- Email: [email protected]
Subsurface damage induced in dental resurfacing of a feldspar porcelain with coarse diamond burs
ARTICLE IN PRESS Journal of Biomechanics 42 (2009) 355–360
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Subsurface damage induced in dental resurfacing of a feldspar porcelain with coarse diamond burs Xiao-Fei Song a, Ling Yin b, a b
School of Mechanical Engineering, Tianjin University, Tianjin 300072, China School of Engineering, James Cook University, Townsville, QLD 4811, Australia
a r t i c l e in fo
abstract
Article history: Accepted 27 November 2008
The primary cause for early failure of ceramic restorations is surface and subsurface damage induced in adjustment and resurfacing using dental handpieces/burs. This paper reports finite element analysis (FEA) modelling of dental resurfacing to predict the degrees of subsurface damage, in combination with experimental measurement using scanning electron microscopy (SEM). The FEA predictions of subsurface damage induced in a feldspar porcelain with coarse diamond burs were in agreement with the SEM experimental measurement. These findings provide dental clinicians a quantitative description of the response of dental resurfacing-induced subsurface damage. The implication of the results for non-destructive evaluation of subsurface damage by FEA modelling will be practically meaningful to clinical dental restorations for high-quality ceramic restorations. & 2008 Elsevier Ltd. All rights reserved.
Keywords: Dental ceramics Dental resurfacing Diamond burs Finite element analysis Subsurface damage
1. Introduction Dental bioceramics are attractive due to their superior biocompatibility, aesthetics and inertness (Ironside and Swain, 1998). These materials are able to be machined quickly to generate ceramic prostheses using dental chair-side computeraided design/computer-aided manufacture (CAD/CAM) systems (Rosenblum and Schulman, 1997). However, ceramics are brittle and subject to premature failure, especially in long-term cyclic loading and moist environments (Lawn et al., 2001). The reported clinical failure rate, approximately 3% per year for all-ceramic crowns, is unacceptably high relative to metal-core crowns (Kelly, 1999; Lawn et al., 2001). Furthermore, analyses of clinically failed crowns have proven that catastrophic fracture had always originated from surface and subsurface damage in the ceramic prostheses (Harvey and Kelly, 1996). In dentistry, ceramic prostheses are intraorally resurfaced and adjusted using dental high-speed handpieces and diamond burs for accurate marginal and occlusal fit. These dental machining processes induce surface and subsurface damage and contribute to clinical failure in ceramic prostheses (Rekow and Thompson, 2007). Therefore, evaluation of dental resurfacing-dependent damage is particularly required in restorative dentistry. Studies were conducted on characterization of in vitro dental resurfacing of ceramic prostheses using dental handpieces and diamond burs (Siegel and von Fraunhofer, 1999; Yin et al., 2003, 2006a).
Corresponding author. Tel.: +6174781 6254; fax: +617 4781 6788.
E-mail address: [email protected] (L. Yin). 0021-9290/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2008.11.031
Extensive chipping damage was found in ceramic prosthetic materials, especially in feldspar porcelains, when using coarse diamond burs (Yin et al., 2003, 2006a). However, little attention was paid to the relations between subsurface damage in ceramic prostheses and dental operational parameters applied in intraoral dental resurfacing. Finite element analysis (FEA) was successfully applied for prediction of subsurface damage depths in ceramic and glass machining using diamond tools (Zhang and Peng, 2000; Subhash and Zhang, 2003; Chuang et al., 2003). The machining-induced depths of subsurface damage in ceramics were associated with the machining parameters (Zhang and Peng, 2000; Subhash and Zhang, 2003; Chuang et al., 2003). In spite of the pervasive use of FEA in engineering, FEA modelling of dental resurfacing processes is little reported. In this paper, we report on FEA modelling of dental resurfacing of a feldspar porcelain with a coarse diamond bur for prediction of subsurface stress distributions and damage degrees, in combination with experimental measurement using scanning electron microscopy (SEM). A two-dimensional FEA was conducted with dental operational parameters and material properties as input variables. The maximum principal stress fields and the depths of subsurface damage were estimated as functions of the dental operational parameters.
2. In vitro dental resurfacing In vitro computer-assisted dental resurfacing operations were conducted using a novel experimental apparatus (Yin et al.,
ARTICLE IN PRESS 356
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
Dental Handpiece
Depth of Cut a
Bur Rotation Vs
Sample Feed Rate Vw Diamond Bur
Fig. 1. Dental handpiece/bur–sample movements in dental resurfacing.
2006b). During the in vitro resurfacing, the computer-assisted apparatus enabled a dental handpiece to realize precise feed and down feed movements. Fig. 1 schematically shows the dental handpiece–bur–prosthesis movements during resurfacing of a ceramic prosthesis. Samples with dimensions 15 12 5 mm3 were Vita Mark II (Vita Zahnfabrik, Bad Sa¨ckingen, Germany), which is a feldspar porcelain. The mechanical properties of the material are: Vickers hardness H ¼ 6.2 GPa, Young’s modulus E ¼ 68 GPa, fracture toughness KIC ¼ 0.9 MPa m1/2, strength s ¼ 100 MPa, and Poisson’s ratio n ¼ 0.2 (Kim et al., 2001; Deng et al., 2002). A new, nickel-coated, 106–125 mm grit diamond bur of diameter 1.4 mm (SF-21, ISO 110523014, Mani, Japan) was used. An air pressure of 0.2 MPa was applied to the handpiece; the unloaded bur rotated at a speed of 318 krpm. The bur was aligned parallel to the 12 5 mm2 surface of the prosthesis. During resurfacing, the dental bur was moved along the prosthetic surface at a depth of cut a and a feed rate vw as shown in Fig. 1. Water was sprayed onto the bur–prosthesis contact zone at a flow rate of 30 ml/min for cooling and cleaning of the resurfacing zone. The in vitro dental resurfacing operations were conducted where the depths of cut were in the range 10–50 mm, and feed rates of 15–60 mm/min. The tangential and normal resurfacing forces and bur speed during resurfacing were in-process measured using a force sensor and a high-speed data acquisition system (Yin et al., 2006b). After dental resurfacing, the cross sections of the resurfaced porcelain samples were polished manually using progressively smaller alumina abrasives of 1200, 1500, 2000, 3000, and 5000 mesh sizes and diamond pastes of 1.5 and 0.5 mm, respectively. The three largest depths of surface damage on the cross section for each sample were measured to determine the average value and the standard deviation using a scanning electron microscope (SEM, XL-30, Philips, Netherlands).
3. FEA modelling In dental resurfacing, the geometry of most practical removal operations is complex. The orthogonal removal model can simplify the model of material removal by neglecting many of the geometric complexities, yet describing the mechanics of the process quite well (Groover, 2007). Although an actual removal process is three-dimensional, the orthogonal model has only two dimensions that play active roles in the analysis. In this investigation, infeed surface removal was conducted, in which the depth of cut of the dental bur, i.e. infeed, was normal to its
direction of motion (feed), as shown in Fig. 1. Thus, the dental resurfacing process became two-dimensional orthogonal resurfacing. The FEA modelling of dental resurfacing could be regarded as a plane problem with a two-dimensional construction. Microcrack formation is an integral part of the process of diamond abrasive surface removal of brittle materials (Quinn et al., 2005). The microcrack response is dominated by residual stresses beneath the contact area between the brittle material and the diamond grits (Marshall et al., 1983). In dental resurfacing, the dental bur of multiple cutting points traversed a porcelain block surface and removed a layer of material. Localized penetrations of the work material by hard, sharp individual diamond grits embedded in the dental bur surface create a complex combination of plastic flow and fracture (Yin et al., 2006a). It is of great interest of an individual chip produced by individual diamond grits. The chip longitudinal shape before cutting and its assumed crosssectional shape are widely documented (Malkin, 1989). Stress fields in the porcelain are generated due to the interactions between the prosthetic material and the diamond grits. Therefore, a physical FEA model was established focusing on subsurface stresses and the damage zone formed beneath the individual chips produced by the diamond grit–prosthesis contact loads. The porcelain material was assumed to be isotropic and homogeneous. When these stresses exceeded a threshold value, e.g., the ultimate strength of the material, subsurface damage was introduced in the prosthetic porcelain. For the FEA geometric model, the dimensions were selected according to Saint-Venant’s principle that stresses far from the points of load application will dissipate (Timoshenko and Goodier, 1970). The horizontal dimension was fixed at 0.8 mm, much larger than any depths of cut in the current study. The vertical dimension was 3 times the bur–porcelain contact arc length, which is a function of depth of cut and diameter of bur. The loads with a set of displacement vectors were applied in the bur–prosthesis contact zone (Chuang et al., 2003). The imposed displacement vectors in magnitude were proportional to the diamond grit depth of cut, i.e., the undeformed chip thickness, which is a function of dental operational parameters. Based on this FEA modelling, all stress components can be obtained using commercial software, ANSYS 8.0 (ANSYS Incorporated, USA). The maximum principal stress st was of particular interest. To quantitatively evaluate the resurfacing-induced subsurface damage, the maximum normal stress criterion associated with the failure of brittle material was applied. According to this criterion, it is assumed that subsurface damage is initiated when the maximum principal stress st exceeds the ultimate strength s of the porcelain material. Thus, the depths of the resurfacinginduced subsurface damage were predicted in the porcelain under different dental operational conditions. A one-way factorial analysis of variance (ANOVA) at a 5% significant level was applied for statistical analyses.
4. Results Fig. 2(a) shows the maximum principal stress distribution under the bur–prosthesis contact zone at the depth of cut of 10 mm and the feed rate of 15 mm/min. It demonstrates that the maximum principal stresses were concentrated under the bur–prosthesis contact zone. The stress values decreased rapidly at increasing distances to the contact surface. Fig. 2(b) shows the detailed stress region near the contact zone. It reveals that the stress values markedly increased when approaching the diamond bur/grit exit point. The maximum principal stress reached the highest value of 875 MPa at the bur/grit exit point.
ARTICLE IN PRESS X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
a
a
Contact
Contact
Zone
Zone
Sample 0
357
10
Sample 50
100
300
500
875
0
10
50
100
300
10 MPa
800 1215
50 MPa
b
b
500
50 MPa
100 MPa
100 MPa
300 MPa
300 MPa
10 MPa
500 MPa
500 MPa
800 MPa Max 1215 MPa
Max 875 MPa
Fig. 3(a) shows the maximum principal stress distribution under the bur–prosthesis contact zone at the depth of cut of 50 mm and the feed rate of 60 mm/min. Similar to the distribution trend in Fig. 2(a), the maximum principal stresses were located under the bur–prosthesis contact zone and reduced quickly when increasing the distance to the contact surface. In comparison with the stress field shown in Fig. 2, the maximum principal stresses at the deeper depth of cut and the higher feed rate were larger in magnitude and in distribution area. The detailed local stress field near the contact zone is shown in Fig. 3(b). The maximum principal stress value of 1215 MPa occurred at the bur/grit exit point. For selected operational conditions in this investigation, it is also found that all the maximum principal stress values were concentrated at the bur/grit exits, in the range 601–1755 MPa. Fig. 4 demonstrates the maximum principal stress st as a function of feed rate Vw for the depths of cut of 10 and 50 mm. The results show that maximum principal stress increased with the feed rate at each of the depth of cut. The increase in feed rate from 15 to 60 mm/min resulted in the doubling of the increase in maximum principal stress for both the depths of cut. Also, at any feed rate, the maximum principal stress values for the depth of cut of 10 mm were larger than those for the depth of cut of 50 mm. This indicates that at the smaller depth of cut, the maximum principal stress values were larger. Fig. 5 shows the FEA-predicted subsurface damage depth as a function of feed rate Vw for the depths of cut of 10 and 50 mm. It shows that subsurface damage depths for both the depths of cut increased approximately linearly with the feed rate, with a coefficient of determination R2 ¼ 0.97. When increasing the feed rate from 15 to 60 mm/min, subsurface damage depths for the depths of cut of 10 and 50 mm increased by 74% and 82%, from 23 to 40 mm and from 44 to 80 mm, respectively. It is also found that
Fig. 3. (a) Contour plot of maximum principal stress distribution, and (b) detailed plot of the stress distribution under the bur–prosthesis contact zone at the depth of cut of 50 mm and feed rate of 60 mm/min.
2000 Maximum Principal Stress σt (MPa)
Fig. 2. (a) Contour plot of maximum principal stress distribution, and (b) detailed plot of the stress distribution at the bur–prosthesis contact zone at the depth of cut of 10 mm and feed rate of 15 mm/min.
Depth of Cut 10 μm
1800
Depth of Cut 50 μm
1600 1400 1200 1000 800 600 400 0
15
35
45
60
75
Feed Rate Vw (mm/min) Fig. 4. Maximum principal stresses versus feed rate for different depths of cut.
subsurface damage depths for the depth of cut of 50 mm at any feed rate were approximately twice those for the depth of cut of 10 mm. Fig. 6 demonstrates a series of SEM micrographs of the cross sections of the resurfaced porcelain samples. Figs. 6a and b show the maximum subsurface damage depths produced at the feed rates of 15 and 60 mm/min for the depth of cut of 10 mm, respectively. They reveal that the maximum depth of subsurface damage at the feed rate of 60 mm/min is larger than that at the feed rate of 15 mm/min. Figs. 6c and d show the maximum subsurface damage produced at the feed rates of 15
ARTICLE IN PRESS 358
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
and 60 mm/min for the depth of cut of 50 mm, respectively. They also reveal that the higher feed rate resulted in deeper subsurface damage. In addition, it is obvious that the maximum depths of subsurface damage at the depth of cut 50 mm were much larger than those at the depth of cut 10 mm. For instance, the subsurface damage depths for the depth of cut 50 mm at the feed rate of 15 and 60 mm/min were about 41 and 74 mm shown in Fig. 6a and b, which are at least 1.46 times and 1.76 times the values of 28 and 42 mm for the depth of cut 10 mm shown in Fig. 6c and d, respectively. To verify the validity of the prediction model, maximum FEApredicted and average maximum SEM-measured depth of subsurface damage as a function of grit depths of cut are plotted in Fig. 7. The details of calculation of the diamond grit depths of cut could
be found in a previous study (Song et al., 2008). The results demonstrate that the maximum FEA-predicted subsurface damage depths are in agreement with the average maximum SEM-measured ones. It could also be found that both predicted and measured subsurface damage depths approximately linearly increased with the grit depth of cut with coefficients of determination Rpre2 ¼ 0.92 and Rexp2 ¼ 0.89, respectively.
5. Discussion Dental resurfacing of ceramic prostheses is typically achieved by multipoint surface abrasive removal. In this process the outer surface of high-speed rotating dental burs traverses and contacts
85
95
75
Depth of Subsurface Damage (μm)
Depth of Subsurface Damage (μm)
Depth of Cut 10 μm Depth of Cut 50 μm
65 55 45 35 25 15
FEA Prediction
85
Experimental Measurement
75 Rpre2 = 0.92
65
Rexp2 = 0.89
55 45 35 25 15
0
15
30
45
60
75
Feed Rate Vw (mm/min) Fig. 5. FEA-predicated depths of subsurface damage versus feed rate for different depths of cut.
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Diamond Grit Depth of Cut (μm) Fig. 7. Predicted and measured depths of subsurface damage versus diamond grit depth of cut.
Fig. 6. SEM micrographs of the maximum depths of subsurface damage induced in the porcelain samples at the depth of cut of 10 mm and the feed rates of (a) 15 mm/min and (b) 60 mm/min; at the depth of cut of 50 mm and the feed rates of (c) 15 mm/min and (d) 60 mm/min.
ARTICLE IN PRESS X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
ceramic prostheses. Abrasive removal of ceramic material is effected mainly by chipping and intersection of cracks from adjacent penetration sites, while cracks generated in an orientation normal to the surface (Marshall et al., 1983; Jahanmir et al., 1998; Quinn et al., 2005). An understanding of the nature of dental resurfacing damage is therefore an essential prerequisite for evaluating the reliability of brittle materials as dental restorations (Rekow and Thompson 2007). In engineering ceramics, the existence of machining-induced cracks oriented normal to the surface has been established in several studies (Marshall et al., 1983; Jahanmir et al, 1998; Quinn et al., 2005). The most definitive evidence comes from fractographic observations (Quinn et al., 2005). In dentistry, the failure rate for ceramic restoration is conspicuously high. The deformation and fracture generated by individual diamond grit contacts during dental resurfacing are expected to resemble the damage associated with sharp diamond indentation and scratching. Based on the nature of resurfacing damage in brittle materials, FEA modelling was constructed for dental resurfacing of a feldspar porcelain with a coarse diamond bur. The FEA results show that the trends for the maximum principal stress distributions under different operational conditions were similar. The maximum principal stresses were distributed near the bur–prosthesis contact zone with a rapid decrease with the distance to the resurfaced surface. The maximum principal stress values were found always at the bur/grit exits in spite of the differences in magnitude. This finding is in agreement with the results of FEA modelling of diamond wheel plunge grinding of silicon nitride (Chuang et al., 2003). A previous study observed extensive edge chipping damage occurring in resurfacing of the same porcelain using coarse burs (Yin et al., 2003). This edge damage may be attributed to stress concentration at bur/grit exits. The relations between the maximum principal stresses and the dental operational parameters indicate that the maximum principal stresses were significantly dependent on both the feed rate and the depth of cut (ANOVA, po0.01). As shown in Fig. 4, maximum principal stresses increased with feed rate. This is because that at a fixed depth of cut, an increase in feed rate does not result in a change in bur–prosthesis contact arc length but results in an increased diamond grit depth of cut. An increase in diamond grit depth of cut could result in an increase in resurfacing force in the contact zone. Thus, the increasing force leads to an increase in the maximum principal stress. Moreover, Fig. 4 also shows that at any feed rate, smaller maximum principal stresses were obtained at the larger depth of cut. It is likely that this can be attributed to the fact that at a fixed feed rate, increasing the depth of cut results both in increases in diamond grit depth of cut and bur–prosthesis contact arc length. Subsurface damage depths also exhibited a significant dependence on the dental operational parameters (ANOVA, po0.01). As shown in Fig. 5, subsurface damage degrees follow upward trends with either feed rate or depth of cut. This indicates that selections of suitable depths of cut and feed rates are both important because severe subsurface damage can result from high depths of cut or feed rate. This finding is also consistent with studies on machining-induced damage in diamond grinding of engineering ceramics (Yui et al., 1989; Maksoud et al., 1999). In addition, both predicted and measured subsurface damage depths increased approximately linearly with the diamond grit depth of cut. This relationship is consistent with the published results in FEA modelling of single grit scratching of ceramics (Subhash and Zhang, 2003). A previous study has reported that the average depths of subsurface damage induced in dental CAD/CAM machining of the same porcelain was 40–60 mm (Sindel et al., 1998). Our FEA prediction of subsurface damage depths are in the
359
range 20–80 mm. Such subsurface damage could not only reduce the accuracy of fit of prostheses, but also lead to reduction in mechanical strength and lifetime of dental prostheses (Shearer et al., 1993; Sindel et al., 1998). For high-quality dental restorations, dental practitioners must select the dental operational parameters to diminish subsurface damage while maintaining reasonable removal rates. The results of the FEA modelling are used, in combination with fractographic observations, to provide a qualitative description of the response of dental resurfacing cracks. Furthermore, the implication of the results for non-destructive evaluation of subsurface damage by FEA modelling will be practically meaningful in clinical dental restorations. Finally it is noted that the current FEA modelling was restricted to static model neglecting dynamic and thermal effects on dental resurfacing. However, this model nonetheless describes the salient features of the dental resurfacing process in its simulated form very well. On the basis of studies on brittle material removal, all the models were established in static regime (Evans and Marshall, 1981; Kirchner, 1984; Bifano et al., 1991; Malkin and Hwang, 1996; Jahanmir et al., 1998; Quinn et al., 2005) although the dynamic nature of removal is widely recognized. Compared to metal removal, whose first static chip formation model was established in early 1940s (Merchant, 1945), studies on brittle material removal can only be traced from late 1970s. In metal removal, until 1997 a modified model for metal chip formation considering a mechanism for thermomechanical feedback was developed (Burns and Davies, 1997). In brittle material removal, establishment of a complete model of chip formation with all dynamic, thermal and wear feedbacks will be a difficult target for the years to come.
