3 Reaction Mechanisms Part (i) Aromatic Compounds ~~ By A. R. BUTLER Department of Chemistry University of St. Andrews St. Andrews 1 Electrophilic Aromatic Substitution Three important reviews of this subject appeared in 1972. Taylor’s contribution’ to Volume 13 of ‘Comprehensive Chemical Kinetics’ comprises an enormous collection of kinetic data and a perceptive analysis of the results. This review will be invaluable to all workers in this field. Ridd’ has provided a valuable study of substitution in four- and five-membered heterocyclics and includes a section on linear free-energy relationships. Intermediates in electrophilic aromatic substitu- tion are the subject of the third review. There are comments on the controversial subject of the intermediacy of 7c-complexes7 on diffusion-controlled reactions and on the symmetry of o-c~mplexes.~ One controversial topic reported last year was the correct explanation for the variable amount of disubstitution obtained in the nitration of bibenzyl by nitro- nium tetrafluoroborate.This was explained by Ridd in terms of the rate of mixing affecting a diffusion-controlled reaction but Olah,4 in defence of his well- known views on the role of 7c-complexes in such reactions suggested that NO; complexes with the ring already substituted the second substitution being the result of an intramolecular migration. However in a further study of this reaction Gastaminza and Ridd’ were unable to find any evidence for complexing and the rate of mixing is the most important factor.There is substantial evidence that the deactivating effect of positive poles in aromatic nitration is transmitted not along a-bonds but directly through space (a field effect). Further evidence in support of this view has appeared. Nitration at the rn-position of (1) is slower than in (2) although the positive pole is separated from the ring by the same number of a-bonds. However the presence of the cyclohexyl ring in (1) holds the pole much closer to the ring and deactivation occurs because of a direct field effect.6 The same phenomenon is illustrated more ‘ R. Taylor in ‘Comprehensive Chemical Kinetics’ ed. C. H. Bamford and C. F. H. Tipper Elsevier Amsterdam 1972 Vol. 13 p. 1. J. H. Ridd in ‘Physical Methods in Heterocyclic Chemistry’ ed.A. R. Katritzky VOl. 4,p. 55. P. Rys P. Skrabal and H. Zollinger Angew. Chem. Internat. Edn. 1972 11 874. G. A. Olah Accounts Chem. Res. 1971 4 244. A. Gastaminza and J. H. Ridd J.C.S. Perkin II 1972 813. A. Ricci and J. H. Ridd J.C.S. Perkin II 1972 1544. 107 108 A. R.Butler (1) dramatically by comparing the rates of nitration of (3) and (4) where n = 2 and m = 4. The presence of the bridge which holds the positive poles close to the ring deactivates (4)by a factor of 100compared with (3). Unfortunately for this simple explanation it was found that the bridged compound with n = 1 where (4) the pole is not brought any closer to the ring by bridging is also deactivated. Undoubtedly factors other than the direct field-effect are operative but accord- ing to the evidence at hand it seems that this is the predominant one.' Steric factors do not appear to be important as the same effects are observed in hydrogen exchange.8 As well as deactivating the system protonation may also change the site of reaction.In acidic solution nitration of 1-methyl-4-phenylpyrazole(5) occurs in the phenyl ring but in acetic anhydride as solvent it is on the pyrazole ring.9 6 N-NMe There have been further studies of ionic reactions in the gas phase including acetylation and nitration. For the former reaction the effect of substituents is the same as that for reactions in solution but for nitration (where the nitrating species is generated from ethyl nitrate) a nitro-group activates the aromatic ring towards attack.It is suggested that the nitrating species is CH,ONOl rather than NO and the first step is electrostatic interaction between the aromatic ring and the nucleophilic oxygen of CH20NO;." Last year it was reported that gas- phase hydrogen exchange may be effected by the use of the helium tritide ion (HeTf).'' Now a study of the reactions of the D2T+ion has been reported. Reaction with arenes results in hydrogen exchange and the pattern of reactivity ' A. Ricci R. Danieli and J. H. Ridd. J.C.S. Perkin IZ 1972 1547. A. Danieli A. Ricci and J. H. Ridd J.C.S. Perkin II 1972 2107. I. J. Ferguson M. R. Grimmett and K. Schofield Tetrahedron Letters 1972 2771. lo R. C. Dunbar J. Shen and G. A. Olah J. Amer. Chem. SOC.,1972,94,6862.'I F. Cacade and G. Perez J. Chem. SOC.(B) 1971 2086; F. Cacace R. Cipollini and G. Ciranni ibid. p. 2089; F. Cacace 'Advances in Physica1,Organic Chemistry' ed. V. Gold Academic Press London 1970. Reaction Mechanisms-Part (i) Aromatic Compounds is the same as that observed with HeT+ i.e.low substrate selectivity (a ratio for reaction with toluene and benzene of only 1.3) but high positional selectivity. These effects are explained by postulating that the attractive electrostatic forces between attacking ion and substrate lead to a long-lived intermediate. Further reaction giving rise to high positional selectivity occurs within the interme- diate.' Such a view is reminiscent of Olah's postulated .It-intermediates. Exactly the same behaviour pattern is found for the reaction of HeT+ with liquid arenes.' Gas-phase electrophilic bromination shows the same lack of substrate selectivity an effect which is not generally observed for bromination in the liquid phase.14 These gas-phase reactions make it possible to compare reactivities under condi- tions where there are no complications due to solvation effects with theoretical predictions.For example MO calculations have shown that the most stable form of protonated benzene has the proton attached directly to a carbon atom but as the authors ppint out,' this assumes no solvation effects and must be com- pared with gas-phase protonation. It is very difficult to measure the reactivity of neutral N-heterocyclics towards electrophiles as they generally react in the protonated form.However the technique of gas-phase pyrolysis of the appropriately substituted ethyl acetate developed so successfully by Taylor does permit this. All the positions in free quinoline have been examined and are less reactive than the corresponding posi- tions in naphthalene but more reactive than those in pyridine." A number of stable a-complexes formed by the halogenation of triaminobenzenes have been isolated and characterized.' ' Nitration of activated substrates in sulphuric acid is thought to be encounter- controlled and should therefore have a low energy of activation (ca.25 kJ mol- '). The observed value of 72 4kJ mol-'in 68.3 % sulphuric acid is due to the large enthalpy change occurring in the step preceding nitration (i.e.HN03-+NO + OH-).Dissociation is complete in 92% acid." The mechanism of nitration in acetic anhydride is still under investigation. Hartshorn Moodie and Schofield" have compared the relative rates of nitra- tion of a number of anilides and aromatic ethers in acetic anhydride and sulphuric acid. The patterns of reactivity are so different that it is clear that dif- ferent mechanisms are operative. One important factor is interaction between substrate and solvent which in an extreme form amounts to protonation. Kinetic studies using acetic anhydride as solvent show that the reaction is of zero or first order in substrate. It is suggested that NO; (the nitrating agent) is formed in the following equilibrium N,O + AcOH * NO + NO;(AcOH) *'F.Cacace R. Cipoilini and G. Occhiucci J.C.S. Perkin II 1972 84. l3 F. Cacace and S. Caronna J.C.S. Perkin II 1972 1604. l4 F. Cacace and G. Stocklin J. Amer. Chem. SOC.,1972 94 2518. Is W. J. Hehre and J. A. Pople J. Amer. Chem. SOC. 1972 94 6901. l6 R. Taylor J. Chem. SOC.(B) 1971 2382. ' P. Menzel and F. Effenberger Angew. Chem. Innternat. Edn. 1972 I I 922. S. R. Hartshorn R. B. Moodie and K. Stead J.C.S. Perkin II 1972 127. l9 S. R. Hartshorn R. B. Moodie and K. Schofield J. Chem. SOC.(B) 1971 2454. 110 A. R.Butler and that with activated substrates the reaction is encounter-controlled and therefore of zero order in substrate. With less reactive substrates NO has time to undergo solvation to give NO:(HNO,) a weaker electrophile and the reaction becomes first order.20 Isomer ratios in the nitration of o-xylene vary with the acidity2’ and this may be due to nitro-group migration in the Wheland intermediate in an acid-catalysed reaction (Scheme 1).22 Nitration of the highly deactivated substrate Scheme 1 2,3-dinitroaniline involves N-nitration to give a nitrosamine with subsequent rearrangement to give ring nitration rather than direct nitration of the proto- nated c~rnpound.’~ Addition may accompany nitration.Nitration of 1,2,3,5-tetramethoxy-benzene gives a mixture of (6) and (7),24 whereas ethylmesitylene and fuming nitric acid give (Q2’ A small amount of (9) is obtained on the nitration of 2,6-dimethylneopentylbenzene.26 There have been no detailed mechanistic studies of these reactions.Electrophilic substitution generally involves replacement of hydrogen but other leaving groups are possible. Nitration of 4-iodoanisole results in the forma- tion of equimolar amounts of 4-nitroanisole and 2,4-di-iodoanisole clearly nitrodeiodination occurs and this is accompanied by iodination of unreacted 4-iodoanisole. The second stage of this reaction is nitrodeiodination of 2,4-di- iodoanisole to give 2-iodo-4-nitroanisole and iodine. Now a mixture of iodine and nitric acid is an iodinating agent and will iodinate 4-nitroanisole formed in the first step at the 2-position to give the same final product (Scheme 2). A 2o S. R. Hartshorn J. C. Hoggett R. B. Moodie K. Schofield and M. J. Thompson J. Chem. SOC.(B) 1971,2461. 21 R.G. Coombes and L. W. Russell J. Chem. SOC.(B) 1971 2443. 22 P. C. Myhre J. Amer. Chem. SOC.,1972 94 7921. 23 J. H. Ridd and E. F. V. Scriven J.C.S. Chem. Comm. 1972 641. 24 B. A. Collins K. E. Richards and G. J. Wright J.C.S. Chem. Comm. 1972 1216. 25 H. Suzuki M. Sawaki and R. Sakimoto Chem. Comm. 1971 1509. 26 A. J. M. Reuvers F. F. van Leeuwen and A. Sinnema J.C.S. Chem. Comm. 1972,828. Reaction Mechanisms-Part (i) Aromatic Compounds 111 ~ - ~ I I NO2 Scheme 2 detailed study of this reaction by Butler and Sander~on~~ has shown that the effective deiodinating species is not nitric but nitrous acid always present as an impurity and there is subsequent oxidation to the nitro-compound. Nitration of Ldiodoanisole with nitronium tetrafluoroborate does not result in deiodina- tion at all but nitration at the 2-position.Thus the situation is that the weak electrophile NO' can displace iodine whereas the strong electrophile NO does not. A similar situation applies to p-tolyltrimethylsilane (10). Nitration occurs at the 2-position but nitrosation results in displacement of the SiMe group and the final product is p-nitrotoluene.28 Azodehalogenation has also been ~tudied.~' Studies of aromatic nitrosation indicate that as an electrophile NO' is lo'* times weaker than NO; .30 There is no evidence for C1+ in electrophilic chlorina- ti~n.~l The active species in Friedel-Crafts reactions has been investigated. Various cornplexe~,~ acetylium ions,3 diacetylium ions,34 oxonium ions,35and alkyl 27 A.R. Butler and A. P. Sanderson J.C.S. Perkin II 1972 989. 28 C. Eaborn Z. S. Salih and D. R. M. Walton J.C.S. Perkin ZI 1972 172. 29 P. B. Fischer and H. Zollinger Helu. Chim. Am 1972 55 2139. 30 B. C. Challis R. J. Higgins and A. J. Lawson J.C.S. Perkin II 1972 1831. 31 C. G. Swain and D. R. Crist J. Amer. Chem. SOC. 1972 94 3195. 32 I. Hashimoto A. Kawasaki and Y. Ogata Tetrahedron 1972 28 217; R. Corriu M. Dore and R. Thomassin Bull. SOC. chim. France 1972 2829. 33 R. Corriu M. Dore and R. Thomassin Tetrahedron 1971 27 5819. j4 A. Germain A. Commeyras and A. Casadevall Bull. Sac. chim. France 1972 3177. 35 L. R. Pettiford J.C.S. Perkin II 1972 52. 