Construction and Building Materials 78 (2015) 26–33

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Effect of coarse aggregate type and loading level on the high temperatureproperties of concrete

http://dx.doi.org/10.1016/j.conbuildmat.2014.12.0960950-0618/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: gyuyongkim@cnu.ac.kr (G. Kim).

Minho Yoon a, Gyuyong Kim a,⇑, Gyeong Choel Choe a, Youngwook Lee a, Taegyu Lee b

a Department of Architectural Engineering, Chungnam National University, 99 Daehak-ro, Yuseong-gu, Daejeon 305-764, Republic of Koreab Department of Construction Technology, DSME Construction Co., Ltd., ELCRU Building, 636 Nonhyun-ro, Gangnam-gu, Seoul 135-010, Republic of Korea

h i g h l i g h t s

� The properties of concrete at high temperature were evaluated by aggregate types.� The thermal properties of concrete are mainly affected by that of aggregate.� LWC indicates better thermal properties than NWC.

a r t i c l e i n f o

Article history:Received 4 May 2014Received in revised form 20 November 2014Accepted 27 December 2014Available online 14 January 2015

Keywords:Artificial lightweight aggregateAggregate densityResidual compressive strengthThermal expansionTotal strainSteady-state creep at high temperature

a b s t r a c t

When concrete is exposed to temperature changes, its durability is reduced because of the decompositionof cement metrics generation of cracks within its structure as its component materials undergo differentvolumetric changes. Coarse aggregates play an important role in such behavior of concrete. We thus eval-uated the influence of coarse aggregates on the fire resistance performance of a concrete structure byconducting a fire experiment under loading on two types of concrete, one with a granite-based coarseaggregate (NWC: normal weight concrete) and the other consisting of a clay-ash lightweight aggregate(LWC: lightweight concrete). LWC displayed a higher residual compressive strength than NWC underthermal load condition. NWC suffered from a large number of cracks at its interior at high temperatures,while the interior of the LWC demonstrated fewer cracks because of the voids in its interior to the mit-igation of thermal expansion stress. When a load equivalent to 20% of its room temperature compressivestrength was applied, both NWC and LWC demonstrated quasi-equilibrium between the thermal expan-sion strain and the loading-induced shrinkage strain. Whereas the 40% loading condition, the specimenexhibited shrinkage strain and its compressive strength was observed to undergo a sharp decrease from500 �C.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Concrete is a mixture of different kinds of materials such ascement pastes and aggregates that have different thermalexpansion coefficients. Hence, when the temperature of concreteis elevated, not only cement hydration products are decomposedbut crack occurs inside of concrete due to different volume changeof component materials consist of concrete, Thus reducing con-crete durability [1–5]. This thermal expansion of concrete is greatlyaffected by coarse aggregates, which account for most of thevolume (see Fig. 1). Studies on the high-temperature propertiesof concrete with various kinds of coarse aggregates have beenconducted by many researchers, especially on those employing

artificial light-weight aggregates that have a small thermal expan-sion coefficient.

To examine the relationship between the concrete microstruc-ture and its strength after exposure to fire, Turker et al. [6] ana-lyzed different concrete microstructures employing three kindsof coarse aggregates, i.e., granite, limestone and pumice stone, afterheating the concrete for four hours at the temperatures of 100, 250,500, 700, and 850 �C. He reported that the concrete using pumicestone, which showed cracks in the aggregates at high temperaturesrather than at the interfaces between the aggregates and pastes,had better interfacial conditions as compared to the concretesusing either of the other two kinds of aggregates. Furthermore,Neville [7] reported that the strength of the concrete using nor-mal-weight granite aggregates, unlike the light-weight aggregateconcrete, begins abruptly decreasing at a temperature higher thanthe approximate value of 430 �C. It decreases by about 50% at

-0.006

-0.003

0.000

0.003

0.006

0.009

0.012

0.015

0.018

0 200 400 600 800Temperature (°C)

Stra

in (m

m / mm)

Coarse AggregateConcreteCement Paste

Fig. 1. Thermal expansion of concrete constituent material.

Fig. 2. Schematic diagram of the moment on structure during a fire.

Table 1Experimental plan.

