3 Reaction Mechanisms Part ( ii) By N. S. ISAACS Chemistry Department The University Reading Carbonium Ions and Solvolytic Reactions.-New aromatic carbonium ions reported include the cyclopropenium ion (1 a) stabilised by alkyl substitution (Me > Pr" > Pr' > Bu').' The trimethylcyclopropenium ion (lb) is one of the most stable hydrocarbon cations known.2 The tetraphenylcyclobutadiene dication (21 stable in highly acidic media has substantial charge delocalisation as shown from I3C n.m.r. measurements which also confirm its ~ymrnetry.~ .928 0.928 RAR (1) (a) R = H (b) R = Me Two bridged bishomotropylium ions (3) and (4) have been reported the presence of a diamagnetic ring-current being evident from the 'H n.m.r. A bridged [lllannulene cation (5) a 10n aromatic system has also been pre- pared.7 The perchlorotrityl cation is extremely stable,* as also are the corre- sponding radical and anion ;chirality in a number of difluoromethyl-substituted ' R.Breslow and J. T. Groves J. Amer. Chem. SOC.,1970,92 984 988. ' J. Ciabattoni and E. C. Nathan Tetrahedron Letters 1969,4997. G. A. Olah and G. D. Mateescu J. Amer. Chem. SOC.,1970,92 1430. ' G. Schroder U. Prange N. S. Bowman and J. F. M. Oth Tetrahedron Letters 1970 3251. M. Roberts H. Hamberger and S. Winstein J. Amer. Chem. Soc. 1970,92 6347. P. Ahlberg D. L. Harris and S. Winstein J. Amer. Chem. SOC.,1970,92 2146. ' E. Vogel R. Feldmann and H. Duwel Tetrahedron Letters 1970 1941. M. Ballester J. Rieva-Figueros and A.Rodriguez-Siurana Tetrahedron Letters 1970 3615. 102 N. S. Isaacs trityl cations has been demonstrated by the temperature-dependence of the 19F n.m.r. ~pectra.~ The doublet spectrum of (6)at room temperature becomes an octet at -71 "Cas rotation is quenched AFS = 48.5 kJmol-' for racemiza- tion. Heats of protonation of 1,l-diarylethylenes show a dependence on CT' (p = 4.59) as do those of benzophenones (p = 6-75) both indicating the import- ance of conjugation in the stabilisation of the cations (7)." The benzenonium ion (8) which at -80 "Cis characterised by a single n.m.r. line at t 1.91 undergoes a degenerate rearrangement which is arrested below -110"C where a more complex spectrum appears.' Protonation of the aceheptylene system (9) occurs under kinetic control at -15 "C at position 2 but at positions 4 and 6 at room temperature under thermodynamic control.l2 The results require a crossing of the two potential-energy pathways predicted as allowed for non- alternant (but not for alternant) hydrocarbons.' The protonation of 9-ethyl- 10-methylanthracene forms almost equal proportions of 9-H and 1 O-H isomers indicating the lack of importance of the Baker-Nathan effect in stabilising the ions.14 A considerable Baker-Nathan effect is reported to be operating in the protonation of P-diketones (lo).' The isomeric 1- and 4-phenylbenzenonium ions are in equilibrium (20:80 % respectively at 0 "C)in superacid solution. l6 J. W. Rakshys S. V. McKenly and H. H. Freedman J. Amer.Chem. Soc. 1970 92 3518. lo E. M. Arnett J. V. Carter and P. P. Quirk J. Amer. Chem. Soc. 1970 92 1771. 'I G. A. Olah R. H. Schlosberg D. P. Kelly and G. D. Mateescu J. Amer. Chem. Soc. 1970,92 2546. lZ E. Hasselbach Tetrahedron Letters 1970 1543. l3 R. D. Brown Quart. Rev. 1952 6 63. l4 D. M. Brouwer and J. A. Van Doorn Rec. Trav. chim. 1970 89 88. J. W. Larsen J. Amer. Chem. Soc. 1970 92 5136. l6 V. A. Koptyug and L. Mozulenko Zhur. obshchei. Khim. 1970,40 102. Reaction Mechanisms-Part (ii) Complex rearrangements of the bicyclo-[2,2,2]- and -[3,2,1]-octyl cations in SbF,-HS0,F lead ultimately to the bicyclo[3,3,0]oct-l-yl cation (1 l) the most stable isomer." The isomeric pentamethylcyclopentenyl cations (12) and (13) OH OH R vR ?-+R R H+ Me Me +-+ -H Me H Me H MeH (12) (13) have been separately observed and their degenerate rearrangements studied by n.m.r.spectroscopy.' Activation parameters for rotation about the partial double bonds in ally1 cations have been measured (Scheme 1) and shown not to be due to prototropy :19 the substituted cyclopropylmethyl cation (14) has the E 73 100 kJ mol-' 65.8 logA 11.8 14.0 Scheme 1 conformation with two non-equivalent methyl groups. The rates of reaction of a series of carbonium ions with carbon monoxide (R+ + CO+ RCO') have been proposed as measures of carbonium ion stability ;while measurements have been carried out at various temperatures the rates span a range of 108 so should be highly sensitive to the stability of the ion R+.20 Cyclic halonium ions are well-established species which can exist as stable entities in superacid media ; the methylene derivatives (1 5) have recently been characterised,2' and a series of dimethylhalonium hexafluoroantimonates (1 6) have been isolated G.A. Olah S. M. Bollinger and D. P. Kelly J. Amer. Chem. Suc. 1970 92 1432. D. M. Brouwer and J. A. Van Doorn Rec. Trav. chim. 1970,89 333. l9 N. C. Deno R. C. Haddon and E. N. Nowak J. Amer. Chem. SOC. 1970,92,6691. 'O H. Hogeveen and C. J. Gaasbeek Rec. Trav. chim. 1970,89 395. J. M. Bollinger J. M. Brinich and G. A. Olah J. Amer. Chem. Suc. 1970,92 4025. 104 N. S. Isaacs as crystalline compounds stable at room temperat~re.~~?~~ 3C n.m.r. indicates increasing positive charge on carbon for the series (16 ha1 = C1 > Br > I); even such weak nucleophiles as ethers displace halide from these compounds.+ Me,O Hb &Me Cf\CH2 Me-hal-Me -+ Me,O+ + Mehal Me H (16) X = C1 Br I (141 (15) Evidence for protonated cyclopropane intermediates has come from several studies ; the rearrangements of many butyl pentyl and hexyl cations in tritiated superacid demonstrate the exchange of a single proton with solvent (Scheme 2).24 Complete carbon scrambling occurs in the rearrangement of the Scheme2 1-propyl cation via (17) in the presence of AlC1 and ZnClz.25-27 Several new degenerate rearrangements of cations have received intensive study. The bicyclo- nonatrienyl cation (18) does not behave as if especially destabilised’* (although classed as anti-bicycloaromatic by Goldstein”) but can pass by a facile route to the 9-barbaralyl cation (19).This species which has been observed directly exhibits a single n.m.r. resonance even at -125 “C due to its rapid degenerate 22 G. A. Olah and J. R. DeMember J. Amer. Chem. SOC.,1970,92,2562. 23 G. A. Olah and J. R. DeMember J. Amer. Chem. SOC.,1970,92,718. 24 G. M. Kramer J. Amer. Chem. SOC.,1970,92,4344. 25 C. C. Lee and D. J. Woodcock J. Amer. Chem. Soc. 1970,92 5991. 26 G. J. Karabatsos C. Ziondron and S. Meyerson J. Amer. Chem. SOC. 1970 92 5996. 21 G. J. Karabatsos C. E. Orzech J. L. Fry and S. Meyerson J.Amer. Chem. SOC.,1970 92 606 614 621. 28 J. B. Grutzner and S. Winstein J. Amer.Chem. SOC.,1970 92 3186. 29 M. J. Goldstein J. Amer. Chem. SOC.,1967 89 6357. Reaction Mechanisms-Part (ii) I05 rearrangement through a series of Wagner-Meerwein shifts and Cope rearrange-ment~.~*-~~ The analogous rearrangements in the 9-methylbarbaralyl cation are somewhat slower although they have an activation energy of only 46kJ mol- 1.34 The dissolution of bicyclo[2,2,l]heptane in HF-SbF initially / H H H results in the formation of the methylcyclohexyl cation (20),which in turn forms the 1,2- and 1,3-dimethylcyclopentyl cations each of which undergoes degenerate rearrangements oia 1,2 hydride and methide migrations to form (21).35 Phenyl cations are believed to result from the decomposition of phenyldiazonium ions in dipolar aprotic solvents ; in the presence of aromatic substrates arylation occurs typical of an electrophilic reaction and no deuterium isotope effect is sho~n.~~-~~ 3.8 30 J.C. Barbarak J. Daub D. M. Follweiler and P. von R. Schleyer J. Amer. Chem. SOC. 1969,91 7760. ” J. C. Barbarak and P. von R. Schleyer J. Amer. Chem. SOC.,1970,92 3184. 32 J. B. Grutzner and S. Winstein J. Amer. Chem. SOC.,1970 92 3186. ” P. Ahlberg D. L. Harris and S. Winstein J. Amer. Chem. Soc. 1970,92,4454. 34 P. Ahlberg J. B. Grutzner D. L. Harris and S. Winstein J. Amer. Chem. SOC.,1970 92 3478. ’’ H. Hogeveen and C. J. Gaasbeek Rec. Trau. chim. 1969,88 1305. 36 M. Kabayashi M. Minato E. Yamada and N. Kabori Bull. Chem. SOC.Japan 1970 43 215 219. ” K. Ishida N. Kabori M.Kabayashi and H. Minato Bull. Chem. SOC.Japan 1970 43 285. ” N. Kabori M. Kabayashi and M. Minato Bull. Chem. SOC.Japan 1970,43 223. Recent molecular orbital calculations by the method of Pople indicate that alkyl cations prefer the staggered conformation (22a) to the eclipsed (22b);39 however the opposite is predicted by Clark and Lilley4* for 1-and 2-fluoroethyl cations the more stable conformer in either case being that with hydrogen eclipsed to fluorine (23). Solvolytic Reactions.-The rate-enhancing effect of an a-methyl group has been proposed as a diagnostic test for the limiting behaviour of a substrate in sol- v01ysis.~~The value of the solvolytic rate constant for a secondary substrate is typically times that of the corresponding a-methyl (tertiary) compound.This is not a maximum value but is reduced by rearside solvent interactions. The maximum effect (k,,condary/ktertiary ca. lo-*) is realised in the 2-adamantyl = system (24) in which no rearside approach by solvent is possible.42 The depar- ture from this value can be taken as a measure of the solvation of the transition state. 2-Adamantyl bromide solvolysis has been used as a means of assessing the extent of solvent participation in the ionisation process to produce a solva- tion parameter analogous to Y. The new values correlate well with Y substan-tiating the essentially limiting behaviour of t-butyl chloride on whose solvolyses the Y scale is based. Solvolytic rates of a series of substituted polycyclic aryl- methyl compounds have been measured.43 Plots of relative rates (variation of substituent) versus (a)relative rates in different solvents of benzyl chloride and (b) relative rates with different leaving-groups show slopes which vary more widely in the latter case.It is concluded that the leaving-group has a greater effect upon rate than does the solvent. The rate of solvolysis of trityl chloride (ca. lo31 mol-s-' in aqueous acetonitrile) is subject to special salt effects by 39 L. Radom J. Pople V. Buss and P. von R. Schleyer J. Amer. Chem. SOC.,1970 92 6380. 40 D. T. Clark and D. M. J. Lilley Chem. Comm. 1970,603. 41 J. L. Fry J. M. Harris R. C. Bingham and P. von R. Schleyer J. Amer. Chem. SOC. 1970,92 2540. 42 D. J. Raber R. C. Bingham J. M. Harris J.L. Fry and P. von R. Schleyer J. Amer. Chem. SOC.,1970,92,5977. 