Professional Documents
Culture Documents
UHPC Beam Fire
UHPC Beam Fire
UHPC Beam Fire
Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct
A R T I C LE I N FO A B S T R A C T
Keywords: This paper presents results from an experimental study on the behavior of ultra high performance fiber re-
Fire resistance inforced concrete (UHPFRC) beams subjected to combined effects of structural loading and fire exposure. Five
Ultra high performance fiber reinforced large-scale UHPFRC beams, fabricated with different batch mix proportions, were tested to evaluate the struc-
concrete tural behavior and spalling performance under ambient and fire conditions. The test variables included the
Fire induced spalling
presence of polypropylene fibers, load level and fire exposure scenario. The test results show that UHPFRC
Steel fibers
Polypropylene fibers
beams are highly susceptible to explosive spalling in the compression zone (on the sides) of the beam section
leading to lower fire resistance, as compared to conventional normal or high strength concrete beams. When
polypropylene fibers are present in UHPFRC, the extent of fire induced spalling decreases, resulting in higher fire
resistance. Results from fire resistance experiments indicate that higher load level aid in release of pore pressure
through tensile cracking and decrease the extent of spalling in UHPFRC beams. In addition, UHPFRC beams with
polypropylene fibers exhibit better performance under design fire scenarios with distinct cooling phase than
under fire scenarios without a decay phase.
⁎
Corresponding author.
E-mail address: kodur@egr.msu.edu (V. Kodur).
https://doi.org/10.1016/j.engstruct.2020.110316
Received 8 November 2019; Received in revised form 24 January 2020; Accepted 30 January 2020
Available online 12 February 2020
0141-0296/ © 2020 Elsevier Ltd. All rights reserved.
S. Banerji, et al. Engineering Structures 208 (2020) 110316
mechanical stresses. Such spalling can be explosive under severe fire (500 × 500 mm) were 3428 mm in length. The columns were fabri-
exposure, and depends on numerous factors such as heating rate, per- cated with UHPFRC mix comprising of hybrid fibers; steel (0.5% by
meability (and associated concrete strength), moisture content, con- volume), nylon (0.2% by volume) and PP fibers (0.2% by volume). Both
crete composition, aggregate type, and loading conditions [15,16]. columns experienced only minor spalling and attained fire resistance of
The potential risk of fire induced spalling has been found to be 3 h and this good performance (minor spalling) was attributed to pre-
higher in concretes with lower permeability and higher compressive sence of hybrid fibers in UHPFRC mix.
strength [17]. In case of UHPFRC, the exceptional compressive strength Kahanji et al. [10] conducted fire tests on seven UHPFRC beams of
and durability properties is achieved through the addition of dense rectangular cross-section (100 × 200 mm) with a span length of
bonding materials such as silica fume and slag leading to compact 2000 mm. Six of these beams were made of UHPFRC batch mix that had
(dense) microstructure of concrete. This densely packed microstructure steel fibers; three beams with 2% (by volume) of steel fibers, and an-
of UHPFRC results in extremely low permeability (two decimal orders other three with 4% (by volume) of steel fibers. The seventh beam was
lower than conventional NSC) [18,19]. Therefore, in fire exposed made of a UHPFRC mix having a combination of steel fibers (2% by
structural members made of UHPFRC, the thermally induced vapor volume) and PP fibers (4 kg/m3). The beams were exposed to standard
pressure cannot easily escape due to low permeability of UHPFRC; thus, ISO-834 fire exposure for 60 min, but the exposure was only on the
making UHPFRC members highly vulnerable to fire-induced spalling as bottom half of the beam (cross-section). All six UHPFRC beams, with
compared to conventional NSC and HSC members. only steel fibers (but without PP fibers), experienced severe explosive
In order to mitigate fire induced spalling in concrete members, re- spalling. The seventh UHPFRC beam (with a high dosage of PP fibers)
searchers have proposed a number of measures including addition of did not experience spalling. However, the addition of PP fibers in the
polypropylene (PP) or steel fibers to the concrete mix [20,21]. In case of seventh beam led to significant reduction in compressive strength of the
concrete with PP fibers, the mechanism through which extent of spal- UHPFRC mix to 100 MPa (from 163 MPa in the case of UHPFRC with
ling can be minimized is attributed to the melting of PP fibers present in steel fibers only).
