Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fracture toughness of hybrid fiber reinforced self-compacting concrete Majdzadeh, Fariborz 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2003-0151.pdf [ 5.81MB ]
JSON: 831-1.0063792.json
JSON-LD: 831-1.0063792-ld.json
RDF/XML (Pretty): 831-1.0063792-rdf.xml
RDF/JSON: 831-1.0063792-rdf.json
Turtle: 831-1.0063792-turtle.txt
N-Triples: 831-1.0063792-rdf-ntriples.txt
Original Record: 831-1.0063792-source.json
Full Text

Full Text

FRACTURE TOUGHNESS OF HYBRID FIBER REINFORCED SELF-COMPACTING CONCRETE by  FARIBORZ MAJDZADEH B.Sc. (Civil Engineering), Amir Kabir University of Technology, Tehran, IRAN, 1986 M.Sc. (Geotechnical Eng.), Amir Kabir University of Technology, Tehran, IRAN, 1988  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CIVIL ENGINEERING  We accept this thesis as conforming to the required standard:  T H E U N I V E R S I T Y OF BRITISH C O L U M B I A March 2003 © Fariborz Majdzadeh, 2003  In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Department of Civil Engineering The University of British Columbia Vancouver, Canada March 10, 2003  ABSTRACT There has been much enthusiasm for hybrid fiber reinforced concrete systems, in which two or more types of fibers are combined in the same concrete matrix. On the other hand, the use of self-compacting concrete (SCC), due to its unique fresh properties and social and environmental benefits, is gaining popularity worldwide. Hence, knowledge on the use of hybrid fibers in SCC deserves attention. The scope of this study encompasses two major research focuses. The first involves the production and evaluation of the fresh properties of high strength fiber reinforced selfcompacting concrete and a discussion of its key attributes in the fresh state. The second involves the assessment of the mechanical properties and potential synergistic effects of various fibers in SCC. In this study, 23 mixes containing mono, double, and triple blended fibers were made. The hybrid systems were different combinations of micro fibers (carbon, polypropylene, and steel) and macro fibers (steel and polypropylene). For each mix, six 100x100x350 mm beam specimens for flexural toughness tests and six 100x200 mm cylindrical specimens for compression tests were made. Fresh properties of concrete were measured by a slump flow test. On the flowability aspects, it was found that carbon and micro-polypropylene fibers, even at low volume fractions, caused significant reduction in the flowability of SCC. However, introduction of macro fibers of steel and polypropylene in SCC did not reduce the flowability of concrete as much as micro-fibers did. The maximum amount of microfibers in the hybrid system to retain the self-compacting properties of the fresh mix was 0.25% for carbon fibers and 0.15% for micro-polypropylene fibers provided that they were used individually in the hybrid system. When hardened properties of hybrid fiber SCC where considered the study focused on flexural toughness improvements due to the presence of hybrid fibers in the SCC. Micropolypropylene fibers in combination with all types of steel fibers showed a synergistic effect, while carbon fibers were efficient only in one combination with the steel fiber. According to the results of this study, in a hybrid fiber composite, micro fibers improve the ductility of the matrix. This enhanced the performance of the macro fibers in the pullout process, and the hybridfiberSCC showed higher flexural toughness as compared to SCC with only one fiber.  ii  T A B L E OF CONTENTS  ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  viii  LIST OF FIGURES  ix  ACKNOWLEDGEMENT  1.4  2.0  xiii  INTRODUCTION  1.1. Background  1  1.2. Scope and objective of the present study  3  1.3. Approach  4  1.4. Organization of this report  4  Literature review on Basic Concepts and fiber Reinforced S C C 2.1 Introduction  6  2.2 Concept of self-compacting concrete  6  2.3 Theoretical background on rheology of concrete  7  2.3.1  Newtonian fluid behavior  2.4 Fresh properties of concrete  7 9  2.5 Fresh properties of SCC  11  2.5.1  Filling ability  11  2.5.2  Resistance to segregation  12  2.5.3  Passing ability  14  2.6 Materials in developing of SCC 2.6.1  15  Cementitious material  15 Cement  15 Fly Ash  16  iii Silica Fume  17  2.6.2  Limestone Powder  18  2.6.3  Admixtures  19 Superplasticizer  19 Viscosity admixture  20  2.7 Mix design  21  2.8 Test methods for fresh SCC  22  2.8.1  Slump flow test  22  2.8.2  V-funnel test  23  2.8.3  L-box test  24  2.8.4  J-ring test method  25  2.9 Role of fibers in behavior of fresh SCC  26  2.9.1  Influence of steel fiber  28  2.9.2  Influence of carbon  2.9.3  Influence of micro-polypropylene  30  2.9.4  Influence of hybrid fibers  31  fiber  2.10 Hardened properties offiberreinforced SCC  29  31  2.10.1 Steelfiberreinforced SCC  32  2.10.2 Polypropylenefiberreinforced SCC  32  2.10.3 Carbonfiberreinforced SCC  34  2.10.4 Hybridfiberreinforced SCC  37  3.0 Experimental program for H y F R S C C 3.1 Introduction  42  3.2 Materials  42  3.2.1 Portland Cement  42  3.2.2 Silica Fume  43  3.2.3 Fly Ash  43  3.2.4 Water  43  3.2.5  43  Coarse aggregates  3.2.6 Fine aggregate  44  iv  3.2.7 Fibers  45  3.2.8 Admixture  45 Superplasticizer  45 Air Entraining Agent  46  3.3. Mix Proportions  4.0  46  3.3.1 The volume of concrete batch  47  3.3.2 Components of hybrid fibers  48  3.3.3 Selection of mixing methods  49  3.3.4  Mixing equipment  51  3.3.5  Specimen preparation  51  Results 4.1 Introduction  53  4.2 Test Results of Fresh HyFRSCC  55  4.3 Test Results of Hardened HyFRSCC  55  4.3.1 Compressive Strength  56  4.3.2 Flexural Strength  57  4.3.3 Flexural Toughness Test  59 Test Apparatus  59 Test Procedure  60 Calculation procedure  61 Techniques for flexural Toughness measurement  62 ASTM C1018 test method and its concerns  62 JSCE standard SF-4 method and its concerns  65 Post Crack Strength (PCS) Method  67 Flexural responses for HyFRCSCC specimens  5.0  69  Results analysis and discussion 5.1 Introduction  82  5.2 Fresh property results  82  v  5.3 Discussion on compressive strength  87  5.4 Flexural Toughness  89  5.4.1 Hybridization of Steel Macro-Fiber and Carbon Fiber  89  5.4.2 Hybridization of Steel Macro-Fiber (flat-end) and MicroPolypropylene fiber  93  5.4.3 Hybridization of HPP (Macro-Polypropylene) Fiber with Carbon Fiber  96  5.4.4 Hybridization of Steel Fiber (crimped) with Micro-Polypropylene  99  5.4.5 Hybridization of Steel Micro-Fiber with Micro-Polypropylene  102  5.4.6 Hybridization of Steel Fiber (flat-end) with Micro-Poly propylene and Carbon Fiber in Low Volume Fraction 5.4.7 Hybridization of Steel fiber (flat-end) with HPP  6.0  105 105  Application of HyFRSCC in repair of RC beams 6.1 Abstract  110  6.2 Introduction  110  6.3 Test program  111  6.3.1  Materials  111  6.3.2  Specimen preparation  113  6.3.3  Test procedure  115  6.4 Test Results  116  6.5 Discussion  117  6.5.1. Bond strength  117  6.5.2. Behavior of the repaired beams  118  6.6 Conclusion  122  vi  7.0 Conclusions 7.1  8.0  Introduction  123  7.2 Production of HyFRSCC  123  7.3  124  Fresh behavior of HyFRSCC  7.4 Compressive strength of HyFRSCC  124  7.5 Flexural toughness of HyFRSCC  125  Recommendations for future studies  127  References  129  vii  LIST O F T A B L E S  Table 2.1: Examples of definition of workability by various societies  9  Table 3.1: Chemical composition of the cement  41  Table 3.2: Chemical composition of the silica fume  42  Table 3.3: Chemical composition of the fly ash  42  Table 3.4: Shows the specification of fiber used in the mixes  44  Table 3.5: Selected SCC mixtures  46  Table 3.6: The proportion of materials for the selected mixes  47  Table 3.7: Hybridization of fibers in the mixes  48  Table 3.8: Preliminary mixing method for SCC  49  Table 4.1: Slump flow results for different hybridization of fiber  53  Table 4.2: Results of compressive strength for different mixes  56  Table 4.3: Modulus of Rapture results for different mixes based on ASTM CIO 18  57  Table 4.4: Flexural toughness results for different mixes based on ASTM C1018  63  Table 4.5: Flexural toughness factor results for different mixes according to JSCE  65  Table 4.6: Flexural toughness results for different mixes according to PCS method  67  Table 5.1: The average amount of superplasticizer used in different mixes  84  Table 5.2: Comparison of FRSCC flowability with different steel fiber contents  86  Table 6.1: Mixture proportion of self-compacting concrete and normal concrete  111  Table 6.2: Specification of the fibers used in the repair materials  111  Table 6.3: Proportion of fibers in repair material  111  Table 6.4: The results of repaired beams under flexural loading  115  viii  LIST O F F I G U R E S  Fig. 1.1: Comparison of SCC with Normal Concrete  3  Fig.2.1: Newton's law of viscous flow  7  Fig.2.2: Bingham's equation for a fluid  8  Fig.2.3: Concrete Rheology Parameters  10  Fig.2.4: Comparison the rheology behavior of different types of concrete  10  Fig.2.5: General approach to achieve self-compacting concrete  12  Fig.2.6: Mechanism of blocking due to arching  14  Fig. 2.7: Relationship between slump flow and elapsed time with different replacement ratios of fly ash  17  Fig.2.8: Measurement of slump flow in SCC  23  Fig.2.9: V funnel Test  24  Fig.2.10: Principle of L shaped box for evaluation of self-consolidating properties  25  Fig.2.11: In the J-ring test the cone is filled with SCC and lifted and the SCC flows through the re-bars  26  Fig.2.12: Non uniform flow patterns  28  Fig.2.13: Blocking is presented based on relation H /Hb Versus re-bar a  opening gap S and fiber factor L/Vf/d  28  Fig.2.14: Rheological parameter g as a function of the fiber volume fraction  30  Fig.2.15: Response obtained from flexural test for carbon fiber composite  35  Fig.2.16: Characteristic stress-strain curves of specimens under uniaxial tension  36  Fig.2.17: Characteristic load-deflection curves for various Vf in CFRC  37  Fig 2.18: Schematic description of load-deflection of hybrid composite  38  Fig.2.19: Hybrid fibers control; cracks at different levels  39  Fig.3.1: Round river gravel used in the SCC mix  43  Fig.3.2: Shows the sieve analysis of fine and coarse aggregate  43  Fig.3.3: Mixer used for the all tests  50  Fig.4.1: The slump flow test  52  Fig.4.2: Compression testing machine  55  Fig.4.3: Schematic of the loading and measuring system  59  Fig 4.4: Arrangement of beam specimen for Flexural toughness test  60  ix  Fig.4.5: The modification of load-deflection curve in flexural toughness test  61  Fig4.6: Description of the flexural toughness according to ASTM C1018  62  Fig.4.7: Flexural toughness factor description according to JSCE in which the area is measured up to a deflection of span/150  64  Fig.4.8: Flexural toughness description according to PCS method  66  Fig.4.9: Flexural response of the FRC beam (Mix 3)  69  Fig.4.10: Flexural response of the FRC beam (Mix 1)  69  Fig.4.11: Flexural response for the FRC beams (Mix 2)  70  Fig.4.12: Flexural response for the FRC beams (Mix 11)  70  Fig.4.13: Flexural response of the FRC beam (Mix 6)  71  Fig.4.14: Flexural response of the FRC beam (Mix 8)  71  Fig.4.15: Flexural response of the FRC beam (Mix 12)  72  Fig.4.16: Flexural response of the FRC beam (Mix 13)  72  Fig.4.17: Flexural response of the FRC beam (Mix 22)  73  Fig.4.18: Flexural response of the FRC beam (Mix 14)  73  Fig.4.19: Flexural response for the FRC beams (Mix 15)  74  Fig.4.20: Flexural response for the FRC beams (Mix 16)  74  Fig.4.21: Flexural response for the FRC beams (Mix 7)  75  Fig.4.22: Flexural response for the FRC beams (Mix 17)  75  Fig.4.23: Flexural response for the FRC beams (Mix 4)  76  Fig.4.24: Flexural response for the FRC beams (Mix 18)  76  Fig.4.25: Flexural response for the FRC beams (Mix 5)  77  Fig.4.26: Flexural response for the FRC beams (Mix 19)  77  Fig.4.27: Flexural response of the FRC beams (Mix 10)  78  Fig.4.28: Flexural response of the FRC beams (Mix 20)  78  Fig.4.29: Flexural response of the FRC beams (Mix 23)  79  Fig.4.30: Flexural response of the FRC beams (Mix 9)  79  Fig.4.31: Flexural response of the FRC beam (Mix 21)  80  Fig.5.1: Relationship between slump flow and compressive strength  83  Fig.5.2: Average slump flow for different types of mixes  84  Fig. 5.3: Compressive strength of different mixes  87  x  Fig.5.4: PCS values for HyFRSCC beams with Steel and Carbon fibers  90  Fig.5.5: Flexural responses for HyFRSCC beams with Steel and Carbon fibers  90  Fig.5.6: ASTM C1018 Method for HyFRSCC beams with Steel and Carbon fibers  91  Fig.5.7: JSCE Method for HyFRSCC beams with Steel and Carbon fibers  91  Fig.5.8: Flexural responses for HyFRSCC beams with steel and micro-polypropylene fibers  93  Fig.5.9: PCS values for HyFRSCC beams with steel and micro-polypropylene fibers  93  Fig.5.10: ASTM C1018 Method for HyFRSCC beams with Steel and Micro-polypropylene fibers Fig.5.11: JSCE Method for HyFRSCC beams with Steel and  94  Micro-polypropylene fibers  94  Fig.5.12: Flexural responses for HyFRSCC beams with HPP and Carbon fibers  96  Fig.5.13: PCS values for HyFRSCC beams with HPP and Carbon fibers  96  Fig.5.14: ASTM C1018 Method for HyFRSCC beams with HPP and Carbon fibers  97  Fig.5.15: JSCE Method for HyFRSCC beams with HPP and Carbon fibers  97  Fig.5.16: Flexural responses for HyFRSCC beams with Steel (crimped) and Micro-polypropylene fibers Fig. 5.17: PCS Method for HyFRSCC beams with Steel (crimped) and Micro-polypropylene fibers  99  Fig.5.18: ASTM C1018 Method for HyFRSCC beams with Steel (crimped) and Micro-polypropylene fibers  100  Fig.5.19: JSCE Method for HyFRSCC beams with Steel (crimped) and Micro-polypropylene fibers  100  Fig.5.20: Flexural responses for HyFRSCC beams with Micro-steel and Micro-polypropylene fibers  99  102  Fig.5.21: PCS values for HyFRSCC beams with Micro-steel and Micro-polypropylene fibers  102  Fig.5.22: ASTM C1018 Method for HyFRSCC beams with Micro-steel and Micro-polypropylene fibers  103  Fig.6.23: JSCE Method for HyFRSCC beams with Micro-steel and Micro-polypropylene fibers  103  Fig.4.24: Flexural responses for HyFRSCC beams with low fiber content  105  Fig.5.25: PCS values for HyFRSCC beams with low fiber content  105  Fig.5.26: ASTM C1018 Method for HyFRSCC beams with lowfibercontent  106  Fig.5.27: JSCE Method for HyFRSCC beams with lowfibercontent  106  xi  Fig.5.28: Flexural responses for HyFRSCC beams with Steel & HPP fibers  107  Fig.5.29: PCS values for HyFRSCC beams with Steel & HPP fibers  107  Fig.5.30: ASTM C1018 Method for HyFRSCC beams with HPP & Steel fiber  108  Fig.5.31: JSCE Method for HyFRSCC beams with HPP & Steel fiber  108  Fig.6.1: a) Detailing of RC beam before repairing, b) Cross section of repaired RC beam Fig.6.2: Pouring the repair materials around the RC beam  112 113  Fig.6.3: Typical section of repaired beam  113  Fig.6.4: Laboratory scale beam test set up  114  Fig.6.5. Baldwin 400 kip universal testing machine  115  Fig.6.6: Load-deflection curves for un-repaired and repaired beams, C: carbon fiber, S: Steel fiber, Mp: micro-polypropylene fiber  116  Fig.6.7: The core gained from the repaired beam  117  Fig.6.7: First crack load comparison in different repaired beams  118  Fig.6.9: Ultimate load in different repaired beams  119  Fig.6.10: Energy absorption in Different repaired beams  120  xii  ACKNOWLEDGEMENT I wish to express my sincere appreciation and profound gratitude to my advisor, Prof. Nemkumar Banthia not only for his continuous guidance and for close supervision throughout this study, but also for his patience and encouragement. Special appreciation and thanks go to Prof. S. Mindess for his invaluable comments and suggestions in reviewing of this thesis.  The experimental part of this study would not have been possible without the prompt assistance provided by specialized technicians particularly Mr. John Wong and Mr. Doug Smith. Thanks are also extended to all graduate students in the materials lab who provided me with precious support in the discussions and the experimental program.  My acknowledgments also go to the companies that provided materials for this study. The Master Builder technology was the major source of chemical admixtures and Conoco and Synthetic Industries were the main providers of fibers for this project.  Finally, I would like to thank my family; my wife, Neda, my son, Ali, and my daughter, Maryam, whose love and support contributed to my study achievement.  xiii  CHAPTER 1 INTRODUCTION 1.1 Background Normal concrete, widely used as a construction material, has many advantages including an ability to be cast, low cost, good durability, fire resistance, energy efficiency, on site fabrication, and aesthetics [1]. However, concrete also has many disadvantages including a low tensile strength, low ductility, and large variability. To improve the performance of concrete, various specialized concretes such as high strength concrete,fiberreinforced concrete, and self-compacting concrete have been developed. High strength concrete has an improved compressive strength and durability, but on the other hand, it is also extremely brittle. Fiber reinforcement enhances the ductility of concrete, but fibers reduce the workability of fresh concrete, thus sometimes requiring specialized processing, and in turn, limiting the application of the material. To improve the quality and reliability of construction in concrete structures, self-compacting concrete, which needs no consolidation on job sites, has been successfully developed and commercialized in Japan [2]. The self-compacting properties include a high deformability, good flowability in the fresh state, and a good cohesiveness against segregation. It can also be high strength to improve its durability.  Recent studies [3] infiberreinforced concrete have shown that a hybrid fiber system is more effective and sometimes more economical than a monofibersystem. One type of hybrid fiber system involves combining macro and micro fibers in concrete. In this system, introduction of macro-fibers in concrete can effectively improve toughness and ductility of concrete, while adding micro fibers may improve not only durability of concrete but also exhibit synergistic effects in toughness. However, in hybrid fiber reinforced concrete, workability of fresh concrete may be of particular concern given the highfibercontent and presence of micro fibers.  1  Thus, the development of self-compacting concrete with hybrid fibers, which combines the self-compacting properties of SCC in the fresh state and ductile behavior of hybrid FRC in hardened state, is expected to be useful in different applications.  The most important issue in development of high performance SCC is the mix proportions. It should possess high strength, improved durability, high ductility and good flowability in the fresh state. Obviously some of these properties are contradictory. For instance, while fiber will improve the durability of concrete, the large surface area of fibers tends to restrain flowability of the fresh concrete. Fiber balling, due to fiber interlocking and poor dispersion, can affect the hardened properties of concrete, which in turn will reduce the durability of concrete. Therefore, without having a good fluidify and a reasonable workability in freshly mixed concrete, it would be impossible to use the full potential of fibers to enhance the mechanical properties of concrete in its hardened state.  Since the 1970s, there have been many attempts to improve the flowability of fresh concrete by controlling interactions between cement particles with various chemical admixtures [4]. The cement particles experience both electrostatic and Van der Waals attractions, which leads to flocculation and a subsequent  loss of flowability.  Superplasticizers are used to disperse the flocculated cement particles and result in reductions of viscosity and yield stress in the fresh concrete mix. Particularly, polycarboxylate-based superplasticizers have become the most important high-range water-reducing agents due to their excellent water reducing, slump retaining, and viscosity reducing effects in concrete with low water to cement ratio. The high-range water reduction and high fluidity of the concrete are attributed to the polycarboxylatebased dispersant which is adsorbed onto the surface of the binding material particles and forms a steric repulsive layer [5].  Rheological studies of self-compacting concrete have been recognized as an important tool to investigate the effects of various types of admixtures, aggregates, binders, and fibers on the flowability. The rheology of SCC is characterized by two parameters: yield stress and plastic viscosity. Each of these two parameters plays a special role in fresh  2  concrete in order to make it flowable and segregation resistant simultaneously. Making a good SCC depends on achieving the desired yield stress and viscosity, which can be obtained by a specific proportion of mixing materials. The yield value shows the ability of a concrete mix to flow. Flowability of SCC increases by decreasing its yield value, but concrete with very low yield values might experience aggregate segregation. Therefore, a good SCC mix should have a viscosity proportional to its yield value to prevent concrete segregation. On the other hand, a high viscosity reduces the flowability of concrete and thus affects the self-compacting nature of SCC.  Self-compacting concrete  Normal concrete  Stress  Yield stress  Shear rate  Fig.1.1. Comparison of SCC with Normal Concrete  1.2 Scope and Objectives of the Present Study  The main objective of this research was to develop a high strength, self-compacting hybrid fiber reinforced concrete. This material is defined as a hybrid FRC maintaining self-compacting properties in its fresh state. It was developed by combining the knowledge gained from self-compacting concrete with that from hybridfiberreinforced concrete. A successful development of this material should improve the workability of FRC and expand its applicability.  3  To achieve this goal, two types of fiber were used in concrete simultaneously: macro fibers including steel and HPP (High Performance Polypropylene) fibers, and micro fibers including carbon and micro-polypropylene fibers. It was expected that hybridization of fibers would improve the mechanical properties of mono-fiber reinforced concrete. The next part of this research involved the application of this material for repair of concrete beams. The high flowability of SCC makes it suitable for repair and retrofit purposes. SCC can be cast without any consolidation, and this facilitates the placement of concrete in restricted area such as in thin repairs. [6] The objective of this part of research was to investigate the effect of SCC repair material on the flexural behavior of reinforced concrete beams. In this section, the concrete beams were retrofitted with different types of hybrid fiber reinforced concrete and tested under flexure. This study compared the effects of using different hybrid fiber combinations on the flexural capacity and energy absorption of the retrofitted beams. 