Loosjes and Vink proposed the pore conduction hypothesis derived from the approximate equality in values of activation energy of conductivity at high temperatures and work function by Richardson plots and also from TV characteristics obtained in their experimental study. Seen from their experimental method, the current in the I-V characteristics is of stationary state value. In the author's experiment, the characteristics are examined for current of d. c. instantaneous value, of stationary state value after the lapse of time, and for a. c. voltages of various frequencies, and also for the case of d. c. and a. c. superimposed voltages.
The characteristics are linear for all d. c. instantaneous values and a. c. values, but for d. c. stationary values, the linearity is lost at high voltages. As to the activity, the linearity holds only for its high values.
The conduction current of oxide cathodes is found to decay in msec-sec as well as the emission current. The deviation from Ohm's law in d. c. stationary state current at high voltages is due to this decay. This dacay of conduction current can be explained by the pore conduction mechanism. It is understood as due to poisoning of gas released from inner oxide walls to pores of crystal intensities as well as the decay of emission current. Thus, when the decay is taken into consideration, all of the I-V characteristics obtained in this experiment are very well explained.
There are two activation energies, one in each of the higher and lower temperature regions. The former is of the pore conduction and the latter is of the bulk crystal. In this case, the values of activation energy of conduction and work function are not always equal which is not in line with the results of Loosjes and Vink. This can, to some extent, be explained by considering the difference in surface activities of outer crystals and inner crystals of oxide cathodes.
