Composites Part B 78 (2015) 431e446

Contents lists available at ScienceDirect

Composites Part B
journal homepage: www.elsevier.com/locate/compositesb

Assessment of exural toughness and impact resistance of

bundle-type polyamide ber-reinforced concrete
Hongseop Kim a, Gyuyong Kim a, *, Nenad Gucunski b, Jeongsoo Nam c, Joongkyu Jeon d
a

Department of Architectural Engineering, Chungnam National University, 99 Daehak-ro, Yuseong-gu, Daejeon 305-764, Republic of Korea
Department of Civil and Environmental Engineering, Rutgers University, Piscataway, NJ 08855-0909, USA
c
Structure Engineering Research Center, Tokyo Institute of Technology, 4259 Nagatsuta, Midori, Yokohama 226-8503, Japan
d
Technology Resrearch Lab., Kolon Global Corp., 46 Jeondae-ro 16-Gil, Pogok-Eup, Cheoin-gu, Yongin-si, Gyeonggi 449-815, Republic of Korea
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 27 November 2014
Received in revised form
6 April 2015
Accepted 8 April 2015
Available online 17 April 2015

This study compares the ber/matrix bonding strength and exural properties of bundle-type polyamide
bers to those of hooked-end steel bers. Their fracture behavior upon impact with a high-velocity
projectile is also assessed in terms of penetration depth, crater diameter and rear-side scabbing. The
results obtained demonstrate that the bundle-type polyamide bers undergo fracture without ber
pullout because of the increased interber gap and specic surface area for bonding, but exhibit poorer
exural fracture behavior with a lower exural strength and fracture energy when compared to hookedend steel bers. Yet despite this, concrete reinforced with bundle-type bers is shown to more effectively
suppress scabbing during high-velocity impact, which is attributed to a more efcacious dispersion of
shock stress due to the increased number of individual bers.
2015 Elsevier Ltd. All rights reserved.

Keywords:
A. Fibres
B. Mechanical properties
B. Fracture toughness
B. Fibre/matrix bond
B. Impact behaviour

1. Introduction
The reinforcing of concrete with short bers is designed to
improve its dynamic loading, impact resistance and explosion
resistance by increasing the tensile strength and toughness, as well
as inhibiting crack formation and growth through the bridging of
the bers. Past studies have conrmed the safety advantages
offered by ber-reinforced concrete, in that it helps to prevent
brittle fracture induced by impact or explosion. In the literature
pertaining to the fracture mechanism of concrete structures, it is
compressive strength (stiffness) that has generally been associated
with projectile penetration, while exural tensile toughness is
related to scabbing.
The use of ber-reinforced concrete in construction and repair/
retrotting applications requires a combination of high exural
toughness, tensile strength, shear strength, crack control and
impact resistance. Performance therefore varies widely depending
on the effectiveness of the ber used, with Fig. 1 providing a
schematic diagram of loadedisplacement relationship for when a
ber is pulled out of the matrix [1]. The high exural tensile
strength and crack resistance offered by steel reinforcing bers

* Corresponding author.
E-mail address: gyuyongkim@cnu.ac.kr (G. Kim).
http://dx.doi.org/10.1016/j.compositesb.2015.04.011
1359-8368/ 2015 Elsevier Ltd. All rights reserved.

makes them a popular choice in construction; however, the high

specic gravity and stiffness of steel bers can cause concrete pump
tubes to rupture and increases the rebound volume of shotcrete,
not to mention the problems associated with corrosion of the bers
[2e5]. To overcome this, the use of lightweight organic bers with
mechanical properties similar to steel, but a higher resistance to
corrosion, has been explored [6,7].

1.1. Literature review

Consequently, it has been shown by Anderson et al. [8] that the
penetration resistance of concrete (i.e., the penetration depth) is
not greatly inuenced by the type of ber type, but only by ber
content within a limited practical range. Moreover, their results
showed that greater ber contents lead to a smaller crater volume.
Ramakrishnan et al. [9] reported that reinforcement with hookedend bers increases the low-velocity impact resistance of shotcrete
during hammer drop testing when compared to plain shotcrete.
Meanwhile, Dancygier and Yankelevsky [10,11] observed that using
steel bers inhibits the brittle fracture of concrete impacted by a
high-velocity projectile, while also tending to suppress crack
propagation and minimize the crater area. This led to investigations
by Mohamed et al. [12], O'Neil et al. [13], Zhang et al. [14], Vossoughi et al. [15], and G.Y. Kim et al. [16,17] into the fracture

432

H. Kim et al. / Composites Part B 78 (2015) 431e446

increase the area of ber/matrix bonding interface, as manufactured by KOLON Co. of Korea. The polyamide ber was 30 mm in
length, 0.5 mm in diameter, with an aspect ratio of 60, tensile
strength of 594 MPa, and density of 1.14 g/cm3. Fig. 2(b) and (c)
shows a schematic diagram and the air nozzle for producing
polyamide bers. The production mechanism is air interlacing of
multi-laments consisting of core yarns and effect yarns under high
pressure. The yarn structure of the effect yarns forming a loop
around the core yarns offers the advantages of less elongation,
resistance against the tension-induced swelling reduction, and
improved ber/matrix bonding strength. As shown in the magnied image of polyamide bers in Fig. 2(d), the interber gaps keep
the bers from intermingling during mixing with concrete, thus
increasing the likelihood of improving the ber/matrix interface
bonding. Fig. 3 shows the improve owability by surface coating of

Fig. 1. Schematic diagram of loadedisplacement relationship for ber pulled out of the
matrix.

resistance of concrete under impact when it is reinforced with

different types and shapes of ber by looking at the change in
exural tensile strength. Many of these researchers have reported
that ber-reinforcement reduces impact-induced crack formation
and scabbing, as the bridging effect of the bers increases the
exural and tensile strength in proportion to the ber volume
fraction.
Some researchers have also reported improved fractureresistance through a hybrid of steel and organic bers, with the
polypropylene/steel hybrid of Almusallam et al. [18] successfully
reducing impact damage and fragmentation by inhibiting impactinduced crack propagation. Beppu et al. [19e23] have reported
that increasing the tensile strength of concrete improves its penetration resistance and tends to prevent scabbing, but means that
any scabbing that does occur is likely to cause brittle fracture.
Contrary to this, however, a polypropylene/steel hybrid exhibits both
an enhanced compressive toughness and exural toughness, which
inhibits the propagation of diagonal tension cracks, and reduces the
depth and diameter of scabbing.
1.2. The purpose of research
In this study, a bundle-type polyamide ber very different from
conventional reinforcing organic bers was developed with a view
to improving the problems of inadequate dispersion often
encountered when mixing with concrete. To this end, the ber/
matrix bonding strength and exural behavior of ber reinforced
concrete was measured, and its behavior upon impact with a highvelocity projectile was assessed by measuring the penetration
depth/diameter on the front side and scabbing depth/diameter on
the rear. These values are compared against those of concrete
reinforced with hooked-end steel bers of the same length that are
currently widely used in construction.
2. Materials and methods
2.1. Experimental bers
As shown in Fig. 2(a), polyamide ber was manufactured by air
injection of 544 micro nylon bers with 19.5 m-diameter to

Fig. 2. Bundle type polyamide ber.

H. Kim et al. / Composites Part B 78 (2015) 431e446

the polyamide ber-reinforced concrete. As shown in Fig. 3(a), ber

ball and bleeding was occurred because polyamide bers are hydrophilic. To solve these problems, polyamide ber surface was
coated with dispersant consisting of 40e50% polyhydric alcohol
esters of lubricant, 30e40% anionic surfactant, and 10e30% antistatic, as shown in Fig. 4. Since this coating agent is composed of a
hydrophilic property and a lipophilic property, during mixing,
owability of polyamide ber reinforced concrete was improved
(Fig. 3(b)).
The hooked-end steel ber used in the study was 30 mm in
length, 0.5 mm in diameter, with an aspect ratio of 60, tensile
strength of 1140 MPa, and density of 7.8 g/cm3 as manufactured by
KOSTEEL Co. of Korea (Fig. 5).

Fig. 4. Surface coating of polyamide ber.

tmax
2.2. Concrete mix and used materials
Table 1 gives the mix proportions used to evaluate the performance of polyamide ber reinforced concrete and hooked-end
steel ber reinforced concrete. Ordinary Portland cement and yash were used in the study. Crushed aggregate (maximum size of
20 mm) was used for the coarse aggregate, and river sand
(maximum size of 5 mm, specic gravity of 2.61 g/cm3) was used
for the ne aggregate, as shown in Table 2.

433

Pmax
pDL

(1)

where tmax is the bonding strength, Pmax is the maximum load, D is

the diameter of the ber, and L is the embedded length of the ber.
Fig. 7 shows exural test setup. The exural properties of the
unnotched beam were performed according to ASTM C1609 [25] to
determine the exural properties of the ber reinforced concrete.
The test was performed on three specimens (100
100
400 mm)
for each test level, and the exural strength was determined using
formula (2).

Pmax $l
bh2

2.3. Experimental designs and methods

fr

Table 3 shows the experimental design for assessing the exural

properties of polyamide ber-reinforced concrete and hooked-end
steel ber-reinforced concrete. The experiments were performed
using the same volume fraction to compare polyamide ber and
hooked steel ber. The ber volume fraction were 0.5, 0.75 and 1.0%
of the volume ratio. We also produced non-reinforced concrete. The
three specimen groups will be referred to as PA (bundle type
polyamide bers), SF (hooked-end steel bers), and plain (nonreinforced concrete).
Table 4 shows the slump, air content and compressive strength.
As the ber volume fraction increased from 0.5 to 1.0%, the slump
also decreased regardless of the ber type. The target slump could
be achieved by using a water-reducing agent. The compressive
strength were measured from 43.80 to 52.30 MPa. These values
satised the design compressive strength at 28 days.
Fig. 6 shows the pullout test setup. Half the length of a ber
(15 mm) was embedded in the middles of the specimen, which
had a cross-sectional area of 25 mm
25 mm. The bond strength
at maximum pullout load was determined using formula (1)
[24].

where fr is exural strength (MPa), Pmax is the load (N), l is the span
length (mm), b is average width of the specimen at the fracture, as
oriented for testing (mm), and h is the height of the fractured crosssection (mm). The equivalent exural strength (MPa) was determined using formula (3).

fr0

Ab $l
dtb bh2

(2)

(3)

where fr0 is the equivalent exural strength (MPa), dtb is 1/150 span
deection (mm), Ab is the area of the load-deection curve to dtb , l
is the span (mm), b is the breadth of the fractured cross-section
(mm), and h is the height of the fractured section (mm).
The exural properties of the notched beam were performed
according to RILEM 50-FMC Draft Recommendation [26] to determine the facture energy of the ber reinforced concrete. Specimen
had notch, with 50 mm depth at the center of beam, which were cut
by using a circular diamond concrete saw. The fracture energy was
determined using formula (4).

