Construction and Building Materials 224 (2019) 639–658

Contents lists available at ScienceDirect

Construction and Building Materials

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

Thermal and mechanical properties of fired clay bricks made by using

grapevine shoots as pore forming agent. Influence of particle size and
percentage of replacement
P. Muñoz a,⇑, M.A. Mendívil b, V. Letelier c, M.P. Morales d
a
Facultad de Ingeniería, Universidad Autónoma de Chile, Chile
b
Escuela Técnica Superior de Ingeniería Industrial de la Universidad de La Rioja, Luis de Ulloa 20, 26004 Logroño, La Rioja, Spain
c
Departamento de Obras Civiles, Universidad de la Frontera, Francisco Salazar 1145, Temuco, Chile
d
Universidad Internacional de La Rioja, Avda. La Paz, 137, INeS Research Group, Spain

h i g h l i g h t s

The technical feasibility for manufacturing fired clay bricks by adding vine shoots has been tested.

The influence of particle size and percentage of vine shoot replacement has been discussed.

The percentage of vine shoot replacement highly impacts on mechanical and thermal properties.

The influence of particle size on mechanical properties increases by increasing percentage of additive.

The proposal environmental impact seems to be positive but further assessments must be conducted.

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

Article history: For decades, the construction industry has focused on developing new materials to minimise environ-
Received 15 February 2019 mental impact and improve building insulation envelopes. The use of certain wastes as additives for con-
Received in revised form 6 July 2019 struction materials (e.g. concrete, clay, steel) has been shown to be a feasible alternative that may
Accepted 9 July 2019
improve both the current waste management strategy and the properties of construction materials.
Among others, the fired clay brick (FCB) industry has been highlighted as an optimal sector for incorpo-
rating such residues, due to the large amount of mass flow involved, the use of firing temperatures above
Keywords:
900 °C and the observed positive effects of FCB thermal conductivity (TC) when organic residues are
Grapevine shoot
Thermal insulation
added. Although the properties of fired clay are closely related to the firing curves and the mineral
Compression strength and chemical composition of raw materials, the effect of particle size is also a key factor that directly
Fired clay brick impacts on the economic and technical feasibility. This paper therefore assesses the addition of wood
Agro-waste chips from pruned grapevine shoots as an additive for manufacturing FCBs, and considers three particle
Valorisation sizes (up to 0.5 mm, above 1.5 mm and between 0.5 and 1.5 mm) and variations in the added percentage
on a dry basis. Several samples were tested in order to determine the mechanical and thermal behaviour
of FCBs, and it was concluded that the maximum of 10% wood chip content is limited by both the com-
pressive strength values and the water absorption. At this percentage, TC is reduced by up to 50%. The
particle size seems to have no effect on the thermal properties, but gives rise to significant variations
in terms of the mechanical response.
Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction
Two main issues are often highlighted in relation to the
sustainable development of modern societies. The first is a search
for feasible ways of improving waste management (i.e. reducing,
⇑ Corresponding author.
E-mail address: pmunozv@uautonoma.cl (P. Muñoz).
reusing and recycling). Although the EU has reduced the amount
of waste produced in overall terms, the residues produced by the
construction sector grew by up to 4% between 2006 and 2016 [23].
Many published studies have presented several ways to reduce,
reuse and recycle waste, and this study is focused on the use of
waste as a replacement for raw materials (i.e. clay). There are sev-
eral reasons for this slow transition to a circular economy strategy
on an industrial scale, but one in particular should be highlighted.

https://doi.org/10.1016/j.conbuildmat.2019.07.066
0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.
640 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658
Unfortunately, most eco-friendly materials made using residues
show worse performance than natural raw materials, in terms of
their mechanical behaviour. Concrete and fired clay bricks (FCBs)
are two examples [29,30,40].
The second issue is that buildings are significant consumers of
energy. For instance, the demands of the residential sector account
for about 26.5% of the total energy consumed in the EU [22]. A large
proportion of this energy is used for heating, ventilating and air-
conditioning. To a great extent, the quality of building enclosures
determines the energy demand involved in achieving thermal
comfort. In Spain, the Código Técnico de la Edificación (Technical
Building Code) [7] has recently been updated, and the thermal per-
formance of the building envelope has been reviewed. Hence,
depending on where the building is located [7], thermal transmit-
tance (Uvalue) requirements currently need to be achieved by devel-
oping new construction materials with lower thermal conductivity
(TC). Thus, a potential solution might be based on the reuse of resi-
dues in the search for better thermal performance. While concrete-
based products tend to be used for structural purposes, FCBs are
commonly part of the envelope system, and have generally been
designed without considering structural loads (wind, snow, seis-
mic loads, etc.). At this point, a decrease in mechanical resistance
(e.g. flexural and compressive strength) does not represent a criti-
cal issue. On the contrary, the enhancement of thermal properties
may strongly impact on the energy performance of the building. In
addition, lightweight bricks for infill walls reduce building struc-
tural requirements and it also seems to improve the seismic
response [11].
The literature gives several examples of clay-based products
made by replacing different percentages of clay matter with resi-
dues [1,55]. Most of these studies focus on physical and mechani-
cal properties, and thermal behaviour is rarely considered [40].
It has been found that the traditional FCBs properties are highly
influenced by used clays (e.g. mineral composition, particle size
distribution, organic content, moisture, etc.) [41] and technological
parameters (e.g. forming technology, conformation pressures, fir-
ing temperature, soaking time or drying processes, etc.) [25].
While inert waste commonly reacts during the firing process
(i.e. at temperatures above approx. 600 °C), residues based on
organic carbon are burned in the pre-heating stage prior to the
transformation of clay into ceramic (for example, quartz reacts at
above approx. 550 °C). This calcination leads to the production of
gas bubbles and ashes, which are generated by the volatility of
the combustible materials during the firing process and which
react with the minerals of the clay matrix and cause the ceramic
properties to vary.
Organic pore-forming additives are generally cheaper than inor-
ganic ones, and are commonly easier to manage (in terms of their
separation, collection, transportation, etc.). In fact, most are
byproducts or even residues from other activities for which new
feasible alternatives need to be found to avoid their disposal (i.e.
reduce, reuse and recycle). In addition to this undeniable advan-
tage, some also possess high calorific value; that is, their combus-
tion during the firing process may involve a valuable heat
contribution, resulting in a reduction in the energy consumption
of the kiln depending on the technology used for firing. Moreover,
the addition of certain residues to the clay also modifies all of these
parameters and leads to better thermal insulation behaviour.
Hence, the selection of residues must focus on both the final
properties of FBCs and the overall economic cost (i.e. storing of
the residue, transportation to factories and pre-processing
requirements, among others). The aim is to develop a new con-
struction material with enhanced properties and similar process-
ing costs. In this context, the residue particle size plays an
important role, since its original state should be modified at
the point of generation in order to allow for economical
transportation and incorporation into the clay mass flow within
the factory process.
Grapevine shoots (GVSs) provide an organic residue that is gen-
erated on a vast scale in all viticulture regions of the world. It is
estimated that solely in the Rioja region of Spain, there is a surplus
of 180,000 metric tons of this matter per annum [36].
Despite its theoretical potential use as a solid biofuel, there are
currently several barriers to this end use (e.g. low price of fuel oils,
contamination of GVSs with earth and stones, traces of sulphur,
which requires a purification process, etc.). The most widespread
existing end use is based on a chipping process, which is followed
by atomisation at the vineyard with the aim of using this as bio-
fertiliser. Despite the positive effects on vineyard yields, the
uncontrolled degradation of organic matter leads to substantial
increases in pests and CH4 emissions.
In this context, and given the fact that GVSs in Rioja constitute a
very abundant residue that incurs costs to the vine grower for its
disposal, its revalorisation as an additive for clay in the brick indus-
try has been proposed. Several previous studies the literature have
been conducted to examine the effects of the addition of similar
residues [12,10].
However, to the best of our knowledge, the effects of both per-
centage and particle sizes have not been reported. Consequently,
the present study aims to achieve the following objectives:
- To study the influence of the proportion and the particle size of
added GVSs on certain physical properties of brick products,
such as the bulk density (BD), water absorption (WA) rate and
TC.
- To determine the sensitivity of the mechanical resistance of
FCBs to variations in the quantity and particle sizes of the
additive.
- To establish the relevant limits on the quantity and particle size
of GVSs added to the clay, according to the current standards for
product manufacturing.
2. Materials and methods
2.1. Materials
GVSs were collected from vineyards owned by the regional gov-
ernment in Rioja, Spain. From the available grape varieties, the
Tempranillo red variety was chosen, as this represents the most
widespread variety in the region. In addition, this variety has
already been the subject of a precise chemical characterisation in
previous studies carried out by the current research team [36].
Despite different types of grape varieties may be considered, sev-
eral authors have concluded that no significant differences may
be highlighted in terms of calorific value and ashes content (Sid-
dique, 2008; [35].
Additionally, differential thermal analysis (DTA) and thermo-
gravimetric analysis (TGA) have been determined by means of a
STA 6000 (PerkinElmerÒ). This, in accordance with the methodol-
ogy for a fast characterization of biomass fuels showed by Saldar-
riaga et al. [46] which allows calculating in a single run the
contents of moisture, volatiles, fixed carbon and ash. From this
TGA analysis two different processes may be distinguished (See
in Fig. 1). The first one is associated with the fast combustion of
the volatiles (i.e. produced by thermal degradation of hemicellu-
lose and cellulose) and the other one is associated with the com-
bustion of the char (i.e. the fixed carbon). The thermal analysis
shows two exothermic peaks that correspond to the combustion
of the volatile compounds (i.e. from 150 °C) followed by non-
volatile components (second peak) (i.e. from 500 °C).
On the other hand, several types of clay are available
worldwide, and their composition is a key factor in terms of the
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658
Fig. 1. DTA-TGA analysis of sawdust from GVSs.
properties of FCBs. Although some studies have considered a mix-
ture of different clays, red clay was used as the sole component in
the present research, with the aim of limiting the number of vari-
ables. Clay was obtained from a quarry owned by an associated
company (i.e. Herederos Cerámicas Sampedro Ltd.). It must be
noticed that XRD pattern of fired clay has been discussed in previ-
ous research [34] while XRD pattern of unfired clay has not been
previously reported.
Hence, dust samples of unfired (Fig. 2a) and fired (Fig. 2b) clay
were analysed in terms of its mineralogical features via X-ray pow-
der diffraction (XRD), using a Bruker D8 Advance device with a
scanning velocity of 0.5° (2h) min1, a phase of 0.15 and Cu-Ka
radiation (40 kV, 30 mA) at the Centro de Investigación Tecnologica
e Innovación de la Universidad de Sevilla (CITIUS).
The unfired clay sample under analysis was composed of
quartz, calcite, feldspar, goethite and phyllosilicates. The largest
proportion of the sample was made up of quartz, calcite and phyl-
losilicates (Fig. 2a). Feldspar, which was less prevalent in the sam-
ple, could not be identified with certainty, while the goethite
appeared as traces (<2%).
The methodology proposed by Schultz [48] was used to achieve
this quantification.
Clay mineral identification was conducted using an oriented
aggregates diagram, and was air-dried, treated with ethylene gly-
col and heated to 550 °C. The following phases were identified:
illite, kaolinite, chlorite and smectite. The existence of a random
I/Sm interstratification, rich in illite, is also possible. Based on the
intensities of the basal reflection (d001Å) of these minerals and
their relative reflecting capabilities, a quantitative estimation was
made [24]. The mineralogical composition and the main phases
of the unfired clay material are showed in Table 1.
The results are similar to the referred quantitative representa-
tion (QR) reported by previous authors [27]. It should be noted that
although QR represents the existing mineral content, it does not
imply an exact measure of the absolute quantity.
641
2.2. Samples
GVSs were naturally dried, chipped and classified, based on the
particle size, into three groups: up to 0.5 mm (Type A), above
1.5 mm (Type C) and between 0.5 and 1.5 mm (Type B). The parti-
cle size measurement was conducted using several lab hand sieves
in order to classify particles into these three different groups
(Fig. 3). At this point, some researchers recommend limiting the
maximum particles sizes to 2 mm [47]. The main reason is mainly
related to the workability and drying process. However, it must be
noticed that particles >1.5 mm were added (e.g. 2.0, 2.5 or even
3.0 mm) with the aim of showing the feasibility of reducing the
chipping process requirements [43].
Clay and biomass were added and mixed in a dry state (i.e. after
24 h at 105 °C). An automatic laboratory mixer (5 L) was then used
to form the blend. The required water was added directly to the
vessel of the laboratory mixer. As it has been well stated by previ-
ous authors [4], the addition of organic residues leads to decrease
the plasticity mixture since the water is retained by the biomass
and less water is available for the clay particles. When the plastic-
ity is inadequate, extrusion failures and heterogeneities develop in
the moulded clay body which, consequently, lead to weak mechan-
ical properties [8]. Hence, in order to keep the feasibility of shaping
samples, the amount of moulding water was increased with the
increasing of the amount of GVS (See in Table 2).
Nine different mixtures corresponding to three different pro-
portions in terms of the weight of the additive (dry basis) and three
different particle sizes (Table 2) were tested to determine the influ-
ence on the amount of GVSs used as additive. The P00 series, in
which specimens were made of clay with no additive, represented
the reference pattern. In order to conduct statistically significant
studies, 10 test specimens were prepared for each mixture of the
additive.
Each blend was set on a square based-mould for test specimen
conformation. The samples were formed by moulding under an
642 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. 2. Powder XRD spectra of a) unfired clay and b) fired clay. Q = quartz, Ca = calcite, I = illite, K = kaolinite, F = feldspar, Cl = chlorite, G = goethite.

