4 Reaction Mechanisms Part (ii) Polar Reactions By J.M. PERCY School of Chemistry University of Birmingham Edgbaston Birmingham B 15 2TT UK 1 Introduction The whole 1992 volume of Advances in Physical Organic Chemistry will interest readers of this chapter. Reviews by Williams,'" Lee,'b Bernasconi," and Ta-Shma and Rappoport Id covered respectively the relationship between effective charge and transition-state structure the use of cross interaction coefficients in the diagnosis of reaction mechanisms the principle of non-perfect synchronization and solvent- induced changes in the selectivity of solvolysis reactions. The first issue of Accounts of Chemical Research contained a number of relevant articles one described Bunnett's work on the dehalogenation of aryl halides by SET processes;2 mechanisms of amide hydrolysis in aqueous media were also disc~ssed.~ Elsewhere the conceptual basis of face-selectivity in carbonyl-group addition reactions was apprai~ed,~ and aspects of aromatic chemistry were reviewed including the electrophilic substitution reactions of heterocyclic compoundsS and vicarious nucleophilic substitutions.6 The roles of ion-pair exchange processes and their manifestation in salt effects were reviewed,' and scales of solvent ionizing power based upon the solvolyses of benzylic substrates were surveyed.8 A number of aspects of nucleophilic aliphatic substitution were summarized and discussed including the intervention of single electron transfer (SET) processes' and theoretical aspects of the S,2 reaction." Other topics which received attention included mechanisms for the formation of ozonides," and the use of the endocyclic restriction test for the elucidation of transition-state geometries.' A set of guidelines for the publication of results in (a)A.Williams Advances in Physical Organic Chemistry 1992,28 1 ;(b)I. Lee ibid.,57; (c)C. Bernasconi ibid. 119; (d) R. Ta-Shma and Z. Rappoport ibid. 239. J. F. Bunnett Acc. Chem. Res. 1992 25 2. R.S. Brown A. J. Bennett and H. Slebocka-Tilk Acc. Chem. Res. 1992 25 481. W. J. LeNoble Red. Trav. Chim. Pays-Bas 1992 111 199. A. R. Katritzky and W. Q. Fan Heterocycles 1992 34 2179. M. Makosza Pol. J. Chem. 1992 66 3. A. Loupy B. Tchoubar and D. Astruc Chem. Rev.1992 92 1141. D.N. Kevill and M.J. D'Souza J. Phys. Org. Chem. 1992 5 287. J.M. Saveant New J. Chem. 1992 16 304. lo S. S. Shaik H. B. Schlegel and S. Wolfe 'Theoretical Aspects of Physical Organic Chemistry. The S,2 Reaction' Wiley New York 1992. R. L. Kuczowski Chem. Soc. Rev. 1992 21. 79. P. Beak Ace. Chem. Res. 1992 25 215. 55 56 J.M. Percy physical organic chemistry was defined.' Books of general interest included the fourth edition of a well-known clas~ic'~ and a short but highly accessible text on polar rearrangements.' A survey of the reactions of C, may prove a useful source work for physical organic chemists wishing to bring the rigour of a quantitative approach to bear on this fascinating and challenging new molecule.'6 2 Solvolysis and Carbocations Diarylmethyl and triarylmethyl cations were generated in aqueous acetonitrile solutions containing amine nucleophiles using the laser flash photolysis technique.' In highly aqueous media amine desolvation became rate determining.Evidence for this included a change of sign (from positive to negative) in fiN with increasing amine basicity. A good correlation was observed between fiN and pK,+ demonstrating that amine nucleophiles do not adhere to the N + constant selectivity relationship. X Id" / + (1) Y = OTS,OBZ,N3 (2) A significant piece of work has bridged the divide between cations that follow the reactivity-selectivity principle and those for which Ritchie's N + scale applies.' * Varying X in (1) allowed precise control of cation reactivity; at the most stable extreme (X = NMe,) behaviour comparable to that of the least stable triarylmethyl cations was detected.As X became less powerfully electron-donating nucleophilic selectivities became increasingly dependent on the electrophile. A recommendation was made' concerning the 9-fluorznyl cation (2). Using combined theoretical and experimental approaches it was shown that the anticipated destabilizing effect of antiaromaticity was very small. The authors concluded that we should no longer refer to (2) as antiaromatic. A number of publications described cations containing cyclopropyl groups. Olah and co-workers2' demonstrated that (3a) and (4a) co-exist at -90 "C while (3b) only commences reorganization to (4b) at -70 "C.The presence of a second cyclopropyl group strongly stabilizes (3c); (4c) only becomes detectable above -20 "C. The authors were unable to detect participation by the C-1-C-6 a-bond in any of these cases. In contrast protonation2 of alkene (5)afforded the symmetrically-bridged cation (6). l3 W. Drenth Pure Appl. Chem. 1992 64 989. 14 J. March 'Advanced Organic Chemistry Reactions Mechanisms and Structure' 4th Edn. Wiley New York 1992. L. Harwood 'Polar Rearrangements' Oxford University Press Oxford 1992. l6 F. Wudl A. Hirsch K.C. Khemani T. Suzuki P. M. Allemand A. Koch H. Eckert G. Srdanov and H. M. Webb ACS Symp. Ser. 1992 481 161. R. A. McClelland V. M. Kanagasabapathy N. S. Banait and S. Steenken J. Am. Chem. SOC.,1992,114 1816.J. P. Richard T.L. Amyes and T. Vontor J. Am. Chem. SOC. 1992 114 5626. l9 T. L. Amyes J. P. Richard and M. Novak J. Am. Chem. SOC.,1992 114 8032. 