Flexural behavior analysis and strength prediction of steel fiber-and-nanosilica reinforced rubber concrete

Abstract

In this study, a series of experiments were conducted to investigate the effects of the hook-shaped steel fiber (SF) and nanosilica (NS) on the flexural properties of the rubber concrete (RC), and the modification mechanisms of the NS on the flexural properties of steel fiber-and-nanosilica reinforced rubber concrete (SFNS-RC) were also revealed by using the X-ray diffraction (XRD) and scanning electron microscope. Based on the RC specimen with a rubber content of 5%, the experimental variables included the SF volume fractions (0%, 0.5%, 1.0%, and 1.5%) and NS replacement ratios (0%, 1%, and 2%). The results indicated that the NS could improve the microstructure of the RC, the peaks of Ca(OH)₂, C₃S, and C₂S steadily reduced as the NS replacement ratio increased, respectively, and the microstructure of SFNS-RC became more compact and uniform because the NS could produce pozzolanic activity to fill the pores of the matrix and strengthen the bond behavior between the SF and matrix. In addition, the RC specimens containing the SF and NS displayed the excellent flexural behaviors, including the ductility and energy dissipation characteristics. Finally, the empirical formulae for predicting the first-peak flexural strength and peak flexural strength of SFNS-RC were proposed through the analysis of experimental data, respectively.

Keywords

rubber concrete nanosilica hook-shaped steel fiber flexural behavior flexural strength prediction

Highlights

• The effects of the SF and NS on the flexural properties of RC were investigated.

• Combined effect of the SF and NS on flexural behavior of SFNS-RC was studied.

• The modification mechanisms of the NS on the flexural properties of SFNS-RC were revealed.

• The empirical formulae for first-peak flexural strength and peak flexural strength of SFNS-RC were proposed, respectively.

Introduction

A survey report showed that large quantities of scrap vehicle tires were generated every year, and approximately one-third of them were disposed to be burned, landfilled, stockpiled, and illegally dumped, which caused a serious threat to environment (Siddika et al., 2019). In recent years, it has been widely received for applying the recycling waste rubber derived from scrap vehicle tires into concrete to produce rubber concrete (RC), which could improve the certain properties of the concrete, such as the ductility, toughness, impact resistance, fatigue strength, energy absorption, and energy consumption (Ismail et al., 2017a).

A brief review of the current studies about RC showed that the rubber could obviously reduce the brittleness of concrete, and enhance toughness and energy absorption of the concrete (Abdelaleem et al., 2018; Ozbay et al., 2011; Zheng et al., 2008). Meanwhile, the rubber particles with high elasticity could also increase the impact resistance and energy dissipation capacity of concrete (Abdelaleem et al., 2018; Ismail et al., 2016a). In addition, the rubber aggregate could significantly contribute to the development of semi-lightweight and lightweight concrete due to the lower density compared to a conventional aggregate. Currently, the usage of waste tire rubber in civil engineering was mostly taken as an additive to Portland cement concrete, the modifiers to asphalt paving mixtures, the lightweight fillers, the crash barriers, the bumpers, and artificial reefs by using whole tires (Aslani, 2016). Therefore, the development of RC in civil engineering provided an effective method for recycling waste tire rubber.

It was found out that the compressive strength, elastic modulus, splitting tensile strength, and flexural strength of RC decreased as the rubber content increased. Antil et al. (2012) found that the compressive strengths of the concrete with a rubber volume fraction of 5%, 10%, 15%, and 20% were 11.1%, 23.5%, 31.9%, and 37.4% less than that of the concrete without rubber, respectively. Al-Tayeb et al. (2012) reported a decrease of 20%, 16.7%, and 22% in the compressive strength, splitting tensile strength, flexural strength, and elastic modulus of RC, respectively, when 20% of the fine aggregate was replaced by rubber particles. The reasons for strength loss could be summarized: (1) the hardness of the rubber particles was less than the surrounding cement paste. (2) Abundant cracks could be observed around the rubber particles. (3) A large number of rubber particles moved toward the top surface of concrete during the molding process (Ganjian et al., 2009; Sofi, 2017).

Adding steel fibers (SFs) to offset the negative effects caused by the incorporation of rubber particles has been tried. It was observed that the SFs could play a key role in many aspects to enhance the behaviors of RC. First, the SF could significantly resist to the cracking behavior, improve the tensile strength, and enhance the durability of RC (Nguyen et al., 2010; Turatsinze et al., 2006). Second, the SF could greatly increase the impact resistance of RC and compensate for the reduction caused by the rubber in splitting tensile strength and flexural strength (Ismail et al., 2017b). Third, the SF could work well with crushed rubber, which made RC exhibit wonderful compression toughness and energy dissipation capacity (Noaman et al., 2016). Fourthly, the hybrid reinforcement of SFs and polypropylene fibers could make the peak load and flexural toughness of RC more remarkable. (Park et al., 2014). Moreover, the SF also enabled RC to develop sufficient flexural strength, which could enhance the strain capacity and post-peak energy absorption behavior (Abaza et al., 2016; Alsaif et al., 2018).

