I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
243–252
A PILOT STUDY OF METHODS FOR MEASURING THE
RESIDUAL PROPERTIES OF CONCRETE EXPOSED TO
ELEVATED TEMPERATURES
PILOTNA [TUDIJA MERJENJA ZAOSTALIH LASTNOSTI BETONA
IZPOSTAVLJENEGA POVI[ANIM TEMPERATURAM
Iva Rozsypalová, Petr Danìk, Hana [imonová, Zbynìk Ker{ner, Dalibor Kocáb
Brno University of Technology, Faculty of Civil Engineering, Veveøí 331/95, 602 00 Brno, Czech Republic
iva.rozsypalova@vutbr.cz
Prejem rokopisa – received: 2017-01-09; sprejem za objavo – accepted for publication: 2017-12-21
doi:10.17222/mit.2017.004
This pilot study and literature review was performed with the purpose of preparing for subsequent extensive research focused on
the design of effective diagnostic methods for determining the current state of concrete structures damaged by high temperatures
during exposure to fire. During the pilot study, specimens were prepared from ordinary concrete with dense aggregate, and were
subsequently loaded by temperatures of (200, 400, 600, 800, 1000 and 1200) °C. In addition, one set of the specimens was kept
as reference specimens without thermal loading. After the thermal loading of the six sets of specimens had cooled, selected
non-destructive tests were performed; i.e., the Silver Schmidt hammer rebound test, the ultrasonic pulse velocity test, and the
impact-echo method. Furthermore, the compressive strength of the specimens was determined destructively to provide reference
data. The experiments indicated the fact that the commonly used non-destructive methods are not suitable for estimating the
compressive strength of concrete exposed to temperatures higher than 600 °C. This indicates that it is necessary to derive new
calibration relationships for such conditions. It was also found that an apparent increase occurred in the hardness and modulus
of elasticity for the tested concrete when it was exposed to temperatures greater than 1000 °C.
Keywords: concrete, high temperature, Schmidt rebound, ultrasonic pulse velocity
S pomo~jo te pilotne {tudije in literaturnega pregleda so se avtorji pripravili na nadaljnjo intenzivno raziskavo oblikovanja
u~inkovite metode dolo~evanja stanja po{kodovanih betonskih konstrukcij po po`arih zaradi njihove izpostavljenosti visokim
temperaturam. Med pilotno {tudijo so pripravili vzorce iz obi~ajnega betona z gostimi agregati in jih nato izpostavili visokim
temperaturam (200, 400, 600, 800, 1000 and 1200) °C. Eno garnituro vzorcev, ki ni bila izpostavljena visokim termi~nim
obremenitvam so obdr`ali kot referen~ni material. Po termi~ni obremenitvi {estih garnitur vzorcev so jih ohladili in izvedli
izbrane metode neporu{nih preiskav in sicer: Silver-Schmidtov odbojni test s kladivom, test ultrazvo~ne pulzne hitrosti in
udarno-odmevno metodo. Nadalje so dolo~ili {e tla~no trdnost vzorcev s standardno poru{no metodo. Preiskave so nakazale, da
neporu{ne preiskave ne dajejo ustreznih vrednosti za oceno tla~ne trdnosti betona, ki je izpostavljen temperaturam nad 600 °C.
To pomeni, da je potrebno izdelati nove kalibracijske algoritme za tak{ne pogoje. Avtorji raziskave so ugotovili, da pride celo do
navideznega povi{anja trdote in elasti~nega modula pri betonu, ki je bil izpostavljen temperaturam vi{jim od 1000 °C.
Klju~ne besede: beton, povi{ane temperature, Schmidtov odboj, hitrost ultrazvo~nih pulzov
1 INTRODUCTION
High temperatures acting upon concrete trigger a
number of physical and chemical processes that result in
transformations in the composite’s structure and undesir-
able changes in its physico-mechanical properties, in
general.
1
A lot of research teams have been investigating
the changes caused by elevated temperatures. Given the
variety of concrete types, parameters and behaviours that
exist under normal conditions, different behaviours
during and after thermal loading can be expected.
1,2
Concrete made with quartz aggregate is thermally more
stable than concrete containing carbonate aggregate.
3
The degradation of compressive strength due to high
temperatures is a property that seems to be receiving
much attention.
1,4
The determination of compressive
strength by non-destructive methods has proved that
different water-to-cement (w/c) ratios have no significant
influence on the dependence of the ultrasonic pulse
velocity on the strength at a temperature of 400–600 °C.
5
The compressive strength and modulus of elasticity of
ultra-high-strength concrete (UHSC) is generally re-
duced by high temperatures to a lesser degree than is the
case with normal and high-strength concretes.
6
The
influence of the addition of various types of fibres is also
often studied. A frequently tested example involves the
addition of polypropylene (PP) fibres to the cement
matrix, with the aim of compensating for the higher pore
pressure that occurs in concrete at elevated temperatures.
The addition of PP fibres to high-strength concrete
(HSC) proved to have a positive effect; their presence
had a greater impact on the compressive strength than on
splitting the tensile strength (at temperatures above
200 °C). A significant improvement in the residual
mechanical properties of HSC was achieved by 2 kg m
–3
of PP fibres in the mixture. Concrete specimens also lose
mass at high temperatures. At up to 800 °C, the mass
loss is relatively smooth (after physically bound water
Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252 243
UDK 655.28.022.1:67.017:621.08:625.821.5 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 52(3)243(2018)

has evaporated). However, beyond this temperature mass
begins to decrease very rapidly and has been found to
decline by as much as 45 %; the mass losses were not
significantly influenced by the w/c ratio nor by the type
of aggregate.
