Ceramics International 32 (2006) 727–732

www.elsevier.com/locate/ceramint

The effect of fired scrap addition on the sintering

behaviour of hard porcelain
Alexander Karamanov a,*, Emilia Karamanova a, Anna Maria Ferrari b,
Fabiola Ferrante a, Mario Pelino a
a
Department of Chemical Engineering and Materials, University of L’Aquila, Monteluco di Roio 67100, Italy
b
Department of Engineering Sciences and Methods, University of Modena and Reggio Emilia, Via Fogliani 1 42100, Italy
Received 22 July 2004; received in revised form 29 April 2005; accepted 20 May 2005
Available online 16 September 2005

Abstract

The effect of the addition of porcelain fired scraps on the sintering, phase composition and properties of a hard porcelain was studied. Two
ceramics were compared: a commercial triaxial porcelain labelled C-0 and one, where 15% of the mass of C-0 was substituted by scraps
yielding a composition labelled C-15. The sintering was evaluated by the linear shrinkage, the water absorption and the variation of closed and
total porosity, while the phase transformations were investigated by DTA-TG, dilatometry and XRD. Samples, sintered at 1350 and 1400 8C,
were used for the evaluation of the Young modulus and the bending strength. The addition of the fired scrap had a positive effect on the quartz
dissolution and on the formation of the liquid phase during the sintering. The sintering and the mechanical properties were improved by heat
treatment in the 1300–1350 8C range. Above this temperature, a significant over-firing with a decrease of the mechanical properties was
observed.
# 2005 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Keywords: A. Sintering; D. Porcelain; Waste

1. Introduction corundum, bentonite, etc. [13,14], glass cullet [15–17] or

waste products [18–20]. This latter aspect acquires
Porcelain is a white, non-porous and partially transparent particular importance due to the reduced availability of
ceramic material, whose production started in China many high quality raw materials. In this regard, the porcelain
centuries ago. Today, porcelain is produced in many industry produces its own residues, part of which can be re-
countries and its technology is well known and described used in the production cycle; a typical example are the fired
in different textbooks and papers [1–5]. Recently, a well- scraps of defected products [12,21,22]. Due to the
sintered tiling material, produced by a short firing cycle and requirement of using raw materials with very low iron
known as porcelain stoneware, was developed [6]. However, oxides content, this possibility has particular importance for
the optimisation of the tableware porcelain and porcelain the production of hard porcelain. In this case, particular
stoneware productions is still ongoing and many new works, attention is paid to the rheological properties of the parent
discussing the ceramic structure [6–8] or the improvement of mass [21,22], because the addition of the waste might
its properties [9–12], are published every year. influence the forming process. This replacement, however,
A consistent part of the new studies is related to the also leads to a variation of the phase composition of the
substitution of the traditional raw material in the composi- green, which may influence the sintering process.
tion (sand, feldspar and kaolin) with others, such as zeolite, The present work investigates the effect of the addition
of fired scraps on the densification process in a hard por-
* Corresponding author. Tel.: +39 0862 434233; fax: +39 0862 434233. celain. Two ceramic compositions are compared: a typical
E-mail address: karama@ing.univaq.it (A. Karamanov). commercial composition of kaolin, sand and feldspar and

0272-8842/$30.00 # 2005 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
doi:10.1016/j.ceramint.2005.05.009
728 A. Karamanov et al. / Ceramics International 32 (2006) 727–732

