Potential of Electrically Conductive Chemical Vapor Deposited Diamond as an Electrode for Micro-Electrical Discharge Machining in Oil and Water

Transcription

1 A. New Sharma Diamond et al. and Frontier Carbon Technology 181 Vol. 15, No MYU Tokyo NDFCT 487 Potential of Electrically Conductive Chemical Vapor Deposited Diamond as an Electrode for Micro-Electrical Discharge Machining in Oil and Water Anurag Sharma *, Manabu Iwai 1, Kiyoshi Suzuki and Tetsutaro Uematsu 1 Nippon Institute of Technology, Miyashiro, Saitama , Japan 1 Toyama Prefectural University, Kosugi, Toyama , Japan (Received 24 July 2004; accepted 16 November 2004) Key words: micro-electrical discharge machining, electrically conductive diamond, chemical vapor deposited diamond thick film, low wear electrode Low electrode wear is the key to achieving high finish and high accuracy machining in micro-electrical discharge machining (micro-edm). Electrically conductive chemical vapor deposited (CVD) diamond has shown almost zero electrode wear, even at short pulse durations, which is extremely important for maintaining the electrode shape throughout the EDM process. This research involves an investigation into the effect of various EDM parameters on the EDM of die steel using electrically conductive CVD diamond with respect to electrode wear, efficiency and ED machined surface properties. The effect of Joule heating due to the high electrical resistivity of electrically conductive CVD diamond, which may cause deterioration of the ED machined surface is also studied. The results indicate that no deterioration of the workpiece surface occurs due to Joule heating when EDM is performed using an electrically conductive CVD diamond electrode. An investigation into the factors responsible for the low wear of the CVD diamond electrode leads us to deduce that an accelerated adhesion of the heat-resolved carbon on the electrode surface along with the high thermal stability of the diamond are mainly responsible. 1. Introduction Electrical discharge machining (EDM) is being increasingly used in the micro-machining of various types of steel as well as other difficult to machine materials. In the case of micro-machining using EDM, it is extremely important not only to impart a good surface * Corresponding author: anuragsharma@hotmail.com 181

2 182 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) finish to the workpiece, but also to reduce electrode wear to increase the shape accuracy of the machined part. In the existing state of the EDM process, it is very difficult to achieve a good surface finish and a high shape accuracy due to the high wear rate of the currently used electrode materials under finishing EDM conditions. Although EDM with very low electrode wear can be achieved as a result of the adhesion of heat-resolved carbon released from the dielectric fluid, which protects the electrode from thermal wear, this phenomenon takes place only at sufficiently long pulse durations in the case of existing conventional electrode materials. (1,2) A surface machined using EDM usually has a matte texture resulting from discharge craters. The size of these craters is directly proportional to the EDM current and the pulse duration. To obtain a good EDM surface finish it is necessary to reduce the depth and size of these craters. Thus lowering the EDM current and making the pulse duration extremely short can produce a good finish. However, at very short pulse durations, the wear rate of the currently used electrodes increases rapidly, and as a result the desired shape accuracy of the workpiece cannot be achieved. Suzuki et al. have found that an electrically conductive CVD diamond thick film (hereafter called EC-CVD diamond) electrode can be used to perform EDM even at a short pulse duration of 3 μs without any significant electrode wear. (3 8) This research deals with further investigations on the application of EC-CVD diamond to carry out EDM on die steel and to understand the factors responsible for the low wear of the diamond electrode. The effect of various EDM conditions on the performance of the EC-CVD diamond electrode is examined. EDM using a diamond electrode is also carried out in de-ionized water. The results of the experiments support the observation of almost zero wear on the EC-CVD diamond electrode when EDM is carried out in oil even for a very short pulse duration of 3 μs. 2. Thermal Behavior of EC-CVD Diamond Table 1 lists the physical properties of EC-CVD diamond along with copper, graphite and Cu-W, which are typical conventional electrode materials. The EC-CVD diamond, which is made electrically conductive by doping with boron during the CVD process, possesses sufficient electrical resistance (ρ=0.4~ Ω. m) to be used in EDM. It also has the highest thermal diffusivity compared to currently used electrode materials, which may Table 1 Properties of EC-CVD diamond thick film. Material EC-CVD diamond * Copper Graphite Cu-W Specific resistance (Ω. m) 0.4~ Specific gravity Young s modulus (N/m 2 ) 1000~ Thermal conductivity (W/m. K) 500~ ~295 Coefficient of thermal expansion Thermal diffusivity (m 2 /s) 0.27~ * CVDITE-CDE from Element Six Co., Ltd

