Experimental and parametrical investigation of pre-stressed ultrahigh-performance fiber-reinforced concrete railway sleepers

Abstract

In this study, ultrahigh-performance fiber-reinforced concrete (UHPFRC) used in a type B70 concrete sleeper is investigated experimentally and parametrically. The main parameters investigated are the steel fiber volume fractions (0%, 0.5%, 1%, and 1.5%). Under European standards, 35 UHPFRC sleepers are subjected to static bending tests at the center and rail seat sections, and the screw on the fastening system is pulled out. The first cracking load, failure load, failure mode, crack propagation, load–deflection curve, load–crack width, and failure load from these tests are measured and compared with those of a control sleeper manufactured using normal concrete C50. The accuracy of the parametric study is verified experimentally. Subsequently, the results of the study are applied to UHPFRC sleepers with different concrete volumes to investigate the effects of the properties of UHPFRC on their performance. Experimental and parametric study results show that the behavior of UHPFRC sleepers improves significantly when the amount of steel fiber in the mix is increased. Sleepers manufactured using UHPFRC with a steel fiber volume fraction of 1% and a concrete volume less than 25% that of standard sleeper B70 can be used under the same loads and requirements, which contributes positively to the cost and surrounding environment.

Graphical abstract

Keywords

pre-stressed concrete sleeper / ultrahigh performance fiber-reinforced concrete / pull-out test / static bending test / steel fiber / aspect ratio / volume fraction

1 Introduction

Sleepers are extremely important components of railway track structures; they are essential for transferring loads from the rails to the ballast. The main functions of sleepers include monitoring the gauge between rails, fixing the rails in the appropriate position and level, and providing stability for the entire track system [1,2]. Wood, steel, and concrete are the three primary materials used for constructing sleepers. Wood and steel present several disadvantages that limit their use in the manufacture of railway sleepers. Timber and steel sleepers have been gradually replaced by reinforced and pre-stressed concrete sleepers [3]. The lifespan of hardwood timber sleepers is approximately 20 years, whereas the potential service life of concrete sleepers is more than 50 years [4,5]. Numerous experiments pertaining to improvements in design methods and loading patterns have been conducted to increase the design life of sleepers [6,7]. Ballast track components can be classified into two main groups: superstructures and substructures. The superstructure comprises rails, fastening systems, rail pads, and railway concrete sleepers, whereas the substructure comprises ground formation, subgrades, and ballasts [8,9]. The main elements of the rail track system are shown in Fig.1.

Steel fibers improve the structural performance of ultrahigh-performance fiber-reinforced concrete (UHPFRC) by increasing the tensile and compressive strengths and eliminating shrinkage cracks. Notably, the compressive strength of UHPFRC concrete is 4%–19% higher than that of concrete without reinforcement, whereas its flexural strength is 3%–81% higher. The most suitable volume fraction of steel fiber in a concrete mix is 0.5%–1.5% that of concrete [10,11]. End-hooked steel fibers offer better mechanical interlocking than other steel fibers. Applying these fibers to concrete mixtures results in higher tensile and compressive loads compared with applying straight fibers [12,13]. End-hooked and corrugated steel fibers should be combined to achieve the highest compressive and flexural strength [14]. Alternatively, a new recycling pathway can be considered that uses waste plastics to generate synthetic fibers, which are then incorporated into concrete. In a previous study, researchers discovered that the bonding, compressive, and splitting tensile strengths of concrete with 6 kg/m³ of fiber increased by 17.8%, 19.4%, and 41.9%, respectively, compared with those of plain concrete [15]. Meanwhile, other authors performed investigations using crumb rubber and glass fiber chopped strands (GFCSs). The crumb rubber in concrete decreased the strength but improved the flexibility. In addition, using GFCSs in concrete increased the porosity and density but decreased flow; this phenomenon has been shown to successfully enhance all mechanical properties [16]. Other authors investigated the effect of carbon nanofibers (CNFs) on ultrahigh-performance concrete (UHPC) and optimized the UHPC mix by selecting the most effective dispersion technique to enhance the mechanical behavior, workability, and permeability. By adding CNFs to UHPC, the maximum flexural strain and maximum flexural stress increased by up to 63% and 7.3%, respectively. Otherwise, the slump decreased by up to 10%; however, increasing the amount of high-range water reducer for dispersion can mitigate the decrease in the slump [17].

