3 Theoretical Chemistry By M. GODFREY Department of Chemistry The University Southampton SO9 5NH 1 Introduction In recent years it has become common for the results of applications of standard computational methods to problems in specialized fields to be included in reports on those fields. Consequently my main criterion for selecting items for this chapter has been the perceived interest to organic chemists as a whole. Most of the relevant material divides quite well into two main categories (i) that which concerns innova- tions in and criticisms of theoretical models the concepts used in them and rules generated from them and (ii) that which gives information on molecular structures and reaction mechanisms through the application of established models.Occasionally a particular aspect of organic chemistry catches the imagination of theoreticians. This year it was the aromaticity of a C60 cluster Buckminsterfullerene or footballene a species which may have been formed in the laser vaporization of graphite and by natural processes in stars. An amusing account of the background to this problem has been given by Kroto.’ My philosophical highlight of the year was an article in which Turro’ explored the intellectual and scientific foundations of organic chemistry and of the methods that provide the field with a platform for making rapid conceptual and experimental advances. It was proposed that these methods involve a geometric and topological approach to scientific reasoning within the framework of scientific paradigms that guide experimental design and execution.2 Models Concepts and Rules Computational Models.-Many organic chemists are confused by the wide variety of basis sets used in MO calculations. Davidson and Feller3 have reviewed basis sets of the type known as ‘contracted Cartesian Gaussians’ from the point of view of the non-specialist who is interested in a practical guide to what is generally available and of the pitfalls to avoid. Ichikawa and his co-workers4 have compared 22 different basis sets for the calculation of molecular electronic structures. They conclude that polarization functions must be added to both heavy atoms and hydrogen atoms in order to produce well-balanced wave functions. Marriott and Topsom’ have found that calculations of barriers to rotation in some mono-’ H.Kroto Roc. R. Znst. 1986 58 45. N. J. Turro Angew. Chem. Znt. Ed. Engl. 1986 25 882. E. R. Davidson and D. Feller Chem. Rev. 1986 86 681. H. Ichikawa K. Sameshima and Y. Ebisawa Bull. Chem. SOC.Jpn. 1986 59 2729. S. Marriott and R. D. Topsom Aust. 1. Chem. 1986 39 1157. 17 18 M. Godfrey substituted benzenes are in better agreement with experiment when minimal basis sets are used for both energy and geometry optimization than when split-valence basis sets are used. Progress made before 1986 on the problem of reducing the errors associated with basis set truncation has been reviewed by Wilson.6 The efficient computation of electron correlation effects is a continuing challenge. Developments include a relatively simple method for calculating correlation energies in large organic molecules which offers a good physical insight into the phenomenon of electron ~orrelation,~ and the application of a novel method to acetylene ethylene and ethane which shows that specific correlation patterns arise from different kinds of bond.8 Locating transition states on potential energy surfaces is also a difficult computa- tional problem.Baker’ has designed a new algorithm for use within the GAUSSIAN 82 program package. It is capable of locating transition states even if the search is started in the wrong region of the surface. Semi-empirical MO methods involving the complete neglect of differential overlap (CNDO) are now rarely used. However some methods involving a degree of NDO are still quite popular.Their usefulness relative to ab initio methods continues to be questioned. Baird and Hadley” have claimed that the apparent superiority of MNDO over ab initio in predicting molecular energetics disappears when the methods are compared on the same basis and McKee Shevlin and Rzepa” have concluded that MNDO and MNDOC cannot be relied upon for the energetics of proton transfer between oxygens or between