Materials 12 03869

materials
Review
Role of Polypropylene Fibres in Concrete Spalling
Risk Mitigation in Fire and Test Methods of Fibres
Effectiveness Evaluation
Izabela Hager * and Katarzyna Mróz
Faculty of Civil Engineering, Cracow University of Technology, 24 Warszawska St., 31-155 Cracow, Poland;
kmroz@pk.edu.pl
* Correspondence: ihager@pk.edu.pl; Tel.: +48-12-628-23-67

Received: 1 November 2019; Accepted: 20 November 2019; Published: 23 November 2019
Abstract: The explosive behaviour of concrete in fire is observed in rapidly heated concrete. The main
factors controlling the occurrence of spalling are related to the material’s low porosity and high density
as well as the limited ability to transport gases and liquids. Thus, for high-strength, ultrahigh-strength,
and reactive powder concrete, the risk of spalling is much higher than for normal-strength concrete.
The paper presents the discussion on the leading hypothesis concerning the occurrence of concrete
spalling. Moreover, the methods for spalling prevention, such as polypropylene fibre application,
which has been found to be an effective technological solution for preventing the occurrence of spalling,
are presented. Various tests and testing protocols are used to screen concrete mixes propensity toward
spalling and to evaluate the polypropylene fibres’ effectiveness in spalling risk mitigation. The most
effective testing methods were selected and their advantages were presented in the paper. The review
was based mainly on the authors’ experiences regarding high performance concrete, reactive powder
concrete testing, and observations on the effect of polypropylene fibres on material behaviour at
high temperature.
Keywords: concrete spalling; temperature; fire; fibres; gas pore pressure
1. Introduction
During a fire, it is possible that concrete, which is an incombustible material [1], will behave
explosively due to the occurrence of spalling. Spalling is defined as an unexpected loss of concrete
cover that has a violent or non-violent character [2,3]. Spalling behaviour may endanger the
load-bearing capacity of concrete elements because of a progressive cross-section loss and expose the
steel reinforcement, which is sensitive to rising temperatures [4]. In a fire, this may lead to a decrease
of the element capacity and even to its collapse [5]. Due to the fact that the effects of fire in concrete
structures may be very severe, a number of experimental studies were carried out to find methods
to reduce concrete propensity to spalling due to fire exposure [6–9]. The addition of polypropylene
fibres (PP) to the concrete mix was found to give promising results. The polypropylene fibres being
present in concrete does not significantly influence its mechanical and physical properties at a normal
temperature [10]. However, when the temperature rises due to fire, the permeability of concrete with
fibres increases as much as 50 times at 200 ◦ C when compared to the concrete without fibres [11]. When
fibres melt, a network of pores is created, which increases the concrete permeability. This facilitates the
transport of water vapour, releases the water vapour pressure, and mitigates the concrete spalling risk
in fire.
Materials 2019, 12, 3869; doi:10.3390/ma12233869 www.mdpi.com/journal/materials

Materials 2019, 12, 3869 2 of 20
Materials 2019, 12, x FOR PEER REVIEW 2 of 21
2.
2. Concrete
Concrete Spalling
Spalling Due
Due to
to Fire
Fire
Spalling
Spallingmay maytake
takethetheform
formofof aa very
very slow
slow detachment
detachmentof of the
the surface
surface layer
layer ofof the
the material,
material, asas well
well
as of pieces of material, which break away in a rapid, explosive manner from the
as of pieces of material, which break away in a rapid, explosive manner from the element [3]. Different element [3]. Different
mechanisms
mechanismsof of spalling
spalling maymay lead
lead toto various
various types
types ofof spalling.
spalling. The main differences
differences between
between spalling
spalling
forms
forms are are related
related toto the
the time
time and
and frequency
frequency of of the
the occurrence
occurrence of of spalling,
spalling, the
the size
size of
of concrete
concrete pieces
pieces
that
that are detached from the structure, and the energy that accompanies the spalling events in the event
are detached from the structure, and the energy that accompanies the spalling events in the event
of
of aa fire.
fire.
Observations
Observations of of fire
fire damage
damage to to real
real structures
structures and
and laboratory
laboratory firefire tests
tests have
have mademade itit possible
possible toto
distinguish five different types of spalling, which were recently summarised
distinguish five different types of spalling, which were recently summarised by the authors [12]. The by the authors [12]. The
following
followingsubdivision
subdivisionwas wasproposed,
proposed,as asshown
shownin inFigure
Figure1.1.The
Thetimes
timest1t1 and
and tt22 indicated
indicated in in the
the figure
figure
relate to different points in time showing the progressive character
relate to different points in time showing the progressive character spalling type. spalling type. In this subdivision,
the
the corner
corner (A),(A), the
the aggregate
aggregate(B), (B),the
thesurface
surface(C),
(C),and
andthetheexplosive
explosivespalling
spalling(D) (D)were
wererepresented.
represented.
Figure 1.
Figure 1. Forms
Formsofofdamagedamageobserved
observedin in various types
various of of
types concrete spalling
concrete in fire
spalling conditions:
in fire (A)
conditions:
Aggregate spalling. (B) Surface spalling (progressive and continuous in nature).
(A) Aggregate spalling. (B) Surface spalling (progressive and continuous in nature). (C) Explosive (C) Explosive
spalling (progressive,
spalling (progressive, discrete nature: tt11,, tt22). (D) Corner
discrete nature: Corner spalling.
spalling. (E) Surface
Surface peeling
peeling away
away during
during
cooling, where
cooling, <2t.2 .
wherett11<t
The
The occurrence
occurrence of of different
differenttypes
typesofofspalling
spallingis is accompanied
accompanied by by characteristic
characteristic sounds—popping
sounds—popping
for
for the aggregate and surface spalling (A, B) and a loud bang for the explosiveSome
the aggregate and surface spalling (A, B) and a loud bang for the explosive one (C). regularity
one (C). Some
in the occurrence
regularity in the of spalling, according
occurrence of spalling,toaccording
the type oftospalling,
the typewas observed.was
of spalling, During the first
observed. minutes
During the
of a fire
first or fireof
minutes test, spalling
a fire or fireis test,
mostspalling
likely toistake place
most in the
likely aggregate
to take place and
in thesurface.
aggregateThen,andafter a few
surface.
minutes,
Then, after a louda fewbang of explosive
minutes, a loudbehaviour appears, and
bang of explosive afterwards,
behaviour appears,cornerandspalling takes corner
afterwards, place.
Usually,
spalling aggregate
takes place. spalling
Usually,(A)aggregate
does not affect the (A)
spalling load-bearing
does not affectcapacity
theofload-bearing
a structure and does of
capacity nota
lead to a significant
structure and does not cross-section loss. It iscross-section
lead to a significant believed that theItmechanism
loss. is believed thatbehindthe aggregate
mechanismspalling
behind
is similar to
aggregate a flaming
spalling stone procedure,
is similar to a flamingwhich
stone results
procedure,fromwhich
quartz transformation.
results from quartzSurface spalling
transformation.
(B) affects
Surface the thin
spalling (B)and flat the
affects fragments
thin andofflat
thefragments
near-surface of thelayers of concrete
near-surface elements,
layers suchelements,
of concrete as walls
and ceilings, which become detached from the surface as a result
such as walls and ceilings, which become detached from the surface as a result of the tensileof the tensile stress σ between
T stress σT
the heated
between the layer, which
heated expands,
layer, and colder
which expands, andconcrete layers. This
colder concrete is usually
layers. not dangerous
This is usually for the
not dangerous
load-bearing capacitycapacity
for the load-bearing of structures, since it generally
of structures, does notdoes
since it generally resultnotinresult
the loss of aloss
in the significant part of
of a significant
the
partcross-section and doesand
of the cross-section notdoesbecomenot exposed
become in reinforcing
exposed steel. Explosive
in reinforcing spalling (C),
steel. Explosive which(C),
spalling is
accompanied by characteristic
which is accompanied explosion sounds,
by characteristic constitutes
explosion sounds,a violent chipping
constitutes off of chipping
a violent concrete pieces,
off of
with largepieces,
concrete surfacewith areaslarge
and spalling
surface depth
areas reaching
and spallingup todepth
aroundreaching
a dozen up centimetres
to around for aa single
dozen
centimetres
piece for concrete.
of spalled a single piece of spalled
Explosive concrete.
spalling Explosive
of concrete spalling
results of concrete
in reinforcing steelresults in reinforcing
becoming exposed
steel becoming exposed and often results in the complete destruction of the concrete layer. Corner
spalling (D) is observed in elements in which two orthogonal surfaces, which form a corner (in
Materials 2019, 12, 3869 3 of 20
and often 2019,

