Ceramics International xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Ceramics International
journal homepage: www.elsevier.com/locate/ceramint

The effect of particle size of body components on the processing parameters

of semi transparent porcelain
⁎
Fazilet Güngöra, , Nuran Ayb
a
Kütahya Porselen Sanayi A.Ş, Kütahya,Turkey
b
Anadolu Üniversitesi, Mühendislik Mimarlık Fakültesi, Eskisehir, Turkey

A R T I C LE I N FO A B S T R A C T

Keywords: The effect of the particle size of quartz and K- feldspar on the sintering behaviour and technical properties of
Semi Transparent Porcelain semi transparent porcelain were investigated. This study was implemented by “the Factorial Experimental
Sintering Design Method” to determine the effect of factors on semi vitreous porcelain production. The sintering behaviour
Particle Size of the samples were examined by using an optical dilatometer. The phases and their respective amounts in the
Ceramic Process Parameters
microstructure were determined by way of Rietvelt X- ray diffraction (XRD) technique. Furthermore, the values
Experimental Design
such as shrinkage, water absorption, bulk density, porosity, thermal expansion, light transmission measurements
Amorphous phase
were taken on the samples and the microstructures were studied by SEM. It was observed that as the particle size
of quartz and K- feldspar decreased; the viscosity of the liquid phase fell down, the relative amount of secondary
mullite crystals increased and the large pores were removed among the particles and the densification rate
increased. Data were analysed by Minitab 13.20 software, and assessed in relation to the amount of glassy phase
and its viscosity. The effect of K- feldspar particle size on the densification of porcelain is less than the change of
heat treatment and quartz particle size.

1. Introduction
Porcelain is a glassy bonded semi-transparent material with a white
colour and low water absorption. In general, a porcelain body is typi-
cally made of clay, feldspar and quartz along with other constituents to
promote the desired characteristics [1,2]. Because of the complex in-
terplay between raw materials, processing routes and kinetics of the
firing process, porcelains represent some of the most complicated
ceramic systems. Because of the complex reactions, obtained micro-
structure controls the final properties of porcelain such as bulk density,
porosity, water absorption, strength, deformation, transparency and
colour [3–5].
The microstructure of porcelain consists predominantly of mullite,
undissolved quartz, glass, and even, rarely, residual feldspar and pores
Mullite starts to form between ~ 1050 and 1200 °C, and its transfor-
mation is completed in the temperature range of 1200–1400 °C under
conventional firing conditions [1, 2, and 6]. The mullite level is pro-
posed to be dependent on the solubility of Al2O3 in the glass phase,
measured to be 1.19 ( ± 0.1) moles of alumina per mole of flux (R2O
+ RO) [7,8]. Dissolution rate of quartz in glass matrix is a function of
firing cycle but relatively insignificant below 1200 °C. If the quartz
grains are very fine, they are highly dissolved in the matrix, whereas if
grains are too coarse, they induce crack generation. The effect of quartz
dissolution in glass matrix on the mechanical and physical properties
were studied by many researchers but still there is an argument re-
garding the positive or negative effect of residual quartz especially on
porcelain strength [9–12]. The glassy phase formed during firing pro-
cess fills the pores in the microstructure and enables the intensification
of the interior. The formation of viscous liquid starts at different tem-
peratures depending on the chemical composition of the porcelain
body. During the firing of a typical porcelain, feldspar particles react
with SiO2 from Meta kaolin to form an eutectic liquid at 990 °C for
potassium feldspar and 1050 °C for sodium feldspar. The glassy phase is
placed in the particle boundaries at 1200–1400 °C [13,14].
An intensive structure can be provided when the proportion of the
open pores is reduced to 0.5%. [15]. The density of the porcelain is

⁎
Corresponding author.
E-mail addresses: faziletgungor@kutahyaporselen.com.tr (F. Güngör), nay@anadolu.edu.tr (N. Ay).

https://doi.org/10.1016/j.ceramint.2018.03.086
Received 20 November 2017; Received in revised form 15 February 2018; Accepted 11 March 2018
0272-8842/ © 2018 Published by Elsevier Ltd.

