Base current of I<sup>2</sup>L gates at low current levels

Abstract

In order to use I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L in the low power range the base current must be minimized. Two previous models have been presented to explain the observed base current values. The first model assumes that the observed base current mainly consists of electrons injected into the inactive metal and oxide covered base, whereas the second model claims that hole injection into the epi and buried layer is the main component of base current. These models have been checked experimentally only at moderate or large current levels and at fixed V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">BE</inf> values. The purpose of this paper is to investigate experimentally the importance of the different base current components over the entire low current range (0.1 nA - 1µA). Conclusions relevant to the magnitude and the temperature dependence of these different components will be given. It will be demonstrated experimentally that surface recombination in the emitter-base sidewall of the I <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> L gates is one of the most important contributions to the base current. This component also causes the degradation of the current gain at lower temperatures. Different methods to eliminate this component will be given. New findings about the influence of the deep n+ isolation in the low current range will be presented. The experimental measurements will be illustrated by computer results of one- and two-dimensional computer-aided device analysis programs.