2o G. A. Olah V. Prakash Reddy G. Rasul and G. K. Surya Prakash J. Ory. Chem. 1992 57 11 14. W. Kirmse H. Landscheidt. and A. Schleich J. Phys. Org. Chem. 1992 5 19. Reaction Mechanisms -Part (ii) Polar Reactions The 1-ferrocenyl- 1 -cyclopropyl cation (7) was observed by NMR spectroscopy and showed extensive charge delocalization into the ferrocenyl moiety.22 (3) a R = Me; (4) a,R=Me; (5) (6) b R = Ph; b R = Ph; c R = cyclopropyl C,R = cyclopropyl The importance of hyperconjugative interactions by the p-silyl group in cation (8) was probed by NMR experiments and ab initio calculation^.^^ The search for observable trivalent silicenium cations was reviewed by two gro~ps.*~*~~ Both concluded that all attempts to generate these species as long-lived intermediates in solution have been unsuccessful.The reduction of carbenium ions by hydrosilanes forms the basis of a useful synthetic reaction. Mayr and co-workers26 have compiled a quantitative scale of hydrosilane reactivity which parallels the anticipated ease of formation of carbenium (and by implication) silicenium-type species. A Hammett correlation was reported (p = -2.46 with up)for the reduction of diarylmethyl cations by aryldimethyl silanes. H Aspects of polyenyl chemistry included the electrocy~lizations~~ of (9a) and (9b) which occurred in dichloromethane at -70 “Cupon treatment with fluorosulfuric acid to afford (10a) and (lob).The cyclizations traversed dicationic intermediates formed via protonation at oxygen. High activation entropy (AS) values were measured and a strong dependence of reactivity upon acid strength was detected consistent with the dicationic mechanism. These reactions are related to the synthetically-useful Nazarov cyclization. Laser flash photolysis was employed to study the reactions of pentadienyl cation (1 1) with nucleophiles. Picosecond laser experiments allowed the detection of contact ion 22 G. K. Surya Prakash H. Bucholz V. Prakash Reddy A. de Meijiere and G.A. Olah J. Am. Chem. Soc. 1992 114 1097. 23 H.-U. Siehl F.-P. Kaufmann and K. Hori J. Am. Chem. Soc. 1992 114 9343. 24 P. D. Lickiss J.Chem. SOC.,Dalton Trans. 1992 1333. ” G. A. Olah G. Rasul L. Heiliger J. Bausch and G. K. Surya Prakash. J. Am. Chem. Soc. 1992,114,7737. 26 H. Mayr N. Basso and G. Hagen J. Am. Chem. Soc. 1992. 114 3060. ” G.R. Elia R. F. Childs and G.S. Shaw Can. J. Chrm. 1992 70 2065. J. M. Percy pairs which dissociated to free ions at longer reaction times in polar solvents. The free carbenium ion obeyed the reactivity-selectivity principle displaying good correlations with Swain-Scott n values.28 The slopes of the correlations were strongly solvent dependent decreasing in magnitude as the water content of the photolysis medium was reduced. An unusual 1,3-hydride shift was claimed to explain the outcome of the annulation reaction depicted in Scheme 1.Isotopic labelling studies allowed an alternative explanation based on consecutive 1,2-shifts to be eliminated.29 OMEMBu SiMe 1.3-shift ' SiMe3 -Aru -bz, TiCb SiMe Scheme 1 3 Other Nucleophilic Substitutions The mechanisms of nucleophilic substitution at allylic and primary alkyl carbon centres were found to have many features in common. A study of ally1 arenesulfonates '' N.J. Pienta and R. J. Kessler J. Am. Chem. Soc. 1992 114 2419. 29 S. R. Angle and H. L. Mattson-Arnaiz J. Am. Chem. SOC. 1992 114,9782. Reaction Mechanisms -Part (ii) Polar Reactions (12) failed to show the anticipated similarities with benzylic species. Lee and co-workers3' reached this conclusion from a study of linear free-energy relationships and derived cross-in teraction coefficients.Cyclophane hosts (13a) and (13b) were shown to be effective catalysts for the methylation of quinolines and the demethylation of dialkylsulfonium cation^.^ Catalysis was attributed to a favourable interaction between these highly-polarizable hosts and the delocalized sN2 transition states. co -02c' The balance between intramolecular SN2 and E2 reactions was investigated to elucidate the controlling effect of strain energy on reaction pathways.32 Scheme 2 shows two systems which straddle the borderline between the two pathways. The determinant quantity is the excess enthalpy differential (EED) the extra strain-energy difference between the acyclic starting material and the cyclic product incurred upon cyclization.A limiting value of 160 kJ mol-was proposed; cyclizations incurring a higher EED were not observed. Nucleophilic displacements at silicon were studied33 in aryloxysilane (14) which undergoes substitution by carboxylate anions in DMF solution. Brsnsted coefficients (& = 1.O PL = 1.9) were reported and despite cross interaction coefficients of zero a concerted mechanism was preferred to a stepwise addition4imination via a penta- covalent intermediate. The effective charge at silicon was calculated for the reaction. Two-electron mechanisms were claimed34 for the nucleophilic ring-opening of cyclic peroxides (15). Thiocyanate anion attacked at the less-substituted oxygen to afford an unstable adduct. Products arising from intramolecular attack at carbon (16) and sulfur (17) were isolated.Other nucleophilic reactions at sulfur include the methan~lysis~~ of H. K. Oh H. J. Koh and I. Lee J. Chem. SOC..Perkin Trans. 