a second one, where part of the sand and part of the feld- Samples A were also used to evaluate the bulk, rb,
spar were substituted by fired porcelain scraps. skeleton, rs, and real, rr, densities. rb was estimated by a dry
flow pycnometer (GeoPyc 1360), while rs and rr by He
displacement Pycnometer (AccuPyc 1330), before and after
2. Experimental procedure crashing and milling the samples below 26 mm, respectively.
The results were used to determine the total, PT, and the
Fired porcelain rejects may be cannibalised in different closed, PC, porosities:
manners. The simplest is the addition to an unchanged batch.
rr  rb
This guaranties no variation of the total chemical composi- PT ¼ 100 (1)
tion but decreases the kaolin percentage, i.e. may influence rr
the plasticity of the parent mass. This way is appropriate for rr  rs
a continuous production cycle and a relatively low PC ¼ 100 (2)
rr
percentage of scrap addition.
Another possibility is the replacement of only the non- GeoPyc 1360 [23] measures the ‘‘enveloped’’ volume of
plastic raw materials. This type of substitution varies the porous samples, placed in a chamber with compressed free-
chemical composition, so that it is more suitable for a flowing dry medium. The used medium, named DryFlo, is
discontinuous production cycle. Since the single replace- made of tiny, rigid spheres and a small amount of dry
ment of a filler increases the alkali oxides content and lubricant. It cannot fill in the open pores so that the measured
decreases the mechanical properties [12], it is more volume corresponds to the bulk density of the samples. The
appropriate to simultaneously substitute part of the sand AccuPyc 1330 [23] is a gas displacement Pycnometer, which
and part the feldspar. determines the ‘‘absolute’’ density by measuring the pres-
In the present work, two porcelains, whose compositions sure change of He in a calibrated volume. When bulk
are reported in Table 1, were studied. The first one, labelled samples are used, the evaluated volume corresponds to
C-0, corresponds to a commercial hard porcelain; in the the mass volume and closed for He pores, while when
second, labelled C-15, part of the sand and feldspar were powder samples are measured the real density is obtained.
replaced by fired scraps, milled under 63 mm in size, The experimental errors associated in the evaluation of
obtained by firing at 1350–1375 8C C-0. rb, rs and rr were estimated as 0.012 g/cm3, 0.003 g/
The chemical compositions of the green and fired cm3 and 0.002 g/cm3, respectively, which correspond to
porcelains were analysed by XRF analysis (Spectro an error of 0.6% and 0.2% for PT and PC, respectively.
XEPOS), while BET specific surface of the parent powders The sintering process was studied by a Differential
were obtained by N2 adsorption (ASAP 2000). The parent Dilatometer (Netzsch 402 ED) and by DTA-TG analysis
batches were prepared directly in a production plant using a (Netzsch STA 409). These experiments were carried out at
discontinuous production cycle, as masses with water 5 8C/min in the 20–1400 8C temperature range. By
content of about 20 wt.% (calculated on dry batch). Then, dilatometry, the thermal expansion curves of samples,
the compositions were dried at 105 8C, crashed, milled and sintered at different temperatures, were also obtained.
sieved at a grain size below 125 mm. After addition of The phase analyses of the C-0 and C-15 porcelains,
10 wt.% water and homogenisation, green samples with sintered at different temperatures, were made by X-ray
initial sizes of 10 mm
10 mm
8 mm (samples A) and diffraction (Philips PW1830 XRD apparatus and Cu Ka
50 mm
4 mm
3 mm (samples B) were obtained by radiation). The quantitative analyses have been carried out
uniaxial pressing at 50 MPa. After drying at room by the combined Rietveld–Reference Intensity Ratio
temperature, the samples were heated at 5 8C/min up to (R.I.R.) method [24,25]. A 10 wt.% of corundum (NIST
different temperatures (1200, 1300, 1350, 1375 and SRM 674a) has been added to all samples as internal
1400 8C), held for 1 h and cooled at 10 8C/min. The degree standard. The mixtures, ground in an agate mortar, have
of sintering was estimated by the linear shrinkage, DL, been side loaded into an aluminum flat holder in order to
related to the initial size of 50 mm (samples B), and by the minimize the preferred orientation problems. Data have
water absorption, W, of samples A [3,4]. The experimental been recorded in the 5–1408 2u range (step size 0.028 and 6 s
error was estimated as 0.05% and 0.03% for DL and counting time for each step). The phase fractions extracted
W, respectively. by the Rietveld–R.I.R. refinements, using GSAS software
[26] have been rescaled on the basis of the absolute weight of
Table 1 the corundum originally added to the mixtures as an internal
Initial compositions of the investigated porcelains (wt.%)
standard, and therefore internally re-normalized.
C-0 C-15 Series of five B samples, sintered at 1350 and 1400 8C
Kaolin 52 52 were used for the evaluation of the mechanical properties:
Quartz 35 24 the Young modulus was measured by means of the non-
Feldspar 13 9
destructive resonance frequency technique (Grindosonic)
Scrap – 15
while the bending strength was evaluated by a three point
 A. Karamanov et al. / Ceramics International 32 (2006) 727–732 729

bending test with 40 mm outer span at a speed of 0.1 mm/ Table 2

min (SINTEC D/10). Chemical compositions of C-0 and C-15 green samples and porcelains
(wt.%)
Al2O3 SiO2 K2 O Fe2O3

3. Results and discussion C-0 green 24.89 63.65 3.16 0.43

C-15 green 26.38 62.30 2.99 0.46
C-0 porcelain 26.58 69.19 3.69 0.47
Table 2 reports the chemical analyses of the green and C-15 porcelain 28.21 67.47 3.52 0.49
fired porcelains. The substitution of 11% quartz and 4%
feldspar with 15% scraps does not vary the alkali content,
while about 1.5 wt.% SiO2 is substituted by Al2O3. The BET higher initial porosity of C-15 (28.9  0.6%) with respect to
specific surfaces of both powders (8.3 m2/g for C-0 and C-0 (26.6  0.6%) and indicates that the addition of reject
7.7 m2/g for C-15) are typical for tableware porcelain and decreases the compressibility. This is probably a result of the
porcelain stoneware [5,11,21]. lower specific surface of the C-15 batch. This difference
The usual methods, used for the evaluation of the degree highlights that the same degree of the densification (i.e. the
of sintering, are the linear shrinkage, DL, and the water same total porosity) in C-15 will be attained with about 1%
absorption, W. These methods, however, give incomplete more linear shrinkage, than C-0.
information about densification: W only measures the open The PT and W values, obtained at 1200 8C, show that the
porosity, while DL is a function of the initial green porosity, densification process is in its initial stage: the porosity is
i.e. it may vary for different batches. In the present work, in only open with a value of about 14.5% for both composition;
order to obtain additional information about the closed W values are also similar (5.95% for C-0 and 5.90% for C-
porosity, density measurements were carried out. 15). This demonstrates that, regardless of the presence of an
The values of DL and W, after sintering at different existent glassy phase in the scraps, at this temperature the
temperatures, are presented in Table 3, together with the waste addition does not improve the densification. This may
corresponding real, rr, skeleton, rs, and bulk, rb, densities. be explained by the too high viscosity of the SiO2-rich melt
Fig. 1 shows the variations of the total, PT, and closed, PC, in the rejects.
porosity as a function of the temperature. At 1300 8C, the sintering is significantly improved and
The right evaluation of the porosity depends on the the transformation of open into closed porosity is observed:
precise measurement of the real density. In some cases, rr PT decreases to 6.8% for C-0 and 5.3 % for C-15, while W to
may be assumed as a constant, but in the porcelain, where 0.96% and 0.87%, respectively. The highest densification
the phase composition varies during the sintering, it must be degree of C-15, corresponding to 4.2% PT is attained at
obtained experimentally for each sample. 1350 8C, while sintering at 1375 and 1400 8C yields an over-
In C-0 and C-15, due to the increasing of the amount of firing, leading to increased closed porosity and decreased
glassy phase with temperature rise, the real density shrinkage. At the same time, the C-0 composition has the
decreases. This is a consequence of the fact, that the glassy best firing temperature at 13758 C, with a total porosity of
structure always has a lower density than the corresponding 5.5%; at 1400 8C C-0 shows over-firing.
mixture of crystal phases. For example, rr of the potassium The C-15 batch has a lower specific surface and the C-15
feldspar, a-quartz and mullite have values of 2.54, 2.65 and green has higher initial porosity. Nevertheless, this
3–3.1 g/cm3, respectively [3,27], while the densities of the composition sinters better at 1300–1350 8C and shows
glasses with feldspar, C-0 and C-15 compositions, estimated over-firing at lower temperatures, respectively, C-0. This
by the Appen method [28], were evaluated as 2.39, 2.35 and behaviour may be explained by the variations in the
2.36 g/cm3, respectively. Some rr decreasing may be also mineralogical compositions during heat-treatment.
observed if re-crystallisation quartz-cristobalite occurs In order to investigate the reactions taking place during
[3,29]. the heat-treatment, the sintering process was studied in non-
The results of the green samples show that the addition of isothermal conditions by Dilatometry and DTA-TG [30,31].
fired scraps decreases rb and increases rr. This leads to a Fig. 2 shows the dilatometric traces of C-0 and C-15, while