4
The currently valid standard governing the assess-
ment of existing structures, ISO 13822:2010,
7
states that
an assessment of the real state of a building must be
conducted with the real properties of the materials from
which the building is made, not the material parameters
described in the building’s original design, or in any
standard or directive. The assessment of the real proper-
ties of these materials must take into account their
degradation and the possible loads and attacks they may
be subjected to during the structure’s service life. It is
also necessary to consider atypical forms of damage,
such as that caused by fire. However, there are no in-
structions for the efficient determination of the material
properties necessary for performing explorations of the
structural conditions of buildings damaged by fire.
This pilot study and literature review was performed
as part of the authors’ preparations for subsequent exten-
sive research focused on the design of effective diag-
nostic methods for determining the current state of
concrete structures damaged by high temperatures during
exposure to fire. This future research will involve a
variety of tests (commonly used for the determination of
the properties of steel-reinforced concrete) performed on
concrete specimens that have been loaded by high
temperatures. The tests will be carried out after the
specimens have cooled down to ambient temperature.
The suitability of each method will be determined from
the degree of change occurring in these properties after
thermal loading compared with the properties of the
reference material, which has not undergone thermal
loading. The pilot study needed to be undertaken prior to
performing this experimental research because of the
need to optimise laboratory test procedures and test the
behaviour of the concrete at elevated temperatures.
The paper focuses on the search for effective test
methods for determining the influence of elevated tem-
peratures on the physico-mechanical properties of con-
crete. Knowledge regarding this influence (possibly
supplemented by mathematical modelling) is essential
for assessing the state and behaviour of concrete struc-
tures after exposure to fire.
2 THE EFFECT OF ELEVATED
TEMEPARTURES ON CONCRETE
Concrete is considered to be a fire-resistant material.
This means that no combustion of the material will occur
if it is exposed to fire. However, it must be understood
that even though concrete is fire-resistant, it will degrade
due to high temperatures and will undergo chemical
changes which affect its physical and mechanical proper-
ties.
2.1 Chemical and physical changes in concrete
Concrete consists of a cement matrix and aggregate.
These components react differently to elevated tempera-
tures, and this causes changes at their interface. For this
reason, the following paragraphs of the paper deal with
an analysis of the effect of fire on each concrete
component, and describe the behaviour of concrete as a
whole (i.e., as a composite consisting of components
with different properties).
2.1.1 The effect of fire on hardened cement paste
The fire gradually heats the cement paste and causes
the evaporation of the water inside. Free water
evaporates first, followed by the physically bound water,
and after that, the water chemically bound in the hyd-
ration products.
1,8
If the cement paste is heated under
confined humid conditions, a hydrothermal reaction
known as internal autoclaving can occur (simultaneous
exposure to high pressure and high temperatures).
1,9
It
can activate changes in the microstructure of hydrates.
The nature of the phase changes depends mainly on the
mineral composition of the cement used, its C/S ratio
(the molar ratio of lime to silicon dioxide; CaO/SiO
2
),
the amount of fine particles the cement contains, the
temperature and pressure, and on the duration of the
thermal loading.
1,8
The mechanical properties of the
cement paste are strongly influenced by the chemical
bonds and attractive forces between the layers of calcium
hydro-silicate (CSH compound). It is assumed that
approximately 50 % of cement paste’s strength comes
from the attraction between the CSH compounds due to
their enormous specific surface area.
1,8
Dehydration
occurs between the CSH layers and causes a decrease in
hydrate volume. This in turn increases the porosity of the
cement matrix, resulting in increased overall pore
volume and size. Deformations in the cement paste are
caused by volume changes in these constituents; the hyd-
rated phase shrinks and the non-hydrated phase expands.
At temperatures around 200 °C, the cement matrix
expands slightly, while beyond this temperature, it
shrinks rapidly.
1,10
The heat first causes ettringite
decomposition; in fact, this occurs even before the tem-
perature reaches 100 °C. CSH compound dehydration
takes place gradually and starts when the heating has
begun. The cement-paste structure is partly damaged by
dehydration at temperatures as low as 105 °C, which is
incidentally a temperature commonly used for drying
building materials.
1
The heating of hardened cement
paste to 500–550 °C results in a rapid decrease in the
content of portlandite (Ca(OH)
2
), which decomposes into
water vapour and CaO.
1,10
The reaction causes an
increase in porosity and reduces the strength of the ce-
ment paste. However, research findings suggest the
hydration process can be modified by the use of pozzo-
lanic materials in concrete. Their hydration products are
calcium hydro-silicate gels, which are resistant to higher
temperatures. Their residues can be found in concrete
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
244 Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252

elements exposed to temperatures of 600–700 °C. For
this reason, it is more suitable to use blended pozzolanic
cements with a lower pozzolanic additive content and a
higher content of fly ash- or slag-based materials.
1,11
Nevertheless, when the fire is being extinguished, a large
amount of water is sprayed onto the concrete. This can
trigger expansive reactions (lime slaking), Equation (1):
CaO+H
2
O  Ca(OH)
2
(1)
and thus cause an increase in volume and damage the
concrete further.
12
2.1.2 The effect of fire on aggregate
Aggregate, like most solids, expands with rising
temperature. The thermal expansion of aggregate is thus
an important factor in the behaviour of concrete during a
fire. A typical feature of thermally stable aggregate is its
chemical and physical stability at high temperatures.
Aggregate suitable for mixing into concrete that is in-
tended to be resistant to high temperatures needs to have
a low thermal expansion coefficient as well as negligible
residual stress. For an aggregate to be considered ther-
mally stable it must not change its mechanical properties
(especially its compressive strength) or its volume at
high temperatures. Such an aggregate must not lose mass
until certain temperatures are reached, and thermal
reactions must not occur within it. The absence of peaks
in the curves obtained from differential thermal analysis
(DTA) is essential.