concrete at relatively low temperatures (in 160–170 °C range). The The above review clearly indicates that there are only limited fire
melting of PP fibers creates channels (pores) inside the concrete mi- resistance studies on UHPFRC beams. Thus, there is a lack of data, in-
crostructure, thus increasing permeability of concrete. These channels cluding detailed observations and recordings of spalling, and response
facilitate dissipation of temperature induced vapor pressure generated of UHPFRC beams under fire exposure. Additionally, it is remarkable to
in a fire exposed concrete member, thus preventing spalling. In case of note that the previous experimental studies used a high dosage of PP
concrete with steel fibers, the enhancement of the tensile strength of fibers (4 kg/m3) for mitigation of spalling. The high PP dosage resulted
concrete (specially at elevated temperatures) is the primary factor that in reduced compressive strength of the concrete mixes in previous
helps in withstanding high pore pressure build-up, and thus minimize studies to levels below that of optimum desired strength of 150 MPa in
spalling, in a fire exposed concrete member [20,22]. There have been UHPC [10,29]. Unlike previously published works, this study seeks the
numerous studies to evaluate the influence of steel and polypropylene incorporation of a balanced dosage of PP fibers in UHPFRC mix to
fibers on HSC structural members’ fire performance [12,23]. However, achieve spalling mitigation, without impacting on compressive strength
only few studies have explored the effectiveness of steel and hybrid and workability. Furthermore, the current fire tests on UHPFRC beams
(steel and polypropylene) fibers in minimizing fire-induced spalling in are under design fire exposure while the reported tests in the literature
UHPC structural members [11,19]. UHPC has an extremely dense mi- were carried out by subjecting UHPFRC members to standard fire ex-
crostructure with much lower permeability than HSC, which can result posure only, without any due consideration to realistic fire scenarios;
in the development of higher pore pressure in UHPC structural mem- that encompass cooling phase.
bers. In addition, UHPC is usually reinforced with steel fibers to attain In addition, the previous fire tests on UHPFRC members were car-
higher tensile strength and improved post-cracking response, and such ried out using concrete batch mixes without any coarse aggregates.
a mix is termed as UHPFRC. Thus, UHPFRC exhibits higher tensile Such UHPFRC mixes, made with fine aggregates and high super-
strength than HSC, and thus can better resist tensile stresses that cause plasticizer and silica fume, incur higher costs and require special mixing
spalling. However, this has not been fully validated through full scale equipment that are not commonly available in many concrete batch mix
fire tests on UHPC or UHPFRC structural members. plants. Recently, a relatively improved mix design for UHPFRC has
To address the above discussed knowledge gaps, five large scale been proposed, wherein fine aggregates can be partially replaced with
beams, made of UHPFRC with different types of fiber reinforcement, coarse aggregates, so as to reduce the dosage of cementitious material
were tested under combined effects of fire and structural loading. and binder, and this can lower the cost of UHPFRC [30,31]. This mix
Detailed thermal and structural response of the beams, including pro- incorporates controlled amount of coarse aggregates in order to attain
gression of spalling and cracking, was traced to discuss the critical dense packing of the concrete without compromising the desired me-
factors influencing the spalling and fire resistance of UHPFRC beams. chanical properties of UHPFRC. Further, this batch mix can be prepared
in a conventional ready-mix truck, and therefore, facilitates the ease of
2. Previous research fabrication at a lower cost, which in turn can promote its market ac-
ceptance [32]. However, no fire tests on UHPFRC beams, made with
A detailed review of literature indicate that there have been a this specialized coarse aggregate UHPFRC batch mix have been con-
number of experimental studies on the fire performance of NSC beams, ducted in the literature. Therefore, a set of fire resistance experiments is
and to a lesser extent on HSC beams. These studies investigated the carried out to develop test data and a comprehensive understanding on
effect of various parameters, such as fire scenario, axial restraint, cross- the behavior of UHPFRC beams under fire conditions.
sectional size, concrete strength, load intensity, reinforcement ratio,
etc., on the fire response of RC beams [24–26]. However, the review 3. Experimental program
shows that there has been only very limited research on UHPC or
UHPFRC members under fire exposure. Most of the previous studies The experimental program consisted of undertaking fire resistance
mainly focused on degradation of strength properties with temperature tests on four ultra high performance fiber reinforced concrete
by testing small scale UHPC or UHPFRC cubes, cylinders or prisms (UHPFRC) beams under simultaneous application of structural loading
[8,17,27,28], and very few fire tests are reported on UHPFRC full-scale and fire exposure. In addition to fire tests, a control beam made of
structural members [9,10]. UHPFRC was also tested under flexural loading to evaluate the response
Lee et al. [9] tested two full-scale UHPFRC columns under ISO-834 of the beam at room temperature. The test variables included load level,
standard fire exposure for 3 h. The square section columns type of fibers (either steel, or a combination of steel and polypropylene)
2
S. Banerji, et al. Engineering Structures 208 (2020) 110316
Table 1
Sectional dimensions and reinforcement details in UHPFRC beams.
