Journal of Constructional Steel Research 121 (2016) 291–310

Contents lists available at ScienceDirect

Journal of Constructional Steel Research

Experimental investigation into the post-fire mechanical properties of

hot-rolled and cold-formed steels
Jie Lu c, Hongbo Liu a,b,c,⁎, Zhihua Chen a,b,c, Xiangwei Liao c
a
State Key Laboratory of Hydraulic Engineering Simulation and Safety, Tianjin University, Tianjin 300072, China
b
Key Laboratory of Coast Civil Structure Safety of China Ministry of Education, Tianjin University, Tianjin 300072, China
c
Department of Civil Engineering, Tianjin University, Tianjin 300072, China

a r t i c l e i n f o a b s t r a c t

Article history: Hot-rolled Q235, Q345, and Q420 steel members and cold-formed Q235 hollow sections are widely used in build-
Received 19 December 2015 ing structures. During fire hazards, steel structures are inevitably exposed to elevated temperatures; provided
Received in revised form 28 February 2016 that structural collapse does not occur after fire events, the residual performance of such structures must be es-
Accepted 2 March 2016
timated accurately to determine whether they should be dismantled, repaired, or directly reused. Therefore, an
Available online xxxx
experimental investigation was conducted to reveal the post-fire mechanical properties of hot-rolled Q235,
Keywords:
Q345, and Q420 steels as well as of cold-formed Q235 steels that underwent different levels of cold working.
hot-rolled steel Specimens were heated to various preselected temperatures up to 1000 °C and subsequently cooled down to am-
cold-formed steel bient temperature via two different methods, namely, air and water cooling. Tensile coupon tests were per-
post-fire mechanical properties formed to obtain the post-fire stress–strain curves, elastic moduli, yield strengths, ultimate strengths, and
predictive equations ductility. Additional tests were also conducted to investigate the effects of cyclic heating–cooling. The post-fire
cooling method mechanical properties of hot-rolled steels changed significantly after exposure to temperatures exceeding ap-
cyclic heating–cooling proximately 700 °C; the corresponding temperature for cold-formed steels was 300 °C. The influences of different
cooling methods were notable, whereas the effects of cyclic heating–cooling were insignificant. Thus, new pre-
dictive equations that incorporated the influences of various cooling methods were developed to evaluate the
post-fire mechanical properties of both hot-rolled and cold-formed steels studied.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
Hot-rolled Q235, Q345, and Q420 steel members have been applied
extensively as load-bearing members in building structures. For
instance, the beams, columns, and joints in most steel residential and in-
dustrial buildings in China are currently made of either Q235 or Q345
steel; meanwhile, Q420 is mainly used extensively in high-rise build-
ings. Q235 cold-formed hollow sections of different shapes (square,
rectangular, or circular) possess advantages such as low cost and a sim-
ple production process; these sections are also increasingly employed in
both high-rise and large-span building structures. Inevitably, building
structures made of steels may be exposed to elevated temperatures dur-
ing fire hazards, which constitute one of the most common and danger-
ous disasters that damage building structures. Unlike structures
composed of reinforced concrete, steel structures are weakly resistant
to fire, i.e., their performance drop significantly within a short time
when exposed to elevated temperatures. Thus, the fire design of steel
structures is highly significant. Extensive studies have been conducted
to investigate the high-temperature performance of steels of various
⁎ Corresponding author at: State Key Laboratory of Hydraulic Engineering Simulation
and Safety, Tianjin University, Tianjin 300072, China.
E-mail address: hbliu@tju.edu.cn (H. Liu).
grades and types [1–9], which revealed that in general the strength
and stiffness of steels significantly reduced with increasing tempera-
ture; furthermore, corresponding recommendations have been provid-
ed in design guides, such as British Standard (BS) 5950-8 [10] and EC3
[11]. Nevertheless, building structures are commonly designed conser-
vatively for safety and bear considerable redundancy (e.g., large-span
steel structures exhibit high degree of statical indeterminacy). Although
the performance of steel decreases remarkably in a fire, entire structures
may not collapse due to internal force redistribution. Provided that
structural collapse does not occur after fire events, the residual perfor-
mance of entire structures and of important load-bearing members
must be evaluated accurately to determine whether the structures
should be dismantled, repaired, or reused directly. Therefore, the post-
fire mechanical properties of steels must be studied to provide an
important basis for assessing the performance of steel structures after
fire events.
At present, increasing but limited studies [12–21] have been
conducted on the post-fire mechanical properties of steels, mainly in
Europe, USA, Australia, and China. Outinen and Makelainen [12,13] pre-
sented an experimental study to determine the mechanical properties
of S355 cold-formed steels at elevated temperatures and after cooling.
Qiang et al. [14,15] conducted experimental studies to estimate the
mechanical properties of high-strength structural steels S460, S690,

