In this letter, we demonstrate that the commonly assumed Arrhenius law is inconsistent with extrapolation of data-retention time-to-failure of nonvolatile memories in highly accelerated life-tests. We argue that the retention time, namely log(t/sub H/), varies linearly with temperature T rather than with 1/T as commonly assumed, yielding an important reduction in the extrapolated time-to-failure. Extensive experimental results demonstrate the physical consistency of the new model. In particular, data-retention of EPROM devices and leakage current of interpoly dielectric and gate oxide have been investigated over a wide range of temperatures. Finally, it is shown that our model reconciles seemingly controversial activation energy data from the literature.
No takes yet. Share an insight, caveat, or question.
Salvo et al. (1999) studied this question.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: