EXPERIMENTAL STUDY ON STEEL AND POLYPROPYLENE FIBERS MORTAR BOND STRENGTH UNDER UNIAXIAL TENSION

The present study aims to investigate the bond behavior between fibers and mortar matrix in fiber-reinforced concrete, which was explored through a uniaxial tension test. The bond behavior was studied for different types of embedded fibers: steel fibers, polypropylene fibers, and hybridizations of steel and polypropylene fibers. The mixture proportion was determined based on the absolute volume method with a water-to-binder ratio (w/b) of 0.29 to achieve high-strength mortar. The high-strength mortar was cast by utilizing an industrial waste by-product of both silica fume and fly ash as a partial cement replacement. The compressive strength and uniaxial tension tests on high-strength mortar were conducted at the ages of 28 and 140 days. The long-term bond-strength behavior was investigated at the age of 140 days. The results showed that the average compressive strengths of hardened mortar on 28 and 140 days were 54.82 MPa and 69.37 MPa, respectively. Whereas, the average fiber-mortar bond strengths with steel, polypropylene, and hybridized fibers were 8.62 MPa, 8.37 MPa, and 7.30 MPa, respectively, at 28 days and 11.22 MPa, 10.21 MPa, and 11.82 MPa, respectively, at 140 days. Compared to the equivalent bond strength of the steel fiber, the polypropylene fiber had an equivalent bond strength of 2.90% and 9.00% lower at the ages of 28 and 140 days. Meanwhile, the equivalent bond strength of hybridized fiber was 15.31% lower than that of steel fiber; however, the long-term behavior of the hybridized fiber showed its performance was about 5,35% higher than that of steel fiber at 140 days.


INTRODUCTION
The use of fiber in concrete has increased over time (Ou et al., 2012).The fiber mostly used is steel fiber (Ou et al., 2012;Bhargava et al., 2006;Chao et al., 2009;Liao et al., 2015Liao et al., , 2017;;Perceka et al., 2016).With the use of fiber, concrete can reduce its brittleness behavior, especially when concrete experiences tension due to either flexural or direct tension (Liao et al., 2017;Liu et al., 2019).In addition, the presence of steel fibers reduces the post-peak behavior of concrete material, particularly high-strength concrete, under compression since it is well-known that concrete turns more brittle as its compressive strength increases (Bhargava et al., 2006;Liao et al., 2015;Perceka et al., 2016).The reduction of brittleness behavior can occur because the steel fibers can provide bridging action across the microcracks inside the mortar and increase crack opening resistance (Liao et al., 2015;Naaman & Najm, 1991;Perceka et al., 2016;Thomas & Ramaswamy, 2007).
This mechanism can occur due to the presence of the bond strength between steel fibers and the surrounding matrix (Liao et al., 2015;Perceka et al., 2016).
In terms of applying steel fiber to RC structural members, as results showed by Perceka et al. (Perceka et al., 2016) conducted a test on high-strength steel fiber reinforced concrete (SFRC) columns, proving that the early spalling of concrete columns and the early buckling failure of columns with wider transverse reinforcement spacing did not exist.Furthermore, according to the shear tests on SFRC beams reported Perceka et al. (2019), the SFRC beam specimens exhibited non-brittle shear failure as ordinary reinforced concrete (RC) beams showed, i.e., the strength shown by the load-mid.The displacement relationship did not decrease suddenly once the peak strength was reached.In terms of bond strength between steel fibers and the surrounding matrix, it has been reported that the bond strength was affected by fiber geometry, concrete or mortar strength, and fiber material (Liao et al., 2015;Xu et al., 2011).
To the authors' knowledge, previous researchers only performed bond-strength tests by using hooked-end steel fiber.No information about the test on bond strength between fiber and surrounding mortar by using fiber made of materials other than steel, such as polypropylene, has been available.Since polypropylene fibers are commercially available, the potential of the use of polypropylene fibers shall be investigated.Another general consideration when using polypropylene fiber is the corrosion potentially occurring on steel fibers.It is well known that corrosion is a common obstacle when using steel material, although no information has yet been available on the corrosion potential of steel fiber in SFRC material.Regardless of the detailed information on the durability of SFRC material with steel fiber, particularly the effect of corrosion of steel fibers, the potential of the use of polypropylene fiber shall be investigated.The basic information that shall be first investigated is the bond strength between polypropylene fiber and the surrounding mortar since it is well-known that the bond strength has become the main role in improving the behavior of RC members with fibers inside.Therefore, in this paper, the bond strength between polypropylene fiber and the surrounding mortar was investigated through pullout tests.The specimens used were dog-bone specimens that contained polypropylene fibers.In addition to those specimens, the dog-bone specimens with steel fiber and a combination of steel and polypropylene (hybridized) fibers were prepared as well.By knowing the pullout-slip behavior and bond strength between polypropylene or a combination of steel fiber and polypropylene fibers, more future research regarding the behavior of RC members with fiber other than steel material can be investigated as well.