Beam designation Width (mm) Depth (mm) Length (mm) Fiber Tensile reinforcement ρt (%) ρv (%)
3
S. Banerji, et al. Engineering Structures 208 (2020) 110316
total) and superplasticizer were added and mixed for 5 min until the tabulated values in Table 3. This can be attributed to the lower moisture
fresh concrete is consistent. Steel fibers were sprinkled and mixed for content adopted in the design concrete mix used for Batch 2 as com-
another 5 min to attain a homogenous mixture. Finally, polypropylene pared to that of Batch 1.
fibers are added and mixed for another 5 min. In order to obtain the
desired workability of UHPFRC, a high range water reducer (HRWR), 3.2. Instrumentation
which is a polycarboxylate based superplasticizer (Chryso 150), was
added to the batch mix [32]. This ensured sufficient workability was The instrumentation mounted in the test beams included thermo-
achieved in the batch mix. couples, displacement transducers and strain gauges. Type-K chromel-
As part of fabrication, plywood forms were assembled to achieve alumel thermocouples, 0.91 mm thick, were installed at two different
required internal dimensions in beams. Thermal curing is essential for cross sections (mid-span and quarter span) in each beam for measuring
the development of a denser microstructure of UHPFRC with the concrete and rebar temperatures. The location and numbering of the
completion of pozzolanic reactions for increased formation of calcium thermocouples in the cross section are shown in Fig. 2. The deflection of
silica hydrate (C-S-H) [33,34]. To attain in-situ high temperature each beam is measured at midspan as well as at the location of the two
curing, from heat of hydration of high cementitious binder contents in point loads using linear variable differential transformers (LVDTs).
UHPFRC, adequate insulation was provided in the beams formwork These LVDTs were placed outside the furnace (on the top of the beam)
using rigid Styrofoam and insulating blankets. Rigid Styrofoam in- since they cannot survive high temperature exposure within the fur-
sulation of 50 mm thickness was installed on two interior sides of the nace. LVDTs are connected to a well-insulated stiff threaded steel rod
framework and bottom side of the framework was provided with rigid attached to mid span and two load points in the beam. The steel rod
Styrofoam of 100 mm thickness. Insulating blankets were used to cover extends vertically to pass through a special opening in the furnace lid.
the casted full-scale beams to prevent heat loss, generated during hy- The strain gauges were mounted on two main longitudinal reinforce-
dration of the binder in UHPFRC. The temperatures developed in ments and one compression reinforcement for beam U-B11 with an
UHPFRC beams curing from the heat of hydration of the cementitious adhesive (glue) application. These strain gauges were of the high-
matrix were monitored. A sustained rise in temperature was observed in temperature foil strain gauge type, which usually is able to provide
the first 25 h of curing time with peak temperatures reaching to about reliable strain readings in temperature range of 20–350 °C. These high-
75 °C; further details on the curing process and data on temperature temperature strain gauges were used to obtain strain data at rebar level
measurements can be found elsewhere [31]. to supplement the data obtained externally though LVDTs.
In fabrication of UHPFRC structures, heat of hydration with surface
cooling effects can generate high temperature gradients with higher 3.3. Test apparatus
temperatures developing in the concrete core (due to heat of hydration
of cementitious components) as compared to regions closer to external The flexural behavior of the control beam U-B3 was evaluated at
surface of the beams [31,35]. Such non-uniform temperature distribu- ambient condition through a two-point specialized loading set-up as
tion can disrupt the hydration process and cause cracking in concrete shown in Fig. 3. In this set-up, two point loads can be applied on the top
from thermal stresses. Besides assisting in thermal curing, the insulation face of beam U-B3 through a displacement controlled actuator (MTS
provided in the formwork also helps to prevent such early age cracking machine with a capacity of 1500 kN). The other four UHPFRC beams
by maintaining relatively uniform temperatures within the casted were tested under combined effect of fire and structural loading using
beams. fire test furnace at Michigan State University (MSU) Civil Infrastructure
During fabrication of UHPFRC beams, cylindrical specimens of di- Laboratory. This furnace is designed to generate typical conditions,
mensions 76 mm by 152 mm were also cast for evaluating compressive such as temperature, structural loading, and heat transfer, to which a
strength. The cylinders were steam cured (80 °C) for 48 h and subse- structural member might be exposed during an actual fire incident. The
quently stored in controlled conditions of air maintained at 25 °C furnace details are illustrated in Fig. 4(a).
