Case Studies in Construction Materials 11 (2019) e00299

Contents lists available at ScienceDirect

Case Studies in Construction Materials

journal homepage: www.elsevier.com/locate/cscm

Underwater abrasion of steel fiber-reinforced self-compacting

concrete
Sallal R. Abid* , Ali N. Hilo, Nadheer S. Ayoob, Yasir H. Daek
Civil Engineering Department, Wasit University, Iraq

A R T I C L E I N F O A B S T R A C T

Article history: Hundreds of tons of concrete may be ruined due to the abrasion erosion phenomenon that
Received 6 September 2019 takes place in several hydraulic structures. Such unfavorable action can impose high
Received in revised form 19 October 2019 maintenance costs and reduce the life span of these structures. The high-velocity water
Accepted 25 October 2019
flow which carries large amounts of sediments is basically the main cause of this
phenomenon. To better understand concrete abrasion in hydraulic structures, an
Keywords: experimental investigation was conducted to study the underwater frictional abrasion
Self-compacting concrete
of self-compacting concrete (SCC) using the ASTM C1138 test method. Six SCC mixtures
SCC
Abrasion
were prepared with different design grades of 30, 40 and 50 MPa and different micro-steel
Underwater abrasion fiber contents of 0, 0.5, 0.75 and 1.0%. Four fresh SCC tests were adopted, while abrasion and
Micro steel fiber control tests were conducted at ages of 7, 28 and 90 days. The ASTM C1138 tests showed
that increasing the mixture strength by 20 MPa improved the abrasion resistance more
effectively than the inclusion of 1.0% of steel fiber. Compared to the 30 MPa plain SCC, the
90-day abrasion resistance improved by 78% for SCC with 50 MPa strength, while the
inclusion of 1.0% of micro-steel fiber enhanced the resistance by only 26%.
© 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC
BY license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction

Erosion of concrete in hydraulic structures can be defined as the gradual deterioration that mainly occurs by three major
causes; abrasion, cavitation, and chemical attack. The main cause of the abrasion erosion is the action of water-borne solids
such as sand, silt, gravel, ice and debris against the surface of concrete during the hydraulic structure operation [1]. Several
strategic hydraulic structures are commonly susceptible to severe abrasion erosion damage, such as stilling basins, spillway
aprons and culverts [1]. The abrasion damage occurs in hydraulic concrete due to saltation, rolling and sliding of solid
particles that conveyed by the water flowing on the concrete surface. This action depends on water velocity and the
transmitted bed load [2–5]. According to this mechanism, waterborne solids create numerous tiny cracks on the concrete
surface. The continuous abrasive action develops these cracks. As a result, it weakens the surface of the structure. The interior
bond between concrete particles is progressively destroyed and the flowing water plucks away the aggregates and cement
binders so that small voids are created. The formation of these voids and the persistent abrasive action permit to enlarge
these voids and increase the damage, which ranges from a depth of few centimeters to one meter or more in the most
dangerous cases [1,2,6].
There are several factors remarkably affect the rate of abrasion erosion phenomenon in hydraulic structures. They can
basically be divided into two groups. Firstly, factors related to the used concrete, such as the compressive and tensile

* Corresponding author.
E-mail address: [email protected] (S.R. Abid).

https://doi.org/10.1016/j.cscm.2019.e00299
2214-5095/© 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/
4.0/).
2 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

strengths, aggregate hardness, aggregate shape, type of cement, w/c ratio, curing of concrete surface and the addition of
fibers and cementitious materials. All these factors play an essential role in determining the abrasion rate [7]. Secondly,
abrasion rate also depends on the characteristics of the transported solids such as their shape, hardness, size and quantity.
Moreover, abrasion damage rate depends significantly on the flowing water velocity. Water with higher velocity can easily
transport sediments with larger size and quantity, therefore, undoubtedly heavier abrasion would be produced. More details
about this process can be found in [4,5,8]. The impact angle of solids and their movement characteristics like their rotation
and saltation affect the wearing rate as well [1,2,7]. It’s obvious that abrasion erosion causes fatal defects for critical hydraulic
structures such as dams. This requires huge, expensive and difficult maintenance works. In Dworshak Dam, it was reported
that more than 1500 m3 of concrete was lost due abrasion action [9]. For example, in order to avoid or minimize these
difficulties, the quality of the used concrete should be improved and tested by one of the several abrasion tests that were
invented to evaluate the abrasion resistance of concrete.
In Japan during the 1980s, researchers produced Self-Compacting Concrete (SCC) which is also known as self-
consolidating concrete in order to reduce compaction problems such as the noise, cost and delay, to enhance the durability of
the structure and to provide safer and more economical environment of working [10,11]. This type of concrete is very
flowable but it is a stable concrete which can spread and fill the required space due to its self-weight without any segregation
with no need for external vibrating [10]. The properties of SCC fresh mixtures must achieve the requirements specified by ACI
237R [11] or the European Specifications EFNARC [10]. According to the several mechanical properties and the considerable
advantages of SCC, it was already used in some hydraulic structures such as the pressure tunnel, the intake and the bottom
outlet of the Upper Gotvand dam in Iran [12]. Consequently, the evaluation of SCC resistance to the abrasion erosion is very
necessary because this property is deeply related with the life span of any hydraulic structure which is subjected to high
speed water flow and high sediment transport.
Recently, several researches were conducted to investigate underwater abrasion erosion of hydraulic structures’ surfaces
using the ASTM C1138 [13–18] or the water jet method [3,6,19–23]. In these published studies, several concretes and
cementitious materials were investigated, while none of which focused on fibrous SCC. However, abrasion of SCC was
investigated by few previous researches using test methods other than underwater methods. In order to study the resistance
of SCC to abrasion erosion damage, Ghafoori et al. [24] compared the abrasion resistance of SCC with the abrasion resistance
of traditional concrete. The test method was based on ASTM C779 [25], procedure C. The test results showed that SCC
presents higher resistance against wear when compared with the ordinary concrete due to its effect on the compressive
strength. For the same cement paste quality and quantity, the compressive strength of SCC was 15–30 higher than the
ordinary concrete, while the final abrasion loss was 50–70% lower than that of ordinary concrete. Moreover, for SCC with
higher w/b ratio, abrasion resistance increased in spite of the noticeable decrement in compressive strength compared with
traditional concrete. Turk and Karatas [26] conducted an experimental study to investigate the abrasion resistance of SCC in
which cement was partially replaced by silica fume and fly ash. The tests were conducted using the ASTM C779 test method.
The test results revealed that SCC specimens exhibited higher resistance to abrasion wear compared to normal concrete. The
authors also addressed that abrasion resistance of SCC increased when cement was replaced by larger quantity of silica fume
but decreased when fly ash was used.
The introduced literature discloses that, till now, there is a big lack in the studies that examine the abrasion resistance of
SCC. The available experimental investigations were conducted using ASTM C779 abrasion method, which includes no water
movement. Although SCC was already used in hydraulic structures as introduced, none of the available experimental works
has investigated the abrasion of SCC surfaces under the effect of water and waterborne materials. Therefore, an experimental
investigation was directed in this work to study the abrasion erosion of plain and fibrous SCC using the ASTM C1138 [27]
underwater abrasion test method, which provides the best simulation for the abrasion erosion of concrete surfaces in
hydraulic structures [15].

