UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hybrid fiber reinforced concrete : fiber synergy in very high strength concrete Gupta, Rishi 2002

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

Item Metadata


831-ubc_2002-0413.pdf [ 15.98MB ]
JSON: 831-1.0063873.json
JSON-LD: 831-1.0063873-ld.json
RDF/XML (Pretty): 831-1.0063873-rdf.xml
RDF/JSON: 831-1.0063873-rdf.json
Turtle: 831-1.0063873-turtle.txt
N-Triples: 831-1.0063873-rdf-ntriples.txt
Original Record: 831-1.0063873-source.json
Full Text

Full Text

HYBRID FIBER REINFORCED CONCRETE: FIBER SYNERGY IN VERY HIGH STRENGTH CONCRETE by RISHI GUPTA B.Eng. (Civil Engineering), Pune University, 1999 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Civil Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 2002 © Rishi Gupta, 2002 In presenting this thesis in partial fulfilment 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 September 5, 2002 Abstract Over the years we have been able to overcome the inherent weaknesses of concrete thereby making it more suitable for a wide variety of applications. One major development has been reinforcement by short randomly distributed fibers that remedy weaknesses of concrete such as low crack growth resistance, high shrinkage cracking, low durability, etc. In fiber reinforced concrete, the use of one type of fiber alone helps to eliminate or reduce the effects of only a few specific undesirable properties. It is believed that the use of two or more types of fibers in a suitable combination would not only help improve more properties of concrete, but would also help provide synergy amongst the fibers. This aspect of combining the fibers, i.e. hybridizing the fibers in a rational manner to derive maximum benefits, was studied in this thesis. High performance fiber reinforced concrete, with a matrix strength of about 85 MPa was used. An attempt was made to make the concrete suitable for practical use, with the required workability, air content, density, etc. This was achieved by making use of proper admixtures including silica fume, superplasticizers and an air entraining agent. The amount and type of fibers to be used in the hybrid composites were planned such that the synergistic behavior of the fibers could be evaluated. The basic property of the hybridized material that was evaluated and analyzed extensively was the flexural toughness of the material. The various fiber types used in diverse combinations included macro and micro fibers of steel, polypropylene and carbon. Control mixes, single fiber reinforced concrete mixes, double fiber hybrids and triple fiber hybrids were investigated. Along with flexural toughness, the size effect of micro fibers, plastic shrinkage resistance, pull-out response of a single macro fiber, impact and shear were also studied. ii Research clearly indicated that there was synergy associated with many fiber types. In particular 2 denier micro polypropylene fibers, when hybridized with the polypropylene macro fibers (HPP) and carbon micro fibers demonstrated maximum synergy. Although significant synergy was observed, it is believed that the synergy is underpredicted in our tests. The minimum volume fraction of macro fibers used for any of the mixes was 0.5% and it appears that this macro fiber volume fraction is too high to observe maximized synergy in the hybrids. This amount of fiber appears to be high enough to make the post peak response of the matrix insensitive to the addition of small dosages of other fibers, such as micro fibers. iii Table of Contents Abstract ii Table of Contents iv List of Tables vii List of Figures viii Acknowledgements xii Chapter 1- Introduction 1 1.1 Overview 1 1.2 Historical Development 2 Chapter 2- Literature Review 5 2.1 Introduction 5 2.2 Types of Fibers used in Hybrids 5 2.2.1 Macro and Micro Fibers 6 2.2.2 Fiber Types 8 2.3 Research in HyFRC- Toughness and Synergy 9 2.4 Other Properties of HyFRC 12 Chapter 3- Materials and Mixes 16 3.1 introduction 16 3.2 Materials Used 17 3.2.1 Cement 17 3.2.2 Aggregate Type 17 3.2.3 Chemical Admixtures 18 Superplasticizers 18 Air Entraining Agent 19 3.2.4 Mineral Admixtures-Silica Fume 20 3.2.5 Fibers 22 iv 3.3 Mixes Investigated 25 Chapter 4- Fresh Properties of Mixes and Preparation of Specimen 27 4.1 Introduction 27 4.2 Production of Mixes and Specimens 27 4.3 Results 31 4.3.1 Slump and Vebe Time 31 4.3.2 Air Content 34 4.3.3 Density 35 Chapter 5- Results and Discussion 37 5.1 Introduction 37 5.2 Test Specimens 37 5.3 Results 38 5.3.1 Compressive Strength 38 5.3.2 Modulus of Rupture (MOR) 39 5.3.3 Four Point Flexure Toughness Test 41 Comparison of Load-Deflection Plots 58 Analysis of Flexural Toughness 65 Test Results 67 Chapter 6- HyFRC: Plastic shrinkage and Single fiber Bond-slip Response 75 6.1 Plastic Shrinkage 75 6.1.1 Introduction 7 5 6.1.2 Test Set-up 75 6.1.3 Experimental Procedure 76 6.1.4 Results 77 6.2 Single Fiber Bond-Slip Response 79 6.2.1 Introduction 79 6.2.2 Pullout Tests- Set up and Discussion 80 6.2.3 Results and Discussion 82 V Chapter 7- Conclusions 84 7.1 Fresh Properties 84 7.2 Hardened Properties 85 Chapter 8- Recommendations for Further Research 89 Bibliography 92 Appendix A 98 Appendix B 102 Appendix C 111 vi List of Tables Table 3.1 Properties of fibers 24 Table 3.2 Details of Final Mixes 25 Table 3.3 Volume fraction of fibers used in the hybrid mixes 26 Table 4.1 Fresh Concrete Properties 32 Table 5.1 Compressive strength results 40 Table 5.2 Modulus of rupture of the different mixes 41 Table 5.3 Toughness Indices and Residual Strength Factors 67 Table 5.4 Mixes in order of their performance 68 Table 5.5 Average MOR and PCS values for the mixes 69 Table 5.6 Synergy Assessment (in percent) 72 Table 6.1 Mix Proportion of Ingredients of Overlay Mortar 77 Table 6.2 Details of the mixes used for the pull-out tests 81 vii List of Figures Figure 1.1 Fiber bridging across cracks 2 Figure 2.1 Effect of Fiber Length 7 Figure 2.2 Pitch and PAN based carbon fibers 8 Figure 3.1 Influence of w/c ratio on cement-paste microstructure 17 Figure 3.2 Microscopic view of the cement particles 18 Figure 3.3 Microscopic view of silica fume 21 Figure 3.4 One-pass finishing process 22 Figure.3.5 HPP crimped fibers, 50mm length 22 Figure 3.6 Dow 2 denier micro poly fibers 22 Figure 3.7 Flat-ended Novotex steel fibers (50mm length) 23 Figure 3.8 Pitch based Conoco carbon fibers, Vi" chopped 24 Figure 4.1 Polypropylene-fiber reinforced paste 27 Figure 4.2 Mixing in Pan mixer 28 Figure 4.3 Silica fume in a paste form 28 Figure 4.4 Omni mixer 29 Figure 4.5 Magnified image of carbon fiber reinforced cement-based composite 29 Figure 4.6 Beam specimens- HyFRC in fresh state on vibrating table 30 Figure 4.7 Beam specimens- HyFRC in finished form 31 Figure 4.8 Slump measurement in a VeBe mould 33 Figure 4.9 VeBe test 33 Figure 4.10 Calibration bowl 35 Figure 4.11 Density vs. Air Content 36 Figure 5.1 Baldwin machine for compression tests 37 Figure 5.2 Failed cylinder specimens 38 Figure 5.3 Cone failure for the cylinder 39 Figure 5.4 Tested beam specimen-fibers bridging the crack 41 Figure 5.5 Arrangement of the beam testing 42 viii Figure 5.6 Complete set-up for flexural toughness test 43 Figure 5.7 Data acquisition system 43 Figure 5.8 Yoke and other accessories 44 Figure 5.9 Representative load vs. deflection curve- open loop 45 Figure 5.10 Representative load vs. deflection curve- closed loop 45 Figure 5.11 Load vs. deflection curve-plain concrete 46 Figure 5.12 Load vs. deflection curve-single fiber (steel) concrete 47 Figure 5.13 Load vs. deflection curve-single fiber (steel) concrete 47 Figure 5.14 Load vs. deflection curve-single fiber (steel) concrete 48 Figure 5.15 Load vs. deflection curve-single fiber (polypropylene) concrete 49 Figure 5.16 Load vs. deflection curve-single fiber (polypropylene) concrete 50 Figure 5.17 Load vs. deflection curve-double fiber (steel + polypropylene) concrete 50 Figure 5.18 Load vs. deflection curve-double fiber (steel + polypropylene) concrete 51 Figure 5.19 Load vs. deflection curve- double fiber (steel + polypropylene) concrete 52 Figure 5.20 Load vs. deflection curve- double fiber (steel + polypropylene) concrete 52 Figure 5.21 Load vs. deflection curve- double fiber (steel + polypropylene) concrete 53 Figure 5.22 Load vs. deflection curve- double fiber (steel + polypropylene) concrete 54 Figure 5.23 Load vs. deflection curve-triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 54 Figure 5.24 Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 55 Figure 5.25 Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 56 Figure 5.26 Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 57 ix Figure 5.27 Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 57 Figure 5.28 Comparison of mixes with different Vf of steel macro fibers 58 Figure 5.29 Comparison of mixes with steel macro fibers and its double fiber (pi) hybrids 59 Figure 5.30 Comparison of mixes with steel macro fibers and its double fiber (p2) hybrids 60 Figure 5.31 Comparison of mixes with steel macro fibers and its double fiber (pi & p2) hybrids 61 Figure 5.32 Comparison of mixes with steel macro fibers and its double fiber hybrids 62 Figure 5.33 Comparison of mixes with steel and polypropylene macro fibers 62 Figure 5.34 Comparison of mixes with polypropylene (HPP) macro fibers and its triple fiber hybrids 63 Figure 5.35 Comparison of mixes with polypropylene macro fibers and its triple fiber hybrid 64 Figure 5.36 Comparison of triple fiber hybrids with polypropylene (HPP) macro fibers 65 Figure 5.37 PCS chart for the mixes N2-N7 70 Figure 5.38 PCS chart for the mixes N8-N13 and N17 71 Figure 5.39 Synergy vs. L/m ratio- mixes containing steel macro and polypropylene micro fibers (pi) 73 Figure 5.40 Synergy vs. L/m ratio- mixes containing steel macro and polypropylene micro fibers (p2) 73 Figure 5.41 Synergy vs. L/m ratio- mixes containing Polypropylene macro (PI & P2) and polypropylene micro fibers (pi) 74 Figure 6.1 Shrinkage chamber with its accessories 76 Figure 6.2 Shrinkage beam specimens with control mix, no fibers 78 Figure 6.3 Shrinkage beam specimens with Vf= 0.1% 78 Figure 6.4 Total crack area vs. volume fraction of fiber 79 Figure 6.5 Schematic of pull-out specimen 80 Figure 6.6 Photograph of specimen within testing apparatus 81 Figure 6.7 Peak pullout loads vs. inclination angle (polypropylene fiber) 82 Figure 6.8 Energy absorption (up to 3 mm of slip) vs. inclination angle (polypropylene fiber) 83 xi Acknowledgements My acknowledgements go out to everyone who was involved in helping me to successfully complete this research project. This also includes people and factors not mentioned below. First of all, I am indebted by Dr. Nemkumar Banthia's scholarly guidance and the long term planning and vision that he had in mind. I also appreciate the confidence that he had in me from the very beginning and the freedom and pressure-free working conditions that he provided me with. I also thank him for giving me an opportunity to work and carry out research on one of the very important and innovative research fields that currently exist. The vast scope of the project was quite overwhelming in the beginning but the support and help of my colleagues was amazing. I wish to thank Emilie who helped me with making concrete through the summer and with a bit of testing later. The technical expertise and continued motivation of my colleagues is really appreciated. I also wish to thank them for the help and inspiration that they provided me with while writing this report. Special thanks to NSERC for their continued support towards innovative research projects like ours. All the support extended by the industries is also appreciated for providing/supplying with materials. The following companies provided fibers for this research project: Synthetic Industries, Dow and Conoco. The materials used for concrete making were bought from Lafarge, Canada. The chemical admixtures were obtained from Master Builder Technologies. xii Thanks are also due to the staff members of the Civil engineering workshop. Help of the supervisory technician- Mr. Harald Schrempp and the engineering technician- Mr. John Wong is greatly appreciated. Last but not the least, I am grateful to my parents (Avinash and Saroj) for giving me this valuable life and for the love and care that they have always given me. My father deserves the credit for exposing me to the field of Civil Engineering at a very early stage in life, and helping me develop a taste for this interesting profession. I also thank my friend Aditi, who provided me with the much needed help. My thanks also go out to my relatives and friends who filled my life with lots of fun and love, which was much needed. xiii Chapter 1- Introduction 1.1 Overview The use of concrete or cement-based materials is quite ancient. With the passage of time the importance of concrete has grown and the limitations of concrete have been gradually eliminated, making the concrete more and more durable and with a higher performance. A very significant development that took place in the history of concrete was the use of rebars in concrete for structural elements. This system was quite efficient in terms of resisting the macrocracks in concrete and in imparting bending strength in flexural members. The purpose was to somehow overcome the low tensile strength of concrete by strategically placing the rebar. Unfortunately, concrete as a material remained weak in tension and brittle. To overcome this, fiber reinforcement was invented. The emergence of Fiber Reinforced Concrete (FRC) is an important achievement in concrete technology. Although the use of straws in bricks and hair in mortar predates the use of conventional Portland cement concrete, the use of advanced fibers in concrete has provided significant opportunities. It is well known that concrete is a very brittle material and that its strain capacity is very low. Due to these problems, the use of FRC has increased over the last twenty years [1]. Fibers are not added to increase the strength of the concrete; instead, they are added to enhance the post crack behavior of concrete. In FRC, cracks are bridged by fibers, inhibiting their growth and providing post-crack ductility. Fibers as reinforcement can be effective in arresting cracks at both micro and macro-levels [2]. When a matrix is strengthened or reinforced due to short fibers, the following improvements can be observed: • Strengthening of the matrix • Fiber bonding and frictional pullout • Bridging of fibers across cracks and crack face stiffness • Stress intensity reduction 1 In Figure 1.1, the bridging of the fibers across the cracks and the path followed by the crack, for a specimen in pure tension is clearly seen. Fiber Figure 1.1: Fiber bridging across cracks The developments in the recent years have led to FRCs that perform like elasto-plastic materials. Also, the different fiber types are being optimized and new ones have been developed to extract maximum benefit from them. Most of the FRCs used today involve the use of a single fiber type. Such concrete can be effective only in a limited deflection range. The science of hybrid composites where two or more fibers are combined to achieve enhancement in the basic properties of the material is underdeveloped, yet very significant. One way of achieving synergy is by judiciously hybridizing or combining different kinds of macro and micro fibers. 1.2 Historical Development Concrete has been under the process of development since a long time. A number of improvements and advancements have taken place with a major leap that came in the form of using rebars in the structural concrete members. Concrete as a material remained brittle and one with low strain capacity. 2 Reinforcement of concrete with short randomly distributed fibers can alleviate some of the concerns stated above. Fibers abate the nucleation and growth of matrix cracks, and bridge them after their creation, thereby providing sources of strength gain, toughness and ductility [1]. A historical perspective to the development of FRC is given below. 1900 asbestos fibers (Hatschek process) 1950 development of concepts and the science of composite materials 1960 FRC 1970 new initiative for asbestos cement replacement 1970 SFRC, GFRC, PPFRC, Fiber Shotcrete 1990 micromechanics, hybrid systems, wood based fiber systems manufacturing techniques, secondary reinforcement, high strength concrete ductility issues, shrinkage crack control 2000+ structural applications, code integration, new products Ludwig Hatschek initiated the modern technology of the reinforcement of brittle cement concrete matrix by asbestos fibers in around 1900. He had experimented with the reinforcement of asphalt and oxychloride cement by asbestos fibers, before he adopted the Portland cement binder. This process proved to be a large commercial success. The concept of using organic fibers to improve the behavior of building materials is however not new. Previous examples include adding straw fibers to sun dried mud. bricks, mud walls, etc. Mud based water-proofing system utilizing agricultural waste product as fibers were also used. The first recorded use of bamboo reinforcement was in China in 1919 and several uses were repeated in U.S.A. and Japan during World War II. In 1963, Romualdi and Batson published their results on steel fiber reinforced concrete. This was followed by publication of pioneering work on reinforcement of cement and concrete by glass fibers by Krenchel in Denmark and by Biryukovich in the then 3 U.S.S.R. Glass fibers, with a sufficient degree of alkali resistance were developed by Dr. A. J. Majumdar at the Building Research Establishment in U K in 1967. Also, polymer fibers such as Polypropylene, Nylon and Polyethylene were investigated in the early 1960s as reinforcement for concrete. Following widespread concerns about the ill-effects of asbestos (causes a lung disease called Asbestosis), the asbestos-cement industry witnessed an irreversible decline which gave further stimulus to alternate fiber types and FRC technology in the early 1980s. FRCs have been used in slabs on grade, shotcrete, in repair material, as precast products and so on. In recent years, researchers have realized the benefits of combining fibers, in terms of extracting synergy and improving the response of the hybridized material. The benefits derived from incorporating organic (polypropylene and nylon) and inorganic fibers (glass, asbestos and carbon) to achieve superior tensile strength and fracture toughness were recognized about 25 years ago by Walton and Majumdar [3]. It is hoped that there is some interaction between the fibers so that the resulting properties exceed the sum of properties provided by individual fibers. This is often termed as "synergy". 