37 lines
1.8 KiB
TeX
37 lines
1.8 KiB
TeX
\documentclass[10pt]{letter}
|
|
\usepackage{UPS_letterhead,color}
|
|
\newcommand{\alert}[1]{\textcolor{red}{#1}}
|
|
|
|
\begin{document}
|
|
|
|
\begin{letter}%
|
|
{To the Editors of the Journal of Chemical Physics, Special Topic on Frontiers of Stochastic Electronic Structure Calculations (K. Jordan, M. Morales, L. Shulenburger and L.
|
|
Wagner)}
|
|
|
|
\opening{Dear Editors,}
|
|
|
|
Please find attached a manuscript entitled
|
|
``Taming the fixed-node error in diffusion Monte Carlo via range separation'', which
|
|
we would like to submit as an invited article in the \textit{Journal of Chemical Physics} for the JCP Special Topic on Frontiers of Stochastic Electronic Structure Calculations.
|
|
|
|
In this manuscript, we combine range-separated density functional theory (RS-DFT) and
|
|
selected configuration interaction using the CIPSI algorithm to build
|
|
multi-determinant trial wave functions for diffusion Monte Carlo calculations.
|
|
Compared to conventional CIPSI wave functions, we are able to reduce by several
|
|
orders of magnitude the number of determinants of the CI expansion, and we can
|
|
obtain fixed-node DMC energies lower than the DMC energy obtained with the full
|
|
CI trial wave function within the same basis set. Benchmarks calculations have shown
|
|
that the proposed scheme can provide a mean absolute error of 2~kcal/mol on the
|
|
atomization energies of the 55 molecules of the Gaussian-1 set with a
|
|
reproducible protocol.
|
|
|
|
Because of the large potential impact of our work, we expect it to be of great
|
|
interest to a wide audience within the electronic structure community.
|
|
We suggest Lubos Mitas, Andrea Zen, Seiichiro Ten-no, Michele Casula, Matus
|
|
Dubecky, Manolo Per, Sandeep Sharma, and Eric Neuscamman as potential referees.
|
|
We look forward to hearing from you soon.
|
|
|
|
\closing{Sincerely, the authors.}
|
|
\end{letter}
|
|
\end{document}
|