Linear Scaling Electronic Structure Methods for Molecules and Materials

普林斯顿大学的艾米丽·A·卡特获得了化学部理论和计算化学项目颁发的奖项，该奖项旨在开发分子和材料的线性扩展方法。对于分子体系，PI采用局域多参考单双激发组态相互作用(MRSDCI)方法及其规模扩展的模拟MR平均耦合对泛函(ACPF)理论。材料的研究是通过无轨道密度泛函理论(of-DFT)进行的。PI将通过将该方法重新表述为“积分直接”O(N)方法来加强线性定标，该方法利用双电子积分的预筛选，并通过平面波辅助基来计算这些积分。她正在继续发展局域赝势(LPS)，并将脂多糖和自旋相关赝势(SDP)方法结合起来，以便有可能在OF-DFT中处理过渡金属。这项工作通过提高可用方法的准确性、效率和通用性，正在产生更广泛的影响，这将对更广泛的科学界和工程界有用。

Emily A. Carter of Princeton University is supported by an award from the Theoretical and Computational Chemistry program within the Division of Chemistry for research to develop methods that scale linearly both for molecules and materials. For molecular systems the PI is using the local multi-reference single and double excitation configuration interaction (MRSDCI) method and its size-extensive analog, MR averaged coupled pair functional (ACPF) theory. Materials are being studied via orbital-free density functional theory (OF-DFT). The PI will enforce linear scaling by reformulating this method into an "integral direct" O(N) approach that exploits pre-screening of two-electron integrals and computes those integrals via an auxiliary basis of plane waves. She is continuing to develop local pseudopotentials (LPSs) and to combine the LPS and spin-dependent pseudopotential (SDP) methods so that it may be possible to treat transition metals in OF-DFT. This work is having a broader impact by enhancing the accuracy, efficiency, and generality of available methods, which will be useful for the broader science and engineering community.