Materials and Structures

, 51:98 | Cite as

Residual flexural behavior of fiber reinforced concrete after heating

  • Piti Sukontasukkul
  • Sittisak Jamnam
  • Manote Sappakittipakorn
  • Kazunori Fujikake
  • Prinya Chindaprasirt
Original Article


In this study, the effects of fire on the flexural performance and residual strength of plain and fiber reinforced concrete are investigated. Three types of concrete are tested: plain, polypropylene (PFRC) and steel fiber reinforced concrete (SFRC). Prior to the flexural test, the specimens were exposed to fire for 15, 30, 45, and 60 min on a furnace. The burnt specimens were then tested under flexural load to measure their toughness and residual strength. Results indicate the reduction of flexural strength for both plain and FRC after being subjected to fire. For FRC, the effect of fire on the flexural response depends mainly on the fiber type and fire exposure duration. For PFRC, the flexural strength is found to drop significantly for every exposure duration, while toughness is found to increase at short exposure duration and then, drop quickly after long exposure duration due to the fiber evaporation effect. For SFRC, the flexural strength and toughness are found to drop gradually for every exposure duration due to the deterioration of cement paste and reduction in bond strength. SFRC exhibits a more consistent ability to maintain load carrying capacity after long exposure to fire than PFRC.


Fire Toughness and residual strength Polypropylene fiber reinforced concrete Steel fiber reinforced concrete 

1 Introduction

Concrete structures may occasionally be subjected to fire accidents which are considered one of the most serious risks for buildings and their residents. Concrete, in general, has considerable fire resistance compared with other construction materials like timber or steel. During a fire incident, several phenomena may occur: pore pressure increase, evaporation of pore-water, hydration of unhydrated cement particles, cracking and spalling of concrete, debonding between cement paste and aggregates due to incompatible thermal expansion [1, 2], etc. The occurrence of stress due to a thermic gradient can also cause internal cracking and lead to the failure of a structural member [3]. These changes affect the mechanical properties of concrete differently, depending on the temperature level. At temperatures below 300 °C, compressive strength may remain unchanged [4] or, in some cases, increase slightly due to the hydration of the unhydrated cement particles. At temperatures between 300 and 550 °C, compressive strength is found to decrease by about 15–40%. Above 550 °C, the decrease becomes significant ranging from 55 to 70% [5, 6].

The use of fibers in order to enhance the flexural performance of fiber reinforced concrete (FRC) has been studied for several decades and the evidence is well documented. In the case of fire resistance, comprehensive studies can be found dating back to the early 2000s. Lau and Anson [7] reported the decrease in both compressive and flexural strength of both plain and 1% steel fiber reinforced concrete (SFRC) subjected to high temperatures ranging from 105 to 1200 °C. However, the SFRC appeared to exhibit higher residual strength than plain concrete. Poon et al. [8] reported the effects of temperatures ranging from 200 to 800 °C on both steel and polypropylene FRC (PFRC). At temperatures below 200 °C, the compressive strength of both plain concrete and FRC remained unchanged. Compressive strength began to drop as the temperature increased above 200 °C. For compression toughness, SFRC exhibited a higher energy absorption than plain concrete even at the highest test temperature while the compression toughness of PFRC was found to decrease quicker than that of plain concrete and SFRC at the highest temperature.

Gao et al. [9] reported the effect of supplementary materials like granulated blast furnace slag (GBFS) on the flexural behavior of FRC after subjecting it to elevated temperatures. The improvement in residual strength of both SFRC and PFRC was found. Zheng et al. [10] indicated that the compressive strength of SFRC mixed with reactive powder remained unchanged when subjected to temperatures below 300 °C. However, as temperature increased above 300 °C, compressive strength began to drop abruptly. Morsy et al. [11] found that after 2 h of exposure to elevated temperature, the addition of steel fiber (from 0.5 to 1.5% volume fractions) enhanced the ultimate compressive load of FRC by 10–15% and the ultimate splitting tensile strength by 25–37% compared with plain concrete. Lee et al. [12] compared the effect of nylon and polypropylene fibers on the spalling resistance of concrete subjected to fire. They found that a combination of the two fibers can improve spalling protection by providing connections between pores with a low fiber content. Bashir and Singh [13] also showed the increase in strength and better fire resistance in concrete containing polypropylene fibers subjected to high temperatures between 300 and 800 °C.

