface removal techniques such as jackhammering,
milling, scarifying, or hand-chipping are very efficient in
removing poor concrete however are likely to cause microcracking of
the substrate. Microcracks, commonly referred to as bruising,
result in zones of weakness that are prone to bond failure. The
extent of bruising is dependent of the surface removal method and
the quality of the concrete, however the bruised layer is typically
of the order of 3 mm thick (Bissonnette et al., 2013).
Water-jetting (hydromilling) and sandblasting are alternative
surface preparation techniques that have been shown to be efficient
in removing concrete and produce substrate surfaces with sufficient
surface texture, without causing microcracking (Silfwerbrand,
1990). It has been shown that mechanical removal, which is
sometimes necessary to remove deteriorated concrete, followed by
water-jetting can achieve good bond strength (Silfwerbrand &
Peterson, 1993).
Substrate Surface Cleaning Substrate surface cleaning refers to
the removal of loose pieces of concrete/debris and any contaminants
such as dust, oil, grease, etc. that may inhibit bond strength. It
is essential not only that the substrate has been cleaned but that
it is clean at time of overlay casting, as the substrate may have
been contaminated by pollutants (from construction and/or the
environment) in the interim.
The use of water-jetting as a surface removal technique provides
a very clean substrate which is dust-free and also removes
corrosion products. It should however be followed by thorough
removal of loose concrete particles to prevent unhydrated cement
particles from bonding to the saturated substrate surface
(Bissonnette et al., 2013).
Substrate Moisture Condition Not only must a substrate be free
of poor-quality and deteriorated concrete, and be clean of any
loose particles and contaminates, but it should also be in an
appropriate moisture condition. A substrate with a surface that is
too wet will cause an increase in the w/b ratio of a thin layer of
the overlay in contact with the substrate lowering its strength.
Additionally, the water may block open pores causing a reduction in
mechanical interlocking and hence lower bond strength. A dry
substrate will absorb water, resulting in a harsh overlay mix that
may not interlock sufficiently with the substrate, and may lead to
less water than that required for full cement hydration. Substrates
are therefore commonly specified to be in a saturated surface dry
(SSD) condition (Beushausen & Alexander, 2009).
Chapter 2: Literature Review
-
9 Bonding Agents Bonding agents, consisting of commercial
products or site-made cement slurries, are specified by many
engineers to improve bond strength. Extensive research has been
carried out on the use of bonding agents which has resulted in
mixed opinions. It is however generally agreed upon that bonding
agents cannot make up for poor substrate surface preparation. They
have been shown to improve bond strength when using overlays of low
workability which cannot easily flow into the pores and rough
surface texture of the substrate however when used to bond a highly
workable overlays they have no beneficial influence on bond
strength (Beushausen, 2010). Bonding agents may cause failure if
used inappropriately, whereby they are allowed to dry/set/cure
prior to overlay placement causing them to act as a bond breaker,
and so should be used with care.
2.2.3.2 Overlay Material Types and Properties The type of
overlay material used and its associated properties in both the
fresh and hardened state are of importance as they strongly
influence the bond strength and durability. Additionally, the
hardened material properties also govern the long-term overlay
performance, as they collectively determine the crack resistance of
the overlay (Beushausen & Alexander, 2009). Commonly used
repair materials are discussed and desired properties of overlays
in the fresh and hardened state are provided in the three
subsections that follow.
Material Types There are a large number of repair materials that
may be used for bonded overlays. These materials can be divided
into cementitious repair materials which include concrete or mortar
made from pure or blended Portland cements, and polymeric repair
materials which include polymer or polymer-modified Portland cement
concrete or mortar (Fowler, 2009). In addition, these repair
materials may be fibre-reinforced.
Cementitious repair materials that are Portland cement-based are
commonly used for concrete repairs as they are widely available,
are low in cost compared to many other repair materials, and are
relatively easy to place and finish (Fowler, 2009). Blended
Portland cements may be used to reduce cost or provide improved
durability of the repair. Cementitious repair materials are however
limited due to shrinkage which may result in cracking of the
overlay (see Section 2.3.3).
Polymer concrete or mortar differ from polymer-modified concrete
or mortar in that the former uses a polymer binder (as opposed to a
cementitious binder) whilst the later uses cement as a binder with
a polymer as an admixture. Polymer and polymer-modified concretes
or mortars have some improved mechanical and durability properties
however are more expensive cementitious repair materials as the
polymers that are used are the most expensive component (Fowler,
2009).