6. Conclusions FEA modelling and SEM measurement were conduced to quantitatively investigate the dental resurfacing-induced subsurface damage in a feldspar porcelain. The results reveal that maximum principal stresses induced by bur–prosthesis interactions were almost all concentrated at bur/grit exits. The maximum values of these stresses increased with the feed rate and decreased with the depth of cut. Increasing either the feed rate or the depth of cut resulted in an increase in subsurface damage depth. The results suggest that both the feed rate and the depth of cut are important parameters in controlling the degrees of subsurface damage in clinical dental resurfacing operations (ANOVA, po0.01). These results provide insights into the prediction of the quality of ceramic prostheses and guidance as to the proper selection of resurfacing operational parameters in dental practice.
Conflict of interest statement The authors have not entered into a commercial relationship with any other individuals or organizations, that might include employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/registrations, and grants or other funding, other than funding by the National Natural Science Foundation of China Project Grant no. 50475115.
Acknowledgements This work was supported by the NSFC Project Grant no. 50475115. We thank Dr. Anthony Flynn of the Australian National University for valuable comments.
ARTICLE IN PRESS 360
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
References Bifano, T.G., Dow, T.A., Scattergood, R.O., 1991. Ductile-regime grinding: a new technology for machining brittle materials. Transactions of ASME—Journal of Engineering for Industry 113, 184–189. Burns, T.J., Davies, M.A., 1997. Nonlinear dynamics model for chip segmentation in machining. Physical Review Letters 79, 447–450. Chuang, T-J., Jahanmir, S., Tang, H.C., 2003. Finite element simulation of straight plunge grinding for advanced ceramics. Journal of the European Ceramic Society 23, 1723–1733. Deng, Y., Lawn, B.R., Lloyd, I.K., 2002. Characterization of damage modes in dental ceramic bilayer structures. Journal of Biomedical Materials Research Part B: Applied Biomaterials 63, 137–145. Evans, A.G., Marshall, D.B., 1981. Wear mechanisms in ceramics. In: Rigney, D.A. (Ed.), Fundamentals of Friction and Wear of Materials. American Society for Metals, Metals Park, OH, USA. Groover, M.P., 2007. Fundamentals of Modern Manufacturing—Materials, Processes and Systems, third ed. Wiley, New Jersey, USA. Harvey, C.K., Kelly, J.R., 1996. Contact damage as a failure mode during in vitro testing. Journal of Prosthodontics 5, 95–100. Ironside, J.G., Swain, M.V., 1998. Ceramics in dental restorations—a review and critical issues. Journal of the Australasian Ceramic Society 34, 78–91. Jahanmir, S., Xu, H.K., Ives, L.K., 1998. Mechanisms of material removal in abrasive machining of ceramics. In: Jahanmir, S., Ramulu, M., Koshy, P. (Eds.), Machining of Ceramics and Composites. Marcel Dekker, New York, USA. Kelly, J.R., 1999. Clinically relevant approach to failure testing of all-ceramic restorations. Journal of Prosthetics Dentistry 81 (6), 652–661. Kim, H-W., Deng, Y., Miranda, P., Pajares, A., Kim, D.K., Kim, H-E., Lawn, B.R., 2001. Effect of flaw state on the strength of brittle coatings on soft substrates. Journal of the American Ceramic Society 84, 2377–2384. Kirchner, H.P., 1984. Damage penetration at elongated machining grooves in hotpressed Si3N4. Journal of the American Ceramic Society 67, 127–132. Lawn, B.R., Deng, Y., Thompson, V.P., 2001. Use of contact testing in the characterization and design of all-ceramic crown-like layer structures: a review. Journal of Prosthetics Dentistry 86, 495–510. Maksoud, T.M.A., Mokbel, A.A., Morgan, J.E., 1999. Evaluation of surface and sub-surface cracks of ground ceramic. Journal of Material Processing Technology 88, 222–243. Malkin, S., 1989. Grinding Technology: Theory and Application of Machining with Abrasives. Willey, New York, USA. Malkin, S., Hwang, T.W., 1996. Grinding mechanisms for ceramics. Annals of CIRP 45, 569–580.
Marshall, D.B., Evans, A.G., Khuri Yakub, B.T., Tien, J.W., Kina, G.S., 1983. The nature of machining damage in brittle materials. Proceedings of the Royal Society of London, Series A, Mathematical and Physical Sciences 385, 461–475. Merchant, M.E., 1945. Mechanics of the metal cutting process: II. Plasticity conditions in orthogonal cutting. Journal of Applied Physics 16, 318–324. Quinn, G.D., Ives, L.K., Jahanmir, S., 2005. On the nature of machining cracks in ground ceramics: part I: SRBSN strengths and fractographic analysis. Machining Science and Technology 9, 169–210. Rekow, D., Thompson, V.P., 2007. Engineering long term clinical success of advanced ceramic prostheses. Journal of Materials Science: Materials in Medicine 18, 47–56. Rosenblum, M., Schulman, A., 1997. A review of all-ceramic restorations. Journal of American Dental Association 128, 297–307. Shearer, A.C., Heymann, H.O., Wilson, N.H., 1993. Two ceramic materials compared for the production of CEREC inlays. Journal of Dentistry 21 (5), 302–304. Siegel, S.C., von Fraunhofer, J.A., 1999. Dental cutting with diamond burs: heavyhanded or hight-tough? Journal of Prosthodontics 8, 3–9. Sindel, J., Petschelt, A., Grellner, F., Dierken, C., Greil, P., 1998. Evaluation of subsurface damage in CAD/CAM machined dental ceramics. Journal of Materials Science: Materials in Medicine 9, 291–295. Song, X.F., Yin, L., Han, Y.G., Li, J., 2008. Finite element analysis of subsurface damage of ceramic prostheses in simulated intraoral dental resurfacing. Journal of Biomedical Materials Research Part B: Applied Biomaterials 85B, 50–59. Subhash, G., Zhang, W., 2003. Finite element analysis of brittle cracking due to single grit rotating scratch. Journal of Applied Mechanics 70, 147–151. Timoshenko, S.P., Goodier, J.N., 1970. Theory of Elasticity. McGraw-Hill, New York. Yin, L., Jahanmir, S., Ives, L.K., 2003. Abrasive machining of porcelain and zirconia with a dental handpiece. Wear 255, 975–989. Yin, L., Song, X.F., Qu, S.F., Han, Y.G., Wang, H., 2006a. Surface integrity and removal mechanism in simulated dental finishing of a feldspathic porcelain. Journal of Biomedical Materials Research Part B: Applied Biomaterials 79B, 365–378. Yin, L., Song, X.F., Qu, S.F., Huang, T., Mei, J.P., Yang, Z.Y., Li, J., 2006b. Performance evaluation of a dental handpiece in simulation of clinical finishing using a novel 2-DOF in vitro apparatus. Proceeding of Institution of Mechanical Engineers Part H: Journal of Engineering in Medicine 220, 929–938. Yui, A., Watanabe, T., Yoshida, Y., 1989. Effect of grinding conditions upon bending strength of alumina ceramics. Grinding Fundamentals and Applications. American Society of Mechanical Engineering PED 39, 229–239. Zhang, B., Peng, X.H., 2000. Grinding damage prediction for ceramics via CDM model. Journal of Manufacturing Science and Engineering 122, 51–58.
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Subsurface damage induced in dental resurfacing of a feldspar porcelain with coarse diamond burs Xiao-Fei Song a, Ling Yin b, a b
School of Mechanical Engineering, Tianjin University, Tianjin 300072, China School of Engineering, James Cook University, Townsville, QLD 4811, Australia
a r t i c l e in fo
abstract
Article history: Accepted 27 November 2008
The primary cause for early failure of ceramic restorations is surface and subsurface damage induced in adjustment and resurfacing using dental handpieces/burs. This paper reports finite element analysis (FEA) modelling of dental resurfacing to predict the degrees of subsurface damage, in combination with experimental measurement using scanning electron microscopy (SEM). The FEA predictions of subsurface damage induced in a feldspar porcelain with coarse diamond burs were in agreement with the SEM experimental measurement. These findings provide dental clinicians a quantitative description of the response of dental resurfacing-induced subsurface damage. The implication of the results for non-destructive evaluation of subsurface damage by FEA modelling will be practically meaningful to clinical dental restorations for high-quality ceramic restorations. & 2008 Elsevier Ltd. All rights reserved.
Keywords: Dental ceramics Dental resurfacing Diamond burs Finite element analysis Subsurface damage
1. Introduction Dental bioceramics are attractive due to their superior biocompatibility, aesthetics and inertness (Ironside and Swain, 1998). These materials are able to be machined quickly to generate ceramic prostheses using dental chair-side computeraided design/computer-aided manufacture (CAD/CAM) systems (Rosenblum and Schulman, 1997). However, ceramics are brittle and subject to premature failure, especially in long-term cyclic loading and moist environments (Lawn et al., 2001). The reported clinical failure rate, approximately 3% per year for all-ceramic crowns, is unacceptably high relative to metal-core crowns (Kelly, 1999; Lawn et al., 2001). Furthermore, analyses of clinically failed crowns have proven that catastrophic fracture had always originated from surface and subsurface damage in the ceramic prostheses (Harvey and Kelly, 1996). In dentistry, ceramic prostheses are intraorally resurfaced and adjusted using dental high-speed handpieces and diamond burs for accurate marginal and occlusal fit. These dental machining processes induce surface and subsurface damage and contribute to clinical failure in ceramic prostheses (Rekow and Thompson, 2007). Therefore, evaluation of dental resurfacing-dependent damage is particularly required in restorative dentistry. Studies were conducted on characterization of in vitro dental resurfacing of ceramic prostheses using dental handpieces and diamond burs (Siegel and von Fraunhofer, 1999; Yin et al., 2003, 2006a).
Corresponding author. Tel.: +6174781 6254; fax: +617 4781 6788.
E-mail address: [email protected] (L. Yin). 0021-9290/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbiomech.2008.11.031
Extensive chipping damage was found in ceramic prosthetic materials, especially in feldspar porcelains, when using coarse diamond burs (Yin et al., 2003, 2006a). However, little attention was paid to the relations between subsurface damage in ceramic prostheses and dental operational parameters applied in intraoral dental resurfacing. Finite element analysis (FEA) was successfully applied for prediction of subsurface damage depths in ceramic and glass machining using diamond tools (Zhang and Peng, 2000; Subhash and Zhang, 2003; Chuang et al., 2003). The machining-induced depths of subsurface damage in ceramics were associated with the machining parameters (Zhang and Peng, 2000; Subhash and Zhang, 2003; Chuang et al., 2003). In spite of the pervasive use of FEA in engineering, FEA modelling of dental resurfacing processes is little reported. In this paper, we report on FEA modelling of dental resurfacing of a feldspar porcelain with a coarse diamond bur for prediction of subsurface stress distributions and damage degrees, in combination with experimental measurement using scanning electron microscopy (SEM). A two-dimensional FEA was conducted with dental operational parameters and material properties as input variables. The maximum principal stress fields and the depths of subsurface damage were estimated as functions of the dental operational parameters.
2. In vitro dental resurfacing In vitro computer-assisted dental resurfacing operations were conducted using a novel experimental apparatus (Yin et al.,
ARTICLE IN PRESS 356
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
Dental Handpiece
Depth of Cut a
Bur Rotation Vs
Sample Feed Rate Vw Diamond Bur
Fig. 1. Dental handpiece/bur–sample movements in dental resurfacing.
2006b). During the in vitro resurfacing, the computer-assisted apparatus enabled a dental handpiece to realize precise feed and down feed movements. Fig. 1 schematically shows the dental handpiece–bur–prosthesis movements during resurfacing of a ceramic prosthesis. Samples with dimensions 15 12 5 mm3 were Vita Mark II (Vita Zahnfabrik, Bad Sa¨ckingen, Germany), which is a feldspar porcelain. The mechanical properties of the material are: Vickers hardness H ¼ 6.2 GPa, Young’s modulus E ¼ 68 GPa, fracture toughness KIC ¼ 0.9 MPa m1/2, strength s ¼ 100 MPa, and Poisson’s ratio n ¼ 0.2 (Kim et al., 2001; Deng et al., 2002). A new, nickel-coated, 106–125 mm grit diamond bur of diameter 1.4 mm (SF-21, ISO 110523014, Mani, Japan) was used. An air pressure of 0.2 MPa was applied to the handpiece; the unloaded bur rotated at a speed of 318 krpm. The bur was aligned parallel to the 12 5 mm2 surface of the prosthesis. During resurfacing, the dental bur was moved along the prosthetic surface at a depth of cut a and a feed rate vw as shown in Fig. 1. Water was sprayed onto the bur–prosthesis contact zone at a flow rate of 30 ml/min for cooling and cleaning of the resurfacing zone. The in vitro dental resurfacing operations were conducted where the depths of cut were in the range 10–50 mm, and feed rates of 15–60 mm/min. The tangential and normal resurfacing forces and bur speed during resurfacing were in-process measured using a force sensor and a high-speed data acquisition system (Yin et al., 2006b). After dental resurfacing, the cross sections of the resurfaced porcelain samples were polished manually using progressively smaller alumina abrasives of 1200, 1500, 2000, 3000, and 5000 mesh sizes and diamond pastes of 1.5 and 0.5 mm, respectively. The three largest depths of surface damage on the cross section for each sample were measured to determine the average value and the standard deviation using a scanning electron microscope (SEM, XL-30, Philips, Netherlands).