Materials results
12, xin
FORthePEER
complete
REVIEWdestruction of the concrete layer. Corner spalling (D) is observed 3 of 21 in
elements in which two orthogonal surfaces, which form a corner (in columns, beams, beam-column
columns,
joints, beams,
beam-wall beam-column
joints, joints, beam-wall
etc.), are heated. In a typicaljoints, etc.), are
fire, corner heated.
spalling In a typical
usually fire, for
takes place corner
about
spalling usually takes place for about 30 min after the fire has started. Corners
30 min after the fire has started. Corners are heated more quickly, which results in higher thermal are heated more
quickly,atwhich
gradients results and,
such points, in higher thermal gradients
consequently, corners are at damaged
such points,moreand, consequently,
quickly corners
as well. The peelingare of
damaged
concrete more
after a firequickly as well. Theheated
(E) characterises peeling of concrete
concrete, where after a fire (E)and
dolomite characterises
calcareousheated concrete,
aggregates occur
where dolomite and calcareous aggregates occur during the cooling of the structural element. This
during the cooling of the structural element. This may take place a few hours or even days after the fire
may take place a few hours or even days after the fire has ended. The peeling of damaged concrete
has ended. The peeling of damaged concrete layers after the fire is the result of an inversed temperature
layers after the fire is the result of an inversed temperature gradient during cooling. Nevertheless,
gradient during cooling. Nevertheless, the main cause of post-fire concrete peeling (concrete sloughing
the main cause of post-fire concrete peeling (concrete sloughing off) is the hydration of calcium oxide,
off) is the hydration of calcium oxide, formed during decomposition of portlandite, calcareous, and
formed during decomposition of portlandite, calcareous, and limestone aggregate, which is
limestone aggregate,
accompanied by thewhich is accompanied
expansion by product
of the reaction the expansion
[13]. of the reaction product [13].
2.1. Spalling Mechanism Hypothesis
2.1. Spalling Mechanism Hypothesis
Several hypotheses concerning the occurrence of spalling have been proposed [3,4]. Currently,
Several hypotheses concerning the occurrence of spalling have been proposed [3,4]. Currently,
it isit believed
is believed that two
that twosimultaneously
simultaneouslyoccurring
occurring and interrelatedmechanisms
and interrelated mechanismsare areresponsible
responsible forfor
itsits
occurrence, which are the thermal-mechanical and hygro-thermal mechanisms.
occurrence, which are the thermal-mechanical and hygro-thermal mechanisms. The main reasons are The main reasons
arebelieved
believedtotobe belinked
linkedtotothe
the thermal
thermalstresses
stresses(thermo-mechanical),
(thermo-mechanical), originating
originating from
from thethe thermal
thermal
gradient
gradient in in
thethe
element
element surface and
surface cold
and inner
cold layer,
inner and
layer, to to
and a water vapour
a water vapourpore pressure
pore pressureincrease
increase[14].
In [14].
Figure 2, we have represented the temperature and moisture profiles as well as the tensile
In Figure 2, we have represented the temperature and moisture profiles as well as the tensile strength
change
strengthprofile.
changeFigure 2 represents
profile. a fire situation
Figure 2 represents and an element
a fire situation heated on
and an element one side.
heated on one side.
Figure
Figure 2. 2. Mechanismofofconcrete
Mechanism concrete spalling
spalling in
in fire
fire and
and the
the pore
pore pressure
pressuredistribution
distributionininnon-fibrous
non-fibrous
concrete and concrete with PP fibres.
concrete and concrete with PP fibres.
During
During a fire,a atemperature
a fire, temperatureincrease
increase leads
leads to
to concrete
concrete drying.
drying. The
Thethermal
thermalgradient
gradientis is
also
also
responsible
responsible forfor water
water migration.As
migration. Asthe
the temperature
temperature increases,
increases,the
thewater
watermoves
movestoward colder
toward zones
colder zones
duedue to the
to the thermo-capillary
thermo-capillary effect.
effect. Moreover,
Moreover, the the
waterwater vapour
vapour that that is produced
is produced during
during heating
heating moves
moves through an open network of concrete pores. The gas transport continues,
through an open network of concrete pores. The gas transport continues, until the thermodynamic until the
thermodynamic conditions are satisfied for vapour condensation. Under this condition, a layer of
conditions are satisfied for vapour condensation. Under this condition, a layer of material, saturated
material, saturated with water, is formed, which is called a moisture clog zone. The theory of a
with water, is formed, which is called a moisture clog zone. The theory of a moisture clog was
moisture clog was confirmed experimentally, in Reference [15], from an experiment, in which, after
confirmed experimentally, in Reference [15], from an experiment, in which, after 15 min of heating, the
15 min of heating, the concrete cubes were split, showing the concrete zones saturated with water.
concrete cubes were split, showing the concrete zones saturated with water. Moreover, the advantages
Moreover, the advantages of neutron tomography have also been proven in studying the mechanism
of the phenomenon [16] and allowing for moisture movement, as well as the local moisture
porous network and evaporates quickly. In denser and less permeable concrete, the moisture clog is
generated sooner and closer to the heated surface. A fully saturated and almost impermeable layer
of the material moves through the material, which we have represented in the pictogram shown in
Figure 3. After some time, in the concrete that is most exposed to a temperature increase, a dry zone
is formed, in which the cement paste is dehydrated. Subsequently, a drying zone, where no free water
Materials 2019, 12, 3869 4 of 20
is present, adjacent to the saturated and impermeable concrete layer, is formed.
As a consequence, the water-clogged zone is more prone to forming in high-strength concretes
(HSC)
of neutron and ultra-high-performance
tomography have also beenconcretes
proven(UHPC), in which
in studying thethe permeability
mechanism is 0.52E-17
of the [17]. Due
phenomenon [16]
to the low permeability of the material, a water clog zone will form quickly. Using
and allowing for moisture movement, as well as the local moisture accumulation, in a heated concrete the mechanism of
spallingobservation.
specimen in an elementUnderheatedthose
on one face, we have
conditions, represented
the water vapouritpore
in Figure 2, which
pressure startsalso takes into
to increase, and
account the material strength reduction in temperature. This is approximately 120 °C, as reported in
the pressure peak is located at the saturated layer. The pressure of the water vapour (P) is exerted on
authors’ research [18]. The reduction of the mechanical performance of concrete is mainly due to
the material skeleton and reaches a maximum level in the water-clogged zone. When the tensile stress
thehydral gradient inside the heated concrete element or specimen.
increases to a critical level, which is higher than the tensile strength of concrete, spalling may begin. In
It needs to be highlighted that the external load has a critical effect on the spalling behaviour of
normal-strength concrete (NSC), the gas permeability measured by authors is 2.22E-16 [17]. Therefore,
a concrete element. It was observed that the external compressive load or pre-stress enhances the risk
theofwater vapour moves with ease through the porous network and evaporates quickly. In denser
spalling [19–21]. An external compressive load might be generated during the fire, as a
and less permeable
consequence of the concrete,
restrainedtheelement
moisture clog is generated
expansion. sooner
The restraint mayandbecloser
due totothe
thepresence
heated surface.
of other A
fully saturated and almost impermeable layer of the material moves through the
elements of the structure or the cold part of the material surrounding the heated zone (localised material, which
fire).we
have represented
In the in the that
case of elements pictogram shown
are tense, it wasinnoticed
Figurethat 3. After
crack some time, insignificantly
development the concretereduces
that is the
most
exposed to a temperature increase, a dry zone is formed, in which the cement paste
occurrence of spalling events [6]. The crack formation allows the water vapour to escape and the gas is dehydrated.
Subsequently,
pore pressure a drying zone, where
to be released. no free
Concrete waterisisthe
spalling present, adjacent
result of to the saturated
pore pressure occurrence,andcompression
impermeable
concrete layer,exposed
in the zone is formed.
to fire, and internal cracks developing in the plane of the heated surface [3].
Figure3.3.Development
Figure Developmentof
of the
the hydral
hydral processes
processes in
inconcrete
concreteexposed
exposedtotofire.
fire.
2.2.As
Pore Pressure Measurements
a consequence, the water-clogged zone is more prone to forming in high-strength concretes
(HSC) and ultra-high-performance
The gas pore pressures can be concretes
measured (UHPC), in which
while heating the pressure
using permeability
gagesisembedded
0.52E-17 [17]. Due
in the
to the low permeability
concrete element duringof the material,
casting. a water
Kalifa et al.clog zone willmeasured
successfully form quickly. Using
the gas thepressure
pore mechanismand of
spalling in an element heated on one face, we have represented it in Figure 2, which
temperature development during heating [22], reporting that the water vapour pore pressure is much also takes into
account the material strength reduction in temperature. This is approximately 120 ◦ C, as reported
higher in HPC than in NSC. The values measured were 3.8 MPa and 1.8 MPa for HPC and NSC,
in authors’ research
respectively. [18].(PTM-
PTM tests The reduction
Pressure, of the mechanical
Temperature, Massperformance
change) allowof forconcrete
the massisloss
mainly due to
of heated
thehydral
slab (M),gradient inside
temperature (T),the
andheated concrete
the pore pressureelement or specimen.Specially designed sintered metal
(P) measurements.
It needs
gauges are to be highlighted
placed in a concretethat the external
element loaddepths
at different has a critical effect
and allow onpressure
pore the spalling
and behaviour
temperature of a
concrete element. It was observed that the external compressive load or pre-stress enhances the risk of
spalling [19–21]. An external compressive load might be generated during the fire, as a consequence of
the restrained element expansion. The restraint may be due to the presence of other elements of the
structure or the cold part of the material surrounding the heated zone (localised fire). In the case of
elements that are tense, it was noticed that crack development significantly reduces the occurrence of
spalling events [6]. The crack formation allows the water vapour to escape and the gas pore pressure
to be released. Concrete spalling is the result of pore pressure occurrence, compression in the zone
exposed to fire, and internal cracks developing in the plane of the heated surface [3].
2.2. Pore Pressure Measurements

