Journal of the European Ceramic Society 24 (2004) 31773185

www.elsevier.com/locate/jeurceramsoc

Incorporation of granite cutting sludge in industrial porcelain

tile formulations
P. Torres, H.R. Fernandes, S. Agathopoulos, D.U. Tulyaganov, J.M.F. Ferreira*
Department of Ceramics and Glass Engineering, University of Aveiro, CICECO, 3810-193 Aveiro, Portugal
Received 27 June 2003; received in revised form 21 October 2003; accepted 25 October 2003

Abstract
Granite wastes in the form of sludge, obtained from granite cutting industry, were incorporated in the batch formulations of
porcelain tiles. The maximum possible substitution of sludge for feldspar was investigated. Samples of dierent formulations, in
the form of pellets or extruded bars, were produced at both laboratory and pilot-plant scales and characterized throughout all the
stages of the production process. The experimental results and their theoretical interpretation show that suitable incorporation of
granite sludge can result in porcelain tiles with superior properties, in terms of water absorption (0.07%) and bending strength
(> 50 MPa). Sludge incorporation had negligible eect on density, shrinkage and plasticity during all stages of tile-production
process, anticipating no modications in the industrial production line.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Extrusion; Firing; Sludge; Porcelain; Traditional ceramics; Waste materials

1. Introduction
Traditional ceramics, such as bricks, or roof and oor
tiles, generally feature high heterogeneity due to the
wide range of the composition of natural clays used as
raw materials in their fabrication. Thereby, there is
a high tolerance for incorporating large amounts
of suitable wastes as raw materials.1
5 This fact attracts
further importance since ceramic industry, which is
classied as heavy industry, consumes huge amounts of
diminishing mineral resources.6
8 Recycling of wastes
has unambiguously benecial environmental and economical impact,9
13 while high temperature ring can
turn wastes, which contain hazardous components, into
inert and safe for the health products.4,14
Granite cutting industry produces large amounts of
solid wastes worldwide, which are expected to increase
owing to the fact that the world production of granite
industry has been increasing annually at a rate of 6% in
the recent years. The wastes of this industrial activity
can reach even 20 or 25 wt.% of the raw granite.15 Evidently, deposition of huge amounts of granite sludge
wastes creates necrotic conditions for ora and fauna
* Corresponding author. Tel. +351-234-370242; fax: +351-234425300.
E-mail address: jmf@cv.ua.pt (J.M.F. Ferreira).
0955-2219/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jeurceramsoc.2003.10.039

while, after drying, ne particles can be deposited in the

lungs of mammals via breath.
Granite sludge is a mixture of debris residue of cut
granite rocks with wear remains of cutting steel blades,
abrasive metallic shot, and hard materials from the
polishing bricks. In the common granite cutting practice,
the abrasive metallic shot is dispersed in Ca(OH)2 aqueous
slurry for cooling. This slurry is continuously pumped
and wets all around the granite block slits. Hence, granite
sludge contains mainly SiO2, Al2O3, Fe2O3 and CaO.
In the light of recycling, this work investigates the
potential prot of incorporation of granite wastes as
raw materials in porcelain-tile industrial formulations,
particularly with regards to the following environmental
and economical aspects:
(a). Maximization of the use of granite-sludge waste
in porcelain tile production, which would reduce both
the environmental impact and the production cost of
porcelain tiles.
(b). Maintenance and if possible improvement of the
properties of the produced porcelain tiles. There is
still no ISO-standard for the extruded porcelain tiles,
which would more interest the present work. Therefore, the produced materials were evaluated in the
light of the ISO standard 13006, Group B-I, which is
directed to the porcelain tiles made by dry-powder

3178

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

pressing, whereby water absorption must be less than

0.5% and three point bending strength higher than 35
MPa.16 Meanwhile, sludge incorporation must not jeopardize materials properties at the intermediate stages
of production process. Otherwise, extended and costly
modications in the industrial production line would be
necessary (e.g. size and speed of caring belts, storing and
transportation facilities from one unit to another etc.).
In this work, the granite sludge (hereafter denoted as
GS) was kindly oered by the industry Incoveca
(Viseu, Portugal). For comparison purposes, the raw
material of commercial porcelain with the trade name
Keratec1 (K) was used as reference material, made in
the porcelain-tile industry Keratec (Vagos, Portugal).
This industry kindly oered the raw materials and their
production line facilities for this work.
The raw materials used in the production of Keratec1
porcelain tiles are feldspar (F) and two types of clays,
denoted as clay-E (E), which aims to provide plasticity
to the green body, and clay-A (A), which provides
strength. This work aimed to maximize the replacement
of the expensive feldspar with granite sludge. Samples of
dierent compositions in the form of pellets and extruded bars were prepared. The crucial properties were
monitored in all the stages of production line until the
nal product, with respect to the aforementioned aspect
(b). The properties of the resulting materials are evaluated in the light of the requirements of the regulations
of the ISO-13006 as well as the inuence of granite
sludge in the mechanism (i.e. phase transformation) of
porcelain formation.
During the past few decades, the replacement of feldspar gained increasing acceptance in industry.17
23
Feldspar is a ux that aims at lowering the vitrication
temperature of ceramic bodies during ring and aiding
the formation of glassy phase. The composition (in
wt.%) of a typical commercial feldspar is  68% SiO2,
19% Al2O3, 7% Na2O, 4% K2O, 1% CaO and 0.08%
Fe2O3. Pure potassium feldspar typically contains
510% K2O.23,24
In the literature, there are few reports dealing thoroughly with the involvement of granite sludge wastes in
porcelain production. Ferreira et al.1,11 have shown that
the physico-chemical characteristics of granite sludge
match well the requirements needed in brick and roof tile
formulations. Thus, their incorporation results in negligible changes in the properties of the nal products.
Menezes et al. have extensively reported the production
of bricks and tiles using granite sludge.5

2. Materials and experimental procedure

The experimental procedure was as follows. The asreceived powder of GS was mixed with water in order to

make possible vibratory sieving (90 mm). The sieved

mud was dried (110  C) and then disagglomerated in a
hammer-mill. Fourteen dierent formulations were prepared by mixing of the collected GS powder with the asreceived powders of clays A and E and feldspar (F)
(Table 1). After this point, commercial powder from
Keratec1 (K), whose composition is upon trade protection, was also similarly used for comparison purposes. The mixtures were homogenized by addition of
water and ultrasonic agitation, dried (110  C), and disagglomerated by gently grinding in an agate mortar.
Pellets (1 20 mm, thickness 2 mm), prepared by cold
uniaxial pressing at 47 MPa, were sintered in a laboratory furnace at 1140, 1180, and 1200  C for 1 h in air
(heating-cooling rate 5 /min). Water absorption was
measured according to the ISO-standard 10545-3, i.e.
weight gain of dried pellets after immersion into boiling
water for 2 h, cooling for 3 h and sweeping of their
surface with a wet towel.25 The best formulations were
selected according to the following criteria: (a) Water
absorption < 0.5%, (b) the maximum incorporation of
GS, and (c) suitable colour for fabricating commercial
porcelain tiles.
Then, the production was upgraded to pilot-plant
scale. Batches of  6 kg of the selected formulations
(and Keratec1 powder) were mixed with water ( 1619
wt.%) and homogenized using a mixer. The extrusion
capability (i.e. plasticity) of the resulting pastes was
evaluated with the Pfeerkorn test (i.e. impact eect on
a cylinder of the paste with standard dimensions).
Extruded cylinders (1  10 mm, length 120 mm) were
obtained according to industrial standards (laboratory
extruder NR Burton on Trent, Rawdon Ltd., Moira,
UK), dried (room temperature-24h, 110  C-24 h), and
red in the industrial kiln (110 m length, total ring
cycle 90 min). The actual sintering temperature in the
kiln was that one given by process temperature control

Table 1
The proportion of the blends for the 14 tested formulations (in wt.%)
Formulation

Clay A

Clay E

1
2
3
4
5
6
7
8
9
10
11
12
13
14

20
30
25
25
20
25
15
30
50% clay
(20% A+30% E)

60% clay
(24% A+36% E)

Feldspar (F)

Sludge (GS)