Please cite this article as: GÜNGÖR, F., Ceramics International (2018), https://doi.org/10.1016/j.ceramint.2018.03.086
F. Güngör, N. Ay Ceramics International xxx (xxxx) xxx–xxx

Table 1 the ratio of the submicron particles. Since the powers of fine-sized
Factors and the levels of the experiments. particles are higher than those of large-sized ones, their mobility is also
higher. Thus, they can react more easily and increase the speed of
Factors Levels
sintering [15–20].
I II III IV Another important effect of the particle size of the porcelain com-
position is the rheology of the porcelain slurry suspension and the
Quartz Quartz 101 Quartz 102 Quartz 103 –
packing density of the structure to be formed at the end of the shaping
K-Feldspar K- feldspar 101 K- feldspar 102 K- feldspar 103
Temperature (°C) 1150 1175 1200 1250 process. Particle packing is directly controlled by the particle size dis-
tribution of the material being processed. For this reason, particle
packing is important to all particulate/fluid systems. The diffusion
Table 2 distance of the particles can be shortened to increase the sintering speed
Experimental heat treatment parameters. with a dense packing [21–23]
A factorial design is a research method that includes two or more
Heating rate Dewell temperature Dwell time
(K min−1) (°C) (Minutes) factors. A factor is an independent variable in an experiment, especially
those that include two or more independent variables. One advantage
10 1150 30 of a factorial design is that it creates a more realistic situation, which
10 1175 30
can be obtained by examining a single factor. In addition, it uses a
10 1200 30
10 1250 30 notation system that identifies both the number of factors and the
number of values or levels that exist for each factor. [24]. In the study
of Bondioli et al., they investigated the mathematical model of the
Table 3 pressing pressure for porcelain tile production, the effect of the max-
Chemical analysis of raw materials. imum firing temperature and the duration of the firing period on the
firing contraction. Two-level multi-factorial experiment design was
SiO2 Al2O3 Fe2O3 TiO2 Na2O K2O CaO MgO L.O.I.
used to model the factors affecting fired shrinkage. The results show
Kaolin 49.65 35.86 0.07 0.01 0.09 1.96 0.08 0.23 12.05 that it is possible to define the effect of different factors on the reduc-
Albite 70.11 18.28 0.10 0.17 10.05 0.3 0.64 0.12 0.18 tion of firing temperature with a limited number of tests. [25].
K- Feldspar 68.30 17.02 0.16 0.47 2.24 10.81 0.16 0.33 0.51 In this study, the bodies were prepared by using quartz and K-
Quartz 98.44 0.92 0.05 0.02 0.16 0.11 0.13 – 0.17
Feldspar in three different particle size distributions in a fixed porcelain
body composition and passed through four different firing cycles. The
Table 4
technical and physical development of bodies during and after sintering
Particle size results of the raw materials. process have been investigated. Experimental studies were carried out
based on the factorial experiment design (the order of construction and
d(0.1)µm d(0.5)µm d(0.97)µm repetition depends on the rules). Considering the significant effect of
Quartz 101 1.52 4.25 12.16
the temperature on the test results, 3241 general factorial experiment
Quartz 102 3.79 21.18 67.16 was designed. The analysis of the experiments was implemented using
Quartz 103 10.12 36.19 141.10 Minitab 13.20 package program.
K-Feldspar 101 1.49 3.97 13.99
K-Feldspar 102 1.93 13.51 50.16
K-Feldspar 103 4.58 29.18 82.19 2. Experimental studies

The factors and levels determined in the experiments are given in

Table 5 Table 1. Test specimens were fired in a laboratory type Forno Ceramica
Chemical analysis of the semi transparent porcelain body composition. kiln at 1150, 1175, 1200 and 1250 °C. Experimental heat treatment
parameters are given in Table 2. The chemical analysis results of the
Na2O MgO Al2O3 SiO2 K2O CaO TiO2 Fe2O3 LOI
raw materials were determined by the fluorescence method (XLab
% 1.15 0.15 22.74 65.52 4.57 0.40 0.09 0.41 4.97 2000-Spectro) and results are given in Table 3.
The particle size distribution of quartz, K- feldspar and albite were
measured by laser diffraction method using Malvern Master Sizer 200 G
increased by increasing the temperature due to the elimination of open model and the results are given in Table 4. The theory based chemical
pores by viscous flow. Viscous flow is controlled by the melt viscosity. analysis of the standard recipe used for every sample is given in Table 5.
The melt viscosity is directly related to the composition and the particle The component contents of the compositions were given in Table 6.
size of the body. A high density structure can be achieved by increasing To prepare the batches, raw materials were dispersed in an aqueous

Table 6
The component contents of the compositions.

Composition code T1011 T1012 T1013 T1021 T1022 T1023 T1031 T1032 T1033

Quartz 101 x x x
Quartz 102 x x x
Quartz 103 x x x
K Feldispar 101 x x x
K Feldispar 102 x x x
K Feldispar 103 x x x
Albite x x x x x x x x x
Kaolin x x x x x x x x x