3 A. Sharma et al. 183 be an added advantage in a faster dissipation of the heat generated during machining. The use of boron as the doping agent has the additional effect of increasing the thermal stability of the EC-CVD diamond compared to the non-conductive CVD diamond. Oxidation tests for EC-CVD diamond (CVDITE-CDE, Element Six Ltd.) and electrically nonconductive CVD diamond (CVDITE-CDM, Element Six Ltd.) were performed for comparison. The tests consisted of heating the CVD diamond thick films ( mm) at temperatures of 825, 850, 875 and 900 C in air for 30 min followed by visual inspection using an optical microscope as well as measuring the change in weight. Figure 1 shows the appearance of the diamond surface after oxidation and Fig. 2 shows the relation between the Fig. 1. Comparison of oxidation of EC-CVD diamond (CVDITE-CDE) with normal CVD diamond (CVDITE-CDM). (Diamond segment: mm, weight before oxidation test=38 mg, in air) Fig. 2. Comparison of weight-loss during the oxidation between EC-CVD diamond (CVDITE- CDE) and normal CVD diamond (CVDITE-CDM).

4 184 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) oxidation temperature and the percentage weight loss due to oxidation. There is no change in the appearance of the EC-CVD diamond surfaces after heating for 30 min at 825 and 850 C, whereas, in the case of the electrically non-conductive CVD diamond thick film, the surfaces appear to deteriorate and lose weight by 2~3% and 8~10% at 825 and 850 C, respectively. However, on heating above 850 C, in the case of EC-CVD diamond, the start of oxidation is observed at 875 C with a 3~5% reduction in weight, which increases to approximately 10% at 900 C. Compared to this, in the case of electrically non-conductive CVD diamond thick film, there is a large reduction of 15~25% in weight at 875 C, which increases drastically to 67% at 900 C. This confirms that, by doping with boron during CVD, thermal stability of the EC-CVD diamond is enhanced in addition to achieving electrical conductivity. 3. Experimental Investigation with EC-CVD Diamond Electrodes in Oil Experiments were performed to investigate the EDM properties of EC-CVD diamond electrodes in oil. Various EDM properties were studied and the results are discussed in the following sections. 3.1 Experimental setup and conditions An electrode segment ( mm) was mounted on the head of a die-sinking EDM machine by clamping onto a copper jig as shown in Fig. 3(a). Figure 3(b) shows the electrode segment used for the EDM. The experimental conditions are listed in Table 2. The experimental electrode segments used for the other conventional electrode materials such as copper, graphite and Cu-W were approximately the same size as the EC-CVD diamond segment electrode. Die steel SKD11 (HRC 61) and stainless steel SUS316 were used as the work samples. Kerosene (Vitol 2) was used as the working fluid. Experiments were conducted for both straight (electrode negative) and reversed (electrode positive) polarity EDM conditions to understand the effect of polarity. (a) Setup on EDM machine (b) Electrode used Fig. 3. Experimental setup on EDM machine.