2. Experimental work

The experimental work of this study consists of the preparing and the testing of SCC mixtures of different design grades
and various micro-steel fiber contents. Cylindrical abrasion specimens and control test specimens were cast and tested.

2.1. Materials and mixtures

Six SCC mixtures were prepared, for which fresh concrete tests were performed to ensure their fulfillment to the
limitations of the fresh SCC properties stated in ACI 237R [11] and the European specifications (EFNARC) [10]. The first three
mixtures were plain SCC without steel fiber reinforcement. They were prepared with different design compressive strengths
(grades) of 30, 40 and 50 MPa, while the remaining three were SCC of grade 30 MPa but were reinforced with different
volumetric contents of steel fiber. The abrasion specimens were tested at three ages of 7, 28 and 90 days. Different dosages of
type R42.5 ordinary Portland cement were used in addition to a fixed quantity of limestone powder with the properties
shown in Table 1. Local siliceous gravel and sand with the sieve analysis listed in Table 2 were adopted for all mixtures. Silica
fume was used only for the 50 MPa design strength mixture. Furthermore, Sika Viscocrete-5930 was added with variable
dosages to reduce the w/c ratio and to achieve SCC mixtures with acceptable fresh properties. Three volumetric contents of
straight shape micro-steel fiber (0.5, 0.75 and 1.0%) with the properties listed in Table 3 were used in this study. A previous
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 3

Table 1
Chemical composition and physical properties of cement and limestone powder.

Property Cement Limestone Powder

SiO2 (%) 21.04 1.38
Fe2O3 (%) 5.46 0.12
Al2O3 (%) 2.98 0.72
CaO (%) 63.56 56.1
MgO (%) 2.52 0.13
SO3 (%) 2.01 0.21
f-CaO (%) 0.76 –
Loss on ignition (%) 1.38 4.56
Specific surface (m2/kg) 362 –
Specific gravity 3.15 –
Fineness (% retain in 45 mm) – –

Table 2
Fine and coarse aggregates sieve analysis results.

Sieve size (mm) Fine (% Passing) Coarse (% Passing)

19 100 100
12.5 100 68
9.5 100 30
4.75 93.2 0
2.36 65 –
1.18 50.6 –
0.6 42 –
0.3 12 –
0.15 0 –

Table 3
Steel fibers properties.

Description Straight
Length 15 mm
Diameter 0.2 mm
Density 7800 kg/m3
Tensile strength 2600 MPa
Aspect ratio 75

study [28] showed that the inclusion of steel fiber with volumetric content up to 1.0% decreased the slump by 36% but
increased the strength noticeably, while adding 1.25% led to further decrease in slump and did not lead to significant
additional increase in strength. This result was also supported by another study [29] which showed that using 1.0% of steel
fiber decreased the slump by 34%, while increased the compressive strength and splitting tensile strength by 16% and 52%,
respectively. On the other hand, increasing the fiber content by further 0.5% decreased the slump without showing
significant improvement in strength. Therefore, a maximum volumetric steel fiber content of 1.0% was adopted in this study.

Table 4
Ingredients of the prepared SCC mixtures.

Material Quantities (kg/m3)

M30-0 M30-0.5 M30-0.75 M30-1.0 M40-0 M50-0

Cement 400 425 425 435 500 550
Sand 1060 1096 1096 1110 990 950
Gravel 586 519 519 494 586 543
Water 185 196.56 196.56 217 200 200
w/ca 0.463 0.463 0.463 0.499 0.4 0.364
w/bb 0.364 0.397 0.397 0.43 0.351 0.29
SPc 9.5 13.55 13.55 15 10.25 17.857
LS powder 70 70 70 70 70 70
Silica fume 0 0 0 0 0 70
Steel fiber 0 38 58.5 78 0 0
a
Water to cement ratio.
b
Water to total cementitious materials ratio.
c
Superplasticizer.
4 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

2.2. Mixing and curing

To prepare the desired SCC mixtures depending on locally available materials, a ready SCC mixture prepared by
Sahmaran et al. [30] was relied. Fresh concrete tests were carried out to check the availability of this mixture.
Modifications were then tried until the acceptable SCC fresh properties were achieved. Table 4 shows the ingredients
weights per one cubic meter. According to an accurate time schedule, SCC specimens were cast in certain dates. All test

Fig. 1. The used abrasion test machine according to ASTM C1138.

Table 5
Results of fresh tests of the six SCC mixtures.