43 M. D. Bentley and M. J. S. Dewar J. Amer. Chem. SOC.,1970,92 3991. Reaction Mechanisms-Part (ii) 107 perchlorate due to ion-pair exchange.44 In aqueous ether the rate of hydrolysis is accelerated 7 x 109-fold by added lithium per~hlorate.~' Charge-transfer excitation energies for a series of substituted benzenes with tetracyanoethylene are used as models to describe the stabilisation of the corresponding benzylic cations by substituent groups (with whose solvolytic rates they ~orrelate).~~ Pressure effects on solvolytic rates of t-butyl chloride have been further in- vestigated :47 the activation volume -2 cm3 mol- * in water at 0 "C becomes progressively more negative as the proportion of ethanol in the solvent is in-creased due to increasingly great nucleophilic interactions between solvent and transition state.This behaviour contrasts with that of a number of unimolecular decompositions to neutral products for which AVt = +9-16 cm3 mol- '. tt-Participation.-2-(Cycloheptatrienyl)ethyl esters (25)solvolyse some 16 times more rapidly than cyclopentylethyl analogues (26) regarded as suitable model compounds incapable of n-participation k25/k26 = 16. Some remote parti- cipation is inferred but less than in the case of 3-~yclopentenylethy1(27) for which kunsaturated/ksaturaled = 95. The products (28) however are those from rearranged TosO OTos CH,CH ,OTo s (26) +OH carbonium ions.t8 Little if any participation by phenyl occurs in the rate- determining ionisation of (29) as its rate is slightly less than the syn isomer (30).However phenyl migration occurs presumably subsequent to i~nisation.~~ Some (Ar-5)-participation is claimed to occur in the solvolysis of (31) since abnormally high rates were obtained for X = Me or OMe.50 Cyclopropane migration occurs in solvolysis of (32) probably assisting the ionisation which occurs at a rapid rate for a primary compound;" the participation of the cyclo- propane ring is stereospecifi~'~ in the ring-expansion reactions (33) -+ (34) and 44 K. T. Leffek Canad. J. Chem. 1970,48 1. 45 Y. Pocker and R. F. Buchholz J. Amer. Chem. SOC.,1970,92 2075. 46 W. Hanstein H. J. Berwin and T. G. Traylor J. Amer.Chem. Sac. 1970 92 829. 47 B. T. Baliga and E. Whalley Canad. J. Chern. 1970 48 528. 48 G. D. Sargent and T. E. McLoughlin Tetrahedron Letters 1970 4359. 49 J. W. Wilt and T. P. Malley J. Amer. Chem. Sac. 1970 92 4747. 5" R. J. Ouellette R. Papa M. Attea and C. Levin J. Amer. Chem. Sac. 1970 92 4893. 51 Y. E. Rhodes and T. Takino J. Amer. Chem. Sac. 1970,92 5270. 52 C. D. Poulter and S. Winstein J. Amer. Chem. SOC.,1970,92 4282. 108 N. S. Isaacs TosO (30) (35) -+(36) (the major product in either case being ~nrearranged).~~ Calcula-tions show that the substituted cyclopropylmethyl cation prefers the conforma- tion (37)rather than (38) and this has been confirmed by n.m.r. studies ;rotation Hg+ AcOH (p -CoAc X X (31) (32) 35% 40% 21 % about the central bond has a surprisingly high activation barrier E = 57.6 kJ mol-This explains the lack of cyclopropyl participation in the rigid bB (33) H (34) OH OPNB 53 C.D. Poulter E. C. Friedrich and S. Winstein J. Amer. Chem. SOC.,1970,92 4274. D. S. Kabakoff and E. Namanworth J. Amer. Chem. SOC., 1970,92 3234. Reaction Mechanisms-Part (ii) system (39),’ whereas in the open 2-cyclopropylethyl esters (40) participation by the three-membered ring is indicated by the large contribution to the rate due 4” H C+-Me / Me4 Me Me (37) (38) to methyl substitution which should make the ring more nucleophilic but could have no purely inductive effect.’ Very marked cyclopropyl acceleration is found in the norbornadiene analogue (41) 1010-10’2 times more reactive than norbornyl p-nitrobenzoate and which leads to a degenerate ion (42) ;57 likewise in ,OPNB OPNB -P (43).’ High-pressure kinetic studies have shown that activation volumes for soholytic reactions of -15 to -20 cm3 mol- ’ are typical for limiting solvolyses probably associated with solvent electrostriction.Much less negative values typify reactions which involve neighbouring groups ;” solvolysisof 2-b-hydroxy-pheny1)ethyl chloride has V = -1 cm3 mol- ’ which may prove a useful diagnostic test. CI-and fi-secondary deuterium isotope effects in solvolysis of 3-phenylbut-2-yl tosylate (44) are not quite equal (1.104 1.097) from which it ’’ B. R. Ree and J. C. Martin J. Amer. Chem.Soc. 1970,92 1660. 56 M. J. S. Dewar and J. M. Harris J. Amer. Chem. SOC.,1970 92 6557. 57 ’’ R. M. Coates and J. L. Kirkpatrick J. Amer. Chem. SOC.,1970 92,4883. M. A.Battiste J. Haywood-Farmer H. Malkins P. Seidl and S. Winstein J. Amer. Chem. Soc. 1970,92,2145. 59 W. J. LeNoble and B. Gabrielson Tetrahedron Letters 1970 45. 110 N. S. Isaacs appears that the transition state for solvolysis is slightly unsymmetrical (though a symmetrical phenonium ion may later form).6o Two mechanistic pathways are inferred in the acetolysis of this compound which could also explain the inequality of the isotope effects. The total rate has been partitioned into assisted (Fk,) and solvent-displacement (&,) components the former giving only reten- tion of configuration.61 The component k is independent of substituents in the aromatic ring while Fk increases with the conjugative ability of the ring.62,63 Ph k 1 YCSH CH,-CH-CH-CH, I OTos (44) Co-ordination of a metal to a n-system can reduce its ability to participate in a solvolytic reaction e.g.compare (45) p = -0.78 and the non-metallated Cr(W3 (45) analogue p = -2.35.64 Iron co-ordination confers retention of configuration on the solvolysis of (46).65 Alicyclic Systems.-Solvolyses of 2-adamantyl compounds (24a) promise to be useful as a mechanistic probe. Rates are moderately fast but the rigid skeleton excludes nucleophilic attack and solvation from the rear side of the reaction centre. The 2-adamantyl system has been proposed as a model having limiting behaviour since it reacts by an unassisted pathway (&,).66 Since this pathway should be independent of solvent the ratio k, [where x refers to a second (ester) substrate] is a measure of the difference in unassisted solvolysis rates 6U S.L. Loukas M. R. Velkon and G. A. Gregoriou Chem. Comm. 1970,251. 61 S. L. Loukas M. R. Velkon and G. A. Gregoriou Chem. Comm. 1969 1199. 62 H. C. Brown C. J. Kim C. J. Lancelot and P. von R. Schleyer J. Amer. Chem. SOC. 1970,92 5244. 63 M. D. Bentley and M. J. S. Dewar J. Amer. Chem. SOC.,1970,92 3996. 64 R. S. Bly R. C. Strickland R. T. Swindell and R. L. Veazey J. Amer. Chem. SOC. 1970 92 3722. 65 N. A. Clinton and C. P. Lillya J. Amer. Chem. SOC.,1970 92 3065.66 P. von R. Schleyer J. L. Fry L. K. M. Lam and C. J. Lancelot J. Amer. Chem. SOC. 1970,92 2542. Reaction Mechanisms-Part (ii) 111 between the second substrate and the 2-adamantyl ester. Thus the value of the rate ratio k (2-propy1):k (2-adamantyl) increases from 10-2'25 in trifluoroacetic acid to lO+ 3'0 in ethanol as nucleophilic participation becomes increasingly important for the 2-propyl substrate. The former value is taken as a provisional limit to the ratio and therefore the (minimum) amount of solvent assistance to the reaction is the difference between 10-2.2sand the appropriate rate ratio. This therefore amounts to 105'5 for ethanolysis of isopropyl tosylate. This approach is substantiated by an examination of the effects of added azide ion on the reaction :67 several related studies in this series have been rep~rted.~*-~' Limiting conditions apparently prevail in the solvolysis of cyclopentyl bromide by aqueous ethanol to judge from the typical isotope effects observed; kH:kDtcis-z,= 1.1533 kH:kDct,ans-2) = 1.1803 kH:kD,, = ~1869.~'Even larger values of isotope effects are found in the solvolysis of 2,4-dimethyl-3-pentyl brosylate (47) suggesting that some H-participation is This would OBros (47) accord with the products even the 3-pentyl (48) system yields a small amount of 2-pentyl products on ~olvolysis.~~ Extensive hydride shifts in the bicyclo- H + -+MeCHCH,CH,Me Me AH/CH2\ dr Me (1 %) (48) '' J.M. Harris D. J. Raber R. E. Hall and P. von R.Schleyer J. Amer. Chem. Soc. 1970 92 5729. 68 J. L. Fry C. J. Lancelot L. K. M. Lam J. M. Harris R. C. Bingham D. J. Raber R. E. Hail and P. von R. Schleyer J. Amer. Chem. SOC.,1970,92,2538. '' J. A. Bone and M. C. Whiting Chem. Comm. 1970 115. 'O L. Baiocchi M. Gianangeli and G.Palazzo Tetrahedron Letters 1970,5025. 71 J. 0. Stoffer and J. C. Christen J. Amer. Chem. SOC.,1970 92 3190. '' V. J. Shiner and J. 0.Stoffer J. Amer. Chem. SOC.,1970,92 3191. 73 H. R. Hudson and D. Ragoonanan J. Chem. SOC.(B),1970 1755. 112 N. S. Isaacs [3,3,2]decyl cation (49) originate from a remote carbon atom.74 A remote double bond in the cyclo-octane series (50) causes a rate retardation in solvolysis com- pared to the saturated analogue (51).75 Presumably n-participation is absent or OoBrOs OOBros OOBros OOBros OBros OBros (51)kre 1700 (50)39 (53)cis 1 cis (52)0.058 trans 4.1 trans 0.65 if it occurs is less effective than transannular hydrogen participation.The effect is somewhat smaller in the diesters (52)and (53). Isomeric cations from the solvo-lysis of cis-and trans-9-decalyl chlorides are known and differ in energy by 6 kJ mol-' similarly isomeric hydrindyl cations (54) have been identified &-+ + o._l"' H H (54) though differing in energy by only 17 kJ mol-1.76 A case of alkyl participation in the steroid series has been investigated i.e.(55)-(56).77Remote n-participa- XCH OH XCHz (55) (56) tion in bicyclic compounds is a potentially useful method for the synthesis of complex ring structures e.g.(57)+(58).78 74 M. P. Doyle and W. Parker Chem. Comm. 1970 755. 7s W. D. Closson J. L. Jernow and D. Grey Tetrahedron Letters 1970 1141. 