In this study, conventional concrete was replaced with UHPFRC in monoblock sleepers to improve their structural performance. After 28 d of curing, UHPFRC offers compressive and tensile strengths of approximately 150 and 20 MPa, respectively [18]. The substitution of coarse aggregates with fine aggregates in cementitious pastes, such as mortar mixes, improves the homogeneity of the mixture. This mix reduces defects such as microcracks and voids in concrete [19]. Three important procedures were performed to achieve the main characteristics of UHPFRC. First, the coarse aggregates in the matrix were minimized to confirm the homogeneity enhancement. Second, the gradation and mixture ratio between the main matrix constituents were optimized to achieve density enhancement. Third, the ductility was enhanced by incorporating fibers into the mixture [20]. Safdar et al. [21] repaired reinforced concrete (RC) beams by adding a layer of UHPFRC at the bottom and top of the beams, which resulted in a higher stiffness and bending capacity. Additionally, they reported an effective method to deform the surface of steel fibers using an electrolyte solution comprising ethylenediaminetetraacetic acid to increase the roughness of UHPC by approximately 10 times. The results showed that the tensile strength and the pull-out energies increased from 17%–36% and 26%–39% owing to increased roughness via surface modification [22]. Furthermore, the authors used dispersed basalt fibers for fabricating UHPC to achieve flexural and compressive strengths exceeding 10 and 120 MPa, respectively. The results showed that the UHPC fabricated presented excellent flowability and that its elastic modulus increased with the fiber content, reaching a value exceeding 50 GPa [23]. Recently, Bae and Pyo [24] fabricated a new type of concrete sleeper using a UHPFRC mix by minimizing the sleeper dimensions; subsequently, the concrete sleeper was tested based on the European standard [25]. In addition, the authors investigated the effect of incorporating steel fibers with different friction volumes on UHPFRC sleepers, and their experimental results satisfied European standards [26,27].

Previous studies show that UHPFRC is ideal for applications where high compressive and tensile strengths, small thicknesses, and high energy absorption capacities are required. The mechanical properties of UHPFRC are governed by the fiber volume fraction and raw material. According to the European standard, several varieties of pre-stressed concrete sleepers are acceptable for the shape, size, and volume of concrete with different train loads and speeds. The most widely used type in Egypt is the type B70 pre-stressed sleeper. Accordingly, UHPFRC is employed in this study to fabricate a standard pre-stressed concrete sleeper type (B70) for the first time using locally available materials. Since steel fibers affect the mechanical properties of UHPFRC mixtures, the structural behavior of UHPFRC railway sleepers with different amounts of steel fiber is investigated in this study based on European standards. In particular, UHPFRC sleepers fabricated using locally available materials, with different volume fractions (0%, 0.5%, 1%, and 1.5%), and with two steel fiber shapes (end hooked and corrugated) are investigated using sleepers manufactured with normal concrete C50. Based on the volume fraction of steel fiber, five categories of specimens are prepared. Subsequently, the structural behavior of UHPFRC sleepers is investigated via two static tests: 1) two sleepers from each category in the center section and three sleepers from each category in the rail seat section are tested based on EN 13230-2 [26]; 2) a vertical load test was performed on the fastening system to pull out the screw from two sleepers in each category based on EN 13481-2 [27]. Subsequently, a parametric study is conducted to examine the effects of the mechanical properties of UHPFRC on concrete sleepers. The accuracy of the parametric study is verified experimentally. The results are then applied to pre-stressed UHPFRC sleepers with different concrete volumes to investigate the effects of the properties of UHPFRC on their performance. Consequently, the optimum properties of UHPFRC for the manufacture of type B70 concrete sleepers are determined, based on which the volume of UHPFRC sleepers can be reduced to satisfy the design requirements of the sleeper, thus rendering the sleeper more economical and environmentally friendly.

2 Experimental method

2.1 Specimen description

Fig.2 shows the geometrical dimensions of monoblock sleepers (B70) with a rail-type UIC 54E1 and a Vossloh rail fastening system (W21) used for the experiment in this study. The gauge and length of the sleepers measured 1435 and 2400 mm, respectively. A cant of 1 in 20 was provided at the top surface of the rail seat. The sleepers were pre-stressed with four steel wires measuring 9.4 mm in diameter and 2337 mm in length, in accordance with EN 10138-2 [28]. The supporting surface area measured 6220 cm², and the estimated weight was 270 kg.

2.2 Raw material and mix design

Recently, UHPFRC has been further developed by several researchers. For example, a different mix of UHPFRC with fine aggregates was developed while preserving the key material advantages of UHPFRC, such as its high flexural and compressive strengths [29]. In that study, the mix design and material characterization were based on a study by Yu et al. [29]. The mix depends on particle packing theory as well as modified Andreasen and Andersen models. The characteristics of fibers, such as their aspect ratio and length, can change the tensile capacity of UHPFRC. Using two types of fibers has been reported to be the best solution for achieving better compressive strengths [14,30]. Hence, in this study, two types of steel fibers (50% end hooked and 50% corrugated) with varying volume fractions (0%, 0.5%, 1%, and 1.5%) were used (see Fig.3). The five mix designs used to cast UHPFRC sleepers and normal concrete sleepers (NC) in this study are shown in Tab.1. Each series name refers to a certain fiber volume percentage.

2.3 Fabrication of ultrahigh-performance fiber-reinforced concrete sleepers

All the specimens were cast in the Siegwart factory in Egypt. Fig.4 shows the mold and production process. Prior to casting the specimens, a mold was prepared via the following steps: 1) the form was lubricated; 2) wires were shielded with plastic ducts; 3) the anchor plate and form were tied with wires and dowels, respectively; and 4) prepare for casting the concrete. A bee mixer (a mixer comprising an oval-shaped bowl with an opening at the top driven by a power-tilting revolver) was poured into the mold.

Concrete casting was performed in layers using a vibrator to achieve complete compaction. The cast sleepers were stored in a steam room at 55 °C for 24 h and demolded after 24 h of curing [31]. The demolded monoblock concrete sleepers were air cured for an additional 24 h, and then 259 kN of pre-stressing force was applied to every wire. The specimens were air cured for at least 28 d before conducting any experimental tests. Four types of UHPFRC sleepers mixed with different amounts of steel fiber were prepared to investigate the effect of steel fiber content on the structural behavior of UHPFRC sleepers, particularly on the rail seat and center sections. The compressive strength values of the mixes designed based on 7 and 28 d of curing were evaluated using standard 15 cm × 15 cm cubes, and the results are shown in Fig.5. The specimens cured for 7 d indicated compressive strength values exceeding 80 MPa.

3 Experimental tests and discussions

3.1 Requirement for performance tests

Several structural experiments were conducted to ensure that the new structural design of the UHPFRC sleeper complied with EN 13481-2 [27] and EN 13230-2 [26] standards. Static bending tests were performed on the center and rail seat sections, which were the most critical sections. In these tests, loads were applied through a steel plate with the same width as that of the 54 UIC rail (E1). A rail pad, which provided a uniform stress distribution at the surface of the sleepers, was used to prevent local failure. Based on a train speed of 140 km/h with 240 kN of axel load and a sleeper spacing of 0.6 m (as stipulated in the Egyptian National Railways specification), the values of the negative design bending moment (M_dc–) and positive design bending moment (M_dr+) were estimated to be 10 and 13 kN·m, respectively, by referring to EN 13230-6 [25] and UIC 713R [32]. Subsequently, the reference test loads of the rail seat (F_r0) and center sections (F_c0n) can be estimated using Eqs. (1) and (2), respectively, as indicated in EN 13230-2 [26].

(1)F_{\text{r}0}=\frac{4M_{\text{dr}+}}{L_{\text{r}}-0.1}=\frac{4\times 13}{0.5-0.1}=130\text{kN},

(2)F_{\text{c}0\text{n}}=\frac{4M_{\text{dc}-}}{L_{\text{c}}-0.1}=\frac{4\times 10}{1.5-0.1}=28.6\text{kN}.

The space between the centers of the rail seat sections (L_c) was 1.5 m. The distance between the center of the rail seat section and the boundary of the sleeper (L_p) was 0.44 m, which is within the range of 0.4 m ≤ L_p ≤ 0.449 m. Therefore, the value of L_r (as shown in Fig.6) was 0.5 m, as indicated in EN 13230-2 [26]. Tab.2 summarizes all the calculated structural requirements for each sleeper position used to evaluate the sleepers. The static coefficient (k_2s) was 2.5, as defined in EN 13230-2 [26] and UIC 713R [32]. The reference load for the pull-out test of the fastening system (P₀) was 60 kN, as indicated in EN 13481-2 [27].

