Citation
Behavior of concrete with internal curing agents

Material Information

Title:
Behavior of concrete with internal curing agents
Creator:
Wang, Jun ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (158 pages). : ;

Thesis/Dissertation Information

Degree:
Master's ( Master of Science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Civil Engineering, CU Denver
Degree Disciplines:
Civil engineering

Subjects

Subjects / Keywords:
Polymer-impregnated concrete ( lcsh )
Fiber-reinforced concrete ( lcsh )
Fiber cement ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
Glass fiber reinforced polymer (GFRP) composite bars are a promising material that can replace conventional steel reinforcement. GFRP is a non-metallic reinforcing material and thus corrosion will not occur during its service life. Given the tensile strength of GFRP bars are significantly higher than the tensile yield strength of steel bars, GFRP may be more suitable for high-performance concrete (HPC) application rather than normal concrete. High-performance concrete inhibits moisture ingress into the core of the concrete due to its low permeability associated with a low water-cement ratio. Self-desiccation of HPC thus takes place and causes autogenous shrinkage. Autogenous shrinkage results in significant cracking of a structural member made of HPC and may lead to premature failure. To avoid this problem, a certain amount of humidity inside the concrete needs to be maintained while the hydration process of the cement is active. Three emerging curing agents are used for the present research to overcome this critical issue respectively. They are light weight aggregate (Hydrocure was the light weight aggregate used in this research), superabsorbent polymer (SAP) and crushed returned concrete aggregate (CCA). This thesis presents the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. Push-out bond tests are conducted to achieve this research objective. Emphasis is given to interfacial capacity, failure mode, and fracture energy. Some other properties of internal curing concrete like coefficient of friction, compressive strength, drying shrinkage, elastic modulus, and rapid chloride ion penetrability are also included in this thesis.
Thesis:
Thesis (M.S.)--University of Colorado Denver. Civil engineering
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Jun Wang.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
904804992 ( OCLC )
ocn904804992

Downloads

This item has the following downloads:


Full Text
BEHAVIOR OF CONCRETE WITH INTERNAL CURING AGENTS
By
JUNWANG
B.S., Northeast Forestry University, 2014
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
Of the requirements for the degree of
Master of Science
Civil Engineering
2014


This thesis for the Master of Science degree by Jun Wang
Has been approved for the Civil Engineering program
By
NY Chang, Chair Yail jimmy Kim Chengyu Li
November 20, 2014
n


JUNWANG
ALL RIGHTS RESERVED


Jun Wang (M.S., Civil Engineering)
Experimental of Internal Curing Concrete
Thesis directed by Associate Professor Yail Jimmy Kim
ABSTRACT
Glass fiber reinforced polymer (GFRP) composite bars are a promising material that can replace conventional steel reinforcement. GFRP is a non-metallic reinforcing material and thus corrosion will not occur during its service life. Given the tensile strength of GFRP bars are significantly higher than the tensile yield strength of steel bars, GFRP may be more suitable for high-performance concrete (HPC) application rather than normal concrete. High-performance concrete inhibits moisture ingress into the core of the concrete due to its low permeability associated with a low water-cement ratio. Self-desiccation of HPC thus takes place and causes autogenous shrinkage. Autogenous shrinkage results in significant cracking of a structural member made of HPC and may lead to premature failure. To avoid this problem, a certain amount of humidity inside the concrete needs to be maintained while the hydration process of the cement is active. Three emerging curing agents are used for the present research to overcome this critical issue respectively. They are light weight aggregate (Hydrocure was the light weight aggregate used in this research), superabsorbent polymer (SAP) and crushed returned concrete aggregate (CCA). This thesis presents the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. Push-out bond tests are conducted to achieve this research objective. Emphasis is given to interfacial capacity, failure
IV


mode, and fracture energy. Some other properties of internal curing concrete like
coefficient of friction, compressive strength, drying shrinkage, elastic modulus, and rapid chloride ion penetrability are also included in this thesis.
The form and content of this abstract are approved. I recommend its publication.
Approved: Yail Jimmy Kim


ACKNOWLEDGEMENTS
I really appreciate the help from my supervisor, Dr. Jimmy Kim, for his invaluable instruction, exhilarated encouragement, and excellent guidance throughout the course of my Masters work. I would also like to thank the technical staff of the Laboratory at University of Colorado Denver, including Tom Thuis, Jack. Thanks are also extended to my fellow graduate students Thushara. siriwardanage for his assistance with casting my specimens. I would like to thank John Martinez for his help with my friction test.
Full-scale tests conducted in this study were made possible with the materials support of Northeast Solite and Barco Fence.
VI


TABLE OF CONTENTS
Chapter
1 Introduction..................................................................1
1.1 Introduction.................................................................1
1.2 Research significance.......................................................3
1.3 Objectives..................................................................4
1.4 Scope.......................................................................5
1.5 Thesis outline..............................................................6
2 Literature review.............................................................7
2.1 Introduction................................................................7
2.2 Different types of internal curing methods..................................9
2.2.1 Substances with chemically bound water..................................9
2.2.2 Substances with physically adsorbed water..............................10
2.2.3 Substances with physically held water..................................10
2.2.4 Substances with unbound water..........................................11
2.3 LWA........................................................................11
2.3.1 The replacement ratio of LWA...........................................12
2.3.2 Water / cement ratio for strength and shrinkage of internal curing concrete.. 12
2.4 CCA........................................................................13
2.4.1 Advantages of CCA......................................................13
2.4.2 The characterization of CCA in different psi...........................14
2.5 SAP........................................................................14
2.5.1 Function and advantage of SAP..........................................15
2.5.2 The characterization of SAP when the adding ratio id different.........15
2.6GFRP.........................................................................16
2.6.1 Advantages of GFRP.....................................................16
2.6.2 Properties of the GFRP used in the research............................17
3 Experimental program.........................................................22
3.1 Introduction...............................................................22
vii


3.2 Materials used for test specimens........................................22
3.2.1 Hydrocure.............................................................23
3.2.2 CCA...................................................................24
3.2.3 SAP...................................................................24
3.2.4 GFRP bars.............................................................25
3.3 Description of test specimens............................................25
3.3.1 Cylinders casted with GFRP bars specimens............................25
3.3.2 Cylinders casted without GFRP bars specimens.........................26
3.3.3 Beam casted for friction tests and frequency tests...................27
3.3.4 Beams casted for shrinkage...........................................27
3.3.5 Cylinders for chloride permeability tests............................28
3.4 Experimental setup and loading...........................................28
3.4.1 Experimental setup for cylinders with GFRP bars.......................28
3.4.2 Experimental setup for concrete compressive testing...................29
3.4.3 Experimental setup for friction testing...............................30
3.4.4 Experimental setup for dynamic elastic modulus testing................30
3.4.5 Experimental setup for shrinkage testing..............................31
4 Experimental results and discussion........................................36
4.1 Introduction..............................................................36
4.2 Compressive stress test results...........................................37
4.2.1 Compressive stress test results for Hydrocure.........................37
4.2.2 Compressive stress test results for CCA...............................38
4.2.3 Compressive stress test results for SAP...............................38
4.3 Chloride ion penetrability tests..........................................39
4.3.1 Chloride ion penetrability tests results for Hydrocure...............39
4.4 Elastic modulus tests.....................................................39
4.4.1 Elastic modulus tests results for Hydrocure...........................39
4.4.2 Elastic modulus tests results for CCA.................................40
4.4.3 Elastic modulus tests results for SAP.................................40
4.5 Friction test results.....................................................41
4.5.1 Friction test results for Hydrocure...................................41
4.5.2 Friction test results for CCA.........................................42
4.5.3 Friction test results for SAP.........................................42
4.6 Push-out test results.....................................................43
4.6.1 Push-out tests results for Hydrocure..................................43
4.6.2 Push-out tests results for CCA.......................................44
4.6.3 Push-out tests results for SAP........................................45
viii


4.7 The fracture energy analysis.............................................46
4.7.1 The fracture energy analysis for Hydrocure...........................46
4.7.2 The fracture energy analysis for CCA.................................47
4.7.3 The fracture energy analysis for SAP.................................48
4.8 Absorptance test.........................................................50
4.8.1 Absorptance test results for Hydrocure...............................50
4.8.2 Absorptance test results for CCA.....................................50
4.8.3 Absorptance test results for SAP.....................................51
4.9 Dry shrinkage test.......................................................51
4.9.1 Dry shrinkage test results for Hydrocure.............................51
4.9.2 Dry shrinkage test results for CCA...................................52
4.8.3 Dry shrinkage test results for SAP...................................52
5 Numerical modelling.......................................................109
5.1 Introduction.............................................................109
5.2 Numerical model for interfacial behavior of internal curing concrete and GFRP 110
5.2.1 Model procedure......................................................Ill
5.2.2 First point (SI, Tmax)..............................................Ill
5.2. 3 Coefficient a.......................................................112
5.2.4 Second point (S2, Tf)...............................................113
5.2.5 Three models of the three types of internal curing agents...........113
5.3 Regression for compressive strength and maximum stress..................117
5.3.1 Regression for Hydrocure.............................................117
5.3.2 Regression for CCA..................................................118
5.3.3 Regression for SAP..................................................118
6 Summary & conclusions.....................................................136
6.1 Summary..................................................................136
6.2 Conclusions.............................................................137
6.3 Recommendations for future work.........................................139
References...................................................................140
IX


LIST OF TABLES
Table
2.1 Properities of Hydrocure................................................18
2.2 Graduation analysis of CCA.............................................19
2.3 Properties of SAP......................................................20
4.1 Compressive strength (Hydrocure).......................................53
4.2 Compressive strength (CCA).............................................54
4.3 Compressive strength (SAP).............................................55
4.4 Displacement at maximum load (Hydrocure)...............................56
4.5 Maximum interfacial load of bonding strength (Hydrocure)...............57
4.6 Maximum stress (Hydrocure).............................................58
4.7 Displacement at maximum load (CCA).....................................59
4.8 Maximum interfacial load of bonding strength (CCA).....................60
4.9 Maximum stress (CCA)...................................................61
4.10 Displacement at maximum load (SAP)....................................62
4.11 Maximum interfacial load of bonding strength (SAP).....................63
4.12 Maximum stress (SAP)..................................................64
4.13 Energy dissipation based on load-displacement ( Hydrocure)............65
4.14 Fracture energy based on shear- displacement (Hydrocure)..............66
4.15 Energy dissipation based on load-displacement (CCA)...................67
X


4.16 Fracture energy based on shear- displacement (CCA)...................68
4.17 Energy dissipation based on load-displacement (SAP)..................69
4.18 Fracture energy based on shear- displacement (SAP)...................70
5.1 The average value of replacement; friction coefficient; elastic modulus; stress and
compressive from the Monte Carlo method generated results for
Hydrocure..................................................................121
5.2 The average value of replacement; friction coefficient; elastic modulus; stress and
compressive from the Monte Carlo method generated results for
Hydrocure..................................................................122
5.3 The average value of replacement; friction coefficient; elastic modulus; stress and
compressive from the Monte Carlo method generated results for
Hydrocure..................................................................123
5.4 Test results for regression analysis..................................123
5.5 Regression analysis of fc'(Mpa) for Hydrocure.........................124
5.6 Regression analysis of maximum stress for Hydrocure...................125
5.7 Regression analysis of fc'(Mpa) for CCA...............................126
5.8 Regression analysis of maximum stress for CCA.........................127
5.9 Regression analysis of fc'(Mpa) for SAP...............................128
5.10 Regression analysis of maximum stress for SAP........................129
XI


LIST OF FIGURES
Figure
2.1 Different kinds of internal curing materials: (a) Hydrocure; (b) Crushed returned
concrete aggregate (CCA); (c) Superabsorbent polymer (SAP)................21
3.1 (a) GFRP bar; (b) Cylinder cast for a concrete compressive test; (c) Cylinders cast
with GFRP bars for push-out tests.........................................32
3.2 Specimen detail for Push-out test: (a) Test setup; (b) Schematic diagram; (c)
Plastic tube for unbonding GFRP bar.......................................33
3.3 Material characterization with various internal curing agents: (a) concrete strength (ASTM C873-10a; (b) chloride permeability (ASTM 014); (c) shrinkage (ASTM C490/C490M-11); (d) coefficient of friction (Luigi M Gratton, 2006); (e) dynamic
elastic modulus (ASTM C215).....................................................35
4.1 Variation of concrete strength with internal curing agents: (a) Hydrocure and CCA;
(b) SAP.........................................................................71
4.2 Chloride penetration for concrete with Hydrocure...........................72
4.3 Frequency and dynamic modulus of concrete with internal curing agents: (a)
Hydrocure; (b) CCA; (c) SAP.....................................................73
4.4 Dynamic modulus (a) comparison of Hydrocure and CCA; (b) SAP..............74
4.5 Acceleration between concrete with Hydrocure and GFRP bar measured from the
friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary average line...........................................76
4.6 Acceleration between concrete with CCA and GFRP bar measured from the
friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e)summary average line............................................78
4.7Acceleration between concrete with SAP and GFRP bar measured from the friction test: (a) 0% adding; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding; (e)summary
average line....................................................................80
4.8 Coefficient of friction: (a) Hydrocure and CCA; (b) SAP....................81
Xll


4.9 Comparison between MTS and linear potentiometer: (a) 0% replacement; (b) 25%
replacement; (c)50% replacement; (d) 75% replacement;...........................82
4.10 Load versus displacement response of push-out tests with various concrete
mixtures with Hydrocure (a) 0% replacement; (b) 25% replacement; (c)50% replacement; (d) 75% replacement; (e) summary...................................83
4.11 Push-out test failure mode: (a) cracking appeared; (b) cracking moved upward, (c)
failure mode 1; (d) failure mode 2..............................................84
4.12 Load versus displacement response of push-out tests with various concrete mixtures with CCA (a) 0% replacement; (b) 25% replacement; (c)50% replacement;
(d) 75% replacement; (e) summary..............................................85
4.14 Load versus displacement response comparison with different internal curing agents........................................................................87
4.15 The Energy dissipation (area A) of push-out specimens with Hydrocure: (a) 0%
replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary...........................................................................90
4.16 The Energy dissipation(area A) of push-out specimens with CCA: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e)
summary.......................................................................91
4.17 The Energy dissipation(area A) of push-out specimens with SAP: (a) 0% adding ;
(b) 25% adding ; (c) 50% adding ; (d) 75% adding ; (e) summary................93
4.18 The Energy dissipation comparison of different internal curing agents....94
4.19 The Fracture energy of shear with Hydrocure: (a) 0% replacement; (b) 25%
replacement; (c) 50% replacement; (d) 75% replacement; (e) summary............96
4.20 The Fracture energy of shear with CCA: (a) 0% replacement; (b) 25%
replacement; (c) 50% replacement; (d) 75% replacement; (e) summary............98
4.21 The Fracture energy of shear with SAP: (a) 0% adding; (b) 25% adding; (c) 50%
adding; (d) 75% adding; (e) summary...........................................100
4.22 The Fracture energy comparison of different internal curing agents......101
4.23 Response comparison for concrete mixed with Hydrocure: (a) energy dissipation
xiii


-displacement; (b) fracture- displacement
102
4.24 Response comparison for concrete mixed with CCA: (a) energy dissipation
-displacement; (b) fracture- displacement.....................................103
4.25 Response comparison for concrete mixed with SAP: (a) energy dissipation
-displacement; (b) fracture- displacement......................................104
4.26 Absorptance of different internal curing agents: (a) Hydrocure and CCA; (b)
SAP............................................................................105
4.27 Dry shrinkage test results for Hydrocure: (a) shrinkage; (b) shrinkage divided by
depth..............................................................107
4.28 Dry shrinkage test results for CCA: (a) shrinkage; (b) shrinkage divided by
depth......................................................................108
4.29 Dry shrinkage test results for SAP: (a) shrinkage; (b) shrinkage divided by
depth......................................................................109
5.1 Model results comparison with tests results with Hydrocure: (a) 0% replacement;
(b) 25% replacement; (c) 50% replacement; (d) 75% replacement.................131
5.2 Model results comparison with tests results with CCA: (a) 0% replacement; (b) 25%
replacement; (c) 50% replacement; (d) 75% replacement..........................133
5.3 Model results comparison with tests results with SAP: (a) 0% adding; (b) 0.2%
adding; (c) 0.4% adding; (d) 0.6% adding......................................135
XIV


1 Introduction
1.1 Introduction
High performance concrete (HPC) has many advantages such as high strength, durability, light weight of structural members, and favorable long-term maintenance costs. Because of these superior characteristics of HPC, modern highway bridges and other types of structural members adopt high performance concrete frequently. However, self-desiccation of high-performance concrete often takes place due to inadequate curing. Then HPC demonstrates significant cracking and may lead to premature failure. Therefore, the inadequate curing problem needed to be solved. A curing method, internal curing, was introduced. By American Concrete Institute (ACI), the concept of internal curing is supply water throughout a freshly placed cementitious mixture using reservoirs, via pre-wetted lightweight aggregate, that readily release water as needed for hydration or to replace moisture lost through evaporation or self-desiccation. There are many internal curing agents, this research will focus on light weight aggregate (LWA), it serves as a source of water-supply inside the concrete and facilitate curing process. Other two internal curing agents SAP and CCA will also be included in this thesis to compare with the results of LWA. With internal curing high-performance concrete will have low cracking. Additionally, this kind of concrete also has many other advantages like lower permeability, early
1


age, and later age compressive strength. Cusson, Lounis and Daigle reported that
internal curing high-performance concrete bridge decks service life is longer than the life of HPC bridge decks by more than 20 years. Zhutovsky and Kovler found that the effect of internal curing was changing for high-performance concrete when the water-cement ratio is different. In this project, to avoid the influence of different water-cement ratio, the water-cement ratio of all mixes is 0.4. As many authors chose the number around 0.4 as water-cement ratio when casting internal curing concrete. Cusson and Hoogeveen focused on the effect of different amount of LWA for internal curing and it turns out that the amount of autogenous shrinkage cracking is different among the changing of the replacement of LWA. That is because LWA has small microscopic pores which enable the aggregate to absorb considerable water to let the high-performance concrete cure from the inside to outside. And different amount of LWA will absorb different amount of water which will affect the performance of HPC. So the replacement levels of LWA for internal curing plays a very important role here. The result of how internal curing concrete combines with GFRP when the replacement of LWA is changing is what the present research focuses on. Since GFRP contacts the internal curing concrete through a mechanical bond, the mechanical bond is really important in this application as it can transfer stresses and keep structural integrity between GFRP and internal curing concrete. In our research, we use Push-out bond tests to achieve the bond stresses. Thus, this paper focuses on the different bond stresses when the amount of LWA of internal curing high-performance concrete curing changes. Because of this test, we can understand the behavior of HPC
2


with internal curing when reinforced with GFRP bars. The objectives of the research
are examining the fundamental material characteristics of HPC and bond stresses between internal curing HPC and GFRP depending upon the degree of internal curing (i.e., strength). This paper deals with an experimental study concerning (1) interfacial stress capacity (2) failure mode (3) fracture energy (4) friction (5) permeability (6) shrinkage and (7) elastic modulus.
1.2 Research significance
Many researches related to internal curing and reinforced concrete were done these days since reinforced concrete is one of the main materials adopted to build structure. Besides, internal curing is a really effective method to deal with self-desiccation problem. But there are very few projects invested the relationship between internal curing concrete and GFRP bars. GFRP is a non-metallic reinforcing material and thus corrosion will not occur during its service life. Given the tensile strength of GFRP bars is significantly higher than the yield strength of steel bars, GFRP may be more suitable for high-performance concrete (HPC) application rather than normal concrete. Therefore, GFRP bars may replace reinforcement to be good partner with high-performance concrete. This urge more studies to be performed about this internal curing concrete together with GFRP bars topic like this research. Since there are many varieties of internal curing agents, the optimal one needed to be
3


picked up to cast high-performance concrete to combine with GFRP bars. The present
study compared the three common internal curing agents, Hydrocure, CCA, SAP, which can be bought from commercial market conveniently and investigated their bond strength with GFRP bars when the internal curing agent replacement amount is different.
1.3 Objectives
The principal objectives of this research program are to investigate the bond strength between GFRP bars and different internal curing agents when the internal curing agents replacement or adding amount changed and compare the results of different types of internal curing agents. Specific objectives addressed in this study include:
1. Investigate the concrete compressive strength of internal curing concrete when the replacement ratios are different.
2. Develop failure modes of the bond tests
3. Evaluate the fracture energy of bond tests.
4. Study the coefficient of friction of internal curing concrete when the replacement amount is different
5. Observe the elastic modulus of internal curing concrete when the replacement amount is different
4


6. Test the drying shrinkage of internal curing concrete casted by LWA when the replacement ratios changed.
7. Get the value of rapid chloride ion penetrability of internal curing concrete casted by LWA when the replacement ratios changed
8. Use the results of these properties of internal curing concrete to perform regressions.
1.4 Scope
The scope of the study consists of an experimental investigation and a numerical model to investigate the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. The aim of the experimental program was to study the effects of varying the amount of internal curing agents replacement or adding and the types of internal curing aggregates.
A numerical model was developed to predict the shear-replacement responses of internal curing concrete bonded with GFRP bars. The model method was CEB-FIP came from Euro-Code. The model assumed four sections to describe the shear-displacement procedure results. To complete the model, a parametric-a is needed to determine. For every kind of internal curing concrete, one value of a is determined.
5


1.5 Thesis outline
The contents of the thesis are briefly outlined below:
Chapter 2: provides a review of literature related to previous research on internal curing concrete, high-performance concrete and GFRP bars.
Chapter 3: presents a detailed description of the experimental program, consisting of design and fabrication of test specimens, instrumentation, test setup, and procedures. .Chapter 4: provides the results of the experimental investigations into the bond strength between internal curing concrete with different replacement ratios, coefficient of friction, compressive strength, drying shrinkage, elastic modulus, and rapid chloride ion penetrability are also included in this thesis.
Chapter 5: presents numerical modeling for interfacial behavior of internal curing concrete and GFRP and regression for compressive strength and maximum stress. Chapter 6: presents conclusions of the study and recommendations for future research into internal curing concrete together with GFRP bars.
References.
6


2 Literature review
2.1 Introduction
This chapter presents the historical development of internal curing methods. Previous research related to GFRP bars is discussed. The first part of this chapter discusses the different types of internal curing methods. Then, research on some certain internal curing aggregates and their advantages and disadvantages. First section investigates Hydrocure. The next section examines CCA. Finally, SAP is studied. After the investigation of internal curing agents, GFRP bars will be followed.
Traditional curing method cures concrete from outside to inside and external curing water is only able to penetrate several mm into low w/c ratio concrete. Therefore, for high performance concrete (HPC) whose permeability is low often can not be cured well by external curing. HPC was introduced to the market in the 1980s. Those days, many state highway agencies begin to implemented HPC for bridge decks to protect reinforcing steel from corrosion as the permeability of HPC is low. It seems like high strength would make concrete less likely to crack. But from the Ohio Department of Transportation, many decks cracked. Besides, Due to hydration of cementitious material self-desiccation occurs as internal humidity decreases (Bentz and Snyder, 2005). Those problems make HPC can not live up to expectations.
To solve this problem internal curing method was introduced. Actually as far back as
7


1957, Paul Klieger found that lightweight aggregate absorb considerable water during
mixing which apparently can transfer to the paste during hydration (Klieger, 1957). Besides lightweight aggregate, other materials that could function as internal curing aggregate also be investigated. Jensen& Hansen found superabsorbent polymers have the ability to absorb significant water to make the concrete cure from inside to outside (Jensen & Hansen, 2001). And the following section briefly describes internal curing methods. By using internal curing methods, many characteristics can be improve, such as shrinkage (including autogenous), cracking, early age and later age compressive strength, lower permeability, resistance to freeze-thaw, minimization of carbonation, densification of the interfacial transition zone, improved mortar strength, and reduced warping. Cusson, Lounis and Daigle reported that internal curing high-performance concrete bridge decks service life is longer than the life of HPC bridge decks by more than 20 years (Cusson, 2010) which clearly show the advantages of internal curing method.
Although internal curing agents were used for the HPC, corrosion still occur to HPC easily which makes GFRP may be a good partner for HPC as GFRP is a non-metallic reinforcing material and thus corrosion will not occur during its service life. The last section of this chapter will give some brief introduction to GFRP.
8


