2 (Part ii) ELECTRON SPIN RESONANCE SPECTROSCOPY By Colin Thomson (Department of Chemistry The University,St. Andrews Fife Scotland) THEuse of electron spin resonance spectroscopy (e.s.r.) in chemistry con- tinues to increase and two important books summarise the present state of this subject. The book by Ayscough provides an extensive survey of the chemical aspects of e.s.r.,’ whilst in ‘Radical Ions’,’ a series of authors review in detail electron spin densities ion-pair equilibria semidione anions aromatic anions and cations ions containing sulphur and Group IV elements orbital de- generacy inorganic radicals complex ions and various aspects of radical ions in radiation chemistry. This book is of immense value to the researcher in e.s.r. This report makes no attempt to cover all aspects of e.s.r.and the references cited are those which seem to the reporter to be of particular significance. Among reviews of interest are the use of 170in e.~.r.,~ a series of reviews in a Discussion of the Royal S~ciety,~ the use of e.s.r. in kinetic^,^ e.s.r. of the triplet state,6 and some more general reviews of e.~.r.~-’ The present report does not deal in detail with irradiated solids polymer e.s.r. gas-phase e.s.r. nor biological applications but concentrates on e.s.r. studies of radicals in solution and their relationship to valence theory. General Theory Hyperfiie Interactions and Spin Density Calculations.- The relationship between the observed hyperfine splitting (h.f.s.) constants a and the spin density distribution in the radical continues to be of interest.For n-electron radicals the McConnell relation :lo between proton splittings (aH)and the spin density (pc) on the contiguous carbon atom or the Karplus-Fraenkel” relationship (for 3C,1’N etc.) have continued to be used. However two recent papers have examined the approxi- mations involved in the McConnell derivation. Malrieu” has extended the P. B. Ayscough ‘Electron Spin Resonance in Chemistry,’ Methuen London 1967. E. T. Kaiser and L. Kevan ‘Radical Ions,’ Interscience New York 1968. B. L. Silver and Z. Luz Quart. Rev. 1967,21,458. Proc. Roy. SOC.,1968 A 302,287-365. Ya.S. Lebedev. UsprXhi. Khini.. 1968,37 934. C. Thomson Quart. Rev. 1968. 1968 22,45 ’ D. H. Eargie jun. Analyt. Chem.1968,40 303R. * G. A. Russell Science 1968,161 423. F. Gerson Chimia (Switz) 1968,22 293. lo H. M. McConnell J. Chem. Phys. 1956,24 764. M. Karplus and G. K. Fraenkel J. Chem. Phys. 1961,35 1312. l2 J.-P. Malrieu J. Chem. Phys. 1967. 46. 1654. 18 Colin Thomson original first-order perturbation theory treatment to second-order. The addi- tional contributions are at least 25% of the first-order and this is of the same order of magnitude as the corrections proposed by Colpa and B01ton.l~ A very detailed study by Davidson and co-~orkers'~ on the C-H fragment is similar to that of Malrieu but numerical calculations were carried out for various basis sets. The authors conclude that Q" depends on both first- and second-order effects and is very sensitive to the orbital exponent of the H atomic orbital.Calculated values of QF. lie in the range -25 < Q,!& < -50 G. These authors conclude that Bolton's13 calculation may be in error. Murrell and Hinchliffe' have calculated fluorine coupling constants uF by use of the Karplus-Fraenkel' ' approach and find reasonable correlations between observed and calculated uF for a series of fluoronitrobenzene anions if the n-electron spin densities are calculated by means of SCF-CI theory. Analysis of the results shows that uF is related to the Q factors by equation 2 i.e. the splitting also depends on the overlap spin densities pCF(these are very small for hydrogen splittings). A similar calculation of QENand Q& has been carried out.l6 Spin polarization mechanisms in aliphatic fragments have been investigated by Luz" using the Dirac vector model. Vincow18 has examined the use of the Colpa-Boltonl and Giacometti Nordio Pavan' relationships in deriving an expression for the total splittings in the cyclic ions C,H,* and the radicals C,HA. An unsatisfactory feature of correlations between h.f.s. and n-electron spin densities has been the semi-empirical nature of the Q factors in view of the difficulty of accurate calculations. The influence of a-electrons has recently been taken explicitly into account in all valence electron calculations which although semi-empirical should provide a more realistic picture of the spin distribution particularly in radicals containing heteroatoms.The extended Huckel Theory2' (EHT) has been used by Drago and Petersen2 in calculations on a-radicals such as vinyl cyclopropyl etc. The calculated coupling constants are in excellent agreement with experiment. A later paper22 has considered nitrogen and proton splittings in the o-radicals Hc0,H2cN and those derived from syn-and anti-acetaldoxime benzaldoxime and 2-pyridyl. For this type of radical EHT is a considerable improvement over Hiickel theory and it has also been applied to long-range couplings in ~emidiones.~~ A related but more l3 J. P. Colpa and J. R. Bolton Mol. Phys. 1963,6 273. l4 S. Y. Chang E. R. Davidson and G. Vincow J. Chem. Phys. 1968,49,529. A. Hinchcliffe and J. N. Murrell Mol. Phys. 1968 14 147. J. C. M. Henning Chem.Phys. Letters 1968 1 678. l7 Z. Luz J. Chem. Phys. 1968,48,4186. G. Vincow J. Chem. Phys. 1967,47,2714. l9 G. Giacometti P. L. Nordio and M. V. Pavan Theor. Chzm. Acta. 1963,l. 404. 'O R. Hoffman J. Chem. Phys. 1963,39 1397. " R. S. Drago and H. Petersen jun. J. Amer. Chem. SOC. 1967,89 3978 5774. 22 R. E. Cramer and R. S. Drago J. Amer. Chem. SOC.,1968,90,4790. 23 G. R. Underwood and R. S. Givens J. Amer. Chem. SOC.,1968,90,3713. Electron Spin Resonance Spectroscopy 19 rigorous theory based on the SCF formalism is Pople’s CND0/2 25 which has been applied to fluoronitrobenzene anions26 but it is also better suited to o-radicals. This disadvantage of both EHT and CND0/2 is that they neglect the 0-n exchange integrals which are involved in the spin polarization mecha- nism in n-electron radicals i.e.in the Q factor. The most notable advance in 1968 has been the application of an extension of the CND0/2 method including such integrals (INDO meth~d),~’ to hyperfine splitting calculations. The split- ting constant is calculated directly and agreement with experiment is very good for N 0,C H and F splittings in a large variety of radicals.28 The observed differences in geometry between eH3 and eF3are accounted for by use of this method.29 Further work with this method should be of great interest. Of the numerous n-electron spin-density calculations which have been reported the use of the unrestricted Hartree-Fock (UHF) theory has in- creased and has been applied to thiophen pyrrole and furan anions,3o and to alkyl radical^.^' Since this method is based on a wavefunction which is not an eigenfunction of S2 most authors have projected out the unwanted spin components after minimisation (PUHF method).Harriman and Sando have. however reported the first n-electron calculations using the Spin Extended HF method (in which projection is carried out before energy minimisation) for allyl pentadienyl benzyl and the anions and cations of naphthalene anthracene and a~ulene.~~ Despite the greater complexity of the method the results appear to be intermediate between those given by the UHF and PUHF methods. Linewidths and Relaxation Theory.-The extensive work on linewidths by Freed and Fraenke133 has been continued by Freed,34 by means of a new method due to K~bo.~’ This method should be of use in studying rotational effects on the g-tensor.Sille~cu~~ has examined quadrupole relaxation which is believed to be important in iodine-substituted iminoxy radicals. He obtained good ON* I 24 J. A. Pople D. P. Santry and G. A. Segal J. Chem. Phys. 1965,43 S129 S136. 25 J. A. Pople and G. A. Segal J. Chem. Phys. 1966,44,3289. 26 D. W. Davies Mol. Phys. 1967 13,465. 27 J. A. Pople D. L. Beveridge and P. A. Dobosh J. Chem. Phys. 1967,47,2026. 28 J. A. Pople D. L. Beveridge and P. A. Dobosh J. Amer. Chem. SOC. 1968,90 4201. 29 D. L. Beveridge P. A. Dobosh and J. A. Pople J. Chem. Phys. 1968,48,4802. 30 N. K. Ray and P. T. Narasimhan Theor. Chim. Acta. 1968,11 156. 31 N. K. Ray and P. T.Narisimhan Chem. Phys. Letters 1968,2 101. 32 J. E. Harriman and K. M. Sando J. Chem. Phys. 1968,48,5138. 33 J. H. Freed and G. K. Fraenkel J. Chem. Phys. 1963,39 326. 34 J. H. Freed J. Chem. Phys. 1968,49 376. ” R. Kubo J. Phys. SOC.Japan 1962,17,1100. 36 H. Sillescu Mol. Phys. 1968 14 381. Colin Thomson agreement between the theoretical and experimental linewidths for the radical (1). Sillescu and Kivelson have studied relaxation theory by use of a classical model for The use of Monte Carlo methods for calculating lineshapes has been further e~plored.~' In experimental work on linewidths the 2,2'- dinitrobiphenyl anion (2)has been studied.39 This anion. produced by reduction with sodium in dimethoxyethane shows linewidth alternation in the spectrum.Different metal cations exchange at different rates between the nitro groups and give rise to linewidth effects which are dependent on the metal solvent and temperature. a;X=O b;X=S 2,2'-Dinitrobiphenyl ether (3a) and sulphide (3b) ions were also studied.40 In the first case the two rings are magnetically equivalent with respect to the protons but not with respect to nitrogen. This is the first case reported where the exchange rate is fast with respect to one coupling constant and slow with respect to the other. A technique for improving the resolution of hyperfine structure in solid-state spectra has been described which is based on Fourier transform technique^.^^ Hydrocarbon Radicals Anions and Cations.-Triphenylmethyl Ph3C has recently been studied by electron nuclear double resonance (ENDOR).42 The ring splitting constants obtained from this study are a, = -2.609 G.a = -1.143G,and up = -2.857 G and were assigned on the basis of McLachlan calculation^^^ and observed and calculated e.s.r. spectra. Several chlorine- and fluorine-substituted derivatives have also been ~tudied.~~9 44 The diphenyl- methyl radical Ph,eH45 has been reported during thermolysis of various compounds which give diphenylmethylene when the reaction is carried out in solvents with abstractable hydrogen. The value of uHfor the C-H fragment is 8.36 G which Vin~ow~~ concludes is much lower than expected on the basis of his work on the structurally similar xanthyls. Further confirmation of this work 37 H.Sillescu and D. Kivelson J. Chem. Phys. 1968,48 3493. 38 M. Saunders and C. S.Johnson jun. J. Chem. Phys. 1968,48,534. 39 J. Subramanian and P. T. Narasimhan J. Chem. Phys. 1968,48 3757. 40 R. K. Gupta and P. T. Narasimhan J. Chem. Phys. 1968,48,2453. 41 A. Hedberg and A. Ehrenberg. J. Chern. Phys.. 1968.48.4822. 42 A. H. Maki. R. D. Allendoerfer. J. C. Danner. and R. T. Keys J. Anw. Chem. SOC.. 1968. 90. 4225. 43 A. D. McLachlan Mol. Phys. 1960,3,233. 44 J. Sinclair and D. Kivelson J. Amer. Chem. SOC. 1968,90 5074. 45 D. R. Dalton S. A. Liebman H. Waldman and R. S. Sheinson Tetrahedron Letters 1968,145. 46 M. D. Sevilla and G. Vincow J. Phys. Chem. 1968,72 3635. Electron Spin Resonance Spectroscopy is needed. Vincow has extended his earlier work on tropenyl radicals C7H7m4’ to the mono- and tri- t- butyltr~penyls.~~~~~ These substituents remove the degeneracy of the unpaired electron orbital and give spectra whose hyperfine splittings are very temperature-dependent.For the tri- t-butyltropenyl the linewidth and g-value are also temperature-dependent. A very detailed analysis is given of the vibronic near-degeneracy problem in these radicals. An interesting but unexplained spectrum is obtained by treating azulene with periodic acid.” The radical is very stable and analysis of the solid indicates that the azulene rings are present. Anion radicals derived from hydrocarbons con- tinue to be of interest. Smentowski’’ has succeeded in obtaining very well- resolved spectra by use of sodium in liquid ammonia as reducing agent.Care must be taken that reduction does not proceed to the dianion. Although C6H; could not be prepared several other hydrocarbon anions such as the anions of anthracene and naphthalene were stable at temperatures -= O”c and the benzophenone anion was stable up to +40°c. A detailed study of deuterium-substituted benzene anions has a~peared.’~ The degeneracy of the benzene molecular orbitals is lifted by deuterium substitution except where the radical has a three-fold or higher symmetry as in the case of [1,3,5-2H3]benzene. The temperature dependence of the split- tings was different from that observed in C6H; and the results were analysed in terms of perturbations of the benzene orbitals as a result of substitution.Radicals ions from several unusual hydrocarbons have been described. Monohomocyclo-octatetraene anion (4)has been re-inve~tigated’~ and the results show that the early work of Katz and TalcotP4 was correct but that the splitting constants of Winstein and co-workers” were in error. (4) (5) A very different spectrum is produced from the related species bicyclo[6,1,0]- nona-2,4,6-triene (3,which agrees with calculations based on structure (5).56 The azupyrene anion (dicyclopenta[ef,kl]heptalene anion) (6)has been observed and the splitting constants a = 0.64 G a3 = -4.23 G and a4 = 0.94 G are 47 G. Vincow M. L. Morrell W. V. Volland H. J. Dauben jun. and F. R. Hunter J. Amer. Chem. SOC.,1965,87 3527. 48 G. Vincow M. L. Morrell F. R.Hunter and H. J. Dauben jun. J. Chem. Phys. 1968,48,2876. 49 w. V. Volland and G. Vincow J. Chem. Phys. 1968,48,5589. 50 A. T. Fatiadi Chem. Comm. 1968,456. ” F. J. Smentowski and G. R. Stevenson J. Amer. Chem. SOC.,1968,90,4661. ’’ R. G. Lawler and G. K. Fraenkel J. Chem. Phys. 1968,49 1126. ” F. J. Smentowski R. M. Owens and B. D. Faubion J. Amer. Chem Soc. 1968,90 1537. s4 T. J. Katz and C. Talcott J. Amer. Chem. SOC. 1966,88,4732. 55 R. Rieke M. Ogliaruso R. McLung and S. Winstein J. Amer. Chem. SOC.,1966,88,4729. 56 G. Moshuk G. PetrowskYi and S. Winstein J. Amer. Chem. SOC.,1968,90,2179. Colin Thomson in good agreement with theoretical calculations.’ An interesting reaction occurs during the alkali-metal reduction of l,l’-binaphthyl(7).The binaphthyl 12 anion initially produced undergoes further reaction to give perylene anion (via the perylene dianion).’* The naphthacene trianion spectrum reported earlier,59 has been re-studied and shown to be due to the 5,12-naphthacene- semiquinone ion,60* 61 which emphasises the important of purity and the ex- clusion of oxygen from these systems. A large number of papers on ion-pair interactions have appeared including two very detailed studies of ion pairing in the naphthalene and anthracene anion-metal cation system^.^^-^^ These papers deal with the structure and stability of the ion-pairs. The various models differentiate between the strong and weak ion-pairs which are observed and the field as a whole has been reviewed by Sym~ns.~’ The pairing theorem for alternant hydrocarbons has been further studied by use of the biphenylene cation and anion66 (8).13C Splittings indicate that the theorem is valid in this case also as it was for anthracene ions,67 but there are large differences in Q“ at the 2-position in the cation and anion which are as yet unexplained. ” A. G. Anderson jun. A. A. MacDonald and A. F. Montana J. Amer. Chem. SOC. 1968 90 2994. S. P. Solodnikov S. T. Ioffe Yu. B. Zaks and M. I. Kabachnik Zzvest. Akad. Nauk S.S.S.R. Ser. khim 1968,442. 59 K. Mobius and M. Plato 2.Naturforsch 1964,199 1240. E. T. Seo J. M. Fritsch and R. F. Nelson J. Phys. Chem. 1968,72 1829. 61 K. Mobius and M. Plato J. Phys. Chem. 1968,72 1830. 62 N. Hirota J. Amer. Chem. SOC. 1968,90 3603.N. Hirota R. Carraway and W. Schook J. Amer. Chem. SOC.,1968,90 3611. 64 C. L. Dodson and A. H. Reddoch J. Chem. Phys. 1968,48,3226. 