Mixture Slump Flow (mm) T50 (s) DJ-ring (mm) Penetration (mm)
M30-0 755 1.98 5 5
M30-0.5 655 2.7 13 5
M30-0.75 630 3.3 14 4
M30-1.0 700 2.7 20 5.5
M40-0 740 2.55 8 4.5
M50-0 615 4.97 9 3

Table 6
Results of control tests of the six SCC mixtures.

Mixture Compressive Strength (MPa) Splitting Tensile Strength (MPa)

7 days 28 days 90 days 7 days 28 days 90 days

M30-0 25.616 28.706 34.229 2.503 3.789 4.395
M30-0.5 21.062 34.387 35.126 3.742 4.5 4.682
M30-0.75 22.325 35.254 36.217 3.776 4.668 5.621
M30-1.0 23.731 37.271 37.5 3.883 4.766 5.734
M40-0 30.11 37.265 39.448 3.406 3.837 4.530
M50-0 33.816 48.321 52.261 4.576 4.937 5.119
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 5

specimens were demolded after 24 h and placed in a suitable curing tank with water temperature ranging between 22 and
25
C until the testing time.

2.3. Fresh concrete tests and control tests

To check the flowability of the prepared SCC mixtures, the slump flow and T50 tests were conducted. The slump flow test
was conducted using a conical mold similar to that used for concrete slump test based on ASTM C143 [31]. The viscosity of the
mixture considerably affects the SCC flow rate. The T50 is a very simple test procedure that evaluates the filling ability and the
viscosity of SCC mixes. It can be carried out during slump flow test by using a stopwatch to determine the required time for
the specimen to achieve an outer diameter of 500 mm from the instant of mold lifting.

Fig. 2. DD-testing time of different grades at age of (a) 7 days, (b) 28 days and (c) 90 days.
6 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

Fig. 3. Abrasion cylinders after 72 h of abrasion test: effect of concrete strength at different ages.

Another governor property of SCC is the passing ability. It gives an obvious indication about the flowing ability of fresh
concrete through a certain obstacle like steel bar reinforcement. In this work, passing ability was investigated using the
J-ring. The J-ring test must be carried out in conjunction with the slump test. Two perpendicular diameters of concrete
should be measured after it flows through a special steel ring with vertical bars. The average value must be compared with
the value of slump flow test and the difference between the two values is the DJ-ring flow, which must not exceed the
acceptable limit. Every SCC mixture should conform the two abovementioned properties in addition to segregation
resistance. The rapid penetration test [32] was used in this study to test the segregation resistance of the adopted mixtures.
The rapid penetration apparatus are fixed above a concrete-filled cone with the free falling cylinder in touch with the
concrete surface. The apparatus rod is left free to let the cylinder penetrates the concrete. After 28–32 s of penetration, the
value of settlement is recorded which should be within the acceptable range.
Compressive strength and splitting tensile strength based on the ASTM C39 [33] and ASTM C496 [34], respectively, were
carried out as concrete quality control tests. Eighteen cylinders of 100 mm diameter and 200 mm depth were cast with each
of the tested abrasion specimens, where three cylinders were used for each test at ages of 7, 28 and 90 days.

2.4. Abrasion test

ASTM C1138 [27] is a standard test that gives a qualitative measurement for the abrasive damage of concrete due to the
effect of waterborne particles such as sand, silt, gravel, etc. The standard test period is 72 h which is divided into six 12-h
intervals. However, the time intervals between each individual mass recording in this research were reduced from 12 to 3 h
to achieve a more accurate indication of abrasion damage behavior. The weight of the tested specimen should be recorded
after each time interval. The test includes the rotation of water that contains seventy chrome steel balls with three different
diameters of (25.4, 19 and 12.7 mm) at a rotation speed of 1200 rpm. These balls are placed above the surface of the 300 mm
diameter and 100 thickness cylindrical specimen to simulate the erodent particles. The test machine shown in Fig. 1 was
fabricated to conform with the requirements of ASTM C1138. As per the recommendations of ASTM C1138, the steel tank of
the machine shown in Fig. 1 was made cylindrical with diameter of 305 mm and height of 450 mm. The impeller was made
from stainless steel with exactly the same detailed dimensions recommended by ASTM C1138. The distance between the test
specimen and the impeller was always kept 38 mm and that between the specimen and the base of the tank was 25 mm.
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 7

3. Results of SCC fresh tests and control tests

Table 5 summarizes the fresh tests results of the six mixtures. The flowability was significantly reduced as the concrete
strength increased, where the slump diameter was reduced by 18.5% for M50-0 mixture compared to the basic M30-0 one.
On the other hand, the T50 time increased from 1.98 to 4.97 s, thus by approximately 150% for the same mixtures. The
decreased water content in the M50-0 mixture is the main reason for such results in spite of the using of higher quantity of
superplasticizer. Comparing the slump flow and T50 results of the first four mixtures in Table 5, it is obvious that the filling
ability decreased due to fiber inclusion. Lower slump flow diameters and higher T50 times were recorded for fibrous mixtures
although of the use of higher water and superplasticizer contents.
The mixtures M40-0 and M50-0 recorded higher DJ-ring values indicating lower passing ability by up to 80% compared to
M30-0 as shown in Table 5. The table also shows that using steel fiber leads to lower passing ability which is reflected from the
significantly higher DJ-ring values of M30-0.5, M30-0.75 and M30-1.0 compared to M30-0. The reduction of the penetration
depth for M40-0 and M50-0 compared to M30-0 reflects the positive effect of concrete strength increment on the segregation
resistance. This result can be attributed to the larger amount of finer materials and the lower water/cementitious materials ratio
of these mixtures compared to M30-0. Table 5 shows that the segregation resistance was comparable for the fibrous mixtures
and the plain one, which might be due to the higher water and superplasticizer contents used in these mixtures.
Table 6 summarizes the results of the control tests at ages of 7, 28 and 90 days. The effect of steel fiber on the compressive
strength was minimal at age of 7 days where the compressive strength of the four M30 mixtures was in the range of
approximately 21 to 25.6 MPa, while at age of 28 days, the fibrous mixtures showed higher compressive strength values than
M30-0 mixture. Such trend was also noticed at 90 days age but with smaller differences between all mixtures. The splitting
tensile strength exhibited the expected continuous increase with both steel fiber content and concrete strength as shown in
Table 6. It is known that fiber increases the tensile capacity of concrete due to fiber bridging activity. However, the M30-0.75
and M30-1.0 showed very close values at all test ages, which indicates the small effect of the increasing of fiber content from
0.75 to 1.0% on splitting tensile strength.