Although the splitting tensile strength and flexural strength of RC could be enhanced by the incorporation of SFs, the poor bond behavior between the hardened cement matrix and rubber particles would reduce the overall mechanical properties of RC. Some scholars had proposed that the micro interface between rubber particles and cement hydrates could been improved by replacing the part of cement with the pozzolanic materials, such as metakaolin, fly ash, and silica fume, which could overcome the inherent microscopic defects of RC (Ozbay et al., 2011; Onuaguluchi, 2015; Gupta et al., 2016; Ismail et al., 2016b). Mohammed et al. (2016) found that the nanosilica (NS) could refine the size of the pores and densify the interfacial transition zone between the cement matrix and the aggregate, which would increase the compressive strength of RC. Similar phenomena were also reported by other scholars (Gao et al., 2020; Mohammed et al., 2018). However, almost no study had been conducted on the combined effect of the SF and NS on the properties of RC. In this study, the effects of the SF volume fraction and NS replacement ratio on the flexural properties of RC were investigated. The combined effects of SF and NS on the flexural behavior of RC were systematically analyzed, which provided a basis for the application of RC with the SF and NS.

Experimental program

Materials

Portland cement labeled 42.5 MPa was used in this study, which met GB175 (2007). The nanosilica (NS) was white powder with an average particle size of 30 nm and an apparent density of 40–60 g/L. The chemical compositions and physical properties of the cement and NS are given in Tables 1 and 2, respectively. The X-ray diffraction (XRD) analysis of NS is shown in Figure 1. It could be observed that a broad peak (2θ deg. varied from 15° to 40°) was obtained, which indicated that the NS was amorphous in nature with a high percentage of SiO₂ (Khaloo et al., 2016). The hook-shaped steel fiber with a length of 35 mm was adopted, which had an aspect ratio of 65 and a tensile strength of 1345 MPa. The coarse aggregate (CA) obtained from limestone had a particle size of 5–20 mm, and the fine aggregate (FA) extracted from riversides had a fineness modulus of 3.0, which were compatible with GB/T14684-2011 (2011). The rubber particle size was 1–2 mm, which was made by the mechanical crushing of the discarded rubber tire. The particle size distributions of the FA, CA, and rubber particles are shown in Figure 2. A commercially available naphthalene-based superplasticizer (SP) with a solid content of 30% and a water-reducing rate of 25% was used to achieve the required workability of the concrete mixes. The amount of the SP was determined as 1.0% weight of cement to ensure the slump of concrete mixtures was within the range of 20–70 mm. The used materials are shown in Figure 3.

Table 1.

Chemical compositions of cement.

Item	Quantity
Sulphate, SO₃ (%)	2.26
Magnesium oxide, MgO (%)	2.21
Chloride ion, Cl (%)	0.015
Specific surface area (m²/kg)	350
Specific gravity (Kg/m³)	3100
Initial setting time (min)	172
Final setting time (min)	230
Loss on ignition (%)	3.82

Table 2.

Chemical compositions of nanosilica (NS).

Item	Quantity
Appearance	High-dispersive white powder
Purity (%)	≥99.5
Diameter (nm)	30
Specific surface area (m²/g)	200 ± 10
Density (g/L)	40–60
PH value	5.0–7.0

Figure 1.

X-ray diffraction pattern of nanosilica.

Figure 2.

Size grading for fine aggregate, coarse aggregate, and rubber aggregate.

Figure 3.

Materials.

Concrete mixture proportions

Six types of concrete with different mixture proportions were produced in the laboratory. For each type of concrete, the rubber particle content and water-to-cement ratio were fixed at 50 kg/m³ and 0.422, respectively. The SF volume fractions varied from 0% to 1.5%. According to a previous study, considering the specimens with a SF volume fraction of 1.0% exhibited excellent working performance and mechanical properties (Gao et al., 2020). Therefore, the specimen fixed at a SF volume fraction of 1.0%, the replacement ratios of cement by NS were set to 0%, 1%, and 2% in weight, respectively. In this study, the concrete with the SF and NS was referred to steel fiber-and-nanosilica reinforced rubber concrete (SFNS-RC). For instance, the SF1NS1-RC represented that the RC had a SF volume fraction of 1% and a NS replacement ratio of 1%. Details of the mixture proportions of SFNS-RC are listed in Table 3.

Table 3.

Mixture proportions of steel fiber-and-nanosilica reinforced rubber concrete (SFNS-RC) (unit: kg/m³).

Mixture	Number of specimens	Rubber	Cement	CA	FA	Water	SP	SF	NS
SF0NS0-RC	3	50	400	703	1004	169	4	0	0
SF0.5NS0-RC	3	50	419	705	1060	177	4.5	39	0
SF1NS0-RC	3	50	438	654	1045	185	5	78	0
SF1.5NS0-RC	3	50	457	604	1029	193	5	117	0
SF1NS1-RC	3	50	433.62	654	1045	185	5	78	4.38
SF1NS2-RC	3	50	429.24	654	1045	185	5	78	8.76

FA: fine aggregate; CA: coarse aggregate; NS: nanosilica; SF: steel fiber; SP: superplasticizer.

The SFNS-RC mixtures were mixed through a shaft mixer. First, all aggregates and SFs were put together and mixed for 2 min. Then, the rubbers and cement were added and mixed for another 2 min. Finally, the water and SP were added to the mixture and mixed for another 2 min. Particularly, for the mixture proportion including NS, the NS was added to water in advance and stirred uniformly. All specimens were cured at the curing room with a temperature of 25°C ± 2°C and relative humidity of 95% RH for 28 days.

Test program

Flexural test

The prismatic specimens with a dimension of 100 mm × 100 mm × 400 mm were tested under flexure by using a 500 kN capacity servo controlled material testing system to obtain the flexural load-deflection curves of SFNS-RC, as shown in Figure 4. A total of 18 prismatic specimens were prepared in this study, in which three prismatic specimens for each group were employed to obtain an average value. In accordance with ASTM C1609/C1690M-19a (2019), all prismatic specimens were loaded with the displacement control at a rate of 0.1 mm/min. Two linear variable differential transformers (LVDTs) were attached at the mid-span on both sides of the specimen with the aid of a steel frame. According to the ASTM C1609/C1690M-19a (2019), the whole testing process terminated when the deflection reached L/150 mm.

Figure 4.

Specimen and test setup.

Microstructure test

In the microstructure test, the fragments of SFNS-RC obtained from the specimens in the flexural test were coated with gold for the scanning electron microscope (SEM) to reveal the evolution of flexural characteristics from the microstructure characteristics. The ZEISS EVO HD15 SEM was used for morphological investigations of the SFNS-RC matrix.

To investigate the formed phase composition of the SFNS-RC matrices, the X-ray diffraction (XRD) test was conducted for the three types of the SFNS-RC mixtures with different NS replacement ratios, and the powder samples obtained from the fragments of the SFNS-RC specimens were tested to evaluate exactly the phase composition of the matrix paste where the chemical reactions of cementitious materials occurred. The XRD measurements were performed in the scanning range of 2θ deg. From 5° to 80°, which had a step size of 0.1°/sec and a scanning speed of 0.007°/sec.

Results and discussion

Flexural failure mode

Effects of steel fiber on the flexural failure mode

Figure 5(a) shows the effects of the SF volume fraction on the flexural failure modes of SFNS-RC. For the specimen without SF and NS (SF0NS0-RC), the catastrophic brittle failure took place with one crack after concrete cracking. For the specimens with SFs, in the initial loading stage, the concrete and SFs resist the load by their integrated action. With the increase of the load, the concrete could not bear the load due to the cracks, and the SF began to debond along with pulling out. Therefore, significant ductile failure characteristics could be observed, and the width and depth of the main crack gradually decreased as the SF volume fraction increased, which indicated that the SF could improve the anti-cracking performance of SFNS-RC under the bending load. Furthermore, abundant tiny vertical cracks around the main crack could also be found, and the distance of cracks decreased as the SF volume fraction increased. Until the end of the test, all specimens with SFs remained intact. For all specimen, the failure occurred in the moment constant zone near the middle of the specimens, where no SFs breakage failure were observed. Similar phenomena could be found in related literature (Park et al., 2014).

Figure 5.

Flexural failure modes of steel fiber-and-nanosilica reinforced rubber concrete with different (a) steel fiber volume fractions and (b) nanosilica replacement ratios.

Effects of nanosilica on the flexural failure mode

Figure 5(b) shows the effects of the NS replacement ratio on the flexural failure modes of SFNS-RC specimens with SF volume fraction of 1%. The width of the main crack and the degree of spalling of the cement mortar gradually decreased as the NS replacement ratio increased. Particularly, the small matrix cracks were not observed on the specimen surface when the NS replacement ratio was 2%. The main reason might be that more C-S-H gel was produced, and Ca(OH)₂ was consumed by NS, which could densify the microstructure of concrete and improve the bond behavior between the SFs and the matrix during the SFs pulling out process, as shown in Figures 16 and 17. This indicated that positive synergy between the SF and NS could enhance the cracking resistance of SFNS-RC.

Load-deflection curves

Effects of steel fiber on the load-deflection curve

Figure 6(a) shows the effects of the SF volume fraction on the load-deflection curve of SFNS-RC. For the specimen without SF and NS (SF0NS0-RC), the load dropped suddenly with increasing deflection after the peak load was reached, which indicated that the SF0NS0-RC specimen exhibited significant brittleness. For the specimens with SFs, the ductility and excellent energy dissipation characteristics could be obviously observed, and the load-deflection curve could be roughly divided into three stages. In the elastic stage (OA), the load increased linearly as the deflection increased. Due to the small deformation in this stage, the SFs in the concrete could not make a significant contribution to bend resistance. In the elastic-plastic stage (AB), the load increased nonlinearly with increasing deflection until the first-peak load was reached at the deformation of approximately 0.04, and the first-peak load increased as the SF volume fraction increased. In the softening stage (BC), for the specimens with a SF volume fraction of less than 1%, the load decreased with increasing deflection until a certain load value was reached. However, for the specimens with a SF volume fraction of 1.5%, the load showed a process of slowly increasing (BD) and then decreasing (DE) when the curve passed the first-peak load, the similar result could be found in previous research (Fu et al., 2019). The main reason was that the curve had a process of deflection-hardening due to the reinforcement of the SFs (Zhang et al., 2017). Thus, it was concluded that the majority of the bend strength was contributed by the SFs in the SFNS-RC.

Figure 6.

Load-deflection curve of steel fiber-and-nanosilica reinforced rubber concrete with different (a) steel fibers volume fractions and (b) nanosilica replacement ratios.

Effects of nanosilica on the load-deflection curve

Figure 6(b) shows the effects of the NS replacement ratio on the load-deflection curves of SFNS-RC specimens with a SF volume fraction of 1%. The load-deflection curves of SFNS-RC consisted of three characteristic stages, including the linear growth portion (OA), nonlinear growth portion (AB) and descending portion (BC). The ascending and descending branches of the curves were similar for all specimens. Moreover, all load-deflection curves displayed a double peak response and straining-softening behaviors nearly parallel. Particularly, for the specimens with NS replacement ratio of 2%, the second peak load on the curves was greater than the first-peak load. The main reason could be that the ITZ between the cement matrix and SFs was strengthened and densified due to the increasing pozzolanic reaction by NS (Mukharjee et al., 2014), as shown in Figures 16 and 17, which further demonstrated that the NS improved the bond behavior between the SFs and the matrix.