2 1992 1981. 3' A. McCurdy L. Jiminez D.A. Stauffer and D.A. Dougherty J. Am. Chem. SOC.,1992 114 10314. 32 S. M. Jeffery S. Niedoba and C.J. M. Stirling J. Chem. Soc.. Chem. Commun.. 1992 650. 33 P. E. Dietze. J. Org. Chem. 1992 57 1042. 34 W. Adam and M. Heil J. Am. Chern. SOC.,1992 114 5591. 35 J. L. Kice and A.G. Kutateladze J. Org. Chem. 1992 57 3298. J. M. Percy PhSO B~OK B~OH 100% 0% EED = 138 kJ mole-’ PhSO2 B~OK B~OH phso20 0% 100% EED = 212 k~ mole-’ Scheme 2 sulfenamides (18). A sulfuramide intermediate (19)was proposed to lie on the pathway for this reaction.0 It Nucleophilic attack at phosphorus by water is an important biological process. It was shown that synthetic host (20),modelled on the active site of on the arginine diad of staphyloccocal nuclease bound the pentacovalent intermediate of phosphodiester hydrolysis more strongly than the phosphodiester monoanion and additionally eased leaving group departure by acting as a general-acid catalyst. An acceleration of phosphodiester hydrolysis was thus achieved.36 36 V. Jubian R. P. Dixon and A. D. Hamilton J. Am. Chem. Soc. 1992 114 1120. Reaction Mechanisms -Part (ii) Polar Reactions 4 Elimination Reactions Alkenes were formed3' as the sole products upon solvolysis of the esters (21) in 80% aqueous acetone.However p-deuterium isotope effects were too small to be consistent with a pericyclic elimination mechanism. Instead an explanation in which proton-loss occurs from an ion-pair intermediate was proposed. A nitrile-forming elimination was proposed38 as the key step in the hydrolysis of carcinogen MNNG (22). A curved region in the pH-rate profile above pH 6.5 required a term that was first-order in hydroxide ion and the conjugate base of MNNG consistent with the mechanism shown in Scheme 3. /I OH-Scheme 3 Decomposition of potassium-(E)-methane diazoate (23) involves an elimination of methanol across an N=N bond from the diazoate (24).The pH-rate profile an uncatalysed region below pH 7 and allowed an apparent pK of 8.63 to be determined for the conjugate acid of (23).Sulfene intermediate (25) was detected during the hydrolyses of methylsulfonyl ~hloride.~' At low pH (56.7) direct attack at sulfur by water was the dominant 37 X. Creary H.N. Hatoum A. Barton and T. E. Aldridge J. Org. Chem. 1992 57. 1887. 38 C. L. Galtress P. R. Morrow S. Nag. T. L. Smalley M. F. Tschantz J.S. Vaugh D. N. Wickens. S. K. Ziglar and J. C. Fishbein J. Am Chem. Soc. 1992. 114. 1406. 3y J. Horvinen and J.C. Fishbein .I. Am. Chem. Soc. 1992 114. 366. 40 J. F King J. Y. L. Lam and S. Skonieczny J. Am. Chern. Soc.. 1992. 114 1743. J. M. Percy pathway. However at high pH (26.7) sulfene formation by elimination of HCl became rate-determining while above pH 11 sulfene formation and interception by hydroxide occurred at comparable rates.The measurement of isotope effects allowed (ElcB),, and (ElcB)irrev mechanisms to be discounted in favour of an E2 process. The choice between concerted and stepwise pathways was encountered41 in the base-catalysed ring opening of (26)to (27). The base-catalysed elimination occurred by the rare (ElcB) mechanism. In acid the (ElcB) mechanism was observed with loss of the neutral phenolic leaving group at rates close to the encounter limit. Both acid and base catalysed eliminations occurred oia transition states that were very early with respect to C-0 bond breaking. The data allowed the nucleofugacity of the phenolic leaving group to be ranked using Stirling’s approach. 5 Addition Reactions The relief of steric strain energy has been implicated as a driving force in many important types of reaction.Shea and Kim42 have quantified the influence of strain release on the rates of alkene epoxidation by mCPBA. A good correlation was obtained between relative epoxidation rates and the calculated (MM2)strain energy released in each reaction (ASE). Attempts to use alkene frontier orbitals to predict reaction rates gave poor results. Transannular cyclizations of acetylenic ketones are useful reactions formally involving an addition reaction across an alkyne. Harding and King43 have addressed the mechanism via a study of an acyclic system and proposed that the cyclization follows the pathway depicted in Scheme 4.Isotopic labelling and careful product analysis studies were conducted to support the claim that the oxete species (28) was the key intermediate.0 Scheme 4 A number of publications dealt with addition reactions of substituted alkenes. Solvent isotope effects indicated that enol phosphate (29)underwent hydrolysis in acid 41 R.G. Button and P.J. Taylor J. Chem. SOC. Perkin Trans. 2 1992 1571. 42 K. J. Shea and J.-S. Kim J. Am. Chem. SOC. 1992 114 3044. 43 C.E. Harding and S.L. King J. Org. Chem. 1992 57 883. Reaction Mechanisms -Part (ii) Polur Reactions 63 uia rate-limiting protonation of the double bond,44 rather than the P-0 cleavage observed for simple alkyl or aryl phosphates. The general acid catalysed hydrolyses of highly nucleophilic ketene acetals (30) and (31) were studied; average Brsnsted a-values were reported for a set of eight ketene acetals (a = 0.44 & 0.08) including (30) and (3 1).The average exponent for a similar set of vinyl ethers had a higher value (a = 0.67 0.08) consistent with lower hydrolytic reactivity. However these suggested that the ketene acetal hydrolysis reactions have an intrinsically higher barrier and that this arises from more extensive charge delocalization in the more highly-oxygenated series. Meo-foMe OMe OMe R\”’ (29) (30) (31) (32) a X = COT b X = CO2H c X = C02Me Vinyl ethers bearing an a-carboxyl function (32a-c) were in~estigated,~~ using excess acidity (X)functions as models for 5-enolpyruvylshikimate-3-phosphate, EPSP (33). The nature of the carboxyl function controlled the hydrolytic reactivity.All the compounds were less reactive than methyl vinyl ether with (32a) being the most reactive suggesting that this is the form of EPSP maintained by the EPSP synthase enzyme. Addition reactions to ketenes continued to attract attention and a number of studies were described. Analysis of the activation parameters and substituent effects for the hydrolyses of a series of ketenes (34) supported a transition state with significant enolate character (35),developed uia attack at the carbonyl group and not at the C=C bond.47 The methylthio group in (36a) slowed the rate of uncatalysed hydration by a factor of 36 relative to phenylketene (36b) whereas a hydroxyl group at the same position (36c) accelerated the same reaction by a factor of 140.It was concluded that the hydroxyl substituent raised the ground-state energy of the ketene and that the bulkier methylthio group sterically hindered the attack of water.48 The reaction of diphenyl ketene with water in the presence of base resulted in direct attack by the base at the carbonyl group rather than general-base catalysed addition of water.49 Nitrone (37) underwent hydrolysis via direct attack by water between pH 4.5 and 10. Acid and base-catalysed pathways were detected and Hammett correlations were reported. 50 Arylnitroso compounds have been identified as active metabolites of arylnitro species and aromatic amines and their reactions with biological nucleophiles have been implicated in a range of toxic effects. Kazanis and McClellandsl have studied the 44 A.S. Kearney and V. J. Stella Pharm. Res. 1992 9 378. 45 A. J. Kresge and M. Leibovitch J. Am. Chem. Soc. 1992 114 3099. 46 A.J. Kresge M. Leibovitch and J.A. Sikorski J. Am. Chem. SOC. 1992 114 2618. 47 A.D. Allen J. Andraos A. J. Kresge M. A. McAllister and T.T. Tidwell J. Am. Chem. SOC.,1992 114 1878. 48 J. Jones and A. J. Kresge J. Urg. Chem. 1992 57 6467. 49 J. Andraos and A. J. Kresge J. Am. Chem. SOC. 1992 114 5643. 50 S. W. Lee C.G. Kwak I. Kwang and K.C. Lee J. Korean Chem. SOC. 1992. 36. 584. 51 S. Kanzanis and R. A. McClelland J. Am. Chrm. Soc. 1992 114 3052. 64 J. M. Percy reaction of nitrosobenzene with glutathione and proposed the mechanism depicted in Scheme 5. Monothioacetal (38) underwent reduction in the presence of glutathione anion via attack at sulfur to afford the corresponding hydroxylamine.In addition a co, I (34) R = alkyl alkynyl alkenyl (33) trifluoromethyl (35) R (36) a R=MeS (37) b,R=H C R = HO rearrangement pathway led to sulfinanilide (39). Uncatalysed and specific and general-acid catalysed pathways were detected for the latter reaction. Cleavage of the N-0 bond was found to be rate determining and isotope tracer and linear free-energy relationship studies (p' = -3.5) implied that nitrenium ion (40) was the key intermediate. This is unlike the course of the Bamberger rearrangement in which nucleophilic attack at the ring precedes N-0 cleavage. SG GSH GSSG H GSH ' L-L' * ArON,OH Ar/"OH GSH = Glutahone (38) H (40) Scheme 5 6 Aromatic Addition and Substitution Rate constants (105-109M-I s-' ) were reported for the reactions between the photogenerated 9-fluorenyl cation and a number of substituted benzenes in CF,CH(OH)CF,.These are formal models for Friedel-Crafts alkylation reactions.52 52 F. Cozens J. Li R.A. McClelland. and S. Steenken Anqew. Chem. Int. Ed. Engl. 1992 31 743. Reaction Mechanisms -Part (ii) Polar Reactions Second-order rate constants ranged from 3.3 x lo5M-Is-for benzene to 1.7 x lo9M-s-' for cation capture by meta-xylene which marked the upper rate limit for these reactions. The reaction with toluene showed high para-selectivity (0:m:p 5 5:90) despite the high reactivity of the nucleophile (k = 1.1 x lo7M-' s-'). A Hammett correlation with a,'(p+ = -5.0) was reported that was consistent with the accepted mechanism via a cyclohexadienyl cation.For arene nucleophiles more reactive than toluene a-adducts were observed directly. Marcus theory has been applied to the nitration and nitrosation reactions of aromatic compounds and allowed some conclusions concerning the nature of the active electrophiles to be drawn.53 The SET model for nitration predicted that H,NO and not NO was the reactive nitrating agent