nitrogens. More sympathetically Lewis” has reviewed the MIND0/3 method Dewar and Reynold~’~ have reported an improved set of MNDO parameters for sulphur and Dewar and DieterI4 have claimed that AM1 seems to be an effective method for studying processes involving deprotonation or protonation of neutral molecules.Schroder and Thiel” have provided some general guidelines for MNDOC studies of thermal organic reactions following a comparative study of correlation effects on activation energies and transition state structures. Although parametrized for neutral closed-shell molecules both MIND0/3 and MNDO methods work well overall in reproducing experimental heats of formation of closed-shell and open-shell cations containing C H N and 0; there are however failures with certain types of structure.16 Regarding other methods for computing molecular structure Lifson” has reviewed the theoretical foundation of the Empirical Force Field approach and Saunders and Jarret18 have described a new method for optimizing molecular structures in which the terms involving bond angles and torsional angles used in the force fields S.Wilson Adu. Chem. Phys. 1987 67 439. A. M. Oles F. Pfirsch P. Fulde and M. C. Bohrn J. Chem. Phys. 1986 85 5183. G. Stollhoff and P. Vasilopoulos J. Chem. Phys. 1986 84 2744. J. Baker J. Comput. Chem. 1986 7 385. 10 N. C. Baird and G. 0. Hadley Chem. Phys. Lett. 1986 128 31. M. L. McKee P. B. Shevlin and H. S. Rzepa J. Am. Chem. Soc. 1986 108 5793. 12 D. F. V. Lewis Chem. Reu. 1986 86 1111. l3 M. J. S. Dewar and C. H. Reynolds J. Comput. Chem. 1986 7 140. 14 M. J. S. Dewar and K. M. Dieter J. Am. Chem. Soc. 1986 108 8075. 15 S. Schroder and W. Thiel J. Am. Chem. SOC.,1986 108 7985. 16 H. Halim N. Heinrich W. Koch J. Schmidt and G. Frenking J. Comput. Chem. 1986 7 93. 17 S.Lifson Gazz. Chim. Ztal. 1986 116 687. 18 M. Saunders and R. M. Jarret J. Comput. Chem. 1986 7 578. Theoretical Chemistry 19 of conventional molecular mechanics are replaced with two-body central forces between atoms with considerable computational advantage. Bonding.-The nature of the covalent bond is not as simple as most textbook discussions s~ggest.'~~~' The accumulation of electron density appears weak or even absent in certain covalent bonds. An explanation for this anomaly is that the structures of some atoms change dramatically as a first step in the bond-formation process.21 Valence bond calculations2' without the imposition of any orthogonality constraints show the equivalent hybrid (banana) description of the double bond in ethylene to be more stable than a u plus 7r bond description.Theories of bonding in charge-transfer complexes have been reviewed.23 Linear conjugation is preferred to cross conjugation for systems of four p-orbitals containing four 7r Studies of the structures and stabilities of fluorinated carbanions provide evidence for negative hyperconjugation by the CF3 group.2s It is known that in pentalene and heptalene an alternating bond structure is more stable than a delocalized bond structure whereas in each corresponding radical anion the opposite is the case. It has been demonstrated26 that electron-donating substituents at some sites tend to relax bond alternation but such substituents at other sites enhance bond alternation. The concept of bond fixation provides no help in understanding the geometry of cyclopropabenzene a molecule which consists of fused cyclopropene and benzene rings.27 Molecular Shape and Size.-Meyer28 has reviewed some of the procedures used to estimate molecular size andz9 has extended his own contribution to the field by proposing new shape and bulk descriptors to deal with steric effects of substituents on reaction rates.Richards3' has developed an approximate method for calculating molecular similarity based on the notion of standard electron densities for molecular fragments. Theory of Reactions.-Papers given at an international conference on the theory of organic reactions held at Gargnano Italy in June 1985 have been published under the editorship of Bernardi.31 Bernardi and Robb3' have analysed the cycloaddition of two ethylene molecules in terms of their own diabatic surface model and have shown that the behaviour of the constituent surfaces can be easily rationalized by using simple MO energy expressions.19 N. C. Baird J. Chem. Educ. 1986 63 660. 