Radoń Mariusz
Faculty of Chemistry, Jagiellonian University, Kraków, Poland.
Phys Chem Chem Phys. 2024 Jul 3;26(26):18182-18195. doi: 10.1039/d4cp01327a.
Accurate computational treatment of spin states for transition metal complexes, exemplified by iron porphyrins, lies at the heart of quantum bioinorganic chemistry, but at the same time represents a great challenge for approximate density functional theory (DFT) methods, which are predominantly used. Here, the accuracy of DFT methods for spin-state splittings in iron porphyrin is assessed by probing the ability to correctly predict the ground states for six Fe or Fe complexes experimentally characterized in solid state. For each case, molecular and periodic DFT calculations are employed to quantify the effect of porphyrin side substituents and the crystal packing effect (CPE) on the spin-state splitting. It is proposed to partition the total CPE into additive components, the direct and structural one, the importance of which is shown to significantly vary from case to case. By knowing the substituent effect, the CPE, and the Gibbs free energy thermodynamic correction from calculations, one can employ the experimental ground-state information in order to derive a quantitative constraint on the electronic energy difference for a simplified (porphin) model of the experimentally characterized metalloporphyrin. The constraints derived in such a way-in the form of single or double inequalities-are used to assess the accuracy of dispersion-corrected DFT methods for 6 spin-state splittings of [Fe(P)(2-MeIm)], [Fe(P)(2-MeIm)], [Fe(P)(THF)] and [Fe(P)] models (where P is porphin, 2-MeIm is 2-methylimidazole, THF is tetrahydrofuran). These data constitute the new benchmark set of spin states for crystalline iron porphyrins (SSCIP6). The highest accuracy is obtained in the case of double-hybrid functionals (B2PLYP-D3, DSD-PBEB95-D3), whereas hybrid functionals, especially those with reduced admixture of the exact exchange (B3LYP*-D3, TPSSh-D3), are found to considerably overstabilize the intermediate spin state, leading to incorrect ground-state prediction in Fe porphyrins. The present approach, which can be generalized to other transition metal complexes, is not only useful in method benchmarking, but also sheds light on the interpretations of experimental data for metalloporphyrins, which are important models to understand the electronic properties of heme proteins.
以铁卟啉为代表的过渡金属配合物自旋态的精确计算处理是量子生物无机化学的核心，但同时也是主要使用的近似密度泛函理论（DFT）方法面临的巨大挑战。在此，通过探究正确预测六种在固态下经实验表征的铁或亚铁配合物基态的能力，评估了DFT方法对铁卟啉自旋态分裂的准确性。对于每种情况，采用分子和周期性DFT计算来量化卟啉侧链取代基和晶体堆积效应（CPE）对自旋态分裂的影响。建议将总CPE分为相加的成分，即直接成分和结构成分，其重要性在不同情况下有显著差异。通过了解取代基效应、CPE以及计算得到的吉布斯自由能热力学校正，人们可以利用实验基态信息，以便对经实验表征的金属卟啉的简化（卟吩）模型的电子能量差得出定量约束。以单重或双重不等式形式得出的此类约束用于评估色散校正DFT方法对[Fe(P)(2-MeIm)]、[Fe(P)(2-MeIm)]、[Fe(P)(THF)]和[Fe(P)]模型（其中P为卟吩，2-MeIm为2-甲基咪唑，THF为四氢呋喃）的6种自旋态分裂的准确性。这些数据构成了结晶铁卟啉自旋态的新基准集（SSCIP6）。在双杂化泛函（B2PLYP-D3、DSD-PBEB95-D3）的情况下获得了最高精度，而杂化泛函，尤其是那些精确交换混合减少的泛函（B3LYP*-D3、TPSSh-D3），被发现会使中间自旋态过度稳定，导致铁卟啉基态预测错误。本方法可推广到其他过渡金属配合物，不仅对方法基准测试有用，还为金属卟啉实验数据的解释提供了启示，金属卟啉是理解血红素蛋白电子性质的重要模型。