applied pressure of 2.5 MPa. In this way, 400  400 mm green test
Table 1 specimens were obtained, with slightly varying thicknesses
Quantitative representation of the main phases and minerals in unfired red clay. according to the proportion of additive in the mixture (between
Phase Content [%] Minerals Content [%] 15 and 20 mm). Samples geometry was designed with the aim of
Quartz 29 Illite 80 improving the precision of thermal measurement by fitting the
Calcite 12 Kaolinite 12 specimens to the thermal test chamber sizes. In addition, it must
Feldspar 2 Clorithe 8 be noticed that the width was limited by the risk of failures and
Phyllosilicates 57 Smectites Traces cracks during the firing process.
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 643

construction material, the WA was calculated by following [17]

Fig. 3. The different particle sizes (A–C) of Tempranillo GVSs used for mixing.
The green test specimens were subject to natural drying in the
lab for 72 h, in a constant atmosphere at 20 °C and 30% relative
humidity (RH). After acquiring the desired consistency, the test
specimens were demoulded and shipped to the associate company.
At their facilities, green specimens were integrated into the origi-
nal production flow of the factory, during which samples under-
went a drying process. This stage consisted of several chambers
in which pieces were exposed to different atmospheres (i.e. from
35 °C at 90% RH to 100 °C and <5% RH). Although rehydration
may be an important issue, the design of the production line makes
water transfer impossible, and this means that the end of the dry-
ing process is the beginning of the firing stage. The temperature of
the kiln was controlled and corrected if necessary at each stage to
obtain standardised conditions. In this manner, all test specimens
experienced a similar firing process (i.e. a firing temperature of
950 °C, a soaking time of 2 h, and rates of temperature
increase/decrease of 62 °C h1 and 115 °C h1 for the heating and
cooling phases, respectively).
After the firing process, the test specimens were subjected to
control trials prior to their testing. In this operation, their general
features were checked, such as the presence or absence of cracks,
calcareous inclusions or efflorescence.
2.3. Assays
Each test specimen was first measured in terms of its width,
length, and height, following the instructions given in [21]. Dry
matter was obtained by carrying out consecutive weightings until
a constant weight was achieved (i.e. a variation of <0.2%). Each
fired specimen was then submerged in a water tank for 24 h.
Underwater weighing was conducted for each piece at intervals
of 30 min, until the variation in the apparent submerged mass
was <0.2%. Immediately following this procedure, superficial water
was removed from the piece using a cloth, and weighing was con-
ducted outside of the water tank. In this way, it was possible to
determine the net volume and percentage of voids and the net
and gross dry density by following [19] and [20], respectively. In
addition, in order to complete the characterisation of the proposed
and the efflorescence behaviour was observed by following [51].
In addition, a Nikon SMZ-2 T (i.e. magnifications up to 63) micro-
scope with a trinocular head that allows to rapidly change from
stereoscopic binocular viewing to photo video capability was used.
Prior to submitting the fired samples to thermal assays, all
pieces were dried again, as explained above. TC measurements
were carried out by means of a guarded hot plate and heat flow
meter, in accordance with the [15] and [16] standards, subjecting
the test specimens to three different thermal leaps of 20 °C, 30 °C
and 40 °C. Following these standards, the TC was calculated in each
of the test specimens at a temperature of 10 °C (k10) via linear
regression extrapolation. In order to minimize the effect of the por-
ous surface, adaptable pads (i.e. THERM-A-GAP 976Ò) were
inserted between the specimen surfaces and both plates (i.e. cold
and hot plates). These high TC gap filler pads lead to absorb any
shortcomings in flatness that might exist. A more detailed descrip-
tion of the methodology is given in previous researches [38].
After the non-destructive assays were being complete, mechan-
ical characterisation was carried out. Compression tests were con-
ducted by means of a universal machine [49].
This equipment incorporates displacement and double pressure
transducers which lead to directly determine strain-stress curves
without any additional gauges but high level of accuracy (i.e. up
to 0.001 mm). Thus, the static modulus of elasticity (MoE) in com-
pression may be calculated, in accordance with [52]. The load was
applied at different rates (i.e. from 0.3 to 0.05 N s1) so that the
maximum load was reached in not less than approx. 1 min, in
accordance with [18]. Samples were machined in a column drill
with the aim of obtaining several cylinders (i.e. of 19 mm ± 2%
diameter) from each original specimen (i.e. 400  400 mm). All
cylinders were dried before testing and measured by a calliper
(i.e. diameter and width) with the aim of determining the ultimate
compressive strength (CS) and the corresponding strain (e). A more
detailed description of the methodology is given in previous
researches [39] and in Appendix A.
From decades various models and fracture criteria have been
proposed with the aim of determining the essential features of
the mechanisms which lead to brittle fracture under compression.
Uniaxial compression test shows nonlinearity due to closing of
some crack-like voids followed by a region of linearity. During this
elastic loading phase, forces rise rapidly with an almost constant
deformation rate. At this point a maximum stress value is carried
out [28]. This value has been the one used (rf), in this research,
for determining both the CS value and MoE. It must be noticed that
after this point (rf, ef), the specimen shows a plastic behaviour
where applied forces may be further increased and the homoge-
neous strain trend highly varies. However these lasts points of
the curve were not considered for calculations.

Table 2
Series characteristics.

ID Claya[%] Additivea[%] Particle size # Samples Plasticityb[%]

P00 100 – – 10 10 (0.2)
A05 95 5 A size < 0.5 10 11.92 (0.8)
A10 90 10 A size < 0.5 10 26.57 (0.03)
A15 85 15 A size < 0.5 10 30.54 (0.1)
B05 95 5 B 0.5 < size < 1.5 10 13.24 (0.7)
B10 90 10 B 0.5 < size < 1.5 10 27.19 (0.7)
B15 85 15 B 0.5 < size < 1.5 10 31.66 (0.2)
C05 95 5 C size > 1.5 10 24.85 (1.4)
C10 90 10 C size > 1.5 10 28.95 (0.9)
C15 85 15 C size > 1.5 10 31.88 (0.4)
a
percentage expressed in dry basis.
b
Average plasticity for all series, between parentheses the standard deviation is reported.
644 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658
All data collected from assays were statistically analysed. First,
basic statistics were calculated (e.g. mean, standard deviation,
coefficient of variation, etc.) in order to describe each case.
Following this, the Kolmogorov-Smirnov methods were applied
with the aim of checking the normality of all distributions. Because
of the inconclusive tests, the Kruskal-Wallis (K-W) method was
considered (i.e. the non-parametric counterpart of the one-way
ANOVA Test) in order to demonstrate the significant differences
among series (i.e. in terms of particle size and percentage of added
residue). The K-W test assesses whether independent samples are
from the same population or from populations with continuous
distribution and the same median for the variable being tested.
The statistical null hypothesis (Ho) (i.e. the one that assumes there
is no significant difference between the specified k-populations) is
accepted when the asymptotic significance (pvalue) is above the
level of significance (a = 0.05). On the contrary, pvalue < a lead us
to reject Ho. Besides, the significance difference was also deter-
mined by means of the one-way analysis of variance (i.e. ANOVA)
test which allows assessing whether or not the null hypothesis
should be rejected when only one grouping variable is present in
the dataset [33].
3. Results and discussion
3.1. Bulk density and water absorption
In the case of the control specimens, the average BD and WA
were 1700 kg m3 and 17.6%, respectively. Similar values were
obtained in previous researches when similar clays were used. In
spite of Demir [8] did not report the particle size of sawdust, he
showed similar relative BD decreases (i.e. 25% for 10% of replace-
ment) and WA increases (i.e. from 17% to 31% for 10% of replace-
ment). Besides, similar increasing/decreasing rates were showed
when authors fired specimens at higher temperature (i.e.
1000 °C) [3].
Fig. 4 illustrates the evolution of BD in terms of amount of addi-
tive and the considered particle size. Despite other authors showed
a sharply drop followed by a slight reduction when further residue
Fig. 4. Bulk density (kg m3) as a function of additive percentage grouped by the particle size.
was added, Fig. 4 shows a linear trend. This may be explained due
to differences regarding the plasticity which highly influence BD
trend.
There are several reasons for this reduction in density. As has
been widely demonstrated, the particle density of the bricks is inver-
sely proportional to the quantity of organic matter added to the mix-
ture. The main reason is related to the amount of water that is added
to the blend to form the samples. Since the amount of water added
varies with the aim of achieving adequate plasticity, the mixture
absorbs more water due to larger pore sizes and higher pore vol-
umes, resulting in a lower density. The firing temperature also
affects the density of the bricks. Hence, the burning of the additive
produces a local increase in temperature that may cause the density
of the overall piece to vary [53]. In addition, the specific gravity value
of the waste material also contributes to a reduction in the density.
In terms of the WA, Fig. 5 depicts the evolution of the test spec-
imens as a function of the quantity and residue particle size. The
percentage WA produced through superficial pores is a function
of the open porosity (OP). As shown in Fig. 5, a strong variation
in WA appeared with an increase in the quantity of additive.
Although the BD and WA are clearly influenced by the percent-
age of additive (i.e. K-W test corroborates that the amount of resi-
due is statistically significant), the dependency of the particle size
appears to be negligible at 5% (pvalue > 0.05). K-W test shows an
increasing of the asymptotic significance from 10% and the effect
of particle size at 15% of additive is considered statistically signif-
icant (pvalue < 0.05) in both cases (i.e. BD and WA).
The WA strongly depends on the connectivity of pores from the
surface to the internal matrix, since the water is introduced by cap-
illarity. Hence, when larger pores are formed, the number of avail-
able paths for water transfer is increased. This can clearly be
observed by analysing both the external surface and the interior
of the test specimens with the aid of a simple optical microscope.
Fig. 6 allows a comparison of test specimens with increasing
amounts of added residue by varying the particle size (i.e. A, B
and C). Fig. 7 shows the interior of the samples, with increasing
numbers and sizes of cavities resulting in a more connected pore
network.
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 645

Fig. 5. Water Absorption (%) as a function of additive percentage grouped by the particle size.

Fig. 6. Outer aspect of the test specimen with increasing amounts of GVSs additive and different particle sizes.

On the other hand, WA percentage is a measurement of the
durability to environmental humidity. The minimal requirements,
regarding WA for FCBs are not expressly determined by the current
normative in Spain. Other standards such as ASTM C20, indicates
that WA should be <17% or 22% for severe or moderate weather
resistance, respectively, Indian standards limit the WA to 20%
[37] and China sets 30% WA as the maximum [31].
The WA indices resulting from the testing of all mixed series
were above 20% (clay with no additive has a WA value of 17%),
meaning that the mixture with GVSs should not be used for the
manufacture of bricks that will not be cladded. However, there is
no contraindication in the case of cladded walls.
Furthermore, OP (i.e. apparent porosity) quantifies the fraction
of volume that may be occupied by water in the interconnected
646 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. 7. Interior of test specimen with increasing amounts of GVSs additive and different particle sizes.

Fig. 8. Open Porosity (%) as a function of additive percentage grouped by the particle size.

porous network. Related to the carried out values (see in Fig. 8),
similar trends may be observed by previous authors [14]. Differ-
ences, in terms of absolute values, may be explained due to the
higher firing temperature followed by Eliche-Quesada et al. [14].
The increased of temperature from 950 °C to 1050 °C produces an
increase in liquid phase amount, which tends to approach the fine
pores contained in the ceramic body and therefore, OP decreases.
However, when similar temperatures are considered the increasing
rates tend to converge [13,8]. Regarding the influence of both
variables (i.e. percentage and particle size) statistics show similar
results than BD and WA. Thus, the effect of percentage is unar-
guable while the particle size impact is increased from 5% of
replacement and becomes statistically significant for 10% and
15% of added residue.
Finally it should be noted that all specimens can be classified as
‘‘not effloresced, regardless of the considered percentage or parti-
cle size. Efflorescence consist on the deposit of water soluble salts
(e.g. Na, K, sulphates and carbonates, Ca, Mg, Al, carbonates, etc.)
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658
that are carried by water through the FCBs and remain either on
the surface or behind the surface of FCBs. These salts are accompa-
nied by increasing local pressure which can eventually result in a
reduction of the FCBs mechanical properties. Another problem
related to the fired clay matrix is the so called black core. This
effect may be promoted by the high proportion of certain carbon-
ates or the uncomplete combustion of the organic matter. When
the OP is reduced the diffusion of oxygen from the outside to the
inside surface is also reduced and gases are trapped inside the fired
clay matrix which causes bloating and cracks. This black core has
not been detected in any specimen.
3.2. Thermal conductivity and compressive strength
The results show some discrepancies from previous studies, in
terms of the TC and CS absolute values while similar results can
be showed in terms of relative decreasing rates. This is as expected,
since both of these parameters are very sensitive to the raw mate-
rials and production procedures used and the measurement
method followed. The TC and CS mainly depend on the mineral
structure, which strongly depends on the raw materials, conforma-
tion pressure, drying and firing process (e.g. increasing/decreasing
temperature rates, soaking time, moisture content, etc.). In addi-
tion, the measurement procedure also influences these values
since the boundary conditions vary greatly from one method to
another [25,9]. For instance, the TC may be determined by several
methods (e.g. hot wire method, hot guarded plate method, tran-
sient analysis, etc.), and the CS values will vary with changes in
the geometry of the samples, the loading speed rate or the mois-
ture content.
Fig. 9 depicts the evolution of TC at 10 °C (ʎ10) with respect to
the amount and particle size of GVSs additive. A reduction in TC
with an increase in the percentage of additive is clearly shown.
This means that GVSs is a very efficient agent in terms of improv-
ing the thermal insulation capabilities of FCBs, giving similar
results to the previously mentioned biomass residue additives.
Hence, Bánhidi and Gömze [2] showed a reduction of TC between
10% and 15% for replacement ratios of 4% and 7%. Saiah et al.
Fig. 9. Thermal Conductivity at 10 °C [W m1 K1) as a function of additive percentage grouped by the particle size.
647
[45] also showed improvements above 15% for 9% replacement
while García-Ten et al. [26] or Shibib et al. [50] used up to 15% of
replacement and highlighted decreasing rates above 40%.
More specifically, it can be stated that the contribution of GVSs
particles reduces TC in a more efficient manner than other organic
residues while being a better combustive with a lower presence of
ash [35].
Similarly to the trends in BD and WA, the TC decreases mark-
edly with the addition of GVSs up to 15%, regardless of the size
of these particles. Therefore, when only the thermal properties
are taken into account, very fine grinding (i.e. size A particle size
up to 0.5 mm) does not appear to be necessary, since test pieces
of size C have a similar thermal behaviour to samples of sizes A
and B. K-W test corroborates it by showing pvalue above the level
of significance in all cases.
Since the brickwork can be used for structural purposes, the rel-
evant codes specify the minimum values to accomplish. According
to the current building regulations in Spain (CTE), the CS of FCBs
used for supporting structural loads must be at least of 5 N mm2.
As it can be observed in Fig. 10, CS values rapidly drop by increas-
ing the percentage of the additive. The obtained mechanical CS
trend verifies that amounts of additive up to 10% do not lead to
CS values below the normative threshold of 5 N mm2. At this
point it can be concluded that an improvement of 65% in thermal
isolation capacity of the samples may be achieved while mechan-
ical behaviour successfully accomplishes the codes in force. In
other words, it can be said that GVSs particles are a very effective
agent in improving insulating capacity of FCBs, achieving similar
results to other biomass additives already referred (paper pulp, rice
husk, coffee grounds, etc.). However when compared CS of FCBs
produced with other organic additives, GVSs originates more pro-
nounced drops in compressive strength than other tested organic
residues [38,39]. Similar decreasing rates were show by previous
authors. For instance, Bánhidi and Gömze [2] showed a reduction
of approx. 25% and 50% (for 4% and 7% of replacement respectively)
while Eliche-Quesada et al. [14] achieved percentages of CS reduc-
tion between 20% and 75% (replacement percentage set at 3% and
10%, respectively).
648 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. 10. Compressive strength [N mm2] as a function of additive percentage grouped by the particle size.

In addition, the statistical analysis lead to consider that particle
size slightly influences the CS variation. The subsequent K-W test
demonstrated that CS and particle size are related from a statistical
point of view (i.e. pvalue < 0.05 in all cases). For a given percentage
of additive in the mixture, the larger the particle size, the lower the
compressive strength. This may be explained due to the significant
reduction of the effective section when larger particles are added.
These introduced voids actually imply internal cracks within the
ceramic matrix that cannot be offset by the mineral structure.
Some authors have demonstrated that pore volume highly impacts
on the average fracture strength while MoE is highly influenced by
the pore size distribution [54]. These discontinuities are increased
due to the internal cracks propagation which seems to increase the
brick ductility by reducing the calculated MoE value (see in Fig. 11)
[6]. Moreover, potential energy absorbed during the stress-strain
assay (which may be calculated from MoE and CS values) decreases
when the GVSs percentage is increased.
3.3. Mineral analysis
The mineral analysis provided by the XRD analysis show
expectable results. The fired clay is basically formed by quartz,

Fig. 11. Static MoE [N mm2] as a function of additive percentage grouped by the particle size.
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 649

calcite, feldspar goethite and phyllosilicates (see Table 1).

Regarding the effect of the additive, the mineral composition of
fired clay, in all samples, keeps constant. As it can be observed in
Figs. 12–14, the ultimate composition is the same, regardless the
added percentage and the particles size. Based on these figures
and the TGA analysis it can be concluded that GVSs has completely
burned and the produced ashes seem to have a negligible effect on
the resulting mineral composition of FCBs.
On the other hand, in accordance with the powder XRD analysis,
the following phases may be highlighted: quartz, mica, melilite (i.e.
gehlenite), wollastonite, anorthite and iron minerals (i.e. hema-
tite). In addition, compared to the control samples XRD pattern
(Fig. 1b) it can be concluded that the additive did not modify the
mineral composition on fired samples which implies that residue
combustion did not change the firing temperature and the entire
organic matter burned during the firing process.

3.4. Environmental impact

Regardless the positive effect, in terms of the aforementioned

TC, the replacement of clay by GVSs has two other major impacts.
On the one hand, raw material depletion is reduced which leads to
decrease the equivalent CO2 emissions (CO2,eq.) from heavy
machine. Moreover, quarries show poor land use efficiency (i.e.
approx. 1.38 m3 m2) and a low rehabilitation degree of deployed
quarries (<1.1%) [5]. Hence, a reduction of required amounts of clay
positively impacts on environmental protection.
On the other hand, the replacement of clay by GVSs leads to
reduce CO2,eq. gas emissions during the firing process. Natural clay
contains MgCO3 and CaCO3 that react during the firing process to

Fig. 13. XRD Pattern of series with 10% of additive whit different particle sizes (i.e.
A–C).

Fig. 12. XRD Pattern of series with 5% of additive whit different particle sizes (i.e.
A–C).
produce CO2 and oxides (i.e. from 600 °C MgCO3 reacts to from
MgO and CO2 and above 900 °C CaCO3 produces CaO and CO2).
Hence, based on the biomass neutral carbon cycle, it can be con-
cluded that, the lower amounts of used clay, the lower amount
of CO2,eq. net emissions.
Regarding the impact in terms of technological process, it must
be considered that firing stage represents the major environmental
impact (from 30% to 50% in terms of global warning). However, this
impact highly depends on both the firing technology (e.g. tunnel
kiln, clamp kilns, Hoffman kiln, fixed chimney, etc.) and the used
fuel (e.g. natural gas, diesel, biomass, coal, etc.). Related to the
energy savings, is widely assumed that the lower heat value of
additive (i.e. approx. 18,717 kJ kg1) may be considered within
the overall firing energy balance and consequently the energy
required by the firing process might decrease (Mendivil et al.,
2015). From TGA analysis, two exothermic peaks can be observed
between 250 and 650 °C which are attributed to the combustion
of organic components in the waste (approx. 96%) that might be
accounted for energy savings [44].
However, such energy contribution must be balanced against
the increased energy demands of the drying process which result
from the increased amounts of moulding water and the energy
needed for pre-treatments (i.e. transportation and chipping).
Besides the used technology clearly impacts on energy savings
and CO2,eq. emissions. Therefore, it is not possible to show a realis-
tic average of savings percentage based on the aforementioned
assumption [32]. In spite of particle size might be increased, which
involves lower energy consumption, further analysis must be
developed prior to determine the actual energy impact of the GVSs
usage.
650 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658
Fig. 14. XRD Pattern of series with 15% of additive whit different particle sizes (i.e.
A–C).
Finally, life cycle assessment should also take into account that
lightweight FCBs reduce the weight of the building system. Conse-
quently, the quantity of structural elements (e.g. concrete, steel)
might be reduced which it redounds in benefit of the environment,
as well. In addition, advantages of lightweight FCBs in construction
include a lower transport costs, easier handling and higher produc-
tivity, since more bricks can be produced per ton of raw materials
[47].
4. Conclusions
The technical feasibility of different percentages of additive
with different particle sizes has been stated. It has been demon-
strated that the use of GVSs as an additive in the manufacturing
of lightened FCBs led to produce lighter FCBs with lower TC. It
has been also highlighted that such reduction of BD and TC go hand
by hand with a reduction of CS and the increasing of WA.
These undesired effects limit the amount of residues that may
be added due to the mechanical requirements. Moreover the WA
values force to rough coat the masonry walls in order to protect
them from severe or harsh weather.
Regarding the effect of particle size, it has been demonstrated
that TC is not affected by such variable while BD and WA seems
to be slightly modified when the amount of residue exceeds 5%.
Besides, CS abruptly dropped with the increase on the amount of
the additive and the greater the size of the additive particles, the
more pronounced the compressive strength decrease was.
Therefore, a limit of 10% in the proportion of GVSs is established
with aims of maintaining above the threshold of 5 N mm2. At this
percentage, the particle size of the GVSs may be set at 1.5 mm
which will reduce the impact of additive pre-treatments. At this
percentage of clay replacement a drop of the TC is achieved in
the fired samples that represents an improvement of 50% com-
pared to the control samples.
Declaration of Competing Interest
None.
Acknowledgements
It must be noticed that this work was supported by the Chilean
National Commission on Research and Development (CONICYT),
Chile, Universidad Autónoma de Chile, [Grant number FONDECYT
REGULAR 1180414]. In addition, I would like to express my very
great appreciation to the staff of Hdros. Cerámica Sampedro S.L.
for his valuable and constructive suggestions and for enabling me
to access and use their facilities and development of this research
work. Technical and human support provided by BIOREN of
Universidad de la Frontera is gratefully acknowledged. Thanks go
to the late Dr. Emilio Galán Huertos (recently deceased) from the
crystallography, mineralogy and chemistry Department of Univer-
sity of Seville who is recently deceased. In sorrow, we dedicate this
manuscript to his memory. More researchers like Dr. Galán are
always necessary.
Appendix A. Detailed methodology to determine mechanical
properties of fired samples
A.1. Samples preparation
In accordance with [52], static modulus of elasticity (MoE) in
compression must be determined from cylindrical samples. There-
fore, the original fired samples (i.e. 400 mm  400 mm) were
machined with the aim of obtaining several cylindrical specimens
(i.e. from each original sample) based on a random selection pro-
cess (Fig. A1). Hence, original samples were milled with a diamond
cutter head (i.e. interior diameter of 19 mm ± 2%) mounted on a
column drill that ensured a perfect right-angle between the gener-
atrix of the test specimen and the load surface which prevented the
appearance of eccentricities that might distort the results (Fig. A2).
These obtained cylindrical specimens (Fig. A3) were measured
(Fig. A4) by means of a calliper and dried until constant mass in
Fig. A1. Selection of compression test specimens from original fired samples.
 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 651

Fig. A2. Machining original fired samples for obtaining cylinders.

Fig. A3. Cylindrical specimens.

Fig. A5. Drying process.

an electric oven (Fig. A5). Then they were placed between the cor-
responding test plates in the aforementioned universal test
machine (i.e. [49] (Fig. A6). The software (i.e. [42] drove each assay
by controlling required key parameters (e.g. load rate, displace-
ment, time, etc.), according to the selected standard (i.e. [52]
(Fig. A7).

