Effects of steel fiber geometry on mechanical properties of concrete

Concrete is still the most commonly used material in construction sector in terms of strength and cost. Besides, workability, resistance against water and high temperature and low maintenance cost can be listed as its advantages. But unfortunately, its brittle behavior, creep and cracks owing to the shrinkage and durability problems such as corrosion, alkali silica reactions etc. do already exist. In recent years, so as to handle these problems, there has been a remarkable increase in the number of researches which are done not only in universities but also in laboratories that are located in developed companies. By the way, plenty of additives for instance, chemical and mineral additives and fibers are used. By adding steel fibers to concrete, brittle behavior of concrete is modified and turns into ductile behavior. Due to this, toughness of composite tends to increase, less cracks occur, the durability problems are prevented or delayed step by step. Generally in Turkey only compressive strength is perceived as the target whereas other paramount properties of concrete are subordinated. As a result of this reason, the service life of structures is dramatically short, at the same time loss of life is possible. But people are becoming more concerned about the other parameters of concrete after natural disasters that happened in Turkey, especially the Marmara Earthquake. Steel fiber reinforced concretes are highly used for tunnel timbering, floorings and particularly very large surfaces, slope stabilization, dam building, seismic retrofitting. Usage areas of steel fibers contribute to tensile strength and flexural strength of composite rather than its compressive strength. This study aims to obtain the difference of effect on concrete between two sorts of hooked steel fiber in terms of fresh and hardened concrete properties such as workability, compressive strength, flexural strength, tensile strength, pull-out strength. To obtain toughness and anchorage strength, "Third-Point Loading" tests and pull-out tests were done respectively. Third-Point Loading was applied to prisms which were sized as 100x100x500 mm. Moreover, the distribution and orientation of steel fibers and the path of crack after loading were acquired via x-ray diffraction. Apart from the pull out test, all experiments were carried out at 28th day. The study comprises of six sections and follows this algorithm respectively, introduction, researches about the steel fiber reinforced concretes, the design of steel fiber reinforced concretes based on performance, experimental study, overview of results and conclusions. Two steel fibers' aspect ratio and their length are the same, 67 and 60 mm. Their geometry is a little bit different. 3D has one hooked and 4D has double hooked at the end of fibers. But flexural strength is different. The former one's flexural strength is 1000 MPa while the latter one's strength is 1500 MPa. Since to be produced easier, fibers are glued and attached to each other. During preparation of concrete, adhesive part of fibers dissolves and fibers release. This study is divided into four groups using different matrice strengths and steel fiber types; First group: this group's concrete had a normal compressive strength and its water/ binder equals to 0.5. Only Portland Cement was selected as binder and 4D fiber type was used. The content of super plasticizer was determined %1.33 of cement weight. Moreover within the various ratios, four aggregates were added to concrete mix. These aggregates were sand, crushed sand, crushed stone I and crushed stone II. This group consists of four subgroups; one control series and three steel fiber reinforced concrete series. Furthermore, fibers were used in the mixture of concrete with a various volumetric ratios: %0 (control specimen), %0.25 (20kg/m³), %0.50 (40kg/m³), %0.75 (60kg/m³). After mixing, each subgroup's concrete was casted into prisms and cylinders until molded. Second group: this group's concrete had a high compressive strength and its water/ binder equals to 0.37. Only Portland Cement was selected as binder and 4D fiber type was used. The content of super plasticizer was determined %4.5 of cement weight. Moreover within the various ratios, three aggregates were added to concrete mix. These aggregates were sand, crushed sand, crushed stone I. Different form first group, crushed stone II was not used because it is not proper to be in the mixture of self-compacting concrete. This group consists of four subgroups as 1st group does; one control series and three steel fiber reinforced concrete series. Similarly, fibers were used in the mixture of concrete with a various volumetric ratios: % 0 (control specimen), %0.25 (20kg/m³), %0.50 (40kg/m³), %0.75 (60kg/m³). After mixing, each subgroup's concrete was casted into prisms and cylinders until molded. Third group: this group's concrete had a normal compressive strength and its water/ binder equals to 0, 5. Only Portland Cement was selected as binder and 3D fiber type was used. The content of super plasticizer was determined %1.33 of cement weight. Besides, the same ratios with first group, four aggregates were added to concrete mix. These aggregates were sand, crushed sand, crushed stone I and crushed stone II. Contrary to this group consists of three subgroups; three steel fiber reinforced concrete series without any control series. As a result of having no control series, fibers were used in the mixture of concrete with a various volumetric ratios: %0.25 (20kg/m³), %0.50 (40kg/m³), %0.75 (60kg/m³). After mixing, each subgroup's concrete was casted into prisms and cylinders until molded. Fourth group: this group's concrete had a high compressive strength and its water/ binder equals to 0.5. Only Portland Cement was selected as binder and 3D fiber type was used. The content of super plasticizer was determined %4.50 of cement weight. Here, this group included three aggregates were added to concrete mix. These aggregates