On the Magnetic Susceptibility of Fluorine

Abstract

The isotropic magnetic susceptibility, χiso, for F2 has been measured experimentally for the first time and found to be (−9.627 ± 0.062) × 10-6 (cgs). The value was measured on a paramagnetic gas analyzer (PGA) configured to be compatible with the aggressive F2. The PGA was shown to respond linearly with χiso and was calibrated using the response from several pure gases and their literature χiso values. The major source of error in the analysis is the inaccuracy of the literature χiso values. The experimental result for F2 was compared to ab initio calculations using both Hartree−Fock (HF) theory and gradient-corrected density functional theory (DFT) theory with a variety of basis sets. DFT calculations were found to converge to a value much closer to the experimental value compared to HF calculations, underlining the importance of the many-electron correlation effect in F2.

Relativistic and electron‐correlation effects on magnetizabilities investigated by the Douglas‐Kroll‐Hess method and the second‐order Møller‐Plesset perturbation theory

“…Ruud et al32 estimated a calibration correction factor of 1.07 for the experimental results of Barter et al33 from their nonrelativistic GIAO including multiconfigurational self‐consistent field (MCSCF) calculations of the magnetizability of Ar. It is expected that the factor would change taking into account relativistic effects 64, 65. However, the corrected experimental values reported for the noble gases are highly consistent with the present DKH2( V )SF and DKH2( V ) values (Table 2).…”

Comment on “On the Magnetic Susceptibility of Fluorine”

“…It has been pointed out, on the basis of ab initio calculations, that there is a calibration error in the data of Barter et al, and that the results of ref should be scaled by a factor 1.07 to correct for the incorrect magnetizability of argon used to calibrate the experimental setup. Rescaling the results of Barter et al would lead to a magnetizability for methane of −18.6 ppm cgs, which is in much better agreement with the value obtained by Cheng et al., − 18.4 ppm cgs, on the basis of their own calibration measurements.…”

“…In summary, we believe that modern ab initio data for small molecules today are of an accuracy higher than that of the older magnetizability measurements still often used to calibrate modern experimental determinations of magnetizabilities. On this basis we have recalibrated the recent magnetizability of fluorine determined by Cheng et al, 1 and we propose that a more accurate value based on their experimental data would be -10.03 ( 0.06 ppm cgs. We have also proposed new experimental estimates for the isotropic magnetizability of CO and CO 2 , and it would clearly be of interest to verify our proposed recalibration through high-level ab initio calculations of the magnetizability of CO, F 2 , and CO 2 .…”

“…In a recent paper, Cheng et al presented an experimental determination of the isotropic magnetizability of the fluorine molecule. In contrast to many earlier experimental determinations of isotropic magnetizabilities, − care was exercised in order to reduce the errors arising from the use of potentially inaccurate magnetizabilities when calibrating the experimental apparatus.…”

“…In Table we have collected the magnetizabilities estimated by Cheng et al from their experimental calibration, together with the experimental reference data they used for the calibration curve. We also report in Table the theoretical reference data available for these molecules, which we have correlated the PGA responses in ref against.…”