112 A. R. Butler cations36 have all been implicated. The benzylation of benzene and toluene with a variety of catalysts has been examined in great detail.The isomer distribution is fairly insensitive to conditions but this is not true of the substrate selectivity. With AlCl and a series of substituted benzyl chlorides it is possible to see how the ortho :para ratio of the products varies with substrate selectivity. In general the p-isomer predominates with greater reactivity. This suggests that the transi- tion state of highest energy (which fixes the isomer distribution) depends on the reactivity. The transition state may be ‘early’ and resemble the reactants (n- complex) or be ‘late’ and resemble the Wheland intermediate (o-complex). So in this rather changed manner n-complexes still have a role as intermediates in electrophilic aromatic sub~titution.~’ 1-Methyltetrazole (11) undergoes hydrogen exchange at the 5-position by a carbanion mechanism and complexing with a transition-metal cation (Cu2 or +-M ~ N ~ N \ N” I (1 1) Zn2+) produces a very substantial increase in rate.This result is significant in understanding the production of carbanions in biological systems where metal ions are known to be required. It may also have important synthetic conse- quences. * 2 Nucleophilic Aromatic Substitution Recent mechanistic studies have been reviewed in concise and perceptive manner by Ross39 in Volume 13 of ‘Comprehensive Chemical Kinetics’. The role of intermediates in such reactions have been di~cussed,~ and Ridd2 has provided a review of the reactivity of heterocyclics towards nucleophiles.The problem of the a-effect (the enhanced reactivity of nucleophiles with a lone pair of electrons alpha to the site of nucleophilicity) has been examined for reactions involving displacement from aromatic compounds. For amines4’ the a-effect is associated with reactions having a fairly large Brransted value i.e. reactions in which considerable bond formation has occurred in the transition state. However this generalization does not apply to nucleophiles other than amines and it seems certain that the phenomenon of the a-effect is due to a variety of causes.41 The reactions of 2,4-dinitrochlorobenzene with butylamine and benzamidine are both first order in amine and the latter is less reactive. However with 4,7-dinitrofluoronaphthalene the reaction is second order in butylamine indicating base-catalysis in the slow removal of the fluoride ion.However the 36 G. A. Olah J. M. DeMember and R. Weiss J. Amer. Chem. SOC.,1972,94 5718. 37 G. A. Olah and S. Kobayashi J. Amer. Chem. SOC.,1971 93 6964; G. A. Olah S. Kobayashi and M. Tashiro ibid. 1972,94 7448. 38 H. Kohn S. J. Benkovic and R. A. Olofson J. Amer. Chem. SOC. 1972 94 5759. 39 S. D. Ross ref. 1 p. 407. 40 J. E. Dixon and T. C. Bruice J. Amer. Chem. SOC.,1972,94,2052. 41 J. E. Dixon and T. C. Bruice J. Amer. Chem. SOC.,1971 93 6592. Reaction Mechanisms-Part (i) Aromatic Compounds reaction is first order in benzamidine and it seems probable that benzamidine is acting as a bifunctional catalyst with a cyclic transition state (12).42 A similar result was obtained on comparing the reactions of imidazole and pyrazole with 2,4dinitrofluorobenzene although with pyrazole the reaction is of mixed order both first and second in pyrazole.However bifunctional catalytic activity of pyrazole is indicated.43 Normally one thinks of the nitro-group activating an aromatic ring towards nucleophilic attack but in certain circumstances this group may be replaced by a nucleophile to give a nitrite ion. This occurs in the reaction of 1,2,4-trinitro- benzene and o-dinitrobenzene with piperidine in benzene,44 and also the reaction of hydroxide ion with 2,4-dinitrobenzenediazonium ions. In the analogous reaction with the 2,6-dichloro-4-nitrobenzenediazoniumion chloride and nitrite are displaced at comparable rates.4s Displacement also occurs on 5-halogeno- 1-methyl-3-nitro-l,2,4-tria~ole.