3.2 Static bending test at rail seat section

The rail seat is typically prioritized in the structural design of sleepers. The rail seat of the sleepers must demonstrate sufficient load resistance because the train load is immediately applied to the rail seat section through the rail and rail pad. Therefore, three sleepers from each mix, as defined in EN 13230-2 [26], were subjected to a positive bending moment at the rail seat section. In fact, the test results can be used to investigate the effect of fiber amount on UHPFRC sleepers. The test was performed on the rail seat sections, and the opposite end of the sleeper was not supported. Two plate supports and resilient pads were placed under the rail seat section with an L_r of 0.5 m, and a vertical load was applied to the rail seat section. Fig.6 shows the configuration adopted for this test, which was based on EN 13230-2 [26], where three linear variable differential transformers were installed on the sleepers.

The load in this test was applied as defined in EN 13230-2 [26], and the loading protocol for this test was classified into three main steps, similar to previous studies [33,34] (see Fig.7). First, the reference load F_r0 was attained by increasing the load at a load rate of 120 kN/min. Second, the load was increased by 10 kN at each step until the first crack measured 15 mm from the bottom of the sleeper. Third, the ultimate failure load F_rB was attained by increasing the load by 20 kN at each step.

The midspan deflection responses vs. the applied load and the relationship between the crack width and applied load based on the bending test are presented in Fig.8 and Fig.9, respectively. The typical failure modes for the S0, S0.5, S1, and S1.5 series are shown in Fig.10. The resulting values and failure modes are summarized in Tab.3. The sleepers in several series of mixes, which were composed of the same material, were designed to have the same capacity. However, their failure modes and resulting values were different. The difference in the failure modes is attributable to the complicated stress distribution over short spans [33].

The load ranges for the first crack F_rr were 255.7–287.9, 307.9–358.9, 367.1–376, and 390.4–401.1 kN for the S0, S0.5, S1, and S1.5 series, respectively, which were higher than the required load, 130 kN. In addition, the average values of flexural load F_rB for the S0, S0.5, S1, and S1.5 series were 422, 451.8, 491.5, and 541.1 kN, respectively, which were higher than the required load, 325 kN. Based on the results of the static bending test, all sleepers for each mix completely satisfied the requirements of European standards [26]. The normal concrete sleepers (SN group) exhibited failure loads ranging from 362.7 to 374.6 kN, where the first crack appeared at loads ranging from 274.8 to 301 kN. Group S0 was considered the weakest group as it did not contain any fiber. S0 specimens 1 and 3 exhibited compression failure at the top of the sleeper and shear failure under the loading plate, as shown in Fig.10(a). Group S0 indicated the largest average deflection and crack width, i.e., 18.2 and 10.9 mm, respectively, as shown in Fig.8 and Fig.9, respectively. For group S0.5, the first and third specimens (S0.5-No.1 and S0.5-No.3, respectively) exhibited an abrupt decrease in load because the wire was displaced from its original position and shear failure occurred under the loading plate, as shown in Fig.10(b). However, the second specimen indicated a decrease in load owing to the compression failure at the top of the sleeper and shear failure under the loading plate, as shown in Fig.10(a). In addition, the average deflection and crack width of this group were 16.7 and 8.4 mm, respectively, as shown in Fig.8 and Fig.9, respectively. For group S1, all three specimens demonstrated an abrupt decrease in load because of flexural-shear failure, as shown in Fig.10(c), with an average deflection of 14 mm and an average crack width of 6.8 mm, as shown in Fig.8. Finally, group S1.5 exhibited the same failure as group S1; additionally, the wire was removed from its original position in specimens 1 and 3, as shown in Fig.10(b). The average deflection and crack width for group S1.5 were 11.4 and 5.7 mm, respectively, as shown in Fig.8 and Fig.9, respectively. All previous results show that increasing the amount of steel fiber in the UHPFRC sleeper increases the stiffness and resistance to bending.