Fig. 3. Improve owability by surface coating.

434

H. Kim et al. / Composites Part B 78 (2015) 431e446

Fig. 5. Hooked-end steel ber.

Table 1
Mix proportions.
W/B Slump (mm) Air (%) S/a (%) Unit weight (kg/m3)
Cement Water Fly-ash Sand Gravel
0.4

120 30

41

55

405

180

45

888

758

Table 2
Mechanical properties of the used materials.
Materials

Mechanical properties

Cement

Ordinary portland cement

Density: 3.15 g/cm3, Fineness: 3200 cm2/g
Density: 2.20 g/cm3, Fineness: 3000 cm2/g
Density: 2.61 g/cm3, Absorptance: 0.81%
Crushed gravel, Maximum size: 20 mm
Density: 2.65 g/cm3, Absorptance: 0.76%
Polycarboxylic acid type

Fly-ash
Sand
Gravel
Super plasticizer

Table 3
Details of concrete mixes.
Mix no.

1
2
3
4
5
6
7

ID.

Plain
PA-0.50
PA-0.75
PA-1.00
SF-0.50
SF-0.75
SF-1.00

Details of ber
Type

Volume fraction (%)

e
Polyamide ber
Polyamide ber
Polyamide ber
Hooked steel ber
Hooked steel ber
Hooked steel ber

0
0.50
0.75
1.00
0.50
0.75
1.00
Fig. 6. Pullout test set up.

Table 4
Fresh properties and compressive strength.
ID.

Slump (mm)

Air (%)

Compressive strength (MPa)

Plain
PA-0.50
PA-0.75
PA-1.00
SF-0.50
SF-0.75
SF-1.00

135
130
125
125
135
135
120

3.9
4.3
4.2
3.0
4.6
4.8
3.4

56.6
55.4
50.2
50.4
51.5
54.5
52.3

.
GF W0 mgd0 Alig

acceleration due to gravity (9.81 m/s2), d0 is deformation at the nal

of the beam (m), Alig is the cross-sectional area of the fracture (m2).
3. Pullout behavior of ber and exural properties of
concrete
3.1. Pullout behavior

(4)

where GF is the fracture energy (N m/m2), W0 is the area of the load

versus CMOD curve (N m), m is m1 2m2 , m1 is the weight of the
beam between the supports (kg), m2 is the weight of the part of the
loading arrangement which is not attached to the machine, g is

Fig. 8 shows the pullout loadedisplacement curve and the

fracture image of the ber. The polyamide ber had a smaller
maximum pullout load and displacement, compared with the
hooked-end steel ber. The polyamide ber achieved maximum
load with increasing displacement, whereas the hooked-end steel
ber achieved maximum pullout load earlier because elongation of
polyamide ber had a greater than that of the hooked-end steel
ber. After the maximum load, polyamide ber was stretched and
cutoff, as shown in Fig. 8. The bond stress generated by the micro
gaps of the bundle-type polyamide bers is assumed to lead to

H. Kim et al. / Composites Part B 78 (2015) 431e446

435

Fig. 8. Pullout loadedisplacement curve.

Fig. 7. Flexural test set up.

cutoff instead ber pullout. In the case of the hooked-end steel ber, after the maximum load, the pullout stress did not decrease
rapidly because the hooked part at the end of ber resisted the
pullout load during the bond test. As such, the ber/matrix interfacial bonding and pullout behavior are associated with the surface
morphology and mechanical properties of bers.
3.2. Flexural properties of ber-reinforced concrete
Fig. 9 shows the exural load-deection curve of unnotched
beam at various volume fraction of ber. The plain concrete specimen exhibited brittle failure at the peak load. The polyamide berreinforced concrete exhibited that exural load was decreased at
the initial crack due to the elongation of ber. But its load increased
again because of the bonding of ber and matrix. In the case of the
ber volume fraction of 1.0%, the load at the second peak load by
bonding of ber and matrix was higher than at the rst peak load.
The maximum exural load of the hooked-end steel berreinforced concrete was higher than that of the polyamide ber-

Fig. 9. Flexural load-deection curve by unnotched beam.

reinforced concrete. The exural test results exhibited a gradually

reduction in the hooked-end steel ber reinforced concrete
loadedeection curves compared with those of polyamide berreinforced concrete because of the pullout behavior of hookedend steel ber and matrix. And the residual strength in the
loadedeection curves of hooked-end steel ber-reinforced concrete was higher than that of polyamide ber-reinforced concrete at
the same ber volume fraction.
Such a exural failure properties was observed to have a similar
pattern in the load versus CMOD curve obtained by performing the
3-point exural load test on the notched beam. Moreover, the
overall exural load was found to be lower due to the stress concentration on the notched part and the reduction of the fractured
cross-sectional area.
Table 5 shows the exural strength and fracture energy determined in the exural test on unnotched and notched beams. The
test results were presented as the mean of the values obtained for
each of the three specimens. The exural strength and facture energy increased as the volume fraction increased. The exural

436

H. Kim et al. / Composites Part B 78 (2015) 431e446

Table 5
Flexural test result (standard deviation).
ID.

Plain
PA-0.50
PA-0.75
PA-1.00
SF-0.50
SF-0.75
SF-1.00

Unnotched beam test

Notched beam test

fr [MPa]

fr0 [MPa]

fr =fr0

5.44
6.18
7.42
8.32
8.45
10.22
11.52

e
3.11
4.50
5.84
6.93
9.27
10.31

e
0.50
0.60
0.70
0.82
0.91
0.89

(0.71)
(0.22)
(0.46)
(0.94)
(0.56)
(0.23)
(1.58)

(0.78)
(1.06)
(1.90)
(0.95)
(0.36)
(1.71)

(0.11)
(0.12)
(0.19)
(0.07)
(0.02)
(0.06)

fe [MPa]

Gf [N$m/m2]

8.24
10.41
11.55
11.25
11.33
12.67
15.76

857.45
2704.86
3796.83
6907.52
5597.48
8223.37
14641.56

(0.52)
(0.80)
(0.29)
(0.47)
(1.01)
(0.97)
(3.82)

(202.31)
(119.68)
(93.83)
(329.70)
(792.22)
(990.73)
(6261.13)

fr : Flexural strength of unnotched beam (MPa), fr0 : Equivalent exural strength (MPa), fr =fr0 : Equivalent exural strength ratio, fe : Flexural strength of notched beam (MPa), Gf :
Fracture energy (N$m/m2).

strength and equivalent exural strength of polyamide ber reinforced concrete was lower than hooked-end steel ber reinforced
concrete. Likewise, the fracture energy, which was calculated by the
area below the load-CMOD curve, was observed to be higher in the
hooked-end steel ber-reinforced concrete. This is ascribed to the
higher post-crack stress and residual stress of the hooked-end steel
ber-reinforced concrete, given that fracture energy greatly depends on the stress-sustaining capacity along with the maximum
exural stress.
Fig. 10 shows the degree of ber dispersion in a specimen with a
1.0% ber volume fraction as determined by visual assessment of
the exural fracture cross-section. The polyamide ber-reinforced
concrete specimens exhibited a regular distribution of bers on
the cross-section, while the hooked-end steel bers tended to
exhibit uneven distribution.
Fig. 11 shows the fracture shape of ber on the exural fracture
cross-section. The polyamide bers were found to be bonded to the
inter-bundle gaps and broken instead of being pulled out from the
matrix. Polyamide bers consist of bundle-type short bers, the
bundled bers are assumed to play the role of holding the concrete
matrix, thus demonstrating a high bonding performance in the
concrete as shown in the image. In the case of the hooked-end steel
ber, which are metal bers, showed a fracture mode of ber being
pulled out from the matrix at the moment of reaching the yield
point and the hooked parts was unfolded.
The ber fracture pattern associated with exural fracture
properties of ber-reinforced concrete is found to be very similar to
the fracture behavior by the ber pullout test. The bonding and
pullout properties of bers have a great inuence in improving
exural and tensile properties of concrete. In the exural fracture
behavior, the maximum exural strength was highly affected by the
ber/matrix bonding strength and the number of bers, the strainsoftening and hardening are substantially affected by the pullout of
ber behavior when bers are pulled out from the matrix.
In this study, even if the number of polyamide bers was 2.46
times higher than that of hooked steel bers, the bonding strength
was 3.35 times smaller than hooked steel bers. Therefore it was
considered that the maximum exural strength of hooked steel
ber reinforced concrete is greater than the polyamide ber reinforced concrete.

4. Impact resistance against a high-velocity projectile

4.1. Preparation of the test specimen
Table 6 shows the experimental design for assessing the impact
resistance performance of the polyamide ber-reinforced concrete
and hooked-end steel ber-reinforced concrete to impact of hard
projectile. The cross-sectional dimension of the concrete specimen
is 200 mm
200 mm, and its thickness varies in the range from

Fig. 10. The degree of ber dispersion in the exural fracture cross-section (Vf 1.0%).

H. Kim et al. / Composites Part B 78 (2015) 431e446

437

Fig. 12. Projectile.

projectile was propelled at 170 m/s and 300 m/s by controlling the
pressure of nitrogen gas. The carrier was detached right before
impact. As shown in Fig. 14, the specimens were fastened to the four
sides by the steel frame. The failure mode of each specimen was
analyzed by visual assessment of the outer morphology of the
specimen impacted by a high-velocity projectile. For the assessment of local damages on the concrete after impact, penetration
depth and crater diameter on the front side, scabbing depth and
crater diameter on the rear side and scabbing limit thickness were
evaluated, as shown in Figs. 15 and 16. The penetration depth and
scabbing limit thickness was estimated by using the modied NDRC
(National Defense Research Committee) formula (5) and (6).

Fig. 11. Fracture shape of ber on the exural fracture cross-section.

 
NM V0 1:8
Gx=D 3:813
105 p
D fc D
 x 2 x
Gx=D
: 2
2D
D
x
x
Gx=D  1 :  2
D
D

(5)

x
 x 2  x

s
7:91
 5:06
< 0:65
D
D
D
D
x 

s
x
2:12 1:36
0:65  < 11:75
D
D
D

(6)

30 mm to 70 mm. The hard spherical shape projectiles used in this

study are shown in Fig. 12. The diameter and mass of projectiles are
10 mm (4.06 g) and 25 mm (66.4 g), respectively. The impact velocity of projectiles varies are 170 m/s and 300 m/s for the 10 mm
projectile and 170 m/s for the 20 mm projectile. The impact velocity
of the projectile are measured by a velocity-sensor installed at the
tip of acceleration tube.

where x is the penetration depth (m), s is the scabbing limit

thickness (m), D is the projectile diameter (m), M is the projectile
mass (kg), V0 is the impact velocity (m/s), fc is the compressive
strength (Pa), and N is the projectile shape factor, G is impact
function.

