- Email: [email protected]
The effect of pre-setting pressure applied flexural strength and fracture toughness of reactive powder concrete during the setting phase
Construction and Building Materials 26 (2012) 459–465
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
The effect of pre-setting pressure applied flexural strength and fracture toughness of reactive powder concrete during the setting phase a,⇑ _ Metin Ipek , Kemalettin Yilmaz b, Mucteba Uysal b a b
Sakarya University, Technical Education Faculty, Construction Education Department, 54187 Sakarya, Turkey Sakarya University, Engineering Faculty, Civil Engineering Department, 54187 Sakarya, Turkey
a r t i c l e
i n f o
Article history: Received 19 January 2011 Received in revised form 9 June 2011 Accepted 18 June 2011 Available online 13 July 2011 Keywords: Reactive powder concrete Pre-setting pressure Fibre Flexural strength Toughness
a b s t r a c t In this study, the effect of pre-setting pressure on reactive powder concrete (RPC) was examined. The fresh (plastic) concrete samples were filled in a specified beam mould and six different pre-setting pressure values (control-without pressure, 5, 10, 15, 20, 25 MPa) were applied. Flexural strength was improved by 34% with very low amounts (5 MPa) of pre-setting pressure. Also, toughness was increased more than thrice. In this study, the maximum flexural strength of 36.4 MPa and toughness of 116.960 N m were obtained under 25 MPa pre-setting pressure. Volume of sample was decreased 7.9% by pre-setting pressure. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction In recent years, high-performance concrete is being assessed not only in terms of high strength, but also in terms of specifications such as strength and ductility. Although high-strength concretes are favourable also because they are impervious, a significant issue with these concretes which emerges as their strength increases has been the problem of fragility. Although the capacity of axial deformation increases in high-strength concretes, once the peak point is reached, stress relaxation is instantaneous and the fragility leads to crashes. Since the beginning of 1960s, in order to resolve the issue of fragility in concrete, it has been a common practice to work fibre into it to ensure its ductility capacity. In order to improve the mechanical features such as tensile strength, resistance to cracks, endurance to wearing and shocks and toughness of plain concrete, it is mixed with steel, glass and polypropylene fibres. The fibres used in concrete increase tensile and flexural strength and also decrease shrinkage cracks [1,2]. Unlike the high performance concrete, reactive powder concrete (RPC) has been produced by working fibre into concrete, replacing aggregates with microgranules, arranging granule distribution so as to form minimum void in concrete, increasing pozzolanic activity and exposing the materials to different production and treatment. Although it was discovered in the first half of
⇑ Corresponding author. Tel.: +90 2642956476; fax: +90 2642956424. _ E-mail address: [email protected] (M. Ipek). 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.06.045
1990s, the RPC has started to be developed and used all over the world [3–5]. RPC is a cement-based ultra high performance concrete with superior mechanical and physical specifications and excellent ductility and durability characteristics. Compression strength of RPC is 150–800 MPa, while its tensile strength changes between 25 and 150 MPa. Its fracture energy, on the other hand, changes between 1200 and 40,000 J/m2. The durability properties of RPC are better than current high performance concrete in terms of magnitude [3–12]. Fine powders such as crushed quartz (100–600 lm) are used instead of coarse aggregate in order to make RPC more homogeneous. The water/cement ratio of RPC is reduced to less than 0.20 by using superplasticizers (SP). Silica fume addition to RPC reduces the total pore volume of the cement paste and the average diameter of the pores. Also, researcher’s reports that application of different heat cure processes improves the mechanical properties of RPC to a great extent after the application of 50 MPa presetting pressure to fresh RPC [3]. Besides, pressure application to fresh RPC during the setting phase of 6–12 h can eliminate some amount of pores caused by autogenous shrinkage. Pre-setting pressure applied during setting stage causes microcracks in the sample. These microcracks are improved as a result of expansion of aggregate after discharging the applied pressure. Pre-setting pressure trials were carried out also in the previous studies [3,6,13,14]. However, no study has been encountered so far where the effect of pre-setting pressure on beam samples are examined. The effects of pre-setting pressure on compressive
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
460
strength have been examined by some researchers. In the study carried out by Dugat et al. [6], they applied a pre-setting pressure of 60 MPa onto the RPB800 concrete during the setting and hardening stages. Also, they applied 90 °C hot water and then heat curing process at 250 °C onto RPB800 samples for 4 days. Moreover, they examined the effect of the fibre content on the fracture energy. In their experiments, they increased the fracture energy up to 40,000 J/m2 and found out the optimum amount of fibre content to be between 2% and 3%. Another study has been carried out by Teichman and Schmidt [13]. In that study, they applied 50 MPa constant pre-setting pressure onto the RPC samples while they are fresh and achieved strength of 487 MPa at the end of the experiment. The last study has been the one performed by us [14]. In this study, the effects of pre-setting pressure on compression strength, the elastic modulus and unit weight have been examined in details. For that purpose, six different pre-setting pressure amounts of control-without pressure, 25, 50, 75, 100, 125 MPa have been applied onto RPC and the compressive strength characteristics of the samples have been examined. Eventually, as the pre-setting pressure increased, compression strength and elastic modulus increased and the unit weight decreased. When compared to control-without pressure, 100 MPa pre-setting pressured had twice better compressive strength. Generally, concrete properties are influenced by placement and compression factor. It is well known that reducing voids positively affect the mechanical and durability properties of hardened concrete. Concrete must be placed with minimum voids [15]. During the application of vibration to RPC, as the samples are examined, the increment in air volume due to air-entrainer property of SP is clearly observed. The RPC workability is similar to self-compacting
Fig. 1. Granule structure of the quartz powder [14].
