Properties of Polypropylene Fiber Reinforced Concrete

Construction and Building Materials 98 (2015) 620–630 Contents lists available at ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat Properties of polypropylene fiber reinforced concrete using recycled aggregates a,⇑ _ Kutalmısß Recep Akça a, Özgür Çakır b, Metin Ipek a b Sakarya University, Faculty of Technology, Civil Engineering Department, 54187 Sakarya, Turkey Yıldız Technical University, Faculty of Civil Engineering, Civil Engineering Department, 34220 Istanbul, Turkey h i g h l i g h t s  Recycled aggregates were incorporated into mixtures by ratio of 25–30–55%.  Compressive strength between 32 and 43 MPa was attained with recycled aggregates.  Optimum fiber content was determined as 1% by volume.  The greatest negative impact of RCA was experienced at water penetration depths.  Use of RCA is more suitable at constructions that have low structural risk factor. a r t i c l e i n f o a b s t r a c t In this study, recycling of rubble obtained during urban transformation and manufacturing new concrete using this material was experimentally studied. Different combinations were generated using the recycled concrete aggregates and polypropylene fiber. Natural aggregates were replaced by recycled concrete aggregates (RCAs) and volume of 0%, 1% and 1.5% fiber were introduced for each series. Although concretes’ physical and mechanical properties were affected negatively by RCA due to RCA’s higher porosity and water absorption capacity, high strength concrete was eventually manufactured. Additionally, although fiber content increases flexural properties, there is no significant difference observed between 1% and 1.5%. Percentage contribution ratios of parameters which influence the results of experiments were also calculated by means of analysis of variance (ANOVA) method. As the result of ANOVA which is carried out on specimens containing fiber and recycled concrete aggregate, main factor on changes of compressive strength were determined as aggregate type, while fiber content were also influential on flexural and splitting tensile strength besides aggregate type. Ó 2015 Elsevier Ltd. All rights reserved. Article history: Received 9 March 2015 Received in revised form 14 July 2015 Accepted 26 August 2015 Keywords: Urban transformation Recycled aggregate Fiber ANOVA 1. Introduction Recycling has become prominent in construction industry in the last decades with the term of sustainable structural materials. In the case of not providing sustainable material flow, it is possible to deplete natural resources, since they are not unlimited. Urban Transformation Project has been put into practice in 2012 and will take approximately 20 years in Turkey [1]. As a result of the project, more than 5 million tons of construction and demolition waste (C&D waste) will be produced annually. Due to the fact that it takes 50% of raw materials from nature, consumes 40% of total energy and creates 50% of total waste [2], ⇑ Corresponding author. E-mail addresses: akca@sakarya.edu.tr _ (Ö. Çakır), metini@sakarya.edu.tr (M. Ipek). (K.R. http://dx.doi.org/10.1016/j.conbuildmat.2015.08.133 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved. Akça), cozgur@yildiz.edu.tr construction has become a critical industry in respect to term of sustainable materials. Amount of recycled or landfilled C&D wastes differs from country to country. In Hong Kong, approximately 20 million tons of C&D waste was produced in 2004. While 12% of the waste was disposed of at landfills, 88% was used as filling materials [3]. 180 million tons of C&D waste is generated in European Countries per year. Only 28% is recycled and reused; rest of it is sent to landfills. Netherlands, Denmark and Belgium are the most accomplished EU countries on waste management via recycling generated wastes 90%, 81% and 87%, respectively [4]. It is necessary to stated that studies for the reuse of waste that is generated during construction or demolition process should be carried out especially in fast developing countries in construction industry (such as Turkey). However, due to the fact that people are not able to abandon customary methods, aggregate being a cheap structural material and lack of recycling consciousness; RCA does not have a wide-spread use in Turkey as well as many countries. RCA is 621 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 Table 1 Chemical composition of the cement (%). Cement SiO2 Al2O3 Fe2O3 CaO MgO K2O Na2O SO3 Cl CEM I 42.5R 19.8 4.7 3.18 64.35 1.27 0.64 0.35 2.49 0.008 100 Percent Passed [%] 80 60 40 Coarse Aggr. (Recycled and Natural) ASTM for coarse aggr. Fine Aggregate ASTM for fine aggr. 20 0 0 5 10 15 20 25 Sieve Size [mm] Fig. 1. Aggregates grading. Table 2 Content of the recycled concrete aggregate. Content Concrete particles Brick&Tile particles Gypsum particles Shell particles Stucco particles River aggregate Glazed tile % 97.53 1.74 0.20 0.17 0.13 0.13 0.10 mostly used as a filling material for building construction, road foundations and hydraulics work spaces [5]. Investigations of RCA usability in structural concrete manufacture are gaining popularity day by day. Rakshvir and Barai [5] and Kartam et al. [6] state that RCA is able to be used in structural concrete rather than its use as filling material. In Spain, use of RCA in structural concretes has been regulated and encouraged with a regulation (EHE-08) that has been put into effect [7]. Committee of American Concrete Institute (ACI) put emphasis on reuse of concrete waste by publishing a document called ‘‘Removal and Reuse of Hardened Concrete” in 2001. Also in Turkey, there is a regulation which indicates that debris is able to be used in the manufacturing of concrete with or without raw materials [8]. In literature, unless natural aggregate is replaced by more than 20–30% by RCA, it is seen that there is no significant negative influence of RCA on physical and mechanical properties of concrete [9– 13]. Despite the fact that loss of