http://dx.doi.org/10.1016/j.jcsr.2016.03.005
0143-974X/© 2016 Elsevier Ltd. All rights reserved.
292 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310
and very high-strength steel S960 after cooling down from elevated
temperatures up to 1000 °C. A similar experimental study was per-
formed by Gunalan and Mahendran [16] to identify the post-fire me-
chanical properties of cold-formed steels G300, G500, and G550 after
exposure to temperatures up to 800 °C. Chiew et al. [17] investigated
the mechanical properties of reheated, quenched, and tempered high-
strength steel plates (grade S690) at elevated temperatures and after
cooling down. Wang et al. [18] conducted an experimental research
on the mechanical properties of high-strength Q460 steel after exposure
to temperatures up to 900 °C and considered both natural air and water
cooling methods. Other works focused on the post-fire mechanical prop-
erties of prestressing steel wires [19], reinforcing steels [20], and stainless
steels [21]. Moreover, in Annex B of BS 5950-8 (2003) [10], some recom-
mendations are available for the reuse of mild steels after fire exposure.
According to the brief review of existing literatures, few studies have
explored the post-fire mechanical properties of the extensively used
hot-rolled Q235, Q345, and Q420 steels and of the cold-formed Q235
steel. Furthermore, no current design guide has provided applicable rec-
ommendations for the reuse of these materials after fire events. Given the
considerable differences in chemical compositions and manufacturing
processes, the results of previous studies on other steels cannot be ap-
plied directly to estimate the post-fire performance of structures made
of hot-rolled Q235, Q345, and Q420 steels and of cold-formed Q235 steel.
In addition, fire guns are employed to extinguish flames when build-
ing structures are exposed to fire. Under such situations, steel members
are cooled down from an elevated temperature by the spraying of water
at a much higher rate than by cooling down in air. Different cooling
methods may induce variations in the post-fire mechanical properties
of steels; thus, various such techniques should be considered in simulat-
ing actual fire events in the study of the post-fire mechanical properties of
steels. Nonetheless, the influence of cooling methods has rarely been
accounted for in previous studies. Furthermore, the residual performance
of the structures must be assessed with great caution given that a few
structures may have been exposed to fire events recurrently without col-
lapsing. Therefore, the effects of cyclic heating–cooling should be ade-
quately considered when evaluating the post-fire performance of steels;
nevertheless, this factor has never been accounted for in previous works.
In general, if the post-fire mechanical properties of hot-rolled Q235,
Q345, and Q420 steels and of cold-formed Q235 steel are not reliably
evaluated and if the influences of the aforementioned factors are not
considered, then the behavior of the structures composed of these steels
after fire events cannot assessed convincingly. The results of such an
evaluation may generate an uneconomical consequence or a potential
safety problem. The current paper presents the details of an experimen-
tal investigation into the post-fire mechanical properties of hot-rolled
Q235, Q345, and Q420 steels as well as those of cold-formed Q235 steels
cut from both the flat and corner regions of square hollow sections. Ten-
sile coupon tests are conducted after these specimens are cooled down
from predetermined elevated temperatures up to 1000 °C (800 °C for
cold-formed steels). Both the air and water cooling methods are consid-
ered here. Associated mechanical properties are obtained, including
stress–strain curves, elastic moduli, yield strengths, ultimate strengths,
and ductility. The influences of the manufacturing processes (with or
without cold working), exposure temperatures, steel grades, and
cooling methods on the post-fire mechanical properties are also
discussed. The effects of cyclic heating–cooling are investigated through
additional tests. Predictive equations that incorporate the influences of
different cooling methods are proposed based on the experimental re-
sults to evaluate the residual behavior of the studied steels.
2. Experimental investigation
2.1. Test materials and specimens
The hot-rolled Q235, Q345, and Q420 steel specimens considered in
this test were cut in the longitudinal direction from hot-rolled steel
sheets of grades Q235, Q345, and Q420, respectively. The steel sheets
utilized comply with GB/T 700 [22] and GB/T 1591 [23]. Q235, Q345,
and Q420 are the abbreviations of the grade designations of these steels;
Q refers to yield strength (in Chinese Pinyin) and 235, 345, and 420 de-
note the corresponding minimum nominal yield strengths of 235, 345,
and 420 N/mm2. The cold-formed Q235 steel specimens were cut in
the longitudinal direction from the flat region (hereafter referred to as
CFS-F) and the corner region (hereafter referred to as CFS-C) of Q235
cold-formed square hollow sections (SHS, 800 × 800 × 20mm). These
sections are produced in accordance with GB/T 6725-2008 [24]. CFS-C
evidently has a higher level of cold working than CFS-F does.
The shapes and dimensions of the test specimens accord with GB/T
228.1-2010 [25] and GB/T 4338-2010 [26], as shown in Fig. 1, Table 1,
and Fig. 2. The dimensions of each specimen were measured with a ver-
nier caliper at three points within the gauge length. The average values
of the measured dimensions were used to calculate the mechanical
properties of the steels.
2.2. Test equipment and procedure
The entire procedure of the experiment mainly comprised two steps.
In the first step, the specimens were initially heated to the preselected
elevated temperatures and subsequently cooled down to ambient tem-
perature. In the second step, a tensile coupon test was conducted on the
specimens at ambient temperature. The heating process was accom-
plished by a temperature-controlled electric furnace (Fig. 3). The
thermocouple located inside the furnace measured the air temperature
in the furnace and relayed this information to the control system to fa-
cilitate the adjustment of the heating rate; thus, a closed control loop
was formed. In this study, 10 elevated temperatures were set for the
Q235, Q345, and Q420 specimens, i.e., 100 °C, 200 °C, 300 °C, 400 °C,
500 °C, 600 °C, 700 °C, 800 °C, 900 °C, and 1000 °C. Owing to the limited
material, four elevated temperatures were selected for the CFS-F and
CFS-C specimens, i.e., 300 °C, 500 °C, 700 °C, and 800 °C, which could
also basically meets the demand of engineering application because in
general, the elevated temperatures steel members may experience dur-
ing a fire hazard will not exceed 800 °C though some local members
may suffer higher temperatures. In the heating process, the furnace
temperature was initially increased at a rate of 15 °C/min to a tempera-
ture of 50 °C less than the target temperature; this temperature was
maintained for 10 min. Subsequently, the furnace temperature was
raised to the target temperature at a rate of 5 °C/min and held for anoth-
er 20 min. This heating process ensures the uniform temperature distri-
bution in the specimens and avoids the exceeding of actual temperature
from the target temperature. The influence of heating duration can be
ignored according to [20,27]. Subsequently, the specimens were re-
moved from the furnace and cooled down to ambient temperature.
Both air and water cooling methods were considered; the specimens
for the air cooling method were exposed to air and allowed to cool
down at their own rates to simulate the situation in which a fire dies
out naturally. The specimens for the water cooling method were cooled
down by water spraying using a water jet to simulate the scenario in
which fire is extinguished by fire guns. The entire heating–cooling pro-
cedure is plotted in Fig. 4. The water volume adopted in the experiment
Fig. 1. Shapes of the hot-rolled Q235, Q345, and Q420 steel specimens.
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 293

Table 1
Dimensions of the hot-rolled Q235, Q345, and Q420 steel specimens (mm).

Specimens t b d R Lc Lt

Q235/Q345 7.5 15 30 15 80 220

Q420 7.0 20 30 15 80 200

was determined based on the principle that the volume sprayed on each
unit area of the specimen surface is proportional to the volume sprayed
on the members in actual fire extinguishment. Given the firefighting pa-
rameters and flux of the water jet, the spraying time can be calculated
with the following:

Q 1T1A
T2 ¼ ; ð1Þ
Q 2 πR2

where T1 refers to the fire extinguishing time in actual fire events (s); Q1
refers to the water flux in actual fire extinguishment (m3/s), R refers to
the cover radius of the fire guns (m); T2 refers to the water spraying
time in this test (s); Q 2 refers to the flux of the water jet (m3/s),
which remained constant during the test at 2 × 10−5 m3/s; and A refers
to the surface area of one specimen (m2). The values of the firefighting
parameters were derived from those suggested in [28], where T1 =
7200s, Q1 = 1.5× 10−2 m3/s, and R = 15 m.
The tensile test was conducted via a universal material testing
machine, which can exert a maximum force of 600 kN in tension. An
electronic extensometer was attached to the specimens to measure
strain precisely. The electronic extensometer was removed from the
Q235, Q345, Q420, and CFS-F specimens when the stress–strain curves
reached the maximum stress value owing to the limited measurement
range; nonetheless, the strain range recorded is sufficient for design
purposes. However, the electronic extensometer with a wide measure-
ment range was used on CFS-C specimens hence their full-range stress–
strain curves were obtained. The tensile load was applied at a constant
stress rate of 5 MPa/s during the elastic stage and, subsequently, at a
constant strain rate of 0.001/s; this application satisfied the requirement
of GB/T 228.1-2010 [25]. All the data obtained from the experiment,
such as force, stroke, stress, strain, and stress–strain curves, were re-
corded by a computer that was compatible with the testing machine.
For comparison, tensile tests were also conducted at ambient tempera-
ture on specimens that had not been exposed to elevated temperatures.
Three specimens that were cooled down from the same elevated tem-
perature with the same cooling method were tested and regarded as a
group to minimize accidental errors. The average value of the three
specimens in each group was adopted to analyze the post-fire mechan-
ical properties of the steels.
Fig. 3. Temperature-controlled electric furnace.
Additional tests were also performed to explore the possible degen-
eration of the mechanical properties of steels as a result of multiple
heating–cooling cycles. The Q345 specimens were heated to preselected
elevated temperatures and cooled down to ambient temperature re-
peatedly for a maximum of three times. Subsequently, the tensile cou-
pon tests were conducted to obtain their mechanical properties and
the effects of cyclic heating–cooling were analyzed. The air cooling
method alone was considered along with three elevated temperatures,
namely, 400 °C, 600 °C, and 800 °C.
3. Experimental results
3.1. Failure modes
Tensile coupon tests were conducted on the steel specimens cooled
down from elevated temperatures until fracture. The failure modes of
the specimens were then obtained and shown in Fig. 5. The failure
modes of the specimens that were not exposed to elevated tempera-
tures were also included for comparison purposes. The hot-rolled steels
Q235, Q345, and Q420 and the cold-formed steels CFS-F and CFS-S all
exhibited obvious ductile failure with necking, and no brittle failure
was observed regardless of the exposure temperature and the cooling