3.3 Static bending test at center section

The center section is one of the critical sections of sleepers. A negative bending moment must be applied to the center section of the sleeper while the train load is exerted on the rail seat components. According to EN 13230-2 [26], the center section of the sleeper must have an adequate impedance to the negative bending moment. Fig.11 shows the test setup at the center section and the support location at the rail seat section, where L_c is 1.5 m. The test load was applied continuously at the center section until the section no longer supported the F_cB. F_c0n is the reference test load, whose maximum value is up to 28.6 kN, based on Eq. (2). The test procedure, as illustrated in Fig.12, was based on EN 13230-2 [26].

Tab.4 presents the results of the tests for all sleepers. The average deflections and applied loads are shown in Fig.13. The relationship between the average crack width and average applied load for the specimens is shown in Fig.14. The failure mode at the center section for all specimens was flexure failure, which satisfied the minimum requirements of EN 13230-2 [26]. Prior to the initial cracking, all sleepers reacted similarly in the ascending branch of the loading response. Subsequently, the average first-crack loads for the SN, S0, S0.5, S1, and S1.5 series were 51, 63.4, 70, 81.2, and 99.9 kN, respectively, which were higher than the required load, 28.6 kN. Additionally, the average failure loads for the SN, S0, S0.5, S1, and S1.5 series were 91.6, 106.7, 118.3, 122.2, and 128.2 kN, respectively. All sleepers in this test completely satisfied the requirements of European standards [26]. The failure modes for each specimen are listed in Tab.4. All flexural cracks initiated from the bottom surface around the midspan, followed by flexural-tension failures with shear cracks, bond splitting, and concrete crushing (see Fig.15). Notably, the applied force increased with the steel fiber content. Therefore, increasing the steel fiber content can prevent crack opening at the center section of the sleepers.

3.4 Pull-out test of the fasting system

A pull-out test was performed on the screws of the fasting system, which were cast into the concrete sleepers based on EN 13481-2 [27]. Vossloh W21 rail fastening was tested in this study (see Fig.16). The test setup is shown in Fig.17. According to EN 13481 [27], the test load should be applied vertically on the screw at a rate of 50 ± 10 kN/min. Hence, a load of 60 kN was applied in this study, which was maintained for 3 min and then removed without shock. The load was applied repeatedly until the screw was pulled out from the sleeper, and the displacement was measured at different loads. This test was performed on two specimens for each group of UHPFRC sleepers.

Tab.5 summarizes the test results for the failure loads. Fig.18 shows the relationship between the vertical load and the displacement of the screw. The pull-out of the screws and the final state are shown in Fig.19. When the pull-out load reached 60 kN and was maintained for 3 min, no cracks were observed in any of the sleepers. All sleepers in this test satisfied the requirements of European standards. Although the S0 group completely satisfied the requirements of European standards [27], the specimens in the group indicated an abrupt decrease in load owing to concrete crushing at the top of the sleepers, as shown in Fig.19(a). By contrast, all other groups with fibers did not exhibit concrete failure; however, they exhibited failure caused by the remaining plastic dowels, as shown in Fig.19(b). This indicates that increasing the fiber amount increases the pull-out resistance of the fasting system.

4 Parametric study for ultrahigh-performance fiber-reinforced concrete sleeper

To assess the effects of the properties of the UHPFRC mix on the static tests and pull-out of the screw, a parametric study was performed for a concrete sleeper. The main parameters were the concrete volume of the sleeper and the amount of steel fiber. A strain compatibility analysis was performed to calculate the nominal moment at the critical sections presented in the previous subsection, and the results are summarized in Tab.6. Similar to cases involving the typical concrete, the compressive stress was assumed to be confined within a rectangular stress block. The neutral axis depth is denoted as c, and the distance between the end edge of the compression side and the top of the steel fiber tensile area is denoted as e. The value of c was varied until the compressive and tensile forces were balanced. Subsequently, e can be estimated using Eq. (5), which is based on ACI 544.4R-88 [35].