76 K. Becker A. F. Boschung and C. A. Grob Tetrahedron Letters 1970 3831. l7 J. G. L. Jones and B. A. Marples Chem. Comm. 1970 126. D. J. Raber G.J. Kane and P. von R. Schleyer Tetrahedron Letters 1970 41 17. Reaction Mechanisms-Part (ii) Aliphatic Systems-Secondary deuterium isotope effects have been measured for solvolyses of the allylic systems (59) and The y-methyl effect is normal indicating that positive charge is accumulating at the y-carbon. The (60) 1.132 inverse effect for P-methyl substitution indicates that this is not the case at the P-carbon. This is therefore taken to indicate the absence of 1,3-interaction in the ally1 cation which would require a contribution from the structure (61).The preferential migration of CH rather than CD in the pinacolic rearrangement of (62) (k,:k = 1.15-1.20) is indicative of positive charge accumulating at the ,CH3 Ph,C-C-CD I OH AH (62) methyl group in the transition state and therefore of methyl migration occurring in the rate-controlling step." Rates and isotope effects on the solvolyses of ethyl (63) and n-propyl trifluoromethanesulphonates (triflates) have been measured in trifluoroethanol and trifluoroacetic acid conditions favouring the SN1 process as far as possible." Both a-and P-isotope effects [la06 and 1.05 Rates in trifluoroacetic CH3CH20Tf 1.37 x lOP5s-' acid at 50 "c CD3CH20Tf 1.19 x 10-5s-1 (63) per deuterium respectively for (63)] are less than a third the full values expected for a limiting solvolysis and indicate that even under these conditions primary substrates react by a largely bimolecular route (ksj.Solvolyses of a-arylethyl acetates in 30% ethanol evidently occur by an S,1 process since the relative log rates correlate with Q+ (p = -5.7).*' A mechanistic change is inferred for solvolyses of some benzylic sulphonate~.~~ The charge on the benzyl carbon in the appropriate 79 R. H. Griffin and J. G. Jewett J. Amer. Chem. SOC.,1970,92 1104. W. M. Schubert and P. H. LeFevre J. Amer. Chem. SOC., 1970,92,7746. G. A. Dafforn and A. Streitweiser Tetrahedron Letters 1970 3159." E. A. Hill M. L. Gross M. Staseiwicz and M. Manion J. Amer. Chem. SOC.,1969 91 7381. 83 A. Streitweiser H. A. Hammond R. H. Jagow R. M. Williams R. G. Jesaitis J. Chong and R. Wolf J. Amer. Chem. SOC.,1970 92 5141. 6' from the break (or curve) in the Hammett plot against 114 N. S. Isaacs cation calculated by the CNDO method is found to correlate with the observed rates. The rate of solvolysis of di-t-butylcarbinyl chloride (64) would be predicted on steric grounds to be some lo4 times greater than that for isopropyl chloride (65). In fact it is somewhat less reactive although the factors responsible for this discrepancy have not been identified with certainty.84 Intermolecular hydride (65) 94-5 kJ mol-' transfer has been shown to occur between a transient secondary cation and a tertiary hydrogen (Scheme 3).85 The rearrangement of the 2-adamantyl cation Scheme 3 (24a) to the 3-adamantyl(66) is presumed to occur by an intermolecular hydride transfer since the rate is reduced by dilution.86 The generation of carbonium H ions in the absence of nucleophilic species can be achieved by the action of nitrosonium tetrafluoroborate on an alkyl azide (67) :87 Ph2CHN3 + NO+BF,-+Ph,CH+BF,-+N2 +NzO (67) Nucleophilic displacements of p-nitrocumyl and a,p-dinitrocumyl chlorides occur by a radical chain process which is more favourable than the carbonium ion mechanism for these deactivated systems.88 A consequence of this radical- initiated reaction is that the addition of a weak nucleophile may be achieved 84 S.H. Liggero J. J. Harper P. von R. Schleyer A. P. Krapcho and D. E. Horn J. Amer. Chem. SOC.,1970.92 3791. 85 D. N. Kirsonov V. N. Setkina V. A. Kurichev R. V. Kudryavtsev and Yu. Lyakhavek- skii Izvest. Akud. Nuuk. S.S.S.R. Ser. khim 1969 2344. 86 P. von R. Schleyer L. K. M. Lam D. J. Raber J. L. Fry M. A. McKervey J. R. Alford B. D. Cuddy V. G. Keizer H. W. Geluk and J. L. M. A. Schlatmann J. Amer. Chem. SOC.,1970 92 5247. M. P. Doyle and W. Weirenga J. Amer. Chem. SOC.,1970 92 4999. ** N. Kornblum R. T. Swiger G. W. Earl H. W. Pinnick and F. W. Stuckal J. Amer. Chem. SOC.,1970,92 5513. Reaction Mechanisms-Part (ii) in the presence of a catalytic amount of a strong initiating nucleophile. Thus azide ion does not react with the dinitrocumyl chloride alone but in the presence of a trace of the 2-nitroisopropyl anion 97 % of the cumyl azide (68) is formed.Me CH,NO Me CH,NO NO Hydrolysis of the diazoketone (69) is general-acid-catalysed and proceeds via a carbonium iong9 H H I I CH3COC-R 5CH3-CO-C-R -+ CH,CO$-R II I N2 N2 + I products (69) Norbornyl Systems.-The ‘H n.m.r. spectrum of the 2-methyl-7-norbornenyl cation (70) has been cited as evidence of the non-classical formulation of this Me Me 7.58 (70) and by analogy the parent carbonium ion (71).” The 3-H resonance is identified at 7 3.53 which is close to that found in the stable norbornenyl cation (71) (72-93). It is argued that the alternating classical formulation (72 r+ 73) would (72) (73) (74) be expected to exist preponderantly in the tertiary cation form (73) and therefore the resonance of 3-H would be much nearer to the appropriate proton (H,) N9 H.Dahn and M. Ballenegger Helv. Chim.Acta 1969 52 2417. ’O R. K. Lustgarten P. G. Gassmann D. S. Patton M. Brookhart S. Winstein H. G. Richey and J. D. Nichols Tetrahedron Letters 1970 1699. 116 N. S. Isaacs in the substituted cyclopropylmethyl cation (74) (5 7-8). The 7-p-anisylnorborn- enyl cation (75) appears to exist in the classical structure as judged by the con- siderable downfield shift of the aromatic protons and lack of such in the olefinic protons.' The chlorine nuclear quadrupole resonances of the 7-norbornenyl (75) (76) chlorides differ by 860 kHz between the syn- and anti-isomers indicating greater s-character of the C-Cl bond in the former possibly on account of Cl-z orbital repulsion as shown in (76).92 The 2-norbornyl cation has been observed in superacid solution at -154 "C; H 3C n.m.r.and Raman spectra have been interpreted in terms of the corner- protonated nortricyclene structure (77).93 All protons are rapidly interchanging Y H -1,2,6 r 5.00 H-3,5,7 8.14 H -4 7-18 above ca. -50 "C by two processes a 3 +2 hydride shift (EA= 45 kJ mol-') and the 6 +1+2 proton migration which is not slowed sufficiently until -154"C (EA= 5.9 kJ mol-I). By contrast 2-halogenonorbornyl cations have the localised structure (78).94 Carbon scrambling in the 2-norbornenyl cation &) 3.38 3.68 4.5 1.76 F (78) (79) has been re-examined and discrepancies in the results of earlier workers have been largely resolved :95 it appears that the skeletal rearrangements are quite slow and for example in acetic acid-potassium acetate at 45 "C for 1h 33 % of 2,3 +1,4,7 rearrangement takes place in accordance with Roberts et al.91 H. G. Richey J. D. Nichols P. G. Gassmann A. F. Fentiman S. Winstein M. Brook-hart and R. Lustgarten J. Amer. Chem. SOC., 1970,92,3783. 92 H.Chihora. N. Nakamura and T. Irie Bull. Chem. SOC.Japan 1969,42 3034. 93 G.A. Olah A. M. White J. R. DeMember A. Commeyras and C. Y. Liu J. Amer. Chem. SOC.,1970,92,4627. 94 G. A. Olah P. R. Clifford and C. L. Jenell J. Amer. Chem. SOC.,1970,92,5531.95 C. C. Lee and B.-S.Halen J. Amer. Chem. SOC.,1970,92,2583. Reaction Mechanisms-Part (ii) while 11 h at 24 “C yields 48 % rearrangement (as found by Cristol and co-workers). By the addition of acetic and trifluoroacetic acids to norbornene the (79) labelled esters with a greater proportion of D at the 3-exo position (80) than at 7-syn are formed.96 This may indicate either an asymmetry in the carbonium ion formed (due possibly to the presence of the anion) or a secondary isotope effect which possibility is supported by the observed lack of temperature depen- dence of the product ratio.” The generation of the norbornyl cation by the thermolysis of exo-2-norbornyl thiocyanate (81) leads to C-14-2 scrambling by a Wagner-Meerwein shift but not of C-6 with C-1 or C-2 suggesting that these hy- dride shifts are slower.’* An inverse /I-deuterium isotope effect (kH:k = 0,958 at 150“C)seems to indicate that C-2 is not bearing positive charge in the transition state.The 1,2-dimethoxynorbornyl cation is stable in fluorosulphonic acid solution and undergoes a rapid degenerate rearrangement to (82) which is sufficiently slow below 0°C for the individual species to be observed (AGS = 55 kJ mol-I).’’ Rearrangements of several (x)-hydroxy-(x)-phenyl-2-norbornyl cations have been Products from the 3-substituted cation (83) 96 H. C. Brown J. H. Kawakami and K-T. Liu J. Amer. Chem. SOC.,1970,92 5536. ” J. L. Holmes D. McGillivray and N. S. Isaacs unpublished data. ’’ L. A. Spurlock and J. E. Parks J. Amer.Chem. SOC.,1970,92 1279. 99 A. Nickon and Y. Lin J. Amer. Chem. SOC.,1969,91 6861. loo C. J. Collins V. F. Raaen and M. D. Sckert J. Amer. Chem. SOC.,1970 92 1787. lo’ C. J. Collins and B. M. Benjamin J. Amer. Chem. SOC.,Q70 92 3182. lo’ B. M. Benjamin and C. J. Collins J. Amer. Chem. SOC.,1970 92 3183. 118 N. S. Isaacs MeO+ OMe (82) depend upon the geometry of the leaving group and indicate the short lifetime of the cation. a-and P-secondary deuterium isotope effects on norbornyl solvolyses have been measured for (84)-(87). The p-effects are quite small especially for the exo-isomer; the a-effect upon the exo-ester is considered too large to support HONO &OH -ocoph'x NH2 18.5 % 22-1 9.7 HONO 0.4 29.2 37.2 (83) a classical formulation of the cati~n.'~~~'~~ The attachment of a methoxy-group to part of a system which is acting as a nuclophile in a solvolytic reaction x = oso,QBr normally enhances the rate to a considerable extent e.g.(88) (89). The methoxyl probe has been systematically applied to the 1 4,5 6 and 7-positions in both exo- and endo-norbornyl tosylates. Their solvolytic rates are not however enhanced (with the possible exception of 4-OMe which is five times faster than '03 B. L. Murr and J. A. Conkling J. Amer. Chem. SOC.,1970,92 3464. Io4 B. L. Murr and J. A. Conkling J. Amer. Chem. SOC.,1970 92 3462. Reaction Mechanisms-Part (ii) OTos RA CH2-0DNB Ril (88) (89) 1 R=H krel = 1 Me 22 11 OMe 2400 791 4-H) and are otherwise retarded due to the inductive effect of the oxygen.This study therefore gives no support to the delocalisation of charge at C-5,6,7.'05 Solvolysis of 7-norbornyl compounds appears to show the greatest electronic demand of many cyclic analogues to judge from Hammett p-values for the series (90a-f).'06 This demand may be satisfied both by a 7-p-anisyl group or (a) (b) (4 (4 (el (f) -4.48 -4.1 -4.65 -5.64 -4.83 -4.54 (90) a cyclopropane ring (91a-~).'~' A four-membered ring is also capable of stabilising the 7-norbornyl cation (92a-f).'08,'09 A sharp mechanistic change krel 1 1014 3 10'7 (91) occurs on solvolysis of the series (93) at the point at which double-bond participa- tion yields to aryl participation. The break in the Hammett plot comes at X = OMe.For substituents with a smaller value of cr' p = -2.30 and for those with larger p = -5.1 which is identical with the solvolytic behaviour of the corresponding saturated compounds (94).' An unusual rearrangement of a '05 P. von R. Schleyer P. J. Stang and D. J. Raber J. Amer. Chem. Sac. 1970,92,4725. lob H. Tanida and T. Tsushima J. Amer. Chem. SOC.,1970 92 3397. lo' P. G. Gassmann and A. F. Fentiman J. Amer. Chem. SOC.,1970,92 2551. log M. Sakai A. Diaz and S. Winstein J. Amer. Chem. SOC.,1970 92 4452. '09 M. A. Battiste and J. W. Nebzydoski J. Amer. Chem. SOC.,1970 92 4450. P. G. Gassmann and A. F. Fentiman,'J.Amer. Chem. SOC., 1970,92 2549. 120 N. S. Isaacs H OTos e!. 608 000 k,, 1 1-2 20 800 8 (e) X = OTos Y = H; 1.1 (f) X = H Y = OTOS 72 700 (92) five-membered to a four-membered ring occurs in part in the solvolysis of 1-methoxynorbornyl brosylate (95)." Some variants on the norbornyl system (93) (94) have been studied; skeletal rearrangement of (96) is deduced to precede sol-volysis."* Solvolyses of both exo-and endo-(97) appear to be abnormally fast 0 (95) (by factors of 250 and 800 respectively) according to predictions based on the Halford-Schleyer-Foote rule,' l3 though the exo:endo rate ratio is less than (96) 'I1 Y.Lin and A.Nickon J. Amer. Chem. SOC.,1970,92 3496. IL2 R. R. Sauers and B. R. Sickler Tetrahedron Letters 1970 1067. J. 0. Halford J. Chem. Phys. 1956 24 830; C. S. Foote J. Amer. Chem. Soc.1964 86 1853; P.von R. Schleyer ibid. p. 1854 1856. Reaction Mechanisms-Part (ii) one.' l4 Solvolysis of 4-tricyclyl triflate (98) proceeds very slowly (estimated as 2.8 x lo4 times less fast than 2-norbornyl triflate at 295 "C); there is no Tf = -OSOZCF, Gb (97) (98) evidence therefore for participation from the face of the three-membered ring.' I Deltacyclyl esters (99)both exo and endo,solvolyse to produce the exo-deltacyclyl products ;a degenerate rearrangement of the cation may be detected by deuterium -RD-& &oBros = +& D D D (99) 1 scrambling e.g. (99)-(100)."7.'18 An am-analogue of the norbomenyl system (101) solvolyses with loss of nitrogen and an exo:endo ratio of 117."' Small Rings.-Very large end0:exo reactivity ratios have been found in the solvolyses of bicyclo[2,1,0]pentyl (102) and bicyclo[2,1,1 Jhexyl (103) tosylates.The reactions are clearly highly assisted compared with cyclobutyl esters and 114 I. Rothberg J. C. King S. Kirsch and H. Skidanow J. Amer. Chem. SOC.,1970 92 2570. 115 S. A. Sherrod R. G. Bergman G. I. Gleicher and D. G. Morris J. Amer. Chem. SOC. 1970,92 3469. 116 R. C. Bingham W. F. Sliwinski and P. von R. Schleyer J. Amer. Chem. SOC.,1970 92 3471. 117 P. K. Freeman and J. N. Blazevich. Chem. Comm. 1969 1357. 118 P. K. Freeman D. M. Balls and J. N. Blazevich J. Amer. Chem. SOC.,1970,92 2051. 119 E. L. Allred and C. R. Flynn J. Amer. Chem. SOC.,1970 1064. 122 N. S. Isaacs participation by the electrons of the four-membered ring is postulated.' 2o Large rate differences have also been recorded between cis-and trans-isomers of (103) exo 1 endo 10' bicyclo[4,2,0]oct-l-yl (104) and bicyclo[3,2,O]hept-l-y1 (105) dinitrobenzoate solvolyses.'2 These effects probably reside in angle strain differences in the derived cations.Deamination of the deuterium-labelled spiropentylamine (106) leads to isomeric methylenecyclobutanols (107) and (108) in which the labelling dCH2- D NH D (107) labelled 100 % 3,4-D,; (108) labelled 75 % 2,2-D, 25 % a,a-D (107) pattern accords best although not precisely with the scheme shown.'*' Exten-sive participation is also proposed for the very facile solvolysis of the bicyclo- [l,l,l]pentyl system (109) which despite the a-phenyl group is more reactive than the cyclobutyl analogue (110a).'23 A 3-ethoxy-group either cis-or trans- retards solvolysis of cyclobutyl esters (1 IOU).The effect cannot be steric as it would then be limited to the trans-isomer the reaction being disrotatory as shown in (1 11).'24 The products of hydrolysis of 3-ethoxybutyl esters are mainly K. B. Wiberg J. E. Hiatt and K. Hseih J. Amer. Chem. SOC.,1970,92 544 553. Iz1 K. B. Wiberg R. A. Fenoglio V. Z. Williams and R. W. Ubersax J. Amer. Chem. SOC. 1970 92 564 568. l2 D. E. Applequist M. R. Johnson and F. Fisher J. Amer. Chem. SOC.,1970 92 4614. '23 A. Padwa and E. Alexander J. Amer. Chem. SOC..1970,92 1796. See 125 refs. 1-5. Reaction Mechanisms-Part (ii) 123 -. the corresponding (but inverted) 3-ols and it seems therefore that the ethoxy- group suppresses cyclobutane ring participation by its -I Differences doTos doTos EtO$x (11oc) (110b) JfOTos EtO” (1lOd) 0-0035 0.0013 in product ratios during solvolysis of cis-and trans-(112) are interpreted in terms of partial specific internal return with differential leakage to a classical cyclopropyl cation.126 Spiro[2,3]hexyl tosylate (I 13) is hydrolysed in part to a @MeCH,OBros ’ CH,OBros de cis-(112) trans-(112) 1 product (114) derived by homoallylic rearrangement.127 Reactions of this type have been examined in fluorosulphonic acid in which the ion (115)is stable lZs I.Lillien and L. Handloser Tetrahedron Letters 1970 1213. ‘*’ J. J. Gajewski R.L. Lyle and R. P. Gajewski Tetrahedron Letters 1970 1189. R. Maurin and M. Bertrand Tetrahedron Letters 1970 5065. 124 N. S. Isuucs though degenerate and exchanges methyl groups by two rearrangements k being rapid and k2 much slower.'28 N-Chloroaziridines (116) ionise by a dis- rotatory process analogous to cyclopropyl cations to form the aza-ally1 cation (117) at a rate which is greatly enhanced by terminal methyl substitution.'29 Participation by the cyclopropane ring and by the double bond are competitive ar"' c1-N in the solvolysis of (118) each process leading to distinct products. The latter is more effective by a factor of 4.l3O The stable degenerate carbonium ion resulting from the dissociation in superacid medium of either cyclobutanol or cyclopropylmethanol shows an n.m.r.spectrum (r 3.5 double quartet; z 5.3 doublet ; r 5.8 doublet) consistent either with the rapidly equilibrating non- classical cyclobutonium ion (119) or a bisected cyclopropylcarbinyl cation (12O).' 31 The methylcyclobutyl cation (121) also undergoes rapid degenerate 12* T. S. Sorensen and K. Ranganayarkava Tetrahedron Letters 1970 659. lz9 P. G. Gassmann D. K. Dygos and J. E. Trent J. Amer. Chem. Sac. 1970,92 2084. J. B. Lambert J. W. Hamersman A. P. Jovanovich F. R. Koeng S. A. Sweet and P. J. Kucinski J. Amer. Chem. SOC.,1970 92 6372. 13' G. A. Olah D. P. Kelly C. L. Jewell and R. D. Porter J. Amer. Chem. SOC.,1970 92 2544. Reaction Mechanisms-Part (ii) do" rearrangement in this medium.'32 Solvolyses of allylic esters are retarded by bond-angle strain at the a-carbon such as produced by a small ring e.g.(122b (125).'337134 However acceleration by electrons from neighbouring highly- strained bonds fits the solvolytic data for a series of such systems as (126H128) 6- CH2--X &FCH2, @cH2x -+ (126) (127) (1 28) (129) which are believed to assist ionisation by conjugation e.g.(129).'35 Vinyl Cations.-Vinyl cations are now well-established intermediates in the solvolysis of particularly a-arylvinyl halides (130) and related compounds. Such solvolyses have been found to proceed more or less readily in highly ionis- ing solvents although at rates quite low compared to the saturated analogues the products being vinyl esters from carboxylic acid solvents and ketones from IJ2 M.Saunders and J. Rosenfeld J. Amer. Chem. SOC.,1970 92 2548. lJJ G. D. Sargent and M. J. Harrison Tetrahedron Letters 1970 3699. IJ4 H. G. Richey R. Fletcher and R. G. Overmayer Tetrahedron Letters 1970 3703. IJ5 N. A. Clinton R.S. Brown and T. G. Traylor J. Amer. Chem. SOC.,1970,92 5228. 126 N. S. Isaacs aqueous solvents. Recent work has been concerned with the verification of the S,1 mechanism (A) and the elimination of other possible contenders (e.g.R-F).'36 A sNI Ph Ph OAc AcOH )=+-Ph --+ R R Ph B S,2 Ph OAc C addition -H Ph Br Ph R Ph H-R Ph (130) AcOH ph E elimination PhCrCPh addition H Ph Ph F aryl participation -I,+'\ AcOH Ph OAc 'A --)M R Ph R Ph Route E can be eliminated since in many cases studied there is no /3-hydrogen.The rates of vinyl solvolyses are greatly accelerated by an aryl group and are sensi- tive to its substituents (p = -3.6);there is a moderate sensitivity to solvents (Grun- wald-Winstein m = 0.3-0-5),137 but first-order kinetics are usually observed and no acceleration by added lyate ion which thereby excludes the direct displace- ment route B. When such solvolyses are carried out in deuteriated solvent there is typically neither a solvent isotope effect of any significance (k,:k = 1.04 in AcOD and 1.3-143 in EtOD) nor is deuterium incorporated in the The latter observation excludes the elimination-addition route An An m 136 Z. Rappoport T. Bassler and M.Hanack J. Amer. Chem. SOC., 1970,92,4985. '" Z. Rappoport and Y. Apeloig Tetrahedron Letters 1970 1817 1845. 13' M. A. Imhoff R. H. Summerville P. von R. Schleyer A. G. Martinez M. Hanack T. G. Dueber. and P. G. Stang J. Amer. Chem. SOC.,1970 92 3802. 139 P. Beltrame M. G. Cattania G. Massolo and M. Simonetta J. Chem. SOC.(B) 1970 453. Reaction Mechanisms-Part (ii) 127 E and the former excludes routes C and D since a rate-determining proton- addition should be accompanied by a large (ca. 7) solvent isotope effect. Rates OTf (1 32) (133) show a dependence on leaving-group similar to that of typical S,1 reactions e.g. in (131) X = Me > H > N02.14' Wagner-Meerwein rearrangements can Me Me Me \ Ht Me + II Me'] .C-C-CH Me*j>>-C=CH2 -+ +C-C=CH2 I Me Me Me k-,"a (134) Me Me \/ ,C-C --Me c1-'/ j Me apparently occur in the 1-adamantylvinyl cation (132) +(133),'38 and in the cation produced by protonation of t-butylacetylene (134).14' P-Aryl groups facilitate reaction ; for instance (135) solvolyses 100 times faster than (136).142 Ph OS02F H OSO2F Ph>=( H>=( Ph Ph (135) (136) The same authors confirmed the lack of solvent isotope effect in these solvolyses (AcOD) the lack of effect of added acetate ion and the rate ratio in solvents of equal ionising power k(AcOH):k(90% aq. EtOH) = 1.6 a value expected to be near unity for route A but only by extreme coincidence to be so for reaction by route D. The P-deuterium isotope effect was found to be 1.45 in [2H]-(136) similar to typical values in known S,1 reactions.The stereochemistry of vinyl solvolyses frequently leads to mixtures of cis-and trans-isomers in the products or to isomerisation of the starting material prior to reaction by ion-pair to a linear cation. Acetolysis of 1 -anisylvinyl bromide is accompanied by no return but extensive bromide ion return occurs in tri- anisylvinyl bromide. '43 cis-and trans-l,2-Dianisylvinyl bromides react in acetic acid at quite similar rates the cis being somewhat faster (which argues I4O 2.Rappoport and J. Karpi J. Amer. Chem. SOC.,1970,92 3220. 141 K. Griesbaum and Z. Rehman J. Amer. Chem. SOC.,1970,92 1417. 