Researchers from the University of Illinois recently advanced gallium nitride (GaN)-on-silicon transistor technology. They reportedly optimized the composition of the device’s semiconductor layers. Working with industry partners IBM and Veeco, the team formed the high electron mobility transistor (HEMT) structure on a 200 mm silicon substrate using a process that they claim will scale to larger industry-standard wafer sizes. The team collaborated with Veeco and IBM to conduct their research at the University of Illinois Micro + Nanotechnology Lab with funding from the Air Force Office of Scientific Research.
￼While using silicon substrates costs less that other substrate options like sapphire and silicon carbide, silicon usage does have challenges including its lattice mismatch with GAN. “When you grow the GaN on top, there’s a lot of strain between the layers, so we grew buffer layers [between the silicon and GaN] to help change the lattice constant into the proper size,” explained ECE undergraduate lead researcher Josh Perozek.
A buffer layer can combat the lattice mismatch which can cause defects such as threading dislocations or holes that can ruin the properties of the device’s 2-dimensional electron gas channel. The HEMTs ability to conduct current and function at high frequencies requires the channel.
Bayram’s team studied three different buffer layer configurations and discovered that thicker buffer layers made of graded AlGaN reduce threading dislocation, and stacking the layers reduces stress. With this configuration, the researchers achieved an electron mobility of 1,800 cm2/V-sec.
“The less strain there is on the GaN layer, the higher the mobility will be, which ultimately corresponds to higher transistor operating frequencies,” said Hsuan-Ping Lee, an ECE graduate student researcher who leads the scaling of these devices for 5G applications.
￼According to Bayram, next, the team plans to produce fully functional high-frequency GaN HEMTs on silicon for use in high bandwidth 5G wireless data networks.
Details of their work can be found at J. Phys. D Appl. Phys. 50 (2017) 055103.