Indexes of the flexural behavior

The flexural behavior of SFNS-RC could generally be classified as either deflection-softening or deflection-hardening (Kim et al., 2008). According to the ASTM C1609/C1690M-19a (2019) and CECS 13-2009 (2009), as illustrated in Figure 7, several indexes on the load-deflection (flexural stress-deflection) curve were used in this study to describe the flexural behaviors as follows.

f_{i} = \frac{P_{i} L}{b h^{2}}

(1)

Where f_i is the calculated flexural strength of SFNS-RC at special point of the load-deflection curve (Mpa), for example, f₁ is the first-peak flexural strength at the first-peak deflection δ₁, f_p is the peak flexural strength at the peak-load deflection δ_p,

f_{600}^{100}

and

f_{150}^{100}

are the residual flexural strength at deflection of L/600 and L/150, respectively; P_i is the load corresponding to the calculated flexural strength (N); L is the span of the specimen (mm); b is the width of the specimen (mm); h is the height of the specimen (mm).

f_{e, 150} = \frac{150 \times T_{150}^{100}}{b h^{2}}

(2)

R_{T, 150}^{100} = \frac{150 \times T_{150}^{100}}{f_{1} b h^{2}} \times 100 %

(3)

where f_e,150 is the equivalent flexural strength (MPa);

T_{150}^{100}

is the area under the load-deflection curve up to the deflection L/150 (J);

R_{T, 150}^{100}

is the equivalent flexural strength ratio (%).

Figure 7.

Definition of the indexes used in this study.

The indexes of the flexural behavior calculated by equations (1)–(3) according to the experimental results are summarized in Table 4. Each value represents the average result of three specimens in the same group.

Table 4.

Indexes of flexural behavior.

Mixture	f₁ (MPa)	δ₁ (mm)	f_p (MPa)	δ_p (mm)	$f_{150}^{100}$ (J)	$R_{T, 150}^{100}$ (%)	$f_{e, 150}$ (MPa)	$f_{600}^{100}$ (MPa)	$f_{150}^{100}$ (MPa)
SF0NS0-RC	5.47	0.033	5.47	0.033	5.73	15.7	0.86	1.33	0
SF0.5NS0-RC	5.58	0.039	5.58	0.039	18.59	50.0	2.79	3.59	2.44
SF1NS0-RC	6.21	0.044	6.17	0.044	24.71	59.7	3.71	4.50	2.78
SF1.5NS0-RC	6.42	0.044	8.10	0.283	47.96	112.0	7.19	7.98	6.15
SF1NS1-RC	6.17	0.068	6.21	0.068	33.25	80.9	4.99	5.72	6.15
SF1NS2-RC	6.22	0.058	6.77	0.311	38.65	93.2	5.80	6.52	3.83

First-peak flexural strength

Effects of steel fiber on the first-peak flexural strength

In this study, the first-peak flexural strength of SFNS-RC specimens increased from 5.47 MPa to 6.42 MPa with increasing SF volume fraction from 0% to 1.5%, as shown in Table 4. The main reason might be that the SF inhibited the production of the micro cracks. Fu et al. (2019) found that the first-peak flexural strength was influenced by both rubber and SF. In order to analyze the effect of the SF on the first-peak flexural strength, abundant test data were collected from this paper and previous researchers (Abaza et al., 2016; Fu et al., 2019; Gao et al., 2018; Soulioti et al., 2009), as plotted in Figure 8(a). It could be observed that the first-peak flexural strength had a linear growth tendency as the SF volume fraction increased, and the relationship between first-peak flexural strength and SF volume fraction could be described as follows (Gao et al., 2018).

f_{1} = f_{0} (1 + α v_{f})

(4)

Figure 8.

Relationship between the first-peak flexural strength of steel fiber-and-nanosilica reinforced rubber concrete and (a) steel fibers volume fraction and (b) nanosilica replacement ratio.

Where f₁ is the first-peak flexural strength of the specimen with SFs; f₀ is the first-peak flexural strength of the specimen without SFs; α is the characteristic coefficient; and v_f is the SF volume fraction. α could be determined through fitting analysis, as shown in Figure 8(a). For the specimens with a rubber content of less than 15%, the value of α varied from 0.10 to 0.18, and the average value was 0.11. However, for the specimens with a rubber content of more than 15%, the value of α varied from 0.04 to 0.06, and the average value was 0.05. This indicated that the influence of SF volume fraction on the first-peak flexural strength gradually decreased as the rubber content increased. The main reason might be the poor bonding performance and weak ITZ between SF and concrete due to the nonpolarity and hydrophobicity of rubber particles (Fu et al., 2019).

Effects of nanosilica on the first-peak flexural strength

Figure 8(b) shows the relationship between the first-peak flexural strength and the NS replacement ratio in this study, this series of specimens had a SF volume fraction of 1%. It could be clearly found that the value of the first-peak flexural strength remained approximately 6.20 MPa, which meant that the NS had no influence on the first-peak flexural strength regardless of SFs content. Thus, the influence of the NS on the overall the first-peak flexural strength should have been minor compared to that of the SF volume fraction.