4.2. Setup and the test procedure

Fig. 17 shows the failure modes of specimens after impact with a

10-mm-diameter projectile at the impact velocity of 300 m/s. The
projectile did not perforate the 30 mm-thickness plain specimen,
but perforation fracture occurred, with the penetration and

The projectile impact test was performed using a gas-pressure

high-velocity impact test device, as shown in Fig. 13. The

4.3. Failure modes

Table 6
Design of impact test.
Impact condition

Specimen condition

Projectile diameter (mm)

Velocity (m/s)

Size (mm)

Thickness (mm)

10
10
25

300
170
170

200
200 (W
H)

30, 40
50, 60, 70

438

H. Kim et al. / Composites Part B 78 (2015) 431e446

Fig. 13. Gas-pressured high-velocity impact test device.

Fig. 14. Specimen setup.

Fig. 15. Test method of failure mode and depth.

Fig. 16. Test method of crater diameter.

H. Kim et al. / Composites Part B 78 (2015) 431e446

scabbing depth exceeding the specimen thickness. Scabbing on the

rear side has been observed in the PA-0.5, SF-0.5 and SF-0.75
specimen. PA-0.75, PA-1.0 and SF-1.0 were observed several radial
cracks from the center of the rear side. Thus, it was veried that
ber reinforcement greatly inhibited scabbing, and its effect

439

improved with the increase in the ber volume fraction. The 40mm-thickness all specimens were observed rear crack on the rear
side.
Fig. 18 shows the failure modes of the specimens after impact
with a 25-mm-diameter projectile (66.4 g) at the impact velocity of

Fig. 17. Failure mode after impact test (Projectile diameter 10 mm, Impact velocity 300 m/s).

440

H. Kim et al. / Composites Part B 78 (2015) 431e446

Fig. 18. Failure mode after impact test (Projectile diameter 25 mm, Impact velocity 170 m/s).

H. Kim et al. / Composites Part B 78 (2015) 431e446

170 m/s. The impact energy (mv2 =2) generated at the moment of
striking against the specimen amounted to 959.48 J, 5 to 16 times
greater than the impact energy generated by a steel ball with
10 mm diameter at the impact velocities of 170 m/s and 300 m/s
(58.81 J and 183.15 J, respectively). Accordingly, all 50-mm-thickness specimens were perforated regardless of the ber reinforcement. In the 60-mm-thickness specimens, scabbing on the rear side
was occurred in plain, SF-0.5 and PA-0.5, whereas SF-0.75, SF-1.0,
PA-0.75 and PA-1.0 were observed big radial cracks from the center
of the rear side without the scabbing. In the case of the 70-mmthickness specimens, plain specimen was occurred scabbing on the
rear side. However, all the ber-reinforced concrete specimen were
occurred radial cracks from the center of the rear side and the rear
cracks were decreased with the increase in ber volume fraction.
The improvement in exural tensile performance by ber reinforcement contributed to reducing the external damages of concrete. The crater diameter on the rear side of the polyamide ber
reinforced concrete is less than that of hooked steel ber reinforced
concrete and the scabbing did not occur in lower ber volume
fraction.
The results of the specimens tested using a 10-mm-diameter
projectile and the impact velocity of 170 m/s are not presented in
this paper. Because they were formed penetration and crater on the
front side without scabbing damage on the rear side.
4.4. Compressive strength and penetration depth
Fig. 19 shows the relationship between the penetration depth
and compressive strength. In the gure, the penetration depth
estimated by the modied NDRC formula is also shown. Estimated
penetration depths by the modied NDRC formula are consistent
with the test results. In the case of same spherical shape projectile,
the penetration depth also increased as the mass, diameter of the
projectile and impact velocity increased. The compressive strength
of the ber-reinforced concrete in this study ranged from 50 to
57 MPa, regardless of the type and volume fraction of the bers. No
noticeable improvement in the compressive strength through ber
reinforcement was observed. The penetration depth yielded similar
depth in the same impact conditions and the effect of exural
tensile performance improvement by ber reinforcement was not
determined.

441

Fig. 19. Relationship between the penetration depth and the compressive strength.

effect. While fragmentation on the front side can occur by the

shock waves generated immediately upon striking, tensile stress
arising from the shock waves reected from the rear side is also
considered to contribute. This allows us to assume that the exural
tensile performance improved by ber reinforcement inhibits
concrete spalling on the front side.

4.6. Scabbing depth and diameter

Fig. 21 shows the scabbing depth and crater diameter on the rear
side. The impact of high-velocity projectile causes severe scabbing
on the rear side due to the exural tensile stress and strain
generated when the shock waves reached the rear side. As for the
plain specimens, although the projectile did not perforate the
specimen, the sum of the penetration depth and rear scabbing
depth was larger than the specimen thickness, with the scabbing
depth (2/3 of the specimen thickness) greater than the penetration
depth. In the case of the ber-reinforced concrete, scabbing depth

4.5. Penetration depth and crater diameter

Fig. 20 shows the relationship between penetration depth and
crater diameter on the front side. The crater diameter on the front
side increased as the diameter of a projectile and the impact velocity increased. It also increased in proportion to the increase in
the penetration depth. From these results, the following regression
equation could be derived.

Plain: Cd 7:18Pd (R2 0.77)

PA0.50: Cd 3:93Pd (R2 0.96)
PA0.75: Cd 3:36Pd (R2 0.92)
PA1.00: Cd 3:11Pd (R2 0.89)

SF0.50: Cd 4:58Pd (R2 0.81)

SF0.75: Cd 3:77Pd (R2 0.88)
SF1.00: Cd 3:40Pd (R2 0.93)

where, Pd : penetration depth, Cd : crater diameter on the front side.

Crater diameters on the front side were reduced by ber reinforcement, and the ber volume fraction was inversely correlated
to the crater diameter. It was thus conrmed that the exural
tensile performance improved by ber reinforcement does not
have a penetration-inhibiting effect but has a spalling-inhibiting

Fig. 20. Relationship between the penetration depth and the crater diameter on the
front side.

442

H. Kim et al. / Composites Part B 78 (2015) 431e446

Fig. 22. Scabbing limit thickness of ber reinforced concrete.

Fig. 21. Scabbing depth and crater diameter on the rear side.

and crater diameter on the rear side were decreased because due to
the improvement of exural tensile performance induced by ber
reinforcement. The improved exural tensile performance exerts a
greater inuence on scabbing inhibition than on the inhibition of
penetration and spalling on the front side. In the case of the polyamide ber-reinforced concrete, the scabbing depth and crater
diameter on the rear side under the same impact conditions were
smaller compared to those in the case of the hooked-end steel
ber-reinforced concrete. This suggests that the degree of
improvement of exural performance is less than that the hookedend steel ber-reinforced concrete, but in terms of impact-related
scabbing inhibition, its impact resistance is as good as, or superior, to that of the hooked-end steel ber-reinforced concrete.
4.7. Scabbing limit thickness
Fig. 22 shows the scabbing limit thickness for different impact
conditions. In this study, the scabbing limit thickness were dened
as the minimum specimen thickness for which concrete scabbing
does not occur despite cracks on the rear side. Under the impact of
the 10-mm-diameter projectile at the velocity of 300 m/s, the
scabbing limit thickness of the plain concrete, PA-0.5, SF-0.5 and
SF-0.75 were exhibited 40 mm. Whereas, the scabbing limit
thickness of PA-0.75, PA-1.0 and SF-1.0 were determined to be
30 mm. Under the impact of the 25-mm-diameter projectile at the
velocity of 170 m/s, even though the impact velocity was lower,

diameter and mass of the projectile increased, the scabbing limit

thickness were also increased. The scabbing limit thickness of the
plain was exhibited 80 mm. In the case of ber-reinforced concrete,
PA-0.5 and SF-0.5 were 70 mm, and in, PA-0.75, PA-1.0, SF-0.75 and
SF-1.0 were determined to be 60 mm. Additionally, it was veried
that scabbing limit thickness decreased in comparison with the
values estimated by the modied NDRC formula. The berreinforced concrete demonstrated that the scabbing limit thickness was decreased by improvement of exural tensile performance. However, given that the modied NDRC formula estimates
the scabbing limit thickness based on the penetration depth, which
is signicantly dependent on the compressive strength of concrete.
Therefore, prediction of the scabbing limit thickness of berreinforced concrete are very limited.

4.8. Relationship between exural properties and impact resistance

induced by ber reinforcement
Fig. 23 shows the relationship between the exural properties
(exural strength and fracture energy) and the scabbing depth on
the rear side. In the case of a low ber volume fraction, the exural
strength is not signicantly higher than that of plain concrete. In
general, since the exural strength increases with the compressive
strength, it is problematic to derive a direct relationship between
the exural strength and scabbing inhibition by ber reinforcement. Moreover, ber-reinforced concrete signicantly contributes
to the improvement in the deformation capacity as well as the increase in the exural tensile strength, so it is assumed that the
fracture energy will be more efcacious for the efcient evaluation
of the enhanced exural tensile performance. Under the same ber
volume fraction, polyamide ber-reinforced concrete was exhibited
lower exural properties compared to hooked-end steel berreinforced concrete, but its scabbing-inhibiting effect was the
same as or superior to that of hooked-end steel ber-reinforced
concrete. This is assumed to be attributable to the differences between the range of strain rate in static test and impact test. Additionally, different effects on ber/matrix bonding strength, crack
control, and absorption of stress and shock waves in the improvement of fracture performance may be considered to have inuence
in relation to a high-velocity impact due to the differences in the
shape and properties of reinforcing bers and surface materials.

H. Kim et al. / Composites Part B 78 (2015) 431e446

443

Fig. 24. Relationship between the number of bers and scabbing depth on the rear
side.

Fig. 23. Relationship between the exural properties and scabbing depth on the rear
side.

Fig. 24 shows the relationship between the number of bers and

scabbing depth on the rear side. Polyamide bers are an organic
ber whose density is lower than that of hooked-end steel bers,
which are metal ber. Consequently, when externally mixed with
respect to volume, the mass of the mixed bers is small, but due to
the light weight, the number of bers is 2.46-fold that of the
hooked-end steel at the same volume fraction, as shown in Table 7.
It appears that a large number of bers absorbs the shockwaves
generated by a high-velocity impact and inhibits scabbing by controlling the crack propagation to the rear side within the concrete.
Fig. 25 shows cross-sections of specimens fractured by highvelocity impact, in which we see that the plain concrete specimens experience considerable diagonal cracking due to the shock
waves generated, and that the propagation of these to the rear
causes scabbing. In the ber-reinforced concrete, on the other hand,
these internal diagonal cracks were less prevalent, and instead
multiple cracks were observed. The lower number and irregular
distribution of the hooked-end steel bers resulted in wider cracks