The gas pore pressures can be measured while heating using pressure gages embedded in the
concrete element during casting. Kalifa et al. successfully measured the gas pore pressure and
temperature development during heating [22], reporting that the water vapour pore pressure is much
changes to be measured at the same time. The specimen, which is a small slab (0.3 × 0.3 × 0.12 m3), is
placed horizontally and is directly exposed to an elevated temperature on the upper surface using
radiant
Materials heaters,
2019, 12, 3869as shown in Figure 4 [10]. Further experiments have confirmed that, in concrete with 5 of 20
a lower moisture content, the measured pressure is lower [23]. However, in a fully saturated
specimen, this tendency is not found. The pressure is released quite quickly due to the macro cracking
higher in HPCwhich
of concrete, than in wasNSC. The values
observed duringmeasured were 3.8
the experiment MPa
[23]. and 1.8itMPa
Recently, was for
alsoHPC and NSC,
reported, in
respectively. PTM tests (PTM- Pressure, Temperature, Mass change) allow
Reference [24], that, when very rapid heating (120 °C/min) is applied, the developed pressures for the mass loss of heated
are
slab (M),times
three temperature (T), and
smaller than thosethe
of pore pressure
concrete heated (P)with
measurements. Specially
a rate of heating designed sintered metal
at 10 °C/min.
gauges Theare placed in a concrete element at different depths and allow
instrumentation and measurements in PTM tests are very sophisticated. Today,pore pressure and temperature
PTM is
changes
considered to be the most advanced material testing method, helping the research 0.3
to be measured at the same time. The specimen, which is a small slab (0.3 × × 0.12 m3to
community ), is
placed horizontally
observe and is directly
the pore pressure exposed
development to conditions
in fire an elevated andtemperature
to understand onthe
thephysical
upper surface
mechanismusing
radiant
behindheaters,
it. Thisastechnique
shown in wasFigure 4 [10].
also usedFurther
to measureexperiments have confirmed
the effectiveness that,addition
of PP fibre in concrete with a
in pore
pressure
lower build-up
moisture in heated
content, concrete and
the measured associate
pressure it with[23].
is lower the spalling
However, propensity
in a fullyofsaturated
certain concretes.
specimen,
thisIntendency
Figure 4 is [10],
notthe PP fibre
found. The action
pressureis presented,
is released which allows the
quite quickly duegas pore
to the pressure
macro to beof
cracking released
concrete,
effectively
which and reduces
was observed duringthethe
pressure picks [23].
experiment to 1 MPa, withit3 was
Recently, kg/malso
3 of fibrillated PP fibres. The PTM
reported, in Reference [24], that,
whentestvery
also rapid
servesheating
as a calibration test for
(120 ◦ C/min) all thermo-hydro-mechanical
is applied, the developed pressures modelsare that
threeare ablesmaller
times to predict
than
the water clog presence and the gas pore pressure
those of concrete heated with a rate of heating at 10 C/min. increase.
◦
Figure 4. 4.
Figure Left:PTM
Left: PTMKalifa
Kalifaset-up
set-up[10],
[10],right: Pressure and
right: Pressure andtemperature
temperaturemeasurements
measurementsinin heated
heated HPC
HPC
concrete:
concrete: left—withPP
Left—with PPfibres,
fibres, right—without
right—without PP PPfibres[10].
fibres [10].
The
2.3. instrumentation
Standard and measurements
Spalling Prevention in PTM tests are very sophisticated. Today, PTM is
Methods and Recommendations
considered to be the most advanced material testing method, helping the research community to
The construction element in case of fire must present a load-bearing capacity for a certain time
observe the pore pressure development in fire conditions and to understand the physical mechanism
to allow for the evacuation of the building. Moreover, the fire brigade must conduct the intervention
behind it. This technique was also used to measure the effectiveness of PP fibre addition in pore
in safe conditions. In some situations, the spalling behaviour of concrete may jeopardise the safety of
pressure build-up in heated concrete and associate it with the spalling propensity of certain concretes.
concrete works. For beams, slabs, and all tensile members, crack occurrence may lead to a release of
In Figure
gas pore4 pressure,
[10], the which
PP fibre action is
decreases thepresented, which
spalling risk. allows the
The surfaces thatgas
arepore pressure
directly exposedto to
befire
released
and
effectively and reduces the pressure picks to 1 MPa, with 3 kg/m 3 of fibrillated PP fibres. The PTM test
work under compression require verification with regard to the spalling of concrete cover. In
also
Reference [25], it is shown that, when the moisture content of concrete is lower than 3% by weight,the
serves as a calibration test for all thermo-hydro-mechanical models that are able to predict
water clog presence
explosive spalling and the gastopore
is unlikely pressure
occur. increase.
Nevertheless, at more than 3%, a more accurate assessment of
the moisture content, nature of the aggregate, permeability of the concrete, and heating rate should
2.3.beStandard Spalling Prevention Methods and Recommendations
considered in order to check the spalling risk. The moisture content threshold may differ from 3%,
and this
The informationelement
construction can be found inof
in case thefire
National Annex atoload-bearing
must present EC2. The recommended
capacity forvalue is 3%.
a certain time to
allow for Forthe
all evacuation
structural elements made of HPC,
of the building. and forthe
Moreover, concretes containing
fire brigade mustsilica fumethe
conduct in intervention
an amount
higher
in safe than 6% atInthe
conditions. design
some stage, the
situations, therisk of spalling
spalling behaviourshouldof be taken into
concrete mayaccount.
jeopardiseOnetheof safety
the
following four methods should be applied. In the first prevention method, a reinforcement
of concrete works. For beams, slabs, and all tensile members, crack occurrence may lead to a release mesh,
withpore
of gas a nominal cover
pressure, of 15 decreases
which mm consisting of wires risk.
the spalling of a diameter of 2 mm
The surfaces thatand
areadirectly
pitch size lower than
exposed to fire
50 × 50 mm , is used. Additionally, the nominal cover of the main reinforcement should be 40 mm. In
2
and work under compression require verification with regard to the spalling of concrete cover. In
Reference [25], it is shown that, when the moisture content of concrete is lower than 3% by weight,
explosive spalling is unlikely to occur. Nevertheless, at more than 3%, a more accurate assessment of
the moisture content, nature of the aggregate, permeability of the concrete, and heating rate should be
considered in order to check the spalling risk. The moisture content threshold may differ from 3%, and
this information can be found in the National Annex to EC2. The recommended value is 3%.
Materials 2019, 12, 3869 6 of 20
For all structural elements made of HPC, and for concretes containing silica fume in an amount
higher than 6% at the design stage, the risk of spalling should be taken into account. One of the
following four methods should be applied. In the first prevention method, a reinforcement mesh,
with a nominal cover of 15 mm consisting of wires of a diameter of 2 mm and a pitch size lower than
50 × 50 mm2 , is used. Additionally, the nominal cover of the main reinforcement should be 40 mm. In
the next method, a concrete mix should be used, to which it has been demonstrated experimentally
that no spalling occurs under fire exposure. The following technique consists of the application of
protective layers, to which it is demonstrated that no spalling of concrete occurs under fire exposure.
Other spalling prevention techniques have been tested on existing structures, such as the
application of heat protection panels, made of wool or gypsum, that slow down the heating process,
which is a commonly used solution [26]. This type of protection is considered to be passive fire
protection. In the case of new concrete structures, steel fibre addition was proposed to increase the
tensile strength of concrete and steel mesh protection in order to reinforce the outer layer of the
concrete structural element, but the effectiveness of this method is limited. In order to prevent concrete
spalling in fire, the application of organic fibres is proven to be the best solution. For this purpose,
polypropylene (PP) fibres are added to the concrete mix. The fibres are used in an amount of 0.1% to
0.2% of the volume of the concrete. Both types—fibrillated and monofilament fibres—were reported to
be efficient in spalling mitigation. The fibrillated fibres are obtained by cutting polypropylene sheets.
Therefore, their cross-section is rectangular. Of the presented fibres, with a section of 50 × 100 µm,
various quantities were investigated, from 0.5 to 3.0 kg/m3 of concrete. Nevertheless, for a high PP
fibre content, a mixing problem is observed, and homogeneity in the concrete mix is difficult to obtain.
It was reported that an increase in the total volume of the cement paste content by about 15% enables
the fibres to be covered with cement paste. The effectiveness of PP fibre addition, as a technological
solution that prevents spalling, was confirmed by numerous experimental studies [27–29]. Moreover,
the addition of 2 kg/m3 of fibres is also recommended by the EUROCODE 2 [4] standard to avoid
fire spalling.
3. Polypropylene Fibre Action
3.1. PP Fibre Behaviour at High Temperatures

PP fibres melt at a temperature of around 170 ◦ C, i.e., lower than the temperature at which the
maximum water vapour pressure in concrete is reached (190–260 ◦ C). In the market, low-temperature
melting PB Eurofiber HPR (Heat Prompt Reaction) fibres can be found, and, for this fibre variation,
melting occurs more quickly because their melting temperature is lower.
At 350 ◦ C, polypropylene burns and turns into CO2 and H2 O. In Figure 5a, a differential thermal
analysis of polypropylene fibres was presented [10]. The peaks on the curve correspond to the melting
point (171 ◦ C), evaporation temperature (341 ◦ C), and carbonation temperature (456 ◦ C). The fibres
melt at 171 ◦ C and are totally or partially absorbed by the porous network of the cement matrix. At
341 ◦ C, the polypropylene vaporises. The melting temperature can be evaluated using an Automated
Melting Point Apparatus MPA100, which allows the melting process to be observed and the range of
temperatures at which melting takes place to be precisely determined, as shown in Figure 5b.
analysis of polypropylene fibres was presented [10]. The peaks on the curve correspond to the
melting point (171 °C), evaporation temperature (341 °C), and carbonation temperature (456 °C). The
fibres melt at 171 °C and are totally or partially absorbed by the porous network of the cement matrix.
At 341 °C, the polypropylene vaporises. The melting temperature can be evaluated using an
Automated Melting Point Apparatus MPA100, which allows the melting process to be observed and
Materials
the range2019,of12, 3869
temperatures 7 of 20
at which melting takes place to be precisely determined, as shown in Figure
5b.
-20
heat flow (µV)
-40
340°C
-60
170°C
-80
457°C
-100
0 100 200 300 400 500
temperature (°C)

(a) (b)
Figure 5. (a) Differential thermal analysis of polypropylene fibres and (b) observation of the melting
Figure 5. (a) Differential thermal analysis of polypropylene fibres and (b) observation of the melting
process and the melting temperature range for polypropylene fibres.
process and the melting temperature range for polypropylene fibres.
In our study, we compared two PP fibres types, known as PP fibrillated polypropylene fibres
In our study, we compared two PP fibres types, known as PP fibrillated polypropylene fibres and
and PB Eurofiber, with a low melting temperature, which have shown advantageous melting
PB Eurofiber, with a low melting temperature, which have shown advantageous melting temperatures
temperatures (162 °C compared to 170 °C for fibrillated PP fibres).
(162 ◦ C compared to 170 ◦ C for fibrillated PP fibres).
After melting, the polypropylene is partly absorbed by the cement matrix. We have represented
After melting, the polypropylene is partly absorbed by the cement matrix. We have represented it
it in an experiment, in which several fibres are heated on a concrete disk. The thermocouple shows
in an experiment, in which several fibres are heated on a concrete disk. The thermocouple shows the
the temperature at the location of the fibres, as shown in Figure 6.
temperature at the location of the fibres, as shown in Figure 6.
Figure 6. Fibrillated polypropylene fibres melting and the absorption of polypropylene by the cement
Figure 6. Fibrillated polypropylene fibres melting and the absorption of polypropylene by the cement
matrix and PP burning.
matrix and PP burning.
Similarly, when fibres melt and are absorbed by a cement matrix, a network of pores is created,
which Similarly,
increaseswhen fibres melt
the concrete and are absorbed
permeability. by a cement
This facilitates matrix,of
the transport a network of pores
water vapour. Theischannels
created,
which increases the concrete permeability. This facilitates the transport of water vapour.
that remain after fibres have melted add up to the natural porosity of the cement paste and the porosity The channels
that
of theremain aftertransition
interfacial fibres have zone melted add up to
(ITZ) between thethe natural
cement porosity
paste of the cement paste and the
and aggregates.
porosity
Nevertheless, as reported by Kalifa et al. [10], several micro-cracksand
of the interfacial transition zone (ITZ) between the cement paste aggregates.
around fibres have been
Nevertheless, as reported by Kalifa et al. [10], several
observed. The cracks observed were 1 µm in size for concrete with fibres and 10 micro-cracks around fibres
µm in have
sizebeen
for
observed. The
non-fibrous cracks observed
concrete. Mercury were 1 μmporosity
intrusion in size for concrete with
measurements fibres anda 10
confirmed μm in size
significant for non-
increase in
fibrous
the concrete.
quantity Mercury
of pores with intrusion
diametersporosity measurements
of less than 1 µm in HPC confirmed
concrete,a significant
when fibres increase in the
were heated
quantity
to 180 ◦ C.ofThepores with observed
change diametersdoes of less
not than
stem1 μm
from in the
HPC concrete, porosity
additional when fibres werebyheated
caused to 180
the melting
°C.polypropylene
of The change observed
fibres butdoes not stem
is, instead, thefrom
resultthe
of anadditional
increase porosity caused by the
in the micro-cracking of melting
the heated of
polypropylene fibres
concrete with PP fibres. but is, instead, the result of an increase in the micro-cracking of the heated
concrete
Twowith PP fibres.
hypotheses concerning this additional cracking in fibrous concrete were put forward. The
Two hypotheses concerning
first is that the fibres, while being this additional
heated, cracking
dilate by 7–10% in fibrous
[29], whichconcrete were local
can generate put forward. The
stresses and
first is that the fibres, while being heated, dilate by 7–10% [29], which can generate
nucleation of cracks. The second is that the fibre bed promotes the local formation of cracks and local stresses and
nucleation of cracks. The second is that the fibre bed promotes the local formation of cracks and
encourages cracking, which we have represented in Figure 7. The crack formation around the fibres
has the beneficial effect of reducing the water vapour pressure in the concrete pores. It is believed
that the fibres’ melted network adds up to the natural porosity of the material, i.e., as estimated by
Bentz [30] from a numerical analysis. The amount of PP fibres when added to the concrete allow the
quantity of pores with diameters of less than 1 μm in HPC concrete, when fibres were heated to 180
°C. The change observed does not stem from the additional porosity caused by the melting of
polypropylene fibres but is, instead, the result of an increase in the micro-cracking of the heated
concrete with PP fibres.
Two
Materials hypotheses
2019, 12, 3869 concerning this additional cracking in fibrous concrete were put forward. The
8 of 20
first is that the fibres, while being heated, dilate by 7–10% [29], which can generate local stresses and
nucleation of cracks. The second is that the fibre bed promotes the local formation of cracks and
percolation threshold to reach 0.1% of fibres, relative to the concrete volume.
percolation threshold to reach 0.1% of fibres, relative to the concrete volume.
(a) (b) (c)
Figure 7. Polypropylene fibre cross-section in concrete (a) non-heated, (b) heated to 150 ◦ C and (c) after
600 ◦ C. SEM image at 50x.
3.2. Change of Transport Properties in Heated Concretes with PP Fibres

