Crystallization of High-strength Fine-sized Leucite Glass-ceramics

Abstract

Manufacturing of leucite glass-ceramics often leads to materials with inhomogeneous microstructures. Crystal-glass thermal mismatches which produce microcracking around larger crystals-agglomerates are associated with reduced mechanical properties. The hypotheses were that fine (< 1 µm) crystal size and uniform microstructure in a thermally matched glass would increase the biaxial flexural strength (BFS). Glass was synthesized, attritor-milled, and heat-treated. Glasses and glass-ceramics were characterized by XRD, SEM, and Dilatometry. Experimental (A, M1A and M2A) and commercial glass-ceramics were tested by the BFS test. Experimental glass-ceramics showed an increased leucite crystal number and decreased crystal size with glass particle size reduction. Leucite glass-ceramics (< 1 µm) showed minimal matrix microcracking and BFS values of [mean (SD) MPa]: M1A = 253.8 (53.3); and M2A = 219.5 (54.1). Glass-ceramics M1A and M2A had higher mean BFS and characteristic strength than the IPS Empress Esthetic glass-ceramic (p < 0.05). Fine-grained, translucent leucite glass-ceramics were synthesized and produced high mean BFS.

Keywords

glass-ceramic flexural strength microstructure crystallization

Introduction

Leucite glass-ceramics are widely used in dentistry to produce dental restorations by a variety of fabrication routes (Cattell et al., 2005), making these materials extremely versatile. Methods of manufacturing these materials include the blending of high and low expansion powdered frits (Weinstein et al., 1962), or of synthetic leucite with glasses (Burk and Burnett, 1978) by fusion methods. The controlled heating of a glass and its surface crystallization has also been used (Höland et al., 2003), although not optimized. Commercial production methods produce glass-ceramics with up to 45% leucite volume fraction (Cattell et al., 2005), and with microstructures that lack homogeneity. Microcracking often results around large, irregular leucite crystals and clusters (≈10 µm), inhomogeneously distributed in a glass matrix (Barreiro et al., 1989). Crystal-matrix microcracking and crystal twinning (Palmer et al., 1988) are associated with a temperature dependent (605-625°C) cubic to tetragonal phase transformation (Mackert, 1988). This is in conjunction with a thermal expansion (TEC) change from 10.1 to 25.1 x 10^-6 K^-1 (Rouf et al., 1978) and a reversible 1.2% volume change (Mackert, 1988). These processes can introduce stress-raising flaws or decouple the leucite crystals from the glass matrix (Binns, 1983). This can reduce the elastic-stored strain energy of the thermal expansion anisotropy and deplete tangential compressive stresses around the crystals, which are key to strengthening in feldspathic dental porcelain (Denry et al., 1996). A critical particle size for glass-ceramic microcracking was explained by a Griffith-type energy balance criterion (Davidge and Green, 1968). In agreement with this theory, microcracking was minimized in tetragonal leucite glass-ceramics by a reduction in the mean leucite particle diameter (< 4 µm) (Mackert et al., 2001). This type of fine-grained microstructure can be achieved by fine powder milling and controlled heat treatments (Cattell et al., 2006), because leucite crystallization is surface-nucleated (Höland et al., 1995). Several studies indicated that a reduction in leucite crystal size, uniformity of microstructure, and minimal matrix microcracking produced flexural strength increases up to 160 MPa (Shareef et al., 1994; Cattell et al., 2001). However, flexural strength values of 200-300 MPa have been found in fine-grained (0.1-1 µm) leucite glass-ceramics (Rouf et al., 1978), although the glass opacity and high viscosity limit their processing and use for aesthetic restorations. In the present work, the objectives were to use a more controlled processing and crystallization regimen of a designed glass to produce translucent high-strength (> 200 MPa) leucite glass-ceramics with fine-grained microstructures. These materials were designed with the objectives of reducing the incidence of restoration fracture, reducing enamel wear (Metzler et al., 1999), and increasing translucency (Beall and Duke, 1969). The hypotheses were that a fine crystal size (< 1 µm) and uniform microstructure in a thermally matched glass were associated with an increased biaxial flexural strength.

