Abstract
This paper examines the current use of the Arrhenius equation and the activation energies associated with integrated circuits. A short introduction to the Arrhenius equation is given and it is shown how activation energies may be computed from experimental data.The paper next addresses the following basic question: is it reasonable to extrapolate from high‐temperature laboratory life‐tests down to use conditions using the Arrhenius equation? The paper shows that the answer to this question in general must be an emphatic no. It is not the validity of the Arrhenius equation as such that is questioned, but its indiscriminate use by reliability practitioners across the electronics industry.This is not a scientific paper, and no ‘proofs’ are given. The discussion is based on ‘typical’ integrated circuits using published data on the temperature dependence of the hazard rate of such circuits. The paper serves as a warning against thinking that ‘well established practices’ are necessarily scientifically sound.

This publication has 1 reference indexed in Scilit: