3 Reaction mechanisms Part (i) Pericyclic reactions By IAN D. CUNNINGHAM Department of Chemistry University of Surrey Guildford UK GU2 5XH 1 Introduction Pericyclic reactions can be categorised as cycloadditions electrocyclisations or sigmatropic rearrangements and this format is adopted here. Several reviews have appeared in 1996 covering density functional theory (DFT) calculations of transition states (TS),1 transition metal-mediated cycloadditions,2 Diels–Alder reactions in nature,3 and metal-catalysed asymmetric 1,3-dipolar cycloadditions. 4 2 Cycloadditions Theoretical Density functional theory (Becke3LYP/6-31G*) has been used to predict energies and isotope e§ects for concerted and stepwise mechanisms of the Diels–Alder reaction; predicted barrier heights and heats of reaction are in good agreement with experimental values with the stepwise barrier predicted to be the higher by 9.6–32.2 kJ mol~1 (2.3–7.7 kcal mol~1).5 A comparison of DFT (Becke3LYP) or RHF calculated kinetic isotope e§ects with experimental values indicates a moderately asynchronous transition state for the Diels–Alder reaction of isoprene and maleic anhydride.6 Neither ab initio (HF/6-31G//AM1) nor frontier molecular orbital (FMO) analysis of the s-trans to s-cis interconversion accounts for the low reactivity of (Z)-1 towards cycloaddition of alkenes compared to (E)-1; this suggests that transition state steric restraint is the predominant factor.7 A comparison of ab initio computational methods (HF MP2 and Becke3LYP) has been made for the Diels–Alder reaction between butadiene and various thiocarbonyl compounds.8 Novel dienes and dienophiles The electron-rich 5-aminopyrazoles 2 have been shown to act as dienophiles in an inverse electron demand Diels–Alder reaction with the 1,3,5-triazine 3 (R\H alkyl; R@\H alkyl thienyl phenyl) (Scheme 1); subsequent retro-Diels–Alder and aromatisation reactions yield pyrazolo[3,4-d]pyrimidine products.9 Royal Society of Chemistry–Annual Reports–Book B 27 Scheme 1 N Et Me N F3C CF3 N N Et Me N N F3C CF3 5 4 Scheme 2 O B O O B O ( E)-1 ( Z)-1 1-(Arylthio)butadienes bearing 4-alkyl 4-alkoxy or 4-amino substituents give endo cycloadducts preferentially with maleic anhydride or maleimide dienophiles; with ethyl acrylate the regioselectivity is controlled by the O or N (where present) rather than by the S.10 Azirines (e.g.4) undergo [4]2] cycloaddition to the 4a,8a-methanophthalazine 5 with high selectivity for the electron-deficient diazadiene portion and for anti approach from the side opposite to the methano bridge (Scheme 2).11 The e§ect of the additional nitrogen on going from a 1,2,4-triazine to a 1,2,4,5- tetrazine is su¶cient to allow a successful inverse electron demand Diels–Alder reaction with electron-rich dienophiles.12 2-Ethynylbuta-1,3-diene was found to undergo cycloaddition to a range of dienophiles with regioselectivity as expected based on FMO considerations; this diene was estimated to be ca.5 times less reactive than isoprene and to undergo competing dimerisation via a diradical mechanism.13 Four mono-adducts of benzyne to [70]fullerene have been isolated; they include a [4]2] adduct across the 7,23-bond in line with earlier theoretical predictions.The lower electrophilicity of [70]fullerene compared to [60]fullerene is proposed to account for its [4]2] cycloaddition.14 Kinetics and mechanism [4]2] cycloadditions Intramolecular [4]2] cycloaddition of a cyclobutadiene to a range of unactivated (e.g. mono- di- and tri-alkyl substituted) alkenes has been reported. Although the 28 Ian D. Cunningham X CO2 Me O O X CO2Me 6 Scheme 3 cyclobutadiene portion was generated in situ by oxidation (e.g. cerium ammonium nitrate) of an iron tricarbonyl-complexed precursor the possibility of a radical cation mechanism for this cycloaddition was not discussed.15 