approximately 20% of compressive strength is observed in the case of 100% RCA replacement, it is possible to come up with discrepancies on achieved results. These discrepancies may be arisen by heterogeneous structure of RCA. Sheen et al. [14] investigated the effect of brick and tiles presence in RCA. As a result of the experimental study, decreasing is observed on compressive strength of concrete due to high water absorption capacity of fine grained RCA. Furthermore, decreasing compressive strength also depends on content of brick and tile. Additionally, a 10–23% loss (arising from brick and tile content) in the flexural tensile strength is observed on concrete prepared with recycled aggregate when compared with control specimen. While Ajdukiewicz and Kliszczewicz [15] stated that concretes manufactured with RCA has 10% lower flexural strength, Topçu and Sßengel [16] observed a reduction of 13% on flexural strength when replacement ratio reaches 100%. In another research, it is stated that failure was generally observed on brick and tile particles that was called the mechanically weakest point [17]. Although some authors such as Matias et al. [4] reporting that replacement ratio (natural aggregate by recycled aggregate) does not have a considerable influence on splitting tensile strength as much as seen on compressive strength, Vazquez et al. [18] observed a reduction within the range of 6–20% when replacement ratio is increased to 100%. However, loss of strength was negligible when RCA incorporation ratio was lower than 50%. Evangelista and De Brito [19] confirmed Vazquez et al. that loss of 5% was observed with replacement of 30%, and 23% with 100%. In the study [13] on recycled aggregate concretes’ (RAC) water penetration depth under pressure, water penetration depths are observed to be around 30 mm independent from the replacement ratio when water/cement (w/c) ratio is lower than 0.45. In another study, it is stated that penetration depths are respectively 15, 16 and 17 mm for replacement ratios of 0%, 20% and 30% [20]. Besides, concretes having 5%, 11% and 19% RCA have a lower modulus of elasticity for incorporation ratios of 20%, 50% and 100%, respectively [13]. Having low resistance to deformation is due to porous structure of RCA which has a significant influence on modulus of elasticity. This study is aimed to investigate usability of polypropylene fiber in recycled aggregate concrete in order to be used primarily in field concrete, since the use of PP fiber in field concrete is gaining popularity nowadays. How fiber usage influences physical, mechanical and durability properties of RAC is not common in the existing studies of literature. For this purpose, different polypropylene fiber contents have been introduced into concretes that have different amount of RCA. Parameters of unit weight, ultrasonic pulse velocity, compressive strength, splitting tensile strength, flexural tensile strength, pull-out behavior of PP fibers, static and dynamic modulus of elasticity and water penetration depth under pressure were investigated. 2. Experimental programme In this study, RCAs (4/16 and 8/32 mm) were incorporated into the mixture by replacing natural coarse aggregates which have the same granulometry. Afterwards, volume of 0%, 1% and 1.5% polypropylene fibers were used in mixtures for each series. In this manner, 12 different mixtures were prepared and notation of mixtures was generated. Letters and numbers are used in the coding, ‘‘N” for natural 622 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 Fig. 2. Content of the recycled concrete aggregate. Table 3 Physical and mechanic characteristics of aggregates. Aggregate size Aggregate type Specific gravity (kg/dm3) Moisture content (%) Water absorption (%) Abrasion loss (LA) (%) Incorporation ratio (%) Coarse NATP1 NATP2 RATP1 RATP2 2.73 2.67 2.52 2.35 0.2 0 3.38 2.95 0.27 0.67 4.34 4.59 30 24 41 25 30 25 30 Fine Sand 2.75 0.34 1 – 45 aggregate, ‘‘R” for recycled concrete aggregate; numbers of 1 and 2 represents aggregate granulometry of 4/16 and 8/32 mm; A, B, C represents volume of 0%, 1% and 1.5% polypropylene fiber content respectively. 2.1. Materials CEM I 42.5R Portland cement was used and the chemical composition of the cement is given in Table 1. Three types of aggregate as grain size were used in mixtures. Aggregates with grain distribution of 0/4, 4/16 and 8/32 were named as Sand, Type 1 (TP1) and Type 2 (TP2) respectively. Fig. 1 shows the grain size distributions of fine and coarse aggregates. Both fine and coarse aggregates are within the curves specified by ASTM C33. In this study, recycled aggregates were derived from urban _ transformation project in Istanbul. Concrete which is the source of used recycled aggregates, had 8–10 MPa of compressive strength [21]. It is well-known that RCA has a heterogeneous structure. Contents of the aggregate can differ depending on the RAC source. That is why the characteristics of the aggregates must be determined by some experiments. Composition of RCA used in this study is given in Table 2 and Fig. 2. Physical (specific gravity, water absorption) and mechanical (Los angeles abrasion loss) characteristics of all aggregate types (natural and recycled) which were determined by experiments and also incorporation ratios for concrete mixes can be found on Table 3. Characteristics of wavy shaped polypropylene fibers (Fig. 3) are also shown on Table 4. As seen on Table 2, RCA used in this study consists of mainly concrete particles where brick and tile content is limited with 1.74%. Besides, it is observed that RCAs have lower specific gravity and abrasion loss but higher water absorption capacity than natural aggregates according to the experiments carried out for aggregates. Although abrasion loss of RCA is higher than natural aggregates, the obtained abrasion loss value is lower than the limit which is indicated by American standards which is used in concrete manufacture [22]. 