Database of Computed Properties for Color Center Defects in Silicon
ORAL
Abstract
Color center defects in silicon are emerging as a promising platform for realizing a number of
applications in quantum information science (QIS), including quantum sensing, single-photon sources,
and integrated quantum communication between quantum computer nodes. Several well-studied
defects such as the G-center, W-center, and T-center, possess some of the necessary attributes for these
applications, including narrow linewidths, emission in the telecommunications band, long electron spin
coherence times, and coupling between spin and optical degrees of freedom. Despite this, no known
defect is perfectly suitable for QIS applications, and in fact different devices can require defects with
expressly distinct sets of properties. We report the publication of a searchable online quantum defect
database containing the computed properties of over 5000 distinct silicon defect structures. Formation
energies, defect energy levels, ground and excited spin states, zero phonon lines, and electric dipole
matrix elements are provided for each defect, which are then used to screen for candidate defects with
emission within the telecommunications band, non-trivial spin state, and strong optical coupling.
Additionally, a machine learning approach is applied to predict defect properties directly from structural
data.
applications in quantum information science (QIS), including quantum sensing, single-photon sources,
and integrated quantum communication between quantum computer nodes. Several well-studied
defects such as the G-center, W-center, and T-center, possess some of the necessary attributes for these
applications, including narrow linewidths, emission in the telecommunications band, long electron spin
coherence times, and coupling between spin and optical degrees of freedom. Despite this, no known
defect is perfectly suitable for QIS applications, and in fact different devices can require defects with
expressly distinct sets of properties. We report the publication of a searchable online quantum defect
database containing the computed properties of over 5000 distinct silicon defect structures. Formation
energies, defect energy levels, ground and excited spin states, zero phonon lines, and electric dipole
matrix elements are provided for each defect, which are then used to screen for candidate defects with
emission within the telecommunications band, non-trivial spin state, and strong optical coupling.
Additionally, a machine learning approach is applied to predict defect properties directly from structural
data.
*This work was supported by the Office of Fusion Energy Sciences and the Molecular Foundry, a DOEOffice of Science User Facility, supported by the Office of Science of the U.S. Department of Energyunder Contract No. DE-AC02-05CH11231.
–
Presenters
-
Vsevolod Ivanov
- Lawrence Berkeley National Lab
- Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
- Lawrence Berkeley National Laboratory