[net.space] Solar cells using dipole antenna arrays

davekn@shark.UUCP (08/10/84)

     A couple of articles concerning this subject have appeared in this
group, thus my comments are sent here.  The reference is a short (1/3
page) news article in the latest Spectrum describing a patent for a
solar cell technology using an array of .18 micron long by .01 micron
wide dipole antennas and rectifiers for converting light into DC
electricity.  Claimed efficiency is 75 to 80%.

     I believe this to be a hoax.  First, even with the lithography
dimensions mentioned in the article, the transit time of electrons
across a diode junction is much more than the 1 femto-second or so
needed to rectify light frequencies.  Second, the intensity of average
sunlight would produce about 30 microvolts across each dipole.  This
is much too small to overcome the forward drop of any diodes.  If the
sunlight is concentrated by 10,000 times (to 1000 watts per square
centimeter), the voltage raises to 3 millivolts - still much too small.
Since sunlight is not coherent, the voltage of several dipoles cannot
be added to improve this situation.  Third, said lithography is just
becoming possible, and would not be economical for large arrays for at
least another 10 years.

                               David Knierim           8/10/84