3. FEA modelling In dental resurfacing, the geometry of most practical removal operations is complex. The orthogonal removal model can simplify the model of material removal by neglecting many of the geometric complexities, yet describing the mechanics of the process quite well (Groover, 2007). Although an actual removal process is three-dimensional, the orthogonal model has only two dimensions that play active roles in the analysis. In this investigation, infeed surface removal was conducted, in which the depth of cut of the dental bur, i.e. infeed, was normal to its
direction of motion (feed), as shown in Fig. 1. Thus, the dental resurfacing process became two-dimensional orthogonal resurfacing. The FEA modelling of dental resurfacing could be regarded as a plane problem with a two-dimensional construction. Microcrack formation is an integral part of the process of diamond abrasive surface removal of brittle materials (Quinn et al., 2005). The microcrack response is dominated by residual stresses beneath the contact area between the brittle material and the diamond grits (Marshall et al., 1983). In dental resurfacing, the dental bur of multiple cutting points traversed a porcelain block surface and removed a layer of material. Localized penetrations of the work material by hard, sharp individual diamond grits embedded in the dental bur surface create a complex combination of plastic flow and fracture (Yin et al., 2006a). It is of great interest of an individual chip produced by individual diamond grits. The chip longitudinal shape before cutting and its assumed crosssectional shape are widely documented (Malkin, 1989). Stress fields in the porcelain are generated due to the interactions between the prosthetic material and the diamond grits. Therefore, a physical FEA model was established focusing on subsurface stresses and the damage zone formed beneath the individual chips produced by the diamond grit–prosthesis contact loads. The porcelain material was assumed to be isotropic and homogeneous. When these stresses exceeded a threshold value, e.g., the ultimate strength of the material, subsurface damage was introduced in the prosthetic porcelain. For the FEA geometric model, the dimensions were selected according to Saint-Venant’s principle that stresses far from the points of load application will dissipate (Timoshenko and Goodier, 1970). The horizontal dimension was fixed at 0.8 mm, much larger than any depths of cut in the current study. The vertical dimension was 3 times the bur–porcelain contact arc length, which is a function of depth of cut and diameter of bur. The loads with a set of displacement vectors were applied in the bur–prosthesis contact zone (Chuang et al., 2003). The imposed displacement vectors in magnitude were proportional to the diamond grit depth of cut, i.e., the undeformed chip thickness, which is a function of dental operational parameters. Based on this FEA modelling, all stress components can be obtained using commercial software, ANSYS 8.0 (ANSYS Incorporated, USA). The maximum principal stress st was of particular interest. To quantitatively evaluate the resurfacing-induced subsurface damage, the maximum normal stress criterion associated with the failure of brittle material was applied. According to this criterion, it is assumed that subsurface damage is initiated when the maximum principal stress st exceeds the ultimate strength s of the porcelain material. Thus, the depths of the resurfacinginduced subsurface damage were predicted in the porcelain under different dental operational conditions. A one-way factorial analysis of variance (ANOVA) at a 5% significant level was applied for statistical analyses.
4. Results Fig. 2(a) shows the maximum principal stress distribution under the bur–prosthesis contact zone at the depth of cut of 10 mm and the feed rate of 15 mm/min. It demonstrates that the maximum principal stresses were concentrated under the bur–prosthesis contact zone. The stress values decreased rapidly at increasing distances to the contact surface. Fig. 2(b) shows the detailed stress region near the contact zone. It reveals that the stress values markedly increased when approaching the diamond bur/grit exit point. The maximum principal stress reached the highest value of 875 MPa at the bur/grit exit point.
ARTICLE IN PRESS X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
a
a
Contact
Contact
Zone
Zone
Sample 0
357
10
Sample 50
100
300
500
875
0
10
50
100
300
10 MPa
800 1215
50 MPa
b
b
500
50 MPa
100 MPa
100 MPa
300 MPa
300 MPa
10 MPa
500 MPa
500 MPa
800 MPa Max 1215 MPa
Max 875 MPa
Fig. 3(a) shows the maximum principal stress distribution under the bur–prosthesis contact zone at the depth of cut of 50 mm and the feed rate of 60 mm/min. Similar to the distribution trend in Fig. 2(a), the maximum principal stresses were located under the bur–prosthesis contact zone and reduced quickly when increasing the distance to the contact surface. In comparison with the stress field shown in Fig. 2, the maximum principal stresses at the deeper depth of cut and the higher feed rate were larger in magnitude and in distribution area. The detailed local stress field near the contact zone is shown in Fig. 3(b). The maximum principal stress value of 1215 MPa occurred at the bur/grit exit point. For selected operational conditions in this investigation, it is also found that all the maximum principal stress values were concentrated at the bur/grit exits, in the range 601–1755 MPa. Fig. 4 demonstrates the maximum principal stress st as a function of feed rate Vw for the depths of cut of 10 and 50 mm. The results show that maximum principal stress increased with the feed rate at each of the depth of cut. The increase in feed rate from 15 to 60 mm/min resulted in the doubling of the increase in maximum principal stress for both the depths of cut. Also, at any feed rate, the maximum principal stress values for the depth of cut of 10 mm were larger than those for the depth of cut of 50 mm. This indicates that at the smaller depth of cut, the maximum principal stress values were larger. Fig. 5 shows the FEA-predicted subsurface damage depth as a function of feed rate Vw for the depths of cut of 10 and 50 mm. It shows that subsurface damage depths for both the depths of cut increased approximately linearly with the feed rate, with a coefficient of determination R2 ¼ 0.97. When increasing the feed rate from 15 to 60 mm/min, subsurface damage depths for the depths of cut of 10 and 50 mm increased by 74% and 82%, from 23 to 40 mm and from 44 to 80 mm, respectively. It is also found that
Fig. 3. (a) Contour plot of maximum principal stress distribution, and (b) detailed plot of the stress distribution under the bur–prosthesis contact zone at the depth of cut of 50 mm and feed rate of 60 mm/min.
2000 Maximum Principal Stress σt (MPa)
Fig. 2. (a) Contour plot of maximum principal stress distribution, and (b) detailed plot of the stress distribution at the bur–prosthesis contact zone at the depth of cut of 10 mm and feed rate of 15 mm/min.
Depth of Cut 10 μm
1800
Depth of Cut 50 μm
1600 1400 1200 1000 800 600 400 0
15
35
45
60
75
Feed Rate Vw (mm/min) Fig. 4. Maximum principal stresses versus feed rate for different depths of cut.
subsurface damage depths for the depth of cut of 50 mm at any feed rate were approximately twice those for the depth of cut of 10 mm. Fig. 6 demonstrates a series of SEM micrographs of the cross sections of the resurfaced porcelain samples. Figs. 6a and b show the maximum subsurface damage depths produced at the feed rates of 15 and 60 mm/min for the depth of cut of 10 mm, respectively. They reveal that the maximum depth of subsurface damage at the feed rate of 60 mm/min is larger than that at the feed rate of 15 mm/min. Figs. 6c and d show the maximum subsurface damage produced at the feed rates of 15
ARTICLE IN PRESS 358
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
and 60 mm/min for the depth of cut of 50 mm, respectively. They also reveal that the higher feed rate resulted in deeper subsurface damage. In addition, it is obvious that the maximum depths of subsurface damage at the depth of cut 50 mm were much larger than those at the depth of cut 10 mm. For instance, the subsurface damage depths for the depth of cut 50 mm at the feed rate of 15 and 60 mm/min were about 41 and 74 mm shown in Fig. 6a and b, which are at least 1.46 times and 1.76 times the values of 28 and 42 mm for the depth of cut 10 mm shown in Fig. 6c and d, respectively. To verify the validity of the prediction model, maximum FEApredicted and average maximum SEM-measured depth of subsurface damage as a function of grit depths of cut are plotted in Fig. 7. The details of calculation of the diamond grit depths of cut could
be found in a previous study (Song et al., 2008). The results demonstrate that the maximum FEA-predicted subsurface damage depths are in agreement with the average maximum SEM-measured ones. It could also be found that both predicted and measured subsurface damage depths approximately linearly increased with the grit depth of cut with coefficients of determination Rpre2 ¼ 0.92 and Rexp2 ¼ 0.89, respectively.
