Lucky-electron model of channel hot electron emission

Abstract

The lucky electron model proposes that an electron is emitted into SiO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> by first gaining enough energy without suffering an energy stripping collision in the channel and then being redirected toward the Si/SiO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> interfact. A closed-form expression for the gate current has successfully reproduced the dependence on V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">g</inf> ,V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</inf> and L. This model also suggests a straightforward correlation with the substrate current. By measuring the substrate current one may be able to replace or supplement long-term stress test for studying a component of channel hot electron emission that causes instability but does not show up in the gate current. The lucky electron model suggests that the channel hot electron effect would be negligible if V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">d</inf> -V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</inf> is less than 2.5 volt no matter how small the channel length or junction depth.