142 W. M. Jones and D. D. Moness J. Amer. Chem. Soc. 1970,92 5457. 143 Z. Rappoport and M.Atidia Tetrahedron Letters 1970 4085. 128 N. S.Isaacs against j?-aryl participation) but the ratio of reactivities is quite affected by added base probably due to a component of reaction leading to elimination. Similarly the solvolytic rates of (137) and (138) are very similar and rearside An An An An X >=(Ph X c! )=+-An Ph x- 1_ )--(Ph An (137) (139) (138) -1ACOH An An Ph OAc>=( participation of the anisyl group in (137) cannot therefore be important. The rates of cis-trans isomerisation and of solvolysis are very similar arguing that both processes occur by way of the ion (1 39). 144 Chlorine exchange in (140) occurs with retention of configuration possibly by direct substitution.145 A cyclopropyl group can stabilise vinyl cations ; (141) reacts with acetic acid in the presence of silver ion at room ternperat~re.'~~ The conclusions to be drawn from intensive studies over the year are that vinyl cations are not especially intrinsically unstable [Rappoport compares the stability of (142)with p-methoxybenzyl The relatively low solvolytic rates appear therefore to be due to dficulties in the ionisation process and sound a warning against attempts to compare carbonium ion stabilities of widely different types by rate measurements in solvolysis reactions.The origin of vinyl halide inertness may lie in the conjugation of the C-ha1 bond with the 7c-system or in high s-character of the C-ha1 bond or in steric hindrance to rearside solvation due to the a-aryl group or a combination of these factors.The subject has been reviewed by Hanack,I4' who with Rappoport and Bassler has discussed and summarised 20 criteria for judging the solvolytic mechanisms of 12 vinylic 144 Z. Rappoport and Y. Apeloig Tetrahedron Letters 1970 1845. 145 P. Beltrame P. L. Beltrame G. Carboni and M. L. Ceveda J. Chem. SOC.(B) 1970 730. 146 D. R. Kelsey and R. G. Bergman J. Amer. Chem. SOC. 1970,92 228. 14' Z. Rappoport and A. Gol Tetrahedron Letters 1970 3233. 148 M. Hanak Accounts Chem. Res. 1970,3 209 149 Z. Rappoport T. Bassler and M. Hanack J. Amer. Chem. SOC. 1970,92,4985. Reaction Mechanisms-Part (ii) Bimolecular Displacements.-Several theoretical papers have appeared descri b- ing the calculation by sophisticated MO methods of the energies of model transition states for S,2 processes.Potential surface calculations of the system (143) suggest that the D, geometry is preferred and energy minimised at bond H (143) lengths a = 201 b = 330 pm the electron density on carbon being significantly lower than in methane.' 50 The corresponding positive ion a model transition state for the SE2reaction still apparently prefers D, geometry over C (143b) (which is usually assumed to account for retention of configuration in these reactions) although by a smaller margin than in the case of the negative ion.' ' An attempt to rationalise substituent effects upon S,2 reactions has been made by summing estimated changes in bending and stretching modes between ground and transition states.' s2 This field is notoriously ambiguous since the electronic perturbation due to a substituent at the central carbon will affect bond-making and bond-breaking processes to opposite but not necessarily equal or constant extents.It is predicted that for the transition state (144)an electron-withdrawing Y I /X-"-C-b-X Ix-f-c-6-z X = 2nd row element J ."\b H' H A Z = 1st row element (1 44) (145) group Y will increase the order of bond b (relative to Y = H) and decrease that of a.153If the two nucleophilic groups are in different rows of the Periodic Table as in (145) the changes in bond-order are predicted to be in the same sense. This seems to imply that for the displacement of a less electronegative atom by a more an electron-withdrawing substituent will shift the transition state to a less 'product-like' appearance.' 53 it seems that secondary a-deuterium isotope effects in the SN2reaction are very small or even inverse while the value of k,:k for the ionisation reaction is about 1.23.Measured values intermediate between these extremes for solvolyses of p-substituted benzyl brosylates have been taken to indicate a mixed mechan- ism. The magnitudes of these isotope effects (in CF,CH,OH-H,O and C,H,OH-H,O) increased with solvent polarity and with electron-donating 15" C. D. Richie and G. A. Chappell J. Amer. Chem. SOC.,1970,92 1819. l5 * N. L. Allinger J. C. Tan and F. T. Win J. Amer. Chem. SOC., 1970 92 579. M. Pahari and R. Baru J. Indian Chem. SOC.,1970,47 364. J. C.Harris and J. L. Kurz J. Amer. Chem. SOC.,1970 92 349. 130 N. S. Isaacs ability of para-substituents as the reaction became progressively more limiting.ls4 Primary chlorine (k,5:k 7) isotope effects have been reported for displacements of chloride from benzyl chlorides by PhO- and PhS- to lie in the range 1-0092-1~0098.'55 The hypothesis that nucleophilic displacements by free and by cation-paired nucleophilic anions can be treated independently -attributed to Acree (1912)ls6-has been examined in the case of iodine-exchange by the alkali-metal iodides on methyl iodide in methan01.l~' The results were analysed taking into account salt effects and the contribution to attack by free ions was separated out. This as required proved to be independent of the nature of the cation.New powerful nucleophilic systems have been examined. In dimethyl sulphoxide aryloxide may be displaced by thiolate ion :Is8 DMSO EtS-+ CH,OAr -EtSCH + OAr-and in molten KSCN benzoate ion may be similarly displaced:'59 ArCO-OCH + KSCN -+ ArCOO-Kf + CH,SCN. Low valence states of cobalt and iron e.g.(146)and (147),provide new and power- ful nucleophiles of potential importance.'60 The apparent displacement of OH-by aniline in the ferrocene derivatives (148)and (149)is remarkable if indeed the mechanism is of the SN2 type.16 Internal displacements (SNi),proceeding 0 154 V. J. Shiner M. W. Rapp and H. R. Pinnick J. Amer. Chem. SOC.,1970 92 232. 155 E. P. Grimsrud and J. W. Taylor J. Amer. Chem. SOC.,1970,92 739. lS6 S.F. Acree J. Amer. Chem. SOC.,1912,48 353. I 57 P. Beronius and L. Pataki J. Amer. Chem. Sac. 1970,92 4518. 15' G. I. Fentrill and R. N. Mirrington Tetrahedron Letters 1970 1327. 159 E. M. Wadsworth and T. I. Crowell Tetrahedron Letters 1970 1085 160 M. D. Johnson unpublished results. Ibl G. Marr B. W. Rockett and A. Rushworth Tetrahedron Letters 1970 1317. Reaction Mechanisms-Part (ii) through intimate ion-pairs have been proposed for the thermolysis of alkyl thiocarbonates (150).'62Partial racemization and '*O scrambling occurs at a rate in the unreacted material greater than the decomposition there being an estimated 80 % ion-pair return.'63 RS Ar \ I SR Ar c=o *-' c +I \/ coo-HT-O H Ph Ph Ph (150) Displacements at Nuclei other than Carbon.-The powerful nucleophile diphenyl- phosphide appears to prefer to attack bromine rather than carbon in propargylic bromides (151).'64 The displacement at sulphur of the aryloxy-group from a sul- -Ph2P-Br-CHR-C-CH -+ Ph2PBr + CHR-C=CH (151) phenic ester (152)has been interpreted as a synchronous process,165 since the rate Ph,C-ScO-PNB -+ Ph,C-S-OAr + OPNB-t ArO-(152) correlation with the basicity of the attacking nucleophile (Brmsted p = 0.25) and of the leaving group (y = -0.97) is satisfactory.However the displacement of chloride from benzenesulphenyl chloride (153) by aniline seems to involve the formation of a moderately long-lived intermediate since the apparent bi- molecular rate-constant increases to a constant value with increasing aniline concentration.' 66 Displacement of chlorine at phosphorus in the conformation- Ph-S-Cl + PhNH2 + [complex] + PhSNHPh + HCl (153) ally stable compound (154) probably occurs by a combination of SN2(P)and S,1(P) processes as judged from the products; that resulting from the bi- molecular process is suppressed by silver ion catalysis which promotes the conformationally mobile cation (155).'67 lb2 J.L. Kice R. L. Scriver E. Koubek and M. Barnes J. Amer. Chem. SOC.,1970 92 5608. 163 J. L. Kice and G. C. Hanson Tetrahedron Letters 1970 2927. 164 W. Hewertson and I. C. Taylor Chem. Comm. 1970 119. 165 L. Senatore E. Guffarin and A. Fava J. Amer. Chem. SOC.,1970 92 3035. lb6 E. Ciuffarin and F. Griselli J. Amer. Chem. SOC.,1970 92 6015."' W. Wadsworth and H. Horton J. Amer. Chem. SOC.,1970 92 3785. 132 N. S.Isaacs MeOH Neighbouring-group Participation.-Competitive participation by oxygen (R20-3) and the homoallylic double bond of (156)is revealed to favour the latter 1 by the relative amounts of the products of acetolysis. (C1-5)-Participation OBros OAc OAc 0 Q-oQ 9+ ao*c + 0 4% cis 69 % trans 22 % in (157) is substantiated by the formation of the halonium ion (158) stable in SbF solution from which the solvolysis products are obtained on addition of water alcohols et~.'~~ The urethane group acts as an ambident neighbouring (157) (158) group. At low pH the carbonyl oxygen is the reactive nucleophile but at high pH the nitrogen participates in the form of the anion (159).'70,17' The hydrolysis of salicylacetals (1 60) occurs by participation of the neighbouring carboxy-group as a specific Brsnsted a~id.'~~,'~~ The rate is independent of pH throughout a considerable range of the acidic region.A type of participation is postulated lb8 L. A. Paquette R. W. Begland and P. C. Storm J. Amer. Chem. Soc. 1970,92 1971. Ib4 P. E. Peterson P. R. Clifford and F. J. Slama J. Amer. Chem. Soc. 1970 92 2840. F. L. Scott and D. F. Fenton Tetrahedron Letters 1970 681 685. F. L. Scott and C. V. Murphy Tetrahedron Letters 1970 1731. B. M. Dunn and T. C. Bruice J. Amer. Chem. SOC.,1970,92 2410. B. M. Dunn and T. C. Bruice J. Amer. Chem. SOC.,1970,92 6589. Reaction Mechanisms-Part (ii) Ph EtO- I (159) to reinforce the nucleophilicity of ethylenediamine (161) which is considerably higher than expected for a simple primary amine.'74 The oxygen bridge of (162) (161) can participate to stabilise the carbonium ion leading to enhanced rates of solvoiysis and rearrangement.I7' Carbanions and Enolisation.-Conrotatory ring-opening of the anion of bicyclo-[6,1,0]nonatriene (1 63) leads to cis,cis,cis,trans-cyclononatetraenyl anion (164) an aromatic species by the n.m.r.criterion; the external protons are strongly '14 W. P. Jencks and K. Solveson Chem. Comm. 1970 548. L. A. Paquette and P. C. Storm J. Amer. Chem. SOC.,1970,92,4295. 134 N S. Isaacs deshielded and the internal one shielded confirming the presence of a dia- magnetic ring-c~rrent."