The growth of self-organised semiconductor dots, such as islands of InAs
surrounded by GaAs, relies on the strain due to their different lattice
constants. Some strain remains after the dot has been buried by further
growth, and experiments have been designed to use this strain to confine
carriers in a nearby quantum well. I have shown that the strain field of a
buried quantum dot has some unusually pleasant properties:
* It can be found from the solution of Poisson's equation with the lattice mismatch playing the role of "charge density"
* The dilation is constant within the dot and vanishes outside it, so
electrons outside the dot feel no deformation potential.
The absence of dilation outside the dot means that electrons feel no
deformation potential and will therefore not be trapped in a nearby quantum
well. Holes and excitons are only weakly affected too. However, strain also
induces a piezoelectric potential, which can trap electrons and holes in
separate regions of space. This may have applications for the storage of
light or charge.
An elastic field can instead be produced by depositing a stressor on the surface. The resulting elastic field is significantly different from that for the buried dot; in particular the dilation no longer vanishes. I shall discuss the implications of these results for recent experiments at Santa Barbara on strain-induced quantum dots, which employed both types of stressor.
This work was carried out while I was a visitor at QUEST in the University of California at Santa Barbara during the academic year 1997-8.