Figure 9 presents the comparison of the first-peak flexural strength calculated by equation (4) and one obtained by the test. The average value and variation coefficient of the ratio of experimental results to predicted results were 1.03 and 0.099, respectively. It indicates that the predicted results are in good agreement with the experimental ones.

Figure 9.

Comparison of the first-peak flexural strength between test and calculated.

Peak flexural strength

Effects of steel fiber on the peak flexural strength

From the experimental results in Table 4, it could be observed that the peak flexural strengths of the specimens with a SF volume fraction of 0.5%, 1.0%, and 1.5% were 2.0%, 12.8%, and 48.1% more than that of the specimen without SFs, respectively. The reason for the increase in peak flexural strength might be owing to that the SFs continuously carried the load after matrix cracking (Li et al., 2018), as analyzed above. Noticeably, for the specimens with a SF volume fraction of 1.5%, the peak flexural strength has a prominent increase. The main reason was that the SF1.5NS0-RC curve had a process of deflection-hardening due to the reinforcement of the SFs.

To analyze the effect of SF on the peak flexural strength, abundant test data were collected from this paper and previous researchers (Duzgun et al., 2005; Fu et al., 2019; Gao et al., 2018; Guler et al., 2019; Ismail et al., 2017a; Li et al., 2017; Soulioti et al., 2009; Thomas et al., 2007; Yazc et al., 2007), which related to the hook-shaped steel fiber reinforced concrete, hook-shaped steel fiber reinforced self-compacting concrete, hook-shaped steel fiber reinforced lightweight concrete and hook-shaped steel fiber reinforced rubber concrete. Figure 10(a) presents the relationship of the normalized peak flexural strength and fiber reinforcing index $R I (R I = v f l_{f} / d_{f})$ , of which the normalized peak flexural strength is the ratio of the peak flexural strength of SFNS-RC to that of RC. It could be seen that normalized flexural strength had a roughly linear growth tendency as the RI increased as follows.

f_{p} = f_{p 0} (1 + λ R I)

(5)

where

f_{P}

, is the peak flexural strength of SFNS-RC;

f_{P 0}

is the flexural strength of RC; RI is the fiber reinforcing index,

R I = v_{f} l_{f} / d_{f}; v_{f}

is the volume fraction of steel fiber;

l_{f} / d_{f}

is the aspect ratio of steel fiber; λ is a characteristic coefficient, which was determined through fitting analysis of experimental results, and was 0.83. The equation (5) fits well with the experimental data, as shown in Figure 10(a).

Figure 10.

Relationship between normalized peak flexural strength and (a) fiber reinforcing index and (b) nanosilica replacement ratio.

Effects of nanosilica on the peak flexural strength

From the experimental results in Table 4, it could be observed that the peak flexural strengths of the specimens with a NS replacement ratio of 1% and 2% were 0.6% and 9.7% more than that of the SF1NS0-RC specimen, respectively. The main reason could be that the ITZ between the cement matrix and SFs was strengthened and densified due to the increasing pozzolanic reaction by NS (Mukharjee et al., 2014), as shown in Figures 16 and 17.

Based on the test data from this paper and previous researchers (Adamu et al., 2017; Amin et al., 2015; Gao et al., 2018; Givi et al., 2011; Li et al., 2015; Mohamed et al., 2016; Mohammed et al., 2018; Mukharjee et al., 2014; Zhang et al., 2014), which related to the steel fiber-and-nanosilica reinforced concrete, nanosilica reinforced high strength concrete and nanosilica modified rubber concrete, the relationship of the normalized peak flexural strength and the NS replacement ratio was summarized in the Figure 10(b). It could be seen that the normalized flexural strength increased linearly as the NS replacement ratio increased. Therefore, the flexural strength of SFNS-RC could be calculated as follows.

f_{p} = f_{p 0} (1 + λ R I) (1 + η v_{n})

(6)

Where v_n is the NS replacement ratio; η is a characteristic coefficient, which could be determined through fitting analysis of experimental results, and was 0.037. Therefore, the flexural strength could be calculated as follows.

f_{p} = f_{p 0} (1 + 0.83 R I) (1 + 0.037 v_{n})

(7)

Figure 11 presents the comparison between the peak flexural strength calculated by equation (7) and one obtained by the test. The average value and variation coefficient of the ratio of experimental results to predicted results by equation (7) were 1.02 and 0.097, respectively. The predicted results were within 85%–115% of the experimental results, which indicated that equation (7) had an excellent accuracy to predict the flexural strength of SFNS-RC.

Figure 11.

Comparison of the peak flexural strength between test and calculated.

Equivalent and residual flexural strength

Figure 12(a) presents the effects of the SF volume fraction on the equivalent and residual flexural strength of SFNS-RC. The equivalent flexural strength $f_{e, 150}$ , residual flexural strength $f_{600}^{100}$ and $f_{150}^{100}$ of SFNS-RC increased significantly as the SF volume fraction increased due to the SF bridged effect after matrix cracking, and similar experimental results could be found in related literature (Li et al., 2017). Figure 12(b) draws the effects of the NS replacement ratio on the equivalent and residual flexural strength of SFNS-RC, this series of specimens had a SF volume fraction of 1%. It could be seen that the equivalent flexural strength $f_{e, 150}$ , residual flexural strength $f_{600}^{100}$ and $f_{150}^{100}$ of SFNS-RC increased linearly as the NS replacement ratio increased. A linear correlation existed between the equivalent and residual flexural strength parameter, as represented in Figure 13, which agreed with results obtained by other authors (Barros et al., 2005).

