Isothermal and isochronal annealing experiments on irradiated commercial power VDMOSFETs

Abstract

The paper presents results of a study of post-irradiation effects in two types of gamma-ray iiradiated power VDMOSFETs, commercially available from different manufacturers. Isothermal annealing with constant and switching gate bias, as well as isochronal annealing have been performed. The application of charge-pumping measurements in commercial three-teiminal VDMOSFETs is also discussed. Besides enabling insight into the post-irradiation response of investigated devices, obtained results point out some interesting post-irradiation phenomena.

Properties of latent interface-trap buildup in irradiated metal–oxide–semiconductor transistors determined by switched bias isothermal annealing experiments

“…2 was measured some 20 h after the switch; however, it takes only several minutes, if not even less, for dramatic changes in ⌬N ot to occur. 10 These changes are shown to be reversible: switching the bias to positive again leads to ⌬N ot values close to the values before the first bias switch. 10 It can be concluded that during phase 2 of annealing, the oxide-trap charge is not permanently neutralized, but rather temporarily compensated by the electrons tunneling from Si under the positive bias.…”

“…10 These changes are shown to be reversible: switching the bias to positive again leads to ⌬N ot values close to the values before the first bias switch. 10 It can be concluded that during phase 2 of annealing, the oxide-trap charge is not permanently neutralized, but rather temporarily compensated by the electrons tunneling from Si under the positive bias. 11 When the bias is switched at later annealing times, i.e., at the end of phase 3 ͑140 and 170°C data͒, ⌬N ot also rebounds, but not quite to its preannealing value, while ⌬N it changes are almost negligible.…”

“…2 and also for the reversible ⌬N ot changes in subsequent bias switches. 10 However, the biasindependent behavior during phase 1 of annealing is more difficult to account for. It might be that the H ϩ ions are, during this period, so deeply trapped that the bias sign has no influence on their transport.…”