As system designers continue to push for higher voltages, greater power density, faster switching speeds, and increased operating frequencies, even today’s leading WBG materials are beginning to encounter performance limitations. This is driving growing interest in ultra-wide bandgap (UWBG) semiconductor platforms, particularly those based on aluminum nitride (AlN), which offers a unique combination of high breakdown strength, excellent thermal conductivity, and strong carrier confinement capabilities.
Within the UWBG landscape, two leading approaches have emerged: GaN-on-AlN heterostructures and fully homoepitaxial AlN-on-AlN devices. Both promise substantial performance advantages over existing technologies, but they represent different stages of technological maturity and present distinct tradeoffs between near-term commercial viability and long-term performance potential.
GaN-on-AlN: Near-Term Performance Gains
Today, GaN-on-AlN possesses a practical path towards commercial adoption, particularly for RF electronics. The materials have a closer structural and thermal matching than incumbent technologies, where performance advantage can be realized by combining the excellent electron transport properties of GaN with the thermal and electrical advantages of joining with AlN.
Recent demonstrations have highlighted several compelling advantages that result in RF device performance comparable to or exceeding state-of-the-art GaN-on-SiC and GaN-on-Si HEMTs.
- Higher breakdown voltage due to the AlN barrier layers
- Improved thermal management when joining with AlN
- Reduced defect density when grown on bulk AlN substrates
One reported AlN-based XHEMT architecture demonstrated nearly 6 W/mm output power with 65% power-added efficiency at 10 GHz, underscoring the potential benefit of the GaN-on-AlN approach for next-generation RF systems. Researchers continue to focus heavily on polarization engineering, defect reduction, and field engineering to fully realize the technology’s potential.
AlN-on-AlN: The Long-Term Vision
While GaN-on-AlN leverages the strengths of both materials, many researchers view fully homoepitaxial AlN-on-AlN devices as the ultimate UWBG platform.
The appeal is obvious. AlN possesses one of the largest bandgaps among commercially relevant semiconductors, with a predicted breakdown field approaching 15 MV/cm. In theory, AlN transistors could operate at much higher voltages while maintaining lower losses than today’s GaN and SiC devices.
Additional advantages for advanced power conversion and defense electronics include:
- High thermal conductivity
- Exceptional voltage handling capability
- Potential for operation in extreme radiation and temperature environments
Despite these compelling attributes, substantial technical hurdles remain before AlN-on-AlN devices can become a practical commercial technology. Unlike GaN, AlN is notoriously difficult to dope effectively, and achieving high carrier concentrations, low resistance ohmic contacts, and sufficient channel conductivity remains one of the largest obstacles to transistor development. Electron mobility in AlN is also generally lower than the highly optimized transport achieved in GaN-based channels. Recent reports are starting to change this paradigm, with significant bulk doping for both p- and n-type AlN achieving carrier concentrations greater than the mid-1018 cm-3 range
Prototype AlN Schottky diodes, MESFETs, and MOSFETs have recently been demonstrated, but it remains clear that AlN-on-AlN technology remains at a much earlier stage of development than GaN-on-AlN platforms.
Which Approach Will Win?
The answer may not be either-or. Over the next decade, GaN-on-AlN is likely to emerge in high-performance RF and power electronics, benefiting from established GaN device physics while exploiting the thermal and electrical advantages of bulk AlN substrates. The availability of larger-diameter single-crystal AlN substrates is accelerating this transition.
Longer term, however, AlN-on-AlN devices may represent the end state of UWBG electronics. If researchers can solve the persistent challenges surrounding doping, contacts, and conductivity, fully AlN-based devices could ultimately surpass GaN and SiC in voltage capability, reliability, and operation under extreme conditions. The parallel development paths of the other UWBG materials, such as diamond and GaO, may present unique opportunities in heterogenous integration and combining material benefits within single systems.
For now, GaN-on-AlN offers the best combination of performance and manufacturability. AlN-on-AlN remains the technology with the highest theoretical ceiling, with many of the technological hurdles being attacked today. The future of UWBG semiconductors will likely be defined by how quickly the industry can bridge the gap between today’s practical GaN-enhanced architectures and tomorrow’s fully realized AlN electronics.
Additional Reading
E. Kim, Y.-H. Chen, N. Pieczulewski, et al. “ XHEMTs on Ultrawide Bandgap Single-Crystal AlN Substrates.” Adv. Electron. Mater. 12, no. 3 (2026): e00393. https://doi.org/10.1002/aelm.202500393
Habib Ahmad, Sangho Lee, Minwoo Cho, et al. “New advances in semiconducting AlN and prospects for future optoelectronic devices”, Proc. SPIE 13913, Light-Emitting Devices, Materials, and Applications XXX, 1391306 (5 Mar 2026); https://doi.org/10.1117/12.3086306