Concretes (cementitious or polymeric), due to the inclusion of
aggregates, are generally used for thicker repairs. It should be
noted that mortars are more susceptible to cracking as they exhibit
considerably higher shrinkage (as opposed to concretes) due to
their higher water and cement (paste) content and lower aggregate
content (Fowler, 2009).
The choice of repair material used is based on many factors
including construction, performance, and durability requirements
(Fowler, 2009). More detailed information on the various repair
material types and the selection thereof are given by the American
Concrete Institute (2006).
Chapter 2: Literature Review
-
10 Fresh Properties The workability of the fresh overlay
influences its ability to fill open cavities/voids of the substrate
surface and therefore determines the effective contact area of the
overlay-substrate bond. Highly workable mixes (used for horizontal
or vertical repairs using formwork), promote capillary suction from
the substrate resulting in a stronger bond through increased
anchorage with the substrate surface. The converse is true for
stiff mortar overlays (commonly used for small patch repairs) which
have much lower capillary suction and corresponding bond strength.
Such overlays may require the use of a bonding agent to improve
adhesion, as discussed in Section 2.2.3.1. In general, concrete or
mortar of conventional workability is sufficient.
Hardened Properties Hardened properties of the overlay that must
be considered are load-bearing (compressive strength, tensile
strength, elastic modulus, etc.), dimensional stability (shrinkage,
creep and relaxation, and thermal deformation) and durability
(permeability, absorption, etc.) properties.
Overlays are not required to have a high compressive strength
and elastic modulus, particularly those used in non-load-bearing
applications. Higher strength concretes with corresponding high
elastic moduli actually have lower resistance to cracking. This is
due to the fact the overlays are deformation governed, and not
force controlled. Their high stiffness leads to higher stresses,
than that of low strength overlays which are considerably less
stiff and lead to lower stresses. However, a high tensile strength
will increase the overlay materials resistance to cracking.
In general, the substrate undergoes significantly less shrinkage
than that of a newly cast overlay and so it is beneficial to lower
the rate and minimise the magnitude of shrinkage of overlay
materials. This minimises differential deformations and thus the
induced stresses that may cause cracking of the overlay. High
levels of relaxation of the overlay which reduce the stresses
induced by shrinkage and low thermal deformations which limit
differential deformation as a result of temperature changes are
also advantageous in this regard (Beushausen & Alexander,
2009).
2.2.3.3 Overlay Application Methods There are various
application methods for bonded overlays used for concrete repair,
each application method being suitable for different repair
applications. The application method should be chosen ensuring that
the repair material fully encapsulates the exposed reinforcing
steel, achieves satisfactory bond with the substrate, and fills the
prepared cavity without segregating. Failure of the application
method to achieve this will result in a poorly performing repair,
irrespective of the quality of the repair material. Selection of
the appropriate application method is therefore crucial for a
successful repair. Brief descriptions and common repair application
methods are provided in the subsections that follow. Further
information on these (and other) application methods is detailed by
Emmons (1994) and the International Concrete Repair Institute
(1996).
Hand-/Trowel-Applied In the hand-/trowel-applied application
method, the repair material is mixed into a troweable, non-sag
consistency. The repair material is pressed into the substrate pore
structure to develop good contact, without the formation of voids.
The repair material is designed to hang in place until subsequent
layers are added. Each layer should be roughened to promote bond to
the next layer. This application method is typically used for small
surface restoration repairs on vertical and
Chapter 2: Literature Review
-
11 overhead locations when reinforcing steel is not present. The
presence of reinforcing steel makes it difficult to consolidate the
repair material and provide complete encapsulation of the
reinforcing steel. Problems associated with this technique
typically include poor bond between layers and voids around
embedded reinforcing steel (Emmons, 1994).
Form and Cast-in-Place For the form and cast-in-place
application method, formwork is used to confine the prepared cavity
within which the repair material is cast. It is important that the
repair material is of adequate flowability and is consolidated by
rodding or conventional vibration to aid in the formation of a good
bond with the concrete substrate. This application method is one of
the most common methods repair of vertical (columns, walls and slab
edges), and in some cases, overhead locations (slab soffits). It
can be used for partial- or full-depth repairs. An advantage of
this application method is that curing can easily be employed by
means of formwork retention. Problems associated with this
technique are that formed surfaces make the placement of bonding
agents difficult and that in some cases complete filling of the
cavity may be difficult (Emmons, 1994).