623 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 where Ed represents the dynamic modulus of elasticity (kg/cm2), V is the ultrasonic pulse velocity (km/s), d is the unit weight (kg/dm3) and g is the gravitational acceleration (m/s2). Besides, percentage contribution ratios of parameters which influence the results of the experiments were also calculated by means of analysis of variance (ANOVA) method. 3. Results and discussion 3.1. Properties of fresh concrete Fig. 3. Shape of polypropylene fiber. 2.2. Mix proportions Conventional structural concrete (C25/30 which is mostly used strength class especially in Turkey) was manufactured according to standards [23,24]. For all specimens, w/c ratio was chosen as 0.53. Desired workability which is the range of 10– 14 cm collapse was obtained by means of superplasticizers. Generated concrete series and mix proportions were shown on Table 5. Amount of water in mixture was arranged for all mixtures by taking into account water content of aggregates while in the case of saturated dry surface. Method of water compensation (adding extra water) is also recommended by [25], rather than pre-saturation of aggregates. In order to control the mixtures; workability (slump), unit weight of fresh concrete and air void content (%) experiments were performed according to corresponding standards [26–28]. 3 concrete specimens were prepared for each experiment. 3.2. Properties of hardened concrete 2.3. Method All specimens are removed from mould 24 h after production and cured in lime saturated water at 20 ± 2 °C until the experiment day in accordance with [29]. The experiments have been carried out 28 days after casting. In order to determine physical characteristics, some experiments were performed such as unit weight, ultrasonic pulse velocity and depth of water penetration under pressure in accordance with the standards [30–32]. Depth of water penetration under pressure is one of the specific experiments which investigates the permeability of concrete under 5 bars of pressure. It also gives an opinion about the concrete’s durability. According to the standard [32], specimens which are at least 28 days old, must be exposed to water column during 72 h. By the end of the experiment, specimens are split and maximum water depth is measured. Compressive strength, splitting tensile strength, flexural tensile strength, static and dynamic modulus of elasticity experiments were conducted in order to determine mechanical performance of the concrete series. Besides, pull-out experiment was carried out in order to determine the adherence between concrete and fiber. Single fiber was centrally embedded into fresh mixture until half of the fiber’s length as seen in Fig. 4(a) [33,34]. 28 days after casting, specimens were mounted by an apparatus and single fiber was pulled out (Fig. 4b). Compressive strength test was carried out with cubic specimens of 150  150  150 mm in accordance with the standard [35]. Also, cylinder specimens (150/300 mm) were used in order to determine splitting tensile strength and static/dynamic modulus of elasticity; prism specimens (75  75  500 mm) were used in case of three point loading flexural tensile strength and toughness. Toughness values were calculated by using the area under the load–deflection curves. Splitting tensile strength and flexural tensile strength tests are both used as methods of investigation of tensile strength which have been done according to corresponding standards [36,37]. Static and dynamic modulus of elasticity were also determined. While static modulus of elasticity was being determined in accordance with [38], load and discharge method was used. The aim of elastic part of stress–strain curve is to determine the modulus of elasticity. Dynamic modulus of elasticity was evaluated for each specimen with equation that is presented below [39]: Ed ¼ 105
V 2
Properties of fresh concrete can be found on Table 6. Due to high water absorption capacity of RCA, workability of concretes which is composed of RCA was determined to be lower than conventional concrete which is composed of NA. This result is confirmed by [14,40,41], as well. When the case of high water absorption capacity of recycled aggregates is taken into account, superplasticizers (Chryso fluid MG) were used. Thus, range of 10–14 slump value was kept constant for all series. Air void content of all series stayed between the range of 1.1–1.7%. High porosity of RCA and impact of fibers on aggregate gradation increased the air void content, because of high fiber content can cause faulty mixing. It is observed that by increasing the replacement ratio, unit weight generally tends to decrease. Increase in fiber content has also similar influence on fresh concrete’s unit weight. In previous studies, while natural concretes’ unit weights were established as approximately 2400 kg/m3, concretes composed of RCA had 2100–2300 kg/m3 unit weight [14,17,42]. Obtained unit weight values from the previous studies show similarity with the results given in Table 6. d g Properties of hardened concrete and standard deviations of strength values of each concrete specimen can be found in Table 7. According to experimental results, it is observed that compressive strength decreases as the incorporation ratio of recycled aggregate is increased. While the reference concrete (N-A) has 53.5 MPa of compressive strength, the lowest strength (32.3 MPa) belongs to the specimen R12-A which has the highest replacement ratio and the loss of compressive strength is approximately 40%. 40% is the greatest compressive strength loss observed in this study and this clarifies how the greatest standard deviation (7.58 MPa) is obtained for series A (Table 8). Additionally, other recycled specimens without fiber (R1-A and R2-A) have reduction of 23.8% and 19.2% respectively. Porous structure of RCA and composition of materials had a lower strength than natural aggregate, induce decreases on compressive strength [9–14]. Although R1-A has lower RCA content ratio than R2-A, due to RATP1 having smaller grain size than RATP2, R1-A’s water demand increases and therefore, greater loss of strength is observed. This fact can be seen in Fig. 5. The same circumstance is observed also on other experimental results given in Table 7. Since PP fibers are disadvantageous, when compared with natural aggregates from the point of strength and granulometry, compressive strength of natural (reference) concrete were influenced negatively with the increase of PP fiber content. Especially for the specimens with RCA, it is seen that there is no significant effect of polypropylene fiber content on compressive strength. This is also confirmed by the result of ANOVA which is presented in Table 9. According to ANOVA analysis, it is observed that fiber has no influence on compressive strength while aggre- Table 4 Characteristics of polypropylene fibers. Fiber Type Length (mm) Diameter (mm) Slenderness ratio Tensile strength (MPa) Specific gravity (kg/dm3) Melting point (°C) Ignition point (°C) Polypropylene 50 1 50 550 0.91 164 >550 624 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 Table 5 1 m3 mix proportions. Concrete type N-A R1-A R2-A R12-A N-B R1-B R2-B R12-B N-C R1-C R2-C R12-C Cement (kg) NATP1 (kg) NATP2 (kg) RATP1 (kg) RATP2 (kg) Sand (kg) Water (kg) Superplast. (kg) PP fiber (vol.%) Recycled aggr. ratio (%) 318 469 576 – – 835 169 6.35 0 0 318 – 576 454 – 835 169 6.67 0 25 318 469 – – 507 835 169 6.35 0 30 318 – – 454 507 835 169 7.00 0 55 318 462 568 – – 823 169 6.35 1.0 0 318 – 567 448 – 823 169 6.67 1.0 25 318 462 – – 500 823 169 6.35 1.0 30 318 – – 448 500 823 169 7.00 1.0 55 318 459 564 – – 818 169 6.35 1.5 0 318 – 563 445 – 818 169 6.67 1.5 25 318 459 – – 496 818 169 6.35 1.5 30 318 – – 445 496 818 169 7.00 1.5 55 Fig. 4. (a) Embedded fiber, (b) pull-out testing apparatus. Table 6 Fresh concrete properties. Concrete Type N-A R1-A R2-A R12-A N-B R1-B R2-B R12-B N-C R1-C R2-C R12-C Workability (slump cm) Air void cont. (%) Fresh unit weight (kg/m3) 14 1.1 2433 13 1.2 2375 12 1.1 2346 11 1.6 2302 12 1.3 2427 11 1.4 2333 11 1.2 2351 10 1.5 2276 12 1.3 2442 10 1.4 2321 11 1.4 2314 10 1.7 2248 gate type has the highest importance with 64%. Fiber’s calculated F value (i.e., 1.0298) is less than the critical F value (i.e., 5.14). This result states that the Fiber’s F value is statistically non-significant and it also means that fiber has no effect on compressive strength. Due to the statistically non-significant fiber effect on compressive strength, error contribution is 36%. Furthermore it can be seen that recycled aggregates has lower effect on compressive strengths of fiber reinforced concrete series than on the samples with no fiber, according to experimental results. In comparison with reference specimens (N-B and N-C) that have volume of 1% and 1.5% fiber content, compressive strength decreases in the range of 17–20% for specimens with the highest replacement ratio and same fiber content ratio (R12-B and R12-C). When replacement ratio is increased, splitting tensile strength decreases. Nevertheless, especially for A and B series, there is no significant effect observed on tensile strength when aggregate RATP2 is used. This circumstance can be explained as the adherence surface between cement paste and RCA being rough [4]. However, splitting tensile strength increases generally when polypropylene fiber is added to the mixtures. Among samples with recycled concrete aggregate, highest splitting tensile strength was obtained from R2-B with 4.62 MPa. Although this corresponds to a decrease of 9.2% in comparison with reference N-B (i.e., 5.09 MPa), it is 13.3% greater than N-A (i.e., 4.08 MPa). The results of ANOVA for splitting tensile strength can be found on Table 10. Considering the ANOVA analysis, it is seen that fiber has an importance (i.e., 21%) as well as aggregate type (i.e., 56%). It is observed that splitting tensile strength increases when compressive strength increases. Changes in especially low fiber containing samples are more reasonable, therefore high correlation coefficients were obtained (Fig. 6). Low correlation coefficients, especially obtained from C series, could be related with faulty mixing arising from high fiber content. The same circumstance is seen in Fig. 7 for flexural tensile strength and this (Figs. 6 and 7) can be explained with the uncertainty of compressive strength of concretes due to fiber addition. While crucial changes in compressive strength were not observed when fiber content is increased, splitting and flexural tensile strengths tended to increase. Besides, it is evidently seen that scattering of flexural tensile strength results of C series affects Fig. 7’s lowest correlation coefficient (i.e., 0.57) as well as highest standard deviation value (i.e., 1.069 MPa) (Table 8) for the flexural tensile strength results. As widely-used two 625 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 Compressive Strength [MPa] 55 N-A B series 50 Series C series R2-A R2-B N-B 45 40 Table 8 Mean values and standard deviations of compressive, splitting and flexural tensile strengths. A series N-C R1-A R2-C R1-B R1-C 35 R12-B R12-C R12-A A B C Compressive strength (MPa) Splitting tensile strength (MPa) Flexural tensile strength (MPa) Mean value Standard deviation Mean value Standard deviation Mean value Standard deviation 42.5 40.5 39.2 7.578 3.233 2.691 3.71 4.24 4.58 0.434 0.766 0.615 4.60 5.31 5.38 0.646 0.846 1.069 30 10 15 20 25 30 35 RCA Rao [%] 40 45 50 55 60 6 Fig. 5. Changes of compressive strengths with RCA. experiments that is to determine concrete’s tensile strength, there is a parallel manner between splitting tensile strength and flexural tensile strength as seen in Fig. 8. However, with flexural tensile strength experiment, approximately 17% higher results are obtained in comparison to the results of splitting tensile strength experiment. When flexural tensile strength results are considered, it is observed that RCA has a negative effect on the flexural tensile strength as in the previous studies [15–17]. R1-A, R2-A and R12-A have 23.7%, 16%, 31.3% loss of flexural tensile strength respectively, when compared with reference (N-A). However, flexural strengths of samples with RCA increases due to the addition of polypropylene (PP) fiber. In the case of PP fiber ratio is increased from 1% to 1.5%, no significant difference is observed on neither flexural tensile strength nor splitting tensile strength. Besides, the result of ANOVA analysis which is given in Table 11, indicates that the aggregate type has a higher level of importance than the fiber has, on flexural tensile strength. Change of pull-out peak loads with replacement ratio can be seen in Fig. 9. Water demand and pores caused by RCA were found influential on pull-out peak loads. It is clearly observed that pullout peak loads decrease when the replacement ratio increases, due to the low mechanical performance of recycled aggregate concretes. This is also confirmed by Figs. 10 and 11 which show the correlation between pull-out peak loads and tensile strengths of concretes. Considering both figures that were drawn for B and C series, pull-out peak loads are generally in accord with both splitting and flexural tensile strength. High pull-out peak load reflect the good bond between matrix and PP fiber. During the experiment, it was observed that fibers usually debonded (a) and rarely severed (b), as seen in Fig. 12. When RCA replacement increased, void content increased and debondings were observed on fibers, related also with fiber geometry. Spling Tensile Strength [MPa] 5 5 4 3 2 A series y = 0,049x + 1,623 1 R² = 0,7206 B series y = 0,1909x - 3,4879 R² = 0,8324 C series y = 0,1381x - 0,8304 R² = 0,4966 0 30 35 40 45 Compressive Strength [MPa] 50 55 Fig. 6. Correlation of compressive strength and splitting tensile strength. 8 Flexural Tensile Strength [MPa] 0 7 6 5 4 3 2 1 0 A series y = 0,0827x + 1,0863 R² = 0,9427 B series y = 0,1958x - 2,6253 R² = 0,7282 C series y = 0,2566x - 4,6725 R² = 0,5696 30 35 40 45 Compressive Strength [MPa] 50 55 Fig. 7. Correlation of compressive strength and flexural tensile strength. PP fiber content increases fracture toughness by means of its positive effect on energy absorption capacity. The toughness values of the specimens are shown in Table 7 while load–deflection Table 7 Hardened concrete properties. N-A R1-A R2-A R12-A N-B R1-B R2-B R12-B N-C R1-C R2-C R12-C Compressive strength (MPa) [St. dev.] Splitting tensile strength (MPa) [St. dev.] Flexural strength (MPa) [St. dev.] Toughness (N m) Ultrasonic pulse velocity (km/s) Unit weight (kg/m3) Static modulus (MPa) Dynamic modulus (MPa) Water penetration (mm) 53.5 40.8 43.3 32.3 44.4 40.1 41.9 35.6 42.1 40.0 39.8 34.8 4.08 3.62 4.09 3.03 5.09 4.24 4.62 3.02 5.60 4.49 4.27 3.97 5.59 4.27 4.69 3.84 6.60 5.04 5.35 4.25 6.96 4.98 5.56 4.01 1.62 0.72 1.51 0.58 20.08 25.71 28.83 16.14 42.59 30.14 27.70 21.55 4.85 4.62 4.75 4.32 4.90 4.64 4.57 4.29 4.87 4.62 4.57 4.21 2391 2354 2305 2287 2412 2317 2337 2254 2426 2305 2297 2221 35,020 25,293 26,995 18,805 33,120 19,490 25,343 15,919 31,672 14,126 23,860 13,306 57,250 51,106 52,943 43,562 59,078 50,945 49,678 42,197 58,665 50,212 48,824 40,185 4 6 4 10 4 11 6 12 4 11 5 11 [1.21] [1.77] [0.94] [0.47] [1.01] [2.15] [1.33] [1.88] [1.06] [0.90] [2.02] [1.76] [0.11] [0.12] [0.21] [0.08] [0.22] [0.17] [0.18] [0.21] [0.04] [0.11] [0.19] [0.16] [0.15] [0.11] [0.10] [0.13] [0.12] [0.17] [0.08] [0.11] [0.11] [0.10] [0.20] [0.15] 626 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 Flexural Tensile Strength [MPa] 8 7 6 5 4 3 A series y = 1,2333x + 0,0262 R² = 0,7006 B series y = 1,004x + 1,0444 R² = 0,8382 C series y = 1,5949x - 1,9263 R² = 0,8455 2 1 0 3 4 5 Experimental Split Tensile Strength [MPa] 6 Fig. 8. Correlation of splitting and flexural tensile strength. Table 9 Results of ANOVA for compressive strength. Parameters Degrees of freedom (df) Sum of square (SS) Variance (V) F Contribution (%) Aggregate Fiber Error Total 3 2 6 11 237.0476 21.7606 63.3945 322.2027 79.0159 10.8803 10.5658 7.4785 1.0298 63.7333 0.1952 36.0715 100 Table 10 Results of ANOVA for splitting tensile strength. Parameters Degrees of freedom (df) Sum of square (SS) Variance (V) F Contribution (%) Aggregate Fiber Error Total 3 2 6 11 3.8444 1.5578 0.7723 6.1745 1.2815 0.7789 0.1287 9.9561 6.0516 56.0089 21.0608 22.9303 100 Table 11 Results of ANOVA for flexural tensile strength. Parameters Degrees of freedom (df) Sum of square (SS) Variance (V) F Contribution (%) Aggregate Fiber Error Total 3 2 6 11 8.7185 1.4953 0.3804 10.5941 2.9062 0.7476 0.0634 45.8442 11.7936 80.5007 12.9172 6.5821 100 graphs are given in Figs. 13–16. Considering all specimens’ load– deflection graphs, a sudden fall is seen after the peak points. Cracks appeared on specimens when there was a sudden fall of loads. Debondings or failures associated with the crack development were rarely observed on fibers which are situated on the bottom section of specimens. When deflection increases, PP fibers at the bottom zone of the specimens provide that specimens to keep on taking load bridging the crack. Graph of experiment results of water penetration depth under pressure is demonstrated in Fig. 17. While RCA specimens which are composed of RATP2 almost have the same penetration depth as the reference (i.e., 4 mm), specimens including RATP1 have higher depth of penetration. However, the maximum water penetration depth is observed in the specimens including both RATP1 and RATP2. The similar result, which pointed out the importance of aggregate type rather than the incorporation ratio, had been observed on compressive strength as well (Fig. 5). In this respect, similarity of Figs. 5 and 17 attract attention and they together confirm an inversely proportional relationship which can be seen in Fig. 18 which shows the correlation between compressive strength and water penetration depth. Void content ratio and pore structure of specimen influences specimen’s permeability. When the void content ratio increases in the specimens, water penetration depth increases, compressive strength decreases and same results are found by [20]. Besides, any crucial impact of polypropylene fibers on penetration depths is generally not established. As seen in Fig. 19, fibers left concrete’s matrix phase by plucking (a), debonding (b) and rarely severing (c). It is observed that ultrasonic pulse velocity (UPV) decreases with RCA replacement (Table 7), as confirmed by [43,44]. In comparison with reference N-A, reductions of 4.8%, 2.1%, 10.9% is observed respectively for R1-A, R2-A and R12-A. Considering all the specimens of the present study, UPV test results stay in the range between 4.21 and 4.90 km/s. This range states that the strength of concrete might be good enough [45,46]. RCA and PP fiber content ratio influence the UPV results negatively. Porous structure of RCA and fibers’ effect on gradation may reduce UPV. In the previous studies [47,48], it is stated that there is a correlation between UPV and compressive strength, related with pore structure of concrete. In this study, the correlation of the two parameters is obtained as well and it is observed that compressive strength increases when UPV increases (Fig. 20). Additionally, UPV shows an increase while water penetration depth decreases and unit weight increase depends on void content and pore structure of concrete (Figs. 21 and 22). Due to the RCA’s specific gravity being lower than natural aggregate (Table 3), concrete which comprises of RCA has a lower unit weight than conventional concrete. As expected by authors in Pull out - Replacement Rao Pull-out Peak Loads [N] 350 300 B series 250 C Series 200 150 100 50 0 0 25 30 Replacement Ratio [%] Fig. 9. Change of pull out peak loads with RCA replacement ratio. 55 627 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 B Series Pull Out Peak Loads [N] 300 250 y = 66.171x - 81.754 R² = 0.6921 200 150 y = 62.685x - 133.64 R² = 0.7542 100 50 0 0.00 1.00 2.00 3.00 Flexural Tensile Strength [MPa] 4.00 5.00 6.00 7.00 Spling Tensile Strength [MPa] Fig. 10. Correlation of pull out peak loads and tensile strengths. C Series Pull Out Peak Loads [N] 350 300 y = 95.376x - 234.69 R² = 0.8943 250 200 y = 50.404x - 68.768 R² = 0.7523 150 100 50 0 0.00 1.00 2.00 3.00 Flexural Tensile Strength [MPa] 4.00 5.00 6.00 7.00 8.00 Spling Tensile Strength [MPa] Fig. 11. Correlation of pull out peak loads and tensile strengths. 6000 5000 Load [N] 4000 3000 2000 N-A N-B 1000 N-C 0 0 2 4 6 8 10 Midspan Deflecon [mm] 12 14 Fig. 13. Load–deflection curves of N-A, N-B and N-C. Fig. 12. Debonding and severing during pull-out experiment. this study, unit weight decreases when replacement ratio increases. Losses of 1.6%, 3.6%, 4.4% were observed on unit weights for R1-A, R2-A and R12-A respectively. The lowest unit weight value which is 2221 kg/m3 is obtained from R12-C that has maximum substitution ratio and polypropylene fiber content, while the reference sample (N-A) has a unit weight of 2391 kg/m3. Static modulus of elasticity (Es) decreases with the increase of ratio of RCA replacement in specimens with fiber or without fiber, as confirmed by [12,13,17]. Decrease is observed up to 46% without 628 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 16 Water Penetraon Depth [mm] 4000 3500 Load [N] 3000 2500 2000 1500 R1-A 1000 R1-B 500 R1-C A series y = -0,2715x + 17,445 R² = 0,6513 B series y = -0,6637x + 34,969 R² = 0,5787 C series y = -0,572x + 30,407 R² = 0,3043 14 12 10 8 6 4 2 0 0 0 2 4 6 8 10 12 14 30 35 Midspan Deflecon [mm] 40 45 Compressive Strength [MPa] 50 55 Fig. 18. Correlation of compressive strength and water penetration depth. Fig. 14. Load–deflection curves of R1-A, R1-B and R1-C. 4500 4000 Load [N] 3500 3000 2500 2000 1500 R2-A 1000 R2-B 500 R2-C 0 0 2 4 6 8 10 Midspan Deflecon [mm] 12 14 Fig. 15. Load–deflection curves of R2-A, R2-B and R2-C. Fig. 19. Types of fibers leave matrix phase after splitting. 3500 60 3000 2000 1500 R12-A 1000 R12-B 500 R12-C 0 0 2 4 6 8 10 Midspan Deflecon [mm] 12 A series 55 Compressive Strength [MPa] Load [N] 2500 50 y = 36,207x - 125,28 R² = 0,8822 B series y = 14,029x - 24,018 R² = 0,7281 y = 11,323x - 12,55 R² = 0,7164 C series 45 40 35 30 25 14 20 4.0 4.5 Ultrasonic Pulse Velocity [km/s] 5.0 Fig. 16. Load–deflection curves of R12-A, R12-B and R12-C. 14 14 R1-B R1-C 12 R12-B R12-C R12-A 10 8 R1-A 6 A series 4 2 B series R2-B R2-C R2-A N-A N-B N-C C series 0 Water Penetraon Depth [mm] Water Penetraon Depth [mm] Fig. 20. Correlation of compressive strength and UPV. 12 10 8 6 4 A series y = -11,741x + 60,311 B series y = -9,8684x + 53,473 C series y = -9,5214x + 51,495 2 0 0 5 10 15 20 25 30 35 40 45 RCA Rao [%] Fig. 17. Water penetration depths under pressure. 50 55 60 4.0 R² = 0,8198 R² = 0,4733 R² = 0,4712 4.5 Ultrasonic Pulse Velocity [km/s] Fig. 21. Correlation of water penetration depth and UPV. 5. 0 629 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630 0.65 2500 0.60 0.55 2300 0.50 Es/Ed Unit Weight [kg/m3] 2400 2200 2100 A series y = 146,6x + 1654,9 R² = 0,4951 B series y = 237,69x + 1236,8 R² = 0,888 C series y = 300,22x + 940,93 R² = 0,9168 0.45 0.40 N- 0.35 R1- 0.30 2000 4 .0 4.5 5.0 R2R12- 0.25 Ultrasonic Pulse Velocity [km/s] A Fig. 22. Correlation of unit weight and UPV. Stac Modulus of Elascity [MPa] that there is a directly proportional correlation. Additionally, more reasonable changes are observed (high correlation coefficients) with less content of fiber. This circumstance can be explained with the uncertainty of behavior of compressive strength during the increasing of fiber content. Despite the fact that there are no major changes observed in void content of the samples, high deformation is still gained. Therefore, samples’ Es/Ed ratios decrease with the increase of fiber content, as seen in Fig. 25. 35000 30000 25000 20000 15000 10000 A series y = 751,35x - 5379,2 R² = 0,9541 5000 B series y = 1510,4x - 37716 y = 1530,8x - 39230 R² = 0,7255 R² = 0,4009 C series 30 35 40 45 Compressive Strength [MPa] 50 4. Conclusions 55 Fig. 23. Correlation of compressive strength and static modulus of elasticity. Dynamic Modulus of Elascity [MPa] 70000 60000 50000 40000 30000 20000 10000 0 10000 A series y = 0,8048x + 29866 R² = 0,9044 B series y = 0,8291x + 31024 R² = 0,7959 C series y = 0,6974x + 35007 R² = 0,6539 15000 C Fig. 25. Es/Ed ratio. 40000 0 B 20000 25000 30000 Stac Modulus of Elascity [MPa] 35000 40000 Fig. 24. Correlation of static and dynamic modulus of elasticity. the fiber content when replacement ratio reaches its highest level. Loss of Es is observed to be up to 62% with the highest RCA and PP fiber content when compared with reference (N-A). Improving the deformation capability of fibers and porous structure of RCA may influence decrease of the modulus of elasticity [13]. According to many widely-used standards [49–51], modulus of elasticity is also calculated depending on compressive strength. When observing the relationship between compressive strength and modulus of elasticity (Fig. 23), it is seen that experimentally calculated Es values of samples without fiber is directly proportional with compressive strength and considerably high correlation value is obtained. However it is observed that the correlation values of samples with fiber tend to decrease. Porous structure of RCA influences dynamic modulus of elasticity (Ed) as well by decreasing UPV. In the case of the comparison between reference (N-A) and RAC which has volume of 1.5% fiber content (R12-C), Ed decreases by 30%. When observing the relationship between Ed and Es (Fig. 24), it is seen Disposal of C&D wastes is becoming more of an issue nowadays. In this study, usability of recycled aggregates derived from C&D wastes was researched with or without fiber. Eventually, the followings are the conclusions drawn from this study:  Although the workability of concrete decreases due to high water absorption tendency of RCA, it is possible to attain desired workability with proper superplasticizer admixture. Due to the fact that fiber content influencing gradation negatively, fiber reinforced concrete series exhibit heterogeneous behavior. Therefore, uncertainties may be observed on changes of some parameters such as compressive strength.  Compressive strength decreases with RCA replacement. In spite of this, all series appeared in structural concrete class even when maximum replacement ratio is utilized. Concrete having a compressive strength of 32–43 MPa is able to be manufactured using recycled aggregate which is derived from concrete having compressive strength of 8–10 MPa. Besides, there is no significant influence observed on the compressive strength arisen by polypropylene fibers, as also seen on the result of ANOVA analysis (0.1952%).  Flexural and splitting tensile strengths show reduction when RCA is used. However, both flexural tensile strength and splitting tensile strength increase with fiber addition on all series. Nevertheless, workability and placing problems may occur depending on fiber content of concrete. When these effects are taken into account, optimum fiber content in this study is recommended as 1%.  Pull-out peak loads decrease when the replacement ratio increases. Besides, pull-out peak loads are in accord with both splitting and flexural tensile strength. High pull-out peak load reflect the good bond between matrix and PP fiber. During the experiment, it was observed that fibers usually debonded and rarely severed.  Incorporation ratio might increase water penetration depth, but the type of RCA has a more significant effect on this parameter. Even if fine grained recycled aggregates such as RATP1 has lower incorporation ratio, it may influence the permeability of concrete more negatively. 630 K.R. Akça et al. / Construction and Building Materials 98 (2015) 620–630  By increasing the RCA replacement and fiber content, modulus of elasticity decreases. However, using RCA is the dominant factor of the reductions rather than the fiber content.  Both Es and Ed tend to decrease when RCA and PP fiber is used. However, more crucial decreasing is observed for Es when compared with the decreases of Ed. Therefore, Es/Ed ratio decreases when RCA and PP fiber content is increased.  Based on ANOVA analysis, efficient factor on compressive strength, splitting tensile strength and flexural tensile strength were found as aggregate type. 5. Recommendations In this study, it is seen that it is possible to manufacture structural concrete with RCA derived from C&D wastes that occur during the construction of the urban transformation project. However, there is a notable risk arising from the RCA to be used. RCA has a heterogeneous structure and its strength depends on the material quality of demolished structures. Thus, it is more suitable to utilize RCA at field concretes, precast elements and road concretes that have low structural risk factor. Besides, fiber usage is seen to have increased for these kinds of applications nowadays. This study will encourage the popularity of the usage of fiber reinforced recycled aggregate concretes. References [1] Ministry of Environment and Urban Planning. 50 Questions and 50 Answers in Urban Transformation, Ankara, 2013. [2] N.D. Oikonomou, Recycled concrete aggregates, Cement Concr. Compos. 27 (2005) 315–318. [3] C.S. Poon, Management of construction and demolition waste, Waste Manage. 27 (2007) 159–160. [4] D. Matias, J. de Brito, A. Rosa, D. Pedro, Mechanical properties of concrete produced with recycled coarse aggregates-influence of the use of superplasticizers, Constr. Build. Mater. 44 (2013) 101–109. [5] M. Rakshvir, S.V. Barai, Studies on recycled aggregates-based concrete, Waste Manage. Res. 24 (2006) 225–233. [6] N. Kartam, N. Al-Mutairi, I. Al-Ghusain, J. Al-Humoud, Environmental management of construction and demolition waste in Kuwait, Waste Manage. 