~ Another new aromatic anion is the tridehydro- H 2 H-1 t 13.52 H-2,9 t2.73 (doublet) H-3-8 T 3-3.6 7 8 (163) (164) [17]annulene anion (165) an 187t species.'77 A number of conjugated carbanions have been studied in liquid ammonia solution.The cyclic dienyl anions (166)- OMe (168) show strongly alternating charges as is predicted by simple HMO theory. The coupling constants for adjacent protons are proportional to the calculated bond orders between the carbons which bear them but no evidence was found for a ring-current as in (169). The allylic systems (170) and (171) are in equilibrium.'78 N.m.r. studies also reveal that the pentadienyl anion (172) and analogues such as (1 73) preferentially adopt the 'W' conformation.Intramolecular 1,6-sigma- H Me +Me NH,-.- Me PhCHJyM" h. v Me gPh-C/ NH-3 p H O.*Me.-(172) (mj G. Boche D. Martens and W. Danzer Angew. Chem. 1970,81 1003. '" J. Griffiths and F. Sondheimer J. Amer. Chem. SOC.,1969,91 7518. H. Klooster and J. A. Van Drune Rec. Trav. chim. 1970 89 368; H. Klooster and G. J. Heiszwol ibid. p. 413. Reaction Mechanisms-Part (ii) tropic shifts can occur in pentadienyl anions (174) and as predicted are antara- facial on thermolysis and suprafacial when ph~to-induced.'~~ The perchlorotrityl H 48% (174) 520; anion is particularly stable; the tetramethylammonium salt (175) has been isolated.' 8o (C,Cl &C- &Me (175) Ionisation constants for a large number of aryldinitromethanes have been measured ;values of pK in general correlate with oo,presumably indicating that resonance contributions from the aryl moiety are unimportant.'81 Substituted fluorenes have been used as indicators to set up an acidity scale appropriate to the highly basic system 4+90 % dimethyl sulphoxide-Me,N+ OH-.These indicators were found to have dissociation constants which were rationally related to a-values of substituent groups and to molecular orbital parameters. /&Substituted vinyldinitromethanes (176) have been shown to possess thermodynamic acidities correlated by the two-term equation :l 84 log k/kO = zO,PI + ORPR showing that both inductive and resonance terms are important. Rates of base-catalysed isomerisation of a-cyano-cis-stilbenes (177) (in dimethyl sulphoxide- X = C02Me pK = 3.14 (O,N)ZCH.CH=CHX SO,Me 1.65 CN 1.93 (176) NO2 0-07 ethanol) correlate with the acidity function H-,appropriate to the medium and with a-values of substit~ents.'~~ A measure of the ability of the methylene H CN He RO CN Ph CN \/ ...I -/ \/ /c=c\ +OR-PC\* /c=c\ Ph Ar Ph Ar H Ar (177) G. J. Heiszwol and J. A. Van Drunen Rec. Trav. chim. 1969 88 1377; R. B. Bates S. Brenner W. H. Deines D. A. McCombs and D. E. Patter J. Amer. Chem. SOC. 1970,92,6345. lM0M. Ballester and G. de la Fuente Tetrahedron Letters 1970,4509. lM1G. I. Kolesetskaya I. V. Tselinskii and L. I. Bagal Reakts. spos. org. Soedinenti 1969 6 387. IMZ "' K. Bowden and A. F. Cockerill J. Chem. SOC.(B) 1970 173.K. Bowden A. F. Cockerill and J. R. Gilbert J. Chem. SOC.(B) 1970 179. L. A. Kaplan N. E. Burlinson W. B. Moniz and C. F. Poronski Chem. Cornrn. 1970 140. D. J. Kroeger and R. Stewart J. Chem. SOC.(B) 1970,217. 136 N. S. Isaacs group to transmit charge comes from a comparison of substituent effects upon the acidity of (178) and (179). Hammett p values for the former are 1.07 and for p-XC H CHNO p-X.C,H,.CH2CHN02 X = C0,Me I Me Me S0,Me CN (178) (179) NO2 the latter 0.4 but correlations of both were made with c rather than 0- giving a linear log-log plot.'86 On the other hand acidities of (180) correlate best with c-,implying a more important contribution from conjugative inter- actions between a carbanionic centre and the substituent group.187 Exchange reactions of vinylic'88 and of aromatic protons [e.g. as in (181)]189 occur when (180) (181) these are suitably activated. Second (and higher)-row elements stabilise adjacent carbanionic centres; sulphur for example is well known to do so. For the halogens the order is I Br > Cl in the acidities of (182).l9O a-Fluorine evidently CH,X I NO2 (182) destabilises dinitromethide anions (183) relative to the corresponding alkyl or chloro-compounds as seen in their relative nucleophilicities towards methyl acrylate."l*' 92 Br onsted plots of kinetic against thermodynamic acidity for a homologous series of carbon acids have been recorded which have slopes greater than unity,193 indicating that as electron-withdrawing capacity increases the X rate (lo4I mol-s-l) -Me 67-1 Et 70.7 c1 98.5 (183) F 164 000 IN* F.G. Bordwell W. S. Boyle and K. C. Yee J. Amer. Chem. SOC.,1970 92 5926. Is' L. A. Kaplan N. E. Burlinson and W. B. Moniz Chem. Comm. 1970 140. D. Daloze H. G. Viehe and G. Chiurdoglu Tetrahedron Letters 1969 3925. R. D. Guthrie and D. P. Wesley J. Amer. Chem. Soc. 1970 92,4057. 190 A. A. Abdullah Y. Iskander and Y. Riad J. Chem. SOC.(B) 1969 1178. 19' S. Wolfe A. Rauk L. M. Tel and I. G. Csizrnadia Chem. Comm. 1970 96. 19' L. A. Kaplan and H. B. Pickard Chem. Comm. 1969 1500. 193 M. Fukuyama P. W. K. Flanagan F. T. Williams L. Frainier S. A. Miller and H. Shechter J. Amer. Chem. SOC.,1970 92 4689. Reaction Mechanisms-Part (ii) 137 effect on the exchange rate is relatively much greater than on the dissociation constant.This has been interpreted by Kre~ge,'~~ using the ionisation of nitro- alkanes by hydroxide ion as a model as typical of a situation in which in the transition state (184) the substituent interacts more strongly with ionic base than with the activating group (NO,). A relatively high barrier to inversion may be inferred from the observation that deuterium exchange in (185) and (186) occurs with retention of configuration to a considerable e~tent.'~~*'~~ H CONH sh P 'Ph '* Ph Similar properties are ascribed to the sulphonyl carbanion which also exchanges hydrogen for deuterium with stereochemical retention. Base-catalysed addition of ethanol to fluorinated olefins probably proceeds via a carbanion e.g.(187) although stereochemical consequences cannot be judged from this experiment.I9' / F r + Ph PkoEt + xF FF F OEt 13 % 31 % Fluoride ion will add to fluorinated olefins such as (188) with the formation of a carbanion which can be captured by mercury(11). 198 Primary isotope effects for proton transfer are at a maximum when the proton is equally co-ordinated ly4 A. J. Kresge J. Amer. Chem. SOC.,1970,92 3210. 195 J. M. Motes and H. M. Walborsky J. Amer. Chem. SOC.,1970 92 3697. 196 H. M. Walborsky and J. M. Motes J. Amer. Chem. SOC.,1970,92 2445. 19' H. F. Koch and A. J. Kielbonia J. Amer. Chem. SOC.,1970 92 729. 19' B. L. Dyatkin S. R. Sterlin B. 1. Martynov and I. L.Knunyants Tetrahedron Letters 1970 1387. 138 N. S. Isaacs CF3 CF,\ HgX, \ F-C=CF2 -+ C-CF Hg(CR q2 / / R R to each base in the transition state when the base-strengths in the solvent used are comparable. This point has been nicely demonstrated by Dixon and Bruice who found k,:k = 10 in the ionisation of nitroethane by ammonia in water while smaller values of the isotope effect were noted for both stronger and weaker bases.'99 Similarly a maximum in the deuterium-exchange rate of menthone (k,:k = 6-5) occurs in aqueous dimethyl sulphoxide at a concentration of 30-40 % dimethyl sulphoxide.200 Camphenilone (189) is known to exchange protons (albeit with drfficulty) the carbanion being supposed to be stabilised by homoconjugation.No such process appears to operate with adamantanone (190),in which no exchange was detected under similar conditions.201 The non- classical carbanion (191) is protonated by methanol on the endo side and by (191) ..M~OD dimethyl sulphoxide on the presumably less-hindered exo side.202 The 7-nor- bornenyl anion shows little destabilisation compared to the saturated analogue (193) undergoes proton exchange at a rate comparable with that of (192). Anti- homoaromatic character is evidently not called into play in this instance.203 rel. kexch 1.0 1.4 lYy J. E. Dixon and T. C. Bruice J. Amer. Chem. SOC.,1970 92 905. *O0 R. P. Bell and B. G. Cox J. Chem. SOC.(B) 1970 194. 20' J. E. Nordlander S. P. Jintal and D. J. Kilko Chem. Comm. 1969 1136. 202 J.M. Brown and E. N. Cain J. Amer. Chem. SOC., 1970 92 3821. 203 R. Breslow R. Pagni and W. N. Washburn Tetrahedron Letters 1970 547. Reaction Mechanisms-Part (ii) On the other hand the system (194a) has its acidity reduced by more than 10” compared to cyclopentadiene since the anion (194b) has cyclobutadienoid character.’04 The substituted cyclopropane (195) undergoes racemisation in polar solvents. The mechanism suggested is by spontaneous heterolysi~.”~ The * ph*N e‘ Ph 3CN Ph+CO2Me ~ Ph C02Me Ph COzMe Ph CN (195) base-catalysed prototropic shift in tropyl compounds presumably occurs via a tropyl anion (196) an 8n (antiaromatic) system though possibly not planar.’06 _7 1_ Q Q r); (196) A transition state resembling a cyclopropenyl anion (197) makes 1,2 anionic shifts relatively unusual; a small amount of rearrangement occurs during the ArCMe2-CH2CI ArCMe2CH,Li + LiCMe,CH,Ph >it (198) p2 p2 (197) Ar.CMe,CH,CO,H PhCH2CMe2C0,H action of lithium on (198).207 1,2-Shifts ofallyl groups however are 6.n-rearrange-ments and are thermally allowed e.g.(199).208,209New polar’ ’* and oxidative” ‘ (199) coupling reactions of carbanions have been reported PhCHCN PhCH,CN Ns Ph-CH-CN [CHzCHzl~PhCHCN -PhCHCN I I I PhCHCN TI:”(SO5)z R’R2CHN02 + R1R2C2NO2 -R1R2C-CR1R2 II NO2 NO2 ’04 R.Breslow and W. N. Washburn J. Amer. Chem. SOC.,1970 92 427. ’OS E. W. Yankee and D. J. Cram J. Amer. Chem. SOC.,1970 92 6328,6329 6331. lob K. Takahashi H.Yamamoto and T. Nozoe Bull. Chem. SOC.Japan 1970,43 200. 207 E. Grovenstein and Y-M. Cheng Chem. Comm. 1970 101. 208 J. E. Baldwin J. de Bernardis and J. E. Patrick Tetrahedron Letters 1970 353. ’09 J. E. Baldwin and F. J. Urban Chem. Comm. 1970 165. lo ’ W. G. Kofron and C. R. Hauser J. Org. Chem. 1970,35,2085. D. J. Edge R. 0.C. Norman and P. M. Storey J. Chem. SOC.(B) 1970 1096. 140 N. S. Isaacs and an intramolecular aromatic cyclisation realised.2 Further evidence for the cyclopropenone intermediate in the Favorskii reaction comes from the I C02Et observation that the same products (200) and (201) result from the base-catalysed fission of the actual proposed intermediate (202) or its acetal as from the Favorskii reaction with the halogenoketones (203) or (204).2 Alkoxyketones which (73 PhCHCICOCH2CH3 PhCHCHCO2H 70% /r b 0 t-. (203) II (200) /“\ OMe-PhCH2COCHC1CH -+ PhCH-CHCH J (204) (202) PhCH2C-CHClCH3 + PhCH2COCHCH3 II I 0 (201) OMe 30 % may accompany the main products are derived from the intermediate chloro- enol by the route (205)+(206).2’4 Methyl substitution can dramatically change the course of a Favorskii reaction; a comparison of the rates of (207 R = H) and (207 R = Me) shows a 200-fold acceleration for the latter and a change in Hammett p value for the reaction from -5.0 to + 1.4 with in addition a considerable increase in the amount of alkoxyketone by-product and deuterium exchange prior to rea~tion.