Journal of Thermal Analysis and Calorimetry, Vol. 75 (2004) 467–473
SYNTHESIS AND THERMAL CHARACTERIZATION
OF ZIRCONIUM TITANATE PIGMENTS
C. E. F. Costa 1, Samara C. L. Crispim 1, S. J. G. Lima 2,
C. A. Paskocimas 3, E. Longo 3, V. J. Fernandes Jr. 4, A. S. Araújo 4,
I. M. G. Santos 1 and A. G. Souza 1*
1LTM, Departamento de Química, CCEN, Universidade Federal da Paraíba, 58059-900 João Pessoa,
Paraíba, Brazil
2LSR, Departamento de Engenharia Mecânica, CT, Universidade Federal da Paraíba,
58059-900 João Pessoa, Paraíba, Brazil
3LIEC, Departamento de Química, CCT, Universidade Federal de São Carlos, Rodovia
Washington Luis, Km 235, 13565-905 São Carlos, São Paulo, Brazil
4Departamento de Química, CCEN, Universidade Federal do Rio Grande Norte, Natal, RN, Brazil
Abstract
The objective of this work was to obtain modern pigments based on zirconium titanate doped with
nickel and cobalt using polymeric precursor method. Thermal evolution of ZrTiO4 phase was deter-
mined. After primary calcination, the material still exhibits two thermal composition events. Calci-
nation was done between 500 and 1000°C, to morphological and structural characterization. Surface
area decreased exponentially with temperature increase, while crystallite size increased. ZTCo and
ZTNi pigments were applied on ceramic tiles and the good quality of the pigment was observed.
Keywords: pigment, polymeric precursors, thermal decomposition, zirconium titanate
Introduction
Ceramic pigments are inorganic compounds, the quality of which depends on its opti-
cal and physical properties. The quality of a ceramic pigment is directly related to its
crystalline structure, its chemical composition, purity and some physical characteris-
tics as particle size distribution, particle morphology others.
Some properties may be observed in relation to a ceramic pigment application,
as capability of developing color (pigmenting capability). Among optical properties,
opacity, which indicates the capability of avoiding light transmission through matrix,
is an important one. White pigments diffract all visible light spectrum more effi-
ciently than absorb it. Black pigments behave exactly in the opposite way. The color
of the pigment is due to the absorption of some wavelength of the visible spectra by
the particles, dispersing the rest [1–3].
* Author for correspondence: E-mail: gouveia@quimica.ufpb.br
1388–6150/2004/ $ 20.00 Akadémiai Kiadó, Budapest
© 2004 Akadémiai Kiadó, Budapest Kluwer Academic Publishers, Dordrecht
468 COSTA et al.: ZIRCONIUM TITANATE PIGMENTS
At present, the research about modern pigments is directed to the development
of pigments easier to reproduce. To attain this objective, chemical synthesis is being
used. The chemical methods (sol-gel, Pechini, others) lead to materials with a good
stoichiometric control, high reproducibility, with pigmenting characteristics [4, 5].
Zirconium titanate is a material with high chemical inertia and thermal stability.
Moreover, it exhibits white color, which permits observing any color change, when
zirconium or titanium are replaced by chromophore ions, favoring its use as ceramic
pigment.
The zirconium titanate phase studied in this work is the high temperature one,
that exhibits an orthorhombic unit cell, with a structure similar to α-PbO2 [6].
The synthesis method used in this work was Pechini [7, 8] or polymeric precur-
sors one that leads to particles with an excellent stoichiometric control. The general
idea in Pechini’s method is to reduce the individuality of the different metallic ions
forming a stable neighborhood of the metallic complexes. With the growth of the
polymeric system, this complex is fixed in the rigid net of the organic polymer, de-
creasing metals segregation. When this polymer is calcined at temperatures above
500°C, metallic oxides are formed.
In this work, zirconium titanate doped with cobalt or nickel was synthesized us-
ing Pechini method. Thermal, structural and morphological characterization was
done and activation energy for particle and crystallite growth was calculated.
Experimental
In a beaker under mixing and heating at 80°C, citric acid was dissolved in distilled
water. Later, titanium isopropoxide was slowly added to this solution leading to the
formation of titanium citrate. Zirconium citrate was synthesized in a similar manner,
using zirconium n-propoxide as precursor. Resin was prepared using zirconium ci-
trate and titanium citrate in equimolar amounts. Dopant (nickel nitrate or cobalt ni-
trate) was added to the solution in order to obtain dopant amounts of 0.1, 0.2, 0.5
and 1 mol%. Later, ethylene glycol was added to the solution. This resin was calcined
at 300ºC/1 h, leading to an expanded resin. This material was powdered and calcined
at temperatures varying from 500 to 1000ºC/1 h.