~~ Reaction of phenols in the presence of pyridine with the benzoate (13) gives the expect aryl ether.However with sterically hindered phenols the products are aryl 3,5-dinitrosalicylates and the reaction is thought to involve intermediate formation of the lactone (14).47 In the alkaline hydrolysis of 2,4-dinitrophenyl esters of benzoic acid nucleophilic attack is on the aromatic ring rather than at the carbo~y-group.~~ 42 G. Biggi F. Del Cima and F. Pietra J.C.S. Perkin 11 1972 188. 43 F. Pietra and F. Del Cima J.C.S. Perkin IZ 1972 1420. 44 F. Pietra and D. Vitali J.C.S. Perkin 11 1972 384. 0. MachaEkova V. StErba and K.Valter Coil. Czech. Chem. Comm. 1972,37,2197. 46 M. S. Pevzner V. Y. Samarenko and L. I. Bagal Khim. geterotsikl. Soedinenii 1972 848. 47 R. Muthukrishnan R. Kannan and S. Swaminathan J.C.S. Chem. Comm. 1972,358. 48 L. S. Prangova L. S. Efros and I. Y. Kvitko Organic Reactiuiry (Turtu) 1971,8 381. 114 A. R.Butler Fendler and Fendler4’ have reported a very complete study of the kinetics of Meisenheimer complex (a-complex) formation between methoxide ion and 2,4,7- trinitromethoxynaphthalene. Crampton and Khanso have shown that the equili- brium constant for 0-complex formation with a number of substituted anisoles depends upon the methoxide concentration. This variation is caused by changes in the rate of complex dissociation and may be due to stabilization of the complex by ion association.A very stable 1,3-complex (15) is formed from 2,6-dinitro-4- trifluoromethylsulphonylanisole,but it reverts to the more stable 1,l-complex.’ O2N(-JpeOMe H SOZCF (15) Bernasconi” has used temperature-jump stopped-flow to study the kinetics of formation of both the 1,3- and 1,l-complexes of 2,4,6-trinitroanisole and methoxide ion. For the former formation is very fast but there is a low equili- brium constant for the 1,l-complex this situation is reversed. It appears that the weak nucleophile aniline can displace rnethoxide ion from the a-complex with trinitrobenzene to give (16).53 H NHPh (16) Hydrogen bonding in Meisenhiemer complexes has been detected in a most elegant manner by Berna~coni.’~ Deprotonation of (17) is effected by hydroxide ion.A study using a temperature-jump technique has shown that the rate of deprotonation is 2 x lo81 mol-s-for the three secondary amines studied which is 100 times less than the diffusion-controlled transfer of a proton from an (17) 49 E. J. Fendler and J. H. Fendler J.C.S. Perkin II 1972 1403. M. R. Crampton and H. A. Khan J.C.S. Perkin IZ 1972 1173. 51 F. Millot J. Morel and F. Terrier Compt. rend. 1972 274 C 23. s2 C. F. Bernasconi J. Amer. Chem. SOC.,1971,93 6975. 53 E. Buncel and J. G. K. Webb Cunud.J. Chem. 1972,50 129. s4 C. F. Bernasconi J. Phys. Chem. 1972 75 3636. Reaction Mechanisms-Part (i) Aromatic Compounds I15 ammonium ion to hydroxide ion. This reduction is thought to be due to hydrogen bonding between the nitrogen of the amine group and the nitro-group in the o-position.Displacement on picryl chloride was thought to be an S,2 reaction but a recent study using DMSO as solvent has shown that the first step is formation of a 1,3-~omplex.~~ The first a-complex (18) of selenophen has been rep~rted.’~ Popp’ has reported an interesting example of an unexpected reaction occurr- ing in a sterically crowded molecule. Electrochemical reduction of (19) results in cyclization to give (21). The mechanism involves two-electron oxidation and formation of the intermediate (20) by intramolecular nucleophilic attack. In the final step the t-butyl group is lost as Me,C=CH,. (19) 3 Acidity Functions Little work appeared in 1972 to remove the confusion surrounding the exact significance of the acidity of a concentrated acid.However Yates and ShapiroS8 showed that the H scale generated by indicators not possessing nitro-groups is the same as that from nitroanilines. This gives confidence in the H scale as a measure of the ability of a concentrated acid to protonate a primary amine. Also activity coefficient measurements have shown thatf, +/’ is a constant for primary aniline indicators as suggested by Hammett.” Postle and Wyatt6’ report that added salts have a very large effect on HR because the forward (k,) and back (kb) reactions are affected in opposite ways causing a large shift in the equilibrium. The value of k,is a linear function of H, so that differences between H,and H are due in part to the effect of anions on k,.55 M R. Crampton M. E. El Ghariani and H. A. Khan Tetrahedron 1972 28 3299. 