3 Reaction Mechanisms Part (ii) Orbital Symmetry Correlations and Pericyclic Reactions By R. GRIGG Department of Chemistry University of Nottingham Nottingham NG7 2RD 1 Orbital Symmetry Correlations The major proponents of alternative treatments of pericyclic processes to that proposed by Woodward and Hoffmann,' have reviewed their theories. Basically these approaches fall into two main categories. Fukui' limits his consideration to the frontier orbitals (the highest occupied molecular orbital and the lowest unoccupied molecular orbital) since these are energetically the most accessible and considers favourable overlap of the frontier orbitals must occur in the region of bond formation for a concerted process to be favoured. This requires a know- ledge of the symmetries of the frontier orbitals but has great utility in interpreting complicated reactions involving extensive bond relocation.The other approach requires no knowledge of MO theory since it utilizes the 'basis set orbitals' i.e. the assortment of orbitals present prior to a MO calculation. This approach has been used both by Dewar3 and by Zimmerman4 and will prove attractive to chemists with no knowledge of MO theory. Their treatment leads to the con- clusion that thermal pericyclic processes proceed preferentially uia aromatic transition states whereas photochemical pericyclic reactions involve anti- aromatic transition states. The idea of an aromatic transition state was first suggested for the Diels-Alder reaction5 some years ago. A number of other reviews have appeared dealing with the interaction of orbitals through space,' orbital symmetry and inorganic chemistry,' and orbital symmetry in thermal and photochemical reactions.* The energetics of photochemical pericyclic processes call for some rethinking of the original Woodward and Hoffmann approach.Although the predictions for photochemical processes have in the main been well substantiated the ' R. B. Woodward and R. Hoffmann Angew. Chem. Internut. Edn. 1969 8 781. K. Fukui Accounts Chem. Res. 1971 4 57. ' M. J. S. Dewar Angew. Chem. Internut. Edn. 1971 10 761. H. E. Zimmerman Accounts Chem. Res. 1971,4 272. ' M. G. Evans Trans. Faraday Soc. 1939 35 824. R. Hoffmann Accounts Chem. Res. 1971 4 1. R. G. Pearson Accounts Chem.Res. 1971 4 152; Pure Appl. Chem. 1971 27 145. H. Katz J. Chern. Educ. 1971 48 84; M. C. Caserio J. Chem. Educ. 1971 48 782. 143 144 R. Grigg mechanics of some of the processes are obscure. Thus the photocyclization of butadiene does not produce the first excited singlet state of cyclobutene which is energetically inaccessible from the first excited singlet state of butadiene instead ground-state cyclobutene is produced. The excited state responsible for the cyclization therefore may not be the first excited singlet state.' A related problem exists in (l),which is unreactive in the first excited singlet state and requires further energy input before opening to (2)." 2 Electrocyclic Reactions Although construction of a correlation diagram for electrocyclic reactions of radical species does not demonstrate any preference for a conrotatory or a disrotatory process extended Huckel calculations led to the prediction that radical reactions should proceed with the same stereochemistry as the corres- ponding anionic species e.g.cyclopropyl radicals should resemble cyclopropyl anions and open in a conrotatory fashion.' However MIND0/2 calculations lead to the prediction' 'that radicals should resemble the corresponding cations e.g. cyclopropyl radicals should undergo disrotatory opening. It is to be hoped that despite the difficulties associated with radical reactions a test of these opposing predictions will be forthcoming. MIND0/2 studies have also been reported for disfavoured ('forbidden') electrocyclic processes.The thermal opening of the oxaziridines (3) produces the nitrones (4) and occurs by both conrotatory modes giving a cis/trans mixture. Low-temperature irradiation reforms the oxaziridine by disrotatory clo~ure.~ The aziridine (5) is sterically unable to undergo conrotatory opening to the dipolar species (6) and the disrotatory process14 occurs at 135 "C as found for the related oxirane (7) which gives @).I5 The solvolysis rates of substituted N-chloroaziridines are as expected for a concerted ionization-ring cleavage process. l6 Thus relative solvolyses rates of (9; R = H) (9; R = Me) and (10) are 1 1490 155 000; W. Th. A. M. van der Lugt and L. J. Oosterhoff J. Amer. Chem. SOC. 1969 91 6042. lo J. Meinwald G. E. Samuelson and M.Ikeda J. Amer. Chem. SOC.,1970 92 7604; J. Michl J. Amer. Chem. SOC.,1971 93 523. 'I M. J. S. Dewar and S. Kirschner J. Amer. Chem. SOC.,1971 93 4290. l2 M. J. S. Dewar and S. Kirschner J. Amer. Chem. SOC. 1971 93 4291 4292. l3 J. S. Splitter T.