than those observed in the polyamide ber-reinforced concrete,

resulting in failure due to the loss of fragments through scabbing
and the pulling out of bers from the matrix. However, although
crack-induced fracture occurred in regions with a low number of
bers, scabbing was still found to be inhibited by whatever partial
bonding of bers was present. The polyamide ber-reinforced
concrete exhibited a far more even distribution of cracks, which
were also narrower than those in the steel ber-reinforced concrete. This can be explained by the large number of reinforcing bers mitigating the shockwaves generated by the high-velocity
impact. In regions where scabbing did occur, the polyamide bers
were found to be broken rather than being pulled out of the matrix,
as is shown in Fig. 26.
The fracture mechanism and strain capacity of the concretereinforcing materials tested were generally strain-rate dependent,
meaning that the strength, strain and ductility could be affected by
both the strain rate and loading rate. Consequently, the strength
and strain of the reinforced concrete tended to be greater under
dynamic loading conditions than static. The fracture properties
were also affected by various factors such as the number of bers,
as well as the bonding strength of each ber type at high strain
rates. The difference in the static exural strength and fracture
properties during impact testing can be attributed to the fact that a
strain of 106e105 was used with the former, whereas a strain rate
of 101e102 was used with the latter [26e31].
Although both polyamide ber and hooked-end steel ber
reinforcement signicantly improved the mechanical properties of
the concrete with increasing strain, the greater number of polyamide bers means that they are more sensitive to strain, and tend
to stretch and break rather than pull out of the matrix. Consequently, although the static mechanical properties of a polyamide
ber-reinforced concrete are inferior to those of a hooked-end steel
ber-reinforced concrete, the impact resistance was greater.

Table 7
Comparison of ber number and volume fraction.
Volume fraction (Vf.%)

Polyamide ber (per m3)

Hooked steel ber (per m3)

0.50
0.75
1.00

1,899,998
2,849,997
3,799,996

780,000
1,170,000
1,560,000

444

H. Kim et al. / Composites Part B 78 (2015) 431e446

Fig. 25. Cross-section of the specimen fractured by high-velocity impact.

Further research is therefore needed to better understand this

impact resistance performance of polyamide ber reinforcement
through an evaluation of the dynamic properties of the bers at
high strain rates.
5. Conclusion
From this comparison of the exural properties and impact
resistance of bundle-type polyamide ber-reinforced concrete

relative to those of conventional hooked-end steel ber-reinforced

concrete, the following conclusions can be drawn:
1) Hooked-end steel bers exhibit pullout behavior from the matrix
after peak pullout load, whereas bundle-type polyamide bers are
stretched and cutoff due to the displacement triggered by a loosening of the bundled bers. This allows the polyamide bers to
disperse stress and reduce stiffness, thereby increasing the interber gap and specic surface area of the ber/matrix bonding.

H. Kim et al. / Composites Part B 78 (2015) 431e446

445

Fig. 26. Fracture shape of ber on the rear side after impact test.

2) The exural strength and fracture energy of bundle-type polyamide reinforced concrete are less than those of hooked-end
steel ber-reinforced concrete; however a post-peak deection-hardening behavior is observed with polyamide ber that
likely occurs as a result of the strong bonding between the micro
gaps of the bers and the concrete matrix.
3) Although bundle-type polyamide ber-reinforced concrete has
a lower exural strength and toughness, its resistance to scabbing during high-velocity impact is equal or superior to that of
hooked-end steel ber-reinforced concrete. This can be attributed to the greater number of bers, which facilitate a more
effective dispersion of impact stress.
4) Polyamide ber-reinforced concrete exhibits an even distribution of cracks following high-velocity impact, as the bers help
to absorb and mitigate the shockwaves generated. In contrast,
the lower of bers in hooked-end steel ber-reinforced concrete
causes an irregular distribution of wide cracks that in extreme
cases leads to scabbing.
5) Polyamide bers in the side opposite the point of impact tend to
fracture rather than pull out, whereas hooked-end steel bers in
the same location experience pullout-induced spalling fracture
on scabbing.
6) Although the resistance of ber-reinforced concrete to impact
load has generally been considered to increase with an
improvement in exural and tensile performance through
increasing ber/matrix bonding, the results of the present study
lead us to believe that the number of reinforcing bers also

exerts a signicant inuence on impact resistance. This is

therefore something that needs to be explored in more detail
through future studies.
Acknowledgment
This research was supported by a grant from a Construction
Technology Research Project (Development of impact/blast resistant HPFRCC and evaluation technique thereof, 13SCIPS02) funded
by the Ministry of Land, Infrastructure, and Transport.
References
[1] Kanda T, Li Victor C. Interface property and apparent strength of high strength
hydrophilic ber in cement matrix. J Mater Civ Eng 1998;10:5e13.
[2] Won JP, Hong BT, Choi TJ, Lee SJ, Kang JW. Flexural behavior of amorphous
micro-steel bre-reinforced cement composite. Compos Struct 2012;94:
1443e9.
[3] Won JP, Hong BT, Lee SJ, Choi SJ. Bonding properties of amorphous micro-steel
bre-reinforced cementitious composite. Compos Struct 2013;102:101e9.
[4] Kim YI, Park DS, Seo CH. Variations of material characteristics of high-strength
concrete according to increase of steel ber volume. J Archit Inst Korea
2005;21:95e101.
[5] Bencardino F, Rizzuti L, Spadea G, Swamy RN. Implications of test methodology on post-cracking and fracture behaviour of Steel Fibre Reinforced
Concrete. Compos Part B: Eng 2013;46:31e8.
[6] Jeon JK, You JO, Moon JH. Durability evaluation of tunnel lining concrete
reinforced with nylon ber. J Korea Concr Inst 2008;20:487e93.
The effects of macro synthetic
[7] Bolat H, Simsek O, ullu M, Durmus G, Can O.
ber reinforcement use on physical and mechanical properties of concrete.
Compos Part B: Eng 2014;61:191e8.

