Basis set dependence of phosphate frequencies in density functional theory calculations

Abstract

The addition of extravalence, polarization and diffuse functions, were studied in order to conclude how they affect the PAO stretching frequencies of several biological relevant phosphate molecules. The results show that the polarization and the diffuse functions have opposite effects on the frequencies: the polarization functions downshift while the diffuse functions upshift the frequencies. The effect of the valence functions was more difficult to interpret. The effect of the conductor-like screening model (CPCM)-continuum model was also studied. The results show that the CPCM-continuum model has a substantial effect on the frequencies for these small molecules. The continuum model's efficiency is mainly due to its effect on the geometries and not on the frequencies.

Quantifying bond distortions in transient enzyme species by a combination of density functional theory calculations and time-resolved infrared difference spectroscopy. Implications for the mechanism of dephosphorylation of the sarcoplasmic reticulum Ca2+-ATPase (SERCA1a)

“…In case of the protein models, the basis set and the electron density (to make sure to obtain convergence to the same state) from the geometry optimization with Jaguar were exported to Gaussian. No scaling factor was applied to the wavenumber calculations with the large basis set since it calculates P\ \O wavenumbers which are very close to experimental results [49]: for the two PO 3 2 − -containing molecules in the studied set of molecules, the calculated wavenumbers of the asymmetric P\ \O stretching vibrations differed by only 3.3 cm −1 from experimental results (averaged absolute value of the deviations). One of these molecules was AcP.…”

“…Frequency calculations on such large systems are computationally very expensive and it is therefore not feasible to use a larger basis set. Furthermore, Rudbeck [49] showed that even though the mean average error of the calculated P\ \O wavenumbers for the 6-31G** basis set is quite high, the phosphate geometry is reasonable. The large basis set was used for the AcP models.…”

“…Wavenumbers calculated with the small basis set were scaled relative to the large basis set using a scaling factor of 0.985. The factor was determined for the asymmetric P\ \O stretching vibrations of AcP from gas phase DFT calculations [49].…”

“…The positions of the interacting HF or H 2 O molecules were adjusted such that the calculated P\ \O wavenumbers agreed with those determined experimentally. The small number of atoms in the AcP models has the advantage that the large basis set 6-311++G(3df, 3pd) could be used, which provides more accurate data than the smaller basis sets, in particular regarding vibrational wavenumbers [49]. In addition, two protein models of the E2P catalytic site were constructed.…”

Two sides of the same coin: How enzymes distort substrates and vice versa. An infrared spectroscopic view on pyruvate kinase and Ca2+-ATPase

“…The advantage of small catalytic site models over more comprehensive ones is the possibility to use large basis sets. The basis set used in our study -6-311++G (3df, 3pd) -was able to calculate the wavenumbers of two PO 3 2− containing molecules in aqueous solution with high accuracy [57]. Therefore, we decided to use this basis set without scaling the vibrational wavenumbers.…”