4 Chapter 2- Literature Review 2.1 Introduction Concrete failure initiates with the formation of microcracks (size in the order of microns), which eventually grow and coalesce together to form macrocracks. The macrocracks propagate till they reach an unstable condition and finally result in fracture. Thus, it is clear that cracks initiate at a micro level and lead to fracture through macrocracking. Fibers, if present, help to abate both the micro and macro cracks from forming and propagating. Hybrid fiber composites have not been investigated extensively. According to Kobayashi and Cho, there was limited activity in the field of hybrids in the 1980s [4], with an increased interest in the hybrids in the 1990s [5-10] and later [11-13]. The most investigated fiber combinations so far are those of steel and polypropylene [11], [4-7 and 9], but other combinations such as polyethylene fibrils and polyethylene pulp [8], steel and carbon [14], mineral fibers (alumina, carbon) and fibrillated polypropylene [10], fibrils of polypropylene and A R glass [12], and steel and steel [13] have also been investigated. 2.2 Types of Fibers used in Hybrids When two or more fibers are simultaneously used, they can be combined in the following ways [1 and 12]: a. Hybrids based on fiber constitutive response: While the stronger and suffer fibers increase the first crack strength and the ultimate strength, the flexible fibers not only help increase the load carrying capacity in the post crack region, but also help increase the toughness and strain capacity of the specimens. There is generally a significant difference in the modulus of 5 elasticity of the two types of fibers mentioned above, and ideally they carry load commensurate to the value of strain the material is subjected to. b. Hybrids based on fiber dimensions: Smaller fibers inhibit the propagation of the micro cracks and their growth into macrocracks, which helps increase the tensile strength of the material. Larger fibers on the other hand help control the propagation of the macro cracks and inhibit the formation of newer ones. This characteristic of the macro fibers increases the post peak fracture toughness of the composite. This is consistent with Banthia's and Shah's [15 and 16] views mentioned earlier. c. Hybrids based on fiber function: While certain fiber types are intended to improve the presetting properties such as resistance to plastic shrinkage, others are intended to improve the hardened/mechanical properties. When these fibers are hybridized, fresh and hardened properties of the composites are simultaneously enhanced. 2.2.1 Macro and Micro Fibers Hybrids can be based on fiber dimensions and can be classified as micro and macro; macro fibers generally being 30-60 mm in length as opposed to about 5 mm size for micro fibers. Based on the size of the fibers, micro and macrocracks can be controlled. The diameter is generally in the range of 0.5 mm for macro fibers and about 20 microns for the micro fibers. Another approach of distinguishing fibers is based on the specific surface area (SSA) defined as the surface area for a unit mass [15]. Mathematically, SSA=2(2l+d)/ldDf\ n.l) where, 1 = length of a circular fiber d = diameter of a circular fiber Df = density of the fiber material 6 In Equation 2.1, note that the diameter of the fiber plays a more important role than its length. Based on this formula, the approximate SSA value for a commonly used steel macro fiber is calculated below. • Steel macro fiber (Novotex) SSA = 2((2 * 5) + 0.1)/(5 * 0.1 * 7.85) SSA = 5.2cm21 gm Similarly, SSA values for some other fibers are: Polypropylene macro fiber (HPP) = 44 cm 2 / gm, Polypropylene micro fibers (Dow 2 denier) = 2616 cm 2 / gm, Carbon micro fibers (Conoco) = 2106 cm 2 / gm. Arbitrarily speaking, macro fibers have an SSA of roughly 10 cm2/gm and micro fibers have an SSA greater than 500 cm2/gm. The micro fiber with a large SSA are expected to reinforce the cement paste and mortar phases, thereby delaying crack coalescence and thus increasing the apparent tensile strength [15 and 16]. The function of the macro fibers, on the other hand, is known to bridge across the macrocracks and induce post-crack ductility in the material. L<le L=lc L>lc Figure 2.1: Effect of fiber length 7 Both micro and macro fibers may exceed the critical length and hence fracture across a crack. Figure 2.1 shows the stress distribution across the length (L) of a fiber, which determines the pull-out or fiber fracture mechanism. In the Figure, l c is the critical length of the fiber. When L< l c , fibers fracture and when L> l c , fibers get pulled out. It is therefore the aspect ratio that governs the possibility of fiber fracture, not the SSA. 2.2.2 Fiber Types Mono fiber systems have been used to produce FRC for a long time, and the properties and enhancements induced by the various types of fibers such as steel, glass, synthetic and natural, in plain cement matrices have been reported by ACI Committee 544 [17]. Besides asbestos, glass and steel, different kinds of synthetic fibers such as polyethylene, polypropylene, polyamide and others are used for cementitious composites [7]. Carbon fibers offer unique, high-performance properties for a broad range of applications. Relative to the other synthetic fiber types, carbon fibers may exhibit a high tensile strength and elastic modulus. Carbon fibers based on a mesophase pitch, which by definition is more ordered, crystalline and has a higher cross-link density, are as strong and as stiff as PAN-based fibers. Pitch-based fibers based on isotropic pitch are generally of low modulus and low strength. The micro structure of PAN and pitch-based fibers can be seen in Figure 2.2 [18]. Pan Figure 2.2: Pitch and PAN based carbon fibers 8 2.3 Research in HyFRC- Toughness and Synergy One measure of evaluating the improvements in fiber reinforced concrete is through flexural toughness tests. Methods of evaluating this property are the A S T M C 1018 [19] and JSCE SF4 [20] techniques, which are the two most widely used methods. Another method is the Post Crack Strength (PCS) method as proposed by Banthia and Trottier [21], which provides a more suitable toughness characterization scheme for hybrids. Double Fiber Hybrid Composites Glavind and Aarre [6] investigated the possibilities of increasing the fracture toughness of high-strength concrete by adding fibers. Tests were conducted with both normal and high-strength concrete containing different amounts of steel and polypropylene fibers and toughness was described by toughness indices. The investigation showed that the addition of steel fibers was effective in increasing the toughness and non-linear load carrying capacity for high-strength concrete. Steel fibers also marginally increased the compressive strength of high strength concrete. The combination of different fibers for thin walled fiber reinforced cement composites for structural applications, with fibers made of metallic, mineral, polymeric or naturally occurring materials was reported by Ramanalingam et al. [22]. The authors stated that different fibers offer different extent of improvement to the mechanical properties of the matrix, but the most sought-after property in a cement-based composite is the strain hardening response that is associated with multiple cracking. The authors believed that in order to achieve this property, different types of fibers should be suitably combined to exploit their unique properties. Another investigation was conducted [23] to study fiber denier (weight in grams of 9000 meters of fiber) effect on toughness of FRC. In this investigation, 3 denier and 6 denier polypropylene micro fibers were added at dosage rates of 0.3% and 0.45% respectively, with 0.3% of flat-ended 50 mm steel fibers. Toughness of high strength HyFRC was determined by A S T M C 1018 [19] and A S T M C 1399 [24] methods. The results of 9 A S T M C 1399 showed that the 6 denier micro fibers gave the hybrid reinforced concrete a greater residual strength than that obtained by the addition of 3 denier fibers. Similar results were obtained from the A S T M C 1018 analysis. Qi et al. [25] tested carbon and polypropylene fiber hybrids and found that hybrid fiber reinforced concrete when tested in tension, resulted in debonding between the matrix and the fiber and a great number of fibers pulled out. The stress was always first transferred from the matrix to the carbon fibers, due to its high modulus. After some displacement, the polypropylene fibers would carry most of the load and a large deformation would induce failure. In the study, it was also noticed that carbon and polypropylene fibers when used in a combination offered toughnening at different structural levels. Significant strengthening and toughening occurred in carbon-polypropylene hybrid fiber reinforced concrete. Horiguchi and Sakai [26] investigated steel and PVA hybrids. They studied the fracture toughness of fiber reinforced concrete in compression as well as flexure. Four different types of steel fiber and two types of PVA fiber individually and in combination form were investigated. The toughness of hybrid fiber reinforced concrete showed different performance from that of mono fiber composites and for the same flexural toughness, the first crack deflection was greater in the hybrids. They also observed that while the compressive strength did not change, the addition of fibers significantly enhanced the compressive toughness. Compressive toughness was found to increase in proportion to the fiber content; the highest compressive toughness occurred for the hybrids. No clear correlation however between the compressive and flexural toughness was noticed. Feldman and Zheng [7] also studied hybrid composites containing steel and polypropylene fibers. From their investigation, they concluded that workability of hybrid fiber reinforced concrete was reduced as steel fiber and/or polypropylene fiber contents were increased. Deformed fibers were found to be more effective than straight fibers in enhancing the flexural and compressive behaviour of concrete. They also found that in hybrid composites, stronger and stiffer steel fibers improved the ultimate strength, while 10 the more flexible and ductile polypropylene fibers led to improved toughness and strain capacity in the post-crack zone. A combination of fibers can be very effective, resulting in some form of synergy. Soroushian et al. reported synergy in hybrid polyethylene fiber reinforced cement composites [8]. They optimized the combined use of two different fiber types (high modulus polyethylene fiber and a fibrillated polyethylene pulp) in cementitious matrices. It was noted that in the case of impact resistance, the favorable effect of each fiber was further enhanced in the presence of the other fiber. For flexural strength and toughness, the combined use of polyethylene fiber and pulp produced highly desirable results. The authors also noted that the negative effect of fibers on compressive strength was less pronounced when the two fiber types were used in combination. Qian and Stroeven [11] investigated the optimization of fiber size, fiber content and fly ash content in hybrid polypropylene-steel fiber concrete with low fiber contents. They found that the presence of fine particles of an admixture helped fiber dispersion and that different sizes of steel fibers contributed to different mechanical properties. Additions of a smaller fiber had a significant influence on the compressive strength, but the splitting tensile strength was only slightly affected. The authors found that the optimum dosage of polypropylene fiber was 0.15% and concluded that synergy existed in their hybrid fiber composites. Triple Fiber Hybrid Composites The combination of micro and macro fibers for reinforcement was explored in an investigation carried out at the University of British Columbia [27]. In this investigation, the fibers used were a combination of synthetic fibers, such as polypropylene 2-denier (manufactured by Dow), polypropylene 3-denier (Stealth, manufactured by Synthetic Industries), high performance polypropylene (HPP) and carbon micro (manufactured by Conoco). The objective of the research was to evaluate the performance characteristics 11 for HyFRC, including flexural strength and toughness. The compressive strengths of HyFRC were not found to be significantly greater than their plain counter-parts. However, HyFRC was found to be substantially tougher, with a much greater resistance to cracking. To evaluate the synergistic effects associated with the different fibers, load-deflection curves and Post Crack Strength (PCS) curves were plotted. The authors preferred PCS curves as they depicted synergy at certain units of deflection and provided an accurate measure of synergy at low and high deflections. The addition of fiber concentrations less than 1.3% by volume did not show any particular decrease in compressive strengths. In the triple fiber hybrids, based on 1.0% of HPP fiber, an addition of only 0.15% volume of carbon micro fiber and 0.15% volume of a micro polymer fiber resulted in a large increase in flexural toughness. Polypropylene 2 denier fibers, were superior to polypropylene 3 denier fibers in hybrids in terms of flexural toughness. In some hybrid mixes, the carbon fiber's ability to prohibit micro-cracks at small deflections and polypropylene fiber's ability to resist micro cracking at large deflections was apparent. Also, when these micro fibers were combined, an increase in the fiber-matrix bond occurred as evident from the pull-out tests. A direct relationship between the magnitude of the first crack and overall toughness was observed. The authors also noted that when the first crack occurred, energy stored in the matrix was suddenly released. This is consistent with the findings of Soleimani [28]. Soleimani noticed that this release is mainly absorbed by the macro fiber and the specimen's structural integrity becomes dependent on the effectiveness of fibers that bridge this first crack. 2.4 Other Properties of H y F R C The benefits of hybridizing fibers are observed not only in flexural and compressive toughness but also in the other properties. The interaction between polyethylene fiber and pulp in controlling the specific gravity, volume of permeable voids, and water absorption capacity of cementitious materials has been found to be either negligible or only moderately significant [9]. 