N-Aryl-2-cyano-1-azadienes undergo ready cycloaddition to a range of electron-rich (e.g. methyl vinyl ether) and electron-deficient (e.g.methyl vinyl ketone) dienophiles with reactivities di§ering typically by a factor of only ca. 1.4. The observed regioselectivities and (to some extent the reactivities) have been rationalised usingFMOtheory withLUMO$*%/% control for reaction with methyl vinyl ether and HOMO$*%/% control for methyl vinyl ketone.16 A study of retro-Diels–Alder femtosecond reaction dynamics has been published.17 The cyclopropylideneacetates 6 (X\Br or Cl) were found to be more reactive than methyl acrylate towards furan (k3%- \16) and 6,6-dimethylfulvene (k3%- \210–230) while the acyclic analogue methyl 1-chloro-3,3-dimethylacrylate does not react (Scheme 3). The low reactivity of the non-halogenated analogue 6 (X\H) and the fluoro compound 6 (X\F) despite the latter’s low energy LUMO was interpreted as mitigating in favour of a diradical or zwittterion mechanism.18 A study of pressure e§ects on the kinetics and diastereoselectivity of the hetero- Diels–Alder reaction of b-aminoenones with highly substituted vinyl ethers has yielded separate values of activation volume *V8 for endo and exo products.For reaction of 4-phthalimido-1,1,1-trifluorobut-3-en-2-one with 2-methoxypropene the values of [37.6cm3 mol~1 for the endo mode and [41.7 cm3 mol~1 for the exo were interpreted as reflecting the relative steric demand of the methoxy vs. the methyl towards the diene trifluoromethyl group in the TS.19 Reaction of the norborna-2,5-diene 7 with furans 8 (R R@\H alkyl) yielded the endo–exo product 9 (Scheme 4) and its endo–endo isomer. The yield of the thermodynamic product the endo–exo increased as alkyl substituents were removed from C2 and C5 of the furan; this was attributed to equilibration between adducts via cycloreversion with a greater di§erence between thermodynamic and kinetic stability for the less substituted adducts.20 A diradical mechanism is proposed for the dimerisation of 2-chlorobuta-1,3-diene (chloroprene) to yield [4]2] and [2]2] dimers.This is based on the observed lack of stereospecificity in the products and the similarity of E!#5 values for [4]2] and [2]2] (the latter assumed to be via diradicals) modes. The reduced stereospecificity compared to butadiene is attributed to the ‘heavy atom e§ect’ converting the initiallyformed singlet diradicals which cyclise rapidly before rotation to triplet diradicals which do not.Low values (small negative) of the ‘volume of concert’ ([22 to [31 cm3 mol~1) are explained by the participation of ‘extended’ diradicals in product formation.21 Mechanisms involving electron transfer The balance between concerted and stepwise mechanisms continues to be of interest. A 29 Reaction mechanisms Part (i) Pericyclic reactions Cl Cl Cl Cl Cl Cl O R' R O R R' Cl Cl Cl Cl Cl Cl 7 8 9 Scheme 4 kinetic study of the reactivity of styrene-derived radical cations (generated by laser irradiation) towards a range of alkene (p2) and diene (p4) substrates has appeared. The reactions lead to cycloaddition products although whether via concerted or stepwise mechanisms is not conclusively stated.22 However the aminium salt or photoinduced electron transfer (PET) initiated cycloaddition of a series of electron-rich allenes to 1,2,3,4,5-pentamethylcyclopentadienewas proposed to proceed via a stepwise mechanism involving a diene radical cation and neutral allene to yield a distonic radical cation product which subsequently cyclises.23 Extending earlier work Hammett–Brown plots for the radical cation cycloaddition of 2,3-dimethylbutadiene to a range of di-substituted stilbenes showed curvature reflecting a change from equilibrium to rate-limiting ionisation of the stilbene with electron-donating aryl substituents (e.g.Me). These results were