Dry or Wet Spraying Dry or wet spraying, also known as dry-mix
or wet-mix shotcrete, involves spraying (at high velocity) of the
repair material onto the repair area using compressed air. For dry
spraying, the repair material is placed dry or slightly damp into
the pan mixer and is then transported via the hose to the exit
nozzle where water and admixtures (if any) are added to the mix. In
contrary, for wet spraying, the pre-mixed repair material is placed
pump and is then transported via the hose to the nozzle where
compressed air and admixtures (if any) are added to the mix.
Dry-spray repair materials should have well-graded aggregates and
well-proportioned mixes to compensate for rebound losses whilst
wet-spray repair materials should be pumpable and should not sag.
The spray application methods are economical for large repair areas
and are therefore typically used for large vertical or overhead
repairs where no reinforcing bars or thin reinforcing bars are
present with minimal congestion. Problems associated with this
technique include sloughing off of the concrete or mortar due to
excessively thick layer placement and the formation of voids behind
reinforcement (Emmons, 1994).
2.2.3.4 Environmental Conditions and Curing The environmental
conditions, more specifically the temperature and relative
humidity, significantly influence the development of the mechanical
and durability properties of concrete and mortar. In cold, wet
environments the cold temperature slows down the rate of hydration
and so the development of mechanical and durability properties are
slowed. In hot, dry environments the high temperature and low
relative humidity increase the rate of drying from the concrete
surface and therefore the rate at which moisture is lost from the
concrete (Banthia & Gupta, 2006). This adversely effects
hydration and so the mechanical and durability properties may not
develop sufficiently resulting in failure of the overlay or more
protection of the underlying reinforcement (Hassan et al., 2000).
It has been suggested that these adverse environmental effects can
be partially mitigated through curing (Beushausen & Alexander,
2009; Bissonnette et al, 2013). This is the focus of the research
at hand and shall be explored in more detail throughout the
remainder of this dissertation.
Chapter 2: Literature Review
-
12
2.3 Failure of Bonded Overlays The failure of bonded overlays is
of discussed herein. An overview of the failure of bonded overlays,
including the typical causes of failure and common failure modes,
is provided in Section 2.3.1. The failure mechanisms and factors
influencing the predominant failure modes, debonding and cracking,
are then discussed in further detail in Sections 2.3.2 and 2.3.3
respectively.
2.3.1 Overview Bonded overlays are a widely used, economical
repair solution for concrete structures that afford many advantages
such as the restoration of structural integrity, durability, and
the improvement of aesthetics of a concrete member, as previously
stated in Section 2.2.2. Despite these advantages, the performance
of bonded overlays is often adversely affected by premature failure
of the overlay which is predominantly manifested as debonding
and/or cracking (Beushausen, 2005; Beushausen & Alexander,
2006b; Crlsward, 2006; Bissonnette et al., 2011; Shin & Lange,
2012).
These failure modes, debonding and cracking, occur as a result
of the shear and/or tensile stresses induced by the differential
deformation between the overlay and substrate. Stresses develop
within the overlay as the contracting movement (shrinkage) of the
newly cast overlay is to some extent restrained by the substrate
and/or the periphery in the case of an enclosed overlay. If the
resulting shear and tensile interface stresses induced are
sufficiently large, they will cause debonding at the interface
between the overlay and substrate whilst if the resulting tensile
stresses induced within the overlay are sufficiently large, they
will cause cracking of the overlay. In some cases, edge lifting
(also known as curling) may occur due to development of a stress
field within the overlay near the free edges which tends to lift
the overlay edges vertically (Crlsward, 2006). These failure modes
are illustrated in Figure 2.2.
Figure 2.2: Failure modes of bonded overlays (adapted from
Crlsward, 2006).
