Limitations on Activation of High Dose Ge implant in beta-Ga2O3

TL;DR

This study investigates the activation limitations of high-dose germanium implantation in beta-Ga2O3, revealing that Ge clustering during annealing reduces electrical activation and impacts device performance.

Contribution

It provides new insights into Ge clustering behavior and its effect on activation efficiency in beta-Ga2O3, highlighting the role of phase formation and annealing conditions.

Findings

Ge activation reaches up to 40% in low damage implants

High damage implants show reduced activation to 23%

Ge clustering and precipitate formation limit activation at high concentrations

Abstract

Among ultrawide bandgap semiconductors, beta-Ga2O3 is particularly promising for high power and frequency applications. For devices, n-type concentrations above 10^19 cm^-3 are required. Ge is a promising alternative n-type dopant with an ionic radius similar to Ga. Homoepitaxial 010 beta-Ga2O3 films were implanted with Ge to form 50 and 100 nm box concentration of 3*10^19 cm^-3 and 5*10^19 cm^-3, with damage ranging from 1.2 to 2.0 displacement per atom. For lower damage implants, optimized anneals in ultrahigh purity N2 at 950-1000 C for 5-10 minutes resulted in Rs of 600-700 omega/sqr, mobilities of 60-70 cm^2/Vs, and Ge activation of up to 40%. For higher damage implants, activation dropped to 23% with similar mobilities. Ge diffusion, measured by second ion mass spectrometry, showed formation of a Ge "clustering peak" with a concentration exceeding the initial implant following anneals in N2 or O2 at 950-1000 C. Beyond this peak, minimal Ge diffusion occurred for N2 anneals at 950 C, but at 1050 C non-Fickian diffusion extended to >200 nm. Electrical activation data suggests that clustered Ge is electrically inactive. To understand Ge clustering, several samples were characterized by synchrotron x-ray diffraction. Second-phase precipitates were observed in as-implanted samples which then fully dissoved after furnace annealing in N2 at 1050 C. Diffraction peaks suggest these implant-induced precipitates may be related to a high pressure Pa-3 phase of GeO2, and may evolve during anneals to explain the Ge clustering. Ultimately, we believe Ge clustering limits activation of implanted Ge at high concentrations.