New Silicate Glass-Ceramic Materials and Composites

Advances in Science and Technology Vol. 68 (2010) pp 1-12 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.68.1

New silicate glass-ceramic materials and composites Dachamir Hotza1,a and Antonio Pedro Novaes de Oliveira2,b 1

1

Department of Chemical Engineering (ENQ), Federal University of Santa Catarina (UFSC), 88040-900 Florianópolis, SC, Brazil

Department of Mechanical Engineering (EMC), Federal University of Santa Catarina (UFSC), 88040-900 Florianópolis, SC, Brazil a

b

[email protected], [email protected]

Keywords: glass-ceramics, pressing, injection moulding, extrusion, casting, rapid prototyping, replication.

Abstract. New silicate glass-ceramic compositions have been investigated due to their interesting chemical, mechanical, thermal, and electrical properties. LZSA glass-ceramics based on βspodumene (Li2O·Al2O3·4-10SiO2) and zircon (ZrSiO4) crystalline phases have shown good chemical resistance, high bending strength as well as high abrasion resistance, when compared with traditional ceramic materials, and coefficient of thermal expansion from 4.6 to 9.1×10-6 °C-1. These features basically depend on the nature, size and distribution of the formed crystals as well as on the residual glassy phase. The nature of the formed crystalline phases and consequently the final properties can be controlled by modifying the chemical composition of the parent glass and also by adequate selection of the heat-treatment parameters. The classical fabrication of glass-ceramic materials consists on the preparation of monolithic glass components followed by heat treatments for crystallisation. However, this technology requires high investments and can be justified only for large production. A viable alternative could be the production of glass-ceramics processed from glass powders and consolidated by sintering using the same equipments of traditional ceramic plants. This work reports the manufacturing and characterization of glass-ceramic materials and composites processed by pressing, injection moulding, extrusion, casting, replication, and rapid prototyping. Introduction Glass–ceramics are relatively new materials specially used due their specific properties such as high bending strength, high abrasion resistance, high hardness and wide range of coefficient of thermal expansion (CTE), which yields to high thermal shock resistance and high chemical resistance. These features basically depend on the nature, size and distribution of the formed crystals as well as on the residual glassy phase [1]. The nature of the formed crystalline phases and consequently the final properties can be controlled by modifying the chemical composition of the parent glass and also by adequate selection of the heat-treatment parameters. Silicate-based glass-ceramics are interesting not only by their properties but also because of the possibility to produce them using low cost raw materials like residues from steel industry, glass wastes and fly ashes, which can be transformed into products with optimized properties for a given application [2-4]. The characteristic properties and relative stability of zirconia and alumina, which give rise to zircon and β-spodumene and to lithium-based glass according to preliminary studies [5], enable homogeneous glasses belonging to the Li2O–ZrO2–SiO2–Al2O3 (LZSA) system to be obtained as precursor materials for glass-ceramics. The usual way for fabrication of glass-ceramics materials consists in the application of glass forming techniques to produce monolithic articles and subsequent crystallisation [6,7]. However, this technology requires high investments and can be justified only for high production volumes. On the other hand, the production of glass-ceramic materials processed from glass powders and consolidated by sintering and crystallisation seems to be a valid alternative. In this case, it is All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the publisher: Trans Tech Publications Ltd, Switzerland, www.ttp.net. (ID: 187.65.209.181-11/09/10,21:50:33)