In order to better understand the thermo-hydric process leading to the spalling phenomenon and
explain the role of polypropylene fibres in spalling mitigation, Kalifa’s PTM experiment was employed.
In this small-scale test, the influence of PP fibres on temperature distribution, pore pressure build-up,
and mass loss in a heated specimen may be tested. According to the experiments conducted on concrete
(fc = 100MPa), with different amounts of PP fibres (0–3kg/m3 ), the amount of 1.1 kg/m3 in concrete
heated on one side leads to a pore pressure development of about 2 MPa, which is close to the one
recorded on NSC (where spalling usually does not occur). Pressure and permeability measurements
suggest that a dosage of around 1 kg/m3 should be sufficient for this type of concrete and within this
range of temperatures. This value may not be sufficient for other types of concrete, with a smaller
porosity, as in UHPC [10]. Similarly, Mindeguia et al., by performing a PTM test, showed that, in
normal-strength concretes (fc = 40 MPa and fc = 60 MPa) exposed to moderate heating, the addition of
2 kg/m3 of PP fibres caused a three-fold reduction in the pore pressure [31].
Experimental investigations have demonstrated the effectiveness of fibres as a means of reducing
the water vapour pressure in concrete. PP fibres at an amount of about 0.2% of the volume makes
it possible to effectively reduce the internal pressures. The addition of 1.75 kg/m3 results in a 70%
reduction in the value of the maximum pressure, which was observed when comparing it with a
non-fibrous reference concrete. The incorporation of the fibres significantly reduces the values of the
maximum pore pressures and decreases the temperature of the pressure peaks (180–200 ◦ C). It has also
been noted that the effect of fibres does not increase significantly when the fibre content is higher than
1.75 kg/m3 .
There is an alternative explanation for the observed positive effect of PP fibres. Their introduction
to concrete decreases the strength after melting by introducing a supplementary porosity that affects the
strength. The reduced tensile strength may allow the formation of thermal cracks in the cross-section
by unloading the thermal compression stresses necessary for spalling on the surface layer. This theory,
highlighted in Reference [6], explains why fibres reduce the spalling of slabs and unloaded specimens
instead of loaded columns, which are made of the same material.
The permeability of concretes with fibres increases rapidly with temperature. In Reference [11],
different lengths and quantities of fibres were used in order to choose the best PP fibre option in
terms of size and quantity, which provides an essential permeability increase. In this study, PP fibres
with a specific gravity of 0.91 g/cm3 , three different amount of PP fibres (0, 0.9 and 1.8 kg/m3 ) and
three lengths (6, 12 and 19 mm) were used. The five series of concrete were tested, designated as
Materials 2019, 12, 3869 9 of 20
“HPC_fibredosage_fibre length.” The increase of the heated concrete gas permeability was measured
using the Cembureau method, described in detail in Reference [17]. The study has shown the effect of
the length and content of PP fibres on the relative permeability of heated HPCs.
The addition of polypropylene fibres in amounts of 0.9 and 1.8 kg/m3 does not strongly affect the
initial properties of concrete (Figure 8). The sharp permeability increase was observed between two
measuring points at 140 ◦ C and 160 ◦ C. A significant increase in residual permeability takes place after
heating to 160 ◦ C, which is a temperature of 3 degrees lower than the melting temperature of the fibres
used. The sharp permeability increase after exposure to 160 ◦ C may be caused by partial absorption of
fibres by cement matrix and due to debonding of fibres from the cement matrix. The length of PP fibres
hasa significant impact on permeability development, as shown in Figure 9. Short fibres (6 mm) seem
to be inefficient in creating an interconnected and permeable pore network. Although the total dosage
of PP fibres was equal, the short fibres of 6 mm provide four-fold lower permeability than fibres of
12 mm in
Materials length
2019, andPEER
12, x FOR eight-fold
REVIEW lower than 19-mm long fibres. 9 of 21
Figure 8.
Figure 8. Relative permeability change of heated HPC and the effect
effect of
of PP fibre amounts
PP fibre amounts of
of 0,
0, 0.9
0.9 and
and
1.8 kg/m
1.8 kg/m 3
3 [11].
Figure 9.
Figure 9. Effect
Effect of
of fibre
fibre length
length onon the
the HPC
HPC relative
relative permeability
permeability of
of heated
heated concrete
concrete with
with 1.8
1.8 kg/m
kg/m33 of
of
PP fibres
PP fibres with
with lengths
lengths ofof 6,
6, 12
12 and
and 19
19 mm
mm [11].
[11].
A practical method
A practical methodfor foradditional
additionalopenopenporosity
porosity evaluation
evaluation in in heated
heated concretes
concretes with with PP is
PP fibres
fibres
the is thewater
surface surface water absorption
absorption test. discs
test. Concrete Concrete discs are
are heated heated totemperature
to different different temperature
levels, which is
levels,towhich
close is closetemperature
the melting to the melting (163 ◦ C in this research,
temperature (163 °C inor
this
160,research,
180, and 160,◦ C).
or200 180,The
andtest
200consisted
°C).
The
of test consisted
measuring of measuring
the mass thespecimen,
gain of the mass gain which
of the specimen,
is in contactwhich
withiswater.
in contact withthe
During water.
test, the
During
water the remained
level test, the water level
constant, remained
and constant,
an increase in the and an mass
sample increasewasinrecorded.
the sample Themass was
measurement
recorded.
set-up The measurement
is shown in Figure 10. set-up is shown
In Figure 11, the in Figure
results 10. investigations
of the In Figure 11, on thetheresults of the
influence of the
investigations on the influence of the PP fibre length on the transport properties of HPC were
presented (designated as “HPC100_fibre dosage_fibre length”). The water surface absorption
was tested, and it showed the increase of the open porosity volume and capillary porosity of the
cement matrix. Consecutive measurements of the water absorption of heated concretes over time
have enabled us to show that the porosity network was more developed in concretes in which
Materials 2019, 12, 3869 10 of 20
PP fibre length on the transport properties of HPC were presented (designated as “HPC100_fibre
dosage_fibre length”). The water surface absorption was tested, and it showed the increase of the open
porosity volume and capillary porosity of the cement matrix. Consecutive measurements of the water
absorption of heated concretes over time have enabled us to show that the porosity network was more
developed
Materials in
Materials2019,
2019, concretes
12,
12,xxFOR inREVIEW
FORPEER
PEER which the longer fibres, favouring percolation, were used (Figure 11).
REVIEW 10
10ofof21
21
Figure
Figure10.
10.The
Thewater
The waterabsorption
water absorptionmeasurement
absorption measurementset-up.
measurement set-up.
set-up.
Figure
Figure11.11.The
TheHPC
The HPCwater
HPC waterabsorption
water absorptionfor
absorption forconcretes
for concretesheated
heatedtoto140,
140,160,
160,180
180and 200◦°C,
and200 C, with
°C, with1.8 kg/m3 3
1.8kg/m
ofofPP
PPfibres.
fibres.Fibre
Fibrelengths:
Fibre lengths:6,
lengths: 6,12
6, 12and
12 and19
and 19mm
19 mm[32].
mm [32].
[32].
3.3.
3.3. The
3.3.The Influence
TheInfluence ofofPP
Influenceof PP Fibres
PPFibres on
Fibreson the
onthe Mechanical
theMechanical Performance
MechanicalPerformance ofofHeated
Performanceof Heated Concrete
HeatedConcrete
Concrete
According
According
Accordingto to some
tosome sources,
somesources,
sources,PPPP fibres
PPfibres atatamounts
fibresat amounts
amountsof of 0.1%
of0.1%
0.1%toto 0.2%
to0.2%
0.2%leadlead
leadtoto
toaaadecrease
decrease
decreasein in the
inthe initial
theinitial
initial
strength
strength of concrete
strengthofofconcrete
concretebyby 10%
by10% to 15%.
10%toto15%.
15%.The The strength
Thestrength decrease
strengthdecrease is
decreaseisisdue due to
duetotothe the
thelower lower stiffness
lowerstiffness of
stiffnessofofthe the
thefibres fibres
fibresinin
in
thethe cementitious
thecementitious
cementitiousmatrix.matrix. Moreover,
matrix.Moreover,
Moreover,the the material
thematerial compaction
materialcompaction
compactionofoftheof the higher
thehigher fibres’
higherfibres’ content
fibres’content is
contentisismore more
more
complicated.
complicated. Therefore,
complicated.Therefore,
Therefore,thethe concrete
theconcrete porosity
concreteporosity
porositymay may
maybe be the
bethe cause
thecause
causeofof the
ofthe fibrous
thefibrous concrete
concretemechanical
fibrousconcrete mechanical
mechanical
performance
performance decrease.
performancedecrease. The
decrease.The influence
Theinfluence
influenceof of the
ofthe heating
theheating temperature
heatingtemperature
temperatureon on
onthethe changes
thechanges
changesin in the
inthe compressive
thecompressive
compressive
strength
strength of the
strengthofofthe reactive
thereactive powder
reactivepowder concretes,
powderconcretes, with
concretes,with and
withandwithout
andwithout PP
withoutPPfibres, was
PPfibres, studied
fibres,was wasstudied[28], and
studied[28], theand
[28], results
and the
the
of this investigation are presented in Figure 12. The RPC concretes with 1 and 2 kg/m 3 of short PP
results
resultsofofthis
thisinvestigation
investigationare arepresented
presentedininFigure
Figure12.12.The
TheRPC
RPCconcretes
concreteswith with11and and22kg/m
kg/m ofofshort
3 3 short
fibres
PP are are
PPfibres
fibres presented, andand
arepresented,
presented, their
and initial