(3)F_{\mathrm{t}}=0.00772\ell /d_{\mathrm{f}}{v}_{\mathrm{f}}{F}_{\mathrm{b}\mathrm{e}},

(4){å}_{\mathrm{s}}=F_{\mathrm{t}}/E_{\mathrm{s}},

(5)e=\left({\epsilon }_{\mathrm{s}}+0.003\right)c/0.003.

where F_{\mathrm{t}} is tensile stress in fibrous concrete, \ell is fiber length, d_{\mathrm{f}} is fiber diameter, v_{\mathrm{f}} is volume friction of steel fibers, F_{\mathrm{b}\mathrm{e}} is bond efficiency of the fiber which varies from 1.0 to 1.2, {å}_{\mathrm{s}} is a tensile strain in steel fiber, E_{\mathrm{s}} is the modulus of elasticity of steel fiber.

Fig.20 shows the dimension of the cross sections at the rail seat and the center section. Stress distribution assumptions at the rail seat section and the center section are shown in Fig.21. The load increase factor \left(\text{ã}\right) and strength reduction factor (\varnothing) were used to evaluate the sleeper sections (\text{ã}=1.85;\varnothing =0.85) [24]. An analysis of the sleeper sections under the external forces exerting on the center and rail seat sections shows that sufficient safety factor values were attained while considering the frequent train loading and the possibility of impact loading.

The safety factor at the center and rail seat sections, as presented in Tab.7, can be calculated after calculating the nominal moment. The results show that increasing the friction volume of steel fiber increased the safety factor at the center and rail seat sections of the concrete sleepers. In addition, the results confirmed that the safety factor for all sleepers with UHPFRC is higher than that of conventional concrete sleepers.

The results of the experimental static test at the center and rail seat sections were compared with the structural design to evaluate the appropriateness of the design mixes. The failure load at the center and rail seat sections can be obtained using calculated values of the nominal moment. The M_{\mathrm{d}\mathrm{r}+} and M_{\mathrm{d}\mathrm{c}-} in Eqs. (1) and (2), respectively, were substituted with \varnothing M_{\mathrm{n}}. The failure loads were calculated based on the nominal moment. As shown in Tab.8 and Tab.9, the structural design results are similar to the experimental results. The results show that increasing the friction volume of the steel fiber increased the rate of both the experimental and reference failure loads, i.e., F_{\mathrm{c}\mathrm{B}}(\mathrm{E}\mathrm{x}\mathrm{p}.) and F_{\mathrm{c}\mathrm{B}}(\mathrm{R}\mathrm{e}\mathrm{f}.), respsectively [26], as shown in Tab.8 and Tab.9.

Based on the previous results and by confirming the consistency between the experimental and structural results, one can reduce the size of the sleeper’s concrete section to satisfy the train speed and axel load requirements. Reducing the volume of the concrete sleeper using UHPFRC mixes enhances the workability and incurs a lower cost for the materials used. Fig.22 shows the sections and percentage of volume reduction for each mixture based on the structural design. The calculated safety factors for each of these sections at the rail seat and center sections are shown in Tab.10.

The failure loads for determining the pull-out resistance of the fasting system were related to the concrete strength. Tab.11 presents the normalized average failure loads corresponding to the UHPFRC strength for each concrete mix (P_{\mathrm{c}\mathrm{r}}/f_{\mathrm{c}\mathrm{u}}). The result shows that increasing the volume fraction of steel fibers benefited the crack control of UHPFRC sleepers and improved the pull-out resistance.

5 Conclusions

A B70 monoblock sleeper was fabricated and developed for the first time in this study using UHPFRC. The effects of different amounts of steel fiber on the structural behavior of UHPFRC sleepers were investigated based on two static bending moments at the center and rail seat sections and the pull-out resistance of the screw on the fastening system based on European testing standards [26,27]. Four levels of steel fiber volume fractions were investigated, i.e., 0%, 0.5%, 1%, and 1.5%. The main conclusions and observations of this study are as follows.

1) The results of static bending tests on the center and rail seat sections of the UHPFRC sleepers showed that the sleepers exceeded the requirements of EN 13230-2 [26].

2) The static bending test at the rail seat sections showed that the average failure load for series S0, S0.5, S1, and S1.5 increased by 48%, 72%, 86%, and 98%, respectively, as compared with the reference value indicated in EN 13230-2 [26].

3) For the static bending test at the center sections, the average failure load for each series S0, S0.5, S1, and S1.5 increased by 42%, 58%, 63%, and 71%, respectively, as compared with the reference value indicated in EN 13230-2 [26].