56 F. Terrier A.-P. Chatrousse R.Schaal C. Paulmier and P. Pastour Tetrahedron Letters 1972 1961. 57 G. Popp J. Org. Chem. 1972 37 3058. 58 K. Yates and S. A. Shapiro Canad. J. Chem. 1972,50 581. 59 T. R. Essig and J. A. Marinsky Canad. J. Chem. 1972,50,2254. 6o M. J. Postle and P. A. H. Wyatt J.C.S. Perkin II 1972 474. 116 A. R. Butler As a criterion for reaction mechanism the acidity dependence of a reaction is often difficult to interpret. There is a curved relationship between HR and H for hydrochloric acid and yet for related reactions log k may be a linear function of H in one case and HRin another.It is argued that this may reflect the nature of the transition state.61 The ionization of 1,3-diphenylindene has been used to generate an acidity scale for sodium methoxide in methanol and the resultant values of H-correlate well with the base-catalysed hydrogen exchange of fluorene.62 However in a detailed analysis More O'Ferra1162" has shown that correlation with H-is a poor guide to reaction mechanism. Plots may be non-linear and reactions with different mechanisms may show the same type of correlation. Linear plots are more fre- quently obtained using the complex term H-pKMeOH -log(^^^^-/^^^^) and the slope of this line may reflect the structure of the transition state. Follow- ing a study of the isomerization of allylbenzenes in aqueous DMSO Bowden and argue cautiously that the slope of the plot of log k us.H-reflects the degree of proton transfer in the transition state and this conclusion is consistent with the hydrogen-isotope effect. 4 Linear Free-energy Relationships There are several possible approaches to the Hammett and related equations. The first and most successful is to treat them as approximate empirical correla- tions of predictive value. What is so astonishing is the variety of phenomena which fit the simple Hammett equation. It applies to peaks in the n.m.r. spectra of pyra~ines,~~ CN stretching frequencie~,~' I3C n.m.r. shifts in benzyl cations,66 chemical shifts of non-aromatic protons,67 barriers to rotation,68 spin-spin coupling constant^,^^ charge-transfer energy,70 and the reactions of p~rphins.~ No extension of the Hammett equation is required for free-radical reactions.72 Once a more precise correlation is sought it is necessary to modify the Hammett equation either generally by the introduction of an extra parameter,73 or for a particular system (e.g.cyclohexyl or silicon substituents7 '). The " A. J. Kresge H. J. Chen and Y. Chiang J.C.S. Chem. Comm. 1972 696; M. Godel A. Jussiaume and F. Coussemant Tetrahedron Letters 1972 23 17. 62 A. Streitwieser C. J. Chang and A. T. Young J. Amer. Chem. SOC.,1972 94 4888. 62aR. A. More O'Ferrall J.C.S. Perkin II 1972 976. 63 K. Bowden and R. S. Cook J.C.S. Perkin lI 1972 1407. 64 G. S. Marx and P. E. Spoerri J. Org. Chem. 1972 37 11 1. 65 T. Funabiki and K.Tarama Bull. Chem. SOC.Japan. 1972,45,2945. 66 G. A. Olah R. D. Porter C. L. Jeuell and A. M. White J. Amer. Chem. SOC.,1972 94 2044. 67 B. Kamienski and T. M. Krygowski Tetrahedron Letters 1972,681. 68 L.-0. Carlsson Acta Chem. Scand. 1972,26,21; T. B. Grindley A. R. Katritzky and R. D. Topsom Tetrahedron Letters 1972 2643. 69 S. Rodmar S. Gronowitz and U. Rosen Acta Chem. Scand. 1971 25 3841. 70 K. Sekigawa Tetrahedron 1972 28 505 515. 71 M. Meot-Ner and A. D. Adler J. Amer. Chem. SOC.,1972,94 4763. l2 A. P. G. Kieboom Tetrahedron 1972 28 1325. 