their
their difference
initial
initial differencein strength
difference waswas
ininstrength
strength 15%
was15% (30 MPa).
15% (30
(30MPa).
MPa).
the cementitious matrix. Moreover, the material compaction of the higher fibres’ content is more
complicated. Therefore, the concrete porosity may be the cause of the fibrous concrete mechanical
performance decrease. The influence of the heating temperature on the changes in the compressive
strength of the reactive powder concretes, with and without PP fibres, was studied [28], and the
results of this investigation are presented in Figure 12. The RPC concretes with 1 and 2 kg/m3 of short
Materials 2019, 12, 3869 11 of 20
PP fibres are presented, and their initial difference in strength was 15% (30 MPa).
Figure 12.Theinfluence of heating temperature on changes in the compressive strength of RPC and
RPCs with PP fibres [28].
In Figure 13, the relative values of the compressive strength evolution for tested heated HPC,
with 0, 0.9 and 1.75 kg/m3 of fibres, are presented [33]. The applied heating rate was 1°C/min, and the
tests were performed 2 h after a plateau had been reached. The relative values of the compressive
strength evolution (relative values refer to the compressive strength of the non-heated material) for
HPC with PP fibres were somewhat identical to those of ordinary concrete (fc = 39.5 MPa), which
confirms
Figurethat the PP fibresofcontribute
12. Theinfluence to the permeability
heating temperature on changes increase.
in the compressive strength of RPC and
The with
RPCs resultsPP for the[28].
fibres compressive strength, tested at 120 °C, presents decreases of about 20%. This
was explained by the presence of a hydral gradient in the middle of the specimen where the core of
the In
specimen
Figure 13, is the
saturated
relativewith
valueswater,
of theand the surface
compressive has dried
strength out. The
evolution for observed decrease
tested heated HPC, in the
with
0,mechanical kg/m3 of fibres,
0.9 and 1.75performance in the
arewater clog zone
presented [33]. is
Theconsidered to contribute
applied heating ◦ C/min,
to 1the
rate was spalling
andbehaviour.
the tests
When
were the melting
performed temperature
2 h after a plateauof hadfibres
beenisreached.
reached,The water evaporation
relative values of is
thepossible. Therefore,
compressive strengthfor
fibrous HPC
evolution andvalues
(relative ordinary
referconcrete, a partial strength
to the compressive strength restoration was observed
of the non-heated material)atfor250
HPC °C.with
The
PPconcrete performance
fibres were somewhat reaches 35%toofthose
identical the initial compressive
of ordinary concrete (fc = 39.5
strength, when thewhich
MPa), temperature
confirmsreaches
that
600PP
the °Cfibres
for allcontribute
tested materials.
to the permeability increase.
120
100
80
fc/fc20°C (%)
60
40
HPC PP fibres 0.9 kg/m3
HPC PP fibres 1.75 kg/m3
20
HPC
NSC
0
0 200 400 600
Temperature (°C)
Figure
Figure13. Relative values
13.Relative valuesofofcompressive
compressivestrength
strengthfor
fornormal-strength
normal-strengthconcrete
concreteand
andHPC,
HPC,with
with0,0,0.9,
0.9,
and 1.75 kg/m 3 3of fibres [33].
and 1.75 kg/m of fibres [33].
The results
4. Testing for the
Methods forcompressive
the Evaluationstrength,
of thetested at 120 ◦ C,of
Effectiveness presents
PP Fibresdecreases of about
in Spalling 20%. This
Mitigation
was explained by the presence of a hydral gradient in the middle of the specimen where the core of
The assessment
the specimen is saturatedof concrete behaviour
with water, and the at surface
a high temperature
has dried out.is carried out using
The observed three types
decrease in theof
testing methods
mechanical at different
performance in thelevels
water of analysis:
clog zone is small-scale,
considered to medium-scale, andspalling
contribute to the full-scale analysis.
behaviour.
Small-scale tests serve to investigate the material behaviour of concrete
When the melting temperature of fibres is reached, water evaporation is possible. Therefore, exposed to an elevated
for
temperature, i.e., the thermal strains, gas pore pressure, temperature gradient,
fibrous HPC and ordinary concrete, a partial strength restoration was observed at 250 C. The concrete ◦ or mechanical
properties. Full-scale
performance reaches 35% fireoftests are carried
the initial out on strength,
compressive full-sizedwhen
concrete elements in reaches
the temperature a real situation,
600 ◦ C forin
which the boundary
all tested materials. conditions, external load, and conditioning correspond to the design. A full-
scale test, as the most representative test, shows the qualitative and quantitative behaviour of
structural concrete elements subjected to fire. The medium-scale test is carried out on concrete slabs
exposed to fire, with a surface area of ca. 1m2. The concrete specimens in medium-scale tests may be
reinforced and loaded or restrained concrete, depending on the laboratory procedure. Medium-scale
tests are performed as a screening test to verify the behaviour of specific concrete mixtures in fire
conditions and select the one that does not tend to spall. To investigate the influence of polypropylene
fibres on concrete spalling behaviour in fire, the different testing methods were developed, and the
most interesting method was described. To date, no standardised document concerning the testing
Materials 2019, 12, 3869 12 of 20
4. Testing Methods for the Evaluation of the Effectiveness of PP Fibres in Spalling Mitigation
The assessment of concrete behaviour at a high temperature is carried out using three types
of testing methods at different levels of analysis: small-scale, medium-scale, and full-scale analysis.
Small-scale tests serve to investigate the material behaviour of concrete exposed to an elevated
temperature, i.e., the thermal strains, gas pore pressure, temperature gradient, or mechanical properties.
Full-scale fire tests are carried out on full-sized concrete elements in a real situation, in which the
boundary conditions, external load, and conditioning correspond to the design. A full-scale test, as
the most representative test, shows the qualitative and quantitative behaviour of structural concrete
elements subjected to fire. The medium-scale test is carried out on concrete slabs exposed to fire, with
a surface area of ca. 1m2 . The concrete specimens in medium-scale tests may be reinforced and loaded
or restrained concrete, depending on the laboratory procedure. Medium-scale tests are performed as a
screening test to verify the behaviour of specific concrete mixtures in fire conditions and select the one
that does not tend to spall. To investigate the influence of polypropylene fibres on concrete spalling
behaviour in fire, the different testing methods were developed, and the most interesting method was
described. To date, no standardised document concerning the testing of the spalling behaviour of
concrete has been proposed to unify the material testing methods in order to allow for the screening of
concrete mixes.
The assessment of the susceptibility of concrete to spalling can also be conducted using a simple
The assessment of the susceptibility of concrete to spalling can also be conducted using a simple
small-scale test that relies on preparing laboratory samples (cylinders, prisms, or cubes), with and
small-scale test that relies on preparing laboratory samples (cylinders, prisms, or cubes), with and
without PP fibres, and exposes them to an elevated temperature in a laboratory electric furnace. Such
without PP fibres, and exposes them to an elevated temperature in a laboratory electric furnace. Such
a test is commonly used to screen the concrete mix in order to check only if it spalls or if it is not
a test is commonly used to screen the concrete mix in order to check only if it spalls or if it is not
under an elevated temperature. In those tests, samples are prepared with different types of fibres and
under an elevated temperature. In those tests, samples are prepared with different types of fibres and
dosages to optimise the minimum effective amount that reduces spalling and, at the same time, does
dosages to optimise the minimum effective amount that reduces spalling and, at the same time, does
not significantly affect fresh concrete properties and enable appropriate mixing and homogenization.
not significantly affect fresh concrete properties and enable appropriate mixing and homogenization.
A small, simple test using cylinders was carried out by Han [9]. Normal-strength concrete (fc = 40MPa)
A small, simple test using cylinders was carried out by Han [9]. Normal-strength concrete (fc = 40MPa)
with three amounts of polypropylene (PP) fibres—0.0, 0.45 and 0.90 kg/m33—was investigated for two
with three amounts of polypropylene (PP) fibres—0.0, 0.45 and 0.90 kg/m —was investigated for two
types of concrete of different water to binder (w/b) ratios: 30% and 40%. The specimens were heated in
types of concrete of different water to binder (w/b) ratios: 30% and 40%. The specimens were heated
a laboratory furnace with the heating curve KS F 2257 [9]. The results show that the addition of the
in a laboratory furnace with the heating curve KS F 2257 [9]. The results show that the addition of the
smallest amount of PP fibres (0.45 kg/m3 , 0.05% in volume) significantly reduces the spalling behaviour
smallest amount of PP fibres (0.45 kg/m3, 0.05% in volume) significantly reduces the spalling
in tested concrete, as shown in Figure 14.
behaviour in tested concrete, as shown in Figure 14.
Figure 14.
Figure 14. Results
Results of
of simple
simple small-scale
small-scale tests
tests on cylinders [9].
on cylinders [9].
Similarly, small samples were also tested by by Li