-M. Su H. Ono and M. Calvin J. Amer. Chem. Soc. 1971 93 4075. l4 J. W. Lown and K. Matsumoto J. Org. Chem. 1971 36 1405. E. F. Ullman and J. E. Milks J. Amer. Chem. SOC.,1962 84 1315; E. F. Ullman and W. A. Henderson J. Amer. Chem. SOC. 1966 88 4942. l6 P. G. Gassman Accounts Chem. Res. 1970 3 26. Reaction Mechanisms-Part (ii) CND0/2 calculations for this process predict the disrotatory opening. l7 Further examples of electrocyclic reactions of 3-membered heterocycles are included in the cycloaddition section.0 0 +/ /\ + ArCH=N Ar CH- NR \ R (3)(3) (4)(4) Ph Ph 1 0 (7) R H R (9) A weight of experimental evidence *suggests the thermal vinylcyclopropane-cyclopentene rearrangement is non-concerted and occurs via a diradical inter-mediate. However vinyl epoxides have an alternative symmetry-allowed pathway. Thermolysis in the gas phase of the cis-epoxide (11) furnishes the 6-electron ylide (12) which partially equilibrates with the isomeric ylide (13) but predominately cyclizes to the dihydrofuran (14).19 The degenerate rearrange-ment of the vinylcyclopropane thujene (15) has been studied using an optically active trideuterio-derivative and shown to be non-concerted.20 " R. G. Weiss Tetrahedron 1971 27 271.M. R. Wilcott and V. H. Cargle J. Amer. Chem. Soc. 1969 91 431 1; A. D. Ketley A. J. Berlin and L. P. Fisher J. Org. Chem. 1966,31 2648; M. J. S. Dewar Tetrahed-ron 1966 suppl. no. 8 p. 75. I9 J. C. Paladini and J. Chuche Tetrahedron Letters 1971 4383. W. von E. Doering and E. K. G. Schmidt Tetrahedron 1971 27 2005. 146 R. Grigg Me Ph 'H f A A 7 q-3 t MewMe (14) Thermal rearrangement of the tricycloheptane (16) occurs by two competing pathways21 to give (17) and (18). The observed energies of activation are difficult to rationalize by a diradical mechanism and in contrast to the thermal rearrange- ment of (19),22a concerted process may be operative. The cyclobutyl-cyclopropylcarbinyl rearrangement occurs with preferential inversion at the reaction site [(20)+ (21)].23 The process may be viewed as a two-electron electrocyclic process analogous to that occurring in the solvolysis of cyclopropyl derivatives.The reaction is sensitive to steric effects and as expected the cis-isomer (20; R = alkyl) solvolyses slower than the trans-isomer (22; R = alk~l).~~ 21 H. M. Frey R. G. Hopkins and L. Skattebol J. Chem. SOC. (B) 1971 539. 22 M. C. Flowers and A. R. Gibbons J. Chem. SOC. (B) 1971 612. 23 I. Lillien G. F. Reynolds and L. Handloser Terrahedron Letters 1968 3475; I. Lillien and L. Handloser ibid. 1969 1035; Z. Majerski and P. von R. Schleyer J. Amer. Chem. SOC. 1971 93 665. 24 I. Lillien and L. Handloser J. Amer. Chem. SOC.,1971 93 1682.Reaction Mechanisms-Part (ii) X -%’ ‘H Thermolysis of the deuteriated cyclobutene (23) leads to conrotatory opening in both directions with a slight predominance of one isomer (24 25 ;52.3 47.7). This is interpreted as a slight preference for a transition state involving D-D H OSiMe Me,SiO D Me,SiO H Me$ D OSiMe interaction [i.e. (26)] as a consequence of the smaller effective size of carbon-bound deuterium compared to hydr~gen.~’ However the isomeric stability of the pure dienes under the reaction conditions was not reported. The thermal opening of benzocyclobutanes to 0-quinodimethanes followed by an intra- molecular Diels-Alder reaction has been utilized in elegant syntheses of fused-ring systems e.g. (27)-(28). The scheme has been applied to the synthesis of ( f)-chelidonine(29).26 25 R.E. K. Winter and M. L. Honig J. Amer. Chem. SOC.,1971 93 4616. 26 W. Oppolzer J. Amer. Chern. SOC.,1971 93 3833 3834; W. Oppolzer and K. Keller ibid. p. 3836. 148 R. Grigg 07 The position of the disrotatory cycloheptatriene-norcaradiene equilibrium [(30)= (31)] is markedly influenced by the C-7 substituents R1 and R2. Substi-tuents containing n-systems more particularly electron-withdrawing substituents appear to be essential for the stabilization of the norcaradiene valence isomer ;a simple theoretical explanation for this has been pr~vided.~' A number of new examples have been reported2* and some studied by l3Cn.m.r. Ability of C-7 substituents to stabilize the norcaradiene valence tautomer is in the order C(CN) > C(Ph)CO,R z C(Ph)PO(OMe) > C(CO,R) .Valence isomerism Thus the in the azocine series is also markedly affected by sub~tituents.~~ equilibrium (32) S(33) lies wholly on the side of (33) for n = 3 or 4. When n = 5 the equilibrium (32) $(33) is established at 100 "C whereas when n = 6 only (32) is present. Some interesting dienophile specificities in intercepting valence H 27 R. Hoffmann Tetrahedron Letters 1970 2907. 28 E. Ciganek J. Amer. Chem. SOC.,1971 93 2207; G. E. Hall and J. D. Roberts J. Amer. Chem. SOC.,1971 93 2203; H. Durr and H. Kober Angew. Chem. Internat. Edn. 1971 10 342; H. Gunther €3. D. Tunggal M. Regitz H. Scherer and T. Keller Angew. Chem. Internat. Edn. 1971 10 563.29 L. A. Paquette J. F. Hansen and J. C. Philips J. Amer. Chem. SOC.,1971 93 152; L. A. Paquette T. Kakihana and J. F. Kelly J. Org. Chem. 1971 36 435; L. A. Paquette Angew. Chem. Internat. Edn. 1971 10 11. Reaction Mechanisms-Part (ii) tautomers have been reported for fluorocyclo-octatetraene.30 Thus tetracyano- ethylene intercepts the valence tautomer (34) whereas the azodienophiles (35 ; R = Me or Ph) intercept (36) exclusively. N The thermal rearrangement cis-bicyclo[6,1,O]nona-2,4,6-trienes continues to attract attention. The problem is that the thermal rearrangement of 9-substituted derivatives (37; R' = R2 = H; R' = H R2 = alkyl) should on simple electro- cyclic ring opening give the cis,cis,cis,trans-cyclononatetraene (38) which on disrotatory closure involving a six-n-electron portion of (38) should furnish a trans-8,9-dihydroindene (39).The products from the parent system (37; R' = R2 = H) and the monoalkyl derivatives are however predominantly the cis-8,9-di h ydroindenes. It seems generally agreed that the all-cis cyclononatetraene (40)is the immediate precursor of the cis-8,9-dihydroindene and indeed (40) has been prepared and shown to undergo the expected cyclization to (41 ;R' = R2 = H).32 A number of 30 G. Schroder G. Kirsch J. F. M. Oth R. Huisgen W. E. Konz and U. Schnegg Chem. Ber. 1971 104 2405. 31 E. Vogel W. Grimme and E. Dinne Tetrahedron Letters 1965 391 ; P. Radlick and W. Fenical J. Amer. Chem. SOC.,1969 91 1560. 32 G. Boche H. Bohme and D. Martens Angew.Chem. Internat. Edn. 1969 8 594; P. Radlick and G. Alford J. Amer. Chem. SOC.,1969 91 6529. 150 R. Grigg valence tautomers of the bicyclo[6,l,O]nonatriene have been suggested as possible intermediates. However some possible rearrangement pathways have been ruled out by the observation that the hexadeuterio-derivative (42) rearranges to (43).33 The importance of the conformation of the bicyclo[6,1,O]nonatriene was first suggested by Staley and Henry,34 who reported that the 9,9-dimethyl derivative (37; R' = R2 = Me) rearranges to the trans-dihydroindene (39; R' = R2 = Me) in contrast to the parent system and the 9-monoalkyl derivatives. Thus the favoured conformation for thermal rearrangement was considered to be (44) leading to cis-dihydroindenes whereas the severe steric interactions in the 9,9-dimethyl derivative would cause it to adopt conformation (45).This view is H R (44) (45) (43) supported by kinetic data showing a higher-energy transition state for (49 and a reinvestigation of the rearrangement of (37; R' = H R2 = Me; R' = Me R2 = H) is in accord with this view.35 Thus the rearrangement of (37; R' = H R2 = Me; AG* = 31 kcal mol-') kinetically resembles that of (45 ;AG* = 32 kcal mol- ') whereas the rearrange- ment of (37; R' = Me R2 = H; AG* = 27.4kcalmol-') with conformation (44; R = Me) favoured was similar to the parent compound (37; R' = R2 = H ; AG* = 28 kcal mol- '). Careful investigation of the reaction mixture from (37; R' = H R2 = Me) indicated some trans-dihydroindene was possibly formed.An in~estigation~~ of the 9-chloro-derivatives (37; R' = D R2 = C1; R' = C1 R2 = D) has shown that (37; R' = C1 R2 = D) in which the folded conformation (44) is readily attained rearranges to the cis-dihydroindene (41 ; R' = D R2 = C1) presumably uia the all-cis cyclononatetraene. However (37; R' = D R2 = C1) in which the extended conformation should be favoured '' J. E. Baldwin and A. H. Andrist J. Amer. Chem. SOC.,1971 93 4055. S. W. Staley and T. J. Henry J. Amer. Chem. Sac. 1969 91 1239. 35 A. G. Anastassiou and R. C. Griffith Chem. Comm. 1971 1301 3b J. C. Barborak T.-M. Su P. v. R. Schleyer G. Boche and G. Schneider J. Amer. Chem. SOC.,1971 93 279; see also A. G. Anastassiou and E. Yokali J. Amer. Chem.Sac. 1971 93 3803. Reaction Mechanisms-Part (ii) undergoes migration of chlorine uia the valence tautomer (46)followed it is suggested by fast disrotatory opening of the cyclopropane ring consequent upon ionization of the chloride giving the allylic ion (47) which recaptures the chloride ion on the same face giving (48). HH c1-C1 (47) (48) In contrast to (37; R' = R2 = H) the cis-bicyclo[6,2,0]deca-2,4,6-triene(49) rearranges initially to a single product (50),as expected from orbital symmetry consideration^.^ Presumably the two-carbon bridge in (49)again favours an extended conformation analogous to (45). Irradiation of undeuteriated (49) gives a rapidly equilibrating mixture of the valence isomers (51) and (52). (51) (52) An interesting steric effect is noted38 in the photochemistry of the substituted cyclohexadienes (53; R' = H R2 = Ph; R' = Ph R2 = H) which open in the conrotatory mode that avoids a cis relationship between the (initial) C-1 and C-6 37 S.W. Staley and T. J. Henry J. Amer. Chem. SOC.,1971 93 1293; 1970,92 7612. 38 A. Padwa L. Brodsky and S. C. Clough Chem. Comm. 1971 417. 152 R. Grigg phenyl substituents giving (54; R' = H R2 = Ph; R' = Ph R2 = H) as intermediates followed by ,4 + ,2 cycloadditions to give (55). A general electrocyclic ylide reaction [(56)-+ (57)] has been proposed and examples utilizing pyridinium ylides reported.39 The conrotatory 16-n-electron electrocyclic closure of the cation (58) occurs at room temperature4' to give the trans-macrocyclic compound (59).Me Me Me Me Et Me Me -N N -+ El Me Me Me Me 3 Cycloaddition Reactions Reviews have appeared on singlet oxygen4' and ketenimine~.~' The main lectures presented at a symposium on cycloaddition reactions have been pub- li~hed.~~ The .2 + 02a cycloreversion of a cyclobutane to two ethylenes would impose severe distortions on the transition state since two of the methylene carbons are required to rotate through ca. 180". Convincing examples of this reaction are rare.44 Conversely thermal electrocyclic opening of a cyclobutene to a butadiene may also be viewed as a 2 + 2 cycloreversion which requires the methylene carbons to rotate through ca. 90". The latter process is comparatively easy and stereospecific.These considerations prompted a of the thermal opening of 2,3-dimethylbicyclo[2,l,0]pentanes,to 2,5-heptadienes in which the cyclo- propane orbital axes [(60) dashed lines] are already canted with respect to the breaking C-2-C-3 bond and thus an angular rotation intermediate between the cyclobutane and cyclobutene cases is required. The experimental results indicate a conrotatory/disrotatory preference of ca. 10,which is 10-30 times greater than 39 Y.Tamura N. Tsunjimoto and M. Ikeda Chem. Comm. 1971 310. 