12 Valle' and Buyukozturk [29] conducted experimental and modeling studies on polypropylene and steel fibers, in beams with or without conventional stirrups. In general, fibers proved to be more effective in high-strength concrete than in normal strength concrete based on ultimate load and overall ductility. Steel fibers were found to be far more effective than polypropylene fibers. In tests, where a combination of fibers and stirrups was used, a slight increase in ultimate load with major improvements in ductility were noted in comparison to beams made of plain concrete. In another investigation [30], floor slabs were tested under static loading. They were related to the toughness characteristics by testing beams according to A S T M C 1399. A 30 MPa plain concrete slab and slabs containing fiber reinforcement (fiber contents of 0.25% and 0.35%) were examined. The fibers used in the investigation were macro steel and micro polypropylene. From the acquired data, maximum peak load, flexural strength and the average residual strength from the beam specimens was calculated. The authors proposed to relate the residual strength of FRC to an increase in structural capacity of slabs. Durability studies were conducted by Larsen and Krenchel [5], who examined steel, AR-glass, synthetic (polypropylene) and natural (sisal and paper pulp cellulose) fiber reinforced concrete for exposure to climatic conditions. They concluded that natural and glass fibers lose strength and ductility when subjected to normal out-door exposure. They also investigated a combination of what they believed to be the best fibers (steel and polypropylene), and found that the ductility and fracture energy of the composites were increased significantly even after a very long period of exposure. The researchers concluded that fracture energy for composites will change with time, depending on the type of reinforcement and the matrix. Under repeated loading, fiber reinforced concrete containing several volume fractions of two types of fibers, polypropylene and steel were tested by Komlos and Nurnbergerova [9]. The authors observed that the increase in steel fiber content led to balling, while the increase in polypropylene fiber content led to a harsh mix. The fracture and impact 13 energy, as well as toughness and ductility were found to be substantially improved by an increase in the volume fraction of polypropylene fibers. In an investigation by Sun, Luo and Qian [31], shrinkage and water permeation properties of high-performance concrete with expansive agents and hybrid fibers (steel, polyvinyl alcohol and polypropylene fibers) were studied. Test results indicated that the hybrid fibers of different types and sizes could reduce the size and amount of cracking at different levels. Kim and Sakai [32] investigated hybrid fiber reinforced concrete containing micro and macro fibers to enhance the resistance to cracking by thermal stresses. The physical properties and crack resistance at early ages was examined. In this investigation, steel fibers and polypropylene fibers 6, 12, and 30 mm in length were used. It was concluded that hybrid fiber reinforced concrete had high resistance to thermal cracks and it was possible to control these cracks. In the case of steel fibers, crack resistance improved monotonically with fiber volume in the range of 0.1-2.5 %. However, for polypropylene fibers, 0.3% was the optimum volume fraction. Synergy in cement-based hybrids is not only in terms of toughness but also in terms of other secondary properties such as electrical conductivity. This is evident from the research carried out by Banthia, Djeridane and Pigeon [33], where synergy was observed in electrical conductivity due to the use of carbon (micro) and steel (micro) fibers. Electrical resistivity measurements were conducted on cement pastes reinforced with conductive micro-fibers of carbon and steel both in mono and hybrid forms, using a high frequency alternating current. Although carbon fibers themselves are far less conductive than steel fibers, cement composites with carbon fibers were found to be far better conductors than those with steel fibers. It was concluded that, more than the conductivity of the material itself, it is the size and the distribution of the fiber in a composite that is of importance. For the same reason, some of the hybrid-fiber composites were found to have conductivities better than their equivalent mono-fiber systems. Improvements in conductivity with carbon fiber were by almost three orders of magnitude as opposed to 14 one order of magnitude increase for steel fibers. Increase in conductivity with increase in the volume fraction of the fibers was also noticed for steel and carbon fibers, but the increase in conductivity because of steel fibers was lower than that due to carbon fibers. They concluded that the two fibers in a hybrid composite had an interactive behavior rather than a purely additive behavior. To summarize then, it is clear that the concept of hybridizing fibers is a sound one and significant improvements have been reported by various investigators. It is also clear that the influence of concrete matrix strength is not well understood. Equally essential is to develop some rational guidelines for fiber hybridization, and some such attempts were made in this thesis. 15 Chapter 3- Materials and Mixes 3.1 Introduction Strategic Highway Research Program SHRP-C-205 [34] has defined high performance concrete (HPC) as follows: HPC shall have the following strength characteristics: • 4-h compressive strength >2500 psi (17.2 MPa) termed as very high early strength concrete (VES), or 24-h compressive strength >5000 psi (34.5 MPa) termed as high early concrete (HES), or 28-day compressive strength >10000 psi (68.9 MPa) termed as very high-strength concrete (VHS) • A durability factor greater than 80% after 300 cycles of freezing and thawing • A water-cementitious material ratio <0.35 "High performance" in high performance concrete may imply high strength, high durability and/or other characteristics needed for a specific application. In this context, toughness is another property implying high performance, specifically for fiber reinforced concrete and the toughness tests provide a quantitative measure of this property. The objective of this project was to develop hybrid fiber composites with steel and polypropylene macro fibers as the primary fibers. The scope of the research project was wide with many single fiber control mixes and double and triple fiber hybrids. The research also targeted quantifying the synergy associated with each mix, and the fiber volume fraction in each mix was judiciously decided so as to be able to quantify synergy. Various admixtures were incorporated in the hybrid mixes to make them workable and at the same time keeping them feasible for practical use. 16 3.2 Materials Used 3.2.1 Cement Cements satisfying the requirements of a certain class might be quite different in fineness and chemical composition. When used in high-strength concrete (same w/c ratio), these cements may result in different fresh and hardened state properties, especially when chemical admixtures and supplementary cementing materials are used. A low w/c ratio, has a considerable influence on the cement paste microstructure itself, as seen in Figure 3.1 [35]. Fresh Cement Paste Hydrated Cement Paste Figure 3.1: Influence of w/c ratio on cement-paste microstructure (Source- Aitcin and Mindess, "High-Performance Concrete: Science and Applications) For the research work, CSA type-10 cement was used in all mixes. This type of cement is most suitable for use when special properties of other types of cements are not required. Figure 3.2 [36] shows a microscopic view of cement particles. 3.2.2 Aggregate Type While in normal strength concrete, the aggregates need to be clean and properly graded, for HPC, all the constituent materials, i.e., cement, aggregate, mineral and chemical admixtures require careful consideration. Moreover, the properties of aggregates, such as 17 grading, particle shape, strength, stiffness and chemical reactivity with cement are all important for the production of high-performance concretes. Figure 3.2: Microscopic view of the cement particles [Courtesy- Barzin Mobasher, Arizona State University] In this investigation, uniformly graded aggregates were used in all mixes to help reduce the void content between the adjacent aggregates. It is generally believed that high fines content is not required because high strength concrete already contains a large volume of cement paste and supplementary cementing materials. For HPC, a fineness modulus in the range of 2.7 to 3.0, or even higher, is often recommended [37]. An earlier attempt to determine the optimum maximum size of coarse aggregates (MSA) for high-strength concrete resulted in an optimum MSA of 10 to 15 mm [38]. Based on this, the aggregate size in between 10 and 14 mm was used for the high-strength concrete mixes. 3.2.3 Chemical Admixtures Superplasticizers One of the serious shortcomings of fiber reinforced concrete is that the workability is greatly reduced due to the addition of fibers, and this can be corrected by adding admixtures such as superplasticizers. Superplasticizers (another term for high-range 18 water reducers) are added to a concrete mix (with a low to normal slump and low water-cementing materials ratio) to help increase its slump and flowability. Superplasticized concrete unfortunately entraps more air than conventional concrete, but most of this air is generally lost during transportation and casting using vibrators. For historical reasons, polynaphthalene sulfonate superplasticizers have been most commonly used in North America and Japan, while polymelamine superplasticizers are commonly used in Europe [35]. In the case of high strength concrete, where fine admixtures such as silica fume are used, it is very important to have an effective superplasticizer which would accommodate the high demand of workability in the mix. For this study a superplasticizer "Glenium 3000NS", formerly Rheobuild 3000 [39], manufactured by "Master Builders Inc" was used. It is a cement dispersant with a specific gravity of 1.087. This superplasticizer is a ready-to-use, high-range water-reducing admixture, based on polycarboxylate chemistry and meets A S T M C 494 [40] requirements for Type A water-reducing, and Type F high-range water-reducing, admixtures. The product is meant to produce a cohesive and non-segregating concrete mixture, with faster strength development. A i r Entraining Agent Air entrained concrete contains numerous microscopic spherical air voids that relieve internal pressure from expansion of water when it freezes. This, in turn, results in concrete that is highly resistant to cycles of freezing and thawing and with a high degree of workability. Air-entrained concrete is produced by using air-entraining Portland cement, or by introducing air-entraining admixtures when the concrete is mixed. The amount of entrained air is usually between 5 percent and 8 percent of the volume of the concrete, but may be varied as required by special exposure conditions. For this study, an air-entraining admixture "MB-VR standard" [41], manufactured by "Master Builders Inc." was used. This is a resin solution of Sodium Hydroxide and uneutralized vinsol resin with a specific gravity of 1.037. It is ready to use in proper concentration for rapid and accurate dispensing, and offers resistance to damage from 19 freezing-thawing, permeability, segregation, bleeding and resistance to scaling from deicing salts. 3.2.4 Mineral Admixtures- Silica Fume High-strength concretes generally contain one or more supplementary cementitious materials such as fly ash, silica fume, blast furnace slag, metakaolin, etc. Until a few years ago, about 40 MPa concrete was considered to be high strength, but today, using silica fume, concrete with compressive strength in excess of 100 MPa can be readily produced [42]. It is possible to achieve compressive strengths in the range of 60-100 MPa without any supplementary cementitious materials [43 and 44], but the advantages of using additives such as silica fume facilitates production, reduces costs, and increases durability. It is interesting to note that the use of about 10% silica fume in high-performance concretes with very low water/binder ratios results in a significant decrease in the superplasticizer dosage required for a given workability [45]. The silica fume used for this research investigation was procured from Lafarge Canada Inc. and it satisfied the requirements of A S T M C 1240 [46]. Silica fume is a type of super pozzolan, a siliceous material (primarily amorphous silicon dioxide) that in the presence of moisture reacts chemically with calcium hydroxide released by the hydration of Portland cement to form compounds possessing cementing properties. Silica fume is a finely divided residue resulting from the production of elemental silicon or ferro-silicon alloys. A few investigations [47 and 48] have shown that the addition of 10-15% silica fume is very effective in enhancing the Interfacial Transition Zone (ITZ) and the cement-aggregate bond by eliminating many of the larger pores in the ITZ, making its structure more homogenous, eliminating the growth of calcium hydroxide, or by transforming the calcium hydroxide to calcium silicate hydrate by the pozzolanic reaction. It is not so much the pozzolanic reaction of the silica fume, but its filler effect that plays a major role in controlling bond, by modifying the rheological properties of the fresh concrete in such a way that the internal bleeding is reduced [49]. 20 Figure 3.3: Microscopic view of silica fume [Courtesy- Barzin Mobasher, Arizona State University] According to the Silica Fume Association, U .SA. [50], silica fume consists of individual particles which are extremely small (approximately 1/100th the size of an average cement particle), as seen in Figure 3.3 [36]. Concretes made using silica fume are resistant to penetration by chloride ions. Also, flatwork containing silica fume concrete generally requires less finishing effort than conventional concrete. This is evident from the "one-pass" finishing process, seen in Figure 3.4, in which the silica-fume concrete is placed, consolidated, and textured with little or no waiting time between operations [42]. Silica fume in FRC can greatly improve the bonding characteristics of the fibers. For example, polypropylene fibers bond in the concrete matrix by a mechanical interlocking and the bond can be improved by using silica fume in the mix. Also, adding silica fume simultaneously with a water-reducing agent is useful for dispersing carbon fibers, thus improving the physical properties of the composite and increasing the paste-to-fiber bond. 21 Figure 3.4: One-pass finishing process 3.2.5 Fibers Various types of fibers were used for this project, including steel and polypropylene macro fibers and carbon and polypropylene micro fibers. These fibers are described in the following section with the physical and mechanical properties of the fibers included in Table 3.1. Polypropylene fibers: These fibers are produced from homo-polymer polypropylene resin and can be fabricated either as fibrillated fibers or as monofilament fibers. Figure 3.5: HPP crimped fibers, 50mm length Figure 3.6: Dow 2 denier micro polymeric fibers 22 As polypropylene is hydrophobic, polypropylene fibers do not bond to the concrete mix. Bonding is therefore accomplished via mechanical interaction. High Performance Polypropylene (HPP) macro fibers, manufactured by Synthetic Industries were used in this study. These fibers were combined with either 2 denier (manufactured by Dow Chemicals) or 3 denier (Stealth, manufactured by Synthetic Industries) micro fibers. HPP fibers were 50 mm crimped (Figure 3.5) and polypropylene micro fibers were 12.5 mm long (Figure 3.6). Steel fibers: Flat-ended steel macro fibers, 50 mm long were used for this research (Figure 3.7). These fibers were flat-ended and were provided by Synthetic Industries. Figure 3.7: Flat-ended Novotex steel fibers (50mm length) Carbon fibers: History of carbon fibers dates back to the 1800s, when they were discovered by Thomas Edison, who "baked" a piece of ordinary cotton thread to produce a pure carbon fiber to use as filament in the first incandescent light bulb [18]. Based on this, many years later a process was developed for carbon fibers to be produced from synthetic fibers. Carbon fibers are produced as strands that contain several thousand individual filaments. In this experiment, a general-purpose micro carbon fiber (manufactured by Conoco) was used. The fibers produced by Conoco Incorporation, shown in Figure 3.8, were lA" chopped and mesophase pitch-based. 23 Figure 3.8: Pitch based Conoco carbon fibers, Vi" chopped The physical and mechanical properties of the different types of fibers used for the research project are summarized in Table 3.1. TABLE 3.1: Properties of fibers Fiber Code Type Dimensions E (GPa) Tensile Strength (MPa) Density (kg/m3) Cross-Sectional Shape Shape L (mm) D Geometry SI Steel fiber Novotex 50 1 mm Flat-end 212 1150 7850 Circular S2 Steel fiber Xorex 50 1 mm Crimped 212 1200 7850 Crescent c Carbon fiber 12.5 9-11 urn Straight 232 2100 1900 Circular PI Macro polypropylene 50 1mm Crimped 3.5 375 900 Rectang-ular Pi Micro polypropylene 12.5 2 denier Straight 3.5 375 900 Circular p2 Micro polypropylene 12.5 3 denier Straight 3.5 375 900 Circular Legend used: SI: Steel fiber (Novotex), PI: HPP macro fiber, 50mm long c: Carbon fiber (with sizing treatment), pi: Dow micro fiber (2 denier) p2: Stealth fiber (3denier) 24 Some mixes were also tried with the "Dow self-fibrillating fiber" (not mentioned in Table 3.1), which were 50 mm long with a straight geometry. The mechanical properties of these fibers were similar to those of the HPP (PI) fiber. 3.3 Mixes Investigated The standard water/cement laws and theories work well for concretes, where the weakness lies in the cement paste-aggregate interface rather than the aggregates themselves. On the contrary, for high strength concrete, the weaker phase may be the aggregate themselves rather than the interfacial zone. In this research, a few preliminary trial mixes were cast to study the fresh properties and compressive strength of plain and hybrid concrete mixes. Six cylinders and beams were cast for most of the trial mixes, including plain concrete mix, which was used to evaluate the matrix compressive strength. Different combinations of macro and micro fibers were cast to study the fresh properties. Suitable adjustments were also made in the volume of silica fume and superplasticizer, to achieve the required strength and workability. Subsequently, a series of final mixes, as listed in Table 3.2 were cast. It may be noted that the mixes N14~, N15~ and N16~ were cast at a later date for completeness of the test series, but somewhat lower matrix strengths were observed. Table 3.2: Details of Final Mixes Mix designation Cement (kg/m3) Sand (kg/m3) Gravel-14mm (kg/m3) Water (kg/m3) Silica fume (kg/m3) Super-plasticizer (cc/kg cement) Air Entrain. (cc/kg cement)* N1,N2, N4, N5, N13 435 643 1029 128 43.5 8 1.3 N3, N6, N7, N8, N9,N10, N11,N12, N14~, N15~, N16~, N17 435 643 1029 123 45 10 1.3 , Note: Dosage of air-entraining agent is based on the total cementitious material (i.e. addition of cement and silica fume) 25 Six cylinders and six beams were cast for all the final mixes. The fiber volume fraction in the mixes is given in Table 3.3. Fresh properties of the mixes are discussed in the next chapter. Table 3.3: Volume fraction of fibers used in the hybrid mixes Mix Type of mix Volume of various fiber types (%) Total Volume fraction SI PI P2 c Pi p 2 N l Plain - - - - - - 0 N2 Single fiber 0.5 - - - - - 0.5 N3 0.75 - - - - - 0.75 N4 1.0 - - - - - 1.0 N5 - 1.0 - - - - 1.0 N6 - - 1.0 - - - 1.0 N7 Double fiber 0.3 - - - 0.2 - 0.5 N8 0.3 - - - - 0.2 0.5 N9 0.5 - - - 0.25 - 0.75 N10 0.5 - - 0.25 0.75 N i l 0.75 - - - 0.25 - 1.0 N12 0.75 - - - - 0.25 1.0 N13 Triple fiber - 0.5 - 0.25 0.25 - 1.0 N14~ - 0.5 - 0.25 - 0.25 1.0 N15~ - 1.0 - 0.15 0.15 - 1.3 N16~ - 1.0 - 0.15 - 0.15 1.3 N17 - - 0.5 0.25 0.25 - 1.0 Legend: SI: Steel fiber (Novotex), PI: HPP macro fiber, 50mm, P2: Self Fibrillating Polymeric fibers, c: Carbon fiber (with sizing treatment), pi: Dow micro fiber (2 denier), p2: Stealth fiber 3d 26 Chapter 4- Fresh Properties of Mixes and Preparation of Specimen 4.1 Introduction From the point of view of placement and long term durability, fresh properties of FRC are of great importance. In the subsequent sections, the actual production of HyFRC is discussed and test results pertaining to the fresh state of the concrete including slump, VeBe time, air content and density are included. In hybrids the use of different kinds of fibers significantly affects the fresh properties of concrete. Fresh properties affect the extent of fiber dispersion in the matrix, thus affecting the micro-structure of the matrix and the bond of the fiber. Figure 4.1 shows a magnified view of a paste reinforced by polypropylene fibers [10]; note the fibrils surrounding the fiber. Mixes with different fresh properties can be expected to have different interfacial characteristics and the fresh properties such as density, air content, workability, etc., may affect the hardened properties such as compressive and tensile strengths, elastic modulus, etc. Figure 4.1: Polypropylene-fiber reinforced paste [Courtesy- Barzin Mobasher, Arizona State University] 4.2 Production of Mixes and Specimens The moisture content of the sand and gravel was determined on each day of mixing in order to determine the corrected weight of wet sand, wet gravel, and water to be used. 27 Samples of sand and aggregates were dried over a flame and the amounts of sand, gravel and water were weighed after correcting for moisture. All the constituent materials were mixed in a pan mixer (Figure 4.2), which had a fixed blade and a pair of rotary blades mounted on a revolving spindle. The pan mixer operated at 12 rpm when empty and at 7 rpm when fully loaded. Figure 4.2: Mixing in Pan mixer Silica fume was added to the concrete mix in a slurry form (Figure 4.3) because it is believed that silica fiime disperses more effectively in this form and also because working with silica fume in the powder form could be a health hazard. Figure 4.3: Silica fume in a paste form In carbon fiber reinforced hybrid concrete, some additional precautions were needed. Carbon fibers are very brittle in nature, and if enough precautions are not taken, the fibers 28 may break down to shorter lengths. For these reasons, a pneumatic "omni mixer" (Figure 4.4) was used with an adjustable speed. Figure 4.4: Omni mixer For the mixes with Carbon fibers, sand, cement and part of the mixing water were first mixed with the superplasticizer in the omni mixer. Carbon fibers were then added to the mortar mix and mixed for a small period of time. This mortar mix was then added to the previously prepared dry mix of aggregates and macro fibers (if any) in the pan mixer. Mixing operation was then continued for about a minute. Figure 4.5: Magnified image of carbon fiber reinforced cement-based composite [Courtesy- Barzin Mobasher, Arizona State University] 29 During the mixing process, it was observed that the carbon fibers worked well with the polypropylene 2 denier micro fibers (pi) and seemed to disperse well in the mix, although it was noticed that the carbon fiber hybrids had a greater water demand. Figure 4.5 shows a magnified image of carbon fiber reinforced cement-based composite after hardening [10]. Although the above mentioned process for mixing of carbon fibers is appropriate in the lab, it is not very practical in the field. Thus, certain modifications in the fiber itself, the mixing method, the mixing sequence and so on are worth investigating. Figure 4.6 depicts the process of finishing concrete on the vibratory table that was employed for all mixes. Good workability and cohesiveness were observed in the hybrid mixes with the recommended dosage (by the manufacturer) of superplasticizer and air entraining agent. Figure 4.7 shows the finished beam specimens being transported on a cart. As stated earlier, good workability was observed in the mixes, which led to good finish and compaction of the hybrid mixes. Figure 4.6: Beam specimens- HyFRC in fresh state on vibrating table 30 Figure 4.7: Beam specimens- HyFRC in finished form Six concrete cylinders and six concrete beams were cast in accordance with A S T M C 192 [51]. The cylinder moulds used were 100 mm in diameter and 200 mm in height. The beam moulds used were 100 mm each in width and in height, and 350 mm in length (Figures 4.6 and 4.7). To prevent evaporation of water, the specimens were covered with a sheet of impervious plastic immediately after finishing. 4.3 Results Sampling of the freshly mixed concrete must be done in accordance with A S T M standards as outlined in A S T M C 172 [52]. The elapsed time between obtaining the first and final portions of the concrete sample must be no greater than 15 minutes. One needs to start the tests for both air content and slump within 5 minutes after obtaining the final portion of the concrete sample. Moulding of specimens for strength tests must begin within 15 minutes after the start of mixing and the samples need to be protected from any sources of rapid evaporation or contamination. Results of the more important fresh properties of concrete, such as the slump, VeBe time (for FRC), density and air content have been discussed in the following sections. 4.3.1 Slump and VeBe Time The workability of the hybrid mixes was determined by a VeBe test which also gives a slump value which is generally low for FRC mixes. In Table 4.1, the Slump, VeBe time, density and air content values are listed for all mixes. By measuring these fresh 31 properties, good control could be exercised on the material. Although an attempt was made to record the values for all mixes, due to certain unavoidable reasons, the values for some could not be recorded. Notice that many mixes had zero slump, which is expected from a slump test for high strength FRC with low workability. When the various mixes were compared, it was observed that the slump varied between 0 mm and collapse, with the VeBe time varying between 0 sec and 10 sec. As can be observed in Table 4.1, all the double and triple fiber hybrids with the exception of Mix N7, had zero slump and longer VeBe times (7-10 sec). Table 4.1: Fresh Concrete Properties Series Slump VeBe Time Density Air Content Value (mm) (sec) (kg/m3) (%) Nl * * * * N2 50 5 * * N3 Collapse 0 * * N4 0 10 * 8.8 N5 0 10 * * N6 50 6 * * N7 50 6 * 5.9 N8 * 5 2508 * N9 0 7 2298 7.5 N10 0 7 2460 5.0 N i l 0 8.5 2337 8.2 N12 0 7.5 2465 7.0 N13 0 10 * 7.8 N17 0 10 2377 7.8 * Unable to record the value Except for the Mixes N4 and N5, all other single fiber mixes had higher slump (collapse or 50 mm) and low VeBe times (0-6 sec). One benefit of using the VeBe time apparatus is that the slump value for concrete is obtained in addition to the VeBe time. A slump test on the fresh concrete is outlined in A S T M C 143 [53]. The dimensions of the conical mold and of the tamping rod (same as for air test) must be in accordance with A S T M guidelines. 32 Figure 4.8: Slump measurement in a VeBe mould The VeBe test for workability was performed immediately after the slump test was completed. Details of the VeBe apparatus, the process of measurement of slump and VeBe time are depicted in Figures 4 .8 and 4 .9 . 33 4.3.2 A i r Content The air content in HyFRC generally increases with an increase in the amount of air entraining agent and also depends on the use of fibers, the amount of superplasticizer used, and the amount of admixtures present. Air content is an important test parameter which affects the workability and the density. There are both positive and negative effects associated with the increase of air content in concrete. Increasing the air content can cause desirable effects such as improving the workability of fresh concrete and increasing the thermal resistance of hardened concrete. Especially in regions with low temperatures, a minimum amount of air content in concrete is often required so that a proper freeze-thaw resistance can be achieved. However, high air content can also cause undesirable effects such as lower density and reduced strengths. According to the Canadian Portland Cement Association [54], "air voids, if present in concrete, relieve the pressure caused by frozen water in concrete. Therefore air-entrainment producing 5% - 8 % air is strongly recommended for concretes subjected to severe exposure conditions". Apart from the dosage of the air entraining agent, two other factors controlling the air content are water/cement ratio and the degree of hydration. In this research, air content was measured using an air meter (Press-ur-meter air meter), manufactured by E L E International Ltd. The configuration of the air meter bowl and head conformed to the A S T M C 231 [55] standards. A constant dosage (1.3 ml/kg of cementitious material) of air entraining agent was added to all mixes. As the air content increased in the HyFRC mixes, a decrease in the compressive strength was seen. The results showed that at 2.7% air, the strength was about 98 MPa, and at 9.0% air, the strength decreased to about 55 MPa. For the above mentioned range of air content, the trend was estimated to be approximately linear implying that the strength decreased in proportion with the increase in air content. Flexural strengths also decreased with an increase in the air content. It can be observed from Table 4.1 that the air content of the concrete varied between 5% (for Mix N10) and 8.8% (for Mix N4). A strong correlation was observed between the air content and workability of concrete. The air content was 34 higher for the mixes which had higher workability, with the exception of Mix N10, which had a low value of 5%. 4.3.3 Density The density of the HyFRC mixes was found to be more or less the same, with a slight variation possibly due to the change in the air content and the workability of the mix which affects the compaction and, in turn the density of the mix. Figure 4.10: Calibration bowl Figure 4.10 shows the calibration bowl that was used to measure the density of HyFRC in the fresh state. It was observed that the air content had a prominent effect on the density of HyFRC, which decreased with the increase in the air content. In this research, it was noticed that the fibers entrapped air and had a substantial effect on the density of concrete. Density and corresponding air content when plotted showed an almost linear variation (Figure 4.11). In this investigation, fresh density of all the mixes could not be recorded, due to which a relationship of concrete density with slump could not be established. It can however be observed in Table 4.1 that the density varied from a maximum value of 2508 kg/m3 to a minimum of 2298 kg/m3. Even though the measured density values are not very accurate, consistent results were observed. 35 36 Chapter 5- Results and Discussion 5.1 Introduction In this chapter, hardened properties of various fiber reinforced composites are described. To characterize their flexural behavior, load-deflection curves were obtained and analyzed. 5.2 Test Specimens The test specimens were prepared according to A S T M C 192 [51]. Compression testing was done in accordance to A S T M C 39 [56]. The specimens were removed from the moulds 24 hours after casting, and moist cured at a temperature of 23°±1.7° C. The specimen ends were ground using a "Hi-kenma Beam Grinder" which was operated at approximately 10 rpm. For all compression testing, a "Baldwin Compression Machine" (Figure 5.1), with a Tate-Emery load indicator and 1750 kN maximum capacity was used. Figure 5.1: Baldwin machine for compression tests 37 A digital display was used to read the exact value of the compressive load. For flexural testing, an Instron machine (Model 4420) was used, which had a reversible load cell, and a 150 kN maximum capacity. These tests are described in details later. 5.3 Results 5.3.1 Compressive strength In this investigation, six cylinders per mix were tested in compression according to A S T M C 39 [56]; the average compressive strength for all the mixes combined was approximately 85 MPa. As expected, not much change was observed in the compressive strength with addition of fibers. This is in accordance with previously published data [57-62], where the strength gain was between 0 and 25%. Even for members that have conventional reinforcement in addition to the steel fibers, the fibers have little effect on the compressive strength [63, 64]. When concrete is loaded under compression, several distinct stages can be defined. During the first stage, micro cracks develop (due to the heterogeneity) randomly throughout the specimen primarily in the direction of the stress applied on the specimen. Since the hardened cement paste and aggregates have different stiffness, local tensile stresses develop and further micro cracking occurs. Figure 5.2 shows some of the tested Figure 5.2 Failed cylinder specimens 38 HyFRC specimens and Figure 5.3 shows a plain concrete specimen with a conical failure. Next, these micro cracks join together forming macro-cracks that run parallel to the direction of applied stress. In this stage, the micro cracks effectively divide the specimen into many slender columns in the direction of loading. Finally, oblique cracks appear and join together to form an inclined plane and the cylinder fails. In Table 5.1, the 28 day compressive strengths are tabulated. It can be noticed that there was some variation in the compressive strength of the mixes, ranging from a maximum of 102.4 MPa to a minimum of 70.0 MPa. With the exception of Mix N8, all mixes with macro polypropylene fibers had lower compressive strengths. As mentioned in Chapter 3, the mixes N14~, N15~ and N16~ were cast at a later stage and they resulted in lower matrix strength. 5.3.2 Modulus of Rupture (MOR) When a concrete specimen is subjected to bending, the tensile behavior will govern its flexural strength because concrete is weak in tension. In the first stage, micro cracks become randomly distributed throughout the specimen and eventually, they begin to join Figure 5.3: Cone failure for the cylinder 39 together to form macro-cracks, cracking being localized. The final stage begins, where the newly formed macro-cracks start to propagate and then quickly lead to unstable propagation and failure. Table 5.1: Compressive strength results Mix Designation Compressive strength-28 day (MPa) Nl 80.8 N2 97.4 N3 92.6 N4 86.6 N5 84.3 N6 77.3 N7 102.4 N8 76.3 N9 83.2 N10 82.3 Nil 97.0 N12 94.5 N13 77.3 N14~ 77.0 N15~ 71.0 N16~ 70.0 N17 86.3 One of the defining points on the load-deflection curve of FRC under flexure is the point of first crack, which corresponds to the modulus of rupture. This is the point at which the load-deflection curve first becomes nonlinear [19]. Fiber reinforcement is effective only after the point of first-crack [65]. In this research project, even though no definite increase in the compressive strength was noted with the addition of micro fibers, there was a clear increase in the MOR of the material under four point bending. In Table 5.2, hybrid mixes are compared with their corresponding controls. For the mixes containing steel macro fibers, the comparison is between Mix N9 and N2, and Mix N l 1 and N3. 40 Among the mixes containing HPP macro fibers, Mixes N13 and N15 are compared with N5. For Mixes N9 and N2 an increase of about 8.8% can be observed in the value of MOR. When Mixes N13 and N15 are compared with its control N5, increases of about 8.5% and 9.9%, respectively, can be observed. Table 5.2: Modulus of rupture of the different mixes Mix Designation N2 (Sl-0.5%) N3 (Sl-0.75%) N5 (Pl-1.0%) N9 (Sl-0.5% + pl-0.25%) N i l (Sl-0.75% + pl-0.25%) N13 (Pl-0.5% + c-0.25% + pl-0.25%) N15~ (PI-1.0% + c-0.15% + p2-0.15%) Modulus of Rupture (MPa) 7.71 8.42 6.48 8.39 10.17 7.03 7.12 5.3.3 Four Point Flexure Toughness Test Flexural tests are the most important tests for FRC. These generate load vs. deflection curves of the material in flexure which will in turn indicate the efficiency of the fibers as reinforcement. Figure 5.4: Tested beam specimen-fibers bridging the crack 41 Figure 5.4 shows a beam specimen, with fibers bridging a crack. In many instances, an increase in the toughness is more desirable than an increase in the flexural strength of the material. Apart from the flexural tests, several other tests (such as toughness in compression, tension, etc.) may be conducted for determining the toughness of FRC and these toughness values are influenced not only by the type of fiber, but also by the matrix quality. In this investigation, all beam specimens were tested to determine the flexural toughness as per A S T M C 1018 [19]. Figure 5.5 shows a test being conducted for a beam specimen with the necessary accessories to record the deflection values. Figure 5.6 shows the entire set up with the data acquisition system. Figure 5.5: Arrangement of the beam testing For all specimens, load vs. displacement plots were obtained. An open loop Instron machine was used. The rate of deflection used was about 0.1 mm per minute. Data were recorded to a total deflection of about 2.5 mm. Load and beam deflections were monitored continuously at frequent intervals (to ensure accurate reproduction of the load-deflection curve) using a software called "Quick Log". 42 Figure 5.6: Complete set-up for flexural toughness test The software records readings from two "Linear Variable Differential Transformers" (LVDTs) and also the load from machine. The data recording was set to a frequency of 1 hertz, which gave a sufficient number of points on the load-deflection curve (Figure 5.7). Channels for acquiring data Panel to control the Instron Electronics for data recording Figure 5.7: Data acquisition system Under four-point bending, deflections are overestimated because of crushing at the supports and load points. Even the loading frame can contribute to greater deflections [2]. To eliminate these spurious displacements, a yoke was used. 43 Figure 5.8: Yoke and other accessories A yoke measures deflections at the neutral axis of the beams. Before a test, specimens were turned 90 degrees on their side, with respect to the position as cast, to avoid testing the beams with variation in depth. In Figure 5.8, the arrangement with the yoke and other accessories that were used to test the beams are seen. The instability during testing occurs due to sudden release of the energy after a specimen reaches the peak load. In FRC, depending upon the number and type of fibers, the fibers across a crack are able to absorb the release of energy to some extent. Other factors controlling instability are the ultimate load carried and the distribution of fibers across a crack. It was observed that for higher peak loads instability was higher, thus affecting the post crack response. The most general way of representing load vs. deflection plots is with the instability region included, but many believe that the region of instability after the peak loads should not be considered. 44 Peak load Point C pulled back Figure 5.9: Representative load vs. deflection curve- open loop In this investigation, the triangular region of instability (ABC in Figure 5.9) was not taken into consideration and the curve was pulled back from point C to B. L o a d Deflection —> Figure 5.10: Representative load vs. deflection curve- closed loop In the following graphs (Figures 5.11-5.27), similar scales have been used for the load axis to enable easier comparison between the various mixes. An average curve for all six specimens has been plotted. The upper bound, lower bound and other individual specimens that stand out in some respect have also been identified. 45 Load-deflection plots for plain concrete are given in Figure 5.11. As seen, the curves are quite consistent. Note also the brittleness of the plain concrete. Load-deflection curve for N1 (Plain concrete) Average -Spec imen 2 0 10 20 30 40 50 60 70 80 90 100 Deflection x 0.001 (mm) Figure 5.11: Load vs. deflection curve-plain concrete Figure 5.12 shows the response of a mix with 0.5% steel macro fibers. Specimen 4 showed a high level of toughness, with the load carrying capacity (after the peak) of about 10 kN greater than the average. The lower bound curve was for Specimen 1. Notice the softening after peak load; this kind of response is expected when concrete is reinforced with steel macro fibers. A small drop in the curve represents fiber fracture or a sudden pull-out. 46 Load-deflection for N2 (0.5% S1) 0.6 0.8 1 1.2 Deflection(mm) 1.4 1.6 18 Figure 5.12: Load vs. deflection curve-single fiber (steel) concrete 40 Load-deflection curve for N3 (0.75% S1) 35 , Specimen 6 Specimen 2 Specimen 1 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Deflection (mm) Figure 5.13: Load vs. deflection curve-single fiber (steel) concrete 47 With an increase in the volume fraction of steel macro fiber from 0.5% to 0.75%, the average load carrying capacity (after peak) of the specimens increased from 15 kN (Figure 5.12) to about 22 kN (Figure 5.13). For Specimens 2 and 6 (Figure 5.13), the load carrying capacity after the peak was quite high, almost equal to the load supported at peak. As is clear from Figure 5.14, a further increase in the volume fraction of the steel fiber to 1.0% resulted in load carrying capacities exceeding the peak load at large deflections (Specimen 3 and 4). It can also be observed that the peak load of the specimens from Mix N4 (Fig. 5.14) is lower than that for Mixes N2 and N3 (Figs. 5.12-5.13). The reason for this could be our inability to properly compact this mix compared to Mixes N2 and N3 (See Table 4.1, Chapter 4). Load-deflection curve for N4 (1.0% S1) 40 35 30 : i — i 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection (mm) Figure 5.14: Load vs. deflection curve-single fiber (steel) concrete The response of Mix N5 is shown in Figure 5.15. It can be seen that the peak load is lower than that for Mixes N2 and N3. This is because, the workability of this mix (VeBe 48 time = 10 sec) was inferior to Mixes N2 and N3, which had VeBe times of 5 and 0 seconds, respectively. Also, the compressive strength of this mix was lower than that for Mixes N2 and N3. As expected, for mixes with polypropylene macro fibers, the load carrying capacity was much lower than for mixes containing steel macro fibers. Specimen 3 and 5 were the upper and lower bound responses for this mix. For Specimen 3, the response could not be recorded beyond a net deflection of about 1.8 mm. Load-deflection curve for N5 (1.0% P1) 40 -I —— 1 35 — 30 25 O J , , , , , 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.15: Load vs. deflection curve-single fiber (polypropylene) concrete Figure 5.16 depicts the response of a Mix N6, a matrix reinforced with self fibrillating polypropylene fibers. The response of this material under flexural loading seemed to be the same as HPP fibers (Mix N5). Note also that this mix had a larger variation in the response as compared to Mix N5. 49 Load-deflection for N6 (1.0% P2) The response of Mix N7 is depicted in Figure 5.17, with Specimen 1 and Specimen 2 being the extremes. The mix is a double fiber hybrid with steel macro fiber and Dow 2 denier polypropylene fiber. Mix N8 was similar, with steel macro fibers, but had Stealth 3 denier polypropylene fiber; its response (Figure 5.18) is similar to that of N7. Load-deflection curve for N8 (0.3% S1+0.2% p2) 40 -i Deflection (mm) Figure 5.18: Load vs. deflection curve-double fiber (steel + polypropylene) concrete Mixes N9 and N10 are similar to Mixes N7 and N8 respectively, with only the total volume fraction being increased from 0.5 to 0.75%. All the hybrid composites are compared to their respective control mixes at a latter stage. For Mix N9, notice a very small variation between individual specimens (Figure 5.19). 51 Load-deflection for N9 (0.5% S1 +0.25% p1) Load-deflection curve for N10 (0.5% S1+0.25% p2) -N10-1 -N10-2 N10-3 - N10-4 -N10-5 -N10-6 •Average Figure 5.20: Load vs. deflection curve- double fiber (steel + polypropylene) concrete 52 The content of steel macro fibers was further increased to 0.75% for Mixes N i l and N12 and the load-deflection plots are given in Figures 5.21 and 5.22. It can be noticed that at 0.75% steel fiber by volume, the amount of instability after the peak reduced drastically. However, no pseudo strain hardening was observed in either of these mixes. For Mix N12, one test data (for Specimen 4) was lost and the average in Figure 5.22 is for five specimens instead of six. Figure 5.21: Load vs. deflection curve- double fiber (steel + polypropylene) concrete Mix N13 (Figure 5.23) was one of the most promising mixes. The average curve plotted is from five specimens. 53 Load-deflection curve for N12 (0.75% S1+0.25% p2) 20 i ii ^Specimen 3 1 / Specimen 1 h / Average / —. , . , r ^ r ^ - ^ — . « m — i ~ > — • • -^y< ' • * — \ - X - t — C — \ ~ — " \ \ — • — - ^ ^ ^ r Specimen 2 Specimen 6 Specimen 4 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.22: Load vs. deflection curve- double fiber (steel + polypropylene) concrete Load-deflection curve for N13 (0.5% P1+0.25% c+0.25% p1) 0 02 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.23: Load vs. deflection curve-triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 54 The curves for N14~, N15~ and N16~ in Figures 5.24-5.26 are averages of four test specimens. These mixes were cast at a later stage. Although the mix design of these mixes was the same as for the previous mixes, the matrix strength was found to be lower than the others. These Mixes (N14~, N15~ and N16~) were cast for completeness, but the data were not critically analyzed because their strengths, for reasons that remain unclear, were low. In Mix N14~, the Dow 2 denier micro fiber was replaced with Stealth 3 denier fiber. Mix N14~ showed poor results when compared to Mix N13, which had Dow 2d fiber, the other reason being that the matrix strength of Mix N14~ was somewhat lower than that of MixN13. Load-deflection curve for N14~ (0.5% P1+0.25% c+0.25% p2) 40 i 1 35 30 Deflection (mm) Figure 5.24: Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 55 Load-deflection curve for N15~ (1.0% P1+0.15% c+0.15% p1) Mixes N15~ and N16~ had the same fibers as Mixes N13 and N14~ respectively, with the total volume of fibers increasing from 1.0% to 1.3%. Large improvements in the flexural response were observed in Mix N16~, when compared with Mix N14~. Figure 5.27 shows the toughness curve for Mix N17 which is a triple fiber hybrid similar to Mix N13, with the type of macro fiber being the only variation. It is evident that the response of Mix N17 is quite poor when compared with Mix N13, thus indicating that the HPP fibers have better performance than the self fibrillating fibers. 56 Load-deflection curve for N16~ (1.0% P1+0.15% c+0.15% p2) [I Specimen 3 Average Specimen 4 'I / - — , ^ — - — 7 — -i ^ s — ^ — — Specimen 2 _ > Specimen 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection (mm) Figure 5.26: Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) Load-deflection curve for N17 (0.5% P2+0.25% C+0.25% p1) A Jj 1 Specimen 2 / Specimen 5 / Specimen 1 [ ——f-~ ' ^ _ _ _ _ _ _ y Average, \ " ^ S p e c i m e n 6 * \ Specimen 4 Specimen 3 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.27: Load vs. deflection curve- triple fiber (polypropylene macro + carbon + polypropylene micro concrete) 57 Figure 5.23 shows the load vs. deflection response of the triple fiber hybrids with HPP macro fiber alone. Although the volume fraction of the macro fibers is reduced to half in the triple fiber hybrid N13, the response is better as compared to the single fiber Mix N5 with 1.0% HPP fibers. This means that by the addition of the micro fibers, not only has the load carrying capacity increased, but the secondary benefits of having microfibers in a mix are also available. Comparison of Load-Deflection Plots As seen previously (Figure 5.11), plain concrete specimens (Mix Nl) failed right at the peak load with no post-crack response, as opposed to the fiber reinforced mixes. Comparison of Load-deflection curves for N2, N3 and N4 (Steel-single fiber concrete) 40 35 30 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.28: Comparison of mixes with different Vf of steel macro fibers 58 Comparison of Load-deflection curves for N2, N7 and N9 (Steel macro fiber & its hybrids) 40 35 30 0-1 , r- , —, , , , , , — I 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.29: Comparison of mixes with steel macro fibers and its double fiber (pi) hybrids Figure 5.28 is a comparison of mixes having different volume fractions of steel macro fiber. When the volume fraction of the fibers is increased from 0.5% to 0.75%, the load carrying capacity at the peak and after the peak is greater. However, when the Vf is further increased to 1.0%, the performance becomes poorer, apparently as a result of the poor workability observed for Mix N4. Thus, there exists a critical volume fraction of the fibers beyond which the performance of the material starts worsening rather than improving, due to difficulties in compaction. Figure 5.29 shows a comparison of mixes with steel macro fibers and its double fiber hybrids (Dow 2 d polypropylene micro fibers were used). Even though Mix N7 had the same total volume fraction as Mix N2, it resulted in a poorer load carrying capacity throughout the deflection range. 59 Mixes which contain micro fibers only can be assumed to have little flexural toughness, thus when the total Vf of Mix N9 is compared to that of Mix N2, clear synergy is observed. The shaded region in Figure 5.29 represents the synergy. Figure 5.30 shows the comparison of mixes with steel macro fiber by itself and its double fiber hybrid with micro-polypropylene 3d fibers (p2). Note minor synergy at large deflections. Comparison of Load-deflection curves for N2, N8 and N10 (Steel macro f ibers & its hybrids) o-l , , , , , , , , , 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection (mm) Figure 5.30: Comparison of mixes with steel macro fibers and its double fiber (p2) hybrids 60 Figure 5.31 was plotted to show the difference in the performance of the two types of polypropylene fibers used in this investigation. Mix N3 is the control mix and clear synergy was observed when mix containing Dow 2 denier fibers was used, at least for beam deflections smaller than 1.4 mm. It is implicitly assumed that pi or p2 fibers will not contribute much to toughness by themselves. No synergy was observed for Mix N12. Comparison of Load-deflection curves for N3, N11 & N12 (Steel macro fiber and its hybrids) 5 o 4 1 , , , , , , 1 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection (mm) Figure 5.31: Comparison of mixes with steel macro fibers and its double fiber (pi & p2) hybrids In Figure 5.32, mixes with steel macro fibers and its double fiber hybrids are compared. This plot helps compare the performance of the two types of micro polypropylene fibers. Except at very small deflections, Mix N8 (with p2 micro-fiber) carried higher loads than Mix N7 which carried pi micro-fibers. A similar trend was observed with the Mixes N9-N10 and N11-N12. pi micro-fiber was more effective than p2 micro-fiber in all the mixes, except when compared with Mixes N7 and N8. 61 Comparison of Load-deflection curves for N7, N8, N9, N10, N11 & N12 (Double f iber hybrids of steel macro & poly micro) Figure 5.32: Comparison of mixes with steel macro fibers and its double fiber hybrids Comparison of Load-deflection curves for N4, N5 & N6 (Single f iber mixes) 0.4 0.6 0.8 1 1.2 Deflection (mm) 1.4 1.6 1.8 Figure 5.33: Comparison of mixes with steel and polypropylene macro fibers 62 Figure 5.33 is a comparison of mixes containing steel and polypropylene macro-fibers. All three mixes (N4 , N 5 and N 6 ) had a volume fraction of 1.0%. Mix N 6 , containing P2 fibers (self fibrillating polypropylene macro-fibers), resulted in a higher load carrying capacity than Mix N 5 with PI fibers (HPP). Also, note that the loads carried by Mixes N 5 and N 6 showed an increasing trend with an increase in deflection. While at other volume fractions the load carried by the steel fiber reinforced concrete was reduced with an increase in deflection, this was not true for Mix N 4 with 1 % steel fiber. Figure 5.34 shows a comparison of mixes with polypropylene (HPP) macro fibers and its triple hybrids (containing polypropylene and carbon micro fibers). The triple fiber hybrids (Mix N 1 3 and N 1 5 ) performed better when when they were compared with the control Mix N 5 . Here, Mix N 1 3 resulted in a high amount of synergy. In Mix N 1 5 , V f of the macro fiber (HPP) was increased and V f of the micro fibers (polypropylene and carbon) was reduced. In this mix, the total volume fraction of the fibers was 1.3%, with 0.3% combined V f of micro fibers, which resulted in an increase in the toughness. Comparison of Load-deflection curves for N5, N13 & N15 (Polypropylene macro f iber & its hybrids) 40 t 1 35 30 25 Figure 5.35 compares the mix with self-fibrillating polypropylene macro fibers (Mix N6) and its triple hybrid (Mix N17). The hybrid produced a higher peak, but had lower load carrying capacity beyond the peak. One reason for the poor performance of the hybrid mix could be its poor workability (a long VeBe time of 10 seconds). Comparison for Load-deflection curves for N6 & N17 (Polypropylene macro f iber and its hybrid) 40 -I 1 35 30 25 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Deflection(mm) Figure 5.35: Comparison of mixes with polypropylene macro fibers and its triple fiber hybrid Figure 5.36 is a comparison of different triple hybrids that contained polypropylene (HPP) macro fibers. While the triple hybrids (Mix N13 and Mix N15) contained pi type polypropylene micro fibers, Mixes N14 and N16 contained p2 fibers. Mixes N13 and N15 showed better performance as compared to Mixes N14 and N16 and thus, it is clear that the pi fiber is more effective than the p2 fiber. 