2

12th INTERNATIONAL CERAMICS CONGRESS PART G

possible to use the same equipments of a ceramic plant for the production of components with very complicated shapes [2,8-10]. The process involves the following steps: (i) glass melting and cooling for producing parent glass frit; (ii) comminuting the frit into powder; (iii) forming by powder technology into many shapes and sizes; (iv) thermal treatment involving sintering and crystallisation to obtain a dense body. Ceramics processing may be usually classified according to forming techniques into three main routes: (i) pressing (both axial and isostatic); (ii) plastic forming (including extrusion, injection moulding, roll pressing, among others); and (iii) casting (slip, tape or gel casting, and other related processes from suspensions or slurries) [11]. All of these techniques may be applied to parent glass powders which will be further sintered and crystallized, making this approach a flexible and powerful tool for producing innovative, low-cost products. This work reviews the manufacturing and characterization of LZSA glass-ceramic materials and composites processed by axial pressing, roll pressing, extrusion, injection moulding, tape casting, replication of polymeric foams and rapid prototyping. Experimental Procedure Glass Preparation and Milling. A glass composition belonging to the system Li2O-ZrO2-SiO2Al2O3 (LZSA) was prepared from commercially available alumina, lithium carbonate, spodumene, quartz and zircon (Colorminas) [12]. The chemical analysis, obtained by atomic absorption spectroscopy (Unicam 969) and by X-ray fluorescence (Philips PW 2400) is presented in Table 1. A batch to produce ~150 kg of glass was placed in a mullite crucible and melted at ~1500°C for 7 h in a gas furnace. The melt was quenched in water and dried. The glass frit was milled for variable periods of time in an aluminous porcelain mill containing alumina grinding media and water. When necessary, a commercial bentonite (Colorminas, average particle size ~4.5 µm) was added to the glass powder to serve as binder [12]. Chemical analysis of bentonite is shown in Table 1, as well. In order to obtain composite compacts by extrusion, commercial zircon (ZrSiO4, average particle size ~4.5 µm) were added to the glass powder. Table 1 Chemical composition (wt.%) of LZSA parent glass and bentonite. LZSA glass Bento -nite

SiO2

Al2O3

ZrO2

Li2O

K2O

Na2O

TiO2

Fe2O3

CaO

MnO

MgO

P2O5

59.4

13.6

15.6

8.6

0.3

0.7

0.1

0.2

0.6

-

0.02

0.82

62.8

20.3

-

-

0.5

2.4

0.1

3.8

1.2

< 0.1

2.3

0.2

Axial Pressing. The powder was formed by uniaxial pressing in a hydraulic press at 40 MPa (1.4 g/cm3 green density) in a steel die (75×25×5 mm). [13] In order to determine the sintering behaviour, the resulting slurry was dried to a water content of around 5.5% and then granulated in a plastic cylinder for 10 min. Granules with 2 mm size were obtained. The granulated glass was first uniaxially pressed using an automatic hydraulic press at 40 MPa (1.45 g/cm3 green density) in steel moulds. Compacted samples of 13 mm diameter and 5 mm thickness were dried at 120°C for 2 h and cooled and stored in a dryer, so that the final water content was constant at about 0.5%. [19] Roll Pressing. Bentonite (7 wt.%) was added to the glass powder and humidified with water (18 wt.%). Subsequently, the mixture was stored for 12 h to homogenize and then it was extruded (Netzsch MA 01) into samples measuring 150×25×9 mm, and roll pressed (Lieme CS-400) into 4 to 5 mm thick samples. Drying was performed in an electrical dryer (60ºC, 2 h). [12] Extrusion. LZSA-ZrSiO4 mixtures ranging from 100%-0% to 30%-70%, in weight, respectively, were prepared using 7 wt.% bentonite and 23 wt.% water. Subsequently, the samples were stored for 12 h to moisture homogenization and then extruded (Netzsch, MA 01) into prismatic samples