1.3 Approach  To develop hybridfiberreinforced self-compacting concrete it is necessary, first, to find the mix proportion for SCC without fibers. Subsequently, when fibers are added to the mix, the constituent materials of the mix should be changed proportionally to the amount, size, and type of fibers added to ensure that the mix will remain self-compacting. Maintaining enough flowability while thefibercontent is increasing is a major concern in making fiber reinforced self-compacting concrete. Ductile behavior is easier to achieve with higher fiber contents and longer fiber aspect ratios. However, these fiber characteristics also lead to poorer compaction and a lower flowability in freshly mixed concretes. Hence, an effort must be made to optimize the constituent of this composite to induce the maximum possible ductility in the hardened state. The preparation of a highly flowable fresh concrete that does not segregate is very important. In fact, achieving both flowable concrete and a desirable dispersion of fibers is very difficult in SCC. Adjustment of the cementitious materials and admixtures are the most significant factors to alter the fluidity of the concrete. Therefore, this research used theflowabilitycontrol  4  test to maintain a fairly constant fluidity in all freshly mixed concrete by changing the proportions of admixtures and cementitious materials at a given volume fraction of fiber.  1.4 Organization of This Report This thesis includes eight chapters. Chapter 1 describes the background, objectives, approach, and organization of the thesis. Chapter 2 presents a literature review on fresh and hardened properties of self-compacting  concrete with and without fiber  reinforcement. Chapter 3 addresses the experimental program for self-compacting concrete with hybrid fibers with a detailed account of materials, mixes, and test program. Chapter 4 shows the test results from fresh and hardened property tests for mono and hybridfiberreinforced self-compacting concrete. This chapter also focuses on toughness test results where the toughness values are calculated by using three different methods. Chapter 5 presents a discussion of the toughness test results of different mixes and compares the efficiency of different hybrid fibers in SCC in terms of toughness enhancement. Chapter 6 demonstrates the application of SCC containing hybrid fibers in repair and retrofitting of structural beams. In this Chapter, test results of the repaired beams with different types of hybrid fibers are given and discussed. Chapter 7 shows overall conclusions from this thesis and Chapter 8 provides recommendations for future research.  5  CHAPTER 2  Literature Review on Basic Concepts and Fiber Reinforced SCC  2.1. Introduction In this chapter, the rheological properties of concrete in general and self-compacting concrete (SCC) in particular are discussed. First, the concepts of rheology, Newtonian fluid behavior, and the Bingham model are presented. Next, the effect of mix composition on rheological properties of concrete with an emphasis on SCC is discussed. In addition, several proposed test methods for evaluation of fresh properties of SCC are introduced. In the end, the hardened properties offiberreinforced SCC are presented.  2.2. Concept of Self-Compacting Concrete The basic concept of self-compacting concrete was first proposed by Okamura at the University of Tokyo. It is defined as concrete that can be placed in the formwork and compacted without any consolidation effort. It flows and fills every corner of the formwork by its own weight [12]. When concrete is able to flow, it has a low yield stress in its fresh state. Increase of water in normal concrete reduces the yield value but causes segregation of concrete ingredients and clustering of coarse aggregate, and concrete loses its mobility. Addition of superplasticizer to concrete reduces its yield stress without reducing the viscosity of the mix. Due to the appropriate viscosity of the mix, concrete ingredients are kept together and flow without segregation as a composite material in narrow passages. However, if the plastic viscosity of concrete is excessively high, because of viscosity resistance, concrete cannot pass through narrow passages.  6  2.3. Theoretical Background on Rheology of Concrete Rheology is defined as the science of the deformation and flow of matter. In the case of concrete, it shows the relationship between shear stress and shear rate, and explains the mobility and flow properties of concrete before setting [4]. Generally, the rheological properties of concrete are important for the construction industry because concrete is usually cast into moulds in the plastic state. In addition, they influence the hardened properties of concrete such as strength and durability [4, 8, 9]. Due to complexities in the material composition of concrete, there are no specific ways of predicting the rheological behavior of concrete from its mix proportions. Even experimental measurements of the rheological parameters are not easy due to the large range of particle sizes usually used in concrete. Therefore, the fluidity of a given concrete is usually measured by using one of the indirect test methods, which do not exactly define the flow properties of the concrete.  2.3.1. Newtonian Fluid Behavior Since fresh concrete consists of concentrated solid particles (aggregates) in a viscous liquid (cement paste), it behaves as a viscous composite and flows as a viscous fluid. Therefore, the Newtonian equation can be applied to fresh concrete. As shown in Fig. 2.1, an applied force to a liquid can induce a velocity gradient in it. The proportionality factor between the force and gradient is called the viscosity and the liquid obeys the following equation:  F_ A  (2.1) Force [F|  Where, y = Shear rate =dv/dy ^ = Viscosity T  = Shear stress  F = Shear force A = Area of plane parallel to force  Fig.2.1. Newton's Law of Viscous Flow  7  Equation (1) applied to viscous fluids with concentrated solid particles relates the shear stress to shear rate with only one factor, n, the viscosity of the whole system. In some other formulations, a second factor, the yield stress, has been considered. The physical interpretation of this factor is that a finite stress needs to be applied to a material to initiate the flow. This mathematical concept of viscosity and yield stress is shown in Fig.2.2. A fluid that follows this behavior is called a Bingham fluid.  Y  (Shear rate)  Fig.2.2. A Bingham Fluid  Many different equations that relate shear stress and shear rate have been proposed for concrete [5]. By examining these equations, one can easily conclude that in all of them at least two parameters have been used to describe the flow of fresh concrete. Two of these two-parameter equations, which are well suited for concrete, are the Herschel-Bulkley and Bingham equations. X = T + T) y o  x  = x + r) y Q  Bingham equation  Herschel-Bulkley equation  (2.2)  (2.3)  The Herschel-Bulkley equation contains three parameters, but n does not represent a physical concept. It has been shown [10] that in some types of concrete, such as selfcompacting concrete, this equation describes the behavior of fresh concrete very well.  8  Nevertheless, the most commonly used equation today is the Bingham equation, because the parameters used in this equation can be measured independently and the flow of concrete seems to follow this equation fairly well in most cases [4].  2.4. Fresh Properties of Concrete Although the various professional societies have tried to define workability as clearly as possible (Table 2.1) these definitions should all be discarded in favor of parameters that are physically measurable.  Table 2.1.  Examples of Definition of Workability by Various Societies [11]  Name of society  Definition  American Concrete Institute  That property of freshly mixed concrete or mortar which determines the ease and homogeneity with which it can be mixed, placed, compacted, and finished  British Standards Institution  That property of fresh concrete, mortar, or the like, which determines the ease with which it can be manipulated and fully compacted  Association of Concrete Engineers, Japan  That property of freshly mixed concrete or mortar which determines the ease with which it can be mixed, placed and compacted due to its consistency, the homogeneity with which it can be made into concrete, and the degree with which it can resist separation of materials  Fig.2.3 shows the importance of considering both Bingham parameters. It explains how two types of concrete with different behavior in the fresh state can have one identical parameter and a very different second parameter. Therefore, it is important to use a test that fully quantifies the concrete flow, by measuring both factors (intercept and slope). Fig.2.4 shows different types of concrete with different rheological parameters. While high strength concrete has higher viscosity and lower yield stress as compared with normal concrete, SCC shows lower viscosity and lower yield stress compared to high  9  strength concrete. The lower yield stress causes SCC flows easily and the viscosity should be proportional to the yield stress in order to prevent segregation in the SCC mix.  Same yield stress but Different Viscosity  Same Plastic viscosity but Different Yield Stress  Fig.2.3. Concrete Rheology Parameters  Unfortunately, most existing test methods measure only one factor, related either to the yield stress or to the viscosity. Tests measuring both parameters exist but are not practical for field use and hence are not widely used.  Shear Stress  Shear rate Fig.2.4. Comparison the Rheological Behavior of Different Types of Concrete  10  It is obvious that these parameters are related to the properties of the components and mix design (i.e., w/c ratio, aggregate content, type of cement, admixture dosage, etc). However, no predictive models for the behavior of fresh concrete based on the properties of the components have yet been successfully developed. One difficulty comesfromthe fact that the size range of the particles used is very wide. In addition, there is no linear relationship between the rheological parameters of cement paste and concrete. Finally, as indicated by Ferraris et al. [12], the rheology of cement paste itself varies with cement content and w/c ratio.  2.5. Fresh Properties of S C C The main functional properties for fresh SCC are described by filling ability, resistance to segregation and its ability to pass through narrow opening during casting. The concepts of these properties can be described as follow [13]: 1- Filing ability is the complete filling of formwork with a homogenous concrete and flow of the concrete horizontally and vertically within the formwork. 2- Resistance to segregation is the ability of fresh concrete to maintain homogeneity throughout the mixing, transportation and casting process. 3- Passing ability is the ability of the concrete to pass through narrow sections in the formwork without being blocked by interlocking of aggregate particles.  2.5.1. Filling ability The 'Filling ability' involves both the flow and rate of flow of concrete. The flow means how far concrete can flow from its discharge point and the rate of flow means with what speed concrete can flow. In a slump test, flow can be evaluated by measuring the final slump flow diameter of the concrete after the concrete stops flowing. The speed of flow can be evaluated by the time in which it reaches a specific diameter. To have a good 'filling ability', there should be a good balance between flow and rate of flow. For instance, if concrete has very low water/cement ratio and a high amount of  11  superplasticizer, it will have a significant flow capacity but very low deformation velocity or speed. Therefore, it will not have a good 'filling ability'. To reach a good filling ability in concrete, two factors should be considered: •  Inter-particle friction  •  Flowability of paste  In order to have a flowable concrete, it is useful to reduce the friction between aggregates including coarse and fine aggregates. Increase of paste content or decrease of aggregate content in the mix can increase the inter-particle distance between aggregates and hence, reduce the friction between aggregates.  Limited size and content for coarse aggregate  Superplasticizer  Low water/powder ratio  High segregation resistance  Self-compacting properties  Fig.2.5 General Approach to Achieve Self-Compacting Concrete  12  Use of naturally rounded aggregate is also helpful to reduce the inter-particle contact especially when the inter-particle distance is not large enough (volume of aggregate is high). To increase the inter-particle distance between particles of fine (powder) materials a surface-active agent, superplasticizer, is needed. The shape of thefinepowder particles is important for determining the water demand and superplasticizer dosage. Using fines with a spherical shape, such as fly ash, is well suited for this purpose. Since the reduction of inter-particle contact is not enough for having a good flowable concrete, the use of flowable paste in concrete is another essential factor to enhance the flowability of concrete. Addition of water to the concrete mix increases the inter-particle distance, which leads to a reduction in the yield value. It also reduces the viscosity of concrete mix, which results in segregation in the mix. Incorporation of a superplasticizer, on the other hand, reduces the yield value (better flowability) and causes onlya limited decrease in viscosity. Therefore, highly flowable concrete can be attained with superplasticizers without any significant reduction in the cohesiveness. Another factor that affects the flowability of concrete is the water/powder (w/p) ratio. Powder includes cementitious materials and various fillers. Reduction in w/p ratio increases the viscosity of the mix, which results in a reduction in the deformation velocity. Overall, in order to have an adequate filling ability in the mix the following points should be considered. •  •  Increase the flowability of the paste by using: •  Superplasticizer  •  Balanced w/p ratio  Reduce inter-particle friction by using: •  Lower coarse aggregate volume  •  Optimum amounts offillerand cement  2.5.2. Resistance to Segregation Segregation in fresh concrete results in non-uniform material distribution, which in turn, causes a concrete structure with variation in properties in its different parts. One source  13  of segregation is the moveable water in the mixture. It leads to water bleeding in the mixture and segregation of paste and aggregate. Another source is the presence of high amount of coarse aggregate or use of large size aggregate in the mix; both of them cause segregation in heavily reinforced members. In order to have an adequate resistance to segregation, the following points should be considered (13). •  •  Reduce separation of solids by using: •  Limited aggregate contents,  •  Aggregate with smaller size,  •  Low water to powder ratio,  •  An appropriate viscosity agent  Minimize Bleeding by using: Low water to powder ratio, Powders with high surface areas, An appropriate viscosity agent  2.5.3. Passing A b i l i t y  While there are narrow openings in the formwork or the reinforcement is congested, another requirement has to be fulfilled to prevent blocking of coarse aggregate. The maximum size and amount of coarse aggregate on the one hand and the minimum clear spacing between reinforcing bars and formwork opening on the other hand are the major factors in blocking of concrete flow. When the cross section of the concrete flow path is reduced, the possibility of contact among the aggregate particles increases, leading to the formation of an arch around the openings. This arch will be developed more easily when the size of aggregate is large and the content of aggregate is high (Fig 2.6).  Fig.2.6. Mechanism of Blocking Due to Arching (13)  14  In order to have an appropriate passing ability, the following points should be considered: •  Compatible clear spacing  •  Coarse aggregate characteristics •  Lower coarse aggregate volume  •  Smaller size of aggregate  2.6. Materials in Development of Self-Compacting Concrete The role of ingredients in SCC is much more important than that in normal concrete because in SCC the compressive strength is not the only criterion for mix design. In fact, both adequate flowability and strength are the main criteria in SCC development. The main difference between SCC and normal concrete in terms of their ingredients is the large amount of cementitious materials used in SCC to increase its flowability. However, other parameters such as aggregate proportion, size, shape and admixture all have an influence.  2.6.1. Cementitious Materials In structural concrete, the cement content is between 250 to 450 kg/m depending on 3  strength and durability requirements. The amount of cementitious materials in selfcompacting concrete is usually in the range of 500 to 600 kg/m . The cement content of 3  the SCC mix can be reduced effectively by partial replacement with supplementary cementitious materials. In addition, when supplementary cementitious materials are used properly in SCC, they will increase its resistance to carbonation, chemical attack, and freeze-thaw cycling [14]. Thus, when concrete durability is a big concern, use of SCC is very beneficial. The powder materials used in the SCC are usually Portland cement, fly ash, limestone powder, blast-furnace slag, and silica fume. Use of greater amounts of cementitious materials increases the paste volume in the mix and induces higher plastic viscosity in the fresh concrete.  15 Cement The high flowability of SCC is initially due to good dispersion of cement particles and it happens because of the presence of chemical admixtures. A substantial amount of chemical admixture is firstly adsorbed by the available C3A and C4AF in the cement within a few minutes after mixing with water. Therefore, for any kind of cement which is rich in C A and C4AF, the adsorption of admixture will not be uniform and dispersion of 3  cement is not adequate. In addition, ettringite, the hydration product of C3A and C4AF with gypsum, binds with cement particles and reduces the flowability of concrete. Therefore, Portland cement with moderate heat has a more effective dispersing action in SCC [15]. However, use of CSA Type 10 Portland cement is also popular in SCC projects. Particle size of cement is important in SCC for two reasons. First, finer cement particle size results in larger specific surface area of cement, which in turn, creates a higher plastic viscosity in the mix. Second, in cements with large specific surface areas, the number of powder particles in a unit volume of paste increases and as a result, the distance between the particles decreases. This increases the collision and contact points between cement particles. Hence, as SCC mix containing finer cement particles needs more superplasticizer to have the same yield value in the fresh state. Fly Ash The spherical form of fly ash particles creates a "ball bearing" effect in fresh concrete, which is very useful for flowability of self-compacting concrete. Replacement of a part of cement in the mix by fly ash shows that it can retain the flowability of concrete for a few hours. This can be considered as one of the useful properties of fly ash in SCC. It has been reported that with the replacement of 15% to 20% of cement with fly ash, the high flowability and segregation resistance in SCC can be maintained for three hours after production [16], Fig. 2.7. Moreover, it is an important factor in avoiding slump flow loss. The vast availability of fly ash at low cost, makes it an ideal cementitious material for use in SCC, and it is possible to replaced up to 50% of the cement [16]. Therefore, when very high strength or high early age strength for concrete is not an issue; use of fly ash is  16  beneficial for both fresh and hardened SCC. Water demand of fly ash in fresh concrete is less than that of cement. Hence, replacement of fly ash with cement in a mix improves the flowability of concrete while its water content is constant or if the workability of concrete needs to be constant, the water content can be reduced. According to the literature, the water content of the concrete mix can be reduced by as much as 13% for a 30% replacement of cement with fly ash [17]. However, the major factor, which influences the workability of concrete, is the particle size of fly ash. It has been shown that substitution of 50% of the cement with fine fly ash particles (<45um) can reduce the water demand by 25% but the same substitution with fly ash with particles greater than 45pm showed no effect on water demand [18] because the particles coarser than 45 um remained unreacted in the mix.  700 650 600  i » c  o  D.  5U0  a  ^  +50 400  1  " SF:(C+F>1.8% ,  350  VA;9G0gi'raCIS-FA -! Steel fibcnl % by volume [  300  0  30  60  >0  l  110 150 ISO 210  Elapsed Time, minutes  Fig. 2.7.Relationship Between Slump Flow and Elapsed Time With Different Replacement Ratios of Fly Ash [16] Silica Fume Although silica fume is not an essential material for self-compacting concrete, it is very useful when a high strength SCC needs to be produced. The higher strength is produced due to the filler effect and the excellent pozzolanic properties of silica fume, which make the transition zone of the paste-aggregate interface stronger [14]. However, use of silica fume may increase the needed dosage of superplasticizer in the SCC mix to maintain the  17  workability of concrete constant. On the one hand, the addition of silica fume reduces friction between the bigger particles and causes a lubricant effect that improves the flowability of an SCC mixture. This effect is more important when the water/binder ratio is low. In addition, silica fume absorbs the mixing water, increases the viscosity of fresh SCC, and makes it sticky. A very high viscosity mix does not have enough flowability and it may not be well consolidated by its own weight. Nevertheless, an appropriate amount of silica fume accompanied by fly ash and a high dosage of superplasticizer in the mix can produce self-compacting concrete with compressive strength exceeding lOOMPa [19]. The optimum dosage of silica fume is very important because a higher amount decreases the flowability of fresh mix, which in turn, causes reduction in strength of hardened SCC. The optimum dosage of silica fume for high strength SCC is depended on other ingredients in the mix such as cement and superplasticizer, and it may vary from 5% to 10% of cementitious materials [19]. SCC with silica fume is less prone to segregation. It changes the rheological properties of SCC and reduces bleeding in the mix. This behavior is primarily due to the high affinity of silica fume for water. Hence, it reduces the amount of available free water for bleeding in the mixture [14]. It has also been shown that silica fume has an accelerating effect on the first hydration reaction and on setting time of SCC [14]. However, use of fly ash simultaneously with silica fume can compensate for the fast setting properties of silica fume in fresh SCC.  2.6.2. Limestone Powder Although this material does not have any pozzolanic effects in concrete, it is used for improving the rheological properties of self-compacting concrete. Limestone powder is usually added into SCC used in massive structures. Since a large amount of cementitious materials is used in SCC, temperature rise of concrete can be a problem in massive structures, where thermal cracks may occur. Therefore, using limestone powder in SCC helps prevent the temperature rise and thermal cracking in massive structures. Arima et al. [15] have combined limestone powder with low heat cement and developed selfcompacting concrete with the adiabatic temperature rise value restricted to less than 25°  18  C. This type of concrete has been used in the foundation of anchorage and framework of the Akashi Kaikyo Bridge in Japan [15]. In addition, cost saving can be another potential benefit in limestone utilization. However, it is not always the case for SCC containing limestone powder. The surface area of limestone powder can influence the rheological behavior of the SCC mix. Limestone powder with very fine particles increases the yield stress of fresh SCC and decreases its slump flow. Hence, more superplasticizer is needed to offset the reduction of SCC flowability. [20].  