The growth of commercially available SiC devices is due in large part to the reduction of materials defects that has occurred over the last decade or more. Further market penetration will depend on continued materials improvements and improved techniques for identifying materials defects that degrade the yield, performance and reliability of SiC power devices.
The understanding and reduction of basal plane dislocations (BPDs) have been major research efforts to improve SiC technology. BPDs cause degradation in a number of power devices including PiN diodes, BJTs and MOSFETs (1-5). While there have been dramatic improvements in reducing the BPD concentration in epitaxial layers, further improvements are still needed. The problems are more pronounced for higher voltage parts that require thicker epitaxy. Until recently, the dominant source of BPDs in epitaxial layers has been from BPDs that propagate from the substrate into the epitaxial layers. A large reduction occurred a decade ago when the wafer offcut angle was changed from 8° to 4°, which strongly increased the tendency of BPDs to convert to threading edge dislocations at the beginning of the epitaxial growth (6). The reduction of BPDs in substrates has also been important (7). Optimizing the epitaxial growth has increased the conversion efficiency at the onset of growth and in many cases can produce a drift layer that is free of BPDs over almost all of the wafer area. As the concentration of BPDs propagating from the substrate has plummeted, other sources of BPDs in the epitaxial layers have become important (8,9). One of these new sources of BPDs is the combination of implantation and activation anneal (10,11). As with all of the alternate BPD sources, these BPDs are just as degrading as BPDs coming from the substrate and must be similarly suppressed. However they originate from the top of the epitaxial layer and drift towards the substrate as well as laterally during the high temperature anneal. This formation of BPDs appears to be sensitive to the details of the implantation and activation anneal. If the penetration of SiC devices into the power electronics applications is to continue, the work to suppress BPDs as well as all other device degrading extended defects must continue. Furthermore, the techniques for quantifying their concentration in wafers that have been recently developed must become a routine part of SiC fabrication monitoring.
1. H. Lendenmann, F. Dahlquist, J. P. Bergman, H. Bleichner and C. Hallin: Mater. Sci. Forum Vol. 389-393 (2002), p. 1259.
2. R.E. Stahlbush, M. Fatemi, J.B. Fedison, S.D. Arthur, L.B. Rowland and S. Wang: J. Elec. Mater. Vol. 31 (2002), p. 370.
3. A. Agarwal, S. Krishnaswami, J. Richmond, C. Capell, S.-H. Ryu, J. Palmour, B. Geil, D. Katsis, C. Scozzie and R.E. Stahlbush: Mater. Sci. Forum Vol. 527-529 (2002), p. 1409.
4. Q. Zhang, A. Agarwal, A. Burk, M. O¢Loughlin, J. Palmour, R. Stahlbush and C. Scozzie: Mater. Sci. Forum, Vol. 645-648 (2010), p. 331.
5. A. Agarwal, H. Fatima, S. Haney and S.-H. Ryu: IEEE Elec. Dev. Lett. Vol. 28 (2007), p. 587.
6. R.L. Myers-Ward, B.L. VanMil, R.E. Stahlbush, S.L. Katz, J.M. McCrate, S.A. Kitt, C.R. Eddy, Jr., and D.K. Gaskill: Mater. Sci. Forum Vol. 615-617 (2009), p. 105.
7. M. Dudley, N. Zhang, Y. Zhang, B. Raghothamachar, S. Byrappa, G. Choi, E.K. Sanchez, D. Hansen, R. Drachev, and M.J. Loboda, Mater. Sci. Forum 645-648, 291 (2010).
8. N.A. Mahadik, R.E. Stahlbush, M.G. Ancona, E.A. Imhoff, K.D. Hobart, R.L. Myers-Ward, C.R. Eddy, D.K. Gaskill, and F.J. Kub, Appl. Phys. Lett. 100, 042102 (2012).
9. R.E. Stahlbush, N.A. Mahadik, M.J. O’Loughlin, Mater. Sci. Forum 778-780,309 (2014).
10. C. Ota, J. Nishio, K. Takao, T. Shinohe, Mater. Sci. Forum 778-780, 851 (2014).
11. R.E. Stahlbush, N.A. Mahadik, Q.J. Zhang, A.A. Burk, B.A. Hull and J. Young, Mater. Sci. Forum 821-823, 851 (2015).