A.2. Calculation procedure

The average Young’s module has been determined in accor-

dance with [52]. Hence, the strain and stress values are calculated
from the average slope of the initial linear trend of the stress-strain
curve. During the assay displacement and applied force are regis-
tered and the curve is split into three zones in accordance with
the elastic, plastic and elastoplastic behaviour. The software (i.e.
[42] automatically calculates the regression line within the consid-
ered elastic behaviour zone. The software uses such linear regres-
Fig. A4. Measuring the samples.
sion to calculate the strain and stressing increments which lead to
652 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. A6. Picture of the cylindrical specimen placed between test plates and sample’s end state.

Fig. A7. Screenshot taken during the software running.

determine the static MoE (Fig. A8) in accordance with Eqs. (1) and
(2), as it is stated in [52].

Dr
MoE ¼ ð1Þ
De a

Dl
ea ¼ ð2Þ
lo

A.3. Results

In order to show the effect of addition and particle size, on the

stress–strain behaviour of control samples (Fig. A9) the following
figures (i.e. Figs. A10–A18) summarize some of the curves carried
Fig. A8. Stress-strain model for brittle materials. out by the aforementioned assays.
P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 653

Fig. A9. Strain-stress curves from control samples.

Fig. A10. Strain-stress curves from type A samples at 5%.

654 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. A11. Strain-stress curves from type A samples at 10%.

Fig. A12. Strain-stress curves from type A samples at 15%.

P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 655

Fig. A13. Strain-stress curves from type B samples at 5%.

Fig. A14. Strain-stress curves from type B samples at 10%.

656 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

Fig. A15. Strain-stress curves from type B samples at 15%.

Fig. A16. Strain-stress curves from type C samples at 5%.

P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658 657

Fig. A17. Strain-stress curves from type C samples at 10%.

Fig. A18. Strain-stress curves from type C samples at 15%.

658 P. Muñoz et al. / Construction and Building Materials 224 (2019) 639–658

References [28] M. Iskander, N. Shrive, Fracture of brittle and quasi-brittle materials in