In a previous study, Sukontasukkul and Pomchiengpin [14] studied the effect of high temperatures up to 800 °C on the post-peak response of FRC under compressive and flexural loadings. Under compression, there was no strength gain found at any temperature level. However, under flexural load, the post-peak response differed depending on the temperature level. Below 400 °C, the post-peak load and the toughness were found to increase slightly, though they later dropped as the temperature increased above 400 °C. Fiber type and content also played an important role on the post-peak behavior of FRC after being subjected to high temperatures. Similar findings were also reported by Li et al. [15] who found that fire damage does not inevitably lead to the deterioration of mechanical properties but that residual flexural properties decrease significantly because of the melting of PVA fibers.

This study is an extension of the authors’ previous work in which the effect of fire direction on the behavior of FRC flexural members was investigated. It is known that the bottom surface of concrete members is a critical region where tensile stresses normally occur, and which has a significant influence on the fire resistance of the flexural member [16, 17]. In both studies, the effect of fire at the bottom surface of a reinforced concrete slab was investigated and analyzed. A robust model was also proposed by Huang [16]. Lo Monte and Felicetti [18] focused on the use of a controlled biaxial load setup which proved to be very effective in assessing the damage of concrete slabs subjected to fire at their bottom surface. The uneven exposure to fire also causes a thermic gradient stress along the cross-section of the structural member which leads to internal cracking and failure of the structure.

In our study, the effect of fire acting directly at the bottom surface of an FRC beam at different exposure durations is investigated. The main purpose is to study the change in the flexural performance of FRC after being subjected to fire at the specific location (bottom surface) and for different durations. A furnace with four gas heads, designed and constructed at the Department of Civil Engineering, KMUTNB, is used to apply fire to the specimen. Three types of concrete (plain concrete, SFRC and PFRC) are cast in form of beam specimens with dimensions of 100 × 100 × 350 mm, cured for 28 days and then subjected to fire from a furnace for 15, 30, 45 and 60 min. The burned specimens are then tested for flexural performance (ASTM 1609) to study the effect of fire and exposure duration.

2 Experimental procedure

2.1 Materials and mix proportion

Materials used in this study consisted of:
  • Type I Portland cement,

  • Fine aggregates: River sand passing sieve no. 4

  • Coarse aggregates: Crushed limestone rock with maximum size of 19 mm.

  • Fibers: Hooked end steel fibers and crimped polypropylene fibers (Table 1).
    Table 1

    Fiber properties


    Specific gravity

    Length (mm)

    Aspect ratio (l/d)

    Tensile strength (N/mm2)











The concrete mix proportion was set at 1:2.4:2.9:0.37 (cement:fine aggregate:coarse aggregate:water). For FRC, the volume fraction of 0.5 and 1% was used (Table 2). The concrete was mixed using a pan mixer and cast in the form of specimens with dimensions of 100 × 100 × 350 mm. The specimens were cured in water and air for 28 and 7 days respectively, prior to being subjected to fire.
Table 2

Mix proportions and number of specimens


Cement (kg/m3)

Fine agg. (kg/m3)

Coarse agg. (kg/m3)

Water (kg/m3)

Fibers (kg/m3)

No. of specimens




































2.2 Burning process

During the casting process, two thermocouple grade wires (type K) were placed near the top and bottom surfaces of the specimen by embedding them in the specimen at about 10 mm from the surfaces. The extruding part of the wire was protected against the flame during the burning process with a heat-treated fiberglass sleeve with an operating temperature range of 600 °C. Prior to the burning process, the thermocouples were connected to the data acquisition system to obtain the change in temperature over time. Three specimens from each type of concrete were burned for 15, 30, 45, and 60 min using a high pressure open furnace constructed at the Department of Civil Engineering, KMUTNB (Fig. 1a). The furnace was constructed from four steel walls with an opening at the top. The base consisted of four high pressure gas-heads with separate controlling valves. The furnace was tested and calibrated in the lab with a temperature–time reference curve as shown in Fig. 1b.
Fig. 1