73 Y. Yukawa Y.Tsuno and M. Sawada Bull. Chem. SOC.Japan 1972 45 1210; Y. Tsuno M. Fujio Y.Takai and Y. Yakawa Bull. Chem. SOC.Japan 1972,45 1519. 74 J. L. Mateos H. Flores and H. Kwart J. Org. Chem. 1972 37 2826.75 J. Lipowitz J. Amer. Chem. SOC.,1972 94 1582. Reaction Mechanisms-Part (i) Aromatic Compounds danger with parametrization is that it is not always possible to distinguish between parameters which have a physical significance and those which are mathematical devices. Another approach when the simple Hammett equation breaks down is to 0'values are required. It is known that cassume that new values for the 2-thienyl and the 1-position of bi~henylene~~ vary with the reaction studied. This may be interpreted in terms of the resonance demands of the reaction. For the reactions of aromatic systems other than benzene it can be assumed that 6'constants will apply and c+orcdifferent values for thi~phen~~ and novalues for polycyclic systems have been determined.79 An entirely new set of c+values based on the rates of protodetritiation of aromatic compounds in TFA have been prepared by Baker Eaborn and Taylor.80 They are fairly successful when applied to electrophilic aromatic substitution.The final approach to the Hammett equation is to use it as a basis for a theoretical analysis of the transmission of substituent effects. Godfrey8 rejects the division into field and inductive effects and has discussed non-linear free- energy plots in terms of his recently proposed field and charge transfer (FCT) treatment. The relative importance of these effects depends upon the reaction considered and a good correlation will be obtained only if the effects are the same as in the reaction used to define 0.The FCT theory has been applied to o-substi- tuents.82 Some work on substituent chemical shifts in p-substituted fluoro- benzenes indicates* that cpis a better measure of resonance interaction between substituent and the aromatic ring than cR.This analysis throws some doubt upon the validity of previous attempts to separate substituent effects into inductive and resonance components. Values of p for the trifluoroacetylation of thiophens furans and pyrroles have been determined by Clementi and mar in^^^ and have been interpreted in terms of varying transition-state structure(a- or n-complex). However Forsyth and NoycegS propose that it is due to varying charge distribution in the transition state. The temperature dependence of the Hammett equation has been consideredg6 and Wold and Sjostrom have performed a major statistical analysis of available data to find the best values of c(i.e.those values which give a linear correlation for most reaction^).^^ This updates Jaffk's classic review of the subject.76 F. Fringuelli G. Marino and A. Taticchi J.C.S. Perkin IZ 1972 158. 77 R. Taylor M. P. David and J. F. W. McOmie J.C.S. Perkin IZ 1972 162; R. Taylor ibid. p. 165. D. S. Noyce C. A. Lapinski and R. W. Nichols J. Org. Chem. 1972,37 2615; D. S. Noyce and H. J. Pavez ibid. pp. 2620 2623. 79 M. Sawada Y. Tsuno and Y. Yukawa Bull Chem. SOC. Japan 1972,45 1206. R. Baker C. Eaborn and R. Taylor J.C.S. Perkin 11 1972 97. M. Godfrey Tetrahedron Letters 1972 753. 82 M. Godfrey Tetrahedron Letters 1972 3203.83 1. R. Ager and L. Phillips J.C.S. Perkin ZI 1972 1975; I. R. Ager L. Phillips T. J. Tewson and V. Wray ibid. p. 1979. 84 S. Clementi and G. Marino J.C.S. Perkin 11 1972 71. D. A. Forsyth and D. A. Noyce Tetrahedron Letters 1972 3893. 86 L. D. Hansen and L. G. Hepler Canad. J. Chem. 1972,50 1030. 87 S. Wold and M. Sjostrom Chem. Scripta 1972 2 49. 118 A. R. Butler Substituent effects display an angular dependence in the dissociation of the acids (22) (23) and (24)88 and in the H-F coupling in substituted fluorocyclo- pro pane^.