J Dent Res 81(7): 487-491, 2002
© 2002 International and American Associations for Dental Research
RESEARCH REPORT
Biomaterials & Bioengineering |
Reliability and Properties of Ground Y-TZP-Zirconia Ceramics
R.G. Luthardt1,*,
M. Holzhüter1,
O. Sandkuhl2,
V. Herold2,
J.D. Schnapp2,
E. Kuhlisch3, and
M. Walter1
1 Dresden University of Technology, Dental School, Department of Prosthetic Dentistry, Fetscherstraße 74, 01307 Dresden, Germany;
2 Friedrich-Schiller-University, Technical Institute, 07743 Jena, Germany; and
3 Dresden University of Technology, Department of Medical Informatics and Biometry, 01307 Dresden, Germany;
* corresponding author, Ralph.Luthardt{at}mailbox.tu-dresden.de
 |
ABSTRACT
|
|---|
Yttria-stabilized zirconia ceramics is a high-performance material with excellent biocompatibility and mechanical properties, which suggest its suitability for posterior fixed partial dentures. The hypothesis under examination is that the strength and reliability of Y-TZP zirconia ceramics are affected by the inner surface grinding of crowns, and vary with the grinding parameter. Flexural strength, surface roughness, and fracture toughness were determined on samples machined by face and peripheral grinding with varied feed velocities and cutting depths. Results have been compared with those on lapped samples. Analysis of variance and Weibull parameter were used for statistical analysis. It was found that inner surface grinding significantly reduces the strength and reliability of Y-TZP zirconia compared with the lapped control sample. Co-analysis of flexural strength, Weibull parameter, and fracture toughness showed counteracting effects of surface compressive stress and grinding-introduced surface flaws. In conclusion, grinding of Y-TZP needs to be optimized to achieve the CAD/CAM manufacture of all-ceramic restorations with improved strength and reliability.
KEY WORDS: zirconia • grinding • strength • fracture toughness • CAD/CAM
|
INTRODUCTION
|
|---|
The application of all-ceramic crowns and fixed partial dentures (FPDs) may benefit from new industrially pre-fabricated advanced ceramics such as yttria-stabilized tetragonal zirconia polycrystals (Y-TZP). In conventional ceramics, the flaw distribution in a given region rather than the thickness of the material dominates the failure of ceramic crowns (Thompson et al., 1994). In contrast to conventional dental ceramics, Y-TZP is composed of many small particles without any glassy phase at the crystallite border and is distinguished by a crack-initiation mechanism. In the stress field of propagating cracks, matrix pressure on the tetragonal particles of Y-TZP is reduced by tensile stress. In addition, shear stress formed in the particles causes a martensitic transformation, inhibiting the opening of the crack and increasing the energy necessary for further crack growth (Stevens, 1986). Sintered Y-TZP showed a mean flexural strength of 
= 900 – 1000 MPa and a Weibull parameter m = 10.7 to 14.9 (Kosmac et al., 1999), which is far superior to those of conventional ceramics applied in dentistry.
The most important types of grinding-induced near-surface characteristics are roughness, plastic deformation, damage, and residual stress (Pfeiffer and Hollstein, 1997). These near-surface characteristics are influenced by the grinding parameter (cutting depth, feed velocity, diamond tool, numerically controlled machine, grinding fluid). In principle, grinding of ceramics can act in two different directions (Giordano et al., 1995). First, it causes residual surface compressive stress, which can considerably increase the mean strength of zirconia-toughened ceramics. Second, it induces surface flaws, which may become strength-determining if they exceed the depth of the grinding-induced surface compressive layer (Pfeiffer and Hollstein, 1997; Tuan and Kuo, 1998; Kosmac et al., 1999). The strength-degradation problem can be reduced by grinding in the ductile mode with a suitable grinding parameter (Liao et al., 1997). The relevance of several grinding parameters to mechanical properties cannot be suitably estimated by measurements of the fracture strength of copings. In analyses of the grinding procedures for crowns and FPDs, internal crown surface grinding is most sophisticated, due to the grinding and cooling process (Luthardt et al., 1997).