64 Comparison of Load-deflection curves for N13, N14-, N15- & N16- (Triple f iber hybrids wi th Polypropylene macro f ibers) 40 35 5" 20 —N13-Average(0.5%P1+0.25%c+0.25%p1) N14-Average(0.5%P1+0.25%c+0.25%p2) N15~Average(1,0%P1 +0.15%c+0.15%p1) N16-Average( 1 0%P1 +0.15%c+0.15%p2) Figure 5.36: Comparison of triple fiber hybrids with polypropylene (HPP) macro fibers Analysis of Flexural Toughness In this section, a comparative study of the values of the toughness indices obtained by the A S T M C 1018 method [19] which is based on first crack determination, and flexural toughness factors as per the JSCE-SF4 Method for Flexural Toughness [20] is carried out. In the A S T M C 1018 method, the toughness indices (I5, I10, etc.) are calculated by dividing the area under the load-deflection curve up to a specified deflection, by the area up to the first crack. From these indices, residual strength factors, that represent the post-crack load retained over a specific deflection interval (as a percentage of the peak load), are calculated. The first-crack load is used to determine the first crack flexural strength, and the first crack deflection value is used to establish the end point deflections for 65 toughness calculations. In this method, determining the point of first crack is arbitrary and subject to human error. Initially, the load-deflection curve is never perfectly linear, and the non-linearity thus begins at quite an early stage with matrix micro cracking. In the JSCE-SF4 [20] method, the apparatus and process used for determining the load vs. deflection plot are the same as in A S T M C 1018, but in this method, the flexural toughness is expressed by a 'flexural toughness factor' (at,*). The flexural toughness (Tb) is determined from the load-deflection curve as the area under the load-deflection curve to a deflection of span/150. The "flexural toughness factor" is determined by the following equation. ab*=(TD*l)/(8tb bh2) where, 8*: deflection of 1/150 of the span i.e. 2 mm when the span is 300 mm. More recently, the Post Crack Strength Method [21] has been proposed. The rationale behind this approach is that the ascending part of the load deflection curve is non-linear much before the peak load and the location of first crack is subject to human error. The technique proposes to locate the peak load and divide the curve into two regions: the pre-peak region and post-peak region. The area under the curve is then calculated up to the peak load and termed as pre-peak energy, i.e., E p r e . In the post-peak region, points are located corresponding to deflections L/m, where'm' could have different values ranging from 3000 to 150. ' L ' is the span of the beam. The area under the curve up to a deflection of L/m is termed as the total energy i.e. E totai,m- The pre-peak energy is subtracted from this total energy to obtain the post-peak values i.e. E p o s t ) m corresponding to a deflection of L/m. The post-crack strength PCS m at a deflection of L/m is given as: P C S m = (Eport.nO xL/((L/m)-5 P eak) x b x h 2 In this investigation, plotting PCS graphs helped quantify synergy at various deflections. 66 Test Results The toughness response of different mixes can be analyzed and compared in two ways. In the first approach, it can be assumed that the micro fibers do not contribute to toughness, which is often seen. In the second approach mixes with the same total volume fraction of fiber are compared. Table 5.3: Toughness Indices and Residual Strength Factors Specimen Design. Toughness Indices Residual Strength Factors A v e r a g e Is A v e r a g e ho A v e r a g e I20 A v e r a g e R5.10 A v e r a g e R10.20 N2 4.5 7.9 14.9 68.3 69.9 N3 3.5 6.8 13.7 65.8 68.5 N4 2.6 4.0 6.7 28.1 27.7 N5 2.3 3.0 4.5 14.3 14.8 N6 2.9 4.4 7.4 29.1 29.7 N7 3.3 5.2 8.6 37.5 34.0 N8 2.9 4.3 7.1 27.9 28.3 N9 2.9 4.7 8.3 35.8 36.2 N10 2.7 4.3 7.4 30.5 44 .6 N i l 3.5 6.4 12.0 57.3 56.8 N12 3.7 6.2 11.0 49 .9 48 .0 N13 2.6 3.7 6.1 21.9 23.9 N17 2.7 3.7 5.7 20.6 19.8 In this investigation, toughness indices (I5, I10 and I20) and the residual strength factors (Rsjo and Ri0,20), were calculated for all the specimens according to A S T M C 1018 (values included in Appendix A). The average toughness indices and the residual strength factors for all mixes are shown in Table 5.3. These values are further sorted in the order of their performance and listed in Table 5.4, where the Mixes N2, N3 show higher and Mixes N4, N8 show lower than expected residual strength values. 67 Table 5.4: Mixes in order of their performance Specimen Designat ion F i b e r T y p e a n d V f (%) Residual St rength Average R5.10 Average R10.20 N2 S1-0.5 68.3 69.9 N3 S1-0.75 65.8 68.5 N11 S1-0.75+p1-0.25 57.3 56.8 N12 S1-0.75+p2-0.25 49.9 48.0 N10 S1-0.5+p2-0.25 30.5 44.6 N9 S1-0.5+p1-0.25 35.8 36.2 N7 S1-0.3+p1-0.2 37.5 34.0 N6 P2-1.0 29.1 29.7 N4 S1-1.0 28.1 27.7 N8 S1-0.3+p2-0.2 27.9 28.3 N13 P1-0.5+C-0.25+p1-0.25 21.9 23.9 N17 P2-0.5+C-0.25+p1-0.25 20.6 19.8 N5 P1-0.5 14.3 14.8 Table 5.5 shows the average Modulus of Rupture (MOR) and Post Crack Strength (PCS) values for all mixes at nine different values of beam deflection. These values are plotted in Figures 5.37 and 5.38. PCS values for all beam specimens are included in Appendix B. 68 Table 5.5: Average MOR and PCS values for the mixes Mix Des. MOR Average PCS (MPa) (MPa) PCS3000 PCS150o PCS10oo PCS750 PCS<joo PCS400 PCS300 PCS200 PCS150 Nl 7.3 - - - - - - - -N2 7.7 5.3 5.9 5.3 5.4 5.2 5.4 5.3 5.0 -N3 8.4 9.0 7.1 7.0 6.9 6.8 6.7 6.6 6.6 7.0 N4 6.4 7.2 5.7 5.7 5.6 5.6 5.5 5.5 5.6 5.6 N5 6.5 3.0 2.7 2.7 2.7 2.7 2.9 3.0 3.2 -N6 6.6 3.6 3.1 3.2 3.1 3.2 3.2 3.3 3.3 4.7 N7 8.3 5.4 4.6 4.2 4.1 4.0 3.9 3.8 3.6 3.0 N8 9.8 4.6 4.1 3.9 3.6 3.9 3.9 3.9 3.9 3.8 N9 8.4 5.8 5.1 5.0 5.0 5.0 5.0 5.0 5.1 4.9 N10 9.7 5.0 4.4 4.4 4.3 4.3 4.4 4.4 4.5 4.5 N i l 10.2 8.5 7.9 7.8 7.7 7.6 7.6 7.5 7.3 7.0 N12 7.9 6.8 6.2 5.8 5.6 5.6 5.5 5.5 5.5 5.4 N13 7.0 4.6 3.4 3.4 3.4 3.4 3.4 3.4 3.6 -N17 6.9 4.3 2.7 2.5 2.4 2.3 2.2 2.1 2.0 -Figure 5.37 shows the PCS values plotted against L/m ratios for single hybrid Mixes N2-N6, whereas Figure 5.38 shows the PCS values for the double and triple hybrid Mixes N7-N13 as well as Mix N17. The PCS values at lower L/m ratios (between 0 and 0.3) can help display the synergy (positive or negative) in the hybrids at low deflection values, which otherwise might be difficult to observe in the load vs. deflection (toughness) curves. For certain mixes, when the load vs. deflection responses are similar, the PCS plots help quantify the synergy if present. 69 Average PCS vs. L/m ratios Figure 5.37: PCS chart for the Mixes N2-N7 It is clear from Figure 5.37 that while the MOR for Mix N2 is higher than that of Mix N4, Mix N4 shows better performance at smaller L/m ratios (approximately between L/m of 0.25 and L/m of 1.75). As seen in Figure 5.38, while N l 1 is one of the stronger mixes throughout the entire deflection range, Mix N17 was a weaker mix for L/m values smaller than 1.5. 70 Average PCS vs. L/m ratios Synergy quantification In the following section, the synergy (in terms of the PCS) between the different mixes has been evaluated and compared as a percentage. As seen in Table 5.6, the PCS value of a hybrid to that of its control is compared. Note that some of these are negative, implying that the hybrid is performing worse than its control. The values presented in the table have been plotted in the form of graphs in Figures 5.39-5.41. In Table 5.6 and in Figures 5.42-5.44, the comparison is between mixes having the same volume fraction of fiber. Notice that in Figure 5.39, where the total volume fraction was compared, Mix N7 showed some positive synergy at L/m ratios less than 0.1 and negative synergy thereafter when compared with Mix N2. Similarly, when Mix N9 was compared with N2, there was positive synergy at small and large deflections and negative synergy elsewhere. 71 Table 5.6: Synergy Assessment (in percent) Deflect ion value=> Mixes compared U MOR 0.1 0.2 0.3 0.4 0.5 0.75 1.0 1.5 2.0 Double f iber hybr ids- steel macro and po lypropy lene micro f ibers (p1) N2-N7 6.5 1.3 -28.1 -26.6 -32.5 -30.1 -39.3 -39.1 -42.2 -N2-N9 8.8 8.4 -13.0 -5.7 -7.4 -3.8 -6.7 -5.3 0.9 _ N3-N11 20.7 -6.0 10.2 11.0 11.5 12.5 13.8 12.7 10.9 -0.2 Double f iber hybr ids- steel macro and po lypropy lene micro f ibers (p2) N3-N12 -6.9 -24.4 -13.4 -17.4 -18.2 -17.7 -17.1 -16.6 -16.7 -23.4 N2-N10 26.1 -7.3 -24.8 -18.2 -20.0 -16.8 -18.3 -16.8 -11.8 -36.6 N2-N8 20.9 -15.7 -45.4 -35.5 -49.7 -33.8 -37.6 -35.3 -29.2 -84.6 Macro pol) ^ propylene mixes and its t r ip le hybr ids N5-N13 12.8 -13.1 -17.9 -4.5 8.0 17.5 33.4 47.0 56.6 _ N5-N15 9.9 2.1 16.3 20.1 22.2 24.1 24.8 26.4 26.3 _ N6-N17 5.1 16.6 -11.6 -23.4 -25.5 -27.8 -30.1 -34.7 -40.1 -With a small addition of 0.25% of pi fibers, a large amount of synergy was observed for Mix N3 vis-a-vis Mix N i l . Little more than 10% positive synergy was evident throughout the deflection range. Figure 5.40 compares Mixes N2-N8, N2-N10 and N3-N12. Except for the modulus of rupture, no positive synergy was observed in any of these three mixes. 72 Synergy in percent associated with various hybrids , L/m Figure 5.40: Synergy vs. L/m ratio- mixes containing steel macro and polypropylene micro fibers (p2) 73 The comparison of the Mixes N6 and N17 (Figure 5.41) containing polypropylene macro (P2) and polypropylene micro fibers (pi), did not show any positive synergy. The results for Mixes N5, N13 and N15 show that except at small L/m ratios, significant positive synergy occurred in hybrids N13 and N15. Figure 5.41: Synergy vs. L/m ratio- mixes containing Polypropylene macro (PI & P2) and polypropylene micro fibers (pi) 74 Chapter 6- H y F R C : Plastic Shrinkage and Single Fiber Bond-Slip Response 6.1 Plastic Shrinkage 6.1.1 Introduction During setting, concrete shrinks and, if restrained, cracks under tensile stresses. Shrinkage crack resistance is one measure of the durability of cement-based materials. The usefulness of fiber reinforcement in improving the cracking resistance of cement-based materials under restrained shrinkage conditions is well documented. As discussed before, in HyFRC, by combining fibers, both the plastic shrinkage and drying shrinkage of the hardened concrete can be improved simultaneously. The presence of fibers is expected to influence the length of the resulting cracks as well as their widths. Here, the effectiveness of Hybrid fiber reinforcement in controlling plastic shrinkage cracking of plain mortar was examined using the UBC overlay method. 6.1.2 Test Set-up The technique involved placing a layer of mortar directly on a fully hardened substrate cured for at least 28 days. The overlay was then examined for development of cracks due to plastic shrinkage. Testing was carried out in a shrinkage chamber (Figure 6.1). The temperature in the chamber was maintained at approximately 50° C and the relative humidity level was kept at a low of 2-5%. The sizes of the openings at the two ends of the chamber were 240 mm x 175 mm. In the chamber, hot air was circulated by a fan at an approximate rate of 9.5 cubic meters per minute, using a heater with an output of about 16500 kilojoules. 75 6.1.3 Experimental Procedure The overlay mortar to be tested for plastic shrinkage was placed directly on a fully hardened substrate. The dimensions of the substrate bases were 40 x 100 x 325 mm. Each substrate base was reinforced with two 15 mm or one 30 mm diameter steel bars to enhance its rigidity. To make the substrate surface rough, after the fresh concrete was placed into the mould, aggregates were evenly placed on the surface. The bases were then removed from the moulds after 24 hours, covered with plastic sheet and then cured in water for 28 days. Fiber reinforced mortar (mix proportions in Table 6.1) with different fiber contents was prepared and poured on top of the hardened substrate bases in a mould measuring 60 x 100x375 mm. As observed earlier (in Chapter 5), Mix N13 demonstrated high synergy in terms of toughness. This mix contained a combination of three fibers with 0.5% by volume of macro HPP fibers and 0.25% by volume each of polypropylene 2 denier and carbon micro fiber. In order to investigate the crack development under plastic shrinkage for this combination of fibers, fiber properties were kept the same and different total fiber volumes (0.05%, 0.08% and 0.1%) were compared with plain mortar. A vibrating table 76 was used to compact the sample and the surface was leveled using a trowel. After placing mortar into the mould, it was transferred in the chamber and demoulded an hour later. It was left for an additional 24 hours in the chamber to enable the shrinkage cracks to form. Table 6.1: Mix Proportion of Ingredients of Overlay Mortar Ingredient Mass by volume (kg/m3) Cement 1200 Water 576 Sand 601 A continuous crack measurement is carried out for the stated 24 hours. A piece of string was used to measure the crack lengths while a microscope (with an accuracy of 0.1 mm) was used to measure the crack widths. For each crack in the specimen, the corresponding width and length were accurately measured and the crack area for the crack was found by multiplying these two values. 6.1.4 Results Figure 6.2 shows three specimens with the control mortar (no fibers). A large number of wide cracks was observed in these three beams. 77 Figure 6.2: Shrinkage beam specimens with control mix, no fibers However, when a fiber dosage rate of 0.1% was used, no visible cracks (cracks wider than 0.01mm) appeared on the overlay surface (Figure 6.3). Figure 6.3: Shrinkage beam specimens with V f = 0.1% The total crack areas for the control and the fiber reinforced specimens (Vf=0.05, Vf=0.08 and VpO.l) are plotted in Figure 6.4. Note a clear decrease in the crack area with an increase in the fiber volume fraction. 78 300.00 250.00 o 200.00 I -u >— 150.00 o H 100.00 343.5S Control 0.05 0.08 Percentage Volume • Trial 1 • Trial 2 0.1 Figure 6.4: Total crack area vs. volume fraction of fiber As the fiber volume fraction was increased, the maximum crack width and crack area were reduced compared to the crack area in plain concrete. Appendix C summarizes these results, where maximum crack widths, rnaximum crack areas, and the number of cracks are shown. 6.2 Single Fiber Bond-Slip Response 6.2.1 Introduction The building block of any fiber concrete response is the fiber-matrix bond which may be measured by conducting fiber-matrix pull-out tests. In this investigation, the fiber combination as used in Mix N13 (Chapter 5) was further studied via such pull-out tests. In particular, the bond-slip response of a single HPP macro fiber embedded in a high strength hybrid concrete matrix (using polypropylene 2d and carbon micro fibers) was investigated. In this investigation, a comparison between the load required for pulling a 79 polypropylene macro-fiber from a hybrid matrix and from a non-hybrid matrix was made. 6.2.2 Pullout tests- Set-up and Discussion Figure 6.5 shows the schematic of a pull-out specimen. Since, in FRC, the fibers are randomly oriented, the macro-fibers were oriented at different angles within the concrete specimens for pull-out. It was expected that the hybridization of a concrete sample with polypropylene and carbon micro-fibers would enhance the pull-out resistance of the polypropylene macro-fiber. Polypropylene fiber Plastic separator cementitious matrix Steel anchorage ring Figure 6.5: Schematic of pull-out specimen (dimensions in mm) Two batches of high strength concrete specimens were cast. The first concrete mix design contained no micro fibers and the other had two different kinds of micro fibers (Table 6.2). 80 Table 6.2: Details of the mixes used for the pull-out tests Concrete Mix Cement Type 10 kg/m3 Water kg/m3 Aggregate 10mm kg/m3 Sand kg/m3 Silica Fume kg/m3 Water-reducing agent, ml/kg (cement) Air-entraining agent, ml/kg (cement) Carbon Fiber (% volume) Micro Polymer Fiber (% volume) Plain matrix 435 138 1029 643 45 7-10 1.3 0 0 Hybrid Matrix 435 138 1029 643 45 7-10 1.3 0.25 0.25 The concrete mix was poured into a metallic specimen mould with the macro fiber clamped into position. Metallic rings were placed inside the moulds, perpendicular to the direction of loading to permit the application of the pullout load. Figure 6.6: Photograph of specimen within testing apparatus The second half of the testing cylinder was cast the following day using an identical mould atop the previously cast half, by using a plastic separator. The length of the fiber was equally distributed within the top and bottom of each specimen, resulting in an embedded length of approximately 25 mm. The discontinuity in the concrete provided by 81 the plastic separator provided a simulated concrete crack. The specimens were cured for a total of 28 days before testing. For both concrete mixes, 100 mm x 200 mm specimens were cast for compression testing. The matrix strength of the hybrid mixes was slightly higher than the other mixes. As stated previously, in order to simulate random fiber distribution in FRC, the macro fibers were oriented at different angles for pull-out. In all, four angles of 90°, 67.5°, 45° and 22.5° were investigated. The pullout tests were performed using a desk mounted Instron testing machine as shown in Figure 6.6. The metal rings formed inside each specimen allowed the machine to grip the concrete specimen. The tensile load was applied through these rings and the vertical crack opening displacement was measured by two LVDTs placed on either side of the specimen across the simulated crack. 