Noise Sources of a-Si[sub 1−x]Ge[sub x]O[sub y] Microbolometers And Their Reduction By Forming Gas Passivation

Abstract

Forming gas annealing was performed on Sii.xGe^Oy microbolometers in order to reduce the electrical 1//^noise. The passivation time and temperature were optimized to obtain least possible noise voltage power spectral density (PSD). To observe the effect of forming gas annealing at different intervals of time on the noise voltage PSD of the devices, the noise voltage PSD was measured before and at the end of each interval of passivation time by tracking individual devices. The microbolometers fabricated from Sii.xGe^Oy were placed inside a rapid thermal annealing chamber in the presence of forming gas at 250 °C with different intervals of time starting from 0.5 hour to 8 hours. The value of flicker noise coefficient . STj, or an/N from the Hooge model was used as a measure of electrical l//^noise. The l//^noise is believed to originate from the defects and dangling bonds in the sensing layer. The hydrogen passivation reduces the electrical l//^noise by repairing the dangling bonds and defects. Passivation done for excess time increased the electrical l//^noise of the device slightly, indicating that the dangling bonds contribute to trap states and scattering centers for the carriers.

Noise Reduction of Amorphous SixGeyO1–x–yThin Films for Uncooled Microbolometers by Si3N4Passivation and Annealing in Vacuum

“…The measured 1/f -noise at 70 Hz was 1×10 −15 V 2 /Hz using 0.3 μA. The calculated 1/f noise coefficient K f was 2.21×10 −10 [51].…”

Gas flow simulation of a fluxless Si solder bonding oven

“…Forming gas is used as an atmosphere for processes that need the properties of hydrogen gas without the explosion hazard. The Forming gas is also widely used in silicon and thin film technology for passivation [6].…”