Exclusive Interview with Dr. Ni Weijiang, Founder of Xinta Electronics: Silicon Carbide Chips Boast Extensive Future Application Scenarios
As a representative of the third-generation wide-bandgap semiconductor materials, silicon carbide (SiC) has significant advantages in indicators such as bandgap width, breakdown electric field, thermal conductivity, and electron saturation velocity. It can meet the needs of modern industry for high power, high voltage, and high frequency, and is mainly used to manufacture high-speed, high-frequency, high-power, and light-emitting electronic components. The downstream application fields of SiC devices include new energy vehicles, photovoltaic power generation, rail transit, smart grids, 5G communications, etc., among which new energy vehicles are the largest end-user market. Nanjing Innovation Investment Group has invited Dr. Ni Weijiang, Founder of Anhui Xinta Electronic Technology Co., Ltd., to share with us the future development trends of SiC devices.
Ni Weijiang holds a PhD from the Chinese Academy of Sciences, is a Senior Engineer, a member of the IEEE, and a reviewer. Dr. Ni is a member of the Beijing Third-Generation Semiconductor Expert Group. He has been engaged in the R&D and industrialization of silicon carbide at the 55th Research Institute of China Electronics Technology Group Corporation, Techown Semiconductor, and CETC Solid State Technology Co., Ltd., with 15 years of experience in the silicon carbide industry.
In October 2020, Dr. Ni founded Anhui Xinta Electronic Technology Co., Ltd. in Wuhu, Anhui Province. Xinta has established R&D centers, application technology centers, and sales centers in Beijing, Hangzhou, and Shenzhen, and a packaging and testing base for silicon carbide power modules with an annual output of 300,000 sets in Huzhou. Xinta Electronics is a national high-tech enterprise focusing on the innovation and localization of third-generation semiconductor power devices and application technologies.
Could you analyze the technology development trend of SiC devices, what are the technical routes of SiC super-junction MOS, and will SiC super-junction MOS have application potential in the future?
In the future, planar, trench, and super-junction technologies will all develop for SiC devices. The greatest advantage of the super-junction structure is that it reduces the resistance of the epitaxial layer, so it has significant advantages in high-voltage devices—for example, it will have greater advantages in 1700V, 2000V, and 3300V devices in the future. There are two technical directions in chip design: "planar + super-junction" or "trench + super-junction". The process route of the super-junction itself refers to that of silicon devices: one is multiple epitaxial growth, and the other is deep trench filling. In silicon carbide, the cost of trench filling is more advantageous.
What are the reasons for the low yield of SiC MOS for main drive high current, and what are the keys to improving the yield in the future?
In recent years, domestic silicon carbide materials and manufacturing technologies have made great progress. Currently, the yield of MOS with medium current specifications can reach more than 90%. An important factor restricting the yield of main drive MOS is the defect density of substrates and epitaxial layers, which is also related to the early failure of chips. Therefore, the progress of single crystal and epitaxial technologies is crucial. The second factor is the manufacturing process and defect control during the manufacturing process.
What are the advantages and disadvantages of enterprises focusing on SiC devices and those focusing on SiC modules in product competition?
If a SiC enterprise only produces discrete devices (single tubes) without manufacturing modules, it will limit the application fields and market expansion of its products, especially in high-power applications. On the other hand, if an enterprise only produces modules without mastering core chip technology, the source of chips will be affected, the supply chain will be passive, and the overall performance and price of products will be uncontrollable. Therefore, in the SiC industry, chips are the core, while modules are the extension and expansion of products and markets.
What advantages does SiC MOS have over silicon-based IGBTs in the fields of photovoltaic inverters and new energy vehicles, and what industry pain points does it mainly solve? Compared with silicon-based IGBTs, what price range of SiC devices/modules can downstream customers accept?
With the maturity of technologies and products, third-generation semiconductor devices (such as SiC MOSFETs) will replace silicon-based products in some vehicle models, greatly increasing the driving range of vehicles, shortening charging time, optimizing the overall vehicle architecture, and improving driving experience. Currently, the application scope of SiC is gradually expanding from high-end vehicle models to mid-to-low-end ones. New energy vehicles with a driving range of less than 500km are expected to gradually adopt SiC power devices.
After photovoltaic inverters adopt SiC power modules, their conversion efficiency can be increased from 96% to over 99%, and energy loss can be reduced by more than 30%. This greatly improves the cycle life of equipment and has the advantages of reducing system volume, increasing power density, extending device service life, and lowering system heat dissipation requirements.
In summary, except for the higher price than IGBTs, SiC has no other shortcomings in terms of performance and customer acceptance. At present, downstream customers, especially high-end ones, can accept a price that is more than 3 times that of IGBTs. However, when the price is twice that of IGBTs, the overall system cost will be lower than that of using silicon-based IGBTs, and the industry will usher in a real explosive growth point. With the rapid advancement of technology and the expansion of industrial scale, this point in time is approaching.
In addition to the current development and penetration of downstream application scenarios such as vehicles, wind power, photovoltaics, and energy storage, will there be other application scenarios for SiC in the future?
SiC chips will have extremely extensive application scenarios in the future. In addition to the fields of photovoltaics, energy storage, charging, and vehicles, SiC chips will also play an important role in military industry, aerospace (including deep space exploration), data center power supplies, industrial frequency conversion, UPS, rail transit, and power grids.
What are the business models of SiC device enterprises, and what will be the focus of competition and living space for various enterprises in the future?
The main business models of SiC device enterprises are Fabless and IDM. Most domestic SiC power device manufacturers, including Fabless and Foundry enterprises, tend to evolve towards the IDM model. However, the IDM model is not suitable for all enterprises. At present, many domestic SiC manufacturers are not large in scale and have not yet achieved profitability. If they build large-scale factories, the operating costs will be too high, which will be a great test for cash flow. There are also problems with the difficulty of process development and customer recognition.
If supported by foundries, Fabless manufacturers do have greater flexibility in design, faster R&D progress in SiC MOS, and are more suitable for entrepreneurial teams with strong R&D capabilities to deploy. At the same time, the qualifications of foundries can endorse the reliability of their supply chains. Considering factors such as the current development stage and time window of the SiC industry, the two business models of IDM and Foundry + Fabless will coexist for a long time in China, and each will meet the different application scenarios of the end market.
Given that the current SiC industry is in the early to mid-stage of development, Fabless + Foundry that can conduct in-depth cooperation—such as forming a C-IDM (Collaborative IDM) model—will have greater advantages. Therefore, technical teams that have domestic Foundry resources, master core design and process capabilities, and possess long-term innovation capabilities will have greater competitive advantages.
Will GaN (gallium nitride) possibly replace the position of SiC in the future?
GaN power devices not only have performance advantages similar to SiC in wide-bandgap materials but also have greater potential for cost control. However, since GaN is a heterogeneous epitaxial material, its inherent defects and reliability will impose certain restrictions on its applications. Therefore, most GaN devices are used in medium-low voltage, high-frequency, and even highly integrated fields, where there will be a large market in the future. Although GaN and SiC have partial competition, they are more complementary in the market.
Thank you for Dr. Ni’s sharing. We believe that the SiC industry will surely have a huge market in the future!
Source: Nanjing Innovation Investment Group
Reviewer: Xue Yao
Publisher: You Yi
