Journal of the Civil Engineering Forum, September 2024, 10(3):239-248

DOI 10.22146/jcef.12439
Available Online at https://jurnal.ugm.ac.id/v3/jcef/issue/archive

Effects of Carbon Fiber on Mechanical Properties of Reactive Powder Concrete

Masdar Helmi* , Ala Ali Qasem Alraimi

Department of Civil Engineering, Universitas Lampung, Lampung, INDONESIA
*Corresponding author: [email protected]

SUBMITTED 5 March 2024 REVISED 2 April 2024 ACCEPTED 23 April 2024

ABSTRACT The use of fiber generally might change the mechanical properties of concrete in flexural or compressive strength. Reactive powder
concrete (RPC) is one of the ultra-high-performance concrete types that has been applied for some constructions. Carbon fiber, having high strength
in tensile, also has the potency to improve the physical characteristics of RPC. The purpose of this study is to evaluate the flexural and compressive
performance of RPC, focusing on the interfacial binding of carbon fiber. Flexural and compressive tests used samples with dimensions of 40 x 40 x
160 mm were tested in accordance with BS-EN-196-1:2011, which allows to use one of the two broken pieces from the flexural test as the sample. The
microstructure surrounding carbon fiber and paste was qualitatively compared using Scanning Electron Microscopy (SEM) in the secondary electron
detector mode. Samples were subjected to static pressure at 8 MPa after 1 hour of pouring and heat curing at 240 °C in a dry oven after demolding.
Results show that the presence of carbon fiber increased the flexural strength of RPC by up to 28.85% for samples without treatments and up to 14.32%
for samples with both treatments. Although carbon fiber increased toughness by 20% and flexural modulus by 6%, it had no effect on the failure mode
after reaching the peak load, which remained brittle. On the other hand, the presence of carbon fiber had little effect on compressive strength. Despite
the pressure and heat, curing treatments had no effect on enhancing the adhesion between carbon fiber and cement paste, which was indicated by the
undamaged surface of carbon fiber. However, the implementation of both treatments on samples might produce RPC with good mechanical properties
in flexure.

KEYWORDS RPC, Carbon Fiber, Mechanical Properties, Flexural Strength, Compressive Strength

© The Author(s) 2024. This article is distributed under a Creative Commons Attribution-ShareAlike 4.0 International license.

1 INTRODUCTION
Reactive Powder Concrete RPC is compositionally dif- ing it suitable for structures near water bodies such as
ferent from plain concrete because the RPC mixture foundations and harbors. According to Ženíšek et al.
uses a high proportion of fine particles <600 µm such as (2016) and Sanjuán and Andrade (2021), RPC was devel-
silica fume, quartz sand, or other admixtures, and is en- oped in France in the late 20th century and known for
tirely without coarse aggregates. The reactive powder its exceptional strength and durability, with a lifespan
components undergo chemical reactions through heat- of up to 200 years and compressive strength exceeding
accelerated hydration of the cement particles, heat- 150 MPa. RPC offers high compressive strength, tough-
accelerated pozzolanic reactions between Portlandite ness, low porosity, and low permeability in comparison
and amorphous silica from the densified silica fume, to high-performance concrete (HPC). It is a durable and
and the transformation of tobermorite to xonotlite dur- sustainable material suitable for various structural ap-
ing heat curing (Cwirzen, 2007). Every variation in plications.
composition and treatment affect the mechanical prop-
erties of RPC, where most studies on RPC have typically Previous study by Shi et al. (2019) describes RPC as an
focused on homogeneity and compacted density. Ac- ultra-high-performance cement composite with excep-
cordance to Ahmed Ali et al. (2021) study, RPC repre- tional strength, toughness, durability compared to reg-
sents a cutting-edge variant of high-performance con- ular concrete and different admixtures influence me-
crete. It distinguishes itself through its elevated ce- chanical properties of RPC and high-temperature cur-
ment content, minimal water content, and the incor- ing can improve its compressive strength. Increasing
poration of fine sand and silica dust. Notably, the re- the fiber volume fraction greatly enhances RPC’s ten-
silience of RPC specimens diminishes as the tempera- sile properties. The flexural properties of plain con-
ture rises. RPC, as described by Richard and Cheyrezy crete generally exhibit a direct correlation with com-
(1995), is an ultra-high-strength and highly ductile pressive strength, where an increase in compressive
concrete achieved through precise aggregate optimiza- strength leads to a corresponding increase in flexu-
tion and the use of fine powders, steel fibers (op- ral strength (Saloma and Agistin, 2019). However, this
tional), and superplasticizers. Its exceptional compact- correlation does not always hold true for RPC, espe-
ness gives it remarkable strength and durability, mak- cially in RPC mixtures containing steel fiber in combi-