Pigments synthesized in this work were characterized in relation to its thermal
aspects using thermogravimetry and differential thermal analysis. To obtain
thermogravimetric curves (TG), a thermobalance (TGA-50, Shimadzu) was used
with a heating rate of 10°C min–1, air atmosphere with a flux of 20 mL min–1 and sam-
ple mass of about 7 mg. For DTA analysis (DTA-50, Shimadzu), a heating rate
of 10°C min–1 was used with static air atmosphere and sample mass of about 8 mg.
To evaluate structural and morphologic aspects, X-ray diffraction (D-5000,
Siemens), BET surface area analyser (Asap 2000, Micromeritics) and scanning elec-
tronic microscope (SEM) were used.
J. Therm. Anal. Cal., 75, 2004
COSTA et al.: ZIRCONIUM TITANATE PIGMENTS 469
Results and discussion
In all thermogravimetric curves analyzed, two thermal decomposition events are ob-
served: the first one is due to water loss and some gases adsorbed in particle surface;
the second one is due to organic material decomposition and cation oxidation, as it
may be observed in Fig. 1.
The second decomposition event starts at about 280°C and finishes at about 500°C,
Fig. 1. This temperature range is observed for all doping amounts, independently of the
cation used as dopant. The total mass loss of the material varies from 50 to 60%. A small
decrease at decomposition temperature is observed as dopant amount increases.
Fig. 1 TG curves of ZrTiO4 with different quantities and types of dopants
In all samples, DTA curves exhibits two exothermic peaks (Fig. 2). The first one
starts at about 230°C and finishes at about 500°C. The intensity of the first exother-
mic peak indicates a great amount of energy released. This is due to a peculiar and
fundamental characteristic of Pechini method: the use of a large amount of organic
compounds. During calcining, organic compounds exhibit an exothermic reaction as-
Fig. 2 DTA curve of ZrTiO4 doped with 0.5 mol% of cobalt
J. Therm. Anal. Cal., 75, 2004
470 COSTA et al.: ZIRCONIUM TITANATE PIGMENTS
sociated to carbonic chain combustion with a high number of thermal decomposition
reactions due to the different organic matrix. Comparing Figs 1 and 2, it may be ob-
served that the first exothermic peak is associated to a high mass loss.
The second exothermic peak, at about 700°C, is observed in all DTA curves, not
associated to any mass loss. According to Bianco [9], this peak may be attributed to
ZrTiO4 crystallization. This indicates that between 500°C (end of decomposition)
and 700°C (crystallization), the material is amorphous.
All samples (with different quantities and types of dopants) exhibit only one
crystalline phase, the high temperature phase of zirconium titanate, without an or-
der–disorder transition, as observed by Bianco [10]. The beginning of crystallization
is observed at 600°C and crystallinity increases gradually up to 1000°C, the tempera-
ture of maximum crystallinity obtained in this work. This is observed for nickel and
cobalt, in all dopant amounts (Figs 3 and 4). An important point is that no intermedi-
ate phase is observed.
Fig. 3 X-ray difractograms of ZrTiO4 doped with 1.0 mol% of cobalt, calcined at 600,
800 and 1000°C
The temperature of crystallization beginning to be observed by X-ray diffraction
(Fig. 3) is different from the temperature observed in DTA curves (Fig. 2). This is due
to the different analysis conditions, as DTA is realized in a dynamic condition (with a
heating rate of 15°C min–1), while samples for XRD are calcined at 600°C for 2 h.
Crystallite size was calculated using the full width to half maximum of the main
peak of the X-ray diffractogram and the Scherrer equation. It is observed that the higher
the calcining temperature, the higher the crystallite size (Table 1). This effect is due to
the fact that crystallite growing is thermally activated. First, crystallite nucleation oc-
curs followed by its growth, which depends on calcining temperature. This behavior is
observed for ZrTiO4 doped with nickel or cobalt, in all doping percentages.
For nickel, crystallite size at 600°C is almost the same for all dopings, but exhib-
its a random variation at other temperatures. For cobalt, a difference of crystallite size
of about 23% is observed at the same temperature. Samples with 0.1 mol% of cobalt
have a smaller crystallite size than samples with 0.2 and 1.0 mol%. Samples with 0.2
and 1.0 mol% do not exhibit meaningful differences between crystallite size with al-
most superimposed results.
J. Therm. Anal. Cal., 75, 2004
COSTA et al.: ZIRCONIUM TITANATE PIGMENTS 471
Table 1 Crystallite size of ZrTiO4 doped with 1.0 mol% of the transition metals
Dopant amount/mol% Co/nm Ni/nm
0.1 206.50 145.08
0.2 225.70 198.57
1.0 234.37 205.18
Surface area results (SBET) show that an exponential decrease of SBET with tem-
perature increase is observed for all dopings, independently of the chromophore ion.
This decrease indicates that a sintering among agglomerates or a crystallite growth
may be occurring. For samples doped with nickel, no meaningful change in surface
area in relation to dopant quantity was observed and curves are almost superimposed,
specially over 800°C. At 500°C, surface area variation in relation to doping is ran-
dom. For samples doped with Co, a small decrease in surface area (and increase in
particle size) with dopant increase is observed. SBET curves in relation to temperature
have the same profile, independently of the amount of dopant.