0
0
5
5
30
25
20
15
10
5
20
15
10
5

50% GS

20
25
30
35
40
45
20
25
30
35

3179

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

rings (TPTCR) (Ferro Electronic Materials B.V, Uden,

the Netherlands), whose accuracy (tested in the laboratory furnace with a Pt/PtRh thermocouple) was  3 .
Thus, the apparent sintering temperature in the kiln was
xed at Tk=1210  C but the actual sintering temperature was TPTCR1185  C.
Several types of characterization tests were conducted
during the entire process. The following techniques,
together with the aforementioned water absorption and
Pfeerkorn tests, were employed. Density of powders
and bulk samples was determined by Archimedes
method with water- and Hg-immersion, respectively.
Particle size distribution was determined by both conventional sieving and light scattering (Coulter LS 230,
UK, Fraunhofer optical model). X-ray diraction analysis (XRD, Rigaku Geigerex D/Mac, C Series, Cu Ka
radiation, Japan). X-ray uorescence analysis (XRF,
Model XRF-1700, Shimadzu Corp., Tokyo, Japan).
Dierential (DTA) and gravimetric (TG) thermal analysis (Labsys Setaram TG-DTA/DSC, France, heating
rate 5 /min, 1 atm owing N2). Dilatometer thermal
analysis (Bahr Thermo Analyse DIL 801 L, Germany,
heating rate 10 /min). Scanning electron microscopy
(SEM, Hitachi S-4100, 25 kV acceleration voltage,
Tokyo, Japan). The presenting results for the threepoint bending strength (Shimadzu Autograph AG 25
TA, 0.5 mm/min displacement) are the average of more
than 10 tested cylinders. Retraction and weight loss
measurements during drying and sintering were also
carried out.
In order to shed light to the path of phase transformation during the production processing, quenching
experiments were carried out by rapid immersion of
pellets, made of the selected formulations and heated at
5 /min rate and soaked at 1200  C for 1 h, into cold
water. These experiments gave also information for the
thermal shock resistance.

3. Results
3.1. Raw materials
In order to understand the inuence of the incorporation of granite sludge as a raw material for porcelain-tile production, the rst stage of this investigation
was directed to the complete characterization of all raw
materials used in the formulations, A, E, F, and GS, as
summarized in Table 2 and Figs. 14.
The powders of the raw materials had similar density
(clay-A 2.65, clay-E 2.50, feldspar 2.54, and granite sludge
2.51 gcm
3). Except A, the size of particles was actually
< 100 mm, following an almost Gaussian distribution of a
single peak, centred between 5 and 20 mm (Fig. 1).
XRD spectra (Fig. 2) showed well crystallized phases.
Quartz (SiO2), albite (sodium feldspar, NaAlSi3O8),

Table 2
Chemical composition of the raw materials (in wt.%)
Oxide

Clay A

Clay E

Feldspar (F)

Sludge (GS)

SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
TiO2
MnO
P2O5

64.69
27.91
2.39
0.19
0.58
0.80
2.43
0.93
0.01
0.07

64.87
29.79
1.75
0.10
0.34
0.65
1.70
0.72
0.01
0.07

72.29
16.72
0.31
0.71
0.20
5.79
3.22
0.02
0.02
0.72

71.65
14.25
2.86
1.83
0.86
3.72
4.43
0.24
0.03
0.13

Fig. 1. Particle size distribution of the as-received A, E, F, and the

sieved GS powders, together with the powders Keratec1 (K) and
formulation 6 (see Table 1).

illite ((K,H3O)Al2Si3AlO10(OH)2) and microcline

(potassium feldspar, KAlSi3O8) were identied in all
powders. Peaks of kaolinite (Al2Si2O5(OH)4) were
detected in GS and both clays A and E. Magnetite
(Fe2+Fe3+
2 O4) was found in GS, while calcite (CaCO3)
in the clays.26 According to Table 2, SiO2 dominates in
the sludge (GS), following by Al2O3, while there are
high amounts of sodium and potassium oxides, together
with calcium and iron oxides too. Evidently, the high
CaO-content in GS is mostly due to the Ca(OH)2 slurry
used to carry the abrasive metallic shot. At this point, it
should be noticed that the composition of F matches the
composition of pegmatite (i.e. the rock formed by feldspar and quartz),23 but for simplicity reasons, in this
article this component will keep being referred as feldspar (F), since this was the trade name used in industry.
The thermal analyses actually sort out two groups,
GS and F, and clay A and E (Fig. 3). The former group
showed higher stability over temperate whereas the
clays showed a strong endothermic peak and signicant
weight loss at 530550  C. In the case of clays, careful
observation might suggest two more weak exothermic

3180

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

Fig. 3. Thermal analysis, (a) DTA, (b) TG, for the A, E, F, and GS.

Fig. 2. XRD spectra of the A, E, F, and GS (the position of the peaks

of the phases are also marked26).

peaks, at 950980  C and 11201150  C (for clay A).

