Examination of the statistics of dielectric breakdown

The use of a compound Weibull distribution in both time and electric field for the description of dielectric breakdown data is examined. It is shown that it is insufficient to measure breakdown experimentally using only a single variable such as a constant electric field, since the evaluation of the minimum number of parameters required to characterise the simplest system requires four independent pieces of information. Applicable statistical relationships are derived and applied to a wide range of experimental data. It is shown that, in general, the information that has been published on dielectric breakdown is limited and does not allow full assessment of the breakdown mechanism. Such information as is available is discussed and some broad conclusions about breakdown mechanisms are proposed.