Power SiC: the cornerstone of ROHM’s business expansion in the next five years – An interview with ROHM

The SiC device market is projected to reach nearly $10 billion by 2029, with a CAGR of 24% from 2023 to 2029, as announced in Yole Group’s annual Power SiC report. The primary driver of this growth is the automotive and mobility sector, particularly BEVs. In 2023, this sector accounted for the largest share of the market, and it is expected to grow even further by 2029. As of 2024, 400V BEVs, such as those produced by Tesla, represent the highest demand for SiC. The introduction of more 800V BEVs by OEMs is accelerating this momentum. Additionally, industrial applications, especially in the energy sector, are emerging as another crucial area for SiC growth.

Yole Group conducts in-depth research into power SiC technologies, covering aspects such as emerging technologies, market trends, supply chain dynamics, and cost-performance analysis. Their latest reports include Power SiC 2024 – Manufacturing, SiC Modules Performance Analysis 2024 Vol 1 and SiC MOSFET Discretes Performance Comparison Analysis 2024 Vol 1, which features an evaluation of ROHM’s SiC MOSFET, SCT3080KLHR.

Indeed, ROHM has unveiled its ambitious plans to significantly increase its power SiC revenue over the coming years by expanding production capacity, transitioning to an 8-inch platform, and launching new generations of power SiC devices and modules.

Yole Group’s analysts had the opportunity to speak with Akifumi Enomoto, Department Manager, Power Devices Business Strategy at ROHM. In this interview, Amine Allouche and Poshun Chiu, Senior Technology & Market Analysts for Compound Semiconductors at Yole Group, delve into ROHM’s business strategies, technological innovations, and product roadmap for the rapidly expanding Power SiC market.

Grab a coffee and dive into this insightful conversation.

Poshun Chiu (PC): In Yole Group’s recent report Power SiC 2024, the revenue forecast for 2029 is nearly $10 billion, with a 24% CAGR. And ROHM has implemented capacity expansion and 8″ capability build-up. Can you comment on your strategies for growing SiC revenue for the ROHM group?

EA: Of course. It is essential that we increase capacity to meet market requirements. Manufacturing on 8-inch substrates is a crucial and essential technical issue from the standpoint of enhancing productivity. ROHM has started production of 8-inch substrates, with the first units scheduled for delivery in 2023. We anticipate the release of devices and other products for mass production and shipment from 2025 onwards. Regarding facility expansion, we are preparing to commence operations at our second plant in Miyazaki, Japan. This plant has a total building floor area of approximately 230,000 m² and is located in Kunitomi-cho, Miyazaki Prefecture. This facility, previously used as a plant by another company, includes a clean room. This allows us to start up quickly and meet the growth of the fast-growing SiC market. In December 2023, the plant received approval from Japan’s Ministry of Economy, Trade, and Industry (METI) under its “Plan for Securing Power Semiconductor Supply.” Consequently, we have been granted a subsidy of approximately one-third of the investment. We are undertaking this expansion with the opening of the new Miyazaki second plant. Currently, the ROHM Group’s SiC process lines are located in Miyazaki and Chikugo. However, following the opening of the Miyazaki second plant in 2023, we anticipate a significant increase in production capacity, enabling us to meet demand up to 2030.

PC: Automotive is the major driver for the Power SiC market, accounting for more than 70% of the market. And we are aware of ROHM’s multiple design-win announcements with multiple OEMs and tier-1s. Can you comment on ROHM’s position in this industry and the outlook?

EA: Yes, as you have noted, automotive is the largest segment of the SiC market. The automotive application that utilizes the greatest quantity of SiC is the traction inverter for xEVs, which drives the motor. A single traction inverter contains a number of SiC chips. In addition to supplying the SiC bare chips, ROHM is expanding its product range to include modules. Regarding SiC modules, in addition to those produced by ROHM itself, they are also available from HAIMOSIC, a SiC module manufacturer, through a joint venture of ROHM with the Zhenghai Group in China. There are a number of design considerations when using SiC chips in a traction inverter. ROHM is focused on providing application-level support, including equipping its own motor testing facilities to address these issues. Furthermore, our company has a team of application field engineers stationed not only at our headquarters but also in Europe, the Americas, and Asia, enabling us to respond quickly to customer needs. We are also focusing on discrete SiC MOSFETs and SiC SBDs for onboard chargers and DC-DC converters and can provide devices in a variety of forms.

As a result of comprehensive evaluations by leading manufacturers of our cutting-edge technology and our business flexibility and stable supply system, we have already secured design wins at more than 130 companies worldwide. For example, in Europe, Vitesco (a member of the Schaeffler Group); in the Americas, Lucid Motor; in China, ZEEKR (a member of the Geely Group); and in Japan, Mazda. These are just a few examples, but we are pleased to report that adoption and collaboration are progressing in a well-balanced manner across all regions.

Amine Allouche (AA): Following the questions about automotive, we are aware of the newly released 2-in-1 TRCDRIVE pack™ integrating ROHM’s 4th Generation SiC MOSFET. Can you comment on the advantages of the power module and the MOSFET and any differentiators that should be highlighted?

