4 Reaction Mechanisms Part (ii) Polar Reactions By D. L. H. WILLIAMS Department of Chemistry University of Durham Durham DHI 3LE 1 Introduction Volume 24 of Advances in Physical Organic Chemistry contains three chapters of interest. Nibbering' discusses the gas-phase reactions of organic ions the review covers the formation of such species as well as many of their reactions including H/ D exchange addition/elimination elimination cycloaddition etc. and stresses the major progress made in the past decade principally resulting from the develop- ment of the necessary sophisticated equipment. Hydride shift and transfer reactions are covered by Watt,* including metal-to-carbon transfers and both anionic and cationic carbon-to-carbon hydride transfers and shifts.The third contribution by Sinnott3 is possibly more controversial and discusses the relative merits of the theory of the principle of least nuclear motion (PLNM) and the antiperiplanar lone pair hypothesis (ALPH) as explanations of factors responsible for stereoelectronic control of reactions. The author argues the case for the former principle mainly on the grounds that the latter is too fundamental and does not cater for observed exceptions to the hypothesis which can be accommodated by PLNM. Two other review articles deal with simple enols one by Capon and co-workers discusses their generation in solution their kinetic stability and also KE determina-tion; the other by Rappoport and Biali' (and dedicated to Professor Saul Patai on the occasion of his 70th birthday) covers the chemistry of sterically crowded simple enols.Here since KE values are generally large the problem in their determination is the accurate measurement of low concentrations of the keto form (the reverse of the more commonly encountered situation). Steric effects alone do not appear to be able to account for the thermodynamic stability of these enols. Two reviews of the importance of SET reactions in organic chemistry have been published. Ashby6 argues that it is a major pathway even for some so-called SN2 processes (but in some cases a2companying the classical mechanism). The SET mechanism is particularly important for substitution in alkyl iodides and is also a major route in reactions of aromatic (but not aliphatic) ketones with some ' N.M. M. Nibbering Adv. Phys. Org. Chem. 1988 24 1. * C. I. F. Watt Ado. Phys. Org. Chem. 1988 24 57. ' M. L. Sinnott Adv. Phys. Org. Chem. 1988 24 113. B. Capon B.-Z. Guce F. C. Kwok,A. K. Siddhanta and C. Zucco Acc. Chem. Res. 1988 21 135. Z. Rappoport and S. E. Biali Acc. Chem. Rex 1988 21 442. E. C. Ashby Acc. Chem. Res. 1988 21 414. 53 D. L. H. Williams nucleophiles. The main experimental evidence comes from the use of well-known cyclizable radical probes. Kochi7 considers all nucleophiles and electrophiles as electron donors and acceptors respectively. Marcus theory is used to produce generalized free energy relationships for electron transfer and their applicability is demonstrated over a wide range of reaction types; the approach confirms for example that a radical-ion pair is a possible intermediate in electrophilic aromatic nitration.The new Journal of Physical Organic Chemistry was first published in 1988; reference will be made to individual papers later in this Report. One issue of Bulletin de la Socie'te' Chimique de France is given over to the proceedings of a successful conference in Paris (July 1987) entitled 'Organic Reactivity'; again individual references will be made later. 2 Solvolysis and Carbocations Previously measurement of the magnitude of the volume of activation AV* has failed to distinguish clearly between sN1 and SN2 mechanisms because at the relatively low pressures used the AV* term is dominated by the electrostriction of the solvent resulting from the high dipole moment of the activated complex.Con- sequently AV* values are negative for both mechanisms. Now it is argued' that at very high pressures (typically up to 70 kbar) the compressibility of the solvent becomes very small and likewise the electrostriction effect. Therefore at these very high pressures A V' for sN1 reactions should become positive whilst for SN2 reactions it remains negative. Results are presented for the reactions of methyl iodide and t-butyl chloride in glycerol which bear out these predictions experimentally. The value of A V* stays negative for the former up to 70 kbar whilst for the latter A V' becomes positive above 16 kbar. The symmetrically bridged structure for the 2-norbornyl cation (1) was first deduced by Winstein and Trifan over 30 years ago and there has been much controversy over it since.The high exo/ endo rate ratio in the 2-norbornyl sulphonates has now been shown to be electronic in origin and much influenced by remote substituents evidence which supports non-classical stabilization of the 2-norbornyl cation. The possibility of homoallylic participation previously predicted by theoreti- cal calculations has now been demonstrated experimentally lo as a large destabilizing effect in the transition state leading to the cyclopent-3-en-1-yl cation in the solvolysis of the sulphonate. Solvolysis of 2-(2-methoxyethoxy)ethyl tosylate occurs by two 'J. K. Kochi Angew. Chem. Int. Ed. Engl. 1988 27 1227. C. Cameron P. S. P. Saluja M. A. Floriano and E.Whalley J. Phys. Chem. 1988 92 3417. 'D. Lenoir Y. Apeloig D. Arad and P. von R. Schleyer J. Org. Chem. 1988 53 661. T. W. Bentley B. Irrgang H. Mayr and P. von R. Schleyer J. Org. Chem. 1988 53 3492. Reaction Mechanisms -Part (ii) Polar Reactions competing pathways," one involving RO-6 neighbouring group participation the other solvent-assisted displacement. The observation of a large kinetic isotope effect (kH/kD = 1.15) for the solvolysis of cis- and trans-2-aryl cyclopentyl tosylates (with D substitution in the 2 position) in HC02H; AcOH and EtOH solvents has been rationalizedI2 in terms of classical intimate ion-pair formation followed by dissoci- ation to a solvent-separated ion pair without participation by the solvent the 2-aryl group or a hydrogen atom at C2.It is argued that 2-adamantyl substrates would make better models for a Y solvent ionizing power scale than t-butyl substrates because the rigid cage structure in the former prevents any backside solvent nucleophilic attack. Such a YPFss scale has now been produced from kinetic data on the solvolysis of 2-adamantyl penta- fluorobenzenesulphonate and the relative merits of such scales discussed.' The kinetics of solvolysis of RC6H,CH(CSNMe2)02CCF3 show that the thioamide group CSNMe2 is very effective at stabilizing the developing positive charge in the transition ~tate.'