2.2 Different types of internal curing methods
The following section briefly describes internal curing methods based on several different physical or chemical principles: chemically bound water, physically adsorbed water, physically held water, unbound water (Ole, 2006). And the followings are some examples of this classification.
2.2.1 Substances with chemically bound water
To serve as potential materials for internal water curing the chemical substance must content a large amount of water and can give the water off easily during the curing period when the surrounding relative humidity decreased. At the same time, the chemical substance must be compatible with the cementitious system. There are many substances can serve as this like C4AH19. Usually, an amount of water of around 22-23% of the weight of the anhydrous cement is chemically bound at complete hydration of cement [Powers & Brownyard 1948], But for these materials, they can contain around 50% water by weight and also can give off the water when surrounding relative humidity drops. They look like perfect materials for internal curing, however, when the relative humidity is high only a part of the water can be released which makes this substances with chemically bound water have not been used technique.
9


2.2.2 Substances with physically adsorbed water
Different from substances with chemically bound water, there are many attempts utilize types of substances with physically adsorbed water. Such as natural bentonite clay which has the ability to absorb several molecular layers of water and SAP whose theoretical maximum water absorption is around 5,000 times their weight (Ole, 2006). And SAP is one of the main internal curing materials that studied in this research. The key of the amount of physically bound water depends on the relative humidity of the pore system. The thickness of the adsorption layer ranges from 1 monomolecular water layer at 20% relative humility to about 6 monomolecular layers at 100% relative humility (Hagymassy et al. 1969, Setzer. 1977, Badmann et al. 1981).
2.2.3 Substances with physically held water
Substances with physically held water serve as a container for internal curing water. That is because they have small microscopic pores whose sizes above around 100 nm which can storage the internal curing water and enable the aggregate to absorb enough water to let the concrete cure from inside to outside. For the aggregates own smaller poles than 100 nm may not work as internal curing material as the pores hold the water so tightly that it can not be released for the cementitious reactions. Pumice, Perlite, Liapor and Leca, Stalite, Diatomanceous earth and
10


Hydrocure all can serve as substances with physically held water as they share the
similar characteristic- their pores sizes are suitable for holding and giving of water at right time. In this project, Hydrocure is used as the internal curing aggregate to investigate the properties of LWA together with GFRP bars.
2.2.4 Substances with unbound water
The last kind of internal curing agents talked here is substances with unbound water. Two examples about it are given here, one is Microencapsulation and the other one is Emulsified water. The first capsule wall has ability to remain water and release it until setting by both chemical and physical ways although the price of it is relatively high. The second type is a kind of colloid dispersed with small particles. This emulsifying agent may work well.
2.3 LWA
The LWA used in the research is Hydrocure as shown in Figure 2.1(a), Table
2.1 shows the properties of Hydrocure investigated here.
11


2.3.1 The replacement ratio of LWA
Since 1990s, research has shown that LWA can provide internal water reservoirs. Philleo proposed that using LWA as a replacement to normal weight aggregate to reduce the effects of self-desiccation. To get the optimal replacement of LWA Dutch performed the work to replace natural weight aggregates with LWAs which show that when the partial replacement amount of natural weight aggregate ranges from 0% to 25% LWA did not have negative effect on the compressive strength. Especially, when w/c is low like 0.33, compressive strength of internal curing concrete with replacement of 10% and 17% even are higher than the concrete with no LWA replacement. Besides, with replacement percentages up to 25% the autogenous shrinkage reduced significantly. Another test used Hydrocure showed that when the water/ Cement ratio is 0.395 the compressive strength increased from the replacement is 0% to 15% and decreased from 25% to 25%, but the strength at 25% replacement ratio was not weaker than 0% replacement ratio.
2.3.2 Water / cement ratio for strength and shrinkage of internal curing concrete.
Internal curing is the process by which the hydration of cement occurs because of availability of additional internal curing water is not part of mixing water (ACI-308). For internal curing concrete water is a key to determine several properties of it, such as strength and elasticity modulus and autogenous shrinkage. Semion
12


Zhutovsky and Konstantin Kovler did test on strength and dynamic elasticity modulus,
shrinkage and total shrinkage of HPC made at w/c of 0.21-0.33 (Semion Zhutovsky and Konstantin Kovler, 2012). And the results show that the for concrete with w/c ratio of 0.33 and 0.25 internal curing makes shrinkage higher, however, internal curing let the shrinkage of concrete with w/c of 0.21become lower. For autogenous shrinkage internal curing method totally eliminated it even when w/c is 0.21.
2.4 CCA
CCA is shown in Figure 2.1 (b) and Table 2.2 shows the properties of CCA used in the research.
2.4.1 Advantages of CCA
The pore structure of CCA is similar to that of cement paste (Ryan, 2008). Both of them have the primary pore- containing component in the crushed aggregates which makes it possible for CCA to serve as internal curing agents. CCA also has a typical gravity of 2.04 (Kim & Bentz, 2008). It is assessed that every year there are about 2 % to 10 % of the 460 million cubic yards of ready-mixed concrete produced
13


in the USA is returned to the concrete plant (Obla et al. 2007) which makes it become
lower price products. Due to the economic value and the value of environmental protection of CCA, it becomes one of the popular internal curing agents that used by commercial markets.
2.4.2 The characterization of CCA in different psi
Haejin Kim and Dale Bentz explored the use of CCA in their article in 2008. This article investigated CCA in different psi from lOOOpsi to 5000psi increased by 2000psi to select the best performing CCA. And the results noted that when the strength of CCA is low the replacement of sand should make a strength reduction. It is also shown that CCA-3000 had the highest strength among the three types of CCA. That is because CCA-5000 had the larger volumetric content in the motor which offset the higher inherent strength, so that lower compressive strength occurred (Dale, 2008)
2.5 SAP
Figure 2.1 (c) shows SAP and Table 2.3 shows the properties of SAP used in the research.
14


2.5.1 Function and advantage of SAP
SAP is a kind of polymetric material which can absorb a large amount of liquid from surroundings and to retain the liquid within its structure without dissolving (Buchholz, 1998) and when the initial free water has been consumed by the hydration reactions, the water absorbed by the SAP will be gradually released. So that SAP is able to be performed as internal curing admixture of concrete. Contrary to LAW SAP permits free design of the pore sharp and the pore size distribution of the concrete. Besides, SAP can avoid the undesirable addition of significant amount of mechanically poor aggregate (Ole, 2001). Furthermore, due to the maintenance of higher internal humidity adding SAP to a concrete can make it have higher ultimate degree of hydration of cement then let the compressive strength become higher at long term (Geiker, 2004).
2.5.2 The characterization of SAP when the adding ratio id different
In 2002, O.M. Jensen and P.F Hansen did the tests of two different SAPs. The SAPs added at a rate of 0-0.6 wt. % of cement (Ole, 2002). And the results noted that the compressive strength of 45* 90 mm mortar cylinders after lday of sealed curing followed by 27 days decreased from 0% SAP to 0.6 % SAP, i.e., the strength was reduced by 19% due to the water entrainment. This article also gave a assumption of this results that is measured lower strength of the water-entrained mortar compared
15


to the reference mortar may be caused by the moisture condition. However, further
research was needed to prove this comment.
2.6 GFRP
GFRP is a fiber reinforced polymer made of plastic reinforced by glass fibers, commonly woven into a mat. It first developed in the mid 1930's and has become a staple in the building industry. In 1967, the attempted destruction of Disneyland's "House of the Future" built in 1956-7 showed the architectural advantages, the futuristic house was no longer deemed necessary and was scheduled to be destroyed in 1967. Amazingly, the wrecking ball merely bounced off the structure as it was entirely built of fiberglass, then the possibilities for GFRP were recognized and began to grow. By 1994, nearly 600 million pounds of composite materials were used in the building industry. Besides field studies also show that glass fiber-reinforced polymer offers a low life-cycle cost option for reinforcement in concrete pavements (Roger H. L, 2008).
2.6.1 Advantages of GFRP
There are many advantages of GFRP are collected such as it has a very high strength
16


to weight ratio, the weight of GFRP is only around 2 to 4 lbs. per square foot which
means faster installation, less structural framing, and lower shipping costs. The resistance of GFRP is also superb, it can resist salt water, chemicals, acid rain and most chemicals. Furthermore, it can let the domes and cupolas be resined together to form a one-piece, watertight structure and can be molded to any shape or form. Researches also showed that after 30 years, there was no loss of laminate properties and GFRP could stand up to category 5 hurricane Floyd with no damage, while nearby structures were destroyed
2.6.2 Properties of the GFRP used in the research
For Aslan 100 Fiberglass rebar use in the investigation there were many benefits. They are impervious to chloride ion and low PH chemical attack, their tensile strength are greater than steel, the weight of them is just l/4th of the steel rebar and they are electrically and thermally non-conductive( Hughes Brother, 2011).
17


Table2.1 Properities of Hydrocure (Ref: http://www.nesolite.com/physicalcharac.htm)
Density
Dry Loose (ASTM C 29) 45 pcf 720 kg/m3
Dry Rodded (ASTM C 29) 50 pcf 800 kg/m3
Saturated Surface Dry Loose (ASTM C-29) 48 pcf 768 kg/m3
Absorption
Saturated Surface Dry, 24 hour (ASTM C-127) 10%
Specific Gravity
Saturated Surface Dry (ASTM C 127) 1.5
Soundness (Loss)
Magnesium Sulfate (ASTM C-88) 1%
Sodium Sulfate (ASTM C-88) 0.50%
After 300 Cycles Freezing and Thawing (AASHTO T103) 1%
Resistance to Abrasion
Los Angeles Abrasion (AAHTO T-96B) <30%
Impurities
Clay Lumps (ASTM Cl42) None
Organic Impurities (ASTM C40) None
Electrical Resistance
Saturated Surface Dry ohm-cm
Aggregate Chemical Characteristics
Ignition Loss (ASTM Cl 14) 0
Stains (ASTM C641) None
Chlorides (NaCl) 0.50 ppm
18


Table 2.2 Graduation analysis of CCA (Ref: http://www.alliedrecycle.com/fmes.aspx)
Aggregate crushing value, % (AS 1141.21) 23.1
Bulk density, kg/m3 (AS 1141.6) 2394
Water absorption, % (AS 1141.6) 5.6
Impurity level, % (AS 1141.32) 0.6
LOI, % 4.9
Graduation analysis of CCA.
Sieve Size Allied Test
1" 100
3/4" 100
1/2" 100
3/8" 100
#4 78
#8 55
#16 36
#30 23
#50 14
#100 9
#200 6.2
19


Table 2.3 Properties of SAP (Ref: www.m2polymer.com)
Product Attributes:
Reduces Waste Disposal Costs Expands in Volume by Less Than 1%. Non-Biodegradable Polymer (40 CFR 264.314 (e)(ii))
Polycarboxylate Polymer Will Not Release Trapped Ionic Contaminants If Solute Evaporation Occurs.
Strong Ion Exchange Capability Allows For Heavy Metal Binding And For Many Solidified Wastes
To Pass TCLP.
When Used Properly, Waste Sludges Will Pass Paint Filter Test (EPA 9095).
Polymer is suitable for Incineration. No halogens.
Freeze-Thaw Tested. The Polymer Will Not Release Free Liquids.
Polymer Will Excellently Absorb Aqueous Wastes of pH > 4. For Highly Acidic Wastes, Neutralization (pH Adjustment) Is Recommended.
Approved Sorbent at Hanford (WA) & WIPP (TRUCON Codes)
Typical Absorptive Properties:
Free Swell in DI Water 400 500 X Free Swell in 1 % NaCl 45 55 X Free Swell in 2 % NaCl 35 40 X Free Swell in 10 % NaCl 19 25 X Free Swell in 1 % CaC12 20 25 x Free Swell in 8N NaOH 24 30 x Bulk Density 4.5 to 6.5 lbs/gallon Liquid Release Test (EPA 9096):
Waste Lock 770 has the ability to both absorb under pressure and to retain absorbed liquids at high
pressures:
At 25X Hydration, PASSES at 50 psi At 10X Hydration, PASSES at 75 psi Bulk Density = 4.5 to 6.5 lbs/gallon
20


(a) (b) (c)
Figure 2.1 Different kinds of internal curing materials: (a) Hydrocure; (b) Crushed returned concrete aggregate (CCA); (c) Superabsorbent polymer (SAP)
21


3 Experimental program
3.1 Introduction
The experimental program investigated the bonding strength between GFRP and internal curing concrete casted with different amount of internal curing agents and some other properties of internal curing concrete. The internal curing agents included Hydrocure, Superabsorbent polymer and crushed returned recycled concrete. The objective was to determine the optimal replacement amount of internal curing agents for different properties of internal curing concrete and compared the results when the variety of internal curing agents is different. To get those properties, for the Hydrocure, 12 cylinders with GFRP bar, 12 cylinders without bar and 8 beams were tested. Same tests were conducted for SAP and CCA. This chapter describes the internal curing materials used to fabricate the test specimens, the fabrication process, the experimental procedure, instrumentation, push-out test for cylinders with GFRP bar, compression test for cylinders without bar, friction test, frequency test, permeability test and shrinkage test.
3.2 Materials used for test specimens
The GFRP used is 13mm as shown in Figure 3.1(a). The ultimate tensile
22


stress is 689 MPa with a modulus of 40.8 GPa. Its maximum bond stress to concrete
is 11.6 MPa. The GFRP was embedded in the high-performance internal curing concrete, the basic constituents of the concrete is water (205 kg/m3), cement (445.65 kg/m3), sand (600kg/m3), and gravel (1095 kg/m3). To proceed the project, natural sand needs partly to be taken place by Hydrocure and CCA and for SAP, we need to add specific amount of SAP according to the weight of cement added.
3.2.1 Hydrocure
Hydrocure is the main internal curing agents that are used. Its bulk damp loose unit weight is 45-55 pcf (720-880 kg/m3) depending on gradation. The gradation of coarse aggregate is 5-20 mm. In Hydrocure many small microscopic pores enable the aggregate to absorb about 20% of its weight with water. Therefore, instead of curing concrete with water from the outside in, internal curing is for curing from the inside out. Before casting concrete the surface of Hydrocure should be dry. Move the Hydrocure to the batching plant and sprinkle with water during 24 hours storage to maintain 24 hour absorption of a minimum of 15 % at 24 hours and a minimum of 9 % to 10 % in 30 minutes. Then adequate saturation of the cement can be maintained during hydration which improves the strength and dimensional stability by reducing shrinkage and warping. Table 13 shows the properties of the Hydrocure used in this research.
23


3.2.2 CCA
Compared with Hydrocure, CCA costs less and has potential as an internal curing agent due to its high absorption capacity. Besides, it is low specific gravity. Those two properties are crucial factors for the internal curing. Therefore, this research also used CCA to do the same tests as Hydrocure and compare the results of the two. The CCA was prepared at Allied Recycled Aggregates. Measured particle size distribution of CCA is shown in table. Its la abrasion is 38% and its modified protector is 118.5@ 12.3.
3.2.3 SAP
SAP- A superabsorbent polymer is able to absorb a significant amount of water from the surrounding and retain the water within itself. That property will provide additional curing water for concrete from inside to outside. Therefore, it is a good resource to cast internal curing concrete. Instead of replacing the amount of sand by SAP, we added SAP according to the weight of cement. Because the effect of SAP is really strong the amount added was smaller compared with Hydrocure and CCA. The SAP used here is Waste Lock 770. It is a solid, granular superabsorbent polymer. This cross-linked polyacrylate material swells and absorbs many times its weight in aqueous solutions and its typical absorptive properties are shown in table. The bulk density of this kind of SAP is 4.5 to 6.5 lbs/gallon. Waste Lock 770 has the ability to both absorb under pressure and to retain absorbed liquids at high pressures.
24


3.2.4 GFRP bars
The GFRP bars used in this investigation were from Hughes Brother and their diameter is 13mm. The ultimate tensile stress is 689 MPa with a modulus of 40.8 GPa. Its maximum bond stress to concrete is 11.6 MPa. Furthermore, they were able to be used to reject corrosion such as exposing to De-icing salts and Marine Salts.
3.3 Description of test specimens
The following section provides details of the test specimens. 12 cylinders with GFRP bar were casted for push-out tests, 12 cylinders without bar were casted for compression test and 8 beams were casted to get the properties of drying shrinkage, elastic modulus, friction, and rapid chloride ion penetrability were got by casted cylinders.
3.3.1 Cylinders casted with GFRP bars specimens.
To study the interfacial behavior of the composite reinforcement when contacted with the internal curing concrete, the cylinders (10.16 cm x 20.32 cm) were casted with the GFRP, as shown in Figure 3.1(c). Compression test showed that 5.08
25


cm GFRP failed before 2.54 cm. Thus, here the GFRP out of the concrete is 2.54 cm
to avoid GFRP failure before rebar being pushed out. Between the concrete and GFRP there is a 15.24 cm long plastic tube which means that the effective bond length is therefore 5.08 cm and GFRP can be pushed out easily. As they are internal curing concrete, all the test specimens were cured in the batching plant without putting them in the water and they were sealed with 152 pm plastic strips followed by 89 pm plastic sheets that were secured with rubber bands. For Hydrocure and CCA, 12 specimens were casted respectively. Replacement amount of sand by internal curing agents ranged from 0% to 75% increased by 25% and for each replacement 3 cylinders with GFRP bar were tested. For SAP, there were also 12 specimen were casted, but the added amount was different from Hydrocure and SAP which ranged from 0% of cement to 0.6% of cement increased by 0.2%.
3.3.2 Cylinders casted without GFRP bars specimens
To get the compressive strength of internal curing concrete when the internal curing agents replacement changed, 12 cylinders without GFRP bars with size of 4x8 inches were casted shown as Figure 3.1(b) when the internal curing agent is LWA which means for each replacement ratio from 0% to 75% increased by 25% 3 cylinders were casted. Since 0% replacement LWA can be the control group when the internal curing agents were CCA and SAP, only 9 cylinders needed to be cast for each
26


one of them.
3.3.3 Beam casted for friction tests and frequency tests
Besides cylinders, beams whose sizes are 3x4x16 inches also needed to be casted. Same as cylinder casted without GFRP bars, 0% replacement concretes can be the control group, they just needed to be casted one time. Since each specimen was able to perform several times tests and both of the two tests can use the same beams only one beam was made for each different replacement. So, ten beams were casted here to achieve the results of coefficient of friction and frequency.
3.3.4 Beams casted for shrinkage
LWA is the main internal curing aggregate to study in the project. For shrinkage, only the material of Hydrocure was considered. Therefore, only 4 beams with size of 4x4x 111/^ inches were casted.
27


3.3.5 Cylinders for chloride permeability tests
Same as shrinkage tests only the internal agent of Hydrocure was studied. 4 cylinders with size of 4><8 inches were casted and then each of them was cut to four 4x2 inches cylinders to complete these tests.
3.4 Experimental setup and loading
The following sections explain the testing apparatus. Methods of loading are also detailed.
3.4.1 Experimental setup for cylinders with GFRP bars
The push-out test was performed using the MTS machine with maximum loading capacity of 89 KN together with linear potentiometer shown as Figure 3.2 (a). Some adjustments were made to MTS machine as it was not designed for push-out test. A circle steel plate with the size of 5x2 inches was settled on the machine base. There was a hole with the diameter of 2 inches in the center of the steel plate to let the GFRP bar be pushed out. To decide the length of the GFRP out of the internal curing concrete the GFRP also was loaded. Compression test showed that two inches GFRP failed before one inch. Thus, here the GFRP out of the concrete is one inch to avoid
28


GFRP fail too quickly. As the total bond strength between the internal curing concrete
and GFRP bars was very high which may let the one inch GFRP bar fail before it be pushed out, a 15.24 cm long plastic tube was added between the concrete and GFRP bar shown as Figure 3.2(c) which means that the effective bond length is therefore 5.08 cm and GFRP can be pushed out easily shown as Figure 3.2 (b). The effective bond length was at the bottom of the GFRP bar that is because if making the top two inches GFRP bar bond with concrete the concrete around the bar would punch the plastic tube. As the thickness of the tube is about 0.5 inch the punching force was very high which will affect the testing results significantly. The load (P) and the slip (s) at the free end of GFRP bars anchored in the test cylinders were measured in order to determine a load-slip relationship. The monotonically increased load was applied by the MTS testing machine. The load was applied with a rate of lmm/min and distributed on the GPRP surface. A laboratorial computer was used to collect test datum including load, time and displacement automatically. When the relative displacement between the GFRP bars and the concrete reached one inch which means the GFRP bar was totally pushed out the test was end.
3.4.2 Experimental setup for concrete compressive testing
Standard compression tests shown as Figure 3.3 (a) were performed in accordance with ASTM C39 (ASTM, 1996) on thirty 4x8 inches concrete cylinders.
29


All cylinders with different internal curing replacement or adding were cured 20 days.
3.4.3 Experimental setup for friction testing
Figure 3.3(d) shows the experimental setup for friction testing. The GFRP bar cut to 13.6 inches and weighted 36. lg was connected through a string to a force weighted 35.5g and then put on the surface of the internal curing beam. The string was connected to an electric motor. When the force pulled the GFRP bar slid on the surface of internal curing beam, the acceleration of the motor can be collected by the laboratorial computer. As the random uncertainties often invalidate the results, each group testing was repeated twelve times to reduce errors. According to the Newtons second law, the coefficient of friction between GFRP bars and different types of internal curing concrete can be solved easily.
3.4.4 Experimental setup for dynamic elastic modulus testing
A conditioned beam with size of 3x4x16 inches was placed on an aluminum support pad. An excitation hammer was used to generate impact signals and then be received by an accelerometer receiver as Figure 3.3(e) shown. It should be make sure that the two premarked locations of impacting and receiving signals were consistent for all test specimens. The response frequency domain waves traveled along the specimen were recorded by a dedicated data acquisition system. The
30


dynamic elastic modulus of each specimen was then determined by the following
equation:
Ed = CMn2 in which C
0.9464
77.3 bt3
where Ed is the dynamic elastic modulus in Pa; n is the frequency; M is the mass of the specimen; L, t, and b are the length, thickness, and width of the specimen, respectively; and T is the correction factor dependent on the radius of gyration to the specimen length and the Poissons ratio( ASTM C215).
3.4.5 Experimental setup for shrinkage testing
Figure 3.3(c) shows the experimental setup for friction testing. This test is based on ASTM C490/C490M-11.
3.4.6 Experimental setup for chloride penetration testing
Figure 3.3(b) shows the experimental setup for test of measuring dying shrinkage of concrete autoclave expansion of Portland cement and potential expansive reactivity of cement aggregate combinations in mortar bars during storage. This test follows ASTM Cl50.
31


(a) (b) (c)
Figure 3.1 (a) GFRP bar; (b) Cylinder cast for a concrete compressive test; (c) Cylinders cast with GFRP bars for push-out tests
32


MTS machine
Data acquisition of
Cylinder bonding with GFRP bar
(a)
(b) (c)
Figure 3.2 Specimen detail for Push-out test: (a) Test setup; (b) Schematic diagram; (c) Plastic tube for unbonding GFRP bar
33


34


Specimen
(e)
Figure 3.3 Material characterization with various internal curing agents: (a) concrete strength (ASTM C873-10a; (b) chloride permeability (ASTM C114); (c) shrinkage (ASTM C490/C490M-11); (d) coefficient of friction (Luigi M Gratton, 2006); (e) dynamic elastic modulus (ASTM C215)
35


4 Experimental results and discussion
4.1 Introduction
In this chapter, the results from the push-out tests, friction tests, compressive strength tests, drying shrinkage test, elastic modulus tests, and rapid chloride ion penetrability tests are presented. The failure mode and fracture energy of internal curing concrete casted with three different types of internal curing agents together with GFRP are discussed in detail. The interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete is evaluated in terms of load-displacement response. The coefficient of friction between different kinds of internal curing concrete with GFRP bars are established from a simple measurement of sliding friction coefficient. The mechanical properties of compressive strength are built from the standard concrete cylinder tests. Furthermore, some other properties like shrinkage, elastic modulus and chloride ion penetrability are also investigated based on the ASTM standards. The primary objective of the experimental program was to determine the optimal replacement or adding amount of the three kinds of internal curing agents based on the results of push-out test, compressive strength tests and the analysis of fracture energy.
36


4.2 Compressive stress test results
Thirty cylinders were tested in compressive stress as discussed in chapter 3. When internal curing agent was Hydrocure three cylinders were tested for each type of replacement ratio from 0% to 75%. Since 0% replacement internal curing concrete is the control specimen, when the internal curing aggregate is CCA or SAP 0% replacement amount or adding amount is not included any more. Therefore, for CCA, three cylinders were tested for each replacement amount from 25% to 75% increased by 25%. And for SAP, three cylinders were tested for the adding amount from 0.2% to 0.6% of cement weight increased by 0.2%.
4.2.1 Compressive stress test results for Hydrocure
The compression strength of internal curing concrete with different percent of Hydrocure is shown as Figure 4.1(a). And the Table 4.1 shows the datum results. The average compressive strength tended to decrease with the adding replacement amount of Hydrocure and the inclusion of Hydrocure actually degrades the concrete performance.
37


