Available online at www.sciencedirect.

com
Construction
and Building

Construction and Building Materials 22 (2008) 948–955

MATERIALS
www.elsevier.com/locate/conbuildmat

Effect of thermal shock due to rapid cooling on residual

mechanical properties of fiber concrete exposed to high temperatures
a,*
Gai-Fei Peng , Song-Hua Bian a, Zhan-Qi Guo b, Jie Zhao a, Xin-Lai Peng a,
Yu-Chuang Jiang a
a
Faculty of Civil Engineering and Architecture, Beijing JiaoTong University, Beijing 100044, PR China
b
Faculty of Civil Engineering and Geosciences, Delft University of Technology, 2628 CN Delft, The Netherlands

Received 5 October 2006; received in revised form 22 November 2006; accepted 1 December 2006
Available online 18 January 2007

Abstract

An experimental investigation was conducted on the effect of thermal shock during cooling on residual mechanical properties of fiber
concrete exposed to elevated temperatures from 200 to 800 °C. Various cooling regimes were used including natural cooling, spraying
water for a series of durations from 5 to 60 min, and quenching in water. The temperature determination results prove that the rapid
cooling regimes such as quenching in water, or water spraying for 30 min or more, caused an action of ‘‘thermal shock’’ to concrete under
elevated temperature, characterized by a high temperature decreasing rate ranged from 25 to 44 min/°C. The experimental results indi-
cate that, compared with natural cooling, thermal shock induced by water quenching and spraying water caused more severe damage to
concrete, in terms of greater losses in compressive strength, tensile splitting strength, and fracture energy. The fact that the impact of
spraying water for 30 min or more on mechanical properties was almost the same as that of water quenching, indicates that spraying
water for 30 min or more could cause thermal shock to a similar degree to water quenching, which is in good agreement with the results
of the temperature determination. Hybrid fiber (steel fiber and polypropylene fiber) can enhance residual strength and fracture energy of
concrete subjected to thermal shock induced by rapid cooling from high temperatures up to 800 °C to room temperature.
Ó 2006 Elsevier Ltd. All rights reserved.

Keywords: High temperature; Thermal shock; Mechanical properties; Fiber reinforcement; High-performance concrete; Fire; Rapid cooling

1. Introduction point, using polypropylene fiber (PP fiber) becomes a com-

As a considerable shortcoming of high-strength con-
crete/high-performance concrete (HSC/HPC), thermally
induced explosive spalling has attracted lots of research
attention during the past two decades [1,2]. It has been
proved that, among a couple of factors influencing occur-
rence of explosive spalling of HSC/HPC, internal vapor
pressure which is associated with both internal moisture
content and dense microstructure of concrete, is a main
factor governing spalling occurrence [3,4]. In light of this
*
Corresponding author. Tel./fax: +86 10 5168 2252.
E-mail address: gfpeng@center.njtu.edu.cn (G.-F. Peng).
mon approach for alleviating or even avoiding explosive
spalling of HSC/HPC [5,6].
In general, if concrete has been well prepared for inhib-
iting explosive spalling, the main damage to concrete
caused by fire or elevated temperatures should be loss in
mechanical properties. Experimental results [7,8] confirmed
that compressive strength can be broadly maintained
within a range of temperature from 20 to 400 °C. Consid-
erable loss in compressive strength occurs between 400
and 600 °C, and most of the original compressive strength
before heating may be lost from 600 to 800 °C. Compared
with compressive strength, tensile splitting strength suffers
a more severe loss under identical temperature, as the latter
is more sensitive to thermally induced cracking [7].

