Construction and Building Materials 151 (2017) 720–731

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Experimental study on the thermal properties of lightweight aggregate

concretes at different moisture contents and ambient temperatures
L.H. Nguyen a, A.-L. Beaucour a,⇑, S. Ortola b, A. Noumowé a
a
University of Cergy-Pontoise, Laboratoire de Mécanique & Matériaux du Génie Civil, EA 4114, F-95000 Cergy-Pontoise, France
b
Université Paris Ouest Nanterre la Défense, Laboratoire Energétique Mécanique Electromagnétisme (LEME), EA4416, 50 rue de Sèvres, 92410 Ville d’Avray, France

h i g h l i g h t s

Twelve mixtures of LWAC made of LWA from three different natures were tested.

Thermal properties of LWAC were measured at different temperatures and moisture content.

The conductivity increases between 5 °C and 35 °C and then stabilizes between 35 °C and 50 °C.

The thermal conductivity and specific heat of LWAC increase linearly with the moisture content.

The increase of the moisture content results in a low variation (4%) of the thermal diffusivity.

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

Article history: This study reports the influence of moisture content and temperature on the thermal properties of light-
Received 9 June 2016 weight aggregate concretes. Seven types of coarse and fine lightweight aggregates from three different
Received in revised form 1 June 2017 natures were used. The thermal tests were performed on concrete samples at three different moisture
Accepted 15 June 2017
conditions: dry state, partial saturation state and saturated state. At the dry state, measures have been
Available online 4 July 2017
taken at four different temperatures: 5 °C, 20 °C, 35 °C and 50 °C. The results show a great dependence
of thermal conductivity and specific heat of lightweight aggregate concretes on the moisture content.
Keywords:
However, concrete thermal diffusivity is not much influenced by the moisture state. These variations
Thermal performance
Conductivity
are discussed as a function of the type of lightweight aggregate and it volume fraction. The thermal con-
Specific heat ductivity increases with the temperature between 5 and 35 °C and then stabilizes between 35 and 50 °C.
Diffusivity Ó 2017 Elsevier Ltd. All rights reserved.
Moisture content
Temperature
Aggregate nature
Structural concrete
Lightweight aggregate

1. Introduction mechanical performance of LWAC. The replacement of fine aggre-

Recent studies have shown the better thermal performance of
lightweight aggregate concrete (LWAC) compared to normal
strength concrete [1–5]. Use of appropriate mixes could lead to
the LWAC with sufficient mechanical strength for structural use
[1–10]. In most cases, structural lightweight aggregate concretes
are made of coarse lightweight aggregate (LWA) and fine normal-
weight aggregate (NWA) [2–11]. LWA used in the structural light-
weight concretes are frequently expanded clay, expanded shale or
pumice [2,6–13]. Nguyen et al. [1] showed that using fine LWA
instead of fine NWA allows reducing the concrete density, thus
increasing its insulating capacity, without much lowering the
⇑ Corresponding author.
E-mail address: anne-lise.beaucour@u-cergy.fr (A.-L. Beaucour).
gate can lead to a decrease up to 70% of the thermal conductivity
of structural LWAC [1].
Up to now, the influence of moisture content on the thermal
properties of LWAC is not much studied. We can mention here
some studies on the properties of wood-concrete and autoclaved
concrete [14–17]. However, in real conditions, new concrete is well
saturated and dries very slowly. Concrete do not remain in satu-
rated state and will not reach the dry state as drying in oven. In
practice, the concrete structure is subjected to variations in tem-
perature and humidity during its life. Furthermore, heat transfer
in a concrete is more complex than in other materials. The concrete
is not only a composite material, but concrete components are also
porous media where there are solid, liquid and air/vapor phases.
The air’s moisture variation could modify the liquid water content
in concrete, by condensation of water vapor in the pores. Thus the