2.6.3. Admixtures It is often emphasized that admixtures play a more important role than that of the other ingredients in SCC. The most important admixtures used in SCC are superplasticizer, air entrainer, and viscosity modifier. Sometimes use of admixtures in the concrete creates problems in a freshly mixed SCC, including fast setting of concrete, variation of slump flow, and large loss of concrete slump. Such phenomena are interaction problems between the cement and the admixtures. The rheological properties of concrete vary largely according to its ingredients, preparation conditions, and the surrounding environment. Dealing with compatibility between cement and superplasticizer is not a simple issue because there are many possible ways in which superplasticizer molecules can interact with different cement phases. There are also interaction problems between the admixtures in a concrete mix. For instance, sometimes there is incompatibility between the superplasticizer and the viscosity modifier agent itself [21]. In addition, it has been seen that using a superplasticizer and air entraining agent at the same time has shown a synergistic effect in producing higher air contents in concrete [22]. Superplasticizer So far, naphthalene sulfonate- or melamine sulfonate-based superplasticizer have been mainly used in concretes, but recently polycarboxylate type superplasticizers have  19  become more popular in self-compacting concrete because they act as highly effective dispersants. Even at small dosage rates, they provide high fluidity, low viscosity and high slump retention for concrete while the w/c ratio is low. In general, the dispersion mechanism of cement particles due to superplasticizers can be divided into two types: 1- Based on electrostatic repulsion 2- Based on steric repulsion Superplasticizers based on electrostatic repulsion include naphthalene sulfonate and melamine sulfonate. They give negative charges to cement particles and stabilize the dispersion. Hence, the cement paste containing cement particles with negative potential is fluidized because the cement particles repel each other.  On the other hand, dispersion mechanisms based on steric repulsion, may be reduced by polycarboxylate superplasticizers, which have polyethylene oxide chains in their chemical formulations. The side chains of polyethylene oxide extend on the surface of the cement particles [23]. They have a strong ability to hold water and create a bulky and thick adsorption layer on the cement surface. This thick adsorption layer generates high steric repulsion. Since a large amount of superplasticizer is used in SCC, there is a tendency for concrete setting to be retarded. Initial setting and final setting are delayed linearly with respect to the dosage of superplasticizer [24]. The variation of w/c has no significant effect on the fluidity of cement paste when w/c ratio is high, whereas it becomes significant at lower w/c because in this case, a slight fluctuation of water content in concrete can cause large variations in its fluidity [24]. Viscosity Admixture When the plastic viscosity is not sufficient to prevent segregation in fresh SCC, the use of a viscosity agent in the mixture is unavoidable. This occurs in concrete when the ratio of water to powder is not low enough or the amount or size of coarse aggregate is more than what it should be in the mix.  20  The mechanisms by which a viscosity agent helps the mix to resist segregation is divided into two types: 1- The mechanisms that act in cement 2- The mechanisms that act in water in concrete In the first type, the adsorptive viscosity agent is attached on the surface of cement particles. The mortar fluidity decreases as its viscosity increases. When more viscosity agent is added, a greater surface area of the cement particles is occupied by the viscosity agent. Hence, less superplasticizer is in turn adsorbed by the cement particle surface, which reduces the fluidity of concrete. In the second type, the apparent plastic viscosity of concrete increases with an increase in the admixture dosage, but the flow value does not change because the viscosity agent does not compete with superplasticizer to occupy the surface of cement particles and the amount of adsorbed superplasticizer does not change. This type of viscosity agent is well suited for SCC mixes.  2.7. Mix Design It should be considered that for normal concrete the mix designers usually change the proportion of materials to make a concrete mix with desirable hardened properties. However, in SCC the initial attention is given to the fresh properties of concrete. Since SCC has more ingredients than conventional concrete, and since the specifications of  powders  (cement,  fly  ash,  silica  fume,  limestone  powder), admixtures  (superplasticizers, air entrainer, viscosity agents), and even aggregate may vary from one place to another, it is very difficult to build up one widely acceptable mix design for SCC. However, a number of SCC mix design methods have been developed by the universities and institutes in Japan, Sweden, Scotland, France, and Canada based on the local chemical admixtures [25]. It can be seen in the literature that these mix designs have been modified over time due to availability of more efficient admixtures. In addition, some admixtures such as viscosity agents have been removed from the list of SCC ingredients.  21  Therefore, a reasonable way for producing SCC is to control the needed characteristics of SCC by doing the related tests. These characteristics, as explained in section 2.5, are flowability, filling/passing ability, and segregation resistance of SCC, all of which can be measured by the proposed test methods. Another approach for producing SCC is by considering the rheological behavior of fresh SCC. The proportions of ingredients can be adjusted by measurements of the rheological parameters of the mix. However, this method has several problems. First, the process of measuring the parameters is very complicated. Second, the equipment is very expensive, and every institute has developed its own equipment, which has not been standardized yet. As a result, different equipment gives different numbers for the rheological characteristics of concrete and this method cannot be considered as an efficient way to develop a mix design. The most effective way of producing SCC with desired characteristics is to make trial batches. However, there are several mix parameters which can be varied in the batches: water/binder ratio, binder content, coarse aggregate content, and chemical admixture dosage. In order to compare and evaluate the results of different batches, several test methods have been developed to confirm that the proposed mix has the desired properties [26]. In the case of fiber reinforced SCC, as a first approximation, a successful mix design for plain SCC can be used. In the next step, the proportions should be modified based of the type and amount of fiber in the mix. However, it is not possible to predict the mix composition of fiber reinforced SCC from a given plain SCC mix design.  2.8. Test Methods for Fresh S C C There are several test methods, which are useful for evaluating the fresh properties of SCC. Although none of them has been standardized yet, and people use different equipment or different sizes of apparatus based on their experience and the project, some of them are very useful and applicable. However, it is not possible to compare the results from different labs and sites. In the following sections, some of the most widely used test methods for evaluating the fresh properties of SCC are reviewed.  22  2.8.1. Slump Flow Test This test method due to its simplicity is very popular for SCC. It evaluates the flowability of fresh SCC from observation of the deformation speed and measurement of the spread diameter of the deformed sample under its own weight. Unfortunately, however, this test cannot be a good indicator for all fresh properties of SCC. For instance, it does not show whether the fresh mix has enough viscosity or not, which is very important for SCC evaluation [27, 28]. This method is also applicable for fiber reinforced SCC. However, adding fibers into plain SCC decreases its flow diameter. Furthermore, in this test method, segregation and dispersion of fibers can also be observed. In a good SCC mix, the spread of the concrete should be circular with homogenous distribution of the solids without any cluster of material at the center of the flow. The flow required for self-compactability, based on the literature is from 600 mm to 700 mm for plain SCC [29, 30].  Slump Flow =(A+B)/2  Fig.2.8 Measurement of Slump Flow in SCC  23  2.8.2. V -Funnel Test  This test evaluates the passing ability of concrete through a narrow opening which indicates with concrete viscosity. The flowing speed of concrete through the funnel under its own weight is a criterion for evaluating the behavior of fresh SCC. In addition, the variation of the flowing speed of concrete in the funnel can also be used to evaluate the segregation resistance of freshly mixed SCC (Fig.2.9).  2.8.3. L - B o x Test Fig 2.10 shows the L-shaped box test apparatus used to evaluate the self-consolidating property of fresh SCC. It can also be used for evaluating other properties such as filling ability, passing ability, and segregation resistance. Segregation and passing ability of concrete can be recognized visually in this test. When segregation occurs, coarse aggregate accumulates behind the reinforcement and the concrete cannot pass easily through the reinforcement [13].  24  The ratio H2/H1, as shown in Fig.2.10, can show the filling ability of concrete. Although there is not an exact number for this ratio, the Swedish experts [13] suggest a value between 0.8 and 0.85. In addition, the time for concrete to flow a distance of 20 cm and 40 cm from discharge point in the box is recorded. These numbers show how easily this concrete can flow. The traveling time for the concrete with very high viscosity is longer than that of SCC with adequate viscosity. High viscosity reduces the filling properties of SCC. The L-box test for fiber reinforced SCC is used without blocking re-bars [30]. Hence, this test shows the self-leveling properties rather than the blockage. The reason for not using the blocking re-bars in SCC with fiber is that the gap distance between re-bars is too small. It may be assumed that fibers in concrete can be replaced with traditional re-bars such as in slabs on ground. Therefore, the blocking criterion is not a deciding factor in the mix design. However, for structures where the fiber is used in combination with re-bars, it is desirable to know the limitation of minimum gap distance.  Fig. 2.10. Principle of L-shaped Box for Evaluation of Self-Consolidating Properties  2.8.4. J-Ring Test Method This test method has been proposed to study the blocking criteria of fiber reinforced SCC [25]. Its apparatus, as shown in Fig.2.11, is a ring, holding a certain number of rebars. It is used in combination with the slump flow test. The concrete flows through the re-bars 25  as the cone is lifted. The J-ring provides a higher degree of freedom in the choice of gap distance between re-bars as compared with L-shaped box. Therefore, to study the passing ability of fiber reinforced mixtures with different fiber types and contents, different bar spacing should be adopted. The height of concrete after lifting the slump cone without using the ring and after using the ring are measured and shown by H and Hb respectively. a  The ratio of H /Hb shows the blocking possibility in the fiber reinforced SCC. The a  passing behavior of a mix depends on the content and maximum size of coarse aggregate, fiber length,fiberfactor (VjLf/df) and stiffening of the mixture. According to Nemegeer [29], blocking is assumed to occur if the difference between the heights of the concrete inside and outside the J-ring is larger than 10 mm. This test method is the only test used for SCC withfiberand is not applicable for plain SCC.  Fig.2.11. In the J-Ring Test the Cone is Filled With SCC and Lifted and the SCC Flows Through the re-Bars.  2.9. Role of Fibers in Behavior of Fresh SCC In general, addition of fibers into a concrete mix causes undesirable changes in the fresh properties of concrete (i.e. it reduces the workability of concrete). Although long fibers are more efficient in terms of toughness, they create constraint in mobility and movement of fresh concrete in the forms. Long and soft fibers have a tendency for balling and it is difficult to disperse them uniformly throughout the mix due to their large surface area. They restrain the flowability  26  of concrete and result in fiber balling in the mix, which can have negative effect on the hardened properties of SCC. While stiff fibers are easier to disperse during mixing, due to fiber interlocking they cause a greater workability reduction in the fresh mix [8]. The addition of fibers has a strong influence on the two important properties of fresh SCC: 1- They reduce the consolidation properties of fresh SCC 2- They reduce the flowability and mobility of SCC  The extent of reduction of concrete workability depends on size, shape, material, and amount of fibers. In addition, it depends strongly on the concrete constituents. Decrease in quality of self-compacting properties of freshly mixed SCC induces a substantial loss in mechanical properties of hardened SCC. Therefore, in order to have a fiber reinforced SCC mix with suitable fresh properties, the amount of powders and admixtures should be adjusted based on the fiber type and content. For instance, for different volume fraction of fibers the constituents of SCC cannot be kept constant and they should be adjusted to produce a good self-compacting properties concrete with desirable flowability. Hence, with any changes of fiber in the mix, (volume, aspect ratio, type), the proportion of ingredients and admixtures may need modifications. On the other hand, changing the mix proportions in SCC may alter some mechanical properties of hardened concrete such as compressive strength. Therefore, if the mechanical properties of hardened concrete need to be kept constant while different fiber volume fractions are used the mixture may need more adjustment for its ingredients to have the desired hardened and fresh properties. Hence, a number of trial batches may be necessary to get the right proportion for a fiber reinforced SCC mix. Moreover, for each type of fiber, there is an upper limit in volume fraction Vj to have desirable self-compacting properties. This can be determined based on the materials and admixtures used in SCC. It means by increasing thefibervolume fraction at a certain Vf, SCC loses its fresh properties and would not be considered as SCC any more. Therefore, it can be defined as a critical fiber volume fraction for fresh SCC. If the mix has more fiber than the critical value, the mixture does not flow uniformly in circular shape and the  27  cluster of material can be seen at the center of the flow circle [29]. This has been shown in Fig.2.12.  I (a)  (b)  Fig.2.12. Non Uniform Flow Patterns (a) Restricted Flow (b) Clustering of Materials  One way of determining the critical value of afiberin fresh SCC is to use the J-ring test method. The results of the J-ring tests show that there is a certain correlation between the ratio of concrete height before and after using the ring (H /Hb), and the ratio between a  opening gap (S) andfiberfactor (LfVf/df). When the mix has enough flowability, (H /Hb) a  is about one. It has been shown that at a certain point, H /Hb increases rapidly and that is a  a critical point for S/(LjVf/dj) [30]. If (S) and aspect ratio Lf/df are considered constant, the critical Vf can be calculated based on the graph gained from the J-ring test [30], as seen in Fig.2.13  Blocking J-Ring Test  7  Critical point  4  s/(L,vyd)  Fig.2.13. Blocking is Presented Based on Relation H /H Versus re-Bar Opening Gap S and Fiber Factor L V /d a  f  f  (H and H are heights of concrete before and after using J-ring) a  b  28  b  2.9.1.Influence of Steel Fiber  SCC in terms of rheological behavior has a lower yield stress and higher viscosity as compared with normal concrete in its fresh state. These properties help steel fibers to be dispersed more uniformly in the mix. Steel fibers are hydrophilic fibers and adsorb more water compare to synthetic fibers. Incorporation of fiber in SCC increases the viscosity and yield stress of the fresh mix. However, the increase of yield stress is much higher than that of the viscosity. Changes in the viscosity are attributed to water adsorption by the fibers and changes of w/c ratio in the mixture. On the other hand, interlocking of fibers increases the yield stress in the fresh concrete. Hence, adding more fiber will increase the viscosity and yield stress in the fresh mix. The stiffness of steel fibers compared with other fibers, creates greater tendency for fibers to be interlocked with coarse aggregate and between themselves. It must be noted that higher aspect ratio induces more interlocking as well. Micro-steel fibers, due to this high surface area, increase the viscosity of the mix. However, using an appropriate proportion of superplasticizer can help to improve the concrete flowability. In the case of using higher volume fraction (V/) of steel fiber in the mix, more superplasticizer should be added but this trend cannot always be helpful. Sometimes adding too much superplasticizer may result in segregation in the mix. Therefore, when high V/ of steel fiber is incorporated in the mix, the proportion of the constituent materials should be modified so that the mixture has self-compacting properties without any segregation.  2.9.2.Influence of Carbon Fiber Carbon fiber is a hydrophobic material. However, it uses a part of the water of fresh concrete because of its fineness (around 10pm diameter), and its high specific area. By adding carbon fiber, the ratio of w/p is reduced and the viscosity of the mixture is increased. On the other hand, the yield stress in the fresh mix starts increasing and since carbon fibers are soft, the rate of this increase is not high as compared with stiffer fibers (i.e. steel fibers). It has been shown that in normal concrete (Fig.2.14), how the  29  rheological parameter g, which is proportional to the yield stress and obtained from a viscometer, increases with an increase in thefibervolume fraction [31].  100 -|  0  ;  ;  0.2  0.4  ;  0.6  1  0.8  Fiber Content % Fig.2.14.  Rheological Parameter g as a Function of the Fiber Volume Fraction [31] (g is a parameter gained from viscometer and proportional to yield stress)  Therefore, it can be concluded that adding carbonfiberin SCC, results in reduction of SCC flowability. However, adding superplasticizer can regain the flowability of carbon fiber reinforced SCC even at 0.5% fiber volume fraction. Superplasticizer can be increased to the extent that segregation does not occur in the mix. For higher fiber volume fractions, a different mix design with less coarse aggregate, more fine aggregate and powder materials is required so that the mix can keep its self-compacting properties. The high viscosity of SCC is not beneficial for uniform dispersion of carbon fibers in the mix. On the other hand, due to the brittleness of the carbonfiber,it breaks easily during the mixing process in concrete and loses its mechanical efficiency. However, due to the low yield stress in fresh SCC, there is less friction between the coarse aggregates. Hence, more carbon fibers can survive during the process of mixing in SCC. Increase in the volume fraction of carbonfiberin the SCC mix induces more air content in the mixture especially when the mix has an air-entraining agent. On the one hand, this is good because it reduces the yield stress in the fresh mix and helps the self-compacting properties to be enhanced. On the other hand, it is not desirable because it compromises the hardened properties of SCC.  30  2.9.3.Influence of Micro-Polypropylene  During mixing of micro-polypropylene fibers in concrete, they tend to separate from each other. Dispersion of such fine fibers in the concrete has a considerable influence on the properties of the fresh mix. It has been shown that adding micro-polypropylenefiberto a concrete mixture gives an immediate reduction in workability [22]. As expected, an increase in fiber content will result in a reduction in concrete consistency and this reduction is greater for longer fibers and higher fiber contents [22]. Since adding fiber reduces the consolidation property of SCC, it entraps a considerable amount of air in the mix, which damages the hardened properties of SCC. Use of micro-polypropylene fibers also has a significant effect on the rheological properties of fresh SCC. This type of fiber increases the viscosity of the fresh mix because of its very fine diameter and high specific area, which in turn, needs more water to wet the surface of fibers. This induces a reduction in available water in the mix. Longer fibers increase the yield stress of fresh SCC and it cannot be flowable anymore. The fibers behave like a network to hold the fresh mix and prevent it from flowing. For solving this problem, more superplasticizer is required to compensate for the reduction in the flowability of SCC. However, the amount of micro-polypropylene which can be used in a SCC mix, according to this study is very limited. To use greater fiber contents a different mix design withfineringredient materials and higher volume of superplasticizer is needed.  2.9.4.Influence of Hybrid Fibers So far, no research is reported in the literature aimed at assessing the influence of hybrid fibers on fresh properties of SCC. However, researchers have shown that adding a single fiber type to SCC reduces its workability and flowability. Usually to improve the mechanical properties of fiber reinforced concrete, the hybrid fiber system consists of both macro and micro fibers. Micro-fibers have more influence on fresh properties than macro-fibers and using both types in the same mix will dramatically reduce the workability of SCC. Therefore, micro-fibers should be limited to less than a few tenths of a percent so that SCC can keep its fresh properties. Using high volume fraction of hybrid  31  fibers in SCC (a total of more than 1%) results in low consolidation and entraps significant air in the mix. Therefore, SCC will need a high volume of superplasticizer to improve its workability and reduce the air content in the mix. In addition, the mix proportions need to be adjusted to counteract the large volume of hybrid fibers in SCC and reduce the yield stress in the mix.  2.10.Hardened Properties of Fiber Reinforced S C C The main function of fiber in SCC, as in normal concrete, is to control the crack propagation and crack widening after matrix cracking. Control of cracking in concrete enhances the material's ability to deform and absorb greater energy. While the effect of fibers on the pre-cracking behavior of concrete is not significant, their influence on cracked concrete is well known. In this stage, fibers suppress crack growth by reducing stress intensity at the crack tip.  2.10.1.Steel Fiber Reinforced SCC Adding a limited volume fraction of steel fiber has no effect on concrete properties such as compressive and tensile strength, Young's modulus, shrinkage and creep, etc. However, the flexural toughness of SFRC is remarkably higher than that of plain concrete. As indicated by Groth (32), comparison between toughness results of steel fiber reinforced SCC and that of SFRC with normal concrete, based on ASTM C1018 toughness indices, showed that both materials performed equally well for equal matrix strength and the same type and amount of fibers. It was also concluded that for fibers with different configuration, the same parameters, which affected SFRC with normal concrete also were effective for steel fiber reinforced SCC. For instance, when the fiber amount increased, the toughness would also increase and an increase in the aspect ratio of fiber enhanced the composite toughness only up to a point where fiber rupture started to occur [30].  