Deep Energy Level Detection in Nanoscale Devices by Cross Spectrum Noise Spectroscopy

D. Sharma, S. Krylyuk, A. Motayed, Q. Li, A.V. Davydov
George Mason University, US

Keywords: silicon, nanowire, defects, noise, deep level, traps

Summary:

It is well recognized that deep energy levels created due to impurity atoms may detrimentally affect the performance of semiconductor devices. However, application of traditional methods to study deep levels, such as deep level transient spectroscopy, photo-induced current transient spectroscopy or low frequency noise (LFN) spectroscopy, often becomes impractical for nanoscale devices. The accurate measurements of the noise signal in low-current nanoscale devices such as filed effect transistors(FETs) are extremely challenging because the device noise, which is proportional to the dc current, becomes comparable with the instrumental noise of the measurement setup. To overcome this issue, we have implemented a LFN measurement method based on dual-channel cross-spectrum analysis technique.We fabricated four sets of FETs using p-type and n-type doped Si nanowires (SiNWs). Noise measurements were performed in 80 K to 320 K temperature range.Temperature-dependent noise analysis indicated that the traps responsible for the generation and recombination (G-R) noise correlate well with the deep-levels introduced by Ni and Au atoms diffused in SiNWs during the growth.This study demonstrates that the optimized noise measurement technique can be very useful in estimating the quality of semiconductor nanoscale devices.