3 Reaction Mechanisms Part (iv) Polar Reactions By N. S. ISAACS Dept. of Chemistry The University Whiteknights Park Reading RG6 2AD 1 Aliphatic Nucleophilic Substitution The exact formulation of the S,2 transition state still occupies a considerable amount of attention. Theoretical studies continue to be made with greater refinement and the normally expected inversion geometry is preferred to 'front- face' attack of the nucleophile.' The system CN-+ CH,-F-+ NC-CH + F-has been studied by partitioning energy changes into contributions from spatially defined fragments.2 Bond-making and -breaking contributions (-141 and 163 kcal mol-' respectively) lead to a predicted activation energy of 22kcal mol-'. Such studies referring to the gas phase aid understanding rather than afford experimentally verifiable values.The idea of pentaco-ordinate carbon as an intermediate rather than a transition state in the S,2 reaction is not new,3 but an attempt has now been made to stabilize such a species using the rigid anthracene frarnew~rk.~ Although (1) undergoes the 'bell-clapper' rearrangement (1) (3)as shown by n.m.r. line broadening no definite evidence for the discrete existence of (2)could be adduced. N.m.r. shifts have also been claimed to show complex formation between chloride ion and benzyl or aryl- thiomethyl halides depicted as (4); K values fall in the ranges 0.001-0.1 and p "e. I +/ Ph \ Me '> / Ph Ph \ IW / s+-c s (1) (2) (3) ' (a)A. Dedieu and A. Veillard J. Amer. Chem. SOC.,1972,94,6730; (b) W.T. A. M. Van der Lugt and P. Ros,Chem.Phys. Letrers 1969,4 389; (c) A. J. Duke and R. F. Bade Chem. Phys. Letters 1970 5 328. R. F. Bader A. J. Duke and R. R. Messer J. Amer. Chem. SOC.,1973,95 7715. W. von E. Doering and H. H. Zeiss J. Amer. Chem. SOC.,1953,75,4733. (a)J . C. Martin and R. J Badelay J. Amer. Chem. SOC.,1973,95,2572; (6)A. J. Parker S. G. Smith I. D. R.Stevens and S. Winstein J. Amer. Chem. SOC.,1970 92 115. 119 120 N. S. Isaacs 0.24.5 re~pectively.~ However it is not clear whether these contact complexes are in any way relevant to sN2 reactions. Still another approach to this question has been attempted by an examination of the Brcansted coefficients (from plots of log k against log K,) for the variation of the nucleophile (BN)and the leaving group (PL)in a displacement reaction.6 It may be reasoned by perturbation theory' that these values reflect the extent of bond formation and bond breaking respectively.Such studies have been reported in displacements at sulphur e.g. in (5) and (6) 0 0 R R OH-II IrS I Ph,C-S I' + Ph,C-S + OAr-X-eS-Y + X-S + Y-\ \ II OAr OH 0 /jN = 0.25 PL= -0.97 (bond making lags far behind bond breaking) and BN = 0.75 -0;PL= -0.71. The indication at least for moderate nucleophiles is of a fairly symmetrical transition state but the abrupt curve in the Brransted plot towards the strong-base end reveals complexities which call into question the whole concept of the use of Brransted coefficients to map sN2 pathways.In recent years a new view ofaliphatic substitution reactions has been proposed by Sneen8- l4 and co-workers which may be summarized by the sequence shown in Scheme 1. In many ways this resembles the Winstein' 'merged mechanism' olefin SOH/ 3-SOH/ AN-\El 1 SOR NR ROS RN olefin products Scheme 1 J.-I. Hayami N. Tanaka N. Hihara and A. Kaji Tetrahedron Letters 1973 385. ' L. Senatore E. Ciufferini A. Fava and G. Levita J. Amer. Chem. SOC.,1973,95 2918. R. F. Hudson Chemie 1963 16 173. R. A. Sneen Accounts Chem. Res. 1973,6,46. H. Weiner and R. A Sneen J. Amer. Chem. SOC.,1965,87 287 292. lo R. A. Sneen and J. W. Larsen J. Amer. Chem. SOC.,1966,88 2593. I ' R. A. Sneen and J. W. Larsen J. Amer. Chem. Soc. 1969 91 362. l2 R. A. Sneen and H.M. Robbins. J. Amer. Chem. SOC.,1969,91 3100. R. A. Sneen and J. W. Larsen J. Amer. Chem. SOC.,1969,91 6031. l4 R. A. Sneen and H. M. Robbins J. Amer. Chem. SOC.,1973,95 7868. IS S. Winstein D. Darwish and N. J. Holness J. Amer. Chem. SOC.,1956 78 2915. Reaction Mechanisms -Part (iu) Polar Reactions or the Doering and Zeiss3 two-step process but the reactive intermediate pro- posed for a great many if not all nucleophilic displacement reactions is the ion-pair I1 ; this may react directly to give products or may proceed first to the solvent-separated ion-pair 111 but it is considered unlikely to lead to the solvated carbonium ion IV. It is proposed that this mechanism is shared by systems pre- viously classified as SN1,SN2 SN2' and El and E2 and may also incorporate neighbouring-group participation and rearrangements.By kinetic analysis of the competitive reactions of solvent and nucleophile with a substrate the product ratio is given (as for conventional SN2 reactions) by and the rate ratio by where ksolvis the rate for pure solvolysis and x = k- Jks. Clearly if this scheme operates equation (2) becomes identical with traditional SN2 and SN1behaviour of the limits of x thus lim (X + ~)keXpt,/ko= 1 + m"] (SN~, k-1 >> ks) lim (x -+ O)k,x,tl/ko = 1 (SNl,k-<< ks) The second case is not radically different from a traditional SN1scheme but the first requires a rate-determining destruction of the ion-pair 11. The substrate is supposed to be in rapid equilibrium with 11 which is only infrequently in- volved in irreversible reaction with solvent or added nucleophile.This perhaps is the point which departs most from older views that the activation step in an SN2 reaction is initiated by attack of the nucleophile. Testing this hypothesis is not easy and is most likely to succeed in the 'border- line' region where the predictions from equations (1)and (2)may be contrasted with those of other mechanisms. Seven secondary and benzylic systems have been analysed in this way and the results shown to be in accordance with the predictions of the unified mechanism and in some instances have been shown to depart from other models. For instance the competitive reaction of a-phenylethyl bromide with azide ion in ethanol gives rate ratios kexpt,/kO as a function of azide ion concentration which deviate widely from the line predicted for two S,2 reactions and fit a curve appropriate to x = 8.50 in equation (2).It is not clear whether the experimental results would be explained by other models such as a mixture of SN1 and S,2 processes ;this system is one in which ionization would be highly probable. The relatively small selectivities in competitive reactions ob-served between reactions of SN2 character and those of SN1 are claimed to be better explained in terms of the unified mechanism. Stereochemical consequences of displacement reactions can be accommodated and