Optical Spectra and Yields of Solvated Electrons and Metal-Electron Species in Alkali Metal Systems BY J. W. FLETCHER AND W. A. SEDDON Physical Chemistry Branch, Chalk River Nuclear Laboratories, AECL, Chalk River, Ontario KOJ IJO, Canada Received 2nd December, 1976 The effect of solvent on the optical spectra of solvated electrons, e; , cation-electron aggregates, (M+, e;), and alkali metal anions, M-, is discussed for a variety of alkali metal systems. With increasing solvent polarity the optical band maxima shift toward the blue with the solvent dependence decreasing significantly in the order (M+, e;)> e; > M-. The relative differences are interpreted in terms of ion-pairing and cation solvation. In ethylamine + ammonia mixtures only the e; band exhibits a pronounced dependence on solvent composition consistent with preferential solvation by NH3 molecules.Initial yields of e; measured by pulse radiolysis in methylamine (MA), ethylamine (EA) and isopropylamine (IPA) correspond to G(e;) N 2, whereas the fraction of electrons escaping the spur decreases markedly in the order MA > EA > IPA. The results indicate that the decay processes for e; in MA (and NH3) are slow compared with the rate of diffusion. In EA and IPA the decay is markedly temperature dependent and at temperatures approaching the freezing point is comparable with that in MA. This leads to an escaped yield approaching that observed for MA. The chemistry of alkali metal solutions has been studied extensively and excellent background information is contained within the proceedings of the Colloque Weyl I-IV.1-4 Pulse radiolysis of alkali metal salts in amine and ether solutions has substantiated that solvated electrons, e ; , cation-electron aggregates of stoichiometry M, and alkali metal anions M- coexist in equilibri~rn.~ e; + M+ + M M + M + M- -+ M+.(1) M + e ; + M - (2) (3) The precise nature of the species M is still controversial '-lo but for the purposes of this paper will be considered as an ion-pair (M+, e;). The optical band maxima for each species are solvent dependent with, in general, the most pronounced shifts with polarity being in the order (M+, e;) > e; > M-. This paper discusses first, the role of the solvent on the various optical spectra and its significance with respect to the structure of e;, (M+, e;) and M- and secondly, the radiolytic yield of e; in the solvent series ammonia, methylamine (MA), ethyl- amine (EA) and isopropylamine (IPA) and its relation to other solvent systems.EXPERIMENTAL Details of the experimental facilities and preparative techniques have been described previously."J2 In brief, solutions were irradiated with 0.3 pus (width at half-height) electron pulses from a 2.5 meV Van de Graaff accelerator or with 20 ns pulses from a Linac.13 Optical absorptions were monitored in quartz cells with an optical path length of 0.5 or 1 .O cm.J . W. FLETCHER AND W. A. SEDDON 19 RESULTS AND DISCUSSION OPTICAL SPECTRA Fig. 1 shows the effect of solvent composition on the relative shifts of the optical band maximum, vmax cm-l for e; in a wide variety of solvent mixture^.^^-^^ The data are plotted as a function of the electron fraction of the more polar molecule to correct for the relative change in energy absorbed by the two components.This scale 0 10 20 30 40 50 60 70 80 90 100 electron fraction % of p o l a r component FIG. 1 .