The differential deformations which induce the stresses that
cause failure of bonded overlays due to debonding or cracking
largely originate from plastic, autogenous, and drying shrinkage,
of the overlay, as well temperature variations, caused by internal
hydration and external environmental conditions (Shin & Lange
2012). These deformations typically occur within the first few
hours to weeks of casting of a new overlay (Beushausen &
Alexander, 2006b, Shin & Lange, 2012) and thus failure of
overlays generally occurs during this time period. These
deformations may also occur due to substrate settlement or
externally applied loads (Denari & Silfwerbrand, 2004). It is
however generally accepted that differential deformation due to
shrinkage, termed differential shrinkage, is
Overlay
Concrete Substrate
Bond between overlay and substrate
Edge Lifting (Curling)
Debonding
Edge Lifting (Curling) Humidity gradient Cracks
Chapter 2: Literature Review
-
13 the primary cause of failure of bonded overlays (Beushausen,
2005; Beushausen & Alexander, 2006b; Bissonnette et al., 2011;
Shin & Lange 2012).
Shrinkage of the overlay occurs as moisture is lost to the
environment or to the substrate through absorption if the substrate
is dry relative to the overlay, or a combination thereof. This is
termed drying shrinkage and results in the occurrence of a hygral
(humidity) gradient in the overlay due to differential drying
conditions (see Figure 2.2). The gradient induces non-uniform
stresses within the thickness of the overlay that encourage edge
lifting (curling). It should be noted that shrinkage may also occur
as a result of moisture lost during cement hydration and/or other
shrinkage mechanisms (see Section 2.4.2), however drying shrinkage
is generally considered to be the most critical type of shrinkage
with regards to bonded overlays (Pigeon & Bissonnette,
1999).
2.3.2 Debonding Debonding is the separation of the overlay from
the substrate to which it was bonded, and is characterised by
delamination and/or spalling (Beushausen & Alexander, 2006b).
It is directly related to the strength of the bond between the
overlay and the substrate, and can therefore be easily minimised,
relative to that of cracking, by ensuring appropriate bond strength
between the overlay and the substrate.
2.3.2.1 Debonding Mechanism Debonding arises due to the shear
and tensile stresses that are induced at the overlay-substrate
interface as a result of the differential shrinkage between the
overlay and substrate, as discussed in Section 2.3.1. It my however
also occur due to the breakdown of the overlay-substrate bond
caused by the penetration of aggressive chemicals through failure
cracks of the overlay. The induced interface stresses occur in a
mixed form of shear and tension and are commonly believed to have
maximum values at the boundaries of the overlay due to the lack of
continuity of stress transfer within the overlay. As such,
debonding is usually initiated at the overlay boundaries such as
edges, joints, and cracks (full-depth separations) when the
interface stresses exceed the bond strength (Carter et al., 2002;
Granju et al., 2004). It has however been argued that the main
reason why debonding commonly occurs at the overlay boundaries is
as a result of overlay slip due to the lack of adjacent restraint
(Beushausen, 2005). The initiation of debonding at overlay
boundaries suggests that debonding is closely related to cracking
(Carter et al., 2002; Beushausen et al., 2006b). It should however
be noted that debonding may also occur in areas within the centre
of the overlay however this is usually as a result of poor surface
preparation, improper overlay placement, or weak substrate concrete
(Knutson, 1990; Carter et al., 2002).
2.3.2.2 Factors Influencing Debonding The bond strength is one
of the main parameters governing debonding (Bissonnette et al.,
2011) as debonding occurs when the interface stresses exceed the
bond strength, as discussed in Section 2.3.2.1. The bond strength,
and therefore the likelihood of debonding, is dependent on a large
number of factors. However, it is commonly agreed upon that
substrate preparation and cleaning (absence of micro-cracks,
removal of the laitance layer, and cleanliness of the substrate
surface,), overlay compaction, and curing are the most important
factors to achieve good mechanical adhesion and bond (Beushausen
& Alexander, 2009; Bissonnette et al., 2011). These factors,
and
Chapter 2: Literature Review
-
14 therefore good bond strength, are clearly all directly
related to the quality of workmanship/construction. Factors of
lessor importance include the overlay mechanical properties (fresh
and hardened properties) and the moisture condition of the
substrate (Beushausen & Alexander, 2009). The environment,
substrate mechanical properties, interface roughness, overlay
placement procedures, and bonding agents are of significantly less
importance in terms of bond strength (Beushausen & Alexander,
2009). The majority of aforementioned factors were discussed in
further detail in Section 2.2.3.