4.2.2 Compressive stress test results for CCA
The compression strength of internal curing concrete with different percent of CCA is shown as Figure 4.1(a). And the Table 4.2 shows the datum results. The average compressive strength tended to decrease with the adding replacement amount of CCA and the inclusion of CCA actually degrades the concrete performance. By comparing the results of CCA and Hydrocure in Figure 6 compressive strength of Hydrocure is lower than CCA
4.2.3 Compressive stress test results for SAP
The compression strength of internal curing concrete with different percent of SAP is shown as Figure 4.1(b). And the Table 4.3 shows the datum results. The average compressive strength tended to decrease with the adding replacement amount of SAP and the inclusion of SAP actually degrades the concrete performance. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 6 is assumed equal which means 42.321 e_0,018*r = 39.867 e_2,044*a where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 113.56*a+3.32. And that means that adding 0.1% SAP has the similar effect of replacing 14.68% Hydrocure for compressive strength. Obviously, the effect of SAP is significant stronger than Hydrocure.
38


4.3 Chloride ion penetrability tests
Only concrete casted with Hydrocure agents tested here.
4.3.1 Chloride ion penetrability tests results for Hydrocure
The test results of Hydrocure are shown in Figure 4.2. The coulombs increased with the adding of Hydrocure replacement.
4.4 Elastic modulus tests
The section presents the frequency tests results and the calculated elastic modulus results.
4.4.1 Elastic modulus tests results for Hydrocure
Figure 4.3(a) shows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for Hydrocure. And Figure 4.4(a) shows the calculated dynamic elastic results. It shows that dynamic elastic decreased with the increasing of replacement ratio for Hydrocure.
39


4.4.2 Elastic modulus tests results for CCA
Figure 4.1(a) shows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for CCA. And Figure 4.4(a) shows the calculated dynamic elastic results. It shows that dynamic elastic decreased with the increasing of replacement ratio for Hydrocure. Furthermore, the dynamic elastic of Hydrocure is lower than CCA.
4.4.3 Elastic modulus tests results for SAP
Figure 4.3(b) shows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for SAP. And Figure 4.4(b) shows the calculated dynamic elastic results. It shows that dynamic elastic decreased with the increasing of adding ratio for SAP. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 7 is assumed equal which means 39.511 e_0,005*r = 35.424 e_1,36*a where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 272*a+21.84. And that means that adding 0.1% SAP has the similar effect of replacing 49.04% Hydrocure for compressive strength. Obviously, the effect of SAP is significant stronger than Hydrocure.
40


4.5 Friction test results
Ten specimens were tested, and for each specimen twelve tests were implemented to get the results. This section presents the friction tests results of acceleration. By the equation of Newtons second law, the friction coefficient between internal curing concrete and GFRP bar was determined for each type of internal curing concrete.
4.5.1 Friction test results for Hydrocure
Acceleration-time plots for Hydrocure, along with photos of test progress, are shown in Figures 4.5. Figure 4.5 presents the friction coefficient-replacement plot when the internal curing agent was Hydrocure. As the figure shown, the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is
36.1 grams. Therefore, the equation established to calculate the friction coefficient should be calculated by the equation of a = (35.5-36. lp)*g / (36.1+35.5). And the calculated results are shown in figure 12. It is evident that adding the replacement amount of Hydrocure the acceleration of the sliding GFRP bar will decrease lightly. Furthermore, friction coefficient will increase slightly when the replacement amount of Hydrocure is increasing
41


4.5.2 Friction test results for CCA
Figure 4.6 presents the friction coefficient-replacement plot when the internal curing agent was CCA. Same as the friction tests for Hydrocure the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is 36.1 grams. Therefore, the equation established to calculate the friction coefficient should be calculated by the equation of a = (35.5-36. lp)*g / (36.1+35.5). And the calculated results are shown in Figure 4.8(a). It is evident that adding the replacement amount of CCA the acceleration of the sliding GFRP bar will decrease lightly when the replacement ratio was from 25% to 75% and the results are different from 0% replacement to 25% replacement. Furthermore, friction coefficient will increase slightly when the replacement amount of CCA is increasing. By comparing the results of CCA and Hydrocure in Figure 12 coefficient of Hydrocure is a little bit higher than CCA
4.5.3 Friction test results for SAP
Figure 4.7 presents the friction coefficient-replacement plot when the internal curing agent was SAP. Same as the friction tests for Hydrocure the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is 36.1 grams. Therefore, the equation established to calculate the friction coefficient should be calculated by the equation of a = (35.5-36. lp)*g / (36.1+35.5). And the calculated results are shown in Figure 4.8(b). It is evident that adding the replacement amount of SAP the
42


acceleration of the sliding GFRP bar will decrease lightly when the adding ratio was
from 0.2% to 0.6% and the results are opposite of 0% replacement to 0.2% adding. Furthermore, friction coefficient will increase slightly when the replacement amount of SAP is increasing.
4.6 Push-out test results
The following sections show the results of the experimental program. The results of push-out tests on thirty cylinders casted with GFRP bars are included. Interfacial capacity, failure mode, and fracture energy is discussed for each specimen, and the optimal internal curing agents replacement amount or adding amount is established from the experimental results. To make sure that the displacement results are tested by MTS machine are accurate linear potentiometer was also included in the tests, Figure 4.9 shows the comparison between the two test methods and the results of them are significant close which means the test results of MTS are exact enough to be used.
4.6.1 Push-out tests results for Hydrocure
Figure 4.10 shows the interfacial load (P) versus displacement at a
43


specific amount of light weight aggregate. The displacement when the GFRP bar
reached the maximum load is shown in Table 4.4. Table 4.5 and Table 4.6 show the maximum interfacial load of bonding strength and maximum stress respectively. It is postulated that the Hydrocure does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessary to generalize this preliminary conclusion.
The first crack appeared at the bottom of the concrete (shown in Figure 4.11(a)). It moved upward slowly as shown in Figure 4.11(b) until it reached about 5.08 cm which was close to the plastic tube. Then it started to move horizontally and merged with the other 2 cracks as it shown in Figure 4.11(c). Figure 4.11(d) shows that the angle between the 3 cracks was about 120 degrees.
4.6.2 Push-out tests results for CCA
Figure 4.12 shows the interfacial load (P) versus displacement at a specific amount of light weight aggregate. The displacement when the GFRP bar reached the maximum load is shown in Table 4.7. Table 4.8 and Table 4.9 show the maximum interfacial load and maximum stress respectively. By compare the push-out test results as Figure 4.14 shown, the maximum load of CCA push-out test is bigger than
44


the maximum load of Hydrocure. But both of the two maximum loads decreased with
the increasing internal curing agent replacement. It is also postulated that the CCA does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessary to generalize this preliminary conclusion.
For the cracking modes, cylinders casted with CCA share the similar cracking procedures with concrete casted with Hydrocure.
4.6.3 Push-out tests results for SAP
Figure 4.13 shows the interfacial load (P) versus displacement at a specific amount of light weight aggregate. The displacement when the GFRP bar reached the maximum load is shown in Table 4.10. Table 4.11 and Table 4.12 show the maximum interfacial load and maximum stress respectively. Just as the previous two types of internal curing agents the maximum push-out load of cylinders casted with GFRP bar when internal curing agent is SAP also decreased with the increasing replacement of SAP. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 4.14 are assumed equal which means 25.594 e_0,013*r = 23.545 e-i.302*a where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 100.15*a+6.42. And that means
45


that adding 0.1% SAP has the similar effect of replacing 16.435% Hydrocure. Obviously, the effect of SAP is significant stronger than Hydrocure. Similarly, it is also postulated that the SAP does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessary to generalize this preliminary conclusion.
For the cracking modes, cylinders casted with CCA share the similar cracking procedures with concrete casted with Hydrocure.
4.7 The fracture energy analysis
Based on the push-out test this section presents the fracture energy analysis for the three types of internal curing concrete casted with GFRP bars respectively.
4.7.1 The fracture energy analysis for Hydrocure
Figure 4.15 shows the interfacial load (P) versus displacement from the load is zero to the maximum value of each specimen. Table 4.13 presents the energy dissipation based on load-displacement of the test cylinders casted with GFRP bars. It
46


is gotten from the area under the load and displacement curve of each specimen which
is calculated by the equations shown in Figure 4.15. And the results show that the energy dissipation tends to increase with the increasing replacement amount as the Figure 4.15(e) shown.
Figure 4.19 shows the shear versus displacement from the shear is zero to the maximum value of each specimen. Table 4.14 summarizes the fracture energy based on shear stress-displacement of the test specimens, which was obtained from the area under the interfacial stress and displacement curve of each specimen and calculated by the quadratic equations shown in Figure 4.19. The trend found is that the measured fracture energy decreased with an increasing replacement ratio. A significant decrease in fracture energy was noticed when the replacement ratio was greater than 50%, while the level of standard deviation tended to decrease. This points out that the failure zone of the test specimen became localized (i.e., reduced variation of the fracture energy).
Figure 4.23(a) and (b) shows the energy dissipation-displacement and fracture-displacement of internal curing concrete mixed with Hydrocure respectively.
4.7.2 The fracture energy analysis for CCA
Figure 4.16 shows the interfacial load (P) versus displacement from the load is zero to the maximum value of each specimen. Table 4.15 presents the energy
47


dissipation based on load-displacement of the test cylinders casted with GFRP bars. It
is gotten from the area under the load and displacement curve of each specimen which is calculated by the equations shown in Figure 4.16. And the results show that the energy dissipation tends to increase with the increasing replacement amount as the Figure 4.16(e) shown. Figure 4.18 shows the energy dissipation comparison of CCA and Hydrocure and as the graph shown the energy dissipation of CCA is higher than Hydrocure.
Figure 4.20 shows the shear versus displacement from the shear is zero to the maximum value of each specimen. Table 4.16 summarizes the fracture energy based on shear stress-displacement of the test specimens, which was obtained from the area under the interfacial stress and displacement curve of each specimen and calculated by the quadratic equations shown in Figure 4.20. The trend found is that the measured fracture energy decreased with an increasing replacement ratio. Figure 4.22 shows the fractural energy comparison of CCA and Hydrocure and as the graph shown the fractural energy of CCA is higher than Hydrocure.
Figure 4.24(a) and (b) shows the energy dissipation-displacement and fracture-displacement of internal curing concrete mixed with CCA respectively.
4.7.3 The fracture energy analysis for SAP
Figure 4.17 shows the interfacial load (P) versus displacement from the load is
48


zero to the maximum value of each specimen. Table 4.17 presents the energy
dissipation based on load-displacement of the test cylinders casted with GFRP bars. It is gotten from the area under the load and displacement curve of each specimen which is calculated by the equations shown in Figure 4.17. And the results show that the energy dissipation tends to increase with the increasing replacement amount as the Figure 4.17(e) shown. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 21 is assumed equal which means 19.805 e_0 021*r = 16.754 e_1'777*a where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 84.62*a+7.97. And that means that adding 0.1% SAP has the similar effect of replacing 16.432% Hydrocure for energy dissipation.
Figure 4.21 shows the shear versus displacement from the shear is zero to the maximum value of each specimen. Table 4.18 summarizes the fracture energy based on shear stress-displacement of the test specimens, which was obtained from the area under the interfacial stress and displacement curve of each specimen and calculated by the quadratic equations shown in Figure 4.21. The trend found is that the measured fracture energy decreased with an increasing replacement ratio. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 4.22 is assumed equal which means 9.3032 e_0,019*r = 8.0163 e_1,777*a where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 93.53*a+7.84. And that means that adding 0.1%
SAP has the similar effect of replacing 17.19% Hydrocure for fracture energy.
49


Obviously, the effect of SAP is significant stronger than Hydrocure.
Figure 4.25(a) and (b) shows the energy dissipation-displacement and fracture-displacement of internal curing concrete mixed with SAP respectively.
4.8 Absorptance test
Internal curing method is a method that let internal curing agents serve as a source of water-supply inside the concrete and facilitate curing process. Therefore the absorptance of water for internal curing agents is a very important property for this research. This section presents the test results of the three different internal curing agents.
4.8.1 Absorptance test results for Hydrocure
Figure 4.26 (a) shows the absorptance of Hydrocure agents.
4.8.2 Absorptance test results for CCA
Figure 4.26 (a) shows the absorptance of CCA agents. By the comparison of Hydrocure, the result is the absorptance of CCA is higher than Hydrocure.
50


4.8.3 Absorptance test results for SAP
Figure 4.26 (b) shows the absorptance of SAP agents. By the comparison of SAP, the result is the absorptance of SAP is much higher than Hydrocure and CCA. That is assumed be the reason why the effect of small amount SAP adding is equivalent to large replacement amount of Hydrocure and CCA.
4.9 Dry shrinkage test
Ten 4X4X111/^ inches beams were tested.
4.9.1 Dry shrinkage test results for Hydrocure
Figure 4.27(a) shows the test results for the shrinkage verse data curve for Hydrocure and Figure 4.27(b) shows the shrinkage divided by depth of the beam for Hydrocure.
51


4.9.2 Dry shrinkage test results for CCA
Figure 4.28(a) shows the test results for the shrinkage verse data curve for CCA and Figure 4.28(b) shows the shrinkage divided by depth of the beam for CCA.
4.8.3 Dry shrinkage test results for SAP
Figure 4.29(a) shows the test results for the shrinkage verse data curve for SAP and Figure 4.29(b) shows the shrinkage divided by depth of the beam for SAP.
52


Table 4.1 Compressive strength (Hydrocure)
Replacement Compressive strength(MPa)
(%) individual averge stdev cov
0%-l 46.8 40.6 5.5 0.14
0%-2 38.5
0%-3 36.4
25%-l 37.8 32.4 9.3 0.29
25%-2 37.8
25%-3 21.7
50%-1 23.8 15.1 7.6 0.50
50%-2 9.7
50%-3 11.8
75%-l 12.1 11.6 3.2 0.28
75%-2 14.6
75%-3 8.2
53


Table 4.2 Compressive strength (CCA)
Replacement Com rressive strength(MPa)
(%) individual averge stdev cov
0%-l 46.8 40.6 5.5 0.14
0%-2 38.5
0%-3 36.4
25%-l 40.6 39.3 1.8 0.05
25%-2 37.3
25%-3 40.0
50%-l 26.0 29.9 3.5 0.12
50%-2 32.8
50%-3 30.9
75%-l 35.3 25.3 11.9 0.47
75%-2 28.5
75%-3 12.1
54


Table 4.3 Compressive strength (SAP)
Replacement Com rressive strength(MPa)
(%) individual averge stdev cov
0%-l 46.8 40.6 5.5 0.14
0%-2 38.5
0%-3 36.4
0.2%-l 23.9 24 0.3 0.01
0.2%-2 23.7
0.2%-3 24.3
0.4%-l 26.2 21.1 4.7 0.22
0.4%-2 16.8
0.4%-3 20.4
0.6%-1 10.4 11 2.4 0.22
0.6%-2 13.6
0.6%-3 8.9
55


Table 4.4 Displacement at maximum load (Hydrocure)
Replacement (%) Displacement at maximum load
individual averge stdev cov
0%-l 1.6 2.1 0.5 0.24
0%-2 2.3
0%-3 2.6
25%-l 1.1 1.5 0.6 0.41
25%-2 1.1
25%-3 2.2
50%-l 1.2 1.4 0.2 0.14
50%-2 1.5
50%-3 1.5
75%-l 1.0 1.2 0.3 0.28
75%-2 1.6
75%-3 1.0
56


Table 4.5 Maximum interfacial load of bonding strength (Hydrocure)
Replacement (%) Max Load (kN
individual averge stdev cov
0%-l 22.6
0%-2 33.7 25.7 7.0 0.27
0%-3 20.7
25%-l 19.0
25%-2 19.1 19.4 0.7 0.03
25%-3 20.2
50%-1 12.8
50%-2 11.6 12.7 0.9 0.07
50%-3 13.5
75%-l 8.4
75%-2 10.1 9.7 1.2 0.12
75%-3 10.7
57


Table4.6 Maximum stress (Hydrocure)
Replacement (%) Max stress (N/mm2)
individual averge stdev cov
0%-l 10.8 12.3 3.4 0.27
0%-2 16.1
0%-3 9.9
25%-l 9.1 9.3 0.3 0.03
25%-2 9.2
25%-3 9.7
50%-1 6.1 6.1 0.5 0.07
50%-2 5.6
50%-3 6.5
75%-l 4.0 4.6 0.6 0.12
75%-2 4.8
75%-3 5.1
58


Table 4.7 Displacement at maximum load (CCA)
Replacement (%) Displacement at maximum load
individual averge stdev cov
0%-l 1.6 1.8 0.4 0.23
0%-2 2.3
0%-3 1.6
25%-l 1.9 2.0 0.3 0.16
25%-2 2.4
25%-3 1.8
50%-1 1.9 2.0 0.5 0.25
50%-2 2.5
50%-3 1.5
75%-l 2.8 2.4 0.5 0.22
75%-2 2.7
75%-3 1.8
59


Table 4.8 Maximum interfacial load of bonding strength (CCA)
Replacement (%) Max load (kN)
individual average stdev cov
0%-l 22.6 25.7 7.0 0.27
0%-2 33.7
0%-3 20.7
25%-l 19.1 18.9 0.3 0.02
25%-2 19.0
25%-3 18.5
50%-1 11.6 13.1 1.4 0.11
50%-2 14.4
50%-3 13.2
75%-l 16.1 17.2 1.1 0.06
75%-2 17.2
75%-3 18.2
60


Table 4.9 Maximum stress (CCA)
Replacement (%) Max stress (N/mm2)
individual average stdev cov
0%-l 10.8 12.3 3.4 0.27
0%-2 16.1
0%-3 9.9
25%-l 9.1 9.0 0.2 0.02
25%-2 9.1
25%-3 8.8
50%-l 7.7 8.2 0.5 0.06
50%-2 8.2
50%-3 8.7
75%-l 5.6 6.3 0.7 0.11
75%-2 6.9
75%-3 6.3
61


Table 4.10 Displacement at maximum load (SAP)
Replacement (%) Displacement at maximum load
individual averge stdev cov
0%-l 1.6 1.8 0.4 0.23
0%-2 2.3
0%-3 1.6
0.2%-l 1.8 1.5 0.4 0.24
0.2%-2 1.1
0.2%-3 1.6
0.4%-l 1.6 1.6 0.1 0.07
0.4%-2 1.7
0.4%-3 1.5
0.6%-1 1.8 1.5 0.2 0.16
0.6%-2 1.4
0.6%-3 1.4
62


Table 4.11 Maximum interfacial load of bonding strength (SAP)
Replacement (%) Max load (kN)
individual average stdev cov
0%-l 22.6 25.7 7.0 0.27
0%-2 33.7
0%-3 20.7
0.2%-l 17.3 16.2 1.5 0.09
0.2%-2 14.5
0.2%-3 16.8
0.4%-l 12.9 14.8 1.8 0.12
0.4%-2 16.6
0.4%-3 14.8
0.6%-1 10.4 10.9 1.3 0.12
0.6%-2 12.4
0.6%-3 10.0
63


Table 4.12 Maximum stress (SAP)
Replacement (%) Max stress (kN/mm2)
individual average stdev cov
0%-l 10.8 12.3 3.7 0.27
0%-2 16.1
0%-3 9.9
0.2%-l 8.3 7.7 0.7 0.09
0.2%-2 6.9
0.2%-3 8.0
0.4%-l 6.2 7.1 0.9 0.12
0.4%-2 7.9
0.4%-3 7.1
0.6%-1 5.0 5.2 0.6 0.12
0.6%-2 5.9
0.6%-3 4.8
64


Table 4.13 Energy dissipation based on load-displacement ( Hydrocure)
Replacement (%) Fracture energy (kNmm)
individual averge stdev cov
0%-l 16.2 19.7 8.4 0.43
0%-2 29.3
0%-3 13.6
25%-l 10.2 14.5 7.9 0.54
25%-2 9.8
25%-3 23.6
50%-l 7.1 6.8 0.6 0.09
50%-2 7.1
50%-3 6.1
75%-l 4.3 4.3 1.0 0.22
75%-2 3.3
75%-3 5.2
65


Table 4.14 Fracture energy based on shear- displacement (Hydrocure)
Replacement (%) Fracture energy (N/mm)
individual averge stdev cov
0%-l 7.8 9.5 4.0 0.43
0%-2 14.1
0%-3 6.6
25%-l 4.9 7.0 3.8 0.54
25%-2 4.7
25%-3 11.4
50%-1 3.4 3.3 0.3 0.09
50%-2 3.4
50%-3 2.9
75%-l 2.1 2.3 0.2 0.10
75%-2 2.4
75%-3 2.5
66


Table 4.15 Energy dissipation based on load-displacement (CCA)
Replacement (%) Energy dissipation (kNmm)
individual averge stdev cov
0%-l 16.2 19.7 8.4 0.43
0%-2 29.3
0%-3 13.6
25%-l 11.9 13.6 1.8 0.13
25%-2 15.5
25%-3 13.4
50%-1 9.9 11.3 3.4 0.30
50%-2 15.2
50%-3 8.8
75%-l 18.8 15.5 2.8 0.18
75%-2 13.8
75%-3 14.0
67


Table 4.16 Fracture energy based on shear- displacement (CCA)
Replacement (%) Fracture energy (N/mm)
individual averge stdev cov
0%-l 7.8 9.4 4.0 0.43
0%-2 14.0
0%-3 6.5
25%-l 5.7 6.5 0.9 0.13
25%-2 7.4
25%-3 6.4
50%-1 4.7 5.4 1.6 0.30
50%-2 7.3
50%-3 4.2
75%-l 9.0 7.4 1.4 0.18
75%-2 6.6
75%-3 6.7
68


Table 4.17 Energy dissipation based on load-displacement (SAP)
Replacement (%) Energy dissipation (kNmm)
individual averge stdev cov
0%-l 16.2 19.7 8.4 0.43
0%-2 29.3
0%-3 13.6
0.2%-l 10.7 9.7 1.6 0.17
0.2%-2 7.9
0.2%-3 10.7
0.4%-l 7.1 9.0 1.7 0.19
0.4%-2 10.5
0.4%-3 9.4
0.6%-1 5.4 6.0 1.8 0.30
0.6%-2 8.1
0.6%-3 4.6
69


Table 4.18 Fracture energy based on shear- displacement (SAP)
Replacement (%) Fracture energy (N/mm)
individual averge stdev cov
0%-l 7.76 9.42 4.01 0.43
0%-2 14.00
0%-3 6.51
0.2%-l 5.10 4.66 0.78 0.17
0.2%-2 3.76
0.2%-3 5.12
0.4%-l 3.41 4.31 0.83 0.19
0.4%-2 5.04
0.4%-3 4.48
0.6%-1 2.58 2.88 0.87 0.30
0.6%-2 3.85
0.6%-3 2.20
70


Compressive strength Compressive strength
(MPa) (MPa)
(a)
(b)
Figure 4.1 Variation of concrete strength with internal curing agents: (a) Hydrocure and CCA; (b) SAP
71


Coulombs
(a)
Figure 4.2 Chloride penetration for concrete with Hydrocure
72


Frequency (Hz) Frequency (Hz)
2000
2000
1500
1000
500
y =-2.8125x+ 1921.9 R2 = 0.9818
Testl 'nT X >. 1500 y = 1919.1 e 002x Testl ATest2
ATest2 XTest3 o c CD 3 O' p 1000 R2 = 0.9689 XTest3 XTest4
XTest4 OTest5 LL 500 OTest5
10 20 30 40 50 60 70 80 Replacement (L
10 20 30 40 50 60 70 Replacementf0/-
I
80
(a)
(b)
Replacement (%)
(c)
Figure 4.3 Frequency and dynamic modulus of concrete with internal curing agents:
(a) Hydrocure; (b) CCA; (c) SAP
73


Replacement (%)
(a) (b)
Figure 4.4 Dynamic modulus (a) comparison of Hydrocure and CCA; (b) SAP
74


LA
CL
Acceleration (mis2)
o -> ro co
Acceleration (mis2)
o ^ ro co
QJ
cr
Acceleration (m/s2)
o - ro
o
cn
3 -
cd cn
ho -
ro
cn
co J

I I I
I i
ro
CD oo -o O) cn 4^
Acceleration (m/s2)
o - ro
o
cn
3
cd cn
ro -
ro
cn
co J
l-v-
I I
i i
I
ccccccccc
^^-^CDOD-slQUl^
ro -> o
co
_i
Run1 ---Run1
Run2 J ] . ..Run2
Run3 ; ---Run3