In Table 2, particle size results are presented for samples with 1 mol% of nickel
and cobalt. It may be observed that samples doped with Co exhibit the highest parti-
cle size. This also indicates that SBET does not depend on the presence and type of
dopant at higher temperatures.
Table 2 Morphological properties of the pigments with 1.0 mol% of Ni and Co
Ni Co
Temp./°C
*D **p/nm Dc/nm Dp/Dc Dp/nm Dc/nm Dp/Dc
600 17.49 14.72 1.19 18.68 17.46 1.07
700 36.74 17.69 2.08 41.53 19.04 2.18
800 66.90 18.88 3.54 94.46 20.15 4.69
900 133.49 19.20 6.95 134.25 20.55 6.53
1000 280.21 20.52 13.65 270.38 23.44 11.53
*D **p – particle size, Dc – crystallite size
According to Table 2, it may also be observed that particles studied exhibit a
particle increase higher than a crystallite increase. This indicates that a sintering
among particles is probably occurring, as confirmed by microscopy analysis (Fig. 4).
Using Arrhenius equation (lnD=lnD0–Ea/RT), activation energy for particle and
crystallite growth was calculated. A graph of lnD as a function of 1/T was done,
where D is the particle or crystallite size (Fig. 5).
As it may be observed in Tables 3 and 4, activation energy for particle growth is
higher than for crystallite growth, as diffusion among crystallites is favoured by
smallest distance among them. In spite of this, particle size increases about 15 times,
while crystallite size increases less than 1.5 times. This may be due to sintering
J. Therm. Anal. Cal., 75, 2004
472 COSTA et al.: ZIRCONIUM TITANATE PIGMENTS
among particles, leading to a highest increase in their size. Comparing activation en-
ergies of zirconium titanate doped with Ni and Co, it may be observed that oxides
doped with Ni exhibit higher activation energy values than oxides doped with Co. In
this sense, ZrTiO4 doped with Ni needs a higher temperature for particle and crystal-
lite growth.
Fig. 4 SEM image of ZrTiO4 doped with 1.0 mol% of nickel calcined at 900°C
Fig. 5 Arrhenius plot for activation energy calculation: a – material doped with Ni,
b – material doped with Co
J. Therm. Anal. Cal., 75, 2004
COSTA et al.: ZIRCONIUM TITANATE PIGMENTS 473
Table 3 Activation energy for particle growth
Material Activation energy for particle growth/J R ∆Ea/%
ZrTiO4 doped with 1% of Ni 62.684 0.992 7.2
ZrTiO4 doped with 1% of Co 7.266 0.996 5.2
Table 4 Activation energy for crystallite growth
Material Activation energy for crystallite growth/J R ∆Ea/%
ZrTiO4 doped with 1% of Ni 7.108 0.960 16.7
ZrTiO4 doped with 1% of Co 6.069 0.961 16.6
Conclusions
Zirconium titanate can be synthesized using the polymeric precursor method with
dopant percentages up to 1 mol% of transition metals (Co and Ni), to obtain only one
phase. Materials with a relatively high surface area were obtained. Above 700°C, an
increase in surface area was observed due to the sintering among particles. Activation
energies for particle growth was higher than for crystallite growth, probably due to
proximity among crystallites, making diffusion easier.
* * *
Authors acknowledge CAPES and CNPq for scholarships and financial support.
References
1 F. Schejbal, D. Vougltanz and B. Camara, Cerámica Información, 220 (1996) 33.
2 P. E. Lópes, J. B. Carda and E. Corconcillo, Esmaltes y Pigmentos Cerámicos,
Faenza Editrice Ibérica, S. L., Castellón 2001.
3 A. Burgyan and R. A. Eppler, Ceramic Bulletin, 12 (1982) 1001.
4 G. Monros, J. Carda, M. A. Tena, P. Escribano and J. Alarcon, Bol. Soc. Esp. Ceram. Vidr.,
29 (1990) 25.
5 G. Monros, J. M. Carda, A.Tena, P. Escribano, M. Sales and J. Alarcon, J. Non-Cryst., Solids,
147–148 (1992) 588.
6 C. Gheorghies, J. Crist., Growth, 213 (2000) 112.
7 M. P. Pechini, U.S. Patent No 3 330 697 (1967).
8 D. S. Gouveia, A. G. Souza, M. A. M. A. de Maurera, C. E. F. da Costa, I. M. G. Santos,
S. Prasad, J. B. de Lima, C. A. Paskocimas and E. Longo, J. Therm. Anal. Cal., 67 (2002) 459.
9 A. Bianco, M. Paci and R. Freer, J. Eur. Ceram. Soc., 18 (1998) 1235.
10 A. Bianco, G. Gusmano, R. Freer and P. Smith, J. Eur. Ceram. Soc., 19 (1999) 959.
J. Therm. Anal. Cal., 75, 2004