ID Aggregatetype

Pre-loadinglevel (�fcu)

Target temp.(�C)

Evaluation items

NWC Granite 0.00.20.4

20, 100, 200,300, 500, 700

� Stress–strainrelation� High

temperature� Compressive

strength� Thermal

expansion� Total strain� High tempera-

ture creep

LWC Clay-ash

Table 2Concrete mixing proportion.

ID W/B(%)

fck

(MPa)Air(%)

S/a(%)

Unit weight (kg/m3)

Wa Ca SFa Sa Ga

NWC 35 63 4 ± 2 40 165 470 – 692 1071LWC 33 59 4 ± 2 40 155 432 38 687 676

a W: water; C: cement; SF: silica fume; S: sand; G: gravel.

M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33 27

600 �C from the compressive strength at room temperature, and byabout 80% above 800 �C, a temperature at which the structure ofthe concrete can collapse. Kong et al. [8] and Abeles and Bard-han-Roy [9] reported that the strength of the light-weight aggre-gate concrete was maintained up to about 500 �C, and thendecreased by about 60% as the temperature increased from500 �C to 800 �C. Uygunoglu et al. [10] evaluated coefficient ofthermal expansion of concrete with limestone and pumice stoneheating up to 1000 �C. Also he reported that since concrete withporous pumice stone has low coefficient of thermal expansion,when it comes to applying to a structural member subjected to ele-vated temperature, spalling and risk of collapse of structure can bereduced. Among others, Al-Sibahy [11], Tanyildizi [12,13],Abdulkareem [14,15] and Sancak [16] examined high temperatureproperties of concrete with LWC.

However, existing studies on the high-temperature propertiesof this light-weight aggregate concrete have mostly focused onthe basic mechanical properties such as the high-temperaturecompressive strength. However, few studies considering designload applied to structural member have been examined. Mean-while, many situations have been reported where vertical mem-bers such as pillars suffered shear failure under a large load, dueto the thermal expansion of horizontal members such as beamsor slabs, as shown in Fig. 2 [17]. Also, creep strain can occur in con-crete members, exposed to a fire for about 120–180 min, of compa-rable magnitude to the creep strain of concrete kept at roomtemperature for about 20–30 years. As shown in these cases, forevaluation of the fire resistance performance of the concrete struc-tures, the strain properties that are introduced in the event of a fireas well as the basic mechanical properties should be consideredwith design load applied to structural member.

Accordingly, in this study, the effects of the different kinds ofcoarse aggregates on the fire resistance performance of concretestructures were evaluated by fire resistance tests under load byappraising the strain characteristics as well as the basic mechanicalproperties of concrete with either granite or clay-ash artificiallight-weight coarse aggregates.

2. Experiment

2.1. Experimental plan and concrete mixing

The experimental plan is described in Table 1, while Table 2 shows the concretemixing proportion. The water-binder ratio (W/B) was set at 35% for the normal-weight aggregate concrete and 33% for the light-weight aggregate concrete, andthe standard design compressive strength was 60 MPa. The compressive strengthat room temperature of NWC and LWC was 68, 69 MPa, respectively.

Loading conditions were set at 20% and 40% of the compressive strength atroom temperature, as well as considering the non-loading condition. The targetheating temperatures were the room temperature (20 �C), 100, 200, 300, 500, and700 �C. At the respective target temperatures, the stress–strain relationship andhigh-temperature compressive strength were measured. The thermal expansionstrain, which occurs during heating to the target temperature, and the steady-statecreep strain at high temperatures, which occurs when the temperature is main-tained at a fixed value, were also measured.

2.2. Materials

The physical properties of the materials and chemical composition ofcoarse aggregates used in this study are described in Tables 3 and 4, respectively.For the normal-weight aggregates, crushed granite gravel was used of up to20 mm in size, 2.65 g/cm3 in density, and water absorption ratio of 0.8%. On theother hand, artificial clay-ash type light-weight aggregates, which were added tothe coal-ash to improve features such as the water absorption ratio, were used,up to 13 mm in size, 1.68 g/cm3 in density, and water absorption ratio of 15.3%[18,19].

The thermal expansion coefficient, which has a great influence on the thermalproperties of concrete, of the materials used in this study are described in Table 5.The thermal expansion coefficient of cement paste started to decrease at around300 �C. However, the thermal expansion coefficient of coarse aggregates have beenincreased with increasing temperature, and especially thermal expansion coeffi-cient of granite aggregate was greater than that of artificial light-weight aggregate.