6.2.3 Results and Discussion While Figure 6.7 shows a plot of peak pullout loads vs. the inclination angle '9', of the macro fiber, Figure 6.8 depicts the energy absorbed. From both Figures, it can be noticed that hybrid mixes carried essentially the same peak pull-out loads and absorbed the same pull-out energies as the control mixes. a> o. 0.0 -I , , ! 22 45 67 90 F i b e r I n c l i n a t i o n A n g l e , 6 Figure 6.7: Peak pullout loads vs. inclination angle (polypropylene fiber) 82 r 22 45 67 90 Fiber Inclination Angle, Degrees Figure 6.8: Energy absorption (up to 3 mm of slip) vs. inclination angle (polypropylene fiber) This trend is the same as that reported by Naaman and Najm [66], who observed that the peak pullout load for identical fibers being pulled out from matrices of different compressive strengths or other properties does not necessarily change. Their reasoning is that the frictional bond strength may be unrelated to the matrix properties. One major limitation of the test program performed here was that a fixed embedment length of 25 mm was investigated. In a real composite, fibers are embedded to lengths between 0 and 25 mm. Also, only a few inclination angles were assessed. That may explain why the pull-out tests failed to predict the improved flexural response of hybrid composites. 83 Chapter 7- Conclusions The primary objective of this research program was to develop high performance hybrid fiber reinforced composites with desirable fresh and hardened properties. The following conclusions were drawn: 7.1 Fresh properties • Workability The slump and the VeBe time of the mixes were measured as indicators of workability. VeBe time was quite useful for most of the mixes. For all double and triple fiber HyFRC mixes, a zero lump was measured, except for mix N7 (containing steel macro and polypropylene micro). These mixes also had higher VeBe times. Thus, for a fixed volume fraction of fibers, the use of micro fibers in the hybrid composites adversely affected the workability. This is expected because the micro fibers have a larger surface area for a given volume fraction than macro fibers. This increases their water demand. • Air Content A strong correlation between the air content and the workability of the hybrid composites was noted; workability was found to be higher for the mixes with higher air contents. Air content increased from a value of 2-3% when no air entraining agent was used to about 7-8% when the recommended dosage was used. It was also concluded that both compressive and flexural strengths decreased with an increase in the air content. • Density No strong relation could be established between the air content and the density of the mixes. The density of the hybrid mixes did not vary significantly when the volume fraction of fibers and the dosage of air entraining agent were changed. It was also 84 concluded that the density did not reduce with increase in air content because of improvement in workability with the increased air content. .2 Hardened properties • Compressive strength Some inconsistencies in the compressive strengths were noticed. With an addition of fibers or with a change in the type of fiber, no particular change in the compressive strength was noticed. Results show that the compressive strength is more dependant on the workability and air content of the mix than on the type and volume fraction of the fibers used. • Modulus of Rupture (MOR) Although the addition of micro fibers did not have any effect on the compressive strength of the mix, there was a clear increase in the MOR of the material under four point bending. Most of the hybrid composites when compared with their controls showed an increase in the value of the MOR by about 10%. • Flexural Toughness It was concluded that in flexural toughness tests, the instability at the peak load was a function of the matrix strength. This unstable release of energy was reduced when fibers were present. It was concluded that the post peak response of the hybrid composites was poor when there was the release of a high amount of energy at the peak. It is more appropriate to exclude the triangular portion of the instability (observed during analysis in the load vs. deflection curve) from the curve. This is, however, expected to give conservative values of toughness. For mixes containing macro steel fibers, the load vs. deflection response followed a softening trend after the peak load. For the mixes with polypropylene macro fibers, there was an increase in the load carrying capacity after the peak, as the cracks 85 become wider. This was expected because of the low modulus of the polypropylene that requires large strains to attain stresses approaching the strength. For steel macro fibers, it was concluded that 0.75% seemed to be the volume fraction at which the load carried by the beams at the peak as well as after the peak was maximum. The response got weaker at higher and lower volume fractions. Thus, 0.75% appears to be the optimal volume fraction as far as the flat ended steel-macro fiber is concerned. It was also concluded that the instability at the peak load reduced with higher amounts of fibers and some pseudo strain hardening was evident. It is believed that a high volume fraction of the fibers (about 1.0%) might be responsible for increasing the effort required for compaction which, in turn, adversely affected the strength and flexural response of the material. The PCS method of analysis was found to very useful for the hybrid composites because the hybrid composites demonstrated different levels of synergy at different deflection values, which could be captured well only by this method. This synergy associated with the hybrid composites (positive or negative) could not easily be identified by observing the load vs. deflection curves alone. When beams were analyzed using A S T M C 1018, the following conclusions were drawn: Fiber type p2 (Stealth 3d), does not contribute much towards toughness enhancement at low dosage rates (0.2 and 0.25%). Fiber type P2 (Dow-Self Fibrillating macro fiber), produced poor toughness at a high dosage of 1.0% because at such a dosage rate fiber fibrillation could not occur. The residual strength factors indicated that the mixes having steel fibers had much higher residual strengths than mixes containing polypropylene macro fibers. High amount of synergy was observed when carbon fibers were used in combination with the polypropylene micro fibers (especially with pi). Also, most of the hybrid composites demonstrated higher residual strength values as compared to their control mixes; this was commensurate with the PCS results. When beams were analyzed using the PCS method, the following conclusions were drawn: All double and triple fiber hybrid composites (except for N12, which 86 contained 0.75% SI and 0.25% p2) had a higher MOR compared with their control mixes (single-macro fiber mixes). For many mixes at L/m ratios of about 0.1 or less, the hybrid composites demonstrated better synergy. For the triple fiber mixes, the synergy that was observed at smaller deflections is possibly due to the contribution from carbon fibers (say, up to L/m ratio of 0.2). Then, there appears to be a transition zone, where the carbon and the polypropylene micro fiber work together and then at higher deflections, just the primary macro fibers contribute to the total toughness. Poor fiber dispersion and large variations in the flexural response were noted with the self-fibrillating macro polypropylene fiber. The responses of the hybrid composites based on the HPP fibers were much better than those based on self-fibrillating polypropylene macro-fibers. Micro polypropylene fibers showed synergy even at low dosages. It was seen that the Polypropylene 2 denier fibers worked better in a hybrid mix compared to the Stealth 3 denier fibers. It is suggested that polypropylene micro fibers enhance the matrix properties and improve the bond of the macro fibers with the matrix. If properly designed, even when the total volume fraction is compared, significant synergy is evident, as in the case of the mix N13 (HPP 0.5% + polypropylene 2d 0.25% + carbon 0.25%). Carbon fibers worked well with the polypropylene 2 denier fibers. Either the toughening occurred due to both carbon and polypropylene micro fibers or due to the polypropylene micro fibers alone. Since previous studies have indicated that carbon fibers were not very effective in enhancing the flexural toughness, one could attribute the enhancement in the load vs. deflection response of the hybrid composites to the polypropylene micro fibers. It is clear from the results that there is significant synergy available in hybrid fiber reinforced concrete. These composites possess much greater toughness and crack growth resistance over conventional FRC and are very useful for various structural 87 and non-structural components. It was concluded that some hybrid composites exhibit greater synergy at smaller deflections and others at larger deflections. In general, it was found that micro fibers enhance synergy at smaller deflections and the macro fibers at larger deflections. The other parameter that would affect the performance in hybrid composites is the modulus of the fiber. Stiffer fibers will carry more load at small deflections and low modulus fibers at large deflections. • Pull-out Macro-fiber pull-out toughness and the peak load capacity increased with the fiber inclination relative to the face of the specimen. The energy absorption capacity of a macro-fiber undergoing pullout in a hybrid matrix was similar to that of a fiber undergoing pull-out in a non hybrid matrix. HPP fibers experienced fracture rather than pull-out when embedded in a high strength matrix • Shrinkage Shrinkage tests show that the hybrid composites have reduced plastic shrinkage cracking and even at a small dosage of fibers, the cracking could be completely eliminated. A dosage of 0.05% fiber volume was found to be effective in reducing the crack area and crack width formed in the cast overlay. The number of cracks increased at higher dosage rates but their crack widths decreased. This resulted in an overall decrease in the crack area. It was also found that the crack area reduced monotonically with an increase in the volume fraction of the fiber. 88 Chapter 8- Recommendations for Further Research 1. The use of different fiber combinations in the hybrid fiber reinforced concrete has been ad hoc. Fracture mechanics based studies are necessary to rationally develop hybrid composites. 2. In this investigation, the effect of denier size of the micro fibers on toughness could not clearly be established. Future studies must explore this aspect further. 3. Data indicate that greater synergy may be expected when smaller volume fractions of the primary macro fibers are used. Thus, it would be worth investigating various fiber combinations at low volume fractions of macro-fibers. 4. Greater synergy may also be expected at low matrix strengths, and this aspect needs to be explored. 5. Tests indicate that low dosages of Stealth polypropylene micro fibers do not work well, hence larger volume fraction of these fibers should be investigated. 6. It is believed that the post peak response of the hybrids might differ when a closed loop system is used, thus it would be interesting to perform tests under a closed-loop system. 7. In this investigation, fibers increased the air content of FRC and the increase in air content depends on the surface area of the fibers. In the case of hybrids, one must suitably decide on the type of fibers, which might be helpful in reducing the entrapment of air in the concrete. Also, defoamers and viscosity modifiers can be used to achieve the desired fresh properties of HyFRC. 89 8. Durability of concrete in many structures has been a major concern, and durability of concrete is intimately related to its permeability. Permeable concrete allows the ingress of water and corrosive agents, which can be detrimental to rebars in concrete, freeze-thaw resistance and overall durability of concrete. Hybrids are known to inhibit the growth of the cracks at both micro and macro levels. This, in turn, helps to reduce the permeability of concrete, making the concrete resistant to ingress of water and other deleterious agents. Tests must be carried out to evaluate the durability and permeability of the hybrids. 9. Some studies indicate that hybrids may depict high synergy when exposed to dynamic loads. Hybrids are expected to perform well under fatigue loading and impact, thus these tests should be performed. 10. Previous studies have shown that FRC has much higher shear strength. Further investigation to study the enhancement in shear for the hybrids might help reduce the stirrup spacing in structural components. One must carry out fracture growth studies in hybrids to study the development and propagation of the cracks and to monitor the stress field at the tip of a propagating flaw. 11. Microscopic examination, where emphasis would be laid on studying the micro-structural development at the fiber matrix interface would be of great help in understanding these hybrid composites. Fiber pull-out tests should be carried out so that the influence of the secondary fibers on the pull-out response of the macro fibers could be precisely understood. 12. Very exciting and promising results were observed under restrained shrinkage when a small volume fraction of fibers was added to the cement-based matrix. Further research must explore other fiber combinations. 90 13. It has also been observed that there are many secondary advantages of hybrids, such as enhanced electrical conductivity (when carbon fibers are used), this must be further studied. 91 Bibliography I] Bentur, A., and Mindess, S., "Fibre Reinforced Cementitious Composites", Elsevier Applied Science, 1990. 2] Gupta, R , Majdzadeh, F., Soleimani, S.M. and Banthia, N., "Hybrid Fiber Reinforced Concrete (HyFRC) for Bridges", Proceedings, Sixth International Conference on Short and Medium Span Bridges, CSCE 2002, pp. 297-304. 3] Walton, P.L., and Majumdar, A.J., "Cement-based Composites with Mixtures of Different Types of Fibres", Composites, Vol. 6, Sept.1975, pp. 209-216. 4] Kobayashi, K., and Cho, R , "Flexural Characteristics of Steel Fibre and Polyethylene Fibre Hybrid-reinforced Concrete", Vol. 13, April 1982, pp. 164-168. 5] Larsen, E.S., and Krenchel, H., "Durability of FRC-Materials", Materials Research Society Symposia Proceedings, Vol. 211, 1991, pp. 119-124. 6] Glavind, M . , and Aarre, T., "High-strength Concrete with Increased Fracture-toughness", Materials Research Society Symposia Proceedings Vol. 211, 1991, pp. 39-46. 7] Feldman, D., and Zheng, A., "Synthetic Fibres for Fibre Concrete Composites", Materials Research Society Symposia Proceedings, Vol. 211, 1991, pp. 123-128. 8] Soroushian, P., Tlili, A., Alhozaimy, A., and Khan, A. "Development and Characterization of Hybrid Polyethylene Fiber Reinforced Cement Composites" ACI Materials Journal, Mar-Apr 1993, Vol. 90, No. 2, pp. 182-190. 9] Komlos, K., Babal, B., and Nurnbergerova, T., "Hybrid fibre-reinforced concrete under repeated loading," Nuclear Engineering and Design, Vol. 156, 1995, pp. 195-200. 10] Mobasher, B., and Li , C.Y., "Mechanical Properties of Hybrid Cement Based Composites, "ACIMaterials Journal, Vol. 93, No.3, 1996, pp. 284-293. II] Qian, C.X., and Stroeven, P., "Development of Hybrid Polypropylene-steel Fiber-reinforced Concrete", Cement and Concrete Research ", Vol. 30, 2000, pp. 63-69. 12] Xu, G., Magnani, S., and Hannant, D.J., "Durability of Hybrid Polypropylene-Glass Fibre Cement Corrugated Sheets", Cement and Concrete Composites, Vol. 20, 1998, pp. 79-84. 92 [13] Banthia, N., Yan, C , and Bindiganaville, V., "Development and Application of High Performance Hybrid Fiber Reinforced Concrete", Proceedings BEFIB, RILEM, Lyons, France, September 2000. [14] Banthia, N., and Sheng, J., "Micro-reinforced Cementitious Materials", Materials Research Society Symposia Proceedings, Vol. 211, 1991, pp. 25-32. [15] Banthia, N., Moncef, A., Chokri, K., and Sheng, J., "Uniaxial Tensile Response of Microfibre Reinforced Cement Composites", Journal of Materials and Structures RILEM, Vol. 28, No. 183, 1995, pp. 507-517. [16] Shah, S.P., "Do Fibers Increase the Tensile Strength of Cement-Based Matrixes?", ACIMaterials Journal, Vol. 88, No. 6, 1991, pp. 595-602. [17] ACI Committee 544, "Fiber Reinforced Concrete", American Concrete Institute, ACI 544.1R-96, May 1997. [18] Conoco Incorporation, < http://carbonfiber.conoco.com/> [19] A S T M C 1018-97, "Standard Test Method for Flexural Toughness and First Crack Strength of Fiber-Reinforced Concrete (Using Beam with Third-Point Loading)", American Society of Testing and Materials, Philadelphia, 1998, Vol. 04.02, Philadelphia, 1998, pp. 506-513. [20] JSCE SF-4, "Method of Test for Flexural Strength and Flexural Toughness of Fiber Reinforced Concrete (SF-4)," Japan Society of Civil Engineers, 1984, pp. 58-66. [21] Banthia, N., and Trottier, J.F., "Concrete Reinforced with Deformed Steel Fibers: Part II: Toughness Characterization," ACI Structural Journal, V. 92, No 2, March-April 1995. [22] Ramanalingam, N., Paramasivan, P., Mansur, M.A. , and Maalej, M . , "Flexural Behavior of Hybrid Fiber-Reinforced Cement Composites Containing High-volume Fly Ash," Seventh ACI/CANMET Conference on Fly Ash, Silica Fume (SP-199), 2001, pp. 147-162. [23] Group 7, "Fiber Denier Size-effect on Toughness of Fiber Reinforced Concrete", Course Civl 321, Department of Civil Engineering, University of British Columbia, 2002. [24] A S T M 1399, "Test Method for Obtaining Average Residual-Strength of Fiber-Reinforced Concrete", Vol. 04.02, Philadelphia, 1999, pp. 677-681. 