32  The design rules for SFRC members such as slabs on grade are based on material toughness properties. Since SFRSCC mixes show no major differences in toughness with SFRC, the same methods and design criteria can be applied for SFRSCC. Casting without vibration is the major difference between normal SFRC and SFRSCC. It was shown that self-compacting capabilities had no damaging effects on the fiber-matrix bond and toughness values [30]. The compressive strength of the SFRSCC has the same general trend as SFRC. The early strength is rather low, probably due to a retardation effect of admixtures, but the final strength is higher as compared with normal concrete with the same w/p ratio, which is due to use of higher amount of powder in SCC [30]. 2.10.2. Polypropylene Fiber Reinforced SCC Polypropylene is one of the most widely used types of fiber reinforcement in concrete. Polypropylene fibers are manufactured by drawing the polymer into thin film sheets, which are then slit and further processed to produce fine fibers. These fibers can be produced in a variety of configurations such as monofilaments,fibrillated,and collated bundles [22]. Monofilaments, which are called micro-fibers, disperse evenly but are somewhat difficult to handle, while collated fiber bundles, which are called as macrofibers can be dispersed easily in concrete. Polypropylene has several unique properties that make it quite suitable in concrete reinforcement. It is chemically inert and stable in the alkaline environment of concrete, and relatively low cost [22]. In general, the addition of macro-polypropylene fibers increases concrete flexural toughness,flexuralfatigue strength, endurance, and impact resistance [22]. Since polypropylene has a low modulus of elasticity, using micro-polypropylene in low volumes in concrete as a single reinforcement cannot provide highflexuraltoughness. However, their fine sizes can be effective in bridging micro-cracks, which provides resistance to fast crack propagation in the concrete matrix.  33  Polypropylene fibers do not significantly increase the compressive strength; in fact, compressive strength may be reduced with the addition of fibers if adequate compaction is not achieved [24]. It seems that permeability is also increased when polypropylene fiber reinforced concrete is not under load because the poor bond between the fibers and the concrete matrix allows passage for external agents [33]. However, a study by Soroushian and Mirza showed no significant change in permeability with 0.1% fiber by volume. They suggested that at slightly higher dosage rates, permeability might be decreased due to the reduction of plastic shrinkage cracks. In addition, since the use of micro-fibers in concrete inhibits the crack growth of micro-cracks in the concrete matrix, their use reduces the permeability [34].  Plastic shrinkage cracking is one of the main reasons for cracking in the concrete. Plastic shrinkage cracks occur during thefirstfew hours of concrete curing due to a quick loss of moisture. It is important to limit the amount of plastic shrinkage cracking in concrete to maintain low permeability and good resistance to deterioration. A crack is an easy way for water, chlorides, and oxygen to invade the concrete and cause freeze-thaw damage, and attack steel reinforcement. When micro-polypropylene fibers are used at low dosages (less than 0.3% by volume), they provide effective secondary reinforcement against plastic shrinkage cracking [22]. In a study by Kovler et al. [35], it was demonstrated that using 0.3% fibers by volume reduced plastic shrinkage cracking to such an extent that no cracks could be observed. In addition, lower volumes of fibers (0.05 to 0.2% by volume) visibly restrained crack width compared to samples that were not reinforced with micropolypropylene fiber.  Because of the adverse effects of micro-polypropylene fibers on the fresh properties of SCC, it cannot be used at large volume fractions. It should be limited to 0.10%~0.25% volume fraction depending on the fineness of the fiber. Even a 0.25% volume of fiber needs a significant adjustment in the proportion of constituents materials in SCC to maintain self-compacting properties.  34  2.10.3.Carbon Fiber Carbon fibers are strong and stiff fibers. They are inert in highly alkaline environment and can be heated up to 2500° C. Due to their fineness the number of fibers in the mix is very high and this explains how they control the micro-cracks more efficiently than the macro-fibers do. They can blunt and arrest micro-cracks before they coalesce into macrocracks leading to fracture. The enhancement of the strain and stress capacity of the composite using carbon microfiber has been shown by Banthia and Sheng as illustrated in Fig.2.15 [36].  700  i J  600  i  ficro reinforcea cement sysi em Cat "bottJtber  500  _400  \ 5%  •D CO  o  300  lr\ \ ° 3  /0  200  100 /A  V  %  D e f l e c t i o n (mm)  Fig.2.15. Load-deflection Response Obtained From Flexural Test for Carbon Fiber Composite  One of the applications of carbon FRC is in thin repair of concrete structures. It ensures strength, toughness and durability for repaired structures [37]. As in any other composite, the matrix strength and the bond betweenfiberand matrix are crucial forfiberefficiency. The interface efficiency betweenfiberand matrix depends not only on matrix strength but also on surface treatment of the fiber. The fiber surface can  35  be modified to increase the adhesion and improve the fiber dispersion, which leads to more energy absorption in the fiber bridging zone of FRC. Researchers have shown that in CFRC, carbon fibers can control the micro-cracks. This issue has been proved by acoustic emission [38]. Banthia[39] has shown that in the case of notched beams reinforced with carbon fibers, specimens carried much higher load at first crack. In addition, the higher values of the stress intensity factor, K i , confirmed c  efficient crack control by carbon fibers. However, it was also shown that at low volume fractions of carbon fibers, i.e. less than 1%, the influence of reinforcement was small and it did not show any noticeable increase in the fracture energy absorption [37]. Flexural behavior of CFRC has also been reported by several other authors [37, 42]. Considerable increase in the flexural strength and fracture energy absorption was observed when the fiber volume fraction exceeded 1%. This is shown in Fig.2.16. In order to quantify the flexural toughness of the composite, ASTM C1018 [32] calculations based on the area under the load-deflection curve, have been used by most researchers [37, 38, 39]. LOAD fN]  0  0.1  0.2  0.3  0.4  0.5  0.6  0.7  0.8  0.9  1  DEFLECTION [mm]  Fig.2.16 Characteristic load-deflection curves for various Vf in CFRC [43]  A slight reduction in compressive strength with respect to the plain matrix has also been reported by several authors. [37, 38, 39]. As a result, no appreciable improvement of the compressive strength may be expected even at high volume fractions of fibers. It seems  36  that adding more carbon fibers further reduces the compressive strength of the composite and apparently results in non-homogeneous distribution of fibers and entrappment of more air voids [37]. The test results of CFRC under uniaxial tension have been reported by several authors [40, 41]. They showed the stiffness of the matrix was increased with increase of fiber content. Also for volume fractions higher than 1%, pseudo-strain hardening was observed and the fracture toughness expressed by the area under stress-strain curve increased as shown in Fig. 2.17.  STRESS [ M P a ]  9 Cmrboo Micro.nbcr  I  I  Matrix: Cement Mortarj  3% Ober  S/C=0.50 W/C = 035  2% fiber  SF/C=0.20  7^1% fiberl No Tiber  0  0.1  0.2  0.3 0.4  0.5 0.6  0.7 0.8 0.9  STRAIN [%]  Fig.2.17. Characteristic stress-strain curves of specimens under uniaxial tension [40]  Short carbon fibers in cement matrix composites also exhibit some other attractive nonstructural properties. They increase the specific heat, reduce the thermal conductivity, enhance electrical conductivity of the composites, and facilitate the cathodic protection of reinforcement in the concrete. Moreover, they show an ability to sense their own strain and temperature [44, 45]. 37  2.10.4.Hybrid Fiber Reinforced SCC Since adding excess fibers in concrete may cause detrimental effects on mechanical properties of the composites, using a large amount of fiber in concrete is not always the most economical way to improve the mechanical properties. Using optimum amounts of different fibers in a hybrid system may be a better way to improve the mechanical characteristics of the composite. In addition, combination of two or more kinds of fibers helps to alleviate different deficiencies in the composite; also, it is useful for precise material design and tailoring the performance of composites. Moreover, the hybrid fiber reinforcement may be less expensive, because the expensivefiberscan be added in smaller amounts and combined with large quantities of a cheaperfiber[37]. According to Mindess and Bentur [22] and Lawler et al. [46], some fundamental approaches behind the use of hybrid systems include: 1. A combination of fiber size such that fine well-dispersed micro-fibers bridge the distributed micro-cracks, leading to an increase in strength by delaying the localization process, and macro-fibers bridge macro-cracks, providing postpeak ductility. 2. A combination offibermaterial properties, where a stronger fiber can add to the strength, while a more ductilefibercan add to overall material ductility. 3. A combination of fiber action over the life cycle of the material, such as prevention of plastic shrinkage at early ages and mechanical performance at later ages. 4. A fiber interaction where one type offibermay improve matrix performance, which increases the effectiveness of otherfibertypes. These approaches are not mutually exclusive and may be combined when an appropriate fiber blend is chosen. A schematic description of the hybrid system is shown in Fig.2.18. It shows that the flexural toughness of HyFRC containing low and high modulus fibers is higher than that of the mono FRCs. [22]  38  Hybrid reinforced  Concrete  Polyethylene + Steel  Fiber  Deformation Fig 2.18 Schematic description of load-deflection of hybrid composite  As soon as load is applied to concrete, short and discontinuous micro-cracks begin to develop in a distributed manner throughout the material [46]. The function of microfibers in concrete is mainly to inhibit the micro-crack initiation and propagation. They may delay the process by which the micro-cracks coalesce to form large macroscopic cracks [46]. The coalescence of the first macro-crack coincides with the peak load and initiates localization, where subsequent deformation is concentrated at the opening of the crack. While they control micro-cracks in different places in the concrete, i.e. in the interfacial zones, the macro-fibers bridge the large cracks, as is shown in Fig. 2.19 [47]. The post peak degradation of the concrete is prolonged and the toughness of the material is increased as these macro-fibers, which carry nearly all the load across a crack in a cracked composite, either pull out or break.  39  Crack development in concrete under tensile load  Microcrack Macrocrack formation formation  Microti bers control Microcracks  c  Macrofibers control Macrocracks  Fig.2.19. Hybrid fibers control; cracks at different levels  The durability of concrete is closely related to its permeability. This is because concrete is susceptible to degradation through the mechanisms that necessitate the ingress of water or other agents. Because cracks significantly increase the permeability of concrete, their presence greatly accelerates the deterioration process. Tsukamoto and Worner [48], and Shah [49] showed that for several reasons including crack resistance, increased tortuosity of individual cracks, crack branching and multiple crack nucleation, the permeability in FRC is less than that in plain concrete. Aldea and Shah [49] showed that while a marginal improvement was seen with macro fiber composites,  micro-fibers reduced the  permeability of cracked composite significantly and the combination of these fibers with macro-fibers further decreased the water flow through the cracked concrete. It was shown that the permeability improvement was due to modification in the crack topography resulting in an increase in multiple cracking in the concrete. The combination of steel and polypropylene fibers and combination of steel and carbon fibers in concrete has been studied by various investigators [3, 46, 47]. It has been reported by Stroeven [3] that introduction of polypropylene fiber into the hybrid system significantly improves the mechanical behavior of the composite. The optimum dosage of polypropylene in combination with steel fiber has been reported to be 0.15% [47]. It has been shown that a good dispersion of micro fibers is critical in a hybrid composite and an excess of fiber content may have an adverse effect on strength and toughness [47].  40  Among the investigations on HyFRC containing carbonfibers,it was shown by Stroeven [50] that a combination of hooked-end steel fibers and PAN based carbon fibers performed efficiently. The carbonfiberscontributed to the post crack behavior of the composite and significantly improved the pull out performance of the steel fibers in the concrete. However, this phenomenon did not occur in the case of pitch-based carbon fibers.  41  CHAPTER 3 Experimental Program for HyFRSCC 3.1. Introduction The  following section consists of a description of the general properties and  characteristics of the various materials used to make self-compacting concrete in this research project. To reach a desirable SCC mix with acceptable fresh properties, several mix designs were proposed and tested. Different combinations of fibers were used and fresh properties of the mixtures were measured to ensure that they could be considered as SCC.  3.2. Materials 3.2.1. Portland Cement The cement used in this project was manufactured by Lafarge Canada Inc. It was classified as CSA Type 10 Normal Portland Cement. In order to avoid using partially hydrated cement and its detrimental effects during the mixing operation, special care was taken to use only fresh cement. The chemical composition of this cement is listed in Table 3.1  Table3.1. Chemical composition of the cement  cs 3  CS 2  C A 3  C AF  CaO  Si0  2  A1 0  (%)  (%)  (%)  (%)  (%)  (%)  (%)  66  13  6.4  11  64.78  20.18  4.72  4  2  3  MgO  so  (%)  (%)  (%)  3.63  1.01  2.74  Fe 0 2  3  Loss  3  on ignition  2.53  Alkalies  (%) 0.55  3.2.2.SiIica Fume In order to get a high strength SCC, silica fume was added to the mixture. According to the supplier, the chemical composition of the silica fiime used in this study is presented in Table 3.2.  42  Table 3.2.Chemical composition of silica fume  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  so (%)  91.7  0.07  0.01  0.04  0.05  0.09  0.05  0.23  0.12  0.6  0.28  Si02  Al 0 2  Ti0  3  2  PiOs  Fe 0 2  3  CaO  CrO  MgO  Na 0  K 0  2  2  3  Loss on Ignition (%) 6.55  3.2.3. Fly Ash The fly ash used in this project was classified as ASTM Class F. It was manufactured in Alberta and supplied by Ocean Concrete Products Limited. Its chemical composition is presented in Table 3.3. Table3.3. Chemical composition of the fly ash  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  (%)  so (%)  52.4  23.4  0.01  0.2  4.7  13.4  0.08  1.3  3.6  0.6  0.2  Si02  Al 0 2  3  Ti0  2  P205  Fe 0 2  3  CaO  Ti0  2  MgO  Na 0  K 0  2  2  3  Loss on Ignition (%) 0.3  3.2.4. Water Potable water was used in the concrete taken directly from the City of Vancouver drinking water supply. 3.2.5. Coarse Aggregate The coarse aggregate used in this research project was round river gravel with a maximum size of 10mm (Fig.3.1). It was obtained from Target Products Ltd. in Burnaby, BC and Lafarge Canada Inc. The specifications of the materials gained from these two sources were almost the same. The specific gravity of this aggregate was 2.69 and the sieve analysis of the aggregate is shown in Fig.3.2.  43  3.2.6. Fine Aggregate Fine aggregate was obtained from Lafarge Canada Inc. The particle size distribution of the sand is also shown in Fig. 3.2. This sand was found to have a relative density of 2.7 and fineness modulus of 2.40.  Fig.3.2. Sieve Analysis of Fine and Coarse Aggregate  44  3.2.7. Fibers The fibers used in this research work were steel, polypropylene and carbon fibers with different geometries. The steel and polypropylene fibers were provided by SI Corporation and carbon fibers were provided by Conoco Corp. The carbon fibers used in the mixes had ozone surface treatment. The other specifications of the fibers are shown in Table 3.4.  Table3.4. Specifications of fibers used in the mixes  Fiber  Si s  2  s  3  c  HPP Mp  Type  Steel fiber Novotex Steel fiber Xorex Steelfiber* Novocon MD-02 Conoco ** Carbon fiber Macro polypropylene Micro polypropylene  Dimensions  Tensile Strength (MPa)  Density  (kg/m )  CrossSectional Shape  212  1150  7850  Circular  Crimped  212  1200  7850  Crescent  0.25 mm  Deformed  212  1200  7850  Circular  12.5  9-11 Um  Straight  232  2100  1900  Circular  50  1mm  Crimped  3.5  375  900  Rectangular  12.5  2 denier  Straight  3.5  375  900  Circular  E  L  D  Geometry  (GPa)  50  1mm  Flat-end  50  1mm  3-10  mm  3  Shape  * Byproduct of tire chord industry ** Pitch-based fiber  3.2.8. Admixtures Superplasticizer A third generation high range water-reducing admixture based on polycarboxylate acid was used. Two types of water reducer were also used: RHEOBUILD 3000 FC and GLENIUM 3000 NS manufactured by Master Builder Technologies. Although according to the producer, they had almost the same properties in the mix, it seemed RHEOBUILD was more efficient than GLENIUM. The RHEOBUILD was used in 70% of the  45  specimens, and then the producer replaced RHEOBUILD with GLENIUM. Therefore, GLENIUM was used in the rest of the specimens. Addition rates of the above admixtures varied depending on the mix composition, type and volume of the fiber. These rates are discussed later in Section 3.3 along with mix proportions. Air Entraining Agent An air-entraining admixture improves the concrete resistance to freezing and thawing, and enhances its plasticity and workability. The air-entraining admixture used in this project was MB-VR Standard, manufactured by Master Builder Technologies. The amount of admixture used varied based on the types of fibers. It was used from 0.4 ml to 0.9 ml per kg of cement. The entrained air measured in the concrete was from 4% to 11%.  3.3. Mixture Proportions Several trial batches were made to reach the proper mix proportions for self-compacting concrete. For making trial batches, a small Hobart mixer was used. Then, the slump flow test was done and the slump flow diameter was measured. Afterward, three cylindrical specimens were cast for compression tests. The main criteria to accept the mix proportions in the trial batches were to have enough flowability (slump flow diameter >600mm) in the fresh state and high strength (compressive strength >80 MPa) in the hardened state. However, by adding fibers in the mixture, sometimes the requirements for flowability and strength of the concrete could not be met. Hence, to meet the requirements, the proportions of the materials had to be changed in accordance with the type and volume fraction of fibers in the mixture. Finally, after making more than ten trial batches and evaluating their properties, two different mix designs were used in this research project. The first one (SCC1) was for plain concrete and hybrid fiber reinforced SCC with macro-fibers and carbon fibers while the second mix (SCC2) was used in hybrids with micro-polypropylene fibers. The proportions of the constituent materials for both mixes are provided in Table 3.5.  46  Table 3.5.Selected SCC mixtures  Mix  Air Entraining (ml/1 Kg cement)  Viscos. Modi fier (ml/1 Kg cement)  Type  Cement Kg/m  Sand Kg/m  Gravel Kg/m  Water Kg/m  SCC1  420  695  1042  123.5  24.7  49.4  10-25  0.9  0.5  477  800  870  133.5  30  90  15-35  0.4  —  SCC2  3  3  3  3  3  Fly Ash Kg/m  Super plasticizer (ml/1 Kg cement)  Silica fume Kg/m  3  Mixture SCC1 was used in the mono macro-fiber reinforced mixes and mixture SCC2 was mostly used in the mixes with hybrid fibers, particularly those which had micropolypropylene. The mixes which had micro-fibers needed more water to maintain their flowability. In order to offset the additional water demand in the mix, more superplasticizer had to be added. However, using more superplasticizer resulted in segregation in the mix. Therefore, by increasing the ratio of sand/aggregate from 0.4 to 0.48 and reducing the ratio of gravel/aggregate from 0.6 to 0.52 in the mixture, the possibility of segregation was alleviated. By increasing the quantity of sand in the mixture, the specific area of the aggregate in the mix increased and more cementitious materials had to be added to produce a desirable flowability. Table 3.6 shows the differences between the two mixes in detail.  3.3.1The Volume of Concrete Batch The total volume of concrete required for each concrete batch to make six beams and six cylinders was: 6x (350x100x100) (Beam) =21 ;r(l00)  Liters  2  6x  -x200  (Cylinder)=9.4  Additional concrete  =3.6  Total volume of concrete  =34  47  Liters  Liters  Table 3.6.The proportions of materials for the selected mixes Mixture  Cement  Fly Ash  Silica Fume  (Kg)  Binder  Binder  Binder  Aggregate  Aggregate  Gravel  Water Cement  Water Binder  SCC1  494  0.85  0.10  0.05  0.40  0.60  0.29  0.24  SCC2  597  0.80  0.15  0.05  0.48  0.52  0.28  0.23  Type  Binder  Sand  3.3.2. Components of Hybrid Fibers Several combinations of hybrid fibers were used in this project. In most of them, one macro fiber was used as the main fiber and one or two micro fibers were used as secondary fibers. The combination of fibers is given in Table 3.7. At the beginning of this research program, the number of combinations planned was more than those listed in Table 3.7. However, some of the combinations did not meet the requirements for the high strength SCC. In addition, some other combinations did not have enough flowability. Therefore, those mixes were excluded from the research program.  