temperature and 50% relative humidity. The average compressive cy- The fire test furnace comprises of a steel framework supported by
linder strength of concrete measured at 28 days and on the day of the four steel columns and a fire chamber of 3.05 m in length, 2.44 m in
testing is tabulated in Table 3. width and 1.78 m in height. A loading frame comprising of steel beams
Despite the ease of manufacturing and lower production cost, and columns to apply realtime support and load conditions encloses the
UHPFRC used in the current study exhibits the characteristic material test furnace. The furnace is equipped with six gas burners, which are
properties (post cracking strain hardening, high compressive and ten- capable of producing maximum heat power of 2.5 MW. These six bur-
sile strength) of conventional UHPC without any coarse aggregates. ners are strategically placed on four walls of the furnace for uniform
Therefore, the change in mixing technique or use of locally available progression of heat energy within the furnace chamber. Six type-K
materials is expected to not have much impact on fire resistance of Chromel–Alumel thermocouples, are also placed on four walls of the
UHPFRC members. The moisture level (relative humidity) was mea- furnace to monitor furnace temperature during fire tests. The input gas
sured at a depth of 50 mm of the beam using a relative humidity probe and ventilation are controlled manually to maintain the average fur-
on the day of the fire test on each beam. The relative humidity of nace temperature consistent with a specified fire curve (standard or
UHPFRC beams with steel fibers (Batch 1) is higher than that of beams design fire exposure). All thermocouple, LVDT and load cell channels
with steel and polypropylene fibers (Batch 2) as can be seen from the are connected to a data acquisition system, which display and record
Table 3
Variables in tested UHPFRC beams.
Beam designation Fiber Fire exposure Applied load kN, (% of capacity) Relative humidity (%) Concrete strength (MPa) Age of specimen (months)
28-day Test day
4
S. Banerji, et al. Engineering Structures 208 (2020) 110316
Fig. 2. Location of strain gauges and thermocouples at various cross-sections in (a) beams U-B1, U-B2, U-B3 and U-B10, (b) beam B-11 (All units are in mm).
5
S. Banerji, et al. Engineering Structures 208 (2020) 110316
Fig. 5. Time-temperature curves for fire scenarios used in the fire tests.
3.4. Test condition and procedure The measured rebar and concrete temperatures at mid-span in
beams U-B1, U-B2, U-10 and U-B11 are plotted in Fig. 6 as a function of
The control beam U-B3 was tested under ambient condition by in- fire exposure time. In the beams subjected to design fire DF1 (U-B1, U-
crementing the loading; in steps, till failure. The fire resistance tests B2 and U-B10), the temperatures in rebar and concrete increase with
were carried out by placing each UHPFRC beam in the furnace under fire exposure time, whereas in the beam subjected to design fire DF2 (U-
simply-supported conditions and exposing them to a desired fire ex- B11), the measured temperatures increase to a peak value and then
posure. Three out of the four tested UHPFRC beams (U-B1, U-B2, U- starts to decrease as per the targeted heating-cooling cycle applied on
B10, and U-B11) were tested under design fire exposure (DF1) to si- this beam. Since the initial portion of the DF2, prior to the decay phase,
mulate a typical office fire without a decay phase [36], till failure. To followed the same time temperature curve as DF1, all four beams had
investigate the effect of fire scenario on the fire response of UHPFRC same fire exposure up to 90 min and thus the rebar and concrete
beam, beam U-B11 was exposed to design fire scenario (DF2) com- temperature rise can be compared directly. The temperature measured
prising of heating phase of 90 min followed by a cooling phase of ap- at the quarter depth (h/4) from the top surface in beam U-B2 is aberrant
proximately 10 °C per minute to represent natural cooling in typical due to malfunctioning of thermocouple.
fires encountered in real life as shown in Fig. 5. It should be noted that In all four beams, the corner rebars experienced higher temperature
the rate of heating in all cases is slightly less severe than that re- rise than middle rebar. This can be attributed to peripheral location of
commended by ASTM E119 [37] to represent typical ventilation-con- corner rebar, which is subjected to heating from the side of the beam, as
trolled conditions encountered in buildings. well as the bottom surface. The concrete temperatures in layers farther
All beams were tested under two point loads, each of which was from the fire exposed bottom surface is lower than the layers closer to
placed at 1.4 m from the end supports as shown in Fig. 3. This setup the fire exposed surface. This trend is on expected lines and is attributed
ensured that the critical span (mid-span) was subjected to pure flexure to the low thermal conductivity and high specific heat of concrete,
6
S. Banerji, et al. Engineering Structures 208 (2020) 110316
Fig. 6. Temperature progression at various depths in beams: (a) U-B1, (b) U-B2, (c) U-B10, (d) U-B11 (All dimensions are in mm).