Materials
In this experimental study, the ternary-blended cementitious materials were composed of Type I ordinary Portland cement (OPC), class F fly ash (FA), and silica fume, which were utilized as binders.Local OPC and FA from Indonesia with specific gravities of 3.03 and 2.13 are being tested (ASTM C188-16, 2016).The commercial densified silica fume had a specific gravity of 2.12 and was tested (ASTM C1240-20, 2020).The fine aggregate had a specific gravity and absorption of 2.49 and 3.09%, which was tested (ASTM C128-15, 2015).The fine aggregate, which conformed to the acceptance range (ASTM C33/C33M-18, 2018), had a fineness modulus (FM) equal to 3.08.The commercial hooked-ended steel and wavy polypropylene fibers were utilized in this study, as shown in Figure 1.The specifications of those fibers are summarized in Table 1.In addition, an acrylic-based superplasticizer (SP) conforming to the requirement (ASTM C494-17, 2017) Type G was used for controlling mortar workability.

Mix Proportions
The mixture proportion mortar was calculated based on the absolute volume method with a water-to-binder ratio (w/b) of 0.29.This number refers to an experimental study conducted by Liao et al. (2015).In addition, Liao et al. (2017) suggested that the total aggregate volume is 0.62 m 3 , and the FA to fine aggregate ratio is 15%.Moreover, the final amount of cement was replaced with silica fume content by mass, in the amount of 10% of the binder mass.In addition, SP based on a modified acrylic polymer was used at a fixed amount of 1.5% of the total powder mass to establish the expected fresh mortar properties.The final mix proportion of the mortar after performing the necessary trial-and-error procedure and modification is listed in Table 2.

Specimen Preparation and Testing Methods
The dog-bone specimens with detailed geometry and dimensions, as shown in Figure 2(a), were used to determine the bond strength between the fibers and mortar matrix.A single dog-bone specimen with ten embedded fibers in the middle section was separated into two sides.There were three configurations of fiber, as demonstrated in Figure 2(b); they were full steel fibers (SF), full polypropylene fibers (PF), and hybridizations of steel and polypropylene fibers (HF).Three dog-bone specimens were prepared for each configuration.An acrylic mold, as shown in Figure 3(a), was used for casting a dog bone specimen.As shown in Figure 3(b), a foam block with clinging film was equipped to mount the fibers in the middle part of the mold.The casting process of the dog-bone specimen was completed within two days.On the first day, the fresh mortar was poured into half of the dog bone specimen until the mounted fibers were completely cast, as demonstrated in Figure 3(c).On the following day, the foam block was carefully removed from the middle part, and the clinging film still stuck in the middle part, as shown in Figure 3(d), then filled the mold with fresh mortar until the other part was completely cast, as demonstrated in Figure 3(e).The bond strength tests were conducted in an electro-hydraulic servo universal testing machine (UTM) with a capacity of 500 kN (brand Hung-Ta model HT9501, serial number 1162) at a displacement rate of 0.6 mm/min.The UTM was equipped with a pair of steel testing grips, as demonstrated in Figure 4, for mounting the dog bone specimen.In addition, a digital single-lens reflex camera was used to digitally capture the pullout slip of the dog-bone specimen.In this study, the mortar-fiber bond strength tests were investigated at 28 and 140 days.At least duplicate 50-mm cubic specimens were made from each side of the casting to determine the compressive strength of mortar specimens in accordance with ASTM C109/C109M (ASTM C109/C109M, 2021).The tests were performed at 28 and 140 days by utilizing a semi-automatic compression machine with a capacity of 2000 kN (ELE model ADR2000) at a constant rate of 0.9 kN/s.According to ASTM C109/C109M (ASTM C109/C109M, 2021), the compressive strength of mortar () is calculated by dividing the load at failure (P) with the cross-sectional area of the mortar cube (A) as described in Equation ( 1).
According to Liao et al. (Liao et al., 2015), the equivalent bond strength ( eq ) of each fiber configuration can be calculated by using Equation (2).It is shown that  eq is the function of fiber pullout energy ( po ) that is expressed by the area under the pullout load-slip curve, fiber diameter ( f ), and the embedded length of the fiber ( emb 2 ).

Compressive Strength
As reported by Liao et al. (2017), the equivalent bond strength of fiber increased as the compressive strength of the mortar matrix increased.Therefore, the bond strength of fibers in the mortar is affected by the matrix compressive strength.In this study, the compressive strength tests for each variation were tested at the age of 28 and 140 days.As summarized in Table 3, the average 28-day compressive strength of the mortar matrix () was 55.08 MPa with a standard deviation () and coefficient of variation (CV) of 2.78 MPa and 5.05%, respectively.Furthermore, it was observed that the 140-day compressive strength was 27.60% higher than the 28-day compressive strength.The incorporation of the silica fume into the mortar matrix induced higher longterm compressive strength due to the pozzolanic reaction that consumes a huge amount of portlandite and forms more C-S-H gel (Tavares et al., 2020;Fu et al., 2021).