40 R. Grigg A. P. Johnson A. W. Johnson and M. J. Smith J. Chem. Soc. (0,1971 2457. 41 D. R. Kearns Chem. Rev. 1971 71 395. 42 G. R. Krow Angew. Chem. Internat. Edn. 1971 10 435. 43 Papers in Pure Appl. Chem. 1971 27 597-679. 44 A.T. Cocks H. M. Frey and I. D. R. Stevens Chem. Comm. 1969,458; J. E. Baldwin and P. W. Ford J. Amer. Chem. Soc. 1969 91 7192. IsJ. A. Berson W. Bauer and M. M. Campbell J. Amer. Chem. SOC.,1970,92 7515. Reaction Mechanisms-Part (ii) 5 the corresponding (2 + 2,)/(2 + 2,) rates in cycl~butanes,~~ but at least five orders of magnitude less than that of a cyclobutene opening.46 A concerted mechanism is favoured for this process in contrast to the analogous thermal opening of bicyclo[2,2,0]hexanes which involves a diradical intermediate.47 Extended Hiickel calculations on the potential surface for nonconcerted opening of cyclobutane to a tetramethylene diradical does not show an energy minimum corresponding to the diradical but rather a large energetically flat region of the potential surface termed a ‘twixtyl’ corresponding in modern collision theory to the intermediate.48 An earlier proposed example of a ,2 + ,2 cycloaddition [(61)+(62)]49 must be reconsidered in the light of studies on (63) which di- merizes to three cyclodimers in which the formal ,2 + ,2 product (64) pre- dominate~.~~ It is concluded that a diradical intermediate (65) is probably involved and that the energy barrier for disrotatory closure of (65) is slightly less than that for conrotatory closure since it enables the coplanarity of the ally1 systems to be maintained.However such an argument clearly cannot apply to (61)-+(62) although the proximate cyclohexadiene n-system might have a related effect. Observations bearing on these problems have been reported by 46 G.A. Doorakian and H. H. Freedman J. Amer. Chem. SOC.,1968,90 5310 6896. 47 E. N. Cain Tetrahedron Letters 1971 1865; L. A. Paquette and J. A. Schwartz J. Amer. Chem. SOC.,1970 92 321 5. 48 R. Hoffmann S. Swaminathan B. G. Odell and R. Gleiter J. Amer. Chem. Soc. 1970 92 709 I. 49 K. Kraft and G. Koltzenberg Tetrahedron Letters 1967 4357 4723. A. Padwa W. Koen J. Masaracchia C. L. Osborn and D. J. Trecker J. Amer. Chem. SOC.,1971 93 3633. 154 R. Grigg -Na-NH Hey,’ who observed surprisingly high stereospecificity in the reductive ring cleavage of cis-1,2-divinylcyclobutanes[e.g. (66)]. Reduction leads mainly to the &,trans-octadienes (67 ; R = H 69 %; R = Me 72 %). This is interpreted as a conrotatory opening of the intermediate (di?)radical-anion to give the allylic anions which are then protonated to yield the thermodynamically more stable disubstituted double bonds.A correlation diagram for this process is provided. The oft-quoted example of a .2 + .2 photochemical reaction the conversion of bicycloheptadienes (68) to quadricyclanes (69) appears to be a non-concerted proce~s.’~ Using a recently developed kinetic model it has proved possible to demonstrate that a two-step singlet photoaddition occurs for (68; X = CH or 0,R’ = Ar R2 = H or Me). Dewar-benzene is not formed from the first excited singlet or triplet states of benzene but from the second excited singlet state in a symmetry allowed pro~ess.’~ Stereospecific 1,2- and 1,4-photocycloaddition of olefins to benzene have been reported :54 cis-but-2-ene gives (70) and (71).However from preliminary observations it appears the first excited singlet state (‘B2J of benzene is involved in an apparently energetically unfavourable” concerted process i.e. concerted 1,3-cycloadditions are favoured from the ‘BZustate whereas the 1,2- and 1,4- processes are not. Although 1,3-photocycloadditions to benzene are known to be H Me ‘ H H (70) (71) H. Hey Angew. Chem. Internat. Edn. 1971 10 132. 52 G. Kaupp Angew. Chem. Internat. Edn. 1971 10 340. 53 D. Bryce-Smith A. Gilbert and D. A. Robinson Angew. Chem. Internat. Edn. 1971 10 745. 54 K. E. Wilzbach and L. Kaplan J. Amer. Chem. SOC.,1971 93 2073. 55 D. Bryce-Smith Chem.Comm. 1969 806. Reaction Mechanisms-Part (ii) stereospecific the question of endo or exo preference of substituents has remained unresolved. An ingenious application of [1,5]-sigmatropic shifts has enabled this problem to be resolved.56 The major isomer from the 1,3-cycloaddition of cis-but-2-ene to benzene is unchanged on heating at 300 "C for 20 min and is assigned the endo-configuration (72; R' = Me R2 = H). In contrast the minor isomer undergoes quantitative isomerization to (73) under these conditions and has the exo-configuration (72; R1= H R2 = Me). Similar studies were carried out on the trans-but-2-ene adducts and on the 1,3-cycloaddition products from benzene toluene and xylenes with cy~lobutene.