The cross-sectional shape of the coarse aggregates is shown in Table 6. It wasverified that the artificial light-weight aggregates had many pores inside, formedduring the manufacturing process. Furthermore, it was confirmed by observingthe interface between the aggregates and cement matrices that the cement pastepermeated the pores on the surface of the artificial light-weight aggregates.

Table 3Physical properties of used materials.

Materials Physical properties

Cement Ordinary Portland cementDensity: 3.15 g/cm3, Specific surface area: 3630 cm2/g

Fine aggregate Washed sandDensity: 2.64 g/cm3, Water absorption ratio: 1.03%

Coarse aggregate Normal Crushed graniteMax size: 20 mm, density: 3.15 g/cm3, water absorption ratio: 0.97%

Light weight Clay-ash type artificial lightweight aggregateMax size: 13 mm, density: 1.68 g/cm3, water absorption ratio: 15.27%

Silica fume Density: 2.23 g/cm3, specific surface area: 200,000 cm2/gAdmixture Polycarboxylic water reducing agent

Table 4Chemical composition of used coarse aggregate.

Aggregate type Chemical composition (%)

SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O

Normal (granite) 72.0 15.5 1.8 2.3 0.1 4.42 3.10Lightweight (clay-ash) 72.8 13.6 3.1 1.5 0.7 1.5 1.2

Table 5Thermal expansion coefficient of used materials (�10�6/�C).

Aggregate Type Temperature (�C)

20 100 200 300 400 500 600 700

Cement Paste 4.4 3.8 2.8 1.8 1.1 �1.0 �3.8 �4.9Normal (granite) 6.1 4.8 7.4 9.8 12.2 14.5 21.1 21.2Lightweight (clay-

ash)4.0 4.0 3.2 3.5 3.9 4.3 5.2 4.9

28 M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33

2.3. Experimental method

2.3.1. Preparation of specimenFor evaluating mechanical properties at a high temperature, ø100 � 200 mm

specimens have been manufactured. The specimens were cured under water forseven days, then air dry curing was conducted up to 180 days at a steady temper-ature and humidity chamber set as 20 ± 2 �C, R.H. 50 ± 5%. Before heating test, theupper and lower surfaces of the concrete were smoothly finished with a grinderand then a heating test was conducted.

2.3.2. Heating apparatus and heating methodThe experimental apparatus used in this study is shown in Fig. 3. For simulta-

neous loading and heating, an electric heating furnace was installed loading appa-ratus with a capacity of 2000 kN. Furthermore, the strain of the test specimensduring heating was measured by transferring the strain to the displacement meterplaced outside through a quartz tube of ø100 mm in diameter, which penetratedthe center of the upper and lower loading jigs.

To increase the temperature inside and outside of the test specimens to thesame level, an indirect heating method was used in which heat was transferredto the test specimens by heating the upper and lower loading jigs. Furthermore,the heating rate was set at 1 �C/min as shown in Fig. 4, and in particular, in the tem-perature range up to 50 �C at the beginning of heating and before reaching the tar-get temperature, the heating rate was set at 0.77 �C/min to maintain thetemperature difference within 5 �C between the inside and outside of the test spec-imens [20].

2.3.3. Strain evaluation methodThe schematic diagram of concrete showing strain behavior during heating and

loading is presented in Fig. 5. Specimens were heated up to targeted temperature intemperature elevating section. And temperature was maintained at the targetedtemperature in temperature maintain section. During all testing period, loadingwas 0%, 20%, 40% of compressive strength at room temperature of specimen. In tem-perature elevating section, thermal expansion and total strain of specimens weremeasured. In temperature maintain section, the steady-state creep strain at high

temperature was measured for 300 min after the target temperature was com-pleted, under condition of load of 20% or 40% of the compressive strength at roomtemperature [20].

3. Results and analysis

3.1. Stress–strain curve

Fig. 6 shows the effects of the initial loading and the differentkinds of coarse aggregates on the stress–strain relationship, atthe heating temperatures of 500 �C and 700 �C, at which the fea-tures are relatively clearly observed. The respective high-tempera-ture compressive strengths of normal-weight aggregate concrete(NWC) and light-weight aggregate concrete (LWC) were 43 MPaand 52 MPa, and the respective strain at maximum stress were7.0 � 10�3 and 5.0 � 10�3 at the heating temperature of 500 �C ina non-loading condition. Consequently, LWC is thought to have asmaller rate of reduction of high-temperature compressivestrength and strain at maximum stress than NWC.

In the case of initial load of 20% or 40% of the compressivestrength at room temperature, the high-temperature compressivestrength increased more than for the case of non-loading, andthe strain at maximum stress was restrained within the range of3.7 � 10�3–4.0 � 10�3, irrespective of the kinds of coarse aggre-gates used. It is considered that the thermal expansion strainwas offset by the shrinkage strain under loading condition shownas Fig. 9 in Chapter 3.3 [21].

In particular, even at loading of 40% of the compressive strengthat room temperature, LWC showed a high-temperature compres-sive strength equal to about 90% of the compressive strength atroom temperature. NWC exhibited a relatively greater decreaseof the high-temperature compressive strength, decreasing to 10%of the compressive strength at room temperature, particularly at700 �C.

It is known that the compressive strength degradation is causedby micro internal crack from different thermal expansion behaviorof materials and decomposition of cement hydrates [4,5]. Com-pared to its general aggregate-based counterpart, concrete thatuses lightweight aggregates, which have low coefficients of ther-mal expansion, exhibits a smaller decrement in high temperaturecompressive strength. This smaller decrease results from a soundinternal structure in which fewer internal cracks form during heat-ing; therefore, lightweight aggregate-based concrete exhibitssmaller strains under maximum load conditions.

3.2. High-temperature compressive strength

Fig. 7 shows the relationship between the residual compressivestrength and the heating temperature of concrete for different kindsof coarse aggregates. For NWC, while the residual compressive

Table 6Cross-sectional shape of used coarse aggregate.

Aggregate type

Cross-sectional shape

Aggregate Concrete

Normal aggregate (Granite)

Artificial lightweight aggregate(Clay-ash)

Table 7Cross-sectional shape of concrete after heating (700 �C).

ID. NWC LWC

Cross-sectional shape

M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33 29

strength decreased to 65% at 100 �C, it increased to 90% at 300 �C. Asthe heating temperature increased to higher than 300 �C, the com-pressive strength decreased, displaying a residual compressivestrength of about 27% at 700 �C. On the other hand, although LWCdemonstrated a similar tendency as NWC up to 300 �C, it displayeda residual compressive strength of 80% on average at 700 �C, whichis higher than that of NWC.

Fig. 8 shows the relationship between the residual compressivestrength ratio and the density of the coarse aggregates at 700 �C.The results indicate that as the density of the coarse aggregatesincreased, the residual compressive strength also decreased,thereby confirming the correlation between the residual compres-sive strength and the density of aggregates within the scope of thisstudy. In the case of pre-loading condition, LWC presented a high

(a) Heating and loading apparatus

(b) Geometry of apparatus

UUpperloading jig

Lowerloading jig

Quartzpipe

Upper LVDT

Lower LVDT

Electricfurnace

Fig. 3. Experimental apparatus used in this study.

Time (min)

Tem

pera

ture

()

30min maintain

0.77 /min

1 /min

30min maintain

0.77 /min

60min maintain

Target temperature

Fig. 4. Heating curve used in the experiment.

Fig. 5. Evaluation method of the strain properties of concrete.

30 M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33

residual compressive strength under the influence of the shrinkagestrain caused by loading. However, NWC, with loading of 20%,showed a residual compressive strength of about 40% of the com-pressive strength at room temperature, which is higher by 10%than for the non-loading condition.

However, at loading of 40%, it exhibited residual compressivestrength as low as about 10% of the compressive strength at roomtemperature. The degradation of strength owing to the thermaldecomposition of cement hydrates resulted from the similaritybetween the two specimens. This similarity, in turn, led to a greaterresidual ratio of the compressive strength under load conditions;therefore, this study focused on the occurrence of cracks ratherthan thermal decomposition.

Table 7 shows the shapes of the aggregates at the interfaces inNWC and LWC after heating at 700 �C. NWC displayed many cracksbetween the aggregates and cement matrices, whereas LWCshowed none. Consequently, it is considered that because LWChas fewer cracks at the interface of the aggregates and in the inter-facial transition zone (ITZ), which are known to be a major cause ofthe decrease in compressive strength of concrete at high tempera-tures, as compared to NWC, it has higher residual compressivestrength [22].