The specific aim of this study was to test the hypothesis that the strength and reliability as well as the surface roughness of Y-TZP zirconia ceramic machined under conditions simulating the inner surface grinding of crowns and FPDs vary, depending on the grinding parameter. Additionally, the hypothesis that strength and surface roughness parameters are dependent on each other for Y-TZP-zirconia ceramic machined under these conditions was tested.
|
MATERIALS & METHODS
|
|---|
Simulated Inner-surface Grinding
Densely sintered Y-TZP zirconia ceramic disks (97 mol% ZrO2, 3 mol% Y2O3; Metoxit AG, Thayngen, Switzerland) of 36-mm diameter and 3-mm thickness were randomly divided into 10 groups of 4. The top surface was single-side-lapped (lapping wheel speed, 40 min-1; pressure, 30 N*cm-2; diamond suspension, D15, grain size, 15 µm). The face and peripheral grinding procedure, according to the multi-pass grinding method defined by spindle speed [ns], width of grinding [ae], feed velocity [vf], and grinding depth [ap], was used to simulate the inner surface grinding of crowns. The surface treatments applied to the lower surface are indicated in Table 1
. A Precident-DCS® milling-machine (DCS Production AG, Allschwil, Switzerland) was chosen due to its long-term clinical application for the fabrication of Y-TZP fixed restorations (Tinschert et al., 2001). When a high-frequency spindle with high stiffness was installed (SC 53H, Precise GmbH, Leichlingen, Germany), its rigidity was improved to reduce vibration during grinding. The disks were horizontally attached to the holder and oriented parallel to the XY-plane. A top-surface flat sinter-diamond tool with mixed grit (range of grit size, 150-250 µm, shape 1A1W-3-6; DCS Production AG, Allschwil, Switzerland), used for bulk material removal, was chosen. The disks were reduced to 2.2 mm thick by the application of feed velocity vf = 100 mm*min-1, cutting depth ap = 0.1 mm, grinding width ae = 1 mm, and spindle speed n = 60,000 min-1. Afterward, the Y-TZP disks of each group were ground to a thickness of 2.00 mm according to the parameters listed in Table 1. A water-based mineral oil emulsion grinding fluid was used as a coolant (Motorex Swisscool 7755 Aero, Bucher AG, Langenthal, Switzerland). The disks of the control group were lapped to a thickness of 2.00 mm according to the parameters applied to the top surface.
View this table:
[in this window]
[in a new window]
|
Table 1. Surface Roughness for Ground and Lapped Y-TZP Zirconia Ceramic Measured Length- and Crosswise to the Grinding Direction
|
|
Surface Roughness
Three surface parameters of the machined disks were measured by means of a profilometer (Form Talysurf, Rank Taylor Hobson, Leicester, UK): first, the arithmetical mean profile deviation [Ra]; second, the mean value of the maximum height of the profile; and third, the perpendicular distances between the highest and lowest points of the filtered roughness profile within the measurement line [Rmax]. The measurement lines (lm = 4 mm) were oriented along and across the grinding direction for the ground samples and in any direction for the lapped samples. Three measurements each were performed for the disks. The Fig. shows scanning electron micrographs representative of the surface finish of lapped and ground surfaces.
View larger version (121K):
[in this window]
[in a new window]
|
Figure. Electron micrographs representing the surface condition of (A) the as-machined samples and (B) the lapped control samples after preparation.
|
|
Sample Preparation and Flexural Strength Testing
Test samples (EN DIN, 1995) were prepared by means of a diamond wheel (1A1R-300-1,2-5-D126-C75-138AG, ALKU, Steinheim, Germany) mounted in a numerically controlled surface grinder (Planomat 408, Blohm GmbH, Hamburg, Germany). The disks were cut perpendicular to the grinding direction to 25 mm in length, then cut parallel to the grinding direction into 6 samples, each 2.5 mm in width, according to feed velocity vf = 20,000 mm*min-1, cutting depth ap = 0.004 mm, and spindle speed of n = 2000 min-1. Fully synthetic, water-soluble grinding coolant free of mineral oil was applied (DAW-AEROLAN VS, DAW Aerocit Schmierungstechnik GmbH, Werdau, Germany). After samples were cut and cleaned, the width and thickness of each sample were calculated as the mean of 3 measurements, by means of a digital caliper (Helio-Red, Helios Messtechnik GmbH & Co. KG, Niedenhall, Germany). The flexural strengths were measured at room temperature by means of a four-point flexural testing facility, with the ground side under tension according to the EN 843-1 standard (EN DIN, 1995). The load to fracture was recorded for each sample, and the flexural strength given by
 | (1) |
was calculated for the measured dimensions of the sample, whereFmis the load to fracture,d the distance between the centers of the supporting and the loading rollers,b the mean width calculated from 3 measurements, andh the mean thickness calculated from 3 measurements. The outer and inner spans of the four-point bending fixture are 20 and 10 mm, respectively. The crosshead speed of the testing machine (Zwick-1435, Zwick GmbH & Co., Ulm, Germany) wasv= 1 mm*min-1. The flexural strength was calculated for the measured dimensions.