239
Journal of the Civil Engineering Forum Vol. 10 No. 3 (September 2024)

nation with high-temperature curing. It has been ob- has found applications in diverse construction projects,
served that steam curing significantly increases com- such as the world’s inaugural RPC structure in France
pressive strength but simultaneously decreases flexu- and the Sherbrook Bridge in Canada.
ral strength (Al-Hassani et al., 2014). RPC is a compos-
ite material known for its impressive mechanical prop- There is limited research on RPC properties when car-
erties, such as high compressive and flexural strength. bon fiber is included in its mixture. The influence of
Achieved through the use of blast furnace slag cement pressure and heat curing on the interfacial bonding of
and optimized mixing, RPC could further be strength- carbon fiber is a critical aspect related to enhancing
ened by adding steel fibers, resulting in increased com- RPC performance. Therefore, it is essential to inves-
pressive and flexural strength (Janus and Grzeszczyk, tigate RPC performance when it contains carbon fiber
2020). and undergoes pressure and heat curing treatments.
The objective of this study is to assess the flexural and
Some researchers have conducted studies examining compressive performance of RPC in relation to its in-
changes in the properties of RPC by incorporating ad- terfacial bonding with carbon fibers.
ditional materials. For instance, they aimed to enhance
the bonding between fibers and paste (Al-Hassani et al.,
2 METHODS
2014) and reduce the cement content by replacing it
with industrial waste (Yazıcı et al., 2008; Bahedh and
Jaafar, 2018). In general, the mechanical properties of 2.1 Material and composition
concrete, whether it contains fibers or not, are actu-
ally influenced by conditions during and after the set- In this study, the composition of the RPC mixture
ting period (Cwirzen, 2007; Tam and Tam, 2012; Zhang adopted as described in Yazıcı et al. (2008), which in-
et al., 2019). The purpose of incorporating fibers is volved replacing 40%, of the weight of cement with
to enhance the performance of cementitious matrices ground-granulated blast furnace slag GGBS. Addition-
in terms of tensile strength, ductility, and durability ally, steel fiber substituted with carbon fiber with a di-
(Johnston, 2006; Alwash and Al-Sultan, 2018). The ameter of 7 µm and a length of 12 mm. This replace-
mechanisms of improvement include an increase in the ment resulted in a significantly higher number of car-
stress at which the matrix begins to crack, an increase bon fibers per unit volume compared to steel fibers,
in strain capacity after reaching the peak load and the given that steel fibers, have a diameter of 120 µm and a
ability to arrest crack development (Kadhum, 2015; Ra- length of 60 mm. In terms of the number of fibers per
heem et al., 2018). The effect of fibers on flexural cubic centimeter, carbon fiber offers 21 times the total
strength is greater than on compressive strength, and surface area of steel fiber. This implies that the fric-
this is dependent on three main factors: the volume of tional surface area of 1.0%, the volume fraction of steel
fibers, the physical properties of fibers and the matrix, fiber is equivalent to that of 0.05% volume fraction of
and the bond between them (Raheem et al., 2018; Han- carbon fiber.
nant, 1998). The addition of fibers in an RPC mixture
can improve its flexural performance, but using them in The composition of the RPC, by mass, was similar to
high proportions can increase the cost and reduce the that of a previous study (Helmi et al., 2018) but included
workability of the mixture (Wang et al., 2013; Kinayekar additional carbon fiber, as indicated in Table 1. The
et al., 2014). components used included Portland Cement (PC) type
1, quartz sand (QS) with three different grades of diam-
In a study conducted by Mahmoud Hama et al. (2021), eter A: 2.36-1.18mm, C: 0.60-0.30mm, E: 0.15-0.09mm,
the addition of steel fibers to RPC was found to improve micro silica fume (SF) grade 940-D, ground granu-
its ductility, changing its failure mode from brittle to lated blast furnace slag (GGBS), water, and a super-
behavior that is more ductile. Steel fibers help hold the plasticizer (SP) made from polycarbonate polymer. The
concrete together, delaying the formation of cracks and high-performance carbon fiber employed was of type
increasing its ability to deform. This increased ductility
is attributed to the steel fibers’ reinforcement of tensile
strength and crack arrest capabilities in RPC. Accord-
ing to Fawzi et al. (2021) study the impact of adding Table 1. The material composition of RPC per 1 m3
carbon fibers to RPC found that the different propor- Material Amount
tions of carbon fibers (0.5%, 1.5%, 2.0%) improved com-
pressive strength, flexural strength, density, and re- Portland cement (kg) 498
duced shrinkage in RPC. Chemical resistance remained Quartz sand (kg) 976
largely unaffected. RPC described as a durable cemen- Silica fume (kg) 208
titious composite with advanced mechanical and phys- GGBS (kg) 332
ical properties, including strength and ductility, supe- Water (liter) 200
rior mechanical and physical characteristics, and high- Superplasticizers (liter) 55
lights RPC’s exceptional strength and ductility, and it Carbon fiber (kg) 1.8

