Re: Holes

[Doug Craigen <dcc@ESCAPE.CA>]

Chuck Britton wrote:
> At 10:02 PM -0700 11/3/99, Jim Green wrote:
> > How does a hole differ from a positron?  Would they act electrically the
> > same in a s/c?
> There is pretty good reason to believe that Light Emitting Diodes
> (LED's) give off a visible photon when an electron and a hole
> combine. If it were to be electrons and POSITRONS combining to
> produce EM radiation, we would expect to see gamma radiation instead
> of visible light.

What we call "holes" are the absence of electrons at the top of the
valence band.  One could consider them to be continually
colliding-annihilating-regenerating-another as the process of moving
around but these are thermal order of magnitude processes.  When they
collide with an electron in the conduction band however, there is
sufficent difference in the energy of the states to produce a photon in
the visible range for appropriate band-gap materials.

To address Jim Green's question, I believe what you are asking is what
electonric states would exist for a positron.  i.e. - if one was to put
an intrinsic semiconductor in the presence of a positron source, would
it essentially behave as if it was p-type.  One does not find positrons
at the top of the valence band - these states exist there because of the
lack of electrons due to thermal processes or pumping in the case of
excess holes generated to produce light.  When a positron enters a
semiconductor it finds defects - typically vacancies/ divacancies etc
and binds to them.  In essence the lack of atoms generates a region of
net negative charge, and the positron binds to that in a Bohr atom like
fashion.  Holes exist in Bloch states characteristic of the lattice,
positrons are trapped in defects.  (When holes are trapped in defects
they are defects with energy levels in the lower part of the band gap
region.)  The positron soon collides with an electron and they
annihilate to generate a pair of 511 keV gammas.  The whole process
typically lasts in the order of hundreds of picoseconds - much shorter
than most electron-hole interactions that we are concerned with in
semiconductors.

()-()-()-()-()-()-()-()-()-()-()-()-()-()-()-()

Doug Craigen
          Latest Project - the Physics E-source
          http://www.dctech.com/physics/