Electron Spin Resonance Studies of Cation Radicals of Some Alkylphosphines BY MASAMOTO KISHIAND TAROISOBEIWAIZUhfI,* TAKASHI The Chemical Research Institute of Non-Aqueous Solutions, Tohoku University, Sendai, Japan AND FUMIOWATARI Department of Applied Science, Faculty of Engineering, Tohoku University, Sendai, Japan Received 7th January, 1975 The effect of alkyl substitution on the geometry of the phosphine cation radical has been investi- gated by e.s.r. spectroscopy. It is shown by the use of Huheey’s electronegativity parameters that the geometry of the alkylphosphine cation radicals is related to the electronegativity of the alkyl groups as in other trigonal radicals. The effects of methyl substitution on the geometry of AIH,‘ and SiH3 are also discussed in terms of Huheey’s electronegativity parameters.It is shown, further, that there is a correlation between the 31Phyperfine coupling constants of the alkylphosphine cation radicals and 31Pn.m.r. chemical shifts for the respective neutral molecules. There have been many investigations which concern the structure of AX3 typeradical^.^'^^ It has been shown 1-6 that the molecular geometry of these trigonal radicals is related to the electronegativity difference between A and X, i.e., the radicals adopt a more pyramidal structure as more electronegative substituents are introduced. Meanwhile, interesting features of the effects of methyl substitution on the geometry of PH;, SiH, and AlH; radicals have been observed. Begum et aL5 have shown that PMel flattens more than PH;, while AlMe; bends further than AIH;, and SiMe, takes nearly the same geometry as SiH3.In their study, the importance of the hyper- conjugation effect of the methyl group has been pointed out. However, despite many investigations of trigonal radicals, the effects of alkyl substitution on the mole- cular geometry of the radicals are not really clear. We have examined the cation radicals of some alkyl-phosphines by e.s.r. In this paper, we present the effect of alkyl substitution on the structure of the phosphine cation radicals, and also show there is a correlation between the 31Phyperfine coupling constants of these phosphine cation radicals and 31Pn.m.r. chemical shifts of the parent neutral molecules.EXPERIMENTAL Methylphosphine and dimethylphosphine were prepared by the deprotonation of phos-phine by KOH-dimethylsulphoxide suspension, followed by reaction with methyl iodide, and purification by fractional condensation.15 Trimethylphosphine, triethylphosphine and tri-isopropylphosphine were prepared by the Grignard method from phosphorus trichloride and the appropriate alkyl halide. These were purified by precipitating the phosphines as the silver iodide adducts and subsequently thermally decomposing the adducts in a vacuum line. Tri-n-butylphosphine was of the highest commercial grade and was used as supplied. The cation radicals of these phosphines were produced by 6oCoy-ray irradiation of de-113 E.S.R. OF PHOSPHINE RADICALS gassed sulphuric acid or dichloromethane solutions of the phosphines.E.s.r. spectra were measured with an Hitachi 771 e.s.r. spectrometer with 100kHz field modulation. Both y-irradiation and e.s.r. measurements were carried out at 77 K. RESULTS y-irradiation of sulphuric acid solutions of phosphines produces predominantly monomer cations, but with dichloromethane solutions it sometimes produces dimer cations or other radical species as well as the monomer cations. The identification of the cation radicals was made by reference to assignments by Symons et al. for the spectra of PMe; and PEtJ.’ Representative spectra observed are illustrated in fig. 1. Although the e.s.r. of cation radicals was not always observed for both the solvent systems, as far as was observed, no appreciable differences were found 1.* VALUES, ‘P HYPERFINE COUPLING CONSTANTS AND p/SRATIO FOR ALKYLPHOSPHINE TABLE CATION RADICALS radical medium Q II 91 AII/G Ai/G Aiso/G plsratio PH; a PH2Me+ H2S04 H2S04 1.993 1.987 2.014 2.012 706 634 423 348 517 443 6.43 7.59 PHMeT H2S04 1.989 2.015 624 316 419 8.67 PMe: H2S04 1.984 2.012 616 294 401 9.47 PEt: CHzClz 1.989 2.012 570 273 372 9.42 P(n-Bu)P(i-Pr): CH2C12 CH2C12 1.990 1.986 2.012 2.010 556 534 257 209 357 317 9.88 12.10 aref.