Fig. 2. Sampling positions and dimensions of the cold-formed Q235 steel specimens.
294 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310
Fig. 4. Heating–cooling procedure.
method used. Notably, the ductility of both CFS-F and CFS-S obviously
increased with the exposure temperature. These observations indicate
that both the hot-rolled and cold-formed steels in this study are able
to avoid brittle failure and keep their original ductility after fire expo-
sure, which are positive for their reuse after fire events.
3.2. Stress–strain relationships
The post-fire stress–strain relationships obtained from this experi-
ment are plotted in Fig. 6 (each curve plotted is one of three curves
derived from the specimens in a group). The post-fire mechanical prop-
erties of the steels were studied based on these curves. In general,
although the stress–strain curves of most of the specimens showed an
obvious plateau in yield at ambient temperature and after cooling
down from various elevated temperatures, the characteristics of the
post-fire stress–strain curves differed considerably with respect to ex-
posure temperature, manufacturing process, steel grade, and cooling
method. The stress–strain curves of hot-rolled steels Q235, Q345, and
Q420 after exposure to temperatures lower than 700 °C were similar
to those obtained at ambient temperature. When the exposure temper-
atures exceeded 700 °C, the stress–strain curves changed significantly.
By contrast, the cold-formed steels CFS-F and CFS-C were highly sensi-
tive to the heating–cooling process, and obvious changes were observed
in their stress–strain curves after cooling from temperatures exceeding
300 °C. Moreover, the stress–strain curves of both CFS-F and CFS-C ex-
hibited long yield plateaus and increased strain values at maximum
stress with increasing exposure temperatures, thereby indicating an
increasing trend in ductility and demonstrating consistency with the
failure modes described in Subsection 3.1.
The influences of different cooling methods also contributed signifi-
cantly to the changes in the post-fire mechanical properties of the stud-
ied hot-rolled and cold-formed steels, particularly when the exposure
temperature exceeded 800 °C. In general, after cooling down from
high fire temperatures, the steels cooled by water spraying showed
higher yield and ultimate strengths than those cooled in air.
3.3. Elastic modulus
The post-fire elastic moduli of the steels tested in this experiment
are defined as the initial slopes of their stress–strain curves (Fig. 7).
The method adopted to calculate the initial slope accorded with GB/T
22315-2008 [29]. To describe the deterioration in the mechanical prop-
erties of steel after fire, the elastic modulus residual factor is defined as
the ratio of the elastic modulus after cooling down from elevated
temperatures (EPT) to that at ambient temperature (E) without fire
exposure. Tables 2 and 3 list the post-fire elastic moduli and the
corresponding residual factors obtained from the experiment for all
five types of steels. The residual factors are also plotted as the function
of exposure temperature in Fig. 8 (each listed and plotted value is the
average value of the three specimens in a group).
The low-grade Q235 and Q345 in hot-rolled steels can regain their
original elastic moduli after exposure to temperatures up to 800 °C;
henceforth, their elastic moduli gradually declined. By contrast, the elas-
tic modulus of high-grade Q420 was not affected after exposure to tem-
peratures below 700 °C; subsequently, however, this value decreased
suddenly. Nonetheless, at least 85% of the elastic modulus was regained
after exposure to elevated temperatures up to 1000 °C for all of the hot-
rolled steels included in the experiment. The elastic moduli of cold-
formed steels CFS-F and CFS-C were almost unchanged after exposure
to temperatures reaching 800 °C. Moreover, the difference in residual
elastic modulus caused by two different cooling methods is insignifi-
cant. The post-fire moduli of the studied steels indicated that the stiff-
ness of both hot-rolled and cold-formed steels did not decrease
significantly after exposure to fire.
3.4. Yield strength
For comparison purposes, the yield strengths of both hot-rolled and
cold-formed steels were determined with the 0.2% proof stress method
although the stress–curves of most of the test specimens exhibited
clearly defined yield plateaus. The 0.2% proof stress f0.2 was determined
from the intersection of the stress–strain curve and proportional line-
offset by 0.2% strain level (Fig. 7). The yield strength residual factor is
defined as the ratio of the yield strength after cooling down from elevat-
ed temperatures (fy , PT) to that at ambient temperature without the
heating–cooling (fy) processes. Tables 4 and 5 list the yield strengths
and residual factors derived from the experiment. The residual factors
are also plotted as the function of exposure temperature in Fig. 9
(each listed and plotted value is the average value of the three speci-
mens in a group).
The yield strengths of both hot-rolled and cold-formed steels
obviously declined after exposure to certain elevated temperatures.
Nonetheless, the reduction characteristics of the corresponding yield
strengths differed. The yield strengths of hot-rolled steels were almost
unaffected until the temperature reached 700 °C and decreased at a
rapid rate thereafter. A reduction of 30% was observed after exposure
to temperatures reaching 1000 °C. It is worth noting that when adopting
water cooling method, Q420 showed an obvious rebound in yield
strength after cooling from temperatures ranging from 800 °C to 1000
°C, which can be attributed to the change in the microstructures of the
steels as a result of rapid cooling. Cold-formed steels are considerably
more sensitive to the heating–cooling process than hot-rolled steels
are. The yield strengths of both CFS-F and CSF-C began to drop after ex-
posure to temperatures exceeding 300 °C. When the exposure temper-
ature reached 800 °C, reductions of approximately 35% and 30% were
observed for CFS-F and CSF-C, respectively.
The influences of different cooling methods on residual yield
strength were significant, particularly when the steels were cooled
down from high exposure temperatures. The water-cooled specimens
of both hot-rolled and cold-formed steels regained more of their origi-
nal yield strengths than the air-cooled specimens did. This difference
is attributed to the strengthening effect induced by water cooling,
which is equivalent to quenching. This observation indicates that
steels can preserve more of their yield strengths after exposure to fires
extinguished by fire guns.
3.5. Ultimate strength
The ultimate strength residual factor is defined as the ratio of the ul-
timate strength after cooling down from elevated temperatures (fu ,PT)
to that at ambient temperature without exposure to fire (fu). Tables 6
and 7 present the ultimate strengths and their corresponding residual
J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 295

Fig. 5. Failure modes of the test specimens.

296 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

factors after air and water cooling, respectively. The residual factors are
also plotted in Fig. 10 as the function of exposure temperatures (each
value listed and plotted is the average value of the three specimens in
a group).
Ultimate strength showed a significantly gentler decreasing trend
with respect to exposure temperature than yield strength did. At least
80% of the original ultimate strengths of the hot-rolled steels could be
regained after air cooling from elevated temperatures up to 1000 °C.
Under water cooling conditions, however, all hot-rolled steels displayed
almost no reduction in ultimate strength regardless of the temperature
they had been exposed to. Furthermore, the ultimate strength of high-
grade Q420 even increased slightly after exposure to fire temperatures

Fig. 6. Stress–strain curves after exposure to various fire temperatures.

