Silicon carbide devices stop at 6.5kV not due to process limits, but because higher voltages are uneconomical for this device type.
A counterintuitive fact: while silicon carbide MOSFETs are considered the mainstream form of wide-bandgap power devices, their commercial voltage ceiling is stuck at 6.5kV. Beyond that, the industry’s chosen solution is a return to IGBTs, a device structure viewed as "legacy" in the silicon era.
Navitas Semiconductor secured a project that fits precisely into this tier. It won the ALATTIS prototype project, standing for Accelerated, Large-Area, 10 kV SiC IGBT. Funded by the US Army Research Laboratory (ARL), which is part of the US Army Combat Capabilities Development Command (DEVCOM) and supported by the Joint Experimentation and Technology Accelerator (JETX), the goal is to develop, execute, and validate an ultra-high-voltage silicon carbide manufacturing process that does not currently exist in the US. The target devices are 10kV and above silicon carbide IGBTs, accompanied by advanced PiN diode technology.
Let’s state the conclusion first.
The key information in this project is not "how much money was awarded," but "why IGBT." The advantage zone for silicon carbide MOSFETs is medium-to-low voltage: fast switching speed and low conduction loss, with the 650V to 1.2kV range already widely adopted in vehicles. When voltage rises above 6.5kV, the channel resistance of MOSFETs increases rapidly with higher voltage requirements, forcing chip area to become very large, making unit cost uneconomical. IGBTs hold their position in the high-voltage range via the conductivity modulation mechanism, at the cost of slower switching speed, which is acceptable in 10kV applications where switching frequency is inherently low.
Navitas’s entry point is its patented Trench-Assisted Planar (TAP) architecture. This architecture currently supports the GeneSiC product line, covering a voltage range from 650V to 6.5kV. The company’s accumulation in this field can be traced back far: in 2010, it produced a 6.5kV silicon carbide thyristor and won the R&D 100 award that year; in 2012, it developed 10 to 15kV PiN diodes; and in 2021, it achieved the commercial release of a 6.5kV silicon carbide MOSFET.
6.5kV
Existing commercial SiC MOSFET limit
≥10kV
Target voltage for this project
650V
GeneSiC coverage lower limit
Three voltage values define the coordinates of this product line: the GeneSiC series starts at 650V and extends up to 6.5kV, covering the mature range for silicon carbide MOSFETs; 10kV and above is the target for this project, requiring a shift in device type to IGBTs plus PiN diodes.
The project aims to solve a process challenge, not a device design issue.
A close reading of the project description reveals a deliberate choice of wording: ALATTIS is tasked with developing, executing, and validating a 'domestic manufacturing process that does not currently exist.' Device design itself is not the hardest part—the industry has been able to produce 10kV silicon carbide switching devices in the lab for twenty years. The difficulty lies in stably manufacturing large-area, ultra-high-voltage devices to achieve usable levels of yield, reliability, and consistency.
The term 'Large-Area' is key here. As voltage increases, the drift region of the device becomes thicker, requiring a larger chip area to achieve the same on-state current. Large-area silicon carbide devices impose amplified requirements on substrate quality, epitaxial uniformity, and defect density. The project roadmap describes an iterative design, manufacturing, and testing loop—precisely the approach used when a process has not yet converged.
Navitas Senior Vice President and Chief Technology Officer Siddarth Sundaresan stated that this project extends the company's technical leadership from silicon carbide MOSFETs to 10kV silicon carbide IGBTs, leveraging its unique TAP architecture combined with advanced PiN diode technology.
The real market for this voltage class is not in electric vehicles.
Applications for 10kV silicon carbide devices include solid-state transformers and flexible transmission in the power grid, rail transit and heavy-duty traction, and variable frequency drives for large industrial applications. Passenger vehicle electric drive platforms currently operate at the 800V level, and there is no near-term reason to require 10kV devices. If a mass-production passenger vehicle electric drive solution centered on 10kV silicon carbide emerges before 2028, this assessment would be invalid.
This is a different tier from the previous piece on silicon carbide.
Building on the latest developments reported on October 1: that article covered the 1200V silicon carbide MOSFET platform launched by L&T Semiconductor Technologies in India, focused on automotive fast charging. The difference between the two lies not in the voltage figures but in the stage of development—1200V is a mature range already entering productization, where competition centers on cost reduction and customer certification; the 10kV segment lacks even a mature process, and the project must solve the question of whether the production line can be built at all.
The pace of advancement for these two lines is therefore completely different. Competition in the mature voltage range relies on price and delivery, while competition in the ultra-high-voltage tier depends on process breakthroughs and supply chain controllability. The latter demands more time and capital, with a longer return cycle.
See you in the comments. If you work for a grid equipment manufacturer, would you wait for the industry to mature 10kV silicon carbide (SiC) processes, or deliver projects now using proven silicon-based IGBT solutions?