CN
C/)
E
c
o
2
0)
(D
o
o
<
(e)
Figure.4.5 Acceleration between concrete with Hydrocure and GFRP bar measured from the friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement;
(d) 75% replacement; (e) summary average line
76


cr
-j
-j
CL


Time (s) Time (s)
Acceleration (m/s2) Acceleration (m/s2)
o - ro co o - ro co
Acceleration (m/s2)
o -> ro to
Acceleration (m/s2)
O ^ K) co


(e)
Figure 4.6 Acceleration between concrete with CCA and GFRP bar measured from the friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary average line
78


-J
VO
CL


Acceleration (m/s2)
5) 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) scccccccccccc
CD
5 :
03 -4- (Q K) CD
cooo-ocncn-p^coro
Acceleration (m/s2)
o ^ ro co
o -i---------1----------1----------1
o
cn
cd cn 'cn
ro -
ro
cn
%*
Sr'
I I I
I
I
ccccccccccc
-^-^-^cDco-viooi-^wro ro - o
Acceleration (m/s2)
ro
Acceleration (m/s2)
93 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) 7) i i i i i i i i i i i i
03 -*
cQ ro
CD
cooo-ocncn-p^coro
Run1


Acceleration(m/s2)
(e)
Figure 4.7Acceleration between concrete with SAP and GFRP bar measured from the friction test: (a) 0% adding; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding; (e) summary average line
80


Coefficient of friction Coefficient of friction
0.8 0.6 0.4 -0.2 -
0
0
v/= 0.6063e0 0009x / R2 = ^:1731
10
0.8 -|
0.6 K-------
0.4 -
0.2 -
= 0.5803e00013x R2 = 0.2034
o Hydrocure = CCA Hydrocure ^14 (Hydrocure)
20 30 40 50
Replacement (%)
60 70 80
(a)
4-
y = -1.5703x3 + 1.8044X2- 0.5134x + 0.6107 R2 = 0.1886
0
0
0.2 0.4
Adding (%)
0.6
(b)
Figure 4.8 Coefficient of friction: (a) Hydrocure and CCA; (b) SAP
81


75% LP 75%
0 5 10 15 20
Displacement (mm)
(c) (d)
Figure 4.9 Comparison between MTS and linear potentiometer: (a) 0% replacement;
(b) 25% replacement; (c) 50% replacement; (d) 75% replacement
82


(a)
(b)
(c)
(d)
Figure 4.10 Load versus displacement response of push-out tests with various concrete mixtures with Hydrocure (a) 0% replacement; (b) 25% replacement; (c)50% replacement; (d) 75% replacement; (e) summary
83


(C) (d)
Figure 4.11 Push-out test failure mode: (a) cracking appeared; (b) cracking moved upward, (c) failure mode 1; (d) failure mode 2
84


Full Text
lightweight aggregates .Cement and Concrete Research 31 (2001): 1-3.
[26] Ryan Henkensiefken. Internal curing in cementitious system made using saturated lightweight aggregate .Faculty of Purdue University^2008): 37-123.
[27] J. F. Chen., J. G. Teng. Anchorage strength models for FRP and Boned to concrete. Journal of structure engineering (2001): 1.
[28] J. Xiao., H. Falkner. Bond behaviour between recycled aggregate concrete and steel rebars. ScienceDirect (2005): 1-2.
[29] Luigi. M. G., Silvia .D. simple measurement of the sliding friction coefficient. IOPscience (2004).\ -2.
[30] Ole Mejlhede Jensen., Per Freiesleben Hansen. Water-entrained cement-based materials I. Principles and theoretical background. Cement and Concrete Research (2001) 647
[31] Ole Mejlhede Jensen., Per Freiesleben Hansen. Water-entrained cement-based materials II. Experimental observations. Cement and Concrete Research (2002): 973-974
[32] N. Nestle., A. Kuhn., K. Friedemann ,,C. Horch., F. Stallmach ,,G. Herth Water balance and pore structure development in cementitious materials in internal curing with modified superabsorbent polymer studied by NMR Microporous andMesoporousMaterials (2009): 51-52.
[33] Bart Craeye., Matthew Geimaert., GeertDe. Schutter Super absorbing polymers as an internal curing agent for mitigation of early-age cracking of high-performance concrete bridge decks. Construction and Building Materials
143


(2011) 1-13
[34] Ole Mejlhede Jensen., Pietro Lura. Techniques and materials for internal water curing of concrete. Materials and Structures. (2006): 825
[35] A. R. Gabr., D. A. Cameron. Properties of Recycled Concrete Aggregate
for Unbound Pavement Construction. Journal of materials in civil engineering (2012) :754.
[36] Ryan Henkensiefken., Javier Castro., Haejin Kim., Dale Bentz, & Jason Weiss. Internal Curing Improves Concrete Performance throughout its Life. Concrete InFocus (2009): 1-2.
144



PAGE 1

BEHAVIOR OF CONCRETE WITH INTERNAL CURING AGENTS By JUN WANG B.S., North east Forestry University, 201 4 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment Of the requirements for the degree of Master of Science Civil Engineering 2 014

PAGE 2

ii This thesis for the Master of Science degree by Jun Wang Has been approved for the Civil Engineering program By NY Chang C hair Yail jimmy Kim Chengyu Li November 20, 2014

PAGE 3

iii 2014 JUN WANG ALL RIGHTS RESERVED

PAGE 4

iv Jun Wang (M.S., Civil Engineering) Experimental of Internal Curing Concrete Thesis directed by Associate Professor Yail Jimmy Kim ABSTRACT Glass fiber reinforced polymer (GFRP) composite bars are a promising material that can replace conventional steel reinforcement. GFRP is a non metallic reinforcing material and thus corrosion will not occur during its service life. Given the tensile streng th of GFRP bars are significantly higher than the tensile yield strength of steel bars, GFRP may be more suitable for high performance concrete (HPC) application rather than normal concrete. High performance concrete inhibits moisture ingress into the core of the concrete due to its low permeability associated with a low water cement ratio. Self desiccation of HPC thus takes place and causes autogenous shrinkage. Autogenous shrinkage results in significant cracking of a structural member made of HPC and may lead to premature failure. To avoid this problem, a certain amount of humidity inside the concrete needs to be maintained while the hydration process of the cement is active. Three emerging curing agents are used for the present research to overcome this critical issue respectively. They are light weight aggregate (Hydrocure was the light weight aggregate used in this research), superabsorbent polymer (SAP) and crushed returned concrete aggregate (CCA) This thesis presents the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. Push out bond tests are conducted to achieve this research objective. Emphasis is given to interfacial capacity, failure

PAGE 5

v mode, and fracture energy. Some other proper ties of internal curing concrete like coefficient of friction, compressive strength, drying shrinkage, elastic modulus, and rapid chloride ion penetrability are also included in this thesis. The form and content of this abstract are approved. I recommend its publication. Approved: Yail Jimmy Kim

PAGE 6

vi ACKNOWLEDGEMENTS I really apprecia te the help from my supervisor, Dr.Jimmy Kim, for his invaluable instruction, exhilarated encouragement, and excellent guidance throughout Laboratory at University of Colorado Denver, including Tom Thuis, Jack. Thanks are also extended to my fellow graduate st ude nts Thushara. siriwardanage for his assistance with casting my specimens. I would like to thank John Martinez for his help with my friction test. Full scale tests conducted in this study were made possible with the materials support of Northeast Solite and Barco Fence.

PAGE 7

vii TABLE OF CONTENTS Chapter 1 Introduction ................................ ................................ ................................ ................. 1 1.1 Introduction ................................ ................................ ................................ .............. 1 1.2 Research significance ................................ ................................ ............................... 3 1.3 Obje ctives ................................ ................................ ................................ ................ 4 1.4 Scope ................................ ................................ ................................ ........................ 5 1.5 Thesis outline ................................ ................................ ................................ ........... 6 2 Literature review ................................ ................................ ................................ ......... 7 2.1 Introduction ................................ ................................ ................................ .............. 7 2.2 Different types of internal curing methods ................................ .............................. 9 2.2.1 Substances with chemically bound water ................................ .......................... 9 2.2.2 Substances with physically adsorbed water ................................ .................... 10 2.2.3 Substances with physically held water ................................ ............................ 10 2.2.4 Substances with unbound water ................................ ................................ ...... 11 2.3 LWA ................................ ................................ ................................ ...................... 11 2.3.1 The replacement ratio of LWA ................................ ................................ ....... 12 2.3.2 Water / cement ratio for strength and shrinkage of internal curing concrete. 12 2.4 CCA ................................ ................................ ................................ ....................... 13 2.4.1 Advantages of CCA ................................ ................................ ......................... 13 2.4.2 The characterization of CCA in different psi ................................ .................. 14 2.5 SAP ................................ ................................ ................................ ........................ 14 2.5.1 Function and advantage of SAP ................................ ................................ ...... 15 2.5.2 The characterization of SAP when the adding ratio id different ..................... 15 2.6 GFRP ................................ ................................ ................................ ...................... 16 2.6.1 Advantages of GFRP ................................ ................................ ....................... 16 2.6.2 Properties of the GFRP used in the research ................................ ................... 17 3 Experimental program ................................ ................................ .............................. 22 3.1 Introduction ................................ ................................ ................................ ............ 22

PAGE 8

viii 3.2 Materials used for test specimens ................................ ................................ .......... 22 3.2.1 Hydrocure ................................ ................................ ................................ ........ 23 3.2.2 CCA ................................ ................................ ................................ ................. 24 3.2.3 SAP ................................ ................................ ................................ .................. 24 3.2.4 GFRP bars ................................ ................................ ................................ ....... 25 3.3 Description of test specimens ................................ ................................ ................ 25 3.3.1 Cylinders casted with GFRP bars specimens. ................................ ................. 25 3.3.2 Cylinders casted without GFRP bars specimens ................................ ............. 26 3.3.3 Beam casted for friction tests and frequency tests ................................ .......... 27 3.3.4 Beams casted for shrinkage ................................ ................................ ............. 27 3.3.5 Cylinders for chloride permeability tests ................................ ........................ 28 3.4 Experimental setup and loading ................................ ................................ ............. 28 3.4.1 Experimental setup for cylinders with GFRP bars ................................ .......... 28 3.4.2 Experimental setup for concrete compressive testing ................................ ..... 29 3.4.3 Experimental setup for friction test ing ................................ ............................ 30 3.4.4 Experimental setup for dynamic elastic modulus testing ................................ 30 3.4.5 Experimental setup for shrinkage testing ................................ ........................ 31 4 Experimental results and discussion ................................ ................................ ......... 36 4.1 Introduction ................................ ................................ ................................ ............ 36 4.2 Compressive stress test results ................................ ................................ ............... 37 4.2.1 Compressive stress test results for Hydrocure ................................ ................ 37 4.2.2 Compressive stress test results for CCA ................................ ......................... 38 4.2.3 Compressive stress test results for SAP ................................ .......................... 38 4.3 Chloride ion penetrability tests ................................ ................................ .............. 3 9 4.3.1 Chloride ion penetrability tests results for Hydrocure ................................ .... 39 4.4 Elastic modulus tests ................................ ................................ .............................. 39 4.4.1 Elastic modulus tests results for Hydrocure ................................ .................... 39 4.4.2 Elastic modulus tests results for CCA ................................ ............................. 40 4.4.3 Elastic modulus tests results for SAP ................................ .............................. 40 4.5 Friction test results ................................ ................................ ................................ 41 4.5.1 Friction test results for Hydrocure ................................ ................................ ... 41 4.5.2 Friction test results for CCA ................................ ................................ ........... 42 4.5.3 Friction test results for SAP ................................ ................................ ............ 42 4.6 Push out test results ................................ ................................ ............................... 43 4.6.1 Push out tests results for Hydrocure ................................ ............................... 43 4.6.2 Push out tests results for CCA ................................ ................................ ........ 44 4.6.3 Push out tests results for SAP ................................ ................................ ......... 45

PAGE 9

ix 4.7 The fracture energy analysis ................................ ................................ .................. 46 4.7.1 The fracture energy analysis for Hydrocure ................................ .................... 46 4.7.2 The fracture energy analysis for CCA ................................ ............................. 47 4.7.3 The fracture energy analysis for SAP ................................ .............................. 48 4.8 Absorptance test ................................ ................................ ................................ ..... 50 4.8.1 Absorptance test results for Hydrocure ................................ ........................... 50 4.8.2 Absorptance test results for CCA ................................ ................................ .... 50 4.8.3 Absorptance test results for SAP ................................ ................................ ..... 51 4.9 Dry shrinkage test ................................ ................................ ................................ .. 51 4.9.1 Dry shrinkage test results for Hydrocure ................................ ........................ 51 4.9.2 Dry shrinkage test results for CCA ................................ ................................ 52 4.8.3 Dry shrinkage test results for SAP ................................ ................................ .. 52 5 Numerical modelling ................................ ................................ .............................. 109 5.1 Introduction ................................ ................................ ................................ .......... 109 5.2 Numerical model for interfacial behavior of internal curing concrete and GFRP ................................ ................................ ................................ ................................ .... 110 5.2.1 Model procedure ................................ ................................ ............................ 111 5.2.2 First point (S1, Tmax) ................................ ................................ ................... 111 5. 2. 3 Coefficient a ................................ ................................ ................................ 112 5.2.4 Second point (S2, Tf) ................................ ................................ .................... 113 5.2.5 Three models of the three types of internal curing agents ............................ 113 5.3 Regression for compressive strength and maximum stress ................................ 117 5.3.1 Regression for Hydrocure ................................ ................................ ............. 117 5.3.2 Regression for CCA ................................ ................................ ...................... 118 5.3.3 Regression for SAP ................................ ................................ ....................... 118 6 Summary & conclusions ................................ ................................ ......................... 136 6.1 Summary ................................ ................................ ................................ .............. 136 6.2 Conclusions ................................ ................................ ................................ .......... 137 6.3 Recommendations for future work ................................ ................................ ...... 139 Reference s .. 140

PAGE 10

x LIST OF TABLES Table 2.1 Properities of Hydrocure ................................ ................................ ........................ 18 2.2 Graduation analysis of CCA ................................ ................................ .................. 19 2. 3 Properties of SAP ................................ ................................ ................................ ... 20 4.1 Compressive strength (Hydrocure). ................................ ................................ ....... 53 4.2 Compressive strength (CCA). ................................ ................................ ................ 54 4.3 Compressive strength (SAP). ................................ ................................ ................. 55 4.4 Displacement at maximum load (Hydrocure). ................................ ....................... 56 4.5 Maximum interfacial load of bonding strength (Hydrocure). ................................ 57 4.6 Maximum stress (Hydrocure). ................................ ................................ ............... 58 4.7 Displacement at maximum load (CCA). ................................ ................................ 59 4.8 Maximum interfacial load of bonding strength (CCA). ................................ ......... 60 4.9 Maximum stress (CCA). ................................ ................................ ........................ 61 4.10 Displacement at maximum load (SAP). ................................ ............................... 62 4.11 Maximum interfacial load of bonding strength (SAP). ................................ ........ 63 4.12 Maximum stress (SAP). ................................ ................................ ....................... 64 4.13 Energy dissipation based on load displacement ( Hydrocure). ............................ 65 4.14 Fracture energy based on shear displacement (Hydrocure). ............................... 66 4.15 Energy dissipation based on load displacement (CCA). ................................ ..... 67

PAGE 11

xi 4.16 Fracture energy based on shear displacement (CCA). ................................ ........ 68 4.17 Energy dissipation based on load displacement (SAP). ................................ ...... 69 4.18 Fracture energy based on shear displacement (SAP). ................................ ........ 70 5.1 The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for Hydrocure ...121 5.2 The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for Hydrocure ...122 5.3 The average value of replacement; friction coefficient; elast ic modulus; stress and compressive from the Monte Carlo method generated results for Hydrocure ...123 5. 4 Test results for regression analysis ................................ ................................ ....... 123 5. 5 Regression analysis of fc'(Mpa) for Hydrocure ................................ ................... 124 5. 6 Regression analysis of maximum stress for Hydrocure ................................ ....... 125 5. 7 Regression analysis of fc'(Mpa) for CCA ................................ ............................ 126 5. 8 Regression analysis of maximum stress for CCA ................................ ................ 127 5. 9 Regression analysis of fc'(Mpa) for SAP ................................ ............................. 128 5. 10 Regression analysis of maximum stress for SAP ................................ .............. 129

PAGE 12

xii LIST OF FIGURES Figure 2.1 Different kinds of internal curing materials: (a) Hydrocure; (b) Crushed returned concrete aggregate (CCA); (c) Superabsorbent polymer (SAP) ................................ .. 21 3.1 (a) GFRP bar; (b) Cylinder cast for a concrete compressive test; (c) Cylinders cast with GFRP bars for push out tests ................................ ................................ ............... 32 3.2 Specimen detail for Push out test: (a) Test setup; (b) Schematic diagram; (c) Plastic tube for unbonding GFRP bar ................................ ................................ .......... 33 3.3 Material characterization with various internal curing agents: (a) concrete strength (ASTM C873 10a; (b) chloride permeability (ASTM C114); (c) shrinkage (A STM C490/C490M 11); (d) coefficient of friction (Luigi M Gratton, 2006); (e) dynamic elastic modulus (ASTM C215) ................................ ................................ .................... 35 4.1 Variation of concrete strength with internal curing agents: (a) Hydrocure and CCA; (b) SAP. ................................ ................................ ................................ ........................ 71 4.2 Chloride penetration for concrete with Hydro cure ................................ ................ 72 4.3 Frequency and dynamic modulus of concrete with internal curing agents: (a) Hydrocure; (b) CCA; (c) SAP ................................ ................................ ...................... 73 4.4 Dynamic modulus (a) comparison of Hydrocure and CCA; (b) SAP ................... 74 4.5 Acceleration between concrete with Hydrocure and GFRP bar measured from the friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary av erage line. ................................ ................................ ...... 76 4.6 Acceleration between concrete with CCA and GFRP bar measured from the friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e)summary average line. ................................ ................................ ....... 78 4.7Acceleration between concrete with SAP and GF RP bar measured from the friction test: (a) 0% adding; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding; (e)summary average line. ................................ ................................ ................................ ................. 80 4.8 Coefficient of friction: (a) Hydrocure and CCA; (b) SAP ................................ ..... 81

PAGE 13

xiii 4.9 Comparison between MTS and linear potentiometer: (a) 0% replacement; (b) 25% replacement; (c)50% replacement; (d) 75% replacement; ................................ ........... 82 4.10 Load versus displacement response of push out tes ts with various concrete mixtures with Hydrocure (a) 0% replacement; (b) 25% replacement; (c)50% replacement; (d) 75% replacement; (e) summary ................................ ........................ 83 4.11 Push out test failure mode: (a) cracking appeared; (b) cracking moved upward. (c) failure mode 1; (d) failure mode 2 ................................ ................................ ............... 84 4.12 Load versus displacement response of push out tests with various concrete mixtures with CCA (a) 0% replacement; (b) 25% replacement; (c)50% replacement; (d) 75% replacement; (e) summary ................................ ................................ .............. 85 4.14 Load versus displacement response comparison with different internal curing agents ................................ ................................ ................................ ........................... 87 4.15 The Energy dissipation (area A) of push out specimens with Hydrocure: (a) 0% replacement ; (b) 25% replacement ; (c) 50% replacement ; (d) 75% replacement ; (e) summary ................................ ................................ ................................ ....................... 90 4.16 The Energy dissipation(area A) of push out specimens with CCA: (a) 0% replacement ; (b) 25% replacement ; (c) 50% replacement ; (d) 75% replacement ; (e) summary ................................ ................................ ................................ ....................... 91 4.17 The Energy dissipation(area A) of push out specimens with SAP: (a) 0% adding ; (b) 25% adding ; (c) 50% adding ; (d) 75% adding ; (e) summary ............................. 93 4.18 The Energy dissipation comparison of different internal curing agents .............. 94 4.19 The Fracture energy of shear with Hydrocure: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary .................... 96 4.20 The Fracture energy of shear with CCA: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary .................... 98 4.21 The Fracture energy of shear with SAP: (a) 0% adding; (b) 25% adding; (c) 50% adding; (d) 75% adding; (e) summary ................................ ................................ ....... 100 4.22 The Fracture energy comparison of different internal curing agents ................. 101 4.23 Response comparison for concrete mixed with Hydrocure: (a) energy dissipation

PAGE 14

xiv displacement; (b) fracture displacement ................................ ................................ 102 4.24 Response comparison for concrete mixed with CCA: (a) energy dissipation displacement; (b) fracture displacement ................................ ................................ 103 4.25 Response comparison for concrete mixed with SAP: (a) energy dissipation displacement; (b) fracture displacement ................................ ................................ 104 4.26 Absorptance of different internal curing agents: (a) Hydrocure and CCA; (b) SAP ................................ ................................ ................................ ............................ 105 4.27 Dry shrinkage test results for Hydrocure: (a) shrinkage; (b) shrinkage divided by depth .107 4.28 Dry shrinkage test results for CCA: (a) shrinkage; (b) shrinkage divided by depth .. 1 08 4.29 Dry shrinkage test results for SAP: (a) shrinkage; (b) shrinkage divided by depth ..109 5.1 Model results comparison with tests results with Hydrocure: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement ........................... 131 5.2 Model results comparison with tests results wi th CCA: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement ................................ ......... 133 5.3 Model results comparison with tests results with SAP: (a) 0% adding; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding ................................ ................................ 135

PAGE 15

1 1 I ntroduction 1.1 Introduction High performance concrete (HPC) has many advantages such as high strength, durability, light weight of structural members, and favorable long term maintenance costs. Because of these superior characteristics of HPC, modern highway bridges and other types of structural members adopt high performance concrete frequently. However, self desiccation of high performance concrete often takes place due to inadequate curing. Then HPC demonstrates significant cracking and may lead to premature failure. Therefore, the inadequate curing problem needed to be solved. A curing method internal curing was introdu ced. By American Concrete Institute cementitious mixture using reservoirs, via pre wetted lightweight aggregate, that readily release water as needed for hydration or to rep lace moisture lost through evaporation or self There are many internal curing agents this research will focus on light weight aggregate (LWA) it serves as a source of water supply inside the concrete and facilitate curing process. Other two internal curing agents SAP and CCA will also be included in this thesis to compare with the results of LWA. With internal curing high performance con crete will have low cracking Additionally, this kind of concrete also has many other advantages like lowe r permeability, early

PAGE 16

2 age, and la ter age compressive strength Cusson, Lounis and Daigle reported that internal curing high life of HPC bridge decks by more than 20 years Zhutovsky and Kovler found that the effect of internal curing was changing for high performance concrete when the water cement ratio is different In this project, to avoid the influence of different water cement ratio, the water cement ratio of all mixes is 0.4. As man y authors chose the number around 0.4 as water cement ratio when cast ing internal curing concrete Cusson and Hoogeveen focused on the effect of different amount of LWA for internal curing and it turns out that the amount of autogenous shrinkage cracking i s different among the changing of the replacement of LWA. That is because LWA has small microscopic pores which enable the aggregate to absorb considerable water to let the high performance concrete cure from the inside to out side And different amount of LWA will absorb different amount of water which will affect the performance of HPC. So the replacement levels of LWA for internal curing plays a very important role here. The result of how internal curing concrete combines with GFRP when the replacement of LWA is changing is what the present research focuses on. Since GFRP contacts the internal curing concrete through a mechanical bond, the mechanical bond is really important in this application as it can transfer stresses and keep structural integrity betw een GFRP and internal curing concrete. In our research, we use Push out bond tests to achieve the bond stresses. Thus, this paper focuses on the different bond stresses when the amount of LWA of internal curing high performance concrete curing changes. Bec ause of this test, we can understand the behavior of HPC

PAGE 17

3 with internal curing when reinforced with GFRP bars. The objectives of the research are examining the fundamental material characteristics of HPC and bond stresses between internal curing HPC and GFR P depending upon the degree of internal curing (i.e., strength). This paper deals with an experimental study concerning (1) interfacial stress capacity (2) failure mode (3) fracture energy (4) friction (5) permeability (6) shrinkage and (7) elastic modulus 1.2 Research s ignificance Many researches related to internal curing and reinforced concrete were done these days since reinforced concrete is one of the main materials adopted to build structure. Besides, internal curing is a really effective method to deal with self desiccation p roblem. But there are very few projects invested the relationship between internal curing concrete and GFRP bars. GFRP is a non metallic reinforcing material and thus corrosion will not occur during its service life. Given the tensile strength of GFRP bars is significantly higher than the yield strength of steel bars, GFRP may be more suitable for high performance concrete (HPC) application rather than normal concrete. Therefore, GFRP bars may replace reinforcement to be good partner with high performance c oncrete. This urge more studies to be performed about this internal curing concrete together with GFRP bars topic like this research. Since there are many varieties of internal curing agents, the optimal one needed to be

PAGE 18

4 picked up to cast high performance concrete to combine with GFRP bars. The present study compared the three common internal curing agents, Hydrocure, CCA, SAP, which can be bought from commercial market conveniently and investigated their bond strength with GFRP bars when the internal curin g agent replacement amount is different. 1.3 Objectives The principal objectives of this research program are to investigate the bond strength between GFRP bars and different internal curing agents when the internal curing agents replacement or adding amount changed and compare the results of different types of internal curing agents. Specific objectives addressed in this study include: 1. Investigate the concrete compressive strength of internal curing concrete when the replacement ratios are di fferent. 2. Develop failure modes of the bond tests 3. Evaluate the fracture energy of bond tests. 4. Study the coefficient of friction of internal curing concrete when the replacement amount is different 5. Observe the elastic modulus of internal curing concrete when the replacement amount is different

PAGE 19

5 6. Test the drying shrinkage of internal curing concrete casted by LWA when the replacement ratios changed. 7. Get the value of rapid chloride ion penetrability of internal curing concrete casted by LWA when the replacement r atios changed 8. Use the results of these properties of internal curing concrete to perform regressions. 1.4 Scope The scope of the study consists of an experimental investigation and a numerical model to investigate the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. The aim of the experimental program was to study the effects of varying the amount of internal curing agents replacement or adding and the types of internal curi ng aggregates. A numerical model was developed to predict the shear replacement responses of internal curing concrete bonded with GFRP bars. The model method was CEB FIP came from Euro Code. The model assumed four sections to describe the shear displ acement procedure results. To complete the model, a parametric a is needed to determine. For every kind of internal curing concrete, one value of a is determined.