~ The cyclobutene adducts all had the endo-configuration (e.g.72 R' = cyclobutyl) and rearranged on heating to derivatives of the tricyclodecadiene (74). Toluene and xylene cycloaddition ,-j";.. l"i H (74) (73) products with cyclobutene show preferential locating of a methyl group at the C-1 position [cf (72)]. The perturbational MO method has been applied5' to the Paterno-Buchi reaction (olefin + ketone or aldehyde +oxetan) and predicts a concerted path will be favoured by electron-withdrawing substituents on the olefin and electron-donating substituents on the carbonyl compound. Reversal of this substituent type is predicted to favour a diradical mechanism. Perturbational MO treatments support the roles of the frontier orbitals in determining the stereochemistry of the Diels-Alder reaction and correctly predict the effect of substituents on the rate of the reaction.59 In a 'normal' Diels-Alder reaction involving an electron-rich diene and an electron-poor dienophile the diene LUMO (lowest unoccupied MO) and dienophile HOMO (highest occupied MO) are closest in energy and result in the dominant stabilizing interaction [Figure l(a)].In the reverse electron demand Diels-Alder reaction the diene-HOMO-dienophile-LUMO interaction is the predominant one [Figure l(b)]. The effect of electron-donating substituents on the diene is to raise both the HOMO and LUMO in energy but the HOMO suffers a greater increase than the LUMO. Conversely electron-attracting substituents on the dienophile lower the energy of the HOMO and LUMO but the LUMO experiences a greater fall in energy.Thus both substitution patterns conspire to produce a more effective 56 R. Srinivasan Tetrahedron Letters 197 1 455 1. 57 R. Srinivasan J. Amer. Chem. Soc. 1971 93 3555. W. C. Herndon Tetrahedron Letters 1971 125; W. C. Herndon and W. B. Giles Mol. Photochem. 1970 2 277. 59 R. Sustmann Tetrahedron Letters 1971 2721; 0. Eisenstein and N. Trong Anh ibid. p. 1191. 156 R. Grigg -LUMO I LUMO-I I I -LUMO I I ----+---------I %HOMO HOMO% I E HOMO# #HOMO I Diene Dienophile. I Diene Dienophile (a) (b) Figure 1 HOMO-LUMO energy ordering for (a) normal Diels-Alder reaction and (b) inverse electron demand Diels-Alder reaction HOMO-diene-LUMO-dienophile interaction and hence facilitate the cyclo- addition.Substitution at the terminal positions of the diene is electronically more effective than substitution at C-2 or C-3. The effect of Lewis acid catalysts on the Diels-Alder reaction involves complexation of the dienophile with consequent lowering of the energy of the dienophile LUMO and acceleration of the reaction. Lewis acid catalysts also affect the isomer distribution6' and it has been suggested that the Diels-Alder reaction involves initial linking of the 'softest' centres.6 ' The suggestion62 that attractive van der Waals (dispersion) forces between methyl groups in the dienophile and unsaturated diene centres may play a role in stabilizing Diels-Alder transition states has been refuted.63 The important interactions are electronic (HOMO-LUMO) and steric (van der Waals re- pulsion between saturated centres) except in reactions of chlorinated dienes [e.g.(75)-(76) + (77) ca. 9 :1I6O in which positive attractive interaction be- tween the C-7 chlorine and the dienophile outweighs the steric factors which would favour (77). c1 c1 + #i cl?i c1 o$ 0 c1 oq0 c1 c1 cl%' (75) 6o K. L. Williamson and Y. L. Hsu J. Amer. Chem. SOC.,1970 92 7385. 61 0. Eisenstein J.-M. Lefour and N. Trong Anh Chem. Comm. 1971 969. 62 Y.Kobuke T. Fueno and J. Furukawa J. Amer. Chem. SOC.,1970 92 6548. 63 K. N. Houk and L. J. Luskus J. Amer. Chem. Sor. 1971,93 4606. Reaction Mechanisms-Part (ii) Rate comparisons for the loss of nitrogen from (78) and (79; n = 1-3) in a retro-homo-Diels-Alder reaction indicate concerted processes are operative for (78)and (79 ;n = 2 or 3) where (79 ;n = 1)decomposes by a diradical mechanism.A concerted mechanism in this case would impose severe strain in the transition state as C-1 C-2 C-3 and C-7 rehybridize to the diene.64 A related double- nitrogen elimination from an intermediate (80) gives the cis-dihydronaphthalene (81).65 An example of the rarely observed homo-Diels-Alder has been reported [(82)+(83)].66 The high ground-state energy and favourable geometry of (82) H a H N CN (83) overcome the normally prohibitively high energy demands of the n2s + 02 + n2 process. A Diels-Alder reaction of ci~-bicyclo[6,l,O]nonatriene provides evidence for the intermediacy of cis,cis,trans,cis-l,3,5,7-cyclononatetraene,since two trans-adducts are obtained [e.g.(84)*(85)p (see electrocyclic section).0 (84) E. L. Allred and A. L. Johnson J. Amer. Chem. SOC.,1971 93 1300. 65 K.-W. Shen Chem. Comm. 1971 391; J. Amer. Chem. SOC.,1971,93 3064. 66 J. E. Baldwin and R. K. Pinschmidt Tetrahedron Letters 1971 935. 67 A. G. Anastassiou and R. C. Griffith J. Amer. Chem. SOC.,1971 93 3083. 158 R.Grigg H H L (86) (87) Cases of borderline6* or non-~oncerted~’ 4 + 2 cycloadditions have been reported. Benzyne adds to cycloheptatriene to give the 2 + 2 adduct (86) presumably by a non-concerted mechanism,70 and not the 6 + 2 adduct as previously th~ught.~’ Photogenerated benzyne is identical in symmetry pro- perties with that prepared by conventional thermal routes and undergoes the same cycl~additions.~~ Chromium pentoxide etherate (87) is not a source of singlet oxygen as previously supposed the observed reactions being caused by traces of chromic acid.73 A study of the ene reaction using I3C-labelled ene components and diazoester enophile~~~ provides evidence for an alternative pathway to the concerted process. Thus (88 ;X = CH) gives (89),indicating a possible free-radical contribu- tion of 30 % (max). The related ene reaction of (88 ;X = N) had a much greater contribution from a non-concerted path (80% max; 40% min). Ene reactions between tetramethylallene and electron-deficient acetylenes are possibly two-step proce~ses.