5 A. Sharma et al. 185 Table 2 Experimental devices and conditions. EDM machine Die sinking EDM machine (AQ35L, Sodick) Dielectric Oil (Vitol 2, Sodick) working fluid Deionized water (2000~4000 Ω. m) 1. Electrically conductive CVD diamond film (CVDITE-CDE, mm, Element Six) Electrodes 2. Copper ( mm) 3. Graphite ( mm) 4. Cu-W ( mm) Workpiece Die steel (SKD11, Hardness HRC61), stainless steel SUS 316 EDM conditions u i = 60, 90 V, i p =1, 3, 6 A, t e = 3, 7, 15, 20, 30, 50 μs, t o =10, 20, 50 μs, EDM depth=0.1, 0.5, 1 mm 3.2 Effect of polarity The effect of polarity on the EC-CVD diamond electrode was investigated and compared with the results from the copper electrodes under the same EDM conditions. The experiment was performed under EDM conditions of an open circuit voltage u i =90 V, a peak current i p =3 A, pulse duration/interval of t e /t o =3/10, 50/50 μs and an EDM area=1.5 mm 2. The results of this experiment are shown in Fig. 4. In the case of the EC-CVD diamond electrode used at reversed polarity (electrode positive), no wear is observed even for a short pulse duration of 3 μs, whereas, when used at straight polarity (electrode negative), a large amount of wear on the order of 170 μm is observed, and the wear depth increases to 350 μm for a long pulse duration of 50 μs. On the other hand, the copper electrode, at a short pulse duration of 3 μs, shows a wear depth of 340 μm at reversed polarity (electrode positive), but at straight polarity (electrode negative) the wear is almost equivalent to the EC-CVD diamond electrode, i.e. 150 μm. However, the copper electrode shows zero wear at a long pulse duration of 50 μs at reversed polarity (electrode positive), which is the same as the wear on the EC-CVD diamond electrode. From these results, it can be said that, for the successful use of EC-CVD diamond as an electrode for EDM in oil, the diamond electrode should be used at reversed polarity (diamond electrode positive) irrespective of the pulse duration. 3.3 Comparison of electrode wear and efficiency To compare the electrode wear depth and EDM efficiency of the EC-CVD diamond electrode with copper, graphite and Cu-W electrodes experiments were performed under EDM conditions such as pulse durations of 3, 7, 15 and 30 μs, an open circuit voltage of 90 V, and a peak current of 6 A. The depth of EDM in each case was 0.5 mm, and a pulse interval of 50 μs was employed to eliminate the possibility of arcing. Figure 5 shows the results of the wear depths and efficiencies for the existing electrodes and the newly proposed EC-CVD diamond electrode. It is well known that the electrode wear is reduced for sufficiently long pulse durations when EDM is performed in oil at reversed polarity (electrode positive) for existing electrodes. It was reported that this phenomenon is mainly due to the adhesion of the heat-resolved carbon released as a result of dielectric breakdown onto the electrode surface. (1,2) However, this phenomenon is not

6 186 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) Fig. 4. Effect of electrode polarity on electrode wear. (u i =90 V, i p =3 A, work=skd11, EDMed depth=1 mm, in oil) (a) Wear depth (b) EDM efficiency Fig. 5. Comparison of wear and efficiency among electrode materials in oil. (u i =90 V, i p =6 A, t o =50 μs, work=skd11, EDM depth=0.5 mm, reversed polarity (electrode +)) observed at short pulse durations due to a reduction in the amount of carbon adhesion during short pulse duration as shown in Figs. 4 and 5(a). On the other hand, the electrode wear depth in the case of the EC-CVD diamond electrode is almost zero even for a short pulse duration of 3 μs, compared to the electrode wear depths of approximately 255, 150 and 105 μm for copper, graphite and Cu-W electrodes, respectively. This indicates that conditions of zero wear in EDM may be achieved with the EC-CVD diamond electrode at short pulse duration or under finishing EDM conditions, which may be impossible in the case of currently used electrode materials. This characteristic of the EC-CVD diamond electrode is particularly important for the die sinking EDM of micro-sized parts under finishing EDM conditions without diminishing the shape accuracy.