PAGE 20

6 1.5 Thesis o utline The contents of the thesis are briefly outlined below: Chapter 2 : provide s a review of literature related to previous research on internal curing concrete, high performance concrete and GFRP bars. Chapter 3 : presents a detailed description of the experimental program, consisting of design and fabrication of test specimens, in strumentation, test setup, and procedures. Chapter 4 : provides the results of the experimental investigations into the bond strength between internal curing concrete with different replacement ratios, coefficient of friction, compressive strength, drying shrinkage, elastic modulus, and rapid chloride ion penetrability are also included in this thesis. Chapter 5 : presents numerical modeling for interfacial behavior of internal curing concrete and GFRP and r egression for compressive strength and maximum stre ss Chapter 6 : presents conclusions of the study and recommendations for future research into internal curing concrete together with GFRP bars. References

PAGE 21

7 2 L iterature review 2.1 Introduction This chapter presents the historical development of internal curing methods. Previous research related to GFRP bars is discussed. The first part of this chapter discusses the different types of internal curing methods. Then, research on some certain internal curing aggregates and their advant ages and disadvantages. First section investigates Hydrocure. The next section examines CCA. Finally, SAP is studied. After the investigation of internal curing agents, GFRP bars will be followed. Traditional curing method cures concrete from outsi de to inside and external curing water is only able to penetrate several mm into low w/c ratio concrete. Therefore, for high performance concrete (HPC) whose permeability is low often can not be cured well by external curing. HPC was introduced to the mark et in the 1980s. Those days, many state highway agencies begin to implemented HPC for bridge decks to protect reinforcing steel from corrosion as the permeability of HPC is low. It seems like high strength would make concrete less likely to crack. But from the Ohio Department of Transportation, many decks cracked. Besides, Due to hydration of cementitious material self desiccation occurs as internal humidity decr eases (Bentz and Snyder, 2005). Those problems make HPC ca n n ot live up to expectations. To sol ve this problem internal curing method was introduced. Actually as far back as

PAGE 22

8 1957, Paul Klieger found that lightweight aggregate absorb considerable water during mixing which apparently can transfer to the paste during hydration (Klieger, 1957). Besides lightweight aggregate other materials that could function as internal curing aggregate also be investigated. Jensen& Hansen found superabsorbent polymers have the ability to absorb significant water to make the concrete cure from inside to outside (Jensen & Hansen, 2001). And the following section briefly describes internal curing methods. By using internal curing methods, many characteristics can be improve, such as shrinkage (including autogenous), cracking, early age and later age compressive strength, lower permeability, resistance to freeze thaw, minimization of carbonation, densification of the interfacial transition zone, improved mortar strength, and reduced warping. Cusson, Lounis and Daigle reported that internal curing high performance concrete b bridge decks by more than 20 years (Cusson, 2010) which clearly show the advantages of internal curing method. Although internal curing agents were used for the HPC, corrosion still occur to H PC easily which makes GFRP may be a good partner for HPC as GFRP is a non metallic reinforcing material and thus corrosion will not occur during its service life. The last section of this chapter will give some brief introduction to GFRP.

PAGE 23

9 2.2 Different t ypes of internal curing methods The following section briefly describes internal curing methods based on several different physical or chemical principles: chemically bound water, physically adsorbed water, physically held water, unbound water (Ole, 2006). And the followings are some examples of this classification. 2.2.1 Substances with chemically bound water To serve as potential materials for internal water curing the chemical substance must content a large amount of water and can give t he water off easily during the curing period when the surrounding relative humid ity decreased. At the same time the chemical substance must be compatible with the cementitious system. There are many substance s can serve as this like C4AH19 Usually, an am ount of water of around 22 23% of the weight of the anhydrous cement is chemically bound at complete hydration of cement [Powers & Brownyard 1948]. But for these materials, they can contain around 50% water by weight and also can give off the water when su rrounding relative humidity drops. They look like perfect materials for internal curing, however, when the relative humidity is high only a part of the water can be released which makes this substances with chemically bound water have not been used techniq ue.

PAGE 24

10 2.2.2 Substances with physically adsorbed water Different from substances with chemically bound water, there are many attempts utilize types of substances with physically adsorbed water. Such as n atural bentonite clay which ha s the ability to absorb several molecular layers of water and SAP whose theoretical maximum water absorption is around 5,000 times their weight (Ole, 2006). And SAP is one of the main internal curing materials that studied in this research. The key of the a mount of physically bound water depends on the relative humidity of the pore system. The thickness of the adsorption layer ranges from 1 monomolecular water layer at 20% relative humility to about 6 monomolecular layers at 100% relative humility (Hagymassy et al. 1969, Setzer. 1977, Badmann et al. 1981). 2.2.3 Substances with physically held water Substances with physically held water serve as a container for internal curing water. That is because they have small microscopic pores whose sizes abov e around 100 nm which can storage the internal curing water and enable the aggregate to absorb enough water to let the concrete cure from inside to outside. For the aggregates own smaller poles than 100 nm may not work as internal curing material as the po res hold the water so tightly that it can not be released for the cementitious reactions. Pumice, Perlite, Liapor and Leca, Stalite, Diatomanceous earth and

PAGE 25

11 Hydrocure all can serve as substances with physically held water as they share the similar characte ristic their pores sizes are suitable for holding and giving of water at right time. In this project, Hydrocure is used as the internal curing aggregate to investigate the properties of LWA together with GFRP bars. 2.2.4 Substances with unbound water The last kind of internal c uring agents talked here is sub s tances with unbound water. Two examples about it are given here, one is Microencapsulation and the other one is Emulsified water. The first capsule wall has ability to remain water and relea se it until setting by both chemical and physical ways although the price of it is relatively high. The second type is a kind of colloid dispersed with small particles. This emulsifying agent may work well. 2.3 LWA The LWA used in the research is Hydrocure as shown i n Figure 2.1(a) Table 2.1 s hows the properties of Hydrocure investigated here.

PAGE 26

12 2.3.1 The replacement ratio of LWA Since 1990s, research has shown that LWA can provide internal water reservoirs. Philleo proposed that using LWA as a replacement to normal weight aggregate to reduce the effects of self desiccation. To get the optimal replacement of LWA Dutch perform ed the work to replace natural weight aggregates with LWAs which show that when the partial replacement amount of natural weight aggregate ranges from 0% to 25% LWA did not have negative effect on the compressive strength. Especially, when w/c is low like 0.33, compressive strength of internal curing concrete with replacement of 10% and 17% even are higher than the concrete with no LWA replacement. Besides, with replacement percentages up to 25% the autogenous shrinkage reduced significantly. Another test u sed Hydrocure showed that when the water/ Cement ratio is 0.395 the compressive strength increased from the replacement is 0% to 15% and decreased from 25% to 25%, but the strength at 25% replacement ratio was not weaker than 0% replacement ratio. 2.3.2 Water / cement ratio for strength and shrinkage of internal curing concrete. Internal curing is the process by which the hydration of cement occurs because of availability of additional internal curing water is not part of mixing water (ACI 308). Fo r internal curing concrete water is a key to determine several properties of it, such as strength and elasticity modulus and autogenous shrinkage. Semion

PAGE 27

13 Zhutovsky and Konstantin Kovler did test on strength and dynamic elasticity modulus, shrinkage and to ta l shrinkage of HPC made at w/c of 0.21 0.33 (Semion Zhutovsky and Konstantin Kovler, 2012). And the results show that the for concrete with w/c ratio of 0.33 and 0.25 internal curing makes shrinkage higher, however, internal curing let the shrinkage of c oncrete with w/c of 0.21become lower. For autogenous shrinkage internal curing method totally eliminated it even when w/c is 0.21. 2.4 CCA CCA i s shown in Figure 2.1 (b) and Table 2.2 shows the properties of CCA used in the research. 2.4.1 Advantages of CCA The pore structure of CCA is similar to that of cement past e (Ryan, 2008). Both of them ha ve the primary pore containing component in the crushed aggregates which makes it possible for CCA to serve as internal curing agents. CCA also has a typical gravity of 2.04 (Kim & Bentz, 2008). It is assessed that every year there are about 2 % to 10 % of the 460 million cubic yards of ready mixed concrete produced

PAGE 28

14 in the USA is returned to the concrete plant (Obla et al. 2007) which makes it become lower price products. Due to the economic value and the value of environmental protection of CCA, it becomes one of the popular internal curing agents that used by commercial markets. 2.4.2 The characterization of CCA in different psi Haejin Kim and Dale Bentz explored the use of CCA in their article in 2008. This article investigated CCA in different psi from 1000psi to 5000psi increased by 2000psi to select the best performing CCA. And the results noted that when the strength of CCA is low the replacement of sand should make a strength reduction. It is also shown that CCA 3000 had the highest strength among the three types of CCA. That is because CCA 5000 had the larger volumetric content in the motor which offset the higher i nherent strength, so that lower compressive strength occurred (Dale, 2008) 2.5 SAP Figure 2.1 (c) shows SAP and Table 2.3 shows the properties of SAP used in the research.

PAGE 29

15 2.5.1 Function and advantage of SAP SAP is a kind of polymetric material which can absorb a large amount of liquid from surroundings and to retain the liquid within its structure without dissolving (Buchholz, 1998) and when the initial free water has been consumed by the hydration reactions, the water absorbed by the S AP will b e gradually released. So that SAP is able to be performed as internal curing admixture of concrete. Contrary to LAW SAP permits free design of the pore sharp and the pore size distribution of the c oncrete. Besides, SAP can avoid the undesirable addition of significant amount of mechanically poor aggr egate ( Ole, 2001). Furthermore, due to the maintenance of higher internal humidity adding SAP to a concrete can make it have higher ultimate degree of hydration of cement then let the compressive strength become higher at long term (Geiker, 2004). 2.5.2 The characterization of SAP when the adding ratio id different In 2002, O.M. Jensen and P.F Hansen did the tests of two different SAPs. The SAPs added at a rate of 0 0.6 wt. % of cement (Ole, 2002). An d the results noted that the compressive strength of 45* 90 mm mortar cylinders after 1day of sealed curing followed by 27 days decreased from 0% SAP to 0.6 % SAP, i.e., the strength was reduced by 19% due to the water entrainment. This article also gave a assumption of this results that is measured lower strength of the water entrained mortar compared

PAGE 30

16 to the reference mortar may be caused by the moisture condition. However, further research was needed to prove this comment. 2.6 GFRP GFRP is a fiber reinforced polymer made of plastic reinforced by glass fibers commonly woven into a mat. It first developed in the mid 1930's and has become a staple in the building industry. In 1967, the attempted destruction of Disneyland's "House of the Future" built in 1956 7 showed the architectural advantages, the futuristic house was no longer deemed necessary and was scheduled to be destroyed in 1967. Amazingly, the wrecking ball merely bounced off the structure as it was entirely built of fiberglass, then the possibilities for GFRP were recognized and began to grow. By 1994, nearly 600 million pounds of composite materials were used in the building industry. Besides field studies also show that glass fiber reinforced polymer offers a low life cycle cost option for reinforcement in co ncrete pavements (Roger H. L, 2008). 2.6.1 Advantages of GFRP There are many advantages of GFRP are collected such as it has a very high strength

PAGE 31

17 to weight ratio, the weight of GFRP is only around 2 to 4 lbs. per square foot which means faster installation, less structural framing, and lower shipping costs. The resistance of GFRP is also superb it can resist salt water, chemicals, acid rain and mos t chemicals. Furthermore, it can let the domes and cupolas be resined together to form a one piece, watertight structure and can be molded to any shape or form. Researches also showed that after 30 years, there was no loss of laminate properties and GFRP c ould stand up to category 5 hurricane Floyd with no damage, while nearby structures were destroyed 2.6.2 Proper ties of the GFRP used in the research For Aslan 100 Fiberglass rebar use in the investigation there were many benefits. They are impervious to chloride ion and low PH chemical attack, their tensile strength are greater than steel, the weight of them is just th of the steel rebar and they ar e electrically and thermally non conductive( Hughes Brother, 2011).

PAGE 32

18 Table 2.1 Properities of Hydrocure (Ref: http://www.nesolite.com/physicalcharac.htm ) Density Dry Loose (ASTM C 29) 45 pcf 720 kg/m3 Dry Rodded (ASTM C 29) 50 pcf 800 kg/m3 Saturated Surface Dry Loose (ASTM C 29) 48 pcf 768 kg/m3 Absorption Saturated Surface Dry, 24 hour (ASTM C 127) 10% Specific Gravity Saturated Surface Dry (ASTM C 127) 1.5 Soundness (Loss) Magnesium Sulfate (ASTM C 88) 1% Sodium Sulfate (ASTM C 88) 0.50% After 300 Cycles Freezing and Thawing (AASHTO T103) 1% Resistance to Abrasion Los Angeles Abrasion (AAHTO T 96B) <30% Impurities Clay Lumps (ASTM C142) None Organic Impurities (ASTM C40) None Electrical Resistance Saturated Surface Dry ohm cm Aggregate Chemical Characteristics Ignition Loss (ASTM C114) 0 Stains (ASTM C641) None Chlorides (NaCl) 0.50 ppm

PAGE 33

19 Table 2.2 Graduation analysis of CCA (Ref: http://www. alliedrecycle.com /fines.aspx) Aggregate crushing value, % (AS 1141.21) 23.1 Bulk density, kg/m 3 (AS 1141.6) 2394 Water absorption, % (AS 1141.6) 5.6 Impurity level, % (AS 1141.32) 0.6 LOI, % 4.9 Graduation analysis of CCA. Sieve Size Allied Test 1" 100 3/4" 100 1/2" 100 3/8" 100 #4 78 #8 55 #16 36 #30 23 #50 14 #100 9 #200 6.2

PAGE 34

20 Table 2.3 Properties of SAP (Ref: www.m2polymer.com ) Product Attributes: Reduces Waste Disposal Costs Expands in Volume by Less Than 1%. Non Biodegradable Polymer (40 CFR 264.314 (e)(ii)) Polycarboxylate Polymer Will Not Release Trapped Ionic Contaminants If Solute Evaporation Occurs. Strong Ion Exchange Capability Allows For Heavy Metal Binding And For Many Solidified Wastes To Pass TCLP. When Used Properly, Waste Sludges Will Pass Paint Filter Test (EPA 9095). Polymer is suitable for Incineration. No halogens. Freeze Thaw Teste d. The Polymer Will Not Release Free Liquids. Polymer Will Excellently Absorb Aqueous Wastes of pH > 4. For Highly Acidic Wastes, Neutralization (pH Adjustment) Is Recommended. Approved Sorbent at Hanford (WA) & WIPP (TRUCON Codes) Typical Absorptive Properties: Free Swell in DI Water 400 500 X Free Swell in 1 % NaCl 45 55 X Free Swell in 2 % NaCl 35 40 X Free Swell in 10 % NaCl 19 25 X Free Swell in 1 % CaCl2 20 25 x Free Swell in 8N NaOH 24 30 x Bulk Density 4.5 to 6.5 lbs/gallon Liquid Release Test (EPA 9096): Waste Lock 770 has the ability to both absorb under pressure and to retain absorbed liquids at high pressures: At 25X Hydration, PASSES at 50 psi At 10X Hydration, PASSES at 75 psi Bulk Density = 4.5 to 6.5 lbs/gallon

PAGE 35

21 (a) (b) (c) Fig ure 2.1 Different kinds of internal curing materials : (a) Hydrocure ; (b) Crushed returned concrete aggregate (CCA) ; (c) Superabsorbent polymer (SAP)

PAGE 36

22 3 E xperimental program 3.1 Introduction The experimental program investigated the bonding strength between GFRP and internal curing concrete casted with different amount of internal curing agents and some other properties of internal curing concrete. The internal curing agent s included Hydrocure, Superabsorbent polymer and crushed returned recycled concrete. The objective was to determine the optimal replacement amount of internal curing agents for different properties of internal curing concrete and compared the results when the variety of internal curing agents is different. To get those proper ties, for the Hydrocure, 12 cylinders with GFRP bar, 12 cylinders without bar and 8 beams were test e d. Same tests were conducted for SAP and CCA. This chapter describes the internal cur ing materials used to fabricate the test specimen s, the fabrication process, the experimental procedure, instrumentation, push out test for cylinders with GFRP bar, compression test for cylinders without bar, friction test, frequency test, permeability test and shrinkage test. 3.2 Materials used for t est s pecimens The GFRP used is 13mm as shown in Figure 3.1(a) The ultimate tensile

PAGE 37

23 stress is 689 MPa with a modulus of 40.8 GPa. Its maximum bond stress to concrete is 11.6 MPa. The GFRP was embedd ed in the high performance internal curing concrete, the basic constituents of the concrete is water (205 kg/m3), cement (445.65 kg/m3), sand (600kg/m3), and gravel (1095 kg/m3). T o proceed the project, natural sand needs p artly to be taken place by Hydr o cure and CCA and for SAP, we need to add specific amount of SAP according to the weight of cement added. 3.2.1 Hydrocure Hydrocure is the main intern al curing agents that are used. Its bulk damp loose unit weight is 45 55 pcf (720 880 kg/m3) d epending on gr adation. The gradation of c oarse aggregate is 5 20 mm. In Hydrocure many small microscopic pores enable the aggregate to absorb about 20% of its weight with water. Therefore, instead of curing concrete with water from the outside in, internal curing is for curing from the inside out. Before casting concrete the surface of Hydrocure should be dry. Move the Hydrocure to the batching plant and sprinkle with water during storage to maintain 24 hour absorption of a minimum of 15 % at 24 h ours and a minimum of 9 % to 10 % in 30 minutes. Then adequate saturation of the cement can be maintained during hydration which improves the strength and dimensional stability by reducing shrinkage and warping. Table 13 shows the properties of the Hydrocu re used in this research.

PAGE 38

24 3.2.2 CCA Compared with Hydrocure, CCA costs less and has potential as an internal curing agent due to its high absorption capacity. Besides, it is low specific gravity. Those two properties are crucial factors for the internal curing. Therefore, this research also used CCA to do the same tests as Hydrocure and compare the results of the two. The CCA was prepared at Allied Recycled Aggregates. Measured particle size distribution of CCA is shown in table. Its la abrasion is 38% and its modified protector is 118.5@ 12.3. 3.2.3 SAP SAP A superabsorbent polymer is able to absorb a si gnificant amount of water from the surrounding and retain the water within itself. That property will provide additional curing water for concrete from inside to outside. Therefore, it is a good resource to cast internal curing concrete. Instead of replacing the amount of sand by SAP, we added SAP according to the weight of cement. Because the effect of SAP is really strong the amount ad ded was smaller compared with Hydrocure and CCA. The SAP used here is Waste Lock 770. It is a solid, granular superabsorbent polymer. This cross linked polyacrylate material swells and absorbs many times its weight in aqueous solutions and its typical abs orptive properties are shown in table. The bulk density of this kind of SAP is 4.5 to 6.5 lbs/gallon. Waste Lock 770 has the ability to both absorb under pressure and to retain absorbed liquids at high pressures.

PAGE 39

25 3.2.4 GFRP bars The GFRP bars use d in this investigation were from Hughes Brother and their diameter is 13mm. The ultimate tensile stress is 689 MPa with a modulus of 40.8 GPa. Its maximum bond stress to concrete is 11.6 MPa. Furthermore, they were able to be used to reject corrosion such as exposing to De icing salts and Marine Salts. 3.3 Description of t est s pecimens The following section provides details of the test specimens. 12 cylinders with GFRP bar were casted for push out tests, 12 cylinders without bar were casted for compression test and 8 beams were casted to get the properties of drying shrinkage, elastic modulus, friction, and rapid chloride ion penetrability were got by casted cylinders. 3.3.1 C ylinders casted with GFRP bars s pecimens. To study the inter facial behavior of the composite reinforcement when contacted with th e internal curing concrete, the cylinders (10.16 cm 20.32 cm) were casted with the GFRP, as shown in Figure 3.1(c) Compression test showed that 5.08

PAGE 40

26 cm GFRP failed before 2.54 cm. Thus here the GFRP out of the con crete is 2.54 cm to avoid GFRP failure before rebar being pushed out. Between the concrete and GFRP there is a 15.24 cm long plastic tube which means that the effective bond length is therefore 5.08 cm and GFRP can be pushed o ut easily. As they are internal curing concrete, all the test specimens were cured in the batching plant without putting them plastic sheets that were secured with rubber bands. For Hydrocure and CCA, 12 specimen s were casted respectively. Replacement amount of sand by internal curing agents ranged from 0% to 75% increased by 25% and for each replacement 3 cylinders with GFRP bar were tested. For SAP, there were also 12 specimen were casted, but the added amount was different from Hydrocure and SAP which ranged from 0% of cement to 0.6% of cement increased by 0.2%. 3.3.2 Cylinders casted without GF RP bars s pecimens To get the compressive strength of internal curing concre te when the internal curing agents replacement changed, 12 cylinders without GFRP bars with size of 4 8 inches were cast ed shown as Figure 3.1(b) w hen the internal curing agent is LWA which means for each replacement ratio from 0% to 75% increased by 25% 3 cylinders were casted. Since 0% replacement LWA can be the control group when the internal curing agents were CCA and SAP, only 9 cylinders needed to be cast for each

PAGE 41

27 one of them. 3.3.3 Beam casted for friction tests and frequency tests Besides cylinders, beams whose size s are 3 4 16 inches also needed to be casted. Same as cylinder casted without GFRP bars, 0% replacement concretes can b e the control group, they just needed to be casted one time. Since each specimen was only one beam was made for each different replacement. So, ten beams were casted here to achieve the results of coefficient of friction and frequency. 3.3.4 Beams casted for shrinkage LWA is the main internal curing aggregate to study in the project. For shrinkage, only the material of Hydrocure was considered. Therefore, only 4 beams with size of 4411 inches were casted.

PAGE 42

28 3.3.5 Cylinders for chloride permeability tests Same as shrinkage tests only the internal agent of Hydrocure was studied. 4 cylinders with size of 4 8 inches were c asted and then each of them w as cut to four 4 2 in ese tests. 3.4 Experimental s etup and l oading The following sections explain the testing apparatus. Methods of loading are also detailed. 3.4.1 Experimental s etup for c ylinders with GFRP bars The push out test was performed using the MTS machine with maximum loading capacity of 89 KN together with linear poten tiometer shown as Figure 3.2 (a) Some adjustments were made to MTS machine as it was not designed for push out test. A circle steel plate with the size of 52 inches was settled on the machine base. There was a hole with the diameter of 2 inches in the center of the steel plate to let the GFRP bar be pushed out. To decide the length of the GFRP out of the internal curing concrete the GFRP also was loaded. Compression test showed that two inches GFRP failed before one inch. Thus, here the GFRP out of the concrete is one inch to avoid

PAGE 43

29 GFRP f ail too quickly. As the total bond strength between the internal curing concrete and GFRP bars was very high which may let the one inch GFRP bar fail before it be pushed out, a 15.24 cm long plastic tube was added between the concrete and GFRP bar shown as F i gure 3.2(c) which means that the effective bond length is therefore 5.08 cm and GFRP can be pushed out easily shown as Figure 3.2 (b) The ef fective bond length was at the bottom of the GFRP bar that is because if making the top two inches GFRP bar bond with concrete the concrete around the bar would punch the plastic tube. As the thickness of the tube is about 0.5 inch the punching force was very high which will affect the testin g results significantly. The load (P) and the slip (s) at the free end of G FRP bars anchored in the test cylinders were measured in order to determine a load slip relationship. The monotonically increased load was applied by the MTS testing machine. The load was applied with a rate of 1mm/min and distributed on the GPRP surface. A laboratorial computer was used to collect test datum including load, time and displacement automatically. When the relative displacement between the GFRP bars and the concrete reached one inch which means the GFRP bar was totally pushed out the test was end. 3.4.2 Experimental s etup for c oncrete c ompressive t esting Standard compression tests shown as Figure 3.3 (a) were performed in accordance with ASTM C39 (ASTM, 1996) on thirty 4 8 inches concrete cylinders.

PAGE 44

30 All cylinders with different internal curing replacement or adding were cured 20 days. 3.4.3 Experimental s etup for f riction t esting Figure 3.3(d) shows th e experimental setup for friction testing. The GFRP bar cut to 13.6 inches and weighted 36.1g was connected through a string to a force weighted 35.5g and then put on the surface of the internal curing beam. The string was connected to an electric motor. When the force pulled the GFRP bar slid on the surface of internal curing beam, the acceleration of the motor can be collected by t he laboratorial computer. As the random uncertainties often invalidate the results, each second law, the coefficient of friction between GFRP bars and different types of in ternal curing concrete can be solved easily. 3.4.4 Experimental s etup for d ynamic e lastic m odulus t esting A conditioned beam with size of 3 4 16 inches was placed on an aluminum support pad. An excitation hammer was used to generate impact signals and then be received by an accelerometer receive r as Figure 3.3(e) shown It should be make sure that the two premarked locations of impacting and receiving signals were consistent for all test specimens. The response frequency domain waves traveled along the specimen were recorded by a dedicated data acquisition system. The

PAGE 45

31 dynamic elastic modulus of each specimen was then determined by the following equatio n: in which where Ed is the dynamic elastic modulus in Pa; n is the frequency; M is the mass of the specimen; L, t, and b are the length, thickness, and width of the specimen, respectively; and T is the correction factor dependent on the radius of gyration to the 3.4. 5 Experimental s etup for s hrinkage t esting Figure 3.3( c ) shows the experimental setup for friction testing. This test is based on ASTM C490/C490M 11. 3.4. 6 Experimental s etup for c hloride penetration t esting Figure 3.3( b ) shows the e xperimental setup for test of measuring dying shrinkage of concrete autoclave expansion of Portland cement and potential expansive reactivity of cement aggregate combinations in mortar bars during storage This test follows ASTM C150

PAGE 46

32 (a) (b) (c) Fig ure 3.1 (a) GFRP bar; (b) Cylinder cast for a concrete compressive test; (c) Cylinders cast wi th GFRP bars for push out tests

PAGE 47

33 (a) (b) (c) Fig ure 3.2 Specimen detail for Push out test: (a) Test setup ; (b) S chematic diagram; (c) Plastic tube for unbonding GFRP bar Data acquisition of MTS machine Cylinder bonding with GFRP bar

PAGE 48

34 (a) (b) (c) (d) 28.58cm 10.16cm 10.16cm 10.16cm 7.62cm 40.64cm 13.6cm

PAGE 49

35 (e) Fig ure 3.3 Material characterization with various internal curing agents : (a) c oncrete strength (ASTM C873 10a ; (b) chloride p ermeability ( ASTM C 114 ) ; (c) s hrinkage (ASTM C490/C490M 11) ; (d) c oefficient of friction ( Luigi M Gratton 2006) ; (e) dynamic elastic modulus ( ASTM C215 ) Data acquisition Receiver Specimen Excitation

PAGE 50

36 4 E xperimental results and discussion 4.1 Introduction In this chapter, the results from the push out tests, friction tests, compressive strength tests, drying shrinkage test, elastic modulus tests, and rapid chloride ion penetrability tests are presented. The failure mode and fracture energy of internal curi ng concrete casted with three different types of internal curing agents together with GFRP are discussed in detail. The interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete is evaluated in te rms of load displacement response. The coefficient of friction between different kinds of internal curing concrete with GFRP bars are established from a simple measurement of sliding friction coefficient. The mechanical properties of compressive strength a re built from the standard concrete cylinder tests. Furthermore, some other properties like shrinkage, elastic modulus and chloride ion penetrability are also investigated based on the ASTM standards. The primary objective of the experimental program was t o determine the optimal replacement or adding amount of the three kinds of internal curing agents based on the results of push out test, compressive strength tests and the analysis of fracture energy.

PAGE 51

37 4. 2 Compressive s tress t est r esults Thirty cylinders were tested in compressive stress as discussed in chapter 3. When internal curing agent was Hydrocure three cylinders were tested for each type of replacement ratio from 0% to 75%. Since 0% replacement internal curing concrete is the control spec imen, when the internal curing aggregate is CCA or SAP 0% replacement amount or adding amount is not included any more. Therefore, for CCA, three cylinders were tested for each replacement amount from 25% to 75% increased by 25%. And for SAP, three cylinde rs were tested for the adding amount from 0.2% to 0.6% of cement weight increased by 0.2%. 4. 2 .1 Compressive s tress t est r esults for Hydrocure The compression strength of internal curing concrete with different percent o f Hydrocure is shown as Fi gure 4.1(a) And the Table 4.1 shows th e datum results. The average compressive strength tended to decrease with the adding replacement amount of Hydrocure and the inclusion of Hydrocure actually degrades the concrete performance.

PAGE 52

38 4. 2 .2 Compressive s tress t est r esults for CCA The compression strength of internal curing concrete with different per cent of CCA is shown as Figure 4.1(a) And the Table 4. 2 shows the d atum results. The average compressive strength tended to decrease with the adding replacem ent amount of CCA and the inclusion of CCA actually degrades the concrete performance. By comparing the results of CCA and Hydrocure in Figure 6 compressive strength of Hydrocure is lower than CCA 4. 2 .3 Compressive s tress t est r esults for SAP The compression strength of internal curing concrete with different percent of SAP is shown as Figure 4.1(b) And the Table 4. 3 show s the datum results. The average compressive strength tended to decrease with the adding replacement amount of SAP and the inclu sion of SAP actually degrades the concrete performance. To compare the effect of SAP and Hydrocure, the equations of them shown in Figure 6 is assumed equal which means = where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 113.56*a+3.32. And that means that adding 0.1% SAP has the similar effect of replacing 14.68% Hydrocure for compressive strength Obv iously, the effect of SAP is significant stronger than Hydrocure.

PAGE 53

39 4. 3 Chloride i on p enetrability t ests Only concrete casted with Hydrocure agents tested here. 4. 3 .1 Chloride i on p enetrability t ests re sults for Hydrocure The test results of Hydrocure are shown in Figure 4.2. The coulombs increased with the adding of Hydrocure replacement. 4. 4 Elastic m odulus t ests The section presents the frequency tests results and the calculated elastic modulus results. 4. 4 .1 Elastic m odulus t ests r esults for Hydrocure Figure 4.3 (a) shows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for Hydrocure. And Figure 4.4(a) sh ows the calculated dynamic elastic results. It shows that dynamic elastic decreased with the increasing of replacement ratio for Hydrocure.

PAGE 54

40 4. 4 .2 Elastic m odulus t ests r esults for CCA Figure 4.1(a) sh ows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for CCA. And Figure 4.4(a) show s the calculated dynamic elastic results. It shows that dynamic elastic decreased with the incr easing of replacement ratio for Hydrocure. Furth ermore, the dynamic elastic of Hydrocure is lower than CCA. 4. 4 .3 Elastic m odulus t ests r esults for SAP Figure 4.3(b) shows the tests results of frequency. It is evident that frequency decreased with the increasing of replacement for SAP. And Figure 4.4(b) shows the calculated dynamic elastic results. It shows that dynamic elastic decreased with the increasing of adding ratio for SAP. To compare the effect of SAP and Hydrocure, the equ ations of them shown in Figure 7 is assumed equal which means = where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 272 *a+ 21.84 And that means that adding 0.1% SAP has the s imilar effect of replacing 49.04 % Hydrocure for compressive strength Obvio usly, the effect of SAP is significant stronger than Hydrocure.

PAGE 55

41 4. 5 Friction t est r esults Ten specimens were tested, and for each specimen twelve tests were implemented to get the results. This section presents the friction tests results of between internal curing concrete and GFRP bar was determined for each type of internal curing concrete. 4. 5 .1 Friction t est r esults for Hydrocure Acceleration time plots for Hydrocure, along with photos of test progress, are shown in Figures 4.5 Figure 4.5 presen ts the friction coefficient replacement plot when the internal curing agent was Hydrocure. As the figure shown, the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is 36.1 grams. Therefore, the equation established to calculate the friction coefficient should be calculated by the equation of a = (35.5 g / (36.1+35.5). And the calculated results are shown in figure 12. It is evident that addi ng the replacement amount of Hydrocure the acceleration of the sliding GFRP bar will decrease lightly. Furthermore, friction coefficient will increase slightly when the replacement amount of Hydrocure is increasing

PAGE 56

42 4. 5 .2 Friction t est r esults for CCA Figure 4.6 presents the friction coefficient replacement plot when the internal curing agent was CCA. Same as the friction tests for Hydrocure the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is 36.1 grams. Therefore, the equation establis hed to calculate the friction coefficient should be calculated by the equation of a = (35.5 g / (36.1+35.5). And the c alculated results are shown in F igure 4.8(a) It is eviden t that adding the replacement amount of CCA the acceleration of the slidi ng GFRP bar will decrease lightly when the replacement ratio was from 25% to 75% and the results are different from 0% replacement to 25% replacement. Furthermore, friction coefficient will increase slightly when the replacement amount of CCA is increasing By comparing the results of CCA and Hydrocure in Figure 12 coefficient of Hydrocure is a little bit higher than CCA 4. 5 .3 Friction t est r esults for SAP Figure 4.7 presents th e friction coefficient replacement plot when the internal curing agent was SAP. Same as the friction tests for Hydrocure the load pulled GFRP bar is 35.5 grams and the weight of GFRP bar is 36.1 grams. Therefore, the equation established to calculate the friction coefficient should be calculated by the equation of a = (35.5 ) g / (36.1+35.5). And the c alculated results are shown in F igure 4.8(b) It is evident that adding the replacement amount of SAP the

PAGE 57

43 acceleration of the sliding GFRP bar will decrease lightly when the adding ratio was from 0.2% to 0.6% and the results are opposite of 0% replacement to 0.2% adding. Furthermore, friction coefficient will increase slightly when the replacement amount of SAP is increasing. 4. 6 Push out t est r esults The following sections show the results of the experimental program. The results of push out tests on thirty cylinders casted with GFRP bars are included. Interfacial capacity, failure mode, and fracture energy is discussed for each specimen, and the optim al internal curing agents replacement amount or adding amount is established from the experimental results. To make sure that the displacement results are tested by MTS machine are accurate linear potentiometer was also included in the tests Figure 4.9 s how s the comparison between the two test methods and the results of them are significant close which means the test results of MTS are exact enough to be used. 4. 6 .1 Push out t ests r esults for Hydrocure Figure 4.10 s how s the interfacial load (P) versus displacement at a

PAGE 58

44 specific amount of light weight aggregate. The displacement when the GFRP bar reached the m aximum load is shown i n Table 4.4. Table 4.5 and Table 4. 6 show th e maximum interfacial load of bonding strength a nd maximum stress respectively It is postulated that the Hydrocure does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessary to generalize this preliminary conclusion. The first crack appeared at the bottom of the concrete (shown in F igure 4.1 1 (a)). I t moved upward slowly as shown in F igure 4.11 ( b ) until it reached about 5.08 cm which was close to the plastic tube. Then it started to move horizontally and merged with the other 2 cracks as it shown in F igure 4.11 ( c ) F igure 4.11 ( d ) shows that the angle between the 3 cracks was about 120 degrees. 4. 6 .2 Push out t ests r esults for CCA Figure 4.1 2 show s the interfacial load (P) versus displacement at a specific amount of light weight aggregate. The displacement when the GFRP bar reached the maximum load is shown i n Table 4.7. Table 4.8 and Table 4.9 show the maximum interfacial load and maximum stress re spective ly By compare the pu sh out test results as Figure 4.14 shown, the maximum load of CCA push out test is bigger than

PAGE 59

45 the maximum load of Hydrocure. But both of the two maxim um loads decreased with the increasing internal curing agent replacement. It is also postulated that the CCA does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessary to generalize this preliminary conclusion. For the cracking modes, cylinders casted with CCA share the similar cracking procedures with concrete casted with Hydrocure. 4. 6 .3 Push out t ests r esults for SAP Figure 4.13 show s the interfacial load (P) versus displacement at a specific amount of light weight aggregate. The displacement when the GFRP bar reached the maximum load is shown i n Table 4.10. Table 4.11 and Table 4. 12 show the maximum interfacial load and maximum stress respectively Just as the previous tw o types of internal curing agents the maximum push out load of cylinders casted with GFRP bar when internal curing agent is SAP also decreased with the increasing replacement of SAP. To compare the effect of SAP and Hydrocure, the equations of them shown i n F igure 4.14 are assumed equal which means = where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 100.15*a+6.42. And that means

PAGE 60

46 that adding 0.1% SAP has the simi lar effect of replacing 16.435% Hydrocure. Obviously, the effect of SAP is significant stronger than Hydrocure. Similarly, it is also postulated that the SAP does not have sufficient adhesion capability when associated with Portland cement so that such unfavorable bond behavior was noticed. This finding indicates that adequate use of the internal curing agent may not benefit the performance of concrete members; however, an excessive dosage appears to be detrimental. Further examinations are necessa ry to generalize this preliminary conclusion. For the cracking modes, cylinders casted with CCA share the similar cracking procedures with concrete casted with Hydrocure. 4. 7 The f racture e nergy a nalysis Based on the push out test this section presents the fracture energy analysis for the three types of internal curing concrete casted with GFRP bars respectively. 4. 7 .1 The f racture e nergy a nalysis for Hydrocure F igure 4.15 shows the interfacial load (P) versus displacement from the load is zero to the maximum value of each specimen. Table 4.1 3 presents the energy dissipation based on load displacement of the test cylinders casted wit h GFRP bars. It

PAGE 61

47 is gotten from the area u nder the load and displacement curve of each specimen which is calcu lated by the equations shown in F igure 4.15 And the results show that the energy dissipation tends to increase with the increasing replacement amount as the F igure 4.15 ( e ) shown. F igure 4 .19 shows the shear versus displacement from the shear is zero to the maximum value of each specimen. Table 4 .1 4 summarizes the fracture energy based on shear stress displacement of the test specimens, which was obtained from the area under the interfacial stress and displacement curve of each specimen and calculated by the quadratic equations shown in F igure 4.1 9 The trend found is that the measured fracture energy decreased with an increasing replaceme nt ratio. A significant decrease in fracture energy was noticed when the replacement ratio was greater than 50%, while the level of standard deviation tended to decrease. This points out that the failure zone of the test specimen became localized (i.e., re duced variation of the fracture energy). F igure 4.23(a) and (b) shows the energy dissipation displacement and fracture displacement of internal curing concrete mixed with Hydrocure respectively. 4. 7 .2 The f racture e nergy a nalysis for CCA F igure 4.16 shows the interfacial load (P) versus displacement from the load is zero to the maximum value of each specimen. Table 4.15 pr esents the energy

PAGE 62

48 dissipation based on load displacement of the test cylinders casted with GFRP bars. It is gotten from the area under the load and displacement curve of each specimen which is calculated by the equations shown in F igure 4.16 And the results show that the energy dissipation tends to increase with the increasing replacement amount as the F igur e 4.16 ( e ) shown. F igure 4.18 shows the energy dissipation comparison of CCA and Hydrocure and as the graph shown the energy dissipation of CCA is higher than Hydrocure. F igure 4.20 shows the shear versus displacement from the shear is zero to the maximum v alue of each specimen. Table 4.1 6 summarizes the fracture energy based on shear stress displacement of the test specimens, which was obtained from the area under the interfacial stress and displacement curve of each specimen and calculated by the quadratic equations shown in F igure 4.20 The trend found is that the measured fracture energy decreased with an increasing replacement ratio. F igure 4.22 shows the fractural energy comparison of CCA and Hydrocure and as the graph shown the fractural energy of CCA is higher than Hydrocure. F igure 4.24(a) and (b) shows the energy dissipation displacement and fracture displacement of internal curing concrete mixed with CCA respectively. 4. 7 .3 The f racture e nergy a nalysis for SAP F igure 4.17 shows the interfacial load (P) versus displacement from the load is

PAGE 63

49 zero to the maximum value of each specimen. Table 4.1 7 presents the energy dissipation based on load displacement of the test cylinders casted with GFRP bars. It is gotten from the area u nder the load and displacement curve of each specimen which is calculated by the equations shown in F igure 4.17 And the results show that the energy dissipation tends to increase with the increasing replacement amount as the F igure 4.17 ( e ) shown. To compa re the effect of SAP and Hydrocure, the equations of them shown in Figure 21 is assumed equal which means = where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equat ion, r= 84.62*a+7.97. And that means that adding 0.1% SAP has th e similar effect of replacing 16.432% Hydrocure for energy dissipation F igure 4.21 shows the shear versus displacement from the shear is zero to the maximum value of each specimen. Table 4.1 8 summarizes the fracture energy based on shear stress displacement of the test specimens, which was obtained from the area under the interfacial stress and di splacement curve of each specimen and calculated by the quadratic equations shown in F igure 4.21 The trend found is that the measured fracture energy decreased with an increasing replacement ratio. To compare the effect of SAP and Hydrocure, the equations of them shown in F igure 4.22 is assumed equal which means = where r represents the replacement amount of Hydrocure and a represents adding amount of SAP. By solving the equation, r= 93.53*a+7.84. And that means that adding 0.1% SAP has the similar effect of replacing 17.19% Hydrocure for fracture energy

PAGE 64

50 Obviously the effect of SAP is significant stronger than Hydrocure. F igure 4.25(a) and (b) shows the energy dissipation displacement and fracture displacement of internal curing concrete mixed with SAP respectively. 4. 8 Absor ptance t est Internal curing method is a method that let internal curing agents serve as a source of water supply inside the concrete and facilitate curing process. Therefore the absor ptance of water for internal curing agents is a very important property for this research. This secti on present s the test results of the three different internal curing agents. 4. 8 .1 Absor ptance t est r esults for Hydrocure Figure 4.26 (a) shows the absorptance of Hydrocure agents. 4. 8 .2 Absor ptance t est r esults for CCA Figure 4.26 (a) shows the absorptance of CCA agents. By the comparison of Hydrocure, the result is the absorptance of CCA is higher than Hydrocure.

PAGE 65

51 4. 8 .3 Absor ptance t est r esults for SAP Figure 4.26 (b ) shows the absorptance of SAP agents. By the comparison of SAP the result is the absorptance of SAP is much higher than Hydrocure and CCA. That is assumed be the reason why the effect of small amount SAP adding is equivalent to large replacement amount of Hydrocure and CCA. 4. 9 Dry shrinkage t est Ten 4 4 11 inches beams were tested. 4. 9 1 Dry shrinkage t est r esults for Hydrocure Figure 4.27(a) shows the test results for the shrinkage verse data curve for Hydrocure and Figure 4.27(b) shows the shrinkage divided by depth of the beam for Hydrocure.

PAGE 66

52 4. 9 2 Dry shrinkage t est r esults for CCA Figure 4.2 8 (a) shows the test results for the shrinkage verse data curve for CCA and Figure 4.28 (b) shows the shrinkage divided by depth of the beam for CCA 4. 8 .3 Dry shrinkage t est r esults for SAP Figure 4.2 9 (a) shows the test results for the shrinkage verse data curve for SAP and Figure 4.29(b) shows the shrinkage divided by depth of the beam for SAP.

PAGE 67

53 Table 4.1 Compressive strength (Hydrocure) Replacement Compressive strength(MP a) (%) individual averge stdev cov 0% 1 46.8 40.6 5.5 0.14 0% 2 38.5 0% 3 36.4 25% 1 37.8 32.4 9.3 0.29 25% 2 37.8 25% 3 21.7 50% 1 23.8 15.1 7.6 0.50 50% 2 9.7 50% 3 11.8 75% 1 12.1 11.6 3.2 0.28 75% 2 14.6 75% 3 8.2

PAGE 68

54 Table 4.2 Compressive strength (CCA) Replacement Compressive strength(MP a) (%) individual averge stdev cov 0% 1 46.8 40.6 5.5 0.14 0% 2 38.5 0% 3 36.4 25% 1 40.6 39.3 1.8 0.05 25% 2 37.3 25% 3 40 .0 50% 1 26 .0 29.9 3.5 0.12 50% 2 32.8 50% 3 30.9 75% 1 35.3 25.3 11.9 0.47 75% 2 28.5 75% 3 12.1

PAGE 69

55 Table 4.3 Compressive strength (SAP) Replacement Compressive strength(MPa) (%) individual averge stdev cov 0% 1 46.8 40.6 5.5 0.14 0% 2 38.5 0% 3 36.4 0.2% 1 23.9 24 0.3 0.01 0.2% 2 23.7 0.2% 3 24.3 0.4% 1 26.2 21.1 4.7 0.22 0.4% 2 16.8 0.4% 3 20.4 0.6% 1 10.4 11 2.4 0.22 0.6% 2 13.6 0.6% 3 8.9

PAGE 70

56 Table 4. 4 Displacement at maximum load (Hydrocure) Replacement (%) Displacement at maximum load individual averge stdev cov 0% 1 1.6 2.1 0.5 0.24 0% 2 2.3 0% 3 2.6 25% 1 1.1 1.5 0.6 0.41 25% 2 1.1 25% 3 2.2 50% 1 1.2 1.4 0.2 0.14 50% 2 1.5 50% 3 1.5 75% 1 1.0 1.2 0.3 0.28 75% 2 1.6 75% 3 1.0

PAGE 71

57 Table 4.5 Maximum interfacial load of bonding strength (Hydrocure) Replacement (%) Max Load (kN) individual averge stdev cov 0% 1 22.6 25.7 7.0 0.27 0% 2 33.7 0% 3 20.7 25% 1 19.0 19.4 0.7 0.03 25% 2 19.1 25% 3 20.2 50% 1 12.8 12.7 0.9 0.07 50% 2 11.6 50% 3 13.5 75% 1 8.4 9.7 1.2 0.12 75% 2 10.1 75% 3 10.7

PAGE 72

58 Table4.6 Maximum stress (Hydrocure) Replacement (%) Max stress (N/mm ) individual averge stdev cov 0% 1 10.8 12.3 3.4 0.27 0% 2 16.1 0% 3 9.9 25% 1 9.1 9.3 0.3 0.03 25% 2 9.2 25% 3 9.7 50% 1 6.1 6.1 0.5 0.07 50% 2 5.6 50% 3 6.5 75% 1 4.0 4.6 0.6 0.12 75% 2 4.8 75% 3 5.1

PAGE 73

59 Table 4.7 Displacement at maximum load (CCA) Replacement (%) Displacement at maximum load individual averge stdev cov 0% 1 1.6 1.8 0.4 0.23 0% 2 2.3 0% 3 1.6 25% 1 1.9 2.0 0.3 0.16 25% 2 2.4 25% 3 1.8 50% 1 1.9 2.0 0.5 0.25 50% 2 2.5 50% 3 1.5 75% 1 2.8 2.4 0.5 0.22 75% 2 2.7 75% 3 1.8

PAGE 74

60 Table 4.8 Maximum interfacial load of bonding strength (CCA) Replacement (%) Max load (kN) individual average stdev cov 0% 1 22.6 25.7 7.0 0.27 0% 2 33.7 0% 3 20.7 25% 1 19.1 18.9 0.3 0.02 25% 2 19.0 25% 3 18.5 50% 1 11.6 13.1 1.4 0.11 50% 2 14.4 50% 3 13.2 75% 1 16.1 17.2 1.1 0.06 75% 2 17.2 75% 3 18.2

PAGE 75

61 Table 4. 9 Maximum stress (CCA) Replacement (%) Max stress (N/mm ) individual average stdev cov 0% 1 10.8 12.3 3.4 0. 27 0% 2 16.1 0% 3 9.9 25% 1 9.1 9.0 0.2 0.02 25% 2 9.1 25% 3 8.8 50% 1 7.7 8.2 0. 5 0 .06 50% 2 8.2 50% 3 8.7 75% 1 5.6 6.3 0. 7 0. 11 75% 2 6.9 75% 3 6.3

PAGE 76

62 Table 4. 10 Displacement at maximum load (SAP) Replacement (%) Displacement at maximum load individual averge stdev cov 0% 1 1.6 1.8 0.4 0.23 0% 2 2.3 0% 3 1.6 0.2% 1 1.8 1.5 0.4 0.24 0.2% 2 1.1 0.2% 3 1.6 0.4% 1 1.6 1.6 0.1 0.07 0.4% 2 1.7 0.4% 3 1.5 0.6% 1 1.8 1.5 0.2 0.16 0.6% 2 1.4 0.6% 3 1.4

PAGE 77

63 Table 4.11 Maximum interfacial load of bonding strength (SAP) Replacement (%) Max load (kN) individual average stdev cov 0% 1 22.6 25.7 7.0 0.27 0% 2 33.7 0% 3 20.7 0.2% 1 17.3 16.2 1.5 0.09 0.2% 2 14.5 0.2% 3 16.8 0.4% 1 12.9 14.8 1.8 0.12 0.4% 2 16.6 0.4% 3 14.8 0.6% 1 10.4 10.9 1.3 0.12 0.6% 2 12.4 0.6% 3 10.0

PAGE 78

64 Table 4. 12 Maximum stress (SAP) Replacement (%) Max stress (kN /mm ) individual average stdev cov 0% 1 10.8 12.3 3.7 0.2 7 0% 2 16.1 0% 3 9.9 0.2% 1 8.3 7.7 0.7 0.09 0.2% 2 6.9 0.2% 3 8.0 0.4% 1 6.2 7.1 0.9 0.12 0.4% 2 7.9 0.4% 3 7.1 0.6% 1 5.0 5.2 0.6 0.12 0.6% 2 5.9 0.6% 3 4.8

PAGE 79

65 Table 4.13 Energy dissipation based on load displacement ( Hydrocure) Replacement (%) Fracture energy (kNmm) individual averge stdev cov 0% 1 16.2 19.7 8.4 0.43 0% 2 29.3 0% 3 13.6 25% 1 10.2 14.5 7.9 0.54 25% 2 9.8 25% 3 23.6 50% 1 7.1 6.8 0.6 0.09 50% 2 7.1 50% 3 6.1 75% 1 4.3 4.3 1.0 0.22 75% 2 3.3 75% 3 5.2

PAGE 80

66 Table 4.14 Fracture energy based on shear displacement (Hydrocure) Replacement (%) Fracture energy (N/mm) individual averge stdev cov 0% 1 7.8 9.5 4.0 0.43 0% 2 14.1 0% 3 6.6 25% 1 4.9 7.0 3.8 0.54 25% 2 4.7 25% 3 11.4 50% 1 3.4 3.3 0.3 0.09 50% 2 3.4 50% 3 2.9 75% 1 2.1 2.3 0.2 0.10 75% 2 2.4 75% 3 2.5

PAGE 81

67 Table 4.1 5 Energy dissipation based on load displacement (CCA) Replacement (%) Energy dissipation (kNmm) individual averge stdev cov 0% 1 16.2 19.7 8.4 0.43 0% 2 29.3 0% 3 13.6 25% 1 11.9 13.6 1.8 0.13 25% 2 15.5 25% 3 13.4 50% 1 9.9 11.3 3.4 0.30 50% 2 15.2 50% 3 8.8 75% 1 18.8 15.5 2.8 0.18 75% 2 13.8 75% 3 14.0

PAGE 82

68 Table 4.1 6 Fracture energy based on shear displacement (CCA) Replacement (%) Fractur e energy (N/mm) individual averge stdev cov 0% 1 7.8 9.4 4.0 0.43 0% 2 14.0 0% 3 6.5 25% 1 5.7 6.5 0.9 0.13 25% 2 7.4 25% 3 6.4 50% 1 4.7 5.4 1.6 0.30 50% 2 7.3 50% 3 4.2 75% 1 9.0 7.4 1.4 0.18 75% 2 6.6 75% 3 6.7

PAGE 83

69 Table 4.17 Energy dissipation based on load displacement (SAP) Replacement (%) Energy dissipation (kNmm) individual averge stdev cov 0% 1 16.2 19.7 8.4 0.43 0% 2 29.3 0% 3 13.6 0.2% 1 10.7 9.7 1.6 0.17 0.2% 2 7.9 0.2% 3 10.7 0.4% 1 7.1 9.0 1.7 0.19 0.4% 2 10.5 0.4% 3 9.4 0.6% 1 5.4 6.0 1.8 0.30 0.6% 2 8.1 0.6% 3 4.6

PAGE 84

70 Table 4.1 8 Fracture energy based on shear displacement (SAP) Replacement (%) Fracture energy (N/mm) individual averge stdev cov 0% 1 7.76 9.42 4.01 0.43 0% 2 14.00 0% 3 6.51 0.2% 1 5.10 4.66 0.78 0.17 0.2% 2 3.76 0.2% 3 5.12 0.4% 1 3.41 4.31 0.83 0.19 0.4% 2 5.04 0.4% 3 4.48 0.6% 1 2.58 2.88 0.87 0.30 0.6% 2 3.85 0.6% 3 2.20

PAGE 85

7 1 (a) (b ) Figure 4.1 Variation of concrete strength with internal curing agents: (a) Hydrocure and CCA; (b ) SAP y = 42.321e 0.018x R = 0.7512 y = 43.315e 0.008x R = 0.4314 0 10 20 30 40 50 0 10 20 30 40 50 60 70 80 Compressive strength (MPa) Replacement (%) Hydrocure CCA Hydrocure y = 39.867e 2.044x R = 0.8759 0 10 20 30 40 50 0 0.2 0.4 0.6 Compressive strength (MPa) Adding (%)

PAGE 86

72 (a) Figure 4.2 Chloride pe netration for concrete with Hydrocure y = 3825.5e 0.0044x R = 0.1628 0 1000 2000 3000 4000 5000 6000 7000 0 10 20 30 40 50 60 70 80 Coulombs Replacement (%)

PAGE 87

73 (a) (b) (c) Fig ure 4.3 Frequency and dynamic modulus of concrete with internal curing agents: (a) Hydrocure; (b) CCA; (c) SAP y = 2.8125x + 1921.9 R = 0.9818 0 500 1000 1500 2000 0 10 20 30 40 50 60 70 80 Frequency (Hz) Replacement (%) Test1 Test2 Test3 Test4 Test5 y = 1808.1e 0.618x R = 0.8352 0 500 1000 1500 2000 0 0.2 0.4 0.6 Frequency (Hz) Replacement (%) Test1 Test2 Test3 Test4 Test5 y = 1919.1e 0.002x R = 0.9689 0 500 1000 1500 2000 0 10 20 30 40 50 60 70 80 Frequency(Hz) Replacement (%) Test1 Test2 Test3 Test4 Test5

PAGE 88

74 (a ) (b ) Fig ure 4.4 Dynamic modulus (a ) com parison of Hydrocure and CCA; (b ) SAP y = 39.511e 0.005x R = 0.9834 y = 33.552e 0.007x R = 0.5471 0 10 20 30 40 50 0 10 20 30 40 50 60 70 80 Dynamic elastic modulus (GPa) Replacement (%) Hydrocure CCA y = 35.424e 1.36x R = 0.8691 0 10 20 30 40 50 0 0.2 0.4 0.6 Dynamic elastic modulus (GPa) Adding (%) SAP

PAGE 89

75 (a) (b) (c) (d) 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12

PAGE 90

76 (e) Fig ure.4.5 Acceleration between concrete with Hydrocure and GFRP bar measured from the friction test : (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary average line y = 1.8501e 0.001x R = 0.1672 0 1 2 3 0 10 20 30 40 50 60 70 80 Acceleration(m/s) Replacement (%)

PAGE 91

77 (a) (b) (b) (d) 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average

PAGE 92

78 (e) Fig ure 4.6 Acceleration between concrete with CCA and GFRP bar measured from the friction test: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement; (e) summary average line y = 6E 06x 3 0.0008x 2 + 0.0273x + 1.8417 R = 0.5288 0 1 2 3 0 10 20 30 40 50 60 70 80 Acceleration(m/s) Replacement (%)

PAGE 93

79 (a) (b) (c) (d) 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 0 1 2 3 0 0.5 1 1.5 2 2.5 3 Acceleration (m/s) Time (s) Run1 Run2 Run3 Run4 Run5 Run6 Run7 Run8 Run9 Run10 Run11 Run12 average

PAGE 94

80 (e) Fig ure 4.7 Acceleration between concrete with SAP and GFRP bar measured from the friction test: (a) 0% adding ; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding; (e) summary average line y = 7.7589x 3 8.9157x 2 + 2.5365x + 1.8417 R = 0.1886 0 1 2 3 0 0.2 0.4 0.6 Acceleration(m/s) Adding (%)

PAGE 95

81 (a) (b) Fig ure 4.8 Coefficient of friction: (a) Hydrocure and CCA ; (b ) SAP y = 0.6063e 0.0009x R = 0.1731 y = 0.5803e 0.0013x R = 0.2034 0 0.2 0.4 0.6 0.8 0 10 20 30 40 50 60 70 80 Coefficient of friction Replacement (%) Hydrocure CCA Hydrocure (Hydrocure) y = 1.5703x 3 + 1.8044x 2 0.5134x + 0.6107 R = 0.1886 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 Coefficient of friction Adding (%)

PAGE 96

82 (a) (b) (c) (d) Fig ure 4.9 Comparison between MTS and linear potentiom e ter: (a) 0% replacement; (b) 25% replacement; (c) 50% replacement; (d) 75% replacement 0 5 10 15 20 25 30 35 0 5 10 15 20 Load (kN) Displacement (mm) 25%LP 25% 0 5 10 15 20 25 30 35 0 5 10 15 20 Load (kN) Displacement (mm) 0% LP 0% 0 5 10 15 20 25 30 35 0 10 20 Load (kN) Displacement (mm) 50%LP 50% 0 5 10 15 20 25 30 35 0 5 10 15 20 Load (kN) Displacement (mm) 75%LP 75%

PAGE 97

83 (a) (b) (a) (c) (d) (e) Fig ure 4.10 Load versus displacement response of push out tests with various concrete mixtures with Hydrocure (a) 0% replacement; (b) 25% replacement; (c)50% replac ement; (d) 75% replacement; (e) s ummary 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0% 1 0% 2 0% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 25% 1 25% 2 25% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 50% 1 50% 2 50% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 75% 1 75% 2 75% 3 y = 25.594e 0.013x R = 0.8907 0 10 20 30 40 0 20 40 60 80 Maximum load (kN) Replacement (%)

PAGE 98

84 (a) ( (a) (b) (c) (d) Fig ure 4.11 Push out test failure mode: (a) cracking appeared; (b) cracking moved upward. (c) failure mode 1; (d) failure mode 2

PAGE 99

85 (a) (b ) (c) (d) Fig ure 4.12 Load versus displacement response of push out tests with various concrete mixtures with CCA (a) 0% replacement; (b) 25% replacement; (c)50% replac ement; (d) 75% replacement; (e) s ummary 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 50% 1 50% 2 50% 3 y = 23.645e 0.007x R = 0.6185 0 10 20 30 40 0 20 40 60 80 Max Load (kN) Replacement (%) 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0% 1 0% 2 0% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 25% 1 25% 2 25% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 75% 1 75% 2 75% 3

PAGE 100

86 (a) (b) (c) (d) (e) Fig ure 4.13 Load versus displacement response of push out tests with various concrete mixtures with SAP (a) 0% adding; (b) 0.2% adding; (c) 0.4% adding; (d) 0.6% adding; (e) summary 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0.2% 1 0.2% 2 0.2% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0.4% 1 0.4% 2 0.4% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0.6% 1 0.6% 2 0.6% 3 0 5 10 15 20 25 30 0 10 20 30 Load (kN) Displacement (mm) 0% 1 0% 2 0% 3 y = 23.545e 1.302x R = 0.7888 0 10 20 30 40 0 0.2 0.4 0.6 Load (kN) Adding (%)

PAGE 101

87 Fig ure 4.14 Load versus displacement response comparison with different internal curing agents y = 25.594e 0.013x R = 0.8907 y = 23.645e 0.007x R = 0.6185 0 10 20 30 40 0 20 40 60 80 Maximum load (kN) Replacement (%) Hydrocure CCA y = 23.545e 1.302x R = 0.7888 0 10 20 30 40 0 0.2 0.4 0.6 Load (kN) Adding(%)

PAGE 102

88 (a) (b) (c) y = 3.2513x 2 + 11.154x 1.068 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 1 y = 4.9086x 2 + 5.1104x 1.7471 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 2 y = 2.7823x 2 + 14.94x 0.705 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 1 y = 4.2251x 2 + 14.227x 1.159 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 2 y = 2.6118x 2 + 15.899x 2.4937 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 3 y = 4.6201x 2 + 7.9771x 0.7001 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 1 y = 5.5099x 2 + 1.2581x 0.2734 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 2 y = 5.9521x 2 0.9261x + 0.0511 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 3 y = 4.1567x 2 + 8.5836x 1.4057 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 3 A A A A A A A A A

PAGE 103

89 (d) (e) Fig ure 4.15 The e nergy dissipation (area A ) of push out specimens with Hydrocure : (a) 0% replacemen t ; (b) 25% replacement ; (c) 50% replacement; (d) 75% replacement ; (e) s ummary y = 19.805e 0.021x R = 0.7988 0 10 20 30 0 20 40 60 80 Energy dissipation (kNmm) Replacement (%) y = 1.0434x 2 + 7.9134x 0.0073 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 1 y = 2.1923x 2 + 9.3391x 0.1747 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 3 y = 4.3509x 2 + 4.4381x 0.3912 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 2 A A A

PAGE 104

90 (a) (b) (c) y = 6.7253x 2 2.222x + 0.0592 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 1 y = 4.1757x 2 1.4196x 0.0437 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 2 y = 0.1997x 2 + 8.9704x 1.7381 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 1 y = 2.0156x 2 + 1.7468x 0.0567 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 2 y = 2.6536x 2 + 6.3539x 0.7534 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 50% 3 y = 3.2513x 2 + 11.154x 1.068 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 1 y = 4.9086x 2 + 5.1104x 1.7471 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 2 y = 4.1567x 2 + 8.5836x 1.4057 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 3 y = 4.2918x 2 + 3.4903x 0.2377 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 25% 3

PAGE 105

91 (d) (e) Fig ure 4.16 The e nergy dissipation(area A) of push out specime ns with CCA: (a) 0% replacement; (b) 25% replacement ; (c) 50% replaceme nt; (d) 75% replacement ; (e) s ummary y = 3.469x 2 2.6994x + 0.6385 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 1 y = 1.8704x 2 + 6.1487x 0.4029 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 2 y = 1.7708x 2 + 15.715x 0.9728 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 75% 3 y = 17.713e 0.008x R = 0.423 0 10 20 30 0 20 40 60 80 Energy dissipation (kNmm) Replacement (%)

PAGE 106

92 ( a) (b) (c) y = 6.4124x 2 0.5947x 0.1459 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.2% 1 y = 5.9353x 2 0.852x + 0.0893 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.4% 1 y = 5.2846x 2 + 1.1139x 0.0408 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.4% 2 y = 4.3195x 2 + 4.4682x 0.224 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.4% 3 y = 3.2513x 2 + 11.154x 1.068 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 1 y = 4.9086x 2 + 5.1104x 1.7471 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 2 y = 4.1567x 2 + 8.5836x 1.4057 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0% 3 y = 4.5699x 2 + 4.088x 0.659 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.2% 3 y = 1.9024x 2 + 11.344x + 0.2609 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.2% 2

PAGE 107

93 (d) (e) Fig ure 4.17 The e nergy dissipation(area A) of push out specimen s with SAP: (a) 0% adding; (b) 25% adding ; (c) 50% adding; (d) 75% adding ; (e) s ummary y = 4.7897x 2 3.3289x + 0.6756 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.6% 1 y = 16.754e 1.777x R = 0.7032 0 10 20 30 0 0.2 0.4 0.6 Energy dissipation (kNmm) Adding(%) y = 6.7249x 2 2.202x + 0.3848 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.6% 3 y = 2.0335x 2 + 6.9429x 0.2815 0 5 10 15 20 25 30 0 1 2 3 Load (kN) Displacement (mm) 0.6% 2

PAGE 108

94 Fig ure 4.18 The e nergy dissipation comparison of different internal curing agents y = 17.713e 0.008x R = 0.423 y = 19.805e 0.021x R = 0.7988 0 10 20 30 0 20 40 60 80 Energy dissipation (kNmm) Replacement (%) CCA Hydrocure y = 16.754e 1.777x R = 0.7032 0 10 20 30 0 0.2 0.4 0.6 Energy dissipation (kNmm) Adding(%)

PAGE 109

95 (a) (b) (c) y = 2.3486x 2 + 2.4452x 0.8359 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 2 y = 1.9889x 2 + 4.107x 0.6726 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 3 y = 2.0216x 2 + 6.8072x 0.5545 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 25% 2 y = 1.2497x 2 + 7.6073x 1.1932 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 25% 3 y = 1.5556x 2 + 5.3368x 0.511 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 1 y = 2.2106x 2 + 3.8168x 0.335 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 50% 1 y = 2.6363x 2 + 0.602x 0.1308 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 50% 2 y = 2.9512x 2 0.568x + 0.0481 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 50% 3 y = 1.3058x 2 + 7.1853x 0.3198 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 25% 1

PAGE 110

96 (d ) (e) Fig ure 4.19 The f racture energy of shear wit h Hydrocure: (a) 0% replacement; (b) 25% replacement ; (c) 50% replaceme nt; (d) 75% replacement; (e) s ummary y = 8.475e 0.008x R = 0.423 0 10 20 0 20 40 60 80 Fractue energy (N/m) Replacement (%) y = 2.0818x 2 + 2.1235x 0.1872 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 75% 2 y = 0.4992x 2 + 3.7863x 0.0035 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm ) Displacement (mm) 75% 1 y = 5.5145x 0.2567 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 75% 3

PAGE 111

97 (a) (b) (c) y = 2.4865x 2 + 3.3191x 0.0532 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 25% 1 y = 1.2863x 2 + 3.4579x + 0.0136 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 25% 2 y = 2.0535x 2 + 1.67x 0.1138 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 25% 3 y = 0.6958x 2 + 3.2907x 0.57 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 50% 1 y = 0.9644x 2 + 0.8358x 0.0271 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 50% 2 y = 1.2596x 2 + 3.0521x 0.3627 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 50% 3 y = 1.5556x 2 + 5.3368x 0.511 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 1 y = 2.3486x 2 + 2.4452x 0.8359 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 2 y = 1.9889x 2 + 4.107x 0.6726 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 3

PAGE 112

98 (d) (e) Fig ure 4.20 The f racture energy of shear with CCA: (a ) 0% replacement; (b) 25% replacement ; (c) 50% replaceme nt; (d) 75% replacement; (e) s ummary y = 0.3323x 2 + 1.5364x + 0.087 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 75% 1 y = 0.8949x 2 + 2.942x 0.1928 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 75% 2 y = 0.8473x 2 + 7.5192x 0.4655 0 5 10 15 20 25 30 0 1 2 3 Shear (N/m/m) Displacement (mm) 75% 3 y = 8.475e 0.008x R = 0.423 0 10 20 0 20 40 60 80 Fractuae energy (N/m) Replacement (%)

PAGE 113

99 (a) (b) (c) y = 3.0681x 2 0.2846x 0.0698 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.2% 1 y = 0.9102x 2 + 5.4279x + 0.1249 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.2% 2 y = 2.1866x 2 + 1.956x 0.3153 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.2% 3 y = 2.8399x 2 0.4077x + 0.0427 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.4% 1 y = 2.5285x 2 + 0.533x 0.0195 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.4% 2 y = 2.0667x 2 + 2.1379x 0.1072 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.4% 3 y = 1.5556x 2 + 5.3368x 0.511 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 1 y = 2.3486x 2 + 2.4452x 0.8359 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 2 y = 1.9889x 2 + 4.107x 0.6726 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0% 3

PAGE 114

100 (d) (e) Fig ure.4.21 The f racture energy o f shear with SAP: (a) 0% adding; (b) 25% adding; (c) 50% adding; (d) 75% adding; (e) s ummary y = 2.2917x 2 1.5928x + 0.3233 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.6 % 1 y = 0.973x 2 + 3.3219x 0.1347 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.6 % 2 y = 3.2177x 2 1.0536x + 0.1841 0 5 10 15 20 25 30 0 1 2 3 Shear (N/mm) Displacement (mm) 0.6 % 3 y = 8.0163e 1.777x R = 0.7032 0 10 20 0 0.2 0.4 0.6 Fractue energy (N/m) Adding (%) Summary

PAGE 115

101 Fig ure 4.22 The f racture energy comparison of different internal curing agents y = 9.3032e 0.019x R = 0.7864 y = 8.475e 0.008x R = 0.423 0 10 20 0 10 20 30 40 50 60 70 80 Fractual energy (N/mm) Replacement (%) Hydrocure CCA y = 8.0163e 1.777x R = 0.7032 0 10 20 0 0.2 0.4 0.6 Fractue energy (N/m) Adding(%) Summary

PAGE 116

102 (a) b Fig ure 4.23 Response comparison for concrete mixed with Hydrocure: (a) energy dissipation displacement; (b) fracture displacement 0 5 10 15 20 25 30 0 1 2 3 Energy dissipation (kN) Displacement (mm) 25% 1 50% 1 75% 1 0% 1 0 5 10 15 20 25 30 0 1 2 3 Fracural energy (N/mm) Displacement (mm) 25% 1 50% 1 75% 1 0% 1

PAGE 117

103 (a) (b) Fig ure 4.24 Response comparison for concrete mixed with CCA: (a) energy dissipation displacement; (b) fracture displacement 0 5 10 15 20 25 30 0 1 2 3 Energy dissipation (kN) Displacement (mm) 25% 1 50% 1 75% 1 0% 1 0 5 10 15 20 25 30 0 1 2 3 Fracural energy (N/mm) Displacement (mm) 25% 1 50% 1 75% 1 0% 1

PAGE 118

104 (a) (b) Fig ure 4.25 Response comparison for concrete mixed with SAP : (a) energy dissipation displacement; (b) fracture displacement 0 5 10 15 20 25 30 0 1 2 3 Energy Dissipation (kN) Displacement (mm) 0% 1 0.2% 1 0.4% 1 0.6% 1 0 5 10 15 20 25 30 0 1 2 3 Fractural energy (N/mm) Displacement (mm) 0.2% 1 0.4% 1 0.6% 1 0% 1

PAGE 119

105 (a) (b) Fig ure 4.26 Absorptance of different internal curing agents: (a) Hydrocure and CCA; (b) SAP 0 10 20 30 40 50 0 1 2 3 4 5 6 7 8 Absorptance Ratio (%) Group Hydrocure CCA CCA average line Hydrocure average line 0 10000 20000 30000 40000 50000 0 1 2 3 4 5 6 7 8 Absorptance Ratio (%) Group SAP SAP average line

PAGE 120

106 (a) (b) Fig ure 4.27 Dry shrinkage test results for Hydrocure: (a) shrinkage; (b) shrinkage divided by depth 0 0.5 1 1.5 2 2.5 3 0 1 2 3 4 5 6 7 Shrinkage (10 4 inch) Time (d) 0% 25% 50% 75% 0 0.1 0.2 0.3 0 1 2 3 4 5 6 7 Shrinkage/Depth (10 4) Time (d) 0% 25% 50% 75%

PAGE 121

107 (a) (b) Fig ure 4.28 Dry shrinkage test results for CCA: (a) shrinkage; (b) shrinkage divided by depth 0 0.5 1 1.5 2 2.5 3 0 1 2 3 4 5 6 7 Shrinkage (10 4 inch) Time (d) 0% 25% 50% 75% 0 0.1 0.2 0.3 0 1 2 3 4 5 6 7 Shrinkage/Depth (10 4 ) Time (d) 0% 25% 50% 75%

PAGE 122

108 (a) (b) Fig ure 4.2 9 Dry shrinkage test results for SAP: (a) shrinkage; (b) shr inkage divided by depth 0 0.5 1 1.5 2 2.5 3 0 1 2 3 4 5 6 7 Shrinkage (10 4 inch) Time (d) 0 0.2 0.4 0.6 0 0.1 0.2 0.3 0 1 2 3 4 5 6 7 Shrinkage/Depth (10 4) Time (d) 0 0.2 0.4 0.6

PAGE 123

109 5 N umerical modelling 5.1 Introduction A numerical model was developed to predict the behavior of the interfacial behavior of the composite reinforcement when contacted with the high performance internal curing concrete. This model referred to local bond stress slip model from CEB FIP MODEL CODE 1990. An important feature of the model was that the loading should be monotonic which was satisfied with condition of the model of this research. The local bond stress s lip mode was about the bond stresses between concrete and reinforcing bar which was different from bond stresses between concrete and GFRP bars as GFRP bars did not own the obvious yield point elongation as reinforcing bars. Therefore, for our model, only three phases were modeled which was less than the four stages of local bond stress slip mode. To established the model well a parametric study was conducted to investigate the effects of key parameters on the behavior since the parameters were changing acc ording to the bond stress slip relationship. For slip, there were two parametric needed to be determined, they were S 1 which was the displacement when the stress reached to its maximum value and S 2 which was the displacement when bond stress began to decre ase very slowly. For stresses, two other parameters were supposed to be determined. The first one T max was the maximum stress value and the second one T f was the stress when the

PAGE 124

110 displacement was S 2 We can use Monte Carlo method to generate additional 9 997 results numerically without actually doing any physical testing based on the information of three tests for each type of internal curing concrete, then use those 10000 results to build models. The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for the three different types internal curing concrete are shown as Table 5.1. Besides, compressive stre ngth of the three types internal curing concrete and the maximum stress of internal curing concrete cast with GFRP were modeled by multiple regression, including all the variables listed in Table 5.1. 5.2 Numerical m odel for interfacial behavior of inter nal curing concrete and GFRP The following sections present the numerical models that were developed to predict the behavior of internal curing concrete and GFRP. The modeling procedure is described and then validated by the results of the experimental program. Because there are three types of internal curing concretes three similar modes needed to be built with the same method but different parameters.

PAGE 125

111 5.2.1 Model p rocedure Detailed procedures for obtaining the stress displacemen t behavior of internal curing concrete casted with GFRP bars are presented in the following sections and needed to be repeated three times to model the three types internal curing concrete. As the parameter of S 2 of local bond stress slip model does not ap pear in this research model, only two points (S 1 T max ) and (S 2 T f ) are remained to determine. Therefore, the values of point (S 1 T max ) and (S 2 T f ) can be obtained from the tests results together with the Monte Carlo method generated results firstly, then another parameter a can be evaluated to model the first stage. And the following section explains how to determine the value of (S 1 T max ), then get the a value followed by the value of (S 2 T f ). By substituting the push out tests results together with the Monte Carlo method generated results of the specific replacement amount or adding amount of internal curing concrete together with GFRP bars, the evaluated tests result s together with the Monte Carlo method generated results of the rest t ypes of internal curing concrete bonding with GFRP bars can be obtained. 5.2.2 First point (S 1 T max ) For each kind of internal curing agent, there are four kinds of replacement amount or adding amount of internal curing aggregate from 0% to 75% increased 25% or 0% to 0.6% increased by 0.2%. And for each replacement or adding amount, three

PAGE 126

112 cylinders with G FRP bars were casted. Therefore, by evaluating the test datum together with the Monte Carlo method generated results 10000 S 1 and three T max can be obtained and here the average value of the three is used. Then for each replacement amount one valid averag e value can be used to model a equation about the relationship between replacement amount and the S 1 or T max After that, the other results of S 1 or T max for the internal curing concrete with any replacement amount internal curing agents can be obtained fr om the equation built before. 5.2. 3 Coefficient a According to the local bond stress slip model of CEB FIP MODEL CODE 1990, the selection of coefficient a started from a bond characteristic with a constant stress ( a =0) up to a bond stress slip relationship with a constant stress ( a = 1). By submitting the tests results into the first stage equation given by the local bond stress slip model of CEB FIP MODEL CODE 1990, for each replacement one average a can be gotten Then average the four a values and get a new average value for the model.

PAGE 127

113 5.2.4 Second point (S 2 T f ) Different from the procedures of getting the first point, we need to make an assumption to obtain the second point. According to the equation of the fourth stage of the local bond stress slip model of CEB FIP MODEL CODE 1990, the value of T f constant. By the analysis of test datum, a particular value of T f can be assumed. Then also by analyzing the tests datum together with the Monte Carlo method generated results and the tests graphs, S 2 can also be evaluated. 5.2.5 Three models of the three types of in ternal curing agents Because the types of internal curing agents are different, the parameters are different. Therefore, for each type of internal curing concrete caste with GFRP bars, one particular model needed to be built 5.2.5.1 Model for Hydrocure When the internal curing agent is Hydrocure, there are four kinds of replacement amount of internal curing aggregate from 0% to 75% increased 25%. And for each replacement amount, three cylinders with GFRP bars were casted. By analyzing the tests datum together with the Monte Carlo method generated results the

PAGE 128

114 first point value is of 0% replacement is (1.82, 12.44). For 25% replacement, the first point is (1.52, 8.99). For 50% and 75% replacement, the first points are (1.28, 6.49) and (1.07 4.69) respectively. Then submit the tests results into the first stage equation given by the local bond stress slip model of CEB FIP MODEL CODE 1990, 10000 values of a was gotten and then average of the 10000 a was used followed. According to the equation of the fourth stage of the local bond stress slip model of CEB FIP MODEL CODE 1990, the value of T f constant. By the analysis of test datum together with the Monte Carlo method generated results T f equals to 1 .5 and then the average value of S 2 was evaluated as 2.5. Therefore the model is followed: (0 S 1 ) T=T max *(S/S 1 )^1.34 S 1 S 2 T=T max (T max T f )*(S S 1 )/(S 2 S 1 ) S 2 T=T f Where T max = S 1 = S 2 = 2.5; T f =1.5; a=1.34 Here a is just selected as the average value of all values of a which is bigger than 1. The results of model also compared with the or igin tests results as the Figure 5.1 shown. And the comparison shows that the model works well.

PAGE 129

115 5.2.5.2 Model for CCA When the internal curing agent is CCA, there are four kinds of replacement amount of internal curing aggregate from 0% to 75% increased 25%. And for each replacement amount, three cylinders with GFRP bars were casted. By analyzing the tests datum together with the Monte Carlo method generated results the first point value is of 0% replacement is (1.93, 11.5). For 25% replacement, the first point is (1.93, 9.64). For 50% and 75% replacement, the first points are (1.93, 8.1) and (1.93, 6.8) respectively. Then submit the tests results into the first sta ge equation given by the local bond stress slip model of CEB FIP MODEL CODE 1990, 10000 values of a was gotten and then average of the three a was used followed. According to the equation of the fourth stage of the local bond stress slip model of CEB FIP M ODEL CODE 1990, the value of T f constant. By the analysis of test datum together with the Monte Carlo method generated results T f equals to 1.5 and then the average value of S 2 was evaluated as 2.5. Therefore the model is followed: (0 S 1 ) T=T max *(S/S 1 )^1.57 S 1 S 2 T=T max (T max T f )*(S S 1 )/(S 2 S 1 ) S 2 T=T f Where T max = S 1 = 1.93 ; S 2 = 2.5; T f =1.5; a=1.57 Here a is just selected as the average value of all values of a which is bigger than 1. The results of model also compared with the origin te sts results as the Figure 5.2 shown. And the comparison shows that the model works well.

PAGE 130

116 5.2.5.3 Model for SAP When the internal curing agent is SAP, there are four kinds of addi ng amount of internal curing aggregate from 0% to 0.6% increased 0.2%. And for each replacement amount, three cylinders with GFRP bars were casted. By analyzing the tests datum together with the Monte Carlo method generated results the first point value is of 0% replacement is (1.65, 11.461). For 0.2% adding amount, the first point is (1.65, 8.79). For 0.4% and 0.6% adding amount, the first points are (1.65, 6.74) and (1.65, 5.16) respectively. Then submit the tests results into th e first stage equation given by the local bond stress slip model of CEB FIP MODEL CODE 1990, 10000 values of a was gotten and then average of the three a was used followed. According to the equation of the fourth stage of the local bond stress slip model o f CEB FIP MODEL CODE 1990, the value of T f constant. By the analysis of test datum together with the Monte Carlo method generated results T f equals to 1.5 and then the average value of S 2 was evaluated as 2.5. Therefore the model is followed: (0 S 1 ) T=T max *(S/S 1 )^1.8 S 1 S 2 T=T max (T max T f )*(S S 1 )/(S 2 S 1 ) S 2 T=T f Where T max = S 1 = 1.65; S 2 = 2.5; T f =1.5; a=1.34 Here a is just selected as the average value of all values of a which is bigger than 1. The results of model also compared with the origin t ests results as the Figure 5.3 shown. And the comparison shows that the model works well.

PAGE 131

117 5.3 Regression for compressive strength and maximum stress The compressive strength and maximum stress of the tested internal curing concrete were modeled by multiple regression We can use Monte Carlo me thod to generate additional 999 7 results numerically without actually doing any physical testing based on the information of three tests for each type of internal curing concrete, then use those 10000 results to do the regression for compressive strength and maximum stress to let the regression analysis more reasonable. 5.3.1 Regression for Hydrocure The compressive strength was fitted using the experimental data, as shown in the following equation s : fc' =0.024*H And the regression analysis is shown in Table 5.2. The maximum stress was fitted using the experimental data, as shown in the following equation: MS=0.001*H And the regression analysis is shown in Table 5.3. where fc compressive stress ( MPa ); MS is the maximum stress between internal curing concrete and GFRP bars; H is the weight of H ydrocure (g); is friction coefficient and D is the dynamic elastic modulus (GPa).

PAGE 132

118 5.3.2 Regression for CCA The compressive strength was fitted using the experimental data, as shown in the following equation: fc'= 10.79*D 709.067 And the regression analysis is shown in Table 5.4. The maximum stress was fitted using the experimental data, as shown in the following equation: MS = 0.63*C+1950.077* 10.39*D 759.87 And the regression analysis is shown in Table 5.5. where fc compressive stress ( MPa ); MS is the maximum stress between internal curing concrete and GFRP bars; C is the weight of CCA (g); is friction coefficient and D is the dynamic elastic modulus (GPa). 5.3.3 Regression for SAP The compressive strength was fitted using the experimental data, as shown in the following equation: fc'= 4.92*D 441.68 And the regression analysis is shown in Table 5.6. The maximum stress was fitted using the experimental data, as shown in

PAGE 133

119 the following equation: MS = 0.8*D 80.48 And the regression analysis is shown in Table 5.7. where fc compressive stress ( MPa ); MS is the maximum stress between internal curing concrete and GFRP bars; S is the weight of SAP (g); is friction coefficient and D is the dynamic elastic modulus (GPa).

PAGE 134

120 Table 5.1 The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for Hydrocure Average value of 10000 results 0% 25% 50% 75% Replacement 0.00026 24.99925 49.99832 74.99902 Friction coefficient 1.84124 1.83637 1.69797 1.67099 Elastic modulus 40.31104 33.66897 29.84817 26.73530 Stress 12.28227 9.29703 6.08938 4.59219 Compressive strength 40.59326 24.00018 21.11742 10.97913

PAGE 135

121 Table 5. 2 The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for CCA Average value of 10000 results 0% 25% 50% 75% Replacement 0.00026 25.00369 50.00182 74.99328 Friction coefficient 1.84124 2.09209 1.83430 1.60174 Elastic modulus 40.31104 22.20001 22.86402 23.04003 Stress 12.28227 9.00000 8.19990 6.30065 Compressive strength 40.59326 39.28999 29.89992 25.30003

PAGE 136

122 Table 5. 3 The average value of replacement; friction coefficient; elastic modulus; stress and compressive from the Monte Carlo method generated results for SAP Average value of 10000 results 0% 0.20% 0.40% 0.60% Replacement 0.00026 0.20603 0.40206 0.61281 Friction coefficient 1.84124 2.05347 1.92651 1.83016 Elastic modulus 40.31104 22.69003 19.75211 17.06003 Stress 12.28227 7.71002 7.10000 5.20011 Compressive strength 40.59326 24.01021 21.09991 11.00034

PAGE 137

12 3 Table 5. 4 Test results for regression analysis type fc'(Mpa) Max stress Hydrocure DEM (GPa) Hydrocure 0 40.6 12.3 0 0.61 40.31 Hydrocure 25 32.4 9.3 150 0.61 33.66 Hydrocure 50 15.1 6.1 300 0.64 29.85 Hydrocure 75 11.6 4.6 450 0.64 26.74 type fc' Max stress CCA(g) DEM (GPa) CCA 0 40.6 12.3 0 0.61 40.31 CCA 25 39.3 9 150 0.56 22.20 CCA 50 29.9 6.3 300 0.61 22.86 CCA 75 25.3 8.2 450 0.66 23.04 type fc' Max stress SAP(g) DEM (GPa) SAP 0 40.6 12.3 0 0.61 40.31 SAP 0.2 24 7.7 0.8913 0.56 22.69 SAP 0.4 21.1 7.1 1.7826 0.59 19.75 SAP 0.6 11 5.2 2.6739 0.61 17.06

PAGE 138

124 Table 5. 5 Regression analysis of fc'(Mpa) for Hydrocure Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 575.67 191.89 Residual 0.00 0.00 65535.00 Total 3.00 575.67 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 285.68 0.00 65535.00 285.68 285.68 Hydrocure 0.02 0.00 65535.00 0.02 0.02 509.57 0.00 65535.00 509.57 509.57 DEM (GPa) 1.64 0.00 65535.00 1.64 1.64

PAGE 139

125 Table 5. 6 Regression analysis of maximum stress for Hydrocure Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 35.33 11.78 Residual 0.00 0.00 65535.00 Total 3.00 35.33 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 27.77 0.00 65535.00 27.77 27.77 Hydrocure 0.00 0.00 65535.00 0.00 0.00 55.58 0.00 65535.00 55.58 55.58 DEM (GPa) 0.46 0.00 65535.00 0.46 0.46

PAGE 140

126 Table 5. 7 Regression analysis of fc'(Mpa) for CCA Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 163.95 54.65 Residual 0.00 0.00 65535.00 Total 3.00 163.95 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 709.07 0.00 65535.00 709.07 709.07 CCA 0.67 0.00 65535.00 0.67 0.67 1940.40 0.00 65535.00 1940.40 1940.40 DEM (GPa) 10.80 0.00 65535.00 10.80 10.80

PAGE 141

127 Table 5. 8 Regression analysis of maximum stress for CCA Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 18.81 6.27 Residual 0.00 0.00 65535.00 Total 3.00 18.81 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 759.87 0.00 65535.00 759.87 759.87 CCA 0.63 0.00 65535.00 0.63 0.63 1950.08 0.00 65535.00 1950.08 1950.08 DEM (GPa) 10.39 0.00 65535.00 10.39 10.39

PAGE 142

128 Table 5. 9 Regression analysis of fc'(Mpa) for SAP Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 452.85 150.95 Residual 0.00 0.00 65535.00 Total 3.00 452.85 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 441.68 0.00 65535.00 441.68 441.68 SAP 53.34 0.00 65535.00 53.34 53.34 1114.65 0.00 65535.00 1114.65 1114.65 DEM (GPa) 4.92 0.00 65535.00 4.92 4.92

PAGE 143

129 Table 5. 10 Regression analysis of maximum stress for SAP Regression Statistic Multiple R 1 R Square 1 Adjusted R Square 65535 Standard Error 0 Observations 4 ANOVA df SS MS Regression 3.00 27.21 9.07 Residual 0.00 0.00 65535.00 Total 3.00 27.21 Coefficients Standard Error t Stat Lower 95% Upper 95% Intercept 80.48 0.00 65535.00 80.48 80.48 SAP 9.55 0.00 65535.00 9.55 9.55 205.05 0.00 65535.00 205.05 205.05 DEM (GPa) 0.80 0.00 65535.00 0.80 0.80

PAGE 144

130 (a) (b) 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0% 1 0% 2 0% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 25% 1 25% 2 25% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 145

131 (c) (d) Fig ure 5.1 Model results comparison with tests results with Hydrocure: (a) 0% replacement; (b) 25% replacement; (c) 50% r eplacement; (d) 75% replacement 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 50% 1 50% 2 50% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 75% 1 75% 2 75% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 146

132 (a) (b) 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0% 1 0% 2 0% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 25% 1 25% 2 25% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 147

133 (c) (d) Fi gure 5.2 Model results comparison with tests results with CCA: (a) 0% replacement; (b) 25% replacement; (c) 50% r eplacement; (d) 75% replacement 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 50% 1 50% 2 50% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 75% 1 75% 2 75% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 148

134 (a) (b) 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0% 1 0% 2 0% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0.2% 1 0.2% 2 0.2% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 149

135 (c) (d) Fig ure 5.3 Model results comparison with tests results with SAP: (a) 0% adding; (b) 0.2% adding; (c ) 0.4% adding; (d) 0.6% adding 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0.4% 1 0.4% 2 0.4% 3 0 5 10 15 20 25 30 0 2 4 6 8 10 Shear (N/mm) Displacement (mm) 0.6% 1 0.6% 2 0.6% 3 model of Monte Carlo values model of Monte Carlo values

PAGE 150

136 6 S ummary & conclusions 6.1 Summary This paper has studied the interfacial behavior of GFRP reinforcing bars embedded in the concrete mixed with three internal curing agents that replaced fine aggregates to a certain extent ranging from 0% and 75% o r added SAP according the weigh t of cement from 0% to 0.6% A push out bond test was conducted to evaluate the performance of this new reinforced concrete concept. With an increasing replacement ratio of the Hydrocure and CCA or the adding ratio of SAP, the compressive strength of the concrete and the bond betwee n the GFRP and the concrete were tested To investigate s ome other properties of internal curing concrete like coefficient of friction, compressive strength, elastic modulus, drying shrinkage, and r apid chloride ion penetrability different kinds of intern al curing concretes were tested and the results are also included in this thesis. An numerical model was developed to predict the stress performance between internal curing concrete and GFRP bar according to CEB FIP MODEL CODE 1990 Several parameters like S 1 ; T max ; S 2 ; T f and were set to build models. T he model was verified by the experimental results and showed good agreement.

PAGE 151

137 6.2 Conclusions The following conclusions were drawn from the experimental and numerical investigations of internal curing concretes at different agents replacement ratio or adding ratio: 1. The average compressive strength tended to decrease with the addi ng replacement amount of Hydrocure and CCA or the increasing amount of SAP 2. The maximum inte rfacial load of bonding stress decreased with an increasing replacement amount of Hydrocure and CCA or the increasing amount of SAP. 3. The degree of scatter in energy dissipation reduced with an increasing replacement amount of Hydrocure and CCA or the increasing amount of SAP. 4. The degree o f scatter in fracture energy reduced with an increasing replacement amount of Hydrocure and CCA or the increasing amount of SAP. 5. The Chloride penetration for concrete with Hydrocure increased with an increasing replacement amount of Hydrocure 6. The frequency and dynamic modulus of concr ete with internal curing agents reduced with an increasing replacement amount of Hydrocure and CCA or the increasing amount of SAP. 7. The acceleration between concrete with Hydrocure and GFRP bar measured from the friction test decreased a little bit with an increasing replacement amount of Hydrocure. For CCA the acceleration increased from the 0% replacement to 25% replacement then decr eased from 25% replacement to 75% replacement. And for SAP the acceleration increased from the 0% replacement to 0.2% adding amount

PAGE 152

138 then decreased from 0.2% replacement to 0.6% adding amount 8. The c oefficient of friction increased a little bit with an increa sing replacement amount of Hydrocure and CCA or the increasing amount of SAP. The following conclusions were drawn from the experimental and numerical investigations of the properties comparison of different types of internal curing concretes: 1. T he compressive strength of internal curing concretes casted with Hydrocure is weaker than internal curing concrete with CCA at same agents replacement ratio. And adding 0.1% SAP has the similar effect of replacing 14.68% Hydrocure for compressive strength 2. T he frequency and dynamic modulus of internal curing concretes casted with Hydrocure is higher than internal curing concrete with CCA at same agent replacement ratio. 3. T he friction coefficient of internal curing concretes casted with Hydrocure is a little b it higher than internal curing concrete with CCA at same agent replacement ratio. 4. T he maximum load of internal curing concretes casted with Hydrocure is weeker than internal curing concrete with CCA at same agent replacement ratio. And adding 0.1% SAP has the similar effect of replacing 16.435% Hydrocure. 5. The energy dissipation of internal curing concretes casted with Hydrocure is

PAGE 153

139 higher than internal curing concrete with CCA at same agent replacement ratio. And adding 0.1% SAP has the similar effect of rep lacing 16.432% Hydrocure for energy dissipation. 6. T he fracture energy of internal curing concretes casted with Hydrocure is higher than internal curing concrete with CCA at same agent replacement ratio. An d adding 0.1% SAP has the similar effect of replacing 17.19% Hydrocure for fracture energy. 6.3 Recommendations for f uture w ork 1. More research is needed to obtain the results of chloride penetration of internal curing concrete when the agent is CCA and SAP and compare the results with the Hydrocure. 2. For the dry shrinkage tests the testing days required to extend to 28 days.

PAGE 154

140 R EFERENCE [1] curing agent for low cracking high Structural Engineering and Engineering Materials SM Report No. 97(2009):142. [ 2 ] Suggested internal curing of concrete specifications 2007. . life cycle cost of high performance concrete bridge decks Cement & Concrete Composite s (2010):1. [ 4 related properties Cement and Concrete Research (2011):3. [ 5 ] G eorge C., H off concrete Northeast Solite Corporation (2002):2. [6] Fiberglass Rebar Manual. May. 2007 < http://aslanfrp.com/Aslan100/Aslan%20100_GFRP_rebar_Flyer.pdf> [ 7 TTCC/NCC (2009):7. surface mounted fiber reinforced polymer bars for concretes strengthening experimental ACI STRUCTURAL JOURNAL Title no. 101 S28:277

PAGE 155

141 Journal of Structural Engineering (1985):506 512. matrix in uniaxially fiber re The Royal Society (2014): 215 217. [11] Chun out stresses of fiber Material Science and Engineering (1990):2 5. shed returned concrete aggregate for NRMCA Concrete Technology Forum on Sustainable Development (2008):1 11. Material and Structures (20 06):1 4. International Journal of Advanced Structures and Geotechnical Engineering (2014):2. [15] C. high performance concrete using super absorbent polymers and light weight Asian Journal of Civil Engineering (2013):1 2. ificance of Tests and Properties of Concrete and Concrete ASTM Stock :551. Faculty of Engineering and Built Environment, University of Cape

PAGE 156

142 T own (2008):104 106. Faculty of Graduate Studies and Research(2004) :138. Destructive evaluation of the penetrability and thickness of the State of the Art Report(2007):91 100. Greenwich Academic Literature Archive (2010):28. [21] Hassan,Ghanem.,Scott,Phelan.,Sanjaya,Senadhee ASCE (2008):1 8. [22] Jianxin Ji.,David D arwin& stainless steels and mmfx microcomposite steel for reinforced concrete bridge Structural Engineering and engineering Materials (2001): 2 3. [23] Ming Te Liang.,Li Hsien Lin& Chih Service Life Prediction of Existing Reinforced Concrete ASCE (2002):1 4. [24] Calibration of resistance factors for reliability based design of externally bonded FRP composi ScienceDirect (2008):665 679. [25] Arnon Bentur., Shin autogenous shrinkage in high strength concrete by internal curing using wet

PAGE 157

143 Cement and Concrete Research 31 (2001):1 3. [26 Faculty of Purdue University (2008): 37 123. Journal of struc ture engineering (2001):1 [28] ScienceDirect (2005): 1 2. [29] IOPscience (2004) :1 2. [30] entrained cement based Cement and Concrete Research (2001) 647 Water entrained cement based Cement and Concrete Research (2002) : 973 974 balance and pore structure development in cementitious materials in internal Microporous and Mesoporous Materials (2009): 51 52. [33] Bart Craeye., polymers as an internal curing agent for mitigation of ea rly age cracking of high Construction and Building Materials

PAGE 158

144 (2011) 1 13 Techniques and materials for internal water curing of concrete. (2006): 825 [3 Journal of materials in civil engineering (2012) :754. [36] Ryan Henkensiefken., Javier Castro., Haejin Kim., Dale Bentz, & Jason Weiss. Concrete InFocus (2009):1 2.