a Four gas head open furnace, and b temperature–time curve

2.3 Flexural performance test

This test was carried out according to ASTM C1609 (flexural performance of FRC) [17], which is the standard for measuring the toughness of a fiber reinforced composite material. Prior to the test, the dimensions and weight of the specimens were measured and recorded. Each specimen was put onto the support rig and loaded under a four-point bending load test as shown in Fig. 2. Two LVDTs were used to measure the deflection. Results from the load–deflection curves were used for calculating values such as the first-peak strength (f) (Eq. 1), residual strengths at deflection of L/600 \((f_{600}^{D} )\) and L/150 (\(f_{150}^{D}\)) (Eq. 2), area under the load–deflection curve from deflection of 0 to L/150 (\(T_{150}^{D}\)), and the equivalent flexural strength ratio (Eq. 3).
$${\text{First-peak strength}},\, f = \frac{PL}{{bd^{2} }}$$
$${\text{Residual}}\;{\text{strength}},\, f_{x}^{D} = \frac{{P_{x} \times L}}{{bd^{2} }}$$
$${\text{Equivalent}}\;{\text{flexural}}\;{\text{strength}}\;{\text{ratio}},\, R_{T,150}^{D} = \frac{{150.T_{150}^{D} }}{{f_{1} bd^{2} }}100\%$$
Fig. 2

Flexural performance test (ASTM C1609)

3 Experimental results

3.1 Temperature pattern and cross-section examination

The temperature patterns for each type of specimen at the top and bottom regions are plotted in Fig. 3. At both locations, regardless of specimen type, the temperature is found to increase with exposure time. The bottom surface which is located nearest to the gas heads exhibited the highest temperature up to about 275–325 °C at an exposure duration of 60 min. The top surface gained a maximum temperature up to about 145–165 °C at an exposure duration of 60 min.
Fig. 3

Temperature versus time along the beam’s thickness

After being subjected to fire, the specimens were cut in half along the crack line to examine the physical changes that occurred to the specimens. Results are shown in Figs. 4, 5 and 6.
Fig. 4

Cross-section of plain concrete specimen after being subjected to fire

Fig. 5

Cross-section of PFRC specimen after being subjected to fire

Fig. 6

Cross-section of SFRC specimen after being subjected to fire

Results for plain concrete specimens show major changes at the heated (bottom) region (Fig. 4). For example, the change in color; as the burning time increases, the grayish cement-like color begins to fade away and is replaced with a burning (brownish/blackish) shade. The burning shade can be seen clearly at the bottom portion close to the fire. Another observation is the increasing porosities or voids at the surface.

In the case of PFRC, beside the change in color, the disappearance (evaporation) of fibers at the heated zone is also observed (Fig. 5). In general, the fibers start to lose their mechanical properties from around 50 °C, and completely melt and begin to disappear around 160 °C. This causes the polypropylene fibers to disappear from the bottom surface first and this “fiber evaporation zone” tends to move up along the height of the beam with increasing exposure time.

For SFRC, both color change and increasing porosity are observed at the heated zone as shown in Fig. 6. Besides the color change of concrete, the change in color of steel fibers from metallic to a dark charcoal is also observed.

3.2 Flexural response

3.2.1 Plain concrete

The flexural responses of plain concrete beams are illustrated in Fig. 7. The flexural strength of the beam decreases with increasing burning time. Several mechanisms such as the non-uniform expansion caused by the differences in the thermal expansion coefficients of the paste and the aggregate which lead to internal stresses, internal cracking, bond strength degradation between aggregate and paste, stress caused by both external loads and thermal gradients, and the pore pressure due to water vaporization are involved in the strength degradation of concrete. For a specimen subjected to bending load, the bottom surface is a critical zone where tensile stress occurs. By adding damage from fire at the bottom surface (tension zone), this lead to a large drop in the load carrying capacity, as shown in Fig. 7. At a fire exposure duration of 15 min, the strength drops by approximately 54% (from 25 to 11.9 kN) and with an increase of fire exposure duration to 60 min, strength falls by approximately 77% (to about 5.9 kN).
Fig. 7