which delays temperature transmission into the inner layers of con- reported in many previous fire tests [24,38]. This difference in sectional
crete. It can be seen from the plotted trends in Fig. 6 that the measured temperature progression in UHPFRC beams can be attributed to fire-
temperature at the mid-depth of concrete is lower than that at the induced spalling that occurs in the compression zone (top layers) of the
quarter-depth (h/4) from bottom exposed surface of concrete. However, beam; which in turn exposes deeper layers (behind the spalled layer on
the temperature plots of UHPFRC beams reveal that the pace of tem- sides) of concrete directly to fire. The spalling that occurred is sche-
perature rise beyond mid-depth (at the three-fourth quarter (3 h/4) matically shown in Fig. 7 and further discussion is laid out in following
depth from fire exposed side and also on the unexposed side at the top sections.
surface) is similar to that of mid-depth, even though the mid-depth is Analysis of measured temperature data from tests show that load
nearer to fire exposed surface than the upper portion of the beam. Such level, fire exposure scenario and presence of PP fibers in concrete have
variation in temperature rise trends with time in upper portion of major influence on the temperature progression in UHPFRC beams. The
UHPFRC beam is quite different from that in NSC and HSC beams, as effect of these factors on the thermal response is through their influence
7
S. Banerji, et al. Engineering Structures 208 (2020) 110316
The deflection at mid span of the control beam U-B3 was measured
at each load increment using LVDTs installed on bottom surface of the
beam. Load-deflection response in beam U-B3, tested under flexural
loading, is plotted in Fig. 9 for the entire range of loading till failure.
The control beam exhibited four distinct stages of response i.e., linear
elastic stage until initiation of tensile cracking, post-cracking stage with
Fig. 8. Comparison of (a) rebar, (b) concrete temperatures as a function of fire enhanced cracking and their progression, onset of yielding in steel re-
exposure time. inforcement, and plastic deformation stage till peak load, followed by
attainment of failure. Tensile cracking in this control beam U-B3 started
with progression of cracking and spalling of concrete. To further illus- at total load level of 26.2 kN (resulting from two point loads). As the
trate variation of temperature trends in different UHPFRC beams, load increased further, stiffness in the beam decreased due to increased
temperature rise at rebars and two concrete locations (mid-depth and macro cracking and progression of these cracks. The strain in reinfor-
top surface) in the four tested beams is compared in Fig. 8. The tem- cing steel at the critical section was also monitored to gauge the
perature rise in beams with PP fibers (U-B10 and U-B11) occur at a yielding point in tensile reinforcement. The steel reinforcement yielded
slower rate than the beams without PP fibers (U-B1 and U-B2). This is at a total load level of 81.1 kN in beam U-B3, beyond which the beam
on expected lines since presence of polypropylene fibers in beams U- exhibited hardening and sustained increase of load carrying capacity.
B10 and U-B11 helped to mitigate spalling thus minimizing loss of cross The maximum load capacity of beam U-B3 was measured to be 97 kN.
section during fire exposure, thereby slowing down temperature rise in After reaching peak load, the beam continued to deform in a plastic
the interior of the section. In beams with PP fibers, the measured manner until failure occurred [31]. The ultimate deflection of the beam
temperature at rebars and top surface follow closely with each other. It was measured to be 107 mm.
can be observed that the temperature at mid-depth in beam U-B11 are The fire response of the four tested UHPFRC beams can be gauged
somewhat lower than that in beam U-B10. Due to the slightly higher through variation of mid span deflection with fire exposure time, as
load level (5%) in beam U-B11, more cracks might have opened up in plotted in Fig. 10(a). Only deflections measured through displacement
beam U-B11 as compared to beam U-B10. The tensile cracks have transducers are presented here because the data obtained through high
possibly contributed in releasing pore pressure, leading to lesser spal- temperature strain gauges was quite inconsistent beyond 200 °C and
ling and thus, slightly slower temperature rise in beam U-B11. thus, was deemed unreliable. Because of erratic nature, the strain gauge
The differences in temperature rise in the beams without PP fibers, data cannot be used to draw definitive conclusions. This type of erratic
U-B1 and U-B2, can be attributed to the extent of spalling, mainly re- measurements in strain gauges (due to instability of strain gauges be-
sulting from different loading levels which is discussed in detail in yond 300 °C) have been reported in numerous previous fire tests
Sections 4.2 and 4.3. Temperatures rise at higher pace in beam U-B1 [20,39]. The progression of deflection in each beam can broadly be
(subjected to lower load level) as compared to other beams due to pore grouped under three stages. In stage 1; i.e. the first 40 min of fire ex-
pressure buildup leading to higher level of spalling in this beam. Some posure, deflections in all four beams increase at a slow and steady pace,
undulations in the form of peaks can be seen in the sectional time- and this is mainly influenced by level of structural loading present in
temperature trends of beam U-B1 plotted in Fig. 6(a) and Fig. 8. Spe- the beam and extent of thermal strains resulting from the development
cifically, the temperature peaks were recorded by thermocouples in of thermal gradients along the beam depth. Beam U-B2, which was
subjected to a higher load level (60% of its room temperature capacity)
8
S. Banerji, et al. Engineering Structures 208 (2020) 110316
Table 4
Summary of fire test results.
