J. Chem. Soc., Faraday Trans. 1, 1982, 78, 3645-3657 Pulse Radiolysis of Solutions of trans-Stilbene Radical-anions and Ion-pairs in Tetrahydrofuran BY JOHN R. LANGAN AND G. ARTHUR SALMON* University of Leeds, Cookridge Radiation Research Centre, Cookridge Hospital, Leeds LS16 6QB Receiced 5th April, 1982 On pulse radiolysis of solutions of trans-stilbene (t-St) in THF the radical-anion of t-St is formed by reaction of e; with t-St [reaction (S)] with k , = (1.16 k0.03) x 10" dm3 mol-' s-l. The transient absorption spectrum observed with I,,, at 500 and 720 nm is attributed to the unassociated St--. The subsequent decay of the radical-anion is accounted for by reaction with the counter-cation of THF formed on radiolysis and with radiolytically generated radicals; rate constants for these processes are estimated.Addition of sodium tetrahydridoaluminate (NAH) results in the radical-anion being associated with Na+ as a contact ion-pair and a shift of Amax to 490 nm. In the presence of the lithium salt the absorption spectrum of the radical-anion reverts to 500 nm and this is interpreted in terms of formation of solvent-separated ion-pairs. On pulse radiolysis of solutions containing NAH the main reaction forming St*- is that of (Na+, e;) ion-pairs with t-St [reaction (12)] with k , , = (1 -29 k0.04) x 10'" dm3 mol-I s-'. In addition there is a delayed formation of St*- over a period of microseconds and a possible mechanism for this is considered. The presence of tetrahydridoaluminate salts also greatly enhances the stability of St*- and at high doses per pulse little decay was observed over 700 p s .Reasons for the enhanced stability of St*- are considered. The variation of G(St'-) with "AH] was studied and was found to attain a plateau value of 2.0 at the higher concentrations. Pulse radiolysis of NAH/THF solutions containing crown ether yielded an absorption spectrum closely resembling that of e;, but kinetic studies on the solutions containing trans-stilbene indicated that the species is an ion-pair between e; and Na+ in which the sodium is complexed with the crown ether, i.e. (Na+C, e;). This species reacts with trans-stilbene with a rate constant of (3.86k0.07) x 10'" dm3 mol-l s-l. Pulse radiolysis of solutions of aromatic compounds (Ar) in tetrahydrofuran (THF) is known to result in the formation of the radical-anion of the aromatic compound by reactions (1) and (2)f-3 THF- THF(H)++e; (1) (2) e;+Ar -+ Ar*-.However, the lifetime of the radical-anions generated in this manner is generally short due to the occurrence of the fast neutralization reaction (3) between the radical-anion and the counter-ion, which is also formed in reaction (1)3 (3) Ar' - + THF(H)+ + Ar(H) + THF. The objective of the present study was to find conditions under which the lifetime of Ar*- is sufficiently extended that it would be possible to study the relatively slow reactions of the radical-anions of polymerizable aromatic olefins which are likely to be of importance in the early stages of the anionic polymerization of these compounds. trans-Stilbene (t-St) was chosen as the solute for this study since its radical-anion, which can be prepared in THF by reduction with metallic sodium, is stable and its absorption spectrum has been well ~haracterised.~* Previously studies1> have indicated that the tetrahydndoaluminate salts of sodium 118 3645 F A R 13646 PULSE RADIOLYSIS OF l'TanS-STILBENE or lithium are able to scavenge the positive ion formed on radiolysis of THF [reaction (4)], thereby extending the lifetime of any transient reduced species: (4) Pulse-radiolysis6-s and flash-photolysisg studies with solutions of alkali-metal salts in THF have demonstrated the formation of alkali-metal-cation-electron ion-pairs, (M+, e;), and alkali-metal anions, M-, which are therefore expected to be inter- mediates in the formation of radical-anions in the presence of