-Relative shift of the optical band maxima, vmax cn1-l for e; in mixed solvent systems. The more polar component-corresponding to larger vmax, is quoted last. A, n-hexane + ethanol and 3-methylpentane + ethanol; By dioxan + H,O; C, tetrahydrofuran (THF) + H20; D, dimethyl- sulphoxide (DMSO) + H20 ; E, hexamethylphosphoramide (HMPA) + H20 ; F, ethylenediamine (EDA) + HzO; G, 2-methyltetrahydrofuran (MTHF) + EDA; H, diethylether (DEE) + EDA; I, THF + EDA; J, HMPA + EDA; K, NH3 + H20 and ND3 + D20; Z, EDA + ethanol.is approximately proportional to the volume fraction.23 Fig. 2 shows a comparable plot for e; in C2H5ND, + ND3 and EA + N2H4 mixtures and M in C2H5ND2 + ND3 and THF + EA solutions. In this case M refers to the (Na+, e;) ion-pair produced in solutions containing the solvent bases NaND2 and NaBD, respectively. It is evident from fig. 1 that in many cases there is a wide departure from ideal behaviour. The effect is particularly pronounced in polar + non-polar mixtures such as alcohol + alkanes 14*15926-29 or water + h y d r o c a r b ~ n l ~ . ~ ~ solutions where the e; spectrum closely resembles that of the polar component, even at very low concentra- tions of the polar molecule. This has been interpreted as being due to presence of small clusters or aggregates of the polar molecules which act as pre-existing trap^.^^,^'*^* As the differences in polarity become less the effect is diminished, indicating that the solvation sphere of e; is more typical of the average composition of the liquid.At the other extreme in NH3 + H20 mixtures some other effect, possibly the formation of NH40H, favours the localisation of e; towards NH3 molecules.20 OPTICAL SPECTRA AND YIELDS OF SOLVATED ELECTRONS Referring now to fig. 2, it is clear that the effect observed for e; in CH3CH2NH2 + NH3 mixtures is analogous to that in alcohol + alkane solutions, even though differences in solvent polarity are by no means as pronounced. We suggest the explanation is similar and relates to the existence of NH3 clusters acting as preferential traps for e;.This also seems to be the case for EA + NzH4 mixtures with, in this case, NzH4 30 forming preferred clusters but with much less specificity. 0 10 20 30 40 50 60 70 80 90 100 e l e c t r o n f r a c t i o n % of polar component FIG. 2.-Relative shift of the optical band maxima, v,,, crn-l, for e; in A, CzH5ND2 3- ND3. B, EA + NzH4 and C,D, for (Na+, e;) in C2H5NDz + ND3 and THF + EA mixtures, respectively; In A and B the shift with increasing NH3 or NzH4 is toward higher energies whereas for C and D the band shifts to lower energies with increasing ND3 or EA. For the (Na+, e;) species in partially deuterated EA + NH, mixtures the deviation from ideality is less pronounced than for e; indicating that the ion-pair has a solvation shell which is more representative of the bulk composition.This is perhaps not too surprising since the spectrum of the aggregate involves the solvation sphere of both the cation and e;. The red shift in v,,, with increasing NH3 con- centration also correlates with the decrease in atomic character observed by electron spin resonance, again implying a looser, well solvated Considering now the M- species, Lok et aZ.32 have shown an excellent correlation between the shift to lower energies in v,,, for Na‘ and K- with a wide variety of solvents. We have utilised this correlation to establish an empirical measure of increasing solvent polarity in the order of di-isopropylether (DIPE) < di-ethylether (DEE) < hexamethylphosphoramide (HMPA) < tetrahydrofuran (THF) < di- methoxyethane (DME) < diglyme < EA < 1,2 propanediamine (PDA) < ethylene- diamine (EDA).Fig. 3 illustrates the absolute shifts of the M’ band maxima5*22g31*32 as a function of solvent polarity for the series Na-, K-, Rb- and Cs-. Corre- sponding values are also shown for e; 5*32 and the charge transfer to solvent (CTTS) band for the iodide ion.32*33 In the latter case the dotted line shown for I- corre-J . W . FLETCHER AND W. A . SEDDON 21 sponds to the relative solvent dependence of 1.65 for e; to I- obtained in more polar solvents.33 In each case a linear correlation with solvating power is obtained. An increase in the slope therefore corresponds to an increasing participation of the solvent in the overall structure of e; or M-.The scatter exhibited by the CTTS iodide band probably reflects the complications due to ion-pairing effects in solvents of low p ~ l a r i t y . ~ ~ * ~ ~ On this basis it is interesting to note that no such deviation D ---n--o------~--- l o t a b c d e f g h i j increase inempirical so [vent polarity FIG. 3.-Absolute shift of the optical band maximum, ymax cm-’, for M-, e; and I- as a function of increasing solvent polarity. A, Na-; B, K-; C, Rb-; D, Cs-; E, e;; F, I-; a, DIPE; b, DEE; c, HMPA; d, THF; e, DME; f, diethylamine (DEA); g, diglyme; h, EA; i, 1,2- propanediamine (PDA); j, EDA. The slope shown for F corresponds to 0.6 that of E observed in more polar solvents (see text). The symbol, +, represents I- data for which v,,, is suspect due to ion-pairing.Similarly, 7, refers to a value for vmax e; which is probably too high because of experimental difficulties due to solvent absorption. For clarity the ordinate for E is displaced towards higher energies by 7000 cm-l and that for F towards lower energies by 27 000 cm-l. occurs for the M- species and furthermore, to our knowledge, no evidence has been obtained which would indicate that the M- band maxima are dependent on either Mf or M- concentration. It would appear therefore that in solution any association with M+ is very loose and solvent ~ e p a r a t e d , ~ ~ ’ ~ ~ a feature which is also consistent with M- being a centrosymmetric species.35 M f M- + M-, M+ (loose). The relatively smaller effect observed for Cs” is consistent with a significant lack of solvation for the Cs+ cation relative to Na+ or K+.34*37 A further point can be derived from fig.3 in that if the correlation extends to22 OPTICAL SPECTRA AND YIELDS OF SOLVATED ELECTRONS NH3 for which vmax (e;) = 5570 cm-1 at 25"C,12 then the Na- peak position should be observed at or about the same wavelength (690 nm) as in EA. Earlier results3' on the optical spectra of Na- observed in EA + NH3 alkali metal solutions containing <7% NH3, extrapolate instead to a band maximum of (1300-1600 nm) for the Na- species in pure NH3. In the EA + NH, mixtures referred to in fig. 2 we observe no shift in v, (Na-) with increasing NH3 concentration but rather a decrease in the equilibrium concentration of Na-. Our observation is in agreement with the results of H o h l ~ t e i n ~ ~ * ~ ~ and Dalton,41 which show no significant shift in vmaX for Na-, but instead a decreasing concentration of Na- in alkali metal solutions containing <40 mol % NH3 in EA.This leads us to the conclusion that the metal-independent band maxima observed in alkali metal ammonia solutions42 do not involve the formation of a species structurally equivalent to M-, but instead represent the formation of a different spin-paired species which absorbs in the infrared. In concluding this section it should be noted that in all cases reported to date the optical spectrum of (M+, e;) lies intermediate between those of e; and M- observed in the same solvent. The band maximum is very solvent dependent with the magnitude of the blue shift from the e; band being associated with a decrease in atomic ~haracter.