2.3.3 Restrained Shrinkage Cracking Cracking of overlays as a
result of the tensile stresses induced by restrained shrinkage
deformations is termed restrained shrinkage cracking. Restrained
shrinkage cracking is related to the interaction between the
time-dependent overlay material properties including shrinkage,
tensile strength, elastic modulus, and tensile relaxation. It has
therefore proven to be difficult to control, relative to that of
debonding, and commonly occurs in practice (Beushausen &
Alexander, 2006a).
2.3.3.1 Restrained Shrinkage Cracking Mechanism Restrained
shrinkage cracking arises due to the tensile stresses induced
within the overlay as a result of the differential shrinkage
between the overlay and substrate, as discussed in Section 2.3.1.
These tensile stresses () develop when the free shrinkage of the
overlay () is restrained to a certain degree () by the substrate.
The tensile stresses induced by the restrained shrinkage of the
overlay are proportional to the stiffness (elastic modulus) () of
the overlay, with stiffer (higher elastic modulus) overlays
inducing greater tensile stresses. A portion of the induced tensile
stresses are however alleviated by tensile relaxation (), a
stress-relieving viscoelastic property of the overlay. The
magnitude of the tensile stresses induced by restrained shrinkage
is therefore:
= (2.1)
where is the restrained shrinkage stress within the overlay, is
the degree of restraint, is the free shrinkage strain of the
overlay, is the tensile modulus of elasticity of the overlay, and
is the tensile relaxation factor (see Section 3.6.1) of the
overlay. It should be noted that this is a simplified
representation of the induced tensile stress as only shrinkage is
considered and the shrinkage considered is assumed to be constant
throughout the overlay thickness, as discussed later in this
section. If the resulting tensile stresses after tensile relaxation
are less than or equal to the tensile strength of the overlay then
the overlay will remain crack-free. However, should the resulting
tensile stresses after tensile relaxation be greater than the
tensile strength of the overlay then the overlay will crack. The
failure criterion for restrained shrinkage cracking is
therefore:
= if if >
(2.2)
It should be noted that, in general, the shrinkage-induced
stresses within common overlay repair materials are often much
greater that the tensile strength of the overlay and so cracking
commonly occurs (Pigeon et al., 1999). The interaction of the
time-dependent material properties and the failure criterion for
cracking are illustrated in Figure 2.3.
Chapter 2: Literature Review
-
15
Figure 2.3: Schematic overview of the time-dependent material
properties governing restrained shrinkage cracking of bonded
overlays (adapted from Troxell et al., 1968).
It should be noted that drying and hence moisture loss
predominantly occurs from the surface of the overlay exposed to the
environment, as discussed in Section 2.3.1 and shown in Figure
2.4a. This causes a hygral gradient to develop throughout the
thickness of the overlay which results in non-uniform drying
shrinkage throughout the thickness of the overlay with the greatest
free shrinkage strains at the exposed surface of the overlay, as
illustrated in Figure 2.4b. The corresponding restrained shrinkage
stresses are therefore also non-uniform throughout the thickness of
the overlay with the maximum stresses occurring at the exposed
surface, as illustrated in Figure 2.4c. Cracking therefore develops
from the surface of the overlay down to the depth at which the
restrained stresses are equal to the tensile strength of the
overlay, as illustrated in Figure 2.4d.
Figure 2.4: Illustration of a simplified representation of (a)
drying conditions, (b) free shrinkage strains, (c) restrained
stresses, and (d) cracking due to restrained shrinkage
of bonded overlays (adapted from Pigeon & Bissonnette,
1999).
2.3.3.2 Factors Influencing Restrained Shrinkage Cracking
Restrained shrinkage cracking occurs when the restrained shrinkage
stresses, which arise due to the shrinkage, elastic modulus, and
tensile relaxation of the overlay repair material, exceed the
tensile
Casting
Stre
ss o
r St
reng
th
Age Age at loading
Age at which cracking develops
( > )
Curing ()
Tensile strength
Cracking starts to develop ( > )
()
Elastic tensile stress after relaxation
(, )
Tensile relaxation
Elastic tensile stress
Overlay
Substrate
Extension Contraction Compression
()
Tensile Strength
Drying Free Shrinkage Strain
Restrained Shrinkage Stress
Restrained Shrinkage Cracking
Interface
Tension
()
Free Shrinkage Strain
()
Restrained Shrinkage
Stress
> Cracked
Uncracked
(a) (b) (c) (d)
Chapter 2: Literature Review
-
16 strength of the overlay repair material, as discussed in
Section 2.3.3.1. Restrained shrinkage cracking is therefore
governed by the shrinkage, tensile strength, elastic modulus, and
tensile relaxation of the overlay repair material, all of which are
time-dependent material properties. As a result, factors that
influence these material properties are likely to have an influence
on restrained shrinkage cracking. Factors that influence restrained
shrinkage cracking therefore include the w/b ratio, water and
cement (paste) content, cement composition and fineness, cement
extenders, aggregate content and properties, admixtures, fibres,
member geometry, environmental conditions, and curing. The
time-dependent material properties influencing restrained shrinkage
cracking and the influence of the aforementioned factors on
restrained shrinkage cracking is discussed in further detail in
Section 2.4.