http://dx.doi.org/10.1016/j.conbuildmat.2017.06.087
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731
LWAC could be more sensitive to ambient humidity than ordinary
concretes with NWA. Because of the higher thermal conductivity of
liquid water compared to that of air, heat transfer by conduction
increases with the moisture content. The works presented in
[14–16] showed that the thermal conductivity of concretes
increases with the moisture content. According to [14], the con-
crete thermal conductivity varies linearly with concrete moisture
content. The influence of moisture content on the conductivity
and diffusivity of the wood-concrete was studied by Taoukil
et al. [16]. The experimental evolution of wood-concrete thermal
properties with moisture content was described by non-linear the-
oretical models.
The study on the thermal behavior of lightweight concretes at
ambient temperature range is still very limited. Some studies [2–
5,17–20] gave the thermal properties of LWAC at one specific con-
dition of temperature and moisture content. Other thermal studies
rather concern the high temperature behavior of concretes [21,22].
Study of Marechal [21] showed that the thermal conductivity of
the concrete made of siliceous aggregates (quartz) increases with
the ambient temperature, reaches a peak at about 50–60 °C and
then decreases. For LWAC, Kim et al. [23] noticed an increase of
the conductivity with temperatures ranging from 0 °C to 50 °C.
Same observations were showed in [24] for normal-weight con-
cretes in the temperature range 20 °C–60 °C. The concrete thermal
conductivity decreases with the temperature over 60–80 °C [21].
The effect of temperature on the conductivity can be explained
by the concrete moisture content [21,25]. The thermal conductivity
of the water increases with temperature, leading to increase in
thermal conductivity of the concrete for temperatures lower than
50–60 °C. For higher temperatures, free water and adsorbed water
escape from the concrete leading to the decrease of concrete ther-
mal conductivity. However, this explanation on the concrete con-
ductivity – temperature relationship makes less sense for
concretes in dry state. The specific heat of concrete also depends
on temperature. Mindess [26] indicated that the temperature has
significant impacts on the specific heat of cement paste and con-
crete. The specific heat of normal-weight concretes can vary from
800 to 1200 J/kg °C following the temperature variation.
In this context, the aim of this study is to analyze the influence
of moisture content and temperature on the thermal properties of
LWAC. The coarse and fine LWA used in this study are expanded
clay, expanded shale and pumice. LWAC have a density lower than
1600 kg/m3 and a compressive strength not less than 20 MPa. In
order to study the influence of moisture content in LWAC on its
thermal conductivity, specific heat and thermal diffusivity, three
different sample conditionings are taken into account. The thermal
properties of LWAC are also analyzed in the temperature range
between 5 °C and 50 °C. In the first part of this paper, the physical
properties, the chemical and mineralogical composition of LWA are
presented. The measurements of density, water absorption and
porosity provide information, not only for the mix design but also
for the knowledge of LWA microstructure. Secondly, the concrete
mixes and testing procedures are explained. An experimental pro-
tocol of LWAC sorption is undertaken in order to study the thermal
properties of LWAC in a partial saturation state. The thermal tests
were carried out using a Hot Disk device. Thermal conductivity and
diffusivity of concretes were measured and their specific heats
were deduced. In the last section, main results on thermal proper-
ties of LWAC depending on the moisture contents and on the tem-
peratures are discussed.
2. Material’s properties
In this study, seven types of LWA of expanded shale, pumice and expanded clay
were used. Aggregate densities and water absorptions were determined. There is no
adequate procedure to measure the thermal properties of LWA because of its small
721
size and its polyhedron form. The knowledge of LWA density and porosity will help
us estimating the influence of LWA on the thermal performances of LWAC. LWA
have usually great difference of thermal properties comparing to NWA and cement
paste. Thus LWA, which occupies from 65 to 75% of concrete volume fraction, has a
great influence on the thermal properties of LWAC and insulating performance of
LWAC structures. Besides, chemical and mineralogical compositions of LWA are also
important data to analyze the influence of LWA nature on thermal properties of
LWAC. Other factors that influence the thermal properties of concrete are the total
porosity and pores distribution of LWA, presented in [1].
2.1. Aggregate density and water absorption
In the following, the studied LWA are named by their sizes and their natures: S
for expanded shale, P for pumice and C for expanded clay. Three fine LWA (0/5 or
0/4 mm) and four coarse LWA (4/10 or 4/8 mm) were tested.
The LWA density and water absorption capacity must be taken into account
when determining the volume of LWA to be added to the mixture. The bulk density
(qv ) was measured according to the standard EN 1097–3. The oven-dry particle
density (qrd ) and the water absorption coefficients in% mass at 24 h (W24
a ) and
48 h (W48a ) of LWA are presented in Table 1. These properties were measured
according to the EN 1097-6. The properties of fine NWA 0/2 mm are added. Three
different samples were tested for each aggregate type and the average value is
reported in Table 1.
For LWA, fine aggregates are commonly heavier than coarse aggregates, with
bulk densities from 800 to 1030 kg/m3 and from 520 to 740 kg/m3, respectively.
0/5 P pumice is the lightest fine LWA. 5/8 P pumice and 4/8 C expanded clay are
heavier than other coarse LWA according to their low grain diameter. The two
aggregate pairs 4/10 S – 4/10 C and 4/8 C – 5/8 P, each of which have a similar size
and density, allow investigating the influence of other physical parameters such as
nature, porosity and pore size distribution of LWA on the properties of LWAC.
The expanded shale aggregates have the lowest water absorption coefficients at
24 h and 48 h. Expanded shale and expanded clay aggregates exhibit a porous inter-
nal structure surrounded by a shell, which is a relatively dense vitrified surface. The
shell of the shale LWA 4/10 S is thicker than that of the clay LWA 4/10 C [6]. This
surface property explains the lower water absorption coefficient of 4/10 S com-
pared to that of 4/10 C. Although fine LWA 0/4 S shale is crushed from bigger aggre-
gates so that its low-permeation outer shell disappears, it has the lowest water
absorption coefficient due to its highest density. Pumice LWA is crushed aggregate.
There is no difference between internal and external structure. Water absorption
coefficients are similar for fine 0/5 P and coarse 5/8 P pumice LWA. With compara-
ble densities, pumice fine LWA and clay fine LWA have similar water absorption
coefficient, which is about 18% at 24 h (Table 1). The water absorption coefficient
of the three expanded clay aggregates increases logically with the decrease of den-
sity. The W48
a coefficient is 19.2, 20.4 and 26.4% respectively for aggregate dry den-
sity of 1410, 1240 and 980 kg/m3, respectively. The measured water absorption
coefficient at 48 h was used to calculate the proportions of the concrete mixtures
(cf. Section 3.1).
2.2. Chemical and mineralogical composition of lightweight aggregates
The chemical composition of the used LWA, provided by producers, is presented
in Table 2. All aggregates contain a high percentage of silica, from 60 to 72%.
Expanded shale and clay have also high percentage of alumina and ferrite oxide.
Pumice has less alumina than the two other aggregates and its ferrite content is
very much lower and even negligible. Due to the higher atomic mass of the iron,
the density of the solid part of the clay and shale LWA should be higher than that
of the pumice LWA.
The mineralogy of the pumice, shale and clay aggregates was determined by
XDR analysis. Samples were scanned with a Philips diffractometer using copper
Ka radiation. The wavelength of the incident X-ray is equal to 1.54 Â. The diffrac-
tometer was running at 40 kV and 20 mA. The step widths was 0.025° from 6 to
66° 2-theta and 0.0025° from 64 to 70° 2-theta. The counting time was 1.5 s per
step. XRD data are can be viewed in [1]. The results showed the vitreous nature
of pumice. Expanded shale and expanded clay aggregates have a crystalline struc-
ture. They contain quartz, iron oxide and feldspar (orthoclase and plagioclase) as
principal mineral phases. The shale and clay aggregates have also iron oxide. The
clay one shows small amounts of amorphous aluminosilicate produced during the
thermal expansion process. This observation allows investigating the influence of
the mineralogy on the thermal properties of LWAC in the last section.
3. Experimental procedure
In this section, the mix design of twelve LWAC is described.
Only experiment procedure for thermal tests on LWA is presented.
Measurement procedure for physical and mechanical properties
can be viewed in [1].
722 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731

Table 1
Physical properties of lightweight aggregates.

Aggregates Size (mm) Aggregates density (kg/m3) W24

a W48
a

qv qrd (%) (%)

0/2 N 0/2 2460 4.54

0/4 S 0/4 1030 1790 8.05 8,6
4/10 S 4/10 520 900 10.01 10,8
0/5 P 0/5 800 1340 18.3 21,0
5/8 P 5/8 680 1180 17.32 20,8
0/4 C 0/4 830 1410 17.88 19,2
4/8 C 4/8 740 1250 19.17 20,4
4/10 C 4/10 590 980 24.76 26,4

Table 2
Chemical composition (%) of lightweight aggregates.

Aggregates SiO2 Al2O3 Fe2O3 CaO MgO K2O Na2O SO3 S Mn2O3 TiO2 P2O5
Expanded shale 63 21 8.5 1.5 – 3.6 1.5 0.008 0.02 – – –
Pumice 71.91 12.66 1.13 1.46 0.32 4.3 3.45 0.03 – – – –
Expanded clay 59.5 17 14.3 2 1.5 3 0.5 – 1 0.2 0.9 0.1

3.1. Mix design
Four series of LWAC were prepared with aggregates of four dif-
ferent natures: coarse and fine LWA of clay, pumice, and shale, as
well as fine NWA. In the following, the two LWAC made of
expanded clay are named OC1 and OC2 respectively, the one with
expanded shale OS and the one with pumice OP. Composition of all
mixtures are presented in Table 3. Portland cement CEM I 52.5 N
was used. The concretes were composed of an ordinary cementi-
tious matrix of which the Portland cement content and the water
to cement ratio (W/C = 0.45) remained the same for all the mixes.
The aggregate was a combination of coarse LWA, fine LWA and
normal fine LWA. The fine NWA was a silico-calcareous river sand
of size 0/2 mm and density 2460 kg/m3. For each LWAC series, the
volumes of fine and coarse aggregates were constant. The only
variable was the substitution ratio of the fine NWA by the fine
LWA: 0%, 50% and 100% by volume which corresponds to the last
number in the mixture names. During this substitution procedure,
fine and coarse LWA were always of the same nature. OC1 and OC2
concretes were made of the same clay fine LWA, but with different
clay coarse LWA 4/10 C and 4/8 C, respectively. For all mixes, the
target range for slump was 160 to 200 mm. The superplasticizer
amount was adjusted to maintain constant workability. Hence 12
mixes were prepared and tested.
LWA were ‘‘saturated” in order to avoid any change of water to
cement ratio due to water absorption of aggregates during mixing
[1]. The more the LWA is ‘‘saturated”, the less it absorbs water dur-
ing mixing. It is thus preferable to immerse LWA for 48 h before
mixing in order to avoid any variation of the W/C ratio of LWAC.
Moreover, the use of saturated aggregates could provide internal
curing to the concrete, which could reduce shrinkage and related
cracking and also increase compressive strength due to a more
effective cement hydration [27–29]. Coarse LWA were immersed
for 48 h and then drained about 30 min until their surface moisture
content become constant. Mixing water was then reduced by the
quantity of water brought by the LWA surface moisture. We sup-
pose that the water absorbed by LWA during immersion was not
available for the cement hydration during the mixing.
Cylindrical specimens Ø160  320 mm were cast for thermal
tests. Concrete moulds were removed at 24 h. Specimens were
stored in water during 90 days to ensure a complete hydration.
The thermal experiments were carried out on ‘‘prismatic” samples
140  70  40 mm cut from Ø160  320 mm specimens. The sam-
ples were polished in order to achieve a good flatness on the
140  70 mm surfaces which are in contact with the thermal
probe. Twelve ‘‘prismatic” samples were prepared for each mix-
ture. The sample conditioning is described in the next subsection.
The cross sections of the concrete without fine LWA are shown
in the Fig. 1. The OS and OP concretes have angular aggregate
shape while the OC1 and OC2 concretes have rounded aggregate
shape. The OC2 concrete which was made with 4/8 clay LWA has
smaller particles on the cross section than the OC1 concrete which
made with 4/10 clay LWA.