3.3. Thermal expansion strain and total strain

Fig. 9 shows the relationship between the thermal expansionstrain and the heating temperature of concrete for different kindsof coarse aggregates. In the case of non-loading condition, as thetemperature increases, NWC shows a great increase in the thermalexpansion strain, and in particular in the temperature rangebetween 500 and 600 �C, it displays a sharp increase. LWC, inwhich the density of coarse aggregates is low, demonstrated asmaller thermal expansion strain than NWC.

Furthermore, in the case of loading of 20% of the compressivestrength at room temperature, the concrete test specimens showedthe smallest strain as the thermal expansion strain is restrained bythe shrinkage stress due to loading, whereas in the case of loadingof 40%, they showed an abrupt shrinkage strain at temperatures

Fig. 6. Stress–strain relation according to the coarse aggregate type.

Fig. 7. Residual ratio of high temperature compressive strength according to thecoarse aggregate type.

Fig. 8. Residual ratio of high temperature compressive strength according to thecoarse aggregate density at 700 �C.

-0.009

-0.006

-0.003

0.000

0.003

0.006

0.009

0.012

0.015

0 200 400 600 800 1000Heating Temperature (°C)

Stra

in (m

m / mm)

Euro_CalcareousEuro_SiliceousNWCLWC Non-

loading

0.2 fcu

loading

0.4 fcu

loading

Fig. 9. Thermal expansion and total strain by coarse aggregate type.

M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33 31

higher than 500 �C because the stress due to loading becomesgreater than that caused by the thermal expansion strain.

Fig. 10 shows the relationship between the values of the strainand loading level at 700 �C. For the non-loading condition, the ther-mal expansion strain of NWC, with high density of coarse aggre-gates, is 1.1 � 10�2, and that for LWC, with low density, is5.0 � 10�3. Therefore, the strain amount of NWC is more thanabout twice that of LWC. In the case of loading of 20% of the com-pressive strength at room temperature, because the thermalexpansion strain is more restrained by the presence of loading thanfor non-loading, irrespective of the different kinds of coarse aggre-gates, the strain is restricted to within the range of �3.0 � 10�3–1.0 � 10�3.

In the case of loading of 40%, because the strain tends to con-strict to the strain at maximum stress at 700 �C, i.e., �8.0 � 10�3,the proof stress is considered to almost disappear because of theloading stress and high temperature. Furthermore, for the casenon-loading condition, although a difference in the thermal expan-sion strain occurs according to the density of the coarse aggregates,the difference decreases as the loading amount is increased. Conse-quently, it is believed that as the loading amount increases, theeffect of the density of the coarse aggregates on the thermal expan-sion strain decreases.

Fig. 10. Relation between thermal strain and loading level at 700 �C.

-0.012

-0.010

-0.008

-0.006

-0.004

-0.002

0.0000.000 0.002 0.004 0.006 0.008 0.010 0.012

Thermal Expansion Strain (mm/mm)

Cre

ep S

train

(mm / m

m)

NWC

LWC

NWC 0.4 fcu

εcr/εth : 0.2~0.7

LWC 0.4 fcu

εcr/εth : 0.4~1.4

NWC 0.2 fcu εcr/εth : 0.1~0.2

LWC 0.2 fcu

εcr/εth : 0.3~0.7

Fig. 12. Relation between steady state creep strain at high temperature andthermal expansion by coarse aggregate type.

32 M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33

3.4. Steady-state creep at high temperature

Fig. 11 shows the relationship between the loading amount andsteady-state creep of NWC and LWC at high temperatures. ForNWC, in the case of loading of 20% of the compressive strengthat room temperature, a similar strain of about �3.5 � 10�4 in thetemperature range between 100 and 300 �C is exhibited, whereasit shows a strain of �1.4 � 10�3, which is 4.2 times greater, at500 �C and 700 �C. LWC shows a similar tendency to NWC until500 �C, whereas it displays a strain amount of �4.1 � 10�3 at700 �C, which is 2.9 times greater than for NWC.

