Investigation on diamond turning of silicon crystal - generation mechanism of surface cut with worn tool

Investigation on Diamond Turning of Silicon Crystal - Generation Mechanism of Surface Cut with Worn Tool

Renato G. Jasinevicius

Jaime G. Duduch

Arthur J. V. Porto

Universidade de São Paulo-EESC

Department of Mechanical Engineering

Caixa Postal 359

13560-970. São Carlos, S.P. Brazil

[e-mail]

This paper discusses the effect of tool wear on surface finish in single-point diamond turning of single crystal silicon. The morphology and topography of the machined surface clearly show the type of cutting edge wear reproduced onto the cutting grooves. Scanning electron microscopy is used in order to correlate the cutting edge damage and microtopography features observed through atomic force microscopy. The possible wear mechanisms affecting tool performance and surface generation during cutting are also discussed. The zero degree rake angle single point diamond tool presented small nicks on the cutting edge. The negative rake angle tools presented more a type of crater wear on the rake face. No wear was detected on flank face of the diamond tools.

Keywords: Wear, diamond machining, silicon crystal, topography, scanning electron microscopy, atomic force microscopy

Introduction

Brittle materials are used for their superior properties and ability to take a polish. In the machining of normally brittle crystals, when extremely sharp tools are used, plastic behavior may take place prior to the onset of fracture when the cut is restricted below critical cutting depth value (Puttick and Franks, 1990, Duduch et al. 1996). In single point diamond turning, the surface roughness is extremely low as a result of the high fidelity reproduction of the cutting edge profile. Because of this, the cutting edge sharpness of diamond tools used in the machining of brittle materials is of extreme relevance in achieving optical quality surface finish and low subsurface damage.

The feasibility of using this process as a finishing operation for optical components from semiconductor crystals has increased significantly. The anomalous plasticity observed in diamond turning of silicon at room temperature may be attributed to a phase transformation induced by pressure/stress generated by the tool edge/tip contact (Morris et al. 1995). Evidences reported on high pressure (Minomura and Drickamer, 1962, Jameison, 1963) and indentation (Gridneva et al. 1972, Clarke et al. 1988) experiments have demonstrated that diamond-cubic silicon transforms to the denser metallic b-tin structure under hydrostatic pressure at room temperature and; upon release of the pressure, a reversion to another form, i.e., an amorphous semiconductor phase, with a cubic structure has been reported (Hu et al.1986). The critical pressure observed to cause a transformation to a metallic b-tin structure is in the range of 11.3-12.5 GPa, under pure hydrostatic conditions (Clarke et al. 1988). However, this is reduced to values as low as 8 GPa when shear stresses are present (Gilman, 1992). Material removal will then be ductile, resultant from the metallic thin layer which is plastically removed.

However, silicon crystal may impose some constrain in terms of changing cutting edge geometry. This can be translated into rapid wear and cutting edge damage. The tool edge bluntness extent can be considered a relevant aspect of influence in respect of the critical cutting dimensions in this machining process. It is worth mentioning that for the brittle-to-ductile transition, the critical thickness of cut and/or critical cutting depth, have been found to be in the range of 50-250 nm for Si (Blake and Scattergood, 1990, Blackley and Scattergood, 1991) using commercially available diamond tools with edge radii larger than 100 nm. Furthermore, a less sharp tool implies in higher contact pressure and higher cutting forces and stresses in the tool’s active zone. When a blunt or worn tool is used, the cutting depth required will be higher in order to generate sufficient amount of stress to drive the phase transformation. Moreover, when larger depth of cut are used, worn tools will probably tend to cause brittle failure before metallization occurance. The control and/or attenuation of wear in diamond tools is a fundamental factor to improve the performance of micromachining in terms of the applicability and economy of the process.