Materials & Methods

Glass/Glass-ceramic Synthesis

The glass batch composition was (mol%): 72.6% SiO₂, 10.7% Al₂O₃, 7.9% K₂O, 2.1% CaO, 0.3% TiO₂, 4.7% Na₂O, 1.1% Li₂O, and 0.5% MgO. The glass batch was heated in an electric kiln (Fredrickson Kiln Co., Alfred, NY, USA) at 10°C/min to 1316°C (7 hours hold), and then the frit was cooled and ground to 125 µm (glass A). Glass A was then wet-ground (distilled water) in an Attritor Mill (Model 1-S Lab Attritor, Union Process, Akron, OH, USA) with 5 mm yttria-stabilized zirconia (YTZ) grinding media (Tosoh Inc., Grove City, OH, USA) at 400 rpm. Glass slurry was withdrawn after 15, 30, 45, 60, 90, and 120 minutes of milling. The 120-minute-milled slurry and the glass A materials were then milled according to the regimens shown in Figure 1a, to produce glasses M1A and M2A. Glasses were heat-treated in a furnace (RHF 1600, Carbolite, Hope Valley, UK) from 23°C at a rate of 10°C/min to 650°C (1 hour hold), then ramped to 1120°C (1 hour hold) and air-quenched.

Glass particle size was measured in a Mastersizer/E particle analyzer (Malvern Instruments, Malvern, Worcestershire, UK). Quantitative elemental analysis of the glass powders was carried out following attritor milling, by inductively coupled plasma mass spectrometry (ICP-MS, Elan 9000, Perkin Elmer, Norwalk, CT, USA).

X-ray Diffraction Analysis

X-ray diffraction (XRD) was carried out on the glasses and glass-ceramics by means of an X’Pert Pro X-ray diffractometer (PANalytical, Almelo, The Netherlands). Flat plate θ/θ geometry and Ni-filtered Cu-Kα radiation (λ₁ = 0.1540598 nm and λ₂ = 0.15444260 nm) were used. Data were continuously collected from 5° to 120° (2θ range) with a step size of 0.0334° and a step time of 200.03 sec. Calibration was carried out using NIST standard reference material 660a (lanthanum hexaboride). The structural model of tetragonal leucite (ICDD: 00-038-1423) was used for phase identification. Crystal strain calculations were performed with the XRD data and Eq. 1 (Cullity and Stock, 2001):

ε_{a} = \frac{a - a_{0}}{a_{0}}, ε_{c} = \frac{c - c_{0}}{c_{0}}

(1)

where ε_a and ε_c are the strain of leucite in the a-axis and c-axis, a and c are the mean unit cell dimensions of tetragonal leucite for the tested glass-ceramics, and a ₀ and c ₀ are the unit cell dimensions of the reference tetragonal leucite (ICDD: 00-038-1423).

Biaxial Flexural Strength Specimen Preparation

The glass-ceramic powders (A, M1A and M2A) were used to fabricate disc specimens (30 per group) by moistening 1.0 g of powder with 0.3 mL of modeling liquid (CH B: 24066, VITA, Bad Säckingen, Germany) and compacting in a cylindrical steel mould with a plunger. Discs were placed in a pre-heated (538°C) porcelain furnace (Multimatt MCII, Dentsply, Konstanz, Germany) and sintered at a rate of 38°C/min to 1040°C for 2 min under vacuum. Specimens were next wet-ground to P1000 grade silicon carbide paper. One simulated stain firing in air from 538°C at a rate of 55°C/min to 780°C without hold and one simulated overglaze firing in air from 538°C at a rate of 55°C/min to 860°C with a 0.3 min hold were then applied.

Thirty IPS Empress Esthetic glass-ceramic (Lot: H22624, ETC2, Ivoclar-Vivadent, Schaan, Liechtenstein) specimens were prepared by sprueing Perspex discs (14 mm x 2 mm) onto muffle bases and investing, pre-heating, and heat-extruding in an Optimal press furnace (Jeneric Pentron, Wallingford, CT, USA) according to the manufacturer’s instructions. The investment was removed by sandblasting with 50 µm glass beads (Bracon Ltd., Etchingham, East Sussex, UK) at 1.5 x 10⁵ Pa pressure. The specimens were wet-ground to P1000 grade silicon carbide paper on the compressive test surface only and fired in a porcelain furnace (Multimatt MCII, Dentsply) according to the manufacturer’s recommended (simulated) firing cycles (2 stain and 2 overglaze firings).

Biaxial Flexural Strength Testing

The biaxial flexural strength was determined using the ball-on-ring test fixture. Disc specimens (14 x 2 mm) were centrally loaded on a 10 mm diameter knife-edge support, via a 4 mm diameter spherical ball indenter, at a crosshead speed of 1 mm/min. The BFS was calculated by Eq. 2 (Timoshenko and Woinowsky-Krieger, 1959):

σ_{max} = \frac{P}{h^{2}} {(1 + v) [0.485 \times \ln (\frac{a}{h}) + 0.52] + 0.48}

(2)

where σ_max was the maximum tensile stress, P was the load measured at fracture, h was the specimen thickness, a was the radius of the knife-edge support, and v was a Poisson’s ratio of 0.25.

Statistical Analysis

Group means were analyzed by one-way ANOVA and Tukey’s multiple-comparison test at p < 0.05 (Sigma Stat, version 2.03, SPSS Inc., Chicago, IL, USA). Weibull analysis of the test groups was carried out with WeibullSMITH software (Fulton Findings, Torrance, CA, USA). Weibull m and characteristic strength values were compared for the overlap of their double-sided confidence intervals (95% level) to determine differences.

Secondary Electron Microscopy

Polished and etched (0.1% HF, 60 sec) specimens were gold coated and viewed by scanning electron microscopy (JSM 6300F, JEOL Ltd., Tokyo, Japan). Micrographs (x 2200, area = 2210.4 µm²) were used to quantify crystal size, number, and area fraction with a light pen (270SD + INT-40, Trackballs, Camarillo, CA, USA) and image analysis software (Sigma Scan Pro 5.0, Systat Software, Inc., Chicago, IL, USA).

Differential Dilatometry

Glass frit was cut into blocks (16 x 6 x 6 mm³) using a plate saw (Struers Accutom 2, Struers Ltd., Glasgow, Scotland, UK), and glass-ceramic specimens were constructed by either heat extrusion or powder compaction and sintering. The TEC of the specimens were measured by a differential dilatometer (DIL 402PC, Netzsch Instruments, Selb, Germany) at a heating rate of 3°C/min (25 - 1200°C).

Results

Particle Size Analysis Results

The glass particle size (mean D ₅₀) vs. milling time is shown in Fig. 1b. A decrease in glass particle size [mean D ₅₀ (SD)] from 11.37 (0.17) to 3.08 (0.04) µm was identified within the first 60-minute milling period. The particle size of glass M1A was 1.79 (0.05) µm, which was further reduced for M2A to 0.45 (0.01) µm. A correlation (exponential fit) between the glass particle size and mean leucite crystal size (r² = 0.906) and leucite crystal number (r² = 0.985) was found (Figure 1c).

Inductively Coupled Plasma – Mass Spectrometry Results

Quantitative elemental analysis of the attritor-milled glasses with ICP-MS showed a correlation (polynomial fit) between zirconium (r² = 0.998), yttrium (r² = 0.999), and hafnium (r² = 0.999) content and milling time (Figure 1d). Trace amounts of chromium (70 ppm), nickel (20 ppm), and tungsten (22 ppm) were also found in the glass M2A.

Differential Dilatometry Results

Glass M1A showed a TEC value of 8.5 x 10^-6 K^-1. The TEC and glass transition temperature (T_g) of the glass-ceramics were: A (19.0 x 10^-6 K^-1, 579.1°C), M1A (18.2 x 10^-6 K^-1, 589.5°C), and M2A (17.2 x 10^-6 K^-1, 589.3°C). A lower TEC value of 16.7 x 10^-6 K^-1 and a higher T_g of 616.3°C were found for the IPS Empress Esthetic glass-ceramic.