interpreted in terms of a concerted mechanism with only a small transfer of positive charge to the diene in the TS.24 Solvent and Lewis acid e§ects Acceleration of the Diels–Alder reaction of the cationic acridizinium bromide 10a (as diene) with cyclopentadiene (as dienophile) in aqueous compared to non-aqueous solvent has been shown to be ca.five-fold due to a hydrophobic e§ect; inclusion of a weak hydrogen-bond acceptor (10b) results in a further ca. two-fold acceleration.25 N Ar O M2+ N R Br – 10a R = H 10b R = CO2Et 11 + An LFER analysis of the inverse electron demand hetero-Diels–Alder reaction between 3,6-di(2-pyridyl)-1,2,4,5-tetrazine and substituted styrenes suggests a more dipolar TS in protic solvent (q\[1.32 in H 2 O–tert-butyl alcohol) than in aprotic solvent (o\[0.51 in toluene).26 Cycloaddition of the Lewis acid-complexed bidentate dienophile 11 (M2`\Co2` Ni2` Cu2` and Zn2`) to cyclopentadiene is accelerated up to 8]104 fold compared to the uncatalysed reaction with only a small further rate increase due to water solvent; endo selectivity is also dominated by the enhancing e§ect of the Lewis acid.27 A series of boronated alumina (BXn–Al 2 O 3 n\1–3) catalysts were found to enhance stereo- and regio-selectivity in the Diels–Alder reactions between methyl acrylate and the dienes cyclopentadiene and isoprene possibly due to steric interactions involving the catalyst surface.28 30 Ian D.Cunningham O C C Ph Ph O Ph Ph O Ph Ph 12 [3,3] [4+2] Scheme 5 Scheme 6 Cumulenes Ab initio calculations (MP3/6-31G*//MP2/6-31G*) on the apparent [2]2] cycloaddition of cyclopentadiene to ketene suggest a pathway via an initial concerted [4]2] addition across the ketene C––O with subsequent [3,3] sigmatropic (Claisen) rearrangement to yield the cyclobutanone product. For the reaction of cyclopentadiene with diphenylketene the intermediate [4]2] adduct 12 is detectable by NMR spectroscopy at low temperature (Scheme 5).29 Several theoretical studies of borderline concerted/zwitterionic imine to ketene,30 imine to isocyanate,31 and aldehyde to ketenimine32 cycloadditions have appeared.1,3-Dipolar cycloadditions The isolation of stable zwitterionic intermediates 13 in the 1,3-dipolar cycloaddition of methanesulfonyl azide 14 to the strongly-polarised alkene 15 is evidence for a nonconcerted mechanism (Scheme 6); the zwitterions yielded the formal cycloaddition products on heating. The possibility of cycloadducts formed via a concerted mechanism following reversion of the zwitterions is excluded by experiments with deuteriated samples.33 The endo product 18 is found exclusively for cycloaddition of the butenoate 17a (R\CH 3 ) to the nitrone 16 while some exo product is found for the addition of trifluorobutenoate 17b (R\CF 3 ) (Scheme 7).From the results of a PM3 analysis this has been attributed to a secondary interaction between the CF 3 group and the nitrone HOMO in a ‘concerted’ TS.34 Asymmetric cycloadditions The p-facial selectivity in the asymmetric Diels–Alder reaction of the chiral diene (S)-19 with maleic anhydride has been rationalised in terms of attack on the conformer which minimises 1,3-allylic and 1,2-eclipsing strains about the diene C2–C3 bond.35 Styrene-based polymers with chiral oxazaborolidinone units were found to catalyse the asymmetric Diels–Alder reaction between methacrolein and cyclopentadiene giving up to 99 1 exo selectivity and up to 95% ee although an explanation for the enhanced stereoselectivity is not given.36 Conversely binding of a Cl 2 Ti- 31 Reaction mechanisms Part (i) Pericyclic reactions N O MeO2C H R R MeO2C N + O – 17a R = CH3 17b R = CF3 16 + 18 Scheme 7 O O O H R¢HN OR H 3 2 19 TADDOLate† (via para positions of aryl groups) to cross-linked polystyrenes Merri- field resins or dendritic molecules led to a decrease in enantioselectivity for the