tetrahydridolaluminate salts.THF(H)+ + A1H; -+ THF + AIH, or THF + AlH, + H,. EXPERIMENTAL The techniques of pulse radiolysis employed in this laboratory have been described else- where.1°-13 Chance-Pilkington filters were placed between the analysing light source and the sample to prevent photolysis of the sample in the cell, which was of 1 cm path length throughout the study. The methods of purification of THF6 (Hopkin and Williams, AnalaR) and trans-stilbenelO (Kodak) have been described. Dicyclohexyl- 18-crown-6 ether (Lancaster Synthesis, PCR Ltd) was used as supplied. The tetrahydridoaluminate salts (Alfa, 95 %) were used as supplied except when used in conjunction with the crown ether, when they were recrystallised from THF.Solutions of the salts were filtered under vacuum before preparing the samples for radiolysis. All solutions were degassed by three freeze-pumpthaw cycles. RESULTS SOLUTIONS OF tYaTZS-STILBENE IN T H F The end-of-pulse absorption spectrum observed on pulse radiolysis of solutions of trans-stilbene in THF, which is shown in fig. 1, agrees with that of the radical anion h/nm FIG. 1 .-(a) End-of-pulse absorption spectrum induced in a solution of lo-* mol dm-3 trans-stilbene in THF by a 50 ns, 15 Gy pulse. (b) Absorption spectrum 1 ps after irradiation of a mol dm-3 solution of trans-stilbene in NAH/THF with a 0.6 ps, 40 Gy pulse. of trans-stilbene.*q5 Using 25 ns, 10 Gy pulses the yield of radical-anion, assuming ~ 5 0 0 = 6.1 x lo4 dm3 mol-1 crnl * increased from G = 0.22 to 0.70 as the concentration of trans-stilbene was varied from 1.5 x mol dm-3.This dependence of G(St'-) on [t-St] was not studied in detail, but results from two effects, namely (i) at low concentrations of trans-stilbene the solute is unable to scavenge all the escaped electrons at the dose used and (ii) at high concentrations of solute additional spur electrons are scavenged. Two small absorption peaks at 360 and 390 nm were also toJ. R. LANGAN A N D G. A. SALMON 3647 observed which are probably due to radicals of trans-stilbene formed in the geminate neutralization process. All the absorptions decayed completely within several microseconds. The rate of reaction of e; with trans-stilbene [reaction ( 5 ) ] was measured by monitoring the formation of the stilbene anion at 500 nm e;+St + St*-.( 5 ) For trans-stilbene concentrations from 1.5 x mol dm-3 the formation of St*- following irradiation with 10 Gy pulses of 25 ns duration was found to obey first-order kinetics. The observed first-order rate constants so obtained were proportional to the concentration of trans-stilbene over the concentration range (2-30) x mol dm-3 and yielded k, = (1.16kO.3) x loll dm3 mol-1 s-l. The decay of the anions would be expected to follow second-order kinetics if they were distributed homogeneously through the medium and reacted only with cations. In fact the decay, monitored at 500 nm, gave second-order plots, but the slopes showed some variation with dose and trans-stilbene concentration and the extent of the decay which gave a linear fit varied from as short as 1.5 up to at least 2.5 half-lives.Comparable results were obtained when monitoring St*- at 710 nm. The deviation from a strict second-order rate law is ascribed, as elsewhere,1.3 to reaction of the radical-anion with radiolytically produced radicals. The observed values of the slopes? of the second-order plots varied from 3.9 x lo6 to 9.2 x lo6 s-l, with a mean value of 6.0 x lo6 s-l. The reaction scheme proposed to account for the formation and decay of St*- in the pulse radiolysis of solutions of trans-stilbene in THF is presented in table 1. to 3 x TABLE REACTIONS AND RATE CONSTANTS IN THE PULSE RADIOLYSIS OF ~~uw-STILBENE IN THF no. reaction rate constant /dm3 mol-l s-' ref. 5 e;+St+St'- 1.16 x 1011 this worka 7 e; + R -+ product 3.0 x 1 O 1 O 8 9 St'-+THF(H)+ 3 product 4.2 x 1011 this workb 