Current Status and Prospects of Silicon Carbide Component Technology

Third-generation semiconductors have been at the center of numerous market discussions in recent years. The most notable among these discussions has involved the business opportunities for the application of silicon carbide (SiC) power components in electric vehicles. When it comes to electric vehicle systems, silicon carbide components are applied mainly in the Inverter, On Board Charger (OBC), and DC-to-DC Converter, etc. Compared to the performance of traditional silicon-based modules, silicon carbide components can reduce power conversion loss by about 50%, lower the production cost of power conversion systems by about 20%, and improve the battery life of electric vehicles by about 4%.

At present, the development of electric vehicle technology is largely being driven by the net-zero carbon emission policies of countries around the world. As such, it has become the focus of development for the global automotive industry and will likely remain so for the next 10 years. Many major manufacturers are already rushing to invest in this field. For example, the Hon Hai Group spent 3.7 billion NT dollars in 2021 to establish “Hongyang Semiconductor”, a wholly-owned subsidiary that will serve as its silicon carbide R&D center, through the acquisition of the Macronix 6-inch wafer fab and to build a complete electric vehicle supply chain thereby. Before that, Tesla took the lead by integrating silicon carbide components into its mass-produced Model 3 vehicles in 2018. 

Furthermore, General Motors (GM), Volkswagen and others have all announced that they will be introducing silicon carbide technologies in their new models in 2022. STMicroelectronics has also made a strategic alliance agreement with Renault-Nissan-Mitsubishi Alliance and BYD Auto for a long-term supply of silicon carbide components for their On Board Chargers. The Renault-Nissan-Mitsubishi Alliance plans to launch as many as thirty electric vehicles by 2030 with an investment of over 20 billion euros in order to seize market shares in the field of electric vehicles. These and other recent market developments make it clear that silicon carbide power components will soon be the stars of the automotive semiconductor industry.

In this issue, MA-tek has specially invited Professor Bing-Yue Tsui, a top scholar in the field of third-generation semiconductor research, to write an article introducing the technological developments and trends of silicon carbide power components for the “New Technology Channel | Collaboration Column” so that we can help our readers understand the progress being made in the academic research of this important scientific field.

Director of R&D Center & Marketing Division, Chris Chen, 2022/02/20

Current Status and Prospects of Silicon Carbide Component Technology

Professor Bing-Yue Tsui  

Institute of Electronics, National Yang Ming Jiaotong University

-

Power semiconductor components are widely used in power systems, power supplies, automotive electronics, motor controls, radio frequency systems, communication equipment, and thin-film transistor liquid-crystal displays, etc. Due to silicon’s low cost and mature technologies, the vast majority of power semiconductor components are currently Si components. However, since the energy band gap of Si is only 1.12 eV, it has some fundamental limitations when it comes to high power applications. These include a low breakdown voltage, high on-resistance (R_on,sp), high reverse bias leakage current, low operating temperature, and more. Also, because of the withstand voltage and on-resistance limitations, bipolar components such as PiN diodes (PND) or Insulated Gate Bipolar Transistors (IGBT) must be used instead of monolithic components such as Schottky Barrier Diodes (SBD) and MOSFETs for high power applications. This makes it impossible to balance power and speed. The aforementioned factors are why wide-bandgap semiconductor materials such as silicon carbide (SiC) and gallium nitride (GaN) have garnered so much attention in the last decade.

SiC has 250 crystal phases. Among them, 4H-SiC has been determined to be the most suitable for making high-power semiconductor components. 4H-SiC has an energy band gap of 3.25 eV even at 300°C, and the intrinsic carrier concentration is lower than that of Si even at room temperature. As such, it is very suitable for operation in high temperature environments. In addition, the breakdown electric field of 4H-SiC is 10 times greater than that of Si. This means that, at 10 times the doping concentration, it is possible to achieve up to 10 times the breakdown voltage. This is because, when the doping concentration is increased, the carrier drift region is shortened, and the on-resistance can be reduced by more than 100 times. GaN has higher carrier mobility than 4H-SiC and is expected to have a higher operating frequency and lower on-resistance. However, the thermal conductivity of 4H-SiC is three times greater than that of GaN, making it more suitable for high power and high temperature environments.