240
Vol. 10 No. 3 (September 2024) Journal of the Civil Engineering Forum

Figure 2 Consistency measurement on RPC fresh mixture

Figure 1 Bundle form of carbon fiber cut in 12mm long

T700SC-120000-50C with a grade of L-Tora-Soficar S.A. of 8 MPa was applied to the surface of the samples 5
France. The factory data sheets provide the following hours after casting and maintained for a duration of 2
physical properties: diameter equal to 7 µm; tensile days. Heat curing was conducted in a dry convection
strength equal to 4.9 GPa; tensile modulus equal to 230 oven at a temperature of 240 °C for 48 hours. Following
GPa; elongation equal to 2.10%; density equal to 1.8 g the heat curing process, the samples were allowed to
cm-3 . The fiber was initially in a thin bundle form and cool and subsequently cured in water until the testing
cut to a length of 12 mm using an automatic cutting time at an average room temperature of 20 °C. Samples
machine (as shown in Figure 1). identified as follows: (A) for samples without pressure
and cured in water; (B) for samples without pressure
The mixing procedure closely followed the method out- and heat cured in a drying oven; (C) for samples with
lined in the previous study (Helmi et al., 2018). Initially, pressure and cured in water; (D) for samples with pres-
the dry materials were placed in a mixer operating at sure and heat cured in a drying oven.
a speed of 120 rpm for a duration of 2 minutes. Sub-
sequently, the superplasticizer mixed with water and Samples were tested in flexure by UTM after 28 days
gradually introduced to the mixer. This is followed by (as shown in Figure 3), with a loading rate of 0.1 mm
further mixing at a speed of 450 rpm for approximately minute-1 applied at mid-span, and a clear distance of
10-12 minutes until the mixture achieves a consistent 100 mm between the simple supports, as specified in
texture. The entire mixing process took approximately BS-EN-196-1:2011. Linear variable differential trans-
12-16 minutes. formers LVDT, were installed on both sides of the mid-
dle span to measure the deflection of the sample during
The consistency of fresh RPC mixture was measured the loading process, enabling the plotting of load and
immediately after mixing in accordance with ASTM deflection data. Additionally, the compressive strength
C230 and following the steps of Nambiar and Rama- test, in accordance with BS-EN-196-1:2011 conducted
murthy (2008) by measuring the spread diameter of at 28 days using one of the two broken pieces from the
fresh RPC and comparing it to an initial diameter of 75 flexural test. The contact pressed area of the sample is
mm. Figure 2 shows that the RPC mixture containing almost 40 × 40 mm, and the height is 40 mm. Tough-
carbon fiber appears as plastic flowing and spreadable ness, which is the ability of a material to absorb energy
around 16 cm which was double than initial diameter before rupture, was accounted from the area under the
(75 mm) and higher than the consistency mixture in load-deflection curve in the flexure test up to 2.5 mm
Yazıcı et al. (2008). Once the mixture reached the de- mid-span (Ghosni et al., 2013). It is also possible to de-
sired consistency, it was poured into oiled steel molds, fine the flexural modulus from this curve using Equa-
which contained three prisms of sample with a dimen- tion 1 (Horners, 2008) as follows:
sion of 40 × 40 × 160 mm.

m.l3
2.2 Treatments and testing Ef f = (1)
4.b.d3

Two treatments were implemented for the fresh sam-

where Ef f is the Flexural modulus of elasticity (MPa),
ples: static pressure and heat curing. A static pressure
m is a gradient of the initial straight-line portion of the