10 St'-+R -+product 3.0 x 109 this workb 6 e; + THF(H)+ -+ product 2.0 x 10'2 2 8 R+R+R, 5 .o ~ 109 8 a Direct measurement; computer fit. Using these data and assuming G(e;) = 0.39, G(R) = 6 and &(St'-) = 6.1 x lo4 dm3 mol-l cm-l, a computer program has been written to calculate the concentration of each species formed as a result of the radiolysis. The program incorporates a least-squares refinement routine in two parameters for the evaluation of unknown rate constants. Reproductions of experimentally observed absorbance changes with the computed points superimposed upon them are given in fig. 2. The good agreement between experiment and calculation suggests that the reaction t k / d is the slope of the plot of A-' against t, where k is the rate constant in units of dm3 rno1-I s-1, E is the molar decadic extinction coefficient of the absorbing species in units of dm3 mo1-I cm-l and 1 is the optical path length of the cell in cm.118-23648 PULSE RADIOLYSIS OF tranS-STILBENE I" (v c 9 8 0 0 0 0 c OD 0 (D 0 U 0 (v 0 0 0 0 ameqiosqeJ. R. LANGAN A N D G. A. SALMON 3649 mechanism in table 1 is essentially correct, although it should be noted that the entity denoted by R may in fact represent more than one radical. The decay of the small absorption peak at 390 nm, which is believed to be due to stilbene radicals, was found to fit a second-order plot, although again there was a small variation of the slopes with dose, which suggests that reactions with stilbene radical-anion and possibly other radicals were occurring.The mean value of the slopes of the second-order plots was found to be 1.7 x lo6 s-l. SOLUTIONS WITH ADDED TETRAHYDRIDOALUMINATE SALTS The pulse radiolysis of a solution of trans-stilbene in THF saturated with sodium tetrahydridoaluminate (NAH/THF) resulted in the spectrum shown in fig. 1. Comparison with the spectrum observed in neat THF reveals a hypsochromic shift in the absorption peak of the radical-anion to 490 nm. The broad band around 710 nm was also observed. When doses per pulse in excess of 40 Gy are used the radical-anion is long lived and very little decay was observed over a period of 7OOp.s. Typical oscilloscope traces recorded during this experiment are given in fig.3 (a) and (b). The L t,,,,,,,,, FIG. 3.-Oscilloscope traces recorded in the pulse radiolysis of trans-stilbene in NAH/THF ordinate : % absorption pulse size dose/Gy I/nm per division (a) 0 . 6 , ~ 40 490 17.9 (b) 0 . 6 ~ s 40 490 17.0 ( d ) 50x1s 10 890 1.4 (e) 0 . 2 p 16 490 7.1 (f) 0.2 PS 16 490 7.1 (c) 50 ns 10 490 4.1 abscissa : ps per division 100 1 0.1 0.1 100 13650 PULSE RADIOLYSIS OF frUnS-STILBENE stability of the anion is indicative of the efficiency with which the AlH, ions remove the counter-ions formed during the pulse [reaction (4)]. The absence of absorptions below 400 nm due to radicals evident in the absence of NAH is probably also a consequence of this reaction. A similar spectrum was recorded when the sodium salt was replaced by lithium tetrahydridoaluminate (LAH), except that A, reverted to 500 nm.From these results it is deduced that in these systems St.- exists as a contact ion-pair (Na+, St.-) when associated with Na+, but as solvent-separated ion-pairs (Li+/ /St' -) when associated with Li+. For comparison purposes similar solutions were subjected to y-radiolysis and the spectra recorded are shown in fig. 4. It was noted that for the solutions containing 3 X/nm FIG. 4.