This paper presents an investigation on the surface generation mechanism involved in single-point diamond turning of single crystal silicon with a worn tool. Possible wear mechanisms affecting tool performance during cutting will be discussed. Scanning electron microscopy (SEM) analysis of tool wear features are presented. The wear pattern of single point diamond tools used to machine single crystal silicon and the effect of this wear on the surface generated are presented. Atomic Force Microscopy (AFM) will be used to characterize the morphology and topography of the machined the machined formed with a damaged tool.

Experimentals

Test Procedure, Experimental Apparatus and Geometry of the Diamond Turning Process

Single-point diamond turning tests were carried out on a commercially available ultraprecision diamond turning machine, the Aspheric Surface Generator Rank Pneumo ASG 2500. Facing cuts were performed on monocrystalline silicon samples (10 x 10 mm2). Single crystal p-type Si was used in this study. Silicon having (100) surface orientation was used. Table I gives the properties of single crystal silicon.

In this work, round nose single crystal diamond tools (Contour Fine Tooling®) were used. Only new diamond tools were used in the cutting tests. The schematic diagram of the profile of the tool/material interaction where brittle-to-ductile transition occurs is depicted in Fig. 1. The chip thickness (t) varies from zero at the tool center up to a maximum t_max at the top of the uncut shoulder. There is a gradual decrease in the severity of the damage along the shoulder. The transition from ductile mode to brittle mode is represented by the parameter y_c , which is the critical depth, and by the critical chip thickness t_c. The position of the transition moves downward along the tool nose as the feed rate increases. This cutting model was first proposed by Blake and Scattergood (1990). The severity of microfracture damage will be higher close to the top of the uncut shoulder. Figure 2 (a) and (b) are, respectively, the photomicrograph made by optical microscopy of the uncut shoulder and a three dimensional schematic representation of the tool/material interaction shown in Fig. 1. It is worth mentioning that large cracks may propagate from near the original surface at the shoulder down into the cut surface, however the formation of thicker chips has not prove to have any different effect on the surface finish. Table II describes the experimental conditions and tool geometry used in the cutting tests.

Sample Preparation for Observation of Surface Features and Cutting Edge

A scanning electron microscope Zeiss, model DSM 960, operated with an electron beam of 20 kV was used for the observations of the tool cutting edge and machined grooves. Metallization coating provided enhanced visualization of details in the machined surfaces at high magnifications. Figure 3 is a scanning electron micrograph of the cutting edge (showing no damage). In this case a metallization enhanced the visualization of the cutting edge, the thickness of the gold film deposited on the tool is 20 nm.

The specimens were cleaved in order to use a surface with no metallization coating for the AFM. The atomic force microscope (AFM) was a Topometrix. It was operated with a standard 50-60^o conical silicon nitride stylus of 15-20 nm radius tip, with cantilever spring constant of ~ 0.06 N/m. Conventional contact mode was employed where the stylus is scanned in contact with the surface with contact forces of typically 10-100 nN.

Results

Figure 4 show images made by SEM and AFM of the worn cutting edge and the surface generated by a worn negative rake angle tool (-25^o), respectively. Crater wear pattern in the rake surface were present, as shown in Fig. 4 (a). In order to make a more precise inspection and produce a better resolution of the reproduction of the cutting edge imperfections, AFM technique was used. The 3D image of this machined surface is shown in Fig. 4 (b). The most significant change in surface appearance due to the wear of the cutting edge was a gradual flattening of the ridges. The feed rate used in this cut was 2.5 mm/revolution and the nominal depth of cut of 5 mm.

The cross-section analysis (Fig. 5) measured the roughness value of Ra = 16.8 nm which is slightly larger than the roughness value range encountered for the as-received sample. The reproduction of the cutting edge imperfection is easily seen in the cross-section profile shown in Fig. 5.