criticisms'6*' concerning '' B. J. Gregory G. Kohnstam M. Paddon-Row and A. Queen Chem. Comm. 1970 1932. " D. J.Raber J. M. Hems R. E. Hall and P. von R. Schleyer J. Amer. Chem. Soc. 1971 93,4821. I22 N. S. Isaacs salt effects have been answered.12 The mechanism envisaged by Sneen may turn out to have wide validity but much more extensive and critical testing is required particularly if the author is to achieve his aim of bringing primary and methyl systems within the scope of a unified displacement mechanism. Careful measurements of the chlorine kinetic isotope effect and its temperature dependence have been made on prototype SN1and SN2 reactions (methanolysis of t-butyl chloride and the reaction of n-butyl chloride with thiophenoxide ion in methanol respectively18 and the values obtained are in accordance with the predictions of Bigeleisen' using conventional S,1 and SN2 model geometries ofthe respective transition states.The former in accordance with the expectedly higher degree of bond breaking shows the greater effect (k3,/k35 = 1.0106 as against 1.0089 for the bimolecular reaction). These values at least seem to argue against a unified type of transition state; one would not expect to observe a primary isotope effect in a rate-determining attack upon an ion-pair. Displacements at carbon attached to the ferrocene nucleus [e.g.as in (7)] seem to be anomalous in that highly nucleophilic leaving groups may be displaced and Me Fe retentionofconfigurationoccurs." It issupposed that acarbonium ion mechanism is occurring here. Halide exchange at bridgehead positions can be readily achieved using AIBr formed in situ21 (though the preformed compound is almost inactive) and a halogenated solvent ;thus(8)gives(9).The mechanism is uncertain but competitive C. R. Turnquist J. W. Taylor E. P. Grimsrud and R. C. Williams J. Amer. Chem. SOC. 1973,95,4133. l9 J. Bigeleisen and M. Wolfsberg Adv. Chem. Phys. 1958 1 15. 2o G. Marr B. W. Rockett and A. Rushworth Tetrahedron Letters 1970 1317. G. W. Gokel D. Marquarding and I. K. Ugi J. Org. Chem. 1972 37 3052; J. F. McKinley R. E. Pincock and W. B. Scott J. Amer. Chem. SOC.,1973,95 2030. Reaction Mechanisms -Part (iu)Polar Reactions 123 elimination is not favourable in these systems. Displacement using organo-copper reagents is of considerable synthetic potential.22 2 Solvolytic Reactions The importance of steric factors in solvolysis is one that is difficult to assess and quantify but in certain instances may be paramount.Rates of ethanolysis of 1-and 2-adamantyl arenesulphonates ( 1Oa.b) (S,1 or k,) show an almost identical (10) a; R' = H R2 = OSO2T0l b; R' = OS02Tol R2 = H response to the effects of leaving groups (p = 1.76 and 1.86 re~pectively).~~ This argues that electronic influences are small and constant leaving steric effects upon ionization to account for the rate differences. Relief of steric strain has been put forward to account for rate differences in methanolysis of methylated adamantyl systems (1 la,b) although the same explanation does not hold in the norbornyl series (12a,b) in which the methylated derivatives react more slowly than the &:' >li R&R (lla) R' = Me R2 = OS02Tol R' = OS0,Tol R2 = Me (1 fb) (12) parent In these systems there is likely to be a fine balance between enthalpy and entropy effects which would be difficult to disentangle.Steric strain relief is most likely the explanation for the decreasing p-factors with increasing bulk ofsubstituents in the series(l3)+15)(PNB = p-nitrobenz~ate).~' Ar(neopentyl),C-PNB ArBu'(neopenty1)C-PNB ArBu:C-PNB (13) (14) (15) P -2.91 -2.64 -1.30 22 C. R. Johnson and G. A. Dutra J. Amer. Chem. SOC.,1973 95 7777 7783; G. H. Posner C. E. Whitten and J. J. Sterling ibid. p. 7788; K. Oshima H. Yamamoto and H. Nozaki ibid. p. 7926. '' D. N. Kevill K. C. Kalwyck D. M. Shold and C. B. Kim J.Amer. Chem. SOC.,1973 95 6022. 24 D. Faulkner M. A. McKervey D. Lenoir C. A. Senkler and P. von R. Schleyer Tetrahedron Letters 1973 705. 25 H. Tanida and H. Matsumura J. Amer. Chern. SOC.,1973 95 1586. 124 N. S. Isaacs The highly compressed di-t-butylaryl compounds derive so much of the driving force to ionization from strain relief that electronic demands upon the aryl sub- stituent are considerably reduced. Rates of unimolecular solvolysis have long been held to give a measure of carbonium ion stability.26 More recently this information has also become avail- able from 13Cn.m.r. spectroscopy since the chemical shift of the positive carbon of the free cation in superacid solution reveals how much charge is delocalized onto substituent groups.” There should therefore be a relationship between 13Cn.m.r.shifts and solvolytic rate data. Unfortunately this seems not to be the case at least in the series (16a-~).*~ The phenyl group appears more electron- releasing thancyclopropyl by then.m.r. probe but less soby thesolvolyticcriterion. R (16) a; R =Me b; R=Bd C; R =Ph It is not clear where the discrepancy lies; probably the use of enthalpies of acti- vation rather than rates at one temperature might resolve the problem but clearly caution must be exercised in interpreting reaction data in terms of carbonium ion stability. Model systems e.g. 2-adamantyl( 10a)29 and pinacolyl(1 7)30esters and halides which solvolyse only by the ionization route (k,) and which by virtue of their Me Me \/ .c -c.Me‘] L*H Me OS0,Ar structure ought not to experience solvent assistance (k,) or neighbouring-group assistance (k,) have been devised in order to estimate the extents of unassisted rates in other systems. If both model systems behave as predicted it might be supposed that they should show a similar solvent dependence. In fact the rate ratio k(l,,a,/k(17)varies quite widely (from 72 to 20 between aqueous ethanol and formic acid) suggesting that the mechanism of at least one model is not ‘pure’ in all 26 C. K. Ingold ‘Structure and Mechanism in Organic Chemistry’ Bell London 1969. 27 G. A. Olah and A. M. White J. Amer. Chem. SOC.,1969,91 5801. 28 H. C. Brown and E. N. Peters J. Amer. Chem. SOC.,1973.95 2400. 29 J. F. Fry C.J. Lancelot L. K. M. Lam J. M. Harris R. C. Bingham D. J. Raber R. E. Hall and P. von R. Schleyer J. Amer. Chem. SOC.,1970,92 2338. 30 V. J. Shiner R. D. Fisher and W. Dowd J. Amer. Chem. SOC.,1969,91 7748. Reaction Mechanisms -Part (iv) Polar Reactions 125 solvents e~amined.