in nitric acid mixtures. The intervention of the electron transfer mechanism has regiochemical consequences; for example attack at C-1 in naphthalene becomes a favoured pathway. Activity coefficient (M,) and excess acidity (X) methods have been applied to diagnose desulfonation mechanisms for aromatic compounds.54 Isotopic exchange in N-aryl-2,5-dimethyl pyrroles (41) showed a low electronic sensitivity (p = -0.90)to substitution in the aromatic ring due to steric impedence of conjugation.Methylation at C-2' and C-6' increased the rate of detritiation by differentially inhibiting solvation of the conjugate acid but not the transition state for the detritiation step.55 Metal-coordinated (q2)-pyrroles showed some interesting acid-base and tautomeric behaviour. Coordination to pentaamineosmium(I1) facili- tated the 1H-to 2H-pyrrole equilibrium (42) to (43) (Keq= 1) and directed the attack of electrophiles to C-fi.It followed that the uncomplexed portion of the ligand was behaving as an enamine with one strongly-screened face resulting in highly stereo- and regioselective protonations.A number of pK values for the various tautomers were reported.56 The reactions of n-excessive heteroaromatics with 4,6-dinitrobenzofuroxan (DNBF) (44) showed that this carbon electrophile is more reactive than either p-nitrobenzene diazonium cation or the pr~ton.~ The reactions with indole 2-methylindole and s3 J.P. B. Sandall J. Chem. SOC.,Perkin Trans 2 1992 1689. 54 E.N. Krylov Zh. Obsch. Khim. 1992 62 147. 55 J. R. Jones S. Hunt F. Terrier and E. Buncel J. Chem. SOC..Perkin Truns. 2 1992 295. 56 W. H. Myers J. I. Koontz and W. D. Harman J. Am. Chem. SOC. 1992 114 5684. " F. Terrier E. Kizilian J.-C. Halle and E. Buncel J. Am. Chem. SOC.. 1992 114 1740. J. M. Percy 1,2,5-trimethyIpyrrole involved rate determining attack at C-7 to afford zwitterionic o adducts from which proton loss was facile.A number of aspects of the S,Ar mechanism were discussed. Even substrates that would normally be considered unreactive undergo substitution at reasonable rates. No2 Hengge used "0 labelling to show that the hydroxyl group of 4-nitrophenol exchanged in dilute (0.55 M) alkaline solution at an observable rate (t+ = 74minutes at 1000C).58Gandler and co-workersS9 measured rate constants for the reactions of picryl chloride and chloro-2,4-dinitrobenzenewith nucleophiles in aqueous and methanolic solutions. A modified Ritchie equation (equation 1 ) logk = S'N + logk (1) was used to treat the results. The electrophile-dependent parameter S+ varied from (0.79 a 0.1 1) for the more reactive picryl halide to (0.95 f0.13)for the less reactive halide consistent with the reactivity-selectivity principle.The reactions of 2,4,6-trinitroanisole with hydroxide methoxide and phenoxide in mixed aqueous/dipolar aprotic solvents afforded a diverse array of products.60 However allowing the systems to reach thermodynamic product mixtures led to simplification. With phenoxide anion the final irreversibly-formed product was (45) whereas with methoxide (46) dominated the equilibrium mixture. A range of kinetic thermodynamic and stereoelectronic factors were discussed to rationalize these observations. Hydrolysis of the C-4 methoxy group in (47) occurred very readily in aqueous sulfuric acid at 25 "C.Protonation of the diazo group was followed by ips0 attack by water at C-4.Rate determining protonation at C-6 initiated the considerably slower hydrolysis of the C-3 methoxy group.61 Excess acidity methods were employed A.C. Hengge J. Am. Chem. SOC. 1992 114 2747. 59 J. R.Gandler 1. U. Setiarakardjo C. Tufon and C. Chen J. Org. Chem. 1992 57 4169. 6o E. Buncel J. M. Dust A. Jouczyk R. A. Manderville and I. Onyido J. Am. Chem. SOC.,1992,114,5610. 61 R.A. Cox I. Onyido and E. Buncel J. Am. Chem. SOC. 1992 114 1358. Reaction Mechanisms -Part (ii) Polar Reactions allowing a number of pKIH and m* values to be obtained for the protonated phenylazopyridines. The mechanism of amino-migration in 0-phenylhydroxylamine was studied.62 In trifluoroacetic acid ortho-rearranged products predominated unlike the Bamberger OMe 6 (47) rearrangement which leads to para-isomers.A Hammett correlation (0') with a large negative slope (p' = -7.8) indicated that initial N-0 cleavage occurred to generate a phenoxenium cation with extensive charge delocalization into the aromatic ring. Recombination of the ion-molecule pair led to ortho-aminophenols; interception of the ion-molecular pair by trifluoroacetate led to catechols and hydroquinones. 7 Proton Transfer and Carbanions Stabilized carbanions reported in 1992 included the conjugate bases of tris(tri- fluoromethanesu1fonato)methane (48) and 4,6-dinitro-7-methylbenzofuroxan(49).A pK of -12 was measured for (48) which is a sufficiently strong acid to protonate diethyl ether.63 The extraordinary charge delocalizing ability of the benzofuroxan system was demonstrated by the low pK (2.50) for C-H ionization exhibited by (49).This was confirmed by the very high intrinsic barrier for the deprotonation reaction (AGt = 85.5 kJ mol- ') consistent with the extensive molecular electronic and solva- tional reorganization required to form a highly delocalized anion.64 I 0-I N+ H+S02CF3 Stable cyclopropyl anions present a synthetic challenge; two groups reported their efforts in the area. Crispino and Bres10w~~ have described several air and water-stable tris(pyridiniumy1)propenyl anions (which are in effect actually dications) including (50). The presence of a cyclopropyl anion intermediate was inferred in the reaction of (51) with cyanide to afford ring-opened products.66 62 N.Haga Y. Endo K. Kataoka K. Yamaguchi and K. Shudo J. Am. Chem. So(..,1992 114 9795. 