Comments on "Structure-enhanced MOSFET degradation due to hot-electron injection"

Abstract

Hsu and Grinolds recently compared channel hot-electron (CHE) stress results of conventional and "extended drain" NMOS FET's. [1]. They observe increasing degradation as the extended drain resistance increases when the drain bias is defined as that which produces a fixed substrate current. A model in which the hot-electron stress induces surface states within the extended drain region is proposed. We argue that the drain bias condition chosen for these measurements does not produce equal numbers of channel hot electrons in all devices as is claimed. Since the ratio of substrate current to source current is a measure of the mean electron energy, we claim that this ratio (and hence the mean electron energy) increases as extended drain resistance increases.