Local failure resistance of polypropylene fiber reinforced concrete plates subjected to projectile impact

Abstract

This study investigated local failure characteristics of polypropylene fiber reinforced concrete (PPFRC) plates subjected to projectile impact. Flexural and compression tests for two types of PPFRC specimens (PPFRC1 and PPFRC2) were conducted to examine mechanical properties of the PPFRC. The average flexural strength of PPFRC1 and PPFRC2 at a strain rate of 10⁻¹/s were 12.2 N/mm² and 10.4 N/mm², respectively. The average compressive strength of PPFRC1 and PPFRC2 at a strain rate of 10⁰/s were 58 N/mm² and 74.9 N/mm², respectively. Projectile impact tests for 60 mm- and 80 mm-thick PPFRC plates were conducted by using a 50 g-mass projectile collided at velocities corresponding to 193–423 m/s. Experimental results exhibited that the PPFRC plate had a higher effect of suppressing local failure than a plain concrete plate. Comparison of the tests results with the modified NDRC formula revealed that the limit scabbing thickness was 15–20% smaller than that of a plain concrete plate.

Keywords

projectile impact polypropylene fiber reinforced concrete local failure fiber reinforced concrete scabbing

Introduction

In recent years, terrorist bombing attack and explosive accidents have occurred frequently in various countries. When a blast terrorism or an explosive accident occurs, the blast pressure damages structures and scattered debris such as concrete pieces generated by the explosive pressure would secondarily collide with surrounding structures. Hence, a design method for global and local failures of structures against the projectile needs to be established specially for socially vital structures such as nuclear power plants, chemical plants, and explosive plants.

Many studies have investigated the failure behavior of concrete structures at velocities corresponding to less than 10 m/s for impacts of a falling rock, debris flow, and manmade disasters such as vehicle and vessel impacts to structures (Kishi et al., 2002; Othman and Marzouk, 2016; Zineddin and Krauthammer, 2007; Miyamoto et al., 1991; Kennedy, 1976; Li et al., 2005). Additionally, the failure behavior of RC slabs under impact load at velocities corresponding to 10–1000 m/s (low-to high-velocity) were also conducted for airplane impact and flying debris impact by explosive accidents (Hughes, 1984; Chang, 1981; Dancygier and Yankelevsky, 1999; Li and Chen, 2003; Li et al., 2005; Chen et al., 2008; Beppu et al., 2008; Werner et al., 2013; Martin, 1994; Shasha et al., 2017). As shown in Figure 1., a concrete plate subjected to low-to high-velocity impact shows local failure. Local failure can be divided into three major failure modes of spalling, scabbing, and perforation. In designing protective structures against low-to high-velocity impact of the projectile, scabbing, and perforation are unacceptable failure modes from the viewpoint of protecting human life inside the structure. Local failure occurs, when tensile stress wave, which derives from reflecting of compressive stress wave by the impact load at a free surface, reaches the strength of tensile strength. In addition, local failure might occur due to local deformation of structural members.

Figure 1.

Classification of local failure of concrete slabs.

Various studies have focused on suppressing the local failure of concrete plates. Dancygier et al. (Dancygier and Yankelevsky, 1996) and Tai (Tai, 2009) conducted high-velocity impact tests on high-strength concrete slabs, and they have investigated the effect of increasing the strength of concrete on the impact resistance. Mohamed (Mohamed and Ahmed, 2014) et al. conducted high-velocity impact tests on RC slabs whose front or back surface was reinforced with a steel plate. The results exhibited that local failure can be suppressed to some extent by reinforcing the back surface with a steel plate, and the suppressing effect of local failure by the back surface reinforcement was greater than that by the front surface reinforcement.

Restraining effect by Fiber Reinforced Cementitious Composite (FRCC) on the local failure has been studied as one of promising methods for improving static or dynamic resistance of structures. FRCC is a cementitious composite made of short fibers such as steel fibers, polypropylene (PP) fibers, polyvinyl alcohol (PVA) fibers, and glass fibers mixed with mortar or concrete to improve tensile strength and toughness by bridging cracks between matrices. Although the fiber reinforced concrete (FRC) with steel fibers mixed with concrete or mortar has been developed mainly to improve the local failure suppressing effect for cementitious plates, synthetic resin fibers such as PP fibers, PVA fibers, and aramid fibers have begun to be used as alternative materials to steel fibers. Synthetic resin has higher corrosion resistance and are lighter than steel fibers, so they have been used for lining to prevent concrete from peeling off in tunnels. Mechanical properties of the organic fibers and FRCs have been studied in the previous studies. Leung et al. (Leung and Li, 1991), Kanda et al. (Kanda and Li, 1999), and Redon et al. (Redon et al., 2001) proposed theoretical or empirical debonding models between a fiber and matrix. They classified the fiber adhesion mechanism into chemical and frictional adhesion, corresponding to the fact that the separation of fibers from matrix can be divided into two mechanisms: chemically peeling the adhesion and mechanically pulling out of the fiber from the matrix. In the literatures, dynamic strength characteristics of FRCC have been investigated. Wang et al. (Wang et al., 1996) conducted a dynamic flexural test on concrete specimens mixed with PP and steel fibers, and revealed that the specimen mixed with steel fibers showed higher energy absorption. Fenu et al. (Fenu et al., 2016) conducted a tension-type improved Hopkinson bar test on cement mortar specimens mixed with glass fibers and basalt fibers to investigate the dynamic mechanical properties. Xu et al. (Xu et al., 2021) conducted a uniaxial tensile test on basalt fiber reinforced engineered cementitious composite specimens, and they reported that multiple cracking behavior was significantly developed in the material.

Impact-resistant performance of FRC plates has also been investigated. Ong et al. (Ong et al., 1999) conducted low-velocity impact tests for concrete slabs mixed with polyolefin, polyvinyl alcohol, and steel fibers, and they concluded that the concrete slabs reinforced with steel fibers had higher impact performance. Ramakrishna et al. (Ramakrishna & Sundarajan, 2005) conducted impact tests for a cement mortar plate mixed with coconut, sisal, jute, and hibiscus fibers, and they revealed that the mortar plate mixed with coconut fibers had the highest impact performance. Wang et al. (Wang and Chouw, 2018) conducted impact tests for a coconut fiber reinforced concrete plate covered with a flax fiber reinforced polymer on the surface, and the relationship between the impact force and displacement was investigated. Ueno et al. (Ueno et al., 2017) conducted projectile impact tests for a FRC slab mixed with PP fibers by using 46 g mass projectile, and they clarified high impact resistance of the FRC slab. Beppu et al. (Beppu et al., 2020) conducted impact tests on ultra-high-strength fiber-reinforced concrete plates, and exhibited their high impact resistance. Almusallam et al. (Almusallam et al., 2013) conducted impact tests for RC slabs mixed with PP and steel fibers, exhibiting that their spalling depth was identical regardless of the type of fiber, even though damaged area was decreased. Wei et al. (Wei et al., 2021) conducted impact tests on FRC plates containing two types of steel fibers with different length, showing their higher impact performance. Liu et al. (Liu et al., 2021) conducted high-velocity impact tests on steel fiber-reinforced geopolymer plates, and they revealed that the penetration limit was improved by the fiber reinforcement. Kheyroddin et al. (Kheyroddin et al., 2021) conducted drop hammer tests on polypropylene fiber reinforced concrete plates covered with a glass fiber sheet, and they concluded that the polypropylene fiber reinforcement and glass fiber sheet reinforcements were effective in mitigating impact damage.

As described above, FRCC and FRC are qualitatively effective in suppressing local failure caused by a projectile impact. Almussalam et al. (Almusallam et al., 2013), Nam et al. (Nam et al., 2016) and Feng et al. (Feng et al., 2019) investigated the impact-resistant performance of PPFRC plates, whose fibers had a diameter of less than 0.05 mm, subjected to projectile impact. However, few studies focused on the impact resistance of PPFRC plates containing fibers with a diameter of greater than 0.5 mm. Hence, examining the relationship between local failure of PPFRC plates and the material properties is included in this study. In this study, impact tests were conducted for FRC plates with two types of polypropylene fibers (PPFRCs), and the suppressing effect of the local failure of the FRC plates was investigated. Flexural tests and uniaxial compression tests were initially conducted for two types of FRC specimens with different fiber dimension and mixing ratio, to examine the static and dynamic flexural or compressive strength characteristics of the PPFRC. Subsequently, projectile impact tests were conducted for three types of PPFRC plates using a high-pressure launching test machine, to examine scabbing resistance of the PPFRC plates. Furthermore, the scabbing limit thickness of the PPFRC plate was evaluated using the modified NDRC formula.

Experimental program

Polypropylene fibers

Table 1 lists the dimension and mechanical characteristics of PP fibers used in this study. Two types of PP fibers were used in fabricating FRC plates, as shown in Figure 2. PP1 has a fiber length of 30 mm and a diameter of 0.5 mm, and PP2 has a length of 12 mm and a diameter of 0.043 mm. PP is made of polyolefin resin with a low specific density, and has excellent fiber formability. PP is processed into an uneven surface to increase the mechanical bonding. PP1 has superior properties in both the tensile strength and the Young’s modulus: the tensile strength of PP1 is approximately 1.3 times that of PP2, and the Young’s modulus of PP1 is twice that of PP2. As the density of both fibers is 0.91 g/cm³, and the number of PP2 in the unit volume is approximately 338 times that of PP1. Assuming that each fiber is columnar, the specific surface area, which is defined as the surface area per unit volume, is 13.9 (mm²/mm³) for PP1 and 321.2 (mm²/mm³) for PP2. Tables 2 and 3 list the constituent of concrete and PPFRC used in this study. The maximum diameter of coarse aggregate was 20 mm. PP1 and PP2 were mixed with concrete 2.0% and 0.4% by volume, respectively. The specimens were prepared according to JCI-SF1(Japan Concrete Institute: JCI Standard Collection, 2004) and JCI-SF2(Japan Concrete Institute: JCI Standard Collection, 2004). The slump of concrete matrix was designed to be approximately 20–25 cm. The mixing ratio of polypropylene fibers was set as high as possible by conducting test kneading in advance. PPFRCs mixed with PP1 and PP2 are referred to as PPFRC1 and PPFRC2, respectively.

Table 1.

Dimension and mechanical characteristics of polypropylene fibers.

Type	Length (mm)	Diameter (mm)	Tensile strength (MPa)	Young’s modulus (MPa)	Density (g/cm³)
PP1	30	1.0	500	10000	0.91
PP2	12	0.043	390	5000	0.91

Figure 2.

Appearance of fibers.

Table 2.

Concrete constituents.

Symbol	Material	Specifications
C	Cement	Normal portland cement, density3.16(g/cm³)
W	Water	Tap water
S	Sand	Pit sand, density:2.62(g/cm³)
G	Coarse aggregate	Crushed limestone2005, density:2.70(g/cm³)
F	Fiber	PP fiber (see Table.1)
SP	Additive	air entraining and water reducing agent, SP8SV BSF Japan

Table 3.

Specified mix constituent.

Type	W/C (%)	s/a (%)	Unit quantity(kg/m³)				F(kg/m³)		SP	Air-volume (%)	Slump (cm)
Type	W/C (%)	s/a (%)	W	C	S	G	PP1	PP2	SP	Air-volume (%)	Slump (cm)
PPFRC1	53	60	195	453	928	666	18.2	0	1.0%	5.7	14.8
PPFRC2	53	60	195	453	928	666	0	3.6	1.0%	2.5	6.4

Flexural test

Static and dynamic flexural tests were conducted according to JCI-SF4 (Japan Concrete Institute: JCI Standard Collection, 2004) and JIS A 1106 (Japan Society of Civil Engineering: Guideline for Experiment on Materials of Civil Works, 2013), using a servo-controlled test machine as shown in Figure 3. The servo-controlled test machine can load with a constant velocity of up to 4 m/s, to investigate the effect of loading rate on the mechanical properties of materials. Figure 4 illustrates an appearance of the flexural test. Load and fulcrum reaction force were measured with load cells in the static and dynamic flexural tests for 100×100×400 mm specimens, as shown in Figure 5. A 120 mm-long strain gauge was attached on the center of the lower face of the specimen, and the strain rate was calculated from the measured strain-time history. The displacement of the L-shaped steel installed at the mid height of the specimen was measured with a laser displacement sensor in the static and dynamic flexural tests. The deflection of the specimen was calculated by subtracting the displacement at the fulcrum of the specimen from the displacement of the L-shaped steel as shown in Figure 5. Table 4 lists test cases. As test parameters in the tests were the type of FRC (PPFRC1 and PPFRC2) and strain rate of 10⁻⁶ and 10⁻¹/s, a total of 18 tests were conducted (4 or 5 specimens in each case).

Figure 3.

Servo controlled impact test machine.

Figure 4.

Appearance of flexural test.

Figure 5.

Schematic diagram of flexural test.

Table 4.

Test cases of flexural test.

Type	Static flexural test	Dynamic flexural test
Strain rate	10^-⁶/s	10^-¹/s
PPFRC1	4 cases	5 cases
PPFRC2	5 cases	4 cases

Compression test

Static and dynamic compression tests were conducted according to JIS A 1108 (Japan Society of Civil Engineering: Guideline for Experiment on Materials of Civil Works, 2013), using the servo-controlled test machine. A schematic diagram of the measurements in the static and dynamic compression tests was depicted as shown in Figure 6. Load was measured with load cells above and below the specimen which has a diameter of 10 cm and a height of 20 cm. To measure the displacement of loading plates installed above and below the specimen, a laser displacement sensor was used in the static compression tests, and an eddy current type displacement sensor was used in the dynamic compression tests. Table 5 lists test cases. As test parameters in the tests were the type of FRC (PPFRC1 and PPFRC2) and strain rate of 10⁻⁵/s and 10²/s, a total of 16 tests were conducted (4specimens in each case).

Figure 6.

Schematic diagram of compression test.

Table 5.

Test cases of compression test.

Type	Static compression test	Dynamic compression test
Loading velocity	0.03 m/s	2.0 m/s
Strain rate	10^-5/s	10⁰/s
PPFRC1	4 cases	4 cases
PPFRC2	4 cases	4 cases

Projectile impact test

Figure 7 illustrates a schematic diagram of a projectile launching machine. This machine consists of an air compressor, an air chamber, and an acceleration tube. By adjusting compressive air pressure, a projectile with a mass of 50–1000 g can be launched at velocities corresponding to 100–500 m/s. The velocity of the projectile can be measured with a resolution of 1 cm/s by laser velocity sensors installed at 50 cm intervals and universal counters at two locations near the launch port. Figure 8 displays a steel projectile used in the tests. The projectile has the tip of hemispherical shape, with a mass of 45 g and a diameter of 25 mm. The projectile was attached to a nylon fixture inserted into the air chamber, and fired with the fixture. Figure 9 illustrates the measuring items of local failure. Perforation depth or spalling depth is the distance from the front or back surface to the deepest position of the dent caused by failure. The spalling diameter and the scabbing diameter were measured in the vertical, horizontal, and diagonal 45-degree directions around the concrete plate, and the average value was taken as the diameter.

Figure 7.

Schematic diagram of projectile launching machine.

Figure 8.

Steel projectile.

Figure 9.

Measuring items.

Table 6 lists test cases. Impact tests for PPFRC1, PPFRC2, and Hybrid Fiber Reinforced Concrete (HFRC) plates were conducted at velocities of 193–423 m/s. HFRC was prepared to examine a combined effect by the PP1 and PP2 fibers on the local failure. As HFRC contains 0.3% by volume of mixing ratio of PP1 and PP2 fibers, a total of 0.6% by volume of two mixed fibers were included. The dimension of the plate specimen was 500×500 mm, and the plate thickness was 60 mm or 80 mm.

Table 6.

Test cases of impact test

(a) PPFRC1 panels
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Projectile velocity (m/s)
1	PPFRC1-6.1	PP1	2.0	6	193
2	PPFRC1-6.2				251
3	PPFRC1-6.3				279
4	PPFRC1-6.4				297
5	PPFRC1-8.1			8	298
6	PPFRC1-8.2				407
7	PPFRC1-8.3				409
8	PPFRC1-8.4				423
(b) PPFRC2 panels
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Projectile velocity (m/s)
9	PPFRC1-6.1	PP2	0.4	6	193
10	PPFRC1-6.2				251
11	PPFRC1-6.3				278
12	PPFRC1-6.4				298
13	PPFRC1-8.1			8	299
14	PPFRC1-8.2				407
15	PPFRC1-8.3				408
16	PPFRC1-8.4				423
(c) HFRC panels
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Projectile velocity (m/s)
17	HFRC-8.1	PP1 PP2	0.3 0.3	8	297
18	HFRC-8.2				374
19	HFRC-8.3				409
20	HFRC-8.4				409

Experimental results and discussions

Flexural tests

Figure 10 displays the failure status of the specimens in the static and dynamic flexural tests. One apparent crack occurred in the pure flexural section in all cases, though several cracks were expected to occur due to the crack dispersion effect. Figure 11 illustrates the relationship between flexural stress and deflection in the static test of PPFRC1, as an example. In the initial gradient of the flexural stress-deflection curve, the initial crack strength f_LOP (LOP: Limit Of Proportion) or the maximum stress point f_MOR (MOR: Modulus Of Rupture) was defined as a point of initial crack or maximum stress. Flexural stress was calculated using the following eq. (1)

σ = \frac{P l}{b h^{3}}

(1)

where P is the load, l is the span length, b is the section width of the specimen, and h is the section height of the specimen.

Figure 10.

Failure states of flexural test.

Figure 11.

Example of Relationship between flexural stress and deflection.

Figure 12 illustrates the relationship between flexural stress and deflection measured in the test. On the whole, the relationship shows that the stress of FRC after the initial crack strength f_LOP increases due to the crack bridging effect, and then reaches the flexural stress f_MOR. The initial crack strength f_LOP and the flexural strength f_MOR increase as the strain rate increases. The softening gradient after the flexural strength f_MOR in the dynamic tests became steep, and the maximum deflection decreased. Although PPFRC1 shows higher ductility failure because of the large dimension of PP1 and crack bridging effect, PPFRC2 shows brittle failure due to the short length of PP2 and weak adhesion with concrete matrix with aggregates.

Figure 12.

Relationship between flexural stress and deflection.

Figure 13 illustrates the relationship between the initial crack strength f_LOP and the strain rate. In the present study, the strain rate was calculated by averaging the strain rate between the origin and one third of the initial crack strength. The average initial crack strength f_LOP of PPFRC1 and PPFRC2 at a strain rate of 10⁻⁶/s were 6.53 N/mm² and 5.88 N/mm², respectively. The average initial crack strength f_LOP of PPFRC1 and PPFRC2 at a strain rate of 10⁻¹/s were 9.48 N/mm² and 10.4 N/mm², respectively. Hence, the dynamic increase factors (DIF_ic, equation (2)) of the initial crack strength f_LOP of PPFRC1 and PPFRC2 at a strain rate of 10⁻¹/s were 1.54 and 1.77, respectively, and the dynamic increase factor of PPFRC2 was approximately 15% greater than that of PPFRC1

D I F_{i c} = \frac{f_{i c d}}{f_{i c s}}

(2)

where f_ics is the average static initial crack strength and f_icd is the average dynamic initial crack strength. Figure 14 illustrates the DIF of the direct uniaxial tensile strength on plain concrete in the previous tests by Ross et al. (Ross et al., 1995) and CEB (CEB/FIP, 1990) and other researchers, comparing with the present test results. Comparing the dynamic increase factors of PPFRC1 and PPFRC2 in this test with the empirical formula by Ross et al. indicated that the DIF of PPFRC1 was decreased by 40%, and the DIF of PPFRC2 was decreased by 10% at a strain rate of 10⁻¹/s. Comparing the dynamic increase factors of PPFRC1 and PPFRC2 with the CEB empirical formula, the results were almost identical at a strain rate of 10⁻¹/s.

Figure 13.

Relationship between initial crack and flexural strengths and strain rate.

Figure 14.

DIF of the tensile strength.

Figure 13 illustrates the relationship between the flexural strength f_MOR and strain rate. The average flexural strengths f_MOR of PPFRC1 and PPFRC2 at a strain rate of 10⁻⁶/s were 9.48 N/mm² and 5.88 N/mm², respectively. The average flexural strengths f_MOR of PPFRC1 and PPFRC2 at a strain rate of 10⁻¹/s were 12.2 N/mm² and 10.4 N/mm², respectively. That is, both of the flexural strength f_MOR PPFRC1 and PPFRC2 plates increased as the strain rate increased. The dynamic increase factors (DIF_fs, equation (3)) of the flexural strength f_MOR of PPFRC1 and PPFRC2 at a strain rate of 10⁻¹/s were 1.27 and 1.87, and the value of PPFRC2 was approximately 50% greater than that of PPFRC1

D I F_{f s} = \frac{f_{f s d}}{f_{f s s}}

(3)

where f_fss is the average static flexural strength and f_fsd is the average dynamic flexural strength. Figure 15 illustrates the relationship between the flexural fracture energy and strain rate at a deflection of 2 mm. The flexural fracture energy (E_f, equation (4)) was calculated by integrating the relationship between the flexural stress and the deflection up to a specified deflection

E_{f} = \int_{0}^{δ} σ d δ

(4)

where E_f is the flexural fracture energy, δ is the deformation of the specimen, and σ is the flexural stress. The average flexural fracture energies of PPFRC1 and PPFRC2 at a deflection of 2 mm and a strain rate of 10^-⁶/s were 44.6 N·mm and 13.6 N·mm, respectively. The average flexural fracture energies of PPFRC1 and PPFRC2 at a deflection of 2 mm and a strain rate of 10⁰/s were 60.1 N·mm and 14.1 N·mm, respectively. PPFRC1 shows higher dynamic flexural fracture energy than that of PPFRC2 by approximately 420%. Figure 16 illustrates the relationship between the dynamic increase factor (DIF_ff, equation (5)) of flexural fracture energy at a deflection of 2 mm and the strain rate

D I F_{f f} = \frac{E_{f f d}}{E_{f f s}}

(5)

where E_ffs is the average static flexural fracture energy and E_ffd is the average dynamic flexural fracture energy. The results show that the dynamic increase factor increases as the strain rate increases. The dynamic increase factors of the flexural fracture energies of PPFRC1 and PPFRC2 at a deflection of 2 mm at a strain rate of 10⁰/s were 1.35 and 1.36, respectively, indicating that the flexural fracture energy of PPFRC1 and PPFRC2 were identical.

Figure 15.

Relationship between flexural fracture energy at 2mm deflection and strain rate.

Figure 16.

Relationship between DIF of flexural fracture energy at 2mm deflection and strain rate.

Compression test

Figure 17 illustrates the failure status of the specimen in the static and dynamic compression tests. The all FRC specimens showed shear failure with diagonal crack or splitting failure in all cases. Figure 18 illustrates the stress–strain curve of the FRC specimen. Test results showed that the compressive strength and the ultimate strain at the compressive strength increased as the strain rate increases. The average compressive strengths of PPFRC1 and PPFRC2 at a strain rate of 10⁻⁵/s were 52.0 kN/mm² and 57.0 kN/mm², respectively.

Figure 17.

Failure states of compression test.

Figure 18.

Stress-strain relationship of the FRC specimen up to the compressive strength.