Fracture Toughness
We used the Vickers indentation technique to estimate the fracture toughness of those disk pieces not used for strength testing. Thirty indentations were made on each sample with a load F = 300 N applied for 90 sec (Hardness-Tester, EMCO, Emco Meier GmbH, Hallein, Austria). The fracture toughness was calculated from the diagonals of the indentations (d1 and d2) and the lengths of the cracks emanating from the Vickers impression c' (c' bottom, c' top, c' right, c' left) measured at a 10-fold magnification (Shimadzu-microscope, Kyoto, Japan). The measured values were analyzed according to the expressions for the stress intensity factor, KIc, given by
 | (2) |
whereE is the Young's modulus,H the Vickers hardness,P the indentation load, andc the mean crack length as measured from the center of the impression.
Statistical Analysis
The overall significance level was set to 5%. The significance level for multiple pairwise comparisons between the cell means is adjusted for the number of tests.
The statistical significance of the differences among (1) the various grinding procedures, (2) the Weibull parameter, and (3) the fracture toughness regarding the feed velocity [vf] and the cutting depth [ap] were analyzed with one-way analysis of variance (ANOVA, SPSS for Windows Release 9.0, SPSS Software Corp., Munich, Germany). The Weibull analysis was performed according to the maximum likelihood method (EN DIN, 1995).
|
RESULTS
|
|---|
Grinding increased the surface roughness compared with the lapped control. The parameter measured across the grinding feed direction exceeded those measured along it (Table 1
). Lapped surfaces and surfaces ground with feed velocity vf = 50 mm*min-1 showed the smallest roughness parameter. Samples ground with feed velocity vf = 75 mm*min-1 showed, in almost all cases, higher roughness than samples with higher or lower feed velocity. Statistically significant differences were found for Ra (p = 0.001), Rz (p = 0.001), and Rmax (p = 0.026), measured along the grinding direction with regard to feed velocity. No significant differences were found for Ra, Rz, and Rmax measured across the grinding direction with regard to the grinding parameter.
Grinding decreased the mean flexural strength by half compared with the lapped control (Table 2). Statistically significant differences (p < 0.05) were found only between the lapped and the ground samples. Statistically significant correlations (p < 0.05) regarding the flexural strength of ground Y-TZP, depending on the machining parameter (feed velocity, cutting depth) and taking into consideration the surface roughness parameter measured along the grinding direction, were found for cutting depth [ap] (p = 0.013). Neither the feed velocity nor the surface roughness parameter (Ra, Rz, and Rmax) significantly influenced the flexural strength of Y-TZP.
View this table:
[in this window]
[in a new window]
|
Table 2. Four-point Flexural Strength, Weibull Parameter (n = 24), and Fracture Toughness KIc for Ground and Lapped Y-TZP Zirconia Ceramics (SD)
|
|
Weibull statistical analysis (Table 2) of the flexural strength data yielded two characteristic parameters for each test group: (1) the characteristic strength O, which corresponds to 63.29% probability of failure; and (2) the Weibull modulus m, indicating the slope of the ln (ln/1-p vs. ln plots), where p is the fracture probability. According to the reduction of flexural strength, the Weibull parameters were lowered by the simulated inner-surface grinding compared with the lapped control (Table 2). While the mean flexural strength and characteristic strength show similar variations, depending on the grinding parameter, the Weibull modulus decreased for samples ground with increased feed velocity (p = 0.055).
The Vickers hardness ranged from 12.17 to 13.70 GPa. With one exception, the fracture toughness of the ground samples exceeded that of the lapped samples (Table 2). The fracture toughness increased with the feed velocity (p = 0.022), while the cutting depth was of minor relevance.
|
DISCUSSION
|
|---|
The hypothesis that strength, reliability, and surface roughness of Y-TZP zirconia ceramics machined under conditions simulating the inner surface grinding of crowns and FPDs vary with grinding parameter (cutting depth, feed velocity) is partly accepted. The reduction in cutting depth shows significant influence on flexural strength, while the Weibull parameter m decreases with increasing feed velocity. These findings are not generally supported by product engineering literature, which shows a linear correlation between feed velocity and strength when diamond wheels of about 150 to 200 mm in diameter are used (Liao et al., 1997). Controversial investigations have found both that depth of cut has an effect on the strength of alumina (Tuan and Kuo, 1998), and that it has no effect (Liao et al., 1997). Nevertheless, the variation in grinding parameter seems not to be suitable for preservation of the strength and reliability of Y-TZP. The flexural strength of the lapped samples exceeds previously published values for as-sintered as well as sandblasted Y-TZP. This finding is confirmed by the examination of the Weibull parameter. The 50% reduction in flexural strength compared with that of the lapped control agrees with reported results (Kosmac et al., 1999). The fracture toughness of lapped samples corresponds to measurements of as-sintered specimens (Kosmac et al., 1999). Lapping ceramics minimizes both surface damage and surface compressive stress. While lapped samples show a fracture toughness of 4.4 MPa m1/2, samples after simulated inner surface grinding show a significant increase.