Advances in Science and Technology Vol. 68

(80×25×5 mm or 100×50×10 mm). After 48 h at 20ºC extruded samples were dried at 110ºC for 2 h. [14] Injection Moulding. The feedstock consisted of the parent glass powder, polypropylene (PP, OPP Petroquímica, H301, melting point 170ºC), paraffin wax (PW, Gewax, 145 P, melting point 56ºC) and stearic acid (SA, Sotex). Four compositions were tested, with glass powder ranging from 50 to 60 wt.%, PP from 35.25 to 28.20 wt.%, PW from 11.75 to 9.40 wt.%, and SA from 3.00 to 2.40 wt.%. The parent glass and additives were mixed while heated in a coupled mixer. The mixtures were then introduced into an injection moulding machine (Arburg, 320 S 500-150), at a pressure of 500 bar, injection speed of 100 mm/s, and temperature of 190ºC into prismatic samples (200×4×24 mm). Binder removal was accomplished by (i) solvent extraction (hexane, 60ºC, 4 h) and (ii) thermal burnout (420ºC, 30 min). [15,16] Tape Casting. A polyvinyl alcohol solution (PVA, 31.5 wt.%, Mowiol 4-88, Kuraray) was used as binder, polyethylene glycol as plasticizer (PEG, PEG 400, Synth) and a blend of modified fatty and alkoxylated compounds as antifoamer (Agitan 354, Munzing). Two commercial ammonium polyacrylates (NH4PA) were tested as dispersants to the glass precursor aqueous suspensions (Darvan C, Vanderbilt; Dolapix CA, Zschimmer & Schwarz). The preparation was carried out in three stages: (i) dispersion of the parent glass powder in distilled water with the dispersant for 24 h; (ii) addition and homogenization of the binder solution for 12 h; (iii) addition of the plasticizer and antifoamer followed by a mixing period of 12 h. The suspensions were cast onto non-covered polyethylene terephthalate (PET) films using a laboratory tape caster with a double doctor blade system, with a casting velocity of 450 mm/min at room temperature. The cast tapes were dried for 48 h. [17,18] Laminated Object Manufacturing (LOM). Cast LZSA tapes were laminated with CW–CO2 laser equipment (Helisys 1015). A retract from 0.10 mm was used, which is related to the distance between the heated roller and the sample surface. Generally, the lower the retraction, the higher the pressure applied. The laser power was 16.8 W. The cutting and roller speeds were 50 and 25 mm/s, respectively. The roller temperature was kept constant at 80ºC. The tapes were laminated using a 5 wt.% aqueous solution of the binder as adhesive agent, applied with a painting brush on the richest organic side from each tape before stacking the subsequent one. Rectangular bars with 270×35×25 mm were obtained as result of 20 green-stacked tapes. [17,18] Replication. A slurry was prepared with two different solvents, water and isopropanol, containing 60 wt.% parent glass powder (LZSA), 1–10 wt.% bentonite used as a binder, and 0.5–2.0 vol.% sodium silicate (Merck, Natronwasserglas 105621), used as a dispersant. The solvent, water, or isopropanol, was first mixed with sodium silicate in a plastic vessel with alumina balls as grinding media for 12 h. Subsequently, the glass powder was added to the slurry and then milled for another 12 h. Bentonite was added to the slurry milled for another 12 h. Commercial polyurethane foams with a monomodal pore size distribution of 55 pores per inch (ppi) and porosity of 95% were used as templates. The polymeric foams were cut into pieces of ~250×30×20 mm and immersed in the LZSA parent glass-ceramic slurry. The impregnated foams were slightly compressed to remove the excess slurry and dried at room temperature for 24 h. [19,20] Heat Treatment. The samples were heated treated stepwise to promote burnout of organics – when necessary, sintering and crystallisation of the glass powder at variable temperatures, holding times and heating rates, summarized in Table 2. In all cases, an electrical furnace in air atmosphere was used. Powder Characterisation. The average particle size distribution of the glass powder was measured by a laser scattering particle size analyzer (Cilas 1064L [13] or Malvern Mastersizer 2000 [18]). The density of the parent glass was measured in powder samples by He-pycnometry (QuantaChrome MVP-4DC [15,16] ). Micromeritics AccuPyc 1330 [17-20]. The specific surface area was determined by the BET method (Micromeritics ASAP 2000 [18]). Slurry Characterisation. Zeta potential measurements were performed in casting slurries using a zeta potential analyzer (Zetameter 3.0). Diluted aqueous suspensions of LZSA glass powder (0.02

3

4

12th INTERNATIONAL CERAMICS CONGRESS PART G

vol.%) with and without 1 vol.% dispersant were tested. Potentiometric titration was performed by an automatic unit. The signal was measured as a function of pH at 3 min time intervals by adding 0.05 mL of 1N HCl and 1N NaOH titrants, depending on the initial pH of the pure suspensions [18]. Viscosity measurements were carried out at 25°C in a rotational rheometer (Paar Physica UDS 200 [17,18]). The effect of the dispersants was also evaluated in suspensions with 70 wt.% solids [18]. Table 2 Heat treatment parameters for LZSA samples. Forming Process [Reference]

