Abstract
Besides randomly walking on the crystal surface the adatoms are assumed to have a drift velocity c=D s F/k T, where D s is the surface diffusion coefficient and F is a force, related in some way to the specimen heating current. In the limiting case of narrow terraces (small interstep distance l) the major process on the whole surface is the surface transport, which results an absence of noticeable conversion of the domain structure when F l/k T≈0.1. Making use of the agreement between this conclusion and the experimentally observed [H. Kahata and K. Yagi: Jpn. J. Appl. Phys. 28 (1989) L858] lack of conversion (under reversal of the current direction) at l≈40–80 nm and T=700°C, the force F is estimated to be in the range 10-14–10-13 N. Experimental verification of the predicted stability of vicinal surfaces with lF has step-up direction and instability (decay to large terraces, separated by steps of multiple height) when F has step-down direction can be a crucial test of the hypothesis for an electromigration of Si adatoms.

This publication has 12 references indexed in Scilit: