Silicon Carbide Technology:Future of SiC

Future of SiC

It can be safely predicted that SiC will never displace silicon as the dominant semiconductor used for the manufacture of the vast majority of the world’s electronic chips that are primarily low-voltage digital and analog chips targeted for operation in normal human environments (computers, cell phones, etc.). SiC will only be used where substantial benefits are enabled by SiC’s ability to expand the envelope of high-power and high-temperature operational conditions such as the applications described in Section 5.3. Perhaps, the only major existing application area where SiC might substantially displace today’s use of silicon is the area of discrete power devices used in power conversion, motor control, and management circuits.

The power device market, along with the automotive sensing market present the largest-volume market opportunity for SiC-based semiconductor components. However, the end consumers in both of these applications demand excruciatingly high reliability (i.e., no operational failures) combined with com- petitively low overall cost. For SiC electronics technology to have large impact, it must greatly evolve from its present status to meet these demands. There is clearly a very large discrepancy between the revolutionary broad theoretical promise of SiC semiconductor electronics technology (Section 5.3) versus the operational capability of SiC-based components that have actually been deployed in only a few commercial and military applications (Section 5.6). Likewise, a large discrepancy also exists between the capabilities of laboratory SiC devices compared with commercially deployed SiC devices. The inability of many “successful” SiC laboratory prototypes to rapidly transition to commercial product demonstrates both the difficulty and criticality of achieving acceptable reliability and costs.

Future Tied to Material Issues

The previous sections of this chapter have already highlighted major known technical obstacles and immaturities that are largely responsible for hindered SiC device capability. In the most general terms, these obstacles boil down to a handful of key fundamental material issues. The rate at which the most critical of these fundamental issues is solved will greatly impact the availability, capability, and usefulness of SiC semiconductor electronics. Therefore, the future of SiC electronics is linked to investment in basic material research toward solving challenging material-related impediments to SiC device performance, yield, and reliability.

The material challenge that is arguably the biggest key to the future of SiC is the removal of dislocations from SiC wafers. As described previously in this chapter and references therein, the most important SiC power rectifier performance metrics, including device ratings, reliability, and cost are inescapably impacted by high dislocation densities present in commercial SiC wafers and epilayers. If mass-produced SiC wafer quality approached that of silicon wafers (which typically contain less than one dislocation defect per square centimeter), far more capable SiC unipolar and bipolar high-power rectifiers (including devices with kilovolt and kiloampere ratings) would rapidly become widely available for beneficial use in a far larger variety of high-power applications. Similar improvements would also be realized in SiC transistors, paving the way for SiC high-power devices to indeed beneficially displace silicon-based power devices in a tremendously broad and useful array of applications and systems (Section 5.3). This advance- ment would unlock a much more rapid and broad SiC-enabled power electronic systems “revolution” compared to the relatively slower “evolution” and niche-market insertion that has occurred since SiC wafers were first commercialized roughly 15 years ago. As mentioned in Section 5.4, recent laboratory results [83] indicate that drastic reductions in SiC wafer dislocations are possible using radically new approaches to SiC wafer growth compared to standard boule-growth techniques practiced by all com- mercial SiC wafer vendors for over a decade. Arguably, the ultimate future of SiC high-power devices may hinge on the development and practical commercialization of low dislocation density SiC growth techniques substantially different from those employed today.

It is important to note that other emerging wide bandgap semiconductors besides SiC theoretically offer similarly large electrical system benefits over silicon semiconductor technology as described in Section 5.3. For example, diamond and some Group III-nitride compound semiconductors (such as GaN; Table 5.1) have high breakdown field and low intrinsic carrier concentration that enables operation at power densities, frequencies, and temperatures comparable to or exceeding SiC. Like SiC, however, electrical devices in these semiconductors are also hindered by a variety of difficult material challenges that must be overcome in order for beneficially high performance to be reliably achieved and commer- cialized. If SiC electronics capability expansion evolves too slowly compared to other wide bandgap semiconductors, the possibility exists that the latter will capture applications and markets originally envisioned for SiC. However, if SiC succeeds in being the first to offer reliable and cost-effective wide bandgap capability to a particular application, subsequent wide-bandgap technologies would probably need to achieve far better cost/performance metrics in order to displace SiC. It is therefore likely that SiC, to some degree, will continue its evolution toward expanding the operational envelope of semicon- ductor electronics capability.

Further Recommended Reading

This chapter has presented a brief summary overview of evolving SiC semiconductor device technology. The following publications, which were heavily referenced in this chapter, are highly recommended as supplemental reading to more completely cover SiC electronics technology development in much greater technical detail than possible within this short chapter.

Reference 11 is a collection of invited in-depth papers from recognized leaders in SiC technology development that first appeared in special issues of the journal Physica Status Solidi (a 162 (1)) and (b 202, (1)) in 1997. In 2003, the same editors published a follow-on book [12] containing additional invited papers to update readers on new “recent major advances” in SiC since the 1997 book.

One of the best sources of the most up-to-date SiC electronics technology development information is the International Conference on Silicon Carbide and Related Materials (ICSCRM), which is held every 2 years (years ending in odd numbers). To bridge the 24-month gap between international SiC meetings, the European Conference on Silicon Carbide and Related Materials (ECSCRM) is held in years ending in even numbers. Since 1999, the proceedings of peer-reviewed papers presented at both the International and European SiC conferences have been published by Trans Tech Publications as volumes in its Materials Science Forum offering, which are available online via paid subscription (http:// www.scientific.net). In addition, the meetings of the Materials Research Society (MRS) often hold symposiums and publish proceedings (book and online editions; http://www.mrs.org) dedicated to SiC electronics technology development. Reference 200 is the proceedings of the most recent MRS SiC symposium held in April 2004, and the next such symposium is scheduled for the 2006 MRS spring meeting in San Francisco.

The following technical journal issues contain collections of invited papers from SiC electronics experts that offer more detailed insights than this-chapter, yet are conveniently brief compared to other volumes already mentioned in this section:

1. Proceedings of the IEEE, Special Issue on Wide Bandgap Semiconductor Devices, 90 (6), June 2002.

2. Materials Research Society Bulletin, Advances in Silicon Carbide Electronics, 30(4), April 2005.

In addition, a variety of internet websites contain useful SiC information and links can be located using widely available internet search engine services. The author of this chapter maintains a website that contains information and links to other useful SiC internet websites at http://www.grc.nasa.gov/ WWW/SiC/.