Organics Burnout Temperature (ºC)

Holding Time (min)

Sintering/Crystallisation Heating Rate (ºC/min)

Temperature (ºC)

Holding Time (min)

Uniaxial Pressing – – – 685-1020 0-180 [13] Roll Pressing – – – 850-1030 10 [12] Extrusion – – – 725-1300 10 [14] Injection Moulding 420 30 0.5 650-850 15-120 [15,16] Tape Casting, LOM 525 * * 700-850 30-60 [17,18] Replication 400-450 1 1 700-900 10 [19,20] *Constant weight loss program of 0.005–0.01%/min. The total binder pyrolysis occurred at 525ºC for 8 h.

Heating Rate (ºC/min)

5 10 10 20 – 5

Thermal Analyses. The glass transition, crystallisation and melting temperatures were determined by differential thermal analysis (DTA). The thermograms were obtained in a simultaneous thermal analyser (Netzsch STA-409), at a heating rate of 5 [12,13] or 10°C/min [1416,19] in dry air. Thermal linear shrinkage and coefficient of thermal expansion of compacted and sintered samples, respectively, were measured using a dilatometer (Netzsch DIL 402PC) at 10°C/min in air, using alumina as reference material [12,14,16]. Density. The theoretical density of the sintered samples was measured by He-picnometry and the apparent density was measured by the Archimedes principle at 20°C. The relative density was calculated from the relation between the theoretical density and the apparent density [12,14]. Mechanical Properties. Three-point bending strength measurements of sintered LZSA samples were performed in a universal testing machine (either EMIC DL 2000[12,14,16] or Instron 4204 [19]) according to ISO 10545-4, at room temperature. The compression strength of glass-ceramic foams was determined at room temperature as well (Instron 4202 [20,21]). Microstructural Analysis. To investigate the crystalline phases formed during heat-treatments, powdered samples were analysed with a X-ray diffractometry (XRD, Philips PW 3710) with Cu Kα radiation. [14,20] Heat treated samples were cut and prepared for scanning electron microscopy (SEM, Philips XL-40) [14]. Results and Discussion Powder Characterization. Table 3 summarizes relevant features of the LZSA parent glass powder. The dispersion in values is related to milling procedures and measuring equipments. The mean particle size is lower for higher milling times (Fig. 1). The morphology of LZSA parent glass powder after 13 days milling is shown in Fig. 2. Agglomerates can be seen, as well as sharp-edged, irregular particles, ranging from 2 to 8 µm. Slurry Characterisation. Figures 3 and 4 show the variation of zeta potential of LZSA glass powder as a function of pH. The isoelectric point (IEP) for the LZSA glass suspension ranged from pH 2.8 to 4.3 [20,18]. The IEP corresponds to the state where a neutral charge surface is achieved. With addition of 5 wt.% bentonite, the IEP of the suspension shifted to pH 3.6 [20]. The addition of 1 wt.% of any dispersant produced no significant effect on either the IEP or the zeta potential of the

Advances in Science and Technology Vol. 68

5

suspensions [18]. A stable suspension can be reached for zeta potential values lower than -10 mV. The pH values of the LZSA suspensions during processing were around 9.5. At this pH, all suspensions exhibited a deflocculated state [20]. Table 3 Properties of LZSA parent glass powder. Specific Surface Area (m2/g)

Mean Particle Size, d50 (µm)

Forming Process [Reference]

Milling time (h)

Uniaxial Pressing [13] Roll Pressing [12] Extrusion [14] Injection Moulding [15,16] Tape Casting, LOM [17,18] Replication [19,20]

3

3.8

20

5.0

Density (g/cm3)

2.67 72

18.6

3.5

312

8.3

2.1

2.58

72

4.9

3.2

2.63

Fig. 1 Particle size distribution of the LZSA powder for different milling times. [18]