Beam Fiber Fire exposure Applied loading kN, Maximum spalling Average maximum Spalled weight Extent of Fire resistance
designation (% of capacity) depth (mm) spalling depth (mm) ratio (%) spalling (min)
*PP: Polypropylene
*DF: Design Fire
9
S. Banerji, et al. Engineering Structures 208 (2020) 110316
However, the average maximum spalling depth is variant in each beam schematic representation of section (with the spalling pattern) is shown
and aligns with the measured spalled weight ratio, providing a more in Fig. 11. Although all four beams were exposed to fire from three
representative measurement of spalling depth. sides, much of the spalling was mainly in upper compression zone on
Spalling in all four beams (including the beams with PP fibers), either side of the beams with only minor levels of spalling or scaling at
started after about 10 min into fire exposure and this was accompanied the bottom surface, as can be seen in Fig. 11. The extent of spalling in
with loud bangs. The spalling was violent as concrete pieces were hit- the tested UHPFRC beams is mainly dependent on the permeability in
ting the walls of the furnace with high velocity. Around 40 min into fire the concrete, as all beams had similar tensile strength to withstand
exposure, this explosive spalling stopped. In beams U-B1 and U-B2, generated stresses, through the addition of steel fibers in same quantity
fabricated without polypropylene fibers, spalling recommenced after in each batch mix. Permeability under elevated temperature exposure is
60 min of fire exposure (second round of spalling). This can be attrib- influenced by the presence of polypropylene fibers and cracking pattern
uted to heating of inner concrete layers, which were subjected to direct that develops, which in turn depends on the level of applied loading
fire exposure after early spalling. Concrete in UHPFRC beams without during fire exposure. As presented in Fig. 11, beam U-B1 experienced
PP fibers have low permeability and when subjected to heating, high maximum spalling (13.2% by weight) due to absence of PP fibers and
levels of pore pressure build-up again due to temperature induced lower cracking developed as this beam was subjected to lower load
moisture migration to interior impermeable parts of the beam cross- level (40%). The extent of spalling in beam U-B2 (7.5% by weight) is
section. However, concrete in beams fabricated with PP fibers, U-B10 much lower due to the higher level of applied loading (60%), which
and U-B11, did not encounter the second round of spalling and this can aided in crack development and this way the pore pressure could be
be attributed to the subsiding of pore pressure through microchannels released. The percentage of spalled concrete in beam U-B10 was also
enabled by melting of polypropylene fibers that occur around 160 °C. lower (7.3%) than the beams without PP fibers due to dissipation of
The overall intensity of spalling and the violent sounds of concrete pore pressure through melting of polypropylene fibers at about 160 °C.
chunks hitting the walls of furnace were significantly less intensive The extent of spalling in beam U-B11 (with PP fibers) is further lower
during the fire tests of beams with PP fibers. (5.4%) and this is due to microcracking induced by relatively higher
The state of UHPFRC beams after the fire tests, together with load ratio (50%) on this beam.
Fig. 11. Fire test results: (a) state of UHPFRC beams after fire tests, (b) bottom surface in tested beams, (c) schematic illustration of spalling and cracking pattern in
tested beams.
10
S. Banerji, et al. Engineering Structures 208 (2020) 110316
11
S. Banerji, et al. Engineering Structures 208 (2020) 110316
portion (compression zone), which is in contrast to NSC and HSC porous ceramic fine aggregate as an internal curing material. Constr Build Mater
beams where majority of spalling is confined to fire exposed bottom 2019;206:219–25.
