Insulation-resistance degradation kinetics of bulk BaTi1−ξAξO3−Δ and defect-chemical origin of acceptor-type(A) and doping-level(ξ) effect

Dearth of the reproducible, consistent observations on insulation-resistance (IR) degradation kinetics of bulk dielectric BaTiO3 may be attributed to their conventional measurement method, two-probe potentiostatic, which would be by no means free from the electrode effect for a finite- dimension specimen in particular. We hereby measured the IR-degradation kinetics galvanostatically by using a series of inner probes on bulk BaTi1−ξAξO3−Δ (A = Al, Mn; ξ = 0.001, 0.003, 0.010) with their high-temperature (1000 °C) equilibrium ionic-defect-structure in air being frozen-in at 250 °C and compared with the kinetics as calculated on the basis of the electromigration of frozen-in oxygen vacancies ( c V o ) in association with the A-ionization or hole-trapping equilibria. It has turned out that the calculated depict sufficiently precisely all the as-observed kinetics as well as the effects of acceptor type(A) and doping level(ξ), thus, quantitatively establishing the correlation between the frozen-in ionic-defect-structure and IR-degradation kinetics with new insights into the degradation inner-workings: IR-degradation is triggered as soon as the oxygen vacancy concentration at the cathode reaches that corresponding to the insulator-to-semiconductor transition ( c V S / I ) and proceeds with the front of just-turned, n-type semiconducting region ( c V = c V S / I ) moving towards the anode at a fixed velocity. The healthy lifetime of the dielectric is, thus, essentially the time duration for the cathode to achieve c V S / I from c V o , and the final stage of degradation is approximated to be the length fraction χs of the semiconductor such that χs = c V o / c V S / I . A new suggestion is finally made to further suppress the IR degradation kinetics of the bulk dielectric BaTiO3.