Table 1 Properties of cement and silica fume. Component
CaO SiO2 C Al2O3 Fe2O3 MgO K2O Cl Na2O P2O5 SO3 H2O Fever loss pH value C3S C2S C3A C4AF Silicate modulus Alumina modulus Hydraulic modulus Total alkaline
Chemical composition (%) Cement
Silica fume
64.47 20.09 – 5.01 2.73 1.72 0.66 0.01 0.21 – 3.03 – 2.11 – 60.7 11.8 8.6 8.3 2.6 1.8 2.3 0.58
0.50 96 1.50 0.70 0.25 0.60 0.85 0.10 0.25 0.10 0.50 0.80 1.50 5.0–8.0 – – – – – – – –
Physical properties Cement
Silica fume
Blaine specify surface Unit volume weight Specific gravity Initial setting time Final setting time Volume expansion Genlesßmesi
5162 cm2/g – 3.14 154 min 191 min 0.8 mm
200,000 cm2/g 0.650 g/cm3 2.26 – – –
2 days 7 days 28 days
Compressive strength of cement (MPa) 39.8 MPa – 54.2 MPa – 61.8 MPa –
Fig. 2. Particle size analysis of granule materials [14].
concrete. In this study, T50 value of the mixture of RPC measured 4 s in slump-flow test. Workability can be brought to a desired level due to high ratio of SP whereas RPC has low water cement ratio. However, water and compressed air of closed pockets were found in the RPC. It was observed that the air bubbles in concrete caused swelling overflow from mould or forming of bubbles at the surface. The voids were formed between fibre and the concrete interface. In such condition, adherence between fibre and concrete interface was affected negatively and consequently the strength was reduced. Because the steel fibres used in RPC is hookless and 0.16 mm in diameter and in order to make these fibres achieve a favourable adherence with concrete, it is mandatory to cover it with a finegrained paste. This necessity stems from the fact that all the materials which compose RPC are at micron level. However, there are voids in the fine-grained paste as compared to normal concrete, even in the least. By reducing the voids to a minimum, there will be achieved more adherence between the fine-grained paste and the micron-sized hookless fibre. In order to increase the adherence between the paste and the fibres and to minimize the voids in the paste, it was decided to apply pre-setting pressure onto the RPC in fresh form. In this study, the effects of pre-setting pressure on RPC’s, unit weight, flexural strength, fracture parameters and cost were investigated. For this purpose, a special moulding system was designed and the mixture in fresh form was filled in these moulds where it was applied pre-setting pressure of six different
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
461
Table 2 Compositions of mixtures [14]. Material
Cement
Silica fume
Q. powder (100-0 lm)
Q. sand (100–300 lm)
Q. sand (300–600 lm)
Water
SP
Air
Steel fibre 4% of all volume)
Weight (kg/m3) 7 days compr. strength 28 days compr. strength
864 145.8 MPa 170.29 MPa
259
346
248
248
216
26
1.8%
287
Fig. 3. The technical drawing and view from above of the specially designed pre-setting pressure mould.
Fig. 4. (a) Filling the fresh RPC into the mould, (b) placing the piston in rectangular prism form, (c) performing pre-setting pressure under the concrete press, (d) tightening the screws which prevent the pressure within the mould from bleeding off.
amounts (control-without pressure, 5, 10, 15, 20, 25 MPa) and the samples were kept under the pressure until they got hardened. At the end of the 28-day period, the samples were tested in terms of unit weight and flexural test. 2. Experimental study 2.1. Materials High performance cement of PC 52.5 CEMIR type was preferred for this study. For RPC, pozzolanic material is needed, which fills voids of microparticulates in binder paste and which contributes to the strength by producing secondary hy-
drates by pozzolanic reaction with the lime resulting from primary hydration [3,12,14,16–19]. Undensified silica fume (SF) provided from Elkem Company in Norway was used in this study. The chemical, physical and mechanical properties of cement and silica fume used in this study are given in Table 1. Two different quartz sands and a powder were used as aggregate, with maximum particle size of 0.6 mm, 0.3 mm and 0.100 mm, respectively. The granule structure of the quartz powder and particle size analysis of granule materials are shown in Figs. 1 and 2. respectively. To fluidify the mixture, a polycarboxylate based superplasticizer was used. This superplasticizer serves to maintain fluidity of fresh concrete and attain high strength in a short time. Also, hookless short brass coated steel fibres, 6 mm long and 0.6 mm wide, were used. The steel fibre had a tensile strength of 2250 MPa, a specific gravity of 7.181 g/cm3 and an aspect ratio of 37.5.
462
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
Fig. 5. (a) The RPC sample that was produced without opening channels on the presetting pressure mould and (b) the RPC sample that was produced by opening channels on the pre-setting pressure mould. 2.2. Experimental procedures The mixture proportions, the mixing procedure and the treatment type were taken from Ref. [14]. Mixture proportions and compressive strength are presented in Table 2 that samples had a length of 200 mm and a diameter of 100 mm and cured in water for 7 and 28 days at 20 °C. Pre-setting pressurized samples were treated for 3 days at 90 °C water steam and then dry air was applied for 12 h at 300 °C and during the remaining days they were hold in water 24 days at 20 °C after demoulding as stated in Ref. [14]. Pre-setting pressure was applied to minimize the adverse effects of autogenous shrinkage and to remove water and air from the RPC. It is needed to keep RPC under pressure to minimize the adverse effects of autogenous shrinkage during the setting phase [3]. Also, the sample was expanded because the pressed grains were pushed against each other. When the pressure was removed, the density of the sample was decreased consequently. The desired compression level was not achieved. During the setting stage, the RPC should be kept under pressure to achieve a desirable compression level. This process was very difficult in terms of applicability. For that reason, some changes in mould design were made to prevent decrease in pressure and setting of RPC under pressure was performed and then plastic RPC in mould was taken away from the press. A mould was specially designed to implement the pre-setting pressure process. As can be seen in Fig. 3 the mould was internally 50 mm wide, 65 mm height, 300 mm long and it was made of 1040 steel hardened by heat treatment. The height of the sample varies to a certain extent in accordance with the pre-setting pressure applied. The mould was filled with the fresh concrete first (Fig. 4a). Then, the piston in rectangular prism form was covered with nylon to ensure it moves conveniently and it was ensured to prevent water, air and the RPC paste come out from unwanted spots of the mould (Fig. 4b). After the desired pressure was applied to the mould, fixture component which helped to maintain the pressure was mounted. In order to tighten the screws on the mould, the fixture in rectangular prism form was temporarily placed between the press and the mould. The pressure to be applied on the concrete was adjusted at such a slow loading speed that it would allow the water come out from the very narrow channels (h = 0.10 mm). Thus, the RPC in fresh form that was filled in the mould was applied pre-setting pressure with a loading speed of 0.13 MPa/s in a controlled manner until the desired load is achieved (Fig. 4c). Keeping it under this load until its deformation is fixed and water and air exit stops (nearly 1 h) and then tightening the six screws of 16 mm in diameter which were hardened by heat treatment and which prevented the piston from going back, the load of the press was relieved (Fig. 4d). During the preliminary trials for the mould design, no water channels were opened in the mould to allow water out and it was assumed that water would come out from the juncture points on the mould. However, it did not come out as
Fig. 7. The effect of pre-setting pressure on RPC’s unit weight. expected and water and air that were kept within the mould caused an increase in the amount of voids. Fig. 5a and b shows the samples produced before and after the mould was provided with water channels. The plastic RPC was applied pre-setting pressure (control-without pressure, 5, 10, 15, 20, 25 MPa) by this mould mechanism. At the end of 28 days, unit weight of samples was measured to control the effects of pre-setting pressure. It is well known that unit weight values of samples increase with the increase in the pre-setting pressure. However, unit weight values may be inconsistent due to eccentricity of piston, friction at the side walls of the mould or unpredictable adverse conditions. Therefore, unit weight values are extremely important in order to perform the experiment in a controlled manner. To define the occurred first crack the sample was divided into parts per 5 cm. By painting the samples in white before the test, the first cracking strength, its graph and progress were monitored. Thanks to the test setup seen in Fig. 6, prismatic samples (50 50 300 mm) were used to evaluate the flexural strength and toughness of concrete. The test samples were used at four points of load flexural test. Flexural test was performed by compact test equipment which could measure automatically load–deflection values during the test period by taking five data per second. The data was saved during the test period by a computer which was connected a loadcell and LVDT were located on the testing equipment to measure load and deflection. The loading speed of the testing equipment was adjusted in order to achieve a deflection of 0.05–0.10 mm/min at the beam midpoint. Having the samples loaded until they get entirely cracked at this speed, the load–deflection graphics were obtained. Using the load–deflection graphics, the cracking strength, flexural strength and toughness were calculated [20–25]. Toughness equivalent to the area under the load–deflection curves up to 10 mm deflection. The area under the load–deflection curve was calculated by a Matlab 7.5.0 software program.