-Radiation-induced absorption spectrum obtained by (a) y-irradiation of 1 0-2 mol dm-3 trans- stilbene in NAH/THF, (b) pulse radiolysis of mol dm-3 rrans-stilbene in NAH/THF, and (c) y-irradiation of 6.3 x lop3 mol dm-3 trans-stilbene in LAH/THF. Each spectrum is normalised to A,,,. NAH, Amax underwent a bathochromic shift with irradiation time to a limiting value of 508 nm.This shift is probably caused by the formation of an absorbing species which gives rise to the shoulder around 570 nm. The size of this shoulder is much smaller in solutions containing LAH or in solutions containing NAH which were irradiated with a pulsed electron beam. High dose rates thus effect a marked reduction in the formation of the species absorbing at around 570 nm. The nature of this species has not been investigated. However, in the y-irradiation of pyrene in LAH/THF1 the absorption of the anion peak (493 nm) was not seen, but instead one at 530 nm was observed which was attributed to the formation of a complex, probably Py*AlH,*THF-. In the pulse radiolysis of that system a post-pulse growth of absorption at 530 nm was noted as partial decay of the anion occurred.It may be that in the radiolysis of trans-stilbene solutions the absorption around 570 nm is due to the complex St*AlH,*THF-. However, no growth related to the absorption at this wavelength was observed when recording the spectrum (fig. 1) by pulse radiolysis. The dependence of the complex formation on dose rate may be explained if AlH, is removed by reaction (1 l), which would be more effective at the high dose rates involved in pulse radiolysis: 2AlH, + Al,H,. (1 1)J. R. LANGAN AND G. A. SALMON 365 1 Using a trans-stilbene concentration of 5 x lop3 mol dm-3 the yield of the radical- anion was measured as a function of NAH concentration using pulse radiolysis and the results are given in fig.5. At the higher salt concentrations a maximum value of G(St'-) = 2.0 _+ 0.1 was reached. This is in accordance with previous work where G(Na+, e;) = 2 was recorded.8 0 0.2 0.4 0 . 6 0 . 8 1 . [NAHl/mol dm-3 FIG. 5.-Yield of trans-stilbene radical-anion as a function of NAH concentration, [t-St] = 5 x mol dm-3. The yield of radical-anions in NAH/THF was shown to be independent of initial trans-stilbene concentration by both gamma and pulse radiolysis. The results are depicted in fig. 6, which also includes some values obtained using LAH/THF. The higher values of G found in LAH/THF are ascribed to the greater solubility and dissociation constant of the lithium salt in THF and are in agreement with other results.1t8 The difference between the values found in gamma and pulse radiolysis is probably due to the higher dose rate of the latter technique which enhances the second-order removal of radicals.For solutions with [t-St] < 5 x lop3 mol dm-3 and using 50 ns pulses and sweep speeds of 100 ns per division it proved possible to observe the broad absorption band of (Na+, e;). Monitoring the decay of this absorption at 880 nm provided a convenient means of following the kinetics of the reaction between (Na+, e;) and trans-stilbene, reaction (1 2) (Na+, e;) + t-St + (Na+, St *-). (12) The decay of (Na+, e;) was matched by the concomitant formation of stilbene radical-anions at 490 nm [see fig. 3(c) and (41. For trans-stilbene concentrations in the range 2.5 x lop4 to 2.2 x lop3 mol dm-3 the decay of absorption at 880 nm was found to obey a first-order rate law and a plot of the observed rate constants against trans-stilbene concentration yielded k,, = (1.29 0.04) x 1O1O dm3 mol-1 s-l.The growth of absorption at 490 nm also obeyed first-order kinetics and yielded3652 PULSE RADIOLYSIS OF t!YLEFZS-STILBENE 3*0* 0 -5 -4 -3 -2 log [trans-stilbene I FIG. 6.-Yields of stilbene radical-anion obtained in gamma and pulse radiolysis of trans-stilbene solutions : 0, pulse radiolysis of NAH/THF solution; x , y-irradiation of NAH/THF solution; 0, y-irradiation of LAH/THF solution; A, pulse radiolysis of LAH/THF solution. k,, = (1.30 & 0.08) x 1O1O dm3 mol-l s-l, in good agreement with the value obtained from the decay of (Na+, e;). As was noted [fig. 3(a) and (b)], there was very little decay of the radical-anion following the pulse.However, this seems to result from the high dose used. On a microsecond timescale, using a 10 Gy pulse, two distinct phases are apparent in the