~ This can occur by formation of cyclized structures and/or by charge delocalization to the sulphur atom. The experimental evidence in the literature as to the relative stabilities of a-oxy- and a-thiocarbocations is inconclusive.Calculations by ab initio methods have shown that both are stabilizing to a comparable degree the net effect of the oxygen substituent being slightly the greater." The importance of the (often neglected) ground state energies of the neutral precursors is stressed. Carbocation stabilization by a-Si- a-Ge- and a-Sn- containing groups has been demonstrated experimentally by the hydrolysis of the a-metalloidal vinyl ethers (equation l) where protonation is rate limiting.I6 The observed sequence is CMe > SnMe,> GeMe > SiMe > H. H+ CH,=C(OMe)MMe3 -CH,&(OMe)MMe (1) Studies on the hydrolysis of the vinyl ether group in homoprostacyclin (2) show that intramolecular catalysis by the carboxy group is much less than it is in the naturally occurring prostacyclin itself (3) which differs structurally only by one I OH OH S.P. McManus R. M. Karaman B. A. Hovanes M. S. Paley and J. M. Harris J. Org. Chem. 1988 53 681. 12 G. Ronco J. P. Petit R. Guyon and P. Villa Helv. Chim. Actq 1988 71 648. l3 D. C. Hawkinson and D. N. Kevill J. Org. Chem. 1988 53 3857. 14 X. Creary and T. E. Aldridge J. Org. Chem. 1988 53 3888. lS Y. Apeloig and M. Karni J. Chem. SOC.,Perkin Trans. 2 1988 625. 16 J. A. Soderquist and A. Hassner Tetrahedron Lett. 1988 29 1899. D. L. H.Williams methylene group,l7 This suggests that prostacyclin has obtained by natural evolution the correct lifetime (in hydrolysis) to be effective as an inhibitor of blood platelet aggregation in the physiological mechanism for the control of bleeding.Surprisingly the extra rigidity of an aromatic over an aliphatic system does not improve the efficiency of intramolecular catalysis (measured by its EM) in vinyl ether hydroly- sis.'* It is suggested that the expected gain from the rigidity of the aromatic system is offset by the reduced conjugation between vinyl ether group and the aromatic ring in the transition state. Abraham and co-worker~'~ have presented a complete analysis of solvent effects for the solvolysis of t-butyl halides for 20-30 solvents using thermodynamic methods. The effect on AG* AH* and AS* has been separated into contributions due to the initial state and the transition state.The decrease in AG* due to solvent dipolarity and also to solvent hydrogen bond acidity arises from a transition state effect. Solvent hydrogen-bond basicity has little effect and the large effects of the solvent Hildebrand solubility parameter largely cancel each other out in the initial and transition states. Solvolysis of thiophene-2-sulphonyl halides in water and acetone-water mixtures (equation 2) shows many mechanistic features in common with the corresponding X 4 xn SOzHal + HzO S03H + HHal S S reactions of benzoyl halides and involves concurrent nucleophile-sulphur bond making and sulphur-halogen bond breaking.20 3 Other Nucleophilic Substitutions Katritzky and Brycki2' have reviewed the mechanistic borderline area between sN1 and S,2.Kinetic evidence with substrates having neutral leaving groups argues in favour of the view that both mechanisms occur independently and simultaneously in the mechanistic borderline region. The results for the sN1 reactions studied can all be accommodated by a mechanism involving either intimate ion-molecule pair or free carbocation intermediates. The dominant reactions of 2-(alky1thio)ethyl derivatives with nucleophiles involve participation by the neighbouring sulphur atom ( kA).A direct sN2 reaction has now been achieved for the first time by the use of very powerful nucleophiles (2-naphthalenethiolate and p-aminothiophenolate) in DMSO solvent.22 Second-order kinetics were found together with the absence of rearranged products using D-labelled reactants.Y. Chiang and A. J. Kresge J. Chem. Soc. Perkin Trans. 2 1988 1083. 18 A. J. Kresge and Y. Yin J. Phys. Org. Chem. 1988 1 247. 19 M. H. Abraham P. L. Grellier A. Nasehzadeh and R. A. C. Walker J. Chem. Soc. Perkin Trans. 2 1988 1717. 2o A. Arcoria F. P. Ballistreri E. Spina G. A. Tomaselli and E. Maccarone J. Chem. Soc. Perkin Trans. 2 1988 1793. 21 A. R. Katritzky and B. E. Brycki J. Phys. Org. Chem. 1988 1 1. 22 M. R. Sedaghat-Herati S. P. McManus and J. M. Harris 1. Org. Chem. 1988 53 2539. Reaction Mechanisms -Part (ii) Polar Reactions 57 The argument concerning the importance of SET in nucleophilic substitution reactions continues. Ashby has reviewed his position.6 Newcomb et ~1.’~ have argued that the earlier results of Ashby and Pham24 do not in fact support the SET mechanism but rather the traditional sN2 pathway; new and literature rate constants for the reduction of alkyl iodides with lithium aluminium hydride are subjected to a detailed kinetic analysis which it is claimed fit the classical mechanism that takes place in competition with a slower radical chain initiation reaction.Reaction of 9-bromoanthracene with thiolate anions in tetraglyme gives in addi- tion to the 9-(phenylthio) derivative anthracene (equation 3).25The presence of the Br SR latter is rationalized in terms of a SET mechanism which is believed to occur in competition with the expected SNAr mechanism. An experimental evaluation of the VBCM model of Shaik and Pross has been attempted.26 Rate constants (k) for reaction between conjugated nucleophiles and alkyl and benzyl halides were measured by cyclic voltammetry.The SET uersus sN2 character of the transition state was obtained from k/k, ratios where &Ex is the rate constant for reaction between an anion radical A-‘ with the same oxidation potential as the nucleophile and alkyl halide in question. The main conclusion is that the VBCM model has reasonably good predictive power for reactions where the transition state is SET-like. The reaction of acetyl chloride with tetraethylammonium fluoride in acetic acid yields an equilibrium concentration of acetyl fluoride. The reaction has been studied kinetically using I9F n.m.r.” The results are in accord with a sequence of reactions set out in equations 4-6 involving rapid formation of acetic anhydride and hydrogen AcCl +AcOH e Ac20+ HCl (4) HCl+F-HF+CI-(5) Ac20+HF * AcF+AcOH (6) fluoride which then react to give acetyl fluoride.The reactions of acyl chlorides with methanol and phenol have also been studied kinetically in acetonitrile solvent.’* Methanolysis is believed to occur by the SN2 mechanism with a loose transition state whereas phenolysis occurs by an electrophilically assisted ionization mechanism. Kevill and F~jimoto*~ have produced a new nucleophilicity scale for anions based on their reactions with the triethyloxonium ion in ethanol. 23 M. Newcornb J. Kaplan and D. Curran Tetrahedron Lett. 1988 29 3451. 24 E. C. Ashby and T. N. Pharn Tetrahedron Lett.1987 28 3197. 2s S. D. Pastor Helu. Chim. Acta 1988 71 859. 26 T. Lund and H. Lund Acta Chem. Scand. Ser. B,1988 42 269. 27 J. Emsley V. Gold F. Hibbert and W. T. A. Szeto J. Chem. SOC.,Perkin Trans. 2 1988 923. 28 D. N. Kevill and C.-B. Kim J. Chem. Soc. Perkin Trans. 2 1988 1353; Bull. SOC.Chim. Fr. 1988 383. 29 D. N. Kevill and E. K. Fujimoto J. Chem. Res. (S) 1988 408. D. L. H. Williams 4 Elimination Reactions Chuchani and his group continue their investigations of elimination reactions in the gas phase. Kinetic measurements on the pyrolysis elimination of methyl 4-bromocr~tonate~~ reveal that the carbomethoxy group provides anchimeric assist- ance to reaction; this and other results are consistent with a polar mechanism now well established for such reactions in the gas phase.Also in the gas phase elimination reactions of cyclic thioethers have been studied by Fourier transform ion cyclotron re~onance.~~ Three reactions occur (in varying proportions depending on the ring size) E2 a’$-elimination and SN2 substitution. The E2 and a’,P-elimination reactions are stereoelectronically controlled (as they are in solution) and the gas-phase acidity of the sulphides decreases as the sulphide becomes more strained. Rate measurements deuterium exchange experiments and kinetic solvent isotope effects show that the elimination reactions of P-cyano thioethers takes place via a carbanion E 1cB mechanism32 (equation 7). The rate-limiting step changes with the HH H NC H II I NC-C-C-R 5NC-C-C-R ‘C=C/ +RS-(7) A‘ AR BH+ A‘ 4R R‘/ ‘R pK of the thiol derived from the leaving group.The second of the two papers discusses the lifetime of the carbanion intermediate. The elimination of hydrogen chloride from fluorene derivatives is discussed with reference to the E lcB-E2 mechanistic b~rderline.~~ Kinetic isotope effects and Brdnsted p-coefficients are measured and the change from ElcB to E2 rationalized in terms of structural changes in the leaving group and the acidity of the substrate proton. A theoretical paper uses ab initio MO calculations to establish transition state geometries of both E2 and SN2 reactions involving ethyl halides and halide ions.34 The transition states are thought to be different as a result of the repulsion of an attacking base from the C,-C region in the E2 case by the incipient .rr-electron cloud.In a communication entitled ‘Isotope Effects on Isotope Effects’ secondary tritium isotope effects have been measured in the E2 reactions of ArCLTCH2X and ArCL2CH2X(L = H or D) and discussed in terms of a contribution due to tunnel- ling.35 Earlier calculations had predicted that the tunnel correction should be smaller when the transferred atom is deuterium rather than protium. This is now supported by the experimental results. 5 Addition Reactions Ab initio MO calculations of the transition state structures in the addition of molecular halogens to ethene have been pre~ented.~~ The results show that the 30 G. Chuchani and I. Martin Int.J. Chem. Kinet. 1988 20 1. 31 L. J. de Koning and N. M. M. Nibbering J. Am. Chem. Soc. 1988 110 2066. 32 J. C. Fishbein and W. P. Jencks J. Am. Chem. Soc. 1988 110 5075 5087. 33 A. Thibblin J. Am. Chem. Soc. 1988 110,4582. 34 T. Minato and S. Yamabe J. Am. Chem. SOC.,1988 110 4586. ’* M. Amin R. C. Price and W. H. Saunders J. Am. Chem. Soc. 1988 110 4085. 36 S. Yamabe T. Minato and S. Inagaki J. Chem. Soc. Chem. Commun. 1988 532. Reaction Mechanisms -Part (ii) Polar Reactions 59 fluorine reaction proceeds via a four-centre transition state whilst both chlorine and bromine react via zwitterionic three-centre transition states. This accounts for the observed stereoselective anti addition of chlorine and bromine and of syn addition for the fluorine reaction.Last year’s Report contained details of the argument in favour of the reversible formation of the bromonium ion in electrophilic bromination. Now the same group3’ has confirmed this view by showing that bromonium ions react with bromide ion at the bromonium Br+ to give free bromine and the corresponding alkene (see equation 8). Br Br + Br2 I X It has been suggested that N-bromosuccinimide and N-bromoamides react with a,P-unsaturated ketones by removing acid and so causing a change from the acid-catalysed mechanism to one involving bromonium ion intermediate formation.38 The evidence is based on the increase in the product ratio Markovnikov anti- Markovnikov as the concentration of the N-bromo reactant is increased.Halogenation of simple enols has been thought of as a classical diffusion- controlled reaction. This view has now been challenged and results presented39 for bromine reactions in water. Using the more accurate recent values for KE it is possible to obtain the second-order rate constants for reaction with the enol with more reliability than hitherto. Although the values for a range of enols are large (typically 14 x lo9 M-’ s-I) they are not quite at the calculated limit and further- more show some substrate selectivity in the