were sand, crushed sand, crushed stone I. Identical with second group, crushed stone II does not exist because of self-compacting concrete. This group consists of three subgroups; three steel fiber reinforced concrete series. Additionally, fibers were used in the mixture of concrete with a various volumetric ratios: %0.25 (20kg/m³), %0.50 (40kg/m³), %0.75 (60kg/m³). After mixing, each subgroup's concrete was casted into prisms and cylinders until molded. To have more accurately results, for each series concrete molded into four prisms that are sized as 100mm x 100mm x 500 mm and six cylinders with 100mm diameter and 300 mm height. Densities of coarse aggregates (crushed stone I and crushed stone II) are 2.74 g/cm³ and 2.68 g/cm³ whereas densities of fine aggregates (crushed sand and sand) are 2.65 g/cm³ and 2.76 g/cm³. The content of aggregates varies from to concrete type to concrete type because crushed stone II was not used in self-compacting concrete. According to this, steel fiber reinforced concrete that has normal compressive strength includes sand %17±1, crushed sand %33±1, crushed stone I %20±2 and crushed stone II %30±2 while in self-compacting concrete mixture, the percentages were obtained as given: sand %20±1, crushed sand %23±1, crushed stone I %57±2. After demolded, approximately 24 hours after molded, samples were put and then kept in curing pool for 27 days in order to reach maximum strength. Curing conditions was at 20 ⁰C. Several hardened concrete tests were done. Compressive test, flexural test, splitting tensile test and pull-out tests were carried out and apart from pull-out tests, in the rest of experiments cylinder samples were used. Prism samples were used for third point loading test and x-ray diffractions. As for fresh concrete tests, to determine workability, slump flow and slump tests and were done. Also, air content was calculated for each mixture. As a result of slump test that had been done for normal concretes, slump varies from 115mm to 220mm. According to another fresh concrete test for self-compacting concrete, ultimate slump flow diameter has an interval of 66cm and 79 cm. When the effect of steel fibers on fresh concrete properties such as workability is examined, it reveals that, 4D fibers show more resistance than 3D does but this difference is slight. Also, when the effect of steel fibers on air content is compared, it can be easily said that, the geometry of steel fibers does not influence this property of concrete because air contents are more or less the same. But air content directly is affected consistently but not linearly by fibers volume. Entrapped air content differs from %0.23 (control specimen) to %3.33 within the increasing fiber volume. After experiments, it is seen that the sort or content of steel fibers embedded by any matrix has no effect on compressive strength or modulus of elasticity (Young's Modulus). However during uniaxial compression test it was seen that using steel fiber can be beneficial in terms of safety. Plain concrete, using no steel fibers, ruptured under compression and noisy failure was observed. Surprisingly, steel fiber reinforced concrete samples and plain samples have approximately the same first crack load in spite of the fact that plain samples fractured very early. Different from the effects of adding steel fibers on compressive behavior, it was proved that fiber reinforced concrete's flexural and splitting tensile strength depend on steel fiber volume and matrix strength. Furthermore, effects of adding steel fibers on flexural behavior are much higher than the other contributions. Load-deflection graphs represent whether such parameters, especially toughness, sharply increases with an outstanding strain hardening. This also leads to an intensive crack control that influences from steel fiber volume and distribution-orientation around weak areas owing to the deflection capacity. Addition to the benefits of steel fibers to flexural strength and splitting tensile strength, first crack load and the level of toughness are directly related to matrix strength but effect of this relation is not as much as steel fiber content does. In third point loading test, fiber reinforced concretes that had % 0.25 steel fibers performed strain softening after first crack load whereas for fiber reinforced concretes including % 0.50 fibers, load-deflection curve reached a valley and remained constant up to 3mm deflection after first crack load. And, fiber reinforced concretes that had % 0.75 steel fibers performed strain hardening after first crack load. It was observed that two significant parameters serviceability limit state (SLS) and ultimate limit state (ULS), calculating according to German Concrete Association, depend on the increase in steel fiber quantity. It is probably due to the anchorage of fibers to matrix and resistance to the elusion. For this reason beam samples indicated a greater performance under third loading test not only in accordance with German Standards but also in accordance with ASTM C 1018 and JSCE. It is understood that both two sorts of steel fibers have non-homogenous distribution. It is highly possible as a result of molding manually and selecting relatively small sample size 100mm x 100mm x 500mm instead of 150mm x 150mm x 600mm. It can be concluded that geometry of steel fibers does not influence on steel fiber distribution and its orientation. Prisms were cut in order to obtain any cross-section and after that, digital imaging procedure was applied. Images were collected by high-resolution camera, processed by image analysis program and observed by x-ray diffractions of the samples. As expected, bonding between samples and 3D steel fiber is weaker than bonding between samples and 4D steel fiber and addition to this, lower water/cement ratio accompanies steel fiber type in terms of anchorage strength. According to outputs of pull out test, specimen that had the lowest water/cement ratio and more complex steel fiber (4D), hereby, resisted much more than other specimens.