~~ In other words a decreasing polarity of the solvent increases the interaction of e; with the cation and is reflected by an increasing blue shift of the (M+, e;) absorption band.ELECTRON YIELDS The concepts of various types of electron states ranging from quasi-free (e;)qf or dry (e-)dry to partially solvated or weakly trapped (e;),,, or (e;)damg to deeply trapped or fully solvated (e;) are currently in v o g ~ e . ~ ~ * ~ ~ In water and methanol the yield of solvated electrons corresponds to G(e;) = 4.6 (100 ps) and 2.3 (30 ps) molecules per 100 eV, r e s p e ~ t i v e l y . ~ ~ ~ ~ ~ Since G (total ionisation) is approximately 5, the fraction of electrons which become solvated and escape the spur decreases significantly with decreasing solvent polarity.Consequently we have examined the effect of polarity on the yield and escape probability of e; in the series NH3, MA, EA and IPA. Picosecond flash photolysis of these solvents at temperatures between -40 and -80°C indicates the solvation process is complete in - 6 ps.47-49 This time scale is much faster than that observed in alcohols at similar low temperature~,~~*~~ but comparable to that observed in water. In ammonia, G(e;) = 3.3 & 0.3 independent of temperature from -75 to 23°C and from 3 to 500 ns.12352953 The yields of e; in MA, EA and IPA, expressed as the product of G(e;) and the extinction coefficient E e s at the absorption maximum, are shown in fig. 4-7 as a function of dose and temperature immediately following a 20 ns or 300 ns pulse.In MA the yields are independent of temperature from 20 to -89°C (m.p. -94°C) and with decreasing dose extrapolate back to a common value of GE = 6.5 & 0.3 x lo4 (fig. 4). In EA and IPA, with decreasing temperature two time domains can be resolved; a fast initial process, independent of dose attributed to reactions in the spur, and a slower second order component due to decay processes in the bulk solution. At low doses (G0.5 x 1019 eV dm-3) or when the initial decay is 290% complete the bulk processes become more pseudo first order presumably due to the presence of impurities or a significant difference in the ratio of remaining e; to oxidising component. The fraction of e; escaping the spur processes increases with decreasing temperature such that in EA (m.p.-84°C) the value for GE in the bulk solution ranges from -2 x lo4 at 20°C to -4.5 x lo4 at -80°C (fig. 5 and 6). Distinct differences are evident from MA to EA and IPA.J . W. FLETCHER AND W. A. SEDDON 23 Corresponding values for IPA (m.p. --10loC) extend from -1 x lo4 at 20°C to 4.5 x lo4 at -97°C (fig. 7). In both EA and IPA the extrapolated value at the end of a 20 ns pulse approaches 5.5 st 0.5 x lo4. At low temperatures the optical band maxima for e; can be clearly resolved with relatively minor differences in band shape between the different amine~.’~ No evidence for spectral shifts with time were observed after a 300 ns pulse. Assuming then that E e s is independent of temperature the observed differences in GE can then be taken to reflect changes in G(e;).1 I 6 5 4 U I E a 3 X Q 2 1 t i m e after p u l s e / s FIG. 4.