Li [34]
[34] using
using Ultra-High-Performance
Ultra-High-Performance Concretes
(UHPCs, fc = 125MPa), with the addition of PP fibres at amounts of 0.0, of
(UHPCs, fc = 125MPa), with the addition of PP fibres at amounts 2.00.0,
and2.0 3 . This
and 4.0
4.0 kg/m kg/m 3. This
research
research
aimed to aimed to evaluate
evaluate the aggregate
the aggregate size. Asimple
size. A small, small,test
simple test on cylinders
on cylinders exposedexposed to the ISO-
to the ISO-834 [35]
834 [35] heating curve in a laboratory furnace showed that, in the case of UHPCs, the spalling
elimination was possible only for 4 kg/m3 of PP fibres for each tested composition, as shown in Figure
15.
These tests are not considered to be representative of a fire situation involving construction
projects when the material is under load. In order to develop a test that is more representative of a
fire situation and to investigate the behaviour of the material from loaded concrete samples, heating
Materials 2019, 12, 3869 13 of 20
heating curve in a laboratory furnace showed that, in the case of UHPCs, the spalling elimination was
possible only for 4 kg/m3 of PP fibres for each tested composition, as shown in Figure 15.
Figure 15. The

Figure 15. The results
resultsof
ofaasmall,
small,simple
simpletest
test
in in a laboratory
a laboratory furnace
furnace on concretes
on concretes withwith different
different PP
PP fibre
fibre dosages
dosages [34]. [34].
In several
These teststests, an acoustic
are not considered emission
to be (AE) setup is additionally
representative assembled
of a fire situation near the
involving specimen.
construction
Fire testswhen
projects with the
an AE setup is
material provide
under essential information
load. In order on micro-cracking
to develop a test that is more development that may
representative of a
differ for fibrous and non-fibrous concretes. Since micro-crack development
fire situation and to investigate the behaviour of the material from loaded concrete samples, heating is considered to enhance
the moisture
the setups forrelease and under
specimens lower the pore pressure,
compressive loads information
are used. Such on aittest
is essential for tests
can be helpful foron the fire
assessing
spalling
the changesof concrete. Huismann
in the mechanical [36] tested
properties cylinderscaused
of concrete made by of HSC (fc = 98temperature,
an elevated MPa) under as anwell
elevated
as to
temperature and uniaxial compressive load
study the PP fibre dosage effect on the change of properties.action, with 2 kg/m 3 and without the addition of PP
fibres.InCylindrical
several tests,specimens
an acoustic of emission
Ø100 mm(AE) andsetup
H300ismm were heated
additionally at 1°C/min
assembled near and loaded with
the specimen. a
Fire
constant ratio. The setup enabled the measurement of the temperature
tests with an AE setup provide essential information on micro-cracking development that may differ changes by thermocouples,
axial deformations
for fibrous by an extensometer,
and non-fibrous concretes. Sinceandmicro-crack
internal damages by AE records,
development is consideredas shown in Figure
to enhance the
16.Such
moisturesmall
releasetests
and led
lower to the
differences in theinformation
pore pressure, development on itofismicro-cracks
essential for testsin the onspecimen to be
the fire spalling
investigated while heating
of concrete. Huismann and the
[36] tested changes
cylinders made of HSC (fc =properties
in mechanical 98 MPa) under of theantested
elevated concrete to be
temperature
shown. Huismann
and uniaxial noticedload
compressive thataction,
the addition
with 2of PP fibres
kg/m 3 resultedthe
and without in more
additionAE events between
of PP fibres. 200 and
Cylindrical
250 °C andoflinked
specimens Ø100 mm it withand intensified
H300 mm were micro-cracking
heated at 1◦and C/min moisture
and loaded release
without of the specimen.
a constant ratio. The
Measurements
setup enabled the of the transient strain
measurement of theenabled the micro-cracking
temperature development axial
changes by thermocouples, to be deformations
confirmed since by
the recorded strains
an extensometer, andwereinternal found to be higher
damages for fibrous
by AE records, concrete
as shown than for
in Figure 16.non-fibrous
Such smallconcrete.
tests led
Huismann
to differences alsoinnoticed that, in HSC
the development with PP fibres
of micro-cracks in heated to 600 to
the specimen °C,bethe cracks werewhile
investigated thinner and
heating
more numerous than the HSC without PP fibres. Micro-cracks in fibrous
and the changes in mechanical properties of the tested concrete to be shown. Huismann noticed concretes are considered to
develop as a result
that the addition ofof
PPincreasing the volume
fibres resulted in more of AE
melted
eventsPP between
fibres and 200theand
expansion ◦
250 C and of the fibreitbed.
linked with
intensified micro-cracking and moisture release out of the specimen. Measurements of the transient
strain enabled the micro-cracking development to be confirmed since the recorded strains were found
to be higher for fibrous concrete than for non-fibrous concrete. Huismann also noticed that, in HSC
with PP fibres heated to 600 ◦ C, the cracks were thinner and more numerous than the HSC without PP
fibres. Micro-cracks in fibrous concretes are considered to develop as a result of increasing the volume
of melted PP fibres and the expansion of the fibre bed.
Figure 16. Example of a small sample test under a compressive load [36].
250 °C and linked it with intensified micro-cracking and moisture release out of the specimen.
Measurements of the transient strain enabled the micro-cracking development to be confirmed since
the recorded strains were found to be higher for fibrous concrete than for non-fibrous concrete.
Huismann also noticed that, in HSC with PP fibres heated to 600 °C, the cracks were thinner and
more numerous
Materials than the HSC without PP fibres. Micro-cracks in fibrous concretes are considered
2019, 12, 3869 14 of to
20
develop as a result of increasing the volume of melted PP fibres and the expansion of the fibre bed.
Figure 16.
Figure 16. Example
Example of
of aa small
small sample test under
sample test under aa compressive
compressive load
load [36].
[36].
A simple weight loss technique was used to establish spalling events and sample mass loss during
heating, as shown in the authors’ research [28]. The specimens were heated in an electric furnace.
During the test, different heating ratios were applied to study the influence of this parameter on
concrete spalling behaviour at a high temperature. The concrete samples were hung above the electric
furnace, and the mass of the sample was continuously recorded during heating. The temperature was
measured using a type K thermocouple placed in proximity to the sample’s surface. The measurement
set-up is shown in Figure 17. From the experiment, the mass loss curve is traced, and the signal
disruption corresponding to the sample explosion is recorded. Moreover, from the mass loss curve, the
permeability of the material can be evaluated by analysing the kinetics of mass loss. For the concrete
with PP fibres, when the polypropylene melts, the mass loss is more pronounced, and the mass loss
rate is higher.
Figure 17.
Figure 17. The
The mass
mass loss
loss set-up
set-up for spalling behaviour
for spalling behaviour observation.
observation.
Figure18,18,
In Figure thethe results
results for ultra-high-performance
for ultra-high-performance RPC reactive
RPC reactive powder were
powder concrete concrete were
presented.
presented.
During slowDuring
heating,slow heating,
i.e., with i.e., with
a heating rate ofa 0.5 ◦ C/min,
heating rate of of
none 0.5the
°C/min,
four RPCnone of the four
specimens RPC
exploded.
specimens
When exploded.
the heating rate When the heating
was increased ◦
to 1.0rate was in
C/min increased to 1.0 °C/min
all four samples, in all
spalling four samples,
occurred spalling
at temperatures
occurred
from 241 at 261 ◦ C. As expected,
to temperatures from 241a to 261 °C.
further As expected,
increase a furtherrate
of the heating (2 ◦ C/min)
increase of thealso
heating rate in
resulted (2
°C/min)
RPC also resulted
showing spallinginbehaviour
RPC showing(T = 253–262 ◦
spalling behaviour
C). As shown(T = in
253–262
Figure°C). Asimpulsive
18, an shown in Figure 18,
explosion
an impulsive explosion occurs when the temperature near the surface of the specimen reaches 250
°C, practically without a mass change.
0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500
0 0 0
T[oC] T[oC] T[oC]
Figure 17. The mass loss set-up for spalling behaviour observation.
In Figure 18, the results for ultra-high-performance RPC reactive powder concrete were
presented. During slow heating, i.e., with a heating rate of 0.5 °C/min, none of the four RPC
Materials 2019, 12, 3869
specimens 15 of 20
exploded. When the heating rate was increased to 1.0 °C/min in all four samples, spalling
occurred at temperatures from 241 to 261 °C. As expected, a further increase of the heating rate (2
°C/min) also resulted in RPC showing spalling behaviour (T = 253–262 °C). As shown in Figure 18,
occursanwhen the temperature nearwhen
the surface of the specimen reachesof250 ◦ C, practically without a
impulsive explosion occurs the temperature near the surface the specimen reaches 250
mass change.
°C, practically without a mass change.
0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500
0 0 0
T[oC] T[oC] T[oC]
-2 -2 -2
weight loss [%]
-4 -4 -4
-6 -6 -6
-8 -8 -8
RPC_0.5 degC/min RPC_1 degC/min RPC_2 degC/min

-10 -10 -10
18. The
FigureFigure 18.mass loss results
The mass for the
loss results for RPC concrete
the RPC were
concrete heated
were with
heated three
with threeheating
heatingrates:
rates: 0.5,
0.5, 1.0,
1.0, and
2.0 ◦ C/min.
and 2.0 °C/min.
This experiment
This experiment has has clearly
clearly shown
shown that,for
that, forRPC
RPC concretes,
concretes, high-density
high-density cementitious
cementitious materials
materials
with a compressive strength reached 250 MPa. In addition, the risk of spalling in fire
with a compressive strength reached 250 MPa. In addition, the risk of spalling in fire is high and is high and PP PP
fibres may prevent spalling. For RPC with 1 kg/m33 of fibre, spalling risk occurs when the heating rate
fibres may prevent spalling. For RPC with 1 kg/m of fibre, spalling risk occurs when the heating rate
is 4 °C/min, and 2 kg/m3 of fibres prevent spalling, even when 8 °C/min is applied (Table 1).
is 4 ◦ C/min, and 2 kg/m3 of fibres prevent spalling, even when 8 ◦ C/min is applied (Table 1).
Table 1. Spalling test results for plain RPC and RPC modified with PP fibres.
Table 1. Spalling test results for plain RPC and RPC modified with PP fibres.
Heating Rate
RPC 0.5 °C/min 1.0 °C/min Heating 2.0 °C/min
Rate 4.0 °C/min 8.0 °C/min
◦ Spalling: Yes/No,
◦ [Tested Spec./Spec. That
◦ Spalled], Spalling◦Temperature Tspall ◦
RPC 0.5 C/min 1.0 C/min 2.0 C/min 4.0 C/min 8.0 C/min
yes [4/4] yes [4/4]
RPC no [10/0] Yes/No, [Tested Spec./Spec. That Spalled], Spalling Temperature
Spalling: - T -
(241, 261, 254, 241°C) (253, 262, 253, 261°C) spall
yes [4/4] yes [4/4] yes [2/1] yes [2/2]