7 A. Sharma et al. 187 EDM efficiency with the EC-CVD diamond electrode is significantly lower compared to the copper electrode at pulse durations of more than 15 μs as shown in Fig. 5(b). The efficiency of the EC-CVD diamond and graphite electrodes is the same for the various pulse durations, but the Cu-W electrode shows the lowest efficiency in this experiment. 3.4 Discharge voltage and current characteristics The voltage and current waveforms during EDM using EC-CVD diamond, copper, graphite and Cu-W electrodes are shown in Fig. 6. In the case of the EC-CVD diamond electrode, the discharge voltage is about 1.5 times that of the copper and graphite electrodes. This is attributed to the high electrical resistivity (ρ=0.4~ Ω. m) of the EC-CVD diamond. It was reported that in the EDM of electrically conductive Si 3 N 4 ceramics, which have high electrical resistance (ρ= Ω. m), Joule heating occurs due to a voltage drop in the material, as a result of which the temperature of the material increases in a region close to the discharge point. (9) This increase in workpiece temperature together with a low thermal diffusivity (k= m 2 /s) of the electrically conductive Si 3 N 4 ceramic causes arcing which results in the deterioration of the quality of the EDMed surface. Therefore, the effects of this high voltage drop in the case of EDM using the EC-CVD diamond electrode were investigated. Figure 7 shows SEM photographs of the workpiece surfaces after EDM using EC-CVD diamond, copper, graphite and Cu-W electrodes at pulse conditions of t e /t o =10/10 and 50/50 μs. In the case of the surface after EDM using the EC- Fig. 6. Current and voltage waveforms during electric discharge. (u i =90 V, i p =3 A, t e /t o =50/50 μs, EDM area=1.5 mm 2, EDM depth=1 mm, reversed polarity (electrode +), in oil)

8 188 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) Fig. 7. SEM photographs of workpiece surface after EDM. (u i =90 V, i p =3 A, EDM area=1.5 mm 2, EDM depth=1 mm, reversed polarity (electrode +), in oil) CVD diamond, copper, and Cu-W electrodes, no evidence of arcing is noted, whereas, in the case of the graphite electrode, a deterioration of the surface due to arcing is seen. Moreover, for the pulse condition of t e /t o =50/50 μs, the discharge craters on the surface of the workpiece after EDM using the EC-CVD diamond electrode appear larger in size but shallower compared to the craters on the workpiece surfaces after EDM using copper and Cu-W. It is believed that, in the case of the EC-CVD diamond electrode, the temperature of the discharge plasma near the electrode surface due to Joule heating is quite high. As a result of the high temperature built up, the rate of spread of the entire plasma column in the radial direction increases causing a larger crater diameter. However, due to the higher thermal diffusivity of the EC-CVD diamond, the temperature at the workpiece surface decreases rapidly as a result of faster heat dissipation into the diamond electrode. This explains the formation of the shallower craters on the workpiece surface after EDM using the EC-CVD diamond electrode compared to the copper and Cu-W electrodes. The faster heat dissipation into the EC-CVD diamond electrode also helps extinguish the discharge plasma column, which prevents arcing and results in a good EDM surface quality. 3.5 Effect of EDM area The influence of the EDM area on electrode wear was examined for copper, Cu-W and EC-CVD diamond electrodes. The thickness of the electrodes under investigation was selected as 0.5 mm for experimental EDM operation on a heat-treated SKD11 workpiece with a set machining depth of 1 mm. EDM conditions included a no load voltage u i =90 V, a peak current i p =3 A, pulse on time t e =3 and 20 μs, and off time t o =20 μs, and with electrode [+]. Four EDM areas of 0.25, 0.5, 1 and 2 mm 2 were targeted. As illustrated in Fig. 8, by changing the electrode overlap length on the workpiece, the area for EDM was varied. The detailed experimental conditions are stated in Table 1. According to the results shown in Fig. 9, at a long pulse duration of 20 μs, the wear rate of Cu-W is the highest among the electrodes used, whereas, for the copper electrode, the