0950-0618/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2006.12.002
 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955 949

Nevertheless, most research data of residual strength after
exposure to high temperature were obtained under condi-
tions of natural cooling [1,2,7,8], which should differ obvi-
ously from cooling regimes in a real fire, where water
spraying is usually used for fire extinguishing and conse-
quently thermal shock is induced to concrete. It has been
reported that water cooling caused more severe decrease
in strength compared to natural cooling [9–11]. Therefore,
the effect of cooling regimes on mechanical properties of
concrete is of great concern, especially after a fire case
was reported [12,13], which happened in the city of Hengy-
ang, Hunan Province, China, November 3rd, 2003. In this
case, an 8-storey reinforced concrete building collapsed cat-
astrophically during fire extinguishing and twenty fire fight-
ers died from the building collapse. Furthermore, apart
from these reports on the effect of cooling on plain concrete
with no fiber [9–13], there is little literature on properties of
fiber concrete subjected to various cooling regimes.
As it has been found in a preceding investigation by the
authors [14] that using hybrid fiber (steel fiber and polypro-
pylene fiber) is an optimum approach for enhancing fire
resistance of HSC/HPC, this paper presents an investiga-
tion on the effect of cooling regimes, including natural
cooling, spraying water for a series of durations from 5
to 60 min, and quenching in water, on residual mechanical
properties of high-performance concrete incorporating
hybrid fiber after exposure to high temperatures from 200
to 800 °C.
2. Experimental details
2.1. Raw materials
Cement: ordinary Portland cement, strength grade
42.5 MPa (which was compressive strength at 28 days,
determined on mortar specimens at a mix proportion
of cement:sand:water = 1:3:0.5 by mass, according to
Chinese Standard GB/T 17671-1999 [15]). Fine aggregate:
medium sand, with the modulus of fineness of 2.5.
Coarse aggregate: crushed limestone, particle size
between 5 and 25 mm. High-range water reducing agent,
at a dosage from 1.0% to 2.5% of cement contents to
maintain slump of mixtures from 100 to 150 mm. The
diameter of the polypropylene fiber (PP fiber) was
20 lm, while the diameter of steel fiber was 2 mm, as
given in Table 1.
2.2. Experimental procedure
Five types of concretes were prepared, of which mix pro-
portions and compressive strength at 3 days and 28 days
are given in Table 2. Cube specimens of 100 mm size were
used for strength determination, and beams of 100 mm
height, 100 mm width and 400 mm length, were used for
fracture energy determination. After demolding, the speci-
mens were cure in water at 20 °C till 58 days.
At 58 days, the specimens were taken from curing water,
and dried under temperature of 105 °C in order to avoid
undesirable explosive spalling of concrete when exposed
to high temperature. The specimens were exposed to high
temperatures in an electric furnace, 200, 400, 600, and
800 °C, respectively. The time–temperature curve of the
furnace is given in Fig. 1. The ISO 834 curve in Fig. 1 is
a standard time–temperature curve for fire resistance test-
ing recommended in ISO standard, ISO 834. For each
exposure to high temperature, the target temperature was
maintained for 1 h. After exposure to high temperatures,
the specimens were subjected to various cooling regimes
as given in Table 3. Under each cooling regime, six cubes
(three cubes a batch for compressive strength, and the
other three cubes a batch for tensile splitting strength deter-
minations, respectively) and two beams (for fracture energy
determination) were used for each type of concrete. After
the specimens cooled down to room temperature, compres-
sive strength, tensile splitting strength, and fracture energy
were measured respectively.

Table 1
Characteristics of fibers
Type Density (g/cm3) Diameter Length (mm) Aspect ratio Tensile strength (MPa) Elastic modulus (GPa)
PP fiber 0.91 20 lm 20 1000 560–770 3.5
Steel fiber 7.80 2 mm 30 15 650–800 200

Table 2
Mix proportions and compressive strength of concretes prepareda
Type of concrete W/B Quantity (kg/m3) Compressive strength (MPa) at 28 days
C W S G SF PP-F S-F
26C 0.26 535 156 597 1153 64 0 0 83.5
26P1 0.26 519 151 579 1122 62 0.91 0 88.6
26P2 0.26 515 150 575 1112 62 2.73 0 83.4
26H1 0.26 527 154 588 1139 63 0.60 80 87.2
26H2 0.26 528 154 589 1139 63 1.82 70 89.3
a
W/B for water/binder ratio (mass), C for ordinary Portland cement (OPC), SF for silica fume, S for sand, G for coarse aggregate, W for water, PP–F
for 20 lm polypropylene fiber (PP fiber), and S-F for steel fiber respectively.
950 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955

1400

1200
Temperature (°C)

1000

800
ISO 834 curve
600

400

200 Experimental curves

0
0 60 120 180 240 300 360 420
Time (min)
Fig. 1. Experimental time–temperature curves compared with the stan-
dard curve in ISO 834.
Fig. 2. Photograph of cube specimens heated inside furnace.