Table 3
Shrinkage, water soaking, bulk, rb, skeleton, rs, and real, rr, densities of C-0 and C-15 after 1 h holding at different temperatures
Shrinkage (%) Water soaking (%) Bulk density (g/cm3) Skeleton density (g/cm3) Real density (g/cm3)
C-0 C-15 C-0 C-15 C-0 C-15 C-0 C-15 C-0 C-15
Initial – – – – 1.962 1.929 2.689 2.731 2.672 2.711
1200 5.83 7.22 5.95 5.90 2.234 2.241 2.589 2.602 2.616 2.625
1300 8.09 9.28 0.96 0.87 2.406 2.452 2.528 2.533 2.581 2.589
1350 8.24 9.95 0.15 0.02 2.413 2.460 2.473 2.488 2.559 2.567
1375 8.29 9.81 0.05 0.03 2.409 2.436 2.421 2.443 2.550 2.558
1400 7.58 8.39 0.04 0.04 2.368 2.363 2.378 2.372 2.542 2.550
730 A. Karamanov et al. / Ceramics International 32 (2006) 727–732

Fig. 1. Total and closed porosities of C-0 and C-15 as function of the
temperature. Fig. 3. Sintering rate of C-0 and C-15 compositions.

Fig. 3 plots the corresponding derivates (i.e. the sintering variations at 1210–1230 8C were also reported in other
rate versus temperature). The two plots in Fig. 2 are similar porcelain compositions [31].
up to 1150–1200 8C, while at higher temperatures C-15 The DTA results are shown in Fig. 4. The endo-effect in
sinters faster. At 550–600 8C and at 950–1100 8C, the the 500–600 8C range is related to 7–7.5% losses in the TG
samples show variations, related to the low-temperature traces and corresponds to the kaolinite dehydration, while
phase transformations. The low-temperature shrinkage is the exo-peak at 990 8C to the metakaolinite transformations
more evident in C-15, probably, due to its higher initial (i.e. the formation of spinel-type aluminosilicate structure
porosity: DL values at 1100 8C are 0.1% and 0.5% for C-0 and amorphous silica) [5,31]. The endo-effect, due to the
and C-15, respectively. At 1200 8C, the shrinkage is 2.9% for feldspar fusion and the partial quartz dissolution, starts at
C-0 and 3.5% for C-15 (i.e. the difference remains similar), about 1170–1190 8C. Notwithstanding the lower feldspar
while at 1350 8C the C-15 shrinkage (10.3%) is significantly content in the batch, this endo-effect is more evident in C-15,
higher in respect of C-0 (8.4%). These results are in highlighting that the milled porcelain waste accelerates the
agreement with the PT variations, presented in Fig. 1. quartz dissolution. Taking the glassy phase already formed
Fig. 3 shows that at 550–600 8C, as a result of the in the scrap into account, it may be inferred that the more
kaolinite de-hydration and metakaolinite formation, some intensive sintering in C-15, as well as the observed over-
shrinkage is observed, followed by a volume increase at firing at lower temperatures, are due to the formation of a
585 8C, due to the transformation of a-quartz in b-quartz. higher amount of liquid phase.
At 980–1000 8C, the transformation of the metakaolinite This assumption was confirmed by the XRD results of
leads to new shrinkage [30] with a maximum transformation samples, sintered at 1350 and 1400 8C. The spectra are
rate at 985 8C. The densification in both samples starts at plotted in Fig. 5, while the quantitative results are
1100–1120 8C and up to 1170–1190 8C the sintering rate is summarised in Table 4. They show lower crystallinity in
similar. Then, the densification rate increases faster in C-15 the C-15 samples and an increase in the amorphous phase
and at 1250 8C its value is 20% higher than C-0. The with the temperature rise. It is also demonstrated that the
observed decrease of the densification rate at 1220 8C in increase in the glassy phase is due to the dissolution of
both ceramics is probably due to the beginning of formation quartz, while the mullite percentage remains constant.
of secondary mullite. Dilatometric traces with similar Similar results were reported by other authors [4,10,21].

Fig. 2. Dilatometric traces of C-0 and C-15 compositions. Fig. 4. DTA traces of C-0 and C-15 compositions.
 A. Karamanov et al. / Ceramics International 32 (2006) 727–732 731

Table 5
Mechanical properties of C-0 and C-15 after 1 h holding at 1350 8C and
1400 8C
Young modulus (GPa) Bending strength (MPa)
C-0 (1350 8C) 68  3 57  3
C-0 (1400 8C) 67  4 56  4
C-15 (1350 8C) 74  4 67  4
C-15 (1400 8C) 74  2 59  3

(20–400 8C) of 5.72 and 5.26
106 deg1 for C-0 and
5.51 and 4.84
106 deg1 for C-15, respectively.
Table 5 summarises the bending strength and Young
modulus of samples, sintered at 1350 8C and 1400 8C. The
results are similar with standard deviations of the same order
of magnitude as the differences between the four series. The
characteristics of both C-0 series are practically identical. At
the same time, the bending strength of C-15, sintered at
1350 8C has a maximum value, while the samples, held at
1400 8C, show some decreases. It may be concluded that the
addition of scraps has a positive effect on the mechanical
properties and that the variations of the properties are mainly
related to the residual porosity.

4. Conclusions

At sintering temperatures below 1200 8C, the addition of

fired scraps had no significant influence on densification. This
may be explained by the high viscosity of the SiO2-rich melt in
Fig. 5. XRD results of C-0 and C-15 porcelains, sintered at 1350 and
the porcelain waste, which was already fired at 1350–
1400 8C. 1375 8C.
By raising the sintering temperature to 1300–1350 8C,
the molten part of the scrap starts to participate in the
The variation of the quartz amount may also be estimated dissolution of the quartz phase and in the densification
by the changes of the thermal expansion coefficient, TEC, process of the porcelain. As a result, the sintering is
of final ceramics. Due to the high TEC of a-quartz improved and products with lower porosity and higher
(15
106 deg1) and the low ones of mullite (5.3
mechanical properties are obtained.
106 deg1) [3] and formed SiO2-rich glassy phase At 1375–1400 8C, the percentage of liquid phase
(4–5.5
106 deg1) [28], the dissolution of the quartz becomes too high, leading to over-firing and a decrease in
leads to a decreasing of the thermal expansion coefficient. the properties.
In fact, the samples, sintered at 1300 and 1400 8C have TEC In conclusion, the addition of 15% porcelain scrap red-
uces the sintering interval, decreases the sintering tempera-
Table 4 ture and helps to obtain better mechanical properties.
Results of Rietveld–R.I.R. quantitative XRD analysis (wt.%) for C-0 and C-
15 samples
1350 8C 1400 8C Acknowledgements
C-0 C-15 C-0 C-15
The authors are grateful to Eng. G. Blasetti and Eng. S.
Phases Costa from Tecnologia Porcellana Mediterranea Ltd. for the
Amorphous 57.8(1) 61.9(1) 63.6(1) 66.2(1) technical support and to Eng. I. Matekovits and Eng. M.
Quartz 21.0(1) 16.4(1) 16.5(1) 11.7(2)
Aloisi for carrying out part of the experimental work.
Mullite 21.1(4) 21.6(1) 19.8(1) 22.0(1)
Total 99.9 99.9 99.9 99.9
2 References
x 5.27 4.22 4.08 3.33
Rwp 0.060 0.053 0.052 0.048
Rp 0.044 0.039 0.039 0.036 [1] W.D. Kingery, Introduction to Ceramics, J. Wiley & Sons, New York,
1976.
732 A. Karamanov et al. / Ceramics International 32 (2006) 727–732