For the GS, a weak endothermic peak might be suggested at 570580  C, probably due to the transformation of a- to b-quartz.27 The clays also underwent a
slight weight loss at  600700  C.
The same grouping (i.e. GSF and AE) seemingly
reects the water absorption capability of pellets sintered between 1140 and 1200  C for 1 h, where the clays
absorb 510 times more water than F or GS (Fig. 4).
3.2. Selection
Fig. 5 depicts the water absorption of pellets made of
the formulations listed in Table 1, together with the
reference composition of Keratec1 (K) sintered at 1140,
1180, and 1200  C for 1 h. Evidently, water absorption
decreases up to 1180  C due to densication process,

Fig. 4. Water absorption for pellets made of the A, E, F, and GS,

sintered at dierent temperatures for 1 h.

and then increases again likely due to an over-ring

eect. In general, all samples satised the three selection
criteria mentioned in the materials and experimental
procedure section, especially those sintered at higher
temperatures. Therefore, a forth selection criterion was
introduced, basing on the lower water absorption level
achieved at the lowest sintering temperature. Accordingly (Fig. 5), the selected formulations were those with
the numbers 3, 5, 6, 7, 8 and 9 (Table 1).

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

Fig. 5. Water absorption of the pellets with the formulations listed in

Table 1 and Keratec1 (K), red at dierent temperatures for 1 h.

3.3. Extruded bars

The particle size distribution curves of the blends
made of the selected formulations resemble those of the
raw materials (Fig. 1, all the selected formulations
showed similar distribution to that of formulation 6). In
accordance to the plasticity (Pfeerkorn) tests (Fig. 6),
extruded cylinders were perfectly obtained from the
moistened pastes, whose water content is shown in
Table 3.
It should be noticed that muscovite ((K,Na)
(Al,Mg,Fe)2(Si3.1Al0.9)O10(OH)2)26 was identied in the
green extruded cylinders (including Keratec1) (Fig. 7),
though it was not identied in the raw materials (Fig. 2).
Therefore, it is suggested that the powders of the raw
materials (mostly the clays) contained randomly orientated
muscovite but the extrusion aligned the muscovite layers
along the extrusion direction, intensifying mainly the
peak at 8.9 , which according to the JCPDS card
07-0042 is assigned to the plane 003. The X-ray diraction of the planes corresponding to the 006, 101 or 100,
112, 115, and 1111 planes might be also suggested

Fig. 6. Plasticity curves for the six selected formulations and Keratec1
(K) (Pfeerkorn test).

3181

(Fig. 7), while the second principal peak of muscovite

(26.7 ) is hardly distinguished from the principal peak
of quartz (26.6 ). In layer structured materials, the
alignment of crystallites has been reported in tape casting or extrusion.28
The extruded cylinders perfectly survived drying and
ring. Table 3 summarizes the values of the properties
of the extruded cylinders and particularly the retraction
and weight loss during drying and sintering, the bending
strength and the density of both green and red bodies
and the water absorption of the sintered ceramics. Evidently, the investigated formulations had superior
properties mainly with regards to water absorption as
well as bending strength. On the other hand, except
formulation 3, sludge incorporation had negligible eect
on the other properties with respect to the commercial
porcelain (K). It should be also underlined that there
was an eort for approaching the optimum sintering
temperature (i.e. 1180  C, which showed the lowest
water absorption values, Fig. 5) also in the industrial
kiln. Therefore, since it was not advisable to change for
long the industrial sintering schedule for merely an
experimental testing, the best approach of this temperature was TPTCR1185  C.
Fig. 7 reveals the phase transformation taken place
during ring and cooling for the formulations 6 and K,
(all the other tested formulations showed similar XRD
spectra to that of formulation 6).26 In general, ring
resulted in the formation of mullite and glassy phase,
with some non-dissolved quartz still remaining in the
sintered bodies. Traces of un-reacted albite were
scarcely identied (not shown here). The XRD patterns of the quenched samples (not shown here)
revealed an intermediate regime (i.e. increasing glassy
phase and presence of traces of the crystalline phases
of the green body together with mullite), evidencing
that the nal crystalline structure is established after
the slow cooling.
Fig. 8 plots the results of the thermal analyses for the
formulation 3 and K. In a qualitative basis, all the selected formulations showed similar behaviour but the clayricher formulations underwent more intensive weight
loss (not shown here). The important temperatures in
both DTA and TG curves coincide with the temperatures mentioned in the presentation of Fig. 3. In general,
the investigated formulations (Fig. 8) resemble rather
the behaviour of the clays than the GS and the F
(Fig. 3). The dilatometry curves of the green samples
(not shown here) appear practically superimposed,
showing a small expansion up to  1000  C where they
start to shrink, following not exactly the same but similar paths afterwards.
Fig. 9 shows characteristic microstructures of red
bars observed at the fracture surface after three point
bending tests. All of them look very similar, exhibiting
good homogeneity when observed at low magnication,

