The reliability of density functional theory (DFT) in the determination of the isotropic hyperfine coupling constants (hfccs) of the ground electronic states of organic and inorganic radicals is examined. Predictions using several DFT methods and 6-31G, TZVP, EPR-III and cc-pVQZ basis sets are made and compared to experimental values. The set of 75 radicals here studied was selected using a wide range of criteria. The systems studied are neutral, cationic, anionic; doublet, triplet, quartet; localized, and conjugated radicals, containing 1H, 9Be, 11B, 13C, 14N, 17O, 19F, 23Na, 25Mg, 27Al, 29Si, 31P, 33S, and 35Cl nuclei. The considered radicals provide 241 theoretical hfcc values, which are compared with 174 available experimental ones. The geometries of the studied systems are obtained by theoretical optimization using the same functional and basis set with which the hfccs were calculated. Regression analysis is used as a basic and appropriate methodology for this kind of comparative study. From this analysis, we conclude that DFT predictions of the hfccs are reliable for B3LYP/TZVP and B3LYP/EPR-III combinations. Both functional/basis set scheme are the more useful theoretical tools for predicting hfccs if compared to other much more expensive methods.

Nitrogen hyperfine coupling constants (hfccs) of organic radicals have been calculated by density functional theory (DFT) methodology. The capability of the B3LYP functional, combined with 6-31G*, TZVP and EPR-III basis sets, to reproduce experimental nitrogen coupling constant data has been analyzed for 109 neutral, cationic and anionic radicals, all of them containing at least one nitrogen atom. The results indicate that the selection of the basis set plays an important role in the accuracy of DFT calculations of hfccs, mainly in relation with the composition of the primitive functions and the quantum number of those functions. The main conclusion obtained is the high reliability of the scheme B3LYP/6-31G* for the prediction of nitrogen hfccs with very low computational cost.

In a previous paper (Hermosilla, L.; Calle, P.; Garcia de la Vega, J. M.; Sieiro, C. J. Phys. Chem. A 2005, 109, 1114), an adequate computational protocol for the calculation of isotropic hyperfine coupling constants (hfcc's) was proposed. The main conclusion concerns the reliability of the scheme B3LYP/TZVP//B3LYP/6-31G* in the predictions of hfcc's with low computational cost. In the present study, we gain insight into the behavior of the above functional/basis set scheme on nuclei of the third row, for which few systematic studies have been carried out up to the present date. The systems studied are neutral, cationic, anionic, localized, and conjugated radicals, containing (29)Si, (31)P, and (33)S nuclei. After carrying out a regression analysis, we conclude that density functional theory (DFT) predictions on the hfcc's of the third-row nuclei are reliable for B3LYP/TZVP by using an optimized geometry with B3LYP/6-31G* combination. By comparison with other much more computationally demanding schemes, namely, B3LYP/cc-pVTZ and B3LYP/cc-pVQZ, we conclude that the B3LYP functional in conjunction with the TZVP basis set is the most useful computational protocol for the assignment of experimental hfcc's, not only for nuclei of first and second rows, but also for those of the third row.

The performance of DFT methodology to predict with accuracy the isotropic hyperfine coupling constants (hfccs) of aromatic radicals containing (14)N nucleus is investigated by an extensive study in which 165 hfccs, belonging to 38 radical species, are obtained from calculations with B3LYP and PBE0 functionals combined with 6-31G*, N07D, TZVP, and EPR-III basis sets, and are compared to the reported experimental data. The results indicate that the selection of the basis set is of fundamental importance in the calculation of (14)N hfccs, whereas there is not so great an influence on the accurate computation of that parameter for (1)H nuclei. The values of the calculated (14)N coupling constants of aromatic nitroxide radicals using DFT methodology are noticeably lower than the experimental ones. A very simple relation to predict these hfccs with high accuracy is proposed on the basis of the present results, as an interesting alternative to the highly computationally demanding integrated approaches so far used.

The sonolysis of water and some organic liquids such as ethylene glycol, methanol and chloroform in the presence of oxygen, at 20 and 475 kHz ultrasound frequencies has been investigated by the ESR-spin trapping technique. 5,5-Dimethyl-1-pyrroline-N-oxide (DMPO), 3,3,5,5-tetramethylpyrroline-N-oxide (TMPO) and N-tert-butyl-alpha-phenyl nitrone (PBN) were able to trap superoxide radical anion, generated as the result of the sonication of the organic media. The addition of superoxide dismutase (SOD) resulted in a dramatic decrease of the ESR signal intensity of the superoxide radical adduct. In addition, the thermolysis of the liquids under ultrasound was shown by ESR detection of the spin adducts of the radicals formed by homolytic fragmentation. Occasionally, the nature of the detected spin adduct was dependent on the sonication time or on the frequency of the ultrasonic radiation. Experiments carried out in the presence of 2-methyl-2-nitrosopropane (MNP) resulted in the detection of radicals originating from thermal decomposition of the spin trap, showing its lability under ultrasonic radiation.