24 (2004) 1049–1059. [7] Gobierno de Espana Ministerio de Fomento EHE-08 Code on Structural Concrete. <http://www.fomento.gob.es/MFOM.CP.Web/handlers/pdfhandler. ashx?idpub=BNW001> 29 September 2013. [8] The Law on Control of the Earthwork and Demolition Wastes <http://www. resmigazete.gov.tr/eskiler/2004/03/20040318.htm#25>. [9] G. Durmusß, O. Sßimsßek, M. Dayı, The effects of coarse recycled concrete aggregates on concrete properties, J. Fac. Eng. Archit. Gazi Univ. 24 (2009) 183–189. [10] J. Xiao, W. Li, Y. Fan, X. Huang, An overview of study on recycled aggregate concrete in China 1996–2011, Constr. Build. Mater. 31 (2012) 364–383. [11] M.C. Limbachiya, T. Leelawat, R.K. Dhir, Use of recycled concrete aggregate in high-strength concrete, Mater. Struct. 33 (2000) 574–580. [12] K. Eguchi, K. Teranishi, A. Nakagome, H. Kishimoto, K. Shinozaki, M. Narikawa, Application of recycled coarse aggregate by mixture to concrete construction, Constr. Build. Mater. 21 (2007) 1542–1551. [13] C. Thomas, J. Setien, J.A. Polanco, P. Alaejos, M.S. de Juan, Durability of recycled aggregate concrete, Constr. Build. Mater. 40 (2013) 1054–1065. [14] Y.N. Sheen, H.Y. Wang, Y.P. Juang, D.H. Le, Assessment on the engineering properties of ready-mixed concrete using recycled aggregates, Constr. Build. Mater. 45 (2013) 298–305. [15] A. Ajdukiewicz, A. Kliszczewicz, Influence of recycled aggregates on mechanical properties of hs/hpc, Cement Concr. Compos. 24 (2002) 269–279. _ Topçu, S. Sßengel, Properties of concretes produced with waste concrete [16] I.B. aggregate, Cem. Concr. Res. 34 (2004) 1307–1312. [17] C. Hoffmann, S. Schubert, A. Leemann, M. Motavalli, Recycled concrete and mixed rubble as aggregates: influence of variations in composition on the concrete properties and their use as structural material, Constr. Build. Mater. 35 (2012) 701–709. [18] E. Vázquez, P. Alaejos, M. Sanchez, F. Aleza, M. Barra, M. Buron, Use of recycled aggregate for the production of structural concrete (in Spanish). Monograph M-11 ACHE, Commission 2, work group 2/5 recycled concrete, Madrid, Spain, 2006. [19] L. Evangelista, J. de Brito, Mechanical behaviour of concrete made with fine recycled concrete aggregates, Cement Concr. Compos. 29 (2007) 397–401. [20] C.J. Zega, A.A. Di Maio, Use of recycled fine aggregate in concretes with durable requirements, Waste Manage. 31 (2011) 2336–2340. _ [21] L. Mazılıgüney, I.Ö. Yaman, F. Azılı, In-situ concrete compressive strength of residential, public and military structures, in: 8th International Congress on Advances in Civil Engineering. September 15–17, Famagusta, Cyprus, 2008. [22] ASTM C33/C33M-13, Standard Specification for Concrete Aggregates, 2013. [23] TS 802, Design Concrete Mixes, 2009. [24] ACI 211.1-91, Standard Practice for Selecting Proportions for Normal, Heavyweight, and Mass Concrete, 2009. [25] L. Ferreira, J. de Brito, M. Barra, Influence of the pre-saturation of recycled coarse aggregates on concrete properties, Mag. Concr. Res. 63 (8) (2011) 617–627. [26] EN 12350-2, Testing Fresh Concrete – Part 2, Slump-Test, 2009. [27] EN 12350-6, Testing Fresh Concrete – Part 6, Density, 2009. [28] EN 12350-7, Testing Fresh Concrete – Part 7, Air Content – Pressure Methods, 2009. [29] EN 12390-2, Testing Hardened Concrete – Part 2: Making and Curing Specimens for Strength Tests, 2009. [30] EN 12390-7, Testing Hardened Concrete – Part 7: Density of Hardened Concrete, 2009. [31] ASTM C597-09, Standard Test Method for Pulse Velocity Through Concrete, 2009. [32] EN 12390-8, Testing Hardened Concrete – Part 8: Depth of Penetration of Water Under Pressure, 2009. [33] A. Beglarigale, H. Yazıcı, Pull-out behavior of steel fiber embedded in flowable RPC and ordinary mortar, Constr. Build. Mater. 75 (2015) 255–265. [34] Q. He, C. Liu, X. Yu, Improving steel fiber reinforced concrete pull-out strength with nanoscale iron oxide coating, Constr. Build. Mater. 79 (2015) 311–317. [35] EN 12390-3, Testing Hardened Concrete – Part 3: Compressive Strength of Test Specimens, 2009. [36] EN 12390-6, Testing Hardened Concrete – Part 6: Tensile Splitting Strength of Test Specimens, 2009. [37] EN 12390-5, Testing Hardened Concrete – Part 5: Flexural Strength of Test Specimens, 2009. [38] TS 3502, Test Method for Static Modulus of Elasticity and Poisson’s Ratio of Concrete in Compression, 1981. [39] B. Baradan, S. Türkel, H. Yazıcı, H. Ün, H. Yiğiter, B. Felekoğlu, et al., Beton, _ University of Dokuz Eylül-Faculty of Engineering Publishing, Izmir, 2012. [40] J. Yang, Q. Du, Y. Bao, Concrete with recycled concrete aggregate and crushed clay bricks, Constr. Build. Mater. 25 (2011) 1935–1945. [41] T. Mukai, M. Kikuchi, Studies on utilization of recycled concrete for structural members, Summ. Tech. Paper Annu. Meet. Arch. Inst. Jpn. 85 (1978) 88. [42] A. Katz, Properties of concrete made with recycled aggregate from partially hydrated old concrete, Cem. Concr. Res. 33 (2003) 703–711. [43] S.C. Kou, C.S. Poon, F. Agrela, Comparisons of natural and recycled aggregate concretes prepared with the addition of different mineral admixtures, Cement Concr. Compos. 33 (8) (2011) 788–795. [44] G. Andreu, E. Miren, Experimental analysis of properties of high performance recycled aggregate concrete, Constr. Build. Mater. 52 (2014) 227–235. [45] IS 13311-1, Method of Non-Destructive Testing of Concrete – Part 1: Ultrasonic Pulse Velocity, 1992. [46] M.S. Güner, Malzeme bilimi-yapı malzemesi ve beton teknolojisi, 16th ed., _ Aktif Yayınevi, Istanbul, 1999. [47] W.H. Kwan, M. Ramli, K.J. Kam, M.Z. Sulieman, Influence of the amount of recycled coarse aggregate in concrete design and durability properties, Constr. Build. Mater. 26 (2012) 565–573. [48] Y.T. Lee, S.U. Hong, H.S. Jang, S.K. Baek, Y.S. Cho, Compressive strength estimation of recycled coarse aggregate concrete using ultrasonic pulse velocity, Appl. Mech. Mater. 147 (2012) 288–292. [49] TS 500, Requirements for Design and Construction of Reinforced Concrete Structures, 2000. [50] ACI 318, Building Code Requirements for Structural Concrete and Commentary, 2011. [51] EN 1992-1-1, Eurocode 2: Design of Concrete Structures – Part 1-1: General Rules and Rules for Buildings, 2004.