Figure 1. Material and Physical Properties of SiC and the Application Advantages of These Properties in High Power Systems [1]

When high power components are in use, the rating voltage of the power component must be greater than the system voltage. Take electric vehicles for example. Currently, the main electric vehicle battery voltage is 400 V. However, after taking the reliability of the power module into consideration, we find that it should be actually using 650 V class power components. Figure 2 shows the ratio of system voltage to rating voltage at FIT=100 [2]. The higher the system voltage, the lower the ratio of system voltage to rating voltage. This indicates the need for power components with higher breakdown voltages. When the battery voltage of an electric vehicle increases to 800 V, it will require 1.2 kV class power components.

Figure 2. Ratio of Applied Voltage to Rating Voltage at a Failure Rate of FIT=100 [2]

Today, large SiC component manufacturers in Europe, America and Japan, such as STMicroelectronics, Infineon, Wolfspeed, Rohm and other companies, already have mature 650 V, 1.2 kV SiC MOSFET products which are widely used in electric vehicles and charging facilities. Higher voltage applications and harsher application environments, however, have greater demands. For example, the voltage of a wind turbine is less than 1 kV, but it needs to use 1.7 kV class power components. The solar power trend will increase this by another 1.5 kV, at which point 3.3 kV power components will be needed. Other applications such as rail transportation, high-speed railways, smart grids, and industrial motors, etc. will also need power components with rating voltages of 3.3 kV or higher, as shown in Figure 3 [3]. Japan’s latest N700 shinkansen train uses a 3.3 kV SBD.

Figure 3. Fields of Application for SiC High Power Components [3]

Discrete power devices can be basically divided into diodes, which are used for rectification, and transistors, which are used for switching. When the current to be turned on is large, the vertical structure is mainly used. So far, the trajectory of the development of SiC power components is basically following that of Si power components. The first to go into mass production was the SBD-type diode. Differences in the detailed structures of these diodes give us the Schottky Barrier Diode (SBD), the Junction Barrier Schottky Diode (JBSD), and the Trench Junction Barrier Schottky (TJBS) etc. The schematic diagram of this structure is shown in Figure 4. Though there is still room for fine-tuning and optimization, this can be considered a mature component. Because the SBD breakdown voltage of SiC can cover the PND of Si, and since the speed of SBDs is higher than that of PNDs, SiC PNDs are only necessary for higher voltage applications. The types and applicable voltage ranges of Si and SiC power components is shown in Figure 5 [4].

There are two types of high power MOSFET structures. The first is the Vertical Double-implantation MOSFET (VDMOSFET). Its cross-sectional structure is shown in Figure 6(a). The second type is the vertical channel trench gate or UMOSFET. Its cross-sectional structure is shown in Figure 6(b). The advantage of the VDMOSFET is that its fabrication processes are simple because the reliability problems that can arise from those processes are simple. The most critical process is the gate oxidation, during which the SiO₂/SiC interface needs to be an unobstructed interface in order to improve carrier mobility, reduce the influence of channel resistance on on-resistance, and achieve sufficient reliability. At present, nitrogen passivation by NO or N₂O annealing has had the best effect. However, the electron mobility can only reach about 30~50 cm²/V-sec—a far cry from the 1000 cm²/V-sec of bulk materials. How the carrier mobility can be raised above 100 cm²/V-sec is one direction of development into which major manufacturers and academic institutions alike have been putting their efforts.

Figure 6. (a)  VDMOSFET Cross-Section

(b)  UMOSFET Cross-Section

The structure of the planar channel makes it difficult to reduce the cell pitch. The junction field effect transistor (JFET) effect caused by the adjacent P-body, or P-base, also increases the on-resistance. At the same process level, the cell pitch of the UMOSFET is only about 60% of that of the VDMOSFET and ideally eliminates the JFET effect, resulting in a lower on-resistance. However, the fabrication processes for UMOSFETs are more difficult than those for VDMOSFETs. The trench etching process and the gate oxidation process of the trench sidewall require precision control. The reduction of the electric field at the bottom of the trench in particular requires the addition of numerous process steps. For example, Rohm has adopted a double-trench structure, as shown in Figure 7(a) [5]. Infineon, on the other hand, uses an asymmetric trench structure, as shown in Figure 7(b) [6]. Another approach is that of adding a P+ shielding layer at the bottom of the trench, as shown in Figure 7(c) [7]. However, in addition to increasing process costs, these practices produce a JFET effect.

Figure 7. (a) Rohm’s Double Trench UMOSFET [5]

(b) Infineon’s Asymmetric UMOSFET [6]

(c) Mitsubishi’s UMOSFET with Additional P+ Shielding Layer at the Bottom of the Trench [7]

Another way to avoid having an excessively strong electric field in the gate oxide layer at the bottom of the trench is to increase the thickness of that oxide layer. Papers on SiC UMOSFET where the thickness of the gate oxide layer at the bottom of the trench is increased have been published by Toyota’s Takaya Group as recently as 2013 [8]. At most, the electric field strength was reduced by 46% and the Q_gd was reduced by 38% while the on-resistance was increased only slightly by about 4%, as shown in Figure 8. However, because the trenches need to first be filled with SiO₂ then etched back to leave only the SiO₂ at the bottom of the trench, the variations in the thickness of the TBOX oxide layer make the variations in device characteristics too great. Until we have better process technologies, the TBOX structure is suitable only for research, not for mass production.

The laboratory headed by Professor Bing-Yue Tsui of Yang Ming Jiaotong University’s Institute of Electronics once proposed a research plan for forming a TBOX by increasing the oxidation rate of the bottom of the trench using the Ar-Ion Implantation Amorphization (Ar-PAI) process. They successfully produced a TBOX structure, as shown in Figure 9(a). Although the fabrication of complete power components has not yet been realized, a R_on,sp of 2.07 mW-cm² was measured for a 1.2 kV class UMOSFET in a single trench test structure, as shown in Figure 9(b). This demonstrated the potential of the Ar-PAI TBOX UMOSFET.

Figure 9. (a) TBOX UMOSFET Implementation Structure Proposed by the Leaders of Sub-Projects One and Three

(b) Turn-On Characteristics of Single Trench TBOX UMOSFET

Because of its complex structure and processes, the UMOSFET’s reliability problems are more severe than those of the VDMOSFET. Therefore, although the UMOSFET has appeared on the technical blueprints published on the Wolfspeed and STMicroelectronics company websites, currently, only Rohm, Infineon, Bosch, and Mitsubishi have launched UMOSFET products. The products of more than ten other companies, including Wolfspeed and STMicroelectronics, are all VDMOSFET products.

To withstand high voltages, a power MOSFET needs to have a low concentration drift region at its drain end. The resistance of that drift region tends to be roughly equivalent to the breakdown voltage raised to the power of 2.5. In other words, the on-resistance will increase rapidly as the voltage increases, leading to severe on-power loss. Figure 10 shows the relationship between the on-resistance of the drift region and the breakdown voltage. This is the theoretical limit of conventional power components. The only way to overcome this limitation is to use a super junction (SJ) structure [9]. This structure is shown in Figure 11 [10]. The main difference between this structure and the traditional high power MOSFET structure is the addition of P-type pillars, or a P-Drift Region, in the drift region. When there is a reverse bias, the junction between the P-type pillar and the N-type drift region depletes the entire drift region. Therefore, it becomes possible to use a thinner drift region. In addition, because the P-type pillar helps the N-type drift region reach depletion, the concentration of the N-type drift region can be higher than that of conventional drift regions. These two factors cause the on-resistance of the SJ drift region to decrease during forward conduction.