Table 7
Depth decrement (mm) and the improvement percentage at age of 7 days for different SCC grades.

Concrete Grade (MPa) 30 40 50

Test time (h) DD DD IP DD IP

12 0.723 0.966 33.712 0.470 34.909
24 1.561 1.956 25.355 0.911 41.614
36 2.601 2.844 9.337 1.323 49.148
48 3.382 3.418 1.069 1.822 46.105
60 4.335 3.901 10.023 2.263 47.791
72 5.260 4.263 18.956 2.734 48.030

Table 8
Depth decrement (mm) and the improvement percentage at age of 28 days for different SCC grades.

Concrete Grade (MPa) 30 40 50

Test time (h) DD DD IP DD IP

12 0.646 0.516 20.101 0.423 34.519
24 1.438 1.426 0.821 0.599 58.350
36 2.084 2.428 16.506 0.740 64.492
48 2.671 3.065 14.762 0.951 64.381
60 3.375 3.369 0.197 1.092 67.639
72 4.080 3.520 13.710 1.268 68.908

Table 9
Depth decrement (mm) and the improvement percentage at age 90 days for different SCC grades.

Concrete Grade (MPa) 30 40 50

Test time (h) DD DD IP DD IP

12 0.516 0.423 18.070 0.277 46.342
24 1.154 0.937 18.840 0.339 70.661
36 1.913 1.420 25.780 0.492 74.259
48 2.612 1.994 23.650 0.585 77.608
60 3.310 2.508 24.244 0.708 78.613
72 3.887 2.961 23.831 0.862 77.829
8 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

4. Under water abrasion test results

This section shows the experimental results of SCC cylindrical specimens that were tested using ASTM C1138 (the
underwater abrasion test method). The Depth Decrement (DD) of each specimen due to the abrasive action, which was
determined based on the weight loss of the specimen after each testing time step, was used to present the abrasion results in
this section.

Fig. 4. Improvement percentages in abrasion resistance due to the increase of concrete design strength at age of (a) 7 days, (b) 28 days and (c) 90 days.
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 9

4.1. Influence of SCC grade

Fig. 2 displays the relation between DD and testing time for three different concrete grades at ages of 7, 28 and 90 days,
respectively. The figure obviously shows that DD increased as the accumulative testing time increased. Another notice is that
at the end of the test, DD reduced as concrete grade increased. Fig. 2(a) shows that at age of 7 days DD of grade 40 MPa was
smaller than DD of grade 30 MPa and that of grade 50 MPa was the smallest. This result is also obvious from the comparison
between the pictures (a)–(c) in Fig. 3, which shows the abraded test cylinders at the end of the 72 h abrasion test. The
pictures show that the total surface area of the 30 MPa specimen suffered significant abrasion loss, where it is seen that
the gravel particles were exposed due to the erosion of the surrounding cement paste. Some cavitations that appear in the
picture also refer to the ripping of some aggregate particles out of the matrix due to the grinding and impact action of the
steel balls and the water. The abrasion was less sever on the surface of the 40 MPa specimen (Fig. 3(b)), while no visual
exposure of aggregate particles on the surface of the 50 MPa specimen was noticed as shown in Fig. 3(c). This can be
considered as a constant behavior for ages of 28 and 90 days as well, but of course with lower abrasion rates for all grades. As
it is agreed, this is a predictable behavior because as compressive strength increases, the hardness of the concrete surface
develops. As shown in Fig. 2(a), the final DDs of grades 30, 40 and 50 MPa were 5.26, 4.26 and 2.73 mm, respectively. It can
also be noticed that during the first 48 h, grade 40 MPa exhibited more DD than grade 30 MPa but then, the abrasion rate of
grade 40 MPa significantly decreased while the DD of grade 50 MPa was always the lowest.
Fig. 2(b) presents the abrasion results at age of 28 days. The final DDs of grades 30, 40 and 50 MPa were 4.08, 3.52 and
1.26 mm, respectively. It is shown that for the 40 MPa specimen, the curve fluctuates widely during the first 60 h but shows a
small constant abrasion rate during the last 12 h of the test. Fig. 2(c) shows that the clearest DD difference of the three SCC
grades occurred at age of 90 days. DDs of grade 30, 40 and 50 MPa were 3.88, 2.96 and 0.86 mm, respectively. It can also be
seen that DD slopes with time is approximately constant along the test time for all grades in this age. The differences
between the three specimens is also shown by the comparison among the pictures (g)–(i) in Fig. 3, where the 30 MPa
specimen still show significantly abraded surface, while minimal surface deformations are shown for the 50 MPa specimen.
In order to enable a better understanding for the effect of concrete grade on the abrasion damage, the test results are
presented in bar charts that relate the Improvement Percentage (IP) with the testing period at each age and for different
concrete grades. The improvement percentage of each grade can be calculated as the difference between DD of grade 30 MPa
and the corresponding DD of the upper grade, divided by DD of grade 30 MPa as shown in Eq. (1), in which DD30 and DDgrade
are the depth decrements for grade 30 MPa at a certain testing time and specific age and the DD of the corresponding higher
grade specimen (40 or 50 MPa), respectively.
 IP ð%Þ ¼ 100ðDD30   DDgrade Þ=DD30 ð1Þ

The DD and the IP of the three different SCC grades for ages of 7, 28 and 90 days are listed in Tables 7–9, respectively. Fig. 4
shows the IP of grade 40 and 50 MPa specimens for the three test ages. Generally, there is a significant improvement due to
the enhancement of concrete grade for all ages. It can be seen that the improvement percentages of grade 40 MPa were
approximately equal for ages of 7 and 28 days while it slightly increased at the age of 90 days. This is because the compressive
strength increases and becomes close to its final value leading to higher surface hardness and thus higher resistance to the
abrasion of waterborne particles. On the other hand, the IPs of grade 50 MPa specimens were noticeably high and increased
considerably as concrete age increased.
Fig. 4(a) presents the IP values at the age of 7 days. The IP of grade 40 MPa started with negative values, it gradually
enhanced till testing time passed the first 48 h, then turned into positive values and increased significantly until they reached
19% at the end of the test. On the other hand, the IP of grade 50 MPa was always positive and reached its maximum value of
49% exactly at the middle of the test and remained approximately constant. Fig. 4(b) shows that at age of 28 days the

Fig. 5. Final abrasion rate after 72 testing hours for different concrete design strengths at ages of 7, 28 and 90 days.
10 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

improvement percentages of grade 40 MPa were better than that of 7 days during the test time but finally reached only 13.7%
which is smaller than that at 7 days, while the IPs of grade 50 MPa were constantly positive and increased with testing time
and finally stabilized at 69%. Fig. 4(c) shows that at 90-day age, the IPs of both grades were positive and reached to 23.8 and
77.8% for grades 40 and 50 MPa, respectively.
Another important notice is that when concrete grade was increased from 30 to 50 MPa, the IPs reached to 48, 69 and
77.8% for ages of 7, 28 and 90 days, respectively, while most of this enhancement took place due to grade increment from
40 to 50 MPa which equals 35.8, 64 and 70.9% for the three subsequent ages, respectively. Thus, a great abrasion
resistance can be achieved by using mixture of grade 50 MPa. This is due to the use of larger quantities of fine

Fig. 6. DD-testing time curves for different steel fiber contents (30 MPa) at age of (a) 7 days, (b) 28 days and (c) 90 days.
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 11

cementitious materials, where higher quantity of cement was used with approximately 11% replacement of silica fume.
The use of such materials significantly improves the concrete properties and minimizes the formation of initial cracks,
consequently they improve the concrete durability [35]. This type of mixtures is more convenient to use in hydraulic
structures that may be subjected to severe abrasion action. Using the water-jet abrasion test method, Liu [20] reported
that 10% replacement of cement by silica fume led to 25% higher abrasion resistance. Other researchers [36] using the
ASTM C1138 showed that only 5% replacement of silica fume could enhance the abrasion resistance significantly and led
to more uniform dispersion of fibers in the matrix, while Kang et al. [16] showed that 7% replacement of silica fume
could reduce the abrasion losses by 47%.
The final abrasion rate in gram/hour is the total weight loss after 72 h of abrasion testing divided by 72. Fig. 5 compares
the final abrasion rates of the three concrete design strengths at the three investigated ages. It is obvious that regardless of
design strength, the rate of abrasion decreases with age due to concrete maturity which leads to harder surfaces, thus, higher
abrasion resistance and lower abrasion weight losses. The figure also shows that the abrasion rate of the 50 MPa mature
specimens (28 and 90 days) was much less than the corresponding 30 and 40 MPa values, which supports the above
discussed findings.

Fig. 7. Abrasion cylinders after 72 h of abrasion test: effect of fiber content at different ages.
12 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

4.2. Influence of steel fiber content

Fig. 6 shows the relationship of DD of different fiber contents (0, 0.5, 0.75 and 1.0%) (30 MPa) with the testing time at
various testing ages (7, 28 and 90 days). Fig. 6(a) shows that at the middle of the 7-day age test, the specimens with 0
and 0.5% exhibited approximately the highest DD, while that with 0.75% fibers exhibited the lowest. The final DDs of
fiber contents of 0, 0.5, 0.75 and 1.0% were 5.26, 5.014, 4.847 and 4.795 mm, respectively with an average DD difference
of 0.155 mm for each fiber content increment. The incorporation of steel fiber minimizes the crack development and
significantly enhances the resistance to impact (reduces DD) and thus increases the durability of concrete and structure
life span [37]. At age of 28 days, Fig. 6(b) shows that DD curves exhibited relatively more stable ascending. After 36 h of
testing, specimens with 0 and 1.0% fiber showed the highest and the lowest DDs of 2.08 and 1.387 mm, respectively. The
final value of DD for 0% fiber was 4.08 mm, and it was equal for 0.5 and 0.75% fiber contents, while the specimen with
1.0% exhibited the lowest DD of 3.32 mm. Fig. 6(c) presents the results of the 90-day age specimens. The final results of
DD for specimens with 0, 0.5, 0.75 and 1.0% fiber contents were 3.89, 3.56, 3.35 and 2.87 mm, respectively. The presented
results show that the differences between the three fiber content are not very significant. Fig. 7 supports this
observation where it is shown in the figure that minor differences can be noticed between the three fiber contents at
ages of 7 (b–d), 28 days (f–h) and 90 days (j–l). However, comparing these differences among the three tested ages, it
was noticed that as specimen age increased, the final difference between DDs due to fiber addition increased. The
average of final DD decrement due to each fiber content increment at age of 90 days was 0.34 mm which increased by
about 55% as compared to that at 7 days age.
From the close examination of the abraded surfaces of the fibrous specimens, it was noticed that the randomly distributed
fibers exhibited different failure mechanisms. Mostly, the fibers were pulled out from the surrounding media, while some
fibers were cut while a part of which still fixed in place. Other fibers, especially groups of three or more fibers, worked as a
shield to protect the behind area in the direction of fiber inclination. Such area was termed in previous researches [14,38,39]
as the shadow zone. An acceptable explanation of the different failure behaviors of the steel fibers is their different
inclination angles in the concrete media due to their random distribution. The grinding balls move horizontally on the

Table 10
Depth decrement (mm) and the improvement percentage at age of 7 days for different fiber contents.