Since no mechanical vibration was used to consolidate the concrete, the mixes with more than 0.25% micro-polypropylene fibers were not easily flowable, and even in some mixes with 0.20% micro-polypropylene, the concrete did not have enough flowability. The proportion of 0.15% by volume was the maximum amount of micro-polypropylene fibers to make an acceptable SCC. In the case of carbon fibers, 0.25% was the maximum amount of fiber. In the triplefibermixes, since the volume of micro fibers was high, they did not easily flow under their own weight. Hence, most of the triple fiber mixes were excluded from the program.  48  Table 3.7 Hybridization of fibers in the mixes  SCC  Type of  Mix  Hybrid  1  Plain  Steel (%) Flat end  2  1  3  0.75  4 5 6  Crimped  Polypropylene (%) Carbon Total Volume HPP Stealth Fraction (%) (%) Deformed (macro) (micro) 1 0.75  1 Single  1 1 0.85  Fiber  —  7  1  8  1 0.85 1  0.2  0.2  9  0.5  0.5  10  0.3  0.3  11  0.75  0.25  1  12  0.5  0.5  1  13  0.75  14  0.5  15 16  0.25 0.5  Double Fiber  1  0.5 0.8  17  0.5 0.2  0.75  18  0.85  19  0.85  20  0.3  21  0.3  22  Triple  23  Fiber  1  0.5 0.5  1 1  0.25  1  0.15  1  0.15  1  0.2  0.5 0.2  0.5  0.25  0.25  1  0.2  0.3  1  3.3.3. Selection of Mixing Methods At an early stage in the development of this testing program, three mixing methods were compared to evaluate their effect on SCC properties. At this stage, only small amounts of SCC were made, using the Hobart, 7-liter laboratory mixer. Table 3.8 shows the procedure used for each of the three methods investigated along with slump flow diameter and compressive strength results. In the preliminary mixes, 1% flat-end steel fibers with 50 mm length were used.  49  There were a large number of possible mixing sequences. However, since it would not be feasible to attempt all of the possibilities, so due to time and practical restrictions, this was narrowed down to three separate sequences. These methods of mixing were compiled from suggestions in the literature. The main source was the ACI 544 report [51] and the literature on two-stage mixing [8].  Table3.8. Preliminary mixing method for SCC Slump Flow Method  Description  Diameter (cm)  7  Day  Strength  1  Coarse aggrega. +fineaggrega. + cement + silica fume + fly ash (mix 1 min) Water (mix 3 min) Add fibers (mix 1 min)  66  76  2  Fine aggrega. + cement + silica fume + fly ash (mix 1 min) Water (mix 3 min) Addfibers(mix 1 min) Coarse aggrega. (mix 2 min)  69  77  3  Coarse aggrega. +fineaggrega. + cement + silica fume + fly ash +fiber(mix 1.5 min) Water (mix 4 min)  62  67  According to Table 3.8, the second method showed better workability and higher compressive strength, while the third method showed lower values. Although the first method did not show a significant difference from the second one, the second method was selected for all mixes. In mixes with carbon fibers, due to brittleness of the carbon fibers, they were introduced in the mix at the last stage (method 1). The most important outcome of these tests was that they confirmed the effect of the mixing process on the properties of SCC. Each mixing method was only used once, so the results are not conclusive on their own. Also these results might be different with different types of fibers.  50  3.3.4. Mixing Equipment  All the mixing was carried out in a forced pan mixer, in the materials laboratory of Civil Engineering Department at the UBC. The effective capacity of the mixer was 10 -40 liters for fresh SCC. However, it could be increased up to 60 liters for normal concrete. The speed of pan was about 18 rpm and the blades rotated at 60 rpm when the mixer was empty. However, it reduced to a minimum of 20 rpm depending on the volume of concrete in the pan and the viscosity of concrete. Effectively, the blades rotated from 38 rpm to 78 rpm relative to the pan. A picture of mixer used in this study is shown in Fig.3.3.  Fig.3.3. Mixer Used for the All Tests  3.3.5. Specimen Preparation For each mix, six beams (100x100x350 mm) and six cylinders (100x200 mm) were cast for toughness tests and compression tests, respectively. The size of the beam specimens was chosen based on ASTM C1018-94 [32] standard for the flexural toughness test  51  In the procedure of mixing, sand, cement, silica fume, and fly ash were mixed in the dry condition for 2 minutes, then the combination of water and admixtures (superplasticizer, air entraining agent, and viscosity modifier) was added to the dry mix. The materials were mixed for 5 minutes, then the fibers were introduced into the mix. In case of hybrid fibers, first micro fibers and then macro fibers were added. After introducing the fibers into the mix and mixing of them for 3 to 5 minutes, the coarse aggregate was added and the mixing continued another 3 minutes. In the case of using carbon fibers in the mix, due to their brittleness they were added in the end.  The slump flow diameter of concrete for all mixes was measured. Then, the concrete was poured into the beam and cylinder moulds without any mechanical vibration. In two mixes with 0.25% micro-Polypropylene fibers, the beam moulds were tapped with a trowel to aid the process of casting. Concrete specimens were demolded after 24 hours and placed in a water tank for 28 days for curing.  52  CHAPTER 4  RESULTS  4.1. Introduction This section first describes the methods employed to measure the fresh and hardened properties of hybrid fiber reinforced SCC. Next test results and other observations are presented here. Thefreshproperty measured for each mix was the slump flow diameter and the hardened properties determined included the flexural toughness test based on ASTM C1018 [32] and compressive strength test based on ASTM C39-86 [52].  4.2. Fresh Properties of Hybrid Fiber Reinforced S C C As mentioned in chapter 2, there are several methods to evaluate the fresh properties of SCC mix. However, slump flow test was chosen for evaluating the fresh mix due to its simplicity and acceptance. Fig.4.1 shows the measurement of slump flow in the lab.  Slump flow Number  Fig.4.1. The Slump Flow Test  53  The slump flow test results are listed in Table 4.1. The slump flow diameter for the plain SCC was 730 mm, so the reduction in workability caused byfiberscan be clearly seen in the mixes with fibers. The reduction in the workability of the mixes was more pronounced when the hybrid fibers were used. In the triplefiberhybrid mixes, the slump flow was less than that of the doublefiberhybrid mixes.  Table 4.1 Slump flow results for different hybridization of fiber SCC Type of Mix Hybrid 1  S1  S2  S3  P  %  %  %  %  C  P %  (%)  Total Slump Flow (mm) (%)  Plain  730  2  1  1  670  3  0.75  0.75  700  1  650  4  1  5  Single  1  1  610  6  Fiber  0.85  0.85  720  1  660  0.2  600  0.5  710  0.3  730  7  1  8  0.2  9  0.5  -----  10  0.3  11  0.75  0.25  1  650  12  0.5  0.5  1  610  13  0.75  1  560  14  0.5  1  640  1  640  1 '  600  1  650  0.15  1  620  0.15  1  580  0.2  0.5  580  0.25 0.5  15  Double  0.5  16  Fiber  0.8  17  0.5 0.2  0.75  18  0.85  19  0.85  0.25  20  0.3  21  0.3  0.2  0.5  620  0.5  0.25 0.25  1  550  0.2  1  540  22  Triple  23  Fiber  0.5  0.3  Sl:  Flat-end steelfiber,50mm  P:  Macro-Polypropylene fiber, 50 mm  S2:  Crimped steelfiber,50mm  p:  Micro-Polypropylenefiber,12.5 mm  S3;  Deformed steelfiber,3-20 mm  C:  Carbonfiber,12.5 mm  54  As shown in Table 4.1, the total volume fraction of fibers in most mixes was 1%. Mixes with macro fibers had higher slump flow numbers as compared with the micro fiber mixes. In addition, among the macro fibers, steel libers showed higher flowability as compared to polypropylene fibers. However, in the mix with 1% steel micro fibers, the slump flow number was not as high as with the other steel fibers (610 mm). The minimum slump flow number was measured in the hybrid fiber mixes containing micropolypropylene fibers. Although in most mixes the volume fraction of micropolypropylene fibers was limited to 0.2%, the corresponding reduction of flowability in the SCC mix was significant. Although carbon fiber was finer than micro-polypropylene fiber, the workability of the SCC mix with carbon fiber was better and showed a higher slump flow number. The reason for this anomalous behavior of fresh FRSCC containing carbon fibers may be attributed to the brittleness of carbon fibers. It appears that carbon fibers were fractured during mixing to smaller lengths. Since the mixing time in the SCC mixture was longer than that of the normal concrete, one may hypothesize that a greater number of carbon fibers were broken in SCC, which had a less drastic effect on the reduction of FRSCC workability.  4.3. Hardened Properties of Hybrid Fiber Reinforced S C C Tests on the hardened concrete consisted of compressive and flexural strengths. However, the main focus was flexural toughness of various mixes.  4.3.1. Compressive Strength  Compression test is a globally accepted method for quality control testing of concrete. Its measurement methods may vary in different standards but this property is used for almost all design purposes. The method used here for compressive strength was that described in ASTM C39 [52]. Although the diameter of the cylinder required to reduce the wall effect for 50mm fibers was 150mm, due to limitations in the capacity of the testing machine the  55  selected size for the cylinders was 100x200 mm. Fig.4.2 shows the machine used for compressive tests. The cylinders were loaded at the rate of 2kN/second.  Fig.4.2. Compression Testing Machine  Six cylinders were cast for each batch of concrete to give a better average strength. The average standard deviation for all mixes was about 7 MPa, which shows a relatively stable result for all mixes. The results of compressive strength for the mixes are presented in Table 4.2.  56  Table 4.2.Results of compressive strength for different mixes SCC  Type of  S1  S2  S3  P  P  C  Total  Compressive strength  Mix  Hybrid  %  %  %  %  %  (%)  (%)  (MPa)  1  Plain  86  2  1  1  98  3  0.75  0.75  60  1  96  4  1  5  Single  1  1  85  6  Fiber  0.85  0.85  90  1  96  0.2  77  7  1  8  0.2  9  0.5  0.5  71  10  0.3  0.3  86  11  0.75  0.25  1  101  12  0.5  0.5  1  74  13  0.75  1  89  14  0.5  1  91  1  82  1  86  1  90  0.15  1  86  0.15  1  92  0.2  0.5  75  0.5  81  0.25 0.5  15  Double  0.5  16  Fiber  0.8  17  0.5 0.2  0.75  18  0.85  19  0.85  0.25  20  0.3  21  0.3  0.2  0.5  0.25 0.25  1  82  0.2  1  78  22  Triple  23  Fiber  Sl: S2: S3:  Flat-end steelfiber,50mm Crimped steelfiber,50mm Deformed steelfiber,3-20 mm  0.5 P: p: C:  0.3  Macro-Polypropylenefiber,50 mm Micro-Polypropylenefiber,12.5 mm Carbon fiber, 12.5 mm  4.3.2.Flexural Strength The flexural strength is also called the modulus of rupture (MOR) and is an indirect measure of tensile strength. It is determined using the maximum load attained on the load-deflection curve. In all mixes, MOR was the point at which the load-deflection 57  curve starts being nonlinear. The results of MOR for different mixes are shown in Table 4.3. Comparison of hybrid mixes with the corresponding mono fiber mixes does not show any noticeable increase in MOR of the hybrid mixes.  Table 4.3.Modulus of Rapture results for different mixes based on ASTM C1018 SCC Type of  S1  S2  S3  P  P  C  Total  Mix  Hybrid  %  %  %  %  %  (%)  (%)  1  Plain  Modulus of Rupture (MOR) (MPa) 10.61  2  1  1  10.80  3  0.75  0.75  8.38  1  9.08  4  1  5  Single  1  1  7.82  6  Fiber  0.85  0.85  8.87  1  10.33  0.2  8.97  7  1  8  0.2  9  0.5  0.5  8.55  10  0.3  0.3  9.12  11  0.75  0.25  1  10.73  12  0.5  0.5  1  9.04  13  0.75  1  9.60  14  0.5  1  9.34  1  8.61  1  9.50  1  10.21  0.15  1  9.81  0.15  1  7.84  0.2  0.5  9.22  0.2  0.5  9.17  0.25  0.25  1  9.20  0.2  0.3  1  8.68  0.25 0.5  15  Double  0.5  16  Fiber  0.8  17  0.5 0.2  0.75  18  0.85  19  0.85  20  0.3  21  0.3  22  Triple  23  Fiber  0.5 0.5  0.25  Sl:  Flat-end steelfiber,50mm  P:  Macro-Polypropylenefiber,50 mm  S2: S3:  Crimped steelfiber,50mm Deformed steelfiber,3-20 mm  p: C:  Micro-Polypropylenefiber,12.5 mm Carbonfiber,12.5 mm  58  4.3.3.Flexural Toughness Test The property of flexural toughness relates to the ability of the concrete to absorb energy, after macro-crack formation, while the fibers hold the matrix together. It is characterized by the post-peak portion of the area under the load-deflection curve obtained during a flexural test on 100x100x350 mm beams in a four-point loading arrangement. Fracture energy absorbed in bending is probably the most appropriate measure of FRC performance because in most instances FRC elements are subjected to bending. In addition, it is relatively easy to test the elements under bending. Unfortunately, however, the method of testing and analysis greatly influences the results.  One parameter that influences the toughness results is the stiffness of the test machine. When a loading system applies the load on a test specimen, a reactive force is created in the loading system. It deforms and stores the energy. The amount of energy in the loading system depends on its stiffness. The softer loading system deforms more and stores greater reactive energy. As soon as the beam cracks, a part of the stored energy is suddenly released back to the specimen, and this sudden release of energy causes unstable propagation of cracks, which influences the post-cracking flexural response. This problem is more pronounced when the concrete is high strength because applied loads would be higher and more energy would be stored in the machine.  The above-mentioned problem will occur when an open loop testing machine with a low stiffness or a very stiff open-loop machine is used. This problem can be solved by using a closed-loop testing machine.  4.3.3.l.Test Apparatus Unfortunately, at the time of testing a closed-loop testing machine was not available. Therefore, open-loop testing was conducted in a 150 kN Instron machine. Fig.4.3 shows a schematic of the loading and measurement system.  59  Fig.4.3. Schematic of the Loading and Measuring System  A steady rate of cross-head displacement of 0.1 mm/min was used. Two rollers were used as supports for the beam specimen, and the loads were applied at the third points of the beam. A Japanese yoke was used around the specimen at the mid height to record the net deflections. Net deflection was measured by fixing a yoke on both side of the specimen, which moved with the specimen during loading. Two electronic transducers (LVDT) were attached to the yoke at the mid span of the beam to record the deflection. Hence, the deflections resulting from the settlement of supports, crushing at the load points and the load fixture deformation were all eliminated, and only the net deflections were recorded. The LVDTs were connected to a data logger and the data were saved in a computer. The recorded data from the LVDTs were averaged and averages were used as the deflections of the beam in data analysis. The data logger was operated through a PC. It picked up the data produced during the test at a frequency of 1 Hz. Fig 4.4 shows the complete arrangement of the test. Test Procedure Once the sample was positioned in the testing machine, it was loaded at a rate of 0.01 mm/min before thefirstcrack and 0.10 mm/min after. The reason for this change was to bring about a stable fracture in the specimen. While a 2 mm deflection of the beam midpoint was enough for toughness calculations, loading was continued until the deflection exceeded 2.5mm. This was done to allow for removal of the instability part  60  from the load-deflection curve to calculate the PCS values [53] and the parameters given in JSCE [54]. Calculation Procedure  The result of a flexural toughness test was a load vs. deflection graph such as that shown in Fig.4.5. As explained before, due to the use of an open-loop system, following the peak load, the load dropped suddenly in an unstable manner. The data logger, even at the high frequency, could not record any data points in this region. Since the curve after the peak load included a large amount of energy originally stored in the testing machine, it did not show the real behavior of the material. To adjust the curve due to unstable failure, the part of curve, which had a sudden drop without recording any data, was shifted back and drawn vertically as shown in Fig.4.5. The flexural toughness for all mixes was calculated based on the modified curves.  61  Load *  Unstable portion  Deflection  Fig.4.5. The Modification of Load-Deflection Curve in Flexural Toughness Test Techniques for Flexural Toughness Measurements  The most common techniques to determine theflexuraltoughness of the FRC beams are the ASTM C1018, JSCE and PCS methods [53]. Although the most widely used standard test methods, are ASTM C1018 [51] and the Japan Society of Civil Engineers (JSCE) standard SF-4 method [54], the PCS method has been emphasized more in this study. Due to concerns with ASTM C1018 and JSCE methods [53], explained in the next part, it seems that the PCS method is more efficient. ASTM C1018 test method and its concerns The ASTM C1018 standard test method is based on determining the amount of energy required to deflect and crack a FRC beam. Toughness indices I5, I10, I20, etc., are then calculated by considering the ratios of the energy absorbed to a certain multiple of the first crack deflection divided by the energy consumed up to the occurrence of the first crack as shown in the Fig.4.6.  62  Area OABCI Area OAJ  20  Area QABEG Area OAJ First Crack Load First Crack Deflection  °  S  35  5.58  10.55  Midpoint Deflection  Zero Load L o a a  Fig4.6. Description of the Flexural Toughness According to ASTM C1018  The calculation of the above mentioned indices requires the first crack energy. However, a precise assessment of the first crack in the load-deflection curve is never possible. The identification of first crack deflection is highly subjective, due to the substantial nonlinearity of the load-deflection curves even prior to the peak load.  Another concern with the ASTM C1018 method is the instability after the peak load. The point of peak load is also the point of instability for the loading machine. If the machine is not stiff enough, it will suddenly unload and release a large amount of energy. This sudden release of energy has major effects on the load-deflection curves immediately following the peak load. Table 4.4 shows the flexural toughness results calculated based on the ASTM C1018 method. The indices were calculated from the modified loaddeflection curves (Fig.4.5). Since the strength of concrete tested was high. The load-deflection curve consisted of an obvious straight line, which went up to a maximum load and a nonlinear part. Almost, in all curves the peak load was considered as afirstcrack.  63  Table 4.4 Flexural toughness results for different mixes based on ASTM C1018 Mix No  Type of Hybrid  1  Plain  ASTM C1018 Toughness  Fibers S1 (%)  S2 (%)  S3 (%)  P (%)  P (%)  C (%)  Total (%)  ho  bo  0  0  0  4.07  7.93  15.37  5.19 10.05  19.58  1  4.73  9.46  18.28  5.66 11.03  20.78  2  1  1  3  0.75  0.75 1  4 5  Single  1  1  6  Fiber  0.85  0.85  4.48  7.87  13.89  1  2.22  3.78  7.11  0.2  1.45  1.45  1.45  1  7  0.2  8 9  0.5  0.5  4.11  7.48  14.49  10  0.3  0.3  4.66 8.64  16.56  11  0.75  0.25  1  4.58  9.14  18.44  12  0.5  0.5  1  3.15  5.56  10.44  13  0.75  1  4.21  8.13  16.15  14  0.5  1  3.85  7.30  14.53  1  2.30  3.59  6.42  1  2.67  4.44  8.37  1  2.54  4.01  7.27  0.15  1  5.39 10.67  20.54  0.15  1  6.00 11.83  22.59  0.2  0.5  4.00 6.78  12.50  0.2  0.5  3.37  5.77  10.77  0.25  0.25  1  2.96  5.21  9.71  0.2  0.3  1  2.75  4.43  8.16  15 16  0.25 0.5 0.5  Double Fiber  0.8  0.25  0.85  18  0.85  19 20  0.3  21  0.3  23  0.2  0.75  17  22  0.5  Triple Fiber  0.5 0.5  Sl: Flat-end steelfiber,50mm  P:  Macro-Polypropylene fiber, 50 mm  S2: Crimped steelfiber,50mm  p:  Micro-Polypropylene fiber, 12.5 mm  S3: Deformed steelfiber,3-20 mm  C:  Carbon fiber, 12.5 mm  64 JSCE standard SF-4 method and its concerns In this method, the area under the load-deflection curve is measured up to a deflection of span/150, which is 2mm for a 300 mm span. This area is then used to calculate the flexural toughness (FT) factor according to JSCE SF-4 as indicated in equation (4.1). FT shows the post crack strength of the material when loaded to a deflection of span/150. The deflection chosen for the calculation of toughness factor is completely arbitrary and is not based on serviceability considerations.  L  1 5 0  Deflection  Fig.4.7. Flexural Toughness Factor Description According to JSCE in Which the Area Is Measured up to a Deflection of Span/150  One of the concerns with this method is that the end point deflection of span/150 is greater than most acceptable deflection limits. In addition, the behavior of the material immediately after the first crack, which may be important in many applications, is not indicated in the FT factor. Finally, the area under the curve, used to calculate FT factor combines the pre-peak and post-peak energies and it is criticized for failing to distinguish between these two areas. Table 4.5 shows the results of the flexural toughness tests based on JSCE standard SF-4 method.  65  Table 4.5 Flexural toughness factor results for different mixes according to JSCE Mix Type of No Hybrid 1  Flexural Toughness  Fibers S1 (%)  S2 (%)  S3 (%)  P (%)  P (%)  C (%)  Total (%)  (JSCE Method) 0.17  Plain  2  1  1  6.69  3  0.75  0.75  5.46  1  6.17  1  4 5  Single  1  1  3.58  6  Fiber  0.85  0.85  2.36  1  3.95  0.2  0.18  1  7  0.2  8 9  0.5  0.5  4.39  10  0.3  0.3  5.09  11  0.75  0.25  1  8.42  12  0.5  0.5  1  3.79  13  0.75  1  6.62  14  0.5  1  6.10  1  2.40  1  3.65  1  3.16  0.15  1  5.86  0.15  1  3.41  0.2  0.5  3.65  0.2  0.5  3.03  0.25  0.25  1  3.58  0.2  0.3  1  3.36  0.25 0.5  15  Double  0.5  16  Fiber  0.8  0.85  18  0.85  19 20  0.3  21  0.3  23  0.2  0.75  17  22  0.5  Triple Fiber  0.5 0.5  0.25  S l : Flat-end steel fiber, 50mm Crimped steel fiber, S2: 50mm  Macro-Polypropylene fiber, 50 P: mm Micro-Polypropylenefiber,12.5 p: mm  S3: Deformed steel fiber, 3-20 mm  C:  Carbon fiber, 12.5 mm  66 Post Crack Strength (PCS) Method In order to simplify the approach, a new method has been proposed by Banthia and Trottier [53], wherein identification of first crack is not required. The procedure according to them is as follows: The load-deflection curve should be measured by use of a Yoke. The area under the loaddeflection curve is then divided into two regions: the pre-peak region and the post-peak region. The area under the curve up to the peak load is termed pre-peak energy E . The pK  areas under the curve are measured up to deflections equal to various fractions of the span  .Etotai,  m  (measured at a deflection of L /m). The post peak energy values of L /m  deflection, isp t, , is obtained by subtracting the pre-peak energy E 0S  values of £  t  o  tai,  m  pie  from the various  - Finally, the post crack strength (PCS) is calculated at a deflection of L  m  /m by using Equation (4.2).  (.Zipost, m)L  PCS = m  (4.2)  JL (  m  Opeak)^  Table 4.6 shows the results of the flexural toughness tests based on the PCS method.  