2.3. Temperature determination inside specimens subjected

to heating and cooling

Temperature was determined on two types of 100 mm

concrete cubes designated by 26C and 26H1, which were
heated in an electric furnace. The target temperature was
maintained at 800 °C for 1 hr. After heating, from three
cubes a batch, one was subjected to quenching by immer-
sion in water, another was subjected to water spraying
for 1 hr, and the other was cooled naturally in the furnace.
The furnace and a typical cube specimen with three ther-
mocouples are shown in Figs. 2 and 3 respectively. Ther-
mocouples were cast at three locations inside the cubes,
i.e. at depths of 1, 3, and 5 cm, as schematically shown in
Fig. 4. The height (h) of the section of each specimen
exposed to quenching or spraying was 85 mm. Tempera-
ture at the various locations inside the cubes were deter-
mined continuously throughout heating and cooling.
Fig. 3. Photograph of a typical cube specimen with three thermocouples
2.4. Method of fracture energy determination subjected to quenching in water.

Notched beam specimens of 100 mm height, 100 mm

width and 400 mm length, were employed for fracture
energy determination. At 58 days, two specimens for each
batch were exposed to 105 °C for about two days, and then
heated to temperatures 400 and 600 °C respectively. Frac-
ture energy was determined according to an RILEM test
method entitled ‘‘FMC1 Determination of the fracture
energy of mortar and concrete by means of three-point
bend tests on notched beams’’ [16]. The configuration of
a specimen is shown in Fig. 5. A notch was prepared during
casting of each specimen, to form a pre-made crack at the
mid-span of the specimen. In a three-point bending test on
a beam specimen, mid-span deflection was recorded during

Table 3
Various cooling regimes used in the experiment
a
Specimen type, high temperatures Natural cooling 5 min Spraying 30 min Spraying 60 min Spraying Quenching in
water
p p p p p
Cubes for compressive strength, 200, 400, 600, and
800 °C
p p p p p
Cubes for tensile splitting strength, 200, 400, 600,
and 800 °C
p p p
Beams for fracture energy, 400, and 600 °C None None
a
Relative temperature of room air and cooling water during the tests was 25 °C.
 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955 951

Thermocouples 1000 Depth 1cm

Depth 3 cm
Natural cooling Depth 5 cm
800 Depth 1cm

Temperature (°C)
Depth 3 cm
Depth 5 cm
600

Section exposed 400

to rapid cooling Depth of thermocouple:
c = 50 mm
c b = 30 mm 200
b Water spraying
a a = 10 mm
h= 85 mm 0
0 60 120 180 240 300 360
Time (min)
Fig. 4. Thermocouple locations at various depths in a cube of 100 mm
size. Fig. 7. Temperature of 26H1 subjected to natural cooling and water
spraying for 60 min.

P 1000 Furnace temperature

W=100 Depth 1 cm
Depth 3 cm
800

Temperature (°C)
Depth 5cm

600
a0=50
400
Quenching
s=260 B=100
200
All in mm

Fig. 5. Configuration of a notched beam specimen. 0

0 60 120 180 240 300 360
Time (min)
the whole loading process until failure. The loading was in Fig. 8. Temperature of furnace and concrete 26C subjected to quenching.
a displacement-controll manner at a rate of 0.05 mm/min.

