The electron affinities of the silicon fluorides SiFn (n=1–5)

Abstract

Several independent density functional methods have been employed to determine the molecular structures and total energies of SiFn and SiF−n (n=1–5). Three significant measures of neutral–anion separation are reported: the adiabatic electron affinity, the vertical electron affinity, and the vertical detachment energy of the anion. The first Si–F ligand dissociation energies D(Fn−1Si–F), D(Fn−1Si−–F), and D(Fn−1Si–F−) as well as the harmonic vibrational frequencies of SiFn and SiF−n are also reported. Trends in the predictions of the different DFT methods are discussed. Self-consistent Kohn–Sham orbitals were obtained using various exchange correlation functionals and a double-ζ plus polarization basis set augmented with diffuse s-type and p-type functions. The method (BHLYP) based upon the Becke half-and-half exchange functional and the Lee–Yang–Parr correlation functional predicts molecular geometries in best agreement with experiment, while the other methods tend to produce bond lengths that are slightly longer. The BHLYP vibrational frequencies are also superior to those obtained via the other three DFT methods utilized. In previous studies of AFn molecules, the density functional methods have predicted electron affinities a few tenths of an electron volt above experiment with the BHLYP value being the best. Although experimental information concerning the electron affinities of the SiFn series is scarce, the BHLYP method does continue to predict electron affinities that are lower than the other methods. The BHLYP adiabatic electron affinities are 0.83 eV (SiF), 0.42 eV (SiF2), 2.50 eV (SiF3), and −0.22 eV (SiF4). The (unknown) Si–F bond distances for the anions are 0.076–0.088 Å longer than their respective neutral counterparts. The SiF−3 anion is more strongly pyramidal than SiF3. The SiF−4 species is predicted to lie energetically above SiF4, despite its experimental identification. No significantly bound minimum was found for the SiF5 molecule, although SiF−5 is predicted to exist and M+SiF−5 has been experimentally observed. The predicted vertical detachment energy of SiF−5 is 8.54 eV with the BHLYP method.

The electron affinities of the perfluorocarbons C2Fn, n=1–6

“…Unlike the exceptional vibrational frequency predictions of the BHLYP method for the SiF n series, 6 the BHLYP method consistently overestimates the vibrational frequencies given in Table IX and Table X. In the neutral C 2 F n series, there are 46 experimentally known fundamental vibrational frequencies.…”

“…DFT has proven valuable in its ability to treat larger molecular systems than conventional ab initio methods, as well as in its ability to more evenly treat charged and uncharged molecules. Our research group has recently investigated DFT predictions of the molecular structures and energies for the following neutral/anion series: ClF n /ClF n Ϫ (nϭ1 -7), 4 SF n /SF n Ϫ (nϭ1 -7), 5 SiF n /SiF n Ϫ (nϭ1 -5), 6 and PF n /PF n Ϫ (nϭ1 -6). 7 The adiabatic electron affinity of C 2 F n (EA ad ), the vertical electron affinity of C 2 F n (EA vert ), and the vertical detachment energy of C 2 F n Ϫ ͑VDE͒ have been calculated as the difference of total energies according to the following formulas:…”

Sputtered molecular fluoride anions (M=B, Si, P, K, Ti, In, W, Pb)

“…2b). Theoretical work [29] predicts the following electron affinities A for the SiF n molecules: 0.83 eV (n = 1), 0.42 eV (n = 2), 2.50 eV (n = 3), À0.22 eV (n = 4); similar values were reported in another study [30] which, in addition, computed for SiF À 5 a very high value of A = 6.5 eV. Interestingly, another theoretical investigation [31] gives A = 5.1 eV for the SiF 6 molecule; this appears to be at variance with the very low intensity of that anion observed in the present work.…”

The selenium oxygen clusters SeOn (n=1–5) and their anions: Structures and electron affinities

“…When predicting molecular energies, structures, and electron affinities, there are many theoretical approaches, but considering both reliability and computational expense, gradient corrected density functional theory is effective for predicting electron affinities of many inorganic species, such as the SiF n /SiF n À , PF n /PF n À , SF n / SF n À , ClF n /CIF n À , GeF n /GeF n À , AsF n /AsF n À , SeF n /SeF n À , and BrF n / BrF n À systems [21][22][23][24][25][26][27][28], The reliability of the predictions for electron affinities with DFT methods was comprehensively discussed in the recent (2002) review of Rienstra-Kiracofe et al [29] They reviewed the theoretical predictions of electron affinities with six DFT methods (BHLYP, B3LYP, B3P86, BP86, BLYP, and LSDA) and showed that the average deviation from experiment for electron affinities with the B3LYP and BLYP methods is only 0.15 eV for a set of 91 molecules. They also suggested that B3PW91 and BPW91 methods might outperform the B3LYP, BLYP, and BP86 functionals.…”