Table 3
The proportions of the concrete mixes (based on SSD conditions).

Mixes Cement (kg/m3) Coarse aggregate (kg/m3) Fine aggregate (kg/m3) Water (kg/m3) SP (g/l)
4/10 S 5/8P 4/10C 4/8C 4/10 N 0/4 S 0/5P 0/4C 0/2 N
OS-0 426 445 0 0 0 0 0 0 0 554 190 4.3
OS-50 426 445 0 0 0 0 213 0 0 277 190 5.0
OS-100 426 445 0 0 0 0 427 0 0 0 190 6.2
OP-0 426 0 626 0 0 0 0 0 0 554 190 3.7
OP-50 426 0 626 0 0 0 0 177 0 277 190 4.3
OP-100 426 0 626 0 0 0 0 354 0 0 190 6.0
OC1-0 426 0 0 558 0 0 0 0 0 554 190 3.4
OC1-50 426 0 0 558 0 0 0 0 189 277 190 2.2
OC1-100 426 0 0 558 0 0 0 0 379 0 190 0.7
OC2-0 426 0 0 0 673 0 0 0 0 554 190 3.2
OC2-50 426 0 0 0 673 0 0 0 189 277 190 2.8
OC2-100 426 0 0 0 673 0 0 0 379 0 190 1.3
 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731 723

Fig. 1. Cross section of concretes (A: OS-0, B: OP-0, C: OC1-0, D: OC2-0).

Table 4
Conditioning of samples for thermal tests.

Objective Sample conditioning Protocol

T (°C) Moisture state
Analysis of the influence of moisture content 20 °C Dry In oven at 80 °C until constant mass
20 °C Equilibrium at 50% RH In oven at 80 °C until constant mass then in a climatic
chamber at 20 °C and 50% RH until constant moisture content
20 °C Saturated In water at 20 °C for 90 days
Analysis of the influence of the temperature 5 °C Dry In oven at 80 °C until constant mass
20 °C Dry
35 °C Dry
50 °C Dry

3.2. Sample preparation
With the aim to analyze the influence of moisture content and
temperature of concrete on their thermal properties, three sample
conditionings were employed. This achieved three different mois-
ture contents in concrete: dry state, saturated state and equilib-
rium at 50% RH (Table 4). Four ‘‘prismatic” samples were used
for each moisture content state.
For the dry state, the samples were firstly cured in water for
three months. After that, they were dried in an oven at 80 °C until
constant weight (variation less than 0.2% after 24 h). Before the
thermal tests, the samples were enclosed in a sealed bag which
was kept in a climatic chamber at 20 °C for at least 24 h in order
to obtain a homogeneous temperature.
For the intermediate state at 50% RH and 20 °C, the samples
were also dried in an oven at 80 °C to constant weight after three
months immersion. Then they were stored in a climatic chamber
at 20 °C and 50% relative humidity until constant moisture content.
The constant moisture content state is verified when the mass of
the samples does not vary more than 0.02% after 24 h [30] during
5 consecutive days. It took from two to three months for ‘‘pris-
matic” sample to reach the moisture content equilibrium. The
higher water-accessible porosity of concrete, the longer time
needed to reach equilibrium will be. The results of moisture
adsorption at 50% RH (W RH50 ) and 20 °C of LWAC are compared
to their water absorption coefficients in Table 5.
The samples used for saturated state testing were cured in
water at 20 °C until the tests. They were dried on surface using a
724 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731

Table 5
Water absorption coefficients, water-accessible porosity at saturated state and equilibrium moisture contents of LWAC at 50% RH.

Concrete Substitution ratio (%) Wa / W RH50

Mean (%) SD(%) RSD (%) Mean (%) Mean (%) SD(%) RSD (%)
OS 0 14.6 0.6 3.92 21.0 2.1 0.3 2.21
50 16.1 1.0 6.30 22.3 2.2 0.4 2.93
100 17.8 0.8 4.71 23.4 2.3 0.6 4.66
OP 0 16.4 1.0 6.26 24.4 2.5 0.2 1.23
50 18.2 0.8 4.43 25.5 2.9 0.2 1.34
100 19.4 0.8 4.00 26.0 3.4 0.7 3.74
OC-1 0 18.9 1.4 7.43 26.7 3.0 0.3 1.84
50 23.2 1.1 4.55 29.7 3.5 0.2 1.32
100 27.1 0.9 3.32 31.8 4.2 0.6 3.53
OC-2 0 17.1 0.3 1.49 25.9 2.6 0.3 1.64
50 19.9 0.3 1.73 27.9 2.8 0.2 1.52
100 22.7 0.6 2.83 29.1 3.4 0.6 4.08