Figure 10. Theoretical limits of Si, SiC, and GaN Semiconductor Materials’ On-Resistance and Breakdown Voltages [7]

Figure 11. Schematic Diagram of SJ MOSFET Cross-Section [9]

The SJ structure is fabricated within the drift region, so it can be matched with either the VDMOSFET or the UMOSFET. Of course, it can also be used with SBD-type power components. However, the fabrication of the SJ structure is rather difficult for SiC because most doping elements do not diffuse in SiC. One way to form a P-type pillar is to first etch away the N-type drift region in the appropriate area then use epitaxy to form the P-type pillars. However, the difficulty of SiC epitaxy is much higher than that of Si. To the best of this author’s knowledge, there is currently no published literature on the successful epitaxial growth from trenches for SiC. Another method of making P-type pillars is to combine epitaxy with multiple ion implantations. Although AIST was the first to present a 3.3 kV SJ MOSFET at the 2021 ISPSD Symposium, it took 16 or 28 epitaxy and ion implantations. As such, this is probably not an acceptable process for mass production [11]. The production method for this structure must be simplified if it is to have an opportunity for commercialization.

A gate driver is required for controlling a MOSFET switch. At present, SiC MOSFET gate drivers are made using Si ICs. However, though SiC components can withstand harsher environments than Si power components, Si ICs cannot. Therefore, a Si IC must be independently packaged and isolated, which increases the physical volume of the power system. Furthermore, signal transmissions are affected by the inductance and capacitance of the package wiring, power loss is increased, and performance is affected. As such, it would still be the most ideal to use a SiC IC driver. Only then can the material advantages of SiC be fully utilized. Because of this, SiC IC technology that can be integrated with vertical MOSFETs is a research topic that has also received a great deal of attention in recent years.

• SiC IC research began in the early 1990s. 6H-SiC was used at the time, and digital and analogue circuit blocks were realized successively [12-15].
• In 1994, Purdue University published the first NMOSFET circuit, demonstrating basic circuit units such as an Inverter, NAND, NOR, XNOR, D-latches, RS flip-flops, binary counter, and half adder, etc. [12].
• In 1999, Purdue University and Cornell University jointly published the first smart gate driver circuit made of 6H-SiC. It included overvoltage, low voltage, short circuit, and open circuit detection functions and was capable of working in an environment of 300°C [16]. Subsequently however, due to the maturing of the crystal growth technology for 4H-SiC, which has high carrier mobility, related research turned its focus to 4H-SiC.
• In 2006, Cree released the first 4H-SiC CMOS Inverter [17].
• From 2011 to 2013, England’s Raytheon Systems Ltd. (RSL) released a 4H-SiC CMOS IC process and basic logic unit successively [18-20]. This high temperature silicon carbide process is referred to as the HiTSiC process.
• In 2016, after the launch of the HiTSiC process, the University of Arkansas published the SiC MAC, NCL Counter, Boolean FSM, DAC Controller and other circuits [21].
• In 2017, the UK’s Newcastle University demonstrated mixed-signal IC applications such as the 555 timer and the 4:1 multiplexer [22]. More recently, the University of Arkansas published a more complex, digitally-controlled PWM Generator circuit [23]. In the same year, Hitachi released a SiC CMOS operational amplifier [24], which demonstrated the SiC CMOS’ ability to withstand radiation using the most basic MOSFET structure.
• In January of 2021, Professor Chih-Fang Huang of Tsing Hua University’s Semiconductor Manufacturing & Design for AI Edge Moonshot Project took the lead by publishing a single-chip integration of an 800 V lateral LDMOSFET and a low-voltage CMOS in the IEEE Electron Device Letters, as shown in Figure 12 [25]. Japan’s AIST published a single-chip integration of a 1200 V-class trench gate MOSFET and a CMOS gate driver circuit at the ISPSD Symposium in the same year. They were the first to integrate a low-voltage CMOS with a vertical-structure power component, as shown in Figure 13 [26].