3. Results and discussion Finding out the unit weight values of the samples, it was tested to see if pre-setting pressure was applied effectively or not. It caused the unit weight value to increase to have water and air discarded from the mould and the particles get closer to each other. While the unit weight value, pre-setting pressure of 5 MPa (2585 kg/m3) increased 3.5% compared to the control-without
Fig. 6. The technical drawing and front view of the flexural test mechanism.
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
463
Fig. 8. The effect of pre-setting pressure on RPC’s microstructure (a) control-without pressure, (b) 5 MPa, (c) 10 MPa, (d) 15 MPa, (e) 20 MPa, (f) 25 MPa.
pressure RPC (2498 kg/m3), it increased almost 1% in each pre-setting pressure interval until the pre-setting pressure of 25 MPa. Between the control RPC and the 25 MPa pre-setting pressurized RPC, it was observed that the unit weight value (2712 kg/m3) increased 8.6% and volume of sample decreased 7.9% in total Fig. 7. These results were taken to be an indication that the compression process was carried out correctly. At the end of the deflection experiment, the fracture surfaces of the samples were examined. It was observed that the fraction occurred in non-pre-setting pressured samples as a result of the fact that the fibres happened to come out easily. In the samples that were pre-setting pressured, on the other hand, it was observed that the fibres hardly managed to come out of the concrete and even some fibres got drifted apart. The specimens taken from the beam
samples were examined also under the microscope. The examination showed that the non-pre-setting pressured samples included voids both in the paste and on the interface between the fibre and the paste Fig. 8a. In the samples which were applied pre-setting pressure, however, it can be seen that the voids in the internal structure and between the fibre and the paste decreased to a great extent Fig. 8b–f. The load–deflection graphics obtained from the flexural test are presented in Fig. 9. Using the load–deflection graphics, the cracking strength, flexural strength and toughness values are presented in Table 3 and their graphical representations are presented in Figs. 10 and 11. When Fig. 10 is examined, the increase ratio between the flexural strength of the control RPC (21.92 MPa) and the flexural strength of 5 MPa pre-setting pressurized RPC (29.28 MPa)
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
464
Fig. 9. The load–deflection graphics of pre-setting pressure and non-pre-setting pressure RPCs.
Fig. 11. Toughness based on pre-setting pressure.
Table 3 Flexural test results of pre-setting pressured RPC. No.
Crack strength (MPa)
Flexural strength (MPa)
Total deflection (mm)
Toughness (N m)
Ctrl 5 10 15 20 25
21.06 24.84 22.00 28.42 29.66 30.20
21.92 29.28 32.20 33.60 35.40 36.40
5.38 11.32 10.80 10.64 10.20 12.60
24.31 82.50 84.38 101.46 103.06 116.96
Fig. 12. The effect of pre-setting pressure on the unit weight, flexural strength and toughness.
Fig. 10. The change in flexural strength based on pre-setting pressure.
seems to be 34%. With a decrease in the increase ratio, it has been 10% between 5 MPa and 10 MPa, 4% between 10 MPa and 15 MPa, 5% between 15 MPa and 20 MPa and 3% between 20 MPa and 25 MPa. It can be seen that the flexural strength between the control RPC and the 5 MPa pre-setting pressurized RPC is higher as compared to the increase in the other pre-setting pressures. It has been observed that the 5 MPa pre-setting pressure that was applied onto the flowable RPC was enough for the large voids and water to come out. When the internal structures views of the samples that were applied pre-setting pressure are examined, it has been observed that there was no definite difference between the internal structures of the RPCs that were produced with pre-set-
Fig. 13. The effect of pre-setting pressure on unit cost.
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
ting pressures bigger than 5 MPa (Fig. 8b–f). Pre-setting pressure also contributed to the flexure parameters by causing an increase in adherence in paste parts of the RPC where there was no fibre. When the toughness graphic (Fig. 11) which is a significant parameter among the flexural test results is examined, the toughness increase ratio between the toughness of the control RPC (24.31 N m) and the toughness of the 5 MPa pre-setting pressurized RPC (82.50 N m) seems to be more than thrice. In the succeeding pre-setting pressures, the increase ratio fell down in each step interval to 2.3%, 20.2%, 1.6% and 13.5% respectively. With the increase in adherence, it was more difficult for the fibres to come out and fracture took place as the load–deflection curve got higher after the maximum load. This caused an increase in the toughness. The correlation between the unit weight, flexural strength and toughness of the pre-setting pressured RPC has been studied (Fig. 12). When the graphic is examined, it has been observed that the pre-setting pressure increased the toughness value more than the flexural strength, while it increased the unit weight less with a total ratio of 8.6%. The water amount that came out of the sample during the application of pre-setting pressure was collected in a separate pot and weighed. When these weight results are examined, it was seen that there was no definite change between the lowest pre-setting pressure (5 MPa) and the highest pre-setting pressure (25 MPa). Therefore, the increase in unit weight values happened to be at a low level. The pre-setting pressure was examined also in terms of cost. The cost value of the RPC with the mixture shown in Table 2 was calculated to be 1676 TL/m3 (TL = Turkish Liras) and based on this figure, the unit flexural strength and toughness costs were calculated as shown in the change graphic in Fig. 13. It seems that the most definite change among the unit costs was between the control RPC and the 5 MPa pre-setting pressurized RPC. Based on the unit cost of flexural strength, when the 5 MPa pre-setting pressurized RPC and the control RPC are compared, a decrease of 25% and in the toughness unit cost, a decrease of 70.5% were determined. It is seen that the ratio of decrease in cost gets lower after the 5 MPa pre-setting pressure.
