Harnessing the quantum nature of donor atoms in silicon may
pave the way for a quantum revolution in the modern digital information era. The idea to combine the exceptional spin coherence properties of donor electron spins in silicon with the prospect
of exploiting technology prevalent in the semiconductor industry
is very appealing. This thesis provides a quantitative limit for
the spin coherence times of phosphorus donor-bound electrons in
silicon, which is a fundamental parameter for spin-based quantum computation. To this end, the spin-lattice relaxation time in
28Si:P is measured with the highest degree of precision to date for
unprecedentedly low temperatures. The measurements yield extremely long spin-lattice relaxation times exceeding twenty hours,
which is orders of magnitude larger than originally determined.
These long spin-relaxation times confirm the latent potential for
devices based on spin manipulation donor electrons in silicon. For
very low temperatures and high magnetic fields, the impact of
the bosonic phonon distribution on the spin-relaxation time is observed for the very first time and with high accuracy which was
predicted by theory more than 60 years ago.
Furthermore, a new method of measuring the bandgap using donor
electrons based on optical spectroscopy of the D0X transition is
presented. This new method can be used to locally detect the lattice temperature via the Si bandgap with exceptional accuracy
and excellent temporal resolution. With the help of this method,
measurements of the bandgap temperature dependence are performed with 7e-10 relative precision. Although the precise
measurements verify the theoretical T^4 limit of the bandgap energy shift with high certainty, a discrepancy of the absolute shift
questions the existing theory of electron-phonon coupling in semi-
conductors in the low temperature limit. Additional time-resolved
experiments facilitate the use of this new method as a precise local
thermometer to be used in 28Si:P based devices
|