^' These observations support transmission of substituent effects by a direct field effect and not along a-bonds. (22) (23) (24) pK 6.261 6.4 16 6.470 There has been a further test of the Westheimer postulate that the k& ratio will be a maximum when the proton is half transferred in the transition state.For hydrogen exchange in indoles with a variety of catalysts the isotope effect was found to be essentially constant over a range of differences in pK and this gives no support to the postulate. The authors conclude with an excellent discus- sion of the problem.’* The significance of the terms tl and /? in the Brsnsted equation has again been the subject of several reports. Murdoch9* has shown by a theoretical treatment based on the Hammond postulate that o! cannot in any precise sense reflect transition-state structure and this has been further confirmed by Bordwell and B~yle~~ in a study of the rate of deprotonation of a number of nitroalkanes and ketones by a variety of amine catalysts.Brernsted CI values have also been examined in terms of the Marcus theory93 for a number of aliphatic diazo- compounds. The value of a-changes as the catalyst is changed but is independent of the diazo-compound. Values of o! outside the theoretical range ((3-1) are explained by destruction of the n-system where a carbon base is protonated. This does not occur when oxygen or nitrogen bases are ~rotonated.’~ A much simpler deduction of transition-state structure from a Brsnsted plot has been made by Rogne9’ in a study of the reaction of aromatic sulphonyl chlorides with anilines. It seems reasonable to suppose that bond formation is less advanced in the transition state when there is an electron-donating substi- tuent at the p-position in the sulphonyl chloride.Now 2-and 2,6-substituted anilines show negative deviations from the Brcansted plot of this reaction for steric reasons but this effect is less pronounced with an electron-donating C. L. Liotta W. F. Fischer E. L. Slightom and C. L. Harris J. Amer. Chem. Soc. 1972,94,2129. B9 K.L.Williamson S. Masser and D. E. Stedman J. Amer. Chem. SOC..1971,93 7208. 90 B. C. Challis and E. M. Millar J.C.S. Perkin I[ 1972 11 16 1618 1625. 9‘ J. R. Murdoch J. Arner. Chem. SOC.,1972 94,4410. 92 F. G. Bordwell and W. J. Boyle J. Amer. Chem. SOC., 1972,94,3907. 93 R. A. Marcus J. Phys. Chem. 1968 72 891. 94 W. J. Albery A. N. Campbell-Crawford and J. S. Curran J.C.S. Perkin 11 1972,2206.95 0. Rogne J.C.S. Perkin /I 1972 472. Reaction Mechanisms-Part (i) Aromatic Compounds substituent at the p-position. This is consistent with the reduced amount of bond formation. The c1 effect has been the subject of a theoretical study using a polyelectron perturbation method,96 and it has been detected among N- heterocyclic^.^' Hibbert and Long98 report that the hydroxide ion is anomalously unreactive as a base catalyst for proton transfer from a carbon acid. Base-catalysed hydrogen exchange in indoles involves formation of an anion and the rate-determining step may be looked upon as attack of an indole anion on water. However a Brsnsted plot based on this (using pK values of substituted indoles) has a slope outside the theoretical range.The reasons for this have been discussed in detail.99 Until now there has not been an entirely satisfactory scale of nucleophilicity. However Ritchie and co-workerstoO have shown that the reaction of a very large number of nucleophiles with various cations (e.g. aryidiazonium ions Crystal Violet Malachite Green) may be correlated by the simple expression logk = logkHz + N where k is the rate of reaction of cation with the nucleophile kH20is the rate of its reaction with water and N is characteristic of the nucleophile and the solvent. Thus N is a quantitative measure of nucleo-philicity with wide-ranging applicability. 96 F. Filippini and R. F. Hudson J.C.S. Chem. Comm. 1972 522. 97 J. A. Zoltewicz and H. L. Jacobson Tetrahedron Letters 1972 189; J.A. Zoltewicz and L. W. Deady J. Amer. Chem. SOC.,1972,94,2765. 98 F. Hibbert and F. A. Long J. Amer. Chem. SOC.,1972,94 2647. 99 B. C. Challis and E. M. Millar J.C.S. Perkin 11 1972 111 1. loo C. D. Ritchie and H. Fleischhauer J. Amer. Chem. SOC.,1972,94 3481; C. D. Ritchie and P. 0.I. Virtanen ibid. pp. 4963 4966.