Diels–Alder reaction between cyclopentadiene and the dienophile 20.The findings have been interpreted as indicating a cationic trigonal-bipyramidal complex (e.g. 21) as the catalytically active species.37 Miscellaneous The zirconium metallacycle 22 (Ar\various para-substituted phenyl; Ar@\2,6- dimethylphenyl) undergoes a [4]2] retro-cycloaddition to yield the a,b-unsaturated imine 23 and products probably derived from Cp 2 Zr––O (Scheme 8).The retrocycloaddition mechanism was proposed on the basis of the similarity of the activation parameters [*H8\110.8 kJ mol~1 (26.5 kcal mol~1) *S8\14.55 J mol~1K~1 (3.48 cal mol~1K~1)] to those commonly found for the retro-Diels–Alder reaction.38 The coupling of graphite with microwave heating as a support for Diels–Alder reactants gave rapid cycloaddition in good yield. For example cycloaddition of 1-(dimethylamino)-3-methyl-1-azabuta-1,3-diene to dimethyl acetylenedicarboxylate unreactive by conventional heating was accomplished (on graphite) in 60% yield with 10 sequential one-minute irradiations.39 †TADDOL\a,a,a@,a@-tetraaryl-1,3-dioxolane-4,5-dimethanol.32 Ian D. Cunningham N Ar Ar ¢ Ph N Zr O Ar ¢ Ph Ar [Cp2Zr=O] 22 23 + Scheme 8 O C N N – O N – O N O N O [4 + 2] [2 + 2] p4 p6 + + + Scheme 9 3 Electrocyclic reactions The reaction of ketene with a,b-unsaturated imine involves initial nucleophilic attack of imine nitrogen on the ketene sp-hybridised carbon followed by p4 or p6 electrocyclic ring closure to give [2]2] or [4]2] cycloaddition products respectively (Scheme 9). The preference of mono-substituted ketene for p4 ring closure ([2]2] product) and the enhancement of p6 ring closure ([4]2] product) for the di-substituted ketene have been rationalised by ab initio (MP2/6-31G*) and semi-empirical (AM1) calculations in terms of torquoelectronic e§ects on the conrotatory (p4) and disrotatory (p6) routes.40 A summary of torquoselectivity (i.e.whether rotation is ‘inward’ or ‘outward’) in the conrotatory electrocyclic ring-opening of 3-substituted cyclobutenes rationalised in terms of interaction of the r bond-localised FM orbitals with donor or acceptor orbitals of the 3-substituent has appeared.41 The theme has been extended to cover a range of 1- and 3-substituted cyclobutenes; calculations are typically at the RHF/6- 31G* level and the relationship between torquoselectivity and the Taft rR 0 parameter has been demonstrated.42 From a similar analysis of torquoselectivity in cyclobutenone ring-opening the smaller e§ect is attributed to di§erences in energies for the r bond-localised FM orbitals.43 An ab initio (RHF/3-21G) prediction of BF 3 –OEt 2 -enhanced ‘inward’ torquoselectivity for ring opening of 3-acetylcyclobutene has been verified by experiment.44 The electrocyclisation of vinylallenes (e.g.24) has been found to proceed with high 33 Reaction mechanisms Part (i) Pericyclic reactions • R R¢ R¢ R E 25 24 Scheme 10 regioselectivity (e.g. for cyclisation of the terminal vinyl allene in 24) (Scheme 10). First-order rate constants are lower for R\CH––O (k\300 h~1 R@\But) than for R\CH 2 OTBDMS‡ (k\900 h~1 R@\But) probably due to loss of conjugation in the TS of the former. Torquoselectivity favours ‘inward’ rotation of the allene to give the E product when R@\But; e.g. E:Z is 83 17 for 25 (R\CH 2 OTBDMS R@\But) but 50 50 for 25 (R\CH 2 OTBDMS R@\Me) so steric factors predominate. The findings were supported by ab initio (MP2/6-31G*//RHF/6-31G*) calculations.45 An ab initio theoretical study (CASSCF and MP2) proposes that Dewar benzene 26 is converted into benzene via an (allowed) conrotatory process despite the strained cis,cis,trans-cyclohexa-1,3,5-triene (Mo� bius benzene) intermediate 27 (Scheme 11).This intermediate is predicted to lie in a shallow basin with a barrier for trans-p-bond rotation to benzene of \12.5 kJ mol~1 (3 kcal