4. Conclusions The pre-setting pressure that was applied onto flowable RPC increased the unit weight, flexural strength and toughness, as a result of discarding the voids and the free water and making the granules get closer to each other. The increase in strength between the flexural strength of the non-pre-setting pressured RPC and the flexural strength of the 5 MPa pre-setting pressurized RPC happened to be at the highest ratio (34%) as compared to the other pre-setting pressured RPCs. Moreover, it was observed that toughness values increased more than thrice. Consequently, with a presetting pressure of 25 MPa, flexural strength of 36.4 MPa and toughness of 116.96 N m were obtained. When compared to the previous fibred concrete studies, this value is considerable flexural strength and toughness. When the internal structure views of the samples from the presetting pressured RPC are examined, no definite difference was observed between the internal structures of the paste and the fibre interface of the RPCs that were produced with pre-setting pressures higher than 5 MPa. The pre-setting pressure application mechanism means an extra cost when compared to the element that was planned to be produced with the initial investment cost. However, when the flexural strength and toughness unit costs are taken into consideration, pre-setting pressure is considered to be economical for prefabricated production. By applying pre-setting pressure, prefabricated
465
elements with thinner internal walls can be produced with RPCs. It is thought that there RPC elements can be an alternative not only to the concrete materials, but also to other materials in industry. Acknowledgements This work was supported by Sakarya University Scientific Research Foundation (Project Numbers: 2007-01-04-003 and 200650-02-055). The authors would like to thank Sakarya University and TCMA due to its financial support for this study. References [1] Bayramov F, Tasßdemir C, Tasßdemir MA. Optimisation of steel fibre reinforced concretes by means of statistical response surface method. Cem Concr Compos 2004;26:665–75. [2] Tasßdemir MA, Bayramov F, Kocatürk, AN, Yerlikaya M. Betonun performansa göre tasarımında yeni gelisßmeler, Beton 2004 Kongresi, June 10–12; 2004. p. 1–34. [3] Richard P, Cheyrezy MH. Composition of reactive powder concrete. Cem Concr Res 1995;25:1501–11. [4] Aitcin PC. Cement of yesterday and today concrete of tomorrow. Cem Concr Res 2000;30:1349–59. [5] Yazıcı H. The effect of curing conditions on compressive strength of ultra high strength concrete with high volume mineral admixtures. Build Environ 2007;42:2083–9. [6] Dugat J, Roux N, Bernier G. Mechanical properties of reactive powder concretes. Mater Struct 1996;29:233–40. [7] Morin V, Cohen-Tenoudji F, Feylessoufi A, Richard P. Evolution of the capillary network in a reactive powder concrete during hydration process. Cem Concr Res 2002;32:1907–14. [8] Matte V, Moranville M. Durability of reactive powder composites: influence of silica fume on the leaching properties of very low water/binder pastes. Cem Concr Compos 1999;21:1–9. [9] Chan Y, Chu S. Effect of silica fume on steel fiber bond characteristics in reactive powder concrete. Cem Concr Res 2004;34:1167–72. [10] Aitcin PC. Concrete the most widely used construction materials. ACI SP, vol. 154; 1995. p. 257–66. [11] Yazıcı H, Yardımcı MY, Aydın S, Karabulut AS. Mechanical properties of reactive powder concrete containing mineral admixtures under different curing regimes. Constr Build Mater 2009;23:1223–31. [12] Richard P, Cheyrezy MH. Reactive powder concretes with high ductility and 200–800 MPa compressive strength. Concrete technology: past, present, and future. In: Proceedings of the V. Mohan Malhotra symposium, ACI SP-144; 1994. p. 507–18. [13] Teichman T, Schmidt M. Influence of the packing density of fine particles on structure, strength and durability of UHPC. Ultra high performance concrete (UHPC). In: International symposium on ultra high performance concrete, September 13–15; 2004. p. 312–23. [14] Ipek M, Yılmaz K, Sümer M, Sarıbıyık M. Effect of pre-setting pressure applied to mechanical behaviours of reactive powder concrete during setting phase. Constr Build Mater 2011;25:61–8. [15] Neville AM. Properties of concrete. 4th ed. England: Longman Group Limited; 1995. [16] Aitcin PC, Sarkar SL, Ranc R, Levy C. A high-silica-modulus cement for high performance concrete. Adv Cem Mater Ceram Trans Gaithersburg 1991;16:103–21. [17] Goldman A, Bentur A. The influence of microfillers on enhancement of concrete strength. Cem Concr Res 1993;23:962–72. [18] Talebinejad I, Bassam SA, Iranmanesh A, Shekarchizadeh M. Optimizing mix proportions of normal weight reactive powder concrete with strengths of 200– 350 MPa. Ultra high performance concrete (UHPC). In: International symposium on ultra high performance concrete, September 13–15; 2004. p. 133–41. [19] Yazıcı H. The effect of curing conditions on compressive strength of ultra high strength concrete with high volume mineral admixtures. Build Environ 2007;42:2083–9. [20] TS 10513. Çelik teller-beton takviyesinde kullanılan. Turkish Standards Institute; 1992. [21] TS 10514. Beton-Çelik Tel Takviyeli-Çelik Telleri Betona Karısßtırma ve Kontrol Kuralları, Turkish Standards Institute; 1992. [22] TS 10515. Çelik Tel Takviyeli Betonun Eg˘ilme Mukavemeti Deney Metodu, Turkish Standards Institute; 1992. [23] ASTM C 1018. Standard test method for flexural toughness and first-crack strength of fiber-reinforced concrete (using beam with third-point loading), annual book of ASTM standards, V 4.02. American Society for Testing and Materials, Philadelphia, PA; 1989. p. 637–44. [24] JSCE standard SF-4, method of test for flexural strength and flexural toughness of fiber reinforced concrete. Japan Society of Civil Engineers (JSCE) Standard; 1984. p. 58–66. [25] Sukontasukkul P. Toughneses evaluation of steel and polypropylene fibre reinforced concrete beams under bending. Thammasa Int J Sci Technol 2004;9(3):35–41.
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
The effect of pre-setting pressure applied flexural strength and fracture toughness of reactive powder concrete during the setting phase a,⇑ _ Metin Ipek , Kemalettin Yilmaz b, Mucteba Uysal b a b
Sakarya University, Technical Education Faculty, Construction Education Department, 54187 Sakarya, Turkey Sakarya University, Engineering Faculty, Civil Engineering Department, 54187 Sakarya, Turkey
a r t i c l e
i n f o
Article history: Received 19 January 2011 Received in revised form 9 June 2011 Accepted 18 June 2011 Available online 13 July 2011 Keywords: Reactive powder concrete Pre-setting pressure Fibre Flexural strength Toughness
a b s t r a c t In this study, the effect of pre-setting pressure on reactive powder concrete (RPC) was examined. The fresh (plastic) concrete samples were filled in a specified beam mould and six different pre-setting pressure values (control-without pressure, 5, 10, 15, 20, 25 MPa) were applied. Flexural strength was improved by 34% with very low amounts (5 MPa) of pre-setting pressure. Also, toughness was increased more than thrice. In this study, the maximum flexural strength of 36.4 MPa and toughness of 116.960 N m were obtained under 25 MPa pre-setting pressure. Volume of sample was decreased 7.9% by pre-setting pressure. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction In recent years, high-performance concrete is being assessed not only in terms of high strength, but also in terms of specifications such as strength and ductility. Although high-strength concretes are favourable also because they are impervious, a significant issue with these concretes which emerges as their strength increases has been the problem of fragility. Although the capacity of axial deformation increases in high-strength concretes, once the peak point is reached, stress relaxation is instantaneous and the fragility leads to crashes. Since the beginning of 1960s, in order to resolve the issue of fragility in concrete, it has been a common practice to work fibre into it to ensure its ductility capacity. In order to improve the mechanical features such as tensile strength, resistance to cracks, endurance to wearing and shocks and toughness of plain concrete, it is mixed with steel, glass and polypropylene fibres. The fibres used in concrete increase tensile and flexural strength and also decrease shrinkage cracks [1,2]. Unlike the high performance concrete, reactive powder concrete (RPC) has been produced by working fibre into concrete, replacing aggregates with microgranules, arranging granule distribution so as to form minimum void in concrete, increasing pozzolanic activity and exposing the materials to different production and treatment. Although it was discovered in the first half of
⇑ Corresponding author. Tel.: +90 2642956476; fax: +90 2642956424. _ E-mail address: [email protected] (M. Ipek). 0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2011.06.045
1990s, the RPC has started to be developed and used all over the world [3–5]. RPC is a cement-based ultra high performance concrete with superior mechanical and physical specifications and excellent ductility and durability characteristics. Compression strength of RPC is 150–800 MPa, while its tensile strength changes between 25 and 150 MPa. Its fracture energy, on the other hand, changes between 1200 and 40,000 J/m2. The durability properties of RPC are better than current high performance concrete in terms of magnitude [3–12]. Fine powders such as crushed quartz (100–600 lm) are used instead of coarse aggregate in order to make RPC more homogeneous. The water/cement ratio of RPC is reduced to less than 0.20 by using superplasticizers (SP). Silica fume addition to RPC reduces the total pore volume of the cement paste and the average diameter of the pores. Also, researcher’s reports that application of different heat cure processes improves the mechanical properties of RPC to a great extent after the application of 50 MPa presetting pressure to fresh RPC [3]. Besides, pressure application to fresh RPC during the setting phase of 6–12 h can eliminate some amount of pores caused by autogenous shrinkage. Pre-setting pressure applied during setting stage causes microcracks in the sample. These microcracks are improved as a result of expansion of aggregate after discharging the applied pressure. Pre-setting pressure trials were carried out also in the previous studies [3,6,13,14]. However, no study has been encountered so far where the effect of pre-setting pressure on beam samples are examined. The effects of pre-setting pressure on compressive
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
460
strength have been examined by some researchers. In the study carried out by Dugat et al. [6], they applied a pre-setting pressure of 60 MPa onto the RPB800 concrete during the setting and hardening stages. Also, they applied 90 °C hot water and then heat curing process at 250 °C onto RPB800 samples for 4 days. Moreover, they examined the effect of the fibre content on the fracture energy. In their experiments, they increased the fracture energy up to 40,000 J/m2 and found out the optimum amount of fibre content to be between 2% and 3%. Another study has been carried out by Teichman and Schmidt [13]. In that study, they applied 50 MPa constant pre-setting pressure onto the RPC samples while they are fresh and achieved strength of 487 MPa at the end of the experiment. The last study has been the one performed by us [14]. In this study, the effects of pre-setting pressure on compression strength, the elastic modulus and unit weight have been examined in details. For that purpose, six different pre-setting pressure amounts of control-without pressure, 25, 50, 75, 100, 125 MPa have been applied onto RPC and the compressive strength characteristics of the samples have been examined. Eventually, as the pre-setting pressure increased, compression strength and elastic modulus increased and the unit weight decreased. When compared to control-without pressure, 100 MPa pre-setting pressured had twice better compressive strength. Generally, concrete properties are influenced by placement and compression factor. It is well known that reducing voids positively affect the mechanical and durability properties of hardened concrete. Concrete must be placed with minimum voids [15]. During the application of vibration to RPC, as the samples are examined, the increment in air volume due to air-entrainer property of SP is clearly observed. The RPC workability is similar to self-compacting
Fig. 1. Granule structure of the quartz powder [14].
Table 1 Properties of cement and silica fume. Component
CaO SiO2 C Al2O3 Fe2O3 MgO K2O Cl Na2O P2O5 SO3 H2O Fever loss pH value C3S C2S C3A C4AF Silicate modulus Alumina modulus Hydraulic modulus Total alkaline
Chemical composition (%) Cement
Silica fume
64.47 20.09 – 5.01 2.73 1.72 0.66 0.01 0.21 – 3.03 – 2.11 – 60.7 11.8 8.6 8.3 2.6 1.8 2.3 0.58
0.50 96 1.50 0.70 0.25 0.60 0.85 0.10 0.25 0.10 0.50 0.80 1.50 5.0–8.0 – – – – – – – –
Physical properties Cement
Silica fume
Blaine specify surface Unit volume weight Specific gravity Initial setting time Final setting time Volume expansion Genlesßmesi
5162 cm2/g – 3.14 154 min 191 min 0.8 mm
200,000 cm2/g 0.650 g/cm3 2.26 – – –
2 days 7 days 28 days
Compressive strength of cement (MPa) 39.8 MPa – 54.2 MPa – 61.8 MPa –
Fig. 2. Particle size analysis of granule materials [14].
concrete. In this study, T50 value of the mixture of RPC measured 4 s in slump-flow test. Workability can be brought to a desired level due to high ratio of SP whereas RPC has low water cement ratio. However, water and compressed air of closed pockets were found in the RPC. It was observed that the air bubbles in concrete caused swelling overflow from mould or forming of bubbles at the surface. The voids were formed between fibre and the concrete interface. In such condition, adherence between fibre and concrete interface was affected negatively and consequently the strength was reduced. Because the steel fibres used in RPC is hookless and 0.16 mm in diameter and in order to make these fibres achieve a favourable adherence with concrete, it is mandatory to cover it with a finegrained paste. This necessity stems from the fact that all the materials which compose RPC are at micron level. However, there are voids in the fine-grained paste as compared to normal concrete, even in the least. By reducing the voids to a minimum, there will be achieved more adherence between the fine-grained paste and the micron-sized hookless fibre. In order to increase the adherence between the paste and the fibres and to minimize the voids in the paste, it was decided to apply pre-setting pressure onto the RPC in fresh form. In this study, the effects of pre-setting pressure on RPC’s, unit weight, flexural strength, fracture parameters and cost were investigated. For this purpose, a special moulding system was designed and the mixture in fresh form was filled in these moulds where it was applied pre-setting pressure of six different
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
461
Table 2 Compositions of mixtures [14]. Material
Cement
Silica fume
Q. powder (100-0 lm)
Q. sand (100–300 lm)
Q. sand (300–600 lm)
Water
SP
Air
Steel fibre 4% of all volume)
Weight (kg/m3) 7 days compr. strength 28 days compr. strength
864 145.8 MPa 170.29 MPa
259
346
248
248
216
26
1.8%
287
Fig. 3. The technical drawing and view from above of the specially designed pre-setting pressure mould.
Fig. 4. (a) Filling the fresh RPC into the mould, (b) placing the piston in rectangular prism form, (c) performing pre-setting pressure under the concrete press, (d) tightening the screws which prevent the pressure within the mould from bleeding off.
amounts (control-without pressure, 5, 10, 15, 20, 25 MPa) and the samples were kept under the pressure until they got hardened. At the end of the 28-day period, the samples were tested in terms of unit weight and flexural test. 2. Experimental study 2.1. Materials High performance cement of PC 52.5 CEMIR type was preferred for this study. For RPC, pozzolanic material is needed, which fills voids of microparticulates in binder paste and which contributes to the strength by producing secondary hy-
drates by pozzolanic reaction with the lime resulting from primary hydration [3,12,14,16–19]. Undensified silica fume (SF) provided from Elkem Company in Norway was used in this study. The chemical, physical and mechanical properties of cement and silica fume used in this study are given in Table 1. Two different quartz sands and a powder were used as aggregate, with maximum particle size of 0.6 mm, 0.3 mm and 0.100 mm, respectively. The granule structure of the quartz powder and particle size analysis of granule materials are shown in Figs. 1 and 2. respectively. To fluidify the mixture, a polycarboxylate based superplasticizer was used. This superplasticizer serves to maintain fluidity of fresh concrete and attain high strength in a short time. Also, hookless short brass coated steel fibres, 6 mm long and 0.6 mm wide, were used. The steel fibre had a tensile strength of 2250 MPa, a specific gravity of 7.181 g/cm3 and an aspect ratio of 37.5.
462
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
Fig. 5. (a) The RPC sample that was produced without opening channels on the presetting pressure mould and (b) the RPC sample that was produced by opening channels on the pre-setting pressure mould. 2.2. Experimental procedures The mixture proportions, the mixing procedure and the treatment type were taken from Ref. [14]. Mixture proportions and compressive strength are presented in Table 2 that samples had a length of 200 mm and a diameter of 100 mm and cured in water for 7 and 28 days at 20 °C. Pre-setting pressurized samples were treated for 3 days at 90 °C water steam and then dry air was applied for 12 h at 300 °C and during the remaining days they were hold in water 24 days at 20 °C after demoulding as stated in Ref. [14]. Pre-setting pressure was applied to minimize the adverse effects of autogenous shrinkage and to remove water and air from the RPC. It is needed to keep RPC under pressure to minimize the adverse effects of autogenous shrinkage during the setting phase [3]. Also, the sample was expanded because the pressed grains were pushed against each other. When the pressure was removed, the density of the sample was decreased consequently. The desired compression level was not achieved. During the setting stage, the RPC should be kept under pressure to achieve a desirable compression level. This process was very difficult in terms of applicability. For that reason, some changes in mould design were made to prevent decrease in pressure and setting of RPC under pressure was performed and then plastic RPC in mould was taken away from the press. A mould was specially designed to implement the pre-setting pressure process. As can be seen in Fig. 3 the mould was internally 50 mm wide, 65 mm height, 300 mm long and it was made of 1040 steel hardened by heat treatment. The height of the sample varies to a certain extent in accordance with the pre-setting pressure applied. The mould was filled with the fresh concrete first (Fig. 4a). Then, the piston in rectangular prism form was covered with nylon to ensure it moves conveniently and it was ensured to prevent water, air and the RPC paste come out from unwanted spots of the mould (Fig. 4b). After the desired pressure was applied to the mould, fixture component which helped to maintain the pressure was mounted. In order to tighten the screws on the mould, the fixture in rectangular prism form was temporarily placed between the press and the mould. The pressure to be applied on the concrete was adjusted at such a slow loading speed that it would allow the water come out from the very narrow channels (h = 0.10 mm). Thus, the RPC in fresh form that was filled in the mould was applied pre-setting pressure with a loading speed of 0.13 MPa/s in a controlled manner until the desired load is achieved (Fig. 4c). Keeping it under this load until its deformation is fixed and water and air exit stops (nearly 1 h) and then tightening the six screws of 16 mm in diameter which were hardened by heat treatment and which prevented the piston from going back, the load of the press was relieved (Fig. 4d). During the preliminary trials for the mould design, no water channels were opened in the mould to allow water out and it was assumed that water would come out from the juncture points on the mould. However, it did not come out as
Fig. 7. The effect of pre-setting pressure on RPC’s unit weight. expected and water and air that were kept within the mould caused an increase in the amount of voids. Fig. 5a and b shows the samples produced before and after the mould was provided with water channels. The plastic RPC was applied pre-setting pressure (control-without pressure, 5, 10, 15, 20, 25 MPa) by this mould mechanism. At the end of 28 days, unit weight of samples was measured to control the effects of pre-setting pressure. It is well known that unit weight values of samples increase with the increase in the pre-setting pressure. However, unit weight values may be inconsistent due to eccentricity of piston, friction at the side walls of the mould or unpredictable adverse conditions. Therefore, unit weight values are extremely important in order to perform the experiment in a controlled manner. To define the occurred first crack the sample was divided into parts per 5 cm. By painting the samples in white before the test, the first cracking strength, its graph and progress were monitored. Thanks to the test setup seen in Fig. 6, prismatic samples (50 50 300 mm) were used to evaluate the flexural strength and toughness of concrete. The test samples were used at four points of load flexural test. Flexural test was performed by compact test equipment which could measure automatically load–deflection values during the test period by taking five data per second. The data was saved during the test period by a computer which was connected a loadcell and LVDT were located on the testing equipment to measure load and deflection. The loading speed of the testing equipment was adjusted in order to achieve a deflection of 0.05–0.10 mm/min at the beam midpoint. Having the samples loaded until they get entirely cracked at this speed, the load–deflection graphics were obtained. Using the load–deflection graphics, the cracking strength, flexural strength and toughness were calculated [20–25]. Toughness equivalent to the area under the load–deflection curves up to 10 mm deflection. The area under the load–deflection curve was calculated by a Matlab 7.5.0 software program.
3. Results and discussion Finding out the unit weight values of the samples, it was tested to see if pre-setting pressure was applied effectively or not. It caused the unit weight value to increase to have water and air discarded from the mould and the particles get closer to each other. While the unit weight value, pre-setting pressure of 5 MPa (2585 kg/m3) increased 3.5% compared to the control-without
Fig. 6. The technical drawing and front view of the flexural test mechanism.
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
463
Fig. 8. The effect of pre-setting pressure on RPC’s microstructure (a) control-without pressure, (b) 5 MPa, (c) 10 MPa, (d) 15 MPa, (e) 20 MPa, (f) 25 MPa.
pressure RPC (2498 kg/m3), it increased almost 1% in each pre-setting pressure interval until the pre-setting pressure of 25 MPa. Between the control RPC and the 25 MPa pre-setting pressurized RPC, it was observed that the unit weight value (2712 kg/m3) increased 8.6% and volume of sample decreased 7.9% in total Fig. 7. These results were taken to be an indication that the compression process was carried out correctly. At the end of the deflection experiment, the fracture surfaces of the samples were examined. It was observed that the fraction occurred in non-pre-setting pressured samples as a result of the fact that the fibres happened to come out easily. In the samples that were pre-setting pressured, on the other hand, it was observed that the fibres hardly managed to come out of the concrete and even some fibres got drifted apart. The specimens taken from the beam
samples were examined also under the microscope. The examination showed that the non-pre-setting pressured samples included voids both in the paste and on the interface between the fibre and the paste Fig. 8a. In the samples which were applied pre-setting pressure, however, it can be seen that the voids in the internal structure and between the fibre and the paste decreased to a great extent Fig. 8b–f. The load–deflection graphics obtained from the flexural test are presented in Fig. 9. Using the load–deflection graphics, the cracking strength, flexural strength and toughness values are presented in Table 3 and their graphical representations are presented in Figs. 10 and 11. When Fig. 10 is examined, the increase ratio between the flexural strength of the control RPC (21.92 MPa) and the flexural strength of 5 MPa pre-setting pressurized RPC (29.28 MPa)
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
464
Fig. 9. The load–deflection graphics of pre-setting pressure and non-pre-setting pressure RPCs.
Fig. 11. Toughness based on pre-setting pressure.
Table 3 Flexural test results of pre-setting pressured RPC. No.
Crack strength (MPa)
Flexural strength (MPa)
Total deflection (mm)
Toughness (N m)
Ctrl 5 10 15 20 25
21.06 24.84 22.00 28.42 29.66 30.20
21.92 29.28 32.20 33.60 35.40 36.40
5.38 11.32 10.80 10.64 10.20 12.60
24.31 82.50 84.38 101.46 103.06 116.96
Fig. 12. The effect of pre-setting pressure on the unit weight, flexural strength and toughness.
Fig. 10. The change in flexural strength based on pre-setting pressure.
seems to be 34%. With a decrease in the increase ratio, it has been 10% between 5 MPa and 10 MPa, 4% between 10 MPa and 15 MPa, 5% between 15 MPa and 20 MPa and 3% between 20 MPa and 25 MPa. It can be seen that the flexural strength between the control RPC and the 5 MPa pre-setting pressurized RPC is higher as compared to the increase in the other pre-setting pressures. It has been observed that the 5 MPa pre-setting pressure that was applied onto the flowable RPC was enough for the large voids and water to come out. When the internal structures views of the samples that were applied pre-setting pressure are examined, it has been observed that there was no definite difference between the internal structures of the RPCs that were produced with pre-set-
Fig. 13. The effect of pre-setting pressure on unit cost.
_ M. Ipek et al. / Construction and Building Materials 26 (2012) 459–465
ting pressures bigger than 5 MPa (Fig. 8b–f). Pre-setting pressure also contributed to the flexure parameters by causing an increase in adherence in paste parts of the RPC where there was no fibre. When the toughness graphic (Fig. 11) which is a significant parameter among the flexural test results is examined, the toughness increase ratio between the toughness of the control RPC (24.31 N m) and the toughness of the 5 MPa pre-setting pressurized RPC (82.50 N m) seems to be more than thrice. In the succeeding pre-setting pressures, the increase ratio fell down in each step interval to 2.3%, 20.2%, 1.6% and 13.5% respectively. With the increase in adherence, it was more difficult for the fibres to come out and fracture took place as the load–deflection curve got higher after the maximum load. This caused an increase in the toughness. The correlation between the unit weight, flexural strength and toughness of the pre-setting pressured RPC has been studied (Fig. 12). When the graphic is examined, it has been observed that the pre-setting pressure increased the toughness value more than the flexural strength, while it increased the unit weight less with a total ratio of 8.6%. The water amount that came out of the sample during the application of pre-setting pressure was collected in a separate pot and weighed. When these weight results are examined, it was seen that there was no definite change between the lowest pre-setting pressure (5 MPa) and the highest pre-setting pressure (25 MPa). Therefore, the increase in unit weight values happened to be at a low level. The pre-setting pressure was examined also in terms of cost. The cost value of the RPC with the mixture shown in Table 2 was calculated to be 1676 TL/m3 (TL = Turkish Liras) and based on this figure, the unit flexural strength and toughness costs were calculated as shown in the change graphic in Fig. 13. It seems that the most definite change among the unit costs was between the control RPC and the 5 MPa pre-setting pressurized RPC. Based on the unit cost of flexural strength, when the 5 MPa pre-setting pressurized RPC and the control RPC are compared, a decrease of 25% and in the toughness unit cost, a decrease of 70.5% were determined. It is seen that the ratio of decrease in cost gets lower after the 5 MPa pre-setting pressure.
4. Conclusions The pre-setting pressure that was applied onto flowable RPC increased the unit weight, flexural strength and toughness, as a result of discarding the voids and the free water and making the granules get closer to each other. The increase in strength between the flexural strength of the non-pre-setting pressured RPC and the flexural strength of the 5 MPa pre-setting pressurized RPC happened to be at the highest ratio (34%) as compared to the other pre-setting pressured RPCs. Moreover, it was observed that toughness values increased more than thrice. Consequently, with a presetting pressure of 25 MPa, flexural strength of 36.4 MPa and toughness of 116.96 N m were obtained. When compared to the previous fibred concrete studies, this value is considerable flexural strength and toughness. When the internal structure views of the samples from the presetting pressured RPC are examined, no definite difference was observed between the internal structures of the paste and the fibre interface of the RPCs that were produced with pre-setting pressures higher than 5 MPa. The pre-setting pressure application mechanism means an extra cost when compared to the element that was planned to be produced with the initial investment cost. However, when the flexural strength and toughness unit costs are taken into consideration, pre-setting pressure is considered to be economical for prefabricated production. By applying pre-setting pressure, prefabricated
465
elements with thinner internal walls can be produced with RPCs. It is thought that there RPC elements can be an alternative not only to the concrete materials, but also to other materials in industry. Acknowledgements This work was supported by Sakarya University Scientific Research Foundation (Project Numbers: 2007-01-04-003 and 200650-02-055). The authors would like to thank Sakarya University and TCMA due to its financial support for this study. References [1] Bayramov F, Tasßdemir C, Tasßdemir MA. Optimisation of steel fibre reinforced concretes by means of statistical response surface method. Cem Concr Compos 2004;26:665–75. [2] Tasßdemir MA, Bayramov F, Kocatürk, AN, Yerlikaya M. Betonun performansa göre tasarımında yeni gelisßmeler, Beton 2004 Kongresi, June 10–12; 2004. p. 1–34. [3] Richard P, Cheyrezy MH. Composition of reactive powder concrete. Cem Concr Res 1995;25:1501–11. [4] Aitcin PC. Cement of yesterday and today concrete of tomorrow. Cem Concr Res 2000;30:1349–59. [5] Yazıcı H. The effect of curing conditions on compressive strength of ultra high strength concrete with high volume mineral admixtures. Build Environ 2007;42:2083–9. [6] Dugat J, Roux N, Bernier G. Mechanical properties of reactive powder concretes. Mater Struct 1996;29:233–40. [7] Morin V, Cohen-Tenoudji F, Feylessoufi A, Richard P. Evolution of the capillary network in a reactive powder concrete during hydration process. Cem Concr Res 2002;32:1907–14. [8] Matte V, Moranville M. Durability of reactive powder composites: influence of silica fume on the leaching properties of very low water/binder pastes. Cem Concr Compos 1999;21:1–9. [9] Chan Y, Chu S. Effect of silica fume on steel fiber bond characteristics in reactive powder concrete. Cem Concr Res 2004;34:1167–72. [10] Aitcin PC. Concrete the most widely used construction materials. ACI SP, vol. 154; 1995. p. 257–66. [11] Yazıcı H, Yardımcı MY, Aydın S, Karabulut AS. Mechanical properties of reactive powder concrete containing mineral admixtures under different curing regimes. Constr Build Mater 2009;23:1223–31. [12] Richard P, Cheyrezy MH. Reactive powder concretes with high ductility and 200–800 MPa compressive strength. Concrete technology: past, present, and future. In: Proceedings of the V. Mohan Malhotra symposium, ACI SP-144; 1994. p. 507–18. [13] Teichman T, Schmidt M. Influence of the packing density of fine particles on structure, strength and durability of UHPC. Ultra high performance concrete (UHPC). In: International symposium on ultra high performance concrete, September 13–15; 2004. p. 312–23. [14] Ipek M, Yılmaz K, Sümer M, Sarıbıyık M. Effect of pre-setting pressure applied to mechanical behaviours of reactive powder concrete during setting phase. Constr Build Mater 2011;25:61–8. [15] Neville AM. Properties of concrete. 4th ed. England: Longman Group Limited; 1995. [16] Aitcin PC, Sarkar SL, Ranc R, Levy C. A high-silica-modulus cement for high performance concrete. Adv Cem Mater Ceram Trans Gaithersburg 1991;16:103–21. [17] Goldman A, Bentur A. The influence of microfillers on enhancement of concrete strength. Cem Concr Res 1993;23:962–72. [18] Talebinejad I, Bassam SA, Iranmanesh A, Shekarchizadeh M. Optimizing mix proportions of normal weight reactive powder concrete with strengths of 200– 350 MPa. Ultra high performance concrete (UHPC). In: International symposium on ultra high performance concrete, September 13–15; 2004. p. 133–41. [19] Yazıcı H. The effect of curing conditions on compressive strength of ultra high strength concrete with high volume mineral admixtures. Build Environ 2007;42:2083–9. [20] TS 10513. Çelik teller-beton takviyesinde kullanılan. Turkish Standards Institute; 1992. [21] TS 10514. Beton-Çelik Tel Takviyeli-Çelik Telleri Betona Karısßtırma ve Kontrol Kuralları, Turkish Standards Institute; 1992. [22] TS 10515. Çelik Tel Takviyeli Betonun Eg˘ilme Mukavemeti Deney Metodu, Turkish Standards Institute; 1992. [23] ASTM C 1018. Standard test method for flexural toughness and first-crack strength of fiber-reinforced concrete (using beam with third-point loading), annual book of ASTM standards, V 4.02. American Society for Testing and Materials, Philadelphia, PA; 1989. p. 637–44. [24] JSCE standard SF-4, method of test for flexural strength and flexural toughness of fiber reinforced concrete. Japan Society of Civil Engineers (JSCE) Standard; 1984. p. 58–66. [25] Sukontasukkul P. Toughneses evaluation of steel and polypropylene fibre reinforced concrete beams under bending. Thammasa Int J Sci Technol 2004;9(3):35–41.