expected sense for a ‘chemically’ controlled reaction. Specifically the rate constants increase as the electron-releasing capacity of substituents in acetophenone enols is increased. The catalytic role of the solvent in electrophilic bromination of alkenes has been examined4’ by consideration of rate data on the bromination of methylideneadaman- tane in various protic solvents and the Y scales of the solvents.The authors conclude that the solvent is involved in the rate-limiting step providing electrophilic assistance as well as a medium effect. The full details have appeared confirming that nitrosation of simple ketones such as acetone takes place via the enol form.41 Either the enolization or the attack of the enol can be rate limiting depending on the enol structure and the experimental conditions. Dimedone which exists primarily in the enol form contains a sufficiently acidic proton as to allow nitrosation even under mildly acidic conditions to occur 37 G. Bellucci R.Bianchini C. Chiappe F. Marioni and R. Spagna J. Am. Chem. Soc. 1988 110 546. 38 V. L. Heasley T. J. Louie D. K. Luttrull M. D. Millar H. B. Moore D. F. Nogales A. M. Sauerbrey A. B. Shevel T. Y. Shibuya M. S. Stanley D. F. Shellhamer and G. E. Heasley J. Org. Chem. 1988 53 2199. 39 R. Hochstrasser A. J. Kresge N. P. Schepp and J. Win J. Am. Chem. SOC.,1988 110 7870. 40 M. F. Ruasse and S. Motallebi Bull. SOC.Chim. Fr. 1988 349. 41 J. R. Leis M. E. Pefia D. L. H. Williams and S. D. Mawson J. Chem. Soc. ferkins Trans. 2 1988 157. D. L. H.Williams (9) XNO\ /XNO NOH via both the enol and enolate forms42 (equation 9) with the latter the more reactive as expected. Nitrosation of aliphatic nitro compounds occurs via the nitronic acid (4) with no contribution from the nitronate anion but kinetic evidence is presented in favour of initial reaction at the oxygen atom followed by rearrangement of the nitroso group to carbon.43 OH +/ R(H)C= N '0-(4) Enol ketonization is currently a much studied reaction and other examples will be discussed in section 8 of this Report.It is worth noting the mechanistic study of the ketonization of a conjugated trienol in a steroidal structure." The kinetics show the expected pH profile for a reaction showing acid- and base-catalysis. Two products are formed (see equation 10). The rate constant for double bond protonation is -200 times less than it is for the corresponding reaction of 1-cyclohexenol presum- ably owing to the extended conjugation in the trienol.J 42 P. Roy and D. L. H. Williams J. Chem. Res. (S) 1988 122. 43 E. Iglesias and D. L. H. Williams J. Chem. SOC.,Perkin Trans. 2 1988 1035. 44 G. D. Dzingeleski S. Bantia G. Blotny and R. M.Pollack J. Org. Chem. 1988 53 1540. Reaction Mechanisms -Part (ii) Polar Reactions 61 Bernasc~ni~~ continues his study of nucleophilic addition to alkenes with kinetic results for the reversible addition of thiolate ion to substituted a-nitrostilbenes (equation 1 l) which are compared with the earlier results with amine nucleophiles. Br@nstedp -coefficients indicate that development of resonance and solvation at the nitro group lags behind carbon-sulphur bond formation in the transition state. COtH / / d eCo2" X X Cyclization of substituted 4-methyl-4-( 2-hydroxyphenyl)pent-2-enoic acid deriva- tives in water (equation 12) results from intramolecular nucleophilic attack on the activated double bond by the phenolate oxygen.46 Activation by C02-and C02H over H are 4000 and lo8 respectively.The stereochemistry of the general acid- catalysed (by protonated amines) reactions contrasts with that of the water-catalysed reaction. The proposed mechanism involves rate-limiting protonation of the mono- or di-anion of the carboxylic acid enolate structures. 6 Aromatic Substitution and Rearrangements A quantitative study is reported of the competitive nitration of benzene and toluene in the gas phase by a range of protonated alkyl nitrates using mass spectrometric and radiolytic technique^.^' The results are consistent with an electrophilic process and the final proton loss is not rate-limiting.The same positional and substrate selectivity is found as for the solution-phase reactions and it is also possible to arrange the conditions such that reactions occur at the encounter rate when substrate selectivity is lost. This is another example in a growing list where well-known mechanistic features of reactions in solution are reproduced in the gas phase. The nitration of aryl iodides in acetic anhydride by nitric acid results in the formation of aryliodine( 111) derivatives rapidly and reversibly by oxidation.48 This accounts for the slowness of the nitration of aryl iodides in this medium whereas in other solvents the expected reactivity pattern is seen.A new kinetic form for nitrous acid-catalysed nitration of naphthalene which is second order in [naphthalene] has 45 C. F. Bemasconi and R. B. Killian J. Am. Chem. Soc. 1988 110 7506. 46 T. L. Amyes and A. J. Kirby J. Am. Chem. Soc. 1988 110 6505. 4' M. Attina and F. Cacae Gazz. Chim. Ztal. 1988 118 241. 48 G. W. Bushnell A. Fischer and P. N. Ibrahim J. Chem. SOC.,Perkin Trans. 2 1988 1281. D. L. H. Williams been observed to occur alongside a first-order term.49 The second-order term has been interpreted in terms of the formation and subsequent reaction of a radical cation containing two naphthalene structures (equations 13-1 5). ArH+ArH+NO+ G (ArH)l’+NO (13) NO+NOl & NO++NO (14) (ArH) +NO + ArNO +ArH +H+ (15) The identification of cyclohexadienone intermediates in electrophilic reactions of phenol continues to attract attention.They have now been identified in brominations using N-bromosuccinimide in CH2C12 and other aprotic solvents.50 Metal ion catalysis has been found in the decomposition of 2-carboxy-2,5-cyclohexadiones in water.’l The origin of the catalysis by Cu2+ and Fe3+ is strong metal ion binding to the incipient di-anion. In aromatic nucleophilic substitution Bunton and co-workers in two papers,52 have proposed radically that the familiar classical S,Ar mechanism involving rate limiting a-complex formation be replaced by a mechanism which includes single electron transfer. This is prompted by the experimental