No abstract

The optical glow of ceramics that becomes established during the constant state of flash, known as Stage III in flash sintering experiments, is investigated. The specimen temperature in this state is obtained from in-situ experiments at the Pohang Light Source II. The measurements of the specimen temperature agree very well with the predictions from the black body radiation model. The optical emission spectrum from the specimen is measured from the visible into the deep infrared, and compared with black body radiation that would have been expected from Joule heating. It is concluded that the specimens radiate by electroluminescence, which is ascribed to electron-hole recombination of excitons. The phenomenon is likely the same as discovered by Nernst at the turn of the twentieth century.

A flash sintering experiment can be carried out by applying an electric field and heating the specimen at a constant rate. The flash event occurs at a specific temperature that depends on the strength of the electric field. Alternatively, the furnace can be held at a constant temperature and the voltage applied as a step function; after an incubation time there is a highly non-linear rise in conductivity. This incubation step is called Stage I. The non-linearity is constrained by switching the power supply to current control. This short transient, during which the sample sinters nearly instantaneously, is the second stage. Under current-control, the (essentially dense) sample remains in a highly excited state indefinitely, which we call Stage III. In this state, the samples are often brightly electroluminescent emitting a green glow; unusual phase transformations occur and the rate of chemical reactions is greatly enhanced. We infer that these manifestations are evidence of a defect catastrophe that includes unusual generation of electrons, holes and point defects, which can produce sintering, electronic conductivity, electroluminescence, and phase transformations, all at the same time. We hypothesize that both Joule heating and electric field are necessary for this catastrophe.

The original flash sintering experiment was carried out by applying an electric field, and switching to current control at the onset of the flash, signaled by a rise in conductivity. Here, we consider experiments where the experiment is controlled from the very start, by injecting current, which is increased at a constant rate. The current rates are varied from 50 mA/min to 5000 mA/min. The experiment is continued until, in all cases, the current density reaches 100 mA/mm 2 . The total duration of the experiment ranged from approximately 7 seconds to 700 seconds. The following comparisons to the earlier voltage-to-current experiments are noted: (a) in both instances, the onset of the flash is signaled by an unusual rise in conductivity; however, since the power supply remains in the current control mode, the increase in conductivity is manifested by a drop in the voltage generated across the specimen; (b) the blackbody radiation model is modified to include the energy absorbed in specific heat, in order to determine the time-dependent change in temperature as the current is increased-this correction is particularly significant at the very high current rates; (c) sintering occurs continuously, reaching full density, in all instances, when the current density reaches~100 mA/mm 2 ; and (d) these early experiments suggest that the current-rate experiments yield fine-grained microstructure across the entire gauge section of the dog-bone specimen, presumably because the highly transient conditions of voltage-to-current flash experiments are sidestepped. The experiments were carried out on 3 mol% yttriastabilized zirconia.