241
Journal of the Civil Engineering Forum Vol. 10 No. 3 (September 2024)

Figure 4 Flexural strength of both RPC non-fiber (NF) and carbon

Figure 3 Configuration flexural testing on UTM
fiber (CF) for all conditions

load-deflection, l is the span of support, b is the width due to the mechanism of carbon fibers, which enhance
of the beam, and d is the depth of the beam. the stress in the matrix before cracking and increase
the maximum load in flexure (Johnston, 2006; Yazıcı,
2018). Moreover, the presence of fibers in reinforced
3 RESULTS concrete can also enhances the bonding between the
matrix and bars by impeding crack growth (Alwash and
3.1 Flexural strength Al-Sultan, 2018).

Flexure testing of RPC prisms employed the three-

point loading method with a rate of 0.1 mm/min and 3.2 Toughness and modulus elasticity
a clear distance of 100 mm between the supports. The
flexural test results for RPC with carbon fiber labeled Flexural testing under all curing conditions exhibited
CF compared with those without fiber labeled NF un- brittle failure modes, with samples rupturing after
der four different conditions. Figure 4 displays the av- reaching their peak load, and only minimal deforma-
erage flexural strength of three samples with and with- tion prior to failure. Consequently, deflection after the
out carbon fiber under the same curing conditions. It peak load could not recorded. To assess the impact of
reveals that carbon fiber had varying influences on flex- carbon fiber in RPC mixtures, the measurements from
ural strength across the treatments, with -2.37% for A, three samples under condition D, both NF and CF, also
16.42% for B, 28.85% for C, and 14.32% for D. Condition plotted in Figure 5.
A exhibits overlapping error bars, suggesting that NC
and NF have similar strengths. Treatment A resulted in The presence of fiber did not seem to have an apparent
the flexural strength of NF being slightly higher than effect on the failure mode, as it continued to exhibit
CF. It is supposed to be due to the presence of an sudden rupture after reaching the peak load, similar to
entrapped void around fiber and a less dense matrix, RPC without fiber brittle failure (as seen in Figure 5).
which then reduces the flexural due to a decrease in However, the gradient of the CF line increased after the
compressive strength. Figure 4 also shows that the flex- first crack (at a point of about 500 N of force), which also
ural strength increased in line with the treatments ap- indicates the increase in stiffness. It supposes that the
plied. It seems that the carbon fiber role slightly in- presence of carbon fiber might hold the crack progres-
creased flexural strength for the same treatments, and sion till due to fiber in pullout resistance mechanism.
treatment D had the most influence on strength. How- In addition, the flexural strength of samples CF in-
ever, regarding overlapping error bars, it suggests that creased after pressure and curing treatments were ap-
NC and NF have similar strengths. plied, which might increase the compressive strength.
It indicates that the pullout resistance was affected by
When comparing RPC with carbon fiber to the same an increase in compressive strength (A. et al., 2022).
mixture under condition A, the commonly used condi- The improved strain capacity after peak load, observed
tions generally improved flexural strength in a range in RPC with carbon fiber, was not evident in the curve,
of 30-40%. It is evident that the addition of carbon highlighting a different behavior from RPC with steel
fiber has a significant impact on the flexural strength fiber (Al-Hassani et al., 2014). This suggests that the
of an RPC, mixture when subjected to any of the al- behavior of carbon fiber after high-temperature curing
tered curing conditions. This effect is believed to be is the primary cause of this rupture mode.

242
Vol. 10 No. 3 (September 2024) Journal of the Civil Engineering Forum

Figure 5 Load and deflection curves for NF and CF with condition

Figure 6 Failure condition of sample after peak load
D showing immediate rupture after peak load