observation of two intermedi- ates which are believed to be (a) a .rr-complex of substrate and OH-and (b) a charge transfer complex of an arene radical ion and OH’ which then forms the u (Meisenheimer) complex.A multistep reaction treatment based on relaxation theory has been developed and the resulting calculated rate and equilibrium constants agree well with the experimental values. The Reporter suspects that not all workers in the field will be convinced by this evidence. In the spiro Meisenheimer complex formation from 3,6-dimethylcatechol 2,4,6-trinitrophenyl ether,53 the kinetics are consistent with the operation of two concurrent mechanisms (a) a trapping mechan- ism with diff usion-controlled proton transfer between a phenolic OH and carboxylate ions and cyclization or ring-opening with no catalysis and (b) a pre-association mechanism in which cyclization/ring-opening involves buffer catalysis by hydrogen bonding.A facile amine-amine exchange has been noted in aromatic nucleophilic substitution reactions of 1-dialkylamin0-2,4-dinitronaphthalene.~~ The dial kylamine group is readily replaced by primary amines and also by pyrrolidine at 30 “C simply by mixing the reactants. Shine’s group have extended their heavy-atom kinetic isotope methods so success-ful in analysis of the benzidine rearrangement to the related quinamine rearrange- ments and also to the Claisen rearrangement. In the former55 (see for example equation 16) the major pathway for the acid-catalysed reaction is a concurrent concerted process ie.is a [5,5J-sigmatropic rearrangement. There is a minor pathway leading to a number of products all derived from a [3,3]-sigmatropic rearrangement. The analogy with the benzidine rearrangement is discussed. 49 J. R. Leis M. E. Pefia and J. H. Ridd J. Chem. SOC.,Chem. Commun. 1988 670. Y. L. Chow D.-C. Zhao and C. I. Johansson Can. J. Chem. 1988,66 2556. 51 0. S. Tee and N. R. Iyengar Can. J. Chem. 1988,66 1194. 52 R. Bacaloglu C. A. Bunton and F. Ortega 1. Am. Chem. SOC.,1988 110 3503 3512. 53 C. F. Bernasconi and D. E. Fairchild J. Am. Chem. Soc. 1988 110 5498. 54 S.Sekiguchi T. Hone and T. Suzuki J. Chem. SOC.,Chem. Commun. 1988 698. 55 B. Boduszek and H. J. Shine J. Am. Chem. SOC.,1988 110 3247. Reaction Mechanisms -Part (ii) Polar Reactions Me o Br P N I -O AceticHCIacid -* H2N 0 (16) Br The Claisen rearrangement of ally1 phenyl ether has been studied by labelling at oxygen a-carbon y-carbon and ~rtho-carbon.~~ The results reveal that in the transition state the C,-0 bond is 50-60% broken whereas the Cy-Corthobond is only 10-20% complete.The 1,3-nitro group rearrangements of nitrocyclohexadienones shown in equations 17 and 18 have been examined for 15Nnuclear polarization using "N-labelled nitro compounds in acetic anh~dride.~' Strong "N nuclear polarization was observed in the first case (equation 17) indicating that the reaction involves a radical pair intermediate but polarization was absent in the second case (equation 18) so the mechanism here (not yet elucidated) must be quite different.7 Proton Transfer The generally accepted view is that carboxylic acids and phenols are stronger acids than alcohols because of a greater degree of stabilization of their anions. Exner5* arrives at this conclusion also using simple thermodynamic ideas and so contradicts a recent alternative view by Siggel and Thomas where the difference was attributed to an unusually high energy content of the undissociated acid molecule. Now further calculation^^^ predict that only 616% of the acidity difference can be attributed to the contribution of resonance to the stabilization of the anions and claim that about 80% of the acidity difference can be accounted for by carbonyl inductive effects. The proposal last year of Arnett and Harrelson:' attributing the high acidity of Meldrum's acid to the unfavourable anti conformations that the ester groups are forced to adopt has prompted two theoretical studies in the quest for the origin of the effect.61962 Both attribute the effect to electrostatic (dipole-dipole) repulsions 56 L.Kupczyk-Subotkowska W. H. Saunders and H. J. Shine J. Am. Chem. SOC.,1988 110 7153. 57 J. H. Ridd J. P. B. Sandall and S. Trevellick J. Chem. SOC.,Chem. Commun. 1988 1195. 58 0.Exner J. Org. Chem. 1988 53 1810. 59 M. R. F. Siggel A. Streitweiser and T. D. Thomas J. Am. Chem. SOC.,1988 110 8022. 60 E. M. Arnett and J. A. Harrelson J. Am. Chem. SOC.,1987 109 809. 61 X. Wang and K. N. Houk J. Am. Chem. SOC.,1988 110 1870.62 K. B. Wiberg and K. E. Laidig J. Am. Chem. SOC.,1988 110 1872. 64 D. L. H. Williams which result in both ester groups in the dilactone taking up the anti conformation thereby decreasing the deprotonation energy sufficiently to account for the acidity. Both show that the loss of a proton from the E-rotamer is easier (by 4.7 kcal in one study and 5.4 kcal in the other) than it is for the 2-rotamer. Acidity measurements in DMSO illustrate the large acid-strengthening effects of polyfluoroaryl sub~tituents.~~ For example C6F5CH2CN is more acidic than is C6H5CH2CN by 5.9 pK units resulting from the powerful field/inductive effects of the five fluorine atoms. The Bordwell group has also measured the acidities of aniline and 26 derivatives in DMSO using an overlapping indicator method.64 A large range is covered from 15.9 for 2,4-dinitroaniline to 30.7 for aniline itself.The meta substituents correlate well with the Hammett u constants; the deviation for a number of para substituents is believed to be due to enhanced solvation of the substituents resulting from a direct conjugation with the anilide ion. Albery Bernasconi and Kre~ge~~ have discussed the nitroalkane anomaly (for example Bronsted a values >1 for the ionization of ring-substituted phenyl nitromethanes) in terms of Marcus-Grunwald theory. The incorrect prediction of the theory can be remedied if solvent reorganization is added to electronic rearrange- ment as the second reaction progress variable. In a multi-author paper66 the quantitative effect of annelation on the acidity and basicity of azoles is discussed both for gas-phase and aqueous solution proton transfers.Annelation increases the acidity