2
F. Güngör, N. Ay
Fig. 1. Positioning of the sample before and after firing for pyroplastic index assay [26].
Fig. 2. Expansion behaviour of the samples.
slip containing 70 wt. percentage solids and 0.01 wt. percentage dis-
persant (Na-silicate) then mixed for 1 h using IKA RW 20 digital mixer.
The slurries were screened by 60 mesh (250 µm) sieves. Slips were
poured into the bar shaped plaster moulds. Dimension of the bars are
3,5 cm × 20 cm × 1 cm. Bulk density and open porosity were de-
termined by water saturation and Archimedes principle (ISO 10545-3).
The Pyroplastic index (PI) of each sample was determined by three-
point test method after firing: where h is the thickness of the body, S is
the maximum deflection and L is the distance between the refractory
supports as shown in Fig. 1 [26].
4h2
PI =
3L4
The sintering behaviours of the structures were investigated using a
Fig. 3. Phases and ratios in the microstructure.
Ceramics International xxx (xxxx) xxx–xxx
double-cam contactless optical dilatometer in the Misira 3.32-ODHT-
HSM 1600/80 (Expert System Solutions) brand and model capable of
applying the industrial firing regime (30 min dwell at 1200 °C, from
cold to cold 130 min). The detection of the phases formed in the fired
product was carried out with a Rigaku Rint 2000 XRD device. A full
profile interpretation by Rietvelt refinement was carried out with the
GSAS-EXP GUI software package. In addition, for microstructural ex-
aminations, secondary electron images were taken with scattered
electron microscopy (SEM, Zeiss Supra 50 VP) at the fractured surfaces
of the samples. Light transmission of the fired samples were measured
by TES 1332 lx meter according to TS EN 1184.
3. Results and discussion
The heat microscopy results of the samples for T1011, T1021,
T1023, and T1033 are given in Fig. 2. The shrinkage for T1011, which
the particle size is the thinnest value for quartz and K- feldspar is 9.61%
while the shrinkage of T1033 sample with the thickest particle size is
5.77%. The shrinkage value for T1023 sample where quartz d (0.5)
21.18 µm and K feldspar d(0.5) 29.18 µm is 7.28%. For T1021 sample,
quartz d(0.5) 21.18 µm and K feldspar d(0.5) 3.97 µm is 7.75%. These
results show us clearly that the effect of quartz on the shrinkage be-
haviour is higher than that of K-feldspar because the amounts of quartz
and K-feldspar are almost the same in the body composition. The
sudden increase in the expansion at 573 °C is the reason of the trans-
formation of quartz from the alpha form to beta form. Densification is
(1)
insignificant from initial temperature until 900 °C. Density then slightly
rises up from 900° to 1000°C resulted from the reduction of specific
3
F. Güngör, N. Ay Ceramics International xxx (xxxx) xxx–xxx