Fig. 3 Zeta potential of the LZSA parent glass aqueous suspension with/without bentonite. [20]

Fig. 2 SEM micrograph of the LZSA powder after 13 days milling. [18]

Fig. 4 Zeta potential of the LZSA parent glass aqueous suspension with/without dispersant. [18]

In ceramic wet processing, a high solids content is desired once it decreases the energy consumption, and some shrinkage problems during the drying process can be avoided. Fig. 5 shows the shear curve of

6

12th INTERNATIONAL CERAMICS CONGRESS PART G

suspensions containing 50, 60, and 70 wt.% solids without dispersant. At lower solids loading (50 wt.%), the suspension viscosity was too low and therefore inadequate for the tape casting process, because high volumetric shrinkage can occur. As the solids loading increased to 60 wt.%, an increase in the suspension viscosity was observed. With 70 wt.% solids, the suspension showed a remarkable thixotropy and dilatant behaviour. This time-dependent behaviour was observed even when 1 wt.% of dispersant was added to the suspension (Fig. 6). [18]

Fig. 5 Shear stress versus shear rate curves of LZSA powder suspensions for different solids loadings. [18]

Fig. 6 Shear stress versus shear rate curves of LZSA powder suspensions with 1 wt.% dispersant and 70 wt.% solids [18]

Thermal Analyses. In Table 4, the temperatures related to glass transition, crystallisation and melting of crystalline phases, respectively, are summarized. Dispersions in values are due to different heating rates and/or powder characteristics. Table 4 Glass transition, crystallisation and melting temperatures of LZSA parent glass powder. Forming Process [Reference]

Heating Rate (°C/min)

Glass Transition Temperature (°C)

Crystallisation Temperature (°C)

Melting Temperature (°C)

Uniaxial and Roll Pressing [13, 12] Extrusion [14] Injection Moulding [15,16] Replication [19]

5

~530

~680

~1050

~780

~1150

10 10

~600

~800

~900

10

~600

~760

~850

A typical differential thermal analysis (DTA) curve is shown in Fig. 7 [15]. When a reinforcing crystalline phase (ZrSiO4) fraction increased, the crystallisation peak intensity slightly decreased (Fig. 8) [14]. This occurred due to the decrease of the glass–ceramic matrix fraction with respect to the reinforcing phase, which means that lower crystal fractions were formed from the glass-ceramic matrix. Moreover, it can be said that the reinforcing crystalline phases did not interfered in the crystallisation peak position.

Advances in Science and Technology Vol. 68

Fig. 7 DTA curve of LZSA indicating the glass transition (Tg), crystallisation (Tc) and melting temperature (Tm) at a heating rate of 10°C/min. [15]

Fig. 8 DTA curves of LZSA glass-ceramic (A), and composites (B), 10 wt.%; (C), 20 wt.%; (D), 30 wt.%; (E), 40 wt.%; (F), 50 wt.%; (G), 60 wt.%; (H), 70 wt.% ZrSiO4, respectively. [14]

A typical linear thermal shrinkage curve is shown in Fig. 9 [12]. Accordingly, densification starts at ~850°C and is completed up to 950°C. Crystallisation starts soon after the completion of densification. On the other hand, when a zircon is added as reinforcement (Fig. 10), densification was apparently affected, i.e., the thermal shrinkage decreased and the shrinkage rate for all studied compositions tends to zero as the temperature increased. This behaviour may be related to the amount reduction of the formed viscous liquid phase as zircon was added. [14]

Fig. 9 Typical thermal linear shrinkage curve of the LZSA glass-ceramic. [12]

Fig. 10 Thermal linear shrinkage curves of LZSA glass-ceramic (A), and composites (B), 10 wt.%; (C), 20 wt.%; (D), 30 wt.%; (E), 40 wt.%; (F), 50 wt.%; (G), 60 wt.%; (H), 70 wt.% ZrSiO4. [14]

The coefficients of thermal expansion (CTE) of the final glass-ceramics are shown in Table 5. For samples prepared at the same conditions, there is no remarkable dispersion of CTE values, which are around 4×10-6 °C-1 [12], even for composite samples with up to 60 wt.% zircon addition [14]. Differences in CTE values are, however, noticeable for samples heated at lower temperatures and longer times [16].