1,8
The mineralogical composition of aggregate deter-
mines its thermal deformation as each material has its
own specific thermal expansion coefficient. Carbonate
aggregates (limestones and dolomites) are stable up to
600 °C; higher temperatures then lead to decomposition
into CaO (portlandite) and CO
2
, which takes place at
around 700 °C. Furthermore, CaO formed by decarbona-
tion can later hydrate when the concrete cools, increas-
ing its volume by 44 %.
1,8,12
At a temperature of 573 °C, siliceous aggregate
transforms from  -t o -quartz modification, bringing
with it a significant increase in volume.
13
Research
shows that the most suitable natural aggregates appear to
be basalt, diabase and andesite. Aggregate composed of
several types of minerals can have a tendency to disinte-
grate as a result of the different thermal expansions of
each component, which then causes inter-crystalline
stress and cracking. This is the reason why fib Bulletin
No. 38
13
does not recommend the use of aggregates con-
sisting of minerals with different thermal expansions. In
order to limit the occurrence of inter-crystalline cracks,
concretes which are at risk of being exposed to tem-
peratures above 700 °C must contain artificial aggregate
which guarantees similar properties throughout its vol-
ume. In order to achieve resistance to temperatures above
1000 °C, crushed chamotte, corundum, silicon carbide,
crushed bauxite or chromite can be used as aggregate.
Most igneous rocks (e.g., granites and diorites) can melt
at such high temperatures.
1,11
2.1.3 Interaction between aggregate and cement paste
If concrete is heated, the volume of the aggregate
increases while the hardened cement paste surrounding
the aggregate shrinks. As is well known, the "interfacial
transition zone" (ITZ) between the cement matrix and
the aggregate is the weakest part within the material’s
structure, even when the concrete is not being subjected
to thermal loading. This is because of the increased
stresses that occur between the aggregate and the cement
matrix. These stresses can be increased by the above-
mentioned counteracting volume changes caused by high
temperatures.
1
Due to these counteracting volume
changes, even greater tensile stresses occur within the
ITZ and the concrete then suffers damage due to
cracking.
1
3 EXPERIMENTAL PART
3.1 Materials and specimens
A total of 21 prism specimens of (100 × 100 × 400)
mm in size were made for the purpose of studying
non-destructive methods of testing concrete damaged by
fire. The mixture consisted of Portland cement, sand
(0–4 mm), aggregate (8–16 mm), superplasticizer and
water. The use of mixture with gap-graded aggregate (the
4–8 mm fraction is absent) should improve the phys-
ico-mechanical properties of structural elements by
reducing shrinkage and creep while increasing the
strength and elasticity for the same cement content. It is
also a suitable way of reducing costs.
14
The composition
of the fresh concrete mixture can be found in Table 1.
The w/c ratio was set at 0.46 (dimensionless).
Table 1: Composition of fresh concrete
Component
Dosage (kg per 1 m
3
)
of fresh concrete
CEM I 42.5 R (Mokrá) 345
Quartz sand (@ab~ice) 0–4 mm 848
Gravel aggregate (Olbramovice)
8–16 mm
980
Sika ViscoCrete 2030 superplasticizer 2.8
Water 160
The prisms were cast and, after demoulding, cured
for 28 d under water. After drying under laboratory
conditions, the specimens were additionally dried in a
ceramic furnace at a temperature of 110 °C for 48 h.
The dry specimens were then placed into a Rohde KE
130B electric-powered laboratory furnace with an 8-kW
output, and were subjected to loading by the following set
of temperatures: (200, 400, 600, 800, 1000 and 1200) °C.
The heating rate was 5 °C/min and the nominal
maximum temperature was maintained for 60 min. Once
their exposure to elevated temperatures was complete,
the concrete specimens were allowed to gradually cool
down in the ambient air. The concrete specimens
underwent thermal loading at the age of 120 d, which is
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252 245

in accordance with RILEM’s recommendation that the
specimen age be at least 90 d.
15
The mechanical pro-
perties of the concrete were then determined 60 d after
the thermal loading took place. The reference specimens
without thermal loading (kept at 20 °C) were also tested.
Each of the seven sets of specimens produced with
the aim of determining the influence of temperature on
concrete properties with the aid of non-destructive
methods consisted of two prisms with the nominal
dimensions (100 × 100 × 400) mm and two cubes with
an edge length of 100 mm. These cubes were cut from
fragments from larger prism specimens with a central
edge notch that had previously been subjected to
three-point bending, which was another task of the
executed project. In total, the residual properties of
concrete after exposure to elevated temperatures were
tested on 14 prisms and 14 cubes.
3.2 Schmidt rebound hammer test
The rebound hammer method is used for measuring
the hardness of a material, i.e., the ability of the material
to resist penetration. The rebound hammer test was
performed using a Silver Schmidt ST/PC Type L ham-
mer due to its low impact energy (0.735 N m), because
the response of concrete after exposure to fire is ex-
pected to be brittle. The manufacturer of this device
gives a possible compressive strength range for concrete
of 10–100 MPa.
16
It is possible to use this hammer to
determine the parameters of specimens with a thickness
of less than 100 mm. The hardness was measured by
taking a minimum of 9 valid readings from the surface of
each test specimen. The distance between the measure-
ment positions was at least 25 mm, and all the positions
were at least 25 mm from the edges of the specimen.