Flexural responses of plain concrete beams after being subjected to fire

3.2.2 SFRC

The effect of gradient heat affected the flexural response of SFRC in several aspects, e.g., the responses of the 1% SFRC beams (both pre- and post-peak) deteriorate with the increase of burning duration as seen in Fig. 8. The reduction of flexural performance is a combined effect of the degradation of concrete strength (as explained previously) which lowers the bond strength between the fiber and paste and the stress gradient caused by the thermic gradient which leads to internal cracking. Also, similar to the plain concrete, the degradation zone appeared to move upward with increasing exposure time.
Fig. 8

Flexural responses of 1% SFRC beams after being subjected to fire

Another important observation is the degradation of the pre-peak response of SFRC. In general, the flexural response of an FRC beam is divided into two parts: pre-peak and post-peak responses. The pre-peak response is controlled by the performance or the properties of concrete while the post-peak response is dominated by the performance of the fibers. These two responses are divided by a small drop of load right after the first crack as seen in the control beam (Fig. 8). At the beginning of loading, the load increases in direct proportion to the deflection and right after the first cracking, the load usually drops slightly before fibers begin to take over and the load begins to rise again.

After exposure to fire, the pre-peak responses degraded slowly with the exposure period. The degradations are in form of a decrease in stiffness (slope) and strength, and disappearance of the first cracking point. This degradation coincides with the obtained results on plain concrete and supported the assumption that internal cracking might have already occurred in concrete during the burning period of 15 min or longer.

3.2.3 PFRC

For the control specimen, the typical pre-peak response shows (1) an increase in flexural load linearly with increasing deflection, (2) a quick drop in load right after the first cracking, and (3) a slow increase in load after the fibers took over.

In the case of PFRC, the mechanism is more complex because polypropylene fibers melt when exposed to heat and begin to disappear at temperatures above 160 °C. When exposed to fire, the fibers closer to the heat source are either melted or disappear before the fibers farther away from the heat source. At short exposure durations, the temperature is not high enough; the fibers only melt and this causes the bond between fibers and paste to improve which leads to an increase in toughness. However, after longer exposure durations, the fibers begin to disappear. This leaves behind large voids and leads to a marked decrease in strength (Fig. 9).
Fig. 9

Flexural responses of 1% PFRC beams after being subjected to fire

In addition to a significant drop of the first peak load, the improvement of the post-peak response is observed in PFRC. The drop in the first peak load is mainly caused by the degradation of the concrete matrix, as mentioned earlier. However, for the post-peak response, the mechanism depended on the exposure duration and temperature level. With short exposure durations (i.e., 15 min), the temperature level for the whole cross-section is below 200 °C (no fiber evaporation). The melting of fibers somehow improves the bond between the fibers and concrete. This causes the post-peak response to improve compared with the control beam. However, as the exposure duration increases, more fibers disappear and the toughness begins to progressively decrease.

3.3 Flexural toughness

Toughness is defined as the area under the load–deflection curve of an FRC beam up to certain deflections. It represents the energy absorption ability of FRC under load. Results of the toughness at deflections of L/600 and L/150 are shown in Figs. 10 and 11, respectively. For SFRC, the toughness is highest in the control specimen (unheated), and decreases with increasing exposure duration. In contrast, PFRC exhibits increasing toughness when subjected to fire for a short period of time due to the increasing bond strength between fibers and the matrix. However, with longer exposure durations, more fibers evaporate and more voids are left in the specimen; subsequently, the toughness decreases.
Fig. 10