compression: a review of the current state of knowledge and a different
approach, Theor. Appl. Fract. Mech. 97 (2018) 250–257.
[1] A. Al-Fakih, B.S. Mohammed, M.S. Liew, E. Nikbakht, Incorporation of waste
[29] V. Letelier, J.M. Ortega, P. Muñoz, E. Tarela, G. Moriconi, Influence of waste
materials in the manufacture of masonry bricks: an update review, J. Build.
brick powder in the mechanical properties of recycled aggregate concrete,
Eng. 21 (2019) 37–54.
Sustainability 10 (2018) art. 1037.
[2] V. Bánhidi, L.A. Gömze, Improvement of insulation properties of conventional
[30] V. Letelier, E. Tarela, P. Muñoz, G. Moriconi, Combined effects of recycled
brick products, Mater. Sci. Forum 589 (2008) 1–6.
hydrated cement and recycled aggregates on the mechanical properties of
[3] L. Barbieri, F. Andreola, I. Lancellotti, R. Taurino, Management of agricultural
concrete, Constr. Build. Mater. 132 (2017) 365–375.
biomass wastes: preliminary study on characterization and valorisation in clay
[31] X. Lingling, G. Wei, W. Tao, Y. Nanru, Study on fired bricks with replacing clay
matrix bricks, Waste Manage. 33 (2013) 2307–2315.
by fly ash in high volume ratio, Constr. Build. Mater. 19 (3) (2005) 243–247.
[4] C. Bories, L. Aouba, E. Vedrenne, G. Vilarem, Fired clay bricks using agricultural
[32] H.A. López-Aguilar, E.A. Huerta-Reynoso, J.A. Gómez, J.A. Duarte-Moller, A.
biomass wastes: study and characterization, Constr. Build. Mater. 91 (2015)
Pérez-Hernández, Life Cycle Assessment of a Traditional Brick Manufacture
158–163.
Improvement, La Revista internacional de contaminación ambiental 35 (1)
[5] M.D. Bovea, U. Saura, J.L. Ferrero, J. Giner, Cradle-to-gate study of red clay for
(2019) 195–206.
use in the ceramic industry, Int. J. Life Cycle Assess. 12 (6) (2007) 439–447.
[33] J.P. Marques de Sá, Applied statistics using SPSS, STATISTICA, MATLAB and R,
[6] L. Ben-Bashat, D. Sherman, On dynamic surface instabilities of cracks in brittle
Springer, Berlin Heidelberg New York, 2007, pp. 111–222.
crystals, Int. J. Eng. Sci. 136 (2019) 78–91.
[34] M.A. Mendívil et al., Grapevine shoots for improving thermal properties of
[7] CTE (2012) Código Técnico de la Edificación. 5th Ed. Asociación Española de
structural fired clay bricks: new method of agricultural-waste valorization, J.
Normalización y Certificación (AENOR).
Mater. Civil Eng. 29 (8) (2017). Article Number: 04017074.
[8] I. Demir, Effect of organic residues addition on the technological properties of
[35] M.A. Mendívil, P. Muñoz, M.P. Morales, M.C. Juárez, Energy potential of vine
clay bricks, Waste Manage. 28 (2008) 622–627.
shoots in La Rioja (Spain) and their dependence on several viticultural factors,
[9] M. Devant, J.A. Cusidó, C. Soriano, Custom formulation of red ceramics with
Ciencia e Investigación Agraria 42 (3) (2015) 443–451.
clay, sewage sludge an forest waste, Appl. Clay Sci. 53 (2011) 669–675.
[36] M.A. Mendívil, P. Muñoz, P. Morales, M.C. Juárez, E. García-Escudero, Chemical
[10] C.N. Djangang et al., Properties of porous ceramics from two Cameroonian
characterization of pruned vine shoots from La Rioja (Spain) for obtaining solid
clays mixed with sawdust, Annales de Chimie-Science des Materiaux 35 (1)
bio-fuels, J. Renew. Sustain. Energy 5 (2013). Art. 033113.
(2010) 1–16.
[37] A. More, A. Tarade, A. Anant, Assessment of suitability of fly ash and rice husk
[11] D. Dominguez, P. Muñoz, L. Muñoz, Impact of using lightweight eco-bricks as
ash burnt clay bricks, Int. J. Sci. Res. Publ. 4 (7) (2014) 1–6.
enclosures for individual houses of one story on zones of high seismicity,
[38] P. Muñoz, M.C. Juárez, M.P. Morales, M.A. Mendívil, Improving the thermal
Materiales de Constr. 67 (328) (2017) e133.
transmittance of single-brick walls built of clay bricks lightened with paper
[12] V. Ducman, T. Kopar, Sawdust and paper-making sludge as pore-forming age
pulp, Energy Build. 59 (2013) 171–180.
for lightweight clay bricks, Indust. Ceram. 21 (2) (2001) 84–86.
[39] P. Muñoz, M.P. Morales, M.A. Mendívil, M.C. Juárez, L. Muñoz, Development of
[13] F. El Fgaier, Z. Lafhaj, C. Chapiseau, Improvement insulating properties of
better insulation bricks by adding mushroom compost waste, Energy Build. 80
ceramic materials by incorporating additives, J. Mater. Civ. Eng. 28 (8) (2016)
(2014) 17–22.
1–7.
[40] V. Muñoz, M.P. Morales, V. Letelier, M.-A. Mendívil, Fired clay bricks made by
[14] D. Eliche-Quesada, F.A. Corpas-Iglesias, L. Pérez-Villarejo, F.J. Iglesias-Godino,
adding wastes: assessment of the impact on physical, mechanical and thermal
Recycling of sawdust, spent earth from oil filtration, compost and marble
properties, Constr. Build. Mater. 125 (2016) 241–252.
residues for brick manufacturing, Constr. Build. Mater. 34 (2012) 275–284.
[41] P.M. Nigay, A. Nzihou, T. Cutard, The impact of the particle size distribution of
[15] EN 12664:2002. Thermal performance of building materials and products.
organic additives on the microstructure of a clay ceramic and its thermal and
Determination of thermal resistance by means of guarded hot plate and heat
mechanical properties, J. Mater. Civil Eng. 30 (4) (2018). art. 04018044.
flow meter methods. Dry and moist products of medium and low thermal
[42] PCD2K Ò, Servosis. Available at: https://www.servosis.com/fileuploads/
resistance.
products/sv-sl-software-de-control-v2-2.pdf (Last access on July 2019)
[16] EN 1745:2013. Masonry and masonry products – methods for determining
(Spanish).
design thermal values.
[43] N. Phonphuak, Effects of additive on the physical and thermal conductivity of
[17] EN 771-1:2011+A1:2016. Specification for masonry units. Part 1: clay masonry
fired clay brick, J. Chem. Sci. Technol. 2 (2) (2013) 95–99.
units.
[44] R. Prateep Na Talang, M. Pizzol, S. Sirivithayapakorn, Comparative life cycle
[18] EN 772-1:2011+A1:2016. Methods of test for masonry units. Part 1:
assessment of fired brick production in Thailand, Int. J. Life Cycle Assess. 22
Determination of compressive strength.
(11) (2017) 1875–1891.
[19] EN 772-3:1999 Methods of test for masonry units. Part 3: Determination of net
[45] R. Saiah, B. Perrin, L. Rigal, Improvement of thermal properties of fired clays by
volume and percentage of voids of clay masonry units by hydrostatic
introduction of vegetable matter, J. Build. Phys. 34 (2) (2010) 124–142.
weighing.
[46] J.F. Saldarriaga et al., Fast characterization of biomass fuels by
[20] EN 772-13:2001 Methods of test for masonry units. Part 13: determination of
thermogravimetric analysis, Fuel 140 (2015) 744–751.
net and gross dry density of masonry units (except for natural stone).
[47] N.A. Sarani, A.A. Kadir, A.S. Abdul-Rahim, A. Mohajerani, Properties and
[21] EN 772-16:2011 Methods of test for masonry units. Part 16: Determination of
environmental impact of the mosaic sludge incorporated into fired clay bricks,
dimensions.
Constr. Build. Mater. 183 (2018) 300–310.
[22] Eurostat (2012) Europe in figures. Eurostat yearbook 2012, available at:
[48] L.G. Schultz, Quantitative interpretation of mineralogical composition from X-
https://ec.europa.eu/eurostat/documents/3217494/5760825/KS-CD-12-001-
ray and chemical analysis data for the Pierre Shale Formation, U.S. Geol.
EN.PDF/032ab046-5604-42cf-b0a5-7e63e1cda8bb (Last access in June, 2019).
Survey. Prof. Paper 391-C (1964).
[23] Eurostat (2018) Waste Statistics, available at: https://ec.europa.eu/eurostat/
[49] ServosisÒ MES150. Available at: https://www.servosis.com/fileuploads/
statistics-explained/index.php/Waste_statistics (Last access in June, 2019).
products/ingles/13-sv-sl-mes-v2.pdf (Last access on July 2019).
[24] E. Galán, J.L. Marín Vivaldi, Caolines españoles. Geología, mineralogía y
[50] K.S. Shibib, H.I. Qatta, M.S. Hamza, Enhancement in thermal and mechanical
genesis: Parte 1, Boletín de la Sociedad Española de Cerámica y Vidrio 12 (2)
properties of bricks, Thermal Sci. 17 (4) (2013) 1119–1123.
(1973) 79–89 (Spanish).
[51] UNE 136029:2019 Clay masonry units. Test for efflorescence. (Spanish).
[25] J. García-Ten, M.J. Orts, A. Saburit, G. Silva, Thermal conductivity of traditional
[52] UNE 22950-3:1990 Mechanical Properties of Rocks. Strength Determination
ceramics. Part I: influence of bulk density and firing temperature, Ceram. Int.
Tests. Part 3: Modules of Elasticity (Young) and Poisson’s Ratio Determination.
36 (2010) 1951–1959.
(Spanish).
[26] J. García-Ten, G. Silva, V. Cantavella, M. Llorente, Utilización de materiales
[53] C.-H. Weng, D.-F. Lin, P.-C. Chiang, Utilization of sludge as brick materials, Adv.
aligerantes en la fabricación de bloques de termoarcilla, Conarquitectura 3
Environ. Res. 7 (2003) 679–685.
(2004) 65–72 (Spanish).
[54] D. Wu, J. Zhou, Y. Li, Mechanical strength of solid catalysts: recent
[27] R.E. Hughes, D.M. Moore, H.D. Glass, Qualitative and quantitative analysis in
developments and future prospects, AIChE J. 53 (2007) 2618–2629.
soils, in: J.E. Ammonette, L.W. Zelazny (Eds.), Quantitative Methods in Soil
[55] Z. Zhang, Y.C. Wong, A. Arulrajah, S. Horpibulsuk, A review of studies on bricks
Mineralogy, SSSA Miscellaneous publication. Soil Science Society of America,
using alternative materials and approaches, Constr. Build. Mater. 188 (2018)
Inc., Madison, Wisconsin, USA, 1994, pp. 330–359.
1101–1118.