Table 7 around the solid particles and fills the pores in the microstructure with
Composition and viscosities of the amorphous phases. capillary effect. Small pore size produces large suction forces a low
viscosity liquid contributes a good wetting effect. The capillary pressure
Sample/ firing T1011/ T1011/ T1022/ T1022/ T1033/ T1033/
temperature 1150 1250 1150 1250 1150 1250 brings the particles closer together, resulting in shrinkage and lowering
porosity consequently development of densification. Depending on the
SiO2 67.06 70.90 60.27 64.83 34.29 58.83 amount and the viscosity of the liquid phase formed, the shrinkage and
Al2O3 23.21 19.24 28.93 25.18 49.00 28.93
density values change [1,7,27–30].
Na2O 1.16 2.19 0.26 2.22 0.40 2.72
K2O 6.97 6.25 8.59 6.33 13.36 7.76 Determined Phases and their percentages in the microstructure are
Fe2O3 0.63 0.56 0.77 0.57 1.20 0.70 given in Fig. 3. Quartz, mullite, albite and amorphous phases are the
TiO2 0.13 0.12 0.16 0.12 0.25 0.15 main phases in the microstructure. The chemical content of the amor-
CaO 0.61 0.55 0.75 0.55 1.17 0.68 phous phases and their viscosity at 1150 and 1250 °C were calculated
MgO 0.23 0.21 0.28 0.21 0.44 0.25
according to Lakatos [31] that is given in Table 7.
Total 100.00 100.00 100.00 100.00 100.00 100.00
Viscosity [31] 5.05 3.92 5.79 4.52 11.70 4.89 In samples with fired at 1150 °C, the viscosity value of the amor-
phous phase varied considerably with the change in particle size. As the
firing temperature increased, the relationship between d(0.5) value and
Table 8 amorphous phase viscosity changed linearly. The effect of change in
Bulk density (BD), water absorption (WA), Pyroplastic index (PI), Light transmission (LT) particle size at 1150 °C on viscosity of the amorphous phase is greater
results of the samples.
than that of 1250 °C.
Sample/temperature BD (gr/ WA. % PI (cm−1(10−5)) L.T. (Lux) By the increasing of d(0.5) value, the increase in the viscosity is due
°C cm3) to feldspar. Feldspar begins to melt at about 1150 °C. As the particle size
of the feldspar thickened, the melting rate decreased and thus the
T1011/1150 2,28 1,16 8,88 21,37
viscosity of the resulting liquid phase increased. Feldspar is the in-
T1011/1175 2,29 0,99 9,83 29,5
T1011/1200 2,3 0,47 11,32 30,5 gredient that begins to melt at the lowest temperature in the micro-
T1011/1250 2,45 0 14,67 49,8 structure. The liquid phase reacts with other constituents and provides
T1012/1150 2,26 1,24 8,12 21,05 condensation in the microstructure. Increase of feldspar ratio decreases
T1012/1175 2,27 1,09 8,97 29,24 vitrification temperature [32]. Increasing the firing temperature in-
T1012/1200 2,29 0,49 10,84 33,66
T1012/1250 2,43 0 14,59 49,76
crease the amount of the liquid phase. The surface energy forces created
T1013/1150 2,25 1,29 7,94 20,96 by thin pores in the ceramic body, the liquid phase tends to get through
T1013/1175 2,26 1,22 8,64 28,59 among the particles, thereby reducing the open porosity value [33].
T1013/1200 2,27 0,54 10,63 32,17 The measurement results of bulk density, water absorption, pyr-
T1013/1250 2,42 0 14,56 49,73
oplastic index and light transmission are given in Table 8. Created cake
T1021/1150 2,21 1,51 6,21 19,95
T1021/1175 2,22 1,44 6,42 27,63 chart according to factorial design by Minitab 13 for bulk density is
T1021/1200 2,25 1,39 7,97 30,11 given in Fig. 4. The parameter that most influence the bulk density
T1021/1250 2,4 0 12,16 48,86 value is the temperature change and its effect is 55.60%. The other
T1022/1150 2,2 1,59 5,99 19,92 effective parameter is the change in quartz particle size. The change in
T1022/1175 2,21 1,53 6,35 27,06
T1022/1200 2,24 1,45 7,24 29,69
particle size of K-feldspar is lesser then the particle size of quartz and
T1022/1250 2,38 0,11 11,94 48,84 temperature. Effect of uncontrollable factors for experimental proce-
T1023/1150 2,15 1,6 5,81 19,87 dure indicated as ‘’Fault’’ on the cake chart is 8.60%. Cake chart for
T1023/1175 2,17 1,57 6,33 26,64 water absorption is given in Fig. 5. The parameter that most influences
T1023/1200 2,23 1,51 7,11 29,08
the water absorption value is the temperature, that is 61.30%. The
T1023/1250 2,37 0,12 11,87 48,79
T1031/1150 2,11 2,16 5,79 18,26 other parameter after temperature is the change in quartz particle size
T1031/1175 2,13 1,98 6,59 24,05 by 32.40%. The effect of interaction of quartz and temperature is
T1031/1200 2,15 1,93 7 27,19 4.60%. Measurement of water absorption value is often used to measure
T1031/1250 2,36 0,37 11,74 45,52 the success of the firing cycle when ceramic product is being done
T1032/1150 2,1 2,27 5,75 18,26
T1032/1175 2,11 2,21 6,26 24,05
[34–36]. When the water absorption values of the samples fired at the
T1032/1200 2,14 2,04 6,99 26,84 same temperature but different particle size are examined, it is seen
T1032/1250 2,36 0,41 11,58 45,51 that the water absorption value decreases with decreasing particle size.
T1033/1150 1,99 2,36 5,73 18,25 There are two main reasons for this. 1. Packing of the particles 2. In-
T1033/1175 1,99 2,31 6,18 23,94
crease of the amount of glassy phase formed. With the use of quartz
T1033/1200 2 2,16 6,94 26,53
T1033/1250 2,34 0,89 11,46 45,11 with a finer particle size, the amount of vitreous phase formed by the
dissolution of fine quartz particles in the glassy phase increases. The
change in the composition of the liquid phase with the amount of quartz
surface area with a negligible of linear shrinkage by surface diffusion entering the liquid phase is due to the decreased particle size of quartz.
mechanism. Sintering of polycrystalline materials, density is increased Cake chart for pyroplastic index result is given in Fig. 6. The effect
due to elimination of open pores by viscous flow sintering or vi- of temperature is higher than median particle sizes of quartz and K-
trification process. The density dramatically jumps from 1000 °C to the Feldspar. Pyroplastic index value increased as temperature increased
peak temperature by flowing of the liquid glass phase, which is pro- and quartz particle size decreased. The viscosity and the amount of
duced for viscous flow sintering [7]. As the particle size of the porcelain glassy phase and pyroplastic index values of the samples of T1011,
batch is decreased the packing density of the shaped material is in- T1022 and T1033 for 1150 and 1250 °C are plotted in the same graph
creased and the diffusion distance is shortened. Shrinkage behaviour is that is given in Fig. 7. The pyroplastic index increased rapidly when the
directly proportional to the surface tension, but inversely proportional amount of glassy phase increased above a certain value. The reason for
to the viscosity because the liquid phase formed during firing wraps this is the increase in temperature and the decrease in particle size of

