Contents 1. Introduction
A number of heavy-ion test conditions can impact DUT response and hence, the results and conclusions of the test. For example, it is possible that one may obtain different results if testing is performed in air or in vacuum. The reason is that, scattering of the beam with materials in the beam's path (e.g., air/vacuum transition foils, degraders and/or detectors, and even the DUT overlayers) can induce significant amounts of energy straggle causing a large distribution in ion energies and effective LETs in the device sensitive volume. The complete energy distribution needs to be taken into account to analyse the device response under irradiation. In the case of power MOSFETs, it has been shown that the single-event burn-out and gate rupture (SEB or SEGR) voltages may be linked to the ion specie, rather than to solely the ion LET [1], [2], [3]. The mechanisms for this “specie” effect are not well understood. In the case of SRAMs and digital ICs, the device sensitivity in the vicinity of the threshold LET has also been shown to depend on ion energy [4], [5], [6]. Past work has shown that high LET secondaries created by nuclear interactions can have an increasingly larger impact on the SEU characteristics as the ion energy is increased in the 10's MeV/a range [6]. However, from previous work, it is not known if the SEU characteristics continually degrade with increasing ion energy or if there is a worst-case ion energy for SEU characterization.