7

8

12th INTERNATIONAL CERAMICS CONGRESS PART G

Table 5 Coefficients of thermal expansion of the final LZSA glass-ceramics. Forming Process [Reference]

Sintering Temperature (°C)

Holding Time (min)

CTE (× ×10-6 °C-1)

Roll Pressing [12]

850 900 950 1000 1030 1150 650 700 750 850

10

4.1 4.0 4.0 4.0 3.9 3.9 10.2 7.8 5.6 6.1

Extrusion [14]* Injection Moulding [16]

10 60

* 40 wt.% LZSA, 60 wt.% ZrSiO4.

Density. Relative densities values increased from about 77% at 850°C to about 96% at 950°C for LZSA samples (Fig. 11) [12]. In this case, the number and size of pores are affected by the amount of crystals and their sizes, since, for samples sintered at 1000-1030°C crystal growing probably took place. In the case of a composite LZSA-zircon, Fig. 12, the relative density at 1150°C was ~95%, which is higher than that for the same composition without bentonite addition. This can be associated to a relatively high content of alkalis in bentonite that contribute to the increase of the amount of viscous liquid phase. Moreover, Fig. 12 shows that densification started at about 650°C and its rate was reduced at 700ºC, probably due to the crystallisation process. [14]

Fig. 11 Relative density of LZSA versus heating temperature. [12]

Fig. 12 Relative density of composite (40 wt.% LZSA, 60 wt.% ZrSiO4) versus heating temperature. [14]

Mechanical Properties. Table 6 summarizes typical values of mechanical strength obtained for LZSA samples manufactured by different shaping methods. The dispersion in values is related to processing (forming, sintering) and/or measuring procedures. For roll pressed samples, it can be seen that results close agree with the sintering and crystallisation behaviour observed, i.e., higher sintering temperatures (850-1030°C) produces a higher crystallinity. Consequently, for temperatures higher than 950°C, a decreasing on densification may be observed, reflecting on mechanical strength [12]. Extruded composite samples presented bending strength variations resulting from compositional and sintering temperature differences. Although there is a relative scatter on the data, the mechanical strength tends to increase with higher amounts of ZrSiO4 fraction. This is probably due to the homogeneity of the dispersion of crystals. Additionally, the presence of finer grains and lower porosity of the composition 40 wt.% LZSA, 60 wt.% ZrSiO4 improved the interaction between the matrix and the reinforcement, which resulted in the highest strength (180 MPa) [14].

Advances in Science and Technology Vol. 68

9

For injection-moulded samples, bending strength test results showed close agreement with the sintering and crystallisation behaviour observed. The highest value (197 MPa) was obtained after sintering at 700°C for 15 min. Increased sintering temperature (650–850°C) produced higher crystallinity. Therefore, there was a decrease of densification that could be appreciated from lower density results and, consequently, lower strengths [16]. Sintered laminates presented bending strengths from 90 to 107 MPa. Tapes with higher tensile strength led to laminates with higher bending strength. No interaction was observed concerning green density and bending strength of sintered laminates [17]. The compressive strength of LZSA foams (0.1–10 MPa) is strongly dependent on their overall porosity and their behaviour could be explained using the Gibson-Ashby model. [21] Moreover, the mean strength was 1.1 MPa for a LZSA foam prepared with isopropanol and 0.5 MPa for the LZSA foam prepared using water as a slurry solvent. The struts of the foam prepared with organic slurry are thicker, which confers these foams a higher mechanical strength. The decrease in the strength of the foam prepared with aqueous slurry is due to the increase in its total porosity, which reflected the non-homogeneity of the particle coating during impregnation [20]. Table 6 Mechanical strength of LZSA sintered samples. Forming Process [Reference] Roll Pressing [12] Extrusion [14] Injection Moulding [16] Tape Casting, LOM [17] Replication [20,21] *Load rate of 1 MPa/s.