RPC
RPCPP 1 kg/m
3 no [10/0]
no [10/0] no [2/0] no [2/0] - -
(241, 261, 254, 241 ◦ C) (253, 262, 253, 261 ◦ C) (331°C) (354, 348°C)
RPC PP 2 kg/m3 no [10/0] no [2/0] no [2/0] yesno [2/0]
[2/1] no
yes[2/0]
[2/2]
RPC PP 1 kg/m3 no [10/0] no [2/0] no [2/0]
(331 ◦ C) (354, 348 ◦ C)
RPC PPThe
2 kg/m3 no [10/0] no [2/0] no [2/0] no [2/0] no [2/0]
small material test does not provide information on element behaviour in fire. For cubes,
cylinders, and prisms, heating is applied on all sides. In a real fire situation, heating on one side
The small material test does not provide information on element behaviour in fire. For cubes,
cylinders, and prisms, heating is applied on all sides. In a real fire situation, heating on one side usually
occurs. In order to more adequately reflect the stress state and boundary condition in a full-scale
concrete element, medium-scale tests are commonly used on loaded or restrained concrete elements
at a reduced scale. Cost-effectiveness is an essential aspect of screening concrete mixes for spalling
propensity. Research laboratories commonly use the testing setup that employs gas burners to expose
concrete to fire.
A useful example of a medium-scale test was proposed in Reference [37]. In this method, a circular
concrete element is moulded in a steel ring, with an external diameter of 300 mm. The testing setup is
developed in order to monitor the internal temperature, the gas pore pressure, and restrained stress of
the concrete specimen. The element is placed on top of the gas furnace and heated from the bottom in a
RABT 30 (RABT—Richtlinienfür die Ausstattung und den Betrieb von Straßentunneln) fire scenario, as
shown in Figure 19. During the test, the steel ring confines the concrete that expands under an elevated
temperature. While the concrete expands due to the elevated temperature, the thermal strains can be
measured by recording the strains of the external steel rings. The thermal stresses of the concrete were
recalculated from the measured strains of the steel ring at a particular temperature.
stress of the concrete specimen. The element is placed on top of the gas furnace and heated from the
bottom in a RABT 30 (RABT—Richtlinienfür die Ausstattung und den Betrieb von Straßentunneln)
fire scenario, as shown in Figure 19. During the test, the steel ring confines the concrete that expands
under an elevated temperature. While the concrete expands due to the elevated temperature, the
thermal strains can be measured by recording the strains of the external steel rings. The thermal
Materialsstresses
2019, 12,of3869
the concrete were recalculated from the measured strains of the steel ring at a particular 16 of 20
temperature.
Figure 19. The scheme of the testing setup of the ring test for the fire spalling of concrete [37].
In 2018, the research team of Akasaka [38] presented comparative test results on the behaviour
of HSC, with and without PP fibres, at 0.1% in terms of the volume. The specimens were heated in a
RABT 30 fire scenario from the bottom. The results showed that the addition of 0.1% of PP fibres did
Figure 19. The scheme of the testing setup of the ring test for the fire spalling of concrete [37].
Figure 19.
not completely The scheme
prevent spallingof the
buttesting
reducedsetupitsofextent
the ring test the
from for the fire spalling
maximum depthof concrete [37].
in non-fibrous HSC
of 35 mm to In 52018,
mmthe in research
the caseteam of HSC with PP,
of Akasaka [38]as shown comparative
presented in Figure 20. test results on the behaviour
Inof2018,
HSC,
the research team
with and without
of Akasaka
PP fibres,in
at the
[38]
0.1%thermal
presented
in terms of
comparative
the volume.
test results
The specimens
on the behaviour
were heated
of
While considering the changes stress of concrete, it was found that in
thea stresses
HSC, with and without PP fibres, at 0.1% in terms of the volume. The specimens
RABT 30 fire scenario from the bottom. The results showed that the addition of 0.1% of PP fibres did were heated in a
induced by HSC with PP are higher than those induced by non-fibrous HSC. This is due to the
RABT not
30 fire scenario
completely fromspalling
prevent the bottom. The results
but reduced its extentshowed
from thethat the addition
maximum depth in of 0.1% of PP
non-fibrous HSC fibres did
expansion of the melting
of 35 mm to
PP fibres.HSC
5 mm inspalling
the case of
Onwith
the PP,
other hand, in the expanding and melting PP fibres provide
not completely prevent but reduced itsasextent
shown from Figure
the20.
maximum depth in non-fibrous HSC
an essentialWhile
network of micro-cracks
considering the of
changes that facilitates
in the thermal the water
stress of transport
it wasthrough the
thetested specimen,
of 35 mm to 5 mm in the case HSC with PP, as shown in concrete,
Figure 20. found that stresses
which induced
contributes by HSCto the
withreduction
PP are higherof the pore
than pressure.
those induced by non-fibrous HSC. This is due to the
expansion of the melting PP fibres. On the other hand, the expanding and melting PP fibres provide
an essential network of micro-cracks that facilitates the water transport through the tested specimen,
which contributes to the reduction of the pore pressure.
Figure 20. Results of the spalling tests for (a) plain HSC, and (b) HSC with PP fibres [38]. The advantages
of the ring tests consist in easily casting and reproducing compressive loading in a similar way to
the existing concrete structures, i.e., lateral compression. Such a test can be performed to screen the
concrete mixture composition, before designing the concrete structures.
While considering the changes in the thermal stress of concrete, it was found that the stresses
induced by HSC with PP are higher than those induced by non-fibrous HSC. This is due to the
expansion of the melting PP fibres. On the other hand, the expanding and melting PP fibres provide
an essential network of micro-cracks that facilitates the water transport through the tested specimen,
which contributes to the reduction of the pore pressure.
The confinement to thermal strains in concrete can also be introduced by the presence of the
unheated part of the concrete slab, which does not expand while heating. Unlike the steel ring, the
unheated concrete part counteracts the expansion of the volume of the tested element that is exposed
to inert fire. The authors have carried out the experimental tests on three different sizes of HSC slabs
(0.6 × 0.6 × 0.15 m3 , 0.8 × 0.8 × 0.15 m3 , and 1.0 × 1.0 × 0.15 m3 ), which were exposed to fire. In each
case, the area exposed to fire was equal to 0.6 m × 0.6 m. For each tested case, the area exposed to
fire was surrounded by an unexposed belt of different widths (0 cm—no cold rim, 10 cm and 20 cm),
which constituted a cold rim, as shown in Figure 21. The slab-like element is simply supported at
four edges and heated from the bottom in an ISO-834 fire scenario. The results of this test show that
concrete is more prone to spalling when a restraint in the form of a cold rim is present. This experiment
unheated concrete part counteracts the expansion of the volume of the tested element that is exposed
to inert fire. The authors have carried out the experimental tests on three different sizes of HSC slabs
(0.6 × 0.6 × 0.15 m3, 0.8 × 0.8 × 0.15 m3, and 1.0 × 1.0 × 0.15 m3), which were exposed to fire. In each
case, the area exposed to fire was equal to 0.6 m × 0.6 m. For each tested case, the area exposed to fire
was surrounded by an unexposed belt of different widths (0 cm—no cold rim, 10 cm and 20 cm),
Materials
which2019, 12, 3869 a cold rim, as shown in Figure 21. The slab-like element is simply supported at 17 of 20
constituted
four edges and heated from the bottom in an ISO-834 fire scenario. The results of this test show that
concrete is more prone to spalling when a restraint in the form of a cold rim is present. This
indicated that, when intermediate scale tests are performed on concrete used for slab-like elements, the
experiment indicated that, when intermediate scale tests are performed on concrete used for slab-like
test elements,
should bethe conducted
test shouldon
beslabs with aoncold
conducted slabsrim.
with a cold rim.
Figure 21. Fire spalling test on HSC concrete slabs with different cold rim widths.
Figure 21. Fire spalling test on HSC concrete slabs with different cold rim widths.
Another type of test that is used to reproduce concrete element behaviour in fire conditions using
Another type of test that is used to reproduce concrete element behaviour in fire conditions using
smaller-scale test procedures consists of testing mechanically-loaded slab-shaped specimens. The
smaller-scale test procedures consists of testing mechanically-loaded slab-shaped specimens. The
external load is introduced in the in-plane lateral direction on a medium-sized slab. The external load
external load
reflects the isstress
introduced
state thatinone the can
in-plane
find inlateral direction(uniaxial
real structures on a medium-sized slab. The By
or biaxial compression). external
loadchoosing
reflects the
the load
stress state
level, that
both theone can find
restraint andinadditional
real structures
external(uniaxial
load mayor bebiaxial compression).
introduced into the By
choosing
testingthe load level,
element. both the
Commonly, restraint andsetup
the experimental additional
consistsexternal
of a gas load may
furnace, be the
with introduced
opening oninto the
testing element.
the top, on whichCommonly, the experimental
the slab specimen setup
is placed. Most consists
often, of aslab-like
squared gas furnace, with
elements arethe opening
used for thison the
top,kind of test.the
on which An slab
external load is applied
specimen is placed.to the slaboften,
Most using squared
hydraulicslab-like
jacks, which are located
elements on the
are used for this
kindsides of the
of test. An slab, as shown
external loadin Figure 22. to the slab using hydraulic jacks, which are located on the
is applied
The slab,
sides of the effectiveness
as shown ofinthe PP fibres
Figure 22. in fire spalling mitigation of concrete elements may be
Materials 2019, 12,
evaluated x FOR PEER
because REVIEWbiaxial load prevents cracking. Thus, the load controls crack opening18 of 21
the external
and moisture release.
Figure 22.
Figure 22. Experimental
Experimental setup
setup for
for testing
testing the fire spalling
the fire spalling of
of concrete
concrete under
under an
an external
external load
load action.
action.
Left: uniaxial and biaxial loading frame at Cracow University of Technology. Right: biaxial
Left: uniaxial and biaxial loading frame at Cracow University of Technology. Right: biaxial loadingloading
frame at
frame at Politecnico
Politecnico di
di Milano
Milano [21].
[21].
The effectiveness
The test, considering
of thethe effectiveness
PP fibres of PP fibres
in fire spalling using
mitigation of loaded
concreteslabs, was may
elements presented by Lo
be evaluated
Monte etthe
because al. external
in 2015 [21]. Theload
biaxial HSCprevents
(fc ± 60MPa) slabs, Thus,
cracking. whichthe
were 0.8controls
load × 0.8 × 0.1 m3 opening
crack in size, were
and
subjected to constant
moisture release. in-plane lateral compression of 10 MPa during the test. In one tested
composition, the addition of 2 kg/m3 of PP fibres was used. Slabs were exposed to an ISO-834 fire
scenario from the bottom side in an area of 0.6 × 0.6 m2. The test results are very similar to those
obtained with the above-mentioned ring test. PP fibres do not prevent spalling but strongly reduce
its occurrence and help protect the reinforcing steel. Both tests confirm that PP fibres can reduce
spalling only to some extent and tests on spalling behaviour and the effective dosage of PP fibres,
Materials 2019, 12, 3869 18 of 20
The test, considering the effectiveness of PP fibres using loaded slabs, was presented by Lo Monte
et al. in 2015 [21]. The HSC (fc ± 60MPa) slabs, which were 0.8 × 0.8 × 0.1 m3 in size, were subjected
to constant in-plane lateral compression of 10 MPa during the test. In one tested composition, the
addition of 2 kg/m3 of PP fibres was used. Slabs were exposed to an ISO-834 fire scenario from the
bottom side in an area of 0.6 × 0.6 m2 . The test results are very similar to those obtained with the
above-mentioned ring test. PP fibres do not prevent spalling but strongly reduce its occurrence and
help protect the reinforcing steel. Both tests confirm that PP fibres can reduce spalling only to some