93 [25] Qi, H.B., Hua, Y. , Jiang, Z.Q., Huang, S.Z. and Zhang, S.B., "Microstructure of the Carbon and Polypropylene Hybrid Fiber Reinforced Concrete Acted by Bending and Tensile Stress", Key Engineering Materials, Vol. 183, 2000, pp. 881-886. [26] Horiguchi, T., and Sakai, K., "Hybrid Effects of Fiber-Reinforced Concrete on Fracture Toughness", Third ACI International Conference (SP-172), Malaysia, 1997, pp. 535-548. [27] Group 2, "Hybrid Fiber Reinforced Concrete- Flexural Toughness and Synergy", Course Civl 321, Department of Civil Engineering, University of British Columbia, 2002. [28] Soleimani, S.M., "Flexural Response of Hybrid Fiber Reinforced Cementitious Composites", Masters Thesis, University of British Columbia, September 2002. [29] Valle, M . , and Buyukozturk, O., "Behavior of Fiber Reinforced High-Strength Concrete under Direct Shear" ACI Materials Journal, Mar-Apr 1993, Vol. 90, No. 2, pp. 122-133. [30] Group 6, "Industrial Floor Design", Course Civl 321, Department of Civil Engineering, University of British Columbia, 2002. [31] Sun, W., Chen, H. , Luo, X., and Qian, H., "The Effect of Hybrid Fibers and Expansive Agent on the Shrinkage and Permeability of High-Performance Concrete", Cement and Concrete Research, Vol. 31, 2001, pp. 595-601. [32] Kim, W., and Sakai, K., "Hybrid Effects of Fiber-reinforced Concrete on Fracture Toughness", Third ACI International Conference (SP-172), Malaysia, 1997, pp. 535-548. [33] Banthia, N., Djeridane, S., and Pigeon M . , "Electrical Resistivity of Carbon and Steel Micro-Fiber Reinforced Cements", Cement and Concrete Research, Vol. 22, 1992, pp. 804-814. [34] High Performance Concretes, a State of the Art Report," Report No. SHRP-C/CF-91-103, Strategic Highway Research Program, National Research Council, Washington, DC, 1991. [35] Aitcin, P.C., and Mindess, S., "High-Performance Concrete: Science and Applications", Materials Science of Concrete, Vol. 5, 1998, pp. 477-511. [36] Mobasher, B., <http://ceaspub.eas.asu.edu/cement > 94 [37] Mehta, P.K., and Aitcin, P.C., "Microstructural Basis of Selection of Materials and Mix Proportions for High-Strength Concrete"; Second International Symposium on High-Strength Concrete, ACI SP-121, American Concrete Institute, Detroit, 1990, pp. 265-86. [38] Larrard de, F., "Are Small Aggregates Really Better Than Coarser Ones for Making High-Strength Concrete?" Cement Concrete Aggregates. [39] Master Builders Technologies <http://www.masterbuilders.conVeprise/main/MBT/Content/Products/pageProduc tDisplav?obiectid=2302> [40] A S T M C 494-98, "Standard Specification for Chemical Admixtures for Concrete", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 253-261. [41] Master Builders Technologies <http://www.masterbuilders.com/eprise/main/MBT/Content/Products/ReadyMix/ pageByProductType?obiectid=2430&seed=65> [42] Silica Fume Association, <http://www.silicafume.ors/seneral-silicafume.html> [43] "Water Tower Place, High-Strength Concrete," Concrete Construction, Vol. 21, No. 5, 1976, pp. 102-104. [44] Bilodeau, A., and Malhotra, V . M . , "Concrete Incorporating High Volumes of A S T M Class F Fly Ashes: Mechanical Properties and Resistance to Deicing Salt, Scaling and to Chloride Ion Penetration", ACI SP-132, American Concrete Institute, Detroit, MI, pp. 319, 335. [45] Rougerson, P., and Aitcin, P.C., "Optimization of the Composition of a High-Performance Concrete," Cement Concrete Aggregates, Vol. 16, No. 2, 1994, pp. 115-24. [46] A S T M C 1240-97b, "Standard Specification for Silica Fume for Use as a Mineral Admixture in Hydraulic-Cement Concrete, Mortar, and Grout", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 635-640. [47] Odler, I., and Zuru, A., "Structure and Bond Strength of Cement-Aggregate Interfaces," Advanced Cement Research, Vol. 1, No. 4, 1998, pp. 21-27. [48] Bentur, A., and Cohen, M.D., "The Effect of Condensed Silica Fume on the Microstructure of the Interfacial Zone in Portland Cement Mortar," Journal of American Ceramic Society, Vol. 70, No. 10, 1987, pp. 738-742. 95 [49] Bentur, A., "Microstructure, Interfacial Effects, and Micromechanics of Cementitious Composites", Advances in Cementitious Materials, Ceramic Transactions Vol. 16, Edited by S. Mindess, The American Ceramic Society, Westerville, OH, 1991, pp. 523-549. [50] Silica Fume Association, <http://wvvw.silicafnme.org/> [51] A S T M 192/C 192M -95, "Standard Practice for Concrete test specimens in the Laboratory", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 112-119. [52] A S T M C 172-97, "Standard Practice for Sampling Freshly Mixed Concrete", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 105-106. [53] A S T M C 143 143M-97, "Standard test method for Slump of Hydraulic-Cement . Concrete", Vol. 04.02, Philadelphia, 1998, pp. 89-91. [54] Canadian Portland Cement Association, "Concrete Design Handbook", 2d ed., Canada: CPCA 1998. [55] A S T M C 231-89 a, "Standard test method for Air content of Freshly mixed concrete by the Pressure method", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 134-141. [56] A S T M C 39-96, "Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens", American Society of Testing and Materials, Vol. 04.02, Philadelphia, 1998, pp. 17-21. [57] Halvorsen, G.T., "Concrete reinforced with plain and deformed steel fibers", Report No. DOT-TST 76T-20, US Department of Transportation, Washington, DC, 1976, pp. 71. [58] Johnston, C D . , "Steel fiber reinforced mortar and concrete: A review of mechanical properties, Fiber Reinforced Concrete", ACI SP-44, American Concrete Institute, Detroit, 1974, pp. 127-42. [59] RILEM Technical committee 19-FRL, Report, Materiaux et Construction (RILEM), 10 (1977), pp. 103-20. [60] Williamson, G.R., "The effect of steel fibers on the compressive strength of concrete", Fiber Reinforced Concrete, ACI SP-44, American Concrete Institute, Detroit, 1974, pp. 195-207. 96 [61] Morris, A.D., and Garrett, G., "A comparative study of the static and fatigue behaviour of plain and steel fibre reinforced mortar in compression and direct tension", International Journal of Cement Composites and Lightweight Concrete, Vol.3, 1981,pp. 73-91. [62] Mangat, P.S., and Azari, M . M . , "Influence of steel fibre reinforcement on the fracture behaviour of concrete in compression", International Journal of Cement Composites and Lightweight Concrete, Vol. 6, 1984, pp. 219-32. [63] Atepegba, D., and Regan, P.E., "Performance of steel fibre reinforced on the fracture behaviour of concrete in compression", International Journal of Cement Composites and Lightweight Concrete, Vol. 3, 1981, pp. 255-259. [64] Mangat, P.S., & Azari, M . M . , "Influence of steel fibre and stirrup reinforcement on the properties of concrete in compression members", International Journal of Cement Composites and Lightweight Concrete, Vol. 7, 1985, pp. 183-92. [65] Furlan, S., and de Hanai, J.B., "Shear Behaviour of Fiber Reinforced Concrete Beams", Cement and Concrete Composites, Vol. 19, 1997, pp. 359-366. [66] Naaman, A.E. , and Najm, H. , "Bond-Slip Mechanisms of Steel Fibers in Concrete," ACI Materials Journal, Vol. 88, No. 2, March-April 1991. 97 Appendix A 98 Calculations for Toughness Indices/Residual Strength Factors Specimen Design. Toughness Indices Residual Strength Factor Is Average Is Iio Average Iio I20 Average I20 1*5,10 Average R5.10 Rl0,20 Average R10.20 N l _ _ _ _ _ - . - -N2-1 3.667 4.506 6.021 7.923 10.391 14.913 47.08 68.34 43.7 69.902 N2-2 3.835 6.942 13.115 62.14 61.73 N2-3 4.974 8.701 16.334 74.54 76.33 N2-4 5.922 11.295 22.978 107.46 116.83 N2-5 * * * * N2-6 4.13 6.654 11.746 50.48 50.92 N3-1 3.029 3.526 5.34 6.814 10.153 13.666 46.22 65.753 48.13 68.517 N3-2 3.989 7.446 14.247 69.14 68.01 N3-3 2.727 6.043 16.738 66.32 106.95 N3-4 3.379 6.602 10.859 64.46 42.57 N3-5 4.132 7.921 15.923 75.78 80.02 N3-6 3.902 7.532 14.074 72.6 65.42 N4-1 * 2.547 * 3.953 * 6.72 * 28.108 27.67 N4-2 3.057 5.034 8.503 39.54 34.69 N4-3 1.948 2.722 4.768 15.48 20.46 N4-4 2.5 4.125 7.331 32.5 32.06 N4-5 2.778 4.28 7.359 30.04 30.79 N4-6 2.454 3.603 5.638 22.98 20.35 N5-1 2.65 2.283 3.659 2.998 5.667 4.48 20.18 14.3 20.08 14.812 N5-2 2.497 3.483 5.537 19.72 20.54 N5-3 2.117 2.928 4.746 16.22 18.18 N5-4 2.049 2.609 3.652 11.2 10.43 N5-5 2.04 2.396 3.114 7.12 7.18 N5-6 2.347 2.915 4.161 11.36 12.46 N6-1 2.463 2.94 3.022 4.397 5.214 7.365 11.18 29.14 21.92 29.688 N6-2 3.251 4.808 7.823 31.14 30.15 N6-3 3.012 5.142 8.786 42.6 36.44 N6-4 2.933 4.398 6.97 29.3 25.72 N6-5 2.56 3.3 4.97 14.8 16.7 N6-6 3.418 5.709 10.429 45.82 47.2 N7-1 2.149 3.299 2.72 5.175 3.706 8.579 11.42 37.52 9.86 34.033 , N7-2 3.875 6.799 10.821 58.48 40.22 N7-3 3.454 5.335 9.153 37.62 38.18 N7-4 3.484 5.914 10.888 48.6 49.74 99 N8-1 3.253 2.9225 4.934 4.3182 8.344 7.1427 33.62 27.913 34.1 28.245 N8-2 4.262 6.098 9.909 36.72 38.11 N8-3 2.036 2.847 4.375 16.22 15.28 N8-4 2.543 3.502 5.418 19.18 19.16 N8-5 2.614 4.254 7.611 32.8 33.57 N8-6 2.827 4.274 7.199 28.94 29.25 N9-1 2.668 2.9183 4.077 4.71 7.04 8.3275 28.18 35.833 29.63 36.175 N9-2 3.071 4.975 8.759 38.08 37.84 N9-3 3.043 5.045 9.135 40.04 40.9 N9-4 2.915 4.799 8.582 37.68 37.83 N9-5 3.031 4.866 8.471 36.7 36.05 N9-6 2.782 4.498 7.978 34.32 34.8 N10-1 2.779 2.7267 4.185 4.253 7.124 7.3717 28.12 30.527 29.39 44.59 N10-2 2.637 4.137 7.207 30 30.7 N10-3 3.122 5.218 9.677 41.92 44.59 N10-4 2.7 4.273 7.419 31.46 31.46 N10-5 2.299 3.553 6.252 25.08 26.99 N10-6 2.823 4.152 6.551 26.58 23.99 N l l - 1 2.6 3.5118 4.44 6.3773 8.182 12.061 36.8 57.31 37.42 56.832 N l l - 2 3.098 5.247 9.623 42.98 43.76 N l l - 3 3.861 7.339 14.443 69.56 71.04 N l l - 4 3.754 7.016 13.516 65.24 65 N l l - 5 4.597 8.986 17.506 87.78 85.2 N l l - 6 3.161 5.236 9.093 41.5 38.57 N12-1 3.342 3.6756 5.566 6.1686 10.072 10.97 44.48 49.86 45.06 48.012 N12-2 4.058 6.735 11.553 53.54 48.18 N12-3 4.086 7.052 12.549 59.32 54.97 N12-4 * * * * * N12-5 3.579 6.351 11.9 55.44 55.49 N12-6 3.313 5.139 8.775 36.52 36.36 N13-1 2.196 2.5588 3.172 3.656 5.023 6.047 19.52 21.944 18.51 23.91 N13-2 1.987 2.778 4.893 15.82 21.15 N13-3 2.983 4.376 7.403 27.86 30.27 N13-4 2.641 3.555 5.459 18.28 19.04 N13-5 2.987 4.399 7.457 28.24 30.58 N13-6 * * * * * 100 N17-1 3.267 4.913 8.392 32.92 34.79 N17-2 1.934 2.546 3.595 12.24 10.49 N17-3 2.172 3.047 4.664 17.5 16.17 N17-4 2.228 2.7097 2.875 3.7413 4.207 5.7198 12.94 20.633 13.32 19.785 N17-5 2.875 4.134 6.367 25.18 22.33 N17-6 3.782 4.933 7.094 23.02 21.61 * Values not included in Calculations 101 Appendix B 102 8 3 3 3 3 3 3 3 103 s 3 3 3 3 3 3 3 3 5 3 104 5 3 3 3 105 s s s s 3 3 S 3 3 3 106 E poSt,m(For Calculations only) Units-kN.mm S u 45.635 1 52.525 31.656 50.077 45.609 1 25.595 42.369 43.839 31.401 36.965 E poSt,m(For Calculations only) Units-kN.mm s a? O 0. 13.603 20.8 19.099 34.294 38.584 18.807 24.763 39.072 33.439 23.113 34.067 34.205 19.327 30.779 31.443 23.87 26.777 11.788 17.025 E poSt,m(For Calculations only) Units-kN.mm 1 GO O a. 9.7081 14.351 12.914 22.503 28.732 11.798 16.111 25.713 21.76 14.573 22.669 22.853 12.645 19.707 19.584 16.042 17.129 7.348 10.213 E poSt,m(For Calculations only) Units-kN.mm co U a. 7.196 10.528 9.683 16.72 21.535 8.46 12.007 ON 15.938 10.623 16.789 17.058 9.51 14.293 14.145 12.05 12.434 5.621 7.158 E poSt,m(For Calculations only) Units-kN.mm to u 0. 4.689 r~ NO 5.242 10.932 12.714 5.314 7.819 12.249 10.189 6.863 10.772 11.189 6.446 8.958 8.867 oo 7.948 3.318 4.336 E poSt,m(For Calculations only) Units-kN.mm ? co u a. 3.701 5.145 4.749 8.755 10.117 4.062 6.064 9.441 8.061 5.418 8.322 8.835 1 5.077 6.911 6.727 6.08 6.164 2.566 3.292 E poSt,m(For Calculations only) Units-kN.mm o CO u a. 2.725 3.662 3.311 6.209 6.545 2.875 4.199 00 NO 5.737 3.944 5.885 6.357 3.738 4.828 4.636 4.328 4.308 1.822 2.323 E poSt,m(For Calculations only) Units-kN.mm © g CO U a. 1.64 2.222 1.927 3.609 5.134 1.624 2.497 3.906 3.574 2.509 3.347 3.464 2.348 2.599 2.479 2.559 2.345 1.115 1.354 E poSt,m(For Calculations only) Units-kN.mm a? y B . 0.4987 0.71 0.511 1.167 2.027 0.458 0.755 1.666 1.465 1.54 1.022 0.729 I 0.858 0.503 0.338 I860 o 0.35 0.45 (kN.mm) 1.004 0.849 1.036 1.024 7.523 1.248 1.605 1.339 1.184 1.063 1.853 1 1.288 3.796 2.858 2.229 4.152 1.224 1.414 Deflection a/ Peak(mm) 0.059 0.059 0.0565 0.0655 0.0595 0.065 0.073 0.058 0.07 0.058 0.06 0.0565 0.0795 0.0595 0.0785 0.085 0.0655 0.059 0.056 0.056 Mix Design. z N2-1 N2-2 N2-3 N2-4 N2-5 N2-6 N3-1 N3-2 N3-3 N3-4 N3-5 N3-6 N4-1 N4-2 N4-3 N4-4 N4-5 N4-6 NS-1 N5-2 107 1 1 30.188 10.834 33.378 15.603 11.213 26.749 27.85 26.849 13.634 15.547 42.559 22.167 31.82 18.258 17.642 11.788 14.832 9.316 15.981 22.009 15.743 13.613 19.609 8.01 26.323 13.898 24.404 9.422 20.522 22.02 20.001 10.41 11.83 31.516 17.129 24.186 11.312 10.907 7.251 9.404 6.204 10.2 14.183 9.801 8.86 12.054 5.315 18.108 11.027 17.062 7.111 13.725 15.109 13.2 7.243 8.07 19.74 11.093 15.754 7.812 7.737 5.155 6.74 4.697 7.108 10.189 7.077 6.606 8.404 3.982 13.419 8.574 12.051 5.652 10.004 10.9 Ov 5.575 6.09 14.227 7.908 11.613 4.639 4.683 3.207 ( N 3.203 4.495 6.323 4.361 4.458 5.122 2.641 8.959 5.755 7.852 3.951 6.376 6.887 6.055 3.797 4.082 8.851 4.916 7.435 3.496 3.584 2.462 3.265 2.585 3.49 4.896 3.375 3.138 3.933 2.12 7.182 4.566 6.024 3.176 5.014 5.321 4.644 3.049 3.248 4.811 3.767 5.765 2.444 2.522 1.716 2.338 1.982 2.547 3.494 2.392 2.158 2.812 1.571 5.514 3.269 4.184 2.309 3.685 3.743 3.294 2.268 4.832 2.676 4.09 VO so T f IT) ( N O m 00 m 2.74 VO VO •3- ( N c n <r; 1.057 3.896 2.045 2.385 1.474 2.371 2.264 1.892 1.365 1.489 3.025 1.626 2.447 0.568 0.589 0.35 0.551 0.615 0.79 0.138 0.823 0.535 0.703 0.514 3.296 0.796 0.672 0.662 1.306 0.7509 0.594 0.468 0.5851 1.071 0.552 0.797 1.416 2.056 2.791 2.106 0.758 0.715 1.915 1.553 1.673 0.63 1.544 1.314 1.144 1.548 1.098 1.081 1.601 1.791 1.75 1.475 1.447 1.421 1.645 0.052 0.0465 0.053 0.054 0.0395 0.042 0.084 0.0735 0.066 0.044 0.0447 0.055 0.056 0.067 0.055 0.053 0.056 0.063 0.059 0.054 0.048 0.062 0.056 N5-3 N5-4 N5-5 N5-6 N6-1 N6-2 N6-3 N6-4 N6-5 N6-6 N7-1 N7-2 N7-3 N7-4 N7-5 N7-6 N8-1 N8-2 N8-3 N8-4 N8-5 N8-6 N9-1 108 co co p CN co CN NO ON t--oq m CN 00 co cO i n CN co i n <n 5-co NO ON ON o CN fN NO ON CO T t r -o CO i n o i n 00 ON ON <N i n »n fn ON fN i n p i n i n NO co ^o co CN CN CO «n fN r-^  fN od en wS CO CN fN co CN ON m CO i n i n * n »n CO ON d i n od fN m i n NO co NO i n NO co OO CO = »n r -r--r r i n NO TJ-co co TJ-OO ON i n fN i n o fN o oq CO ON <n ON CO oo oo o - N O fN fN NO i n NO NO fN od fN CN CN fN 00 d CN CN >n CN \6 od d m i n CN CN fN NO fN NO d i n m 1 O N fN fN fN co r-~ 00 00 ON vo oo i n r -ON so CN co NO NO fN r -i n ON r -co fN co ON fN CO 00 OO i n NO OO oq i CO CO p •5 ON ON ON fN O OO fN N O m ON t-; r-^ NO <n i n ™" fN d CN od fN od CN CM r n fN ON fN • O N co oo co c-i co co 00 00 co co <n co * n «n »n co r -fN i n fN <n r~-NO O fN ^O oq CN fN oq o CO NO i n rn oo fN p O fn <N T t CN fN d d ON d fN - NO od ON <n fN fN fN d cs CN NO od ^ * i n p cO oq i n 00 ON NO ON NO ON fN SO i n CN i n i n ON en CN ON i n O NO »n o OO ON CO r -fN O >n i n 00 NO NO d O NO t ON >n 00 o fN od NO NO r--' NO r-^ i n i n ON 00 od ON ON i n fN ON ON O i n oo CN oo CN ON 00 i n O NO r-; 00 o NO fN ON CN NO fN CO ro ON CN CO fN fN <n o 00 o 00 fN fN i n p ON NO i n i t CN. ON ffl NO ^o i n i n i n co «n i n i n •*t NO o\ NO od t-^ O N • NO i n co i n fN co CN oq NO ON -3-oq NO CO NO o i n co r -ON TJ-ON ON co <n p fN p CN T t «n NO r -«n 00 ON NO NO m NO o NO N O 00 NO i - ~ co co CO CN co co co co co in" TT' r--* NO »ri NO NO en in co co i n oo co CN CN CN i n co p ON fN fN fN co co 00 oq i n ON fN NO i n 00 CN 00 CN Tl- o so ON m i n ON o oo fN CN CN CN CN CN ~* CN fN fN —* — CN fN fN ro f*S 1 <N fN 00 ON »n 00 CN ON Tl" i n NO 00 <n <n CN i n co ON NO NO r--CO ON ON o OO i n CO ON 00 NO 00 ON i n oo p m ON oo oo N O p NO C> " n NO i n d d d d d ^ > d d d d d d d — —' d ~- d d — d d ON <n NO NO r-- CN co r -o NO ON O co ON NO CN oq NO ON CN NO ON ON oo ON CN >n oo oo r~~-»n ON i n NO NO CO NO NO fN ON »n s NO ON oo <—i — co —. ~~ ~* d — ~" ~* ~ ~ (N — (N 1 — fN ON i n o CN SO O »n o i n o i n NO o ON ^o o o 3 o i n i n o oo CO o i n CN NO o »n o 00 NO o i n oo NO o »n ON NO O CN «n o i n i n NO o o NO NO o ON N O O oo o d d d d d d d d d d d d d d d d d d d d 1 d d N9-2 co ON 2 ON Z <n ON 2 VO ON Z i o fN 1 © Z co o Z o Z m o Z NO o Z i Z fN i z CO Z z •n i z NO i z i fN z (N I fN z fn fN z T fN z i n <N z NO CN z 109 3 3 3 3 3 3 110 Appendix C 111 Summary of Crack Results Total crack Max crack Number of Fiber content Mix Specimen no. area (mm2) width (mm) cracks Control (0.0%) 1 414.03 1.94 4 2 327.50 1.58 4 3 289.15 1.41 3 Average 343.56 1.64 4 0.05% Mix 1 1 257.64 0.90 5 2 140.07 1.08 4 3 188.88 0.90 5 Average 195.53 0.96 5 Mix 2 1 287.98 1.44 4 2 184.4 0.70 6 3 N/A N/A N/A Average 236.19 1.07 5 0.08% Mix 1 1 200.86 1.28 2 2 137.80 0.78 3 3 180.42 0.76 5 Average 173.03 0.94 3 Mix 2 1 198.9 0.9 4 2 55.20 0.89 4 3 93.7 0.68 3 Average 115.93 0.82 4 0.1% Average 0.0 0.0 0 112 


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