Figure 6 show images made by SEM and AFM of the worn cutting edge of the 0^o rake angle and surface generated, respectively. A careful inspection in Fig. 6 (a) shows that small changes occur in the cutting edge geometry and they will in fact represent modification in the machining grooves formed. In this tool only small nicks on the cutting edge were observed. Fig.ure 6 (b) is a 3D image made by AFM illustrating the roughness within the grooves created by cutting edge imperfections corresponding to the surface machined grooves produced by the worn 0^o rake tool The tool edge defects are replicated rather faithfully from groove to groove. These ridges lead to a diffraction effect from the otherwise mirror-like surface. Despite the wear of the tool cutting edge, the roughness value of the turned mirror is Ra = 6.1 nm. The material removal occurred mostly in the ductile mode. The cutting condition were feedrate of 1.5mm/revolution and nominal depth of cut of 5 mm. Although the cutting and rake face have shown some deleterious wear, nothing could be detected in the clearance face of both the 0 ^o and -25 ^o rake tools.

During machining with worn tool an interesting damage pattern was observed. Certain areas of the sample had a ductile response and others brittle responses as shown in Fig. 7. Figure 7 (a) illustrates the cracks and damage pattern formed. Figure 7 (b) show detailed view of both regimes, i.e., brittle and ductile. The top image revealed cracks running perpendicular to the cutting direction. The bottom image shows the surface finish in the ductile mode. The cutting condition were feedrate of 1.25 mm/revolution and nominal depth of cut of 0.1 mm using negative rake angle tool.

Discussion

Negative rake tool a suffer a more severe deterioration on the cutting edge. Nevertheless, it is worth mentioning that this tool geometry presents superior performance in terms of material removal in the ductile regime. A possible explanation to the drastic damage caused to the cutting edge is that negative rake angle tools generate more compressive stress in the cutting zone. This stress is maximum in the vicinity of the cutting edge. In this case, friction forces tend to increase which possibly contributes to the wear involving an abrasion mechanism.

It is well established that silicon and germanium can be diamond machined, however silicon presents a more aggressive edge wear (Blake and Scattergood, 1990) when compared with germanium. The estimate of wear rates for the pair diamond/silicon is of great relevance for the machining of this material under feasible production rates. Specific wear rates for diamond tips and silicon were determined in sliding tests (Khurshudov et al. 1997) performed in Atomic Force Microscope. The estimated values of the wear rate of diamond in contact with silicon were 1.4 x 10 ^-9 mm³ N^-1 m^-1 and 4.5 x 10^-4 mm³ N^-1 m^-1, respectively.

The relationship between the material hardness and wear can be considered a factor of particular relevance to the analysis of the performance of a single point diamond tool and the development of wear(Hurt and Decker, 1984). The reported wear rate for diamond/silicon interaction is comparable with the rate reported for the wear of diamond in contact with hard steel (Thornton and Wilks, 1978)(2.0 x 10^-9 mm³ N^-1 m^--1). The high wear rate with hard steel is due to the large chemical affinity of the iron with the carbon of the diamond. Ferrous metals in contact with diamond tools can rapidly wear the cutting edge. Diamond shows a considerable weight loss by oxidation or graphitization when exposed to high temperatures (Davies and Evans, 1972). These particular observations give support to the thermomechanical process which is an important mechanism of wear of a diamond cutting edge. According to Hurt & Decker (1984), temperatures during diamond machining can easily reach 427 ^oC (or 700 ^oK). This only causes a moderate reduction in diamond strength (Brookes, 1979). During single point diamond turning of non-ferrous metals the temperature on the cutting edge and rake face is not sufficiently high to thermally reduce diamond strength. Furthermore, the maximum temperature found on the rake face of a diamond tool was approximately 190 ° C for aluminum and 220 ° C for copper at cutting speed of 620 m/min (Ueda et al. 1998). A successfully attempt has been made by Abdel-Aal et al. (1997) to calculate the temperature development during the simulated machining of silicon and germanium wafers. The results indicated that the workpiece surface temperature rise is well bellow the thermal softening temperatures for these materials under the conditions applied in the tests. Particularly the temperature rise was of about 80 ^oC for Ge and less than 400 ^oC for Si. These can be considered of moderate magnitude and are proportional to the loads and speeds used in the tests (idem).

