| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RESEARCH REPORT |
1 Department of Dental Prosthetics, Section of Dental Materials, University RWTH Aachen, Germany; and
2 Department of Ceramics and Refractory Materials, University RWTH Aachen, Mauerstrasse 5, D-52064 Aachen, Germany;
*corresponding author, h.fischer{at}rwth-aachen.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: dental ceramics • surface roughness • failure stress • roughness-free strength
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A parameter which affects the flexural strength of a ceramic component, besides the natural microscopic defect population, is surface roughness. The surface roughness of a ceramic restoration depends on the final preparation of the veneering material. This is important, since almost every dental restoration is selectively ground to correct static and dynamic occlusal interferences (Denbo, 1990; Olthoff et al., 2000; Olsson and Lindqvist, 2002). If the peak-to-valley height of the surface roughness is in the range of the critical defect size value, the roughness can affect the flexural strength. The fundamental fracture mechanics equation (Griffith criterion) is given by
 |
where ac is the critical defect size, 
c is the critical (tensile) stress, KIc is the fracture toughness, and Y is a constant based on the geometry of the microscopic defect. From this equation, it can be deduced that, with an increased defect size, a decrease in the critical (failure) stress occurs. Since ceramic materials fail due to the ’weakest-link principle’, the maximum (critical) microscopic defect will cause failure at the critical stress.
The critical defect size can be estimated if the (flexural) failure stress of a brittle material is known. Based on the equation, there should theoretically be no risk of failure, if the specimen consists only of microscopic defects below this estimated critical value. Our hypothesis is that surface roughnesses below this microscopic defect size—based only on fracture mechanics considerations—also affect the flexural strength distribution of ceramic materials. We created different defined surface roughnesses on flexural strength specimens to prove this hypothesis. The strength was evaluated in a four-point bending test arrangement. Based on these results, the correlation between surface roughness and failure stress was evaluated. We also discuss the significance of these results for the dental field.
|
MATERIALS & METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subsequently, all specimens were annealed at a temperature 100°K below the respective transition temperature of the material in a ceramic oven (Austromat 300; Dekema, Freilassing, Germany) for 10 hrs to minimize any residual stresses induced by the ceramic process or by the surface treatment (air-abrading and polishing). After the annealing process, the specimens were very slowly cooled to room temperature in the closed oven overnight.
The surface roughness of all specimens was examined by means of a stylus analyzer (Perthometer PRK/S6P; Feinprüf Perthen, Göttingen, Germany). The roughness values Ra, Rz, and Rmax were evaluated on those sides of the specimens which were loaded by maximum tensile stresses during the subsequent flexural strength test between the inner roller spans of the four-point bending test arrangement. The value Rmax was taken as a criterion for judging the influence of the roughness on the strength. This value is defined as the maximum of the peak-to-valley heights of the measured (reference) section.
The flexural strength values were evaluated in a four-point bending test arrangement on a universal testing machine (Z030; Zwick, Ulm, Germany) according to DIN EN 843-1 (1995). The outer and inner roller spans were 24 mm and 12 mm, respectively. The crosshead speed was 0.5 mm/min. The mean strength values and their standard deviations were determined. Moreover, the characteristic strength values 0 and the Weibull modulus values m were conducted based on Weibull theory (Weibull, 1939; Thoman et al., 1970).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The specimens of the ’Polished’ batch showed the smoothest, and those of the ’Korox 250’ batch the roughest surfaces, respectively. The mean failure stress values were in the range of 65.3 to 102.7 MPa (characteristic strength: 68.3 to 108.3 MPa) for Empress 1, 48.1 to 88.0 MPa (50.5 to 93.7 MPa) for Empress 2, 49.2 to 72.9 MPa (52.7 to 77.8 MPa) for Symbio Ceram, and 48.6 to 72.4 MPa (52.0 to 77.2 MPa) for Vita Akzent. A linear correlation between the roughness values Rmax and the mean failure stress values m, as well as between the roughness values Rmax and the characteristic strength values 0, was found for all 4 investigated materials (Fig.). The correlation coefficients for the linear regression curves (least-square fits) were in the range of 0.9796 to 0.9998 for the analysis based on a Gauss stress distribution, and between 0.9725 and 0.9923 based on a Weibull stress distribution (Table 2).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). These results were not influenced by residual stresses, since the specimens were carefully annealed after the surface treatments. The strength of the specimens decreased continuously with the increase in surface roughness. This finding is in good agreement with those of another study (de Jager et al., 2000), in which this correlation was also found on dental ceramics, although the mean surface roughness value Ra was taken as the criterion for describing the surface condition in that study. We decided to use the Rmax value as a criterion, considering the ’weakest-link principle’ of the failure behavior of ceramic materials. Moreover, in the study mentioned, the strength was evaluated in a biaxial test, which induces a very complex state of stress in the specimens. The four-point bending test arrangement induces a much more homogenous state of stress.