In the case of loading at 40% of the compressive strength atroom temperature, the strain was similar to or a little higher thanfor the case with loading of 20% in the temperature range between

(a) NWC

Fig. 11. Steady state creep strain at high t

100 and 300 �C, irrespective of the kinds of coarse aggregates,whereas the strain greatly increased at temperatures higher than500 �C. The respective strains of NWC and LWC were �4.8 � 10�3

and �2.1 � 10�3 at 500 �C, demonstrating that the strain of NWCis about 2.3 times larger than that of LWC. Furthermore, the strainof NWC could not be measured as the NWC structure collapsed atabout 560 �C during heating to 700 �C, whereas LWC displayed astrain of �8.6 � 10�3. Studies on creep at room temperature [23]have reported that LWC, with low density of coarse aggregates,shows a larger creep strain at room temperature than NWC.

However, the present study contends that because in the case ofloading at 40% of the compressive strength at room temperature,

(b) LWC

emperature by coarse aggregate type.

M. Yoon et al. / Construction and Building Materials 78 (2015) 26–33 33

the proof stress of NWC sharply decreases at the high temperaturesof 500 �C and 700 �C, the steady-state creep strain at high temper-atures is larger than that of LWC.

Fig. 12 shows the relationship between the thermal expansionstrain and the steady-state creep strain at high temperatures fordifferent kinds of coarse aggregates. NWC, in the case of loadingat 20% of the compressive strength at room temperature, demon-strates a smaller steady-state creep strain at high temperaturesthan the thermal expansion strain. It presents a ratio of steady-state creep strain at high temperatures to thermal expansion strain(ecr/eth) of 0.1–0.2. The ecr/eth of LWC was in the range of 0.3–0.7,showing that it tends to have a larger high-temperature creepstrain than NWC at the same level of thermal expansion strain.

However, in the case of loading at 40% of the compressivestrength at room temperature, the ecr/eth of NWC was 0.7, greaterthan the 0.6 of LWC, at 500 �C; on the other hand, the ecr/eth ofLWC was 1.4, but that of NWC could not be measured at 700 �C,as it had collapsed before its creep strain could be measured.

4. Conclusions

1) Concrete that consists of lightweight aggregates of smallerdensity and thermal expansion coefficient showed higherresidual compressive strength at high temperatures thanits general aggregate-based counterpart owing to the forma-tion of internal pores during the manufacturing process. Theconcrete that used general aggregates, of high thermalexpansion coefficient, exhibited many cracks in the aggre-gate interface and ITZ at temperatures greater than 300 �C,and, as such, its compressive strength was degraded.

2) The thermal strain of concrete can be explained as thermalexpansion strain under non-loading conditions, and asshrinkage owing to confinement and strength degradationunder load conditions. Under loading of 20%, the room tem-perature compressive strength, the shrinkage strain owingto the load, almost achieves equilibrium with the thermalexpansion strain; however, a rapid shrinking phenomenonwas observed under 40% loading condition.

3) In the scope of the present study, NWC that was subjectedwith 20% compressive strength at room temperature hadhigh density aggregates since its thermal expansion strainwas large and the high temperature creep strain (shrinkagestrain) was small; however, LWC showed higher high tem-perature creep strain despite the relatively high residualratio of the modulus of elasticity at high temperatures. Nev-ertheless, NWC showed greater high temperature creepstrain than LWC owing to the sharp degradation in the inter-nal strength of NWC at 40% load and temperatures greaterthan 500 �C. Therefore, the compressive strength at hightemperatures should be considered in the evaluation of hightemperature creep strain.

4) This study validated the excellent performance of concretewhich consists of clay-ash-based artificial lightweight aggre-gates by comparing and evaluating its high temperatureproperties with those of the concrete that contains granite-based aggregates. Therefore, using lightweight aggregate-based concrete in structures such as a slab should improvethe safety of the structure during fires.

Acknowledgement

This research was financially supported by the Ministry of Edu-cation, Science Technology (MEST) and National Research Founda-tion of South Korea (NRF) through the Human Resource TrainingProject for Regional Innovation.

References

[1] Schneider, U. Behaviour of concrete at high temperatures, RILEM Committee44-PHT Draft Paper 1, 1982.

[2] Poon CS, Shui ZH, Lam L. Compressive behavior of fiber reinforced high-performance concrete subjected to elevated temperatures. Cem Concr Res2004;34(12):2215–22.

