SciELO - Scientific Electronic Library Online

vol.5 issue1Transição de Fase Difusa e Propriedades Dielétricas de Cerâmicas Ferroelétricas de (Pb, Ba)Nb2O6Structural Evaluation and Performance of Portland Cement Concretes After Exposure to High Temperatures author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Materials Research

Print version ISSN 1516-1439On-line version ISSN 1980-5373

Mat. Res. vol.5 no.1 São Carlos Jan./Mar. 2002 

Crystallization Mechanism and Kinetics of BaO-Li2O-ZrO2-SiO2 Glasses

Cristian Berto da Silveiraa, Sílvia Denofre de Camposa, Elvio A. de Camposa, Antônio Pedro Novaes de Oliveirab

aDepartamento de Química, C.P.476
bDepartamento de Engenharia Mecânica, Labmat
Universidade Federal de Santa Catarina, 88040-900 Florianópolis - SC, Brazil


Trabalho apresentado no 14° CBECIMAT, Águas de São Pedro, Dezembro de 2000.

Receive: September 27, 2000; Revised: November 24, 2001


Differential thermal analysis and scanning electron microscopy were used to determine the influence of the addition of BaO on the crystallization mechanism of Li2O-ZrO2-SiO2 systems. As the concentration of BaO in the samples increased, a transition occurred in the predominant crystallization mechanism, which passed from superficial to volumetric. To determine the maximum nucleation rate, the crystallization kinetics of the sample containing 20 mole % BaO, which showed the most uniform crystallization, was studied by counting the nuclei with an image analyzer. The first nuclei appeared at the first endothermic inflection point (at the start of Tg, at 440 °C), while the maximum number of nuclei was counted at the midpoint of the glass transition region (446 °C). These results are similar to those observed for other materials that crystallize in volume, and confirm scanning electron microscopy data.

Keywords: glass ceramics, barium oxide, crystallization, Li2O-ZrO2-SiO2 glass

1. Introduction

A variety of materials with specific properties can be produced by controlling glass crystallization. By identifying the crystallization mechanism of a glass-ceramic, its potential uses can be controlled and evaluated. The crystallization mechanism may be of three types: superficial, volumetric or mixed. When superficial crystallization takes place, a crystalline layer is formed on the surface of the material1. The volumetric crystallization mechanism causes crystallization of the material as a whole, while the mixed crystallization mechanism results in a material with the characteristics of both superficial and volumetric crystallization.

The maximum nucleation rate, maximum crystal growth rate and activation energy can be determined by measuring the kinetics of crystallization. The process of crystallization kinetics is favored by the different stages of nucleation and crystal growth during cooling of the glass or by a specific thermal treatment of the already cooled glass under controlled conditions2. The nucleation and crystallization processes are crucial to control the formation of glass-ceramics and to determine their technological applications.

The crystallization mechanism and the kinetic process can be identified through differential thermal analysis (DTA), complemented by microscopy techniques2.

The purpose of this study is to verify the influence of the addition of barium oxide on the crystallization mechanism of Li2O-ZrO2-SiO2 systems and to determine the maximum nucleation rate, using DTA and scanning electron microscopy (SEM).

From the standpoint of cost, these systems are very interesting because of the low fusion temperature used in the process and the fact that silica oxide is the most abundant component in the composition.

2. Experimental

Glasses with general compositions of 30% Li2O: 5% ZrO2: x% BaO: (65% - X SiO2), in which x = 0, 5, 10, 15 and 20 mole % of barium oxide, were synthesized as described in previous reports3.

Differential thermal analysis of the powder and monolithic samples was performed using a Netzsch Termische Analyze STA-409 Cell thermal analyzer with a 10 °C/min heating rate, at temperatures ranging from 25 °C to 1250 °C in a dry air atmosphere. The samples were analyzed in powder form, with particle sizes of 0.315 mm to 0.500 mm, and in monolithic form, using pieces of approximately 30 mg4.

For the microscopy study, the samples were placed on a metal support and covered with a thin layer of gold in a metallizer SEM coating unit E5000 (Polaron Equipment Ltd). The surface morphology of the materials was analyzed through micrographs obtained by scanning electron microscopy (SEM), with a Philips XL 30 microscope, and energy dispersive spectroscopy (EDS), using an EDAX spectroscope.

The samples were subjected to a 60 min. thermal treatment at 600 °C to determine the crystallization mechanism. The pieces were cut in cross sections with a diamond disc and the samples subjected to scanning electron microscopy2.

The glass samples were subjected to a series of 60 min thermal treatments, using nucleation temperatures of 430, 440, 450, 466, 470, 480, 490 and 500 °C, to determine the maximum nucleation rate. The temperatures used for grain growth were 560 °C for 30 min. After thermal treatment, cross sections cut with a diamond disk were forced into an unheated phenolic resin, and smoothed and polished using an aluminum paste with a 1 mm grain size. The material was chemically treated with a mixture of HCl (0.12 in mol.L-1) and HF (0.50 in mol.L-1) for 15 s. The samples were analyzed by SEM and the granules counted with an image analyzer, using the software program AnalySis Pro Version 2.111.

3. Results and Discussion

DTA and SEM were used to determine the crystallization mechanism of the glasses5, and DTA was also used to verify the glass transition (Tg), crystallization (Tc) and melting (Tm) temperatures.

The thermal analysis curves produced by DTA of the samples in powder form are shown in Fig. 1. The glass transition, crystallization and melting temperatures were found to decrease with increasing BaO concentrations (Table 1). These lower temperatures were ascribed to the fact that the Ba2+ ions penetrated the material's structure, weakening the interaction between silicate lattices.