4) Based on the results of the pull-out test, the incorporation of steel fibers in all the specimens prevented the initiation of cracks on the UHPFRC sleepers and increased the pull-out resistance via an increase in the friction volume of the steel fiber, thus satisfying the requirements of EN 13481-2 [27].

5) The safety factors obtained by implementing the UHPFRC mix in railway sleepers for series S0, S0.5, S1, and S1.5 were 2.45, 2.59, 2.73, and 2.86 for the rail seat sections and 1.67, 1.79, 1.92, and 2.04 for the center sections, respectively, which were reasonable values based on a comparison with the reference value indicated in EN 13230-2 [26].

6) The UHPFRC mix improved the structural performance and incurred a lower material cost by reducing the volume of the standard sleeper (B70) while satisfying the speed and axial load requirements, thus rendering it economical and suitable for the environment.

7) Sleepers fabricated using UHPFRC with a steel fiber volume fraction of 1% and a concrete volume less than 25% that of standard sleeper B70 can be used under the same loads and requirements, which contributes positively to the cost and the surrounding environment.

This investigation is expected to serve as a basis for the application of UHPFRC to concrete sleepers and to provide information regarding the extent to which its application reduces the volume of type B70 standard concrete sleepers. In the future, further investigations should be conducted to optimize the shape of sleepers using UHPFRC via dynamic and fatigue tests while satisfying the requirements of EN 13230-2 [26] to fully exploit the advantages of UHPFRC and to conduct field applications that reflect the dynamic responses afforded.

References

[1] Tolou Kian A R, Sadeghi J, Zakeri J A. Influences of railway ballast sand contamination on loading pattern of pre-stressed concrete sleeper. Construction & Building Materials, 2020, 233: 117324

[2] Berbey Alvarez A, Guevara-Cedevo J. Railway Engineering. Oxford: Oxford University Press, 2021, 371–395

[3] Sadeghi J, Barati P. Comparisons of the mechanical properties of timber, steel and concrete sleepers. Structure and Infrastructure Engineering, 2010, 8: 1151–1159

[4] Bae Y, Pyo S. Effect of steel fiber content on structural and electrical properties of ultra high performance concrete (UHPC) sleepers. Engineering Structures, 2020, 222: 111131

[5] Remennikov A M, Murray M H, Kaewunruen S. Conversion of AS1085. 14 for prestressed concrete sleepers to limit states design format. In: AusRAIL PLUS 2007 Conference & Exhibition. Sydney: Australian Railway Association (ARA), 2007, 1–18

[6] LutchR H. Capacity optimization of a prestressed concrete railroad tie. Thesis for the Master’s Degree. Houghton, MI: Michigan Technological University, 2009

[7] Sadeghi J M, Babaee A. Structural optimization of B70 railway prestressed concrete sleepers. Iranian Journal of Science & Technology, Transaction B, Engineering, 2006, 30(B4): 461–473

[8] Kaewunruen S, Remennikov A M. Progressive failure of prestressed concrete sleepers under multiple high-intensity impact loads. Engineering Structures, 2009, 31(10): 2460–2473

[9] Kaewunruen S, Remennikov A M. Dynamic crack propagations in prestressed concrete sleepers in railway track systems subjected to severe impact loads. Journal of Structural Engineering, 2010, 136(6): 749–754

[10] Vairagade V S, Kene K S, Patil T R. Comparative study of steel fiber reinforced over control concrete. International Journal of Scientific and Research Publications, 2012, 2(5): 1–3

[11] Yazıcı Ş, Inan G, Tabak V. Effect of aspect ratio and volume fraction of steel fiber on the mechanical properties of SFRC. Construction & Building Materials, 2007, 21(6): 1250–1253

[12] Wu Z, Shi C, He W, Wu L. Effects of steel fiber content and shape on mechanical properties of ultra high performance concrete. Construction & Building Materials, 2016, 103: 8–14

[13] Beglarigale A, Yazici H. Pull-out behavior of steel fiber embedded in flowable RPC and ordinary mortar. Construction & Building Materials, 2015, 75: 255–265

[14] Mahakavi P, Chithra R. Impact resistance, microstructures and digital image processing on self-compacting concrete with hooked end and crimped steel fiber. Construction & Building Materials, 2019, 220: 651–666

[15] Abousnina R, Premasiri S, Anise V, Lokuge W, Vimonsatit V, Ferdous W, Alajarmeh O. Mechanical properties of macro polypropylene fibre-reinforced concrete. Polymers, 2021, 13(23): 1–25

[16] Yu P, Manalo A, Ferdous W, Abousnina R, Salih C, Heyer T, Schubel P. Investigation on the physical, mechanical and microstructural properties of epoxy polymer matrix with crumb rubber and short fibres for composite railway sleepers. Construction & Building Materials, 2021, 295: 123700

[17] Li L, Wang B, Hubler M H. Carbon nanofibers (CNFs) dispersed in ultra-high performance concrete (UHPC): Mechanical property, workability and permeability investigation. Cement and Concrete Composites, 2022, 131: 104592

[18] Pyo S, Kim H K, Lee B Y. Effects of coarser fine aggregate on tensile properties of ultra high performance concrete. Cement and Concrete Composites, 2017, 84: 28–35

[19] SchmidtMFehlingEGeisenhanslükeC. Ultra High Performance Concrete (UHPC): Proceedings of the International Symposium on Ultra High Performance Concrete. Kassel: Kassel University Press, 2004

[20] ACI239R-18. Ultra-High-Performance Concrete: An Emerging Technology Report. Farmington Hills, MI: American Concrete Institute, 2018

[21] Safdar M, Matsumoto T, Kakuma K. Flexural behavior of reinforced concrete beams repaired with ultra-high performance fiber reinforced concrete (UHPFRC). Composite Structures, 2016, 157: 448–460

[22] Yoo D Y, Jang Y S, Chun B, Kim S. Chelate effect on fiber surface morphology and its benefits on pullout and tensile behaviors of ultra-high-performance concrete. Cement and Concrete Composites, 2021, 115: 103864

[23] Lam N N, Van Hung L. Mechanical and shrinkage behavior of basalt fiber reinforced ultra-high-performance concrete. GEOMATE Journal, 2021, 20(78): 28–35

[24] Bae Y, Pyo S. Ultra high performance concrete (UHPC) sleeper: Structural design and performance. Engineering Structures, 2020, 210: 110374

[25] EN13230-6. Railway Applications-Track-Concrete Sleepers and Bearers—Part 6: Design. Brussles: European Committee for Standardization (CEN), 2014

[26] EN13230-2. Railway Applications-Track-Concrete Sleepers and Bearers—Part 2: Prestressed Monoblock Sleepers. Brussles: European Committee for Standardization (CEN), 2009

[27] EN13481-2. Railway applications-Track-Performance Requirements for Fastening Systems—Part 2: Fastening Systems for Concrete Sleepers. Brussles: European Committee for Standardization (CEN), 2014

[28] EN10138-2. Prestressing Steels—Part 2: Wire Armatures. Brussles: European Committee for Standardization (CEN), 2000

[29] Yu R, Spiesz P, Brouwers H J H. Mix design and properties assessment of ultra-high performance fibre reinforced concrete (UHPFRC). Cement and Concrete Research, 2014, 56: 29–39

[30] Pyo S, Wille K, El-Tawil S, Naaman A E. Strain rate dependent properties of ultra high performance fiber reinforced concrete (UHP-FRC) under tension. Cement and Concrete Composites, 2015, 56: 15–24

[31] Mo Z, Gao X, Su A. Mechanical performances and microstructures of metakaolin contained UHPC matrix under steam curing conditions. Construction & Building Materials, 2021, 268: 121112

[32] UIC 713R. Design of Monoblock Concrete Sleepers. Paris: International Union of Railways, 2004, 30

[33] Yang J M, Shin H O, Yoon Y S, Mitchell D. Benefits of blast furnace slag and steel fibers on the static and fatigue performance of prestressed concrete sleepers. Engineering Structures, 2017, 134: 317–333

[34] Yoo D Y, Lee J Y, Shin H O, Yang J M, Yoon Y S. Effects of blast furnace slag and steel fiber on the impact resistance of railway prestressed concrete sleepers. Cement and Concrete Composites, 2019, 99: 151–164

[35] ACI544.4R-88. Design Considerations for Steel Fiber Reinforced Concrete. Farmington Hills, MI: American Concrete Institute, 1999

RIGHTS & PERMISSIONS

Higher Education Press