最近开发的模型化学(表示为junChS [Alessandrini等人J. Chem. Theory Comput. 2020,16,988-1006])已经扩展到显式相关(F12)方法的使用。这使我们提出了一个家庭的有效的,可靠的和无参数的计划,计算精确的相互作用能的分子复合物统治的非共价相互作用。基于广泛的相互作用的全面基准测试表明,所谓的junChS-F12模型,采用具有成本效益的revDSD-PBEP 86-D3(BJ)参考几何结构,相对于其传统的对应物具有更好的性能,并且优于众所周知的模型化学。在不使用任何经验参数和负担得起的计算成本的情况下,junChS-F12达到了亚化学精度。分子复合物的精确表征通常限于能量学。为了向前迈出一步,传统的和F12复合计划开发的相互作用能已扩展到结构的测定。基准研究表明,最有效的选择是在fc-CCSD(T)-F12/jun-cc-pVTZ几何结构中加入MP2-F12核价相关校正,而不需要恢复基组叠加误差和外推到完整基组。
英文摘要:
A recently developed model chemistry (denoted as junChS [Alessandrini et al J. Chem. Theory Comput. 2020, 16, 988-1006]) has been extended to the employment of explicitly-correlated (F12) methods. This led us to propose a family of effective, reliable and parameter-free schemes for the computation of accurate interaction energies of molecular complexes ruled by non-covalent interactions. A thorough benchmark based on a wide range of interactions showed that the so-called junChS-F12 model, which employs cost-effective revDSD-PBEP86-D3(BJ) reference geometries, has an improved performance with respect to its conventional counterpart and outperforms well-known model chemistries. Without employing any empirical parameter and at an affordable computational cost, junChS-F12 reaches sub-chemical accuracy. Accurate characterizations of molecular complexes are usually limited to energetics. To take a step forward, the conventional and F12 composite schemes developed for interaction energies have been extended to structural determinations. A benchmark study demonstrated that the most effective option is to add MP2-F12 core-valence correlation corrections to fc-CCSD(T)-F12/jun-cc-pVTZ geometries without the need of recovering the basis set superposition error and the extrapolation to the complete basis set.
Improved accuracy and transferability of molecular-orbital-based machine learning: Organics, transition-metal complexes, non-covalent interactions, and transition states
Attaining high accuracy for charge-transfer excitations in non-covalent complexes at second-order perturbation cost: the importance of state-specific self-consistency
Accurate non-covalent interaction energies via an efficient MP2 scaling procedure
利用观测到的CCSD(T)和MP2相关相互作用能的比例关系[I. Grabowski,E. Fabiano,F. Della Sala,Phys.Chem.Chem.Phys.15,15485(2013)],我们提出了一种简单的定标程序来计算非共价复合物的精确相互作用能。我们的方法利用MP2和CCSD(T)相关能,在相对较小的基组计算,并拟合比例系数产生几乎完整的基组限制CCSD(T)质量的相互作用能。由于计算中涉及的标度系数具有良好的可移植性,因此对于不同的分子间距离可以容易地获得良好的结果。
Learning to Dock: Geometric Deep Learning for Predicting Supramolecular Host-Guest Complexes
Shanghaitech University(上海科技大学);State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech University(先进医用材料与设备国家重点实验室,上海科技大学);School of Information Science and Technology, ShanghaiTech University(上海科技大学信息科学与技术学院);University of Science and Technology of China(中国科学技术大学);State Key Laboratory of Precision and Intelligent Chemistry, University of Science and Technology of China(精密与智能化学国家重点实验室,中国科学技术大学)
Organic-Inorganic Hybrid CH3NH3PbI3 Perovskite Solar Cell Nanoclusters: Revealing Ultra-Strong Hydrogen Bonding and Mulliken Inner Complexes and Their Implication in Materials Design
甲基铵碘化铅(CH 3 NH3 PbI 3)钙钛矿太阳能电池由于其作为光捕获半导体的出色器件基础性能而在太阳能电池技术的光伏历史上产生了显著的突破。虽然在过去的4年里,在固态下对该系统的实验和理论研究已经有了大量的报道,但其基本的团簇物理还有待开发。为此,本研究进行了理论研究,使用DFT-M06-2X/ADZP检查的主要几何,电子,拓扑和轨道性质的CH 3 NH3 PbI 3纳米团簇块。这些集群被发现是异常强烈的约束,与位于93.53和125.11千卡mol-1之间的结合能(超过共价极限,40千卡mol-1),使我们能够表征为超强型的潜在的相互作用。基于此,连同异常高的电荷转移,强烈的超共轭相互作用,复杂的拓扑结构的电荷密度,和短的分子间距离发现,我们已经确定了CH 3 NH3 PbI 3作为Mulliken内配合物。此外,这些后果,以及超强相互作用,在设计新的功能纳米材料进行了简要讨论。在这项研究中获得的各种新的结果并不完全符合那些已经报道的实验(Nat. Commun. 2015,6,7124)和计算上(Chem.Commun.,2015,51,6434; Sci. Rep. 2016,doi:10.1038/srep21687; Chem. Mater. 2016,28,4259; J. Mat. Chem A 2016,4,4728; J. Phys. Chem. Lett…
Single-electron Fluorine Bonding. I. The SEFB-15 and CSFB-14 Benchmark Datasets, Reference Binding Energies and Intermolecular Bond Distances: Assessment of M06-2X density functional
Inorganic and Physical Chemistry, Indian Institute of Science(印度科学研究所无机与物理化学)