[9] Lee J-H, Sohn Y-S, Lee S-H. Fire resistance of hybrid fibre-reinforced, ultra-high-
surface and bottom corners. This pattern of spalling in compression strength concrete columns with compressive strength from 120 to 200 MPa. Mag
zone results in faster temperature rise in the inner compressive Concr Res 2012;64:539–50.
layers of concrete (including compression rebars). [10] Kahanji C, Ali F, Nadjai A. Explosive spalling of ultra-high performance fibre re-
• Load level and fire scenario have influence on the extent of spalling
reinforcement on fire resistance of high-strength concrete columns. J Struct Eng
2003;129:253–9.
and fire performance of UHPFRC beams. The extent of spalling is [13] Kodur VKR, Dwaikat MB. Effect of fire induced spalling on the response of re-
less severe in UHPFRC beams under higher load levels due to alle- inforced concrete beams. Int J Concr Struct Mater 2008;2:71–81.
[14] Bazant ZP. Analysis of Pore Pressure, Thermal Stress and Fracture in Rapidly Heated
viation of pore pressure resulting from increased cracking in the
Concrete. International Workshop on Fire Performance of High-Strength Concrete.
tension zone of the beam. Also, UHPFRC beams exhibit better per- Proceedings. Appendix B: Workshop Papers. B10, Gaithersburg, MD: NIST; 1997, p.
formance under design fire exposure with distinct cooling phase, as 155–64.
compared to standard fire exposure. [15] Anderberg Y. Spalling phenomenon of HPC and OC. NIST SP 919, Gaithersburg,
MD: NIST; 1997, p. 69–75.
[16] Harmathy TZ. Fire safety design and concrete; 1993.
CRediT authorship contribution statement [17] Ahmad S, Rasul M, Adekunle SK, Al-Dulaijan SU, Maslehuddin M, Ali SI.
Mechanical properties of steel fiber-reinforced UHPC mixtures exposed to elevated
temperature: effects of exposure duration and fiber content. Compos B Eng
Srishti Banerji: Methodology, Formal analysis, Investigation, Data 2019;168:291–301.
curation, Writing - original draft. Venkatesh Kodur: [18] Shi C, Wu Z, Xiao J, Wang D, Huang Z, Fang Z. A review on ultra high performance
Conceptualization, Methodology, Writing - review & editing, concrete: Part I. Raw materials and mixture design. Constr Build Mater
2015;101:741–51.
Supervision, Funding acquisition. Roya Solhmirzaei: Investigation, [19] Li Y, Tan KH, Yang E-H. Influence of aggregate size and inclusion of polypropylene
Writing - review & editing. and steel fibers on the hot permeability of ultra-high performance concrete (UHPC)
at elevated temperature. Constr Build Mater 2018;169:629–37.
[20] Khaliq W, Kodur V. Effectiveness of polypropylene and steel fibers in enhancing fire
Declaration of Competing Interest resistance of high-strength concrete columns. J Struct Eng 2018;144:04017224.
[21] Kodur VKR. Guidelines for fire resistance design of high-strength concrete columns.
The authors declare that they have no known competing financial J Fire Prot Eng 2005;15:93–106.
[22] Kodur VKR, Phan L. Critical factors governing the fire performance of high strength
interests or personal relationships that could have appeared to influ-
concrete systems. Fire Saf J 2007;42:482–8.
ence the work reported in this paper. [23] Ali F, Nadjai A, Silcock G, Abu-Tair A. Outcomes of a major research on fire re-
sistance of concrete columns. Fire Saf J 2004;39:433–45.
Acknowledgements [24] Dwaikat MB, Kodur VKR. Response of restrained concrete beams under design fire
exposure. J Struct Eng 2009;135:1408–17.
[25] Agrawal A, Kodur V. Residual response of fire-damaged high-strength concrete
The authors wish to acknowledge the support of U.S. Airforce beams. Fire Mater 2019;43:310–22.
Research Laboratory (AFRL), Metna Company, and Michigan State [26] Choi EG, Shin YS. The structural behavior and simplified thermal analysis of
normal-strength and high-strength concrete beams under fire. Eng Struct
University for undertaking this research. Any opinions, findings, con- 2011;33:1123–32.
clusions, or recommendations expressed in this paper are those of the [27] Liang X, Wu C, Yang Y, Li Z. Experimental study on ultra-high performance concrete
authors and do not reflect the position, policy; or views of the sponsors. with high fire resistance under simultaneous effect of elevated temperature and
impact loading. Cem Concr Compos 2019;98:29–38.
The authors would like to thank Sia Ravanbakhsh for his help in [28] Kahanji C, Ali F, Nadjai A, Alam N. Effect of curing temperature on the behaviour of
carrying out experimental studies. Also, the authors would like to thank UHPFRC at elevated temperatures. Constr Build Mater 2018;182:670–81.