2o K. Jug N. D. Epiotis and S. Buss J. Am. Chem. SOC.,1986 108 3640. 21 K. L. Kunze and M. B. Hall J. Am. Chem. SOC.,1986 108 5122. 22 W. E. Palke J. Am. Chem. SOC.,1986 108 6543. 23 C. J. Bender Chem. SOC.Rev. 1986 15 475. 24 S. Inagaki K. Iwase and N. Goto J. Org. Chem. 1986 51 362. 2s D. A. Dixon T. Fukunaga and B. E. Smart J. Am. Chem. SOC.,1986 108 4027. 26 M. Kataoka T. Ohmae and T. Nakajima J. Org. Chem. 1986 51 358. 2' Y. Apeloig and D. Arad J.Am. Chem. SOC.,1986 108 3241. 28 A. Y. Meyer Chem. SOC.Rev. 1986 15 449. 29 A. Y. Meyer J. Chem. SOC.,Perkin Trans. 2 1986 1567. 30 E. E. Hodgkin and W. G. Richards J. Chem. SOC.,Chem. Commun. 1986 1342; P. E. Bowen-Jenkins and W. G. Richards Int. J. Quantum Chem. 1986 30 763. 31 THEOCHEM 1986 31 Nos. 1/2. 32 F. Bernardi M. Olivucci M. A. Robb and G. Tonachini J. Am. Chem. SOC.,1986 108 1408. 20 M. Godfrey An alternative to the FMO method for predicting regiochemical preferences has been presented by Hehre and his co-worker~.~~ Measures of differential reactivity of the various sites in nucleophiles towards electrophiles and in electrophiles towards nucleophiles are obtained by allowing the test electrophile H+ and the test nucleophile H- to interact with the electron density surfaces of the substrates.Preferred sites of interaction between actual nucleophiles and electrophiles are then predicted by matching these differential reactivities. Alston and his co-~orkers~~ have demonstrated that the regio-selectivity in the Diels- Alder reaction of 1 -substituted- 1,3-butadienes can be completely explained by FMO theory when the FMOs of the reactants in the concerted transition state as opposed to the ground state are considered. Bachrach and Streitwieser3’ have concluded that the FMO approximation of HOMO-LUMO interactions dominating the transition state is too severe HOMO-HOMO interactions are also important. The configuration mixing model of Pross and Shaik36 has been used by Pros3’ to classify polar reactions as ‘allowed’ and ‘forbidden’.Thus direct nucleophilic attack is forbidden with respect to radical cations but allowed with respect to dications. The same model has been used by Shaik3* to give a unified understanding of the factors controlling the mechanistic choice of a given reactant pair the variation of the reaction barriers and the geometries of the transition states in a set of nucleophilic substitution reactions. The dynamical excitation of high vibrational states in the direct scattering process has been given as the most probable reason for the low efficiency of gas-phase SN2 reactions with no activation barriers.39 Increasing understanding of biological redox systems has led to the need for detailed information regarding the effects of mutual orientation and separation distance of the redox partners on the rate of electron transfer.Marcus4’ has developed a semi-classical model for orientation effects and4’ has given a theoretical treatment for the effects of intramolecular vibrational and diff usive solvent orientational motions. War~hel~~ has developed a semi-classical trajectory approach to the micro- scopic simulation of electron transfer reactions in polar solvents. Aromaticity.-One of the most nebulous concepts in organic chemistry is aromaticity. Everyone is agreed that benzene is highly aromatic but even in this case the electronic basis of the property is uncertain. Cooper Gerratt and Rairn~ndi~~ have argued that the T electrons of benzene are essentially localized and that the aromatic characteristics arise from the symmetric coupling of the electron spins around the 33 S.D. Kahn C. F. Pau L. E. Overman and W. J. Hehre J. Am. Chem. SOC.,1986 108 7381; S. D. Kahn C. F. Pau and W. J. Hehre ibid. 1986 108 7396; S. D. Kahn and W. J. Hehre ibid. 1986 108 7399. 34 P. V. Alston R. M. Ottenbrite 0. F. Guner and D. D. Shillady Tetrahedron 1986 42 4403. 35 S. M. Bachrach and A. Streitwieser J. Am. Chem. Soc, 1986 108 3946. 36 A. Pross and S. S. Shaik Acc. Chem. Res. 1983 16 363. 37 A. Pross J. Am. Chem. SOC.,1986 108 3537. 38 D. Cohen R. Bar and S. S. Shaik 1. Am. Chem. SOC.,1986 108 231. 