(5) 105G j3300G lb FIG.1.-First derivative e.s.r. spectra for (a) PH2CHi in sulphuric acid and (b) P(i-Pr): in dichloro- methane at 77 K. x : signals from the dimer cation radical.between the hyperfine coupling constants in sulphuric acid and in dichlorornethane solutions. The e.s.r. parameters obtained are shown in table 1. Breit-Rabi correc- tions have been made for the determination of g and hyperfine coupling constants. The p and s spin densities were calculated from the isotropic and anisotropic parts of the 31P hyperfine tensors using the atomic parameters given in the literature.16 Table 1 also contains the p/sratio obtained for the unpaired electron orbital. M. IWAIZUMI, T. KISHI, T. ISOBE AND F. WATARI DISCUSSION It is seen in table 1 that there is an appreciable effect of alkyl substitution on the 31P hyperfine interaction. The result shows a marked contrast to SiH3 where no appreciable effects of methyl substitution were observed.If the hyperconjugation effect were dominant as was pointed out by Begum et uZ.,~ the pis ratios of PH;, P(i-Pr); and PMe; would increase in this order. However, the ratios increase in the order of PHT, PMe: and P(i-Pr);, suggesting that hyperconjugation is not necessarily a dominant effect for the determination of molecular geometry. The sequence of the pjs ratio shown in the table rather suggests that the electronegativity of the substituents may be more important. We have examined the relation between the electronegativities of the substit uents and the molecular geometry by using Huheey’s electronegativity parameters. l7 Huheey showed the method of calculating the electronegativity of groups based on the Hinze and Jaffk concepts for the orbital and bond electronegativities.In the calculation, he assumes that the electronegativity of an atom or group is equalized by displacement of charge upon covalent bond formation ; his electronegativity is expressed by the two terms, a+66, where a corresponds approximately to the fixed electronegativity by Mulliken’s definition, b is a constant which may be termed the charge coefficient, and 6 is the formal charge on the atom or group resulting from I I I I I I I .-~ -0.1 0.0 0.1 FIG.‘.-Plot of the pis ratio for the unpaired electron orbital against the formal charge on phos-phorus, Sp, of some alkylphosphine cation radicals. replacement of the charge by bond formation. By the use of his electronegativity parameters, and solving the following simultaneous equations, the formal charge on phosphorus in the radicals of the type of PR’R”R”’+ can be calculated.ap bp8p = a,. + bRfdRt = aR*p+ bRntdR** = + bR~~~dR~t~ 6p + &*+6R)f + 8R”’ = + 1. Fig. 2 shows a plot of pls ratio of the unpaired electron orbital against dp obtained in this way.lg The figure shows well the correlation between the p/s ratio and aP, and indicates that as the density of CT bonding electrons decreases on the phosphorus, the phosphorus 3p orbitals are more favoured for CJ bonding. This result is consistent with Pauling’s explanation of the structure of trigonal radicals and apparently 116 E.S.R. OF PHOSPHINE RADICALS indicates that the electronegativity of the alkyl groups makes a dominant contribution to the determination of the molecular geometry of the alkylphosphine cation radicals as in the cases of other trigonal radicals.It can be shown that the effects of methyl substitution on the structures of AIH;, SiH, and PEP: radicals, which are mentioned above, can also be well explained in terms of Huheey’s electronegativity parameters. Table 2 shows the formal charges on the central atoms for these radicals calculated by a similar method to that used above.2o It is seen that for AlHF the central atom loses electrons by methyl sub- stitution but in PH; it gains more electrons and it neither gains nor loses in SiH,, showing good correspondence to the observed changes in the p/sratios of the radicals by methyl substitution.It is apparent, therefore, that the structures of these radicals are also explained by the electronegativity difference argument. TABLE2.