The correlation coefficients of the least-squares fits through the respective data points (Fig.; Table 2) confirm a linear correlation between the roughness value Rmax and the mean failure stress c as well as the characteristic strength 0. This linear correlation gives the opportunity for "roughness-free" failure stress values to be extrapolated, i.e., strength values of the respective material with an ideal smooth-surface quality. These values allow for an easier comparison of different ceramic materials with respect to their "basic" strengths. The roughness-free failure stress is given by the value of the abscissa in the equation of the linear regression curve (Fig.). The equation of the regression curve based on the least-squares fit reveals another material characteristic: The smaller the value for the slope, the greater the strength decrease with an increase of surface roughness. Based on the two parameters of the regression curve equation, the veneering material of the ceramic system Empress 1 exhibits a higher roughness-free strength (abscissa: 111.2 MPa), but it is more sensitive to roughnesses with respect to the strength decrease (slope: -1.056) than are the other 3 ceramic veneering materials investigated (slope: -0.625 to -0.983). The results reveal that the final preparation of a ceramic restoration has a decisive influence on the strength of the material. Roughnesses in the magnitude of 40 µm can decrease the strength from 37% (Symbio Ceram) up to 48% (Empress 2) compared with the mean roughness-free failure stress, respectively. The well-known strategy of re-firing of rough ceramic surfaces does not increase the flexural strength (Griggs et al., 1996). Therefore, it is recommended that the ceramic veneering material be carefully polished as the final step of the selective grinding process so that very smooth—"roughness-free"—surfaces can be achieved.
The value Rmax is the roughness parameter that indicates the magnitude of the peak-to-valley height introduced by the surface treatment. It varied between 2.5 and 45.2 µm (Table 1). If we assume a typical fracture toughness value of 1.1 MPam0.5 for glass ceramic veneering materials (Marx et al., 2000, 2001) and a value of 1.13 for the geometry constant Y in the equation (Munz and Fett, 1999), critical microscopic defect sizes between 89.8 and 409.6 µm can be estimated for the veneer ceramics used. The maximum defects were significantly below these "theoretical critical microscopic defect" sizes. This re-states our hypothesis that surface roughnesses below this microscopic defect size—based only on fracture mechanics considerations—can also affect the flexural strength distribution of ceramic materials. A reason for the discrepancies may be notch effects under the assumption that the surface roughness profile can be interpreted as a population of (microscopic) notches. It is known, from technical publications, that microscopic notches can significantly decrease strength (Hertel et al., 1998; Fett et al., 1999). This means that the mechanical behavior will be superposed by statistical effects. Consequently, the application of the fundamental fracture mechanics equation is limited for the prediction of strength of specimens with rough surfaces. The extrapolation of a (mean) roughness-free failure stress value is the more reliable method for determining the (basic) strength of rough ceramic specimens.
|
ACKNOWLEDGMENTS |
|---|
Received February 3, 2003; Last revision July 7, 2003; Accepted September 8, 2003
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS & METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Denbo JA (1990). Malocclusion. Dent Clin North Am 34:103–109.[ISI][Medline]
DIN EN 843-1 (1995). Advanced technical ceramics—monolithic ceramics—mechanical tests at room temperature—Part 1: Determination of flexural strength. Berlin: Beuth.
Fett T, Hertel D, Munz D (1999). Strength of notched ceramic bending bars. J Mater Sci Lett 18:289–293.
Green DJ (1998). An introduction to the mechanical properties of ceramics. Cambridge: Cambridge University Press.
Griggs JA, Thompson JY, Anusavice KJ (1996). Effects of flaw size and auto-glaze treatment on porcelain strength. J Dent Res 75:1414–1417.
Hertel D, Fett T, Munz D (1998). Strength prediction for notched alumina specimens. J Eur Ceram Soc 18:329–338.
Marx R, Tinschert J, Fischer H (2000). Subcritical crack growth of dental ceramic materials. In: Materials for medical engineering. Stallforth H, Revell P, editors. Weinheim: Wiley-VCH, pp. 116-119.
Marx R, Fischer H, Weber M, Jungwirth F (2001). Crack parameters and Weibull moduli: subcritical crack growth and durability of all-ceramic materials. Dtsch Zahnärztl Z 56:90–98.
Munz D, Fett T (1999). Ceramics: mechanical properties, failure behaviour, materials selection. 1st ed. Berlin: Springer.
Olsson M, Lindqvist B (2002). Occlusal interferences in orthodontic patients before and after treatment, and in subjects with minor orthodontic treatment need. Eur J Orthod 24:677–687.
Olthoff LW, Van Der Zeel JM, De Ruiter WJ, Vlaar ST, Bosman F (2000). Computer modeling of occlusal surfaces of posterior teeth with the CICERO CAD/CAM system. J Prosthet Dent 84:154–162.[ISI][Medline]
Ritter JE (1995). Predicting lifetimes of materials and material structures. Dent Mater 11:142–146.[ISI][Medline]
Thoman DR, Bain LJ, Antle CE (1970). Maximum likelihood estimation, exact confidence interval for reliability, and tolerance limits in the Weibull distribution. Technometrics 12:363–371.
Weibull W (1939). A statistical theory of the strength of materials. Ingeniörs Vetenskaps Akademien-Handlingar 151.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| IADR Journals | Advances in Dental Research ® |
| Journal of Dental Research ® | Critical Reviews (1990-2004) |