The exact mechanism responsible for the wear of single point diamond tools used in the micromachining of silicon has not yet been completely explained. Paul and collaborators (1996) have not found any strong evidence of chemical wear caused by silicon/diamond tool interaction. According to them, because silicon contains no d electrons, no chemical wear is predicted by electron complex mechanism. Hence, processes that result in scratches or gouges are signs of non chemical wear (idem). It is believed that hard particle may be formed during micromachining process. Consequently, this is likely to promote the polishing of the cutting edge. Recently, it was reported(Puttick et al. 1995)the detection of thin superficial film of oxide residues on the virgin surface of as-received polished silicon wafers and nanoturned surfaces (0.5 nm and 1.5 nm thick of SiO₂, respectively). In the case of the nanoturned specimen a significant amount of carbon was also found, possibly transferred from the diamond tool. Molecular dynamics simulation techniques, applied to the understanding of the mechanisms involved in the nanomachining process of silicon, have detected the presence of SiC particles as well as the graphitization of the diamond tool cutting edge (Boercker et al. 1992). On the other hand, no evidence of the formation of this abrasive material (SiC) was detected through XPS technique on single point diamond turned monocrystalline silicon surfaces (Kunz et al. 1996). The presence of small particles formed by silicon debris found in the tool edge/material interface can aid the promotion of abrasive wear of diamond.

The microcracking pattern observed in Fig. 7 had definite orientation dependence and corroborates well with the results reported on the occurrance of surface damage in single point diamond turning of Ge [Blackley and Scattergood, 1990].This damage may be related to crystallographic dependence since it occurred in preferential directions [110] as schematically illustrated in Fig. 8. It is known that single crystals presents certain planes which have lower cleavage stresses than others; the preferential damage was correlated to the maximum amplitude of the resolved tensile stress on the (111) cleavage planes as shown in Fig 9. According to Morris and Callahan (1994), for a blunt tool to produce higher stresses (on contact - compared with a sharp tool) and a considerable increase in depth of cut is needed. At the pressures required to promote phase transformation, the depth of cut may be very close to or exceed the Griffith flaw size (idem). Consequently, in subsequent cuts at lower depths of cut cracking would propagate to relieve the stress concurrent with transformation with the decrease in depth of cut.

Conclusions

The mechanisms involved in the wear of diamond tools were discussed. It was shown that worn tools can remove the material in the ductile regime. Worn tools generate high stress levels with the increase of penetration depth. If the stresses are high enough, the onset of the phase transformation takes place prior to cracking. It was shown that the ductile mode material removal is governed by a phase transformation induced by hydrostatic pressure and shear stresses generated by the tool tip/edge interaction with workpiece material along with dislocation movement and activation of slip systems. Chemical wear in the interaction of diamond/silicon was discarded based on the findings of Paul and collaborators³⁶. The rapid and deleterious wear of the diamond tool when cutting silicon is considered a constrain factor in the production of large surfaces. The mechanisms of tool wear during machining of silicon and how this wear can be controlled is not completely understood. The reduction of wear and consequently, the improvement of surface finish characteristics of silicon optical components is open to receive more contributions. In industry, ductile grinding is becoming more widely used (as well as chemo-mechanical polishing) for large surfaces, due to this rapid and deleterious diamond tool wear.

Acknowledgments

The first author would like to acknowledge that his work is financially supported by brazilian agency FAPESP (Proc. # 98/11300-3) (Brazil). The authors also would like to acknowledge the financial support by CNPq FAPESP. The authors thank Dr. P.S.P. Herrmmann for providing the AFM analyses at CNPDIA, EMBRAPA – São Carlos - SP – Brazil.