behaviour of the absorption after the pulse. Initially, over ca. 5 ps, there is an increase in the absorption followed by a decay lasting for several hundred microseconds. These features are illustrated in fig. 3(e) and (f). Three spectra were recorded, at the end-of-pulse, at 5 ps and at 600ps after the pulse, respectively. The maximum absorbance for all three occurred at 490 nm in NAH/THF and 500 nm in LAH/THF. At the longer sweep speeds the decay at 570 nm appeared to be offset by the formation of the complex referred to previously. The growth in absorbance at Amax after the pulse gave linear plots when fitted to a first-order rate equation and some results are given in table 2.The subsequent decay obeyed neither first- nor second-order kinetics and was greatly influenced by the dose. TABLE 2.-RATE CONSTANTS FOR THE POST-PULSE GROWTH OF ABSORBANCE [ trans-stilben-el solvent /mol dm-3 dose/Gy k'/s-' % slow growth ~ ~~ ~~ ~~ NAH/THF 1.13 x 15 4.07 x 107 24 6.02 x 10-3 7 8.63 x 107 15 LAH/THF 9:23 x 10-3 12 7.8ox 107 25 8.85 x 10-3 12 6.66 x lo7 __J. R. LANGAN AND G. A. SALMON 3653 Similar observations have been made elsewherel in the pulse radiolysis of pyrene in LAH/THF. The kinetic evidence, taken together with the fact that at high dose rates neither effect is observed, strongly suggests that the reactions involve radicals.Pulse-radiolysis studies of amine solvents containing alkali cations have described post-pulse growths of absorption lasting several microsecond~.'~ Further proof of the stability of the ion-pair formed in the pulse irradiated solutions was provided by e.s.r. The width of the signal and the line splitting were consistent with the published In addition the solutions were able to initiate the polymerization of butadiene. The resulting polymer was found not to have incorporated stilbene showing that initiation was by electron transfer rather than addition. EFFECT OF ADDED CROWN ETHER Addition of dicyclohexyl-l8-crown-6 (DCH) to a solution of NAH in THF is expected to result in the complexation of the sodium cations and hence in the in- creased dissociation of the salt.Pulse radiolysis of a solution containing NAH (6.1 1 x mol dm-3) and DCH (1.03 x lo-* mol dmP3) yielded a spectrum which was similar to that of the solvated electron2T8 with GE = 2.5 x lo4 at 2000 nm.? The absorption band of (Na+, e;) with 3,,,, = 890 nm which is observed in NAH/THF solutions was absent. For kinetic reasons (see below) it is believed that the spectrum resembling that of e; is not due to e;, but is the spectrum of (Na+ C, e;), i.e. an ion-pair between e; and the sodium cation complexed with crown ether. The absorption decayed completely over a period of ca. 10 p s [see fig. 7 (a)] by first-order kinetics with an observed rate constant of 2.51 x lo5 s-l. This value, measured at a dose of 43 Gy, is similar to that for e; at the same dose.8 Pulse radiolysis of a solution of trans-stilbene in NAH/DCH/THF is expected to yield the spectrum of the solvent separated ion-pair of stilbene anion, and this was found to be so.To ensure that the appearance of LmaX at 500 nm was not merely due to the effect of a lower salt concentration a spectrum was also recorded using only NAH at a similar concentration. The absorption maximum was then observed to be at 485 nm. The behaviour of St*- in this system followed the same pattern as in the NAH/THF system. The post-pulse changes were still present [see fig. 7(b)-(d)] and again were greatly reduced by the use of large doses per pulse. The decay observed at lower dose per pulse [fig. 7 ( d ) ] was found to fit a second-order plot with k/E = 1 x lo5 cm s-1 under the conditions used.The kinetics of formation of St*- were followed using the decay of the species at 2000 nm and the growth in absorbance at 500 nm. For concentrations of trans-stilbene from 5 x mol dm-3 the decay of absorption at 2000 nm and