The average compressive strengths of PPFRC1 and PPFRC2 at a strain rate of 10⁰/s were 58.9 kN/mm² and 74.3 kN/mm², respectively. Hence, the dynamic increase factors (DIF_c, equation (6)) of the compressive strength of PPFRC1 and PPFRC2 at a strain rate of 10⁰/s were 1.13 and 1.30, respectively, and the value of PPFRC2 was approximately 15% greater than that of PPFRC1

D I F_{c} = \frac{f_{c d}^{'}}{f_{c s}^{'}}

(6)

where f’_cs is the average static compressive strength, f ^′_cd is the average dynamic compressive strength. Figure 19 illustrates the DIF of the compressive strength in the previous tests of plain concrete by other researchers (Watstein, 1953; Tang et al., 1992; Dilger et al., 1984), and the CEB formula. Compared with the CEB formula, the dynamic increase factor of the compressive strength of PPFRC1 and PPFRC2 at each strain rate range was approximately decreased by 10–30%. Figure 20 illustrates the DIF of the compressive strength in the previous tests of FRC (Chen et al., 2019; Liu et al., 2020). Chen et al. (Chen et al., 2019) conducted split Hopkinson bar tests on FRC specimens containing 0.1% volume of polypropylene fibers with a length of 19 mm and a diameter of 26 μm. Liu et al. (Liu et al., 2020) conducted split Hopkinson bar tests on FRC containing 0.1% volume of carbon fibers which had a length of 6 mm and a diameter of 7 μm. Compared with the results by Chen et al. (Chen et al., 2019) and Liu et al. (Liu et al., 2020), the DIF of the compressive strength of PPFRC1 at a strain rate of 10⁰/s and the DIF of the compressive strength of specimen by Chen et al. (Chen et al., 2019) at a strain rate of 10²/s were almost identical.

Figure 19.

DIF of compressive strength.

Figure 20.

Relationship between DIF of compressive strength and strain rate.

The average ultimate strain of PPFRC1 and PPFRC2 at a strain rate of 10⁻⁵/s were 4800 μ and 5330 μ, respectively. The average ultimate strain of PPFRC1 and PPFRC2 at a strain rate of 10⁰/s were 5480 μ and 6290 μ, respectively. The previous studies reported that the average ultimate strain of plain concrete at strain rates of 10⁻⁵/s and 10⁰/s were 2029 μ and 2321 μ (Iwane & Experimental, 2017). The results show that the ultimate strain of PPFRC1 and PPFRC2 is approximately 250% greater than that of plain concrete.

Figure 21 illustrates the relationship between stress and deformation after the compressive strength. The residual stress after the peak stress remains in the static compression test even when the strain reaches 5 mm. Both PPFRCs shows small residual strain and steeper stress-softening as the strain rate increases. Figure 22 illustrates the average compressive fracture energy at a deformation of 1 mm. The compressive failure energy was calculated by integrating the stress-deformation curve up to a specified deformation. The average static compressive failure energies of PPFRC1 and PPFRC2 were 43.6 N/mm and 35.3 N/mm, respectively. The average dynamic compressive energies of PPFRC1 and PPFRC2 were 33.1 N/mm and 28.1 N/mm, respectively. Figure 23 illustrates the relationship between the dynamic increase factor (DIF_cfe, equation (7)) of the compressive fracture energy at a deformation of 1 mm and the strain rate. The dynamic increase factors of PPFRC1 and PPFRC2 at a strain rate of 10⁰/s were 0.76 and 0.80, respectively

D I F_{c f e} = \frac{E_{c f d}}{E_{c f s}}

(7)

where E_cfs is the average static compressive fracture energy, E_cfd is the average dynamic compressive fracture energy. The dynamic increase factor of the compressive energy becomes small as the strain rate increases due to the brittle failure.

Figure 21.

Relationship between compressive stress and strain on post-peak range.

Figure 22.

Relationship between compressive fracture energy and strain rate.

Figure 23.

Relationship between DIF of compressive fracture energy and strain rate.

Projectile impact tests

Table 7 lists test results of local failure obtained from the impact tests. Figure 24 –26 show failure status of each FRC plates. Figure 24 shows the failure status of PPFRC1 plates. The failure mode of 6 cm-thick PPFRC1 plates was spalling at the impact velocities corresponding to 193 m/s and 251 m/s. As the failure mode of 6 cm-thick PPFRC1 plate was scabbing at the impact velocity corresponding to 279 m/s, the scabbing limit velocity was between 251 and 279 m/s. The failure mode of 8 cm-thick PPFRC1 plate at the impact velocities corresponding to 407–409 m/s was spalling. The failure mode of 8 cm-thick PPFRC1 plate was scabbing at the impact velocity of 423 m/s.

Table 7.

Failure states and local failure dimension of FRC plate

(a) PPFRC1 plates
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Velocity (m/s)	Failure State	Spalling Depth (cm)	Spalling Diameter (cm)	Scabbing Depth (cm)	Scabbing Diameter (cm)
1	PPFRC1-6.1	PP1	2.0	6	193	Spalling	1.7	8.1
2	PPFRC1-6.2				251	Spalling	1.7	8.1
3	PPFRC1-6.3				279	Spalling	1.9	8.5	3.2	15.8
4	PPFRC1-6.4				297	Spalling	1.9	8.3	3.9	14.4
5	PPFRC1-8.1			8	298	Spalling	2.4	8.3
6	PPFRC1-8.2				407	Spalling	2.8	9.0
7	PPFRC1-8.3				409	Spalling	2.5	9.7
8	PPFRC1-8.4				423	Scabbing	2.7	13.7	4.5	19.4
(b) PPFRC2 plates
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Velocity (m/s)	Failure State	Spalling Depth (cm)	Spalling Diameter (cm)	Scabbing Depth (cm)	Scabbing Diameter (cm)
9	PPFRC1-6.1	PP2	0.4	6	193	Spalling	4.3	6.9
10	PPFRC1-6.2				251	Spalling	1.5	7.9
11	PPFRC1-6.3				278	Scabbing	2.0	9.9	2.8	16.6
12	PPFRC1-6.4				298	Scabbing	2.3	9.8	3.5	20.9
13	PPFRC1-8.1			8	299	Spalling	2.0	10.8
14	PPFRC1-8.2				407	Scabbing	2.2	12.8
15	PPFRC1-8.3				408	Spalling	2.2	12.3
16	PPFRC1-8.4				423	Scabbing	2.9	12.6	4.6	17.9
(c) HFRC plates
No	Case	Fiber	Mixing ratio (%)	Thickness (cm)	Velocity (m/s)	Failure State	Spalling Depth (cm)	Spalling Diameter (cm)
17	HFRC-8.1	PP1 PP2	0.3 0.3	8	297	Spalling	2.0	8.2
18	HFRC-8.2				374	Spalling	2.3	10.1
19	HFRC-8.3				409	Spalling	1.8	12.1
20	HFRC-8.4				409	Spalling	2.4	10.0

Figure 24.

Failure state of PPFRC1.

Figure 25.

Failure states of PPFRC2.

Figure 26.

Failure state of HFRC.

Figure 25 displays the failure status of the PPFRC2 plates. The failure mode of 6 cm-thick PPFRC2 plate was spalling at the impact velocities corresponding to 193–251 m/s. As the failure mode of 6 cm-thick PPFRC2 plates was scabbing at the impact velocities corresponding to 278–298 m/s, the scabbing limit velocity was between 251 and 278 m/s. The failure mode of 8 cm-thick PPFRC2 plate was spalling at the impact velocity of 299 m/s. The failure mode of 8 cm-thick PPFRC2 plate was scabbing limit at the impact velocity of 408 m/s. Although the slight difference of failure limit between PPFRC1 and PPFRC2 plates might be caused by the difference in fiber orientation, and their scabbing limit and the perforation limit thickness of FRC were identical. For reference, the failure mode of 8 cm-thick PPFRC2 plate was scabbing at the impact velocity of 423 m/s. At the impact velocities corresponding to 408–423 m/s, flexural cracks also occurred in 8 cm-thick PPFRC2 plate.

Figure 26 shows the failure status of the HFRC plate. The failure mode of 8 cm-thick HFRC plate were all spalling at the impact velocity corresponding to 297–409 m/s. Compared with the failure status of PPFRC1 and PPFRC2, the back surface of HFRC had more cracks than those of PPFRC1and PFRC2, indicating their higher energy absorption.

Average penetration depth of PPFRC1 was 15% higher than that of PPFRC2. This is due to the difference of compressive strength at the strain rate of 10⁰/s.

Experimental results showed that the average penetration depth of PPFRC1 plate was approximately 10% greater than that of PPFRC2 plate. From the compressive strength test, the static and dynamic compressive strength of PPFRC1 was approximately 10–20% less than that of PPFRC2, and the compressive failure energy of PPFRC1 was 20% greater than that of PPFRC2. Assuming that the penetration depth could be caused by plastic deformation of concrete, the strength and the absorbed energy after compressive strength affect the penetration depth. Whereas, as the stress state in the penetration process remains in the tri-axial condition, the effect of the uniaxial mechanical properties on the penetration depth would be limited. The experimental result revealed that the compressive strength had a greater effect on the penetration depth than the fracture energy.

The failure modes of PPFRC1 and PPFRC2 plates, which had different strengths and fiber volumes, were almost identical. However, fragmentation of concrete in the rear face was more suppressed in PPFRC2 plate. From the flexural test results, the static and dynamic flexural strengths of PPFRC1 were 60% and 20% greater than those of PPFRC2, respectively. The static and dynamic tensile fracture energies of PPFRC1 were three to four times greater than those of PPFRC2. However, the cross-section of the PPFRC1 plate exhibited a wide opened crack, while multiple cracks were observed in PPFRC2 plate. In other words, PPFRC2 plate generated a large number of fine cracks and absorbed the impact energy. As two PPFRC plates had the identical suppressing effect, it is necessary to investigate the relationship between the mechanical properties of the material and the crack dispersion capability, and the influence of the properties on the impact resistance of the plate.

Comparison of scabbing limit thickness by the modified NDRC formula

The previous studies have reported that the FRC plates has a higher effect of suppressing local failure than that of plain concrete and mortar plates. In the present study, the modified NDRC formula was used to evaluate the scabbing limit of the FRC plates.

Equations (8), (9), and (10) show the formula for the penetration depth by the modified NDRC formula, and equations (11) and (12) show the evaluation formula for the scabbing limit plate thickness

\frac{x}{d} = 2 G^{\frac{1}{2}} \frac{x}{d} \leq 2

(8)

\frac{x}{d} = G + 1 \frac{x}{d} \geq 2

(9)

G = 3.8 \times 10^{- 5} \frac{N M}{d \sqrt{f_{c}}} {(\frac{V_{0}}{d})}^{1.8}

(10)

\frac{s}{d} = 7.91 (\frac{x}{d}) - 5.06 {(\frac{x}{d})}^{2} \frac{x}{d} \leq 0.65 o r \frac{s}{d} \leq 3

(11)

\frac{s}{d} = 2.12 + 1.36 (\frac{x}{d}) 0.65 \leq \frac{x}{d} \leq 11.75 o r 3 \leq \frac{s}{d} \leq 18

(12)

where x is the penetration depth (m), d is the diameter of projectile (m), N is the shape modulus of projectile (0.84 for hemisphere shape), M is the mass of projectile (kg), V₀ is the impact velocity corresponding to projectile (m/s), f_c is the compression strength of concrete (N/mm²), and s is the limit scabbing thickness (m).

Figure 27 shows the comparison of the experimental penetration depth with those evaluated by the modified NDRC formula in the relationship between the penetration depth and impact velocity. The results show that PPFRC has slight effect in suppressing penetration depth. For reference, the penetration depth of plain concrete calculated by the modified NDRC formula is multiplied by the reduction modulus α to match the experimental results of the PPFRC1 and PPFRC2 plates. The reduction modulus, α = 0.9, were determined to be compatible with the failure mode. This reduction modulus indicates suppressing rate of penetration depth. It can be observed that the penetration depth of PPFRC1 and PPFRC2 plates is approximately 10% smaller than that of plain concrete plate. Figure 28 illustrates the failure mode of the FRC plates in this test, marked in the relationship between the plate thickness and the impact velocity. In Figure 28, the scabbing limit of a plain concrete plate calculated by the modified NDRC formula is multiplied by the reduction modulus α1 and α2 to match the experimental results of the PPFRC1 and PPFRC2 plates. Specifically, reduction modulus, α1 = 0.86 and α2 = 0.75, were determined to be compatible with the failure mode. This reduction modulus indicates suppressing rate of the scabbing limit thickness of the plain concrete plate. It can be observed that the scabbing limit slab thickness of PPFRC1 and PPFRC2 plates is approximately 15%–25% smaller than that of plain concrete plate. By comparing with similar type of FRC which contains 2.0% volume of PP fiber has a length of 0.7 mm and a diameter of 30 mm, performed by Ueno et al. (Ueno et al., 2017), it was found that it has the identical effect of suppressing local failure.