39 M. V. Basilevsky and V. M. Ryaboy Chem. Phys. Lett. 1986 129 71. 40 R. J. Cave S. J. Klippenstein and R.A. Marcus J. Chem. Phys. 1986 84 3089. 41 H. Sumi and R. A. Marcus J. Chem. Phys. 1986 84,4894. 42 A. Warshel and J.-K. Hwang J. Chem. Phys. 1986 84 4938. 43 D. L. Cooper J. Gerratt and M. Raimondi Nature (London) 1986 323 699. Theoretical Chemistry 21 carbon ring framework. BairdU has argued against the conclusion of Hiberty Shaik Ohanessian and Lefour which I reported last year that the symmetric structure for benzene is preferred because of the a-bond energy but the latter authors have disagreed with his arg~ment.4~ Buckminsterfullerene appears to be a spherical aromatic molecule. The many reported calculation^^^ on this species illustrate the difficulties in establishing a precise definition of degree of aromaticity. By the Huckel rule planar cis-[ 12lannulene is anti-aromatic but high level ab initio MO calculations surprisingly indicate a small positive resonance energy.47 The concept of o-aromaticity helps in explaining why cyclopropane has a very similar strain energy to cycl~butane.~~ The role of homoaromaticity in stabilizing appropriate species has been further explored.49 SINDO/ 1 calculations have been used to establish that excited states of antiaromatic molecules can exhibit aromatic ~haracter.~' Finally it has been shown that aromatic transition-state theory and the general rule of the principle of conservation of orbital symmetry can be regarded as the same concept described in different termin~logies.~' Linear Free Energy Relationships and Substituent Effect Analysis.-The disputation between Wold and Sjostrom who regarded LFERs as locally valid linearizations of complicated functional relationships and Kamlet and Taft who believe that LFERs reflect fundamental and rather simple physiochemical relationships con- tinue~.~~ A chemometric investigation of substituent effects on chemical and spectro- scopic data sets leads to the conclusion that individual components of the two effects cannot be determined q~antitatively.~~ There is still little understanding of short-range substituent effects.A statistical analysis54 has demonstrated that Mullay's group electronegativity scale (reported last year) linearly correlates with (J,which is widely regarded as a good measure of the field effect of a substituent.The apparent dissimilarity of the two scales is due to a large constant term in the electronegativity scale. There is no precise relationship between a,and the HCC bond angles at the ips0 carbon in monosub- stituted ethylenes as determined by ab initio MO calculation^.^^ This result is in 44 N. C. Baird J. Org. Chem. 1986 51 3907. 45 P. C. Hiberty S. S. Shaik G. Ohanessian and J.-M. Lefour J. Org. Chem. 1986 51 3908. 46 R. L. Disch and J. M. Schulman Chem. Phys. Lett. 1986 125 465; P. W. Fowler and J. Woolrich ibid. 1986 127 78; R. C. Haddon L. E. Brus and K. Raghavachari ibid. 1986 125 459; 1986 131 165; A. D. J. Haymet J. Am. Chem. Soc. 1986 108 319; B. A. Hess and L. J. Schaad J. Org. Chem. 1986 51 3902; D. J. Klein T. G. Schmalz G. E. Hile and W.A. Seitz J. Am. Chem. Soc. 1986 108 1301; D. S. Marynick and S. Estreicher Chem. Phys. Lett. 1986 132 383; M. D. Newton and R. E. Stanton J. Am. Chem. Soc. 1986 108 2469; A. J. Stone and D. J. Wales Chem. Phys. Lett. 1986 128 501. 47 R. C. Haddon hre Appl. Chem. 1986 58 129. 48 D. Crerner and J. Gauss J. Am. Chem. Soc. 1986 108 7467. 49 P. v. R. Schleyer E. Kaufmann A. J. Kos H. Mayr and J. Chandrasekhar J. Chem. SOC. Chem. Commun. 1986 1583; A. B. McEwen and P. v. R. Schleyer J. Org. Chem. 1986 51 4357. 50 E. J. P. Malar and K. Jug Tetrahedron 1986 42 417. '' Z. Hua-ming and W. De-xiang Tetrahedron 1986 42 515. 52 S. Wold and M. Sjostrom Acta Chem. Scand. Ser. B 1986,40 270; M. J. Kamlet and R. W. Taft ibid. 1986 40 619. 53 S. Alunni S.Clementi C. Ebert P. Linda G. Masumarra U. Edlund M. Sjostrom and S. Wold Gazz. Chim. Ital. 1986 116 679. 54 E. Sacher Tetrahedron Lett. 1986 42 4683. 5s S. Marriott and R. D. Topsom THEOCHEM 1986 32 101. 