The DTA curves for the samples without BaO and with 5 mole % BaO (Figs. 1a and 1b, respectively) showed similar characteristics: an exothermic peak below 200 °C, associated with the loss of water, an endothermic inflection at 475 °C, which is characteristic of glass transition, a second exothermic peak between 630 °C and 650 °C, associated with the crystallization temperature, and an endothermic peak at close to 1000 °C, corresponding to the melting temperature of the material. The curve of the sample containing 10 mole % of BaO (Fig. 1c) shows a tendency to form a third exothermic peak close to 800 °C, suggesting the crystallization of a second phase.

The curves of the samples containing 15% and 20 mole % of BaO (Figs. 1d and 1e, respectively) are longer, showing new melting peaks at 900 °C and 1100 °C. The tendency to form a second crystallization peak, the lengthening of the curve and the appearance of new melting peaks indicate a heterogeneous crystallization process, which is favored by the higher concentration of barium oxide in the samples6.

Figure 2 illustrates the DTA curves of the monolithic samples. These curves also show a decrease in glass transition, crystallization and melting temperatures as the concentration of BaO in the samples increased. None of the DTA curves for the monolithic samples show an exothermic peak below 200 °C, indicating that the material in this form does not adsorb water, probably because of its smaller surface area than that of the powdered material.



The exothermic peak appearing in the DTA curves of the non-powdered samples between 600 °C and 800 °C is a strong indicator of volumetric crystallization5. The DTA curves for monolithic samples of Li2O-ZrO2-BaO-SiO2 systems (Fig. 2) showed a better defined exothermic crystallization peak and a decrease in the interval of the crystallization temperature with higher concentrations of BaO. This behavior suggests that the predominant crystallization mechanism tends to change from superficial to volumetric5.

A comparison of the crystallization temperatures of the powdered samples with those of the monolithic samples (Table 1) shows a variation of the Tc values. Although this variation is characteristic of surface crystallization, the fact that Tc variations decreased as the concentration of BaO increased reinforces the assumption that a transition from superficial to volumetric crystallization occurred.

Figure 3 shows a micrograph of the sample without BaO obtained by SEM, while Fig. 4 shows the images of the samples with 5% and 10 mole % BaO obtained by the same method. These micrographs reveal the presence of small crystals scattered over the surface of the sample without BaO and, in the case of the sample with 5 mole % BaO, a crystallized surface layer of about 66 mm. Increasing the concentration of BaO from 5% to 10 mole % caused the crystallized layer to grow to about 130 mm.





The SEM micrographs of the samples with 15% and 20 mole % BaO (Figs. 5a and 5b, respectively) show that these samples crystallized completely. The excessive growth of the crystals in the sample containing 15% BaO, which reached a size of 500 mm, caused it to break up into fragments, while the sample with 20% BaO presented a regular volumetric crystallization. The transition of the crystallization mechanism from superficial to volumetric became evident when the BaO concentration increased from 10% to 15%, confirming the DTA results.



The material with 20% of BaO, whose volumetric crystallization was uniform, was used to determine the maximum nucleation rate and to study the crystallization kinetics. The image of a cross section of that sample after being subjected to a 60 min nucleation temperature of 466 °C and to a 30 min grain growth temperature of 560 °C (Fig. 6) reveals the formation of nuclei preceding the formation of crystals.



The graphic representation of the maximum nucleation rate for the sample with 20% BaO, shown in Fig. 7, was obtained by counting the nuclei at different nucleation temperatures. Signs of crystallization were found close to the first endothermic inflection point at 440 °C, while the greatest number of crystals was observed close to the midpoint of the glass transition region at 466 °C. This behavior has already been observed in similar materials that crystallize in volume7.



4. Conclusion

The crystallization mechanism of Li2O - ZrO2 - SiO2 was directly related to the concentration of BaO in the samples. A transition from superficial to volumetric crystallization occurred as the concentration of BaO increased.

The measurement of the crystallization kinetics allowed for the maximum nucleation temperature of 466 °C to be determined for the sample with 20% BaO. This maximum nucleation temperature was close to the midpoint of the glass transition region, which is characteristic of glasses that crystallize in volume.


C.B. da Silveira thanks the Brazilian research funding institution CAPES for the scholarship granted to him.


1.Oliveira, A.P.N. de; Alarcon, O.E.; Manfredini, T.; Pellacani, G.C; Siligardi, C. Phys. Chem. Glasses, v. 41, n. 2, p.100-3, 2000        [ Links ]

2.Folgueras, M.V. Obtenção de Vitro-Cerâmicos Sintetizados a partir da Combinação de Escórias Siderurgicas e Cinzas Volantes, Doctoral Thesis presented to the Technological Center, Post-Graduate Course in Mechanical Engineering - UFSC, 2001.        [ Links ]

3.Silveira, C.B. da; Campos, S.D. de; Castro, S.C. de; Kawano, Y. Mater. Res. Bull, v. 34, n. 10, p. 1661-1671, 1999        [ Links ]

4.Ray, C.S.; Day, D.E. J. Am. Ceram. Soc, v. 80, n. 12, p. 3100-108, 1998.        [ Links ]

5.El-Shennavi, A.W.A.; Morsi, M.M.; Khateer, G.A.; Abtel-Hameed, S.A.M. Journal of Thermal Analysis, v. 51, p. 553-560, 1998.        [ Links ]

6.Stranad, Z. Glass Ceramic Materials: Glass Science Technology 8, Ed. Elsevier, N.Y., p. 145-150, 1986.        [ Links ]

7.Uhlmann, D.R.; Kreidel, N.J. Glass Science and Technology, Academic Press, INC., v. 1, p. 8-35, 1983.        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License