Deflection  Fig.4.8. Flexural Toughness Description According to the PCS Method  67  Table 4.6 Flexural toughness results for different mixes according to PCS method Mix Type of No Hybrid  1  Si  s  s  (%)  (%)  (%)  2  P  3  3  (%)  Total  0.1 0.2 0.3 0.4 0.5 0.75  (Vo)  0  7.86 7.97 8.00 7.80 7.59 7.21 6.96  6.60  6.62  1  1  6.45 6.29 6.05 5.82 5.61 5.16 4.81  4.08  3.53  0.85  0.85  4.82 4.55 4.29 4.09 3.92 3.54 3.20  2.67  2.29  1  2.90 2.91 3.00 3.07 3.14 3.30 3.46 3.71 3.91  0.2  0  0  0  0  0  0  0  0  0  0.5  4.15 4.22 4.26 4.33 4.39 4.51 4.54  4.49  4.36  0.3  0.3  5.32 5.31  5.25  5.08  —  0.5 0.75  0.8  0.85 0.85  19  0.5  1  4.24 4.06 4.05 4.04 4.04 4.01 3.98 3.89 3.75 7.51 7.36 7.29 7.32 7.34 7.38 7.33 7.05 6.66  1  6.05 5.95 6.03 6.09 6.14 6.22  1  1.72  1  2.57 2.68 2.80 2.91 2.99 3.15 3.28 3.48 4.61  1  2.15 2.35 2.48 2.58 2.66 2.80 2.92 3.04 3.07  0.15  1  8.62 8.32 8.13 7.92 7.81 7.44 7.03  0.15  1  6.60 6.48 6.31 6.05 5.78 5.22 4.74  3.95  3.36  0.2  0.5  3.51 3.46 3.48 3.53 3.57 3.63 3.67  3.68  3.72  0.5  3.73 3.68 3.68 3.74 3.82 3.77 3.63  3.28  2.95  0.5 0.2  0.75  17 18  9.25 9.08 9.07 9.14 9.20 9.21 9.13 8.84 8.50  1  0.5  Fiber  0.3  5.20 5.16 5.19 5.27  1  0.5  15 Double  5.19  0.25  0.25  0.5  22 Triple Fiber 23  8.21 8.30 8.30 8.27 8.23 8.08 7.90 7.33 6.72  0.5  0.75  21  0  1  0.2  20  0  5.48  8  14  0  5.61  1  13  0  8.21 6.11 6.04 5.99 5.95 5.84 5.76  7  12  0  0.75  Fiber  11  0  2  0.75  5 Single  10  0  1.5  1  1  9  0  1  1  4  16  P  (%) (%)  c  Plain  2  6  Ll VI (mm)  Fibers  0.25  1.81  1.91  6.23 6.23 6.11  1.98 2.01 2.09 2.15 2.26 2.35  6.42  5.87  0.3  0.2  0.5  0.25 0.25  1  4.2 4.03 4.00 3.99 3.97 3.92 3.85 3.70 3.53  0.2  1  2.25 2.35 2.33 2.55 2.64 2.82 2.95 2.19 3.31  0.5  0.3  51 : Flat-end steel fiber, 50 m  P: Macro-Polypropylene fiber, 50 mm  52 : Crimped steel fiber, 50mm  p: Micro-Polypropylene fiber, 12.5 mm  53 Deformed steel fiber, 3-20 mm  C: Carbon fiber, 12.5 mm  68 Flexural Responses of HyFRSCC Specimens  Flexural toughness curves for all 23 mixes are shown in Fig. 4.9 to Fig. 4.31. Each graph contains six curves and one average curve. Some graphs have four curves with one average curve.  Figs. 4.9, 4.11, 4.12, 4.15, 4.16, 4.17, 4.18, represent the load-deflection curves for mixes reinforced with hybrid fibers. Flat-end steel fibers were used as the primary fibers in the hybrid composites and carbon and polypropylene were added as secondary fibers. The total fiber volume fraction in these mixes was 1%. In all of them, the load increased to a peak value around 30 kN and then dropped suddenly to about 20 kN. Afterwards, during the 2 mm deflection, the load decreased very slowly.  Figs. 4.10 and 4.14 show the load-deflection curves for plain concrete and for the mix with 0.2% micro-polypropylene. The load, after reaching its peak value, had a sudden drop to zero with no toughness. Figs. 4.19, 4.20, 4.21, 4.22, and 4.29 show the curves for 1% hybrid FRC. In these composites, macro-polypropylene fibers were used as the primary fiber while carbon and micro-polypropylene were added as secondary fibers. In these composites, the sudden drop after peak load was very severe and the load dropped to around 22 kN. Figs. 4.13, 4.25, and 4.26 show the load-deflection graphs for mixes with two microfibers, micro-steel and micro-polypropylene. The specimens with micro-steel did not carry as much load in the post-peak region, as the macro-steel fibers did. However toughness improvement due to use of micro-polypropylene with micro-steel was noticeable. Figs. 4.23 and 4.24 show the load-deflection graphs for crimped steel fibers and their combinations with micro-polypropylene fibers. The total volume fraction in these mixes was 1%. Figs. 4.30, 4.27, 4.28, and 4.31 demonstrate the load-deflection curves for specimens with low volume fractions of fiber. The totalfibercontent was 0.5%.  69  70  40  0 I 0  1 0.2  0.4  0.6  0.8  1  1.2  1.4  1.6  1.8  Deflection (mm)  Fig.4.11 Flexural Response for the FRC Beams (Mix 2)  Fig.4.12 Flexural Response for the FRC Beams (Mix 11) 71  2  40 ,  Deflection (mm)  Fig.4.13 Flexural Response of the FRC Beam (Mix 6)  35  30 4  -#1 -#2 #3 #4 •Average  0.2% Micro-polypropylene  25  20 "D  ra o  15  10  0  0.2  0.4  0.6  0.8  1  1.2  1.4  1.6  Deflection (mm)  Fig.4.14 Flexural Response of the FRC Beam (Mix 8)  72  1.8  2  Fig.4.19 Flexural Response for the FRC Beams (Mix 15)  35  0.8%HPP + 0.2% m icro-polypropylene  Fig.4.20 Flexural Response for the FRC Beams (Mix 16)  7  5  -Seriesl -Series2  1  0.8  1.2  1.6  Deflection (mm)  Fig.4.21 Flexural Response for the FRC Beams (Mix 7)  Fig.4.22 Flexural Response for the FRC Beams (Mix 17)  76  1.8  35  10  5  o J — 0  0.2  ,  ,  ,  ,  ,  ,  ;  0.4  0.6  0.8  1  1.2  1.4  1.6  Deflection (mm)  Fig.4.23 Flexural Response for the FRC Beams (Mix 4)  77  .—; 1.8  2  0  0.2  0.4  0.6  0.8  1  1.2  1.4  1.6  1.8  2  Deflection (mm)  Fig.4.25 Flexural Response for the FRC Beams (Mix 5)  0  0.2  0.4  0.6  0.8  1  1.2  1.4  1.6  Deflection (mm)  Fig.4.26 Flexural Response for the FRC Beams (Mix 19)  78  1.8  2  0.8  1  1.2  Deflection (mm)  Fig.4.27 Flexural Response of the FRC Beams (Mix 10)  35  0.3%Steel (flat-end) + 0 . 2 % Micro-polypropylene  0.8  1  1.2  Deflection (mm)  Fig.4.28 Flexural Response of the FRC Beams (Mix 20)  7  9  35 0.5%HPP  + 0.3%Carbon +  0.2%Micro-polypropylene  Fig. 4.29 Flexural Response of the FRC Beams (Mix 23)  35 ,  0 J 0  ,  ,  ,  ,  ,  ,  ,  0.2  0.4  0.6  0.8  1  1.2  1.4  D e f l e c t i o n (m m )  Fig.4.31 Flexural Response of the FRC Beam (Mix 21)  81  1 1.6  ! 1.8  ! 2  CHAPTER 5  Analysis of Results and Discussion 5.1. Introduction This section contains the analysis of data from experiments on the fresh and hardened properties of HyFRSCC. Slump flow diameter measured in the fresh state is discussed. The effects of different fiber types and different fiber systems (mono, double and triple fiber) on the fresh properties of HyFRSCC are compared and discussed. In the hardened state, flexural toughness, compressive strength, and modulus of rupture (MOR) measured on different mixes are compared. In addition, the flexural toughness results of each mix are presented based on three methods, ASTM C1018, JSCE SF-4, and the PCS method and compared with each other. Finally, the synergy gained in flexural toughness by using different hybrid fibers is analyzed.  5.2. Fresh Property Results As mentioned before, concrete with a higher slump flow diameterflowsmore easily and fills the form more efficiently. It has enough workability to be compactable without any external vibration and can achieve the required strength in the hardened state. Usually, in plain SCC, higher slump flow indicates higher compressive strength because it has better filling ability and compaction. However, this cannot always be the case for fiber reinforced SCC. The minimum slump flow value to show sufficient flowability and compactability for plain SCC, is 600mm [29, 30], but in the case of FRSCC, no specific number has been mentioned in the literature. According to the results of this research, no minimum number could be recommended for slump flow to show the required compactability because it is highly dependent on the type and volume fraction of fibers. In a few cases, the results showed that some mixes with a low volume fraction of macro fibers even with a high slump flow number, had lower compressive strengths than those with a higher volume offiberand lower slump flow diameter. For instance, in Mix 2 with  82  1% steel fiber, the slump flow number was 670 mm and f =98 MPa while, in Mix 9 with c  0.5% steel fiber, the slump flow number was 710mm and f =71 MPa. That is, Mix 9 had c  higher slump flow but had lower compressive strength than Mix 2. Therefore, it appears that in the mixes with different fiber volume fractions, the slump flow number cannot be a good indicator of compactability. Only when the fiber content and the other constituent material are same in the mixes, can slump flow be a good indicator of compactability.  Mixes with micro fibers showed lower slump flow than those with macro fibers. This reduction came from the increase of viscosity in the mix due to the use of micro fibers, but it did not necessarily mean that reduction in slump flow also led to a reduction in the compressive strength. For instance, the slump flow number in Mix 13 with 0.25% micropolypropylenefiberwas low, 560 mm; whereas the compressive strength of this mix was rather high: 89 MPa. In many instances, though, mixes with micro fibers showed lower compressive strength as compared with those containing macro fibers. Micro fibers also increased the air content in the mix and reduced the self-compacting properties of the mix. If the data of mixes with very low volume fractions of macro fibers and mixes with high volume fractions of microfibersare excluded from the results, it can be seen that there is an approximate relationship between the slump flow diameter of the mix and its compressive strength. Mixes with low volume fraction of macro fiber (Mixes 9 and 10) showed a high slump flow without showing a corresponding high compressive strength. In addition, some mixes with high volume fraction micro-fiber, >0.25%, (i.e. Mixes 22, 23, andl3) did not demonstrate low strength in spite of a low flowability. The relationship between the slump flow diameter and compressive strength is shown in Fig 5.1. The mixes which are in the ellipse demonstrate a relationship between the slump flow and compressive strength. However, there are eight data points outside of the ellipse, which did not show a good relationship. They are as follows: 1- Mono fiber mixes with low volume fractions of steel fiber (i.e. 0.3%, 0.5%, and 0.75%). The use of lower volume fractions of steel fiber in those mixes caused higher slump flow as compared to the mixes with 1 %fiber,without creating a higher strength. 2- Doublefibermixes with high amount of micro-polypropylene. 3- Triple fiber mixes  83  with high amount of micro fibers (carbon + micro-polypropylene). In the last two cases, the use of micro fibers caused a major reduction in the slum flow.  110  2 100 E  I) e s »  • Mono fiber RS Double fibers • Triple fibers  .....yff^m7^\  /  9 0  80  A  ik  /  L  ••7  rf f/ • ra  •  m.My  4  A  \T  500  550  600  650  700  750  Slump flow diameter (mm)  Fig.5.1. Relationship Between Slump Flow and Compressive Strength  Fig 5.2 presents the relation between slump flow diameter and number of fiber types in a hybrid mix. In this chart, slump flow numbers have been averaged and shown for plain, mono-fiber, double-fiber, and triple-fiber mixes. Although the slump flow value for each group in this graph is an average of mixes with different fiber types, it roughly shows that by increasing the number of fiber types in a hybrid mix the workability of HyFRSCC reduces. However, the main reason for low slump flow number in the triplefibermix was the addition of 0.5% micro-polypropylene fibers.  The results showed that the average volume of superplasticizer needed in the mixes depending on the number of fiber types in the concrete was between 10 to 40 ml/kg of cementitious material. In the plain SCC the minimum amount and in the HyFRSCC mixes with micro fiber the maximum amount of superplasticizer was required. The average amounts of superplasticizer used in different mixes are shown in Table 5.1.  84  800  T  700 ----% 600 $ 500 - - 5 400-  .  i£ 300 — -  •  |  -.' -  .,  ,  , - - - - - .. -  200  "  100 0  ••:  .  I 1  1  Plain  :  1 1 Single fiber  /  :  '  ; -  •  \— 1 : L 1 Double fiber Triple fiber  Fig.5.2. Average Slump Flow for Different Types of Mixes  Table 5.1. Average amount of superplasticizer used in different mixes Mix Type  Superplasticizer (ml/kg Cementitious)  Plain  Single fiber  10  20  Double fiber  33  Triple fiber  40  Although in this study no viscometer was used to measure the rheological parameters of the mixes visually, mixes with micro-fibers displayed higher viscosity as compared with mixes containing macro-fibers. Among the mixes containing micro-fibers, mixes with micro-polypropylenefiberhad higher viscosity as compared with the other mixes.  Using a higher amount of superplasticizer in the mixes with low w/p ratio increased the viscosity of the fresh mix. Due to thefinenessof micro-polypropylene fibers and this high specific area, a part of water in the mix was used to wet the surface of fibers and hence reduced the workability of the mix. It was observed that, even by adding more superplasticizer, the mix did not get more flowability but became more viscous. The reason for this behavior was due to low w/p ratio (w/p=0.24), high volume pozzolanic materials, and presence of the fine fibers (i.e. micro-polypropylene) in the mix.  85  Although carbon fibers are finer than micro-polypropylene fibers, mixes with carbon fibers needed less superplasticizer and showed more flowability as compared with mixes containing the micro-polypropylene fibers. Mix 11 with 0.75% steel fiber and 0.25% carbon fiber had a 650 mm slump flow, whereas mix 13 with 0.75% steel fiber and 0.25% micro-polypropylene fiber had a 560mm slump flow. This behavior can be attributed to the brittle characteristics of carbon fibers. Since the mixing time for SCC was longer than that of normal concrete, carbon fibers broke into very small pieces during mixing and became less effective in reducing the flowability of SCC. A washout test was carried out on the carbonfibermixture to specify the amount of visible carbon fiber after mixing in the SCC. The results showed that after five minutes of mixing, no carbon fiber was visible to the naked eye.  The appearance of freshly mixed concrete with macro-fibers showed that macro-fibers did not have a serious effect on the viscosity of concrete. As mentioned before, the slump flow test essentially presents the yield stress of the fresh concrete [5, 55]. The results of slump flow tests of mixes with macro-fibers showed that macro-fibers reduced the slump flow of concrete and therefore, increased the yield stress of the fresh mix. It was observed that reduction of slump flow and increasing of the yield stress in the mixes with steel fiber was more pronounced than that with HPP, and it was due to stiffness of steel fibers, which induced more interlocking between fibers and created higher yield stress in the fresh state. Results of slump flow in mixes 2, 9, and 10 showed that interlocking of fibers was the major reason behind a reduction in flowability of SCC, Table 5.2.  Table 5.2. Comparison of FRSCC flowability with different steelfibercontents SCC Mix Steel fiber content Slump flow (mm)  2  9  10  1%  0.5%  0.3%  670  710  730  86  Unlike the mixes with micro-fiber which needed a large amount of superplasticizer, mono macro-fiber mixes with 15 to 20 ml superplasticizer per kg of cementitious materials gained higher slump flow values leading to a reduction in the yield stress in the fresh concrete.  In the case of hybrid fibers a combination of micro and macro fibers restrained the flowability of fresh concrete effectively. It had high viscosity due to the use of microfibers and high yield stress because of the macro-fibers. Both were restricting factors in the flowability of fresh concrete. Therefore, as shown before, a large amount of superplasticizer (30 to 40 ml/kg cementitious materials) was needed to create selfcompacting properties. It was noticed that, in HyFRSCC only the use of correct amount of admixture and other required materials for the mix would give the desired results. Insufficientflowabilitycan damage the composite in two different ways. First, it cannot fill the form completely, which may lead to honeycombing. Second, it can reduce the mechanical properties of hardened concrete.  5.3. Discussion on Compressive Strength One of the objectives in this study was to get a high compressive strength (>80 MPa) for HyFRSCC. Although the compressive strength in most mixes reached 80 MPa or higher, a few of the mixes had compressive strengths that were less than the target value of 80 MPa. Thefluctuationof compressive strength observed in different mixes is shown in Fig 5.3.  87  120  o  I LJ LJ L-l, LI, r 1 III 1  i  i  2  3  4  5  6  I, I' I I I, LJ IJ , 11, LI ,11, II, H , I i , M II, r I, II, r-l, II 7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  SCC Mix  Fig. 5.3 Compressive Strength of Different Mixes  The reasons for the fluctuation of compressive strength can be attributed to several factors. One of them might be an error in measuring the water content of the sand used in the concrete. The water content of the sand was considered based on the saturatedsurface-dry (SSD) condition and it was not easy to obtain a true SSD condition. Therefore, the real amount of water in the mix was different from the calculated number and the ratio of water to cementitious material used in the mix was varied in different mixes. Furthermore, because of using a high amount of superplasticizer with low w/p ratio in the SCC mixes, even small amounts of extra water caused high variation in the compressive strength of concrete. Another reason for low compressive strength in some mixes was their low flowability and hence their low self-consolidation properties. This happened in mixes with high viscosity. The high viscosity of the mixture did not allow the entrapped air in the concrete to be removed easily and it led to a reduction in compressive strength of concrete. The method of casting can be another reason for the variation in compressive strength. There is no standard for casting of cylindrical specimens of SCC, and since no external vibration is utilized in SCC, the ASTM C31 method used for normal concrete could not be applied for SCC. One of the casting methods used in this study was to pour SCC into a cylinder and fill the mold in one stage The other method was to pour SCC into a cylinder in three layers and after pouring of each layer, allow the air voids to leave the concrete. Although in most mixes the second  88  method was applied, a small number of specimens were cast incorrectly with the first method at the start of this program.  5.4. Flexural Toughness In this part, the flexural toughness values are calculated for mixes with mono, double and triple fibers with three different methods: ASTM C1018, JSCE, and PCS. Then they are compared and discussed. Figs. 5.4 to 5.31 show the comparison graphs of different mixes to evaluate the toughness improvement due to use of hybrid fibers in the mix. For each group of the load-deflection curve, PCS values at different L/m ratios, ASTM C1018 indices:  I5,110,120,  and JSCE SF-4 values are presented. In most cases, each curve is an  average of six specimens.  5.4.1 Hybridization of Steel Macro-Fiber (Flat-End) With Carbon Fiber  The flexural toughness results of four different mixes are presented and discussed in Figs 5.4 to 5.7. In this comparison, the steel and carbonfiberwere considered as major and minor fibers, respectively. The mixes included HyFRSCC containing 0.75% steelfiber+ 0.25% carbon, 0.5% steelfiber+ 0.5% carbon, 0.5% steelfiber,and those containing 1% steel fiber.  As shown in the Fig. 5.4, hybridization of 0.25% by volume carbonfiberwith 0.75% by volume steel fiber created a noticeable improvement in the flexural response of the specimens, compared to those containing 1% steelfiber.Since the MOR and compressive strength for both mixes were identical one can assume that the bulk properties of the matrix were also the same. Therefore, it can be concluded that addition of carbonfiberin this mix increased the toughness of the interfacial zone between steel fibers and matrix. However, it was the only hybrid mix with carbonfiberwhich showed flexural toughness improvement. This efficient performance can be attributed to the effective dispersion of carbonfibersand use of short mixing time. This improvement is also shown clearly by the PCS method indicated in Fig. 5.5. The PCS curves for these mixes showed that the  89  hybrid mix had noticeable strength enhancement at different deflections as compared with the mono-fiber mixes.  Hybridization of 0.5% steel with 0.5% carbon by volume did not show any improvement in the toughness. As seen in Figs. 5.4 and 5.5, the fracture toughness of the hybrid mix was even less than that of the mix with 0.5% steel. The reason for this behavior was due to the use of a high volume fraction of carbonfiber(0.5%) in the mix, which reduced the flowability of the mix and increased the air content in the mix. Both these factors caused a reduction in the compressive strength of the matrix, which in turn reduced the quality of the bond between matrix and steel fiber.  Flexural toughness comparisons in those mixes were also carried out as per the ASTM C1018 method and the JSCE method, Figs. 5.6 and 5.7, respectively. They demonstrated the same trend as shown in load-deflection and PCS curves in Figs. 5.4 and 5.5. However, the performance of mixes based on ASTM C1018 method, Fig.5.6, was not in proportion to that of the other methods.  90  40 0 . 7 5 % S t e e l (Flat-end) + 0 . 2 5 % C a r b o n 1 % Steel (Flat-end)  0.8  1  1.2  1.4  1.6  1.8  Deflection (mm)  Fig.5.4. Flexural Responses for HyFRSCC Beams with Steel and Carbon Fibers  12.00 0 . 7 5 % S t e e l (Flat-end) + 0 . 2 5 % C a r b o n 10.00  8.00  6.00  4.00  , 0 . 5 % Steel (Flat-end) + 0 . 5 % C a r b o n  \  0 - 5 % Steel (Flat-end)  2.00 Plain  0.00 0.2  0.4  0.6  0.8  1  1.2  1.4  1.6  1.8  L/M (mm)  Fig.5.5. PCS Value for HyFRSCC Beams with Steel and Carbon Fibers  91  25  0.75%Steel (Flat-end) + 0.25% Carbon  U  '10  I20  ASTM C1018 Indices  Fig.5.6. ASTM C1018 Method for HyFRSCC Beams with Steel and Carbon Fibers  8.42  0.75%Steel (Flat-end) + 0.25% Carbon  1% Steel (Flat-end)  5.46 0.75% Steel (Flat-end) 0.5% Steel (Flat-end) 4.39  0.5% Steel (Flat3.79 end) + 0.5%Carbon  Hybrid Fiber Mixes  Fig.5.7. JSCE Method for HyFRSCC Beams with Steel and Carbon Fibers  92  5.4.2 Hybridization of Steel Macro Fiber (Flat-End) with Micro-Polypropylene Fiber The results in Figs. 5.8 to 5.11 show the results of hybridization of 0.75% by volume of steel (flat-end) fibers and 0.25% by volume of micro-polypropylene fibers compared to the results of mono-fiber reinforced SCC mixes containing 1% steel, 0.75% steel, 0.5% steel, and 0.2% micro-polypropylene fibers. Fig 5.8 shows the corresponding loaddeflection curves. It was observed that the mix containing 0.2% by volume of micropolypropylene fiber did not possess toughness on its own. However, hybridization of 0.25% micro-polypropylene+0.75% steel displayed higher fracture toughness compared to the mono-fiber mix containing 0.75% steel. Hybridization of macro and micro fibers demonstrated an evident synergy in this mix. In addition, the improvement of load carrying capacity of the hybrid mix was more noticeable at lower deflections and it showed the benefit of using micro-polypropylene in restricting unstable crack propagation in the matrix. By looking at the PCS graph, Fig. 5.9, on could notice that the curves in this graph had the same order as shown in the load-deflection graph. Additionally, comparison of the mentioned hybrid mix with FRSCC containing only 0.75% steel fiber showed a consistent improvement in the fracture toughness at all the deflections after the peak load. Toughness values of mix 2 (1% steel), mix3 (0.75% steel), and mixl3 (0.75% steel+ 0.25% micro-polypropylene) calculated by three different methods (Figs 5.9, 5.10, 5.11), showed that these three methods produced different results. In the PCS method, the values for different L/M ratios of mix 2 were placed at the top, values of mix 13 at the middle, and mix 3 at the bottom, Fig.5.9. However, in JSCE method, the fracture toughness factors for mixes 2 and 13 were almost equal (Fig. 5.11). This difference was due to calculation of fracture toughness in the JSCE method only to a 2 mm deflection. On the other hand, the results based on ASTM C1018 demonstrated a completely opposite performance as compared with the PCS method, Fig. 5.10. As per the ASTM C1018 method, mix 3 and 2 showed the highest and the lowest toughness indices respectively. However, in the two other methods and even in the load-deflection curves, an opposite conclusion could be drawn. Therefore, it showed the results gained by ASTM  93  CIO 18 were so far from the reality. The source of this problem is in determining of first crack in the load-deflection curve as previously mentioned in section 40  0.75%Steel (Flat-end) + 0.25% MicroPolypropylene 35 30  25 1  Fig.5.8. Flexural Responses for HyFRSCC Beams with Steel and Micro-Polypropylene Fibers  12.00 0.75%Steel (Flat-end) + 0.25% MicroPolypropylene 1% Steel (Flat-end)  10.00  • •-  8.00 ro Q.  ~  tn  6.00  o  0.  -K-  4.00 , 0.5% Steel (Flat-end) 2.00 0.2% micro polypropylene 0.00  ^IC  ^IC  0.2  )iC  0.4  ))C  0.6  0.8  1  1.2  1.4  1.6  1.8  U M (mm)  Fig.5.9. PCS Values for HyFRSCC Beams with Steel and Micro-polypropylene Fibers  94  25  0.75%Steel (Flat-end) + 0.25% Micro-polypropylene  Ii  '20  ho  ASTM C1018 Indices  Fig.5.10. ASTM C1018 Method for HyFRSCC Beams with Steel and MicroPolypropylene Fibers  6.62  0.75%Steel (Flat-end) + 0.25% Micro-polypropylene  5  4  6  0.75% Steel (Flat-end)  0.5% Steel (Flat-end) 4.39  Hybrid Fiber Mixes  Fig.5.11. JSCE Method for HyFRSCC Beams with Steel and Micro-Polypropylene Fibers  95  5.4.3 Hybridization of HPP (Macro-Polypropylene) Fiber and Carbon Fiber Figs. 5.12 to 5.15 show the results of HyFRC with HPP as a major fiber and carbon as a minor fiber. Mixes with 0.5%HPP, 1%HPP, 0.5%HPP + 0.5% carbon, 0.75%HPP + 0.25% carbon, and 0.5%HPP + 0.3% carbon+ 0.2% micro-polypropylene are compared. Load-deflection curves in Fig. 5.12 demonstrated an increase in load carrying capacity with an increase in deflection in all mixes in the post crack region. Comparison of FRSCC with 0.5%HPP and HyFRSCC with 0.5%HPP +0.5% carbon, Fig. 5.12, indicated that carbonfiberdid not play any important role in fracture toughness improvement of the composite. Even use of different volume fraction of carbonfiber,0.75%HPP + 0.25% carbon, mix 17, did not show any improvement. However, the triple fiber mix with 0.5%HPP + 0.3% carbon + 0.2% micro-polypropylene, showed a noticeable improvement in fracture toughness Fig.5.12. This improvement was due to the existence of micro-polypropylene in the mix. The graph of PCS values, Fig.5.13, showed the same behavior as explained for all mixes. It displayed more clearly the deficiency of carbonfiberin different deflections. Although, the JSCE method quantifies the fracture toughness just for a 2mm deflection, the comparison of the mix results in this method demonstrated a good agreement with PCS results (Fig.5.15). In contrast, the results gained from ASTM C1018 demonstrated different performances for the mixes compared with the two other methods, Fig.5.14. While the mix with 1% HPP in JSCE and PCS methods showed the best performance among the other mixes in this group, the ASTM method showed the triplefibermix had a better performance than the other mixes. In another comparison, I and Iio toughness indices for the mix with 5  0.5% HPP were higher than that of the mix with 1%HPP, which meant the mix with 0.5% HPP had higher performance than the mix with 1% HPP, which was not correct at all. In addition, mixes with 0.5% HPP and 0.5%HPP+0.5% carbon in PCS method and JSCE method demonstrated the same level of performance, whereas the ASTM method showed a different level of performance for these two mixes, Fig 5.14. Therefore, the evaluation of this group of mixes with PCS, JSCE and ASTM showed that there is no conformity between ASTM C1018 and the two other methods.  96  40 35  Deflection (mm)  Fig.5.12. Flexural Responses for HyFRSCC Beams with HPP and Carbon Fibers  12 i  0 J 0  , 0.2  0.4  0.6  ,  ,  ,  ,  ,  ,  0.8  1  1.2  1.4  1.6  1.8  L/M (mm)  Fig.5.13. PCS Values for HyFRSCC Beams with HPP and Carbon Fibers  97  2  ri  (To ASTM C1018  '20  Indices  Fig.5.14. ASTM C1018 Method for HyFRSCC Beams with HPP and Carbon Fibers  3.95 1 % H P P 0 . 5 % H P P + 0.2% Micro-polypropylene + 0.3%Carbon  3.36 3.16  0.75% H P P + 0.25%Carbon  0.5% H P P + 0.5%Carbon  2.4  2.41  0.5% HPP  0.5  Hybrid Fiber Mixes  Fig.5.15. JSCE Method for HyFRSCC Beams with HPP and Carbon Fibers 98  5.4.4 Hybridization of Steel Fiber (Crimped) with Micro-Polypropylene Figs. 5.16 to 5.19 describe the results of FRSCC and HyFRSCC containing macro-steel fibers (crimped) and micro-polypropylene. As shown in Fig. 5.16, addition of 0.15% micro-polypropylene to 0.85% crimped steel fiber created a noticeable improvement in the fracture toughness of mono-fiber FRSCC containing 0.85% steel fiber. The PCS graph in Fig.5.17 demonstrate that the hybrid mix had even higher toughness values than the mono-fiber mix with 1% steel for deflections less than 1 mm. By looking carefully at the graph, it can be recognized that the improvement at lower deflections was more pronounced than that at higher deflections. Therefore, the efficiency of micropolypropylene at lower deflections with smaller cracks was more significant.  The JSCE method in Fig. 5.19 indicates that the mix with 1% crimped steelfiberhad higher toughness than the hybrid mix with 0.85% steel + 0.15% micro-polypropylene, while in PCS method, the hybrid mix, up to 1 mm deflection showed better performance and from 1 mm to 2 mm deflection showed inferior performance. Therefore, it can be said that the PCS method presents a better fracture toughness evaluation at different levels of deflection.  On the other hand, the results of the mixes using the ASTM C1018 method, Fig. 5.19, were completely different from the two other methods. While, based on the PCS and JSCE methods, the mix with 0.85% steel crimped had the lowest performance among the mixes in this group, the ASTM method showed the highest performance for this mix.  99  Fig.5.16. Flexural Responses for HyFRSCC Beams with Steel (crimped) and MicroPolypropylene Fibers  100  0.85%Steel (Crimped) 5.60  Hybrid Fiber Mixes  Fig.5.19. JSCE Method for HyFRSCC Beams with Steel (crimped) and MicroPolypropylene Fibers  101  5.4.5 Hybridization of Steel Micro-Fiber with Micro-Polypropylene Figs. 5.20 to 5.23 show the toughness results for hybridization of two micro-fibers: micro-steel and micro-polypropylene fibers. The mixes in this group are HyFRSCC with 0.85% micro-steel + 0.15% micro-polypropylene and FRSSCs with 1% micro-steel fiber, and 0.85% micro-polypropylene fiber. As seen in Fig.5.20, the rate of drop of load in the post crack region of mixes with microsteel fiber was greater than that in mixes containing macro-steel fibers. However, as compared with mixes with other micro-fibers (i.e. carbon and micro-polypropylene) it revealed a higher toughness. Moreover, addition of 0.15% micro-polypropylene to 0.85%micro-steel exhibited a considerable improvement in toughness performance of FRSCC with 0.85%o micro-steel fiber. In this case, it was noticeable that micropolypropylene fibers with low modulus of elasticity improved the load carrying capacity and at the same time enhanced the crack-bridging capacity of the matrix. This comparison showed that a combination of micro-polypropylene not only with macro-steel fiber but also with micro-steelfiberimproved the fracture toughness.  Evaluation of these mixes with the PCS method and the JSCE method, Figs 5.21 and 5.23, demonstrated that these two methods matched well. Conversely, the ASTM method showed a different behavior for these mixes. The mono-fiber mix with 0.85% micro-steel in all three methods had the lowest performance. Two other mixes, FRSCC containing 1% micro-steel  and HyFRSCC  containing 0.85%  micro-steel  +0.15% micro-  polypropylene, had almost identical performance based on the PCS method, Fig.5.21, whereas ASTM method and JSCE method, Fig. 5.22 and 5.23 respectively, presented opposite performance for these two mixes. On the other hand, ASTM method, Fig. 5.22, indicated that the toughness enhancement in the hybridfibermix was more pronounced at higher indices, but in the PCS method the improvement was higher at lower L/M ratios, which is more realistic.  102  30  0  4  0  ,  0.2  ,  0.4  ,  0.6  ,  ,  1  0.8  ,  1.2  ,  1.4  ,  1.6  ,  1.8  1  2  Deflection (mm)  Fig.5.20. Flexural Responses for HyFRSCC Beams with Micro-Steel and MicroPolypropylene Fibers  10 -  1I  0  0.2  0.4  0.6  1  0.8  1.2  1.4  1.6  1.8  2  L/M (mm)  Fig.5.21. PCS Values for HyFRSCC beams with Micro-Steel and Micro-Polypropylene Fibers  103  25  0.85% Micro-Steel + 0.15%Micro-Polypropylene  T»T  I10  '20  ASTM C1018 Indices  Fig.5.22. ASTM C1018 Method for HyFRSCC Beams with Micro-Steel and MicroPolypropylene Fibers  0.85% Micro-Steel + 0.15%Micro-Polypropylene  1% Micro-steel  3.58  3.41  0.85% Micro-steel  2.36  Hybrid Fiber Mixes  Fig.6.23. JSCE Method for HyFRSCC Beams with Micro-Steel and MicroPolypropylene Fibers  104  5.4.6 Hybridization of Steel Fiber (Flat-End) with Micro-Polypropylene and Carbon Fiber at Low Volume Fraction Figs. 5.24 to 5.27 demonstrate the toughness results for hybrid fiber mixes containing maximum 0.5% fibers by volume fraction. The mixes in this category are HyFRSCCs with 0.3% steel (flat-end)+0.2% carbon and 0.3% steel (flat-end)+0.2% micropolypropylene; as well as FRSCC with 0.3% steel (flat-end). Both HyFRSCC mixes showed inferior performance than FRSCC. The reduction in the performance of HyFRSCC containing 0.2% carbon fiber is more visible than that of HyFRSCC containing 0.2% micro-polypropylene Fig.5.24. Therefore, there was no synergy effect in hybridization of fibers at low volume fractions. This can be attributed to the high strength of the matrix and the low volume fraction of fibers. Since the volume fraction of macrosteel fibers in this category was low (0.3%) the instability and sudden drop after the peak load was much larger than in those mixes containing higher volume fractions of macrosteel fibers. The large drop in the load-deflection curve created the condition of dynamic loading for the micro-polypropylene fiber. Therefore, the micro-polypropylene fibers fractured and did not play any role in performance enhancement of HyFRSCC.  In this category, all three toughness evaluation methods, PCS, JSCE, ASTM, displayed a good agreement with each other. In all these methods, FRSCC with 0.3% steel fiber had the best performance, then HyFRSCC with micro-polypropylene fiber and HyFRSCC with carbon fiber, Figs. 5.25 to 5.27.  5.5.7 Hybridization of Steel Fiber (Flat-End) with HPP This category presents the hybridization results of two macro-fibers, steel (flat-end) accompanied by HPP (macro-polypropylene), Figs. 5.28 to 5.31. It can be seen that there is no synergy effect in HyFRSCC in this category. Adding 0.5% HPP to 0.5% steel fiber, produced toughness much less than of produced by FRSCC containing 1% steel fiber. Hence, there was no benefit in combining these two macro-fibers. As in the other categories, it was found that there was a good agreement between PCS and JSCE methods, Figs. 5.29 and 5.31, whereas the results by the ASTM method, Fig. 5.30, did not show any conformity with the results of the two other methods.  105  40  35  30  25  20 re o  T3  0 . 3 % Steel (Flat-end)  /  0 . 3 % Steel (Flat-end) + 0 . 2 % ftlicro-polypropylene  1 0.8  1  0 . 3 % Steel (Flat-e nd) + 0 . 2 % Carbon  1.2  Deflection (mm)  Fig.4.24. Flexural Responses for HyFRSCC Beams with Low Fiber Content  0.8  1  1.2  L/m (mm)  Fig.5.25. PCS Values for HyFRSCC Beams with Low Fiber Content  106  0.3% Steel (Flat-end) 5.09  0.3% Steel (Flat-end) + 0.2%Micro-Polypropylene 3.65 0.3% Steel (Flat-end) + 0 . 2 % Carbon 3.03  Hybrid Fiber Mixes  Fig.5.27. JSCE Method for HyFRSCC Beams with Low Fiber Content  107  o L-i  0  ,  ;  ;  ,  0.2  0.4  0.6  0.8  1  1.2  1.4  ,  ,  1  1.6  1.8  2  Deflection (mm)  Fig.5.28. Flexural Responses for HyFRSCC Beams with Steel & HPP Fibers  ASTM C1018 Indices  Fig.5.30.  ASTM C1018 Method for HyFRSCC Beams with HPP & Steel Fiber  Hybrid Fiber Mixes Fig.5.31.  JSCE Method for HyFRSCC Beams with HPP & Steel Fiber  109  CHAPTER 6  Application of Hybrid Fiber Reinforced SCC in Repair of Reinforced Concrete Beams 6.1. Abstract Five types of repaired beams and virgin reinforced concrete beams were tested under static loading in this study. The beams, with a rectangular section 100x120 mm, were repaired with HyFRSCC and tested in flexure in a simply supported mode on a span of 900 mm under four point loading. These include control virgin beams, and beams repaired by plain SCC, FRSCC containing 1% steel fiber, HyFRSCCs containing 1% steel+ 0.15% micro-polypropylene and 1% steel+0.25% carbon. The influence of bonding between repair material and substrate, casting method and the presence of hybrid fibers on the mechanical behavior of repaired beams was investigated. The results indicate that the mechanical properties of the repaired beams were improved by the use of hybrid fibers in the repair material. The results showed an improvement in flexural capacity of the repaired beams containing hybrid fibers, and exhibited significant enhancement in the energy absorption capacity of the beams in the post peak region.  6.2. Introduction Deterioration of concrete structures such as bridges, multilevel parking garages, tunnels, etc., is a major problem around the world. The deterioration is mostly in the form of spalling of concrete cover, caused by freezing and thawing cycles and corrosion of steel bars within the concrete. This has been attributed to the inadequate cover thickness, high porosity and poor quality of cover and the presence of cracks caused by the shrinkage and temperature. Therefore, repairs are needed not only to improve the aesthetic appearance and reinforcement protection but also to increase the load carrying capacity and energy absorption of damaged members. In addition, in many cases, the repair itself is sufficient to rehabilitate the structure, and it is thus not necessary to rebuild. Hence, repair is often the most cost effective solution. High strength self-compacting concrete (SCC) with high  110  flowability and good segregation resistance is a good option for thin repairs [6]. Since SCC has high flowability, it does not need any consolidation effort to facilitate the process of casting in restricted areas such as in thin repairs. To produce SCC, large amounts of cementitious materials and superplasticizer are used to reduce the yield stress and create high flowability. The yield stress is defined as the necessary stress to initiate the flow of concrete [4]. Both yield stress and plastic viscosity are used to characterize the fresh concrete properties. Also, this concrete shows good durability to freezing and thawing when it has low air-void spacing factors [6]. The mechanical behavior and the modes of failure in repaired concrete structures under loading are affected by the mechanical properties of the repair material and the bond between the repair material and substrate. Introduction of hybrid fibers with different sizes and types in the repair material can not only enhance the durability of the material but also increase the toughness performance of the repaired beam in the post crack region. A hybrid fiber system can inhibit the initiation and propagation of cracks at different levels and reduces the size and amount of cracks in concrete [46]. As a result, it makes the material more ductile, which brings about higher energy absorption for the repaired beam.  6.3.  Test program  6.3.1. Materials Ordinary Portland cement, natural sand, and 10 mm maximum size round aggregates, were used in the preparation of repair material. To get a higher strength, silica fume was added and to have self-compacting properties fly ash and two types of chemical admixtures were used in the mix. Carbon, micro-polypropylene and steel fibers used to create hybrid fiber system in the concrete. The mix proportion and specification of the fibers are presented in Table 6.1 and 6.2 respectively. In addition, the composition of the normal concrete, which was used in un-repaired beams, is shown in Table 6.1.  Ill  Table 6.1.Mixture proportion of self-compacting concrete and normal concrete Mixture Type  Kg/m  Sand Kg/m  Gravel Kg/m  Water Kg/m  Silica fume Kg/m  Fly Ash Kg/m  Super plasticizer (ml/1 Kg cement)  Air Entraining (ml/1 Kg cement)  420  695  1042  123.5  24.7  49.4  10-25  0.9  250  600  1120  125  ....  ....  ....  Cement 3  SCC Normal Concrete  3  3  3  3  —  3  (MPa)  76  35  Table 6.2.Specification of the fibers used in the repair materials Fiber  Type  S  Steel fiber Novotex  C Mp  L  mm  Conoco Carbon fiber Micro polypropylene  50 12.5  12.5  Dimensions D Geometry 1mm 9-11 u.m 2 . denier  E (GPa)  Tensile Strength (MPa)  CrossSectional Shape  Density (kg/m ) 3  Flat-end  212  1150  7850  Circular  Straight  232  2100  1900  Circular  Straight  3.5  375  900  Circular  Shape  Table 6.3.Proportion of fibers in repair material Test  Repair material specifications  Type of beam  Slump flow  f  (cm)  (MPa)  Plain SCC  65  68  1.0% Steel fiber  64  79  61  82  58  75  58  75  Fiber type Bl B2 B3 B4 B5  Repaired with SCC Repaired with FRSCC Repaired with HyFRSCC Repaired with HyFRSCC Intentionally broken & Repaired with HyFRSCC  1.0% Steel +0.25%Carbon 1.0% Steel +0.15%Micropoly 1.0% Steel +0.15%Micropoly  112  c  6.3.2. Specimen Preparation Six 100x120x1000 mm reinforced concrete beams were cast with normal concrete in wooden formwork. The detailing of rebar reinforcement is shown in Fig.6.1.a, and the mix proportions of the concrete used are given in Table 6.1. To determine the compressive strength of the concrete, six 100x200 mm cylinders were cast accompanying each batch. After demolding the beam and cylinder specimens, they were placed in the water tank. After 28 days of curing, three sides of the beams were surface roughened using mechanical chipping. In all, six different beams were tested. One of them was a control beam and the other five were repaired beams, which were upgraded with HyFRSCC. One of the repaired beams was intentionally damaged before repairing to observe the efficiency of the repair method on a damaged beam.  1000 mm 300  300 mm  300 n  50mm  30mm  I  SUPPORT  *3  380 mm 3@75 mm  H—I*  mm I  2* 10 r  1  90mm  30mm  H—k- Old Concrete ffc=30MPal Hybrid Fiber SCC trc=70MPa)  110» 40mni  SUPPORT  35 mm  Fig.6.1. a) Detailing of RC Beam Before Repairing, b) Cross Section of Repaired RC Beam  All the beams that needed to be repaired were placed in a 150x150x1000 mm wooden mold in such a way that repair material could encase the sides and bottom of the RC beams, Fig.6.1.b. The thickness of the layers of the repair material for the sides and bottom of the beam were 30 mm and 40 mm respectively. After adjusting the RC beams in the molds, the repair material was prepared based on Table 6.1 and 6.2 and was poured into the molds through the 30mm width of opening between the beam and the mold. Due to the high fluidity of the repair material, it moved down very fast, filled the bottom of the mold and rose from the other side. No vibration was used during the casting process and concrete was compacted under its own weight. The repair material encased the  113  hardened RC beams from three sides. The combination of the fibers in the different repair material mixes has been shown in Table 6.3. Comparison of the fresh properties of the HyFRSCC mixes by using the slump flow test showed that the mix with micro-polypropylene fiber had less flowability than the other mixes. The flowability of the mix was not enough to fill the mold from one side. Hence, the mold was first filled from one side and left for a few minutes so that the air voids were removed from the mix, then the rest of concrete was poured from the other side to fill the form completely. The repaired beams were demolded after one day and placed in a water tank for 28 days. A typical section of the repaired beam is shown in Fig.6.3.  114  6.3.3.Test Procedure All of the beam specimens were tested in third-point loading using a Baldwin 400 kip universal testing machine. These beams were 1 m in total length and were tested over a 900 mm span, using the loading arrangement shown schematically in Fig. 6.4.  1000 mm 300 mm  LVDT1  300 mm  300 mm  Load LVDT2I  Repaii e d B e a m  Suppoit  LVDT3  I  I Support  Fig.6.4. Laboratory Scale Beam Test Set Up  During each test, mid-span deflections and support deformations were monitored using 3 LVDTs, Fig. 6.5. To compare the performance of the different beams, data acquisition software was used to monitor and collect both the applied load and the three LVDT readings. After correcting the measured deflection values for extraneous deflection, loaddeflection curves were produced for each specimen. From these curves, the first crack load and the ultimate loads were identified. Another quantity calculated from these data was the energy absorbed by the specimen during loading, as represented by the area under the load-deflection curve. Thefinalpiece of information that was reported for each specimen was the mode of failure, which explains the event causing the load to drop after reaching its ultimate value. However, in many cases, the beams after failure were still capable of carrying load.  115  Fig.6.5. Baldwin 400 kip Universal Testing Machine  6.4. Test Results The load-deflection curves were drawn for all beams. The results obtained from the beam test were: the first crack load, ultimate load, failure mode, and the energy absorbed by the beam. The aforementioned results for control beam Bo (virgin beam) and the other beams Bi, B2, B3, B4, B5 are summarized and shown in Table 6.4. The load-deflection curves of the six beams are presented in Fig.6.6.  Table.6.4. The results of repaired beams under flexural loading Beams  B  Beam energy  Failure mode  Type of fiber in  First crack load  Ultimate load  repair materials  (kN)  (kN)  10  53  1076  Flexural-shear  0  absorption (N.m)  Bl  Plain SCC  19  57  1441  Bond-flexural  B2  1.0% Steel fiber  27  66  1957  Flexural  B3  1.0% Steel +0.25%Carbon 1.0% Steel +0.15%MicropoIy  25  69  2003  Flexural  28  62  1910  Flexural  1.0% Steel +0.15%Micropoly  20  61  1395  Flexural-shear  B4 B5 (Broken)  116  0 1 ^ — 0  i  5  1  1  1  [  1  1  10  15  20 Deflection  25  30  35  -I 40  (mm)  Fig.6.6. Load-Deflection Curves for Un-repaired and Repaired Beams, C: Carbonfiber,S: Steelfiber,Mp: Micro-Polypropylene Fiber  6.5. Discussion 6.5.1. Bond Strength Bond strength between repair material and substrate is an important issue, which influences the durability of the repaired structures. Without an adequate bond, a repair layer is ineffective irrespective of its quality. In order to increase the bond strength between the substrate and the repair material, the surface was prepared by mechanical chipping to produce higher interlock in the interfacial zone of two materials. To observe the quality of the bond between the repair material and substrate, several cylindrical specimens with diameters of 50 mm were cored from the repaired beams and compared with each other, Fig.6.7. The observation showed the bond between the repair material and substrate in vertical surfaces, on the lateral sides of the beams, was almost perfect. However, bonding between the two materials on the bottom side of the beam was weak,  117  and it was noticeable with the naked eye. This was due to the presence of entrained air voids on the top layer of the SCC which was in contact with old concrete, which created a weak interface between the old concrete and the repair material. It seem the lateral sides of the beam provided enough bonding between the two materials in the composite beam so that it could perform as an integrated beam element.  