63 Y. L. Yagupol'skii T. I. Savina N. 0. Pavlenko A. A. Pankov. and S. A. Pazenok Zh. Ohshch. Khim. 1991 61 1512. 64 F. Terrier D. Croisat A. P. Chatrousse M. J. Pouet and J.-C. Halle J. Org. Chem. 1992 57 3684. G. R. Crispino and R. Breslow J. Org. Chem. 1992 57 1849. 66 A. S. Feng S.G. DiMagno M. S. Konings and A. Streitweiser J. Org. Chem. 1992 57 2902. J.M. Percy Me The formal antiaromatic carbanion (52) was generated via an unusually facile photochemical decarboxylation rea~tion.~’ This may prove an interesting method for revealing fundamental properties of carbanions in the way that laser flash photolysis has enriched our knowledge of carbocations.An unusual carbanionic rearrangement occurred upon treatment of (53)with LDA. Vinylsilane (54) was obtained as a mixture of E and 2 isomers following a 1,2-migration of the silyl group with concerted leaving group departure. An alternative mechanism involving migration of the phenyl group was discounted on the basis of labelling studies.68 The question of how enzymic systems deal with proton transfers to and particularly from carbon remains the subject of dispute. There are two main areas of contention and they concern the nature of the intermediates arising from formal proton transfer at enzyme active sites and how enzymes achieve high reaction rates for proton transfers.Two contrasting observations were reported concerning the status of enolates. Gerlt and Gassman concluded that enzyme-catalysed p-elimination reactions to afford a$-unsaturated carbonyl compounds (vinylogous E2 elimination) occurred via the formation of en01s.~~ A combination of enforced concerted general-acid and general- base catalysis from active site residues was proposed to prevent the formation of unstable intermediates. The conclusions arose from considerations of the pK,s of substrates and active site residues and the general model was discussed with reference to a number of specific enzyme-catalysed reactions. An alternative approach sought to generate simple models of putative intermediates away from the enzyme and establish their chemical stability.This approach led Amyes and Richard to generate thioester enolate (55) in quinuclidine-buffered aqueous ~olution.’~ A pK was estimated (20.4-2 1.5) and an 6’ E. Krogh and P. Wan J. Am. Chem. SOC.,1992 114 705. 68 S. Menichetti and C.J. M. Stirling J. Chem. SOC..Perkin Trans. 2 1992 741. 69 J.A. Gerlt and P.G. Gassman J. Am. Chem. SOC. 1992 114 5928. ’O T. L. Amyes and J.P. Richard J. Am. Chem. SOC. 1992. 114 10297. Reaction Mechanisms -Part (ii) Polar Reactions 69 intimate ion pair with the quinuclidinium cation of lifetime 10-9-10-10 s was detected. The authors therefore concluded that enzyme-catalysed Claisen condensa- tions do not occur by enforced concerted mechanisms. On the question of efficiency an effective molarity of 104-105 M was estimated for exchange of the methine proton in (56) an extremely facile reaction even at -80 "C in toluene.71 Examination of the crystal structure of (56) revealed a close contact between 0-the nitrogen atom and the exchanging proton; the authors concluded that the combination of proximity and reactivity supported the concept of spatiotemporal control.8 Carbonyl Derivatives The molybdate dianion was shown to be a more reactive nucleophilic catalyst for the hydrolysis of 4-nitrophenyl acetate and thioacetate than the basicity of the dianion would lead one to predict. It was suggested that either the anion was weakly solvated or that the carbonyl oxygen coordinated to the molybdenum atom providing some Lewis-acid activation.A more complex synthetic catalyst (57)was described by Breslow and Zhang.73 The B-cyclodextrin units provided a binding site for the adamantyl group holding the carbonyl group of ester (58) close to the divalent zinc ion which acted as a Lewis acid and as a source of nucleophile via a coordinated water molecule. The catalyst achieved a rate enhancement of 2.2 x 105-fold over the uncatalysed hydrolysis at pH 7 and a turnover of at least 50. p-CDX p-CDX OpNP (57) p-CDX = P-cyclodextrin (58) pNP = paranitrophenol The aminolysis of 4-nitrophenyl acetate in chlorobenzene was accelerated by glyme co-solvents. Recognition of the zwitterionic tetrahedral intermediate (59)is possible in triglyme via the formation of bifurcated hydrogen bonds to the ammonium protons.Proton transfer to the aryloxide leaving group occurred in a subsequent and non-rate-determining 71 F. M. Menger and K. Gabrielson .I. Am. Chem. SOL... 1992 114 3574. '' B. Wikjord and L.D. Byers J. Am. Chem. SOL..,1992 114 5553. '' R. Breslow and B. Zhang J. Am. Chem. Soc. 1992 114 5882. 