mol~1); calculations also support the existence of trans-Dewar benzene 28 in a high energy basin 660 kJ mol~1 (158 kcal mol~1) above benzene.46 Reaction profiles for the conrotatory (allowed) and disrotatory (forbidden) ringopenings of some cyclobutenes (e.g. compounds 29 and 30) have been constructed using values of activation parameters obtained froexperiments involving competition trapping by O 2 and NO.Values of *H8 [e.g. 178.1 kJ mol~1 (42.6 kcal mol~1) for 29] for the disrotatory process yielded *H& 8 values [e.g. 314.8 kJ mol~1 (75.3 kcal mol~1) for 29] similar to those calculated for the orthogonal diradicals derived from the ‘diene’ products suggesting a non-concerted mechanism for this process. The value of *Hf 8 (disrotatory) for compound 30 TS was a little lower than the diradical value suggesting here a true forbidden disrotatory ring opening.47 The ring-opening of aminoazirinium ions such as 31 has been studied. The lack of products derived from the achiral ring-opened intermediate 32 and the lack of racemisation in the products which are found excludes such an electrocyclic opening as a significant process in the overall reaction (Scheme 12).48 4 Sigmatropic rearrangements Ab initio calculations indicate a low barrier of 53 kJ mol~1 (12.7 kcal mol~1) for the [1,3] chlorine migration 33]33@ via an ‘in-plane’ pathway involving interaction of the chlorine lone pair with the keteneLUMOlobe on the central carbon (Scheme 13).This ‡TBDMS\tert-butyldimethylsilyl. 34 Ian D. Cunningham H H H H 27 26 28 conrotatory disrotatory conrotatory disrotatory p-bond rotation Scheme 11 30 29 N Me N Me Me But N NMe But Me Me 32 ( R)-31 + + Scheme 12 C Cl C C O O H C C C* Cl O O H C* C C Cl O O H 33 33¢ LUMO l.p. Scheme 13 route termed ‘lone pair–LUMO mediated pericyclic reaction’ by the authors is calculated to be substantially lower in energy than the true pericyclic antarafacial [1,3] chlorine shift in the related 3-chloroprop-1-ene; the ‘lone pair–LUMO’ mechanism is also calculated to be favoured for the vinylogous [1,5] chlorine migration.49 The rate constants for the conversion of deuteriated trans-1-ethenyl-2-phenylcyclopropane 34 into 3-phenylcyclopentene 35 were determined (Scheme 14) i.e.k4* for suprafacial migration across the p-system with inversion at the migrating carbon k!3 for antarafacial migration across the p-system with retention at the migrating carbon etc. The reaction flux via the formally forbidden ([p24 ]r24] and [p2! ]r2!]) modes as indicated by k43 ]k!* was found to be almost the same as via the allowed modes suggesting that this reaction is not controlled by orbital symmetry factors.50 A theoretical treatment of the ‘allylic 1,3-strain e§ect’ with relevance to the ene reaction (an intermolecular [1,5] hydrogen migration) has appeared.51 A theoretical 35 Reaction mechanisms Part (i) Pericyclic reactions D D D D D5C6 D D D5C6 D D etc.34 35 via ksi Scheme 14 C C C Me3Si C N N SiMe3 R R H R N R CH2 H SiMe3 SbCl6 – SbCl6 – H 36 'inverse' TS + + Scheme 15 R B R B R 37 38 Scheme 16 study using semi-empirical (e.g. AM1) ab initio (e.g. HF MP2) and density functional (e.g. Becke3LYP) methods of the retro-ene reaction of methyl prop-2-ynyl ether indicates a TS with the hydrogen lying equidistant between the two carbon atoms and an aromatic TS.52 An inverse electron demand ene reaction in which the iminium salt 36 (R\alkyl) normally an enophile behaves as the ene component (H-donor) while allyltrimethylsilane (normally an ene) behaves as the enophile (Scheme 15) has been reported.This ‘reversal’ is attributed to the ability of Me 3 Si in the ‘inverse’ TS to stabilise the developing]charge on C2 of the enophile; in addition the ‘normal’ TS would involve some unfavourable steric interactions.53 Activation barriers for the [3,3] sigmatropic rearrangement (Cope) of the 9- borabarbaralanes 37 (Scheme 16) have been determined to be 45.7 kJ mol~1 (10.9 kcal mol~1) (R\But) and 40.5 kJ mol~1 (9.7 kcal mol~1) (R\Ph) by NMR spectroscopy; barriers for other substituents (R\H Me NH 2 ) were calculated (Becke3LYP/6-311G**) and found to be 41.1–35.0 kJ mol~1 (9.8–8.4 kcal mol~1).54 A dynamic NMR investigation of the related rearrangements in substituted bullvalenes 38 (R\F CN CO 2 H) gives values of E!#5 for interconversion of the major isomers similar to those found for the parent bullvalene ca.54.8 kJ mol~1 (13.1 kcal mol~1).55 The oxyanion 39 undergoes a [3,3] (Cope) rearrangement via the conformation 36 Ian D. Cunningham O – CH3 O – H RO H 7 5 4 2 40 41 39 Scheme 17 O O O O O O O O •• •• •• •• • • Scheme 18 shown. However the analogue 40 does not probably due to the destabilising e§ect of the steric interaction shown on the chair TS; instead a [1,3] carbon shift is observed (Scheme 17). There is increasing debate about how common true pericyclic [1,3] carbon shifts really are (vide supra) and the mechanism here is not explored in detail although the inversion of the migrating carbon (compare the cis stereochemistry of the Oatom at C2 relative to the methyl at C4 in 40 with the transOatom at C5 relative to methyl at C7 in 41) is consistent with a [p24 ]r2!] mechanism.56 Ab initio calculations (using e.g.MP2/6-311G**//RHF/6-31G*) were carried out to explore the [3,3] sigmatropic rearrangements of 3-oxa- -aza- -thia- and -phospha-1,5- dienes and of allyl phenyl ether and its hetero-analogues.57 The importance of using di§use functions and the value of the DFT method for theoretical treatments of [3,3] shifts has been noted.58 A DFT study of the 1,3-acyloxy shift in allyl formate and the 1,2-acyloxy shift in the ethyl formate radical favours a nucleophilic attack on the alkene or radical by the oxygen lone pair rather than a normal [3,3] mechanism involving the carbonyl p orbital particularly for the open shell (radical) case (Scheme 18).59 The aza-[2,3]-Wittig reaction continues to attract attention mainly for its synthetic potential.Incorporation of a trimethylsilyl group in 42 accelerated the rearrangement (Scheme 19). As with many rearrangements there is debate over concerted vs. stepwise mechanisms. Here the trimethylsilyl e§ect is consistent with the TS 43 based on the model of Houk60 in which a d~ charge develops on C2 while the variation in diastereoselectivity is attributed to increased preference for the less crowded 43@ rather than 43 TS.61 On the other hand the trans-2-alkyl-3-alkenylaziridines bearing a tert-butoxycarbonylmethyl N-substituent [e.g. 44 (R\alkyl; R@\H Me)] were found to yield mainly cis-2,6-disubstituted tetrahydropyridines by an aza-[2,3]-Wittig reaction (Scheme 20). While this can be explained by consideration of a TS based on the Houk model (e.g. 45) the fact that the cis isomers (i.e.R is cis relative to the alkene) of the aziridines gave a mixture of cis and trans tetrahydropyridines is taken to indicate a more complex mechanism with the possibility of an anionic ring-opened intermediate. 62 A theoretical study of the relative importance of the [1,2] (Stevens) and the [2,3] (Sommelet–Hauser) rearrangements of the ammonium ylideN-methyl-3-propenylam- 37 Reaction mechanisms Part (i) Pericyclic reactions N Me3Si R H Boc Ph N Me3Si R H Boc Ph N SiMe3 Ph Boc R SiMe3 R N Ph Boc SiMe3 R N Ph Boc 2 42 d– d– d– 43¢ d– d– d– 43 •• – •• •• – – Scheme 19 N CO2But H R H R¢ N H R R¢ CO2But N H CO2But H R¢ LDA 44 45 •• – H R Scheme 20 monium methylide found the two processes to be very close in energy; N-substitution was calculated to favour a dissociative ([1,2]) mechanism while substitution and therefore delocalisation of the double bond prefers a pericyclic [2,3] mechanism.63 An analysis of allyl methyl thioether anion rearrangements using mass spectrometry has given a value of E!