-The effect of dose per pulse on the yield of e; in methylamine at 20°C; A, 0.3; B, 0.5; C , 1.2; D, 3.6; E, 6.4; F, 9.0; G, 17.6; H, 23.0 x eV dmA3, respectively. Values for G(e;) can be estimated on the basis of the solute and temperature independence of the yields in MA. Assuming G(e;) = 2,” then E ~ ; - 3 x lo4 for each amine. These values are summarized in table 1 along with the corresponding yields in H20, alcohols, N2H4 and NH, for comparison. Within the ammonia-amine series the results exhibit a gradual decrease in G(e;) with decreasing solvent polarity and static dielectric constant (0,). However the effect of increasing D, within the series is more pronounced than observed, for example, in polar organic liquids including the alcohol^.^^^^^ Comparing the amine and liquid NH3 data one must infer that there is very little spur decay of e, in NH,.This is probably due to a combination of the relatively high mobility of the ammoniated electron58 (approximately ten times greater than H,O) and its slow rate of reaction with the positive i o 1 P This would also seem to apply to MA where the diffusion rate is apparently faster than the competing rate processes for spur decay. In EA and IPA the corresponding spur processes appear to be much faster than in MA. However, with decreasing temperature these rates are slowed down, thereby enhancing the escape probability, giving an observed G(e;) comparable with MA.The rela- tively high values obtained for G(e;) in the lower dielectric amines, as compared with the alcohols, presumably reflects the slower decay within the spur for the amine series. Table 2 summarizes the measured second order rate constants for the decay24 OPTICAL SPECTRA AND YIELDS OF SOLVATED ELECTRONS U I 52 x w c;l 10-8 10-7 10-6 10-5 lo-" time a f t e r p u l s e / s FIG. 5.-The effect of dose per pulse on the yield of e; in ethylamine at 20 and -80°C. A-D, -80°C at 1.6, 5.6, 9.6 and 23.0 x eV dm-3, respectively. E-J, 20°C at 0.5, 1.3, 3.2, 6.8, 8.9 and 23.0 x 10'' eV dm-3, respectively. 1 1 I I U 1 s1 x w u 1 I 10-8 10-6 10-L t i m e after p u l s e / s FIG, 6.-The effect of temperature at doses G0.5 X 1019 eV dm-3 on the yield of e; in methylamine and ethylamine.A and B, methylamine at -89 and 20°C, respectively. C-G, ethylamine at -80, -70, -50, -30 and 20°C, respectively.J . W. FLETCHER AND W. A . SEDDON 25 t i m e after pulse / s FIG. 7.-The effect of temperature and dose per pulse on the yield of e; in isopropylamine. A-E, -93°C at 0.1,0.25,0.8,1.5 and 23 x lo1' eV dm-3; F-H, -70°C at 0.7,2.2 and 23 x 10'' eV dm-3; I-K, -33°C at 0.8,4.4 and 23 x 10'' eV dm-3; L, -14°C at 23 x 10'' eV dm-3; M-0,20°C at 2.4, 6.6 and 23 x 10'' eV dm-3. TABLE ~.-COMPAFUSON OF THE INITIAL SOLVATED ELECTRON YIELD, G(e;) AND THE YIELDS ESCAPING SPUR PROCESSES, G(e;) ESCAPED, IN VARIOUS SOLVENTS. IN MA, EA AND PA WE ASSUME AN EXTINCTION COEFFICIENT Ees- = 3 x lo4 dm3 mol-' crn-l. SOURCES OF THE OTHER DATA ARE REFERRED TO IN THE TEXT temp ./"C G (total ionisation) G(G-) G(e ;) escaped 20 - 15 -30 - 60 - 80 H2O N2H4 CH30H NH3 MA EA IPA 5.4 4.6 4.8 4.8 4.8 4.7 3.4" 2.3 3.3 2.2 1.8 1.8 2.7 2.6 1.2 3.3 2.2 0.66 0.33 3.3 2.2 3.3 2.2 0.9 0.66 3.3 2.2 1.26 1.03 2.2 1.53 1.43 a Ref.(61). TABLE 2.