The reliable prediction of Cu(II) hyperfine coupling constants remains a challenge for quantum chemistry. Until recently only density functional theory (DFT) could target this property for systems of realistic size. However, wave function based methods become increasingly applicable. In the present work, we define a large set of Cu(II) complexes with experimentally known hyperfine coupling constants and use it to investigate the performance of modern quantum chemical methods for the prediction of this challenging spectroscopic parameter. DFT methods are evaluated against orbital‐optimized second‐order Møller‐Plesset (OO‐MP2) theory and coupled cluster calculations including singles and doubles excitations, driven by the domain‐based local pair natural orbital approach (DLPNO‐CCSD). Special attention is paid to the definition of a basis set that converges adequately toward the basis set limit for the given property for all methods considered in this study, and a specifically optimized basis set is proposed for this purpose. The results suggest that wave function based methods can supplant but do not outcompete DFT for the calculation of Cu(II) hyperfine coupling constants. Mainstream hybrid functionals such as B3PW91 remain on average the best choice.

中文翻译：

---

密度泛函法和相关波函数法预测Cu（II）超细耦合常数的比较

Cu（II）超细耦合常数的可靠预测仍然是量子化学的一个挑战。直到最近，只有密度泛函理论（DFT）才能针对实际大小的系统指定此属性。但是，基于波动函数的方法变得越来越适用。在当前的工作中，我们定义了一大批具有实验已知的超精细偶合常数的Cu（II）配合物，并用它来研究现代量子化学方法在预测这一具有挑战性的光谱参数方面的性能。DFT方法是根据轨道优化的二阶Møller-Plesset（OO-MP2）理论以及包括单次和双次激发的耦合簇计算（由基于域的局部对自然轨道方法（DLPNO-CCSD）驱动）进行评估的。对于本研究中考虑的所有方法，要特别注意基集的定义，该定义应充分收敛到给定属性的基集限制，并且为此目的提出了专门优化的基集。结果表明，基于波动函数的方法可以代替但不超过DFT来计算Cu（II）超精细耦合常数。平均而言，主流混合功能（例如B3PW91）仍然是最佳选择。