16,17
The experiment was carried out on seven sets of spe-
cimens, each of which consisted of two prisms and two
cubes. The specimens were placed into a universal
testing machine press and stressed by a compressive load
equal to 10 % of their compressive strength. The prisms
was clamped at both ends and their hardness tested by
impacts of the plunger on the two opposing faces with
the larger surface area of the prisms (in the direction of
compaction) – each face was impacted 12 times. Valid
measurements were evaluated according to standard EN
12504-2.
17
The evaluation of the rebound value always
had at least 10 valid readings available for each
measured face. Hardness values were determined for
every tested face; these were then averaged to represent
the specimen as a whole.
The cubes were tested in a similar manner, with the
exception of the amount of faces tested; a total of three
faces of each specimen were impacted – the cross sec-
tion, the face opposite the cross-section and the bottom
face (in the direction of compaction). The evaluation of
the hardness of the cube specimens was in accordance
with the method described for the prisms. The rebound
value for one temperature was then calculated as an
average of the mean rebound value of a prism and the
mean rebound value of a cube for that temperature.
Concrete hardness expressed as a rebound value is
commonly used for estimating concrete strength. For the
SilverSchmidt Type L hammer, the relevant standard
17
recommends that the strength should be estimated using
the formula given by the manufacturer, i.e., Proceq. For
the given range of concrete strengths and rebound values
R, the dependence of compressive strength f
c,R
on the
rebound value is expressed as Equation (2):
16
f
c,R
= 1.9368·exp(0.0637·R) (2)
3.3 Ultrasonic pulse velocity method
An ultrasonic wave travelling through a material
causes its particles to vibrate in different directions. The
velocity of the ultrasonic wave depends on the material
properties (especially the modulus of elasticity) of the
specimen being tested. The velocity of ultrasonic waves
in materials is often determined by means of the ultra-
sonic pulse velocity method. The principle of the method
consists of sending ultrasonic pulses into a material and
recording them once they have passed through. The time
the pulse front needs to pass from the transmitter to the
receiver is then used for calculating the ultrasonic pulse
velocity. The modulus of elasticity is then determined
from the pulse velocity and the material’s bulk density.
18
The experiment was carried out on the same seven
sets of specimens as in the case of the Schmidt rebound
hammer test. The transit times of the ultrasonic longi-
tudinal waves pulse were measured using a Proceq
Pundit Plus with two 54-kHz transducers.
19
A sufficient
acoustic coupling between the ultrasonic transducers and
the specimen was achieved by applying plastic modell-
ing clay. Each measurement started with the calibration
of the testing instrument using a calibration rod.
18,20
In order to determine the dependence of the modulus
of elasticity on the temperature to which the concrete is
exposed, it was first necessary to determine the bulk
density of the specimens loaded by high temperatures.
The bulk density was determined from the mass and
dimensions (volume) of each specimen using the method
presented in EN 12390-7.
21
3.4 Impact-echo method
The impact-echo method is based on the principle of
analysing the reflection of an impulse-induced mecha-
nical wave. The wave generated by the impact of a
mechanical hammer on the specimen surface passes
through the structure of the material and reflects from
the defects inside the specimen or on its surface, return-
ing as vibration.
22
A receiver on the surface detects the
vibration in the time domain, which is then converted
into the resonance frequency using the fast Fourier
transform.
22,23
The experiment was carried out on seven
sets of cube specimens as in the case of the Schmidt
rebound hammer test.
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
246 Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252

3.5 Compressive cube strength tests
In the final stage of the experiment, compression
tests were performed in accordance with the corres-
ponding standards governing laboratory compressive
strength testing. The cubes, which had first been tested
by non-destructive methods, were subjected to a com-
pression test according to EN 12390-3
25
in order to
determine a trend in strength changes as a result of
exposure to elevated temperatures.
4 RESULTS AND DISCUSSION
After exposure to high temperatures, the specimens
changed their appearance. Discolorations and cracks
were visible on the surface at high temperatures. The
appearance of the specimens after thermal loading is
documented in Figure 1, where colour changes and the
formation of cracks are visible.
The normal colour of reference unheated concrete
was dark grey. After heating up to 200 °C, no colour
changes were observed. At temperatures of 400 °C and
600 °C, the surface of the concrete became slightly pink.
After heating up to 800 °C, the concrete became light
grey. After heating to 1000 °C, the grey was very light,
almost white-grey. A significant change occurred after
the concrete was heated to 1200 °C, when the concrete
became brown. These changes in colour are the result of
dehydration of the cement matrix and transformations of
the aggregate minerals.
1
There were no cracks on the surface of the specimens
heated up to 400 °C. Thermal loading up to 600 °C
caused the appearance of small micro-cracks. After the
concrete was loaded up to 800 °C, the first cracks
appeared, but they still did not cover the whole surface of
the test specimen. The most apparent cracks could be ob-
served on the specimens loaded up to 1000 °C, where the
cracks covered the whole surface, and their width
increased up to 1 mm. Near the edges of these specimens
there were small local spalls in the concrete surface that
revealed the presence of red-pink aggregate. The sur-
faces of specimens loaded up to 1200 °C were also
covered with cracks, though there were fewer than the
amount at 1000 °C. In the case of the specimens heated
to 1200 °C, it was even possible to observe a degree of
deformation – the specimens were bent into arches, with
a deflection of approximately 3 mm. The cracks created
in the specimens were caused by volume changes occurr-
ing during their heating due to dehydration of the cement
matrix and transformations of the aggregate minerals.