#5 for the [1,4] process for 46 of 112.9 kJ mol~1 (27 kcal mol~1); this is consistent with a homolytic S–Me cleavage stepwise mechanism.A [2,3] process was observed for 47 and this appears to follow a concerted mechanism.64 S S 46 47 •• •• – – 5 Miscellaneous The cyclisation of 1-(prop-2-ynyloxy)naphthalene and related compounds which involves a Claisen rearrangement [1,5] hydrogen shift and p6 electrocyclisation was 38 Ian D.Cunningham C C N N H H H H CH2 NH C NH + 48 •• Scheme 21 Cl Cl Cl Cl Cl Cl N N Ar Ar H H Cl Cl Cl Cl Cl Cl N N Ar Ar H H Scheme 22 O N COR N O R O N R N COR 51 p4 electrocyclisation 50 49 retro-Diels–Alder p6 electrocyclisation Scheme 23 found to be greatly accelerated by microwave irradiation.65 The [2]1] cycloreversion of a model aziridinimine 48 (Scheme 21) has been studied by ab initio methods [CISDQ and QCISD(T)/6-311G(d,p)//MP2/6-31G(d,p)].The reversion mirrors the cheletropic [2]1] cycloaddition of isocyanide to imine in that the C–C bond is a§ected less than the C–N bond in the TS although the positive calculated *S8 is at variance with the negative values found experimentally.66 Rate constants for intramolecular dyotropy (e.g. Scheme 22) have been measured and compared to values for similar compounds. The trends in rate constant and the linearity of lnk vs. 1/T plots indicate a concerted ([r4]p2]) mechanism with no tunnelling.67 Evidence is presented for the formation of the 1,3-benzoxazines 49 rather than the N-acyl-1,2-dihydrobenzazetes 50 from the thermal decomposition of the benzoxazines 51 (R\alkyl phenyl) (Scheme 23); overall E!#5 values are 146–175 kJ mol~1 (34.9–41.8 kcal mol~1).The mechanism is proposed as a retro-Diels–Alder extrusion of formaldehyde followed by a p6 (rather than a p4) electrocyclisation; AM1 calculations are consistent with this although calculated E!#5 values are higher than experimental values.68 39 Reaction mechanisms Part (i) Pericyclic reactions References 1 O. Wiest and K. N. Houk Top. Curr. Chem. 1996 183 1. 2 M. Lautens W. Klute and W. Tam Chem. Rev. 1996 96 49. 3 S. Laschat Angew. Chem. Int. Ed. Engl. 1996 35 289.4 K.V. Gothelf and K. A. Jorgensen Acta Chem. Scand. 1996 50 652. 5 E. Goldstein B. Beno and K. N. Houk J. Am. Chem. Soc. 1996 118 6036. 6 B.R. Beno K. N. Houk and D. A. Singleton J. Am. Chem. Soc. 1996 118 9984. 7 G. Ohanessian Y. Six and J.-Y. Lallemand Bull. Soc. Chim. Fr. 1996 133 1143. 8 V. Barone R. Arnaud P. Y. Chavant and Y. Valle� e J. Org. Chem. 1996 61 5121. 9 Q. Dang B. S. Brown and M. D. Erion J. Am. Chem. Soc. 1996 118 5204. 10 J. Maddaluno O. Gaonac’h A. Marcual L. Toupet and C. Giessner-Prettre J. Org. Chem. 1996 61 5290. 11 J. Laue and G. Seitz Liebigs Ann. Chem. 1996 645; ibid. 1996 773. 12 G. A. McLean B. J. L. Royles D. M. Smith and M. J. Bruce J. Chem. Res. (S) 1996 448; (M) 2623. 13 H. Hopf H. Ja� ger and L. Ernst Liebigs Ann. Chem. 1996 815. 14 A. D. Darwish A.G. Avent R. Taylor and D. R. M. Walton J. Chem. Soc. Perkin Trans. 2 1996 2079. 15 J. A. Tallarico M. L. Randall and M.L. Snapper J. Am. Chem. Soc. 1996 118 9196. 16 N. J. Sisti I. A. Motorina M.-E. Tran Huu Dau C. Riche F. W. Fowler and D. S. Grierson J. Org. Chem. 1996 61 3715. 17 B. A. Horn J. L. Herek and A. H. Zewail J. Am. Chem. Soc. 1996 118 8755. 18 A. de Meijere S. Teichmann F. Seyed-Mahdavi and S. Kohlstruk Liebigs Ann. Chem. 1996 1989. 19 M. Buback G. Kuchta A. Niklaus M. Henrich I. Rothert and L. F. Tietze Liebigs Ann. Chem. 1996 1151. 20 K. Mackenzie E. C. Gravett J. A. K. Howard K. B. Astin and A. M. Tomlins J. Chem. Soc. Perkin Trans. 2 1996 1233. 21 R. A. Firestone Tetrahedron 1996 52 14 459. 22 N. P. Schepp and L. J. Johnston J. Am. Chem. Soc. 1996 118 2872. 23 W.Schmittel C. Wo� hrle and I. Bohn Chem. Eur. J. 1996 2 1031. 24 W. Yeuh and N. L. Bauld J. Chem. Soc. Perkin Trans. 