2.4 Restrained Shrinkage Cracking of Bonded Overlays Restrained
shrinkage cracking of bonded overlays is discussed herein. A brief
overview of restrained shrinkage cracking is provided in Section
2.4.1. The material properties that govern restrained shrinkage
cracking, namely shrinkage, tensile strength, elastic modulus, and
tensile relaxation, are then discussed briefly in Sections 2.4.2 to
2.4.5 respectively, after which the factors influencing restrained
drying shrinkage cracking are discussed in the subsections of
Section 2.4.6.
2.4.1 Overview Bonded overlays are subjected to differential
deformations that can cause cracking of the overlay, as discussed
in detail in Section 2.3. The stresses develop within the overlay
as a result of the contracting movement (shrinkage) of the newly
cast overlay which is to some extent restrained by the substrate
and/or periphery in the case of an enclosed overlay. The induced
stresses are therefore due to the inherent shrinkage of the overlay
and are proportional to the elastic modulus of the overlay
material. A portion of the induced tensile stresses are however
alleviated by tensile relaxation, a stress-relieving viscoelastic
property of the overlay. If the resultant tensile stresses induced
within the overlay are greater than the tensile strength of the
overlay then cracking of overlay, termed restrained shrinkage
cracking, will start to develop.
As previously noted, restrained shrinkage cracking occurs due to
shrinkage which induces tensile stresses in the overlay that are
proportional to the elastic modulus overlay however they are the
somewhat reduced by tensile relaxation of the overlay. Shrinkage of
concrete is therefore the driving force behind the cracking. There
are a variety of different types of shrinkage (see Section 2.4.2),
and as a result, restrained shrinkage cracking may be attributed to
different types of shrinkage. Restrained shrinkage cracking
commonly occurs due to either plastic shrinkage which results in
restrained plastic shrinkage cracking, or drying shrinkage which
results in restrained drying shrinkage cracking. These two types of
restrained shrinkage cracking are briefly discussed respectively in
the paragraph that follows.
Restrained plastic shrinkage cracking occurs within a few hours
of placing concrete when the concrete is still in its plastic
state. It is largely as a result of plastic shrinkage which occurs
when rapid evaporation of bleed water which leads to drying of the
concrete surface, causing menisci to form between solid particles
and capillary tension forces to be set up resulting in
contraction
Chapter 2: Literature Review
-
17 (shrinkage) of the cement paste (see Section 2.4.2.1).
Restrained drying shrinkage cracking may start to occur after
approximately 24 hours of the placement of the concrete when the
concrete is in its hardened state. It is largely as a result of
drying shrinkage which occurs when water is lost to the environment
resulting in capillary tension, disjoining pressure, surface
tension and movement of interlayer water, all of which lead to the
contraction (shrinkage) of the cement paste (see Section 2.4.2.4).
Attention will largely be paid to restrained drying shrinkage
cracking in the sections that follow.
2.4.2 Shrinkage The inherent shrinkage of concrete is the key
deformation that leads to failure of bonded overlays (cracking or
debonding) (see Section 2.3) (Pigeon & Bissonnette, 1999;
Beushausen & Alexander, 2006). This occurs when the shrinkage
of bonded overlays is restrained and stresses that are sufficiently
large to cause cracking develop within the overlay, as discussed in
Section 2.3.3.1. The shrinkage of a repair material is therefore an
important material property with regards to restrained shrinkage
cracking as it directly influences the magnitude of the stresses
induced within the overlay.