Equivalent Bond Strength
The equivalent bond strength average (̅ eq ) was calculated by averaging the results obtained from three tests, as summarized in Table 4.The symbols SF, PH, and HF, respectively, stood for steel fibers, polypropylene fibers, and hybrid fibers.The equivalent bond strengths of those three different configurations were investigated at days 28 and 140. Figure 5 demonstrates the pullout load-slip curves of all specimens.In each test, the  po was determined by calculating the area under the pullout load-slip curve, and the  eq was determined by using Equation ( 2).As presented in Table 4, the SF exhibited the highest bond strength, with an average value of 7.30 MPa at 28 days and 9.70 MPa at 140 days.In the meantime, the average bond strengths of PF specimens at 28 days and 140 days, respectively, were 6.74 MPa and 8.57 MPa.In addition, the 28-day and 140-day bond strengths of HF specimens were 7.13 MPa and 9.15 MPa, respectively.
Compared to the average bond strength of the SF specimen, it can be seen that the 28-day average bond strength of the PF and that of the HF specimens, respectively, were 7.67% and 2.33% lower.In addition to the comparison for days 28, the average bond strength of PF and that of HF specimens at 140 days decreased by 11.65% and 5.67 %, respectively, when compared to the average bond strength of SF at days 140.It can be noted that the SF specimen exhibited the highest bond strength due to the mechanical anchorage of the hooked end at each edge of the fiber (Abdallah et al., 2018).Unlike the hooked-end steel fiber, polypropylene, in general, relies only on the friction force between the fiber and the surrounding mortar.On the other hand, the HF specimen demonstrated a higher bond strength compared to the PF specimen; however, the bond strength of the HF specimen was lower than that of the SF specimen.In the HF specimen, the friction steel fibersurrounding mortar and the mechanical anchorage action due to hooked-end steel fiber still exist.Therefore, the HF specimen could exhibit higher strength compared to the PF specimen.In the meantime, the reduction in the number of steel fibers in the HF specimen led to a reduction in bond strength when compared to the SF specimen.Moreover, at the age of day 140, the average bond strengths of SF, PF, and HF increased by 32.88%, 27.15%, and 28.33%, respectively, compared to the strengths at day 28.Therefore, the improvement of mortar strength led to the improvement of bond strength generally (Ding et al., 2021).In addition, the PF specimens tested on day 28 failed due to the slip of fiber in the mortar.In contrast, after 140 days, it was found that two polypropylene fibers of the PF specimen got broken, as shown in Figure 6(d).This phenomenon indicated the bond strength might be higher than the tensile strength of polypropylene fiber in the long-term period since the mortar strength could still develop after days 28 (Ding et al., 2021).A broken fiber was only found in the PF specimen, while no broken fiber occurred in the SF and HF specimens, as seen in Figures 6(e) and (f).It must be noticed that the tensile strength of steel fiber is much greater than that of polypropylene fiber.Therefore, by keeping steel fiber inside the HF specimen, the bond strength between the single fiber and surrounding mortar could be lower and could be broken before the broken fiber occurred.The final appearance of the steel fiber in this study also confirmed the mechanism of the steel fiber pullout test reported by Liao et al. (2015).

CONCLUSIONS
The bond strengths of steel fibers, polypropylene fibers, and hybridizations of steel and polypropylene fibers on cement-based high-strength mortar matrix incorporated with silica fume and fly ash were successfully investigated by involving different fiber configurations in the dog-bone specimens and performing tests at different times in order to capture the short-and long-term effects.The SF specimens exhibited the highest equivalent bond strength among the three type configurations since the equivalent bond strength of the SF specimen is affected by both the steel fiber surface-surrounding mortar friction force and the mechanical anchorage action due to hooked-end steel fiber.In contrast, the equivalent bond strength of the fiber and surrounding mortar of the PF specimen only relied on the fiber surface-surrounding mortar friction force.Furthermore, in the HF specimen, due to the presence of steel and polypropylene fibers, the equivalent bond strength was affected by the fiber surface-surrounding mortar friction force that appeared from steel and polypropylene fibers and the mechanical anchorage action due to hooked-end steel fiber.In terms of failure Broken-fiber mode, the failure mode of SF and HF specimens was a bond-slip failure, while the failure mode of PF specimens was the rupture of fibers.In addition, according to the failure modes shown by SF, PF, and HF specimens, it can be noticed that the presence of steel fibers affected the failure mode and tended to turn the rupture of fiber into bond-slip failure.This study presents the long-term effects as well.It is known that the equivalent bond strength at day 140 was greater than that at day 28, regardless of the fiber configuration.Since polypropylene fiber is commercially available, further study in terms of the use of polypropylene fiber in FRC structural members may be necessary.However, based on the fiber pullout behavior, the value of equivalent bond strength, and referring to previous studies of FRC structural members, the steel fibers could exhibit better performance in FRC structural members compared to the polypropylene fibers.

Figure 1 .
Figure 1.Physical Appearance of Steel and Polypropylene

Figure 2 .
Figure 2. (a) The Geometry and Dimensions of Specimen; (b) Fiber Configurations

Figure 6
Figure 5. Equivalent bond strength of SF, PF, and HF

Figure 6 .
Figure 6.Typical failure mode of dog-bone specimens after pullout test