[3] Kalifa P, Chéné G, Gallé C. High-temperature behaviour of HPC withpolypropylene fibres: from spalling to microstructure. Cem Concr Res2001;31(10):1487–99.

[4] Arioz O. Effects of elevated temperatures on properties of concrete. Fire Saf J2007;42(8):516–22.

[5] Zega CJ, Di Maio AA. Recycled concrete made with different natural coarseaggregates exposed to high temperature. Constr Build Mater2009;23(5):2047–52.

[6] Turker P, Erdogdu K, Erdogan B. Investigation of the various type of aggregatemortar exposed to fire. J Cem Concr World 2001;6(31):52–69.

[7] Neville AM. Properties of Concrete. 4th ed. New Jersey, USA: Prentice Hall;1995.

[8] Kong FK, Evans RH, Cohen E, Roll F. Handbook of StructuralConcrete. London: Pitman Books Limited; 1983.

[9] Abeles PW, Bardhan-Roy BK. Prestressed Concrete Designer’sHandbook. London: New Fetter Lane; 1981.

[10] Uygunoglu T, Topçu _IB. Thermal expansion of self-consolidating normal andlightweight aggregate concrete at elevated temperature. Constr Build Mater2009;23(9):3063–9.

[11] Al-Sibahy A, Edwards R. Thermal behaviour of novel lightweight concrete atambient and elevated temperatures: experimental, modelling and parametricstudies. Constr Build Mater 2012;31:174–87.

[12] Tanyildizi H, Coskun A. The effect of high temperature on compressivestrength and splitting tensile strength of structural lightweight concretecontaining fly ash. Constr Build Mater 2008;22(11):2269–75.

[13] Tanyildizi H, Coskun A. Performance of lightweight concrete with silica fumeafter high temperature. Constr Build Mater 2008;22(10):2124–9.

[14] Abdulkareem O, Abdullah M, Hussin K, Ismail K, Binhussain M. Mechanical andmicrostructural evaluations of lightweight aggregate geopolymer concretebefore and after exposed to elevated temperatures. Materials2013;6(10):4450–61.

[15] Abdulkareem OA, Mustafa Al Bakri AM, Kamarudin H, Khairul Nizar I, Saif AeA.Effects of elevated temperatures on the thermal behavior and mechanicalperformance of fly ash geopolymer paste, mortar and lightweight concrete.Constr Build Mater 2014;50:377–87.

[16] Sancak E, Dursun Sari Y, Simsek O. Effects of elevated temperature oncompressive strength and weight loss of the light-weight concrete with silicafume and superplasticizer. Cement Concr Compos 2008;30(8):715–21.

[17] Iguchi S, Okuwaki K, Shinozuka W, Hirashima T. Fire resistance of mixedstructural composite of R/C columns and steel beams (Part 11: Shear force inR/C columns subjected to action caused by thermal elongation of beams infire). Summaries of Technical Papers of Annual Meeting of ArchitecturalInstitute of Japan 2012:213–214.

[18] González-Corrochano B, Alonso-Azcárate J, Rodas M. Effect of prefiring andfiring dwell times on the properties of artificial lightweight aggregates. ConstrBuild Mater 2014;53:91–101.

[19] Bernhardt M, Justnes H, Tellesbø H, Wiik K. The effect of additives on theproperties of lightweight aggregates produced from clay. Cement ConcrCompos 2014;53:233–8.

[20] RILEM TC 129-MHT. Test methods for mechanical properties of concrete athigh temperatures: Part 8: Steady-state creep and creep recovery for serviceand accident conditions. Mater Struct 2000;33:6–13.

[21] Hertz KD. Concrete strength for fire safety design. Mag Concr Res2005;57(8):445–53.

[22] Andiç-Çakır Ö, Hızal S. Influence of elevated temperatures on the mechanicalproperties and microstructure of self consolidating lightweight aggregateconcrete. Constr Build Mater 2012;34:575–83.

[23] Wu B, Siu-Shu Lam E, Liu Q, Yuk-ming Chung W, Fung-yuen Ho I. Creepbehavior of high-strength concrete with polypropylene fibers at elevatedtemperatures. ACI Mater 2010;107(2):176–84.