[2] J.S. Reed, Principles of Ceramic Proceedings, J. Wiley & Sons, New [17] F. Matteucci, M. Dondi, G. Guarini, Effect of soda-lime glass on
York, 1995. sintering and technological properties of porcelain stoneware tiles,
[3] J. Hlavac, The Technology of Glass and Ceramics: An Introduction, Ceram. Int. 28 (2002) 873–880.
Elsevier, Amsterdam, 1983. [18] P. Torres, H.R. Fernandes, S. Agathopoulos, D.U. Tulyaganov, J.M.F.
[4] E. Gerasimov, S. Bachvarov, Technology of the Ceramics Product, Ferreira, Incorporation of granite cutting sludge in industrial porcelain
Technica, Sofia, 1977 (in Bulgarian). tile formulations, J. Eur. Ceram. Soc. 24 (2004) 3177–3185.
[5] W.M. Carty, U. Senapati, Porcelain—raw materials, processing, phase [19] K. Dana, S. Das, S.K. Das, Effect of substitution of fly ash for quartz in
evolution, and mechanical behavior, J. Am. Ceram. Soc. 81 (1) (1998) triaxial kaolin–quartz–feldspar system, J. Eur. Ceram. Soc. 24 (2004)
3–20. 3169–3175.
[6] T. Manfredini, G. Pellacani, M. Romagnoli, L. Pennisi, Porcelainized [20] F. Andreola, L. Barbieri, A. Corradi, I. Lancellotti, T. Manfredini,
stoneware tile, Am. Ceram. Soc. Bull. 74 (5) (1995) 76–79. Utilisation of municipal incinerator grate slag for manufacturing of
[7] Y. Iqbal, W.E. Lee, Microstructural evolution in triaxial porcelain, J. porcelainized stoneware tiles manufacturing, J. Eur. Ceram. Soc. 22
Am. Ceram. Soc. 83 (12) (2000) 3121–3127. (2002) 1457–1462.
[8] Y. Iqbal, W.E. Lee, Fired porcelain microstructures revisited, J. Am. [21] S.M. Olhero, G. Tari, J.M. Ferreira, Feedstock formulations for direct
Ceram. Soc. 82 (12) (1999) 3584–3590. consolidation of porcelains with polysaccharides, J. Am. Ceram. Soc.
[9] C. Leonelli, F. Bondioli, P. Veronesi, M. Romagnoli, T. Manfredini, G. 84 (4) (2001) 719–725.
Pellacani, V. Cannillo, Enhancing the mechanical properties of por- [22] I. Agote, A. Odriozola, M. Gutierrez, A. Santamaria, J. Quintanilla,
celain stonewere tiles: a microsrtuctural approach, J. Eur. Ceram. Soc. P. Coupelle, J. Soares, Rheological study of waste porcelain feed-
21 (2001) 785–793. stocks for injection moulding, J. Eur. Ceram. Soc. 21 (2001) 2843–
[10] O. Isik Ece, Z. Nakagawa, Bending strength of porcelains, Ceram. Int. 2853.
28 (2002) 131–140. [23] http://www.micromeritics.com.
[11] E. Sanchez, M.J. Orts, J. Garcia-Ten, V. Cantavella, Porcelain tile [24] M. Bellotto, C. Cristiani, Quantitative X-ray diffraction Rietveld
composition effect on phase formation and end products, Am. Ceram. analysis of low temperature coal ashes, Mater. Sci. Forum 79–82
Soc. Bull. 80 (6) (2001) 43–49. (1991) 745–750.
[12] G. Stathis, A. Ekonomakou, C.J. Stournaras, C. Ftikos, Effect of firing [25] A.F. Gualtieri, G. Artioli, Quantitative determination of chrysotile
conditions, filler grain size and quartz content on bending strength and asbestos in bulk materials by combined Rietveld and RIR methods,
physical properties of sanitaryware porcelain, J. Eur. Ceram. Soc. 24 Powder Diffr. 10 (1995) 269–277.
(2004) 2357–2366. [26] A.C. Larson, R.B. von Dreele, GSAS: General Structure Analysis
[13] R. Gennaro, P. Cappelletti, G. Cerri, M. Gennaro, M. Dondi, G. System LANSCE, MS-H805, Los Alamos National Laboratory, Los
Guarini, A. Langella, D. Naimo, Influence of zeolites on the sintering Alamos, NM, 1998.
and technological properties of porcelain stoneware tiles, J. Eur. [27] W. Deer, R. Howie, J. Zussman, An Introduction to the Rock-Forming
Ceram. Soc. 23 (2003) 2237–2245. Minerals, second ed., Longman Scientific & Technical, 1992.
[14] M.F. Abadir, E.H. Sallam, I.M. Bakr, Preparation of porcelain tiles [28] A. Appen, Chemistry of Glasses, Himiq, Leningrad, 1974 (in Russian).
from Egyptian raw materials, Ceram. Int. 28 (2002) 303–310. [29] U. Mücke, Ch. Ullner, G. Nolze, Microsrtructure, internal stresses and
[15] L. Tucci, E. Esposito, C. Rastelli, E. Palmonari, Rambaldi, Use of fracture mechanics properties of quartzose silicate materials for white-
soda-lime scrap-glass as a fluxing agent in a porcelain stoneware tile wares, cfi/DKG 78 (7) (2001) 31–36.
mix, J. Eur. Ceram. Soc. 24 (2004) 83–92. [30] J. Blumm, Recent advanced in dilatometry, Am. Ceram. Soc. Bull. 79
[16] S.R. Bragança, C.P. Bergmann, Traditional and glass powder porce- (11) (2000) 64–68.
lain: technical and microstructure analysis, J. Eur. Ceram. Soc. 24 [31] M. Paganelli, Using the optical dilatometer, Am. Ceram. Soc. Bull. 81
(2004) 2383–2388. (11) (2002) 25–30.