The low frequency noise characteristics of HgCdTe photoconductors are very important to the weather satellite user community because of the extremely long integration time used in these sensor applications. There are several detector technologies critical to the reduction of 1/f noise including surface passivation, bulk material selection, defect-free wafer thinning, contact metallurgy, off-HgCdTe bonding,and planar/low pressure substrate mounting. Each of these technical building blocks is discussed. When integrated to form a combined process, these critical technologies lead to a significant improvement in 1/f noise. Tow widely accepted empirical models of 1/f noise are reviewed. Experimental results validate Kruse's model but repudiate Broudy's model, namely, 1/f noise is inversely proportional to the square root of the detector volume but does not depend on the gr noise of the detector. Furthermore, we find no correlations between 1/f noise and total detector surface area or the detector contact effects. A novel test structure is presented which suggests that by using innovative detector geometries, the detector designers may be able to increase D* at low frequencies compared to the conventional square or rectangular detector structures.

Date Published: 18 October 1996
PDF: 12 pages
Proc. SPIE 2812, GOES-8 and Beyond, (18 October 1996); doi: 10.1117/12.254098

Published in SPIE Proceedings Vol. 2812:
GOES-8 and Beyond
Edward R. Washwell, Editor(s)