Prof. Parviz Soroushian for his guidance on UHPFRC mix design and [29] Hou X, Ren P, Rong Q, Zheng W, Zhan Y. Comparative fire behavior of reinforced
RPC and NSC simply supported beams. Eng Struct 2019;185:122–40.
fabrication of UHPFRC beams.
[30] Chen Y, Matalkah F, Yu Y, Rankothge W, Balachandra A, Soroushian P.
Experimental investigations of the dimensional stability and durability of ultra-
Appendix A. Supplementary material high-performance concrete. Adv Mater Sci 2017;6:1–8.
[31] Kodur V, Solhmirzaei R, Agrawal A, Aziz EM, Soroushian P. Analysis of flexural and
shear resistance of ultra high performance fiber reinforced concrete beams without
Supplementary data to this article can be found online at https:// stirrups. Eng Struct 2018;174:873–84.
doi.org/10.1016/j.engstruct.2020.110316. [32] Sbia LA, Peyvandi A, Lu J, Abideen S, Weerasiri RR, Balachandra AM, et al.
Production methods for reliable construction of ultra-high-performance concrete
(UHPC) structures. Mater Struct 2017;50:7.
References [33] Chen T, Gao X, Ren M. Effects of autoclave curing and fly ash on mechanical
properties of ultra-high performance concrete. Constr Build Mater
[1] Yoo D-Y, Yoon Y-S. A review on structural behavior, design, and application of 2018;158:864–72.
ultra-high-performance fiber-reinforced concrete. Int J Concr Struct Mater [34] Bahedh MA, Jaafar MS. Ultra high-performance concrete utilizing fly ash as cement
2016;10:125–42. replacement under autoclaving technique. Case Stud Constr Mater 2018;9:e00202.
[2] Gu C, Ye G, Sun W. Ultrahigh performance concrete-properties, applications and [35] Sbia LA, Peyvandi A, Harsini I, Lu J, Abideen SU, Weerasiri RR, et al. Study on field
perspectives. Sci China Technol Sci 2015;58:587–99. thermal curing of ultra-high-performance concrete employing heat of hydration.
[3] Yoo D-Y, Banthia N, Yoon Y-S. Flexural behavior of ultra-high-performance fiber- ACI Mater J 2017;114:733–44.
reinforced concrete beams reinforced with GFRP and steel rebars. Eng Struct [36] Shakya AM, Kodur VKR. Response of precast prestressed concrete hollowcore slabs
2016;111:246–62. under fire conditions. Eng Struct 2015;87:126–38.
[4] Wang W, Wu C, Liu Z. Compressive behavior of hybrid double-skin tubular columns [37] ASTM E119-19. Standard Test Methods for Fire Tests of Building Construction and
with ultra-high performance fiber-reinforced concrete (UHPFRC). Eng Struct Materials. West Conshohocken, PA: American Society of Testing Materials; 2019.
2019;180:419–41. [38] Wu HJ, Lie TT, Hu JY. Fire resistance of beam slab specimens: experimental studies.
[5] Wang W, Wu C, Li J, Liu Z, Zhi X. Lateral impact behavior of double-skin steel Internal Report/National Research Council of Canada, Institute for Research in
tubular (DST) members with ultra-high performance fiber-reinforced concrete Construction; No 641; 1993.
(UHPFRC). Thin-Walled Struct 2019;144:106351. [39] Williams B, Kodur V, Green MF, Bisby L. Fire endurance of fiber-reinforced polymer
[6] Wang W, Wu C, Li J, Liu Z, Lv Y. Behavior of ultra-high performance fiber-re- strengthened concrete T-beams. ACI Struct J 2008;105:60.
inforced concrete (UHPFRC) filled steel tubular members under lateral impact [40] Lie TT. Structural fire protection. American Society of Civil Engineers 1992.
loading. Int J Impact Eng 2019;132:103314. [41] Kodur VKR, Alogla SM. Effect of high-temperature transient creep on response of
[7] Japan Society of Civil Engineers (JSCE). Recommendations for design and con- reinforced concrete columns in fire. Mater Struct 2017;50:27.
struction of ultra high strength fiber reinforced concrete structures; 2006. [42] Dwaikat MB, Kodur VKR. Fire induced spalling in high strength concrete beams.
[8] Ozawa M, Subedi Parajuli S, Uchida Y, Zhou B. Preventive effects of polypropylene Fire Technol 2010;46:251.
and jute fibers on spalling of UHPC at high temperatures in combination with waste [43] Hertz KD. Limits of spalling of fire-exposed concrete. Fire Saf J 2003;38:103–16.
12