X-ray Diffraction Results

Experimental glasses were largely amorphous, with some tiny visible crystalline peaks for the 60 min A, 90 min A, 120 min A, and M1A glasses (Figure 1e). These 2θ positions matched with the hkl reflections 112 and 400 for tetragonal leucite. Tetragonal leucite and an amorphous glass phase were identified for all experimental and IPS Empress Esthetic glass-ceramics (Figure 1f). The mean unit cell dimensions and calculated crystal strains are given in Table 1.

Table 1.

The Mean Unit Cell Dimensions and Crystal Strains of the Glass-ceramics

Glass-ceramics	Mean a-axis Unit Cell Dimension (SD) (nm)	Mean c-axis Unit Cell Dimension (SD) (nm)	Mean Unit Cell Volume (SD) (nm³)	ε_a	ε_c
A	1.31089 (0.00009)	1.37071 (0.00015)	2.35547 (0.0012)	0.33%	−0.35%
M1A	1.31229 (0.00010)	1.36963 (0.00015)	2.35865 (0.0015)	0.44%	−0.43%
M2A	1.31172 (0.00009)	1.36916 (0.00013)	2.35579 (0.0011)	0.40%	−0.46%
IPS Empress Esthetic	1.31069 (0.00006)	1.37170 (0.00009)	2.35645 (0.0032)	0.32%	−0.28%

ε_a = crystal strain in the a-axis; ε_c = crystal strain in the c-axis.

Secondary Electron Microscopy Results

Experimental glass-ceramics (Figure 2a, 2b, 2c) showed fine and evenly distributed leucite crystals and minimal matrix microcracking, with a leucite area fraction between 24.9% and 25.4%. The experimental and control glass-ceramics showed a mean (SD) leucite crystal size and crystal number of: A, 0.99 (0.59) µm², 558; M1A, 0.39 (0.28) µm², 1434; M2A, 0.16 (0.10) µm², 3378; and IPS Empress Esthetic, 1.72 (1.98) µm², 364. The area fraction of Empress glass-ceramic was 31.2%, with microcracking present both within the larger leucite crystals and the glass matrix (Figure 2d).

Biaxial Flexural Strength Results

Glass-ceramic M1A and M2A groups showed significantly higher BFS and characteristic strength (σ₀) values (p < 0.05) compared with those of IPS Empress Esthetic and the glass-ceramic A groups (Table 2). There was no significant difference between the Weibull m values for IPS Empress Esthetic and experimental glass-ceramic groups.

Table 2.

Biaxial Flexural Strength and Weibull Analyses Results of the Tested Glass-ceramics

Glass-ceramics	Mean Flexural Strength (MPa)	SD (MPa)	m	C.I. for m (95%)	σ₀ (MPa)	C.I. for σ₀ (95%)	r ²
A	153.2^a	21.7	8.5^a	6.9 - 10.4	162.0^a	156.0 - 168.2	0.941
M1A	253.8^b	53.3	5.4^bc	4.3 - 6.8	274.9^b	259.2 - 291.7	0.971
M2A	219.5^c	54.1	4.7^bc	3.8 - 5.8	239.5^c	223.8 - 256.3	0.972
IPS Empress Esthetic	165.5^a	30.6	6.3^ac	5.0 - 7.9	177.5^d	168.8 - 186.6	0.977

C.I. = confidence interval, m = Weibull modulus, σ₀ = characteristic strength. Different superscript letters indicate significant differences within categories (p < 0.05).