3. Results and discussion 1000 Depth 1 cm

Depth 3 cm
Natural cooling Depth 5 cm
3.1. Temperature in specimens during heating and cooling 800 Depth 1 cm
Temperature (°C)

Depth 3 cm
Depth 5 cm
The temperature determination results are shown in 600
Figs. 6 and 7 for concrete 26C, and in Figs. 8 and 9 for
26H1. These results indicate that the inner temperature 400
decreasing rate of a specimen during cooling, either under
quenching or under water spraying for 60 min, was signif- 200
Water spraying

0
Furnace temperature 0 60 120 180 240 300 360
1000
Depth 1cm Time (min)
Depth 3 cm
Depth 5 cm
Temperature (°C)

800 Fig. 9. Temperature of 26C subjected to natural cooling and water

spraying for 60 min.
600

400
icantly higher than that of the specimen subjected to natu-
Quenching ral cooling. In general, the specimens subjected to
200 quenching, or water spraying for 60 min, could cool to
100 °C within 20 min. For 26C subjected to quenching,
0 temperature at 10 mm depth dropped from 754 to 103 °C
0 60 120 180 240 300 360 within 15.5 min; temperature at 30 mm depth dropped
Time (min) from 745 to 102 °C within 17.5 min; and temperature at
Fig. 6. Temperature of furnace and concrete 26H1 subjected to 50 mm depth dropped from 727 to 102 °C within
quenching. 19.5 min. These results prove that the cooling regimes such
952 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955

as quenching in water, or water spraying for 30 min or 120 Natural cooling

more, caused an action of ‘‘thermal shock’’ to concrete 5 min spray
under elevated temperature, characterized by a high rate 100 30 min spray

Residual compressive
of temperature decreasing ranged from 25 to 44 min/°C, 60 min spray

strength (MPa)
80 Water quenching
which was calculated from Figs. 6–9. It can be concluded
that, in terms of thermal shock, water spraying for dura- 60
tion of 30 min or more is in consistency with quenching
in water. Nevertheless, if the duration for water spraying 40
was too short, for example 5 min, it can be seen from Figs.
20
7 and 9 that temperature would drop from 761 to 582 °C at
10 mm depth in 26H1, and from 743 to 681 °C at 10 mm 0
depth in 26C, after which the specimens would still 0 200 400 600 800 1000
undergo natural cooling from 582 to 681 °C, respectively. Temperature (°C)
It can be found from the results in Figs. 6–9 that there
Fig. 11. Residual compressive strength of polypropylene fiber concrete,
was only a slight difference between the temperature drop- 26P1.
ping rate of the specimens under quenching in water and
that under water spraying for 60 min, i.e. quenching caused
the temperature dropping rate slightly higher. Moreover, 100 Natural cooling
26H1 incorporating both PP fiber and steel fiber had both 5 min spraying

Residual compressive
a temperature rising rate during heating and a temperature 80 30 min spraying
60 min spraying

strength (MPa)
dropping rate during cooling slightly higher than those of
Water quenching
plain concrete 26C, as steel fiber could increase thermal 60
conductivity of hybrid fiber concrete 26H1.
40
3.2. Effect of cooling regimes on residual compressive
strength 20

0
The results of residual compressive strength of the five 0 200 400 600 800 1000
types of concrete subjected to various cooling regimes are Temperature (°C)
given from Figs. 10–14, respectively. It is obvious in these
figures that residual compressive strength of concrete was Fig. 12. Residual compressive strength of polypropylene fiber concrete,
significantly influenced by the cooling regimes. For each 26P2.
type of concrete under an identical target temperature,
among the five cooling regimes, natural cooling maintained
a relatively higher value of residual compressive strength, 140 Natural cooling
Residual compressive

while both quenching in water or spraying over 30 min 5 min spray

120
30 min spray
caused compressive strength approximately 20 MPa lower.
strength (MPa)

100 60 min spray

It should be noted that, as a few specimens were severely Water quenching
80
damaged by thermal shock, so that these specimens were
unavailable for strength testing and their results are 60
40
20
Natural cooling
120 0
5 min spraying 0 200 400 600 800 1000
Residual compressive

30 min spraying Temperature (°C)

100
60 min spraying
strength (MPa)

