Comments on "Mobility degradation due to the gate field in the inversion layer of MOSFET's"

Fu's model [ibid., EDL-3, pp. 292-293, Oct. 1982] is based on the assumption that the thermal energy of electrons or holes in the inversion layer may be increased by the application of a gate field. In the absence of a tangential field (VDS = 0), however, carriers are in thermal equilibrium, regardless of the gate field, and hence can be characterized by a constant fermi level through the inversion layer and substrate. Indeed, as Smith (ref. [4] in Fu's paper) himself points out, the rate of increase in the thermal energy is the (dot) product of the field and the average velocity. If the average velocity is zero, there can be no heating. In fact, even in the case of an arbitrary tangential field E (VDS > 0), so large that the drift velocity is no longer simply pE, the gate field still plays no direct role in increasing the thermal energy, since it is normal to the current flowing in the channel. (To be sure it does affect the size of the current by altering the scattering rates, and hence, in this manner indirectly influences the heating of the carriers.)