 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310
Fig. 6 (continued).
above 700 °C. Similarly, the ultimate strengths of cold-formed steels de-
creased slightly. Both CFS-F and CFS-C regained at least 85% of their orig-
inal ultimate strengths even after exposure to 800 °C though they began
to lose these strengths when the exposure temperature exceeded 300
°C, which is much earlier than hot-rolled steels.
The influences of different cooling methods on post-fire ultimate
strength were similar to those on yield strength. Both hot-rolled
Fig. 7. Determination of the elastic modulus and yield strength.
297
and cold-formed water-cooled specimens exhibited higher ultimate
strengths than air-cooled specimens did as a result of the aforemen-
tioned strengthening effect of quenching.
3.6. Ductility
The ductility of steel is an indicator reflecting its plastic deformation
capacity, which is defined in this study according to the percentage
elongation of fracture. The percentage elongation of fracture is calculat-
ed as the ratio of the gauge length after fracture to the original gauge
length. The post-fire ductility residual factor is defined as the ratio of
the percentage elongation of fracture after cooling down from elevated
temperatures (δu,PT) to that at ambient temperature without exposure
to fire (δu). Tables 8 and 9 present the post-fire percentage elongation
of fracture obtained through the experiment and the corresponding re-
sidual factors for air and water cooling, respectively. The residual factors
are also plotted in Fig. 11 as the function of exposure temperature (each
listed and plotted value is the average value of the three specimens in a
group).
The post-fire ductility of the hot-rolled steels was significantly influ-
enced by cooling methods. Under air cooling, the ductility of both Q235
and Q345 steels gradually increased after exposure to temperatures up
to 1000 °C; while the ductility of Q420 remained almost unchanged
after fire exposure. Under water cooling, the ductility obviously de-
creased after exposure to temperatures exceeding 800 °C for all Q235,
Q345 and Q420 steels. Therefore, great caution should be given in
298 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Table 2
Post-fire elastic moduli and residual factors after air cooling.

Temperature Elastic moduli (GPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 203.1 207.8 204.2 208.1 204.5 1.000 1.000 1.000 1.000 1.000
100 206.8 206.1 202.2 — — 1.018 0.992 0.990 — —
200 206.6 206.7 206.1 — — 1.017 0.995 1.009 — —
300 204.2 203.9 208.9 201 206.3 1.005 0.981 1.023 0.966 1.009
400 203.4 204.9 205.5 — — 1.001 0.986 1.006 — —
500 204.7 209.5 207.5 203.2 207.1 1.008 1.008 1.016 0.976 1.013
600 200.8 206.1 206.8 — — 0.989 0.992 1.013 — —
700 202.4 207.6 196.0 199.8 204.9 0.997 0.999 0.960 0.960 1.002
800 205.1 208.8 178.3 203.6 207.4 1.010 1.005 0.873 0.978 1.014
900 195.6 193.0 176.0 — — 0.963 0.929 0.862 — —
1000 190.2 180.5 176.4 — — 0.936 0.869 0.864 — —

reusing hot-rolled steels cooled by fire guns. On the contrary, the ductil-
ity of cold-formed steels increased significantly with exposure temper-
ature under both air and water cooling conditions. In fact, the ductility
of CFS-F and CFS-C rose by 60% and 45%, respectively, after cooling
down from temperatures up to 800 °C. This observation is conducive
to the reuse of structural members made of cold-formed Q235 steel
after fire exposure.
3.7. Effects of cyclic heating–cooling
Additional tensile coupon tests were conducted on the Q345
specimens after multiple heating–cooling cycles to explore whether
possible degeneration in the mechanical properties of steels exists;
elevated temperatures of 400 °C, 600 °C, and 800 °C were considered.
Related mechanical properties, such as yield strength, ultimate
strength, elastic modulus, and ductility, were obtained after cyclic
heating–cooling. The degeneration factors are defined as the ratio
of the mechanical properties after multiple heating–cooling cycles
(E MPT, fy , MPT , fu , MPT, and δu , MPT ) to that after one heating–cooling
cycle (EPT , fy , PT, fu , PT, and δu , PT). Tables 10–12 list the mechanical
properties after multiple heating–cooling cycles obtained in the
experiment and their corresponding degeneration factors. The de-
generation factors are also plotted in Fig. 12 as the function of the
heating–cooling cycle index.
The elastic moduli, yield strengths, ultimate strengths, and ductil-
ity of Q345 remained almost the same after multiple heating–cooling
cycles as those after only one heating–cooling cycle regardless of the
exposure temperature. This observation indicates that the effects of
cyclic heating–cooling on the mechanical properties of hot-rolled
Q345 steels are so insignificant that it can be ignored. Considering
that the hot-rolled steels tested in this paper showed similar post-
fire mechanical properties, we may speculate that cyclic heating–
cooling will not significantly affect the mechanical properties of
hot-rolled steels.
4. Discussion
As detailed in this section, the residual factors of the mechanical
properties of the hot-rolled Q235, Q345, and Q420 steels obtained in
this study after exposure to elevated temperatures were compared
with those of the hot-rolled high-strength structural steel S460 [14]
and Q460 [18]. The results of the cold-formed Q235 steels CFS-F and
CFS-C were contrasted with those of the cold-formed G300 steel [16].
Only the experimental results derived under air cooling condition
were compared because the water cooling method was not included
in most previous studies. The experimental results were also contrasted
with the recommendations made in BS 5950-8 [10] for the reuse of mild
steels after fire in the form of residual factors. According to this stan-
dard, S235 and S275 steels should be capable of regaining at least 90%
of their original mechanical strength regardless of the exposure temper-
atures. For S355 steel, the guide recommends that at least 75% of its
strength can be regained after cooling down from temperatures above
600 °C.
In addition, the effects of different levels of cold working on the post-
fire mechanical properties of steels were discussed in this section based
on the experimental results obtained for CFS-F, CFS-C, and the hot-
rolled Q235 without cold forming process.
4.1. Comparison of post-fire elastic moduli
Fig. 13 shows the comparison of the post-fire elastic moduli obtained
in this study with those of S460, Q460, and G300 in existing studies. The
elastic modulus residual factors of all hot-rolled steels remained almost
the same when the exposure temperatures were below 600 °C; hence-
forth, significant differences were observed. The residual factors of both
Q235 and Q345 were similar to that of Q460 but differed obviously from
that of S460. The elastic modulus of Q420 with increasing temperature
decreased less than that of S460 but more than that of Q460. With re-
spect to the cold-formed steels, the results of CFS-F, CFS-C, and G300

Table 3
Post-fire elastic moduli and residual factors after water cooling.

Temperature Elastic moduli (GPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 203.1 207.8 204.2 208.1 204.5 1.000 1.000 1.000 1.000 1.000
100 200.6 209.2 209.5 — — 0.988 1.007 1.026 — —
200 198.1 207.0 201.9 — — 0.975 0.996 0.989 — —
300 203.5 207.2 206.4 204.1 201.3 1.002 0.997 1.011 0.981 0.984
400 201.7 206.8 205.9 — — 0.993 0.995 1.008 — —
500 201.6 202.2 204.3 207.6 201.8 0.993 0.973 1.000 0.998 0.987
600 201.2 205.7 202.6 — — 0.991 0.990 0.992 — —
700 204.3 204.5 201.9 205.4 208.1 1.006 0.984 0.989 0.987 1.018
800 199.9 203.2 186.3 210.1 206.6 0.984 0.978 0.912 1.010 1.010
900 195.7 195.4 171.4 — — 0.964 0.940 0.839 — —
1000 188.1 182.5 174.0 — — 0.926 0.878 0.852 — —
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 299

Fig. 8. Post-fire elastic modulus residual factors.

almost do not vary; their elastic moduli remained almost unchanged
after cooling down from elevated temperatures up to 800 °C.
4.2. Comparison of post-fire yield strengths
As depicted in Fig. 14, the post-fire yield strengths obtained in this
study were compared with those of S460, Q460, and G300 as well as
with the recommendations of BS 5950-8 for the reuse of mild steels
after fire exposure. With increasing exposure temperatures, the yield
strength residual factors of Q235, Q345, and Q420 were similar to that
of Q460 but varied markedly from that of S460 when the fire tempera-
tures exceeded 700 °C. Given cold-formed steels, both CFS-F and CFS-C
shared reduction characteristics with G300. In addition, adopting the
recommendations in BS 5950 for steels S235, S275, and S355 will cause
the post-fire yield strength of the steels tested to be overestimated
when they are exposed to high fire temperatures; thus, such recommen-
dations are not suitable for the reuse of both hot-rolled and cold-formed
steels after fire exposure.