4
F. Güngör, N. Ay
Fig. 4. Cake chart of bulk density.
Fig. 5. Cake chart of water absorption.
Fig. 6. Cake chart for pyroplastic index results.
the raw materials used. It is thought that the pores found in the mi-
crostructure having fine particle size are in small and they are easily
removed during the firing process. Viscosity should not be so low as to
cause distortion by the gravitational effect during the time required for
densification. In this case, the particle size should be in the appropriate
value to provide higher surface stress than the stresses of the gravity
force [37–40]. Viscous liquid silicate forms at the firing temperature
5
Ceramics International xxx (xxxx) xxx–xxx
and acts as a binder for the body. The amount and viscosity of the liquid
phase should be acceptable and sufficient for densification. In excess
amount and formation of vitreous glassy phase, slippage occurs be-
tween the particles. The glassy phase formed by the temperature in-
creases and the particles becomes soft. Also, viscosity decreases, ac-
celerating the creep behaviour, causing the product to deform and
warping. The measurement results of shrinkage and deformation are
important parameters in terms of determining the ceramic body com-
position and firing cycle [37].
The optimum vitrification interval is provided when the amount of
water absorption approaches zero and linear shrinkage reaches its
maximum value. Firing over than vitrification range increased the force
of abandonment of the trapped gases and created negative effects such
as blistering on the surface of the fired product. In the study carried out
at 1250 °C of the sample of T1011, bubbles were observed on the final
product because of the firing above the vitrification interval.
The thermal expansion behaviours of the samples fired at 1200 °C
were measured. Thermal expansion results of the samples in the range
of 30–550 °C and the effect of d(0.5) values of quartz and K-feldspar are
represented in Fig. 8. Although all of the recipe components are iden-
tical, the change in particle size caused a change in the thermal ex-
pansion coefficient. It is seen that the thermal expansion coefficients of
the samples containing fine quartz (quartz-101) are lower than those of
medium and thick quartz 102 and quartz 103. By the decrease in the
particle size of quartz, the amount of residual quartz decreased and
confirmed with XRD analysis results. Crystalline and amorphous forms
of quartz have different thermal expansion coefficients. For the crys-
talline quartz, the thermal expansion coefficient for the range of
20 – 750 °C is 23 × 10–6, while the glassy phase thermal expansion
coefficient is 3 × 10–6 [32]. Theoretical thermal expansion coefficient
values of the samples fired at 1250 °C were calculated to determine the
effect of increasing the quartz solubility and the increase in amorphous
phase on the thermal expansion coefficient. Eq. (1) was used for the
calculation. The thermal expansion coefficient for the mullite phase was
taken as 4 × 10−6, 23 × 10−6 for crystalline quartz and 3 × 10−6 for
the amorphous phase [4,32]. The relationship between the calculated
thermal expansion coefficients and the amorphous phase ratio percen-
tage and the crystal quartz ratio percentage are plotted in Fig. 9. The
solubility of quartz in the glassy phase depends on its particle size. In
the porcelain glass, the quartz resolution lasts until reaching the sa-
turation limit, and after reaching this limit, the resolution ends. Studies
show that the bodies have fine quartz particle size the solubility of
quartz during firing is increased and the thermal expansion coefficient
is decreased [38,41–43].
(∝ap × Vap) + (∝q × Vq) + (∝m × Vm)
∝sample =
Vap + Vq + Vm (2)
(∝ap : Thermal expansion coefficient of the amorphous phase, ∝q :
Thermal expansion coefficient of quartz, ∝m : Thermal expansion coef-
ficient of mullite, Vap : Percentage of amorphous phase in the micro-
structure, Vq : Percentage of quartz in the microstructure, Vm :
Percentage of mullite in the microstructure)
The cake and the basic effect graph for light transmission are shown
in Fig. 10 and Fig. 11 respectively. When cake chart is examined, it is
seen that the most effective factors are temperature and quartz particle
size. The effect of temperature is more than quartz particle size. The
basic effect graph shows that the increase in temperature and the de-
crease in quartz particle size increased the light transmittance value.
Although the reduction of K-feldspar particle size increases the light
transmittance value, this effect is not as great as the change in tem-
perature and quartz particle size.
F. Güngör, N. Ay
Fig. 7. Effect of amount of glassy phase and its viscosity on pyroplastic index.
Fig. 8. Thermal expansion results and d(0.5) values of quartz and K-feldspar.
Fig. 9. Effect of amount of amorphous and quartz phases on thermal expansion.
If the crystals that enter the liquid phase are at the same refractive
index value as the liquid phase, they become invisible because the light
path is not diffracted. Porcelain; glass, crystals, air gaps and the reason
for the lack of transparency; glass, crystals and air have different re-
fractive indices. However, porcelain is semi-transparent. Because the
6
Ceramics International xxx (xxxx) xxx–xxx
Fig. 10. Cake chart for light transmission.
difference of the refractive indices values between the phases in the
porcelain are small. The phases and refractive indices in hard porcelain
after firing are given in Table 9 [44].
In a dense porcelain body microstructure, fine needle shaped mul-
lite crystals and undissolved or partially dissolved quartz crystals are
found in glassy matrix. Mullite particle size is small in micron range
while quartz particle size is bigger. In porcelain microstructure, the
mullite phase is the main phase, which provides scattering due to the
large difference in particle size and refractive indices, and therefore the
transparency decreases with the increase of the mullite phase amount.
The basic process for increasing the transparency is to increase the
amount of vitreous phase and reduce the amount of mullite phase [44].
The light transmittance value is plotted in Fig. 12 versus the mullite
phase ratio for the 1250 °C determined by XRD analysis. The result
shows that the increase of the mullite phase ratio increases the light
transmission. Because the amount of mullite phase increases as the
amount of glassy phase increases, the amount of glassy phase was en-
ough for eliminating the negative effect of mullite on light transmission.
Reflected electron images of the polished surfaces of the specimens
produced after firing at the specified temperatures are given in Fig. 13.
F. Güngör, N. Ay Ceramics International xxx (xxxx) xxx–xxx

Fig. 11. Basic interaction graph for light transmission.

means that free quartz contains amorphous phase.

Table 9 The effect of amorphous phase viscosity on the amount of open
Refractive indices of phases in porcelain microstructure porosity is given in Fig. 14. The decrease in the viscosity value reduced
[44].
the amount of open porosity in the microstructure. Increasing the sin-
Phase Refractive indices tering time and temperature accelerates the viscous flow by reducing
the viscosity of the vitreous phase present in the matrix, and conse-
Glass 1.48 quently the porcelain body is condensed. Sintering with glassy phase
Quartz 1.52
formation increases rapidly and as the total porosity decreases, the
Mullite 1.64
density increases, the activation energy decreases and diffusion thus
becomes easier.
Secondary electron images taken from fractured surfaces of T1033,
T1011 and T1022 fired at 1150 and 1250 °C are given in Fig. 15. It is
seen that for the sample T1033, only primary mullite crystals are ob-
served in the microstructure fired at 1150 °C. The formation and
amount of secondary mullite crystals are increased by decreasing the
particle size and increasing the temperature. Formation of secondary
mullite crystals requires low viscosity liquid formation [2,32]. It is seen
that the amount of secondary mullite crystals formed by thinning of
particle size increases and vitrification progresses with increasing firing
temperature. At high temperatures, the viscosity of the liquid phase
falls and the large pores removes among the particles.

Fig. 12. Effect of mullite on light transmission. 4. Conclusion

Densification of porcelain body is strongly affected by the heat

In the reflected electron image, the heavy elements appear to be whiter
treatment parameters and the particle sizes of quartz and K- feldspar in
than the other elements. The microstructure images show quartz par-
the porcelain batch. As the particle size of porcelain batch decreases,
ticles with dark grey shades and porosities with black shades. The dif-
the packing density of the shaped material increases and the sizes of the
ferences in the amount of residual quartz and pores are different ac-
pores between the particles decreases. As the area of the pores de-
cording to the difference in the firing temperature and median particle
creases, diffusion distance decreases and densification can be im-
size. The amount and the size of the pores was decreased as the particle
plemented.
sizes of the samples became smaller. In addition to this, the residual
The effect of quartz particle size on the shrinkage behaviour is
quartz particles formed in the structures was both decreased in size and
higher than that of K-feldspar. Quartz is the main determinant of the
amount. Again, due to the shrinkage of the particle size of the sludge,
viscosity and the amount of liquid phase, density and the shrinkage
the residual is reduced in the cracks formed around the quartz particles
value. These factors dramatically changes as the particle size of quartz
and the structure of the structures becomes more homogenous. The
changes.
obtained microstructure images show that the decrease in the amount
There is a linear relationship between particle size and the amor-
of residual quartz occurs as the particle size increases, which is de-
phous phase. The thickness of particle size also increased the amor-
termined by the XRD results. It also supports the results of dilatometry
phous phase viscosity. The effect of particle size on viscosity of the
analysis and the decrease in the coefficient of thermal expansion, which
amorphous phase at 1150 °C is greater than that of 1250 °C.

7
F. Güngör, N. Ay Ceramics International xxx (xxxx) xxx–xxx

Fig. 13. Reflected electron images of the polished surfaces of the specimens (a) T1033 sample fired at 1150 °C, (b) T1022 sample fired at 1150 °C, (c) T1033 sample fired at 1250 °C, (d)
T1022 sample fired at 1250 °C.