by more than it does the basicity possibly because the azoles are electron-rich systems which can more readily accommodate positive rather than negative charge. The sterically hindered 2,6-di-t-butylpyridine is an unusually weak base in DMSO (pK 0.81) whereas its gas-phase basicity is much greater.67 It is argued that this arises from the much reduced hydrogen bonding between the sterically hindered protonated form and a relatively large DMSO solvent molecule. Generally it is agreed that the oxygen atom is the more basic site in amides. However it is now suggested68 that in hindered amides the nitrogen atom becomes protonated first.The evidence comes from the position of charge transfer bands and from comparison of the formation constants for I,-amine and I,-amide complexes. The explanation offered is that steric hindrance reduces conjugation of the CO and NR2 groups in N,N-disubstituted amides and so reduces the basicity of the oxygen atom sufficiently to allow protonation at nitrogen. The very high basicity of aromatic diamine proton sponges is discussed in a review article.69 The abnormal basicity is discussed in terms of the extreme steric strain in these systems and of the destabilizing effect of the overlap of the nitrogen lone pair of the neutral diamine and the strong N--H--N hydrogen bonds formed on monoprotonation which leads to much reduced steric strain.63 F. G. Bordwell J. C. Branca J. E. Bares and R. Filler J. Org. Chem. 1988 53 780. 64 F. G. Bordwell and D. J. Algrirn J. Am. Chem. Soc. 1988 110 2964. 65 W. Albery C. F. Bernasconi and A. J. Kresge J. fhys. Org. Chem. 1988 I 29. 66 J. Catalan R. M. Clararnunt J. Elguero J. Laynez M. Menendez F. Anvia J. H. Quian M. Taagepera and R. W. Taft J. Am. Chem. Soc. 1988 110 4105. 67 R. L. Benoit M. Frkchette and D. Lefebvre Can. J. Chem. 1988 66 1159. 68 G. Guiheneuf J.-L. M. Abboud and A. Lachkar Can. J. Chem. 1988 66 1032. ’’ H. A. Staab and T. Saupe Angew. Chem. In?. Ed. Engl. 1988 27 865. 65 Reaction Mechanisms -Part (ii) Polar Reactions Finally in this section it is worth noting that the basicities of a number of common dipolar non-hydroxylic solvents have been determined by an n.m.r.technique which measures the protonation equilibrium in aqueous sulphuric acid.70 A relative basicity scale -amides > DMSO > acetone >> acetonitrile > sulphones > nitromethane -has been established. 8 Carbonyl Derivatives A theoretical treatment of nucleophilic reactivity in addition reactions to the carbonyl group has been given.71 The value of AG* for XF attack in esters correlates with the vertical ionization potentials of XF. This is interpreted to mean that an important aspect of the activation process is the single electron transfer from XT to the substrate. The structures and some properties of the adducts between HO-and MeO- and a range of carbonyl compounds have been studied by FT-ICR mass spe~trometry.~~ The main conclusion is that tetrahedral intermediates are involved (confirming an earlier theoretical prediction); here is another example of similarities between gas-phase and solution reactions.Nucleophilic addition of aniline to methyl formate has been studied mechanisti- ally.^^ The kinetic results are explained in terms of the trapping mechanism given in the reaction scheme 19. Steps 1 2 or 3 can be made rate-limiting by varying the strength of the base. General base catalysis occurs together with diff usion-controlled proton transfer. Terrier and co-~orkers~~ have examined the reaction of a-keto-aldoximate ions with p-nitrophenyl acetate (equation 20) particularly with reference to the effect of solvation on the reactivity of a-nucleophiles in aqueous solution./\ +I I H-N + C=OeH-N-C-O-\/ II U. B (step 1) \BH++ / IN-C-0-I fast \ I / I Y N-C-OH+B 02N-(3- OCOCH3 + RCH2COCH=NO- (20) __+ 04CH3C + O2N 0 0 - ON=CHCOCH2R At low pK values (cf:phenoxide anions) the Brsnsted plot is linear but at pK > 7.8 there is a very strong downward curvature. This is explained in terms of the need for desolvation of the oximate anion before nucleophilic attack and the fact that desolvation is more difficult with the more basic oximate anions. 70 A. Bango and G. Scorrano J. Am. Chern. Soc. 1988 110,4577. 7' E. Buncel S. S. Shaik I.-H. Urn and S. Wolfe J. Am. Chem. SOC.,1988 110 1275. 72 H. van der We1 and N.M. M. Nibbering Red. Trau. Chim. Pays-Bas 1988 107 491. 73 C. C. Yang and W. P. Jencks J. Am. Chem. SOC.,1988 110 2972. 74 F. Terrier F. Degorre D. Kiffer and M. Laloi Bull. SOC.Chim. Fr. 1988 415. D. L. H.Williams The measurement of the acidity of the keto-enol system continues to attract a lot of attention. Estimates of the pK values for the OH acidity in protonated simple carbonyl compounds (equation 21) have been made7s from KE values and CH acidity constants obtained by the application of the Marcus equation to the ketoniz- ation process. Two gro~ps'~,~~ have independently measured the acidity of 2- indanone (equation 22) from measurements of the rate of enol formation. The acidity (pK 12.2) is much higher than expected and is attributed to the extra stabilization of the enolate by interaction of the Ph and 0-groups via the enolate double bond.It is estimated that the intrinsic electronic effect of a Ph group upon ketone acidity is ca. 106-107. The Marcus equation has been applied to the deprotonation of some unusually acidic benzylic ketoces [e.g. (5) and (6)],78which constitute examples of the principle of non-perfect synchronization presented by Bernasconi in recent years. Various aspects of ester hydrolysis have been examined. The ElcB mechanism is now established for the hydrolysis of esters containing acidic protons at the LY -position (e.g.acetoacetates malonates etc.).Additional evidence for this pathway comes from the sign of the volume of activation (AV") for such reactions,79 which is positive whereas A V' is negative for the more common B,,2 mechanism.Alkynyl benzoates tosylates and phosphates have recently been prepared and their hydroly- sis mechanisms investigated.80 A number of interesting pathways have been estab- lished including those involving both electrophilic and nucleophilic attack of the triple bond. Hydrolysis of esters of phosphorous acid have also been studied.*l Trialkyl phosphites hydrolyse in alkali hundreds of times faster than do the corre- sponding phosphates whilst in acid solution the phosphites are ca. lo'* times faster than the phosphates; the enormous effect in the acid-catalysed reactions is thought 75 J. Toullec Tetrahedron Lett. 1988 29 5541. 76 J. R. Keeffe A. J.Kresge and Y. Yin J. Am. Chem. Soc. 1988 110 1982. 77 A. M. Ross D. L. Whalen S. Eldin and R. M. Pollack J. Am. Chem. Soc. 1988 110 1980. 78 J. W. Bunting and D. Stefanidis J. Am. Chem. Soc. 1988 110 4008. 79 N. S. Isaacs and T. S. Nagem J. Chem. Soc. Perkin Trans. 2 1988 557. 8o A. D. Allen T. Kitamura K. A. Roberts P. J. Stang and T. T. Tidwell J. Am. Chem. Soc. 1988 110 622. " F. H. Westheimer S. Huang and F. Covitz J. Am. Chem. Soc. 1988 110 181. Reaction Mechanisms -Part (ii) Polar Reactions to stem from the existence of an unshared pair of electrons on the phosphorous atom in the phosphite. Intramolecular participation by both the amide and hydroxy groups in ester hydrolysis reactions of naphthalene structures has been as has the effect of metal ions (Cu2+ Ni2+ Co2+,and Zn2+) on the intramolecular participation by the acetamido In the latter case a general mechanism for metal ion catalysis in reactions where C-0 bond breaking is rate limiting has been proposed which contains a transition state where the metal ion is simultaneously bound to two oxygen atoms and a nitrogen atom of a heterocyclic system.In the same area the effect of neighbouring pyridinium groups on the base-catalysed hydrolysis of arylbenzoates has been examined.8s The ortho substituents are more effective than the para and the catalytic effect is based on interaction of a negatively charged site in the transition state with an electron deficient 7r-system of the pyridinium ring. Novel methods for the generation of simple enols continue to be found.One such procedure involves the photo-oxidation of alcohols by carbonyl compoundsB6 (equation 23). This procedure led to KE values in good agreement with other recent determinations. A simple stable enol has also been produced by isomerization (equation 24) using a rhodium carbonyl triphenylphosphine complex.87 Again ketonization rate constants are in agreement with literature values. Both (Z)-and (E)-1-hydroxybutadienes have been generated in solution for the first time from their trimethylsilyl derivatives in slightly acidic aqueous acetonitriIe.88 Both were characterized by n.m.r. and the kinetics of the ketonization reported. Further studies with 2-indanone are reported leading to a pKE value of 3.84 and a pK value of 8.36 for the enol acting as an oxygen acid.89 Enol addition equilibria and keto-enol tautomerization of 9-formylfluorene have also been examined.” The enol of aceto-phenone has been generated by photohydration of phenylacetylene and photoelimi- nation and rates of ketonization obtained;” the results have been analysed by Marcus theory and comparisons made with an earlier treatment with the enol of isobutyrophenone.The reactivity (in enolization) of amino ketones is much greater than that of simple unsubstituted ketones. Kinetic analysis9* reveals that two factors 82 F. Hibbert and R. J. Sellens J. Chem. Soc. Perkin Trans. 2 1988 399. 83 F. Hibbert and K. J. Spiers J. Chem. Soc. Perkin Trans. 2 1988 571. 84 T. H. Fife T. J. Przystas and M.P. Pujari J. Am. Chem. Soc. 1988 110 8157. 85 J. F. J. Engbersen G. Geurtsen D. A. DeBie and H. C. Van der Plas Tetrahedron 1988,44 1795. 86 J. R. Keeffe A. J. Kresge and N. P. Schepp J. Am. Chem. Soc. 1988 110 1993. 87 C. S. Chin S. Y. Lee J. Park and S. Kim J. Am. Chem. Soc. 1988 110 8244. 88 B. Capon and B. Guo J. Am. Chem. Soc. 1988 110 5144. 89 J. R. Keeffe A. J. Kresge and Y. Yin J. Am. Chem. Soc. 1988 110 8201. 90 M. P. Harcourt and R. A. More O’Ferrall Bull. Soc. Chim. Fr. 1988 407. 91 Y. Chiang A. J. Kresge J. A. Santaballa and J. Win J. Am. Chem. Soc. 1988 110 5506. 92 B. G. Cox P. DeMaria and L. Guerzoni J. Chem. Soc. Perkin Trans. 2 1988 163. 68 D. L. H. Williams are responsible (a) the influence of the positive charge in the N-protonated (or methylated) derivatives and (b) intramolecular general base catalysis by the neutral nitrogen atom which is at a maximum for the &substrates when a six-membered ring transition state is possible.Enolization rate constants for malonic acid and methylmalonic acid have been obtained by an isotopic exchange reaction followed by ‘Hn.m.r.93 Finally here the preparation ionization energies and heats of formation of unstable enols in the gas phase have been described.94 9 Other Reactions The kinetics of the N-chlorination of secondary amines by N-chlorosuccinimide show that the process is reversible and second order in both direction^.'^ The results are consistent with a process which involves direct transfer of chlorine to the amine and do not support a mechanism involving a prior hydrolysis of N-chlorosuccinimide.The Ar+--N2 pair which is the first intermediate form in the dediazoniation reactions of arenediazonium ions can be trapped with carbon monoxide in water to give arenecarboxylic acids.96 This is a model for the reverse of dediazoniation since CO is isoelectronic with N2. The yields from substituted diazonium ions correlate with the Taft dual substituent parameters to give positive pF and negative pR reaction constants as expected for addition of N to aryl cations. A book has been published entitled ‘Nitr~sation’~’ which describes the various reagents that can bring about nitrosation with discussion (mainly from a mechanistic viewpoint) of nitrosation at C N 0 S and metal sites.The kinetic identification of NO+ as the effective reagent in dilute acid solution has been achieved for the nitrosation of alcohols and a thiol using cither alkyl nitrites or nitrous acid as the reagent.98 Reactions in acetonitrile solvent are truly zero order in the nitrosatable substrates and so must involve rate-limiting formation of some species such as NO+. The analogy with the well-known situation in aromatic nitration is obvious. The solvent kinetic isotope effect on the equilibrium formation of NO+/H2N02+ in water (K,/K,) has been determined as 2.55 * 0.28 in the pH range 1-5.99 Two pathways have been identified in the nitrosation of phenylureas,loO one leading to the N-nitroso derivative and the other to diazonium ion formation.The reaction between nitrous acid and N,N’-dialkylthioureas occurs by S-nitrosation followed by a slow re- arrangement to give the N-nitroso product.”’ The possibility of a direct N-nitrosa- tion is ruled out. In a re-examination of the nitrosation (diazotization) of aniline at high acidity it is proposed102 that deprotonation of the di-cation complex occurs before and not after the NO+ group rearrangement to nitrogen (equations 25 and 93 E. W. Hansen and P. Ruolf J. Phys. Chem. 1988,92 2641. 94 F. Turecek L. Brabec and J. Korvola J. Am. Chem. SOC.,1988 110 7984. 95 J. M. Antelo F. Arce J. Franco M. C. G. Lopez M. Sanchez and A. Varela Znt. J. Chem. Kinef. 1988 20 397. 96 M. D. Ravenscroft P. Skrabal B. Weiss and H. Zolinger Hefv.Chim. Acta 1988 71 515. 97 D. L. H. Williams ‘Nitrosation’ Cambridge University Press 1988. 98 M. J. Crookes and D. L. H. Williams J. Chem. SOC.,Chem. Commun. 1988 571. 99 A. Castro M. Mosquera M. F. R.Prieto J. A. Santaballa and J. V. Tato J. Chem. SOC.,Perkin Trans. 2 1988 1963. 100 A. Castro M. Gonzalez F. Meijide and M. Mosquera J. Chem. SOC.,Perkin Trans. 2 1988 2021. 101 F. Meijide and G. Stedman J. Chem. SOC.,Perkin Trans. 2 1988 1087. 102 H. Zollinger Helv. Chim. Acta 1988 71 1661. Reaction Mechanisms -Part (ii) Polar Reactions NO+ NO+ I I I NO+ 26). The carbanion of ethyl cyanoacetate reacts with the nitroprusside ion in two stages (equations 27 and 28) to give the oxime pr~duct."~ Intramolecular nucleophilic catalysis by neighbouring OH has been demonstrated in the acid- and base-catalysed hydrolysis of aromatic sulphonamide~'~~ (equation 29).There is a large Thorpe-Ingold gem-dialkyl effect where relief of initial-state steric strain is a major effect. .OH &02NMe2 -OSO2+Me2NH Primary and tertiary nitronate ions react with sulphonyl halides to give various products which depend on the nature of the halide the nitronate structure and the Some of these reactions take place by single electron transfer mechanisms as the addition of p-dinitrobenzene (a single electron acceptor) results in the inhibition of some products. 10 Some Probes of Polar Mechanisms Jencks'06 urges caution in the interpretation of structure-reactivity coefficients since solvation or desolvation of a reacting group often plays a part and the effects of substituents in different parts of a molecule often lead to imbalances between estimates of reaction progress in the transition state.Additional information regard- ing the nature of the transition state (which can be described by theoretical reaction 103 A. R. Butler A. M. Calsy and C. Glidewell J. Chem. SOC.,Perkin Trans. 2 1988 1179. I04 A. Wagenaar and J. B. F. N. Engberts J. Org. Chem. 1988,53 768. lo' P. E. Pigou and C. J. M. Stirling J. Chem. SOC.,Perkin Trans. 2 1988 725. 106 W. P. Jencks Bull. SOC.Chim. Fr. 1988 218. 70 D. L.H. Williams surfaces or by empirical energy-contour diagrams) can however be obtained from changes in structure-reactivity coefficients i.e.from the second derivatives of log k The limitation of the use of Hammett p values in assessing relative bond tightness in transition states has previously been stressed. Now Lee and co-~orkers''~ argue that the cross-interaction constant pxy (given by equation 30) can be used more reliably to characterize the transition state structure. Experimental results for some nucleophilic substitution reactions are analysed in these terms. log(kxYlkH*) = PXUX+ PYUY + PXYUXUY (30) Fadhil and GodfreyIo8 examine the theory behind quantitative linear substituent correlations and conclude that the pattern of deviations is remarkably simple if the choice of standard substituent constants is based on I3Csubstituent chemical shifts at the /3 site in ring-substituted styrenes.This can also be put in the more familiar language of field and resonance substituent effects if it is realized that the resonance effect modifies the field effect by an amount that can vary up to a maximum. This approach indicates that there are serious flaws in commonly used dual substituent parameter equations. The same authors extend these ideas to cover Brflnsted and related plot^.''^ Taft and (many) co-workers"' have examined the effects of molecular structure on gas-phase proton transfer equilibria. The summation of simple product terms for substituent inductive polarizability and resonance effects fit well with the observed substituent effects over a wide range of structures. The reactivity-selectivity principle continues to attract attention.Johnson and Stratton"' argue that such a link between selectivity and reactivity is not necessary since variations in the slope of linear free energy relationships can account for the range of observed selectivity ratios. A review discusses the physical basis of intramolecularity. A simple extra-thermodynamic treatment of enthalpy and entropy changes for cyclization of short-chain compounds both for gas-phase and solution-phase reactions offers a ready interpretation of the magnitude and variation of the observed 'effective molarities'. Unusually large kinetic isotope effects often coupled with anomalous temperature dependences of these effects are usually thought to result from quantum tunnelling. Thibblin"3 has shown that these effects can also arise from reaction branching i.e.from partitioning of an intermediate. This may be the explanation in some cases. 107 I. Lee C. S. Shim S. Y. Chung H. Y. Kim,and H. W. Lee J. Chem. Soc. Perkin Trans. 2 1988 1919. 108 G. F. Fadhil and M. Godfrey J. Chem. Soc. Perkin Trans. 2 1988 133. '09 G. F. Fadhil and M. Godfrey J. Chern. Soc. Perkin Trans. 2 1988 139. I10 R. W. Taft J. L. M. Abboud F. Anvia M. Bertheiot M. Fujio J.-F. Gal A. D. Headley W. G. Henderson I. Koppel J. H. Qian M. Mishima M. Taagapera and S. Ueji J. Am. Chem. Soc. 1988 110 1797. 'I1 C. D. Johnson and B. Stratton J. Chem. Soc. Perkin Trans. 2 1988 1903. L. Mandolini Bull. Soc. Chim. Fr. 1988 173. A. Thibblin J. Phys. Org. Chem. 1988 1 161.