Using Stuttgart basis set improves reproducibility



This paper describes the density functional theoretical study on oxorhenium(V) complexes incorporating quinoline or isoquinoline carboxylic acids. The study is aimed at understanding causes for the difference in stabilities of particular isomers of [ReOX(N–O)2] species, where N–O symbolizes carboxylate ligand chelating to the oxorhenium core through N and O atoms. The structural, spectroscopic, and electronic properties were investigated by means of DFT to corroborate with the available experimental data from literature.

Original Procedure:

The study is focussed on identifying the most stable isomer of [ReOX(N–O)2] species. The selection parameters for the model complexes is mentioned by the authors as “For studies, we choose four compounds [ReO(OMe)(2-qc)2] (1), [ReOCl(2-qc)2] (2), [ReO(OMe)(1-iqc)2] (3), and [ReOCl(1-iqc)2] (4), reported previously. Three of them 2, 3, and 4 were found to be isomers of type D, whereas complex 1 showed the geometry A”.

Geometry optimizations for all the molecules without any symmetry restrictions were done at B3LYP/ LANL2DZ level of theory with additional d (with exponent a = 0.3811) and f (with exponent a = 2.033) functions for Re and 6-31G basis set for the other atoms. For Cl, P, O, and N diffuse and polarization functions were added. The optimized parameters were compared to the data available for such systems.

For geometry optimizations and TD-DFT calculations, the solvent effect (MeOH, MeCN, and THF) was simulated using the polarizable continuum model (PCM).

Practical observations:

The values for Re=O bond lengths are reproducible for all the complexes as listed in Table 1. The v(Re=O) values for all the complexes shifted by about 10% similar to what the authors have listed in Table 2. When I applied Stuttgart basis set for Re, the results were much closer to the experimental values in both cases. Further, the impact of the relative positions of Ooxo and X is more prominent in this case. The error percentage in Re-X is also found to be lower.

The stability of isomer A in case of complex 1 was reported by the authors as “For complex [ReO(OMe)(2-qc)2], the energy difference between isomers denoted as A and D is remarkably small (0.88 kcalmol-1), indicating that formation of isomer A is kinetically rather than thermodynamically controlled”. Using ST basis set for Re, the difference was found to be 0.93 kcalmol-1 which is slightly higher, but still confirming the kinetic stability of A.  Similar results were obtained for complex 2.

The TD-DFT studies were found to be quite similar to the ones reported by the authors. Use of St basis set also did not show much difference.

The HOMO and LUMO for all the complexes assessed by the methods indicates that the role of Re=O bond is significant in the global chemical reactivity parameters. In light of this, use of ST basis set is recommended.


In general, the calculations are well reproducible and helpful in establishing the kinetic stability of the oxo-rhenium molecular structures in crystals.

Use of ST instead of LANL2DZ improves the Re=O bond length parameter without significantly diverging from the effectiveness of the latter in predicting the stabilities.

Global chemical reactivity parameters could be assessed better by ST basis set.

Overall, it is evenly matched between the LANL2DZ and ST for predicting the different parameters and I would recommend using both of these.

Leave a Comment