~” Despite this loss of C1 is rate-determining for ’’ R.Filler and A. Fiebig Chem. Comm. 1970 546. C. Rappe L. Knutsson N. J. Turro and R. B. Gascosian J. Amer. Chem. SOC.,1970 92 2032. ’I4 F. G. Bordwell and M. W. Carson J. Amer. Chem. SOC., 1970,92 3377. 215 F. G. Bordwell and M. W. Carson J. Amer. Chem. SOC.,1970,92 3374. 141 Reaction Mechanisms-Part (ii) (207 R = Me) and suggests a high degree of carbonium ion character on the halogen-bearing carbon possibly in a dipolar transition state (208). Elimination Reactions.-Studies on the halide- and thiophenoxide-ion-induced elimination reactions of 2-p-anisylbutyl menthyl and neomenthyl systems have been made.216 These reactions initiated by strong carbon bases on compounds with weakly acidic 8-protons and good leaving groups are believed to fall into the E2C mechanistic category involving the rather loose transition state (209).B B B (209) (210) (211) (212) A strong preference towards anti-elimination and the formation of the most stable olefin (Saytzev product) is shown by all systems. The similarity between the E2 (210) and Elcb (211) transition states might lead one to suppose that a gradation of mechanism could extend between these extremes i.e. (212). Elimination from 9-hydroxymethylfluorene (21 3) indicates that the Elcb process occurs (a-exchange more rapid than elimination k,:k = 7-2 solvent deuterium isotope effect k, k = 042) but detailed examination has suggested interpreta- tion in terms of a competing E2 process in addition to the Elcb mechanism.It is argued that in this case the borderline region between these two mechanisms is met. If a mechanistic continuum exists the intermediate carbanion should be of comparable energy to the E2 transition state and should be destroyed in a diffusion-controlled process. It is observed that the lifetime of the carbanion is ca. 10 times greater than this and hence it is inferred that the two mechanisms occur as independent pathways of the rea~tion.~”*~~* The Elcb mechanism typically occurs in systems with a rather acidic 8-hydrogen and a poor leaving group. Crosby and Stirling have examined Elcb reactions of a series of com- pounds XCH,CH,OPh where X is an electron-withdrawing gr~up.~’~,~~’ The + 4-most effective group (NO,) is closely followed by -PPh and -SMe, but -&Me is less effective by a factor of lo5,which points to the great effectiveness ’I6 G.Biale A. J. Parker S. G.Smith I. D. R. Stevens and S. Winstein J. Amer. Chem. Suc. 1970 92 115. ’” R. A. More O’Ferrall and S. Slae J. Chem. Sac. (B) 1970 260 268. ’” R. A. More O’Ferrall J. Chem. Sac. (B) 1970 274. ’I9 J. Crosby and C. J. M. Stirling J. Chem. Sac. (B) 1970 671. *’O J. Crosby and C. J. M. Stirling J. Chem. Sac. (B),679. 142 N. S. Isaacs of second-row elements in stabilising an adjacent carbanion. The a-deuterium secondary isotope effect k,:k = 0.67 is more typical for a reaction involving a relatively long-lived carbanion formed in a rapid pre-equilibrium. Elimination of methanol from (214) also shows Elcb characteristics,221 and a similar mechan- ism has been proposed for the hydroxide-catalysed elimination of benzoic OH (213) acid from (215) to give butenone.222 Electronic effects upon the SN2 E2C and S,l-El reactions of the series of compounds (216) have been compared.223 0II (a) Y = NO CH3 C-CH -CH 2 Ye yHCH2CH (b) H I (4 Me H Br (d) OMe (215) (2 16) The conditions were Bu4NfC1- in dimethylformamide (S,2) Bu4N+C1- in acetondutidine (E2C) and acetone-water (SN1-El).Hammett plots indicate that the E2C process is far less sensitive to conjugating substituents in the aryl group than the El [102.'-fold rate difference from (a)to (d) for the former 107'4- fold for the latter]. Rates of elimination of HCl from (217) by triethylamine and the ethanolamines are in the order of Brcansted base strength; values of the Hammett p decrease with decreasing basicity of the reagent which may be due (217) to a decrease in carbanionic character with base strength.224 Bordwell et.al. have classified Elcb reactions according to whether or not the fast prior ionisa- tion is reversible and if it is rate-determining. Attempts have been made to characterise examples according to this clas~ification.~~~ Hofmann eliminations of P-phenethylammonium salts (218) occur by a concerted (E2) process as judged by the lack of hydrogen-exchange with solvent and the strictly trans stereochemistry.226 The primary nitrogen isotope effect kI4:kl5 may be more 221 F. G. Bordwell K. C. Yee and A. C. Knipe J. Amer. Chem. SOC.,1970,92 5945.222 R. C. Cevestri and L. R. Fedor J. Amer. Chem. SOC.,1970,92,4610. 223 D. J. Lloyd and A. J. Parker Tetrahedron Letters 1970 5029. 224 Y. Yano and S. Oae Tetrahedron 1970,26 27. 225 F. G. Bordwell M. M. Vestling and K. C. Yee J. Amer. Chem. SOC.,1970,92 5950. 226 A. N. Bourns and A. 1. Frosst Canad. J. Chem. 1970,48 131. Reaction Mechanisms-Part (ii) or less than unity depending upon the substituent in the /3-aryl group and pre- sumably depends upon the degree of bond-breaking in the transition state.227 X (a) R = H QD (b) R = Me /+ CH -CH -NR D yMe3 / D (218) (219) Significant proportions of syn-elimination are reported from (219) induced by OH-[(a) 1-4 %; (b) 4h-50 %] and t-butoxide [(a) 46-50 % ;(b) ca.70 syn and anti-Eliminations appear to have quite different transition states charac- terised by primary isotope effects of 2 and 4.75 respectively.229 However the formation of both cis-and trans-2-butenes by HBr elimination from threo-2- brom0[3-~H]butane occurs by anti-elimination mechanisms. Eliminations from ex0-2-[3-~H]norbornyI tosylate (220a) and its 7,7-dimethyl derivative (220b) occur by an almost exclusively syn process presumably on account of steric hindrance towards the bulky base (2-cyclohexylcyclohexanolate) attacking the endo side.230 The transmission of electronic effects through sulphur is greater than through oxygen as evidenced by rates of elimination of (221) (a) and (b) with a variety of substituents Y for which values of pz=s = 0-37 and pz=o = ca.0 were found. It is suggested that sulphur participation is stabilising the transition state in the former case as shown in (222).231 E2 Mechanisms are proposed for YY V (a) Y = H (a) 2 = S (b) Y = Me (b) Z = 0 the formation of thiones from (223)232 and (224).233 Primary isotope effects of 3.0 and 6.1 respectively and the absence of hydrogen exchange with the solvent before reaction are further cited as evidence for the proposed mechanism. 227 P. J. Smith and A. N. Bourns Canad. J. Chem. 1970,48 125. 228 K. C. Brown and W. H. Saunders J. Amer. Chem. SOC.,1970,92,4292. 229 R. A. Bartsch Tetrahedron Letters 1970 297. 230 H. C. Brown and K. T. Liu J. Amer. Chem. SOC., 1970,92 200. 231 Y. Yano and S. Oae Tetrahedron 1970 26 67.232 A. Ceccon U. Miotti U. Tonellato and M. Padovan J. Chem. SOC.(B) 1969 1084. 233 U. Miotti U. Tonellato and A. Ceccon J. Chem. SOC.(B) 1970 325. 144 N. S. Isaacs Ph Ph \ % \ C=S + ROH + CN-/Y-s-cN Ar / Ar H (223) p = +3-5 Steric factors associated with the 6-substituent are probably responsible for the relative ease of elimination of the P-D-lyxofuranoside (225) ;no corresponding reaction occurs under the same conditions with the riboside (226),in which both Na. benzoate DMF 120"C possible ,&hydrogens (2 and 3) are shielded by the 6-tosyloxy-gro~p.~~~ A new pyrolytic cis-elimination [of tosyl carbamates (227)]occurs at lo@-150 "Cfrom secondary and at room temperature from tertiary substrates. As the carbamates are readily prepared from tosyl isocyanate this should prove a useful synthetic route to the ~lefin.~~~ A pyrolytic trans-elimination (228) -+ (229) is claimed and rationalised by invoking neighbouring-group participation by the methyl ester It was not established whether epimerisation at the P-proton was taking place (which should be fairly facile at the high temperature and especially in the presence of a trace of base) which could lead to a concerted cis-elimination.This elimination would be expected to take place more readily than usual since the neighbouring ester group labilises the P-proton. In fact an Elcb elimination is not ruled out. Eliminations from 2,3-diphenyl-2-propanol 234 J. Hildesche A. Gandemer and S. D. Gero Chem. and Ind.1970 94. 2JJ L. C. Roach and W. H. Daly Chem. Comm. 1970 606. 23h E. E. Suissman J. P. Li and M. W. Creese J. Org. Chem. 1970 35 1352. 145 Reaction Mechanisms-Part (ii) CH3 I C ?OdCH3 0’ +o I 350 “C + CH3C02H and its derivatives lead to some 40-55% of a-benzylstyrene (230) as the kinetically-controlled product a surprising result for a reaction which would be expected to obey the Saytzev rule. The thermodynamically formed product is a-me t hyls til bene (23 1).’ ’ Me Me tH2 I I Ph-C-CH,Ph -+ Ph-C-CH2Ph Ph-C=CHPh I X (230) (231) (a) X = OH ; TosOH benzene (b) X = OMe; TosOH MeCN (c) x = C1; pyridine (d) X = NH,; NH,- Additions to Unsaturated Systems.-The cornplex nature of additions of HCI to olefins in acetic acid is illustrated by a study of the kinetics and products from [1,3,3-2H3]cyclohexene leading to anti-addition of HC1 (third-order kinetics) syn-addition of HCI (second-order kinetics) and anti-addition of acetic acid (third-order kinetics) in three competing proces~es.~~~,’~~ The chloride :acetate ratio depends upon free chloride ion concentration showing that the reaction does not occur exclusively by way of the carbonium ion but probably by at least two distinct carbonium ion pairs.The addition of acetic acid to cyclohexene is general-acid-catalysed obeys the Brsnsted Law well and has a solvent isotope effect (k,:k = 1.42) interpreted as showing a slow protonation step analogous to that found in hydrati~n.’~’ Brown and co-workers have postulated that processes involving single-step addition to the double bond of norbornene are strongly retarded by 7,7-dimethyl substitution whereas those occurring by a two-step mechanism via a trivalent carbon intermediate are not generally re- tarded and yield endo-substituted products.On this basis catalytic hydrogena- tion hydroboration epoxidation and silver ion co-ordination come into the first category while oxymercuration HC1 addition and free-radical thiophenol 23’ I. Ho and J. G. Smith Tetrahedron 1970 26 4277. 