446

H. Kim et al. / Composites Part B 78 (2015) 431e446

[8] Anderson WF, Watson AJ, Rmstrong PJ. Fiber reinforced concretes for the
protection of structures against high velocity impact. In: Proceedings of the
international conference on structural impact and crashworthiness; 1984.
p. 687e95.
[9] Ramakrishnan V, Coyle WV, Fowler LJ, Schrader EK. A comparative evaluation
of ber shotcretes. Report SDSM&T-CBS 7902. USA: Civil Engineering
Department, South Dakota School of Mines and Technology. 1979.
[10] Dancygier AN, Yankelevsky DZ. High strength concrete response to hard
projectile impact. Int J Impact Eng 1996;18:583e99.
[11] Dancygier AN, Yankelevsky DZ, Jaegermann C. Response of high performance
concrete plates to impact of non-deforming projectiles. Int J Impact Eng
2007;34:1768e79.
[12] Mohamed AK, Ahmed F. Effect of reinforced on the response of concrete
panels to impact of hard projectile. Int J Impact Eng 2014;63:1e17.
[13] O'Neil EF, Neeley BD, Cargile JD. Tensile properties of very-high-strength
concrete for penetration-resistant structures. Shock Vib 1999;6:237e45.
[14] Zhang MH, Shim VPW, Lu G, Chew CW. Resistance of high-strength concrete
to projectile impact. Int J Impact Eng 2005;31:825e41.
[15] Vossoughi F, Ostertag CP, Monteiro PJ, Johnson GC. Resistance of concrete
protected by fabric to projectile impact. Cem Concr Res 2007;37:96e106.
[16] Hwang HK, Nam JS, Kim HS, Lee IC, Hiroyuki Miyauchi, Kim GY. Evaluation on
the blast resistance performance of ber reinforced mortar with ber type
and PVASTF panel installed intermediate space thickness. J Archit Inst Korea
2012;28:115e22.
[17] Nam JS, Kim HS, Lee IC, Hiroyuki Miyauchi, Kim GY. Evaluation on the impact
resistance resistance performance of ber reinforced concrete by highvelocity projectile and contacted explosion. J Korea Concr Inst 2013;25:
104e7.
[18] Almusallam TH, Siddiqui NA, Iqbal RA, Abbas H. Response of hybrid-ber
reinforced concrete slabs to hard projectile impact. Int J Impact Eng
2013;58:17e30.
[19] Beppu M, Miwa K, Itoh M, Katayama M, Ohno T. Damage evaluation of concrete plates by high-velocity impact 2008;35:1419e26.

[20] Beppu Masuhiro, Miwa Koji, Ohno Tomonori, Shiomi Masanori. An experimental study on the local damage of concrete plate due to high velocity
impact of steel projectile. Jpn Soc Civ Eng E 2007;63:178e91.
[21] Miwa Koji, Beppu Masuhiro, Ohno Tomonori, Itoh Masaharu,
Katayama Masahide. An estimation method of local damage in concrete plates
by the modied the theoretical model. Jpn Soc Civ Eng A 2009;65:844e58.
[22] Morita Takeshi, Beppu Masuhiro, Suzuki Makoto. Impact resistance performance of polypropylene short ber reinforced concrete subjected to high
velocity impact. J Struct Constr Eng AIJ 2013 Feb;78:319e27.
[23] Morita Takeshi, Beppu Masuhiro, Suzuki Makoto. An experimental study on
the effect of reinforcement and short ber reinforcement in the local fracture
properties of the concrete plate subjected to high-velocity impact of rigid
projectile. JCI Proc 2013;35:667e72.
[24] Kang SH, Kim JJ, Kim DJ, Chung YS. Effect of sand grain size and sand-tocement ratio on the interfacial bond strength of steel ber embedded in
mortar. Constr Build Mater 2013;47:1421e30.
[25] ASTM C1609/C1609M-12. Standard test method for exural performance of
ber-reinforced concrete (Using beam with third-point loading).
[26] RILEM 50-FMC Draft Recommendation. Determination of the fracture energy
of mortar and concrete by means of three-point bend tests on notched beams.
[27] Bischoff PH, Perry SH. Compressive behaviour of concrete at high strain rates.
Mater Struct 1991;24(6):425e50.
[28] Habel K, Gauvreau P. Response of ultra-high performance ber reinforced
concrete (UHPFRC) to impact and static loading. Cem Concr Compos
2008;30(10):938e46.
[29] Wang S, Zhang MH, Quek ST. Mechanical behavior of ber-reinforced highstrength concrete subjected to high strain-rate compressive loading. Constr
Build Mater 2012;31:1e11.
[30] Tran TK, Kim DJ, Choi E. Behavior of double-edge-notched specimens made of
high performance ber reinforced cementitious composites subject to direct
tensile loading with high strain rates. Cem Concr Res 2014;63:54e66.
[31] Su H, Xu J, Ren W. Mechanical properties of ceramic ber-reinforced concrete
under quasi-static and dynamic compression. Mater Des 2014;57:426e34.