extent and tests on spalling behaviour and the effective dosage of PP fibres, before designing concrete
structures, are strongly recommended.
5. Conclusions
This paper presents an overview of current knowledge about the concrete spalling in fire and
the role of polypropylene fibres in its effective mitigation. The main mechanisms leading to spalling
occurrence include the thermal-mechanical and hygro-thermal mechanisms. The risk of concrete
spalling arises together with an increase in concrete strength and density and is practically absent in
low-grade concretes characterised by high degrees of open porosity. New-generation high-performance
and ultra-high-performance concretes are particularly susceptible to the risk of fire spalling. The use of
polypropylene fibres is recommended in order to limit spalling occurrence.
Despite the numerous experiments conducted, the complex numerical hygro-thermal models
developed. It is still not possible to effectively determine the risk of fire spalling for concrete with a
given composition and properties. Spalling is a complex process that is affected by the composition of
concrete (w/c ratio, type of aggregate) and its properties (gas permeability, density, and open pores
network) and the spalling risk must be evaluated for a given application.
The effectiveness of polypropylene fibres depends on their melting temperature, geometry length,
and cross-section. We have shown that melting processes of fibres cause the creation of micro-cracks
and contributes to the lower water vapour pressure. The melting temperature of PP fibres (170 ◦ C)
is very close to the peak measured by a number of experimental peaks of pore pressure created in a
concrete section exposed to fire. Therefore, the addition of PP fibres to concrete provide successful
results in reducing spalling risk due to fire.
The quantity of fibres enabling the percolation of open pores in the matrix, which limits the gas
pore pressure picks and prevents spalling, must be determined, and the experimental techniques
are commonly used for this purpose. In addition, the manner in which concrete is loaded and the
stress conditions during fire (presence of compressive stresses and restraint thermal expansion) plays
an important role that needs to be considered when spalling risk is evaluated. Several examples of
different approaches for assessing spalling risk in concrete due to fire were presented. It is undeniable
that determining mechanical and physical properties of concrete under elevated temperature are vital
in the design process. However, in the case of structures being at high risk of fire occurrence, it is
recommended to provide the assessment of concrete susceptibility to spall due to fire exposure. The
most appropriate approach for the fire test is to ensure the conditions that reflect the intended use of a
targeted concrete element.
The complex nature of the problem discussed, the absence of standard procedures for testing the
susceptibility of concrete to fire spalling, and the stochastic nature of the process will undoubtedly
result in further work on spalling and methods of spalling risk mitigation.
Author Contributions: Conceptualization, I.H. Data curation, I.H. and K.M. Formal analysis, K.M. Funding
acquisition, I.H. and K.M. Investigation, I.H. and K.M. Methodology, I.H. Supervision, I.H. Validation, K.M.
Writing—original draft, I.H. and K.M., Writing—review and editing, K.M. and I.H.
Funding: The study was financially supported. The National Science Centre, Poland (NCN, grant
number: 2016/23/N/ST8/01155), by the EMMAT project, E-mobility and sustainable materials and technologies
(PPI/APM/2018/1/00027), and by the Polish National Agency for Academic Exchange (NAWA) partly funded this
research. The author, K. Mróz, obtained financial resources as part of the financing of a doctoral scholarship from
the National Science Centre, Poland (NCN 2018/28/T/ST8/00352).
Materials 2019, 12, 3869 19 of 20
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Neville, A.M. Właściwości Betonu, 1st ed.; Arkady: Warszawa, Poland, 1977; p. 342.
2. Khoury, G.A. Polypropylene Fibres in Heated Concrete. Part 2: Pressure Relief Mechanisms and Modelling
Criteria. Mag. Concr. Res. 2008, 60, 189–204. [CrossRef]
3. Khoury, G.A.; Anderberg, Y.; Both, K.; Fellinger, J.; Hoj, N.; Majorana, C. Fire Design of Concrete
Structures—Materials, Structures and Modelling. Fib Bull. 2007, 38, 6245.
4. EN 1992-1-2. Eurocode 2: Design of Concrete Structures—Part 1–2: General Rules—Structural Fire Design;
Authority: The European Union Per Regulation 305/2011, Directive 98/34/EC, Directive 2004/18/EC; European
Union: Brussels, Belgium, 2004.
5. Himmelwright, A.L.A. The San Francisco Earthquake and Fire; a Brief History of the Disaster; a Presentation of
Facts and Resulting Phenomena, with Special Reference to the Efficiency of Building Materials, Lessons of the Disaster;
The Roebling Construction Company: New York, NY, USA, 1906.
6. Hertz, K.D. Limits of Spalling of Fire-Exposed Concrete. Fire Saf. J. 2003, 38, 103–116. [CrossRef]
7. Jansson, R.; Boström, L. Experimental Study of the Influence of Polypropylene Fibres on Material Properties
and Fire Spalling of Concrete. In Proceedings of the 3rd International Symposium on Tunnel Safety and
Security (ISTSS), Stockholm, Sweden, 12–14 March 2008.
8. Pimienta, P.; Anton, O.; Mindeguia, J.; Avenel, R.; Cuypers, H.; Cesmat, E. Fire Protection of Concrete
Structures Exposed to Fast Fires. In Proceedings of the 4th International Symposium on Tunnel Safety and
Security (ISTSS), Frankfurt, Germany, 17–18 March 2010; pp. 235–247.
9. Han, C.G.; Hwang, Y.S.; Yang, S.H.; Gowripalan, N. Performance of Spalling Resistance of High Performance
Concrete with Polypropylene Fiber Contents and Lateral Confinement. Cem. Concr. Res. 2005, 35, 1747–1753.
[CrossRef]
10. Kalifa, P.; Chéné, G.; Gallé, C. High-Temperature Behavior of HPC with Polypropylene Fibers from Spalling
to Microstructure. Cem. Concr. Res. 2001, 31, 1487–1499. [CrossRef]
11. Hager, I.; Mróz, K.; Tracz, T. Contribution of Polypropylene Fibres Melting to Permeability Change in Heated
Concrete—The Fibre Amount and Length Effect. IOP Conf. Ser. Mater. Sci. Eng. 2019, 706.
12. Mróz, K.; Hager, I. Causes and Mechanism of Concrete Spalling under High Temperature Caused by Fire.
Cem. Wapno Beton 2017, 6, 445–456.
13. Hager, I.; Tracz, T.; Śliwiński, J.; Krzemień, K. The Influence of Aggregate Type on the Physical and Mechanical
Properties of High-Performance Concrete Subjected to High Temperature. Fire Mater. 2016, 40, 668–682.
[CrossRef]
14. Khoury, G.A.; Anderberg, Y. Concrete Spalling Review. Fire Saf. Des. 2000. [CrossRef]
15. Jansson, R.; Boström, L. Fire Spalling in Concrete—The Moisture Effect, Part II. MATEC Web Conf. 2013,
6, 03003. [CrossRef]
16. Dauti, D.; Weber, B.; Dal Pont, S.; Tengattini, A.; Toropovs, N.; Briffaut, M. First Results on Fast Neutron
Tomography of Heated Concrete. In Proceedings of the 5th International Workshop on Concrete Spalling
due to Fire Exposure, Boras, Sweden, 12–13 October 2017; pp. 249–258.
17. Hager, I.; Tracz, T.; Choińska, M.; Mróz, K. Effect of Cement Type on the Mechanical Behavior and Permeability
of Concrete Subjected to High Temperatures. Materials 2019, 12, 3021. [CrossRef] [PubMed]
18. Bulletin of the Polish Academy of Sciences. Behaviour of Cement Concrete at High Temperature. Report
2013, 61, 1–10. [CrossRef]
19. Carré, H.; Pimienta, P.; La Borderie, C.; Pereira, F.; Mindeguia, J.-C. Effect of Compressive Loading on the
Risk of Spalling. MATEC Web Conf. 2013, 6, 01007. [CrossRef]
20. Miah, M.J. The Effect of Compressive Loading and Cement Type on the Fire Spalling Behaviour of Concrete.
Ph.D. Thesis, University of Pau and Pays de l’Adour, Pau, France, 2017.
21. Lo Monte, F.; Felicetti, R. In-Plane Loaded Concrete Slabs Subjected to Fire: A Novel Test Set-up to
Investigate Spalling. In Proceedings of the IFireSS—International Fire Safety Symposium, Coimbra, Portugal,
20–22 April 2015.
22. Kalifa, P.; Menneteau, F.D.; Quenard, D. Spalling and Pore Pressure in HPC at High Temperatures. Cem. Concr.
Res. 2000, 30, 1915–1927. [CrossRef]
Materials 2019, 12, 3869 20 of 20
23. Mindeguia, J.-C.; Pimienta, P.; Hager, I.; Hélène, C. Influence of Water Content on Gas Pore Pressure in
Concrete at High Temperature. In Proceedings of the 2nd International RILEM Workshop on Concrete
Spalling, Delft, The Netherlands, 5–7 October 2011; Volume I, pp. 113–121.
24. Felicetti, R.; Lo Monte, F.; Pimienta, P. A New Test Method to Study the Influence of Pore Pressure on Fracture
Behaviour of Concrete during Heating. Cem. Concr. Res. 2017, 94, 13–23. [CrossRef]
25. Meyer-Ottens, C. The Question of Spalling of Concrete Structural Elements of Standard Concrete under Fire
Loading. Ph.D. Thesis, Technical University of Braunschweig, Braunschweig, Germany, 1972.
26. Mróz, K.; Hager, I.; Korniejenko, K. Material Solutions for Passive Fire Protection of Buildings and Structures
and Their Performances Testing. Procedia Eng. 2016, 151, 284–291. [CrossRef]
27. Jansson, R.; Boström, L. The Influence of Pressure in the Pore System on Fire Spalling of Concrete. Fire
Technol. 2010, 46, 217–230. [CrossRef]
28. Hager, I.; Zdeb, T.; Krzemień, K. The Impact of the Amount of Polypropylene Fibres on Spalling Behaviour
and Residual Mechanical Properties of Reactive Powder Concretes. MATEC Web Conf. 2013, 6, 02003.
[CrossRef]
29. Pasquini, N. Polypropylene Handbook, 2nd ed.; Hanser Publishers: Munich, Germany, 2005.
30. Bentz, D.P. Fibers, Percolation, and Spalling of High Performance Concrete. ACI J. 2000, 97, 351–359.
31. Mindeguia, J.-C.; Pimienta, P.; Carré, H.; La Borderie, C. Experimental Study on the Contribution of Pore
Vapour Pressure to the Thermal Instability Risk of Concrete. In Proceedings of the 1st International Workshop
on Concrete Spalling due to Fire Exposure, Leipzig, Germany, 3–5 September 2009; pp. 150–167.
32. Hager, I.; Tracz, T. The Impact of the Amount and Length of Fibrillated Polypropylene Fibres on the Properties
of HPC Exposed to High Temperature. Arch. Civ. Eng. 2010, LVI, 57–68. [CrossRef]
33. Hager, I.; Pimienta, P. Wpływ Dodatku Włókien Polipropylenowych Na Właściwości Mechaniczne Betonów
Wysokowartościowych Poddanych Działaniu Wysokiej Temperatury. Cem. Wapno Beton 2003, 8/70, 263–266.
34. Li, Y.; Tan, K.H. Synergistic Effect of Polypropylene Fiber and Aggregate Size on Spalling Behaviour of
Ultra-High Performance Concrete. In Proceedings of the 5th International Workshop on Concrete Spalling
due to Fire Exposure, Borås, Sweden, 12–13 October 2017; pp. 83–91.
35. ISO 834-1:1999. Fire-Resistance Tests—Elements of Building Construction—Part 1: General Requirements; ISO:
Geneva, Switzerland, 1999.
36. Huismann, S.; Weise, F.; Meng, B.; Schneider, U. Transient Strain of High Strength Concrete at Elevated
Temperatures and the Impact of Polypropylene Fibers. Mater. Struct. Constr. 2012, 45, 793–801. [CrossRef]
37. Tanibe, T.; Ozawa, M.; Lustoza, D.R.; Kikuchi, K.; Morimoto, H. Explosive Spalling Behaviour of Restrained
Concrete in the Event of Fire. In Proceedings of the 2nd International RILEM Workshop on Concrete Spalling
due to Fire Exposure, Delft, The Netherlands, 5–7 October 2011; pp. 319–326.
38. Akasaka, H.; Ozawa, M.; Parajuli, S.S.; Sugino, Y.; Akutsu, Y.; Murakami, M. Preventive Effect on Fire
Spalling of High-Strength Concrete With Jute Fibre in Ring-Restraint Specimen. IOP Conf. Ser. Mater. Sci.
Eng. 2018, 431, 042001. [CrossRef]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Materials 12 03869

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Materials 12 03869

Uploaded by

Copyright:

Available Formats

materials

Keywords: concrete spalling; temperature; fire; fibres; gas pore pressure

Materials 2019, 12, 3869; doi:10.3390/ma12233869 www.mdpi.com/journal/materials

and often 2019,

2.2. Pore Pressure Measurements

3. Polypropylene Fibre Action

3.1. PP Fibre Behaviour at High Temperatures

Materials 2019, 12, x FOR PEER REVIEW 7 of 21

(a) (b) (c)

3.2. Change of Transport Properties in Heated Concretes with PP Fibres

Materials 2019, 12, x FOR PEER REVIEW 11 of 21

Similarly, small samples were also tested by by Li

Figure 15. The

RPC_0.5 degC/min RPC_1 degC/min RPC_2 degC/min

yes [4/4] yes [4/4] yes [2/1] yes [2/2]

Conflicts of Interest: The authors declare no conflict of interest.

You might also like