~' One possibility is that the pinacolyl esters derive some assistance to ionization (kb)from the migrating /I-methyl group. Vinyl halides and esters are well established as able to undergo unimolecular solvolysis especially if an a-aryl group is present ;1,Zshifts of /I-anisyl group can occur [(18)+ (19)].32 An a-cyclopropyl group is greatly rate-enhancing but ArHph[ArH1:] AcOH + AcOq ArHAr + Ar Br Ar Ph AcO Ph (18) Ar = p-MeOC,H (19) rather like /I-phenyl a /I-cyclopropyl group is less effective than methyl :33 both groups are more effective when trans to the leaving group than when cis.Also the deuterium secondary isotope effect is somewhat greater from the trans$- position than from cis-B (kH/kDct,,,, = 1.57; kH/kD(eis)= l.10).34 Isotope effects in halogenoallene solvolysis have been measured ;35 the a-effect (kH/kD = 1.22) is comparable with effects in vinyl systems. y-Methyl isotope effects are also quite considerable (k,,CH3/ky-CD3= 1.25). The allene (20) solvolyses 6500 times faster than the structurally similar (21).36 The mechanisms are S,1 (m,= 0.73 AH = 20 kcal mol-' AS = -11cal K-') but the origin of this effect is obscure probably steric. Me Me The long-controversial question of solvolysis of 2-norbornyl derivatives has been reviewed by BrownJ7 and cogent arguments have been presented for the classical formulation of the 2-norbornyl cation.Further evidence supporting the classical nature of 6-arylnorborn-2-enyl cations (22) has been presented in the form of a comparison of Hammett p-values of -4.17 and -4.21 respectively for ionization rates of exo- (22a) and endo-compounds (22b).38 If significant n-participation were occurring (in the exo-case only) the electronic demand from 31 J. E. Nordlander R. R. Gruentzmacher and F. Miller Tetrahedron Letters 1973,927. 32 Z. Rappoport A. Gal and Y.Harminer Tetrahedron Letters 1973 641. 33 D. R. Kelsey and R. G. Bergman J.C.S. Chem. Comm. 1973 589. 34 D. D. Maness and L. D. Turrentini Tetrahedron Letters 1973 755.35 M. D. Schiavelli and D. E. Ellis J. Amer. Chem. Soc. 1973 95 7916. 36 M. D. Schiavelli P. L. Timpanaro and R. Brewer J. Org. Chem. 1973 38 3054. " H. C. Brown Accounts Chem. Res. 1973 6 377. 38 E. N. Peters and H. C. Brown J. Amer. Chem. Soc. 1973,95 2398 7920. 126 N. S. Isaacs (22) a; R' = OPNB R2= Ar b; R'= Ar R2= OPNB the aryl group should be lower than for endo-solvolysis. Rate enhancement is at any rate comparable with values for cyclopentenyl esters (23)-(25).38 Secondary deuterium isotope effects have been shown to be a function of the leaving group the upper limit increasing in the order I- (1.09) C1- (1.15) RS02-(1.22).36,29*40 One interpretation is that the transition from tight to solvent- separated ion-pair is kinetically significant even rate-determining.Interpretation of the magnitudes of secondary isotope effects is a highly developed art; for example a value of 1.15 for the a-effect in the ethanolysis of cyclopentyl brosylate is taken41 to indicate the rate-determining step as being solvent separation of the ion-pair rather than attack by solvent k (expected isotope effect for an SN2 reaction is 1.22). In solvents of higher polarity (ethanol-water trifluoroethanol- water) the isotope-effect values rise (1.18 and 1.25) which is interpreted as solvent attack becoming kinetically important despite the lower nucleophilicity of these solvents. Solvolysis of deuteriated cyclobutyl mesylates gives a p and y secondary isotope-effect ratios of 1.102 0.934 and 1.055 respectively ;the normal y-effect is taken to indicate a 1,3-interaction in the transition state.42 By deuterium labelling rates of solvolysis of cyclohexyl tosylate have been partitioned into rates of formation of both substitution and elimination products both that from a direct reaction and after a 1,2 hydride shift.The ratios of these four types of product vary somewhat with solvent (an ion-pair mechanism is supported) but 20 % of the substitution product and 5-50 % of the cyclohexene has suffered rearrangement.43 39 V. J. Shiner and W. Dowd,J. Amer. Chem. SOC.,1971,93 1029. 40 V. J. Shiner M. W. Rapp E. A. Halevi and M. Wolfsberg J. Amer. Chem. SOC.,1968 90 7171. 41 K. Humski V. Sendijarivic and V. J. Shiner J. Amer. Chem. SOC.,1973 95 7722.42 B. Goricnic Z. Majerski S. Borcic and D. E. Sunko J. Org. Chem. 1973 38 1881. 43 J. B. Lambert and G. J. Putz J. Amer. Chem. SOC.,1973.95 6316. 127 Reaction Mechanisms -Part (iv) Polar Reactions 3 Neighbouring Group Participation The effect of an a-methyl group in solvolyses of 7-sulphonates in the norbornane series (23H25)decreases dramatically as the electronic demands of the reaction become satisfied by n-parti~ipation.~~ Participation may also be important in the solvolysis of (26),but in any event rearranged products are found.45 .OAc OTos It is known that the reactivity ratio B-phenylethyllethyl (solvolysis of tosylates) increases rapidly as the ionizing power of the solvent is increased (e.g. by 7500-fold between ethanol and trifluoroacetic acid).This is undoubtedly due to an increase in the phenyl-assisted component k of the phenylethyl reaction relative to k or k for both reactions. This increase in k might in turn be due to a dissimilar response of the two processes to solvent polarity (m,>m,)or an increase in the extent of aryl bridging brought about by solvent change. It is found that the solvent response is similar to that of neophyl systems (27) in which bridging (k,) is the only term of significance;consequently the latter term is ~nimportant.~~ Ph \7 Ph \+ Ph-C-CH + /C-cHzPh Ph’ Ph d& The presence of aryl assistance in fi-phenylethyl ester solvolysis has been de- monstrated in (28) by the presence of a 14C-secondary kinetic isotope effect (I4Clocated at position 1 of the aromatic ring); a maximum isotope effect would depend upon there being no internal return a condition obtaining in neophyl Me OBros 44 P.G. Gassman and J. M. Pascoe J. Amer. Chem. SOC.,1973,95 7801. 45 R.M. Coates and K.Yano J. Amer. Chem. SOC.,1973,95 2203. 46 F. L. Schadt and P. von R. Schleyer J. Amer. Chem. SOC.,1973 95 7860. 128 N. S. Isaacs systems. It is reported that the isotope effect in neophyl solvolysis is very close to that in 2-(panisyl)ethyl systems both being greater than in P-~henylethyl.~’ On the basis of rate data it is concluded that P-aryl participation occurs in the solvolysis of /I-azulenylethyl sulphonates (29) whether the azulene system is bonded at 2- 4- 5- or 6-p0sitions.~*,~’ Entropies of activation are typically highly negative (4-azulenyl -25 ;6-azulenyl -20 cal K-I) and similar to the values for p-phenylethyl.Rates however are 68000 times greater than for fi-phenylethyl.50 Other effective participating groups (rates relative to p-phenylethyl) are 3-indolyl(7200) ferrocenyl(1400) and thiophenyl(21). Cyclo-octatetraene on the other hand shows no tendency to act as a nucleophilic neighbouring group by kinetic criteria though rearrangements of P-(cyclo-octatetraeny1)ethyl sulphonates (30) to (3 1)and (32) occur on solvolysis suggesting .c-c-H -1 H __ OAc H-l‘OTos H that some contribution from k or k is important.” Trifluoroacetolysis of the deuteriated 2-butyl tosylate (33) reveals the products to be (34) and (39 with no detectable amounts of (36) or (37).52 This is the expected result if hydrogen participation occurs with the formation of the transition state rather than a rapidly equilibrating 2-butyl cation.Solvolysis with concomitant hydrogen migration is also proposed in the bicyclo[3,3,1]nonane series (38) on the basis of 47 Y. Yukawa S. G. Kum and H. Yamataka Tetrahedron Letters 1973 373. 48 R. N. McDonald N. L. Wolfe and H. E. Petty J. Org. Chem. 1973 38 1106. 49 R. N. McDonald N. L. Wolfe H. E. Petty R. G. Cooks and P. T. Cranor J. Org. Chem. 1973,38 1 114. 50 D. S. Noyce and R. L. Castenson J. Amer. Chem. SOC., 1973 95 1247. 51 L. A. Paquette and K. A. Henzel J. Amer. Chem. SOC.,1973,95,2724,2726. 52 J. J. Dannenberg D. H. Weinwurzel K.Dill and B. J. Goldberg Tetrahedron Letters 1972 1241. Reaction Mechanisms -Part (iv) Polar Reactions 129 CF CO H CH,CH,-CDCD CH,CH,CDCD + CH,CHCDHCD, I I I OTos OCOCF OCOCF (33) (34) (35) CH,CHDCHCD CH,CDCH,CD, I I OCOCF OCOCF (36) (37) a non-linear Hammett plot (substitution in Ar) indicating a change in mechanism as electron-releasing substituents are placed in the aromatic nucleu~.~~*~~ A similar consideration may also apply in hydrolysis of longifoline derivatives (39). U (39) Participation by cyclopropane rings depends acutely on the orientation with respect to the reaction centre e.g. little or no participation occurs in solvolysis of (40) although rearrangement of the cations so formed does OCCU~.~~,~~ The D (40) OPNB series of solvolytic rates for (41H44)indicates that participation is a continuous function of the dihedral angle between cyclopropane ring and leaving group ;' a 60" angle seems particularly unfavourable.Large effects of the cyclopropane ring on solvolysis of endo-compounds (45)-(47) are recorded but there is no evidence of any contribution from (48) a bishomo-antiaromatic structure which should destabilize the cati~n.~~?~~ Participation by the cyclopropane ring is likely to be of importance in the system (49).60 53 L. Stehelin L. Kannellias and G. Ourisson J. Org. Chem. 1973 38 847. 54 L. Stehelin L. Kannellias and G. Ourisson J. Org. Chem. 1973 38 851. 55 E. C. Friedrich M. A. Solch and S. Winstein J. Org. Chem. 1973 860.56 E. C. Friedrich and M. A. Solch J. Amer. Chem. SOC.,1973 95 2617. 57 Y. E. Rhodes and V. G. diFate J. Amer. Chem. SOC.,1972,94 7582. 58 P. G. Gassmann and X. Creary J. Amer. Chem. SOC.,1973 95 6852. 59 P. G. Gassmann and X. Creary J. Amer. Chem. SOC.,1973 95 2729. 6o J. B. Lambert A. P. Jovanivich J. W. Hamersma F. R. Koenig and S. S. Oliver J. Amer. Chem. SOC.,1973 95 1570. 130 N. S. Isaacs (42b) kJk 4 x 2.5 x TosO (43b) 4 x 103 The carbonyl group is not one to manifest neighbouring-group participation most strongly yet it is claimed to do so in the following examples. Enhanced endolexo solvolytic ratios in (50a) and (Sob) are taken as evidence forparticipation of the 7-keto-gro~p.~' [The value of this ratio for the parent system (5 1) is 0.17 R;,R' 8OTos / A1 R2 (50) a; R'R2 = =O (51) a; R' = H R2 = OTOS b; R' = OTOS,R2 = H b; R' = OTOS R2= H 61 R.Baker and J.C. Salter J.C.S. Perkin 11 1973 150. Reaction Mechanisms -Part (iu) Polar Reactions compared to for the norbornyl tosylates.] The analogous ketals appear also to derive some driving force from Me0-4 participation (known to be negligible in acyclic systems62) though this is claimed for (53)on the basis of an enhanced endolexo ratio and more negative entropy of a~tivation.~~ The bridgehead oxygen in (54) apparently does not assist endo-solvolysis ;both endo- and exo-isomers suffer from inductive deceleration of solvolytic rates.64 Amide carbonyl participation evidently occurs in the hydrolysis of (55);65 the intermediate cation (56) is isolable as the tetrafluoroborate.Also the rapid hydrolysis of the ester (57) produces the isolable intermediate (58).66 R-C. I R-CO (55) (56) HR The carboxylic acid group may act as either a nucleophilic or an electro- philic participating group. As the former the hydrolysis of 0-and p-carboxy- benzal chlorides6' (59) and (60) may be cited and as the latter the hydrolysis of 62 P. G. Gassmann and J. L. Marshall J. Amer. Chem. SOC.,1966 88 2822. 63 P. G. Gassmann J. L. Marshall and J. G. McMillan J. Amer. Chem. SOC.,1973 95 63 19. b4 L. A. Paquette and I. R. Dunkin J. Amer. Chem. SOC.,1973 95 3067. 6s D. A. Tomalia and J. N. Paige J. Org. Chem. 1973 38 422. 66 K. Bowden and A.M. Last J.C.S. Perkin II 1973 3S1. " V. P. Vitullo and N. R. Grossman J. Org. Chem. 1973 38 179. 132 N. S. Isaacs disalicylacetals (61);68 these fast hydrolyses are independent of hydronium ion catalysis. 0Q Phenolic hydroxy-group participation swings carbamate hydrolysis from the normal addition4imination mechanism via an isocyanate iri favour of an internal nucleophilic displacement (62a)69 and similarly for the corresponding o-hydroxy- methyl compound (62b).70 (-CH,OH) I:B (62) a; Z = OH b;Z = CH,OH Ring-opening of the epoxide (63)is strongly assisted by the 7 a-hydroxy-group acting in an electrophilic ~apacity.~' N3-''0 0 @ \H' (63) 68 E. Anderson and T. H. Fife J. Amer. Chem. SOC.,1973,95 6437. 69 J. E.C. Hutchins and T. H. Fife J. Amer. Chem. SOC.,1973 95 2282. 'O J. E. C. Hutchins and T. H. Fife J. Amer. Chem. SOC.,1973 95 3786. '' D. H. R. Barton and Y. Hauminer J.C.S. Chem. Comm. 1973 839. Reaction Mechanisms -Part (iv) Polar Reactions 4 CarboniumIons A survey of carbocation chemistry in superacid media has been published72 and also information concerning Hannett acidity functions in these highly protonating media.73 Values as high as H = 19 are recorded for the system HS03F-SbF5. New carbonium ion species which have been observed include the bkration (64),’* protonated naphthalene (65),75 acenaphthene (66),76 and the nortricyclyl cation (67).7 The tricycloundecyl cation (68) undergoes a five-fold degenerate R rearrangement with respect to both ally1 moieties [i.e.behaves as (69)].78 Other degenerate rearrangements observed include those of the bicyclo[2,l,l]hexyl cation (70),” bicyclo[3,2,l]octadienyl (71),*’ and deltacyclyl (72).8 Several rearrangements of cyclopropylallyl systems8* e.g.(73)4(74) and cycloallyl 72 G. A. Olah Angew. Chem. Internat. Edn. 1973 12 173. 73 R. J. Gillespie and T. E. Peel J. Amer. Chem. SOC.,1973 95 5 173. 74 G. A. Olah G. Liang P. von R. Schleyer E. M. Engler M. J. S. Dewar and R. C. Bingham J. Amer. Chem. SOC.,1973,95 6829. 75 G. A. Olah G. D. Mateescu and Y. K. Mo J. Amer. Chem. SOC.,1973 95 1865. 76 G. A. Olah G. Liang and P. Westerman J. Amer. Chem. SOC.,1973,95 3698. 77 G. A. Olah and G. Liang J. Amer. Chem. SOC.,1973,95 3792. ’’ M. J. Goldstein and S.A. Kline J. Amer. Chem. SOC.,1973 95 935. 79 G. Seybold P. Vogel M. Saunders and K. B. Wiberg J. Amer. Chem. SOC.,1973,95 2045. ‘O H. Hart and M. Kazuya J. Amer. Chem. SOC.,1973,95,4096. P. K. Freeman and B. K. Stevenson J. Amer. Chem. SOC., 1973.95 2890. 82 K. Rajaswari and T. S. Sorensen J. Amer. Chem. SOC.,1973,95 1239. 134 N. S. Isaacs (to bicycloalkyl) e.g. (75)+(76),have been studied.83 Strained cyclobutyl cations such as (77) rearrange stereospecifically (77) +(78).84 Me3cQ Me &Me Me Me 'Me (75) (76) 83 L. Huang K. Ranganayakulu and T. S. Sorensen J. Amer. Chem. SOC.,1973.95 1936. P. G. Gassman and E. A. Armour J. Amer. Chem. SOC.,1973,95 6129. Reaction Mechanisms -Part (iv) Polar Reactions CH,OTos (77W Protonated hydrazobenzene (80)and azoxybenzene (79) have been identified presumably species identical with intermediates in the benzidine and Wallach rearrangement^.^' Bridgehead carbonium ions are notably difficult to form.Two compounds which ionize considerably faster than t-butyl esters are (81) and (82);86the related biscation (83)has now been observed as a long-lived species at 0 0C.87 & eb B~'C~ 3.757H-fB (81) (82) H 3.417 H 2.557 krcl 10' 104 1 (83) Dealkylation of t-butylbenzene (84) a reverse Friedel-Crafts reaction can occur in superacid solution,88 and protolysis of C-C and C-H bonds can also occur e.g. (85)-P (86).89 BU' Bu' H (84) " G. A. Olah K. Dunne D. R. Kelly and Y. K. Mo,J. Amer. Chem. SOC.,1972,94,7438.86 A. de Meijere and 0.Schallner Angew. Chem. Internat. Edn. 1973 12 399. A. de Meijere 0.Schallner and C. Weitmeyer Angew. Chem. Internat. Edn. 1971,10 404. G. A. Olah and Y.K. Mo,J. Org. Chem. 1973.38 3221. 89 G. A. Olah and Y. K. Mo J. Amer. Chem. SOC..1973,95,6827. 136 N. S. Isaacs ,Bu' Bu' + 2Me3C+ Proton migration in benzenonium ions is well recognized. Protonated fluoro- benzene a mixture of the three isomers at 33 "C,becomes only the 3-and 4-fluoro- compounds at -10 "C and only the 4-fluoro-compound the most stable isomer at -84 OC90 Evidence of the mobility of nitro-groups in aromatic substitution reactions comes from the acid-catalysed rearrangement of (87).91 OAc Many carbonium ion rearrangements can be explained by postulating the intermediacy of protonated cyclopropanes this concept has been re~iewed.~' A reaction difficult to reconcile with other mechanisms is the carbon and hydrogen scrambling of the 2-propyl cation (88) which occurs in superacid solution.93 'CH $ CH CH /\ /+\ 'LA CH3 CH3 H3C-CH2 / Careful kinetic and product studies accord with this mechanism but not with others in which the 1-propyl cation is required to be an intermediate.Hydrogen-exchange and alkylations of alkanes in superacid are presumed to occur uia 'five-co-ordinate' carbon species ; the scope of these reactions [for example (89)-+ (go)] has been further extended.9L97 Similar intermediates 90 G. A. Olah and Y. K. Mo J. Qrg. Chem. 1973,38 3212. 91 P. C. Myhre J.Amer. Chem. SOC.,1973,95 7921. 92 M. Saunders P. Vogel E. L. Hagen and J. Rosenfeld Accounts Chem. Res. 1973 6 53. 93 C. C. Lee A. J. Cessna E. C. F. KO and S. Vassie J. Amer. Chem. SOC.,1973 95 5688. 94 G. A. Olah J. R. deMember and J. Shen J. Amer. Chem. SOC.,1973,95,4952. 95 G. A. Olah J. Shen and R. H. Schlosberg J. Amer. Chem. Soc. 1973,95,4957. 96 G. A. Olah Y. K. Mo and J. A. Olah J. Amer. Chem. SOC.,1973 95 4939. 97 G. A. Olah Y. Halpern J. Shen and Y. K. Mo J. Amer. Chem. SOC.,1973,95,4960. Reaction Mechanisms -Part (iu) Polar Reactions are postulated for hydride abstractions by carbonium ions and evidence in favour of this scheme is provided by the exchange of deuterium with the transferring hydrogen in reactions of perdeuteriotrityl cation (91) with complex hydrides or tr~pylidene.~~ Recent structural investigations of carbonium ions include the I9F n.m.r.spectra of p-fluorobenzyl cations ;98 the higher-field chemical shifts in the order acyclic monocyclic bicyclic may indicate steric restriction on sol-vation. Similar considerations may be important in the physical processes responsible for solvent shifts of fluorophenyl compounds used as a solvation parameter.99 '3C n.m.r. spectra of protonated carboxylate esters (92)"" ''' OH p.p.m. 177 169 164 158 17.5,( CH,-CH2-CH2-CH2-C I+ \ 66+ 6+ OMe indicate a smooth diminution of charge density along the carbon chain. This contrasts with the charge densities alternating along the chain from the perturbing substituent"' as predicted by CNDO calculations.The same technique shows that cyclic allyl cations bear almost all the charge at the allyl termini (93),'03 and benzoyl cations have considerable charge delocalization into the aromatic ring owing to contributions from structures such as (94).' O4 Experimental charge densities in polyenylic cations have been compared with predictions from several types of MO calculation.'05 That which appears best to reproduce experimental 98 G. A. Olah and J. J. Svoboda J. Amer. Chem. SOC.,1973,95 3794. 99 R. W. Taft and L. D. McKeever J. Amer. Chem. SOC.,1965,87 2488; D. G. Farnum and D. S. Patton J. Amer. Chem. SOC.,1973 95 7728. loo G. A. OlahandY. K. Mo J. Org. Chem. 1973,38 353. lo' G. A. Olah and P. W. Westerman J. Org.Chem. 1973 38 1986. lo* J. A. Pople and M. Gordon J. Amer. Chem. SOC.,1967 89,4253. Io3 G. A. Olah and G. Liang J. Amer. Chem. SOC.,1972,94,6434. '04 G. A. Olah and P. W. Westerman J. Amer. Chem. SOC.,1973 95 3706. lo5 H. V. Navangal and P. E. Blatz J. Amer. Chem. SOC.,1973,95 1508. 138 N.S.Isaacs 0 -41 0 -41 p.p.m. +48 + (93) (94) results is the semi-empirical o-technique though correct trends are apparent even using simple Huckel theory. Protonation of phenols and anisoles has been studied ;Io6 methylation of anisole by CH,F-SbF gives the oxonium ion (95) which is stable at -70°Cbut is Me ?+ ,Me ?Me OMe (95) converted into methylated anisoles at higher temperatures by an intermolecular rearrangement.'" Heats of formation of t-butyl and t-pentyl cations (169 and 161 kcal mol-' respectively) have been determined in the gas phase by the equilibrium (96) S (97) ;(AAG" = -2.9 kcal mol- ',AAH" = -3.6 kcal mol-H + + I Me,CH + Me,CEt S Me,C + Me,CEt (96) (97) AAS' = 2.3 cal K-').'O* Further examples of cycloadditions of allylic cations to olefins and dienes have a~peared,"~ as has a review of this topic.'" The 2-methoxyallyl cation (98) is especially readily formed and reactive in Diels-Alder reactions.Potential-surface calculations have been carried out on ally1 cations 0 '06 G. A. Olah and Y. K. Mo J. Org. Chem. 1973.38 2212 353. lo' G. A. Olah and E. G. Melby J. Amer. Chem. SOC.,1973,95,4971. IonJ. J. Solomon and F. H. Field J. Amer. Chem. SOC.,1973.95.4483.'09 A. E. Hill G. Greenwood,and H. M.R. Hoffmann J. Amer. Chem. SOC.,1973 95 1338. 'lo H. M. R. Hoffmann Angew. Chem. Inrernar. Edn. 1973 12. 819. Reaction Mechanisms -Part (iu) Polar Reactions 139 and predictions made concerning the mechanisms of geometrical isomeriza- tion0f(99).'"*"~ Two mechanisms may be considered ;simple bond twisting (a) and intermediate isomerization to a cyclopropyl cation (b). It is predicted that X %kr (99) 2-fluoro and 2-methylallyl cations undergo isomerization by path (b) but 2-hydroxy- and 2-aminoallyl cations must use path (a)since the corresponding cyclopropyl cations are more stable than the ally1 isomers. The cyclopentadienyl cation (100) is known to be a very difficult species to form solvolytically,' '' and MO calculations of varying sophistication have been used on this species.' ''-'16 The ion has been formed at low temperature in superacid solution and shown by itse.s.r.spectrum to be a triplet in theground state as predicted by simple theory.' '' A convenient alcohol -+amine conversion (101) +(102) evidently proceeds via a carbonium ion intermediate or ion-pair (lola). Yields can be high when a reasonably stable carbonium ion is formed.' The subject of carbonium ions n-bonded to metals has now an enormous literature. A brief review of this topic serves to introduce the reader to this subject of increasing significance.' ' 'I1 L. Radom J. A. Pople and P. von R. Schleyer J. Amer. Chem. SOC.,1973,958193. 'I2 L. Radom P. C. Hariharan J.A. Pople and P. von R. Schleyer J. Amer. Chem. SOC. 1973,95 6531. 'I3 R.Breslow and J. M. Hoffmann J. Amer. Chem. SOC.,1972,94,2110. 'I4 M. J. S.Dewar and R. C. Haddon J. Amer. Chem. SOC. 1973,95 5836. ''' H. Kollmer H. 0.Smith and P. von R. Schleyer J. Amer. Chem. SOC.,1973,95 5834. ' l6 W. J. Hehre and P. von R. Schleyer J. Amer. Chem. SOC.,1973,95 5837. ''' M. Saunders R. Berger A. Joffe J. M. McBride J. ONeill. R.Breslow J. M.Hoff-mann C. Pevchonock E.Wasserman R. S.Hutton and V.J. Kuck J. Amer. Chem. SOC.,1973,95 3017. 'IBJ. B. Hendrickson and I. Joffee J. Amer. Chem. SOC.,1973,95,4083. 'I9 M. H. Chisholm and H. C. Clark Accounts Chem. Res. 1973,6,202; T. H. Whitesides R. W. Arhat and R. W.Slaven J. Amer. Chem. SOC.,1973,95 5792. 1 40 N.S. Isaacs - ROH + R-0 i=R+O -P R+NH -+ R-NH \ \ I I c=o S0,CI S0,Cl / HN HN +co pzo I I S0,CI S0,CI (101) (101a) 5 Addition Reactions Applications of linear free energy relationships (LFER) to polar additions have been made; Charton and Charton have correlated some 28 systems using the extended Hammett relationship,' *' logk,, = an + PoR + const where uIand uRare 'inductive' (localized) and 'resonance' (delocalized) substituent constants. Systems for which a >> /3 are deduced to react via bridged intermediates e.g. (103) while those for which a << B are supposed to form open carbonium-ion intermediates such as (104). Values of p span the range -12 to -35 for electro- philic additions and + 14 to +35 for nucleophilic additions e.g.to give (105). ---* R-f' R\\ + Br2 -RT/Br+Br- a = -13 P = -3.56 (103) Br R R - R CH,< + H,O+ + CH,ACH CH,\-cH,; a = -4.3 P= -32.9 OH LFER analysis of the addition of bromine to stilbenes has been made in terms of a combination of substituent effects from the two rings the reaction constants p and pa being -5.07 and -1.40 respectively. The fact that the reaction responds differently to substituents in each ring indicates the transition state to be non- symmetrical (106) even if a bridged intermediate (107) is duly formed.I2' IZo M.Charton and B. I. Charton J. Org. Chem. 1973 38 1631. J. E. Dubois and M.-F.Ruasse J. Org. Chem. 1973 38 493. 141 Reaction Mechanisms -Part (iv) Polar Reactions (106) Dipolar aprotic solvents are capable of being captured during addition reactions and stable adducts such as (108) are isolable.'22 Hypohalous acid additions to dienes123 and to cinnamic acid derivativeslZ4 have been studied and the stereo- chemistries of the products established with precision.The trans-cinnamate esters give a mixture of erythro-and threo-a-halogeno-fi-hydroxyphenyl-propionic acids. Steric effects of 7,7-dimethyl substitution upon products of addition of nor- bornene are quite severe. The methyl groups force addition in most cases from essentially exo to largely or completely endo [e.g. (109-(l10)].'25 Addition of fluorine to olefins (111) (or at least to diphenylethylenes) may be achieved by xenon difluoride in the presence of acid.'26 (110a) (1 lob) R=H 99.5 0.5 R = Me 5 95 "' C.Anselmi G. Certi B. Macchia F. Macchia. and L. Monti Tetrahedron Letters 1972 1209. D. R. Dalton and R. M. Davies Terruhedron Letters 1972 1057. L24 P. B. D. de la Mare and M. A. Wilson J.C.S. Perkin II 1973 653. H. C. Brown J. H. Kuwakami and K. T. Lin J. Amer. Chem. SOC.,1973,95 2209. 126 M. Zupan and A. Pollak J.C.S. Chem. Comm. 1973 845. 142 N. S. Isaacs The kinetic form of the substitution by secondary amines in (1 12) is consistent with an addition-elimination mechanism in which deprotonation (k,) of the XeF, > >=( HF FF Ar CN Ar CN ";kc" R,NH+ x..)-<.-x; )=( X CN CN CN +NH2 YH I (X = CN F OR) RI R (1 12) (1 13) intermediate (113) is kinetically ~ignificant.'~' It is found that k,(obs) = k' + k"[R,NH].An analogous mechanism is proposed for acetylenic halide substitu- tion by thiophenoxide ion (1 14)+(1 1S).' 28 Large positive p-values (p = 3.4 for ArS;AkSAr __+ ArCEC-hal -hal-ArC=C-SAr ha1 (1 14) (115) chloride displacement and 3.9 for bromide) are consistent with rate-determining attachment of nucleophile to the acetylenic carbon. Also chloride is more readily displaced than bromide (kC,/kBr = 2.54.2) typical of a reaction in which C-halogen bond fission is not rate-determining. Addition of arylsulphenyl chloride to the allene (116) is proposed to occur via the bridged intermediate since there is a considerable inverse secondary deuterium isotope effect for the 3-deuterio-compound but a very small effect for the l-deuterio.I2' It would appear that the transition state involves the terminal carbon becoming more nearly sp3 hybridized but there being little change in hybridization at C-1.Additions of alcohols carboxylic acids etc. to the C=N bond in (117) have been reported.' 30 12' Z. Rappoport and P. Peles. J.C.S. Perkin II 1973. 