van der Spoel David, Ghahremanpour Mohammad Mehdi, Lemkul Justin A
Department of Biochemistry , Virginia Tech , 303 Engel Hall, 340 West Campus Dr. , Blacksburg , Virginia 24061 , United States.
J Phys Chem A. 2018 Nov 15;122(45):8982-8988. doi: 10.1021/acs.jpca.8b09867. Epub 2018 Nov 6.
Spectroscopic analysis of compounds is typically combined with density functional theory, for instance, for assigning vibrational frequencies, limiting application to relatively small compounds. Accurate classical force fields could, in principle, complement these quantum-chemical tools. A relatively simple way to validate vibrational frequencies is by computing thermochemical properties. We present such a validation for over 1800 small molecules using the harmonic approximation. Two popular empirical force fields (GAFF and CGenFF) are compared to experimental data and results from Gaussian-4 quantum-chemical calculations. Frequency scaling factors of 1.035 (CGenFF) and 1.018 (GAFF) are derived from the zero-point energies. The force field calculations have larger deviation from experiment than the G4 method for standard entropy, but for heat capacity the results are comparable. For internal thermal energy and zero-point energy the deviations from G4 are relatively small. The work suggests that with some tuning force fields could indeed complement DFT in spectroscopical applications.
化合物的光谱分析通常与密度泛函理论相结合，例如用于确定振动频率，但这限制了其应用于相对较小的化合物。原则上，精确的经典力场可以补充这些量子化学工具。一种相对简单的验证振动频率的方法是计算热化学性质。我们使用谐波近似对1800多个小分子进行了这样的验证。将两种流行的经验力场（GAFF和CGenFF）与实验数据以及高斯-4量子化学计算结果进行了比较。从零点能量得出频率缩放因子1.035（CGenFF）和1.018（GAFF）。对于标准熵，力场计算与实验的偏差比G4方法大，但对于热容，结果相当。对于内部热能和零点能量，与G4的偏差相对较小。这项工作表明，经过一些调整，力场确实可以在光谱应用中补充密度泛函理论。