22 M. Godfrey accord with earlier work on monosubstituted benzenes. An MNDO indicates that inductive effects of a-substituents have no special importance in determining substituent stabilization effects on carbocations. The Marcus relationship between free energy of activation and free energy of reaction has been applied extensively to electron proton and group transfer reac- tions in organic chemistry over the years. A special issue of the Journal of Physical Chemistry” to commemorate the contributions of Marcus to chemistry includes papers from several leading physical organic chemists.This year Form~sinho~~ has presented an extension of the Marcus relationship that involves two new independent variables. Within this extended relationship it is possible to describe the progress of sigmatropic shifts and cycloadditions in terms of one progress variable. Marcus theory breaks down when the bond order of the transition state varies with the nature of the reaction and/or when the effect of the configuration entropy is significant. Elsewhere59 Formosinho has proposed a general inter-relationship between transition state bond extensions and the energy barrier to reaction. The model introduces the concept of a variable bond order over the reaction path if reactants or products have anti-bonding or non-bonding electrons which can have a bonding character at the transition state.That Marcus theory can sometimes be successful in reproducing quantum mechanical data has been demonstrated in an MO study6’ of the effects of a-alkyl substituents on proton transfers between pairs of nitrogen atoms and between pairs of oxygen atoms. Another probe into the relationship between transition state structures and reac- tivities through an MO study of specific reactions this time the additions of amines to carbonyl compounds gives support to the BEP (Bell-Evans-Polanyi) theory in that the activation energy falls as the reaction becomes more exothermic.61 On the other hand transition state structures and kinetic isotope effects are virtually invariant in spite of a wide variation in reactivity.The significance of the Bransted parameter for transition state structure remains a subject for conjecture. Shaik6* has defined a new intrinsic selectivity factor for identity S,2 reactions which yields information about (a) the average looseness of the transition state geometry in a reaction series; and (b) the sensitivity of the reaction series to geometric loosening. Lee and S~hn~~ have given mechanistic significance to values of cross interaction constants pij in an extended Hammett relationship for simultaneous variation of two substituents in different part of a reaction complex. The relative gas-phase acidities and basicities of substituted amines mercaptans alcohols etc. are believed to be strongly dominated by charge/induced-dipole stabilization of the ion formed by protonation or deprotonation.A scale of substituent effects on this factor has been obtained64 from ab initio MO calculations on a model 56 A. M. Aissani J. C. Baum R. F. Langler and J. L. Ginsburg Can. J. Chem. 1986 64 532. 57 J. Phys. Chem. 1986,90 No. 16. 58 S. J. Formosinho Tetrahedron 1986 42 4557. 59 A. J. C. Varandas and S. J. Formosinho J. Chem. SOC.,Faraday Trans. 2 1986 82 953. 6o S. Scheiner and P. Redfern J. Phys. Chem. 1986 90 2969. 61 H. Yamataka S. Nagase T. Ando and T. Hanafusa J. Am. Chem. SOC.,1986 108 601. 62 S. S. Shaik Can. J. Chem. 1986 64 96. 63 1. Lee and S. C. Sohn J. Chem. SOC.,Chem. Commun. 1986 1055. 64 W. J. Hehre C. F. Pau A.D. Headly R. W. Taft and R. D. Topsom J. Am. Chem. Soc. 1986,108 1711. The0retica1 Chemistry 23 system. The variation of the gas-phase basicities of amines can be analysed by using two parameters one global and one local.65 Other Electronic Effects.-It has been concluded66 that merostabilization of carbon- centred radicals is important in media of high dielectric constant but less so in the gas phase. In the gas phase only this merostabilization (or capto-dative) effect has been ~alculated~~ to be greater in nitrogen-centred radicals than in carbon-centred analogues. Ab initio studies68 have indicated that intramolecular long distance electron transfer between A and A2 in [A1-G-A2]- proceeds mostly by through-bond interaction involving the spacer G and does so in a very stereospecific manner.Other ab initio studies69 indicate that relay orbitals play an important role in the determination of the energy levels in molecules containing a pair of perpendicular T systems (e.g. spiropentadiene). New interpretations have been reported for the antiperiplanar effect in eliminations from ethane derivatives and for the cis effect in difluoroethylenes. In the former case,7o continuity of phase of the bonding and anti-bonding orbitals of the antiperi- planar bonds and the bonding orbital of the intervening C-C bond is the key factor. In the latter case,’ attraction of one fluorine nucleus for electrons of the other fluorine atom is considered to be relatively very important. 