-sECOND ORDER RATE CONSTANTS FOR THE NON-SPUR DECAY PROCESSES OF e, IN NH3 AND AMINES AS A FUNCTION OF TEMPERATURE. EXPERIMENTAL VALUES OF k/c ARE ALL CORRECTED WITH RESPECT TO cmax = 3 x lo4 dm3 mol-1 cm-l rate constant, k x 10-1°/dm3 mol-' s-l temp./"C NH3 MA EA IPA 20 2 10 150 300 - 30 8 60 70 -60 6 10 20 -80 4 5 726 OPTICAL SPECTRA AND YIELDS OF SOLVATED ELECTRONS of e; in the bulk solution. If these values are considered to be representative of the spur processes then the relative loss of e; in the spur increases significantly in the order of IPA > EA > MA.Interestingly, these rate constants are all comparable at -80°C. It is also significant that while G(e,) is -2 in the pure amines the formation of Na- and biphenyl anion observed in basic s o l u t i o n ~ ~ ~ ~ ~ corresponds to a yield of e; of 4.8. Thus there appears to be either an initial yield of " dry " electrons, G(e-d,y) -2.8, which in the absence of base do not become solvated, or alternatively, the additional source of e; results from the reaction of the solvent base with radicals,59 or possibly H atoms. Further studies are in progress on this aspect but at the present time the question remains unresolved. On the basis of the results in H20, alcohols and the amines it would seem reasonable to expect G (total ionisation) in NH3 of -5. To-date we have no experimental evidence for such a yield.This merits further study on a picosecond time scale. Finally it is worth reporting that preliminary results in EA + NH3 mixtures give G(e;) E ~ ; - 2 x lo4 at concentrations of NH3 <2% electron fraction (\(5 mol dm-3). However, over this same concentration range the optical band has clearly shifted toward that of e; in NH3 (fig. 2). In pure ammonia,55 GE = 1.5 x lo5 so it appears, as in the alcohol + alkane mixtures,27 that the yield does not change in concert with the optical spectrum. At this ammonia concentration the preferential solvation process should be complete on a picosecond time scale27*60 indicating that the escape probability of e; is not related to the trap depth as monitored by vmaX but rather by the bulk dielectric properties.The authors acknowledge the cooperation and assistance of Dr. H. A. Gillis in Thanks are also due to Mr. F. C. Sopchyshyn the use of the NRC linear accelerator. and Mr. J. J. Jevcak for their technical assistance. Metal-Ammonia Solutions, Colloque Weyl I, ed. G. Lepoutre and M. J. Sienko (Benjamin, New York, N.Y., 1964). Metal-Ammonia Solutions, Colloque Wevl 11, ed. J. J. Lagowski and M. J. Sienko (I.U.P.A.C.) (Butterworths, London, 1970). Electrons in Fluids, Colloque Weyl 111, ed. J. Jortner and N. R. Kestner (Springer-Verlag, Berlin, 1973). Electrons iri Fluids-the Nature of Metal-Ammonia Solutions, Colloque Weyl IV, ed. B. Craw ford, J.Phys. Chem., 1975, 79, 2789. J. W. Fletcher and W. A. Seddon, J. Phys. Chenz., 1975, 79, 3055. K. Bar-Eli and T. R. Tuttle, Jr., J. Clzem. Phys., 1964, 40, 2508. J. L. Dye and L. R. Dalton, J. Phys. Chem., 1967,71, 184. R. Catterall and P. P. Edwards, J. Phys. Chem., 1975, 79, 3010. Wiley and Sons, New York, 1974), vol. 2, p. 133. * T. R. Tuttle, Jr., J. Phys. Chenz., 1975, 79, 3071. lo M. Szwarc and J. J. Grodzinski, Ions and Ion-Pairs in Organic Reactions, ed. M. Szwarc (J. l1 J. W. Fletcher, W. A. Seddon and F. C. Sopchyshyn, Canad. J. Chenz., 1973, 51, 2975. l2 W. A. Seddon, J. W. Fletcher, J. Jevcak and F. C . Sopchyshyn, Canad. J. Chem., 1973, 51, 3653. l3 N. V. Klassen, H. A. Gillis and 6. G. Teather, J. Plzys. Chern., 1972, 76, 3847. l4 R. R. Hentz and G .A. Kenney-Wallace, J. Phys. Chem., 1972, 76, 2931. l5 R. R. Hentz and G. A. Kenney-Wallace, J. Phys. Chem., 1974, 78, 514. l6 T. J. Kemp, G. A. Salmon and P. Wardman, Pulse Radiolysis, ed. M. Ebert, J. P. Keene, A. J. Swallow and J. H. Baxendale (Academic Press, London, 1965), p. 247. l7 J. H. Baxendale and M. A. J. Rodgers, J. Phys. Chem., 1968,72, 3849. Is L. M. Dorfman, F. Y. Jou and R. Wageman, Beu. Bunsenges. Phys. Chem., 1971,75, 681. l9 F. Y. Jou and L. M. Dorfman, J. Chem. Phys., 1973,58,4715. 