PyFLOSIC: OpenSource Implementation of the Fermi-Löwdin-Orbital Self-Interaction-Correction for use in PySCF

我们为社区提供了一个Fermi-Löwdin轨道自相互作用校正（FLO-SIC）方法的开源实现，称为PyFLOSIC。此外，我们建立了费米轨道描述符（FOD），FLO-SIC的核心，常见的化学键理论之间的联系。此外，我们指出，SIC需要一个更紧密的数值参数空间（数值求积和基组）比底层的密度泛函。这些发现对于SIC在现代（Meta）GGA-泛函之上的应用尤其重要。需要进一步的调查，以了解SIC的数值需求。更重要的是，几乎被遗忘的林内特双四重态（LDQ）理论在FLO-SIC的背景下复活了。基于LDQ的FOD可以提高单粒子密度，并为联合收割机SIC与化学键概念的结合提供了一个全新的研究视角。

We provided the community with an open source implementation of the Fermi-Löwdin orbital self-interaction correction (FLO-SIC) method called PyFLOSIC. In addition, we established the connection between the Fermi-orbital descriptors (FODs), being the core of FLO-SIC, to common chemical bond theories. Further, we pointed out that SIC has the need for a tighter numerical parameter space (numerical quadrature and basis set) than the underlying density functionals. These findings are especially important for application of SIC ontop of modern (meta-) GGA-functionals. Further investigations are needed to understand the numerical demands of SIC. More importantly, the nearly forgotten Linnett double-quartet (LDQ) theory was revived in context of FLO-SIC. FODs based on LDQ may lead to improved single particel densities and offer a completely new research perspective to combine SIC with chemical bonding concepts.