Mode of Force Application

Sample Dimensions (mm)

Testing Speed (mm/min)

Mechanical Strength (MPa)

Bending (3-point test) Bending (3-point test) Bending (3-point test) Bending (3-point test) Compression

100×50×10

*

65–110

100×50×10

*

60–180

30×5×3

1

100–197

270×35×25

0.5

90–107

20×12×12

1

0.1–10

Microstructural Analyses. Fig. 13 shows an XRD pattern from the LZSA sintered at 900ºC for 10 min. The reflections associated with the sintered sample were assigned to the crystalline phases of zirconium silicate ((ZrSiO4, JCPDS 6-266), lithium metasilicate (Li2SiO3, JCPDS 29-828), and β-spodumene (LiAlSi2O6, JCPDS21-503). The β-spodumene phase shows a low CTE. Although zirconium silicate shows a higher CTE, this phase contributes to a high abrasion and chemical resistance of the material. Lithium metasilicate was detected as a minor phase. [20] Fig. 14 shows XRD patterns related to composite samples (40 wt.% LZSA, 60 wt.% ZrSiO4) sintered at 1125, 1150 and 1175°C for 10 min and at 1175°C for 60 min, respectively. With increasing temperatures, the peak diffraction intensities related to the β-spodumene crystalline phase increased, while those associated to the zircon crystalline phase decreased. With the increase on the holding time (Fig. 14(d)) no significant variation in the X-ray diffractions was observed. [14]

10

12th INTERNATIONAL CERAMICS CONGRESS PART G

Fig. 13 XRD patterns of LZSA sintered at 900°C for 10 min. M= lithium metasilicate; Z= zirconium silicate; S= β-spodumene. [20]

Fig. 14 XRD patterns of composites sintered at (a)1125ºC, (b) 1150ºC, (c) 1175ºC for 10 min, and (d) at 1175ºC for 60 min. E= β-spodumene; Z= zirconium silicate; M= lithium metasilicate, Zo= zirconia. [14]

Fig. 15 shows typical micrographs of samples sintered at 1125 and 1150 for 10 min. The zircon particles exhibit good interaction with the glass–ceramic matrix, i.e., zircon particles were well wetted by the matrix so that the formed interface exhibited low noticeable porosity. The particles are homogeneously distributed over the glassy matrix for all observed samples. [14]

Fig. 15 SEM micrographs of composites sintered at (a) 1125ºC, (b) 1150ºC for 10 min. [14] Finally, as examples of the flexibility of shapes and applications of glass-ceramics obtained through the powder processing route, two products are presented. In Fig. 16, the morphology of glass-ceramic replicated foams was analysed by SEM (Fig. 16). After heat treatment, the polymeric and the glass-ceramic foams presented very similar microstructures. Those structures may be applied to radiant burners or catalyst substrates. [19]

Advances in Science and Technology Vol. 68

(a) (b) Fig. 16 SEM micrographs of (a) PU sponge, (b) glass-ceramic foam. [19] Fig. 17 presents two gear wheel samples fabricated by LOM from LZSA tapes. Each part consists of approximately 20 laminate tapes. After heat treatment, 20% shrinkage was observed in all directions. No surface flaws or inhomogeneous areas were detected. The sintered part maintained the curved edges and internal profile after heat treatment. In this case, the possibility of fabricating complex geometry, defect-free, laminate glass-ceramic materials by LOM of cast tapes has been proven. [18]

(a)

(b)

Fig. 17 Photographs of LOM samples: (a) green tapes; (b) gear wheel geometries: green laminate (left); sintered LZSA glass-ceramics (right). [18] Summary Glass-ceramics from parent glass powders may be produced by different processing routes. As an example, glass-ceramic parts from the LZSA systems were manufacture by pressing, plastic forming or casting, mostly in environmental friendly conditions, using water-based additives and low temperature heat treatments. The microstructures of the LZSA glass-ceramics showed a homogeneous distribution of porosity and the main phases were identified as β-spodumene, zirconium silicate and lithium metasilicate. After crystallisation the densification increased by reduction of the glassy phase viscosity as the temperature increased. A further temperature increase resulted in an expansion of the material caused by the melting of the glass-ceramic matrix. LZSA is a potential candidate to produce sintered silicate glass-ceramic materials and composites for several applications.