5. Discussion Dental resurfacing of ceramic prostheses is typically achieved by multipoint surface abrasive removal. In this process the outer surface of high-speed rotating dental burs traverses and contacts
85
95
75
Depth of Subsurface Damage (μm)
Depth of Subsurface Damage (μm)
Depth of Cut 10 μm Depth of Cut 50 μm
65 55 45 35 25 15
FEA Prediction
85
Experimental Measurement
75 Rpre2 = 0.92
65
Rexp2 = 0.89
55 45 35 25 15
0
15
30
45
60
75
Feed Rate Vw (mm/min) Fig. 5. FEA-predicated depths of subsurface damage versus feed rate for different depths of cut.
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Diamond Grit Depth of Cut (μm) Fig. 7. Predicted and measured depths of subsurface damage versus diamond grit depth of cut.
Fig. 6. SEM micrographs of the maximum depths of subsurface damage induced in the porcelain samples at the depth of cut of 10 mm and the feed rates of (a) 15 mm/min and (b) 60 mm/min; at the depth of cut of 50 mm and the feed rates of (c) 15 mm/min and (d) 60 mm/min.
ARTICLE IN PRESS X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
ceramic prostheses. Abrasive removal of ceramic material is effected mainly by chipping and intersection of cracks from adjacent penetration sites, while cracks generated in an orientation normal to the surface (Marshall et al., 1983; Jahanmir et al., 1998; Quinn et al., 2005). An understanding of the nature of dental resurfacing damage is therefore an essential prerequisite for evaluating the reliability of brittle materials as dental restorations (Rekow and Thompson 2007). In engineering ceramics, the existence of machining-induced cracks oriented normal to the surface has been established in several studies (Marshall et al., 1983; Jahanmir et al, 1998; Quinn et al., 2005). The most definitive evidence comes from fractographic observations (Quinn et al., 2005). In dentistry, the failure rate for ceramic restoration is conspicuously high. The deformation and fracture generated by individual diamond grit contacts during dental resurfacing are expected to resemble the damage associated with sharp diamond indentation and scratching. Based on the nature of resurfacing damage in brittle materials, FEA modelling was constructed for dental resurfacing of a feldspar porcelain with a coarse diamond bur. The FEA results show that the trends for the maximum principal stress distributions under different operational conditions were similar. The maximum principal stresses were distributed near the bur–prosthesis contact zone with a rapid decrease with the distance to the resurfaced surface. The maximum principal stress values were found always at the bur/grit exits in spite of the differences in magnitude. This finding is in agreement with the results of FEA modelling of diamond wheel plunge grinding of silicon nitride (Chuang et al., 2003). A previous study observed extensive edge chipping damage occurring in resurfacing of the same porcelain using coarse burs (Yin et al., 2003). This edge damage may be attributed to stress concentration at bur/grit exits. The relations between the maximum principal stresses and the dental operational parameters indicate that the maximum principal stresses were significantly dependent on both the feed rate and the depth of cut (ANOVA, po0.01). As shown in Fig. 4, maximum principal stresses increased with feed rate. This is because that at a fixed depth of cut, an increase in feed rate does not result in a change in bur–prosthesis contact arc length but results in an increased diamond grit depth of cut. An increase in diamond grit depth of cut could result in an increase in resurfacing force in the contact zone. Thus, the increasing force leads to an increase in the maximum principal stress. Moreover, Fig. 4 also shows that at any feed rate, smaller maximum principal stresses were obtained at the larger depth of cut. It is likely that this can be attributed to the fact that at a fixed feed rate, increasing the depth of cut results both in increases in diamond grit depth of cut and bur–prosthesis contact arc length. Subsurface damage depths also exhibited a significant dependence on the dental operational parameters (ANOVA, po0.01). As shown in Fig. 5, subsurface damage degrees follow upward trends with either feed rate or depth of cut. This indicates that selections of suitable depths of cut and feed rates are both important because severe subsurface damage can result from high depths of cut or feed rate. This finding is also consistent with studies on machining-induced damage in diamond grinding of engineering ceramics (Yui et al., 1989; Maksoud et al., 1999). In addition, both predicted and measured subsurface damage depths increased approximately linearly with the diamond grit depth of cut. This relationship is consistent with the published results in FEA modelling of single grit scratching of ceramics (Subhash and Zhang, 2003). A previous study has reported that the average depths of subsurface damage induced in dental CAD/CAM machining of the same porcelain was 40–60 mm (Sindel et al., 1998). Our FEA prediction of subsurface damage depths are in the
359
range 20–80 mm. Such subsurface damage could not only reduce the accuracy of fit of prostheses, but also lead to reduction in mechanical strength and lifetime of dental prostheses (Shearer et al., 1993; Sindel et al., 1998). For high-quality dental restorations, dental practitioners must select the dental operational parameters to diminish subsurface damage while maintaining reasonable removal rates. The results of the FEA modelling are used, in combination with fractographic observations, to provide a qualitative description of the response of dental resurfacing cracks. Furthermore, the implication of the results for non-destructive evaluation of subsurface damage by FEA modelling will be practically meaningful in clinical dental restorations. Finally it is noted that the current FEA modelling was restricted to static model neglecting dynamic and thermal effects on dental resurfacing. However, this model nonetheless describes the salient features of the dental resurfacing process in its simulated form very well. On the basis of studies on brittle material removal, all the models were established in static regime (Evans and Marshall, 1981; Kirchner, 1984; Bifano et al., 1991; Malkin and Hwang, 1996; Jahanmir et al., 1998; Quinn et al., 2005) although the dynamic nature of removal is widely recognized. Compared to metal removal, whose first static chip formation model was established in early 1940s (Merchant, 1945), studies on brittle material removal can only be traced from late 1970s. In metal removal, until 1997 a modified model for metal chip formation considering a mechanism for thermomechanical feedback was developed (Burns and Davies, 1997). In brittle material removal, establishment of a complete model of chip formation with all dynamic, thermal and wear feedbacks will be a difficult target for the years to come.
6. Conclusions FEA modelling and SEM measurement were conduced to quantitatively investigate the dental resurfacing-induced subsurface damage in a feldspar porcelain. The results reveal that maximum principal stresses induced by bur–prosthesis interactions were almost all concentrated at bur/grit exits. The maximum values of these stresses increased with the feed rate and decreased with the depth of cut. Increasing either the feed rate or the depth of cut resulted in an increase in subsurface damage depth. The results suggest that both the feed rate and the depth of cut are important parameters in controlling the degrees of subsurface damage in clinical dental resurfacing operations (ANOVA, po0.01). These results provide insights into the prediction of the quality of ceramic prostheses and guidance as to the proper selection of resurfacing operational parameters in dental practice.
