Abstract
One of the most significant drawbacks of the all-electron ab initio diffusion Monte Carlo (DMC) is that its computational cost drastically increases with the atomic number ($Z$), which typically scales with $Z^{\sim 6}$. In this study, we introduce an algorithm based on a very efficient implementation of the Lattice Regularized Diffusion Monte Carlo (LRDMC), where the conventional time discretization is replaced by its lattice space counterpart. This scheme enables us to conveniently adopt a small lattice space in the vicinity of nuclei, and a large one in the valence region, by which a considerable speedup is achieved, especially for large atomic number $Z$. Indeed, the computational performances of our algorithm can be theoretically established by using the Thomas-Fermi model for heavy atoms, yielding an almost affordable scaling with the atomic number, i.e., $Z^{\sim 5}$. This opens the way for efficient and accurate all-electron ab initio DMC in electronic structure calculations.
Highlights
In recent years, the grand challenge in materials modeling is to provide extremely accurate reference energetics often well beyond the standard benchmark provided by the density functional theory (DFT) that notoriously is not enough predictive in several materials of both scientific [1,2] and technological interests [3,4]
We have introduced a powerful strategy to deal with the different time and length scales in electronic systems that has allowed us to establish an improved doublegrid lattice regularized diffusion Monte Carlo (LRDMC) with careful balance between speedup and accuracy, yielding unprecedented computations even for large atomic number Z
The speedup of the LRDMC is predicted theoretically within the standard Thomas-Fermi model for atoms with large atomic number, and the calculation is accelerated in practice by a large amount, especially for large atomic number Z
Summary
The grand challenge in materials modeling is to provide extremely accurate reference energetics often well beyond the standard benchmark provided by the density functional theory (DFT) that notoriously is not enough predictive in several materials of both scientific [1,2] and technological interests [3,4]. Devised a method for the diffusion Monte Carlo (DMC) [31] by reducing the velocity in the vicinity of nuclei to prevent from overshooting electrons Despite that this improves the accuracy by a sizable amount, the major drawback of the conventional DMC is that the imaginary time step has to remain necessarily the same both for the valence and the core region [31] because it is difficult to decompose the propagation in imaginary time into the sum of multiple operators acting on different length scales. We introduce a strategy to handle different length and time scales in electronic ground-state calculations by having in mind the Thomas-Fermi theory, that is quite general and allows in particular a much more powerful doublegrid LRDMC After this appropriate treatment, this method becomes applicable to any element without introducing significant bias, while definitely accelerating the computation especially for large atomic number Z
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
