Defects responsible for charge carrier removal and correlation with deep level introduction in irradiated β-Ga2O3

Abstract

Carrier removal rates and electron and hole trap densities in β-Ga2O3 films grown by hydride vapor phase epitaxy (HVPE) and irradiated with 18 MeV α-particles and 20 MeV protons were measured and compared to the results of modeling. The electron removal rates for proton and α-radiation were found to be close to the theoretical production rates of vacancies, whereas the concentrations of major electron and hole traps were much lower, suggesting that the main process responsible for carrier removal is the formation of neutral complexes between vacancies and shallow donors. There is a concurrent decrease in the diffusion length of nonequilibrium charge carriers after irradiation, which correlates with the increase in density of the main electron traps E2* at Ec − (0.75–0.78) eV, E3 at Ec − (0.95–1.05) eV, and E4 at Ec − 1.2 eV. The introduction rates of these traps are similar for the 18 MeV α-particles and 20 MeV protons and are much lower than the carrier removal rates.

Hydrogen plasma treatment of β -Ga2O3: Changes in electrical properties and deep trap spectra

“…After annealing at 450 C, the diffusion length was 70 nm, and after 550 C, it was 80 nm. The donor concentration and the E2 Ã and E3 trap concentrations in the reference sample were an order of magnitude higher than in HVPE films studied by us previously, 20,24,25 which correlates with the lower L d in the reference sample compared to previously reported for HVPE b-Ga 2 O 3 [L d $ 450 nm (Refs. 20, 24, and 25)].…”

“…With these densities, such acceptors should be detectable in low frequency, high temperature C-V measurements with illumination. 25,26 The most likely candidates seem to be deep acceptors with the optical ionization threshold near 2.3 eV and 3.4 eV. 25,26,29 Their presence in very high concentration is not confirmed in our high temperature/low frequency C-V measurements with illumination.…”

“…25,26 The most likely candidates seem to be deep acceptors with the optical ionization threshold near 2.3 eV and 3.4 eV. 25,26,29 Their presence in very high concentration is not confirmed in our high temperature/low frequency C-V measurements with illumination. The proposed hydrogen passivation of donors would be in contrast to theoretical predictions and to some experimental results obtained for H introduction at high temperatures during growth.…”

“…DLTS spectra showed two major peaks belonging to electron traps with levels near E c -0.8 eV and E c -1.05 eV, with signatures close to those of the electron traps E2 Ã and E3. [23][24][25] Figure 2 shows the spectra obtained with a reverse bias of À5 V pulsed to 0 V for 3 s for DLTS time windows t 1 /t 2 ¼ 1.75 s/17.5 s. The y-axis in the figure represents the DLTS signal DC/C multiplied by 2 N d and divided by the DLTS spectrometer function F ¼ exp(Àt 2 /s) À exp(Àt 1 /s), with s ¼ (t 2 À t 1 )/ln (t 2 /t 1 ). 18 The convention is that positive peaks correspond to electron traps (capacitance increases during the capacitance transient 18 ).…”

“…Proton irradiation increases the density of the E2 and E3 electron traps, 16,25 and so it cannot be ruled out that these centers can contain hydrogen as a constituent. However, the steady increase in the centers' density with proton fluence suggests involvement of native point defects.…”