~ The reaction of allylboranes with olefins gives 1,3-dienes probably by an ene reaction [(90)+ (91)].76 * Et0,C-N=N -CO,Et 15*” 85*% PhCH=X-CH2Ph Ph-CH=CH-CH-Ph I (88) N-C02Et HNC0,Et (89) (90) 68 G.Kresze and W. Kosbahn Tetrahedron 1971,27 1931 ;G. Kresze H. Saitner J. Fir1 and W. Kosbahn ibid. p. 1941. 69 I. J. Westerman and C. K. Bradsher J. Org. Chem. 1971 36 969. 70 L. Lombard0 and D. Wege Tetrahedron Letters 1971 3981 ; P. Crews M. Loffgren and D. J. Bertelli ibid. p. 4697. H. Yamada Z. Yoshida and H. Kuroda Tetrahedron Letters 1971 1093. 72 M. Jones and M. R. Decamp J. Org. Chem. 1971,36 1536. 73 J. E. Baldwin J. C. Swallow and H. W.-S. Chan Chem. Comm. 1971 1407. 74 M. M. Shemyakin L. A. Neiman S. V. Zhukova Y. S. Nekrasov T.J. Pehk and E. T. Lippmaa Tetrahedron 1971 27 28 11. 75 H. A. Chia B. E. Kirk and D. R. Taylor Chem. Comm. 1971 1144. l6 B. M. Mikhailov and Y. N. Bubnov Tetrahedron Letters 1971 2127 2153. Reaction Mechanisms-Part (ii) The steric course and regioselectivity in 1,3-dipolar cycloadditions of diazo- acetic esters to cQ-unsaturated esters has been ~tudied.~ Interaction between substituents on the lY3-dipole and the dipolarophile can be attractive (7t-overlap dipole-dipole interaction) or repulsive (van der Waals strain) and interplay of these factors determines the preferred transition state. A perturbational MO treatment of substituent effects on rates of 1,3-dipolar cycl~additions~~ provides a rationale for the observed trends in terms of HOMO-LUMO interactions similar to those described for the Diels-Alder reaction (Figure 1).Numerous reports of the thermal and photochemical generation and trapping of lY3-dipolar species from 3-membered heterocycles have a~peared.'~ Calculations of the tendency to diradical character in open forms of 3-membered rings [(92) -+ (93) or (94)] indicate a maximum tendency for cyclopropane (80%)and a minimum for cyclopropyl anion (8%) with oxiran (38 %) and aziridine (30 %) intermediate in value.80 A + (96) 2-Am-allyl-lithium compounds (95) undergo anionic .4 + .2 1,3-cycloadditions to >C=C< >C=N- -N=N- and -C=C-in preparatively useful yields [e.g.(95)+(96)].*' 4 Sigmatropic Reactions The remarkable methylenecyclopropane rearrangement [e.g.(97)+(98) + (99)] has been investigated82 using optically active (97; R = Me). cis-trans Isomerism 77 P. Eberhard and R. Huisgen Tetrahedron Letters 1971 4337 4343. 78 R. Sustmann Tetrahedron Letters 197 1 27 1 7. 79 R. Huisgen and H. Mader J. Amer. Chem. SOC.,1971 93 1777; H. Hermann R. Huisgen and H. Mader ibid. p. 1779; E. Brunn and R. Huisgen Tetrahedron Letters 1971,473; R. Huisgen and W. Scheer ibid.,p. 481 ; R. Huisgen V. Martin-Rarnos and W. Scheer ibid. p. 477; J. H. Hall and R. Huisgen Chem. Comm. 1971 1187; J. H. Hall R. Huisgen C. H. Ross and W. Scheer ibid. p. 1188; K. Burger and J. Fehn Angew. Chem. Internat. Edn. 1971 10 728 729; J. J. Pommeret and A. Robert Tetrahedron 1971 27 2977; H. Harnberger and R. Huisgen Chem. Comm. 1971 1190; A.Dahrnen H. Hamberger R. Huisgen and V. Markowski ibid. p. 1192; A. Padwa and J. Smolanoff J. Amer. Chem. Soc. 1971 93 548; H. Gotthardt Tetra-hedron Letters 1971 1277; H. Gotthardt and F. Reiter ibid. p. 2749; M. Marky H.-J. Hansen and H. Schmid Helu. Chim. Acta 1971 54 1275; C. S. Angadujavar and M. V. George J. Org. Chem. 197 1 36 1589. E. F. Hayes and A. K. Q. Siu J. Amer. Chem. SOC., 1971,93,2090. T. Kauffmann and R. Eidenschink Angew. Chem. Internat. Edn. 1971 10 739. 82 J. J. Gajewski J. Amer. Chem. Soc. 1971 93 4450. 160 R. Grigg (97) competes with the rearrangement and suggests the reaction proceeds by way of an orthogonal diradical (loo),as first discussed by von Doering and Roth for the rearrangement of optically active Feist's ester (97; R = C0,Me).83 The pro- posed sigmatropic z2s + ,2 process' cannot be ruled out as a parallel pathway resulting in some product.MIND0/2 calculation^^^ predict the most stable singlet structure is the orthogonal diradical. The orthogonal diradical path also receives support from studies on related hydrocarbon^^^ such as (101) and (102). The photochemically allowed methylenecyclopropane rearrangement (1,3-suprafacial sigmatropic shift) is accompanied by fragmentation to alkylidene- carbenes.86 Me Me Me Me Me TAMe R Ar H H The full paper on the proposed antara-antara Cope rearrangement [e.g. (103)+(104)]has appeared.87 Baldwin and Kaplan" present a well reasoned case for interpreting such rearrangements as involving an initial conrotatory opening of the cyclobutene moiety followed by conrotatory closure involving the other double bond and provide an example [(105)-+ (106)] not containing 83 W.von E. Doering and H. D. Roth Tetrahedron 1970 26 2825. 84 M. J. S. Dewar and J. S. Wasson J. Amer. Chem. Soc. 1971 93 3081. 85 M. E. Hendrick J. A. Hardie and M. Jones J. Org. Chem. 1971 36 3061 ; T. B. Patrick E. C. Haynie and W. J. Probst Tetrahedron Letters 1971,423; W. R. Dolbier A. Akiba J. M. Riemann C. A. Harmon M. Bertrand A. Bezagnet and M. Santelli J. Amer. Chem. Soc. 1971,93 3933; M. C. Flowers and A. R. Gibbons J. Chem. Soc. (B) 1971 362; W. R. Roth and Th. Schmidt Tetrahedron Letters 1971 3639. 