Figure 12. Research Team Led by Professor Chih-Fang Huang of Tsing Hua University’s Semiconductor Manufacturing & Design for AI Edge Moonshot Project Presented an 800 V Lateral MOSFET and CMOS Integration in the IEEE EDL [26]

Figure 13. AIST’s Single-Chip Integration of a 1200 V-class Trench Gate MOSFET and CMOS Gate Driver Circuit, Presented at the 2021 ISPSD Symposium [24]

With the support of the Semiconductor Manufacturing & Design for AI Edge Moonshot Project, this laboratory developed the local oxidation isolation process, the double gate oxide thickness process, the P-type polysilicon gate process, the junction isolation process for high and low voltage components, and other key technologies and successfully completed a single-chip integration of a 10 V CMOS + 20 V Gate driver +60 V VDMOSFET in 2021. The results will be presented in VLSI-TSA and ISPSD workshops in 2022 [27, 28]. Figure 14 is a schematic diagram of the cross-sectional structure. Even before the ISPSD, these results indicate that the single-chip integration of CMOS driver circuits and power MOSFETs is going to be a future trend. The quest for component performance optimization and more integrated process technology will continue to drive innovation.

Figure 14. Schematic Diagram of the Cross-Sectional Structure of the Single-Chip Integration of a CMOS Driver Circuit and a VDMOSFET Proposed by the Moonshot Project

In summary, the application of SiC components in high power applications is an inevitable trend. As the technology matures and wafer sizes increase, component costs will continue to decline, resulting in improved cost performance. This will in turn lead to a rapid increase in product penetration. As for the development of SiC ICs, it is still in the early stages of research, but it is already showing promise. The key is to improve performance and reduce costs. It may be three to five years before it has practical value. To that end, it will require active input from our research institutions.

References:

[1] M. E. Levinshtein, S. L. Rumyantsev, and M. S. Shur, Eds., “Properties of Advanced Semiconductor Materials GaN, AlN, InN, BN, SiC, SiGe“ New York: Wiley, 2001, pp.31-47.

[2] A. Bolotnikov et al., "Overview of 1.2 kV – 2.2 kV SiC MOSFETs targeted for industrial power conversion applications," 2015 IEEE Applied Power Electronics Conference and Exposition (APEC), 2015, pp. 2445-2452.

[3] F. Roccaforte, G. Greco and P. Fiorenza, "Processing Issues in SiC and GaN Power Devices Technology: The Cases of 4H-SiC Planar MOSFET and Recessed Hybrid GaN MISHEMT," 2018 International Semiconductor Conference (CAS), 2018, pp. 7-16.

[4] T. Kimoto, “Material science and device physics in SiC technology for high-voltage power devices,” Japaness J. Appl. Phys., vol.54, p.040103, 2015

[5] T. Nakamura, Y. Nakano, M. Aketa, R. Nakamura, S. Mitani, H. Sakairi, and Y. Yokotsuji, “High performance SiC trench devices with ultra-low ron,” in Proc. IEEE Int. Electron Devices Meeting (IEDM), 2011, pp.599-601.

[6] R. Siemieniec et al., "A SiC Trench MOSFET concept offering improved channel mobility and high reliability," 2017 19th European Conference on Power Electronics and Applications (EPE'17 ECCE Europe), 2017, pp. P.1-P.13

[7] T. Kojima, S. Harada, Y. Kobayashi, M. Sometani, K. Ariyoshi, J. Senzaki, M. Takei, Y. Tanaka, and H. Okumura, “Self-aligned formation of the trench bottom shielding region in 4H-SiC trench gate MOSFET,” J. J. Appl. Phys., vol.55, 04ER02, 2016.

[8] H. Takaya, J. Morimoto, K. Hamada, T. Yamamoto, J. Sakakibara, “A 4H-SiC Trench MOSFET with Thick Bottom Oxide for Improving Characteristics, “in Proc. of Int. Symp. on Power Semi. Dev., 2013, pp.43-46.

[9] X. B. Chen US Patent 5216275.

[10] B. J. Baliga, Advanced Power MOSFET Concept, Springer, 2010

[11] M. Baba, T. Tawara, T. Morimoto, S. Harada, M. Takei and H. Kimura, "Ultra-Low Specific on-Resistance Achieved in 3.3 kV-Class SiC Superjunction MOSFET," 2021 33rd International Symposium on Power Semiconductor Devices and ICs (ISPSD), 2021, pp. 83-86.

[12] W. Xie, J. A. Cooper, and M. R. Melloch, “Monolithic NMOS digital integrated circuits in 6H-SiC,” IEEE Electron Device Lett., vol. 15, no. 11, pp. 455–457, Nov. 1994.

[13] D. M. Brown et al., “High temperature silicon carbide planar IC technology and first monolithic SiC operational amplifier IC,” in Proc. 2nd Int. High-Temperature Electron. Conf., 1994, pp. XI-17–XI-22.

[14] D. B. Slater et al., “Demonstration of 6H-SiC CMOStechnology,” in Proc. 3rd Int. High-Temperature Electron. Conf., 1996, pp. XVI-27–XVI-32.

[15] S. Ryu and K. T. Kornegay, “Design and fabrication of depletion load NMOS integrated circuits in 6H-SiC,” in Proc. Inst. Phys. Conf., London, U.K., 1996, pp. 789–792.

[16] J.-S. Chen, K. T. Kornegay, and S.-H. Ryu, “A silicon carbide CMOS intelligent gate driver circuit with stable operation over a wide temperature range,” IEEE J. Solid-State Circuits, vol. 34, no. 2, pp. 192–204, Feb. 1999.

[17] Brett A. Hull, Sei-Hyung Ryu, Husna Fatima, Jim Richmond, John W. Palmour, and James Scofield, “Development of A 4H-SiC CMOS Inverter,” Mater. Res. Soc. Symp. Proc. Vol. 911, B13-02, 2006.

[18] D. T. Clark, et al., “High Temperature Silicon Carbide CMOS Integrated Circuits,” Materials Science Forum, vol. 679-680, pp. 726-729, Mar 2011.

[19] E. P. Ramsay et al., “Digital and analogue integrated circuits in silicon carbide for high temperature operation,” in Proc. IMAPS High Temperature Electron. Conf., Albuquerque, NM, USA, 2012, pp. 373–377.

[20] R. A. R. Young, et al., “High Temperature Digital and Analogue Integrated Circuits in Slicion Carbide,” Materials Science Forum, vol. 740-742, pp. 1065-1068, Jan 2013.

[21] M. H. Weng, D. T. Clark, S. N. Wright, D. L. Gordon, M. A. Duncan, S. J. Kirkham, M. I. Idris, H. K. Chan, R. A. R. Young, E. P. Ramsay, N. G. Wright, and A. B. Horsfall, “Recent advance in high manufacturing readiness level and high temperature CMOS mixed-signal integrated circuits on silicon carbide,” Semicond. Sci. Technol. 32 (2017) 054003.

[22] N. Kuhns, L. Caley, A. Rahman, S. Ahmed, J. Di, H. A. Mantooth, A. M. Francis, and J. Holmes, “Complex High-Temperature CMOS Silicon Carbide Digital Circuit Designs,” IEEE Trans. on Devices and Materials Reliability, vol.16, no.2, 2016, pp.105-111.

[23] S. Roy, R. C. Murphree, A. Abbasi, A. Rahman, S. Ahmed, J. A. Gattis, A. Matt Francis, J. Holmes, H. A. Mantooth, and J Di, “A SiC CMOS Digitally Controlled PWM Generator for High-Temperature Applications,” IEEE Trans. on Industrial Electronics, vol.64, no.10, 2017, pp.8364-8372.

[24] M. Masunaga , S. Sato , R. Kuwana, N. Sugii, and A. Shima, “4H-SiC CMOS Transimpedance Amplifier of Gamma-Irradiation Resistance Over 1 MGy,” IEEE Transactions on Electron Devices, vol. 67, no. 1, pp. 224-229, Jan. 2020.

[25] J.-Y. Jiang, J. -C. Hung, K. -M. Lo, C. -F. Huang, K. -Y. Lee and B. -Y. Tsui, “Demonstration of CMOS Integration with High-voltage DMOS in 4H-SiC,” IEEE Electron Device Lett., vol.42, no.1, pp.78-81, 2021.

[26] M. Okamoto, A. Yao, H. Sato, and S. Harada , "First Demonstration of a Monolithic SiC Power IC Integrating a Vertical MOSFET with a CMOS Gate Buffer," in Proc. 33th Int. Symp. Power Semiconductor Devices ICs (ISPSD), Jun. 2021, pp. 71-74.

[27] B. Y. Tsui, C. L. Hung, T. K. Tsai, L. J. Lin, T. W. Wang, and P. H. Chen, “Dual Gate Oxide CMOS Process on 4H-SiC,” to be presented in the 2022 International Symposium on VLSI Technology, Systems and Applications (VLSI-TSA), 2022.

[28] B. Y. Tsui, C. L. Hung, T. K. Tsai, Y. C. Tsui, T. W. Wang, Y. X. Wen, C. P. Shih, J. C. Wang, L. J. Lin, C. H. Wang, K. W. Chu, and P. H. Chen, “First Integration of 10 V CMOS Logic Circuit, 20 V Gate Driver, and 600 V VDMOSFET on a 4H-SiC Single Chip,” to be presented in the 34th IEEE International Symposium on Power Semiconductor Devices and ICs (ISPSD), 2022.

Compared to traditional silicon-based semiconductor components, SiC and GaN-based third-generation semiconductors are not only able to withstand higher power levels, frequencies of use and ambient temperatures but also have good heat dissipation characteristics. Therefore, it has definite technical application advantages in a variety of industrial areas ranging from 5G-base stations, electric vehicles, and low-orbit satellites to solar energy sources and various high value Industry 4.0 etc. In the past, limited by the high costs and difficulty of wafer fabrication, third-generation semiconductors were used only in national defense, aerospace and other such fields. It was only in more recent years that technological advances have significantly reduced the production costs, enabling these components to become widely used in the industrial, automotive, and consumer electronics industries.

At present, the top three applications of silicon carbide semiconductor components are electric vehicles at 61%, solar power generation and energy storage at 13%, and charging facilities at 9%. Compared to the original silicon-based power components, SiC components can provide more efficient electronic conversions and are better at saving energy. TrendForce market research predicted that, from 2020 to 2025, the silicon carbide component market will grow from 6.8 hundred million USD to 33.9 hundred million USD, with a compound annual growth rate (CAGR) of as high as 38%. Even though, at present, the silicon carbide semiconductor market is still small, the growing global demands for electric vehicles and energy applications are sure to create new opportunities for the development of the semiconductor field in the near future.

The research team led by Professor Bing-Yue Tsui of Yang Ming Jiaotong University is among the most well-known in the country when it comes to third-generation semiconductor research. The team has been an active participant in the “Ministry of Science and Technology’s Moonshot Project – Silicon Carbide Single-Chip Power System Platform” initiative since 2018. In this time, in addition to completing numerous innovative research studies, they have actively cooperated with the industry on the practical applications of R&D results. Their work has contributed greatly to enhancing Taiwan’s competitive advantages in the field of third-generation semiconductors. This year, MA-tek is honored to be able to work with Professor Tsui on industry-university collaborations by providing all the analysis services needed for the research of silicon carbide power devices and the related integrated modules. With MA-tek’s comprehensive testing equipment and professional testing experience, we can fully meet the various testing and analysis needs of compound semiconductors’ process, packaging and failure analysis.