Figure 12.

Equivalent and residual flexural strength of steel fiber-and-nanosilica reinforced rubber concrete with different (a) SF volume fractions and (b) nanosilica replacement ratios.

Figure 13.

Relationship between equivalent and residual flexural strength.

Flexural toughness

Figure 14(a) shows the effects of the SF volume fraction on the flexural toughness and equivalent flexural strength ratio of SFNS-RC. With increasing SF volume fraction from 0% to 0.5%, 1.0%, and 1.5%, the $T_{150}^{100}$ increased markedly from 5.73 J to 18.59 J, 24.71 J, and 47.96 J, respectively. The $R_{T, 150}^{100}$ increased observably from 15.7% to 50.0%, 59.7%, and 112.0%, respectively. This was attributed to the fact that SFs could inhibit the development of cracks in concrete and hinder the development of macro-cracks. The stress concentration in SFNS-RC could be avoided by changing the direction of major fractures through the bridging action of SFs (Mohammed et al., 2016).

Figure 14.

Flexural toughness of steel fiber-and-nanosilica reinforced rubber concrete with different (a) SF volume fractions and (b) NS replacement ratios.

Figure 14(b) shows the effects of the NS replacement ratio on the flexural toughness and equivalent flexural strength ratio of SFNS-RC, this series of specimens had a SF volume fraction of 1%. With increasing NS replacement ratio from 0% to 1% and 2%, the $T_{150}^{100}$ increased steadily from 24.71 J to 33.25 J and 38.65 J, respectively. The $R_{T, 150}^{100}$ increased gradually from 59.7% to 80.9% and 93.2%, respectively, which meant that both the SFs and NS could substantially enhance energy absorption and dissipation of SFNS-RC. This phenomenon indicated that the high pozzolanic reactivity of NS could enable it to react and consume the surplus Ca(OH)₂ and produce extra C-S-H gel which was the main compound for the strength densification of the ITZ between the cement matrix and SFs.

Microscopic properties

Surface characteristics

Figure 15 exhibits the surface characteristics of SF1NS1-RC specimen. It could be found that abundant rubber particles rose to the top of the concrete, which could form a weak area on the upper surface because of their uneven distribution, as shown in Figure 15(a). Similar phenomenon could be found by other scholars (Li et al., 2014; Fakhri et al., 2016). In addition, plentiful stomata were observed on the side and bottom surface of the concrete due to the nonpolarity and hydrophobicity of rubber particles, as shown in Figure 15(b) and (c).

Figure 15.

Surface of SF1NS1-RC specimen: (a) upper surface; (b) side surface; and (c) bottom surface.

SEM analysis

Figure 16 presents the microstructures of the matrix at the interfacial transition zone (ITZ) captured by SEM. The porous paste structures at ITZ for the specimen without NS were shown in Figure 16(a), where the crystalline calcium hydroxide Ca(OH)₂, amorphous calcium-silicate-hydrate (C-S-H) gel and stomata with different sizes could be clearly recognized. For the specimen with a NS replacement ratio of 1%, a more uniform and dense microstructure could be obtained due to the conversion of Ca(OH)₂ to C-S-H under the action of NS, as shown in Figure 16(b).

Figure 16.

Microstructure morphologies at the interfacial transition zone: (a) SF1NS0-RC (b) SF1NS1-RC.

XRD analysis

Figure 17 illustrates the XRD patterns of SFNS-RC with different NS replacement ratios. The main peaks observed were calcium hydroxide (Ca(OH)₂, 2θ deg. 18.1, 28.7, 47.1, and 50.1), alite (C₂S, 2θ deg. 32.1, 34.4, and 41.2), belite (C₃S, 2θ deg. 29.3, 32.5, and 34.4), quartz (2θ deg. 20.8, 26.6, 36.5, 39.4, 42.4, 45.7, and 54.8) and ettringite (2θ deg. 8.9) (Hakamy et al., 2013; Kontoleontos et al., 2012). As shown in Figure 17, the Ca(OH)₂, C₂S, and C₃S peaks steadily reduced as the NS replacement ratio increased. For instance, with increasing NS replacement ratio from 0% to 1% and 2%, a peak of Ca(OH)₂ (2θ deg. 18.1) decreased steadily from 597 to 388 and 252, respectively. A peak of C₂S (2θ deg. 41.2) reduced gradually from 330 to 255 and 208, respectively. A peak of C₃S (2θ deg. 29.3) diminished equably from 405 to 328 and 282, respectively. This was attributed to the NS could produce pozzolanic activity and accelerate cement hydration (Zhang et al., 2017). In addition, the CaO·SiO₂·H₂O (C-S-H) gel was the hydration product of SiO₂, which was obtained by the following reaction (Khaloo et al., 2016).

{SiO}_{2} + {mH}_{2} O + nCa {(OH)}_{2} \to nCaO \cdot {SiO}_{2} \cdot (m + n) H_{2} O

(8)

Figure 17.

X-ray diffraction pattern of the steel fiber-and-nanosilica reinforced rubber concrete with different NS replacement ratios.

On the one hand, more C-S-H gel was produced during the hydration reaction, which improved the density of the matrix; on the other hand, more Ca(OH)₂ were refined and consumed by NS. These newly generated products could be crystallized into fine particles to fill the pores of SFNS-RC, making the microstructure of composites dense and uniform (Quercia et al., 2014).