2 o J. L. Dye, M. G. DeBacker and L. M. Dorfman, J. Chem. Phys., 1970,52, 6251.J . W . FLETCHER AND W. A . SEDDON 27 'l E. A. Shaede, L. M. Dorfnian, G. T. Glynn and D. C. Walker, Canad. J. Cliem., 1973, 51, 22 P. S. Childs and R. R. Dewald, J. Pliys. Chem., 1975, 79, 58.23 E. I. Mal'tsev, A. M. Koulkes-Pujo, A. V. Vannikov and N. A. Bakh, High Energy Chem., 24 R. Olinger and U. Schindewolf, Ber. Bunsenges. Phys. Chem., 1971, 75, 693. " T. K. Cooper, D. C. Walker, H. A. Gillis and N. V. Klassen, Canad. J. Chem., 1973, 51,2195. 26 J. R. Brandon and R. F. Firestone, J. Phys. Chem., 1974,78, 792. "J. H. Baxendale and E. J. Rasburn, J.C.S. Faraduy I, 1974, 70, 705. 28 J. H. Baxendale, J. P. Keene and E. J. Rasburn, J.C.S. Faraday I , 1974, 70, 718. 29 J. H. Baxendale and P. H. G. Sharpe, Chem. Phys. Letters, 1976, 41, 440. 30 W. A. Seddon, J. W. Fletcher and F. C. Sopchyshyn, Canad. J. Chem., 1976,54, 2807. 31 W. A. Seddon, J. W. Fletcher and R. Catterall, Canad. J. Chem., 1977, 55, 2017. 32 M. T. Lok, F. J. Tehan and J.L. Dye, J. Phys. Chem., 1972,76,2975. 33 M. F. Fox and E. Hayon, Chem. Phys. Letters, 1974, 25, 511. 34 J. Smid, Ions and Ion-Pairs in Organic Reactions, ed. M. Szwarc (Wiley Interscience, New York, 35 J. L. Dye, C. W. Andrews and J. M. Ceraso, J. Phys. Chem., 1975,79, 3076. 36 M. Szwarc, Ions and Ion-Pairs in Organic Reactions, ed. M. Szwarc (Wiley Interscience, New 37 D. N. Bhattacharyya, C. L. Lee, J. Sniid and M. Szwarc, J. Phys. Chem., 1965, 69, 608. 38 S. Matalon, S. Golden and M. Ottolenghi, J. Phys. Chem., 1969, 73, 3098. 39 von G. Hohlstein and U. Wannagat, Z. Anorg. Chem., 1956, 284, 191. 40 von G. Hohlstein and U. Wannagat, Z. Anorg. Chem., 1956, 288, 193. 41 L. R. Dalton, J. D. Rynbrandt, E. M. Hansen and J. L. Dye, J. Chem. Phys., 1966,44, 3969. 42 J. L. Dye, Metal-Ammonia Solutions, ed. J. J. Lagowski and M. J. Sienko, Colloque Weyl 11, 43 R. K. Wolff, J. E. Aldrich, T. L. Penner and J. W. Hunt, J. Phys. Chem., 1975,79,210. 44 W. J. Chase and J. W. Hunt, J. Phys. Chem., 1975,79,2835. 45 C. D. Jonah, M. S. Matheson, J. R. Miller and E. J. Hart, J. Phys. Chem., 1976,80, 1267. 46 D. W. Johnson and G. A. Salmon, Canad. J. Chem., 1977, 55,2030. 47 D. Huppert and P. M. Rentzepis, J. Chem. Phys., 1976, 64, 191. 48 D. Huppert, P. M. Rentzepis and W. S. Struve, J. Phys. Chem., 1975,79,2850. 49 D. Huppert, W. S. Struve, P. M. Rentzepis and J. Jortner, J. Chem. Phys., 1975, 63, 1205. J. H. Baxendale and P. Wardman, Chem. Comm., 1971,429. 51 J. H. Baxendale and P. Wardman, J.C.S. Faraday I, 1973,69, 584. 52 Farhataziz and L. M. Perkey, J. Phys. Chem., 1975, 79, 1651. 53 J. Belloni, P. Cordier and J. Delaire, Chem. Phys. Letters, 1974, 27, 241. 55 W. A. Seddon, J. W. Fletcher, F. C. Sopchyshyn and J. Jevcak, Canad. J. Chem., 1974, 52, 56 E. Hayon, J. Chem. Phys., 1970,53,2353. 57 G. R. Freeman and J. M. Fayadh, J. Chem. Phys., 1965,43, 86. st? J. Jortner, Ber. Bunsenges. Phys. Chem., 1971, 75, 696. 59 J. L. Dye, M. G . DeBacker, J. A. Eyre and L. M. Dorfman, J. Phys. Chem., 1972,76,839. 6o A. Mozumder, J. Phys. Chem., 1972,76, 3824. 3905. (Eng. Trans.), 1975, 9, 168. 1972), vol. 1, p. 86. York, 1972), vol. 1, p. 1. (I.U.P.A.C.) (Butterworth, London, 1970), p. 1. W. A. Seddon and J. W. Fletcher, in preparation. 3269. J. Delaire, P. Cordier, J. Belloni, F. Billiau and M. 0. Delcourt, J. Phys. Chem., 1976, 80, 1687.