2 1996 1761. 25 G. K. van der Wel J. W. Wijnen and J. B. F. N. Engberts J. Org. Chem. 1996 61 9001. 26 J. W. Wijnen S. Zavarise and J. B. F. N. Engberts J. Org. Chem. 1996 61 2001. 27 S. Otto F. Bertoncin and J. B. F. N. Engberts J. Am. Chem. Soc. 1996 118 7702. 28 M.B. Mcginnis K. Vagle J. F. Green L. Choo Tan R. Palmer J. Siler R.M. Pagni and G. W. Kabalka J. Org. Chem. 1996 61 3496. 29 S. Yamabe T. Dai T. Minato T. Machiguchi and T. Hasegawa J. Am. Chem. Soc. 1996 118 6518. 30 (a) D.-C. Fang and X.-Y. Fu Int. J. Quantum Chem. 1996 57 1107; (b) R. Lopez M. F. Ruizlopez D. Rinaldi J. A. Sordo and T. L. Sordo J. Phys. Chem. 1996 100 10 600; (c) D.-C. Fang and X.-Y. Fu Chin.J. Chem. 1996 14 97. 31 D.-C. Fang and X.-Y. Fu J. Mol. Struct. 1996 365 219. 32 D.-C. Fang and X.-Y. Fu Chem. Phys. Lett. 1996 265. 33 H. Quast M. Ach A. Ivanova E.-M. Peters K. Peters and H. G. von Schnering Liebigs Ann. Chem. 1996 1551. 34 K. Tanaka T. Imase and S. Iwata Bull. Chem. Soc. Jpn. 1996 69 2243. 35 G. T. Crisp and M. G. Gebauer J. Org. Chem. 1996 61 8425. 36 K. Kamahori K. Ito and S. Itsuno J. Org. Chem. 1996 61 8321. 37 D. Seebach R. E. Marti and T. Hintermann Helv. Chim. Acta 1996 79 1710. 38 T. A. Hanna A. M. Baranger and R. G. Bergman J. Org. Chem. 1996 61 4532. 39 B. Garrigues C. Laporte R. Laurent A. Laporterie and J. Dubac Liebigs Ann. Chem. 1996 739. 40 B. Lecea I. Arrastia A. Arrieta G. Roa X. Lopez M. I. Arriortua J. M. Ugalde and F. P. Cossý� o J. Org.Chem. 1996 61 3070. 41 W.R. Dolbier Jr. H. Koroniak K. N. Houk and C. Sheu Acc. Chem. Res. 1996 29 471. 42 S. Niwayama E. A. Kallel D. C. Spellmeyer C. Sheu and K. N. Houk J. Org. Chem. 1996 61 2813. 43 S. Niwayama E. A. Kallel C. Sheu and K. N. Houk J. Org. Chem. 1996 61 2517. 44 S. Niwayama J. Org. Chem. 1996 61 640. 45 S. Lopez J. Rodrigu� ez J. G. Rey and A. R. de Lera J. Am. Chem. Soc. 1996 118 1881. 46 R. P. Johnson and K. J. Daoust J. Am. Chem. Soc. 1996 118 7381. 47 W.R. Roth V. Rekowski S. Bo� rner and M. Quast Liebigs Ann. Chem. 1996 409. 48 H. Quast S. Aldenkortt B. Freudenreich P. Scha� fer E.-M. Peters K. Peters H. G. von Schnering and E.-U. Wu� rthwein Liebigs Ann. Chem. 1996 87. 49 R. Koch M.W. Wong and C. Wentrup J. Org. Chem. 1996 61 6809. 50 J. E. Baldwin and S.J. Bonacorsi Jr. J. Am. Chem. Soc. 1996 118 8258. 51 L. F. Tietze and G. Schulz Liebigs Ann. Chem. 1996 1575. 52 P. Va� rnai and G. Keseru� J. Org. Chem. 1996 61 5831. 53 A. R. Ofial and H. Mayr J. Org. Chem. 1996 61 5823. 40 Ian D. Cunningham 54 G. L. Herberich H.-W. Marx S. Moss P. vR. Schleyer and T. Wagner Chem. Eur. J. 1996 2 458. 55 R. Poupko H. Zimmermann K. Mu� ller and Z. Luz J. Am. Chem. Soc. 1996 118 7995. 56 L. A. Paquette S. Liang and H.-L. Wang J. Org. Chem. 1996 61 3268. 57 S. Yamabe S. Okumoto and T. Hayashi J. Org. Chem. 1996 61 6218. 58 P.M. Warner J. Org. Chem. 1996 61 7192. 59 H. Zipse J. Chem. Soc. Perkin Trans. 2 1996 1797. 60 K. Mikami T. Uchida T. Hirano Y.-D. Wu and K. N. Houk Tetrahedron 1994 50 5917; Y.-D. Wu K. N. Houk and J. A. Marshall J. Org.Chem. 1990 55 1421. 61 J. C. Anderson D. C. Siddons S. C. Smith and M. E. Swarbrick J. Org. Chem. 1996 61 4820. 62 J. A ! hman T. Jareva" ng and P. Somfai J. Org. Chem. 1996 61 8148. 63 G. L. Heard and B. F. Yates J. Org. Chem. 1996 61 7276. 64 M.R. Ahmad G. D. Dalke and S. R. Kass J. Am. Chem. Soc. 1996 118 1398. 65 F. Matloubi Moghaddam A. Sharifi and M.R. Saidi J. Chem. Res. (S) 1996 338. 66 M.T. Nguyen A. Van Keer and L. G. Vanquickenborne J. Chem. Soc. Perkin Trans. 2 1996 299. 67 K. Mackenzie J. A. K. Howard R. Siedlecka K. B. Astin E. C. Gravett C. Wilson J. Cole R. G. Gregory and A. S. Tomlins J. Chem. Soc. Perkin Trans. 2 1996 1749. 68 S. A. Glover K. M. Jones I. R. McNee and C. Rowbottom J. Chem. Soc. Perkin Trans. 2 1996 1367. 41 Reaction mechanisms Part (i