Conclusions

In this study, a series of experiments were conducted to investigate the effects of the SF volume fraction and NS replacement ratio on the flexural properties of the RC. The following conclusions could be drawn:

The peaks of Ca(OH)2, C3S, and C2S in SFNS-RC steadily reduced as the NS replacement ratio increased, respectively, and the microstructure of SFNS-RC became more compact and uniform because the NS filled the pores of the matrix and strengthened the bond behavior between the SF and matrix.

The SF and NS could significantly improve flexural behaviors of SFNS-RC, including the peak flexural strength, equivalent flexural strength, residual flexural strength and flexural toughness of SFNS-RC. However, the SF had a slight influence on the first-peak flexural strength and the NS had almost no effect on it.

The SFNS-RC specimens containing SFs and NS could resist preferably generation and development of macro and micro cracks, and exhibit obvious ductility and excellent energy dissipation characteristics. Finally, the empirical formulae for predicting the first-peak flexural strength and peak flexural strength of SFNS-RC were proposed through the analysis of experimental data, respectively.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the financial support of the National Natural Science Foundation of China (NO. 52008377, U1704254, and 51808508).

ORCID iD

Tao Zhang

References

Abdelaleem

Ismail

Hassan

AAA

(2018) The combined effect of crumb rubber and synthetic fibers on impact resistance of self-consolidating concrete. Construction and Building Materials 162: 816–829.

Abaza

Hussein

(2016) Flexural behavior of steel fiber-reinforced rubberized concrete. Journal of Materials in Civil Engineering 28(1) 04015076.1–04015076.10.

Adamu

Mohammed

Shafiq

(2017) Flexural performance of nano silica modified roller compacted rubbercrete. International Journal of Advanced and Applied Sciences 4(9): 6–18.

Alsaif

Koutas

Bernal

, et al. (2018) Mechanical performance of steel fibre reinforced rubberised concrete for flexible concrete pavements. Construction and Building Materials 172(30): 533–543.

Al-Tayeb

Bakar

BHA

Ismail

, et al. (2012) Impact resistance of concrete with partial replacements of sand and cement by waste rubber. Polymer-Plastics Technology and Engineering 51(12): 1230–1236.

Amin

Abu El-Hassan

(2015) Effect of using different types of nano materials on mechanical properties of high strength concrete. Construction and Building Materials 80: 116–124.

Antil

Verma

Singh

(2012) Rubberized concrete made with crumb rubber. International Journal of Science and Research 3(5): 1481–1483.

Aslani

(2016) Mechanical properties of waste tire rubber concrete. Journal of Materials in Civil Engineering 28(3).

ASTM C1609/C1690M-19a (2019) Standard Test Method for Flexural Performance of Fiber Reinforced Concrete.

10.

Barros

JAO

Cunha

VMCF

Ribeiro

, et al. (2005) Post-cracking behaviour of steel fibre reinforced concrete. Materials and Structures 38(1): 47–56.

11.

CECS 13-2009 (2009) Test Methods for Fiber Reinforced Concrete.

12.

Düzgün

Gül

Aydin

(2005) Effect of steel fibers on the mechanical properties of natural lightweight aggregate concrete. Materials Letters 59(27): 3357–3363.

13.

Wang

, et al. (2019) Evolution of mechanical properties of steel fiber-reinforced rubberized concrete (FR-RC). Composites Part B: Engineering 160: 158–166.

14.

Fakhri

Saberi

(2016) The effect of waste rubber particles and silica fume on the mechanical properties of roller compacted concrete pavement. Journal of Cleaner Production 129: 521–530.

15.

GB175 (2007) Common Portland Cement.

16.

GB/T14684-2011 (2011) Sand for Construction.

17.

Gao

Zhao

Chen

(2018) Flexural Behavior of Fiber and Nanoparticle Reinforced Concrete at High temperaturesFire and Materials.

18.

Gao

Zhang

Wang

, et al. (2020) Analysis and prediction of compressive properties for steel fiber-and-nanosilica-reinforced crumb rubber concrete. Advances in Civil Engineering 2020(2): 1–15.

19.

Ganjian

Khorami

Maghsoudi

(2009) Scrap-tyre-rubber replacement for aggregate and filler in concrete. Construction and Building Materials 23(5): 1828–1836.

20.

Gupta

Chaudhary

Sharma

, et al. (2016) Mechanical and durability properties of waste rubber fiber concrete with and without silica fume. Journal of Cleaner Production 112: 702–711.

21.

Guler

Yavuz

Korkut

, et al. (2019) Strength prediction models for steel, synthetic, and hybrid fiber reinforced concretes. Structural Concrete 20(1): 428–445.

22.

Naji Givi

Abdul Rashid

Aziz

FNA

, et al. (2011) The effects of lime solution on the properties of SiO2 nanoparticles binary blended concrete. Composites Part B: Engineering 42(3): 562–569.

23.

Hakamy

Shaikh

Low

(2013) Microstructures and mechanical properties of hemp fabric reinforced organoclay-cement nanocomposites. Construction and Building Materials 49(12): 298–307.

24.

Ismail

Hassan

(2016a) Impact resistance and acoustic absorption capacity of self-consolidating rubberized concrete. Aci Materials Journal 113(6): 725–736.

25.

Ismail

Hassan

AAA

(2016b) Use of metakaolin on enhancing the mechanical properties of self-consolidating concrete containing high percentages of crumb rubber. Journal of Cleaner Production 125(1): 282–295.

26.