3182

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

Table 3
Values (and standard deviation) of properties of the selected formulations in comparison with the commercial Keratec1 (K) (for samples
identication see Table 1)
Property

Green bodies
Water content (wt.%)
Retraction during drying (%)
Three-point bending strength (MPa)
Density (gcm
3)

Formulation
3

18.80
5.04 0.23
3.22 0.18
1.79 0.06

17.50
3.88 0.14
3.53 0.27
1.81 0.01

18.88
5.34 0.25
3.43 0.15
1.77  <0.01

16.79
3.560.21
2.750.16
1.800.01

16.90
3.870.16
3.260.16
1.830.01

16.82
2.920.14
3.280.23
1.820.01

16.91
3.200.12
3.640.08
1.85 <0.01

0.13 <0.01
5.960.11
5.400.04
59.212.41
2.23 <0.01

0.210.08
6.570.15
4.760.03
60.263.29
2.250.02

0.270.01
6.780.16
4.620.03
59.792.54
2.27 <0.01

1.310.10
6.260.22
5.140.06
47.843.05
2.27 <0.01

Samples red in the industrial kiln (Tk=1210  C, TPTCR1185  C, 90 min)

Water absorption (%)
0.08  <0.01
0.26 0.01
0.07  <0.01
Retraction during sintering (%)
7.13 0.14
5.56 0.12
6.45 0.20
Weight loss during sintering (%)
4.64 0.03
5.30 0.03
5.36 0.05
Three-point bending strength (MPa) 58.262.85
60.021.92 54.792.60
Density (gcm
3)
2.26 0.01
2.22 0.01
2.20 0.01

Fig. 8. Thermal analyses (DTA and TG) of the formulation 3 and

Keratec1 (K). Qualitatively, all the selected formulations showed
similar curves, but the clay-richer formulations underwent more
intensive weight loss (not shown here).

Finally, it is worthy noting that all the selected formulations showed remarkable thermal shock resistance
since they survived quenching and not even a single crack
was observed at the surface of the quenched pellets.

4. Discussion

Fig. 7. Crystalline structure (XRD26) of dry green body and sintered

samples of the formulation 6 and Keratec1 (K). All the selected formulations showed very similar patters.

while, at higher magnication (insets), there were

observed spherical pores, typically formed in the glassy
phase. These porous microstructures probably reect a
somewhat over-ring eect, which might cause increasing water absorption (Fig. 5).

This work is actually directed to the eld of designing

new ceramic compositions, which can meet certain
requirements. Among the wide family of commercial
and traditional ceramics, porcelain tiles must satisfy the
highest and most strict quality standards, which apparently depend on the high quality of the raw materials
and the optimization of the processing parameters. This
is however a dicult task due to the use of natural raw
materials. Granite is a magmatic and one of the most
abundant natural rocks and approximately contains 66
wt.% quartz, (which exposes its acidic behaviour), 26%

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

3183

Fig. 9. Characteristic SEM microstructures of extruded cylinders of (a) the formulation 8 and (b) Keratec1 (K), observed at the fracture surfaces
after three point bending test at low and high magnication (insets). All the selected formulations showed similar microstructures.

feldspar, and 8% mica, together with smaller amounts

of clay- and iron-containing minerals. In this particular
work, the characteristics of the incorporated granite
wastes likely depend on the specic source. For example, the iron content in granite sludge can increase due
to residues of cutting steel blades and abrasive metallic
shot. Moreover, CaO content may be easily changed if
sludge from granite is mixed in any proportion with
sludge from marble rocks.
Apparently, each one of the four raw materials mixed
in the tested batches (Table 1) contains dierent amount
of uxes and thus it has a particular contribution to the
sintering process. Hence, the optimization of tile formulation should take into account both the chemical
composition and the crystalline structure of the raw
materials. The proportion of the raw materials in the
batch is another important factor. This is clear because
of the following fact. This work aimed to maximize the
substitution of sludge for the expensive feldspar. However, the best results were achieved when both feldspar
and sludge were present, whereas absolute substitution
(samples 1 and 2 of Table 1) resulted in materials with
poorer properties (Fig. 5).
In order to nd a relationship between the composition and the sintering ability of the tested batches,
Table 4 presents the composition (in mol%) of the main
(K2O, Na2O) and the auxiliary uxes (CaO, MgO,
Fe2O3) in F and GS. The total molar fraction for K2O,
Na2O, CaO, and MgO uxes is 9.57% for the F and
10.62% for the GS, while GS is also richer than F in
another auxiliary ux, i.e. Fe2O3 (probably derived
from the wearing of metallic cutting steel blades and
abrasive metallic shot). Hence, substitution of GS for
Table 4
Content of uxes in the feldspar (F) and the granite sludge (GS) (in
mol%)
Material

CaO

MgO

Na2O

K2O

Fe2O3

F
GS

0.84
2.15

0.33
1.41

6.15
3.96

2.25
3.10

0.13
1.18

F can improve the sintering behaviour because the

higher content of uxes evidently decreases the viscosity of the glassy phase and accelerates the sintering
process of ceramic body.29 On the other hand, the
DTA curve of F seemingly ends with a plateau at
 1150  C probably assigned to the melting of feldspar
(Fig. 3a). Thus, feldspar melting should precede the
melting of granite sludge. Hence, the optimization of
the composition requires the presence of both the
sludge and the feldspar.
In a qualitative basis, the results of the thermal
analyses and the XRDs showed that the commercial
Keratec1 paste, the clays A and E, and the six selected
formulations behave in the same manner (Figs. 3, 7 and
8). Therefore, the ring process should comprise the
following stages. Since the clays are constituted of very
ne plates with lateral dimensions of 0.13 mm and
thickness 0.52 mm,22,23,30 the water adsorbed at the high
specic surface area clay particles is eliminated until
200  C. Oxidation of organic components takes place
until 700  C, depending on their chemical nature. At
573  C, quartz is transformed to its high temperature b
form. Between  450 and  600  C, kaolinite looses the
chemical bonded water and turns into metakaolinite.
The slight weight loss at  600700  C might be ascribed
to illite and muscovite minerals.27 At  950  C, the lattice
is totally collapsed, silica is eliminated and spinel is
formed (i.e. 2(Al2O3.2SiO2)!2Al2O3.3SiO2+SiO2).
Spinel is transformed to mullite at  1150  C (i.e.
3(2Al2O3.3SiO2)!2(3Al2O3.2SiO2)+5SiO2). It has been
reported, however, that the formation of spinel depends
on the presence of other admixtures.23,29
Theoretically, liquid phase should form at the eutectic
temperature of potassium feldspar (KAlSi3O8) and
quartz (985  C).23,29 Amorphous silica, formed during
the thermal decomposition of kaolinite, reacts with
potassium feldspar. At high temperatures, quartz is dissolved in the liquid phase. In this work, there was no
addition of quartz in the batches during the process but
quartz was intrinsically incorporated in the clays, the
granite sludge, and the feldspar (Fig. 2). For pure

3184

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

potassium feldspar, complete melting should take place

at its melting temperature  1150  C. It is also mentioned that the eutectic point of sodium feldspar and
silica is  1062  C, while the melting point of pure albite
is  1118  C.27
As far as the fabrication and commercialization of porcelain tiles are concerned with the incorporation of granite
sludge, this work has shown that optimal proportion of
the raw materials in the batch and ring at  1180  C
potentially result in superior porcelain tiles. In particular, the powder of granite sludge featured a symmetrical and wide particle size distribution (between 1 and
100 mm, Fig. 1), which facilitates high compaction of the
nal product while mixing in suitable proportions with
the other raw materials.31 The incorporation of the
sludge had negligible eect in density, shrinkage and
plasticity of materials during all stages of production
process. Therefore, expensive modications in the
industrial production line are omitted. On the other
hand, the presence of granite sludge considerably lowered the water absorption (< 0.25%) and rmly
increased of mechanical strength (> 55 MPa), comparing to the commercial porcelain (Table 3).
These values make further sense if compared with the
earlier work reported by Menezes et al.5 In that work,
the incorporation of granite sludge resulted in bricks and
tiles with relatively high water absorption values and
poorer exural strength than the produced porcelains in
this work (i.e. 817% and 823 MPa upon ring at
1150  C; 28% and 1936 MPa upon ring at 1175  C;
0.43.0% and 2436 MPa upon ring at 1200  C). The
superior behaviour of the investigated formulations is
also evidence by the small dierences in water absorption shown for several selected formulations sintered
between 1140 and 1200  C (Fig. 5). In ceramic industry
for tile production, this is very important because the
crucial goal is the secure broadening of sintering range,
with regard to the maintenance of the quality of the
product.
At this point it must be underlined that Keratec1
commercial porcelains exhibit water absorption denitely lower than 0.5% because they obviously satisfy
the ISO regulations. The higher amount of water
absorption measured in this work with the K samples is
due to the dierent experimental procedure (i.e. mixing,
drying and sintering conditions), comparing to the strict
industrial fabrication. The experimental conditions followed in this work purposefully increased the gap
between the properties of the tested formulations and the
commercial one and thus facilitated their comparison.
The construction sector is the best target market (i.e.
bricks, wall and oor tiles) for absorbing and consuming products made of recycling solid wastes which are
produced at enormous amount, such as granite or other
cutting rock wastes.1,5 In the particular case of granite,
beyond reducing the harmful environmental impact of

deposition of wastes and lowering the production cost

of porcelain, it is worthy noting that granite sludge
contains no hazardous chemicals but true natural
materials, anticipating, therefore, safe for the health an
environmentally friendly recycling products.

5. Conclusions

(a). The incorporation of granite sludge wastes in

porcelain tile production anticipates safe for the
health an environmentally friendly recycling products.
(b.) Provided that there is optimal proportion of the
raw materials in the batch and ring temperature is
 1180  C, the resulting extruded porcelain items
satisfy the closest ISO standard (13006, Group B-I),
and can exhibit superior properties comparing to
the commercial porcelains, with regards to water
absorption, exural strength, and widening of sintering range.
(c). The incorporation of granite sludge has negligible
eect on the other properties during the entire process, anticipating no costly modications in the
industrial production line.

Acknowledgements
The nancial support of the Project PRAI-Centro,
Caracterizacao e reciclagem de lamas derivadas do
corte de rochas naturais e ornamentais, Programa
Regional de Accoes Inovadoras do Centro de Portugal
is acknowledged. S.A. acknowledges the nancial
support of the Portuguese Foundation of Science and
Technology under the PRAXIS-XXI grant SFRH/
BPD/1619/2000 and D.T. CICECO for the post-doctoral grant.

References
1. Ferreira, J. M. F., Torres, P. M. C., Silva, M. S. and Labrincha,
J. A., Recycling of granite sludges in brick-type and oor tile-type
ceramic formulations. Euroceram News, 2003, 14, 15 http://
www.euroceram.org/en/news/news_14_1.cfm).
2. Ribeiro, M. J., Ferreira, J. M. F. and Labrincha, J. A., Incorporation of clay-based ceramic formulations containing dierent
solid wastes. Euroceram News, 2003, 14, 14 http://www.
euroceram.org/en/news/news_14_2.cfm.
3. Knirsch, M., Penschke, A., Krebs, S., Rub, W., Mortel, H.,
Mayer, W. A. and Mayer-Pittro, R., Application of brewery
wastes in the production of bricks. Tile & Brick Int., 1998, 14, 93
101.
4. Ferreira, J. M. F., Alves, H. M. and Mendonca, A. M., Inertization of galvanic sludges by its incorporation in ceramic products.
Boletin de la Sociedad Espanola de Ceramica y Vidrio, 1999, 38,
127131.
5. Menezes, R. R., Ferreira, H. S., Neves, G. A. and Ferreira, H. C.,

P. Torres et al. / Journal of the European Ceramic Society 24 (2004) 31773185

6.

7.

8.
9.

10.

11.

12.

13.
14.

15.