65 J. Burgess and M. C. R. Symons Quart. Rev. 1968,22,276. 66 P. R. Hindle J. dos Santos Veiga and J. R. Bolton J. Chem. Phys. 1968,48,4703. J. R. Bolton and G. K. Fraenkel J. Chem. Phys.. 1964,40. 3307. Electron Spin Resonance Spectroscopy 23 Cation dimers first observed for naphthalene by Lewis and Singer,68 have been produced from coronene and pyrene by use of the BF3-S02 oxidising 70 An excess of reagent gives the monomer but an excess of hydro- carbon gives the dimer. Other oxidising systems have also been reported to give the cation dimer of coronene such as BF3-CH2C12 S03-S02 AlC1,-MeN02.70 The hyperfine structure is well resolved and the splitting constant is temperature-dependent.There is also a species present in this system which may be the cation trimer.70 There have been several investigations of substituted hydrocarbons. An extensive study of alkoxy bezene anions has shown that vibronic mixing is small in these ions.7’ The metal splittings observed have been used to obtain in- formation on the structure of ion-pairs. The cation radicals of p-bis(methy1thio) benzene and p-bis(ethy1thio)benzene have been prepared by reaction with A1C1,-MeN02.72 The spectra were interpreted in terms of a superposition of spectra due to the cis-(9)and truns-(10) isomers with the splitting constants shown. /Me 5.30 G /Me 5.44G S S 0 1*03G \ 1.79 G S Me/s ‘Me (9) (10) McLachlan spin-density calculations support this assignment and different g-values are observed for the two species.72 The radical PhCOi- can be produced in a flow system by reaction of benzoic acid with sodium in liquid ammonia.73 The use of this technique for other systems should be interest.Radicals Containing Halogens.-The perfluoro-analogues of aliphatic and aromatic hydrocarbons have been prepared and extensively studied in recent and in view of the interest in the mechanism of fluorine hyperfine interactions the preparation of the anions and cations of perfluoro-species is of considerable interest. This year has seen the first studies of this type of radical. Although naphthalene does not give CloH; on oxidation only the dimer cation,68 perfluoronaphthalene CloFg can be oxidised to C,,F;.Bazhin et al. reported that this cation was formed in oleum SO3,or SbF,-(Me0)2S02 but their analysis of the spectrum was in error owing to the failure to observe I. C. Lewis and L. S. Singer J. Chem. Phys. 1965,43 2712. 69 J. T. Cooper and W. F. Forbes Canad. J. Chem. 1968,46,1158. 70 H. van Willigen E. de Boer J. T. Cooper and W. F. Forbes J. Chem Phys. 1968,49 1190. J. K. Brown and D. R. Burnham Mol. Phys. 1968,15 173. ’’ W. F. Forbes and P. D. Sullivan Canad. J. Chem. 1968,46,317. 73 A. R. Buick T. J. Kemp G. T. Neal and T. J. Stone Chem. Comm.. 1968. 1331. 74 R. E. Bank ‘Fluorocarbons and their derivatives,’ Oldbourne Press London 1964. Colin Thomson the outside lines.’ Independently Thomson and MacC~lloch~~ observed a strong 21-line spectrum by use of the new oxidising medium SbClS-SO2 with splitting constants uorF= 19.0 G ugF= 4-75 G i.e. uaF -4ugP The analysis of these spectra in terms of theory and of the pronounced linewidth effects which are observed should hopefully shed more light on fluorine hyperfine interactions. Several perfluoro-anions have been observed despite the tendency of such species to lose fluoride ion F-. The ketyls of (CF,),CO perfluorodimethyl keten and trifluoroacetophenone have been studied by Russian and (CF&CO independently by Jan~en.~~ Second-order splittings and y-fluorine h.f.s. are observed in hexafluorocyclobutanone.79The anions of several fluoro- and difluoro-benzophenones have been prepared electrolytically but decafluorobenzophenone does not give the simple anion.80 Correlations with Hiickel and McLachlan spin densities have been made for fluorine splittings in fluoro benzosemiquinones.However the above results all show that the relationship between uF and pc is not of the simple McConnell type aspredicted by Murrell and Hinchcliffe.” The anion radical from perfluorobenzoselenadiazole (11) and its H-analogue have been studied.82 In this case uF is twice a for the corresponding position in the benzoselenadiazole anion. The anion radicals of several difluoro- F 0. 0-II C Fi-N -N -C F3 biphenyls have been prepared by alkali-metal reduction. 83 Various fluoro- nitroaromatic radicals have been studied by Brown and Williams,84 who also reported that perfluoroaromatic hydrocarbons could not be reduced to the anion radicals because of loss of F-.Other workers have studied fluorine splittings from CF groups in several substituted nitro benzene^.^' An interest- ing reaction occurs during electrolysis of CF,NO and CF,N02.86The radical ” N.M. Bazhin N. E. Akhmetova L. V. Orlova V. D. Shteingarts L. N. Shchegoleva and G. G. Yakobson Tetrahedron Letters 1968 4449. l6 C. Thomson and W. J. MacCulloch Tetrahedron Letters 1968 5899. l7 V. V. Bukhtiyarov and N. N. Bubnov Teor. i. eksp. Khim. 1968,4,413. l8 E. G. Janzen and J. L. Gerlock J. Phys. Chem. 1967,71,4577. 79 J. L. Gerlock J. Phys. Chem. 1968,72 1832. 8o P. H. H. Fischer and H. Zimmerman Z.Naturforsch 1968,239 1339. 81 P. H. H. Fischer and H. Zimmerman 2.Naturforsch 1968,239 1399. 82 J. Fajer B. H. J. Bielski and R. H. Felton J. Phys. Chem. 1968 72 1281. 83 A. L. Allred and L. W. Bush Tetrahedron 1968,24,6883. 