Table 4
Post-fire yield strengths and residual factors after air cooling.

Temperature Yield strengths (MPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 342 389 447 375 397 1.000 1.000 1.000 1.000 1.000
100 347 394 446 — — 1.015 1.013 0.998 — —
200 341 390 441 — — 0.997 1.003 0.987 — —
300 333 386 453 364 395 0.974 0.992 1.013 0.971 0.995
400 337 403 451 — — 0.985 1.036 1.009 — —
500 349 393 437 322 353 1.020 1.010 0.978 0.859 0.889
600 345 396 446 — — 1.009 1.018 0.998 — —
700 335 383 450 300 297 0.980 0.985 1.007 0.800 0.748
800 302 328 382 268 259 0.883 0.843 0.855 0.715 0.652
900 282 307 386 — — 0.825 0.789 0.864 — —
1000 244 273 314 — — 0.713 0.702 0.702 — —
300 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Table 5
Post-fire yield strengths and residual factors after water cooling.

Temperature Yield strengths (MPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 342 389 447 375 397 1.000 1.000 1.000 1.000 1.000
100 347 397 448 — — 1.015 1.021 1.002 — —
200 352 390 449 — — 1.029 1.003 1.004 — —
300 347 388 468 372 403 1.015 0.997 1.047 0.992 1.015
400 344 398 459 — — 1.006 1.023 1.027 — —
500 350 393 489 336 375 1.023 1.010 1.094 0.896 0.945
600 353 393 482 — — 1.032 1.010 1.078 — —
700 329 395 476 317 336 0.962 1.015 1.065 0.845 0.846
800 328 354 337 289 301 0.959 0.910 0.754 0.771 0.758
900 315 329 365 — — 0.921 0.846 0.817 — —
1000 283 329 409 — — 0.827 0.846 0.915 — —

4.3. Comparison of post-fire ultimate strengths
As illustrated in Fig. 15, the post-fire ultimate strengths derived in
this study were compared with those of S460, Q460, and G300 as
well as with the recommendations of BS 5950-8. The ultimate
strength residual factors of all hot-rolled steels were similar when
they were cooled down from temperatures up to 700 °C; however,
obvious differences were observed thereafter. The decreases in the
ultimate strengths of low-grade Q235 and Q345 were less than
those of Q460 and S460. By contrast, the residual ultimate strength
of high-grade Q420 remained unchanged at temperatures of up to
900 °C and varied considerably from all other grades of hot-rolled
steels. Both CFS-F and CFS-C exhibited ultimate strength residual fac-
tors similar to those of G300. Furthermore, the recommendations in
BS 5950 for mild steels S235 and S275 are not applicable to the reuse
of both hot-rolled and cold-formed steels; while the recommenda-
tion for mild steel S355 is safe but too conservative when applied
to the reuse of these steels.

Fig. 9. Post-fire yield strength residual factors.

 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 301

Table 6
Post-fire ultimate strengths and residual factors after air cooling.

Temperature Ultimate strength (MPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 468 503 595 450 469 1.000 1.000 1.000 1.000 1.000
100 488 509 583 — — 1.043 1.012 0.980 — —
200 472 508 577 — — 1.009 1.010 0.970 — —
300 462 512 571 440 468 0.987 1.018 0.960 0.978 0.998
400 469 515 573 — — 1.002 1.024 0.963 — —
500 485 511 572 413 439 1.036 1.016 0.961 0.918 0.936
600 465 502 571 — — 0.994 0.998 0.960 — —
700 463 487 554 399 404 0.989 0.968 0.931 0.887 0.861
800 427 462 585 395 422 0.912 0.918 0.983 0.878 0.900
900 423 463 575 — — 0.904 0.920 0.966 — —
1000 404 443 474 — — 0.863 0.881 0.797 — —

4.4. Effects of cold working
In this study, the cold-formed Q235 steels CFS-F and CFS-C were
cut from the flat and corner regions of the Q235 square hollow sec-
tions (SHS, 800 × 800 × 20mm), respectively. During the process of
producing these sections, different levels of plastic deformation
were generated that would influence the material response of the
final cold-formed steels. In general, the key effects involved an in-
crease in yield strength, a reduction in ductility, and the formation
of residual stress [30]. To explore the effects of cold working at differ-
ent degrees on the post-fire mechanical properties of steels, the
post-fire yield strengths, ultimate strengths, and ductility of the
CFS-F, CFS-C, and hot-rolled Q235 steel without cold working were
compared, as displayed in Figs. 16 and 17.
The enhanced strength as a result of cold working was clearly ob-
served in the yield strengths of cold-formed steels at ambient tempera-
ture but was not evident in ultimate strengths. The yield strengths of
CFS-F and CFS-C increased by 9.7% and 16.1%, respectively, comparing
to Q235 steel without cold working. Nonetheless, these enhancements
began to diminish with increasing exposure temperature. Under air
cooling, the enhancement effects in CFS-F and CFS-C disappeared at
500 °C, and both CFS-F and CFS-C presented close yield strength;
while under water cooling, the enhancement effects remained until
the temperature reached 700 °C. Thus, the water cooling method is con-
ducive to the reservation of strength enhancements in cold-formed
steels.
Moreover, ductility reduction is also observed in cold-formed steels
at ambient temperature. The percentage elongation of fracture of both
CFS-F and CFS-C was almost 20% lower than that of Q235 without cold
working. As in the case of the strength enhancements, the ductility re-
duction effect began to diminish with an increase in exposure tempera-
ture. After cooling down from temperatures exceeding 500 °C, both CFS-
F and CFS-C exhibited ductility similar to or even higher than that of the
hot-rolled Q235, thereby indicating the disappearance of ductility re-
duction induced by cold working.
5. Predictive equations
As discussed in Section 4, the residual factors of the post-fire mechan-
ical properties of the hot-rolled steels Q235, Q345, and Q420 in this ex-
periment differ significantly from those of the Q460 and S460 steels
reported in previous literatures. For cold-formed steels, although both
CFS-F and CFS-C showed similar reduction trend to that of G300 for air
cooling, there is no available mechanical property residual factors under
water cooling condition. Therefore, the existing predictive equations pro-
posed for other steels are insufficiently accurate to evaluate the post-fire
mechanical properties of the hot-rolled Q235, Q345, and Q420 steels as
well as of the cold-formed Q235 steel, which are used extensively. More-
over, the recommendations in BS 5950-8 for the reuse of mild steels are
not accurate enough to effectively estimate the post-fire performances
of these steels. Thus, new predictive equations that are simple in form
yet have considerable precision were proposed to evaluate the post-fire
mechanical properties of both hot-rolled and cold-formed steels. As the
exposure temperature is the main factor causing the deterioration of me-
chanical properties, the predictive equations are developed as a function
of the highest exposure temperature T. The influences of different cooling
methods are also incorporated into these equations.
5.1. Elastic modulus
5.1.1. Hot-rolled Q235, Q345, and Q420 steels
According to the experimental results, the following piecewise pre-
dictive equations are established to describe the reductions in the elastic
moduli of Q235, Q345, and Q420 steels after cooling down from elevat-
ed temperatures. The influences of different cooling methods were con-
sidered in these formulas.

Table 7
Post-fire ultimate strengths and residual factors after water cooling.

Temperature Ultimate strength (MPa) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 468 503 595 450 469 1.000 1.000 1.000 1.000 1.000
100 466 508 583 — — 0.996 1.010 0.980 — —
200 487 507 581 — — 1.041 1.008 0.976 — —
300 483 503 593 442 476 1.032 1.000 0.997 0.982 1.015
400 472 507 589 — — 1.009 1.008 0.990 — —
500 485 506 596 435 458 1.036 1.006 1.002 0.967 0.977
600 483 508 597 — — 1.032 1.010 1.003 — —
700 456 495 598 411 443 0.974 0.984 1.005 0.913 0.945
800 458 479 613 414 448 0.979 0.952 1.030 0.920 0.955
900 459 488 604 — — 0.981 0.970 1.015 — —
1000 451 498 626 — — 0.964 0.990 1.052 — —
302 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Fig. 10. Post-fire ultimate strength residual factors.

Under air cooling condition, High-grade Q420 steel is calculated as follows:

Low-grade Q235 and Q345 steels are computed as follows:
EPT
20  C≤T ≤600  C; ¼ 1; ð3aÞ
 EPT E
20 C≤T ≤800 C; ¼ 1; ð2aÞ
E
EPT
600  CbT ≤1000  C; ¼ 2:051−2:51  10−3 T þ 1:32  10−6 T 2 : ð3bÞ
EPT E
800  CbT ≤1000  C; ¼ 2:148−2:15  10−3 T þ 9:02  10−7 T 2 : ð2bÞ
E
Under water cooling condition,

Table 8
Post-fire percentage elongation of fracture (%) and residual factors after air cooling.

Temperature Percentage elongation of fracture (%) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 29.5 28.5 31.3 23.6 22.7 1.000 1.000 1.000 1.004 1.000
100 30.5 29.2 30.3 — — 1.034 1.023 0.968 — —
200 32.7 28.7 30.3 — — 1.107 1.006 0.968 — —
300 32.7 29.8 30.5 24.1 26.6 1.107 1.047 0.975 1.026 1.172
400 32.7 30.5 30.3 — — 1.107 1.070 0.970 — —
500 30.3 30.5 31.3 30.5 28.2 1.028 1.070 1.001 1.298 1.242
600 34.5 32.7 31.0 — — 1.169 1.146 0.991 — —
700 35.0 34.3 30.3 33.0 32.8 1.186 1.205 0.970 1.404 1.445
800 35.8 32.2 28.2 34.5 32.6 1.215 1.129 0.903 1.468 1.436
900 36.2 34.0 28.7 — — 1.226 1.193 0.919 — —
1000 34.8 33.8 31.0 — — 1.181 1.187 0.991 — —
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 303

Table 9
Post-fire percentage elongation of fracture (%) and residual factors after water cooling

Temperature Percentage elongation of fracture (%) Residual factors

(°C) Q235 Q345 Q420 CFS-F CFS-C Q235 Q345 Q420 CFS-F CFS-C

20 29.5 28.5 31.3 23.6 22.7 1.000 1.000 1.000 1.000 1.000
100 32.2 29.2 29.8 — — 1.090 1.023 0.953 — —
200 31.7 29.3 30.3 — — 1.073 1.029 0.967 — —
300 31.3 31.7 29.0 23.1 25.9 1.062 1.111 0.928 0.979 1.141
400 29.8 28.8 30.5 — — 1.011 1.012 0.974 — —
500 27.8 30.3 31.1 27.0 26.3 0.944 1.064 0.995 1.144 1.159
600 33.0 32.2 29.2 — — 1.119 1.129 0.933 — —
700 32.5 32.2 29.5 32.9 28.2 1.102 1.129 0.943 1.394 1.242
800 26.8 31.3 26.1 34.2 33.1 0.910 1.099 0.835 1.449 1.458
900 29.5 18.6 26.0 — — 1.000 0.651 0.830 — —
1000 22.5 23.2 22.9 — — 0.763 0.813 0.732 — —

Low-grade Q235 and Q345 steels are computed as follows: High-grade Q420 steel is calculated as follows:

EPT EPT
20  C≤T ≤800  C; ¼ 1; ð4aÞ 20  C≤T ≤700  C; ¼ 1; ð5aÞ
E E

EPT EPT
800  C b T ≤1000  C; ¼ 2:891−4:27  10−3 T þ 2:23  10−6 T 2 : ð4bÞ 700  C b T ≤1000  C; ¼ 2:891−4:27  10−3 T þ 2:23  10−6 T 2 : ð5bÞ
E E

Fig. 11. Post-fire ductility residual factors.

304 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Table 10 Table 12
Mechanical properties and degeneration factors of Q345 after cyclic heating–cooling Mechanical properties and degeneration factors of Q345 after cyclic heating–cooling
(400 °C). (800 °C).

Cycle index EMPT (GPa) EMPT fy,MPT f y;MPT fu ,MPT f u;MPT δu,MPT δu;MPT Cycle index EMPT (GPa) EMPT fy,MPT f y;MPT fu ,MPT f u;MPT δu ,MPT δu;MPT
EPT (MPa) f y;PT (MPa) f u;PT (%) δu;PT EPT (MPa) f y;PT (MPa) f u;PT (%) δu;PT

1 204.9 1.000 403 1.000 515 1.000 30.5 1.000 1 208.8 1.000 328 1.000 462 1.000 32.2 1.000
2 202.4 0.988 405 1.005 515 1.000 31.2 1.023 2 207.9 0.996 329 1.003 462 1.000 34.0 1.056
3 203.2 0.992 405 1.005 519 1.008 30.7 1.007 3 208.1 0.997 330 1.006 446 0.965 33.7 1.047

Figs. 18(a) and 18(b) compare the post-fire elastic modulus residual All Q235, Q345, and Q420 steels can be computed as follows:
factors predicted using Eqs. (2a, 2b)-(5a, 5b) with the experimental re-
sults, and a good agreement can be observed. f y;PT
20  C≤T ≤700  C; ¼ 1:0; ð9aÞ
fy
5.1.2. Cold-formed Q235 steels
The elastic moduli of both CFS-F and CFS-C remain almost un- f y;PT
700  CbT ≤1000  C; ¼ 1:6−8:88  10−4 T: ð9bÞ
changed after fire exposure under both air and cooling conditions. fy
Therefore, two options were proposed to predict their post-fire elastic
modulus residual factors. Under water cooling condition,
For simplicity, conservative percentage of 95% of the original elastic Low-grade Q235 and Q345 steels can be calculated as follows:
modulus after exposure to temperatures up to 800 °C is recommended
for both cold-formed Q235 steel CFS-F and CFS-C in Option 1. Thus, f y;PT
Eq. (6) was developed for both air and water cooling conditions. 20  C≤T ≤600  C; ¼ 1:007 þ 2:17  10−5 T; ð10aÞ
fy