Fiber Content 0% 0.5% 0.75% 1.0%

Test Time (h) DD DD IP DD IP DD IP

12 0.723 0.812 12.32 0.721 0.17 0.585 19.06
24 1.561 1.623 4.00 1.385 11.26 1.520 2.58
36 2.601 2.609 0.29 2.164 16.81 2.398 7.82
48 3.382 3.391 0.29 3.116 7.85 3.421 1.17
60 4.335 4.232 2.38 3.982 8.16 4.211 2.88
72 5.260 5.014 4.67 4.847 7.85 4.795 8.84

Table 11
Depth decrement (mm) and the improvement percentage at age of 28 days for different fiber contents.

Fiber Content 0% 0.5% 0.75% 1.0%

Test Time (h) DD DD IP DD IP DD IP

12 0.646 0.436 32.47 0.617 4.52 0.578 10.48
24 1.438 0.959 33.30 1.086 24.47 0.896 37.70
36 2.084 1.715 17.70 1.820 12.65 1.387 33.43
48 2.671 2.326 12.93 2.554 4.37 2.081 22.09
60 3.375 2.965 12.16 3.142 6.93 2.717 19.51
72 4.080 3.517 13.78 3.494 14.36 3.324 18.53

Table 12
Depth decrement (mm) and the improvement percentage at age of 90 days for different fiber contents.

Fiber Content 0% 0.5% 0.75% 1.0%

Test Time (h) DD DD IP DD IP DD IP

12 0.516 0.434 15.88 0.553 7.11 0.609 18
24 1.154 0.840 27.24 1.019 11.73 1.073 6.98
36 1.913 1.419 25.85 1.630 14.81 1.538 19.63
48 2.612 2.056 21.29 2.212 15.31 1.944 25.57
60 3.310 2.866 13.41 2.823 14.71 2.350 29.01
72 3.887 3.561 8.39 3.347 13.90 2.872 26.11
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 13

specimen surface with the direction of water rotation, however, with fluctuated impact due to the high speed rotated water
and the restriction of the tank walls. Thus, fibers that have higher inclination angles may be subjected to direct concentrated
impact forces that weakens these parts of the fibers leading to its breaking. Yet, in some cases, groups of fibers held in place
resisting higher grinding and impact forces and shielding behind a shadow zone. The fibers that lays approximately

Fig. 8. Improvement percentages in abrasion resistance due to different fiber contents at age (a) 7 days, (b) 28 days and (c) 90 days.
14 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

horizontally or with low inclination angles were mostly exposed due to the continuous erosion of the covering matrix
leading finally to the delamination of these fibers. Similar abrasive behavior was also recorded by previous researcher [14]
where the shadow zone creation, the cutting of the fibers with higher inclination angles and the delamination of low
inclination angle fiber were distinguished. One difference is that in literature [14], the shadow zone was recorded behind
single fibers, while it was observed behind groups of fibers in the current research. The inability of single micro-steel fibers

Fig. 9. DD-testing time at different ages for grade (a) 30 MPa, (b) 40 MPa and (c) 50 MPa.
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 15

(used in this research) to form a shadow zone may be attributed to their shorter length (15 mm) compared to the 30 and
50 mm length fibers [14].
To achieve a better imagination of the steel fiber addition impact on SCC abrasion resistance, results are displayed in the
form of bar charts which relates IP with testing time. The IP magnitude of any fiber content at a certain testing time and age

Fig. 10. DD-testing time at different ages for steel fiber content of (a) 0.5%, (b) 0.75% and (c) 1.0%.
16 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299

can be determined directly depending on the DD of the 0% fiber specimen. Considering DD0% and DDfiber% as the DD of 0% and
any fibrous specimen at the same test time and age, respectively, Eq. (2) can be used to determine IP for any fiber content,
IP ð%Þ  ¼ 100ðDD0%   DDfiber% Þ=DD0% ð2Þ
Tables 10–12 present the DDs and the IPs of specimens with different steel fiber contents at ages of 7, 28, 90 days,
respectively, while Fig. 8 shows the IPs in abrasion resistance due to steel fiber inclusion at different ages. From Fig. 8(a) it can
be seen that the addition of steel fiber improves the abrasion resistance of SCC. At this age and especially at the first hours of
the tests, the IPs of fibrous specimens showed negative values, while at the end of the tests, specimens of 0.5, 0.75 and 1.0%
showed positive IP values of 4.7, 7.9 and 8.8%, respectively. The negative and relatively small IP values for 7-day age
specimens took place because of the use of superplasticizer with higher dosages for fibrous mixtures which reduced the
compressive strength and surface hardness at the early ages. Fig. 8(b) shows that at age of 28 days, the IPs of fibrous
specimens increased considerably as compared with that of 7 days. For 0.5% fiber, the IPs were in the range of 12.9–33.3%. On
the other hand, the IPs of 0.75 and 1.0% fiber contents were in the range of 4.3–24.5% and 10.5–37.7%, respectively. Fig. 8(c)
shows the data of 90-day age tests. During the first 12 h of the experiments, the IP of 0.5% fiber showed a positive value of 15%,
while specimens of 0.75 and 1.0% fiber showed negative values of 7 and 18%, respectively. This behavior was changed at the
middle of the tests and totally inverted during the last 24 h. Finally, the IPs of 0.5, 0.75 and 1.0% fibers were 8, 14, 26%,
respectively.
Horszczaruk [14] showed that mixtures with 50 mm (1.0 mm diameter) steel fiber suffered higher abrasion than the same
mixture without fiber, while that with fiber 30 mm (0.5 mm diameter) steel fiber exhibited the lowest abrasion among the three
mixtures. The author attributed the fall of abrasion resistance in the case of 50 mm fibers to their lower aspect ratio (50), which
increased their rigidity. Where the delamination of these fibers was accelerated due to the plastic deformation of the fibers under
the repeated impact of the steel balls, which led to the grooving of the contact zone with the surrounding matrix, resulting in
higher abrasion than the corresponding plain concrete. The results of the current study supports this conclusion, where the used
micro-steel fibers have an aspect ratio of 75 and the resulted total abrasion was lower than that of plain SCC mixture. Yet, the
percentage improvement due to fiber inclusion was less than 10% at age of 7 days and less than 20% at age of 28 days. Such
improvement percentage can be considered uneconomical when the high price of fibers is considered.