9 A. Sharma et al. 189 Fig. 8. Schematic of the experiment to investigate the influence of EDM area on electrode wear. Fig. 9. Influence of EDM area on electrode wear. (u i =90 V, i p =3 A, short pulse: t e /t o =3/20 μs, long pulse: t e /t o =20/20 μs) wear remains zero for all the EDM areas except for 0.25 mm 2. In the case of the diamond electrode, the length of the electrode increases by more than 10 μm for all the EDM areas and, especially when the EDM area is 0.25 mm 2 (representing high current density), the length of the electrode increases by about 25 μm. The increase in the size of the electrode (shown as [ ] in the graph) is considered to be due to the adhesion of heat-resolved carbon from the EDM oil onto the electrode surface. At a short pulse duration of 3 μs, the copper electrode showed a wear of 130~150 μm, whereas for the Cu-W electrode it was about 50~65 μm. In comparison, the EC-CVD diamond electrode showed only slight wear for the EDM area of 0.25 mm 2, and for EDM areas more than 0.5 mm 2, the wear rate remained zero. 3.6 EDM performance using knife-edge diamond electrode From the influence of the EDM area on the electrode wear, there is an increased possibility of EC-CVD diamond being used for the EDM of very small areas such as in micro-machining. A knife-edge with an edge radius of 6 μm was machined using EDM on the EC-CVD diamond segment shown in Fig. 10; this knife-edge EC-CVD diamond

10 190 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) (a) Cross section of electrode after EDM (b) SEM photo of the knife-edge electrode Fig. 10. Knife-edge manufactured on EC-CVD diamond electrode. (u i =60 V, i p =1 A, t e /t o =20/20 μs, straight polarity (electrode: )) segment was used as an electrode for ED machining on a stainless steel SUS316 workpiece. Figures 10(a) and 10(b) show the cross section and 3-D views of the EC-CVD diamond segment made using EDM on a copper electrode. For comparison, ED machining was also performed on the same workpiece using a similar knife-edge shaped copper electrode. The EDM conditions selected for this experiment were u i =60 V, i p =1 A, t e /t o =20/20 μs, and machined depth=0.1 mm. The results of the experiment are shown in Fig. 11. The wear depth in the case of the EC-CVD diamond electrode is almost zero, whereas a significant amount of wear on the copper electrode is observed. A sharper radius at the bottom of the groove can be achieved in the case of the EC-CVD diamond electrode compared to the copper electrode even after the ED machining of 6 successive grooves. In the case of the EC- CVD diamond electrode, however, the depth of the groove is increases successively, i.e. the depth of the sixth groove is greater than the depth of the first groove. This may be due to an increase in the electrode size because of the accretion of heat-resolved carbon from the dielectric fluid along with molten workpiece material onto the edge of the EC-CVD diamond electrode. 4. EDM Experiments with EC-CVD Diamond Electrodes in Water 4.1 Electrode wear depth and EDM efficiency The EDM of SKD11 die steel using an EC-CVD diamond electrode was performed in de-ionized water as dielectric working fluid and compared with the results for copper, graphite and Cu-W electrodes. Pulse durations of 3, 7, 15 and 30 μs, a peak current of 6 A, an open circuit voltage of 90 V, a pulse interval of 50 μs, and reversed polarity were used. Figure 12(a) shows the results for wear depths of EC-CVD diamond, copper, graphite and Cu-W electrodes. Although a maximum wear depth of approximately 30 μm is observed for the EC-CVD diamond electrode at a pulse duration of 3 μs, it is significantly lower than the wear depths of 174, 239 and 87 μm for the copper, graphite and Cu-W electrodes,

11 A. Sharma et al. 191 Fig. 11. EDM using knife-edge manufactured on EC-CVD diamond and copper electrodes. (u i =60 V, i p =1 A, t e /t o =20/20 μs, work=stainless steel SUS316, reversed polarity (electrode: +), ED area=0.5 2 mm, machined depth: 0.1 mm) (a) Comparison of wear depth (b) EDM efficiency Fig. 12. Comparison of wear and efficiency among electrode materials in water. (u i =90 V, i p =6 A, t o =50 μs, work=skd11, EDM depth=0.5 mm, reversed polarity (electrode +)) respectively. The significantly lower wear of the EC-CVD diamond electrode is thought to be due to the high thermal stability as well as the high thermal diffusivity of the EC-CVD diamond compared to copper, graphite and Cu-W electrode materials. Figure 12(b) shows the EDM efficiencies of the EC-CVD diamond, copper, graphite and Cu-W electrodes. The EDM efficiency of the graphite electrode is the highest, followed by the copper, diamond and Cu-W electrodes at a pulse duration of 7 μs. For the pulse durations of 3, 15 and 30 μs no significant difference in the EDM efficiency is observed.

12 192 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) 5. Changes on EC-CVD Diamond Electrode Surfaces after EDM To investigate the changes occurring on the EC-CVD diamond electrode surface, Raman spectral characteristics of the surface of the diamond used for EDM were studied. Figure 13(a) shows the Raman spectrum of an unused diamond surface. Figures 13(b) and 13(c) show the Raman spectra of the EC-CVD diamond electrode surface after EDM in oil and water, respectively. The Raman spectrum of the EC-CVD diamond electrode surface after EDM in oil shows the presence of graphitic carbon. This is considered to be heat-resolved carbon released from the dielectric oil. The Raman spectrum in Fig. 13(c) shows that a damaged crystal structure of the EC-CVD diamond is detected when EDM is performed in water. In the absence of oxygen, diamond transforms into graphite when heated above 1200 C, but oxygen accelerates this process. (10) In this case, because the Raman spectrum of the diamond electrode after EDM in water does not show the presence of graphite on its surface, it may be surmised that the wear of the EC-CVD diamond electrode is caused by the oxidation due to the oxygen ions produced as a result of the ionization of water. 6. Mechanism of Low Wear of Diamond Electrodes 6.1 Accelerated adhesion of heat-resolved carbon One of the most important reasons for the low wear of the EC-CVD diamond electrode at reversed polarity, when oil is used as the dielectric medium, is the accelerated adhesion of carbon compared to other electrode materials. In the case of conventional electrode materials, the wear of the anode is lower than the wear of the cathode when EDM is Fig. 13. Raman spectra of diamond surface after EDM.

13 A. Sharma et al. 193 performed in oil. This is mainly due to the adhesion of the heat-resolved carbon from the dielectric oil onto the electrode surface, which protects the electrode from thermal wear. However, this mechanism of prevention of electrode wear is active at pulse durations more than 20 μs. (1,2) In the case of the EC-CVD diamond electrode at reversed polarity (electrode positive), there is an almost zero wear even at the short pulse duration of 3 μs compared to a large amount of wear in the case of copper, graphite and Cu-W electrodes, as shown in Fig. 4. This indicates that in the case of EC-CVD diamond the adhesion of heat-resolved carbon from the dielectric oil onto the electrode surface is faster compared to other electrode materials. 6.2 Thermal stability That the EC-CVD diamond shows higher resistance to oxidation compared to the nonconductive CVD diamond indicates that the EC-CVD diamond has a high thermal stability. This is confirmed from the results of EDM experiments using an EC-CVD diamond electrode in water as shown in Fig. 12(a), which shows a very low wear of the diamond electrode compared to copper, graphite and Cu-W. This property of diamond may also be useful in the case of EDM with the EC-CVD diamond in oil. 7. Conclusions In conclusion, the following can be pointed out. 1. The electrode polarity in the case of EDM using the EC-CVD diamond electrode plays a very important role, i.e. at straight polarity (electrode ), rapid electrode wear is observed, and at reversed polarity (electrode +), almost zero wear of the EC-CVD diamond electrode is observed. 2. The ED machining ability of the EC-CVD diamond electrode without any wear even at very short pulse durations gives it a great advantage toward becoming an important electrode material for the machining of micro-parts. 3. The discharge voltage rises significantly by about 9 V in the case of the EC-CVD diamond resulting in an increase in the plasma temperature due to Joule heating; however, no damaging effects of this on the ED machined workpiece surface are observed. 4. The EC-CVD diamond electrode shows an increased wear depth when water is used as the dielectric fluid; however, this wear depth is significantly lower than that of copper, Cu-W and graphite electrodes, possibly because of higher thermal stability of the EC- CVD diamond. 5. The extremely low wear of the EC-CVD diamond electrode even at short pulse durations may be due to the acceleration in the rate of heat-resolved carbon adhesion onto the EC-CVD diamond electrode surface as well as to the high thermal stability of the EC-CVD diamond.

14 194 New Diamond and Frontier Carbon Technology, Vol. 15, No. 4 (2005) Acknowledgments The authors would like to thank Element Six Co., Ltd. for providing the EC-CVD diamond thick films. The authors also acknowledge the support and advice from Mr. S. Sano and Mr. K. Karato of Sodick Co., Ltd. The authors express their heartfelt thanks to Prof. M. Kunieda and Mr. K. Shoda for their valuable advice. The authors appreciate the support received from Osawa Scientific Studies Grants Foundation. The authors also thank Mr. M. Suzuki of the R&D Center for Advanced Materials and Technology of N.I.T. for assistance in spectroscopy. References 1) H. Xia, M. Kunieda and N. Nishiwaki: IJEM 1 (1996) 45. 2) N. Mohri, M. Suzuki, M. Furuya and N. Saito: Annals of the CIRP 44 (1995) ) K. Suzuki, M. Iwai, A. Sharma, T. Uematsu, K. Shoda and M. Kunieda: Electrical Discharge Machining Using Electrically Conductive CVD Diamond as an Electrode, Seventh Applied Diamond Conference / Third Frontier Carbon Technology Joint Conference (ADC/FCT 2003) (2003) pp ) M. Iwai, A. Sharma, K. Suzuki, M. Kunieda and T. Uematsu: Application of Electrically Conductive Diamond to Removal Machining, ABTECH (2003) pp (in Japanese). 5) K. Suzuki, M. Iwai, A. Sharma and T. Uematsu: Electrical Discharge Machining Using Electrically Conductive Diamond as an Electrode, Proc. JSPE Autumn Conference (2003) p ) K. Suzuki, M. Iwai, A. Sharma, T. Uematsu and M. Kunieda: Low-Wear Diamond Electrode for Micro-EDM of Die-Steel, in Advances in Abrasive Technology VI, Key Engineering Materials Vols. (2003) pp ) A. Sharma, K. Suzuki, M. Iwai and T. Uematsu: EDM Using Electrically Conductive Diamond Electrode - 1st report: EDM property, Proc. JSPE Spring Conference (2004) pp ) K. Suzuki, A. Sharma, M. Iwai, T. Uematsu, K. Shoda and M. Kunieda: New Diamond and Frontier Carbon Technology 14 (2004) 35. 9) T. Saeki, M. Kunieda, M. Ueki and Y. Satoh: Influence of Joule Heating on EDM Process of High-Electric Resistivity Materials, Transport Phenomena in Materials Processing and Manufacturing, ASME, (1996) HTD-336: ) H. O. Pierson: Handbook of Carbon, Graphite, Diamond and Fullerenes (Noyes Publications, New Jersey, 1993).