**1. Introduction**

Concrete is widely used in architectural structures and social infrastructure facilities because it is economical and has high compressive strength and durability. However, concrete is characterized by brittle fracture due to low bending and tensile strengths and weak crack resistance compared to the high compressive strength [1–6]. Recently, studies have been conducted to develop high-performance construction materials with excellent performance by improving the disadvantages or maximizing the advantages of concrete. Attempts are being made to mechanically increase strength or improve ductility, because strength and ductility have opposite properties [7].

With the development of construction technology, the construction of high-rise of buildings and long structures, the use of 100 MPa or higher ultra-high-strength concrete is increasing [8]. However, although ultra-high-strength concrete has high compressive strength, it has low tensile strength, bending strength, and toughness, and weak resistance to cracks [8]. Above all, ultra-high-strength concrete has the intrinsic problem of brittle fracture to the peak stress. To overcome this problem, active studies are being conducted on high-performance fiber-reinforced cementitious composites (HPFRCCs), in which brittle fracture to ductile fracture are induced [8].

HPFRCCs are characterized by improved tensile force, strain capacity, and energy absorption capacity, which are the weaknesses of conventional concrete [9,10]. However, the maximum fiber volume fraction of conventional HPFRCCs and fiber-reinforced concrete is limited to 2.0% due to fiber balling [9,10]. A majority of fiber balling occurs during the fiber addition process due to excessive fibers during the mixing of HPFRCCCs and fiber-reinforced concrete. Thus, the limited fiber volume fraction has been one of the biggest disadvantages. Studies are being actively conducted to understand the direct tensile behavior characteristics of HPFRCCs [9,10]. To overcome these disadvantages, an increase in the steel fiber volume fraction may improve tensile strength and energy absorption capacity. Therefore, HPFRCCs reported thus far have exhibited deflection hardening behavior under flexural tensile load, rather than direct tensile behavior and strain hardening behavior with multiple micro-cracks [9,10]. Therefore, it is very difficult to obtain a tensile stress-strain curve under direct tensile load to acquire information on multiple cracks [9,10].

To maximize the mechanical properties of HPFRCCs and overcome the limitation of fiber volume fraction, this study developed slurry-infiltrated fiber-reinforced cementitious composites (SIFRCCs), which can incorporate a high volume of steel fibers. The SIFRCCs can incorporate up to 8% fiber volume fraction, thus maximizing tensile strength, energy absorption capacity, and strain capacity, which are shortcomings of the conventional concrete and fiber-reinforced concrete. An experimental research on the tensile behavior characteristics was conducted with respect to the fiber volume fraction of high-performance SIFRCCs through a direct tensile test.

#### **2. Existing Works Related to Direct Tensile Test**

In a study on the flexural tensile strength of fiber-reinforced concrete members, the ductile behavior improved after cracking at 80 MPa or lower compressive strength. However, studies on structural behavior analysis of members of 150 MPa or higher compressive strength are relatively insufficient and predictions of the bending strength are limited [8,11]. Flexural tensile tests of fiber-reinforced concrete involve many difficulties, but flexural tensile test are mainly conducted for direct tensile tests because of the problems of slip phenomenon of specimens in the drawing process [8]. However, attempts at direct tensile tests are being made continuously because reliability can be compromised due to many assumption conditions in the process of estimating tensile strength through flexural tensile tests. To address this problem, French regulations have presented a method of performing tests by directly installing notches in the specimens [12].

According to a research report on ultra high performance concrete (UHPC), crack review appears to be unnecessary for UHPC considering its higher tensile strength than conventional concrete due to the action of steel fibers, and the characteristic of low crack width relative to the tensile load also appears to be unnecessary [8,13,14]. The report mentioned that even though the crack width is small, it is necessary to examine crack behavior under various load conditions [8,15]. It also stated that because the crack behavior can vary under the working load and extreme load due to the ductile tensile behavior, it is necessary to validate the crack examination of UHPC with the existing design standards through direct tensile test [8,15]. Furthermore, unlike the direct tensile test, indirect tensile tests, such as tensile test through bending, are burdensome because they require inverse analysis using numerical analysis [8,15]. Although direct tensile tests are not widely used in the measurement of the tensile behavior of concrete, they have been considered to be the most direct and proper method owing to the characteristics of UHPC that exhibits ductile behavior after cracking [8,15].

#### **3. Experiment Overview**

#### *3.1. SIFRCCs*

The SIFRCCs developed in this study can maximize the steel fiber volume fraction of the existing fiber-reinforced concrete and HPFRCCs. With its high fiber volume ratio, high tensile strength, energy absorption capacity, and strain capacity can be expected. Unlike fiber-reinforced concrete and HPFRCCs, SIFRCCs can be produced by the following steps. First, the mold is filled with steel fibers in advance. Second, high-performance slurry is prepared after mixing. This slurry should be poured to avoid mixing the concrete matrix and steel fibers. The high-performance slurry should be pour from one end to the other within the cluster of steel fibers such that there are no bubbles. This is done to avoid voids. SIFRCCs are characterized by the omission of coarse aggregates for the high-performance slurry to achieve sufficient filling performance between the steel fibers [9,16].

#### *3.2. Experiment Method*

To analyze the compressive strength of SIFRCCs, fiber volume fractions of 4%, 5%, and 6% were considered. The compressive strength test was performed in accordance with Korea Standards (KS F 2405) [17]. The cylinder specimens used had a diameter of 100 mm and a height of 200 mm.

To analyze the tensile behavior characteristics with respect to the fiber volume fraction of SIFRCCs, the characteristics of direct tensile behavior were experimented. In the case of direct tensile test, the test method is not clearly defined. Therefore, we conducted an experimental study on the tensile behavior characteristics such as energy absorption capacity through the direct tensile strength, which indicates the maximum tensile stress, strain capacity at the direct tensile strength, and stress-strain curve based on literature review.

To analyze the tensile behavior characteristics of SIFRCCs, fiber volume fractions of 4%, 5%, and 6% were considered. The direct tensile test was performed by displacement control method at the rate of 1 mm/min. A direct tensile test specimen appropriate for the dedicated tensile jig was fabricated as shown in Figure 1. The tensile behavior characteristics were analyzed using the 300-ton class universal testing machine shown in Figure 2. The cross-section of the direct tensile test specimen has a width of 50 mm and a height of 25 mm. The scope of gauge length for tensile performance measurement was set to 50 mm. To induce multiple cracks of SIFRCCs within the range of gauge length, a wire mesh was installed outside the gauge length and used facilitate crack inducement. Furthermore, to derive the tensile stress-strain curve of the SIFRCCs, a linear variable differential transformer (LVDT) (Tokyo Sokki, Tokyo, Japan) that can take measurements up to 25 mm was installed on either side of the tensile jig. The direct tensile strength was calculated using the following Equation (1) [7]:

$$f = \frac{P\_{\text{max}}}{bh} \tag{1}$$

where *Pmax* is the maximum load (N), *b* is the width (mm) at the gauge length, and *h*. is the height at the gauge length (mm).

**Figure 1.** Schematic of the specimen.

**Figure 2.** Experimental setup.

#### *3.3. Materials*

This study used type 1 ordinary Portland cement. Table 1 lists the physical and chemical properties of the used cement. In this study, silica fume was used to achieve high-performance and high strength of the slurry. Table 2 lists the physical and chemical properties of the used silica fume.


**Table 1.** Physical and chemical properties of the used cement.


**Table 2.** Physical and chemical properties of silica fume.

Furthermore, fine aggregates with a diameter of 0.5 mm or less were used to improve the filling performance of the high-performance slurry and to reduce material separation. Coarse aggregates were not used to secure filling performance. To improve the filling performance of the slurry, a high-performance polycarboxylic acid water-reducing agent was used. The admixture used in this experiment has high strength and high flow characteristics and has excellent unit water quantity reduction property and material separation resistance.

For steel fibers, double hook steel fibers for conventional concrete with a diameter of 0.75 mm, a length of 60 mm, and an aspect ratio of 80 were used. Regarding physical properties, the steel fibers have a density of 7.8 g/cm3 and a tensile strength of 1200 MPa. Figure 3 shows the shape of the used steel fibers.

**Figure 3.** Shape of the used steel fibers.