Figure 27.

Comparison of penetration depth.

Figure 28.

Comparison of limit thickness.

Li et al. (Li et al., 2006) derived the relationship between the limit thickness and impact energy by adapting the modified NDRC formula. The relationship between the scabbing limit thickness and the impact energy will be discussed by considering the reduction factors associated with the PPFRC plates that were proposed by Li et al. (Li et al., 2006).

Li et al. (Li et al., 2006) derived the impact energy that causes scabbing or perforation by arranging the scabbing and perforation limit thicknesses of the modified NDRC formula. The relationship between the limit thickness and the impact energy can be obtained by arranging equations (11) and (12), as shown below.

The relationship between the scabbing limit thickness [m], H_s and impact energy [J], E_k is shown as

(i) \frac{x}{d} \leq 0.65 o r \frac{H_{s}}{d} < 3

\frac{E_{k}}{d^{3} f_{c}^{'}} = {(\frac{V_{0}}{d})}^{0.2} f_{c}^{' \frac{1}{2}} {(4.851 \times 10 - \sqrt{c_{1}})}^{2}

(13)

where

C_{1} = 2.395 \times 10^{3} - 7.748 \times 10^{2} (\frac{H_{s}}{d})

(14)

(ii) 0.65 < \frac{x}{d} < 2 o r 3 < \frac{H_{s}}{d} < 4.84

\frac{E_{k}}{d^{3} f_{c}^{'}} = {(\frac{V_{0}}{d})}^{0.2} f_{c}^{' \frac{1}{2}} {[4.851 \times 10 (\frac{H_{s}}{d}) - 9.760 \times 10]}^{2}

(15)

(iii) 2 < \frac{x}{d} < 11.75 o r 4.84 < \frac{H_{s}}{d} < 18.0

\frac{E_{k}}{d^{3} f_{c}^{'}} = {(\frac{V_{0}}{d})}^{0.2} f_{c}^{' \frac{1}{2}} {[1.151 (\frac{H_{s}}{d}) - 2.558]}^{2}

(16)

Figure 29 illustrates the relationship between the plate thickness and the impact energy, and Figure 30 shows the comparison of scabbing limit energies between a plane concrete plate and the PPFRC plate. The scabbing limit energies of 6 cm-thick and 8 cm-thick of plain concrete plate were 429J and 1822J, respectively. The scabbing limit energies of 6 cm-thick and 8 cm-thick of the PPFRC plate were 1822J and 4368J, respectively. Therefore, the scabbing limit energy of the 6 cm-thick of PPFRC plate was approximately 330% greater than that of a 6 cm-thick plain concrete plate. The scabbing limit energy of 8 cm-thick of PPFRC plate was approximately 130% greater than that of 8 cm-thick plain concrete plate.

Figure 29.

Comparison of scabbing limit energy.

Figure 30.

Scabbing limit energy of FRC plate.

Conclusion

This study investigated local failure resistance of polypropylene fiber reinforced concrete plates by conducting projectile impact tests. In the test, two types of PP fibers with different dimensions were employed and the effect of suppressing local failure due to the mixing of PP fibers was examined. The conclusions drawn from this study are listed as follows.

1. From the flexural test, the static and dynamic flexural strengths of PPFRC1 were 60% and 20% greater than those of PPFRC2, respectively, and the static and dynamic tensile fracture energies of PPFRC1 were three to four times greater than those of PPFRC2. Although their tensile properties were significantly different, the failure modes of PPFRC1 and PPFRC2 plates were almost identical. Observation of the cracks in the cross-section showed that a wide opened crack was observed in the cross-section of PPFRC1 plate, while multiple cracks were observed in the cross-section of PPFRC2 plate. Hence, crack dispersion capability affected the impact-resistant performance of the PPFRC plates.

2. From the compressive test, the static and dynamic compressive strengths of PPFRC1 was approximately 10–20% less than that of PPFRC2, and the compressive failure energy of PPFRC1 was 20% greater than that of PPFRC2. Considering the impact test result that the average penetration depth of PPFRC1 plate was approximately 10% greater than that of PPFRC2 plate, it was revealed that the compressive strength had a greater effect on the penetration depth than the fracture energy.

3. From the projectile tests, it was found that the FRC plate containing 2.0% volume of PP1 fibers and the FRC plate containing 0.4% volume of PP2 fibers had identical suppressing effect of local failure. From the comparison with the modified NDRC formula, the scabbing limit thickness of the PPFRC1 and PPFRC2 plates was approximately 15–25% smaller than that of plain concrete plate. The scabbing limit energy of 8 cm-thick of PPFRC plate was approximately 130% greater than that of 8 cm-thick plain concrete plate.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Koki Mori

References

Almusallam

Siddiqui

Iqbal

, et al. (2013) Response of hybrid-fiber reinforced concrete slabs to hard projectile impact. International Journal of Impact Engineering 58: 17–30.

Beppu

Kataoka

Ichino

, et al. (2020) Failure characteristics of UHPFRC plates subjected to projectile impact. Composite Part B 182: 107505.

Beppu

Miwa

Itoh

, et al. (2008) Damage evaluation of concrete plates by high-velocity impact. International Journal of Impact Engineering 35(12): 1419–1426.

CEB/FIP (1990) CEB/FIP Model Code for Concrete Structure, pp. 48–51.

Chang

(1981) Impact of solid missile on concrete barriers. Journal of Structural Division ASCE 107(ST2): 257–271.

Chen

Huang

, et al. (2008) Normal perforation of reinforced concrete target by rigid projectile. International Journal of Impact Engineering 35(10): 1119–1129.

Chen

Zhong

, et al. (2019) Experimental study on dynamic compressive behaviour of recycled tyre polymer fibre reinforced concrete. Cement and Concrete Composites 98: 95–112.

Dancygier

Yankelevsky

(1996) High strength concrete response to hard projectile impact. International Journal of Impact Engineering 18(6): 583–599.

Dancygier

Yankelevsky

(1999) Effects of reinforced concrete states on resistance to hard projectile impact. ACI Structura Journal 96(2): 259–267.

10.

Dilger

Koch

Kowalczyk

(1984) Ductility of plain and confined concrete under different strain rates. ACI J 81: 73–81.