Manuscript received: September 2000, Technical Editor: Alisson Rocha Machado.

• Abdel-Aal, H.A, Smith, S.T. and Patten, J.A. 1997, "On the Development of Surface Temperatures in Precision Single Point Diamond Abrasion of Semiconductors", Int. Comm. Heat Mass Transfer, Vol. 24, pp.1131-1140.
• Blake, P.N. and Scattergood, R.O., 1990, "Ductile-Regime Machining of Ge and Si", J. Am. Ceram. Soc., Vol.73, No. 4, pp.949-957.
• Blackley, W.S., Scattergood, R.O., 1990, "Crystal Orientation Dependence of Machining Damage - Stress Model", J. Am. Ceram. Soc., Vol.73, No. 10, pp.3113-3115.
• Blackley, W.S. and Scattergood, R.O., 1991, "Ductile-Regime Machining Model for Diamond Turning of Brittle Materials", Precision Engg., Vol. 13, No. 2, pp. 95-103.
• Boercker, D.B., Belak, J., Stowers, I.F., Donaldson, R.R., Siekhaus, W.J., 1992, "Simulation of Diamond Turning of Silicon Surfaces", Seventh ASPE Annual Meeting, Orlando, Florida , USA, pp.45-48.
• Brookes, C.A , 1979 in "Properties of Diamond" ed. J.E. Field, London, New York, Academic Press 674 p..
• Clarke, D.R., Kroll, M.C., Kirchner, Cook, R.F. and Hockey, 1988, "Amorphization and Conductivity of Si and Ge During Indentation", Phys. Rev. Lett., Vol. 60, No. 21, pp.2156-2159.
• Crompton, D., Hirst, W., Howse, M.G.W., 1973 "The Wear of Diamond". Proc. Royal Soc. (London). Vol. A.333, pp.435-454.
• Davies, G. and Evans, T., 1972, "Graphitization of Diamond at Zero Pressure and at High Pressure", Proc. R. Soc. (London), Vol. A. 328, pp.413-427.
• Duduch, J.G., Gee, A E., Porto, A J. V., Jasincevicius, R.G., Carpinetti, L.C.R., 1996 "Model of Brittle Materials Single Point Machining with Removal Rates". Revista Brasileira de Ciências Mecânicas (Journal of the Brazilian Society of Mechanical Sciences), Vol. 18, No. 1, pp.33-39.
• Gerk, A. P., Tabor, D., 1978 "Indentation Hardness and Semiconductor-Metal Transition of Germanium and Silicon", NATURE, Vol.271, p.732-733.
• Gilman, J.J., 1992, "Insulator-Metal Transitions at Microindentations", J. Mater. Res. Vol. 7, No. 3, pp.535-538.
• Gridneva, I.V., Milman, Y.V., Trefilov, M., 1972 "Phase Transformation in Diamond-Structure Crystals During Hardness Measurements", Phys. Status Solidi A, Vol. 14, pp. 177-182.
• Hiatt, G.D. and Strenkowski, J.S., 1991, "Fracture Mechanics Applied to the Ductile Machining of Brittle Materials", Proc. 7th ASPE Annual Meeting, Greenleaf, FLA, USA, pp.29-32.
• Hu, J.Z., Markle, L.D., Menoni, C.S., Spain, I. L., 1986, "Crystal Data for High-Pressure Phases of Silicon", Phys. Rev. B., Vol. 34, No. 7, pp. 4679-4684.
• Hurt,H.H., Decker, D.L., 1984 "Tribological Considerations of the Diamond Single Point Tool", Proc. of the SPIE, Production aspects of single point machined optics, Vol. 508, pp.126-131.
• Jameison, J.C., 1963, "Crystal Structures at High Pressures of Metallic Modifications of Si and Ge", Science, Vol.139, pp.762-764.
