Self-Consistent Identification of Ni-Related Deep Levels in Silicon using Multi-Window DLTS
This work presents a comprehensive and self-consistent investigation of deep-level defects in nickel-doped silicon using deep level transient spectroscopy (DLTS). The study integrates capacitance–temperature characteristics, time-resolved capacitance transients (C–t), multi-window DLTS spectra, and Arrhenius analysis of carrier emission rates to establish a physically consistent description of defect states. The exponential character of the measured C–t transients over a wide temperature range confirms the dominance of a single emission process and validates the application of the single-level emission model. This enables reliable extraction of characteristic relaxation times and corresponding emission rates. The DLTS spectra reveal a dominant defect level with a maximum at Tmax ≈ 220–225 K, accompanied by an additional peak in the temperature range ≈245–255 K and a low-temperature feature near 82 K attributed to shallow traps. Multi-window DLTS analysis demonstrates the stability of peak positions under variation of rate windows, indicating that the observed features correspond to distinct defect levels rather than measurement artifacts. Arrhenius analysis based on independently obtained emission rates exhibits high linearity (R² ≈ 0.996), confirming the thermally activated nature of carrier emission. The activation energy of the dominant defect level is determined as Ea ≈ 0.203 ± 0.01 eV. The higher-temperature peak is tentatively attributed to Ni-related defect complexes, based on its slower emission kinetics and pronounced dependence on rate window selection. However, this interpretation remains indirect and requires further structural verification. The main contribution of this work is the implementation of a cross-validated methodology in which defect identification is supported simultaneously by transient analysis, spectral stability, and kinetic consistency. This integrated approach significantly reduces ambiguity in systems with overlapping defect signals and provides a robust framework for the investigation of transition-metal-related defects in silicon.