1
4.1 Compressive strength tests
Figure 2 shows there was a relatively uniform de-
crease in the compressive strength of the concrete. The
compressive strength determined for the reference un-
loaded concrete specimens was approx. 70 MPa. Expo-
sure to a temperature of 200 °C caused this strength to
drop by 5.7 % (down to 65.9 MPa), while a temperature
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252 247
Figure 2: Development of compressive strength in relation to tempe-
rature
Figure 1: Colour changes of heated concrete (maximal temperatures 20–1200 °C)

of 400 °C resulted in a 21.2 % decrease (to 55.0 MPa),
600 °C decreased it by 45.3 % (to 38.2 MPa), 800 °C
caused a 53.2 % drop (to 32.7 MPa), 1000 °C resulted in
a 78.8 % decrease (to 14.8 MPa), and the strength of the
concrete damaged by the highest temperature of 1200 °C
decreased by 80.2 % (to 13.8 MPa). The decreasing
strength of the tested concrete cubes can be described by
the following Equation (3):
f
c
= –0.0529·T + 73.345 (3)
where f
c
is the compressive strength of the concrete
(MPa) and T is the maximum nominal thermal load
(°C). The strength shows a strong negative dependence
on the increasing temperature.
The decrease in the compressive strength is caused
by the high thermal loading, which results in chemical
and physical changes and subsequent gradual disintegra-
tion of the material’s internal structure. Section 2 dis-
cusses this issue in more detail. One of the main causes
of the decrease in strength can be attributed to the cracks
that form and propagate as a consequence of these
changes.
4.2 Schmidt rebound hammer tests
Figure 3 shows the rebound values of the thermally
loaded concretes. The rebound value of the reference
concrete was 55.9 Q. As the temperature increased, the
rebound values changed. The temperature of 200 °C
caused a very slight decrease in the rebound value; it was
reduced by 1.2 % to 55.2 Q. The temperature of 400 °C
reduced it by 8.7 % to 51.1 Q, while 600 °C caused it to
decrease by 19.1 % to 45.2 Q, 800 °C reduced it by 43 %
to 31.9 Q, 1000 °C caused a 53.5 % decrease to 26.0 Q,
and heating up to 1200 °C reduced the rebound value by
23.9 % of the original, so down to 42.5 Q. This trend is
visible in Figure 3, which shows the results of the
rebound value measurements. The temperature of 200 °C
causes no significant changes in the rebound value,
which means that the hardness of the concrete is not
reduced. Further increases in temperature slowly reduce
the rebound value up to a temperature of 1000 °C. The
slow decrease in concrete hardness is caused by gradual
transformations in the internal structure and the appear-
ance of micro-cracks and cracks. On the other hand,
thermal loading of the concrete above 1000 °C and up to
1200 °C results in a relatively sharp increase in the
rebound values, and thus in hardness as well. The values
approach the values measured for concrete heated to
600 °C. The increase in hardness is caused by changes in
the material structure – new crystalline phases are
formed. The high temperatures cause the transformation
of an important component of concrete – calcite (calcium
carbonate CaCO
3
) – into a mineral known as wollasto-
nite (CaSiO
3
). According to the Mohs scale of mineral
hardness, calcite has a hardness of 3, whereas the newly
formed wollastonite is far harder, with a hardness of
5.
26,27
The compressive strength values calculated from the
rebound values are displayed in Figure 4, which shows
the trends of compressive strength development in rela-
tion to maximum nominal temperature – the black line
shows the development of compressive strength
measured by the compression test, while the grey line
shows changes in compressive strength calculated from
the rebound values obtained by the rebound hammer test.
Only the compression test can be considered conclusive
in this case. The rebound hammer test proved unsuitable
for temperatures higher than 600 °C here. The values at
800 °C are too low, while on the other hand, at tempe-
ratures above 1000 °C, the test returns values which are
much higher compared to those obtained by destructive
tests (due to the formation of new minerals and struc-
tures). At temperatures above 600 °C, the use of the
rebound hammer test cannot be recommended for
estimating the compressive strength of concrete since its
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
248 Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252
Figure 4: Comparison of the concrete compressive strengths cal-
culated from the rebound values (according to the formula by Proceq)
with the strengths determined by the compression test – average
strength values are plotted in relation to thermal load
Figure 3: Rebound values measured from cubic and prismatic
specimens in relation to temperatures and average rebound values for
concrete heated up to selected temperatures

behaviour is significantly different in terms of physical
quantities and hardness. Hardness is generally consid-
ered a good indicator of concrete strength. However,
these test results indicate the necessity to limit the scope
of this assumption – after taking a broader spectrum of
specimens into account.
17
The usability of this method can be considerably
limited in common practice due to the necessity of per-
forming the tests on a flat surface. Such surfaces may not
always be available due to the danger of explosive
spalling (i.e., the breakdown of the surface layer), which
can occur during fires even at relatively low temperatures
due to the thermal expansion of water.
4.3 The ultrasonic pulse velocity method
The first step was to determine the bulk density of the
thermally loaded specimens.
21
Figure 5 shows the
gradual decrease in bulk density across the temperature
scale; this decrease is caused by the disappearance of
water. First, free and physically bound water was re-
moved in a dryer. This is why there are no significant
decreases in bulk density at lower temperatures in the
ceramic furnace. Compared with the value obtained for
the reference concrete of 2340 kg m
–3
, there is only a
2.6 % reduction to 2280 kg·m
–3
at temperatures below
600 °C. Above this temperature, chemically bound water
is removed as well, especially via the dehydration of
cement hydration products. This is why both bulk den-
sity and mass decrease: at a temperature of 1200 °C the
bulk density reached 2040 kg m
–3
, which meant a
reduction of 12.8 % compared with the reference bulk
density.
The main parameter observed by the method was the
ultrasonic pulse velocity. The results are shown in Fig-
ure 6. The ultrasonic pulse front travelled through the
reference concrete at an average velocity of 4420 m s
–1
.
17
As the thermal loading was increased up to 200 °C, the
average velocity decreased by 5.7 % to 4280 m s
–1
, while
during loading up to 400 °C, the value decreased by
25.8%to3280ms
–1
. At a temperature of 600 °C, it de-
creased by 48.2 % to 2290 m s
–1
, at 800 °C, it decreased
by 67.2 % to 1450 m s
–1
, at 1000 °C, it dropped by
76.0 % to 1060 m s
–1
, and at 1200 °C, it decreased by
51.4 % from the reference value to 2150 m s
–1
. Figure 6
shows a constant decrease in ultrasonic pulse velocity
until the temperature reached 1000 °C. At higher tempe-
ratures, the structural changes cause the ultrasonic wave
velocity to increase again.
The overall structure of the concrete changes when
its temperature exceeds 1000 °C. Due to the melting of
cement paste, the conversion of hydraulic bonds into
ceramic ones is initiated. For a new material with
ceramic bonds, it is expected to increase the material’s
elasticity. The increase in ultrasonic wave velocity is
attributed to this phenomenon.
28,29
Figure 6 clearly shows the different development of
ultrasonic pulse velocity in relation to specimen shape –
a cube with an edge length of 100 mm and a prism with a
width of 100 mm and a length of 400 mm. The standard
describing the ultrasonic pulse velocity method
18
states
that the short pulse velocity created by vibration is inde-
pendent of the size and shape of specimens as long as the
side dimensions are above the acceptable limit. The
standard
18
lists the minimum dimensions with regard to
the pulse velocity in concrete for a 54-kHz transducer
(used in this experiment), Table 2. Despite the fact that
the side dimensions of all specimens were greater than
those stated by the standard, the specimen shape still had
an effect on short pulse velocity. No explanation of this
inconsistency has been found in the currently available
literature, which is why it would be beneficial to
investigate the issue of the real influence of specimen
shape and size on ultrasonic pulse velocity.
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252 249
Figure 6: Changes in ultrasonic pulse velocity in relation to thermal
loading; specimen shape taken into account
Figure 5: Decrease in bulk density in relation to increasing thermal
load

Table 2: Influence of specimen size on ultrasonic pulse velocity
18
Transducer
frequency
(kHz)
Ultrasonic pulse velocity in concrete (km s
–1
)
3.50 4.00 4.50
Smallest acceptable side dimensions (mm)
54 65 74 83
Subsequently, the obtained ultrasonic pulse velocity
values were used in the determination of the concrete’s
modulus of elasticity. Figure 7 shows the dependence of
the dynamic modulus of elasticity on the maximum
nominal temperature to which the concrete was exposed.
Almost the entire temperature spectrum shows a con-
siderable decrease in the modulus of elasticity. Only at a
temperature of 1000 °C is there a slight increase. The
reduction in the material’s modulus of elasticity is
caused by the overall degradation of the concrete due to
cracking. The slight increase at temperatures above
1000 °C is caused by sintering of the material and by the
formation of a new structure with wollastonite.
4.4 The impact-echo method
The dynamic modulus of elasticity was also deter-
mined by means of the impact-echo method. The test
results are plotted in Figure 7, which shows that the
losses in modulus of elasticity correspond to the results
determined by means of the ultrasonic pulse velocity
method.
4.5 Plans for the future experiments
Subsequent experimental investigations will aim to
design effective methods for the assessment of the
condition of concrete in concrete structures damaged by
fire. In addition to the standard mechanical parameters,
basic fracture properties will also be determined. Both
destructive and non-destructive tests are expected to be
performed on statistically significant sets of concrete
specimens.
Future research will also focus on obtaining suitable
data to aid in the design of methodology for determining
the residual concrete properties in steel-reinforced con-
crete structures. Thus, instead of the common system of
manufacturing and testing small concrete specimens (e.g.
cubes with an edge length of 150 mm or prisms of the
nominal dimensions (100 × 100 × 400) mm, the tests
will be performed on large steel-reinforced slabs. This
should better simulate the material’s behaviour in a real
structure.
Seven steel-reinforced concrete panels with dimen-
sions of (2300 × 1300 × 150) mm will be prepared. The
fresh concrete formula will be designed to correspond to
a concrete with ordinary fire resistance; it will be a
C30/37 strength class concrete, exposure class XC3,
containing limestone aggregate. The concrete panels will
be made with reinforcement at the bottom face. Type-K
thermocouples will be embedded in the panels in order
to measure the temperature during testing. The panels
will be identified according to the target thermal load as
T020, T200, T400, T600, T800, T1000, and T1200. For
7 d after casting, the slabs will be wrapped in a plastic
film that prevents them from drying, and stored in an
environment with a temperature of 20±2 °C. Afterwards,
they will be demoulded and stored in the ambient air for
84 d, since according to the Recommendations of
RILEM TC 200-HTC,
15
the age of the tested concrete
should be more than 90 d.
A total of 6 fire tests will be performed. The test
panels will be removed from the environment in which
they aged, and will be placed onto the top of a gas-burn-
ing fire resistance test furnace used for the study of
material behaviour at elevated temperatures. This fur-
nace is operated by the AdMaS Research Centre at Brno
University of Technology. The construction of the
furnace allows the heating of one (bottom) side of the
panel. The top face will be left to cool spontaneously in
the surrounding environment.
The thermal loading of the panels will follow the
standard temperature curve stating gas temperatures g
in
Equation (4):
 g
=20+345·log
10
(8t + 1) (4),
where t represents the time since the start of the fire in
minutes. The equation of the curve written in EN
1991-1-2
30
represents a common type of fire which may
affect a building in rare cases. Each specimen will be
heated up to its target temperature with regard to the
defined temperature curve (200, 400, 600, 800, 1000,
1200) °C. The given thermal load will then be main-
tained for 60 more minutes so that the heat may spread
uniformly throughout the whole panel. Once the panel
has been exposed to the elevated temperatures, it will be
left on top of the furnace until it cools down to the
ambient temperature.
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
250 Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252
Figure 7: Dependence of the dynamic moduli of elasticity E
bu
, E
br
on
thermal loading

4.5.1 Non-destructive testing of the panels
The Silver Schmidt L rebound hammer will be used
to determine the rebound number according to EN
12504-2
17
. This number will then be recalculated to the
compressive strength using the calibration relationship
provided by the manufacturer as well as the standard.
Using the ultrasonic pulse velocity test in accordance
with EN 12504-4,
18
the transit time of the ultrasonic
pulse through the specimen will be measured at the same
positions on the surface of the panels. This value is
required for the calculation of the velocity of the ultra-
sonic pulse. The dynamic modulus of elasticity will be
calculated from the ultrasonic pulse velocity, and the
bulk density determined according to EN 12390-7.
21
4.5.2 Destructive tests
Each panel will be cut into a set of specimens, which
will then be tested in accordance with the corresponding
standards:
• 6 core specimens of 50 mm in diameter and cut to a
length of 100 mm – for the determination of cylinder
compressive strength (EN 12390-3),
25
• 6 core specimens of 100 mm in diameter and cut to a
length of 100 mm – for the determination of splitting
tensile strength (EN 12390-3),
25
• 6 prisms with dimensions of (100 × 100 × 400) mm –
for the determination of flexural strength (EN
12390-5),
31
• 6 prisms with dimensions of (100 × 100 × 400) mm
and provided with an initial central notch – for
three-point bending tests. The data from the fracture
tests (load vs. displacement diagrams) will be applied
in the evaluation of the modulus of elasticity values.
32
In addition, the fracture energy
33
and effective
fracture toughness
32
values will be determined. Apart
from the determination of commonly used values,
attention will be paid to the application of the
Double-K fracture model
34
in order to identify the
initiation component of the stress intensity factor (to
be used in the calculation of the load at which stable
crack propagation starts in an initial stress
concentrator).
4.5.3 Evaluation of the planned tests
Percentage differences between the individual
physico-mechanical properties of the concrete specimens
and panels after high thermal loading and subsequent
cooling will be determined, making use of different
methods. The parameters determined for the unheated
concrete (T20) will be used as a reference.
The methods will be arranged according to the per-
centage differences between the values measured for
high temperature exposure and the results of tests
performed on specimens not exposed to fire. In cases
when a method provides significant changes in proper-
ties, it will be recommended for the testing of concrete
structures exposed to elevated temperatures.
Calibration relationships will be determined between
the destructive and non-destructive methods.
5 CONCLUSIONS
The pilot research described in this article has
gathered the most important information about the beha-
viour of concrete during a fire, especially from the
material perspective. The data from this research will be
used during upcoming extensive research that will focus
on searching for effective test methods for determining
the influence of elevated temperatures on the physico-
mechanical and mechanical fracture properties of con-
crete. This knowledge (possibly supplemented by mathe-
matical modelling) is essential for assessing the state and
behaviour of concrete structures after exposure to fire.
The experiment results have revealed the following
facts:
The necessity to investigate the real influence of
specimen shape and size on ultrasonic pulse velocity.
• All the non-destructive tests performed on concrete
exposed to elevated temperatures have proved the fact
that many of its properties improve to different
extents at temperatures above 1000 °C. At these
extremely high temperatures, minerals undergo trans-
formations which cause the hardness of concrete as
well as its dynamic modulus of elasticity to increase.
However, compressive strength continues to degrade
at these temperatures.
• Concrete hardness determined using the Silver-
Schmidt rebound hammer, as well as the modulus of
elasticity determined via the ultrasonic pulse velocity
or impact-echo method, correspond to the destruc-
tively determined compressive strength of concrete at
temperatures of up to 600 °C. However, at tempe-
ratures higher than that, the dependencies begin to
deviate. For this reason, the non-destructive methods
used here cannot be recommended for the deter-
mination of the compressive strength of concrete
exposed to temperatures above 600 °C.
• It will be necessary to find suitable, effective, and
commonly applicable methods for the determination
of the properties of concrete. One solution could be
to use calibration relationships which more closely
correspond to the real properties of a material
subjected to such a high thermal load.
Acknowledgment
This outcome has been achieved with the support of
the specific research programme at Brno University of
Technology, project No. FAST-S-17-4304. The financial
support of Czech Science Foundation project No.
16-18702S (AMIRI) is also gratefully acknowledged.