Conflict of interest statement The authors have not entered into a commercial relationship with any other individuals or organizations, that might include employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/registrations, and grants or other funding, other than funding by the National Natural Science Foundation of China Project Grant no. 50475115.
Acknowledgements This work was supported by the NSFC Project Grant no. 50475115. We thank Dr. Anthony Flynn of the Australian National University for valuable comments.
ARTICLE IN PRESS 360
X.-F. Song, L. Yin / Journal of Biomechanics 42 (2009) 355–360
References Bifano, T.G., Dow, T.A., Scattergood, R.O., 1991. Ductile-regime grinding: a new technology for machining brittle materials. Transactions of ASME—Journal of Engineering for Industry 113, 184–189. Burns, T.J., Davies, M.A., 1997. Nonlinear dynamics model for chip segmentation in machining. Physical Review Letters 79, 447–450. Chuang, T-J., Jahanmir, S., Tang, H.C., 2003. Finite element simulation of straight plunge grinding for advanced ceramics. Journal of the European Ceramic Society 23, 1723–1733. Deng, Y., Lawn, B.R., Lloyd, I.K., 2002. Characterization of damage modes in dental ceramic bilayer structures. Journal of Biomedical Materials Research Part B: Applied Biomaterials 63, 137–145. Evans, A.G., Marshall, D.B., 1981. Wear mechanisms in ceramics. In: Rigney, D.A. (Ed.), Fundamentals of Friction and Wear of Materials. American Society for Metals, Metals Park, OH, USA. Groover, M.P., 2007. Fundamentals of Modern Manufacturing—Materials, Processes and Systems, third ed. Wiley, New Jersey, USA. Harvey, C.K., Kelly, J.R., 1996. Contact damage as a failure mode during in vitro testing. Journal of Prosthodontics 5, 95–100. Ironside, J.G., Swain, M.V., 1998. Ceramics in dental restorations—a review and critical issues. Journal of the Australasian Ceramic Society 34, 78–91. Jahanmir, S., Xu, H.K., Ives, L.K., 1998. Mechanisms of material removal in abrasive machining of ceramics. In: Jahanmir, S., Ramulu, M., Koshy, P. (Eds.), Machining of Ceramics and Composites. Marcel Dekker, New York, USA. Kelly, J.R., 1999. Clinically relevant approach to failure testing of all-ceramic restorations. Journal of Prosthetics Dentistry 81 (6), 652–661. Kim, H-W., Deng, Y., Miranda, P., Pajares, A., Kim, D.K., Kim, H-E., Lawn, B.R., 2001. Effect of flaw state on the strength of brittle coatings on soft substrates. Journal of the American Ceramic Society 84, 2377–2384. Kirchner, H.P., 1984. Damage penetration at elongated machining grooves in hotpressed Si3N4. Journal of the American Ceramic Society 67, 127–132. Lawn, B.R., Deng, Y., Thompson, V.P., 2001. Use of contact testing in the characterization and design of all-ceramic crown-like layer structures: a review. Journal of Prosthetics Dentistry 86, 495–510. Maksoud, T.M.A., Mokbel, A.A., Morgan, J.E., 1999. Evaluation of surface and sub-surface cracks of ground ceramic. Journal of Material Processing Technology 88, 222–243. Malkin, S., 1989. Grinding Technology: Theory and Application of Machining with Abrasives. Willey, New York, USA. Malkin, S., Hwang, T.W., 1996. Grinding mechanisms for ceramics. Annals of CIRP 45, 569–580.
Marshall, D.B., Evans, A.G., Khuri Yakub, B.T., Tien, J.W., Kina, G.S., 1983. The nature of machining damage in brittle materials. Proceedings of the Royal Society of London, Series A, Mathematical and Physical Sciences 385, 461–475. Merchant, M.E., 1945. Mechanics of the metal cutting process: II. Plasticity conditions in orthogonal cutting. Journal of Applied Physics 16, 318–324. Quinn, G.D., Ives, L.K., Jahanmir, S., 2005. On the nature of machining cracks in ground ceramics: part I: SRBSN strengths and fractographic analysis. Machining Science and Technology 9, 169–210. Rekow, D., Thompson, V.P., 2007. Engineering long term clinical success of advanced ceramic prostheses. Journal of Materials Science: Materials in Medicine 18, 47–56. Rosenblum, M., Schulman, A., 1997. A review of all-ceramic restorations. Journal of American Dental Association 128, 297–307. Shearer, A.C., Heymann, H.O., Wilson, N.H., 1993. Two ceramic materials compared for the production of CEREC inlays. Journal of Dentistry 21 (5), 302–304. Siegel, S.C., von Fraunhofer, J.A., 1999. Dental cutting with diamond burs: heavyhanded or hight-tough? Journal of Prosthodontics 8, 3–9. Sindel, J., Petschelt, A., Grellner, F., Dierken, C., Greil, P., 1998. Evaluation of subsurface damage in CAD/CAM machined dental ceramics. Journal of Materials Science: Materials in Medicine 9, 291–295. Song, X.F., Yin, L., Han, Y.G., Li, J., 2008. Finite element analysis of subsurface damage of ceramic prostheses in simulated intraoral dental resurfacing. Journal of Biomedical Materials Research Part B: Applied Biomaterials 85B, 50–59. Subhash, G., Zhang, W., 2003. Finite element analysis of brittle cracking due to single grit rotating scratch. Journal of Applied Mechanics 70, 147–151. Timoshenko, S.P., Goodier, J.N., 1970. Theory of Elasticity. McGraw-Hill, New York. Yin, L., Jahanmir, S., Ives, L.K., 2003. Abrasive machining of porcelain and zirconia with a dental handpiece. Wear 255, 975–989. Yin, L., Song, X.F., Qu, S.F., Han, Y.G., Wang, H., 2006a. Surface integrity and removal mechanism in simulated dental finishing of a feldspathic porcelain. Journal of Biomedical Materials Research Part B: Applied Biomaterials 79B, 365–378. Yin, L., Song, X.F., Qu, S.F., Huang, T., Mei, J.P., Yang, Z.Y., Li, J., 2006b. Performance evaluation of a dental handpiece in simulation of clinical finishing using a novel 2-DOF in vitro apparatus. Proceeding of Institution of Mechanical Engineers Part H: Journal of Engineering in Medicine 220, 929–938. Yui, A., Watanabe, T., Yoshida, Y., 1989. Effect of grinding conditions upon bending strength of alumina ceramics. Grinding Fundamentals and Applications. American Society of Mechanical Engineering PED 39, 229–239. Zhang, B., Peng, X.H., 2000. Grinding damage prediction for ceramics via CDM model. Journal of Manufacturing Science and Engineering 122, 51–58.