6.5.2 Behavior of the Repaired Beams The test results shown in Table 6.4 indicated that there were improvements in the mechanical properties of all repaired beams. Comparison of the first crack load in Bo (control beam) with the other repaired beams showed the first crack loads in repaired beams (i.e. B, to B ) increased 90%, 170%, 150%, 180%, and 100%, respectively. A 5  portion of this improvement is due to increase of the beam cross section. For instance, Bi without any reinforcement in the repair material showed 90% improvement. However, other beams withfiberin the repair material demonstrated more improvement in the first crack load. Even the beam which was intentionally damaged, showed a 100% improvement in thefirstcrack load. Fig.6.8 compares thefirstcrack load improvement in the different beams. Thefirstcrack was considered the point at which the load-deflection curve started being nonlinear.  118  28 27  Un repaired Beam  Repaired with Plain SCC  1.0% Steel fiber  1.0% Steel +0.25%Carbon  1.0% Steel +0.15%Micropoly  1.0% Steel +0.15%Micropoly {Broken Beam)  Fig.6.8. First Crack Load Comparison for Different Repaired Beams  As the applied load increased, the failure mode was not identical for all six beams. For the control beam, Bo, a few vertical cracks first occurred at the bottom of the beam. The width of these cracks increased with an increase in the applied load. However, shear cracks appeared at higher deflections in the middle portion of the beam under the concentrated loads and propagated downward until failure occurred. The failure mode for B 5 (intentionally broken) beam was almost similar to that of Bo. For Bi (unreinforced repair material), cracks appeared in the middle portion, on the bottom side of the beam. When a crack reached the interface, it propagated along the interface until failure of the beam occurred. Debonding and slip were observed between the substrate and the repair material with further loading. The observed debonding can be attributed to the strength of repair material, which was not as high as that of the other repair materials. The mode of failure in B2, B 3 , B 4 was flexural, which showed higher ultimate load than the other beams did. Compared to Bo (control beam), the improvements in the ultimate loads for  119  B2, B3, B4 were 25%, 30% and 17%, respectively. The comparison of ultimate load in different repaired beams is shown in Fig.6.9.  I  5 0  "O  ra  ^  O  ra E  £  3  40  30  Un repaired Beam  Repaired with Plain SCC  1.0% Steel fiber  1.0% Steel +0.25%Carbon  1.0% Steel +0.15%Micropoly  1.0% Steel +0.15%Micropoly (Broken Beam)  Fig.6.9. Ultimate Load in Different Repaired Beams  In addition, the load-deflection curves in Fig. 6.5 indicated that after dropping the load, the beams repaired with hybrid fiber reinforced material carried higher loads, as much as twice that of one repaired with plain SCC. This is due to the presence of steel fibers in the repair material.  Fig.6.10 shows the energy absorption in the repaired beams. The area under the loaddeflection curve was used as an indicator to describe the beam energy absorption. Although the deflection considered for calculation of the beam energy absorption was larger than the serviceability requirement of the beam, the idea was to compare the fiber efficiency in different repaired beams after yielding of steel rebars in concrete. The beam B i (with plain repair material) showed 34% improvement in energy absorption as compared with the un-repaired beam (control). However, the improvements for B2, B3,  120  and B 4 were 80%, 86%, and 76%, respectively. Therefore, the enhancement of the beam energy absorption was noticeable due to use of fibers in the repair material. In addition, the use of carbon fibers in the hybrid fiber reinforcement repair material, B 3 , showed higher ductility than that of the mono steel fiber in the repair material, B2. However, B 4 which was repaired with 1% steel fiber + 0.15% micro-polypropylene fiber, exhibited less ductility than B (1% steel fiber). This function could be attributed to workability 2  reduction of the repair material due to use of micro-polypropylene, which in turn decreased the mechanical properties of the composite beam.  1910  c  0  S-  1200 -  o  JO ro >% O) 900 0)  c E ro w  600  m  Un repaired Beam  Repaired with Plain SCC  1.0% Steel fiber  1.0% Steel +0.25%Carbon  1.0% Steel +0.15%Micropoly  Fig.6.10. Energy Absorption in Different Repaired Beams  121  1.0% Steel +0.15%Micropoly (Broken Beam)  6.6. Conclusion Based on the limited test results reported here, the following conclusions can be drawn:  •  HyFRSCC repair material performed well in terms of fresh properties, flowability through the narrow spaces, andfillingability in the form. Due to its high fluidity it should save energy and is well suited for cast-in situ repairs  •  The repair materials had a good bond with the substrate on the lateral beams sides, while the bonding between the two materials was weak at the bottom of the beams.  •  The efficiency of using hybrid fibers in repair materials was noticeable. The mechanical properties of the repaired beams were significantly affected by the mechanical properties of the repair materials and the fibers used in them. It was also noticed that the mechanical properties of the beams repaired with the composites containing fibers were significantly higher than that of the beam repaired with plain SCC.  •  The beam repaired with hybrid fiber composite materials containing 1% steel fiber + 0.25% carbon fiber, demonstrated the highest energy absorption and ultimate load capacity among all other repaired beams.  •  It should be pointed out that the present investigation was conducted using beam specimens that were smaller than beam members commonly used in the field. Therefore, additional tests using larger size beam specimens are needed to show the influence of beam size on debonding of repair material. Furthermore, selfcompacting concrete is a new material, and further research is necessary to characterize the bond between SCC and old concrete.  122  CHAPTER 7  Conclusions  7.1. Introduction This research focused on the use of self-compacting concrete in combination with hybrid fibers. It was found that it is possible to produce SCC containing hybrid fibers without having detrimental effects on its workability. One of the most important factors related to HyFRSCC performance is its production. This composite was successfully developed and studied with regard to its feasibility in practical applications. Several properties of the composite were studied in fresh and hardened stages. It was also shown how the different types and combinations of macro and micro fibers affected the flowability and compactability of concrete in the fresh state and the compressive strength, MOR, and flexural toughness in the hardened state. The significant conclusions drawn from this study can be grouped into the following categories.  7.2. Production of H y F R S C C 1- The sequence of adding of materials to the mixer was an important factor in producing high flowable concrete. Mixing of fine materials with water and admixtures first, followed by fibers and finally the addition of coarse aggregate produced the most efficient mixing for fresh SCC.  2- Due to high viscosity in the SCC fresh mix, the speed of the mixer was reduced. Therefore, the mixing time of SCC materials was longer than that of normal concrete. The mixing time for SCC materials was 15 to 20 minutes. The mixer therefore should be designed to have enough power to mix the highly viscous materials fast and to prevent fiber balling.  123  7.3. Fresh Behavior of H y F R S C C 1 - Introducing micro-fibers in concrete resulted in an increase in viscosity. This was due to the high specific surface area of micro-fibers. It was necessary to increase the amount of superplasticizer in the mix to maintain a high flowability of the concrete. Thus, there was a limited opportunity for using micro-polypropylene and carbonfibersin the concrete while maintaining a desirableflowability.At the most, 0.15% by volume of micro-polypropylene fiber and 0.25% by volume of carbon fiber could be used. Mixes with higher volume fractions of micro-fibers showed a remarkable reduction in the slump flow. 2- Though less than the micro-fibers, macro-fibers also increased the yield stress and reduced theflowability.This behavior was more pronounced in the case of stiffer fibers (i.e. steel fibers), longer lengths and higher volume fractions of fibers.  7.4. Compressive Strength of H y F R S C C 1- Due to the low water to cementitious ratio (0.24) of the mixes, the compressive strength of concrete was highly sensitive to minor errors in controlling the content of water. 2- It was found that there was an approximate relationship between slump flow diameter and compressive strength in hybrid fiber mixes. This correlation was stronger if only one fiber type was used in the mix. The reason for this behavior was that in hybridfibermixes when a high volume of micro-fiber was introduced, this lowered the slump flow but resulted in relatively higher strengths. On the other hand, mixes with low volume fractions of macro-fibers due to reduced fiber interlocking showed higher slump flow but did not necessarily increase the strength. 3- Incorporation of carbon and micro-polypropylene fibers at low volume fractions did not change the mode of failure from brittle to ductile, and it performed like plain concrete. However, the mode of failure for macro-fibers was not brittle, and the specimens kept its integrity even after failure.  124  7.5. Flexural toughness of H y F R S C C 1- The use of the open loop machine for flexural toughness tests, for all mixes, showed a sudden drop in load in an uncontrolled and unstable manner immediately following the peak load. This was particularly true for high strength mixes. 2- Toughness performance of the mixes was characterized by three methods: ASTM CIO 18, JSCE SF4, and the PCS methods. The results showed that these three methods agreed with each other only in some cases. In a majority of cases, however, there was a good agreement between the results of the JSCE method and the PCS method. Only in a few mixes, was disagreement observed between the JSCE and PCS methods. The ASTM C1018 method did not show a good conformity with the two other methods and in some cases, it even exhibited an opposite performance trend in comparison to other methods. The reason for this behavior was attributed to sensitivity of this method to determination of the first crack in the load-deflection curve. 3- Toughness results demonstrated that the PCS method was the best suited to evaluate the toughness performance and therefore, the concluding remarks in toughness performance of HyFRSCC are summarized based on the PCS method as follow: •  Addition of 0.25% of carbon fibers to a mix containing 0.75% by volume of flat-end macro steel fibers, showed remarkable synergy in the composite. However, other hybrid mixes with carbonfibersdid not show synergy.  •  Incorporation of micro-polypropylene, even at a lower dosage rate of 0.15% by volume, with all macro-fibers (i.e. steel and HPP fibers), exhibited synergy in toughness. The synergy was more apparent in hybrid mixes with flat-end steel macro-fibers and steel micro-fibers.  •  No synergy was noted in composites with low totalfibervolume fraction hybrid composites (0.3% steel + 0.2% polypropylene for instances). The  125  reason of this behavior can be attributed to the high strength of the matrix and very unstable failure after peak load. Combination of two macro-fibers (steel and HPP fibers) did not show any synergy in toughness improvements. It can be concluded that in the SCC reinforcing matrix with two macro-fibers, both acting in the post crack stage, there is no synergy. Hence, it is concluded that for having synergy, it is required that micro fibers be added to enhance the matrix performance at the micro-crack level, and macro-fibers be added to increase the composite toughness induced by fiber pullout.  126  CHAPTER 8 Recommendations for Future Studies  It has been more than a decade since self-compacting concrete has come to the market and has been used in some special projects. However, many questions remained unanswered. For instance, in the production phase, there are no recommendation for mix design, mixing sequence, mixing time, mixer requirement, optimum quantity of cement, and adjustment for local aggregates. In testing and quality control, there are no standardized methods to identify whether or not SCC has the three key properties: filling ability, passing ability, and resistance to segregation. Although, there are a number of test methods developed for specific applications, there is no general agreement on the most suitable one. In Addition, the following aspects of SCC need attention:  •  Dispersion of fibers throughout the matrix in FRSCC depends on the rheological properties of matrix. It is recommended that proper rheological studies be carried out to find the optimum fiber type and volume fraction for yield stress and plastic viscosity that give the best fiber dispersion in the composite.  •  The interfacial transition zone (ITZ) between matrix and fiber in SCC and conventional concrete is expected to be different. Use of a fiber pull out test and scanning electron microscope (SEM) observation will help clarify these differences.  •  Testing high strength concrete in an open-loop machine caused a very sudden drop after peak. To remove this undesirable condition, and have stable conditions, the test should be done in closed-loop machine. The results from the closed-loop tests can better identify fibers and fiber combinations that provide synergy.  •  Durability performance of hybrid fiber composites with SCC needs to be studied.  127  The efficiency of using micro-polypropylene fibers in concrete to reduce the plastic shrinkage cracking has been proven. Evaluation of micro-polypropylene fibers on shrinkage cracking of HyFRSCC should be studied in detail.  Since fibers hybridization in some mixes showed synergy for static condition, it is recommended that studies be carried out to investigate the effectiveness of fibers hybridization in impact loading and fatigue.  In the current research, it was observed that the bond strength between repair material and normal concrete at different interfacial surfaces showed different characteristics. It is recommended that further studies be carried out to characterize the interfacial zone of two concretes by use of scanning electron microscope (SEM) observations.  128  References  1- Mindess, S. and Young, J.F., Concrete, Prentice-Hall, 1981. 2- Nagamoto, N. and Ozawa, K. "Mixture proportions of self-compacting high performance concrete". High performance concrete: design, Materials and Advances in concrete Technology, ACI international, SP-172, 623-666, Dec. 1997. 3- Stroeven P., Shui Z., Qian C , Properties of Carbon-steel and polypropylene-steel hybrid fiber concrete in low volume fraction range, 5th ACI/Can. Conf. Adv. In Cone. Tech. (SP200), 2001, pp 713-732. 4- Tattersall, G.H., and Banfill, P.F.G., The rheology of fresh concrete, Pitman, Advanced Publishing Program, 1989. 5- Ferraris, C.F., Measurement of the rheological properties of high performance concrete, Journal of research of the national institute of standards and technology, Vol. 104, N5, 1999. 6- Lacombe P., Beaupre D., Pouliot N. Rheology and bonding characteristics of self leveling concrete as a repair material, Materials & Structures, Vol 32, Octl999, pp 593-600. 7- Okamura H., Ouchi M. Self-compacting high performance concrete. Progress in Structural Engineering and Materials 1 (4), 1998, pp378-83. 8- Bartos P., Fresh Concrete: Properties and Tests, Elsevier 1992. 9- Banfill P.F.G., Rheology of fresh cement and concrete, E& FN Spon, London 1991. 10- Larrard F., Ferraris, C.F., and Sedran, T. Fresh concrete: A Herschel-Bulkley material, Materials and Structures, 31, 494-498 (1998). 11- Iwasaki N., Estimation of workability- Why has the slump remained beingused so long, Concrete Journal, 21 (10), 4-12 (1983). 12- Ferraris C. and Gaidis J., Connection between the rheology of concrete and the rheology of cement paste, ACI J. 88, 388- 393 (1992).  129  13- Skarendahl A., Petersson O. Self-compacting concrete, State of the art report of RILEM, Report 23, 2000. 14- Malhotra V. M . & Mehta P. K., Advances in concrete technology. Volumel, Pozzolanic and cementitious materials, Gordon and Breach Publishers, 1996, pl91. 15- Nawa T., Izumi T., Edamatsu Y. State of the Art report on materials and Design of SCC, international workshop on SCC, 1998, Japan, pp 160-190. 16- Kurita M., Nomura T., Highly-Flowable steel fiber reinforced concrete containing fly ash, ACI Int. Conf. Bangkok, 1998, ppl59-175. 17- Bartos P., Fresh concrete properties and tests, Elsevier, 1992, p224. 18- Owens, P.L. Fly ash and its usage in concrete; Journal concrete society, England, Vol.13, No. 7,pp21-26, 1979. 19- Sari M., Prat E., Labasrire J. F., High strength self-compacting concrete original solutions associating organic and inorganic admixtures, Cement and Concrete Research, 29 (1999), pp813-818. 20- Shi Yunxing, Matsui Isamu, Feng Naiqian, Effect of compound mineral powders on workability and rheological property of HPC, Cement and Concrete Research, 32 (2002) pp71-78. 21- Ramachandran  V.S., Malhotra V.M., Jolicoeur  C.  and  Spiratos  N.  Superplasticizers: properties and applications in concrete. CANMET publication, 1998. 22- Mindess S., Bentur A., Fiber reinforced cementitious composites, Elsevier Science Publishers Ltd, 1990. 23- Hanehara S., Yamada K., Interaction between cement and chemical admixture from the point of cement hydration, absorption behavior of admixture, and paste rheology. Cement and Concrete Research, 29 (1999) ppl 159-1165. 24- Yamada K., Takahashi T., Hanehara Shunsuke, Matsushisa M., Effects of the chemical structure on the properties of Polycarboxylate-type superplasticizer, Cement and concrete research, 30 (2000), ppl97-207.  130  25- Bartos J.P., Measurement of Key Properties of Fresh SCC, Workshop on measurement,  testing and standardization: Future needs in the field of  construction materials, Paris, June 2000. 26- Uomoto T. and Ozawa K., Recommendation for self-compacting concrete, JSCE Concrete Engineering Series 31 1999, 77pages 27- Beaupre D., Lacombe P., and Khayat K.H., Laboratory investigation of rheological properties and scaling resistance of air entrained self-consolidating concrete. Materials and Structures, Vol. 32, April 1999, pp235-240. 28- Peterson O., Flowing Concrete, Proceeding of an International RILEM Seminar on Workability of Special Concretes and Efficiency of Concrete Mixers, May 1995, Stockholm, Sweden, 5p. 29- Grunewald S., Walraven J.C, Parameter-study on the influence of steel fibers and coarse aggregate content on fresh properties of self-compacting concrete, Cement and concrete research, 31 (2000) ppl793-1798. 30- Groth P., Nemegeer D., The use of steel fibers in self-compacting concrete, International. Symposium on Self-Compacting Concrete (Sweden), 1999, pp497507. 31- Kucharska L., Logon D., Rheological properties of high performance mortars with carbon fibers, Proceeding 42 Conference KILWPAN and KNPZITB, vol. nd  6, 1996,pp37-44. 32- ASTM C1018-94 Standard Test Method for Flexural Toughness and First-crack Strength of Fiber-reinforced Concrete, American Society for Testing and Materials, Volume 04.02, 1998, pp 506-513. 33- Soroushian P. Mirza F., Permeability characteristics of polypropylene fiber reinforced concrete, ACI Material Journal. V.92, No. 3, 1995, pp291-295. 34- Shah S.P., Workshop on Fracture and Transport Performance of Concrete, Italy, 2001. 35- Kovler, K., Sikuler, J., and Bentur, A. Free and restrained shrinkage of fiber reinforced concrete with low polypropylene fiber content at early age, Fiber Reinforced Cement and Concrete, edited by Swamy, E & FN Spon, London, 1992, pp 91-101  131  36- Banthia N. and Sheng J. Micro-reinforced cementitious materials, Materials Research Society Symposium Proceedings, Vol. 211, S. Mindess and J. Skalny editors, Boston, 1990, pp25-32. 37- Kucharska L., Brandt A.M., Pitch based carbon fiber reinforced cement composites- A Review, Archives of Civil Engineering, 1997, pp 165-187. 38- Brandt A.M. Glinicki, Flexural behavior of concrete elements reinforced with carbon fibers, Proceedings of International RILEM Workshop "High Performance Fiber Reinforced Cement Composites" 1991, pp288-299. 39- Banthia N., Genois I., Pitch- based carbon fiber reinforced cement composite, Proceeding 2  nd  CANMET/ACI, International symposium, SP154, Las Vegas,  1995. 40- Banthia N., Moncef K., Chokri J., Micro fiber reinforced composites, Uniaxial tensile response, Canadian Journal of Civil Engineering, 21,6, 1994, pp999-1011. 41- Akihama T. Banno T., Mechanical properties of carbon fiber reinforced cement composites, International Journal of cement comp. and lightwt. Cone, 8, 1, 1986, ppl-21. 42- Chen P.W., Chung D.D.L. Concrete reinforced with up to 0.2% volume of short carbon fiber, Composites, 24, 1, 1993, pp33-52. 43- Kucharska L., Brandt A. M., Microstructure and mechanical properties of cement mortars reinforced with pitch based carbon fibers, Proceeding IV Int. Workshop th  on High Performance Concrete- Characteristic Material Properties and Structural Performance, 269-281, Weimar 1995. 44- Chung D.D.L., Cement reinforced with short carbon fibers: multifunctional material, Composites (Part B engineering), 31, 2000, pp511-526. 45- Chung D.D.L. and Xuli F., Carbon fiber reinforced mortar as an electrical contact material for cathodic protection, Cement and Concrete Research Vol.25, No. 4, pp689-694, 1995. 46- Lawler J.S. Zampini D. Shah S.P., Permeability of cracked hybrid fiber reinforced mortar under load, ACI Materials Journal, Vol. 99, No. 4, 2002, pp379-385.  132  47- Stroeven P., Qian C, Fracture properties of concrete reinforced with steelpolypropylene hybrid fibers, Cement and Concrete Composites 22 (2000), pp343351. 48- Tsukamoto M., and Worner, J. D. Permeability of cracked fiber reinforced concrete, Darstadt Concrete, V. 6 pp. 123-135. 49- Aldea C. M , Shah S.P. and Karr A. Permeability of cracked concrete, Materials and Structures, V.32, pp.370-376. 50- Li V.C., Mihashi H., Wu H.C., Micromechanical model of mechanical response of HPFRCC, in High Performance Fiber Reinforced Cement Composites 2, E & FN Spon, London, 1996, pp43-100. 51- A O Committee 544, State-of-the-art report on fiber reinforced concrete, American Concrete Institute, Detroit, Michigan, 1996. 52- ASTM C39, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimen, American Society for Testing and Materials, Volume 04.02, 1998, pp 17-21. 53- Banthia N. and Trottier J. F., Test Methods for Flexural Toughness Characterization of Fiber Reinforced Concrete: Some Concerns and a Proposition, ACI Materials Journal, V. 92, No. 1, 1995, pp48-57. 54- Method of Test for Flexural Strength and flexural Toughness of Fiber Reinforced Concrete (SF-4) Japan Society of Civil Engineers, 1984, pp58-66. 55- Ferraris C. F. and de Larrard F., Testing and modeling of fresh concrete rheology, Building and Fire Research Laboratory, National Institute of Standards and Technology, NISTIR 6094, 1998.  133  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items