Discussion

Attritor milling of the experimental glasses, followed by controlled crystallization heat treatments, produced fine-sized 0.16 (0.10) µm² leucite crystals in the same size range as synthesized by sol-gel and hydrothermal techniques (Zhang et al., 2008). The present study allowed for the synthesis and control of a range of glass-ceramics with decreasing leucite crystal size and increasing crystal number, which correlated with the reduction of glass particle size (Figure 1c). Rapid glass particle size reduction was found in the first 60 min of attritor milling (Figure 1b), and major reductions were then achieved only with the use of finer grinding media (McLaughlin, 1999). The resultant fine powder sizes achieved increased heterogeneous nucleation sites. The increased surface area of the glass powder encouraged a large number of sharp edges and a change in particle shape (Müller et al., 2000), and reduced the inhibiting effects of surface elastic strains (Schmelzer et al., 1993). The dramatic increase in crystallite number from 558 to 3378 after sustained milling supports these assumptions. Glass particle size and heat treatments were previously shown to modify crystallization, with the present nucleation step adopted to increase the number of crystallites (Cattell et al., 2006).

The different milling protocols and longer milling times for the M2A glass led to an increase in zirconium, yttrium, hafnium, and other trace elements, associated with the continuous wear of the YTZ grinding media (Figure 1d). The reduction in mean BFS compared with the M1A material could be due to the local agglomeration of these products or the formation of another species and limits this process. Leucite glass-ceramic powder doped with nano-sized (30-60 nm) zirconia (60 wt%) previously produced a bending strength of 214 MPa (Conrad and Meyer, 2007), in the same range as found for the M2A material. When suitably dispersed, zirconia, yttria (> 3 wt%), and hafnia can act as nucleating agents, influencing the viscosity and crystallization kinetics of fine-scale (40-250 nm) glass-ceramics (Rouf et al., 1978; Zheng et al., 2007; Wondraczek and Pradeau, 2008). Trace elements in the M1A glass, if well-dispersed, could encourage the heterogeneous growth of the leucite crystalline phase (James, 1982). Micro- or nano-immiscibility in these glasses is also likely (Cattell et al., 2005), as is the formation of titanates (Rouf et al., 1978). These phases can induce heterogeneous crystallization of a metastable phase, which at increased temperature breaks down into the stable crystalline phase (Pinckney and Beall, 2008). The experimental glasses had some trace crystalline peaks before crystallization heat treatments (Figure 1e). These fine isostructural sites have the potential to reduce the energy barrier for crystallization and provide low-energy crystallization sites (Zhang et al., 2007).

The mean BFS and σ₀ of the experimental glass-ceramics were significantly (p < 0.05) increased following milling and crystallization heat treatments (Table 2). This increase in strength was associated with the reduction of leucite crystal size and uniformity of the microstructure. The lack of matrix microcracking at mean leucite crystal sizes of less than 1 µm² also supports previous work (Mackert et al., 2001). The homogeneous and fine-scale crystallization could also influence the local depletion of the glass structure, resulting in more favorable residual glass-crystal interfacial stresses. The glass-ceramic M1A had a-axis and c-axis strains of a similar magnitude (a = 0.44% and c = -0.43%, Table 1), which caused no matrix microcracking and depletion of generated stresses. Changes to the anisotropic stress field and the generation of favorable residual stresses in the glass have been previously linked to strength increases (Denry et al., 1996). Mackert (1988) discussed the presence of microcracks parallel to the c-axis in leucite glass-ceramics, partially decoupling crystallites and reducing the bulk thermal expansion. This was evident for the control group, with a reduced c-axis strain associated with crystal-matrix microcracking and a lower TEC (16.7 x 10^-6 K^-1, 100-400°C) and mean BFS. A higher area fraction of the high-expansion tetragonal leucite phase (31.2%) compared with the experimental glass-ceramics (24.9-25.4%) also suggests that strain dissipation may have occurred.