84 J. K. Brown and W. G. Williams Trans. Faruday SOC. 1968,64,298. ” J. W. Rogers and W. H. Watson J. Phys. Chem. 1968,72 68. 86 J. L. Gerlock and E. G. Janzen. J. Amer. Chem. SOC..1968.90. 1652. Electron Spin Resonance Spectroscopy produced is the bis(trifluoromethy1)semidiazoxide (12).Of considerable interest is the report that Cl,’ is produced on dissolving ClF in SbF or in Olah’s ‘magic acid’ FS03H-SbF,-S02. A seven-line spectrum characteristic of two equivalent I = 3/2 nuclei is observed with a = 2.5 G. The reaction mechanism is not clear but Cll is in equilibrium with another paramagnetic species (with may be ClF;) at higher temperature.This is the first observation of a diatomic radical cation in solution.87 Radicals Containing S Se Si Ge Sn and P.-Among recent Lvork on ring compounds containing S and Se we have mentioned perfluorobenzoselena- diazole82. In an attempt to assess the dependence of sulphur x-bonding on molecular geometry the non-planar molecule dibenzo[b,f]thiepin (13) was reduced to the anion by potassium in dimethoxyethane.88 The spin distribution is similar to that in the stilbene anion8’ and the authors conclude that conjugation through S is weaker than conjugation through the vinyl residue. The natural abundance of 33S(0.74%) is low but 33Sh.f.s.can be observed in some systems if the linewidths are small. The A1C13-MeN02 oxidising system has been used to study the 33S splittings (as)in the thianthrene (14) cation.g0 Linewidth studies show that the sign of a is positive. Sullivan has extended his studies to 33S splitting in several cations of thiepin (15) 1,4-benzothiepin (16) and phenoxathiin (17).9’ The theory of the 33Ssplittings is similar to the Karplus-Fraenkel theory of 3C splittings. Shineg2 has studied the oxidation of aromatic sulphides with persulphuric acid. Radicals are ob- tained whose g-values indicate that they are not the simple cation radicals but may be hydroxylated molecules. The trithienyl radicals (18) and (19) have been prepared and their solution e.s.r.spectra recorded although not completely analy~ed.’~ ’’ G. A. Olah and M. B. Comisarow,J. Amer. Chem. SOC.,1968,90 5033 88 M. M. Urberg and E. T. Kaiser J. Amer. Chem. SOC.,1967,89 5931. 89 C. S. Johnson,jun. and R. Chang,J. Chem. Phys. 1964,41,3272. H. J. Shine and P. D. Sullivan,J. Phys. Chem. 1968,72 1390. 91 P. D. Sullivan,J. Amer. Chem. SOC.,1968,90 3618. 92 H. J. Shine M. Rahman H. Nicholson and K.K. Gupta Tetrahedron Letters 1968,5255. 93 A. Mangini. G. F. Pedulli. and M. Tiecco. Tetrahedron Letters. 1968 4941. Colin Thomson R2 I An interesting recent observation of a small radica in solution is tile electro-lytic generation of In radicals containing Group IV elements several Ph,Si- substituted polyenes benzenes and naphthalenes have been prepared.95 Many of these are very stable and 29Si hyperfine structure (4.7% natural abundance) can be observed. The results provide some evidence for carbon n-orbital to silicon delocalisation. An independent study of Ph,Si- and Ph,Ge-substituted anions has appeared.96 Dimroth has extended his earlier work on cation radicals derived from 2,4,6-trisubstituted phosphor in^^^ with the preparation of 1,l-dialkoxy- and 1,l-diaryloxy-phosphorincations and anions (20).98An impor- tant recent report of the observations of PC12 and PCl radicals in u.v.-irradiated PC1 and MePC1299 at 77°K. has appeared. These small radicals should be of interest to theoreticians concerned with phosphorus hyperfine splittings. In a similar study U.V. irradiated triethyl phosphite at 77"~ gives (EtO),PO radicals.loo Radicals Containing Nitrogen.-(a) a-radicals.The carbamoyl radical (2 1) has been produced by reaction of H*CO*NH with OH in a flow system.''' These authors differ in their conclusions from earlier work,lo2 in which the radical was believed to be H-CONH and support their assignment with EHT calculations which give very good agreement with experiment. 94 K. P. Dinse and K. Mobius Z. Naturforsch. 1968,239,695. 95 F. Gerson J. Heinzer H. Bock H. Alt and H. Seidl Helv. Chim. Acta. 1968 51 707. 96 A. L. Allred and L. W. Bush J. Amer. Chem. SOC.,1968,90,3352. 97 K. Dimroth N. Grief W. Stade and F. W. Steuber Angew. Chem. Internat. Edn. 1967,7 711. 98 K. Dimroth and W. Stade Angew. Chem. Internat. Edn.1968,7 881. 99 G. F. Kokoszka and F. E. Brinkman Chem. Comm. 1968,349. loo K. Terauchi and H. Sakurai LW.Chem. SOC.Japan 1968,41 1736. T. Yonezawa I. Noda and T. Kawamura &ll. Chem. SOC. Japan 1968,41,766. lo' P. Smith and P. B. Wood Canad. J. Chem. 1966,41 3085. Electron Spin Resonance Spectroscopy 27 Electrochemical oxidation of nitrosobenzene and its derivative^"^ gives the radical cations (22) which are o-radicals since the splitting constants shown are typical of such species i.e. a large nitrogen coupling (aN= 37.0 G) and one large proton coupling (rneta) and other smaller proton splittings due to inter- action with one or three protons. Iminoxy radicals derived from halogen-substi- tuted acetophenones exhibit halogen splittings which are very dependent on temperature and solvent.'04 1,3-Dicarbonyl compounds react with C(NO,) to give various iminoxy radicals.' O5 (b) Nitrogen heterocyclic ions.The use of Pb(OAc) as an oxidising agent has been studied in the production of cation radicals of carbazole