Xu Qiang, Prendergast David, Qian Jin
Chemical Science Division, Lawrence Berkeley National Laboratory, Berkeley, California94720, United States.
Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California94720, United States.
J Chem Theory Comput. 2022 Sep 13;18(9):5471-5478. doi: 10.1021/acs.jctc.2c00474. Epub 2022 Aug 29.
We systematically studied a real-space pesudopotential method for the calculation of 1 core-electron binding energies of second-row elements B, C, N, and O within the framework of Kohn-Sham density functional theory (KS-DFT). With Dirichlet boundary conditions, pseudopotential calculations can provide accurate core-electron binding energies for molecular systems, when compared with the results from all-electron calculations and experiments. Furthermore, we report that with one simple additional nonself-consistent calculation as a refinement step using a hybrid exchange-correlation functional, we can generally improve the accuracy of binding energy shifts, promising a strategy for improving accuracy at a much lower computational cost. The specializations in the present approach, combined with our efficient real-space KS-DFT implementation, provide key advantages for calculating accurate core-electron binding energies of large-scale systems.
我们在Kohn-Sham密度泛函理论（KS-DFT）框架下，系统地研究了一种实空间赝势方法，用于计算第二周期元素B、C、N和O的1s芯电子结合能。在狄利克雷边界条件下，与全电子计算和实验结果相比，赝势计算能够为分子系统提供准确的芯电子结合能。此外，我们报告称，通过使用混合交换相关泛函进行一个简单的额外非自洽计算作为细化步骤，我们通常可以提高结合能位移的精度，这有望成为一种以低得多的计算成本提高精度的策略。本方法的特点，结合我们高效的实空间KS-DFT实现，为计算大规模系统的准确芯电子结合能提供了关键优势。