11.

Feng

Gao

, et al. (2019) Penetration resistance of hybrid-fiber-reinforced high-strength concrete under projectile multi-impact. Construction and Building Materials 202: 341–352.

12.

Fenu

Forni

Cadoni

(2016) Dynamic behavior of cement mortars reinforced with glass and basalt fibres. Composite Part B 92: 142–150.

13.

Hughes

(1984) Hard missile impact on reinforced concrete. Nuclear Engineering and Design 77: 23–35.

14.

Iwane

Experimental (2017) Study on Dynamic Characteristic of Fiber Reinforced Cementitious composite, Graduation Master Thesis of. National Defense Academy of Japan. [in Japanese].

15.

Japan Concrete Institute: JCI Standard Collection , pp.57–72, 2004.

16.

Japan Society of Civil Engineering: Guideline for Experiment on Materials of Civil Works , pp.97–180, 2013.

17.

Kanda

(1999) Effect of fiber strength and fiber-matrix interface on crack bridging in cement composites. Journal of Engineering Mechanics 125: 290–299.

18.

Kennedy

(1976) A review of procedures for the analysis and design of concrete structures to resist missile impact effects. Nuclear Engineering and Design 37(2): 183–203.

19.

Kheyroddin

Arshadi

Ahadi

, et al. (2021) The impact resistance of fiber-reinforced concrete with polypropylene fibers and GFRP wrapping. Materials Today: Proceedings 45(6): 5433–5438.

20.

Kishi

Konno

Ikeda

, et al. (2002) Prototype impact tests on ultimate impact resistance of PC rock-sheds. International Journal of Impact Engineering 27(9): 969–985.

21.

Leung

CKY

(1991) New strength-based model for the debonding of discontinuous fibres in an elastic matrix. Journal of Materials Science 26: 5996–6010.

22.

Chen

(2003) Dimensionless formulae for penetration depth of concrete target impacted by a non-deformable projectile. International Journal of Impact Engineering 28(1): 93–116.

23.

Reid

Wen

, et al. (2005) Local impact effects of hard missiles for concrete targets. International Journal of Impact Engineering 32(1–4): 224–284.

24.

Reid

Ahmad-Zaidi

(2006) Critical impact energies for scabbing and perforation of concrete target. Nuclear Engineering and Design 236(11): 1140–1148.

25.

Liu

Bai

, et al. (2020) Dynamic compressive mechanical properties of carbon fiber-reinforced polymer concrete with different polymer-cement ratios at high strain rates. Construction and Building Materials 261: 119995.

26.

Liu

, et al. (2021) Projectile impact resistance of fibre-reinforced geopolymer-based ultra-high performance concrete (G-UHPC). Construction and Building Materials 290: 123189.

27.

Martin

(1994) Modeling of local impact effects on plain and reinforce concrete. ACI Structural Journal 91(2): 178–187.

28.

Miyamoto

King

Fuji

(1991) Nonlinear dynamic analysis of reinforced concrete slabs under impulsive loads. ACI Struct J 88(4): 538–545.

29.

Mohamed

Ahmed

(2014) Effect of reinforcement on the response of concrete plates to impact of hard projectiles. International Journal of Impact Engineering 63: 1–17.

30.

Nam

Shinohara

Atou

, et al. (2016) Comparative assessment of failure characteristics on fiber-reinforced cementitious composite panels under high-velocity impact. Composites Part B 99: 84–97.

31.

Ong

KCG

Basheerkhan

Paramasivam

(1999) Resistance of fibre concrete slabs to low velocity projectile impact. Cement and Concrete Composites 21(5–6): 391–401.

32.

Othman

Marzouk

(2016) An experimental investigation on the effect of steel reinforcement on impact response of reinforced concrete plates. International Journal of Impact Engineering 88: 12–21.

33.

Ramakrishna

Sundararajan

(2005) Impact strength of a few natural fibre reinforced cement mortar slabs: a comparative study. Cement and Concrete Composites 27(5): 547–553.

34.

Redon

, et al. (2001) Measuring and modifying interface properties of PVA fibers in ECC matrix. Journal of Materials in Civil Engineering 13(6): 399–406.

35.

Ross

Tedesco

Kuennen

(1995) Effects of strain rate on concrete strength. ACI Materials Journal 92(1): 37–47.

36.

Shasha

Hoang

TNL

Leong

, et al. (2017) Effect of high strain rate on compressive behavior of strain-hardening. Construction and Building Material 136: 31–43.

37.

Tai

(2009) Flat ended projectile penetrating ultra-high strength concrete plate target. Theoretical and Applied Fracture Mechanics 51: 117–128.

38.

Tang

Malvern

Jenkins

(1992) Rate effects in uniaxial dynamic compression of concrete. Journal of Engineering Mechanics 118(1): 108–124.

39.

Ueno

Beppu

Ogawa

(2017) A method for evaluating the local failure of short polypropylene fiber-reinforced concrete plates subjected to high-velocity impact with a steel projectile. International Journal of Impact Engineering 105: 68–79.

40.

Wang

Chouw

(2018) Experimental and theoretical studies of flax FRP strengthened coconut fibre reinforced concrete slabs under impact loadings. Construction and Building Materials 171: 546–557.

41.

Wang

Mindness

(1996) Fibre reinforced concrete beams under impact loading. Cement and Concrete Research 26(3): 363–376.

42.

Watstein

(1953) Effect of straining rate of the compressive strength and elastic states of concrete. ACI J 49(8): 729–744.

43.

Wei

, et al. (2021) Hybrid fibre reinforced ltra-high performance concrete beams under static and impact loads. Engineering Structures 245: 112921.

44.

Werner

Thienel

Kustermann

(2013) Study of fractured surfaces of concrete caused by projectile impact. International Journal of Impact Engineering 52: 23–27.

45.

Song

Feng

, et al. (2021) Development of basalt fiber engineered cementitious composites and its mechanical properties. Construction and Building Materials 266: 121173. Part B.

46.

Zineddin

Krauthammer

(2007) Dynamic response and behavior of reinforced concrete slabs under impact loading. International Journal of Impact Engineering 34(9): 1517–1534.