Shrinkage is a non-load induced, time-dependent decrease in the
volume of concrete, during its fresh and hardened state, due to the
movement of moisture within or out of the concrete. The total
volume decrease of concrete as a result of moisture movement or
loss is known as the total shrinkage. It comprises of five broad
types of shrinkage, namely plastic, chemical, autogenous, drying,
and carbonation shrinkage. The two main contributors to the total
shrinkage of bonded concrete overlays, however, are plastic and
drying shrinkage which are both due to moisture diffusion
(Alexander & Beushausen, 2009). Furthermore, the focus of this
research is related to restrained shrinkage cracking due to drying
shrinkage. For these reasons, drying shrinkage is discussed in more
detail.
The various types of shrinkage (plastic, chemical, autogenous,
drying, and carbonation shrinkage) are discussed in Sections
2.4.2.1 through 2.4.2.5 and are presented in the order in which
they typically occur in hardened concrete. For each type of
shrinkage, the associated mechanisms and the influencing factors
are noted.
2.4.2.1 Plastic Shrinkage Plastic shrinkage, also known as
early-age or capillary shrinkage, is the decrease in volume of
concrete due to the rapid removal of water from concrete during
early ages when the concrete is still in its plastic state. It
therefore typically occurs within the first few hours of placement
but may occur up to around 24 hours after placement of the concrete
(Banthia & Gupta, 2009; Theiner & Hofstetter, 2012).
Plastic shrinkage is often accompanied by surface cracking, known
as plastic shrinkage cracking.
In fresh concrete, the solid particles (particles and
aggregates) are covered by a layer of bleed water and the spaces
between the solid particles form a system of interconnected pores
that are completely filled with water (Mindess et al., 2003;
Schmidt & Slowik, 2013). The layer of bleed water may be
removed by evaporation, absorbed by dry substrates and/or lost to
absorbent
Chapter 2: Literature Review
-
18 formwork. If it is removed faster than it can be replaced by
bleeding water (i.e. when the rate of evaporation is greater than
the rate of bleeding), it will begin to decrease in thickness until
it eventually diminishes. When the solid particles are no longer
covered by the layer of bleed water, menisci are formed between
solid particles at the surface of the concrete. The curvature of
the water surface as a result of the menisci formation causes
negative capillary pressures to develop. The pressure consolidates
the solid particles and forces pore water to the surface of the
concrete, resulting in a volume reduction of the fresh concrete
known as plastic shrinkage (Mindess et al., 2003; Schmidt &
Slowik, 2013).
The rate of water evaporation of the bleed water, and hence the
magnitude of plastic shrinkage, is strongly dependent on the
environmental conditions, increasing with higher concrete
temperatures, higher ambient temperatures, lower relative
humidities, and higher wind velocities (see Section 2.5.2). It
should however be noted that there is still a risk of plastic
shrinkage in cool weather if the wind velocities are high
(Kellerman & Crosswell, 2009). To minimise the likelihood and
magnitude of plastic shrinkage, it is generally agreed upon that
evaporation rates should not exceed 1.0 kg/m2/h (Neville, 2004). It
should however be noted that the relationship between the rate of
water loss and the magnitude of plastic shrinkage is not
straightforward as the magnitude of plastic shrinkage is not only
dependent on the rate of water loss but also the rigidity of the
concrete mix in its plastic state (Neville, 2004). Set retarding
admixtures therefore have the effect of allowing more bleeding by
delaying the setting time however result in increased plastic
shrinkage as the concrete is in a plastic state for longer
(Neville, 2004). In general, plastic shrinkage is increased for
concrete mixes with higher water or cement contents due to the
increased paste content in which the shrinkage occurs (Neville,
2004). Additionally, concrete mixes that have lower w/b ratios
and/or contain fine fillers would be expected to bleed less and
experience a greater magnitude of plastic shrinkage (Neville,
2004).
2.4.2.2 Chemical Shrinkage Chemical shrinkage, also known as
hydration or endogenous shrinkage, is the absolute volume reduction
of concrete due to cement hydration. It occurs as a result of the
reactions between cement and water, known as cement hydration,
which results in a volume reduction. The basic reactions of
Portland cement clinker involve four reactions of tricalcium
silicate (C3S), dicalcium silicate (C2S), tricalcium alumina (C3A),
and tetracalcium alumina ferrite (C4AF). Each of these reactions
requires water (H2O), is exothermic and is associated with a
decrease in volume (Holt, 2