Study of a Solar-Blind Photodetector Based on an IZTO/β-Ga2O3/ITO Schottky Diode

An InZnSnO2 (IZTO)/β-Ga2O3 solar blind Schottky barrier diode photodetector (PhD) exposed to 255 nm, 385 nm and 500 nm light wavelengths was simulated and compared with measurement. The measured dark photocurrent at reverse bias and responsivity were successfully reproduced by numerical simulation by considering several factors such as conduction mechanisms and material parameters. Further optimizations based on reducing trap densities and insertion of a 50-nm Al0.39Ga0.612O3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\left({\mathrm{Al}}_{0.39}{\mathrm{Ga}}_{0.61}\right)}_{2}{\mathrm{O}}_{3}$$\end{document} passivation layer between IZTO and β-Ga2O3 are carried out. The effect of reducing bulk traps densities on the photocurrent, responsivity and time-dependent photoresponse (persistent conductivity) were studied. With decreasing traps densities, the photocurrent increased. Responsivity reached 0.04 A/W for low β-Ga2O3 trap densities. The decay time estimated for the lowest ET(0.74,1.04eV)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${E}_{{T}}\; (0.74, 1.04\; \mathrm{eV})$$\end{document} densities is ∼0.05s\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sim 0.05\; \mathrm{s}$$\end{document} and is shorter at ∼0.015s\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sim 0.015\; \mathrm{s}$$\end{document} for ET(0.55eV)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${E}_{{T}}\; (0.55\; \mathrm{eV})$$\end{document}. This indicates that the shallowest traps had the dominant influence (ET=0.55eV\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${E}_{{T}}=0.55\; \mathrm{eV}$$\end{document}) on the persistent photoconductivity phenomenon. Furthermore, with decreasing trap densities, this PhD can be considered as a self-powered solar-blind photodiode (SBPhD). The insertion of a Al0.39Ga0.612O3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\left({\mathrm{Al}}_{0.39}{\mathrm{Ga}}_{0.61}\right)}_{2}{\mathrm{O}}_{3}$$\end{document} passivation layer increases the photocurrent which is related to a recombination decrease and the photogenerated carrier increase, and hence the increase of the internal quantum efficiency.