EPT f y;PT
20  C≤T ≤800  C; ¼ 0:95: ð6Þ 600  CbT ≤1000  C; ¼ 1:313−4:75  10−4 T: ð10bÞ
E fy

For precise evaluation, Eqs. (7) and (8) were established in Option 2 High-grade Q420 steel can be computed as follows:
for air and water cooling conditions, respectively.
Under air cooling condition, both CFS-F and CFS-C can be predicted f y;PT
20  C≤T ≤700  C; ¼ 0:998 þ 9:60  10−5 T; ð11aÞ
as follows: fy

E PT f y;PT
20  C≤T ≤800  C; ¼ 1:001−5:02  10−5 T þ 4:90  10−8 T 2 : ð7Þ 700  CbT ≤1000  C; ¼ 41:658−0:137T þ 1:52
E fy
 10−4 T 2 −5:57  10−8 T 3 : ð11bÞ
Under water cooling condition, both CFS-F and CFS-C can be predict-
ed as follows:
5.2.2. Cold-formed Q235 steels
E PT Eqs. (12)–(14) are proposed to predict the yield strength residual

20 C≤T ≤800 C; 
¼ 1:001−8:04  10−5 T þ 1:19  10−7 T 2 : ð8Þ
E factors of cold-formed Q235 steels in which the influences of different
cooling methods are considered. Fig. 19(c) and (d) compare the post-
Figs. 18(c) and 18(d) compare the elastic modulus residual factors fire yield strength residual factors predicted using these equations
predicted using Eqs. (7) and (8) with the experimental results and with the experimental results, and a good agreement can be seen.
there is a good agreement between them. Under air cooling condition,
CFS-F can be computed as follows:
5.2. Yield strength
f y;PT
20 C≤T ≤800  C; ¼ 1:004−5:76  10−5 T−3:72  10−7 T 2 : ð12Þ
5.2.1. Hot-rolled Q235, Q345, and Q420 steels fy
Piecewise Eqs. (9a, 9b)–(11a, 11b) were developed to determine the
reduction trends of the post-fire yield strengths of Q235, Q345, and CFS-C can be calculated as follows:
Q420. The influences of the different cooling methods were incorporat-
f y;PT
ed as well. Figs. 19(a) and 19(b) compare the post-fire yield strength re- 20  C≤T ≤800  C; ¼ 0:999 þ 1:90  10−4 T−7:82  10−7 T 2 : ð13Þ
sidual factors predicted using these equations with the experimental fy
results, and a good agreement is shown.
Under air cooling condition, Under water cooling condition,
Both CFS-F and CFS-C can be computed as follows:

Table 11 f y;PT
Mechanical properties and degeneration factors of Q345 after cyclic heating–cooling
20  C≤T ≤800  C; ¼ 0:999 þ 1:40  10−4 T−5:36  10−7 T 2 : ð14Þ
fy
(600 °C).

Cycle index EMPT (GPa) EMPT fy,MPT f y;MPT fu ,MPT f u;MPT δu,MPT δu;MPT
EPT (MPa) f y;PT (MPa) f u;PT (%) δu;PT
5.3. Ultimate strength
1 206.1 1.000 396 1.000 502 1.000 32.7 1.000
2 207.3 1.007 403 1.018 515 1.026 33.1 1.012 As in the case of yield strength, predictive equations that consider
3 207.1 0.992 392 0.990 504 1.004 33.3 1.018 the influences of different cooling methods were also developed to
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 305

Fig. 12. Degeneration factors of Q345 steel after multiple heating–cooling cycles.

predict the post-fire ultimate strength residual factors of both hot-rolled Low-grade Q235 and Q345 can be calculated as follows:
and cold-formed steels in this experiment.
f u; PT
20  C≤T ≤1000  C; ¼ 0:999 þ 1:59  10−4 T−2:89
fu
5.3.1. Hot-rolled Q235, Q345, and Q420 steels
 10−7 T 2 : ð15Þ
Under air cooling condition,

Fig. 13. Comparison of post-fire elastic modulus residual factors.

306 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Fig. 14. Comparison of post-fire yield strength residual factors.

High-grade Q420 can be computed as follows: f u; PT

800  CbT ≤1000  C; ¼ −4:344 þ 1:27  10−2 T−7:57  10−6 T 2 :
fu
ð16bÞ
f u; PT
20  C≤T ≤800  C; ¼ 1:004−2:26  10−4 T þ 2:49
fu
 10−7 T 2 ; ð16aÞ Under water cooling condition,

Fig. 15. Comparison of post-fire ultimate strength residual factors.

Fig. 16. Comparison of the post-fire yield and ultimate strengths of Q235, CFS-F, and CFS-C.
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310 307

Fig. 17. Comparison of the post-fire ductility of Q235, CFS-F, and CFS-C.

Low-grade Q235 and Q345 can be calculated as follows: High-grade Q420 can be computed as follows:

f u; PT f u; PT
20  C≤T ≤1000  C; ¼ 0:990 þ 2:57  10−4 T−5:91 20  C≤T ≤1000  C; ¼ 0:991−3:59  10−5 T þ 8:97
fu fu
 10−7 T 2 þ 3:16  10−10 T 3 : ð17Þ  10−8 T 2 : ð18Þ

Fig. 18. Comparison of the predicted elastic modulus residual factors with the experimental results.
308 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310

Fig. 19. Comparison of the predicted yield strength residual factors with the experimental results.