the growth at 500nm obeyed first-order rate laws and the observed rate constants increased linearly with trans-stilbene concentration, from which we deduce k,, = (3.86k0.07) x lo1" dm3 mol-1 s-l to 2.2 x (Na+C, e;) + t-St -+ (Na+C, St * -). (13) This rate constant is significantly different from that for the reaction between e; and trans-stilbene. Thus, although the spectrum of e; is not modified by the presence of Na+C, this latter ion is exerting a significant influence on the reactivity of e; and it seems appropriate to ascribe this to the formation of the ion-pair (Na+C, e;).The effect of NAH concentration of the yield of St*- is given in fig. 8. This shows t Gc is the product of the radiation chemical yield in units of molecules (100 eV)-' and the molar decadic extinction coefticient in units of dm3 mol-1 cm-'.3654 PULSE RADIOLYSIS OF tTaYtS-STILBENE la I l CI FIG. 7.-Oscilloscope traces recorded in the pulse radiolysis of NAH/DCH/THF [(b)-(d) with ca. mol dmV3 trans-stilbene added]. ordinate: % absorption abscissa : pulse-length/ps dose/Gy i/nm per division ps per division (a) 0.2 43 2100 2.3 1 (b) 0.2 35 500 7.2 1 (c) 0.6 140 500 17.8 100 (d) 0.2 35 500 7.3 100 2 .c h I c-' F", u 1 . 0 I 1 1 I I 1 2 3 4 ! [ NAHJ / 1 O-* mol dm-3 FIG. 8.-Yield of stilbene radical-anion as a function of NAH concentration in NAH/DCH/THF; 0.2 ,us pulses, [t-st] = 5 x lop3 mol drnp3.J .R. LANGAN A N D G. A. SALMON 3655 that G(St*-) = 2, which corresponds to the maximum value in NAH/THF solutions, is reached at a much lower salt concentration than in the absence of crown ether. This implies that the greater dissociation of the salt has enhanced the scavenging of the counter-cation [THF(H)+] by the AlH, ions. Using the dissociation constant of NAH in THF,8 the concentration of AlH, in the NAH/THF system was estimated to be 3 x mol dm-3. If it is assumed that the salt is completely dissociated in the presence of an excess of crown ether then the evidence suggests that undissociated NAH is also able to scavenge the positive ion of THF.In an experiment where sodium tetrahydridoborate was substituted for NAH, G(St'-) was found to be 1.65. However, no stabilization of St*- resulted and it seems that, as in the case of sodium tetraphenylboron,8 the reaction of BH, with the cation of THF gives a radical which can react with St*-. THERMAL GENERATION OF ANIONS During the course of this work it was noticed that when solutions of trans-stilbene in THF containing the tetrahydridoaluminate salts were left to stand at room temperature a red colour developed. Similar behaviour in pyrene solutions has been reported16 where the presence of the anion was ascertained by e.s.r. A solution of trans-stilbene (ca. l OP4 mol dmP3) in NAH/THF was allowed to stand in the dark until it had developed red colour, approximately one week, after which it was found to exhibit the absorption and e.s.r.spectra of the stilbene anion. The rate of coloration quickened when the solutions contained crown ether. DISCUSSION The spectrum of the radical-anion of stilbene recorded in the pulse radiolysis of solutions of trans-stilbene in neat THF may readily be assigned to that of the free ion. In the presence of tetrahydridoaluminate salts the anion will probably be associated with the metal cation. In the case of the sodium salt the effect of this can be seen in the occurrence of a hypsochromic shift of Amax. The absence of such a shift in solutions containing Li+ suggests that solvent-separated ion-pairs are formed in this system. The difference in the behaviour of the two cations is explicable in terms of the larger solvation energy of the small Li+ ion.For larger cations, interaction with an anion as a contact ion-pair is more favourable than interaction with solvent molecules. The most important property of a solvent in promoting the formation of solvent separated ion-pairs is its ability to coordinate alkali-metal cations, which is measured by its donicity rather than by bulk properties such as permitti~ity.'~~ The solvation of ions and ion-pairs and its effects on anionic polymerization have been described comprehensively elsew here. Computer simulations of the pulse radiolysis of solutions of trans-stilbene in NAH/THF showed that lO-l5% of the electrons were scavenged directly by the hydrocarbon. Any reaction of St*- formed in this way with Na+ in the solution to give ion-pairs would be expected to proceed with a rate constant 2 loll dm3 mol-1 s-l and hence be unobservable under the conditions of the experiment.The absorption spectrum recorded in NAH/THF will be due to the species in the following equilibria (Na+, St*-)tighte(Na+, St'-)loose e N a + + S t ' - . The small difference in A, seen here compared with that of the anion prepared by sodium-metal reduction4 may be due to the presence of some free ions or loose ion- pairs. It is known that radical-anions of hydrocarbons in THF at concentrations < ca. mol dmP3 may exist as free ions17 l8 and the concentration of anions3656 PULSE RADIOLYSIS OF trCInS-STILRENE produced by pulse radiolysis will not exceed this figure. Solvent-separated pairs are not distinguishable optically from the free ions.However, in flash-photolysis studies of extremely dilute solutions of sodium pyrenide in THF, uiz. (2-4) x mol dm-3, the dissociation of ion-pairs into free ions was suppressed by the presence of excess sodium tetraphenylboron.18 Therefore, it seems likely that the concentration of free ions in our solutions should be negligible. The reduced interaction of the solvated electron with the Na+ cation when it is complexed by crown ether is indicated by the effect of crown ether on the transient absorption spectrum and, as may be expected from this result, the spectrum of the stilbene anion in a solution containing crown ether is that of the loose ion-pair. The rate constants for the formation of radical-anions and ion-pairs of trans-stilbene are given in table 3 and are similar to those reported for other aromatic hydrocarbons.2* 7 7 l8 The rate of reaction (1 3) is consistent with the spectral data, which shows that the interaction between the sodium cation and the solvated electron is reduced by the presence of crown ether.TABLE 3.-RATE CONSTANTS FOR THE FORMATION OF STILBENE RADICAL-ANIONS reacting species rate constant/lO1° dm3 mol-’ s-’ e, 1 1.6 0.03 (Na+C, e;) 3.86 0.07 “a+, e,) 1.3 k 0.08 The variation in the observed yield of stilbene radical-anions with the concentration of trans-stilbene in THF followed the pattern established in previous pulse-radiolysis studies.l In solutions containing tetrahydridoaluminate salts the total reducing power depends on the concentration of the salt and consequently this factor governs the yield of radical-anions formed in the radiolysis.The salt increased the yield of ion-pairs by scavenging THF cations within the spur, thus making electrons available in the form of (Na+, e;). The subsequent stability of the radical-anion is also due to the AlH; ion scavenging the counter-ion as described earlier. Complexation of the Na+ cation by crown ether led to the expected increase in scavenging of the counter-ions by the AlH; ions. It seems from comparison of yields of St-- in NAH/THF with those in NAH/DCH/THF that the undissociated salt also acts as a scavenging agent. Alternatively, it may be that in NAH/DCH/THF some of the Na+ cations and AlH, anions are associated in crown-separated pairs whose reactivity lies between that of the free ions and the undissociated salt.Computer simulations of the pulse radiolysis of trans-stilbene in NAH/THF indicated that over 99.904 of the electrons reacted with trans-stilbene as either e; or (Na+, e;). This is a reflection of the extreme rapidity of these reactions, which thus prevent significant decay of the negative species by other possible pathways. The values of GE for (Na+, St.