4.3. Influence of concrete age

The age impact on grade 30 MPa specimens is shown in Fig. 9(a), while Fig. 9(b) and (c) show that impact on the
specimens of grade 40 and 50 MPa, respectively. It can obviously be seen in these figures and supported by Fig. 3 that the DD
of 7-day age specimens was distinctly greater than that of 28-day age specimens which was greater than that of 90-day age
specimens. That is undoubtedly an expected behavior because the abrasion resistance develops with age due to the
compressive strength and surface hardness development by maturation. One more important notice can be focused on,
which is as concrete grade increased, the DD differences between each two successive ages widely increased. For grade
30 MPa, the DDs were 5.2, 4.1 and 3.9 mm at ages of 7, 28 and 90 days, respectively. Similarly, the DDs of grades 40 and
50 MPa at the same age sequence were 4.3, 3.5 and 2.9 mm; 2.7, 1.3 and 0.86 mm, respectively. Fig. 10 shows concrete age
impact on the resistance of abrasion for the three fibrous specimens (0.5, 0.75 and 1.0%). It can be noticed that the final DD of
the 7-day age curve was always the highest while the DDs of 28 and 90 days are too close for the specimen of 0.5% fiber but
this difference gradually increased as fiber content increased. These results can also be distinguished in Fig. 7 by the
comparison between the three pictures along each row of pictures.

5. Conclusions

Based on the results obtained from the current experimental investigation, the followings can be concluded,

1 For all abrasion specimens, abrasion damage increases with testing time. However, the rate of abrasion majorly depends
on concrete grade, concrete age and steel fiber content. For all tested mixtures, the abrasion loss is considerably reduced as
concrete age increases. Generally, the gained abrasion resistance due to the age development form 7 to 28 days represents
the greater part of the age development from 7 to 90 days. For mixture of grade 50 MPa, the abrasion improvement that
took place due to the age development from 7 to 28 days represents 78% of the abrasion enhancement that took place from
7 to 90 days.
2 For all ages, the abrasion resistance increases as the concrete grade increases. At the age of 90 days, the improvements in
abrasion resistance were 23.8 and 77.8% for grades of 40 and 50 MPa, respectively, compared to grade 30 MPa specimen.
From the fresh SCC tests and the underwater tests, it can be concluded that the mixture of grade 50 MPa is of acceptable
fresh properties and considerably higher abrasion resistance than the other mixtures due to the inclusion of silica fume.
Thus, it is convenient to be adopted in hydraulic structures where SCC is required to be used.
3 The inclusion of micro-steel fiber improves the abrasion resistance. Relatively small improvement percentages due to
micro steel fiber inclusion were obtained at the ages of 7 and 28 days. On the other hand, at 90 days age, the improvement
percentages for 0.5, 0.75 and 1.0% steel fiber inclusion were 8, 14 and 26%, respectively. Comparing the effect of fiber
 S.R. Abid et al. / Case Studies in Construction Materials 11 (2019) e00299 17

inclusion and concrete strength improvement, the 30 MPa mixture with 1.0% steel fiber showed much lower abrasion
resistance than the 50 MPa plain mixture and slightly greater abrasion resistance than that of the 40 MPa one by not more
than 6%. Despite this fact, the fibrous mixture with 1.0% steel fiber is practically more expensive than the 40 MPa and
50 MPa plain mixtures. Moreover, it has lower fresh SCC properties. Thus, it can be said that normal strength steel fiber-
reinforced SCC mixtures are not favorable to be used in the casting of huge and far parts of hydraulic structures.

Declaration of Competing Interest

The authors have no conflicts of interest to declare.

