Vol.3 No.4 2011

26/72

Research paper : Development of single-crystalline diamond wafers (A. Chayahara et al.)−267−Synthesiology - English edition Vol.3 No.4 (2011) Discussions with Reviewers1 Clarification of the development goal and comparison of the existing technologiesComment (Kazuo Igarashi, Institute of National Colleges of Technology, Japan and Hisao Ichijo, Tsukuba Center, Inc.)In this paper, the ways of using the single-crystal diamond wafer is not stated clearly. For example, if the target is for use in the power device substrates, I think you will have a paper with better scenario scheme by stating the size you set as the target for size increase as indicated in the subtitle, where you stand now in achieving the target, and the superiority of the direct wafer fabrication technology and its applicability to size increase.Also, there are technological developments to address the topics of high-speed growth, low defect crystal, and size increase. I think people can better understand the result of this research if you provide numerical comparison (such as how much faster it is) with the existing or other technologies.Answer (Akiyoshi Chayahara)The objective of this development is the utilization in the substrate of next-generation power devices. We aim to create the 2-inch wafer needed for the manufacturing line of the prototype device. Currently, we can fabricate a maximum 1 cm single-crystal substrate by ultrahigh-pressure synthesis, and this is called the Ib type that costs 1 to 2 million yen. The size and cost make its use difficult. The direct wafer technology can be applied to a substrate with 2-inch diameter for polycrystal. We believe it is possible to attain the target size, and a one-digit decrease in price can be expected.For the numerical comparisons with existing or current technologies, other than the crystallization property, I added Nobuteru TsubouchiCompleted the master’s course at the Department of Physics, Graduate School of Science, Osaka University in 1994. Joined the Osaka National Research Institute, Agency of Industrial Science and Technology in 1994. Worked at the R&D Lab for Purity Control Material, AIST and the Diamond Research Center. Senior Researcher at the Diamond Research Laboratory in 2010. Doctor of Engineering. In this paper, was in charge of crystal evaluation.Hideaki YamadaCompleted the doctorate course at the Graduate School of Science and Technology, Niigata University in 2002. Doctor of Science. Industry-academia collaborative researcher at the Graduate School of Energy Science, Kyoto University in 2002. AIST Postdoctoral Researcher of the Diamond Research Center, AIST in 2004. Researcher of the Diamond Research Center in 2006. Has been involved in the theoretical and simulation study of nuclear fusion plasma and process plasma, as well as the synthesis of single-crystal diamond. In this paper, was in charge of the plasma CVD simulation and the fabrication of mosaic crystal.sentences that contain the numerical values in the appropriate places in the text to make the comparisons easier.2 Quality of the waferQuestion (Kazuo Igarashi)While it depends for what the single-crystal diamond wafer is used, I think the quality (transparency, impurity concentration, concentration of crystallizations/defects, etc.) must be guaranteed. What is the level of the current wafer fabricated by the microwave plasma CVD method in terms of quality? If further improvement of the quality is needed, what key technologies are required?Answer (Akiyoshi Chayahara)When synthesizing the single-crystal using the microwave plasma CVD method, we add ultralow volume of nitrogen to control the macro defect called the abnormal nucleus. The crystallization to the same degree as in the ultrahigh-pressure synthesis Ib type has been obtained from the half-value width of the x-ray locking curve. For semiconductor use, this level is sufficient as the replacement for use in the R&D of the devices that conventionally used the Ib type substrate. To further develop the research, the reduction of the dislocation density is necessary, and we are aware that the pretreatment of the epitaxial growth is important. Currently, the disadvantage of the direct wafer technology using the electrolysis etching is that the conductive substrate cannot be separated. Therefore, a technology that enables this is awaited. For optical use, there is a problem since nitrogen-related absorption occurs in the visible range. When growth is done without nitrogen addition, transparency can be obtained, but currently the growth rate is 10 m/h or less. Or, the transparency can be obtained by treating the nitrogen added diamond with ultrahigh-pressure. In either case, cost will be an issue.3 Selection and integration in the wafer technologyComment (Hisao Ichijo)Since you explain the reasons for employing the microwave plasma CVD, if you briefly explain the selection and integration of the various technologies for wafer creation, I think the importance as a Synthesiology paper will become clearer.Answer (Akiyoshi Chayahara)In the wafer fabrication, unless there is some innovation in the cutting technology, there is no room for selection other than laser cutting and “direct wafer” technologies. In the case of laser cutting, about 1 mm width disappears as the reserve portion to cut diamond to 1 cm size. Even if we achieve high speed, the cost will be a problem at the current growth rate of 50 m/h, and when over 1 inch is cut, reserve portion cannot be tolerated. Therefore, at this point, we believe the “direct wafer technology” is the most realistic and optimal manufacturing method. We hope the people of the processing fields will become aware that such issues are present, and propose an applicable cutting technology.For mosaic wafer, we plan to continue the size increase of the single crystal by the repeated growth method, but at this point we do not have any ideas for further accelerating the size increase. In contrast, bonding is done in the lateral direction in the mosaic method, and is a method that matches well with the CVD that enables deposition on large surface area, and immediate size increase is possible. Also, if the defects are reduced in the crystal fragments, these can be “copied” and bonded, and the mosaic wafer, in principle, can be increased in terms of quality instantly. Therefore, we shall work on the mosaic technique as our immediate topic.

※このページを正しく表示するにはFlashPlayer9以上が必要です