It is known that steel fiber has a thermal coeffi- play a role during the heat curing process (Pradere and
cient of expansion similar to that of plain concrete Sauder, 2008; Neville, 2018).
(Neville, 2018), which is approximately 13.51 ×10-6 /°C,
whereas carbon fiber has a lower coefficient of expan-
sion (Pradere and Sauder, 2008), around 2.1 ×10-6 /°C. 3.3 Compressive strength
When the RPC sample cured at high temperatures and
then cooled in the air, the expansion and shrinkage of The compression test results for NF and CF under all
carbon fiber is less than that of cement paste or steel curing conditions are depicted in Figure 7. In condition
fiber. This behavior may diminish the bonding mech- A, the compressive strength of CF is marginally higher
anism between the paste and the carbon fiber, subse- than that of NF. Despite the relatively weak bond be-
quently reducing the fiber’s pullout strength. More- tween carbon fiber and cement paste, this improvement
over, the pullout strength of carbon fiber at ambient is actually likely attributed to the role of carbon fiber in
temperature is also lower than that of steel fiber, with preventing random crack propagation. However, since
1.29 MPa for carbon fiber and 5.48 MPa for steel fiber the error bars overlap, it can be assumed that the com-
(Katz et al., 1995). The use of carbon fiber in low- pressive strength of both NF and CF is similar.
volume fractions of 0.1%, is believed to be an additional
factor contributing to the failure occurring shortly af- Under condition B, the presence of carbon fiber in the
ter reaching the maximum load (Johnston, 2006). It ap- RPC mixture resulted in a marginal 2% increase in com-
pears that the weak bonding between carbon fiber and pressive strength. However, this difference is not sta-
the cement paste, possibly caused by the heat curing tistically significant when considering the error bars.
process, is a primary factor leading to brittle failures in Notably, the compressive strength of RPC with carbon
RPC (as shown in Figure 6). fiber, subjected to heat curing in a dry oven, shows
distinct results when compared to samples with steel
Toughness quantified as the area under the load- fibers that have undergone dry oven curing or steel
deflection curve up to 2.5 mm of deflection, and the fibers subjected to autoclave curing (Tam and Tam,
flexural modulus calculated using Equation 1 refer to 2012). Despite dry oven curing also leading to the for-
Table 2. The introduction of carbon fiber into the RPC mation of xonotlite (Tam and Tam, 2012); it does not
mixture led to a significant 20% increase in toughness significantly enhance compressive strength when the
and a slight 6% increase in the flexural modulus. This sample contains carbon fiber. This might be attributed
improvement is likely attributed to the fiber’s role in to the varying thermal expansion coefficients between
impeding crack propagation and absorbing energy be- carbon and steel fiber, resulting in higher internal ther-
fore the sample undergoes complete failure (Johnston, mal stresses with carbon fiber (Szoke, 2006).
2006; Raheem et al., 2018). However, the presence of
carbon fiber did not appear to have an effect on the Under condition C, there was a slight increase in com-
failure mode, as the samples still exhibited immedi- pressive strength, approximately up to 6%. This in-
ate, brittle failure following the peak load. This Be- crease might attributed to the application of pressure,
havior may be linked to an increase in porosity within which could have resulted in the repositioning of the
the paste and at the interface with the carbon fibers. binder particles around the fiber, making them closer
Additionally, the difference in thermal expansion co- and more compacted, reducing the void ratio. This,
efficients between carbon fiber and the cement paste in turn, could enhance the bond between the cement

243
Journal of the Civil Engineering Forum Vol. 10 No. 3 (September 2024)

Table 2. Toughness and flexural modulus result for NF and CF

with condition D
No Load (N) Deflection (10-2 mm) Toughness (N.mm) Ef (MPa)
Type NF
1 6448 28.1 584 4547
2 6047 27.4 519 4527
3 6448 27.3 500 4330
Average 535 4468
Type CF
1 7345 29.1 694 4774
2 6847 27.4 631 4639
3 6719 27.3 606 4742
Average 644 4718 Figure 7 Compressive strength NF and CF for all conditions show-
ing the overlapping of error bars between both.

paste and the fiber. Moreover, the application of pres- promotes pozzolanic reactions and transforms C-S-H
sure may also reduce the porosity of the cement paste phases from tobermorite to xonotlite, (Cwirzen, 2007).
(Cwirzen, 2007). This observation supports the expla-
nation provided for condition B, suggesting that the
differential expansion of the cement paste and carbon 3.4 Interfacial binding between matrix and fiber carbon
fiber during heat curing might be a factor contributing
to the relatively lower compressive strength. In general, pressure curing reduces the bulk density of
the paste, whereas heat curing further fills micro-pores
Under condition D, the measured compressive strength with hydration products resulting from the pozzolanic
exhibited a slight increase of up to 7%. However, the in- reaction. To assess the effects of both treatments, the
fluence of carbon fiber, in this case, is somewhat dimin- microstructure of the interfacial zone Surrounding car-
ished due to the significant difference in the thermal bon fibers qualitatively compared using Scanning Elec-
expansion coefficient between the carbon fibers and tron Microscopy (SEM) in secondary electron detector
the cement paste, as previously explained, and believed mode, as illustrated in Figure 8.
to be an additional factor contributing to the failure oc-
curring shortly after reaching the maximum load (John- Figure 8 (a) displays the surface of the carbon fiber
ston, 2006). It appears that the weak bonding between at the fracture surface of the specimen without any
carbon fiber and the cement paste, possibly caused by treatment (A). The fibers surface appears remarkably
the heat curing process, is a condition B. These com- smooth without any scratches or signs of deboned
pressive strength results are statistically similar, as in- paste, indicating a poor adhesion between the cement
dicated by the overlapping error bars. Regarding the paste and the carbon fiber. In Figure 8 (b), the surface of
result of NF and CF in Figure 7, the presence of car- the paste depicted after the carbon fiber has pulled out
bon fiber had little effect on compressive strength, and during the prism’s rupture. This surface shows numer-
the increase in strength was mostly affected by treat- ous micro-pores with a diameter of around 0.1 190 µm.
ments. The compressive strength of both NF and CF Applying pressure and heat curing still appears to have
without treatments (condition A) was less than 100 no significant effects on the surface of the carbon fiber,
MPa and increased after treatments, becoming over 100 as seen in Figure 8 (c). However, both treatments have
MPa. Even both treatments (condition D) resulted in resulted in the cement paste becoming notably denser,
the compressive strength of RPC without carbon fiber along with longitudinal crack propagation, as observed
(NF) being better than C-CF, B-CF, and A-CF. It indi- in Figure 8 (d).
cates that condition treatment has most affected on
compressive strength than CF presence. Nonetheless, These observations align with the previous discussion
condition D is totally recommended as a method to pro- regarding the behavior of RPC in flexural tests, where
duce RPC samples with favorable properties from both brittle rupture and a loss of strength were then ob-
a mechanical and microstructural perspective. served after reaching the peak load. It appears that
the treatments applied to RPC specimens have limited
There are two distinct advantages to applying condi- effectiveness in enhancing the adhesion between car-
tion D: first, the application of pressure can compact bon fibers and the cement paste. Furthermore, the un-
the matrix, thereby enhancing the capacity of the added damaged surface of the carbon fiber suggests a primar-
carbon fiber to mitigate the initiation of micro defects ily friction-based bonding mechanism with a constant
within the concrete and prevent subsequent bridging rate of fiber pullout (Katz et al., 1995). The main lim-
cracking (Chen and Liu, 2004). Second, heat curing iting factor in brittle rupture appears to be the adhe-

244
Vol. 10 No. 3 (September 2024) Journal of the Civil Engineering Forum

Figure 8 Interfacial carbon fiber for treatment A and D: (a) and (b) are surfaces of carbon fiber and cement paste in treatment A; (c) and
(d) are the surface of carbon fiber and cement paste in treatment D.

sion between carbon fiber and cement paste, while the ence of carbon fiber also increased the toughness by
pressure and heat treatments only contribute to the en- 20% and the flexural modulus by 6%. This enhance-
hancement of paste properties rather than the com- ment would be attributed to the role of fibers in sup-
posite as a whole. This limitation might potentially pressing crack progression, which resulted in the ab-
to improved by implementing treatments that focus on sorption of rupture energy before the complete sepa-
the preparation or functionalization of the carbon fiber ration of the sample occurred. Despite these improve-
surface, such as heating, ozonation, or NaOH solution ments in flexural properties, the influence of carbon
etching (Cwirzen, 2007). fiber appeared to have a limited effect on compressive
strength under all treatment conditions. The treat-
ments applied in this study had little effect on en-
4 CONCLUSION hancing the adhesion between carbon fibers and ce-
ment paste. Consequently, brittle rupture occurred af-
The addition of carbon fiber to the RPC mixture signif- ter reaching the peak load, ultimately leading to fiber
icantly improved the flexural strength of RPC, with an pullout as the mode of failure. However, it should be
increase of up to 40% observed for D treatments. How- noted that both pressure and heat-curing treatments
ever, it did not have an apparent effect on the failure could yield low fiber fraction RPC mixes with favorable
mode of RPC, which continued to exhibit sudden rup- flexural properties. Pressure treatment during curing
ture after reaching the peak load brittle failure, similar compacts the matrix and enhances the capacity of the
to RPC without carbon fiber. This suggests that the dif- added carbon fiber to mitigate the initiation of micro-
ference in thermal expansion coefficients between car- fractures and prevent subsequent bridging cracking.
bon fiber and cement paste during heat curing condi-
tions were the cause of this brittle rupture. The pres-

245
Journal of the Civil Engineering Forum Vol. 10 No. 3 (September 2024)

DISCLAIMER Helmi, M., Hall, M. R. and Rigby, S. P. (2018), ‘Effect

of pressure and heat treatments on the compressive
The authors declare no conflict of interest. strength of reactive powder concrete’, MATEC Web of
Conferences 147, 01006.
URL: https://doi.org/10.1051/matecconf/201814701006
REFERENCES
Horners, S. (2008), Activity 20 - Bendy Wafer, Pearson
A., R., Putra, J. W. and Saputra, J. (2022), ‘Chemical an- Education, Essex, United Kingdom.
chor pullout force modeling with variation of anchor Janus, G. and Grzeszczyk, S. (2020), ‘Reactive pow-
embedment length in concrete and concrete strength’, der concrete with blastfurnace slag cement’, Cement,
Journal of the Civil Engineering Forum 8(3), 277–290. Wapno, Beton 2020(4), 306–315.
URL: https://doi.org/10.22146/jcef.3769
URL: https://doi.org/10.32047/cwb.2020.25.4.5
Ahmed Ali, M., Elthakeb, A., Mohamed, U. and
Johnston, C. D. (2006), Fiber-Reinforced Cements and
Taha Noaman, M. (2021), ‘Behavior of reactive powder
Concretes, Taylor & Francis, New York.
concrete beams exposed to fire’, Diyala Journal of Engi-
neering Sciences 14(4), 1–10. Kadhum, M. M. (2015), ‘Studying some mechanical
URL: https://doi.org/10.24237/djes.2021.14401 properties of reactive powder concrete using local ma-
Al-Hassani, H. M., Khalil, W. I. and Danha, L. S. (2014), terials’, Journal of Engineering 21(7), 113–135.
‘Mechanical properties of reactive powder concrete URL: https://doi.org/10.31026/j.eng.2015.07.09
with different steel fiber and silica fume contents’, ATC Katz, J., Li, V. C. and Kazmer, A. (1995), ‘Bond proper-
Bulletin of Engineering Tome VII, 48–58. ties of carbon fibers in a cementitious matrix’, Journal
URL: https://doi.org/10.30684/etj.31.16A.8 of Materials in Civil Engineering 7(2), 125–128.
Alwash, J. J. and Al-Sultan, S. M. B. (2018), ‘Flexural be- URL: https://doi.org/10.1061/(ASCE)0899-
havior of reactive powder concrete beams with various 1561(1995)7:2(125)
types of fibers’, International Journal of Civil Engineering
Kinayekar, S. M., Gundakalle, V. D. and Kulkarni, K.
and Technology 9(4), 32–44.
(2014), ‘The effect of the addition of carbon fibers on
Bahedh, M. A. D. R. and Jaafar, M. S. (2018), ‘Ultra high- the mechanical properties of high-strength concrete’,
performance concrete utilizing fly ash as a cement re- International Journal of Innovative Research in Science,
placement under autoclaving technique’, Case Studies Engineering, and Technology 3(1), 8777–8784.
in Construction Materials 9.
URL: https://doi.org/10.1016/j.cscm.2018.e00202 Mahmoud Hama, S., Hamdullah, D. N. and Hama, S. M.
(2021), ‘Reactive powder concrete beam’s behavior in
Chen, X. and Liu, J. (2004), ‘Residual strength of hybrid- flexural: Review’, Diyala Journal of Engineering Sciences
fiber-reinforced high-strength concrete after exposure 14(4), 113–131.
to high temperatures’, Cement and Concrete Research URL: https://doi.org/10.24237/djes.2021.14410
34(6), 1065–1069.
URL: https://doi.org/10.1016/j.cemconres.2003.11.010 Nambiar, E. and Ramamurthy, K. (2008), ‘Models for
strength prediction of foam concrete’, Materials and
Cwirzen, T. (2007), ‘The impact of the heat-treatment structures 41(2), 247–254.
process on the characteristics of reactive powder con- URL: https://doi.org/10.1617/s11527-007-9234-0
crete’, Advance Cement Research 19(1), 25–33.
URL: https://doi.org/10.1680/adcr.2007.19.1.25 Neville, M. (2018), Properties of concrete, 5th edn, Pear-
son Education Limited.
Fawzi, N. M., Aljalawi, N., Waisi, B., Hassan, A. A., Waisi,
B. I., Saeed, S. S. and Megbel, S. M. (2021), ‘Effect of Pradere, P. and Sauder, C. (2008), ‘Transverse and lon-
nanomaterial on properties of reinforced reactive pow- gitudinal coefficient of thermal expansion of carbon
der concrete by carbon fiber’, Turkish Journal of Physio- fibers at high temperatures (300- 2500 k)’, Carbon
therapy and Rehabilitation 32(3). 46(14), 1874–1884.
URL: https://doi.org/10.1016/j.carbon.2008.07.035
Ghosni, N., Samali, B. and Vessalas, K. (2013), Energy
absorption and flexural toughness evaluation of fiber- Raheem, H. A., Mahdy, M. and Mashaly, A. A.
reinforced polymer modified concrete, in ‘Proceeding (2018), ‘Mechanical and fracture mechanics proper-
of VIII International Conference on Fracture Mechanics ties of ultra-high-performance concrete’, IOSR Jour-
of Concrete and Concrete Structures FraMCoS- 8 (Van nal of Mechanical and Civil Engineering (IOSR-JMCE)
Mier et al Eds)’, Toledo, Spain. 15(5), 33–39.
Hannant, D. (1998), Fibre-reinforced cements and con- URL: https://doi.org/10.1016/j.conbuildmat.2019.03.298
cretes, in J. Illston, ed., ‘Construction Materials-Their
Nature and Behavior’, E & FN Spon, London.