616. 128 P. Beltrame P. L. Beltrame. M. G. Cattania. and M. Simonetta. J.C.S. PerkinII 1973 63. K. Izawa T. Okuyama and T. Fueno J. Amer. Chem. SOC.,1973,95,4090. J. L. Zollinger. C. D. Wright J. J. McBrady D. H. Dybvig F. A. Fleming G. A. Kurhajec. R. A. Mitsch and E. W. Neuvar J. Org. Chem. 1973 38 1065. Reaction Mechanisms -Part (iv)Polar Reactions (F,N),C=NF +ROH -+ (F,N),C-NHF I (117) OR 6 Elimination Reactions Bredt's rule excluding the formation of bridgehead bicyclic olefins on grounds of steric strain has proved a challenge to synthetic chemists and it seems now estab- lished that bicyclo[x,v,z]alk-1-enes with (x +y +z) =S 3 7 are stable com- pounds which can be prepared e.g.(118)-(119). Compounds for which S <7 are unstable but some have been prepared transiently. Thus bicyclo[2,2,l]hept-l- ene may be generated by the route (120) -P (121) and trapped by its Diels-Alder reaction with furan. The same ratio of the two stereoisomeric adducts (122) and (123) results with considerable variation of conditions and starting materials supporting a reaction with a common intermediate. The subject of Bredt's rule has been reviewed.I3' Eliminations of sulphonates adjacent to heteroatoms should be facile on account of the resonance stabilization of an intermediate carbonium ion (124).However in compounds such as (125) the bridgehead nitrogen compounds react a. +/ + x-c 4-b x=c / /\/\ 10-20 times slower than the corresponding carbon analogues.' 32 This could be due to the unfavourable strain introduced by the bridgehead double bond. G. Kobisch Angew. Chem. Internal. Edn. 1973 12 464. 132 P. G. Gassman R. L. Cryberg and K. Shudo J. Amer. Chem. SOC.,1974 in press. 144 N. S. Isaacs The reaction of the bromosulphone (126) with t-butoxide ion leads to substitution by an elimination-addition mechanism.' 33 The subject of mechanism in base-induced /3-eliminations has become anything but clear in the past few years with the introduction of the E2C-E2H and syn-anti mechanistic dichotomies.' 34 Ethoxide-induced elimination of (127) shows both \+ NMe3 Q-(secondary) and p-(primary) isotope effects which vary in magnitude smoothly with electronic effects of aryl substituents (p =0.95) the values of kdk being of the order respectively 1.04 and 4.5.' 35 Eliminations of P-bromoethylbenzenes (p =2.85) show primary isotope effects k& =7-9.5 which also depend on aryl sub~titution.'~~ These systems can be interpreted in terms of an El42 continuum and tunnelling is probably important in the latter case.Conversion of cyclohexyl tosylate into the dimethyl analogue (128) a 'neopentyl' system retards the rate of elimination (by C1-) by only a factor of 12 which the authors deem to be insufficient to support any degree of E2C ~haracter.'~' The Brnrnsted coefficient for thiolate- induced eliminations of 2-halogeno-2-methylbutanes (129) p =0.134.16 is ,OTos c1 / CH,CH,CCH, \ CH3 interpreted as indicative of very weak bonding between nucleophile and p-proton in the transition state despite the fact that this system would be expected to be of the E2H type.' 38 Products of elimination and consequently the pathways used are sensitive to the nature of the base ;the introduction of cation-complexing crown ethers affects the ratio of syn-to anti-elimination products in reactions of C.B. Quinn J. R. Wiseman and J. C. Calabrese J.Amer. Chem. SOC.,1973,95 6121. 134 N. S. Isaacs Annual Reports (B) 1972,69 180. '35 P. J. Smith and S. K. Tsui Tetrahedron Letters 1972 917. 136 L. F. Blackwell P. D. Buckley J. W. Jolley and A. K. H. McGibbon J.C.S. Perkin ZI 1973 169. 13' J. F. Bunnett and D. L. Eck J. Amer. Chem. Soc. 1973,95 1897 1900. 138 D. S. Baily and W. H. Saunders J. Org. Chem. 1973 38 3363. Reaction Mechanisms -Part (iv) Polar Reactions 145 cyclodecane derivative^'^^ and also the cisltrans and orientation ratios in elimi- nations of 2-bromobutane and the acenaphthene derivative (130).140,'41 This test is usually considered to differentiate reactivity due to a solvated ionic base and to an ion-pair. Dehydration of di-t-butylcarbinol (1 3 1) by hexamethylphosphoric triamide (HMPT) yields the rearranged olefin (132) which is explained in terms of a carbonium ion intermediate not a species usually associated with reactions in this ~olvent.'~' Sulphuranes (133) will also cause efficient dehydration of an Me (131) (1 32) Ar,S-OBu' -/C-CH, Ar,SR + Me,COH-a Ar,SR -a+ CH\ -Ar,SO I (133) OBu' CH3 alcohol even at low temperatures (p = -1.68).'43 Oxalate proves to be especially favourable towards elimination of isopropyl tosylate compared with other carboxylate ions which lead principally to substitution.' 44 Mesylation followed by pyridine-induced elimination can generate olefins in high yield'45 from iodo- hydrins (134).MeS0,CI py. -20°C O'fYR I OH ( 134) 139 M. Svoboda J. Halpala and J.Zavada Tetrahedron Letters 1972 265. I4O D. H. Hunter Y. Lin A. L. Mclntyre D. J. Shearing and M. Zvagulis J. Amer. Chem. SOC. 1973 95 8327; D. H. Hunter and D. J. Shearing J. Amer. Chem. SOC. 1973,95 8333. 141 R. A. Bartsch G. M. Pruss D. M. Cook R. L. Buswell and K.-E. Wiegers J. Amer. Chem. SOC.,1973,95,6745. 14* J. S. Lomas D. S. Sagatys and J. E. Dubois Tetrahedron Letters 1972 165. 143 L. J. Kaplan and J. C. Martin J. Amer. Chem. SOC.,1973 95 793. 144 E. J. Corey and S. Terashima Tetrahedron Letters 1972 111. 14' E. J. Corey and P. A. Grieco Tetrahedron Letters 1972 107. 146 N.S. Isaacs 7 Carbanions The predicted conrotatory ring-opening of a cyclopropane anion (1 35) has been shown to take place.146 A (very small) diatropic character has been found in the anion (136) but not in the related species (137) and (138).It is concluded that there is some homoaromatic character and delocalization of charge into the cyclo- propane ring.' 47-149 The relative rates of proton exchange of (139) and (140) indicate that the homoaromatic anion (141) is more stable than (142) and no additional stabilization from bicycloaromaticity need be considered.' The mono- and di-anions of (143)have been prepared ;the n.m.r. spectrum of the latter A-H (139) (140) H (144) shows that the charge is delocalized over the carbonyl group and the C-6-C-7 double bond making this a bishomo-analogue of the cyclopentanone 146 M. Newcombe and W. T. Ford J. Amer. Cliem. SOC.,1973,95 7186. 147 S.W. Staley and W. G. Kingsley J. Amer. Chem. SOC.,1973 95 5804. 14' S. W. Staley and G. M. Cramer. J. Amer. Chem. SOC.,1973 95 5051. 149 S. W. Staley. G. M. Cramer and W. G. Kingsley J. Amer. Chem. SOC.,1973,95 5052. Is' M. V. Moncur and J. B. Grutzner. J. Amer. Chem. SOC.,1973 95 6449. Reaction Mechanisms -Part (io) Polar Reactions dianion (145).15' Proton-exchange rates in the bicyclic ketone series'52 (146) and in 7-cyanonorbornenes' 53 indicate that ring-strain can accommodate the data without invoking exotic homoaromatic concepts. The I3C n.m.r. spectra 0- of a number of delocalized carbanions e.g. (147) have been recorded and the chemical shifts shown to correspond with a high charge density at the odd carbon positions as predicted.' 54 Rates of reaction with electrophiles (indicated by product structures) are not entirely according to the charge densities but attack at internal positions is somewhat preferred.+142 p.p.m. Acidity measurements by the equilibration method using cyclohexylamide have been made on thiophene (a-H pK = 38.42) benzothiophen (a-H pK = 37.05),benzofuran (a-H pK = 36.84) isothiazole (2-H pK = 29.50) and benz[d J-isothiazole (2-H pK = 28.08). Referred to 9-phenylfluorene (pK = 18.49) the internal precision of these measurements is extremely high.'55 Acidities (by the kinetic method) of phenylmethanes and related compounds have been determined. A satisfactory Br~rnsted relationship between pK for this reaction and pK for known thermodynamic acidities has permitted a value for the pK of toluene (40.9) to be obtained by extrap~lation.'~~*'~~ An other potentially very useful relation- ship has been found between heats of deprotonation of weak Br~rnsted acids (by DMSO-) and thermodynamic acidities over a range of 50 pK units and in- cluding carboxylic acids alcohols and weak carbon acids.' 58 G.B. Trimitsis E. W. Crowe G. Slomp and T. L. Hills J. Amer. Chem. SOC.,1973 95 4333. G. A. Abad S. P. Jindal and T. J. Tidwell J. Amer. Chem. SOC.,1973 95 6326. Is' D. D. Davies and W. B. Bigelow Tetruhedron Letters 1973 149. R. B. Bates S. Brenner C. M. Cole E. W. Davidson G.D. Forsythe D. A. McCombs and A. S. Roth J. Amer. Chem. SOC.,1973 95 926. A. Streitweiser and P. J. Scannon J. Amer. Chem. SOC.,1973,95 6273.Is' A. Streitweiser P. H. Owens G. Sonnichsen W. K. Smith G. R. Ziegler H. M. Niemayer and T. L. Kruger J. Amer. Chem. SOC.,1973.95 4254. Is' A. Streitweiser M. R. Granger F. Mores and R. A. Wolf J. Amer. Chem. SOC.,1973 95. 4257. E. M. Arnett T. C. Moriarty L. E. Small J. P. Rudolph and R. P. Quirk J. Amer. Chem. SOC.,1973.95 1492. 148 N. S. Isaacs Cyclo-octatetraene and some derivatives equilibrate with their dianions to form anion-radicals. Thermodynamic parameters for some of these equilibria have been obtained and indicate the entropy term to account largely for differences in K.ls9 Carbanions undergo arylation by a chain mechanism involving inter- mediate anion-radicals (SRNl mechanism) ( 148).160 PhBr + In' -+ PhBr; + In PhBr; + Ph' + Br-Ph' + -* ...-:*.-Ph A carbanionic elimination (Elcb) is proposed for dehydrohalogenations of (149) in which there is a negligible primary isotope effect a positive reaction constant (p = 3.94) and P-hydrogen exchange prior to reaction.16' The base- catalysed halogenation of acetylenes shows third-order kinetics rate = k[RC-CH] [OH-] [OCl-1 a very small solvent isotope effect and a moderate PhCHClCF,X (149) positive p-value (0.8).162It is proposed that the fast reversible ionization of the acetylene is followed by co-ordination of hypohalite ion to form a (structurally unspecified) dianion R-C-COC12- which then reacts with water etc.to give the halogenoacetylene. Base-catalysed dehydration of P-ketols such as 9-hydroxy-1O-methyl-2-decalone has been studied ; the mechanism proposed involves a general-base- catalysed a-proton removal and expulsion from the enolate ion which may be rate-determining at high catalyst concentration.The kinetics are complex as there are probably several steps of comparable rate.'63 8 Carbonyl Reactions Relaxation techniques have been used to measure rates of the fast amine addition reactions to pyridine-4-carboxaldehyde ; the value for piperazine is more than lo51mol-' s-' 164 Attention has been drawn to a number of carbonyl reactions G. R.Stevenson and J. G. Concepcion J. Amer. Chem. SOC.,1973 95 5692. I6O R.A. Rossi and J. F. Bunnett J. Org. Chem. 1973 38 3020. 16' H. F. Koch D. B. Dahlberg P. G. Toczko and R.L. Salsky J. Amer. Chem. SOC.1973,95 2029. R.-R.Liu and S. I. Miller J. Amer. Chem. Suc. 1973 95 1602. D. J. Hupe M. C. R. Kendall G. T. Sinner and T. A. Spenser J. Amer. Chem. SOC. 1973.95 2260 2271. 164 H. Diebler and R. N. F. Thorneley J. Amer. Chem. Suc. 1973.95 896. Reaction Mechanisms -Part (iv) Polar Reactions 149 in which the rate-determining steps are encounter-limited,16’ usually proton transfers. For example the thiolester (150) hydrolyses by the mechanism shown 4-0 CF3Cq; + RSH 0 for which estimated rate constants are k-x lo9 k x 4 x lo91mol-’ s-’. The nucleophilic fission of the a-disulphone (151) by 20 nucleophiles shows a rate dependence parallel to their reactivity towards the carbonyl group. Rates 0 II Ph-S02-SOz-Ph Ar0-S -C1 II (151) 0 are fast necessitating stopped-flow measurements.’ 66 A normal reactivity scale of nucleophiles towards the arylchlorosulphates (152) is found and the favoured mechanism is of a displacement at chlorine.’ 67 Hydrolysis of imidate esters (153) shows differences in behaviour depending upon the basicity of the OR2 R’-C / ORz R’+Nh R3 ORz R‘+-NHR >NHR3 OH OH OR2 11 + 11 -+ 11 -+ Rf+NHR3 ORZ ORZ 0- R’-C / ‘NR3 (153) \ RtTNHR3 R’.+YH /O- NH R/ R3NH2+ R1COzR2 or R1CONHR3+ RzOH Scheme 2 165 R.E. Barnett Accounts Chem. Res. 1973 6,41. J. L. Kice and E. Legan J. Amer. Chem. SOC.,1973,95 3912. 16’ E. Buncel A. Raoult and L. A. Lancaster J. Amer. Chem. SOC.,1973 95 5964. 150 N.S. Isuacs amine moiety. Esters derived from strongly basic amines show a decreasing amount of the amine product with pH whereas the opposite is true for those derived from weakly basic amines.16* The reasons no doubt lie in the complex reaction scheme (Scheme 2) in which a series of intermediates differing in their position and state of protonation can each give rise to amine or amide products.A sulphurane (154)has been shown to cleave amides efficiently a hydrogenation step is needed to complete the rea~ti0n.l~~ N.m.r. spectra ofa series of protonated 0 / PhC \ H -Pd Ph,S i:" OC-CF3 Ph,S=NPh *Ph,S + PhNH CF 2 carboxylic acid anhydrides which are stable species in superacid solution have been reported maleic anhydride for example protonates at the carbonyl group to give (155).l7' +OH ''* T.Okuyama T. C. Pletcher D. J. Sahn and G. S. Schmir J. Amer. Chem. Soc. 1973 95 1253. J. A. Franz and J. C. Moitu J. Amer. Chem. SOC.,1973,85 2017. "O G. A. Olah Y. K. Mo,and J. L. Grant J. Org. Chem. 1973,38 3207.