-p/S RATIO, FORMAL CHARGES ON CENTRAL ATOMS, &, AND CHANGES IN 6, BY METHYL SUBSTITUTION AIH; AlMe; SiH3 SiMe3 PH; PMeS pls ratio 3.37a 2.3ga 5.2a 5.7 a 6.43 9.47 &Vf -0.25 +0.14 -0.03 -0.02 +0.15 -0.06 SM(Me)-BM(H) -k0.39 +0.01 -0.2 1 a from ref. (5). On the other hand, it was found that there is a linear relation between the p/s ratios of alkylphosphine cation radicals and 31Pn.m.r. chemical shifts for the corre- sponding parent neutral molecules (fig. 3). It is known that the 31Pn.m.r. chemical 0 50 100 150 200 (ppm) 31Pchemical shift FIG.3.-Plot of the p/sratio for the unpaired electron orbital of some alkylphosphine cation radicais against the 31Pn.m.r.chemical shifts @.p.m. from H,P04) for the parent neutral molecules. shifts for trivalent phosphorus compounds are determined mainly by the bond angles at the phosphorus and the ionic character in the phosphorus-ligand bonds.21 To see the effect of ionic character in the phosphorus-ligand bonds, calculation of charge distribution was attempted for PH3 and PMe, by the use of Huheey’s electronega- tivity parameters, and by assuming 95 and 86 % p characters, respectively, for their c bonds based on the observed bond angles.22 The result showed that the effect of M. IWAIZUMI, T. KISHI, T. ISOBE AND F. WATARI methyl substitution on the charge distribution is very small in such neutral molecules ; the change in the formal charge on phosphorus by methyl substitution was only -0.06, suggesting that the changes in the n.m.r.chemical shifts may not be due to the changes in the ionic character for the present case. The correlation shown in fig. 3 may imply rather that the geometry of the parent neutral molecules varies in similar manner to the cation radicals, and the resulting changes in the hybridization of the phosphorus make a contribution to the observed changes in the 31Pn.m.r. chemical shifts. Quantitative explanation, however, is difficult at present. L. Pauling, J. Chem. Phys., 1969, 51, 2767. D. L. Beveridge, P. A. Dobosh and J. A. Pople, J. Chem. Phys., 1968, 48,4802. L. J. Aarons, I.H. Hillier and M. K. Guest, J.C.S. Faraduy 11, 1974, 70,167. I. Biddles and A. Hudson, Mol. Phys., 1973, 25, 707. A. Begum, A. R. Lyons and M. C. R. Symons, J. Chem. SUC. A, 1971, 2290. A. Begum, J. H. Sharp and M. C. R. Symons, J. Chem. Phys., 1970,53,3756.'J. H. Sharp and M. C. R. Symons, J. Chem. SOC.A, 1970,3084. A. R. Lyons and M. C. R. Symons, J. Amer. Chem. Suc., 1973,95,3483. T. Cole, H. 0.Pritchard, N. R. Davidson and H. M. McConnell, Mol. Phys., 1958, 1, 406. lo K. Morokuma, L. Pedersen and M. Karplus, J. Chem. Phys., 1968, 48, 4801. R. W. Fessenden, J. Phys. Chem., 1967, 71, 74. l2 C.Hesse, N. Leray and J. Roncin, J. Chenz. Phys., 1972, 57, 749. l3 R. L. Morehouse, J. J. Christiansen and W. Gordy, J. Chem. Phys., 1966,45,1751.l4 T. A. Claxton, M. J. Godfrey and N. A. Smith, J.C.S. Faraday 11, 1972, 68, 181. l5 Inorganic Synthesis XI, ed. W. L. Jolly (McGraw-Hill, New York, 1968), p. 124. l6 P. W. Atkins and M. C. R. Symons, The Structure of Inorganic Radicals (Elsevier, Amsterdam, 1967)."J. E. Huheey, J. Phys. Chem., 1965, 69,3284. (a)J. Hinze and H. H.Jaffe, J. Amer. Chem. SOC.,1962,84,540 ; (b) J. Hinze, M. A.Whitehead and H. H. Jaffk, J. Amer. Chern. Soc., 1963,85, 148; (c) J. Hinze and H. H. Jaffk, J. Phys. Chem., 1963, 67, 1501. l9 Inthe calculation, orbital hybridization the same as that estimated experimentally was used for phosphorus. 2o For each central atom, the same hybridization as that estimated experimentally was used. 21 N. M. Crutchfield, C. H. Dungen, J. H. Letcher, V. Mark and J. R. Van Wazer, Topics in Phosphorus Chemistry, vol. 5, 31PNucIeccr Magnetic Resonance (Interscience, New York, 1967), and references therein. 22 G. M.Kosolapoff and L. Maier, Organic Phosphorus Compounds (Wiley-Interscience, New York, 1972), vol. 1. (PAPER 5/025)