80 Water quenching Fig. 13. Residual compressive strength of hybrid fiber concrete, 26H1.

60 unavailable in the present figures, e.g. the results for water

quenching of 26P1 under 800 °C in Fig. 11 and for that of
40
26P2 in Fig. 12.
20 Water spraying for duration over 30 min can have an
effect on compressive strength similar to quenching in
0 water. Under natural cooling after exposure to 200 °C,
0 200 400 600 800 1000
residual compressive strength was even higher than original
Temperature (°C)
strength at room temperature, which might be attributed to
Fig. 10. Residual compressive strength of plain concrete, 26C. additional hydration of residual cement in concrete acti-
 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955 953

120 Natural cooling 6 Natural cooling

Residual compressive

5 min spraying 5 min spraying

100 30 min spraying
strength (MPa)

splitting strength (MPa)

5 30 min spraying
60 min spraying

Residual tensile
80 60 min spraying
Water quenching
4
60 Water quenching

40 3

20 2

0 1
0 200 400 600 800 1000
Temperature (°C) 0
0 200 400 600 800
Fig. 14. Residual compressive strength of hybrid fiber concrete, 26H2.
Temperature (°C)

Fig. 16. Residual tensile splitting strength of polypropylene fiber concrete,

vated by elevated temperature [17]. However, the increase 26P1.
in compressive strength was rather insignificant in the case
of quenching, or water spraying over 30 min, after expo-
sure to 200 °C. It is estimated that, as these specimens Natural cooling
6
cooled down rather rapidly, such an additional cement 5 min spray

splitting strength (MPa)

hydration occurred to a fairly little degree. 5

Residual tensile
30 min spray
4 60 min spray
3.3. Effect of cooling regimes on residual tensile splitting Water quenching
strength 3

2
The results of residual tensile splitting strength of the
five types of concrete subjected to various cooling regimes 1
are given from Figs. 15–19 respectively. Similar to the
0
results of residual compressive strength, residual tensile 0 200 400 600 800
strength under natural cooling was still relatively higher Temperature (°C)
than that under the other cooling regimes at each target
Fig. 17. Residual tensile splitting strength of polypropylene fiber concrete,
temperature. Concrete incorporating PP fiber lonely, iden-
26P2.
tified by 26P1 or 26P2, suffered a loss in tensile strength
more considerably than concrete incorporating hybrid fiber
(steel fiber and PP fiber), identified by 26H1 or 26H2.
It is shown in Figs. 16 and 17 that 26P1 or 26P2 concrete Natural cooling
incorporating PP fiber lonely had a low value of residual 8 5 min spraying
30 min spraying
tensile strength under 400 °C, which was almost equal to 7
splitting strength (MPa)

60 min spraying
that under 600 °C, implies that concrete incorporating PP
Residual tensile

6 Water qunching
fiber was rather severely damaged by exposure to 400 °C. 5
Compared with the results in Figs. 16 and 17, hybrid
4
fiber concrete performed significantly better, characterized
by a much slower decrease in tensile strength for tempera- 3
2
1
7 Natural cooling 0
Spray-5 min 0 200 400 600 800 1000
splitting strength (MPa)

6
Spray-30 min Temperature (°C)
Residual tensile

5 Spray-60 min
Water quenching Fig. 18. Residual tensile splitting strength of hybrid fiber concrete, 26H1.
4

3
ture increasing from 200 to 800 °C, as shown in Figs. 18
2 and 19. It is known from previous researches [1] that PP
1 fiber can help HSC/HPC improve resistance to thermally
induced explosive spalling. Nevertheless, the results in the
0
0 200 400 600 800 present investigation indicate that hybrid fiber (steel fiber
Temperature (°C) and PP fiber) can enhance fire resistance of HSC/HPC in
the form of relatively high residual strength, at the mean-
Fig. 15. Residual tensile splitting strength of plain concrete, 26C.
954 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955

6 Natural cooling This result is in good consistency with previous reports

[14,18–21]. Nevertheless, fracture energy of hybrid fiber
splitting strength (MPa)

5 5 min spray
concrete increased for 400 °C, but decreased for 600 °C.
Residual tensile

30 min spray However, under thermal shock induced by rapid cooling

4
such as water spraying for 30 min, the tendency of the
3 change in fracture energy with temperature increasing is
2 quite different from that under natural cooling. Under
water spraying for 30 min after exposure to 400 and
1 600 °C, although fracture energy of hybrid fiber concrete
0 decreased dramatically, but it was still as high as
0 200 400 600 800 1000 5317 W/m2. This result also confirms that hybrid fiber
Temperature (°C) (steel fiber and PP fiber) can significantly enhance fire resis-
Fig. 19. Residual tensile splitting strength of hybrid fiber concrete, 26H2.
tance of HSC/HPC even under a condition of thermal
shock caused by water spraying.
The fact that the impact of spraying water for 30 min or
time hybrid fiber HSC/HPC is still resistant to explosive more on mechanical properties was almost the same as that
spalling due to the incorporation of PP fiber, which has of water quenching, indicates that spraying water for
been proved in a preceding experimental investigation [14]. 30 min or more could cause thermal shock to a similar
degree to water quenching, which is in good agreement
3.4. Effect of cooling regimes on residual fracture energy with the results of temperature determination presented
in Section 3.1.
The results of residual fracture energy of two types of The results of both mechanical strength and fracture
concrete, identified by 26C and 26H1, subjected to various energy in this investigation reveal that fire properties of
cooling regimes, are given in Figs. 20 and 21 respectively. It concrete might be overestimated in most previous literature
can be clearly seen that hybrid fiber has a remarkable [1–3,7,14,18–22], where natural cooling rather than rapid
toughening effect on concrete. At room temperature, cooling was usually employed in experimental investiga-
hybrid fiber concrete had fracture energy of 14486 W/m2, tions. Thus, proper evaluation of fire resistance of concrete
while that of plain concrete was only 247 W/m2. needs experimental data obtained under rapid cooling
After exposure to natural cooling, residual fracture regimes such as water spraying or water quenching. On
energy of plain concrete 26C for target temperature of the other hand, it can be recommended from this investiga-
400 or 600 °C increased more or less, compared with the tion that a media other than water should be used for fire
original value at room temperature, as shown in Fig. 20. extinguishing, which will neither induce any thermal shock
like the case of water spraying or quenching in water, nor
cause any considerably negative effect on concrete
400 Room temperature properties.
Residual fracture

Natural cooling
energy (W/m2)

300 5 min spraying 4. Conclusions

200 30 min spraying
The temperature determination results for various
100 locations in concrete, prove that the cooling regimes such
0
as quenching in water, or water spraying for 30 min or
20 400 600 more, caused an action of ‘‘thermal shock’’ to concrete
Temperature (°C) exposed to high temperature, characterized by a high rate
Fig. 20. Residual fracture energy of plain concrete, 26C.
of temperature decreasing ranged from 25 to 44 min/°C.
In terms of thermal shock, water spraying for duration
of 30 min or more is in consistency with quenching in
water.
20000 Room temperature The experimental results indicate that cooling regimes
Residual fracture

Natural cooling significantly influences residual properties of concrete, such

energy (W/m2)