• Jasinevicius, R.G., Duduch, J.G., Porto, A J.V. 1998 "Two Fracture Modes Observed Through SEM in Brittle-to-Ductile Regime Diamond Turning Of Monocrystalline Silicon", ACTA MICROSCOPICA, Vol. 7 (Suppl. A), pp. 417-420.
• Kunz, R.R., Clark, H.R., Nitishin, P.M., Rothschild, M., Ahern, B.S., 1996 "High Resolution Studies of Crystalline Damage Induced by Lapping and Single-Point Diamond Machining of Si(100)", J. of Mater. Res., Vol. 11, No. 5, pp.1228-1237.
• Khurshudov, A.G., Kato, K., Koide, H., 1997, "Wear of the AFM Diamond Tip Sliding Against Silicon", WEAR, Vol.203-204, pp.22-27.
• Minomura, S., Drickamer, H.G., 1962 "Pressure-Induced Phase Transformations in Si, Ge, and Some III-V Compounds", J. Phys. Chem. Solids, Vol. 23, pp.451-456.
• Morris, J.C., Callaham, D.L., 1994 "Origins of the Ductile Regime in Low Load Scratching in Si", J. Mater. Res.,Vol.9, No. 11, pp.2907-2913.
• Morris, J.C., Callaham, D.L.,Kulik, J., Patten, J. A. and Scattergood, R.O., 1995, "Origins of the Ductile Regime in Single Point Diamond Turning of Semiconductors", J. Am. Ceram. Soc., Vol. 78, No. 8, pp.2015-2020.
• Nakasuji, T., Kodera, S., Hara, S., Matsunaga, H., Ikawa, N. and Shimada, S., 1990, "Diamond Turning of Brittle Materials for Optical Components", Annals of the CIRP, Vol.39, No. 1, pp.89-92.
• Paul, E., Evans, C.J., Manganmelli, A, McGlauflin, M. L., Polvani, R. S., 1996 "Chemical Aspects of Tool Wear in Single Point Diamond Turning", Precision Engg., Vol. 18, pp.4-19.
• Pizani, P. S., Jasinevicius, R.G., Duduch, J.G. and Porto, A J.V. 1999, "Ductile and Brittle Damage in Single Point Diamond Turned Silicon Probed by Raman Scattering", Journal of Materials Science Letters, Vol. 18, pp. 1185-1187.
• Puttick, K.E., Franks, A., 1990, "The Physics of Ductile-Brittle Machining Transitions: Single Point Theory and Experiment", Jap. Soc. Prec. Engg., Vol. 56 , No. 5, pp.12-16.
• Puttick, K.E., Whitmore, L. C., Gee, A.E., Chao, C.L., 1995 "Energy Scaling Transitions in Machining of Silicon by Diamond". Trybology International, Vol. 28, No. 6, pp.349-355.
• Suzuki, T., Ohmura, T, 1996, "Ultramicroindentation of Silicon at Elevated Temperatures", Phil. Mag. A, Vol. 74, No. 5, pp.1073-1084.
• Thornton, A.G., Wilks, J., 1978, "Clean Surface Reactions Between Diamond and Steel". Nature, Vol.274, No. 24, pp.792-793.
• Ueda, K., Sugita, T, Hiraga, H., 1991 "A J-Integral Approch to Material Removal Mechanism in Microcutting of Ceramics", Annals of the CIRP, Vol. 40, No. 1, pp.61-64.
• Wortman, J.J. and Evans, R.A, 1965, "Young’s Modulus, Shear Modulus, and Poissons’s Ratio in Silicon and Germanium", J. Appl. Phys., Vol. 36, No. 1, pp.153-156.
• Ueda, T., Sato, M. and Nakayama, K., 1998, "The Temperature of a Single Crystal Diamond Tool in Turning" Annals of the CIRP, Vol. 47 , No. 1, pp.41-44.

Publication Dates

• Publication in this collection
  24 Oct 2001
• Date of issue
  2001

History

• Received
  Sept 2000