EA: The key requirement for traction inverters is a power module that is both compact and capable of handling high currents, which necessitates high power density. The TRCDRIVE pack™ features a 2-in-1 configuration, and its multiple SiC chips allow for higher current throughput than discrete products. As traction inverters must handle large currents of 300A or more, using high-current-density SiC modules enables the design of compact traction inverters. ROHM’s TRCDRIVE pack™ offers industry-leading SiC module power density thanks to its unique structure, which separates the primary current and control signals, and an optimized internal layout. Specifically, it achieves approximately 1.5 times the power density of existing SiC molded modules. Furthermore, the separated control pins, exiting at the top of the mold, simplify connection to the gate driver substrate, reducing design work.

PC: Industrial applications, such as EV chargers, PV inverters, converters, and power supplies are important applications for Power SiC as of today. We also see heat-pump and motor drive for Power SiC to gain more penetration. What are ROHM’s views on industrial applications in terms of market size and the added value of ROHM’s SiC devices?

EA: PV inverters account for a significant portion of the SiC demand in industrial applications. Increasingly, PV inverters raise the operating voltage of DC systems to 1500V to improve conversion efficiency and reduce transmission losses. ROHM has developed a 2kV SiC MOSFET specifically targeting these 1500V DC systems. In addition, ROHM has already lined up SiC SBD products ranging from 650V to 1,700V. These products, available as either discrete devices or bare chips, are being adopted for the PFC sections of EV chargers, uninterruptible power supplies (UPS), and PV inverters. In the automotive field, ROHM is also accelerating its business model of providing SiC bare chips to module manufacturers. For example, we have concluded a partnership with Semikron regarding SiC, and their eMPack® product uses ROHM’s SiC MOSFET bare chips. We are also strengthening our collaboration with Semikron in the industrial equipment field and have already started providing SiC and IGBT bare chips.

AA: With SiC devices increasing globally and becoming more cost-efficient, we also see other applications showing interest in SiC devices. What applications could ROHM target after automotive and the current industrial applications? For example, high-voltage applications, 3.3kV, 6.5kV, or above.

EA: We anticipate that the adoption of SiC will continue to gain momentum in high-voltage industrial applications in the future. Specifically, the targets will include PV inverters, energy storage systems, pulse power supplies, and solid-state transformers, as previously mentioned. We have developed 2kV SiC MOSFETs for 1,500VDC PV inverter systems, and ROHM has been engaged in the research and development of SiC technology for 3.3kV and higher for an extended period. Our current strategy is to expand our product range further to include 2kV-rated components for PV inverters.

AA: We’d like to understand more about the product roadmap for SiC devices and bare dies/discretes/power modules. Is ROHM expecting any technological breakthroughs in the next few years? (e.g., we are aware of the 6-in-1 molded module with heat sink presented by ROHM), or new device technology in development, etc.? Any specific challenges that need to be highlighted?

EA: As previously stated, one of the key advantages of ROHM’s SiC business model is its flexibility, allowing it to accommodate various delivery formats, including bare chips, discrete devices, and modules. We are currently developing a 5^th generation SiC MOSFET, with an anticipated release in 2025. This new product is expected to achieve a 30% reduction in on-resistance at high temperatures compared to the fourth generation. Furthermore, we are developing next-generation products, including the sixth and seventh generations, with support from the Japanese government: these next-generation products are scheduled for release in 2027 and 2029, respectively. Typically, it takes four to six years for a generation to evolve; however, by releasing products every two years, we anticipate being able to respond to market changes and demands with greater speed.

I would, however, like to discuss the development of SiC molded-type modules, including the TRCDRIVE pack™. In addition to further improving power density, our objective is to achieve a breakthrough that enhances heat dissipation and reduces on-resistance, both of which are often traded off with power density. Specifically, we are developing high-performance resins to support next-generation bonding materials with excellent heat dissipation properties, such as Ag sinter. Additionally, we are planning to produce SiC molded-type modules with a current capacity of up to 900 Ams by incorporating the 5^th generation SiC bare chip mentioned earlier. Furthermore, we plan to expand the lineup of molded SiC modules, including the introduction of DIP-type (HSDIP) modules and DOT-247 package modules.

AA: Following the challenges in SiC, cost competitiveness is also driving penetration in the electrification of various systems. Does ROHM have any strategies and innovations for increasing SiC’s cost competitiveness? In terms of yield, how do you consider the current level and the future 8″ platform?

EA: Given that SiC is entering a period of widespread adoption, we are acutely aware of the importance of cost competitiveness. We are confident that we can strengthen cost competitiveness by three elements – 1) significantly improving yield rates through the integration of substrate technology, epitaxial device processing technology, and device design technology 2) developing next-generation devices 3) enhancing productivity with the use of 8-inch wafers. As a result, the electrification of systems will accelerate across a broader range of markets.
We maintain a cohesive development and production system encompassing everything from substrates to device modules. Additionally, we possess the expertise and technology to ensure stable supply by increasing the diameter of our wafers from 3 inches to the current 6 inches. In fact, we were the first in the world to successfully mass-produce SiC MOSFETs in 2010.

Source: www.rohm.com