74 J.C. Hogan and R. D. Gandour J. Org. Chem. 1992 57. 55. J. M. Percy The presence of divalent barium or strontium cations accelerated the ethanolysis of (60)700-fold. It was proposed that metal cations were bound strongly by the transition state but only weakly by the reactant and that the catalysis arose from the utilization of this differential binding energy.75 Hengge studied the hydrolysis of 4-nitrophenyl acetate using a 5Nisotopic probe in the nitro It was argued that the size of the observed isotope effect was consistent only with a mechanism in which leaving group departure occurred in the rate-determining step.A mechanism involving tetrahedral intermediate formation and breakdown would not satisfy this criterion. The author therefore favoured an AND description for this mechanism. Acyl phosphate (61)was synthesized by a novel proton pump system. The transfer of citraconic anhydride from acid solution (pH 0.3) across a chloroform 'membrane' into alkaline solution (pH 10)drove the synthesis of the high-energy intermediate by the transfer of two protons.77 The hydrolyses of orthocarbonates (62a and b) were studied and shown to be subject to general-acid catalysis. The hydrolysis of (62a) occurred ten times more slowly than that of triphenyl orthoformate while the latter compound and (62b) hydrolysed at the same rate and with very similar sl values.The authors argued that the slow orthocarbonate hydrolysis was due to the instability of trioxacarbenium ions. Stabilization was provided by the electron-releasing methoxy groups in (62b) raising the hydrolysis rate.78 0 (61) (62) a X = H b X = OMe '' D. Kruft R. Cacciapaglia V. Bohrner A.A. El-Fadl S. Harkema L. Mandolini D. N. Reinhoudt W. Verboorn and W. Vogt J. Org. Chem. 1992 57 826. 76 A. Hengge J. Am. Chem. SOC.,1992 114 6515 77 I. J. Colton and R. J. Kazlauskas J. Org. Chem. 1992 57 7005. 70 P. Kandaanarachchi and M. L. Sinnott J. Chem. SOC..Chem. Commun.1992 777. Reaction Mechanisms -Part (ii) Polar Reactions Keto-enol equilibria in the pyruvic acid system have been studied. Acidity constants for ionization of the pyruvate enol (63) (pK = 11.55) at the hydroxyl group and pyruvic acid (64) (pKf = 16.58) ionizing at carbon were rep~rted.'~ The enol was generated directly in aqueous solution by hydrolysis of bis(trimethylsily1) precursor (65)or by rapid dilution of a DMSO solution of (64). The second method exploited the relatively high stability of simple enols in DMSO a good hydrogen-bond accepting solvent. Ketonization of the enol accounts for 47% of the free energy liberated by the conversion of phosphenol pyruvate to pyruvic acid. 9 Other Reactions The mechanism of the Wittig reaction was studied using isotope effects and Hammett correlations.For the reaction between substituted benzaldehydes and benzylidene triphenyl phosphorane (66) in THF large reaction constants (p = 2.77) were obtained when lithium bromide was absent. In the presence of the salt the reaction constant was lower (p = 1.38) leading to the conclusion that the reaction passed through an earlier transition state in the presence of the salt." Azomethine ylides are useful species for the construction of nitrogen-containing heterocycles by dipolar cycloaddition. A method for the generation of these 1,3-dipoles involved' an iodide-catalysed intramolecular alkylation of oxazole (67) in the presence of cyanide anion which then attacked at the hard immonium carbon of (68). Fragmentation of the heterocyclic ring then occurred to generate the ylide (69).Softer nucleophiles (thiophenolate thiocyanate) attacked directly at the S,2 centre in (67).A number of publications dealt with processes related to the Favorskii rearrange- ment. Cordes and Berson described the thermal interconversion of diastereoisomeric 79 Y. Chiang A.J. Kresge and P. Pruszynski J. Am. Chem. SOC. 1992 114 3103. 8o H. Yamataka K. Nagareda K. Ando and T. Hanafusa J. Org. Chem. 1992 57 2865. 81 A. Hassner and B. Fischer J. Org. Chem. 1992 57 3070. J. M. Percy cyclopropanones (70a and b) via oxyallyl (71) (only one rotamer shown).82 Recycliz- ation with inversion of the spiro centre completed the interconversion which was facile (70) a X = CH2 Y = C=O; (71) b X= C=O Y = CH2 in dichloromethane at -80 “C (tt = 80 minutes).A related process occurreds3 in the conversion shown in Scheme 6. Mechanisms involving [3,3]-rearrangement or &2‘ displacement of mesylate from (72) were excluded. Interception of nitrogen-containing oxyallyl (73) with other anionic nucleophiles was possible. a NMe -% NMe NMe Y I Y I Y I OH OMS OMS (72) Y NHMe C1 (73) Scheme 6 A novel functionahzation of c-3 of the /?-lactam ring occurred via the sN2’ displacement of an arenesulfonate leaving group as shown in Scheme 7.84 However it appears possible that the reaction may have proceeded via (74). In any case this mechanistically-interesting transformation may find extensive application in /?