-) found in this work (see fig. 5 ) are consistent with the reported values of G(Na+, e;)* and the extinction coefficient for the radical-anion of ~tilbene.~ Hence the value G(anion) = 2.0 0.1 may be assumed for any hydrocarbon of similar electron affinity to trans-stilbene whose reaction with (Na+, e;) proceeds at a similar rate.Thus, the absorption spectra and extinction coefficients of radical-anions which are not sufficiently long lived to be observed when prepared by reduction by alkali metals may be determined and the results of such work will be reported in a subsequent paper.J . R . L A N G A N A N D G. A. SALMON 3657 As indicated earlier, the fast formation of St*- in solution containing the tetra- hydridoaluminate salts can be attributed to reactions ( 5 ) and (1 2). However, the slow formation of St*-- observed on microsecond timescales is more difficult to interpret. A similar delayed formation of e; has been observed in the pulse radiolysis of amines containing bases.14 Since the effect is reduced when larger doses per pulse are used it is tempting to speculate that radiolytically produced radicals are involved in the formation of e; by reaction with A1H; ions [reaction (14)] +AIH, + (THF-) -+ e; + AIH, (14) 3T-T 0 but further experimentation is required to establish the mechanism of this slow formation of St*-.A further difficulty concerns the effect of dose on the partial decay of St. in the NAH/THF solutions. Again theeffect ofdose suggests the involvement of radiolytically produced radicals [reaction (1 5)] R* +St*- -+ St+product. (1 5 ) However, it is thought that several cationic species are formed on the radiolysis of THFl9 and it is possible that the slow reaction with St. -- involves reaction with a cation derived from THF which is not readily scavenged by NAH. Again further experimentation is required to resolve this problem.We thank the S.E.R.C. for the award of a CASE Studentship to J. R. L., the Ministry of Defence for laboratory facilities at Waltham Abbey and Dr D. H. Richards for his interest in the work and for fruitful discussions. I J. H. Haxeniiale, D. Beaumond and M. A. J. Rodgers, Trans. Faraday Soc., 1970, 66, 1996. '? F. Y. Joii ;id L. M. Dorfman, J . Them. Phys., 1973, 58, 4715. E. A. Shaede. H. Kurihara aiid L. M. Dorfman, Int. J . Radiut. Phys. Chem., 1974, 6, 47. E. R. Zabolotny and J. F. Garst. J . .4m. C!zrm. Soc., 1964, 86, 1645. E. .4. Robinson and G. A. Salmon, J. Phys. C'hen7.. 1975, 82, 382. G. A. Salmon and W. A. Seddon, Chem. Phys. Z,ett., 1974, 24, 366. B. Bockrath and L. M. Dorfman, J . Phys. Chem.. 1973, 77, 1002. G. A. Salmon, W. A. Seddon and J. W. Fletcher, Cun. J . Chem., 1974, 52, 3295. L. J. Giling, J. G. Kloosterboer, R. P. H. Rettschnick and J. D. W. van Voorst, C'hem. Phys. Lett., 10 11 12 1 :1 14 15 18 1 7 18 19 1971, 8, 457. F. S. Dainton, E. A. Robinson and G. A. Salmon, J . Phys. Chem., 1972, 76, 3897. T. J. Kemp, J. P. Roberts, G. A. Salmon and G. F. Thompson, J . Phys. Chem., 1968, 72, 1464. D. H. Ellison, G. A. Salmon and F. Wilkinson, Proc. R. Soc. London, Ser. A, 1972, 328, 23. G. V. Buxton, J. Kroh and G. A. Salmon, J . Phys. Chem., 1981, 85, 2021. J . A. Delaire, M. 0. Delcourt and J. Belloni, Radiat. Phys. Chem., 1980, 15, 255. R. Chang and C. S. Johnson, J . C'hem. Phys., 1964, 41, 3273. D. H. Paskovich, A. H. Reddoch and D. F. Williams, J . Chem. Soc., Chem. Commun., 1972, 1195. (a) Ions and Zon Pairs in Organic Reactions, ed. M. Szwarc (John Wiley and Sons, New York, 1972), vol I. (h) M. Szwarc, Carbanions, Living Polymers and Electron Transfer Processes (Interscience, New York, 1968). M. Fisher, G . Ramme, S. Claesson and M. Szwarc, Chem. Phys. Lett., 1971, 9, 306. J. H. Baxendale, D. Beaumond and M. A. J. Rodgers, Int. J . Radiat. Phys. Chem., 1970, 2, 39. (PAPER 2/58])