derivatives.'06 In the presence of acids as catalysts N-methyl- N-isopropyl- and N-phenyl- carbazole give stable radicals believed to be (23) although the spectra were not resolved. L R R J (23) (24) The behaviour of acridine (24) during reduction with sodium in tetrahydro- furan is complex but a thorough study of the species involved has detected the acridine anion which had not been reported previously. '07 Reaction of formazans (RN=NCR=NNHR) with tetrazolium salts (25) results in stable tetrazolinyl radicals (25a).'08 logA variety of derivatives were (25 ) (25d studied and McLachlan spin-density calculations were carried out.The results are consistent only with the cyclic structure (25a). Several new stable nitroxides have been reported e.g. (26),'" and the a-nitronylnitroxides (27)." '* 'l2 These radicals can be readily protonated with lo3 G. Cauquis M. Gemies H. Lemaire A. Rassat and J. P. Ravet J. Chem. Phys. 1967,47,4642. '04 B. C. Gilbert and R. 0.C. Norman J. Chem. SOC.(B),1968 123. lo' C. Lagercrantz and K. Torssell Arkiv. Kemi 1968 29,203. lo6 D. H. Iles and A. Ledwith Chem. Comm. 1968,498. lo' S. Niizuma M. Okuda and M. Koizumi hll. Chem. SOC.Japan 1968,41,795. F. A. Neugebauer Tetrahedron Letters 1968,2129.F. A. Neugebauer and G. A Russell J. Org. Chem. 1968,33,2744. L. B. Volodarsky G. A. Kutikova R. Z. Sagdeev and Yu. N. Molin Tetrahedron Letters 1968 1065. J. H. Osiecki and E. F. Ullman J. Amer. Chem. SOC. 1968 90,1078. 'I2 D. G. B. Boocock R. Davey and E. F. Ullman J. Amer. Chem. SOC. 1968,90 5945. Colin Thomson 0-+I I :0>: -b-0 CF,CO,H in benzene. Nitroxides (28) from binitrones have been described,' and a large number of azomethin nitroxides (29).l l4 A series of papers by Hudson and Hussain have examined the use of hydrogen peroxide to oxidise amines to nitroxides.' '-'l8 Very stable aliphatic nitroxides such as Me,NO can be produced and a large number of nitroxides derived from aliphatic amines cyclic amines etc.have been studied. In the latter case for instance with pyrrolidine nitroxide (30),interesting linewidth effects and strong temperature dependence of the spectra yield details of the conformation of the radicals. 2-Monosubstituted 1.3-dicarbonyl compounds react with I 0' C(NO,) to give nitroxides but the unsubstituted compounds give iminoxy radicals. Photolysis of alkyl nitrites in hydrocarbon solvents gives dialkyl nitroxides and alkoxy-alkyl nitroxides.' 2o However several interesting reactions occur in these systems by use of different solvents.'21 '13 A. R. Forrester R. H. Thomson and G. R. Luckhurst J. Chem. SOC.(B),1968 1311. '14 H. G. Aurich and F. Baer Chem. Eer. 1968,101 1770. A. Hudson and H. A. Hussain J. Chem. SOC.(B),1967 1299.A. Hudson and H. A. Hussain J. Chem. SOC.(B),1968 351. A. Hudson and H. A. Hussain J. Chem. SOC.(B),1968,953. '18 A. Hudson and H. A. Hussain J. Chem. SOC. (B),1968,1346. 'I9 C. Lagercrantz and K. Torssell Acta. Chem. Scand. 1968,22 1935. 120 A. Mackor Th. A. J. W. Wajer and Th. J. de Boer Tetrahedron 1968,24 1623. 12' S. Shih. R. J. Pritchett. and J. M. Rivero. Tetrahedron Letters. 1968. 4897. Electron Spin Resonance Spectroscopy (c) Nitro-and NH,-substituted radicals. Interest in nitroanions has centred on the study of ion-pair interactions particularly in 0-,rn- and p-dinitrobenzene anions reduced with alkali metals in a variety of solvents.'22* 123 The dependence of uN on counter ions temperature and solvent yields interesting information on solvent effects and ion-pair equilibria.Other systems of this type which exhibit interesting linewidth effects are 3,5dinitrobenzoate anions' 24 and s-trinitroben~ene.'~~ Fessenden has extended his in situ irradiation technique'26 with 2-8 Mev electrons (previously applied to alkanes and fluoroalkanes) to the study of nitroalkanes.' 27 Nitroalkane anions are produced and their spectra were interpreted. Tertiary nitroaliphatics can be reduced electrolytically to their anions' 28 which further react to give dialkyl nitroxides. An investigation into the photolysis of various substituted nitrobenzenes in tetrahydrofuran has shown that Ward's',' original interpretation of the radical as Phfi0,H is probably in~orrect.'~' The radical is believed to be (31) an adduct of nitrobenzene and a solvent radical.6 R' I I R = alkyl or H Ph'-N-0-C-R I Ph' = substituted phenyl Solvent = R1R2HCOR3 (31) R2 Several nitrobenzenes enriched with "0 have been studied and the signs of' Qo and of the splitting constant are negati~e.'~' Most nitro-compounds will not give the cation radical ; an exception is N,N-dimethyl-p-nitroaniline where both anion and cation have been studied.'32 Aryldinitromethanes can be reduced to divalent radical anions,'33 as well as to the dinitromethane radical (32) which results from secondary reactions. NO* H-C' 'NO (32) "' R. F. Adams and N. M. Atherton Trans. Faraday Soc. 1968,64 7. J. Oakes and M. C. R. Symons Chem. Comm. 1968,294. 124 W. E. Griffiths C.J. W. Gutch G. F. Longster J. Myatt and P. F. Todd J. Chem. SOC.(B) 1968,795. 125 R. K. Gupta J. Subramanion N. K. Ray and P. T. Narasimhan Chem. Phys. Letters 1968,2 150. R. W. Fessenden and R. H. Schuler J. Chem. Phys. 1963,39,2147. 127 K. Eiben and R. W. Fessenden J. Phys. Chem. 1968,72 3387. 12' H. Say0 and M. Masui Tetrahedron 1968,24,5075. R. L. Ward J. Chem. Phys. 1963,38 2588. 