wet rag before the measure. Water absorption coefficients and
water accessible porosity of concrete were deduced from concretes
densities measured on saturated samples according to the NF EN
12390-7. Water accessible porosity, which is the volume ratio of
absorbed water to concrete sample, is given by:
/ ¼ V water =V concrete ¼ ðqd  qs Þ=qeau
where qd and qs are oven-dried density and saturated density of
concrete. The water absorption coefficient is given by:
W a ¼ mwater =mconcrete ¼ /  qeau =qd
Furthermore, the thermal properties were measured on dried
samples at different temperatures. The tests were conducted at
5 °C, 20 °C, 35 °C and 50 °C.
3.3. Thermal tests
Thermal experiments were conducted on concrete samples
using a Hot Disk probe TPS1500 (Fig. 2). The probe was used both
as heat source and as temperature sensor. The sensor was sand-
wiched between two identical 140  70  40 mm samples of con-
crete. Each sample must have a flat surface in order to avoid
contact defect with the sensor. The sensor consists of a very fine
nickel double spiral (10 lm of thickness) covered with two thin
layers of electrically-insulating material. A kapton probe of
9.87 mm radius was used for measurement in climatic chamber
and a mica probe of 9.72 mm radius for measurement in the fur-
nace. The system is based on Transient Plane Source techniques
(TPS). The concrete sample is initially kept at thermal equilibrium.
Then a heating pulse is applied by the Hot Disk sensor to create a
thermal disturbance on the sample. The thermal resistance of the
probe is monitored during the time of measurement. This allows
to determine the temperature increase DT(t) of the sensor. For
the Hot Disk sensor, DT(t) is given by the following equation, from
which the thermal conductivity k and diffusivity a can be obtained
[31]:
P0
ð1Þ DTðsÞ ¼ DðsÞ ð3Þ
p3=2 rk
where P0 is the total output power, k is the thermal conductivity of
the sample, and r is the radius of the sensor. DðsÞ is the theoretical
ð2Þ expression of the time dependent temperature increase, which
describes the conducting pattern of the Hot Disk sensor, This
parameter DðsÞ is function of the thermal diffusivity of the sample
[31]. Analysis of the data was performed using the Hot Disk Thermal
Constants Analyser software.
The thermal conductivity k and specific heat cp and thermal dif-
fusivity a are related by the following expression: a ¼ qkcp , where q
represents the density of the material.
Thermal tests at 5 °C and 20 °C were carried out in the climatic
chamber at the corresponding temperature. Once the temperature
of the sample was stabilized, this one was installed on the Hot Disk
sample holder inside the climatic chamber. After each measure, the
probe was moved to another position on sample in order to avoid
the influence of heterogeneity in the concrete. For each concrete
mix, five tests were performed. The standard deviation of each con-
crete mix is less than 6% for thermal conductivity, thermal diffusiv-
ity and specific heat.
Temperatures at 35 °C and 50 °C were obtained using the Hot
Disk furnace (Fig. 3). For these measures, the samples at 20 °C were
put in the sample holder in the furnace. The Hot Disk device used
The Thermal Constants Analyzer program to control and stabilize
the furnace temperature before measurement. For each sample,

Fig. 2. Hot Disk device. Fig. 3. Measurements at 35 °C and 50 °C.

 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731 725

thermal tests at 35 °C and 50 °C were carried out consecutively.
There were therefore two temperature increases of 15 °C and two
stabilization stages at 35 °C and at 50 °C. The heating rate was
about 0.25 °C/minute which was executed automatically by Hot
Disk program. The stability criterion was set at a maximum varia-
tion of 0.5 °C for 30 min. Two pairs of samples were prepared for
each concrete mix. A total of six measurements were performed
at a target temperature for each mix.
The oven dry density and thermal properties of LWAC are
shown in Table 6 for measurements at different moisture contents
and in Table 7 for measurements at different temperatures.
coefficient. The water content is determined by the ratio of the
weight gain to the oven-dry sample weight.
Fig. 4 shows the evolution of the equilibrium moisture content
of LWAC according to the fine aggregate substitution rate. The
measurements are performed on four samples per formulation.
The average relative standard deviation is 3% which shows a low
dispersion of the results. At equilibrium state, LWAC with 100%

4. Discussion

The thermal properties of LWAC are discussed in this section.

Firstly, we present the moisture adsorption of LWAC at equilibrium
at 50% RH and 20 °C. Secondly, the thermal behavior of LWAC at
different moisture contents and the influence of mix design param-
eters are investigated. Finally, the influence of the temperature
increasing from 5 °C to 50 °C is discussed.

4.1. Moisture adsorption of lightweight aggregate concrete at 50% RH

and 20 °C

Under this curing condition (Table 4), the concretes are in a

sorption state. Concrete moisture content increases due to the pen-
etration of moist air and water vapor condensation. In order to
analyze the impact of moisture content on the thermal properties
of LWAC, water content at 50% RH was calculated comparing to the Fig. 4. Equilibrium moisture concrete LWAC at 50% RH and 20 °C according to the
dry state, in the same way that is calculated the water absorption substitution ratio of fine LWA.

Table 6
Thermal properties of LWAC at different moisture states.

Mixes qd (kg/m3) k (W/mK) cp (J/kgK) qps (kg/m3) k (W/mK) cp (J/kgK) qs (kg/m3) k (W/mK) cp (J/kgK)

Dry conditioning 50%RH conditioning Immersion conditioning

OS-0 1440 0.71 1021 1470 0.74 1053 1650 0.97 1243
OS-50 1380 0.64 1055 1410 0.67 1088 1610 0.90 1281
OS-100 1320 0.57 1094 1350 0.60 1113 1550 0.81 1355
OP-0 1490 0.70 996 1530 0.74 1048 1730 1.05 1341
OP-50 1410 0.56 1039 1450 0.60 1061 1660 0.91 1337
OP-100 1340 0.46 1077 1390 0.50 1131 1600 0.78 1338
OC1-0 1410 0.67 1049 1450 0.73 1111 1680 1.00 1368
OC1-50 1290 0.53 1075 1330 0.59 1128 1580 0.88 1441
OC1-100 1170 0.43 1077 1220 0.51 1193 1490 0.83 1539
OC2-0 1520 0.73 1024 1560 0.79 1066 1780 1.08 1259
OC2-50 1400 0.61 1048 1440 0.66 1105 1680 0.95 1349
OC2-100 1280 0.49 1060 1330 0.55 1136 1570 0.84 1437

Table 7
Thermal properties LWAC at different temperatures.

Mixes qd (kg/m3) k (W/mK) cp (J/kgK) k (W/mK) cp (J/kgK) k (W/mK) cp (J/kgK) k (W/mK) cp (J/kgK)