3 Studies of Molecular Structures and Reaction Paths General.-Applications of computational methods to the study of biological systems have been revie~ed.~~,~~ have continued to produce their Ohno and M~rokuma~~ annual bibliography of ab initio calculations.Ab initio Structures.-The ability to perform calculations with sophisticated basis sets and with allowance for electron correlation permits the reinvestigation of structural problems concerning stable molecules. The preference of 1,2-difluoroethane for the gauche rather than the trans conformation can be explained at the MP2/6-31G** level but not at the Hartree-Fock Polarization functions in the basis set reduce the rotational barriers in formamide and acetamide by about 5 kcal mol-’ at the Hartree-Fock level; electron correlation yields less than 1 kcal-mol-’ further red~ction.~~ The effect of including polarization functions on the geometrical parameters for pyridine has been determir~ed.~’ 65 W.Yang and W. J. Mortier J. Am. Chem. SOC.,1986 108 5708. 66 A. R. Katritzky M. C. Zerner and M. M. Karelson J. Am. Chem. SOC.,1986 108 7213. 67 D. Kost M. Raban and K. Aviram J. Chem. SOC.,Chem. Comrnun. 1986 346. 68 K. Ohta G. L. Closs K. Morokuma and N. J. Green J. Am. Chem. Soc. 1986 108 1319. 69 K. Kanda T. Koremoto and A. Imamura Tetrahedron 1986 42 4169. 70 S. Inagaki K. Iwase and Y. Mori Chem. Lett. 1986 417. 71 M. M. Heaton and M. R. El-Talbi THEOCHEM 1986,32 61. 12 ‘Theoretical Chemistry of Biological Systems’ ed. G. Naray-Szabo Elsevier Amsterdam 1986.73 Y.B. Luzhkov and G. N. Bogdanov Russ. Chem. Rev. (Engl. Transl.) 1986 55 272. 74 THEOCHEM 1986 33 Nos. 3/4. 75 G. F. Smits M. C. Krol P.N. Van Kampen and C. Altona THEOCHEM 1986 32 247. 76 P. G. Jasien W. J. Stevens and M. Krauss THEOCHEM 1986 32 197. 17 D. A. Dixon T. Fukunaga and B. E. Smart J. Am. Chem. SOC.,1986 108 1585. 24 M. Godfrey Molecules containing second row atoms can now be studied by ub initio methods. It has been that d polarization functions on sulphur play an essential role in the formation of the disulphide bond in CH3SSCH3. The conformational behaviour of bithienyls is similar to that of bif~ranyls.'~ Methyl and trifluoromethyl hyperconjugation," and the stabilization of a-substituted carbocations" and car- banions,82 with substituents based on second-row atoms have also been studied.Other species investigated include fluorine substituted phosphonium ylidess3 and the thioxyallyl radi~al.'~ Dicationic species have attracted much attention. The linear structure of C4H22f is markedly less stable than the 4-membered ring.85 Back donation of fluorine lone-pair electrons is very important in determining the shape of fluorine-substituted ethylene dications.86 Bipyridine dications have markedly twisted conformations while the corresponding radical cations are nearly planar this may be significant for herbicidal action.87 Barriers to the dissociation of ylide dications H2C=XH2+ (X = NH, OH F PH2 SH and C1) are substantial and hence these species may be observable.88 Lischka and KarpfenS9 have begun a systematic study at different levels of sophistication of electronically excited states in the polyene and polyyne series.A study of the low-lying electronic states of ketene has added to the understanding of the dissociation of ketenes to carbenes." There has been continuing interest in carbenes" and diradi~als,~ and in solvent- solute interaction^.^^ Ab initio Potential Energy Surfaces Transition States and Mechanisms.