In the result of factorial design, the parameter that most influences

Fig. 14. Effect of viscosity of amorphous phase on open porosity.
the bulk density value is the temperature change and the other para-
meter is the change in quartz particle size. The change in particle size of
K- feldspar on the effect of water absorption is less than the change of
temperature and quartz particle size. Pyroplastic index value increases
as temperature increases and quartz particle size decreases.
For light transmission value, the most effective factors are tem-
perature and the particle size of quartz. Although the reduction of K-
feldspar particle size increases the light transmittance value but this
effect is not as great as the change in temperature and quartz particle
size. Amount of mullite phase increases as the viscosity of glassy phase

8
F. Güngör, N. Ay Ceramics International xxx (xxxx) xxx–xxx

Fig. 15. Secondary electron images of the samples (a) T1033 fired at 1150 °C, (b) T1033 fired at 1250 °C (c) T1022 fired at 1150 °C, (d) T1022 fired at 1250 °C, (e) T1011 fired at 1150 °C
(f) T1011 fired at 1250 °C.

9
F. Güngör, N. Ay
decreases. Despite the high refractive indices of mullite, light trans-
mission was high because the amount of glassy phase increases as the
particle size of porcelain batch decreases and its amount is enough for
eliminating the negative effect of mullite.
References
[1] J. Marquez, J.M. Rincon, M. Romero, Effect of firing temperature on sintering of porce-
lain stoneware tiles, Ceram. Int. 34 (2008) (2017) 1867–1873.
[2] W.E. Lee, Y. Iqbal, Influence of mixing on mullite formation in porcelain, J. Ceram. Soc.
21 (2001) 2583–2586.
[3] W.M. Carty, U. Senapati, Porcelain, raw materials, processing, phase evolution and me-
chanical behaviour, J. Am. Ceram. Soc. 81 (1) (1998) 3–20.
[4] S.R. Braganca, C.P. Bermann, H. Hübner, ‘Effect of quartz particle size on the strength of
triaxial porcelain’’, J. Eur. Ceram. Soc. 26 (2006) 3761–3768.
[5] Y. Kobayashi, O. Ohira, E. Kato, ‘‘Effect of firing temparature on bending strength of
porcelains for tableware’’, J. Am. Ceram. Soc. 75 (1992) 1801–1806.
[6] J. Martín-Márquez, J.M. Rincón, M. Romero, Effect of microstructure on mechanical
properties of porcelain stoneware, J. Eur. Ceram. Soc. 30 (2010) 3063–3069.
[7] W. Lerdprom, Firing of Porcelain, Alfred University, Master of Science in Ceramic
Engineering, 2014.
[8] W.M. Carty, Observations on the glass phase composition in Porcelains, Ceram. Eng. Sci.
Proc. 23 (2) (2002) 79–94.
[9] J.M. Marquez, J.M. Rincon, M. Romero, Effect of microstructure on mechanical proper-
ties of porcelain stoneware, J. Eur. Ceram. Soc. 30 (2010) 3063–3069.
[10] G. Stathis, A. Ekonomakou, C.J. Stoutnaras, C. Ftikos, Effect of firing conditions, filler
grain size and quartz content on bending strength and physical properties of sanitaryware
porcelain, J. Eur. Ceram. Soc. 24 (2004) 2357–2366.
[11] W.M. Carty, B.M. Pinto, Effect of filler size on the strength of porcelain bodies, Ceram.
Eng. Sci. Proc. 23 (2) (2002) 95–105.
[12] L. Mattyasovszky, Zsolnay, Mechanical strength of porcelain, Am. Ceram. Soc. 40 (1957)
299–306.
[13] E. Rambaldi, W.M. Carty, S. Tucci, L. Esposito, Using waste glass as a partial flux sub-
stitution and pyroplastic deformation of a porcelain stonewate tile body, Ceram. Int. 33
(2007) 727–733.
[14] W. Lerdprom, R.K. Chinnam, D.D. Jayaseelan, W.E. Lee, Porcelain production of direct
sintering, J. Eur. Ceram. Soc. 36 (2016) 4319–4325.
[15] T. Aydın, A. Kara, Effect of spodumene addition on pyroplastic deformation of porcelain
stoneware, J. Ceram. Process. Res. 15 (6) (2014) 486–491.
[16] A.M. Bernardin, D.S. Medeiros, H.G. Riella, Pyroplasticity in porcelain tiles, Mater. Sci.
Eng. A 427 (2006) (2006) 316–319.
[17] Reed, J., Principles of Ceramics Processing, U.S.A, 1995.
