The native and metastable defects and their joint density of states in hydrogenated amorphous silicon obtained from the improved dual beam photoconductivity method
Künye
Mehmet Güneş, Jimmy Melskens, and Arno H. M. Smets "The native and metastable defects and their joint density of states in hydrogenated amorphous silicon obtained from the improved dual beam photoconductivity method", Journal of Applied Physics 133, 125702 (2023) https://doi.org/10.1063/5.0138257Özet
In this study, undoped hydrogenated amorphous silicon (a-Si:H) thin films deposited under moderate dilution ratios of silane by radio frequency plasma-enhanced chemical vapor deposition (RF-PECVD) have been investigated using steady-state photoconductivity and improved dual beam photoconductivity (DBP) methods to identify changes in multiple gap states in annealed and light-soaked states. Four different gap states were identified in annealed state named as A, B, C, and X states. The peak energy positions of these Gaussian distributions are consistent with those recently identified by Fourier transform photocurrent spectroscopy (FTPS). After in situ light soaking, their density increases with different rates as peak energy positions and half-widths remain unaffected. The electron-occupied A and B states located below the dark Fermi level and their density and ratios in the annealed and light-soaked states correlate well with those defects detected by time-domain pulsed electron paramagnetic resonance (EPR) experiments. The A, B, and X states located closer to the middle of the bandgap anneal out at room temperature in dark and define the "fast"states. However, the C states show no sign of room temperature annealing such that they must define the "slow"states in undoped a-Si:H. The results found in this study indicate that the anisotropic disordered network is a more appropriate model than previously proposed defect models based on the continuous random network to define the nanostructure of undoped a-Si:H, where multiple defects, D0 and non-D0 defects, can be identified by using the improved DBP method.