Changes to the lattice parameters for the experimental glass-ceramics with glass particle size reduction may be due to changes in glass diffusivity, mobility (Freiman et al., 1971), and crystallization mechanisms (Tošić et al., 2008). The unit cell dimensions (M1A and M2A, Table 1) are also in the same range (a = 1.312-1.315 nm, c = 1.369-1.374 nm) as reported for solid solutions of tetragonal leucite (Hermansson and Carlsson, 1976). A limited substitution of sodium for potassium is possible, and some geological leucites indicate up to 2.4 sodium atoms per unit cell (16 KAlSi₂O₆ per unit cell) (Deer et al., 2004). The incorporation of lithium and sodium into the leucite crystal lattice can cause reduced TEC and modification to the cubic to tetragonal transformation temperature (Ota et al., 1993). The relationship between the transformation temperature, T_g, and glass properties is important to the maintenance of residual stresses and mechanical properties (Lee et al., 1997).

The present study produced leucite glass-ceramics with higher mean flexural strengths (253.8 MPa) than those reported for current leucite glass-ceramics (120-140 MPa) (Höland et al., 2003), and in the same range (200-300 MPa) as reported for an opaque leucite glass-ceramic (Rouf et al., 1978). The original hypothesis can therefore be upheld, because the fine (< 1 µm²) leucite crystal size and uniformity of microstructure in a thermally matched glass produced high mean biaxial flexural strengths. These new fine-grained leucite glass-ceramics also have other advantages in terms of improved translucency, aesthetics, processability, and reduced enamel wear (Metzler et al., 1999), which will be reported in a future study.

Footnotes

Acknowledgements

This research was carried out in partial fulfillment for the degree of PhD in the Institute of Dentistry, Barts and The London School of Medicine and Dentistry, 2009. The authors acknowledge Support for Oral Science, Barts and the London, School of Medicine and Dentistry and the Den-Mat Corporation for the support of this research.

References

Barreiro

Riesgo

Vicente

(1989). Phase identification in dental porcelains for ceramo-metallic restorations. Dent Mater 5:51-57.

Beall

Duke

(1969). Transparent glass-ceramics. J Mater Sci 4:340-352.

Binns

(1983). The chemical and physical properties of dental porcelains. In: Dental ceramics: proceedings of the First International Symposium on Ceramics. McLean

, editor. Chicago: Quintessence Publishing Co., pp. 41-56.

Burk

Burnett

, inventors (1978). Leucite-containing porcelains and method of making same. US Patent 4101330.

Cattell

Chadwick

Knowles

Clarke

Lynch

(2001). Flexural strength optimisation of a leucite reinforced glass ceramic. Dent Mater 17:21-33.

Cattell

Chadwick

Knowles

Clarke

(2005). The crystallisation of an aluminosilicate glass in the K₂O-Al₂O₃-SiO₂ system. Dent Mater 21:811-822.

Cattell

Chadwick

Knowles

Clarke

Samarawickrama

(2006). The nucleation and crystallisation of fine grained leucite glass-ceramics for dental applications. Dent Mater 22:925-933.

Conrad

Meyer

, inventors (2007). Leucite glass ceramic doped with nanoscale metal oxide powder, method for producing the same, and dental materials and dental products formed therefrom. US Patent 7172649.

Cullity

Stock

(2001). Elements of x-ray diffraction. 3rd ed. New Jersey: Prentice-Hall, pp. 439-441.

10.

Davidge

Green

(1968). The strength of two-phase ceramic/glass materials. J Mater Sci 3:629-634.

11.

Deer

Howie

Zussman

Wise

(2004). An introduction to rock-forming minerals. 2nd ed. Harrow, UK: Prentice-Hall, pp. 304-315.

12.

Denry

Mackert

Holloway

Rosenstiel

(1996). Effect of cubic leucite stabilization on the flexural strength of feldspathic dental porcelain. J Dent Res 75:1928-1935.

13.

Freiman

Onada

Pincus

(1971). Spherulitic crystallization in glasses. In: Advances in nucleation and crystallization in glasses. Hench

Freiman

, editors. Columbus, OH: The American Ceramic Society, pp. 91-112.

14.

Hermansson

Carlsson

(1976). High and low temperature forms of leucite. In: Proceedings of the Eighth International Symposium on the Reactivity of Solids. Gothenberg, Sweden: Swedish Institute for Silicate Research, pp. 541-545.

15.