[18] D.W. Richerson, Modern Ceramic Engineering, The University of Utah, Marcel Dekker,
INC., 1992.
[19] R.M. German, Fundamentals of sintering, Ceram. Glasses Eng. Mater. Handb. 4 (1996)
260–269.
[20] S. Villegas-Palacio, D.R. Dinger, PSD effects on firing properties on firing properties of
porcelains, J. Am. Ceram. Soc. 75 (1996) 71–83.
[21] S.M. Olhero, J.M.F. Ferreira, Influence of particle size distribution on rheology and
10
Ceramics International xxx (xxxx) xxx–xxx
particle packing of silica based suspensions, Powder Tech. 139 (2004) 69–75.
[22] F.F. Lange, Powder procesing science and technology for increased reliability, J. Am.
Ceram. Soc. 72 (1989) 3–15.
[23] D.R. Dinger, J.E. Funk, Particle Packing Phenomena and their Application in Material
Processing, MRS Bulletin, 1997.
[24] F.J. Gravetter, L.B. Forzano, Research methods for the behavioural sicences, 5th ed.,
Cengage Learning, 2012.
[25] R.M. German, Fundamentals of Sintering, 4 Ceramic and Glasses, Engineered Materials
Handbook, 1996, pp. 260–269.
[26] L. Conserva, F. Melchiades, S. Nastri, A. Boschi, M. Dondi, G. Guaribi, M. Raimondo,
C. Zanelli, ‘Pyroplastic deformation of porcelain stoneware tiles: wet vs dry processing, J.
Eur. Ceram. Soc. 37 (2017) 333–342.
[27] M.N. Rahaman, Ceramic Processing and Sintering, Taylor and Francis Group, U.S.A,
2003.
[28] G.W. Scherer, Viscous Sintering of Inorganic Gels, Surface and Colloidal Science, Plenum
Publishing Corporation, Wilmington, A.B.D., 1987, pp. 266–296.
[29] R.M. German, Fundamentals of Sintering, Ceramic and Glasses, 4 Engineered Materials
Handbook, 1996, pp. 260–269.
[30] E.M.H. Sallam, A.C.D. Chaklader, Sintering characteristics of porcelain, Ceram. Int. 11
(1985) 151–161.
[31] T. Lakatos, L.G. Johansson, B. Simmingsköld, Viscosity temperature relations in the glass
system SiO2-Al2O3-Na2O-K2O-CaO-MgO in the composition range of technical glasses,
Glass Technol. 13 (3) (1972) 88–95.
[32] Y. Iqbal, E.W. Lee, Microstructural Evolution in Triaxial Porcelain, J. Am. Ceram. Soc. 83
(2000) 3121–3127.
[33] J.M. Marquez, J.M. Rincon, M. Romero, ‘Mullite development on firing in porcelain
stoneware bodies’’, J. Eur. Ceram. Soc. 30 (2010) 1599–1607.
[34] W.M. Carty, Observations on the glass phase composition in porcelains, Ceram. Eng. Sci.
Proc. 23 (2) (2002) 79–94.
[35] F. Bondioli, A.M. Ferrari, M. Romagroli, The effect of sintering conditions on linear
shrinkage in porcelainized stoneware tiles as studied in a two level full factorial design,
Tile Brick Int. (2006) 18–21.
[36] A. Salem, S.H. Jazayeri, E. Rastelli, G. Timellini, Dilatometric study of shrinkage during
sintering process for porcelain stoneware body in presence of nepheline syenite, J. Mat.
Process Tech. 209 (2009) 1240–1246.
[37] W.D. Kingery, H.K. Bowen, D.R. Uhlmann, Introduction to Ceramics, New York, 1976.
[38] J.L. Amoros, J. Garcia, M.J. Orts, S. Mestre, J. Garcia-Ten, C. Feliu, J. Eur. Ceram. Soc. 30
(2010) 17–28.
39] L. Bachero, Effect of quartz particle size on the thermal expansion of porous whiteware
bodies, Qualicer (2000) 35–37.
[40] E.S. Vilches, Technical considerations on porcelain tile products and their manufacturing
process, Qualicer 1 (2002) 57–83 (2002).
[41] F.J.S. Arantes, D.F. Galesi, E. Quinterio, A. Boschi, Closed porosity in porcelain tile,
Qualicer (2002) 139–142.
[42] A.D.N. Junior, D. Hotza, V.C. Soler, E.S. Vilches, Analysis of the development of micro-
scopic residual stresses on quartz particle in porcelain tile, J. Eur. Ceram. Soc. 28 (2008)
2629–2637.
[43] T.K. Mukhopadhyay, S. Ghosh, S. Ghatak, H.S. Maiti, Effect of pyrophyllite on vitrifica-
tion and on physical properties of triaxial porcelain, Ceram. Int. 32 (2006) 871–876.
[44] P. Rado, An Introduction to the Technology of Pottery, Pergamon Press, Oxford, 1988.