Co-analysis of Variables
The hypothesis that strength and surface roughness parameters depend on each other is accepted for the latter measured in the grinding direction. The co-analysis of variables finds complex relationships among several factors. The fracture toughness of the ground samples increases with feed velocity, while the Weibull parameter m decreases with increasing feed velocity. These counteracting effects could explain the lack of correlation between grinding parameter and flexural strength. A similar effect was described for sandblasted Y-TZP that showed significantly increased strength compared with as-sintered samples, while the Weibull parameter m decreased (Kosmac et al., 1999). The lack of correlation between grinding parameters and flexural strength may be due to machining-induced near-surface characteristics, because the Griffith strength relation used for the calculation of fracture toughness does not take into account any residual stresses (Liao et al., 1997). The measurements are most likely affected by two counteracting mechanisms previously described for conventional dental porcelain (Giordano et al., 1995) and alumina (Tuan and Kuo, 1998). The introduction of residual surface compressive stress (Liao et al., 1997; Tuan and Kuo, 1998; Kosmac et al., 1999) increases the surface toughness of zirconia-toughened ceramics considerably. Grinding introduces deep surface flaws that act as stress concentrators and may determine fracture toughness if their length largely exceeds the depth of the grinding-induced surface compressive layer (Sindel et al., 1998; Tuan and Kuo, 1998). The large standard deviation of the surface roughness parameter indicates different types of material removal in samples ground with identical parameters. These differences are likely to be related to the diamond tool, especially the numbers and shapes of the active diamond grains.
Within the limitations of this study, the conclusion could be drawn that methods for the fabrication of Y-TZP restorations should be optimized to fulfill the aim of CAD/CAM-manufactured crowns and FPDs of Y-TZP with improved strength and reliability. There is a need for further investigation to examine the influence of outer surface veneering and bonding of the restoration with composite resin that may reduce the strength and reliability degradation induced by grinding.
|
ACKNOWLEDGMENTS
|
|---|
This study was supported by the TMWFK (Thuringian Ministry of Science, Research and Culture), Grant B 403-97005, and in part by Girrbach-Dental GmbH, Pforzheim, Germany. This paper is based in part on a poster presented at the 35th Annual Meeting of the Continental European Division of the International Association for Dental Research (IADR/CED), Montpellier, France, September, 1999.
Received July 23, 2001;
Last revision May 1, 2002;
Accepted May 15, 2002
|
REFERENCES
|
|---|
EN DIN (1995). Advanced technical ceramics—monolithic ceramics—mechanical properties at room temperature. Part 1: Determination of flexural strength. 843-1. Berlin: Beuth-Verlag.
Giordano R, Cima M, Pober R (1995). Effect of surface finish on the flexural strength of feldspathic and aluminous dental ceramics. Int J Prosthodont 8:311–319.[Medline]
Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L (1999). The effect of surface grinding and sandblasting on flexural strength and reliability of Y-TZP zirconia ceramic. Dent Mater15:426–433.[Medline]
Liao TW, Li K, Breder K (1997). Flexural strength of ceramics ground under widely different conditions. J Mater Process Tech 70:198–206.
Luthardt RG, Rieger W, Musil R (1997). Grinding of zirconia-TZP in dentistry—CAD/CAM technology for the manufacturing of fixed dentures. In: Bioceramics 10. Proceedings of the 10th International Symposium on Ceramics in Medicine, Oct. 5-8, 1997, Paris, France. Sedel L, Rey C, editors. Oxford, UK: Elsevier Science Ltd, pp. 437-440.
Pfeiffer W, Hollstein T (1997). Characterisation and assessment of machined ceramic surfaces. Interceram 46:98–102.
Sindel J, Petschelt A, Grellner F, Dierken C, Greil P (1998). Evaluation of subsurface damage in CAD/CAM machined dental ceramics. J Mater Sci-Mater Med 9:291–295.
Stevens R (1986). Zirconia and zirconia ceramics. 2nd ed. Twickenham: Magnesium Electron Publication No. 113.
Thompson JY, Anusavice KJ, Naman A, Morris HF (1994). Fracture surface characterization of clinically failed all-ceramic crowns. J Dent Res 73:1824–1832.[Abstract/Free Full Text]
Tinschert J, Natt G, Mautsch W, Spiekermann H, Anusavice KJ (2001). Marginal fit of alumina-and zirconia-based fixed partial dentures produced by a CAD/CAM system. Oper Dent 26:367–374.[Medline]
Tuan WH, Kuo JC (1998). Effect of abrasive grinding on the strength and reliability of alumina. J Eur Ceram Soc 18:799–806.
TOP
ABSTRACT
INTRODUCTION