-lactam chemistry.An unusual example of a reaction in which a sulfonyl oxygen acted as a nucleophile was reported.85 Treatment of bissulfone (75) with bromine and silver tetrafluoroborate afforded an epimeric mixture of y-sultinium ions (76).The cyclic adducts were opened readily by nucleophiles to afford em-syn products. Attempts to deprotect enantiomeri- cally pure ketal (77) in acid lead to racemization.86 The problem was traced to a pseudo-Smiles rearrangement of diol (78) shown in Scheme 8 activated by N-protonation of the heterocycle. This had implications for the correct choice of deprotection conditions; brief exposure of the ketal to strong acid allowed deprotec- ” M.H. Cordes and J.A. Berson J. Am. Chem. Soc. 1992 114 11 010. 83 R.V. Hoffmann N. K. Kayyar and B.W. Klinekole J. Am. Chrm. Soc. 1992 114 6262. 84 C. M. Gasparski M. Teng and M. J. Miller J. Am. Chem. SOC. 1992 114 2741. 85 V. Lucchini G. Modena and L. Pasquato J. Chem. SOC.,Chem. Commun. 1992 293. 86 J. J. Barlow M. H. Block J. A. Hudson. A. Leach J. L. Londridge. B.G. Main and S. Nicholson J. Org. Chem. 1992 57 5158. 73 Reaction Mechanisms -Part (ii) Polar Reactions -TsN~ EtjN 0'~TS -0 OTs &-N "FI' NpwR-N3gR -0 0 -0 (74) Scheme 7 PhO& pho& ' &?,-0 S0,Ph 0 tion with minimal racemization. This exploited the specific-acid catalysed pathway for ketal hydrolysis which is rapid at very low pH. Below the pK of the heterocycle racemization became pH independent allowing the rate difference between wanted and unwanted reactions to be maximized.(77) R-(+) Scheme 8 In a similar vein a method allowing the yields of certain important reactions to be maximized was prescribed.*' Ketone alkylation and Schotten-Baumann reactions formed the subject of the study which showed how control of pH allowed mono- or dialkylations to be performed or similar functional groups to be acylated selectively. J.F. King R. Rathore J.Y. L. Lam Z. R. Guo and D. F. Klassen. J. Ory. Chem. 1992 57. 3028. J. M. Percy 10 Probes of Polar Reactions Bentley and Jones88 have described new rate-product correlations for reactions in binary solvent mixtures involving general-base catalysis. The selectivities (S) for a number of well-known systems were derived allowing deviations due to mechanistic change to be identified in solvolysis studies.A refinement in the Grunwald-Winstein procedure was proposed which took account of variations in the solvation of groups adjacent to the reaction centre. A Ysimparameter was therefore derived for a range of binary solvent mixtures by measuring solvolysis rates for compounds with alkyl alkenyl aryl and alkynyl groups next to the reaction ~entre.~’ These parameters are anticipated to aid the prediction of changes in reaction rates caused by changing the solvolysis medium. Advantages were claimed for aqueous acetonitrile as the binary mixture of choice for the study of solvolysis reactions over aqueous acetone and aqueous dioxan mixtures. Four different solvent structures were detected in aqueous acetonitrile.” Changes in the structure of aqueous acetonitrile were also detected during the reaction of 4-nitrophenyl acetate with anionic nucle~philes.~~ Solvent effects on the solvolyses of neophyl tosylates were summarized in a new solvent scale (Y,) for P-aryl assisted (k,) solv01yses.~~ The new scale was shown to be more suitable than the more established YoTsfor dealing with processes of this type.Multiparameter approaches to the correlation of solvent effects were described by Gaje~ski’~ and drag^,'^ while Blokzijl and Engberts described a quantitative approach to understanding the hydrophobic acceleration of organic reactions in water and aqueous solvent mixture^.^' Substituent effects were discussed by a number of groups.Lee related the magnitude of cross-interaction coefficients to differences in force constants (AFI) between reactant and product state~,’~ while Hoz presented a non-traditional interpretation of LFER data.97 Exner and co-workers reported that the nature of the effects exerted by meta- and para-substituents on reaction centres are quite different and advised caution in the use of small sets of compounds for constructing Hammett correlations when both positional isomers were present.98 A new fast electrochemical technique was described which allowed SET and S,2 mechanisms to be distinguished.” Bethel1 and Parker showed that predictive relationships exist between charge-transfer transition energies and the reactivity of electrophile-nucleophile reactions.O0 ’’ T. W. Bentley and R.O. Jones J. Chem. SOC.,Chem. Commun. 1992 743. 89 T.W. Bentley J.-P. Dau-Schmidt G. Llewllyn and H. Mayr J. Org. Chem. 1992 57 2387. 90 A. Wakisaka Y. Shirnizu K. Nishi K. Tokurnaru and H. Sakuragi J. Chem. Soc.,Faraday Trans. 1992 88 1129. 9’ I.H. Urn G.J. Lee H. W. Yoon and D.S. Kwon Tetrahedron Lett. 1992 33 2023. 92 M. Fujio M. Goto K. Funatsu T. Yoshino Y. Saeki K. Yatsugi and Y. Tsuno Bull. Chem. SOC.Jpn. 1992 65 46. 93 J.J. Gajewski J. Org. Chem. 1992 57 5500. 94 R.S. Drago J. Org. Chem. 1992 57 6547. 95 W. Blokzijl and J. B. F. N. Engberts Process Technol. 1992 2 49. 9h I. Lee J. Phys. Org. Chem. 1992 5 736. 97 S. Hoz Acta Chem. Scand. 1992 46 503. 98 M. Ludwig S. Wold and 0.Exner Acta Chem. Scand.1992 46 549. 99 D. L. Zhou P. Walder R. Scheffold and L. Walder Helv. Chim. Acta 1992 75 995. loo D. Bethel1 and V.D. Parker J. Phys. Org. Chem. 1992 5 317.