Höland

Frank

Rheinberger

(1995). Surface crystallization of leucite in glasses. J Non-Cryst Solids 180:292-307.

16.

Höland

Rheinberger

Schweiger

(2003). Control of nucleation in glass ceramics. Phil Trans R Soc Lond: Math Phys Eng Sci 361:575-589.

17.

James

(1982). Nucleation in glass-forming Systems. A review. In: Advances in ceramics. Nucleation and crystallization of glasses, 83rd Annual Meeting of the American Ceramic Society. Simmons

Uhlmann

Beall

, editors. Washington, DC: The American Ceramic Society, pp. 14-48.

18.

Lee

Kon

Asaoka

(1997). Influence of modification of Na₂O in a glass matrix on the strength of leucite-containing porcelains. Dent Mater J 16:134-143.

19.

Mackert

(1988). Effects of thermally induced changes on porcelain-metal compatibility. In: Perspectives in dental ceramics, proceedings of the Fourth International Symposium on Ceramics. Preston

, editor. Chicago: Quintessence Publishing Co., pp. 53-64.

20.

Mackert

Twiggs

Russell

Williams

(2001). Evidence of a critical leucite particle size for microcracking in dental porcelains. J Dent Res 80:1574-1579.

21.

McLaughlin

(1999). Bead size and mill efficiency–producing inorganic nanoparticles by milling with ultrafine ceramic beads. Ceramic Industry 149:34-40.

22.

Metzler

Woody

Miller

3rd Miller

(1999). In vitro investigation of the wear of human enamel by dental porcelain. J Prosthet Dent 81:356-364.

23.

Müller

Zanotto

Fokin

(2000). Surface crystallisation of silicate glasses: nucleation sites and kinetics. J Non-Cryst Solids 274:208-231.

24.

Ota

Takahashi

Yamai

Suzuki

(1993). High thermal expansion polycrystalline leucite ceramic. J Am Ceram Soc 76:2379-2381.

25.

Palmer

Putnis

Salje

EKH

(1988). Twinning in tetragonal leucite. Phys Chem Miner 16:298-303.

26.

Pinckney

Beall

(2008). Microstructural evolution in some silicate glass-ceramics: a review. J Am Ceram Soc 91:773-779.

27.

Rouf

Hermansson

Carlsson

(1978). Crystallisation of glasses in the primary phase field of leucite in the K₂O-Al₂O₃-SiO₂ system. Trans J Br Ceram Soc 77:36-39.

28.

Schmelzer

Pascova

Möller

Gutzow

(1993). Surface-induced devitrification of glasses: the influence of elastic strains. J Non-Cryst Solids 162:26-39.

29.

Shareef

Van Noort

Messer

Piddock

(1994). The effect of microstructural features on the biaxial flexural strength of leucite reinforced glass-ceramics. J Mater Sci Mater Med 5:113-118.

30.

Timoshenko

Woinowsky-Krieger

(1959). Theory of plates and shells. 2nd ed. New York: McGraw-Hill, pp. 70-71.

31.

Tošić

Živanović

Grujić

Stojanovic

Nikolić

(2008). A study of the primary crystallization of a mixed anion silicate glass. J Non-Cryst Solids 354:3694-3704.

32.

Weinstein

Katz

Weinstein

, inventors (1962). Fused porcelain-to-metal teeth. US Patent 3052982.

33.

Wondraczek

Pradeau

(2008). Transparent hafnia-containing b-quartz glass ceramics: nucleation and crystallization behavior. J Am Ceram Soc 91:1945-1951.

34.

Zhang

Rao

(2007). Nanocrystalline seeding effect on the crystallisation of two leucite precursors. J Am Ceram Soc 90:2390-2398.

35.

Zhang

Rao

(2008). Mechanical properties of dental porcelain with different leucite particle sizes. J Am Ceram Soc 91:527-534.

36.

Zheng

Cheng

Tang

Quan

Cao

(2007). Effect of Y₂O₃ addition on viscosity and crystallization of the lithium aluminosilicate glasses. Thermochim Acta 456:69-74.