11

12

12th INTERNATIONAL CERAMICS CONGRESS PART G

Acknowledgements The authors are thankful for the financial support from CAPES, CNPq and DAAD. The Institute of Glass and Ceramics at the University of Erlangen-Nuremberg, Germany, is also acknowledged for supporting part of the experimental work. References [1] Z. Strnad: Glass Science and Technology, Vol. 8 (Elsevier, New York, 1996). [2] E.M. Rabinovich: J. Mater. Sci. Vol. 20 (1985), p. 4259-4297. [3] S.D. Yoon and Y.H. Yun: J. Mater. Process. Technol. Vol. 168 (2005), p. 56-61. [4] Y.H. Yun, S.B. Kim, B.A. Kang, Y.W. Lee, J.S. Oh and K.S. Hwang: J. Mater. Process. Tech. Vol. 178 (2006), p. 61-66. [5] A.P. Novaes de Oliveira, T. Manfredini, G.C. Pellicani, A.B. Corradi and L. Di Ladro: Proc. Cimtec’98: 9th Int. Conf. on Modern Materials and Technologies, Florence, Italy, June 1998. [6] P.W. McMillan: Glass Ceramics, 2nd Ed. (Academic Press, New York, 1979). [7] J.H. Simmons et al.: Advances in Ceramics, Vol. 4: Nucleation and Crystallization in Glasses (American Ceramic Society, Ohio, 1982). [8] D.M. Miller. U.S. Patent 3,926,648 (1975). [9] E.A. Takher et al.: Glass and Ceramics, Vol. 34 (1977), p. 445-452. [10] C.I. Helgesson: Science of Ceramics, Vol. 8 (British Ceramic Society, London, 1976). [11] J.S. Reed: Principles of Ceramics Processing, 2nd Ed. (Wiley, New York, 1995). [12] G.M. Reitz, O.R.K. Montedo, O.E. Alarcon, D. Hotza and A.P. Novaes de Oliveira: Adv. Sci. Tech. Vol. 45 (2006), p. 442-446. [13] O.K. Montedo F.J. Floriano, J. Oliveira Filho, A.M. Bernardin, D. Hotza and A.P. Novaes de Oliveira: submitted to Int. J. Appl. Glass Tech. (2010). [14] F.M. Bertan, O.R.K. Montedo, C.R. Rambo, D. Hotza and A.P. Novaes de Oliveira: J. Mater. Process. Tech. Vol. 209 (2009), p. 1134-1142. [15] L. Giassi, O.R.K. Montedo, D. Hotza, M.C. Fredel and A.P. Novaes de Oliveira: Glass Tech. Vol. 46 (2005), p. 277-280. [16] L. Giassi, D. Hotza, O.E. Alarcon, M.C. Fredel and A.P. Novaes de Oliveira: Am. Ceram. Soc. Bull. Vol. 84 (2005), p. 9301-9306. [17] C.M. Gomes, A.P.N. Oliveira, D. Hotza, N. Travitzky and P. Greil: J. Mater. Process. Tech. Vol. 206 (2008), p. 194-201. [18] C.M. Gomes, C.R. Rambo, A.P. Novaes de Oliveira, D. Hotza, D. Gouvea, N. Travitzky and P. Greil: J. Am. Ceram. Soc. Vol. 92 (2009), p. 1186-1191. [19] E. Sousa, C.B. Silveira, T. Fey, P. Greil, D. Hotza and A.P. Novaes de Oliveira: Adv. Appl. Ceram. Vol. 104 (2005), p. 22-29. [20] C.R. Rambo, E. Sousa, A.P. Novaes de Oliveira, D. Hotza and P. Greil: J. Am. Ceram. Soc. Vol. 89 (2006), p. 3373-3378. [21] E. Sousa, C.R. Rambo, D. Hotza, A.P. Novaes de Oliveira, T. Fey and P. Greil: Mat. Sci. Eng. A Vol. 476 (2008), p. 89-97.