References

[1] ACI Committee 210, Erosion of Concrete in Hydraulic Structure (ACI 210R-03), American Concrete Institute, USA, 2003, pp. 1–24.
[2] A. Kryžanowski, M. Mikoš, J. Šušterši9 c, V. Ukrainczyk, I. Planinc, Testing of concrete abrasion resistance in hydraulic structures on the lower Sava river, J.
Mech. Eng. 58 (2012) 245–254.
[3] S.R. Abid, M.S. Shamkhi, N.S. Mahdi, Y.H. Daek, Hydro-abrasive resistance of engineered cementitious composites with PP and PVA fibers, Constr. Build.
Mater. J. 187 (2018) 168–177.
[4] L.S. Sklar, W.E. Dietrich, A mechanistic model for river incision into bedrock by saltating bed load, Water Resour. Res. 40 (2004) 1–21.
[5] C. Auel, I. Albayrak, T. Sumi, R.M. Boes, Sediment transport in high-speed flows over a fixed bed. 2: particle impacts and abrasion prediction, Earth Surf.
Process. Landf. 42 (2017) 1365–1383.
[6] Y.W. Liu, S.W. Cho, T.H. Hsu, Impact abrasion of hydraulic structures concrete, J. Mar. Sci. Technol. 20 (2012) 253–258.
[7] A. Kryžanowski, M. Mikoš, J. Šušterši9 c, V. Ukrainczyk, I. Planinc, Abrasion resistance of concrete in hydraulic structures, ACI Mater. J. 106 (2009) 349–356.
[8] C. Auel, R.M. Boes, T. Sumi, Abrasion prediction at Asahi sediment bypass tunnel based on Ishibashi’s formula, J. Appl. Water Eng. Res. 6 (2018) 125–138.
[9] ACI Committee 210.1R-94, Compendium of Case Histories on Repair of Erosion-Damaged Concrete in Hydraulic Structures (ACI 210.1R-04), American
Concrete Institute, USA, 1999, pp. 1–33.
[10] EFNARC, Specifications and Guidelines for Self-Compacting Concrete EFNARC, UK, (2002) , pp. 1–32.
[11] ACI Committee 237, Self-Consolidating Concrete, (ACI 237R-07), American Concrete Institute, USA, 2007, pp. 1–30.
[12] A.Z. Zadeh, B. Zafari, M. Yaminpour, Multifunctional use of self-compacting concrete as a fundamental material in dam construction: upper Gotvand
Dam, Key Eng. Mater. 629–630 (2015) 391–398.
[13] T. Yen, T.H. Hsu, Y.W. Liu, S.H. Chen, Influence of class F fly ash on the abrasion–erosion resistance of high-strength concrete, Constr. Build. Mater. 21
(2007) 458–463.
[14] E. Horszczaruk, Hydro-abrasive erosion of high performance fiber-reinforced concrete, Wear 267 (2009) 110–115.
[15] E. Horszczaruk, Abrasion resistance of high-performance hydraulic concrete with polypropylene fibers, Tribologia 1 (2012) 63–72.
[16] J. Kang, B. Zhang, G. Li, The abrasion-resistance investigation of rubberized concrete, J. Wuhan Univ. Technol.-Mater. Sci. Ed. 27 (2012) 1144–1148.
[17] R. Mohebi, K. Behfarnia, M. Shojaei, Abrasion resistance of alkali-activated slag concrete designed by Taguchi method, Constr. Build. Mater. 98 (2015)
792–798.
[18] S. Abid, A. Hilo, Y. Daek, Experimental tests on the underwater abrasion of engineered cementitious composites, Constr. Build. Mater. 171 (2018) 779–792.
[19] Y.W. Liu, T. Yen, T.H. Hsu, Abrasion erosion of concrete by water-borne sand, Cem. Concr. Res. 36 (2006) 1814–1820.
[20] Y.W. Liu, Improving the abrasion resistance of hydraulic-concrete containing surface crack by adding silica fume, Constr. Build. Mater. 21 (2007) 972–977.
[21] Z.J. Grdic, G.A.T. Curcic, N.S. Ristic, I.M. Despotovic, Abrasion resistance of concrete micro-reinforced with polypropylene fibers, Constr. Build. Mater. 27
(2012) 305–312.
[22] N. Ristic, G.T. Curcic, D. Grdic, Abrasion resistance of concrete made with micro fibers and recycled granulated rubber, Zastita Materijala 56 (2015) 435–445.
[23] R. Dandapat, A. Deb, A probability based model for the erosive wear of concrete by sediment bearing water, Wear 350–351 (2016) 166–181.
[24] N. Ghafoori, M. Najimi, M.A. Aqel, Abrasion resistance of self-consolidating concrete, J. Mater. Civ. Eng. 26 (2014) 296–303.
[25] ASTM C779 / C779M-00, Standard Test Method for Abrasion Resistance of Horizontal Concrete Surfaces, ASTM International, West Conshohocken, USA,
2000, pp. 1–5.
[26] K. Turk, M. Karatas, Abrasion resistance and mechanical properties of self-compacting concrete with different dosages of fly ash/silica fume, Indian J.
Eng. Mater. Sci. 18 (2011) 49–60.
[27] ASTM C1138-97, Standard Test Method for Abrasion Resistance of Concrete (Underwater Method), ASTM International, West Conshohocken, USA, 1997,
pp. 1–4.
[28] F.H. Arna’ot, A.A. Abbass, A.A. Abualtemen, S.R. Abid, M. Özakça, Residual strength of high strength concentric column-SFRC flat plate exposed to high
temperatures, Constr. Build. Mater. 154 (2017) 204–218.
[29] A.A. Abbass, S.R. Abid, M. Özakça, Experimental investigation on the effect of steel fibers on the flexural behavior and ductility of high-strength
concrete hollow beams, Adv. Civ. Eng. 2019 (2019) 1–13.
[30] M. Sahmaran, A. Yurtseven, I.O. Yaman, Workability of hybrid fiber reinforced self-compacting concrete, Build. Environ. 40 (2005) 1672–1677.
[31] ASTM C143/C143M, Test Method for Slump of Hydraulic-cement Concrete, ASTM International, West Conshohocken, USA, 2014, pp. 1–4.
[32] ASTM C 1712, Standard Test Method for Rapid Assessment of Static Segregation Resistance of Self-Consolidating Concrete Using Penetration Test, ASTM
International, West Conshohocken, USA, 2014, pp. 1–6.
[33] ASTM C 39/C 39M, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens, ASTM International, West Conshohocken, USA,
2002, pp. 1–5.
[34] ASTM C 496/C 496M, Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens, ASTM International, West Conshohocken,
USA, 2004, pp. 1–5.
[35] M. Gesog lu, E. Güneyisi, E. Özbay, Properties of self-compacting concretes made with binary, ternary and quaternary cementitious blends of fly ash,
blast furnace slag and silica fume, Constr. Build. Mater. 23 (2009) 1847–1854.
[36] T.-C. Cheng, A. Cheng, R. Huang, W.-T. Lin, Abrasion properties of steel fiber reinforced silica fume concrete according to Los Angeles and water abrasion
tests, Mater. Sci. (Medžiagotyra) 20 (2014) 498–502.
[37] B. Felekog lu, S. Türkel, Y. Altuntaş, Effect of steel fiber reinforcement on surface wear resistance of self-compacting concrete repair mortars, Cem.
Concr. Compos. 29 (2007) 391–396.
[38] X.G. Hu, A.W. Momber, Y. Yin, Hydro-abrasive erosion of steel-fiber reinforced hydraulic concrete, Wear 253 (2002) 848–854.
[39] X.G. Hu, A.W. Momber, Y. Yin, H. Wang, D.M. Cui, High-speed hydrodynamic wear of steel-fiber reinforced hydraulic concrete, Wear 257 (2004) 441–450.