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252 251

6 REFERENCES
1
I. Hager, Behaviour of cement concrete at high temperature, Bull.
Pol. Acad. Sci.-Te., 61 (2013), 145–154, doi:10.2478/bpasts-2013-
0013
2
EN 1992-1-2:2004 – Eurocode 2: Design of concrete structures, Part
1-2: General rules, Structural fire design, Brussels
3
G. A. Khoury, Compressive strength of concrete at high tempe-
ratures: a reassessment, Mag. Concrete Res., 44 (1992), 291–309,
doi:10.1680/macr.1992.44.161.291
4
O. Arioz, Effects of elevated temperatures on properties of concrete,
Fire Safety J., 42 (2007), 516–522, doi:10.1016/j.firesaf.2007.01.003
5
H. Yang, Y. Lin, Ch. Hsiao, J.-Y. Liu, Evaluating Residual Compres-
sive strength of concrete at elevated temperatures using ultrasonic
pulse velocity, Fire Safety J., 44 (2009), 121–130, doi:10.1016/
j.firesaf.2008.05.003
6
M.-X. Xiong, J.Y. Richard Liew, Mechanical behaviour of ultra-high
strength concrete at elevated temperatures and fire resistance of
ultra-high strength concrete filled steel tubes, Mater. Design, 104
(2016), 414–427, doi:10.1016/j.matdes.2016.05.050
7
ISO 13822:2010 – Bases for design of structures, Assessment of
existing structures, Brussels
8
Q. Ma, R- Guo, Z. Zhao, Z. Lin, K. He, Mechanical properties of
concrete at high temperature – A review, Constr. Build. Mater., 93
(2015), 371–383, doi:10.1016/j.conbuildmat.2015.05.131
9
J. Piasta, Heat deformations of cement paste phases and the
microstructure of cement paste, Mater. Struct., 17 (1984), 415–420,
doi:10.1007/BF02473981
10
E. Menéndez, L. Vega, C. Andrade, Use of decomposition of port-
landite in concrete fire as indicator of temperature progression into
the material: Application to fire-affected builds, J. Therm. Anal.
Calorim, 110 (2012), 203–209, doi:10.1007/s10973-011-2159-4
11
J. Válek, Research and development of composites materials with
higher resistance to high temperatures, doctoral thesis, Brno Univer-
sity of Technology, Brno 2014, 188
12
P. Rovnaník, A. Dufka, Effect of the aggregate type on the properties
of alkali-activated slag subjected to high temperatures, Mater.
Tehnol., 49 (2015), 709–713, doi:10.17222/mit.2014.116
13
fib Bulletin No. 38, Fire design of concrete structures, Materials,
structures and modelling, fédération internationale du béton,
Stuttgart 2007, 106
14
M. Collepardi, The new concrete, 2
nd
ed., Grafiche Tintoretto, 2010,
436
15
Recommendation of RILEM TC 200-HTC: Mechanical concrete
properties at high temperatures – modelling and applications, Mater.
Struct., 40 (2007) 9, 855–864, doi:10.1617/s11527-007-9286-1
16
Proceq, Schmidt rebound hammers for concrete and rock testing.
https://www.proceq.com/
17
EN 12504-2:2012 – Testing concrete in structures, Part 2: Non-des-
tructive testing, Determination of rebound number, Brussels
18
EN 12504-4:2004 – Testing concrete, Part 4: Determination of
ultrasonic pulse velocity, Brussels
19
Proceq, Pundit ultrasonic pulse velocity and pulse echo tomography
testing. https://www.proceq.com/
20
U. Dilek, Ultrasonic Pulse Velocity in nondestructive evaluation of
low quality and damaged concrete and masonry construction, J.
Perform. Constr. Fac., 21 (2007), 337–344, doi:10.1061/(ASCE)
0887-3828(2007)21:5(337)
21
EN 12390-7:2009 – Testing hardened concrete, Part 7: Density of
hardened concrete, Brussels
22
K. Krzemieñ, I. Hager, Post-fire assessment of mechanical properties
of concrete with the use of the impact-echo method, Constr. Build.
Mater., 96 (2015), 155–163, doi:10.1016/j.conbuildmat.2015.08.007
23
F. Schubert, B. Köhler, Ten Lectures on Impact-Echo, J. Nondestruct.
Eval., 27 (2008), 5–21, doi:10.1007/s10921-008-0036-2
24
L. Topoláø, P. Rypák, K. Tim~áková-[amárková, L. Pazdera, P. Rov-
naník. Non-traditional non-destructive testing of the alkali-activated
slag mortar during the hardening, Mater. Tehnol., 50 (2016), 7–10,
doi:10.17222/mit.2014.130
25
EN 12390-3:2009 – Testing hardened concrete, Part 3: Compressive
strength of test specimens, Brussels
26
Webmineral, Wollastonite-1A Mineral Data. http://www.webmine-
ral.com/data/Wollastonite-1A.shtml#.WgwAO0qnE2w
27
Webmineral, Calcite Mineral Data. http://webmineral.com/data/
Calcite.shtml#.WgwAlUqnE2y
28
Z. Ba`ant, M. F. Kaplan, Concrete at high temperatures: Material
properties and mathematical models, 1
st
ed., Longman Group
Limited, Essex 1996, 412
29
L. Pazdera, L. Topolar, K. Mikulasek, J. Smutny, H. Seelmann,
Non-linear characteristics of temperature degraded concrete at high
temperature, Procedia Eng, 190 (2017), 100–105, doi:10.1016/
j.proeng.2017.05.313
30
EN 1991-1-2:2002 – Eurocode 1, Actions on structures, Part 1-2:
General actions, Actions on structures exposed to fire, Brussels
31
EN 12390-5:2009 – Testing hardened concrete, Part 5: Flexural
strength of test specimens, Brussels
32
B. L. Karihaloo, Fracture Mechanics and Structural Concrete, Long-
man Scientific & Technical, 1995 New York, 346
33
RILEM TC-50 FMC (Recommendation): Determination of the
fracture energy of mortar and concrete by means of three-point bend
test on notched beams, Materials & Structures, 18 (1985), 285–290
34
S. Kumar, S. V. Barai, Concrete Fracture Models and Applications,
Springer, Berlin 2011, 406
I. ROZSYPALOVÁ et al.: A PILOT STUDY OF METHODS FOR MEASURING THE RESIDUAL ...
252 Materiali in tehnologije / Materials and technology 52 (2018) 3, 243–252