Toughness at a deflection of L/600

Fig. 11

Toughness at a deflection of L/150

3.4 Flexural strength and residual ratio

According to the ASTM C1609 standard, flexural strength (residual strength) up to a deflection of L/600 and L/150 can be calculated using Eqs. 2 and 4. The residual strength ratio is not a true stress but an engineering stress reflecting the behavior of FRC under static flexural loading and it indicates the ability of a specimen to carry load after first cracking up to a certain deflection.
$$f_{L/600}^{D} = \frac{{P_{L/600} L}}{{bd^{2} }}\;{\text{and}}\;f_{L/150}^{D} = \frac{{P_{L/150} L}}{{bd^{2} }}$$
where P, value of load at a deflection of L/600 and L/150; b, width of the beam; d, depth of the beam; L, clear span.
The residual strengths at deflections of L/600 and L/150 are shown in Fig. 12. For SFRC, the residual strengths decreased with increasing exposure time due to the degradation of matrix strength and interfacial bond strength between fibers and the matrix. For PFRC, residual strengths after short exposure durations increased slightly due to the increase in interfacial bond strength between fibers and the matrix. As the exposure durations increased, the bond became weak, the internal voids increased because of fiber evaporation and subsequently, the residual strengths began to decrease.
Fig. 12

Flexural (residual) strength at deflections of a L/600 and b L/150

The equivalent flexural strength ratio is the ratio between the toughness calculated up to a deflection of L/150 and the first peak strength (Eq. 1). It represents the equivalent flexural strength of the specimen at a deflection of L/150 as compared with the first peak strength, expressed as a percentage. It can be calculated using Eq. 3 and the results are presented in Fig. 13.
Fig. 13

Equivalent flexural strength ratio

The obtained results show that the equivalent flexural strength ratios for SFRC are quite consistent and remain more or less constant at around 85–90% of the first peak strength for all fire exposure durations. This indicates that the calculated flexural peak strength and the toughness are reduced in a similar proportion.

In the case of PFRC, the equivalent flexural strength ratio was found to decrease with increasing exposure duration. The 0.5% PFRC exhibited a quicker rate of decrease than the 1.0% PFRC. The decrease in the equivalent flexural strength ratio in the case of PFRC implied that the reduction rate of toughness is greater than the reduction rate of the first peak flexural strength. This is due to the effect of fiber evaporation and increasing void volume as mentioned earlier.

4 Conclusions

Fire causes changes in the physical and mechanical properties of plain concrete and FRC differently. Physically, a change in color is observed together with increasing voids or porosities. At the zones closest to the fire source, the color of the concrete turned from a grayish into a brownish shade. For SFRC, the change in color of steel fibers is observed. For PFRC, the evidence of fiber evaporation is observed in large voids left behind.

In case of the beams’ flexural response, the reduction of first crack strength after exposure to fire is commonly found in both plain concrete and FRC. However, for FRC, the effect on the post-peak response depends mainly on the fiber type and content. For SFRC, the reduction in post-peak response is due to the deterioration of matrix properties after being subjected to heat. In case of PFRC, the effect of fiber melting causes an improvement in the post-peak response after short exposure durations; however, the effect of fiber evaporation (at longer exposure durations) causes the post-peak response to drop significantly.

For flexural performance such as toughness and residual strength, the two types of FRC behaved differently. For PFRC, both toughness and residual strength increased slightly when the exposure time was shorter than 15 min and decreased thereafter. As for SFRC, gradual decreases in both toughness and residual strength were observed right from the beginning of fire exposure.

For the equivalent flexural strength ratio, two trends are observed. For SFRC, the equivalent flexural strength ratio remains almost constant during the exposure period, while for PFRC, it is found to gradually decrease. This implies that SFRC is more consistent in maintaining load carrying capacity when subjected to fire.

The test results here present the effect of a gradient heat on the flexural behavior of FRC. It must be noted here that in this kind of test, the results are also affected by the other factors such as specimen geometry and heating rate. Therefore, for the other test configurations, the results may be different.



This project is funded by King Mongkut’s University of Technology North Bangkok under contract no. KMUTNB-60-GOV-003 (Prof. Piti Sukontasukkul) and TRF Distinguished Research Professor Grant No. DPG6180002 (Prof. Prinya Chindaprasirt). The authors declare that they have no conflict of interest. The authors would like to thank SR. Fiber Co., Ltd. for providing steel and polypropylene fibers.


  1. 1.
    Georgali B, Tsakiridis PE (2005) Microstructure of fire-damaged concrete. A case study. Cem Concr Compos 27:255–259CrossRefGoogle Scholar
  2. 2.
    Guerrieri M, Sanjayan J, Collins F (2009) Residual compressive behavior of alkali-activated concrete exposed to elevated temperatures. Fire Mater 33(1):51–62CrossRefGoogle Scholar
  3. 3.
    Fernandes B, Gil AM, Bolina FL, Tutikian BF (2017) Microstructure of concrete subjected to elevated temperatures: physico-chemical changes and analysis techniques. Rev IBRACON Estrut Mater 10(4):838–863CrossRefGoogle Scholar
  4. 4.
    Handoo SK, Agarwal S, Agarwal SK (2002) Physicochemical, mineralogical, and morphological characteristics of concrete exposed to elevated temperatures. Cem Concr Res 32:1009–1018CrossRefGoogle Scholar
  5. 5.
    Powers-Couche L (1992) Fire damaged concrete-up close. Concr Repair Digest 1:241–248Google Scholar
  6. 6.
    Gustafero AH (1983) Experiences from evaluating fire-damaged concrete structures—fire safety of concrete structures. American Concrete Institute SP-80Google Scholar
  7. 7.
    Lau A, Anson M (2006) Effect of high temperatures on high performance steel fibre reinforced concrete. Cem Concr Res 36:1698–1707CrossRefGoogle Scholar
  8. 8.
    Poon CS, Shui Z, Lam L (2004) Compressive behavior of fiber reinforced high-performance concrete subjected to elevated temperatures. Cem Concr Res 34:2215–2222CrossRefGoogle Scholar
  9. 9.
    Gao D, Yan D, Li X (2014) Flexural properties after exposure to elevated temperatures of a ground granulated blast furnace slag concrete incorporating steel fibers and polypropylene fibers. Fire Mater 38:576–587CrossRefGoogle Scholar
  10. 10.
    Zheng W, Li H, Wang Y (2012) Compressive stress–strain relationship of steel fiber-reinforced reactive powder concrete after exposure to elevated temperatures. Constr Build Mater 35:931–940CrossRefGoogle Scholar
  11. 11.
    Morsy NMA-E-F, El Kady HMG, Mokhtar ASA, El Nawawy OAM (2016) Enhancement of fire resistance of reinforced concrete beams using steel fibers. ARPN J Eng Appl Sci 11(19):11782–11789Google Scholar
  12. 12.
    Lee G, Han D, Han MC, Han CG, Son HJ (2012) Combining polypropylene and nylon fibers to optimize fiber addition for spalling protection of high-strength concrete. Constr Build Mater 34:313–320CrossRefGoogle Scholar
  13. 13.
    Bashir J, Singh K (2017) Experimental inquest for improving the fire resistance of concrete by the addition of polypropylene fibers. Int J Civ Eng Technol 8(8):129–139Google Scholar
  14. 14.
    Sukontasukkul P, Pomchiengpin W (2010) Post-crack (or post-peak) flexural response and toughness of fiber reinforced concrete after exposure to high temperature. Constr Build Mater JCBM 24:1967–1974CrossRefGoogle Scholar
  15. 15.
    Li Q, Gao X, Xu S, Peng Y, Fu Y (2016) Microstructure and mechanical properties of high-toughness fiber-reinforced cementitious composites after exposure to elevated temperatures. J Mater Civ Eng. Google Scholar
  16. 16.
    Huang Z (2010) The behaviour of reinforced concrete slabs in fire. Fire Saf J 45(5):271–282CrossRefGoogle Scholar
  17. 17.
    Allam SM, Elbakry HMF, Rabeai AG (2013) Behavior of one-way reinforced concrete slabs subjected to fire. Alex Eng J 52(4):749–761CrossRefGoogle Scholar
  18. 18.
    Lo Monte F, Felicetti R (2017) Heated slabs under biaxial compressive loading: a test set-up for the assessment of concrete sensitivity to spalling. Mater Struct. Google Scholar

Copyright information

© RILEM 2018

Authors and Affiliations

  1. 1.Construction and Building Material Research CenterKing Mongkut’s University of Technology North BangkokBangkokThailand
  2. 2.Department of Civil and Environmental EngineeringNational Defense Academy of JapanYokosukaJapan
  3. 3.Sustainable Infrastructure Research and Development Center, Department of Civil EngineeringKhon Kaen UniversityKhon KaenThailand

Personalised recommendations