238 R. C. Fahey M. W. Monahan and C. A. McPherson J. Amer. Chem. SOC.,1970,92 2810. 239 R. C. Fahey and M. W. Monahan J. Amer. Chem. SOC.,1970,92 2816. 240 R. Corrin and J. Guenzet Tetrahedron 1970 26 671.146 N. S. Isaacs addition come into the second ;for example (232).241 However [c~rboxy-~H]- acetic acid and trifluoro[~arboxy-~H]acetic acid add exclusively em e.g. (233). Me Me Me Me \ H C1 (232) The mechanism is shown to involve exo-protonation followed after hydride and Wagner-Meerwein shifts by exo-addition of trifluoroacetate or acetate V \/ ions.242 syn-7-Bromobenzonorbornadiene (234) suffers exclusive endo-attack by diborane and exo-attack by acetic acid again presumably for steric reasons.243 YB‘ 05vCOCH3 Additions to acetylenes have been reviewed.244 The additions of halogen hydracid to propiolic acid follow the rate law rate = k,[CH-CHCO,H] [hal-] and lead to predominantly trans-addition.It is proposed that the slow step 241 H. C. Brown and J. H. Kawakami f. Amer. Chem. SOC.,1970,92 201; H. C. Brown and K-T. Liu f.Amer. Chem. SOC.,1970,92 3502. 242 H. C. Brown J. H. Kawakami and K-T. Liu f. Amer. Chem. SOC.,1970 92 3816. 243 R. Caple and C. S. Ilenda J. Amer. Chem. SOC.,1970 92 3817. 244 E. Winterfeldt Chem. Acetylenes 1969 267. Reaction Mechanisms-Part (ii) in this reaction is the nucleophilic attack of halide ion (I-> Br-> Cl-) on the 0-protonated substrate to yield a carbanion (235) which rapidly shifts a proton.245 The addition of acyl chloride-aluminium chloride complex to (235) acetylenes leads to mixtures of p-chloro-ap-unsaturated ketones.246 The cis:trans ratio from benzoyl chloride and 3-hexyne is 6.6 decreasing in polar solvents;those for propionyl chloride and acetyl chloride with the same alkyne are 2.5 and 0.33.Transition states (236) and (237) are proposed for the cis-and pans-addition processes respectively ;the latter resembles an internally-stabilised d+ - - -AlCI R-c.. I* i *Cl 8 I, vinyl cation hence it might be surmised that more trans product would result if R = Ar. The rates of HC1 addition to phenylallene (238a) 1-methyl-3-phenyl- allene (238b) and 1-methyl-1-phenylallene (238c) are in the ratio 1 :200 :4000 (a) R' = RZ= H; (b) R' = Me R2 = H; (c) R' = H RZ= Me from which it is inferred that the transition state resembles the localised ally1 cation (239) although no p value was Vinyl cations are inferred as intermediates in the addition of DCl to allene from the structures of the products (total 0-83 %) (240) (241) and the cyclobutane (242) formed by cyclo- addition of the vinyl cation to alle~~e.~~' The bromination of 4-substituted 245 K.Bowden and M. J. Price J. Chern. SOC.(B) 1970 1466 1472. 246 H. Martens and G. Hoornaert Tetrahedron Letters 1970 1821. 24' T. Okuyama K. Izawa and T. Fueno Tetrahedron Letters 1970 3295. 248 B. S. Charleston C. K. Dalton S. S. Washburne D. R. Dalton and S. R. Schroeder Tetrahedron Letters 1969 5 147. I48 N. S. Isaacs stilbenes gives a curved Hammett plot showing that there is no sharp mechan- istic change but that the reagent attacks both vinylic carbons competitively throughout a wide range of sub~tituents.~~’ No rate enhancement at n = 3 was observed250 in the bromination and iodination of a series of olefins (243,n = 1-4) which might have been expected in the case of bromine stabilisation of the (243) (244) carbonium ion for n = 3 (244).Of course in this system bromine is not a par- ticularly efficient neighbouring group and (Br-3) stabilisation may indeed be more effective than (Br-5). None the less there seems every reason for seeking examples of anchimeric assistance during addition reactions. An example probably occurs in the addition of trifluoroacetic acid to the acetylene (245) which yields a rearranged bromotrifluoroacetate (246).” Michael addition of CG CF,CO,H H*ococF3 ,c/-71 cQ-+ R’ Br H’h B,’ R Br (245) (246) alkoxide to activated olefins is enhanced by polar aprotic solvents and retarded by hydrogen-bonding species.252 For alkoxide addition the rate law k[olefin] [OR-] rate = [ROHY is observed.The rate is clearly related to the effective nucleophilicity of the 249 M.-F. Ruasse and J.-E. Dubois Tetrahedron Letters 1970 1163. 2so E. Bienvenue-Goetz J.-E. Dubois D. W. Pearson and D. H. L. Williams J. Chem. SOC.(B) 1970 1275. 251 P. E. Peterson R. J. Bopp and M. M. Ajo J. Amer. Chem. SOC.,1970,92 2834. 252 B. A. Feit and Z. Bigon J. Org. Chem. 1969 34 3942. Reaction Mechanisms-Part (ii) 149 alkoxide ion. The parameters obtained from observations of the pressure depen- dence of the rate of acid-catalysed hydration of acrylic acid have been interpreted as ruling out a slow attack by water on a protonated substrate (247).The choice left is between a simultaneous process or a rate-determining protonation involv- ing a tautomeric change.253 The order of reactivity towards Michael addition OH HO CH,=CHCO,H H’ CH2=CH-Cf+ -&+CHz-CHZC02H \-I OH OH (247) of morpholine and pyrrolidine to CH,=CHX is shown to be X = PhCO > PhS02 > CHO > MeCO > C02Ph > CN > CONH2,254but is not identical to the order of carbanion stabilisation found for instance in Elcb elimination (ref. 211). Isotope effects (a,P-cis,and p-trans) for the addition of styrene to tetracyano- ethylene oxide (248) are identical indicating a process in which the two new bonds are equally developed in the transition state.,’ Homolytic processes are presumed to occur in the alkyl hypobromite addition to olefins (249) because the orientation of addition is affected by oxygen.256 Additions to olefins PhCH=CH p? PhCH-CH2 NC,HF-~H;CN NC /CN -3 NC\; j/CN --* c-c c--c c\ ,c NC’ 0 ‘CN NC’ &’ \CN NC’ ‘CN (248) of iodine azide and iodine nitrate INO, occur by stereospecific routes to the trans adducts no doubt by way of cyclic iodonium ions but bromine azide adds non-stereo~pecifically.~ Chlorine acetate is less stereospecific than 732 chlorine in its additions to 01efins.~~’ Whether a halonium ion is able to form or whether a cis molecular addition occurs may well depend on the ease of leaving of the nucleophile.The bromonium ion which is produced by the action of N-bromosuccinimide on stilbene (250) in dimethylformamide reacts with the latter to give a bromo-formate (251).260 New addition processes which have ”’ S.K. Bhattacharya and C. K. Das J. Amer. Chem. SOC.,1969,91 6715. 254 H. Shenhav 2.Rappoport and S. Patai J. Chem. SOC.(B) 1970,469. ’ W. F. Bayne and E. Snyder Tetrahedron Letters 1970 2263. 25b V. L. Heasley C. L. Frye G. E. Heasley K. A. Martin D. A. Redfield and P. S. Wilday Tetrahedron Letters 1970 1573. 257 A. Hassner F. P. Boerwinkel and A. B. Levy f. Amer. Chem. SOC.,1970,92 4879. U. E. Diner and J. W. Lourn Chem. Comm. 1970 333. 259 P. B. D. de la Mare C. J. O’Connor M. J. Rosser and M. A. Wilson Chem. Comm. 1970 731. 260 D. R. Dalton R. C. Smith and D. G. Jones Terrahedron 1970 26 575.150 N. S. Isaacs Bu LMeOBr_ By-/+ Bu)7 (249) Me0 Br Br OMe absence of O2 29 71 % presence of O2 71 29 % + OCH=NMe, /Ph / /o-cH CH=CH + CHBr-CH / NBS / + /CHBr-CH Ph Ph \Ph Ph \Ph (250) (251) been reported include those of mercury trifluoroacetate (252):261 K = 826 BuCH=CH + Hg(OCOCF,) ,a BuCHCH,HgOCOCF (?) I OCOCF, (252) chromyl chloride for which the intermediate is (253):262 R R,C-CH R2C-CH2 I/ 0 0 II Or O\CCo'I+ RCH2C-R I R H Cr OCl CI/\c1 (253) and N-chlorocarbamates in the presence of chromous the product being (254) 0 4 CrCl + RO-C-NHCI (254) cis:trans = 2-6.7 Carbonyl Reactions.-Evidence has been presented that many B-keto-esters undergo hydrolysis by the Elcb mechanism.264 Loss of the a-proton is followed by p-elimination of carboxylate ion to give an intermediate keten (255) which rapidly hydrolyses.The existence of secondary deuterium isotope effects on acetal and orthoformate (256) hydrolysis is typical of carbon undergoing a hybridisation change in the transition state.265 The hydrolysis of orthoacetates 26' H. C. Brown and M. H. Rei Chem. Comm. 1969 1296. 262 F. Freeman P. D. McCart and N. J. Yamachica J. Amer. Chem. SOC.,1970,92,4621. 263 U. Lessard and J. M. Paton Tetrahedron Letters 1970 4883. 264 R. F. Pratt and T. C. Bruice J. Amer. Chem. SOC.,1970,92 5956. z65 H. Bull T. C. Pletcher and E. H. Cordes Chem. Comm. 1970 527. Reaction Mechunisms-Part (ii) 151 0 0II IIR-C-CH2C-OAr 0 -0 I '--II5 R-C-CH-LC-Ar 0 I1 + R-C-CH=C=O + OAr- (255) R-CCH2C02H (257) is preceded by a rapid alkoxy-group exchange,266 and a slow deuterium (D)H ,OEt kH:k = 1.15 s'.\ NO2 (256) uptake from the solvent in the alkyl moiety. Acid-catalysed cleavage of ortho- esters yields oxocarbonium ions (258) whose reactivity towards amines (com- CH,-C-OR \ \ OR \OR 0R' (257) etc. R' = R or C2D petitively with water) depends upon their base strength.267 The pH-rate profile for the ammonolysis of some or-substituted phenyl acetates has been interpreted MeO + ,OMe C Ph C0,Me Ht Ph CONHR (258)MeOH as indicating the occurrence of a carbanion intermediate at moderate to high pH (259).268Tetrahedral intermediates are inferred in the aminolysis of phenyl (259) L.R. Schroeder J. Chem. SOC.(B) 1970 1789. z6' K. Koeler and E. H. Cordes J. Amer. Chem. SOC.,1970,92 1576. "' T. C. Bruice A. F. Hegarty S. M. Felton A. Danzel and N. G. Kundu J. Amer. Chem. SOC.,1970,92 1370. /OR C,D,OD +-' /OC2D /OR' CH3C-OR CHZDC-OR' 152 N. S. Isaacs acetates from rate data269 and directly observed by spectrophotometry in hydrolysis of the amidine (260).270The hydrolysis of (261) occurs with no loss n OH-A A Ph-N+N-Ph PhN NPh * NPh PhlN\H A, 255 nm c=o HkbH / (260) H of and rates of vinyl ether hydrolysis correlate with a*.272Hydrolyses of the cyclic sultones (262) show a moderate dependence on electronic influences (p = 1-23).273 (261) (262) The hydrolyses of some thiochloroformates exhibit activation parameters characteristic of an SN1reaction; for example the change in heat capacity AC; = -360 J mol-’ deg-1.274 26’ F.M. Menger and J. H. Smith Tetrahedron Letters 1970 4163. 270 D. R. Robinson J. Amer. Chem. SOC.,1970,92 3138. ’” L. H. Brannigan and D. S. Tarbell J. Org. Chern. 1970 35 693. 272 B. A. Trofinov I. S. Emel’yanov A. S. Atavin B. V. Prokop’ev and A. V. Gusarov Reakts. spos. org. Soedinenii 1969 334 351. 273 0. R. Zabarski and E. T. Kaiser J. Amer. Chem. SOC.,1970,92 860. 274 A. Queen T. A. Naur M. N. Paddon-Row and K. Preston Canad. J. Chem. 1970 48 522.