Background events in the DRIFT-IId dark matter detector, mimicking potential
WIMP signals, are predominantly caused by alpha decays on the central cathode
in which the alpha particle is completely or partially absorbed by the cathode
material. We installed a 0.9 micron thick aluminized-mylar cathode as a way to
reduce the probability of producing these backgrounds. We study three
generations of cathode (wire, thin-film, and radiologically clean thin-film)
with a focus on the ratio of background events to alpha decays. Two independent
methods of measuring the absolute alpha decay rate are used to ensure an
accurate result, and agree to within $10\%$. Using alpha range spectroscopy, we
measure the radiologically cleanest cathode version to have a contamination of
$3.3\pm0.1$ ppt $^{234}$U and $73\pm2$ ppb $^{238}$U. This cathode reduces the
probability of producing an RPR from an alpha decay by a factor of $70\pm20$
compared to the original stainless steel wire cathode. First results are
presented from a texturized version of the cathode, intended to be even more
transparent to alpha particles. These efforts, along with other background
reduction measures, have resulted in a drop in the observed background rate
from 500/day to 1/day. With the recent implementation of full-volume
fiducialization, these remaining background events are identified, allowing for
background-free operation.