86 A. S. Kende Z. Goldschmidt and R. F. Smith J. Amer. Chem. Soc.1970 92 7606; J. C. Gilbert and J. R. Butler ibid. p. 7493. T. Miyashi M. Nitta and T. Mukai J. Amer. Chem. Soc. 1971 93 3441. J. E. Baldwin and M. S. Kaplan J. Amer. Chem. Soc. 1971 93 3969. Reaction Mechanisms-Part (ii) [b] D the magical C-1 methoxy-substituent so important in (103). Cope rearrangement of meso-3,4-diphenylhexa- 1,Sdiene gives 37 % of transpans- 1,6-diphenylhexa- 1,5-diene on rearrangement via a boat transition state,89 indicating the conse- quence of steric perturbations on the chair transition state. An example of the sulpho-Cope rearrangement using (107) has been reported,” and a variety of fascinating [3,3]-sigmatropic rearrangements in fused-ring systems continue to be disc~vered.’~ Thus the hydrocarbon hypostrophene (108) undergoes a degenerate Cope rearrangement at 35 “C although at higher temperatures an irreversible diradical rearrangement supervene^.^^ A new general ylide sigmatropic hydrogen shift [(109) -+ (1 101 has been proposed93 and examples provided for (109 ; X = S,Y = 0)94 and (109 ;X = N Y = O).93 These processes occur at room temperature and below.An alternative and probably diradical path giving olefin and X=Y competes when the mole- cular geometry is not favourable for the hydrogen-transfer reaction. A study of the [2,3]-sigmatropic shift occurring when optically active (1 11) is treated with butyl-lithium has established that a supra-supra shift occurs.9s Initial results on the effect of transition state geometry on [2,3]-sigmatropic shifts have been rep~rted.’~Thus the quaternary salts (112; R = CH,CH=CHPh or CH2Ph) form stable ylides (113) owing to the forced orthogonal relationship between 89 R.P. Lutz S. Bernal R. J. Boggio R. 0. Harris and M. W. McNicholas J. Amer. Chem. SOC.,1971,93 3985. 90 J. F. King and D. R. K. Harding Chern. Cornm. 1971 959. 91 E. Vedejs Chem. Cornm. 1971 536; J. W. Hanifin and E. Cohen J. Org. Chem. 1971 36 910; J. P. Synder L. Lee and D. G. Farnum J. Amer. Chem. SOC.,1971,93 3816. 92 J. S. McKennis L. Brener J. S. Ward and R. Pettit J. Amer. Chem. SOC.,1971 93 4957. 93 J. E. Baldwin A. K. Bhatnagar S. C. Choi and T. J. Shortridge J. Amer. Chem. SOC. 1971,93,4082. 94 A. Kondo and A. Negishi Tetrahedron 1971 27 4821; J. E. Baldwin G. Hofle and S.C. Choi J. Arner. Chern. SOC.,1971 93 2810. 95 J. E. Baldwin and J. E. Patrick J. Amer. Chem. SOC.,1971 93 3557. 96 S. Mageswaran W. D. Ollis I. 0. Sutherland and Y. Thebtaranonth Chem. Comm. 1971 1494. 162 R. Grigg (1 11) (1 12) (1 13) the N-R bond and the enolate mystem. In contrast the more-flexible azabi- cyclo[3,3,l]nonane system (114; R = CH,CH=CHPh) in which the N-R bond and the enolate mystem are at an angle of ca. 30° does undergo a [2,3]- shift giving (115). However the Stevens rearrangement (radical pair mechanism) of (114; R = CH,Ph) does not occur. An anionic version [(116) *(117)] of a (1 16) (117) [2,3]-sigmatropic shift has been reported and occurs at low temperature in good yield. ' The rearrangement (118) -+ (119) has been studied by deuterium labelling and occurs by intramolecular Diels-Alder reaction giving (120) followed by a U" (119) &+HH&+ (120) (121) 97 J.F. Biellmann and J. B. Ducep Tetrahedron Letters 1971 33; V. Rautenstrauch Helv. Chim. Acta 1971 54 739. Reaction Mechanisms-Part (ii) Me hv 7 Me*Ph double [1,5]-hydrogen shift [(120) -+(121) -+ (1 19)].98 Photochemical [1,3]- suprafacial shifts with retention of configuration at the migrating centre have been reported for several systems [e.g. (122)-+ (123)].99 Surprisingly the thermal reversal (123) *(122) occurs with 90 % retention of configuration at the migrating centre. This apparent violation of orbital symmetry is attributed to the very unsymmetrical nature of the dicyanoallyl framework across which the a-phenyl- ethyl group migrates.The node in the HOMO of the x framework no longer passes through C-2. This observation invoked a cation against the uncritical application of orbital symmetry rules to highly perturbed systems. 5 Cheletropic Reactions In contrast to the ready decarbonylation of benzonorbornadienones the hydra- zone (124) is remarkably thermally stable.'00 Failure to achieve cheletropic elimination of the isocyanide is thought to be due to the steric effect of the N-benzoxazoline substituent retarding the attainment of a linear geometry by N-R (124) R = I the bridging C-N-N- array of atoms. In contrast what appears to be a cheletropic isocyanide addition involving a 1,3-dipolar species [(125) -+ (126)] has been reported."' On treating dihydrothiophenium salts [e.g.(127)] with 98 J. A. Berson R. R. Boettcher and J. J. Vollmer J. Amer. Chem. Soc. 1971 93 1540. " R. C. Cookson and J. E. Kemp Chem. Comm. 1971 385; R. C. Cookson J. Hudec and M. Sharma ibid. pp. 107 108. loo R. S. Atkinson A. J. Clark and R. E. Overill Chem. Comm. 1971 535. lo' J. A. Deyrup Tetrahedron Letters 1971 2191. 164 R. Grigg butyl-lithium proton abstraction competes with S-alkylation. The S-alkylated products [e.g. (128)] undergo a cheletropic reaction furnishing methyl butyl sulphide and the hexadienes (129). Cheletropic elimination of SO from the corresponding dihydrothiophen sulphones generates only the trans,trans-2,4- hexadiene. The large amount of cis&-diene (33.5%) generated from (128) arises from steric interactions of the ring methyl groups with the S-alkyl groups in the transition state which adversely affect the disrotatory mode leading to the trans,trans-diene (58.5% produced).lo2 6 Homogeneous Catalysis and Pericyclic Processes A number of possible functions of metal ions in catalysing pericyclic processes merit consideration (a) the metal ion could convert symmetry-'forbidden' processes into symmetry-allowed concerted processes by interaction of the MO systems of substrate(s) and metal ion (b) the symmetry-'forbidden' process could occur stepwise via labile o-bonded organometallic intermediates or (c) both allowed or 'forbidden' [operating by either (a) or (b)]reactions (especially multicomponent processes) could be facilitated by prior co-ordination to a metal ion reducing unfavourable entropy factors.Numerous theoretical treatments of (a)have appeared. '03-lo6 More particu- larly the major focus of attention has been the metal-catalysed disproportiona- tion of olefins [(130) (131)]. The results of the theoretical treatments while R'CH CHR~ R'CH=CHR~ I1 II s R'CH CHR~ R~CH=CHR~ agreeing that 'forbidden'-to-allowed catalysis can occur differ in the way in which this is visualized to occur. These differences are still the subject of conten- tionlo4 and result from different choices of theoretical models. One scheme"' considers the reaction to occur via a cyclobutane and utilizes a pair of orthogonal d-orbitals one filled and one empty to transfer a pair of electrons from a substrate orbital which is antibonding at the site of reaction to one which is bonding in the region of reaction.An alternative treatment'06 suggests simultaneous rupture of B. M. Trost and S. D. Ziman J. Amer. Chem. SOC.,1971 93 3826. lo3 F. D. Mango Adv. Catalysis 1969 20 291; G. L. Caldow and R. A. MacGregor J. Chem. SOC.(A) 1971 1654. lo4 W. Th. A. M. van der Lugt Tetrahedron Letters 1970 2281. lo5 F. D. Mango Tetrahedron Letters 1971 505. lo6 W. B. Hughes J. Amer. Chem. SOC.,1970 92 532; G. S. Lewandos and R. Pettit Tetrahedron Letters 1971 789. Reaction Mechanisms-Part (ii) 7~-and o-bonds occurs and considers that an intermediate cyclobutane is not involved. Evidence supporting this contention is presented for a heterogeneous catalytic system which shows little tendency to convert cyclobutane to ethylene.One area attracting much experimental attention is the metal-catalysed re- organization of strained-ring systems by formally 'forbidden' reactions such as the Ag'-catalysed rearrangements of cubane and related derivatives [( 132) and (133)]lo7 to systems containing cyclopropane rings e.g. (133) +(134).'08 I C0,Me C@,Me R' Metal catalysis [Ag' Rh' Ir' Pd" or Ru"] of theformal u2a+ u2aconversion of bicyclo[l,l,O]butanes to butadienes [e.g. (135)+ (136)]'09 has been the subject of numerous studies. 'lo Related Rh'-catalysed rearrangements of olefinic epoxides and bicyclo[6,1,O]nonatrienes [e.g. (137) -+ (138)] have also been reported.'" The initial theories suggested to account for these extensive bond relocations reflect the ingenuity of the chemists involved but Katz and Cerefice' l2 provided (137) lo' W.G. Dauben C. H. Schallhorn and D. L. Whalen J. Amer. Chem. SOC.,1971,93 1446; L. A. Paquette R. S. Beckley and T. McCreadie Tetrahedron Letters 1971 775; L. A. Paquette and J. C. Stowell J. Amer. Chem. SOC.,1971 93 2459. H. H. Westberg and H. Ona Chem. Comm. 1971,248. M. Sakai H. Yamaguchi H. H. Westberg and S. Masamune J. Amer. Chem. SOC. 1971 93 1043; L. A. Paquette S. E. Wilson and R. P. Henzel ibid. p. 1288. lo P. G. Gassman and F. J. Williams J. Amer. Chem. Soc. 1970,92,7631 ; P. G. Gassman T. J. Atkins and F. J. Williams ibid. 197 1,93 18 12 ; P. G. Gassman and F.J. Williams Tetrahedron Letters 1971 1409; P. G. Gassman and T. J. Atkins J. Amer. Chem. SOC. 1971 93 1042; L. A. Paquette R. P. Henzel and S. E. Wilson ibid. p. 2335; L. A. Paquette Accounts Chem. Res. 1971 4 280. R. Grigg and G. Shelton Chem. Comm. 1971 1247; R. Grigg R. Hayes and A. Sweeney ibid. p. 1248. 'I2 T. J. Katz and S. A. Cerefice J. Amer. Chem. SOC. 1971 93 1049. 166 R. Grigg kinetic evidence for a stepwise process and some organorhodium complexes have been isolated in the isomerization of quadricyclane to norbornadiene. Maitlis Childs and Kaiser' l4 suggested these catalysts were functioning as weak Lewis acids and provided many examples of such catalysts which effected transformations of tri-t-butylprismane (139).Even trinitrobenzene was effective in this case! Other workers'" subsequently concurred with the Lewis acid interpretation although the variqtion of product distribution with metal catalyst still calls for further study and the role of oxidative-addition processes in these rearrangements remains to be clarified. 'l6 'I3 L. Cassar and J. Halpern Chem. Comm. 1970 1082. K. L. Kaiser R. F. Childs and P. M. Maitlis J. Amer. Chem. SOC.,1971 93 1270. P. G. Gassman and T. J. Atkins J. Amer. Chem. Soc. 1971 93 4597; M. Sakai and S. Masamune ibid. p. 4610; M. Sakai H. H. Westberg H. Yamaguchi and S. Masamune ibid. p. 4613. 'I6 J. E. Byrd L. Cassar P. E. Eaton and J. Halpern Chem. Comm. 1971 40.