The use of granite wastes as ceramic raw materials. Ceramica,

2002, 48, 92101 (in Portuguese).
Martins, O. R., Grandes linhas do comportamento mundial da
industria e do comercio das rochas ornamentais em 1996. Boletim
de Minas, 1998, 35, 123149 (in Portuguese).
Vieira, T. and Sobreiro, M. J., Rochas ornamentais e industriais
Portuguesas. (elementos estatsticos de 1997). Boletim de Minas,
1999, 36, 375390 (in Portuguese).
Moura, A. C., A pedra natural ornamental em Portugalnota
breve. Boletim de Minas, 2001, 38, 161177 (in Portuguese).
Boccaccini, A. R., Schawohl, J., Kern, H., Schunck, B., Rincon,
J. M. and Romero, M., Sintered glass ceramics from municipal
incinerator y ash. Glass Technol., 2000, 41, 99105.
Gorokhovsky, A., Gorokhovsky, V. and Mescheryakov, D.,
Glass-ceramic materials based on complex utilization of industrial wastes: structural features. Glass Sci. Technol., 2000, 73,
374377.
Ferreira, J. M. F., Torres, P. M. C., Silva, M. S. and Labrincha
J. A., Recycling of sludges generated from natural stones cutting
processes in ceramic formulations. In Proceedings of the Recycling and Waste Treatment in Mineral and Metal Processing:
Technical and Economic Aspects, ed. B. Bjorkman, C. Samuelsson, and J.O. Wikstrom. Lulea University of Technology, LuleaSweden, 2002, pp. 389396.
Ferreira, J. M. F., Guedes, P. J. S. and Faim, P. F., Recycling of
industrial residues: the best strategy for waste management. AlAzhar Bulletin of Science, 2003, March, 293305.
Ferreira, J. M. F. and Olhero, S. M., Al-rich sludge treatments
towards recycling. J. Eur. Ceram. Soc., 2002, 22, 22432249.
Ferreira, J. M. F., Olhero, S., Lemos, A. and Xavier V., Suitability of the ceramic matrixes for recycling industrial inorganic wastes. In Proceedings of the Recycling and Waste
Treatment in Mineral and Metal Processing: Technical and
Economic Aspects, ed. B. Bjorkman, C. Samuelsson and J.O.
Wikstrom. Lulea University of Technology, Lulea-Sweden,
2002, pp. 369375.
Fernandes, H. R., Torres, P., Agathopoulos, S., Tulyaganov, D.
and Ferreira, J. M. F., Utilization of solid wastes from granite

16.

17.

18.
19.
20.

21.
22.
23.
24.
25.
26.

27.
28.
29.
30.
31.

3185

cutting processing in porcelain industry. Al-Azhar Bulletin of

Science, 2003, March, 3343.
Ceramic TilesDenition, classication, characteristics and
marking. International Organization for Standardization, ISO
13006, 1998.
Sanchez, E., Garcia, J., Sanz, V. and Ochandio, E., Raw material
selection criteria for the production of oor and wall tiles. Tile &
Brick, 1990, 6, 1521.
Sandoval, F. and Ibanez, A., Fast ring wollastonite-based wall
tile bodies. Am. Ceram. Soc. Bull., 1999, 78, 7275.
Taspinar, B. and Ay, N., Use of perlite in wall tile production.
Am. Ceram. Soc. Bull., 1999, 78, 8687.
Rokhvarger, A. E. and Khizh, A. B., Conveyer technology for
tile and other thin-walled ceramics. Am. Ceram. Soc. Bull., 1998,
77, 5975.
Bozadgiev, L., Porcelain on bulgarite basis. Interceram, 2000, 49,
810.
Ryan, W., Properties of Ceramic Raw Materials, 2nd ed. Pergamon press, New York, 1978.
Avgustinik, A. I., Ceramics, 2nd ed. Stroiizdat, Leningrad,
Russia, 1975 (in Russian).
Annual Minerals Review, Am. Ceram. Bull., 1999, 78, 131.
Ceramic TilesPart 3. International Organization for Standardization, ISO 10545-3, 1995.
JCPDS cards: quartz: 46-1045; microcline: 19-0932; albite: 090466; ilite: 26-0911; kaolinite: 29-1488; calcite: 05-0586; magnetite: 19-0629; mullite: 15-0776; muscovite: 07-0042.
Jouenne, C. A., Traite de Ceramiques et Materiaux Mineraux.
Editions Septima, Paris, 1975.
Grim, R. E., Applied Clay Mineralogy. McGraw-Hill, New York,
1962.
Budnikov, P. P., The Technology of Ceramics and Refractories.
MIT press, Cambridge, Mass, 1972.
Carty, W. M., The colloidal nature of kaolinite. Am. Ceram. Soc.
Bull., 1999, 78, 7279.
Ortega, F. S., Pileggi, R. G., Sepulveda, P. and Pandolfelli, V. C.,
Optimizing particle packing in powder consolidation. J. Am.
Ceram. Soc., 1999, 78, 106111.