Login / Signup

Kohn-Sham potentials from electron densities using a matrix representation within finite atomic orbital basis sets.

Xing ZhangEmily A Carter
Published in: The Journal of chemical physics (2018)
We revisit the static response function-based Kohn-Sham (KS) inversion procedure for determining the KS effective potential that corresponds to a given target electron density within finite atomic orbital basis sets. Instead of expanding the potential in an auxiliary basis set, we directly update the potential in its matrix representation. Through numerical examples, we show that the reconstructed density rapidly converges to the target density. Preliminary results are presented to illustrate the possibility of obtaining a local potential in real space from the optimized potential in its matrix representation. We have further applied this matrix-based KS inversion approach to density functional embedding theory. A proof-of-concept study of a solvated proton transfer reaction demonstrates the method's promise.
Keyphrases
  • human health
  • magnetic resonance
  • density functional theory
  • computed tomography
  • risk assessment
  • minimally invasive
  • deep learning
  • molecular dynamics
  • big data
  • neural network
  • double blind