5 °C 20 °C 35 °C 50 °C
OS-0 1440 0,62 1046 0.71 1021 0,76 1067 0,77 1077
OS-50 1380 0,55 1077 0.64 1055 0,70 1108 0,71 1127
OS-100 1320 0,47 1136 0.57 1094 0,60 1128 0,61 1199
OP-0 1490 0,66 1096 0.70 996 0,76 1034 0,77 1081
OP-50 1410 0,52 1127 0.56 1039 0,62 1093 0,63 1141
OP-100 1340 0,41 1186 0.46 1077 0,51 1137 0,51 1173
OC1-0 1410 0,61 1125 0.67 1049 0,74 1110 0,73 1158
OC1-50 1290 0,49 1174 0.53 1075 0,58 1132 0,60 1224
OC1-100 1170 0,39 1237 0.43 1077 0,49 1161 0,50 1294
OC2-0 1520 0,68 1056 0.73 1024 0,80 1065 0,80 1101
OC2-50 1400 0,56 1124 0.61 1048 0,67 1099 0,68 1147
OC2-100 1280 0,45 1170 0.49 1060 0,55 1099 0,56 1202
726 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731
fine NWA have moisture content between 2% and 3% while LWAC
with 100% fine LWA have moisture content between 2% to 4%. This
value is about 1.5% for concrete with NWA according to [30].
The OC2-0 and OP-0 concretes, which have similar densities
(Table 6), have almost identical moisture adsorption (2.5%). In con-
trast, the moisture adsorption of OS-0 concrete (2%) is lower than
that of OC1-0 concrete (3%) despite their similar densities. The
same cementitious matrix (volume, W/C) was used in all concretes,
so the difference in concrete porous network is related to that of
aggregates. As saturated LWA aggregates are used the effect of
aggregate absorption on the hydration water content at the aggre-
gate/cement interfacial zone is minimised so the porosity of the
cement matrix would remain the same for all types of concretes.
The low moisture of shale LWAC is explained by the low water-
absorption coefficient (low percentage of open porosity compared
to total porosity) of shale aggregates 4/10 S despite its low density
[1]. The moisture adsorption of OC1 clay concrete is the highest,
which correspond to the high water-accessible porosity of 4/10 C
aggregate. So the adsorbed water content seems clearly linked to
the open porosity of aggregate. The substitution of fine NWA by
fine LWA leads to change of equilibrium moisture content accord-
ing to the nature of the fine LWA. The complete substitution of the
fine NWA by fine shale LWA (OS), which is the heaviest fine LWA,
provides additional moisture content of 0.3%. Following the com-
plete substitution, the increasing of moisture content is between
0.8 and 1.2% for clay concretes (OC1 and OC2) and pumice concrete
(OC) (Fig. 4).
The water contents in saturated state and at 50% RH are com-
pared in Fig. 5. The three points of each curve correspond from left
to right to the concretes with 0%, 50% and 100% of fine LWA. The
comparison shows that the open porosity is not the only factor
explaining the different moisture contents of LWAC at 50% RH.
The water absorption by immersion is linked to the water transfer
of liquid phase into the pores (capillary action, gravity, external
pressure) when the sample is immersed. Furthermore, the mois-
ture content at a specific humidity is related to the diffusion of
water vapor and its adsorption in the pores. Water vapor con-
denses to liquid water when the vapor pressure in the pore is high
enough. Indeed, the equilibrium condensation – evaporation at a
given temperature of water is governed by the pore size according
to Kelvin’s law and Laplace’s law [32,33]. The adsorbed water also
depends on the surface forces (Van der Waals forces, hydrogen
Fig. 5. Equilibrium moisture content at 50% RH versus water absorption by
immersion at 20 °C for LWA concretes.
bonds. . .). Fig. 5 shows a higher slope of the curve of pumice con-
crete (OP) than that of other concretes. This means the pumice fine
aggregate (0/5 P) has a greater hygroscopic character than clay
(0/4 C) fine aggregate despite similar water absorption coefficient.
The comparison of pore size distribution of both aggregates doesn’t
show finer pores for the pumice one [1]. But, the pumice aggregate
consists of amorphous silica and silicates. These chemical com-
pounds can bind hydroxyl groups –OH (silanols) which are
strongly hydrophilic. This higher hydrophilicity could lead to a
higher propensity of the vapor-water to be fixed on pore surfaces.
4.2. Thermal properties of lightweight aggregate concretes at different
moisture states
4.2.1. Influence of the moisture content on the thermal properties
This section presents the evolution of the thermal conductivity,
the specific heat and the thermal diffusivity of LWAC according to
their moisture content. Results are presented for the minimum and
maximum volume fractions of fine LWA, i.e. LWAC without fine
LWA and with 100% fine LWA. The thermal properties of LWAC
were investigated for three different conditionings: oven-dried
(dry state), at equilibrium at 50% RH (partial saturation state)
and immersion (saturated state).
Thermal conductivity of LWAC increases almost linearly with
the water content (Fig. 6). At the saturated state, LWAC with
100% fine LWA have equivalent thermal conductivity to oven-dry
LWAC without fine LWA. The total substitution of fine aggregate
reduces the thermal conductivity of LWAC by 0.22 W/mK on aver-
age while the saturation by immersion increases the thermal con-
ductivity of LWAC with 100% fine LWA for 0.33 W/mK on average.
From the dry state to the saturated state, there is an increase of 37%
–93% of the thermal conductivity depending on the nature and the
volume fraction of LWA. For 50% RH conditioning, the increase in
thermal conductivity of LWAC ranges between 4% and 19%. In fact,
when the moisture content increases in concrete, the ‘‘dry” air in
the pores is replaced by water and moist air. Thermal conductivity
of water (0.601 W/mK at 20 °C) is much greater than that of air
(0.026 W/mK at 20 °C) [34,35]. Thus the thermal conductivity of
LWAC logically increases with the saturation degree.
Similar results were confirmed by other authors on concrete
with normal weight aggregates. The linear relation between mois-
ture content and thermal conductivity of lightweight concretes
Fig. 6. Thermal conductivity of LWAC according to the increase of moisture content
(% mass).
 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731 727

was presented in [14]. Kim et al. [23] have found an increase of

0.33–0.43 kcal/hmC (about 0.43–0.5 W/mK) between the dry state
and the saturated state at 20 °C of normal weight concretes (mixes
C-GC1 and C-GC2 with W/C of 0.4). Normal weight concretes have
lower porosity than LWAC. However, the comparison shows that
the normal weight concretes have higher conductivity increase
(0.43–0.5 W/mK) due to the saturation than that of LWAC (0.27–
0.39 W/mK). This means that the water proportion is not the only
factor that influences thermal conductivity increase. As an example
the OC1-100 concrete has a water absorption coefficient of 27.1%,
equivalent to a water accessible porosity of 31.8%”, yet the thermal
conductivity increases by 0.39 W/mK which is much higher than
the value given by a general rule of mixtures (about 0.17 W/mK).
The used LWA have different properties and different pore struc-
ture which were investigated in [1,6]. The contribution of low
resistance pathway (created by water) doesn’t only depend on
the volume fraction of the phases but also on their pore structure.
Hence, the thermal conduction could be affected by the percolation
threshold of the thermal conduction in LWAC.
The evolutions of specific heat of LWAC according to the mois-
ture content are shown in Fig. 7. The specific heat increases with Fig. 8. Thermal diffusivity of LWAC according to the increase of moisture content.
the moisture content. A quasi-linear relationship between these
two factors is also observed. Between the dry state and the satu-
mal diffusivity of LWAC is about 4% on average. As a result, the
rated state, the specific heat increases by 21%–43% depending on
transient thermal performance of LWAC elements and therefore
the nature and the volume fraction of the aggregates. The increase
thermal inertia, which is dominated by concrete thermal diffusiv-
of specific heat of LWAC at the partial saturation state (50% RH
ity, is not much affected by the increase of concrete moisture con-
conditioning) is only 5% on average. These evolutions of the speci-
tent. However, the steady thermal performance of LWAC, which
fic heat are also linked to the replacement of ‘‘dry” air, which have
depends only on concrete thermal conductivity, decreases with
a low specific heat (1005 J/kgK), by water phases, which have
the moisture content increasing.
higher specific heats (4181.3 J/kgK for liquid water and 2080 J/
kgK for water vapor).
The thermal diffusivity is the ratio of thermal conductivity to 4.2.2. Influence of the nature of coarse aggregates
the specific heat and the density. It varies from 0.32 to Figs. 9 and 10 show the relative increase of thermal conductiv-
0.48 mm2/s for the studied LWAC in the dry state. The variations ity of LWAC at saturated state according physical properties of
of thermal conductivity, specific heat and density of LWAC as a coarse LWA. The error bar shows the mean value ± standard devi-
function of moisture content have the same linear growing ten- ation. For more information about the influence of the coarse LWA
dency. The increases of these properties are compensated when nature, the results on concrete mixes with 100% of fine NWA were
calculating thermal diffusivity. This leads to a low increase or considered. The OS shale concrete has the lowest increase of ther-
decrease of the thermal diffusivity of LWAC following the increase mal conductivity despite the low particle density of 4/10 S aggre-
of moisture content (Fig. 8). The minor variations of the thermal gate (Fig. 9). This aggregate showed the lowest water accessible
diffusivity do not exceed 10% (except for concrete OP-100) porosity of all coarse LWA (Fig. 10). This is related to the pore
between the dry state and the saturated state. At the maximum structure and density of shale aggregate. Experimental investiga-
moisture content (obtained by immersion), the variation of ther- tion in [1,6] shows that 4/10 S aggregate has a high proportion of

Fig. 9. Relative increase of thermal conductivity of LWAC at saturated state

Fig. 7. Specific heat of LWAC according to the increase of moisture content. according to the coarse LWA particle density.
728 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731

Fig. 12. Relative increase of thermal conductivity of LWAC at equilibrium at 50%

Fig. 10. Relative increase of thermal conductivity of LWAC at saturated state RH, 20 °C.
according to the coarse LWA water-accessible porosity.

closed pores, thanks to its thicker dense-outer-shell than that of

other coarse LWA. Therefore, this feature diminishes the conduc-
tivity rise at saturated state for the OS shale concrete. The increases
of conductivity are thus more related to water accessible porosity
(Fig. 10) than to grain density (Fig. 9). The OC1 concrete made of
4/10 C clay aggregate has the highest conductivity increase with
the moisture content. This is due to the highest water-accessible
porosity of the 4/10 C aggregate (Fig. 10). The OP pumice concrete
has a similar increase of conductivity as that of OC 2 clay concrete.
This can be explained by their similar water-accessible porosities.

4.2.3. Influence of the nature and the proportion of fine lightweight

aggregates
The influence of fine aggregate substitution on the thermal con-
ductivity of LWAC at different moisture states can be observed in
Figs. 11–13. The concrete density decreases with the fine LWA pro-
portion, leading to a decrease of thermal conductivity of concrete
Fig. 13. Relative increase of thermal conductivity of LWAC at saturated state at
for all three sample conditionings. For dry LWAC, the reduction
20 °C.
of conductivity ranges between 20 and 35%. In general, the replace-

ment of fine NWA by fine LWA increases the influence of sample

Fig. 11. Thermal conductivity of dry and saturated LWAC according to its density.
conditioning on the thermal conductivity of LWAC (Figs 12 and
13). Concretes containing more fine LWA are more porous than
concrete with fine NWA. They have higher moisture content than
LWAC with fine NWA in the same environmental condition. That
leads logically to a higher increase of thermal conductivity. At sat-
urated state, the thermal conductivity increases by 45% for LWAC
without fine LWA and by 70% for LWAC with 100% of fine LWA
(Fig. 13). At equilibrium state at 50% RH, the increases of thermal
conductivity are 7% and 11% on LWAC before and after the fine
aggregate substitution, respectively (Fig. 12).
The influence of fine LWA nature is examined through his-
tograms in Figs 12 and 13. The error bar shows the mean
value ± standard deviation. It will be recalled that fine and coarse
LWA of the same nature are used in each mix. The 0/4 S aggregate
has the lowest water absorption coefficient at 8% while these coef-
ficients are about 18% for both clay and pumice fine aggregates
(0/4 C and 0/5 P). As expected, the substitution of fine NWA by
the 0/4 S shale aggregate induces a low increase of thermal con-
ductivity of OS concrete, either at partial saturation state or at sat-
urated state. With 100% of shale fine LWA, the k=kdry ratio is about
 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731
1.05 at partial saturation state and 1.42 at saturated state (Figs. 12
and 13). These two values are respectively 1.1–1.17 and 1.7–1.9 for
others fine LWA. This could be attributed to the low water accessi-
ble porosity of fine shale aggregate [1]. Furthermore, OP and OC2
concretes have similar increase of thermal conductivity in both
partially saturated and saturated states. These two concretes use
fine lightweight aggregates of similar density, similar water
absorption coefficient, i.e. similar water accessible porosity
(Table 3). This confirms the role of aggregate density and water
accessible porosity on the thermal conductivity of concrete.
However, the substitution of fine NWA by the same fine LWA
(0/2 N by 0/4 C in the clay concretes) leads to a faster conductivity
increase in the OC1 for both saturated and 50% RH conditionings
(Figs. 12 and 13). The OC1 concrete has indeed a higher increase
of water-accessible porosity (8.2%) than that of OC2 concrete
(5.6%) following the total substitution of fine aggregate (Table 5).
Adding porous fine LWA could more improve the pore interconnec-
tion in OC1 concrete than in OC2 concrete. The access to aggregate
open pores could be easier for OC1 than for OC2. The thermal con-
ductivity depends not only on the volume of solids/water/air but
also on the pore structure of these phases. This may lead to higher
thermal conductivity of moisturized LWAC if there is a good pore
‘‘connection”. Indeed, in presence of water, the contribution of
water conductivity and moist air conductivity to the concrete con-
ductivity is higher than that of air to the dry concrete conductivity.
Good pore connection favors the heat transfer in water phase by
conduction and convection. The higher connected porosity of con-
crete, the higher influence of ‘‘moist” conditioning on the concrete
conductivity is.
4.3. Influence of the temperature on the thermal properties of
lightweight aggregate concretes
Fig. 14 shows the changes in thermal conductivity of LWAC
according to the temperature increase. The thermal conductivity
increases from 5 °C to 35 °C than it levels off after 35 °C. In the tem-
perature range from 5 °C to 50 °C, the thermal conductivity of
LWAC without fine LWA increases by 0.12–0.15 W/mK (17% to
24%). The rise of conductivity ranges from 0.1 W/mK to 0.14 W/
mK (about 24–30%) for LWAC made of 100% fine LWA.
In LWA, the water-accessible porosity can reach up to 32% [36].
There may be also closed pores which contain air and/or trapped
Fig. 14. Thermal conductivity of dry LWAC according to the increase of the ambient
temperature.
729
water. According to [34], the thermal conductivity of the dry air
increases almost linearly from 0.0245 to 0.028 W/mK (about
14%) for temperature ranging from 5 °C to 50 °C. The air conductiv-
ity contribution is negligible in the rise of concrete thermal con-
ductivity. Another factor that may influence the increase in
thermal conductivity is the presence of trapped water in the pores.
The physically bound water is not all removed after drying in oven
at 80 °C. In the temperature range between 5 °C and 50 °C, the
thermal conductivity of water increases from 0.5715 W/mK to
0.6445 W/mK (about 13%) according to [35]. The higher conductiv-
ity increase of LWAC with 100% fine LWA highlights the part taken
by trapped water on the relationship between LWAC conductivity
and temperature. However, the low increase of water conductivity
with temperature doesn’t fully explain the rise of LWAC thermal
conductivity. Some other researches have mentioned the radiation
contribution to the thermal conductivity of materials [37–39]. The
modeling of radiation conductivity by Ochs [38] showed a slight
increase of thermal conductivity with temperature in expanded
clay and expanded glass aggregates. Ochs pointed out an increase
of 0.2 W/mK of all studied expanded clay aggregates following a
temperature rising from 20 °C to 80 °C. According to [40] the con-
tribution of radiation on thermal conductivity is not negligible for
materials with a high porosity and a pore size superior to 100 mm.
However, the increase of conductivity due to radiation and high-
lighted by Och [38] can’t explain all of the increase in conductivity
of concretes between 5 and 35 °C. The thermal expansion may also
reduce the pore sizes and contribute to the increase in thermal
conductivity.
The stabilization of thermal conductivity between 35 °C and
50 °C may be explained by the loss of conductive bonds in con-
crete. This could be linked to the departure of physically and chem-
ically bound water [21]. The occurrence of microcracks at the
interfacial transition zone could also contribute to the stabilization
of the conductivity at that temperature range. These microcracks
could be generated by the differential thermal expansion and the
occurrence of stresses at the paste-aggregate interface. This poros-
ity increase compensates for the increasing trend of thermal con-
ductivity rise with temperature.
The specific heat of LWAC decreases from 5 °C to 20 °C (Fig. 15).
Then there is an increase of this thermal property when the tem-
perature rises up to 50 °C. The increase rate is from 6% to 10% for
shale and pumice concretes and from 10% to 14% for clay concrete
Fig. 15. Specific heat of LWAC according to the increase of the ambient
temperature.
730 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731
Fig. 16. Thermal diffusivity of LWAC according to the increase of ambient
temperature.
(except the OC1-100). The variation of the specific heat as a func-
tion of temperature is significant for LWAC. The dependence of the
specific heat on temperature has been noticed in [25,26] for the
cement pastes and normal weight concretes. The rise of specific
heat with temperature increase is due to the specific heat evolu-
tions of the mineral phases in concrete. XDR analysis on the same
LWA has shown that expanded shale and expanded clay aggregate
contain a lot of mineral phases while pumice aggregate contain
almost amorphous silica [1,36]. These aggregate have from 60 to
72% of quartz in their composition (Table 2). According to several
authors [41–43], the specific heat of minerals is a thermodynamic
parameter varying as a function of temperature. Hemingway [41]
has presented in his work an increase of specific heat of quartz
by 20% while temperature rises from 298°K to 400°K. Bruckner
[42] has shown that between 300°K and 500°K, the specific heat
of vitreous silica increase about 30%. Otherwise, the specific heat
of other minerals such as hematite, magnetite and plagioclase
increases between 16% and 18% in the temperature range from
298°K to 400°K [43]. These are mineral phases that have been
observed in the expanded shale and expanded clay aggregates
[1,36].
The high values of LWAC specific heat at 5 °C could be explained
by the presence of water vapor in the pores. Indeed, trapped water
in concrete may exist in liquid-vapor equilibrium. During the sam-
ple conditioning at 5 °C, the temperature could fall below the dew
point, leading to an excess of water condensation. Therefore, the
amount of condensed water on the pore walls would be greater
at 5 °C than that between 20 °C and 50 °C. Since the specific heat
of liquid water (4185 J/kgK) is more than 2 times greater than that
of water vapor (2026 J/kgK) and that of dry air (1012 J/kgK) [44],
specific heat of concrete increases with the temperature decrease
from 20 °C down to 5 °C.
The thermal diffusivity evolution results from those of thermal
conductivity and the heat capacity (Fig. 16). LWAC diffusivity
increases by 11–38% according to temperature increase from 5 °C
to 35 °C. The diffusivity decreases slightly from 35 °C to 50 °C. This
is related to the conductivity stabilization and the specific heat
increase in this temperature range.
5. Conclusion
In this study, the thermal properties of LWAC at several
moisture contents and ambient temperatures were investigated.
Twelve mixtures of LWAC made of LWA from three different nat-
ures were tested. The densities of the studied LWAC are between
1170 kg/m3 and 1520 kg/m3. At the dry state, LWAC have a thermal
conductivity varying from 0.43 to 0.73 W/mK, a specific heat varying
from 1000 to 1100 J/kgK and a thermal diffusivity ranging from 0.32
to 0.48 mm2/s. The study shows a non-negligible dependence of
LWAC thermal properties on the moisture content and on the