15000
Spray-5 min as compressive strength, tensile splitting strength, and frac-
10000 Spray-30 min ture energy. The fact that the impact of spraying water for
30 min or more on mechanical properties was almost the
5000
same as that of water quenching, indicates that spraying
0 water for 30 min or more could cause thermal shock to a
2 400 60 0
similar degree to water quenching, which is in good agree-
Temperature (°C)
ment with the results of the temperature determination.
Fig. 21. Residual fracture energy of hybrid fiber concrete, 26H1. Hybrid fiber (steel fiber and PP fiber) can enhance both
 G.-F. Peng et al. / Construction and Building Materials 22 (2008) 948–955
residual strength and fracture energy of concrete subjected
to thermal shock induced by rapid cooling from elevated
temperature up to 800 °C.
For each type of concrete under identical target temper-
ature, among the five cooling regimes, natural cooling
maintained a relatively higher value of residual compres-
sive strength. The similar results were also obtained on ten-
sile splitting strength and fracture energy. This means that
residual mechanical properties might be overestimated in
most previous literature, where natural cooling rather than
rapid cooling was usually employed.
Acknowledgements
The authors gratefully acknowledge the financial sup-
port of the National Science Foundation of China (Project
No. 50478060) and the PanDeng Foundation of Beijing
JiaoTong University. Shanghai Bekaert-Ergang Company
Ltd. is acknowledged for providing steel fiber and TaiAn
XianDai Plastic Products Company is acknowledged for
providing polypropylene fiber.
References
[1] Phan LT. Fire performance of high-strength concrete: a report of the
state-of-the-art, NISTIR 5934. Building and Fire Research Labora-
tory, National Institute of Standards and Technology, Gaithersburg,
MD; 1996.
[2] Schneider U. Concrete at high temperatures – a general review. Fire
Safety J 1988;13:55–68.
[3] Chan YN, Peng GF, Anson M. Fire behavior of high-performance
concrete made with silica fume at various moisture contents. ACI
Mater J 1999;96:405–9.
[4] Kalifa P, Chene G, Galle C. High-temperature behavior of HPC with
polypropylene fibers from spalling to microstructure. Cem Concr Res
2001;31:1487–99.
[5] Bentz DP. Fibers, percolation and spalling of high performance
concrete. ACI Mater J 2000;97:351–9.
[6] Purkiss JA. High-strength concrete and fire. Concrete (UK)
2000:49–50.
955
[7] Chan YN, Peng GF, Anson M. Residual strength and pore structure
of high-strength concrete and normal strength concrete after exposure
to high temperatures. Cem Concr Comp 1999;23:23–7.
[8] Khoury GA. Compressive strength of concrete at high temperatures:
reassessment. Mag Concr Res 1992;44:291–309.
[9] Luo X, Sun W, Chan SYN. Effect of heating and cooling on residual
strength and microstructure of normal strength and high-perfor-
mance concrete. Cem Concr Res 2000;30:379–83.
[10] Ishihara S, Gshima T, Nomura K. Crack propagation behavior of
cermets and cemented carbides under repeated thermal shocks by the
improved quench test. J Mater Sci 1999;34:629–36.
[11] Nassif AY. Postfiring stress–strain hysteresis of concrete subjected to
various heating and cooling regimes. Fire Mater 2002;26:103–9.
[12] Anonymous. Problems arising from the Hengyang fire case to the
philosophy of fire fighting in China (in Chinese). China News Weekly
2003; November 17.
[13] Chen JQ. Key issues in the ‘‘11-03’’ building fire and collapse case in
the city of Hengyang (in Chinese). Fire Sci Technol (China)
2004;23:99–102.
[14] Peng GF, Yang WW, Zhao J, Liu YF, Bian SH, Zhao LH. Explosive
spalling and residual mechanical properties of fiber-toughened high-
performance concrete subjected to high temperatures. Cem Concr Res
2006;36:723–7.
[15] Chinese Standard, GB/T 17671-1999 Method for determination of
strength of cement mortar (ISO method). pp. 1–13.
[16] RILEM, FMC1 determination of the fracture energy of mortar and
concrete by means of three-point bend tests on notched beams. In:
RILEM technical recommendations for the testing and use of
construction materials, London: E&FN SPON; 1994. pp. 99–101.
[17] Peng GF, Chan SYN, Anson M. Chemical kinetics of C–S–H
decomposition in hardened cement paste subjected to elevated
temperatures up to 800 °C. Adv Cem Res 2001;13:47–52.
[18] Nielsen CV, Bicanic N. Residual fracture energy of high-performance
and normal concrete subjected to high temperatures. Mater Struct
2003;36:515–21.
[19] Baker G. The effect of exposure to elevated temperatures on the
fracture energy of plain concrete. Mater Struct 1996;29:383–8.
[20] Zhang B, Bicanic N, Pearce CJ, Balabanic G. Residual fracture
properties of normal- and high-strength concrete subject to elevated
temperatures. Mag Concr Res 2000;52(2):123–36.
[21] Prokopski G. Fracture toughness of concretes at high temperatures. J
Mater Sci 1995;30:1609–12.
[22] Chen B, Liu JY. Residual strength of hybrid-fiber-reinforced high-
strength concrete after exposure to high temperatures. Cem Concr
Res 2004;34:1065–9.