246
Vol. 10 No. 3 (September 2024) Journal of the Civil Engineering Forum

Richard, P. and Cheyrezy, M. (1995), ‘Composition of re- heating regimes’.

active powder concrete’, Cement and Concrete Research URL: https://doi.org/10.1680/macr.2012.64.3.259
25(7), 1501–1511.
URL: https://doi.org/10.1016/0008-8846(95)00144-2 Wang, J. Y., Chia, K. S., Liew, J. Y. R. and Zhang, M. H.
(2013), ‘Flexural performance of fiber-reinforced ultra-
Saloma, H. and Agistin, V. (2019), ‘Analysis of mechan- lightweight cement composites with low fiber content’,
ical properties of reactive powder concrete with curing Cement and Concrete Composites 43, 39–47.
temperature variation’, IOP Conf. Series: Materials Sci- URL: https://doi.org/10.1016/j.cemcon-
ence and Engineering 620, 012045. comp.2013.06.006
URL: https://doi.org/10.1088/1757-899X/620/1/012045
Yazıcı, H. (2018), ‘Prediction of the strength proper-
Sanjuán, M. A. and Andrade, C. (2021), ‘Reactive powder ties of carbon fiber-reinforced lightweight concrete ex-
concrete: Durability and applications’, Applied Sciences posed to high temperature using artificial neural net-
(Switzerland) 11(12). work and support vector machine’, Advances in Civil En-
URL: https://doi.org/10.3390/app11125629 gineering 2018, Article ID 5140610.
Shi, R. J., Xiang, Z., Liu, S. G., Wang, D. H. and Ju, Y. Z.
Yazıcı, H., Yiğiter, H., Karabulut, A. S. and Baradan, B.
(2019), ‘Mechanical properties of reactive powder con-
(2008), ‘Use of fly ash and ground granulated blast fur-
crete: A review’, IOP Conference Series: Materials Sci-
nace slag as an alternative silica source in reactive pow-
ence and Engineering 629(1).
der concrete’, Fuel 87(12), 2401–2407.
URL: https://doi.org/10.1088/1757-899X/629/1/012009
Szoke, S. (2006), Resistance to fire and high tempera- Zhang, X., Ge, L., Zhang, Y. and Wang, J. (2019),
tures, in J. Lamond and J. Pielert, eds, ‘Significance of ‘Mechanical properties of carbon fiber rpc and design
Tests and Properties of Concrete and Concrete-Making method of carbon fiber content under different curing
Materials’, ASTM International, West Conshohocken, systems’, Materials 12, 3759.
PA. Ženíšek, M., Vlach, T. and Laiblová, L. (2016), ‘Flexural
Tam, C. M. and Tam, V. W. Y. (2012), ‘Microstructural strength of reactive powder concrete’, Solid State Phe-
behavior of reactive powder concrete under different nomena 249, 108–111.

247
Journal of the Civil Engineering Forum Vol. 10 No. 3 (September 2024)

[This page is intentionally left blank]

248