T. A. CLA?LTON and J. OAKES Deparimenr
Received 1 June 1967 Variation in the sodium hyperfine preted in terms of an equilibrium
coupling constant of sodium 2 : 6-dimethylbenzosemiqumone between two structurally different contact ion-pairs.
Variation of the sodium hyperfine coupling constant in the electron spin resonance spectrum of the sodium salt of 2: 6-dimethylbenzosemiquinone in t-pentanol has been observed. The central quartet is not complicated by overlap with neighbouring hyperfine lines and has been studied in some detail. This quartet (fig. 1) proved interesting for two reasons: i) The linewidths of the outer components were broader than the inner components. This pattern of lines is characteristic [l] in ion-pairs where the gegen-ion migrates between inequivalent sites. In this case the sites are almost certainly analogous to those occurring in the contact ion-pairs  of the sodium salts of P-benzosemiquinone (aNa = 0.28 gauss) (31 and durosemiquinone @Na = 0.85 gauss)  in t-pentanol. This is the first example of an equilibrium between two types of structurally different contact ion-pairs. Solvent participation has always been invoked to explain similar behaviour in other ion-pair systems [5-73. ii) The separation between the inner components is larger than that for the other,adjacent components of the quartet. At 20°C this difference is 0.03 gauss which decreases as the temperature increases. This is the first time that this sort of
variation in the hyperfine coupLing constants has been observed, and is a function of the rate of migration of the gegen-ion between inequivalent sites of the radical. The observed line shifts are probabIy best interpreted as arising from a distortion of the LQrent&m line shape due to the exchange process. Since the ‘crossover’ in the first derivative representation of the absorption spectrmn is normally defined as the Iine pa&ion, a slight distortion of the line shape will shift the position of the ‘crossover’. The migrationsof the sodium ion induces a 0.07 0.06
= 0.05 % & ;;; 0.04 8 z! 20 0.03 a 0.02
Fig. 1. Central quartet of the electron spin resonance spectrum of 2 : 6-dimethyibenzosemiquinone in bpentanol at 20°C.
Fig. 2; Difference in the separations of adjacent components of a quartet with a hyperfine coupLing constant 0.5 gauss as a function of the lsetime r = rArB /i:rAfTB), -&here 7A and TB are the lifetimes of the two structut~lly different ion-pairs. The fractionaL population ratio for each curve is indicated.
T. A. CLAXTON and J. OAKES
change in its hyperfine coupling constant of about 0.5 ga!lss if the ion pairs involved are analogous to their symmetrical counterparts. We have used the equation of Gutowsky and Holm  to plot the .difference in the separations of adjacent members of the quartet as a function of the lifetimes of each ion pair (fig. 2). A complete analysis of the line width variation of the whole spectrum  gives tt.e fractional population ratio of the sites as 9:l in favour of the site remote from the methyl groups, with a lifetime of 10-5 set (T =10-y set), which shows that the observations reported here are not only plausible but also quantitative. It is possible that this type of observation will be quite rare, since some of the necessary requirements are: i) a group of lines.which are not overlapped with other components, ii) the sites involved in the exchange must be inequivalent, iii) T should be about 0.15 x 1tY6 set, iv) the fractional population ratio should be large.
Thanks are due to the Science cil for financial support to J. 0.
REiF?ZRJZNCES [l] M.C.R.Symons, J.Phys.Chem.71 (1967) 172.  T.R.Griffithe and M.C.R.Symons, Mol.Phys.3 (1960) 90.  T.A.Claxton, J.Oakes and M.C.R.Symons, unpublished.  T.E.Gough and M.C.R.Symons, Trans.Faraday Soc.62 (1966) 269.  N.Hirota, J.Phys.Chem.71 (1967) 12’7.  P.B.Xyscough and F.P.Sargent, J.Chem.Soc.(S) (1966) 900.  R.F.Adams, N.M.Atherton, A.E.Goggins and C.M.GooId, Chem.Phys.Letters l(1967) 48.  H.S.Gutowsky and C.H.Holm, J.Chem.Phys.25 (1956) 1228.