Figs. 20(a) and 20(b) indicate the comparison of the post-fire ulti- 6. Conclusion
mate strength residual factors predicted using Eqs. (15)–(18) with the
experimental results. It can be seen there is a good agreement between This paper presents a detailed experimental study on the post-fire
them. mechanical properties of hot-rolled Q235, Q345, and Q420 steels and
of cold-formed Q235 steels cut from the flat and corner regions of
5.3.2. Cold-formed Q235 steels square hollow sections. The standard steel specimens were initially
Under air cooling condition, both CFS-F and CFS-C can be predicted heated to various predetermined elevated temperatures up to 1000 °C
as follows: (800 °C for cold-formed steels) and subsequently cooled down to
ambient temperature by air or by water spraying. Subsequently, tensile
coupon tests were conducted to obtain the associated mechanical prop-
f u; PT
20  C≤T ≤1000  C; ¼ 0:992 þ 3:97  10−4 T−1:76 erties, such as stress–strain curves, elastic moduli, yield strengths,
fu ultimate strengths, and ductility. Additional tensile tests were also per-
 10−6 T 2 þ 1:37  10−9 T 3 : ð19Þ formed to explore the effects of cyclic heating–cooling on the post-fire
mechanical properties of steel. The test results showed that the post-
fire mechanical properties of the steels were significantly influenced
Under water cooling condition, both CFS-F and CFS-C can be predict-
by the manufacturing process (with or without cold working), exposure
ed as follows:
temperature, steel grade, and the cooling method adopted, whereas the
effects of cyclic heating–cooling are negligible. Hot-rolled Q235, Q345,
f u; PT and Q420 steels all can regain at least 85% of their original elastic moduli
20  C≤T ≤1000  C; ¼ 0:995 þ 2:55  10−4 T−9:89
fu after exposure to elevated temperatures up to 1000 °C. Their post-fire
 10−7 T 2 þ 7:19  10−10 T 3 : ð20Þ yield strengths also remained unchanged until 700 °C; however, a re-
duction of 30% in yield strength was observed when the exposure tem-
perature increased thereafter up to 1000 °C. The reduction in ultimate
Figs. 20(c) and 20(d) compare the post-fire ultimate strength resid- strength was much gentler, and all the hot-rolled steels regained at
ual factors predicted using Eqs. (19) and (20) with the test results, and least 80% of their original ultimate strengths after fire exposure. Their
there is a good agreement between them. ductility did not degenerated after air cooling, but significant reductions
 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310
Fig. 20. Comparison of the predicted ultimate strength residual factors with the experimental results.
were observed after water cooling, hence great caution should be given
in reusing hot-rolled steels cooled by fire guns. With respect to the Q235
cold-formed steels, both CFS-F and CFS-C exhibited similar reduction
characteristics after fire exposure although they underwent different
levels of cold working. The corresponding elastic moduli remained un-
changed after cooling down from temperatures up to 800 °C; however,
the strengths were sensitive to the process of heating–cooling. Both
yield and ultimate strengths began to decrease when the exposure tem-
peratures exceeded 300 °C. Notably, the ductility of cold-formed Q235
steels increased remarkably after fire exposure under both air and
water cooling conditions. Moreover, the yield strength enhancements
and ductility reduction induced by cold working diminished with in-
creasing exposure temperatures.
The influences of different cooling methods were significant, partic-
ularly in terms of residual strength and ductility. Water-cooled, hot-
rolled steels displayed higher yield and ultimate strength but lower duc-
tility than air-cooled steels. With regard to cold-formed steels, strength
enhancements alone were observed. As discussed previously, suitable
recommendations for the reuse of the common steels examined in
this study are not provided in current design guides and in existing
literatures. Accordingly, new predictive equations that consider the in-
fluences of different cooling methods were developed to evaluate the
post-fire elastic moduli, yield strengths, and ultimate strengths of the
studied steels precisely. The application of the predictive equations
established in this study can assist in accurately evaluating the post-
fire performance and the safe reuse of steel structures.
309
Acknowledgements
This study was financially supported by the Fok Ying-Tong Educa-
tion Foundation. The authors also appreciate the financial support pro-
vided by the Chinese Scholarship Council (File No. 201506255034),
which enables the visiting research scholar to collaborate with Prof.
Luke Bisby at the University of Edinburgh.
References
[1] J. Outinen, J. Kesti, P. Makelainen, Fire design model for structural steel S355 based
upon transient state tensile test results, J. Constr. Steel Res. 42 (3) (1997) 161–169.
[2] P. Makelainen, J. Outinen, J. Kesti, Fire design model for structural steel S420M based
upon transient-state tensile test results, J. Constr. Steel Res. 48 (1) (1998) 47–57.
[3] J. Outinen, O. Kaitila, P. Makelainen, High-temperature testing of structural steel and
modelling of structures at fire temperatures, Helsinki University of Technology Lab-
oratory of Steel Structures, Helsinki, Finland, 2001.
[4] J. Chen, B. Young, B. Uy, Behavior of high strength structural steel at elevated tem-
peratures, J. Struct. Eng. 132 (12) (2006) 1948–1954.
[5] J. Lange, N. Wohlfeil, Examination of the mechanical properties of the microalloyed
grain refined steel S460 at elevated temperatures, Bautechnik 84 (10) (2007)
711–720.
[6] J. Chen, B. Young, Experimental investigation of cold-formed steel material at elevat-
ed temperatures, Thin-Walled Struct. 45 (1) (2007) 96–110.
[7] T. Ranawaka, M. Mahendran, Experimental study of the mechanical properties of
light gauge cold-formed steels at elevated temperatures, Fire Saf. J. 44 (2) (2009)
219–229.
[8] F. McCann, L. Gardner, S. Kirk, Elevated temperature material properties of cold-
formed steel hollow sections, Thin-Walled Struct. 90 (2015) 84–94.
[9] Yongfeng Luo, Xiyu Wang, Qiang Xuhong, Liu Xiao, Progress in application of high
strength steel to engineering structures, J. Tianjin Univ. 48 (2015) 134–141 (Suppl.).
310 J. Lu et al. / Journal of Constructional Steel Research 121 (2016) 291–310
[10] BS 5950, Structural use of steelwork in building, Part 8: code of practice for fire re-
sistant design, Institution BS, 1998.
[11] Eurocode 3, Design of steel structures, Part 1-2: general rules-structural fire design,
CEN, 2005.
[12] J. Outinen, P. Makelainen, Mechanical properties of structural steel at elevated tem-
peratures and after cooling down, Fire Mater. 28 (2-4) (2004) 237–251.
[13] J. Outinen, Mechanical properties of structural steels at high temperatures and after
cooling down [Ph.D. thesis], Helsinki University of Technology, Helsinki, Finland,
2007.
[14] X. Qiang, F.S.K. Bijlaard, H. Kolstein, Post-fire mechanical properties of high strength
structural steels S460 and S690, Eng. Struct. 35 (2012) 1–10.
[15] X. Qiang, F.S.K. Bijlaard, H. Kolstein, Post-fire performance of very high strength steel
S960, J. Constr. Steel Res. 80 (2013) 235–242.
[16] S. Gunalan, M. Mahendran, Experimental investigation of post-fire mechanical prop-
erties of cold-formed steels, Thin-Walled Struct. 84 (2014) 241–254.
[17] S.P. Chiew, M.S. Zhao, C.K. Lee, Mechanical properties of heat-treated high strength
steel under fire/post-fire conditions, J. Constr. Steel Res. 98 (2014) 12–19.
[18] W. Wang, T. Liu, J. Liu, Experimental study on post-fire mechanical properties of
high strength Q460 steel, J. Constr. Steel Res. 114 (2015) 100–109.
[19] J.M. Atienza, M. Elices, Behavior of prestressing steels after a simulated fire: Fire-
induced damages, Constr. Build. Mater. 23 (8) (2009) 2932–2940.
[20] Z. Tao, X.Q. Wang, B. Uy, Stress-strain curves of structural and reinforcing steels after
exposure to elevated temperatures, J. Mater. Civ. Eng. 25 (9) (2012) 1306–1316.
[21] X.Q. Wang, Z. Tao, T.Y. Song, et al., Stress–strain model of austenitic stainless steel
after exposure to elevated temperatures, J. Constr. Steel Res. 99 (2014) 129–139.
[22] GB/T 700-2006, Carbon structural steels, Standards Press of China, 2006.
[23] GB/T 1591-2008, High strength low alloy structural steels, Standards Press of China,
2008.
[24] GB/T 6725-2008, Cold forming steel sections, Standards Press of China, 2008.
[25] GB/T 228-2010, Metallic material-tensile testing, Part 1: Method of test at room
temperature, Standards Press of China, 2010.
[26] GB/T 4338-2010, Metallic material-tensile testing at elevated temperature, Stan-
dards Press of China, 2010.
[27] C.I. Smith, B.R. Kirby, D.G. Lapwood, et al., The reinstatement of fire damaged steel
framed structures, Fire Saf. J. 4 (1) (1981) 21–62.
[28] GB 50045-95, Code for fire protection design of tall buildings, Standards Press of China,
2005.
[29] GB/T 22315-2008, Metallic material-Determination of modulus of elasticity and
Poisson’s ratio, Standards Press of China, 2008.
[30] S. Afshan, B. Rossi, L. Gardner, Strength enhancements in cold-formed structural
sections—Part I: Material testing, J. Constr. Steel Res. 83 (2013) 177–188.