temperature. Relationships between thermal properties and
environmental conditions are influenced by the volume fraction,
the density and the pore structure of the aggregates.
There is a quasi-linear increase of the thermal conductivity and
specific heat of LWAC with the moisture content in mass The
increase ranges from 1,8 102 W/mK to 2.1 102 W/mK per mass
% of water for concretes with lightweight coarse aggregates and
normal weight fine aggregates and is about 1,5 102 W/mK per
mass% of water for concretes with fines and coarse lightweight
aggregates. At the saturation state, the thermal conductivity
increase is 60% in average. At the equilibrium state of 50% RH,
the thermal conductivity increases by 9% in average with a mois-
ture content ranging from 2.1 to 4.2%. Increasing of the moisture
content leads to an increase of the LWAC specific heat by 30% in
average at saturated state and by 5% in average at the intermediate
state of 50% RH. The increase of the moisture content results in a
low variation (4% in average) of the thermal diffusivity for all stud-
ied mixes.
The dependence of LWAC thermal properties on moisture con-
tent is due to the replacement of the air phase by a water phase
in the concrete pores. The water-accessible porosity of LWA is
the key parameter which influences the increase of LWAC thermal
conductivity with the moisture content. Aggregates pore structure
within the concrete could generate different ‘‘interconnections” in
the microstructure, which could have an impact on the thermal
conductivity of LWAC.
The evolutions of the thermal properties as a function of the
temperature are not linear in the studied range 5 °C–50 °C. The
temperature has a slightly adverse effect on the thermal behavior
of LWAC. In the range 20 °C–50 °C, the thermal conductivity
increases with the temperature, and then stabilizes around a mean
value after 35 °C. The thermal conductivity increases by 17% to 30%
depending on the mineralogical composition of the aggregates. The
specific heat of LWAC decreases with the temperature in the range
5 °C–20 °C, and then increases from 5 to 20% in the range 20 °C–
50 °C depending on mineralogical aggregate composition. The
effect of temperature in building isolation purpose must take into
account the evolution of theses thermal properties.
References
[1] L.H. Nguyen, A.-L. Beaucour, S. Ortola, A. Noumowé, Influence of the volume
fraction and the nature of fine lightweight aggregates on the thermal and
mechanical properties of structural concrete, Constr. Build. Mater. 51 (2014)
121–132.
[2] T.S. Yun, Y.J. Jeong, T.S. Han, K.S. Youm, Evaluation of thermal conductivity for
thermally insulated concretes, Energy Build. 61 (2013) 125–132.
[3] U.J. Alengaram, B.A. Al Muhit, M.Z. Jumaat, M.Y.J. Liu, A comparison of the
thermal conductivity of oil palm shell foamed concrete with conventional
materials, Mater. Des. 51 (2013) 522–529.
[4] M.Y.J. Liu, U.J. Alengaram, M.Z. Jumaat, K.H. Mo, Evaluation of thermal
conductivity, mechanical and transport properties of lightweight aggregate
foamed geopolymer concrete, Energy Build. 72 (2014) 238–245.
[5] F. Koksal, O. Gencel, M. Kaya, Combined effect of silica fume and expanded
vermiculite on properties of lightweight mortars at ambient and elevated
temperatures, Constr. Build. Mater. 88 (2015) 175–187.
[6] Y. Ke, A.-L. Beaucour, S. Ortola, H. Dumontet, R. Cabrillac, Influence of volume
fraction and characteristics of lightweight aggregate concrete on the
mechanical properties of concrete, Constr. Build. Mater. 23 (2009) 2821–2828.
[7] J.A. Bogas, A. Gomes, Compressive behavior and failure modes of structural
lightweight aggregate concrete – characterization and strength prediction,
Mater. Des. 46 (2013) 832–841.
[8] N.U. Kockal, T. Ozturan, Strength and elastic properties of structural
lightweight concretes, Mater. Des. 32 (2011) 2396–2403.
 L.H. Nguyen et al. / Construction and Building Materials 151 (2017) 720–731
[9] H.Z. Cui, T.Y. Lo, S.A. Memonb, W. Xu, Effect of lightweight aggregates on the
mechanical properties and brittleness of lightweight aggregate concrete,
Constr. Build. Mater. 35 (2012) 149–158.
[10] E. Güneyisi, M. Gesoglu, T. Özturan, S. Ipek, Fracture behavior and mechanical
properties of concrete with artificial lightweight aggregate and steel fiber,
Constr. Build. Mater. 84 (2015) 156–168.
[11] Y. Ke, S. Ortola, A.-L. Beaucour, H. Dumontet, Micro-stress analysis and
identification of lightweight aggregate’s failure strength by micromechanical
modelling, Mech. Mater. 68 (2014) 176–192.
[12] T.Y. Lo, W.C. Tang, H.Z. Cui, The effects of aggregate properties on lightweight
concrete, Build. Environ. 42 (2007) 3025–3029.
[13] M.H. Zhang, O.E. Gjorv, Characteristics of lightweight aggregates for high
strength concrete, ACI Mater. J. (1991) 150–158.
[14] D. Gawin, J. Kosny, K. Wilkes, Thermal conductivity of moist cellular concrete –
experimental and numerical study, in: Proceedings of ASHRAE Thermal IX
Conference, Clearwater Beach, FL, 2004.
[15] M. Jerman, M. Keppert, J. Vyborny, R. Cerny, Hygric, thermal and durability
properties of autoclaved aerated concrete, Constr. Build. Mater. 41 (2013)
352–359.
[16] D. Taoukil et al., Moisture content influence on the thermal conductivity and
diffusivity of wood–concrete composite, Constr. Build. Mater. 48 (2013) 104–
115.
[17] H. Uysal, R. Demirboga, R. Sahin, R. Gul, The effects of different cement
dosages, slumps, and pumice aggregate ratios on the thermal conductivity and
density of concrete, Cem. Concr. Res. 34 (2004) 845–848.
[18] L. Gunduz, The effects of pumice aggregate/cement ratios on the low-strength
concrete properties, Constr. Build. Mater. 22 (2008) 721–728.
[19] O. Unal, T. Uygunoglu, A. Yildiz, Investigation of properties of low-strength
lightweight concrete for thermal insulation, Build. Environ. 42 (2007) 584–
590.
[20] R. Demirboga, R. Gul, The effects of expanded perlite aggregate, silica fume and
fly ash on the thermal conductivity of lightweight concrete, Cem. Concr. Res.
33 (2003) 723–727.
[21] J.C. Marechal, Contribution à étude des propriétés thermiques et mécaniques
du béton en fonction de la température, Annales de l’Institut Technique du
Bâtiment et des Travaux Publics 274 (1970) 122–146.
[22] M.A. Othuman, Y.C. Wang, Elevated-temperature thermal properties of
lightweight foamed concrete, Constr. Build. Mater. 25 (2011) 705–716.
[23] H.K. Kim, J.H. Jeon, H.K. Lee, Workability, and mechanical, acoustic and thermal
properties of lightweight aggregate concrete with a high volume of entrained
air, Constr. Build. Mater. 29 (2012) 193–200.
[24] K.H. Kim, S.E. Jeon, J.K. Kim, S. Yang, An experimental study on thermal
conductivity of concrete, Cem. Concr. Res. 33 (2003) 363–371.
[25] A. Neville, Propriétés des bétons, Eyrolles, 2000.
[26] S. Mindess, J.F. Young, D. Darwin, Concrete, 2e ed., Pearson Education, 2003.
731
[27] D.P. Bentz, K.A. Snyder, Protected paste volume in concrete Extension to
internal curing using saturated lightweight fine aggregate, Cem. Concr. Res. 29
(1999) 1863–1867.
[28] A. Bentur, S. Igarashi, K. Kovler, Prevention of autogenous shrinkage in
highstrength concrete by internal curing using wet lightweight aggregates,
Cem. Concr. Res. 31 (2008) 1587–1591.
[29] T.Y. Lo, H.Z. Cui, Z.G. Li, Influence of aggregate pre-wetting and fly ash on
mechanical properties of lightweight concrete, Waste Manage. 34 (2004) 333–
338.
[30] Y. Billard, Contribution à l’étude des transferts de fluides au sein d’une paroi en
béton, Thèse de doctorat de l’INSA de Lyon, 2004.
[31] S.E. Gustafsson, Transient plane source techniques for thermal conductivity
and thermal diffusivity measurements of solid materials, Rev. Sci. Instrum. 62
(1991) 797–804.
[32] J.F. Daian, J. Saliba, Détermination d’un réseau aléatoire de pores pour
modéliser la sorption et la migration d’humidité dans un mortier de ciment,
Int. J. Heat Mass Transfer 34 (8) (1991) 2081–2096.
[33] R. Evans, U.M.B. Marconi, Capillary condensation an adsorption in cylindrical
and slit-like pores, J. Chem. Soc. Faraday Trans. 2 (82) (1986) 1763–1787.
[34] K. Stephan, A. Laesecke, The thermal conductivity of fluid air, J. Phys. Chem.
Ref. Data 14 (1985) 227–234.
[35] M.L.V. Ramires et al., Standard reference data for the thermal conductivity of
water, J. Phys. Chem. Ref. Data 24 (1995) 1377–1381.
[36] L.H. Nguyen, Béton de structure à propriétés d’isolation thermique améliorées:
approche expérimentale et modélisation numérique, Thèse de doctorat de
l’Université de Cergy-Pontoise, 2013.
[37] S.K. Ghosh, O. Anwar Beg, Theoretical analysis of radiative effects on transient
free convection heat transfer past a hot vertical surface in porous media,
Nonlinear Anal. Modell. Control 13 (2008) 419–432.
[38] F. Ochs, W. Heidemann, H. Muller-Steinhagen, Effective of thermal
conductivity of moistened insulation materials as a function of temperature,
Int. J. Heat Mass Transfer 51 (2008) 539–552.
[39] A.V. Kuznetsov, D.A. Nield, The Cheng-Minkowycz problem for cellular porous
materials: effect of temperature-dependent conductivity arising from
radiative transfer, Int. J. Heat Mass Transfer 53 (2010) 2676–2679.
[40] A. Helte, Radiative and conductive heat transfer in porous media: Estimation of
the effective thermal conductivity, J. Appl. Phys. 73 (1993) 7167–7173.
[41] B.S. Hemingway, Quartz: Heat capacities from 340 to 1000 K and revised
values for the thermodynamic properties, Am. Mineral. 72 (1987) 273–279.
[42] R. Bruckner, Properties and structure of vitreous silica I, J. Non-Cryst. Solids 5
(1970) 123–175.
[43] E.C. Roberson, Thermal properties of rocks, Open-file report 88–441, U.S.
Department of the Interior, Geological Survey, 1988.
[44] A. Melling, S. Noppenberger, M. Still, H. Venzke, Interpolation correlations for
fluid properties of humid air in the temperature range 100°C to 200°C, J. Phys.
Chem. Ref. Data 26 (4) (1997) 1111–1123.