130 D. J. Cowley and L. H. Sutcliffe Chem. Comm. 1968,201. 13' W. M. Gulick jun, W. E. Geiger jun. and D. H. Geske J. Amer. Chem SOC.,1968,90,4218. 13' R. F. Nelson and R. N. Adams J. Phys. Chem. 1968,72,740. 133 K. Torssell C. Lagercrantz and S. Wold Arkiv. Kemi 1968,29 219. Colin Thomson (d) Other nitrogen-containing radicals. Diphenylpicrylhydrazyl has been re-investigated at K band and the most probable coupling constants (deter- mined by computer simulation) indicate that both nitrogens are equivalent.' 34 A number of studies of the oxidation of aniline have appeared.In basic solution PhNO- is formed.' 35 Oxidation by copper(I1) acetate however gives a complex species.' 36 Different radicals are produced under acid and basic conditions when amino- acids react with TiCl,-H,O,. In basic solutions glycine gives NH,cHCO; but a different species is produced at pH 3 which was not positively identified.' 37 A thorough study has appeared of the different methods of preparing azo- type radical anions.' 38 Several stable free radicals derived from thermolysis of aliphatic diazo compounds have been reported.These are believed to be biradicals or biradical polymers resulting from secondary reactions of the initially formed carbene.' 39 Radicals Containing Oxygen.-A detailed study of a variety of xanthyl radicals (33)by Sevilla and Vincow' 4"-1 42 indicates weak conjugation through the oxygeq. H The 9-phenyl xanthyl is non-planar and 9-alkyl xanthyls exhibit considerable hindered rotation. These are structurally related to diphen~lmethyl.~~ Ketyl radical ions well known in solution have also been prepared in the solid state by use of the rotating cryostat method.'43 Linewidth effects in semi- quinone anion system^'^^'^^ have been studied by several authors and like the nitrobenzenes (refs. 122 and 123) have yielded interesting information on 134 Z.Haniotis and Hs. H. Gunthard Helv. Chim. Acta. 1968,51 561. 13' R.Konaka K. Kuruma and S. Terabe J. Amer. Chem. SOC.,1968,90,1801. 136 A. van Heuvelen and L. Goldstein J. Phys. Chem. 72,481. 137 R. Poupko B. L. Silver and A. Lowenstein Chem. Comm. 1968,453. 138 G. A. Russell R. Konaka E. T. Strom W. C. Danen K. Y. Chang and G. Kaupp J. Amer. Chem. SOC.,1968,90,4646. lJ9 L. S. Singer and I. C. Lewis J. Amer. Chern. SOC., 1968,90,4212. 140 M. D. Sevilla and G. Vincow. J. Phys. Chern.. 1968.72. 3635. 14' M. D. Sevilla and G. Vincow. J. Phys. Chem.. 1968 72. 3641. M. D. Sevilla and G. Vincow J. Phys. Chem. 1968,72,3647. 143 J. E. Bennett B. Mile and A. Thomas J. Chem. SOC.(A),1968,298. 144 P. S. Gill and T. E. Gough Trans. Faraday SOC.,1968,64 1997.145 P. S Gill and T. E. Gough Canad. J. Chem. 1968,46 656. 146 T. A. Claxton J. Oakes and M. C. R. Symons Trans. Faraday SOC. 1968,64 59. 14' T. A. Claxton and J. Oakes Trans. Faruday SOC. 1968,64 607. 148 J. Oakes and M. C. R. Symons Trans. Faraday SOC.,1968,64,2579. 149 D. H. Chen E. Warhurst and A. M. Wilde Trans. Faraday SOC.,1968,64,2561. A. W. Rutter and E. Warhurst Trans. Faraday SOC.,1968,64,2338. Electron Spin Resonance Spectroscopy the structure and properties of the ions and ion pairs. Activation energies for the two-site jumping process involving the cation have also been determined.' 5' The AlC13-MeN02 system has been used to prepare the cation radicals of several dimethylhydroquinones. These can exist in both cis and transforms.' 52 Russell has extended his detailed studies of the radical anions of diketones the semidiones.' Of particular interest is his study of 7Ohyperfine structure' 54 in which the results indicate an equation of the type ao = Q%o which describes the results quite well if Qg = -40 & 4 G.The parameter Q" has been determined from the spectrum of the semidione (34) and is (0.4) G.lS5 The value of this parameter has been in dispute for some time. (34) Radicals in Flow Systems.-The Ti3+-H202 system for generating 'OH' radicals continues to be widely used although the controversy over the nature of the oxidising radical is still not resolved. Florin et conclude from kinetic data that OH is not present but that the species involved is an ionized complex of Ti3+ with H6,.This interpretation differs from that of Takakura' 57 who concludes that 6H and OH are present both co-ordinated with Ti" The organic radicals produced are however usually well defined. Amino- acids undergo abstraction from C-H bonds distant from the -fiH3 group.'58 Ti3+ together with t-butylhydroperoxide gives eH3 which reacts with added organic and inorganic molecules.' A potentially useful reaction is the production of R6 radicals during photo- lysis of di-t-butylhydroperoxide in iso-octane containing alcohols.' 6o Alcohol radicals ReHOH are produced as in the usual experiments but this reaction can be carried out in non-polar organic solvents and should be readily extended to other molecules. Reaction of Ti3+-H202 with anisole and acetanilides does not result in the expected OH adducts although C,H,F does add OH like benzene and 15' J.C. Chippendale and E. Warhurst Trans. Faraday SOC..1968 64 2332. lS2 P. D. Sullivan and J. R. Bolton J. Amer. Chem. SOC. 1968,90 5366. Ref. 2 p. 87. G. A. Russell and G. R. Underwood J. Phys. Chem. 1968,72 1074. 15' G. A. Russell J. McDonnell and C. Myers J. Phys. Chem. 1968,72 1386. lS6 R. E. Florin F. Sicilio and L. A. Wall J. Phys. Chem. 1968,72 3154. 15' K. Takakura and B. Ranby J. Phys. Chem. 1968,72 164. H. Taniguchi,K. Fukui S. Ohnishi H. Hatano H. Hasegawa and T. Maruyama J. Phys. Chem. 1968,72 1926. D. Mickewich and J. Turkevich 1.Phys. Chem. 1968,72,2703. 160 J. Q. Adams J. Amer. Chem. SOC. 1968,90 5363. 32 Colin Thomson phenols.'61 An interesting reaction occurs during addition of Grignard reagent to quinobromides in that phenoxy radicals (35)are formed'62 i.e.RQR RMgX RQR R (35) Radicals Produced by Photolysis in Solution.-Several interesting photo- chemical studies have appeared. The n + n* triplet state of biacetyl has been used to test electrophilic and radical-like properties of the carbonyl n -+ n* states with benzenoid compounds.163 With benzene at pH 0-5,the protonated semidione is produced. Nitrobenzene gives the protonated derivative PhN0,H. In a similar study of benzophenone and ben~aldehyde'~~ H abstraction occurs and Ph2t0H and PheHOH are observed. No solvent radicals were detected. Zeldes and Livingston have continued their earlier work on liquids during photolysis and have examined the radicals from acid amides imide~,'~~ and aliphatic nitroanions.'66 Triplet States.-The study of organic triplet states in both glasses and single crystals continues to be of interest.The observed spectra give the parameters D and E which can be calculated from the electronic wavefunction. Several such calculations have appeared including zero field splitting in polyphenyl- benzenes,'67 and an extensive calculation on the C1,H triplet including accurate two- three- and four-centre integrals.'68 The results are not much better than for more approximate calculations and the desirability of all- electron calculations remains. In experimental work on triplet states several nitrogen heterocyclics biphenyl acenaphthene and tetramethylpyrazine have been ~tudied.'~' The latter is a n + n* state unlike the pyrazine triplet which is n -+ n*.Benzene has been investigated in a single crystal'70 and the non-trigonal symmetry was confirmed.Usually the low-field transition (Am= 2) do not show hyperfine structure. An exception is found with 1-and 2-fluoronaphthalene where aPrFcan be mea~ured.'~'The value is close to that reported for C,,F; in solution.76 16' C. R. E. Jefcoate and R. 0.C. Norman J. Chem. SOC.(B) 1968,48. 16' A. A. Volod'khin M. V. Tarkhanova A. L. Buchachenko and V. V. Ershov Izvest. Akad. Nauk S.S.S.R. Ser. khim 1968 57. 163 E. J. Baum and R. 0.C. Norman J. Chem. SOC.(4,1968,227. 164 R. Wilson J. Chem. SOC.(4,1968,84.R. Livingston and H. Zeldes J. Chem. Phys. 47,4173. 166 H. Zeldes and R. Livingston J. Amer. Chem. SOC.,1968,90,4540. M. K. Orloff and J. S. Brinen J. Chem. Phys. 1968,47,3999. 16* C.Thomson Theor. Chim. Acta. 1968 11 165. 16' Y. Gondo and A. H. Maki J. Phys. Chem. 1968,72,3215. 170 M. S. de Groot I. A. M. Hesselmann and J. H. van der Waals Mol. Phys. 1968 13 583. "' P. H. H. Fischer and K. H. Hawser Chem. Phys. Letters 1968,1 665. Electron Spin Resonance Spectroscopy 33 An interesting observation on the triplet dianions of 1,3,5-triphenylbenzene has been re~0rted.I~~ The spectra show that three triplet states are present and these are believed to be due to unperturbed dianion and to two new species in which successive solvation of the dianion by alkali-metal counter-ions occurs.The important technique of optical spin polarisation has been success- fully used in the study of the emissive e.s.r. lines from the naphthalene tri~1et.I~~ Miscellaneous.-Quintet states,' 74 quartet states,' 75 and even a septet state (in a nitrene) have been detected by e.~.r.'~~ Many papers have appeared dealing with radicals produced by high-energy radiation but two seem to be of particular importance. Pulse radiolysis techniques have been used to study radical formation and decay in systems where the radical half life is of the order of mse~.'~~ Using a similar technique Avery et al. have reported the e.s.r. spectrum of H30+ (lifetime ca. 10 msec.) in aqueous methan01.l~~ Radicals whose lifetime in solution is too short for them to be observed directly may be trapped and identified by reaction with nitroso-compounds' 79 or phenyl-t-butyl nitrone.180 Both these give stable nitroxides and from the spectrum the nature of the short-lived radical may be inferred.Finally two recent papers have described methods by which high-resolution e.s.r. spectra of radicals with lifetimes of ca. 1 msec. may be observed.18',182 "' J. A. M. van Brockhoven H. van Willigen and E. de Boer Mol. Phys. 1968,15 101. 173 M. Schwoerer and H. C. Wolf Chem. Phys. Letters 1968,2 14. 174 K. Itoh H. Konishi and N. Mataga J. Chem. Phys. 1968,48.4789. 17' E. Wasserman K. Schueller and W. A. Yager Chem. Phys. Letters 1968 2 259. 176 G. R. Luckhurst and E. R. Rozantsev Izvest.Akad. Nauk S.S.S.R. Ser. khim 1968 1708. 177 R. Hirasawa T. Mukaibo H. Hasegawa N. Odan and T. Maruyama J. Phys. Chem. 1968 72 2541. 17' E. C. Avery J. R. Remko and B. Smaller J. Chem. Phys. 1968,49,951. 179 C. Lagercrantz and S. Forschult Nature 1968,218 1247. E. G. Janzen and B. J. Blackburn J. Amer. Chem. Soc. 1968,90,5909. P. W. Atkins K. A. McLauchlan and A. F. Simpson Chem. Comm. 1968 179. lS2 P. W. Atkins K. A. McLauchlan and A. F. Simpson Nature 1968,219 927.