Relativistic Exact-Two-Component Core-Valence-Separated Algebraic Diagrammatic Construction Theory For Near L-edge X-ray Absorption Spectra
Somesh Chamoli, Sudipta Chakraborty, Xubo Wang, and Achintya Kumar Dutta

TL;DR
This paper introduces an efficient relativistic CVS-ADC(2) method for core-excitation calculations, optimized for heavy elements, with demonstrated accuracy and computational savings over traditional approaches.
Contribution
The authors develop a two-component relativistic CVS-ADC(2) implementation using SA-FNS and Cholesky decomposition, reducing computational cost while maintaining accuracy.
Findings
Benchmarking confirms reliability against four-component calculations.
Close agreement with canonical results demonstrates accuracy.
Application to transition-metal compounds and a ruthenium complex shows practical utility.
Abstract
We present an efficient implementation of the second-order two-component relativistic core-valence-separated algebraic diagrammatic construction method (CVS-ADC(2)) for core-excitation calculations. The approach employs state-averaged frozen natural spinors (SA-FNS) to reduce the number of floating-point operations, together with the Cholesky decomposition (CD) technique, which lowers the storage requirements associated with two-electron integrals. These reductions make the method particularly well-suited for systems containing heavy elements. Systematic benchmarking against four-component reference calculations confirms the reliability and robustness of the two-component (X2CMP/X2CAMF)-based framework. The close agreement with canonical results further demonstrates that the SA-FNS-based CVS-ADC(2) approach achieves comparable accuracy at only a fraction of the computational cost.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