Ismail

Hassan

AAA

(2017a) Use of steel fibers to optimize self-consolidating concrete mixtures containing crumb rubber. Aci Materials Journal 114(4): 581–594.

27.

Ismail

Hassan

(2017b) Impact resistance and mechanical properties of self-consolidating rubberized concrete reinforced with steel fibers. Journal of Materials in Civil Engineering 29(1). 04016193.1–04016193.14.

28.

Kontoleontos

Tsakiridis

Marinos

, et al. (2012) Influence of colloidal nanosilica on ultrafine cement hydration: Physicochemical and microstructural characterization. Construction and Building Materials 35: 347–360.

29.

Khaloo

Mobini

Hosseini

(2016) Influence of different types of nano-SiO2 particles on properties of high-performance concrete. Construction and Building Materials 113(15): 188–201.

30.

Kim

Naaman

El-Tawil

, et al. (2008) Comparative flexural behavior of four fiber reinforced cementitious composites. Cement and Concrete Composites 30(10): 917–928.

31.

Chi

, et al. (2018) Experimental investigation on the flexural behavior of steel-polypropylene hybrid fiber reinforced concrete. Construction and Building Materials 191: 80–94.

32.

(2017) Experimental study on performance of rubber particle and steel fiber composite toughening concrete. Construction and Building Materials 146: 267–275.

33.

Ruan

Zeng

(2014) Mechanical properties and constitutive equations of concrete containing a low volume of tire rubber particles. Construction and Building Materials 70: 291–308.

34.

Huang

Cao

, et al. (2015) Effects of nano-silica and nano-limestone on flowability and mechanical properties of ultra-high-performance concrete matrix. Construction and Building Materials 95: 366–374.

35.

Mohammed

Awang

Wong

, et al. (2016) Properties of nano silica modified rubbercrete. Journal of Cleaner Production 119: 66–75.

36.

Mukharjee

Barai

(2014) Influence of nano-silica on the properties of recycled aggregate concrete. Construction and Building Materials 55(3): 29–37.

37.

Mohammed

Adamu

(2018) Mechanical performance of roller compacted concrete pavement containing crumb rubber and nano silica. Construction and Building Materials 159: 234–251.

38.

Mohamed

(2016) Influence of nano materials on flexural behavior and compressive strength of concrete. Hbrc Journal 12(2): 212–225.

39.

Nguyen

Toumi

Turatsinze

(2010) Mechanical properties of steel fibre reinforced and rubberised cement-based mortars. Materials & Design 31(1): 641–647.

40.

Noaman

Abu Bakar

Akil

(2016) Experimental investigation on compression toughness of rubberized steel fibre concrete. Construction and Building Materials 115: 163–170.

41.

Noaman

Abu Bakar

Akil

, et al. (2017) Fracture characteristics of plain and steel fibre reinforced rubberized concrete. Construction and Building Materials 152(15): 414–423.

42.

Ozbay

Lachemi

Sevim

, et al. (2011) Compressive strength, abrasion resistance and energy absorption capacity of rubberized concretes with and without slag. Materials and Structures 44(7): 1297–1307.

43.

Onuaguluchi

(2015) Effects of surface pre-coating and silica fume on crumb rubber-cement matrix interface and cement mortar properties. Journal of Cleaner Production 104(1): 339–345.

44.

Park

Abolmaali

Mohammadagha

, et al. (2014) Flexural characteristic of rubberized hybrid concrete reinforced with steel and synthetic fibers. Advances in Civil Engineering Materials 3(1): 495–508.

45.

Quercia

Spiesz

Hüsken

, et al. (2014) SCC modification by use of amorphous nano-silica. Cement and Concrete Composites 45(1): 69–81.

46.

Sofi

(2017) Effect of waste tyre rubber on mechanical and durability properties of concrete-a review. Ain Shams Engineering Journal 9(4): 1–10.

47.

Siddika

Mamun

MAA

Alyousef

, et al. (2019) Properties and utilizations of waste tire rubber in concrete: A review. Construction and Building Materials 224(10): 711–731.

48.

Soulioti

Barkoula

Paipetis

, et al. (2009) Effects of fibre geometry and volume fraction on the flexural behaviour of steel-fibre reinforced concrete. Strain 47(1): 535–541.

49.

Turatsinze

Granju

Bonnet

(2006) Positive synergy between steel-fibres and rubber aggregates: Effect on the resistance of cement-based mortars to shrinkage cracking. Cement and Concrete Research 36(9): 1692–1697.

50.

Thomas

Ramaswamy

(2007) Mechanical properties of steel fiber-reinforced concrete. Journal of Materials in Civil Engineering 19(5): 385–392.

51.

Yazc

Nan

Tabak

(2007) Effect of aspect ratio and volume fraction of steel fiber on the mechanical properties of SFRC. Construction and Building Materials 21(6): 1250–1253.

52.

Zheng

Huo

Yuan

, et al. (2008) Strength, modulus of elasticity, and brittleness index of rubberized concrete. Journal of Materials in Civil Engineering 20(11): 692–699.

53.

Zhang

Wan

Wang

, et al. (2017) Influence of nano-SiO 2 on properties of fresh and hardened high performance concrete: A state-of-the-art review. Construction and Building Materials 148(1): 648–658.

54.

Zhang

Zhao

Liu

, et al. (2014) Combined effect of nano-SiO2 particles and steel fibers on flexural properties of concrete composite containing fly ash. Science and Engineering of Composite Materials 21(4): 597–605.