The "next next thing" in power semiconductors is already in motion. SiC and GaN are just starting to transform EVs and data centers, but a Japanese university startup is already eyeing what comes after, a material called germanium dioxide (GeO2) with properties that could leapfrog today's cutting-edge chips. With $1 million in fresh funding, the journey from lab to factory begins.

What Is Patentix?

Founded in December 2022 as a spin-off from Ritsumeikan University in Shiga Prefecture, Patentix is a deep-tech startup focused on commercializing germanium dioxide (GeO2) power semiconductors. The company was co-founded by CEO Toyosuke Ibi, a veteran of university-born semiconductor ventures including Kyoto University's FLOSFIA and the University of Tokyo's Gaianixx, and Professor Kentaro Kaneko of Ritsumeikan University.

In April 2026, Patentix announced a Series A1 (extension round) raise of approximately $1 million (150 million yen), led by Mitsubishi UFJ Capital and TMH. This brings the company's total funding since inception to roughly $8 million (1.209 billion yen).

Why GeO2? How It Stacks Up Against SiC and GaN

Power semiconductors are the electronic components that convert and control electricity, from your phone charger to EV motor controllers to power grid infrastructure. The material they're made from determines how efficiently they can handle voltage, heat, and switching speed.

The current mainstream material, silicon (Si), has a bandgap of about 1.1 eV (electron volts), think of this as the material's "toughness" against electrical breakdown. The wider the bandgap, the higher the voltage a device can handle and the less energy it wastes as heat.

Third-generation materials like SiC (silicon carbide, ~3.3 eV) and GaN (gallium nitride, ~3.4 eV) offer roughly three times silicon's bandgap, enabling more efficient and compact power devices. They're already being adopted in EVs and fast chargers.

GeO2 goes further. Rutile-type germanium dioxide (r-GeO2) has a bandgap of approximately 4.5–4.8 eV, placing it in the "ultra-wide bandgap" (UWBG) semiconductor category. In theory, this means even higher voltage tolerance and lower power losses than SiC or GaN.

Material Bandgap Generation Status
Si (Silicon) 1.1 eV 1st gen Current mainstream
SiC (Silicon Carbide) 3.3 eV 3rd gen Mass production underway
GaN (Gallium Nitride) 3.4 eV 3rd gen Expanding in high-frequency
GeO2 (Germanium Dioxide) 4.5–4.8 eV Next gen R&D stage

GeO2 also has a unique advantage: unlike gallium oxide (Ga2O3), another UWBG candidate, GeO2 can be doped to create both p-type and n-type conductivity. This bipolar controllability is essential for building complex power device structures like MOSFETs. Additionally, GeO2 films can potentially be grown on inexpensive silicon substrates, offering a path to cost competitiveness.

Milestones Achieved in Just Three Years

Since its founding, Patentix has racked up a string of world-first achievements:

  • September 2024: First confirmed n-type doping in r-GeO2
  • November 2024: First Schottky barrier diode (SBD) operation on r-GeO2 single-crystal film
  • Summer 2025: Successfully formed r-GeO2 thin films on silicon substrates; grew bulk crystals using the floating zone (FZ) method
  • Fall 2025: Achieved n-type conductivity via ion implantation, opening the door to complex device architectures
  • March 2026: World's first demonstration of MOSFET transistor operation using r-GeO2

Going from basic crystal growth to working transistors in about three years is a remarkable pace for a material this new.

Japan's Power Semiconductor Shakeup

Patentix's funding comes at a pivotal moment for Japan's power semiconductor industry. In March 2026, three major Japanese companies, Rohm, Toshiba, and Mitsubishi Electric, announced they are entering negotiations to merge their power semiconductor operations. If completed, the combined entity would become the world's second-largest power semiconductor player by market share.

This consolidation push reflects a harsh reality: Germany's Infineon dominates with roughly 23% global share, and even combining all four of Japan's major power semiconductor makers (Mitsubishi Electric, Fuji Electric, Toshiba, and Rohm) doesn't match Infineon's revenue alone. Meanwhile, Chinese competitors are rapidly scaling up SiC production with massive state backing.

In this landscape, Patentix represents a different kind of bet, not competing in today's SiC race, but staking out a position in the generation beyond. Notably, major Japanese industrial players including AISIN, DENSO, Mitsubishi Electric, and Toray all invested in Patentix's earlier Series A round in 2025, signaling that Japan's industrial establishment sees GeO2 as a strategic hedge for the future.

Applications: EVs, Renewables, and Space

Patentix is targeting applications across automotive, power infrastructure, and even space, environments where high voltage, extreme temperatures, and reliability are non-negotiable.

The demand for more efficient power conversion is only growing. As EVs and AI data centers drive global electricity consumption upward, and new power plants take decades to build, reducing conversion losses at the device level becomes a global imperative. UWBG semiconductors like GeO2 could be a key piece of that puzzle.

The company plans to ship 10mm square samples by the end of 2025 and aims for 6-inch wafer samples by 2027. With this latest funding, Patentix has explicitly declared its transition from "basic research phase" to "commercialization development phase."

The SiC era has barely begun, and here's a Japanese startup already building the technology that could succeed it. What's the conversation about next-generation power semiconductors like in your country? We'd love to hear your perspective.

References