-Bernardi and R~bb~~ have published an essay on transition state structure computations and their analysis covering the pre- 1986 literature. The mechanisms of pericyclic reactions have continued to attract interest. There is more evidence for the synchronous concerted mechanism in the Diels-Alder reaction of butadiene with eth~lene.9~ The cycloaddition of HF to ethylene is easier 78 M.Honda and M. Tajuna THEOCHEM,-1986,29,93. 79 V. Barone F.Lelj N. Russo and M. Toscano J. Chem. SOC.,Perkin Trans. 2 1986 907. 80 E. Magnusson J. Am. Chem. SOC.,1986,108,11. F. Bernardi A. Bottoni and A. Venturini J. Am. Chem. SOC. 1986 108 5395. 82 F. Bernardi A.Bottoni A. Venturini and A. Mangini J. Am. Chem. SOC.,1986 108 8171. D. A. Dixon and B. E. Smart J. Am. Chem. SOC.,1986 108 7172. 84 T. Furuhata and W. Ando Tetrahedron Lett. 1986,27 4035. 85 K. Lammertsma J. A. Pople and P. v. R.Schleyer J. Am. Chem. SOC.,1986,108 7. 86 G. Frenking W. Koch and H. Schwarz J. Comput. Chem. 1986,7,406. H.-J. Hofmann R.Cimiraglia and J.Tomasi THEOCHEM,1986,32,213. 88 B. F.Yates W. J. Bouma and L. Radom. J. Am. Chem. Soc.. 1986 108 6545. 89 H.Lischka and A. Karpfen Chem. Phys. 1986 102,77; A. Karpfen and H. Lischka ibid. 1986,102.91. 90 W. D. Allen and H. F. Schaefer J. Chem. Phys. 1986,84 2212. 91 E.A.Carter and W. A. Goddard J. Phys. Chem. 1986,90,998;D. A.Dixon ibid. 1986,90 54; G. E. Scuseria M. Duran R.G. A. R.Maclagan and H. F. Schaefer J. Am. Chem. SOC.,1986,108 3248; H. Tomioka T. Sugiura Y. Masumoto Y. Izawa S. Inagaki and K. Iwase J. Chem. Soc. Chem. Commun. 1986 693; J. S. Yadav and J. D. Goddard J. Chem. Phys. 1986,85 3975. 92 P. Du D. Hrovat and W. T. Borden J. Am. Chem. SOC.,1986 108 8086. 93 C. A. Deakyne M. Meot-Ner C. L. Campbell M. G. Hughes and S. P. Murphy J.Chem. Phys. 1986 84 4958;J. Gao D. S. Gamer and W. L. Jorgensen J. Am. Chem. SOC.,1986 108 4784; P. G.Jasien and W. J. Stevens J. Chem. Phys. 1986,84 3271. 94 F. Bernardi and M. A. Robb Adu. Chem. Phys. 1987 67 155. 95 K. N. Houk Y.-T. Lin and F. K. Brown J. Am. Chem. SOC.,1986,108,554. Theoretical Chemistry 25 with the H-Fs-aH-F dimer than with a single H-F m01ecule.~~ There is no barrier in the gas phase to the symmetry forbidden [3 + 21 cycloaddition of the butadiene radical cation to eth~lene.~' Dewar's suggestion which I reported last year that tunnelling from a twisted form of the reactant should play an important part in the mechanism of the thermal [1,5]H shift in cis-1,3-pentadiene is strongly supported the reaction is found most likely to proceed via a transition state of C2 symmetry.98 Ab initio MO calculations support the predictions of resonance theory for the preferred protonation sites in furan and in vinyl alcohol.99 The proton-transfer properties of C=O in the carboxyl group of formic acid are very similar to those of C=O in formaldehyde.The primary effect of the OH group is to increase the proton affinity of the C=O. The proton affinity of OH is lowered by its association with C=O.loo Formic acid pyrolyses to yield carbon monoxide and carbon dioxide by competing processes. It appears that the water molecules produced along with the carbon monoxide catalyse the formation of carbon dioxide."' Calculated substituent effects on neutral and ionized C=C and C=O bonds imply that the variation of the keto :enol ratio for first-row substituents ought to be very different with radical cations than with neutral compounds.'"2 The enhanced rate of sN2 displacement of chloride ions due to neighbouring group effects is similar in origin in a-chloroacetaldehyde and in ally1 ch10ride.l'~ A PMO analysis of ab initio results for identity &2 reactions involving various a-substituents and various nucleophiles/leaving groups indicates that the T interac-tions between the a-substituent and an occupied orbital associated with the reaction co-ordinate axis are of major importance in governing the transition state energy.lo4 The activation energy for 0-alkylation of acetaldehyde enolate by methyl fluoride is lower than that for C-alkylation even though the latter is favoured thermodynami- cally.This and other results imply that a free enolate reacts exclusively at oxygen regardless of the hardness or softness of the alkylating agent or whether the transition state is early or late. The preference for C-alkylation of metal enolates in solution must arise from co-ordination of the enolate oxygen by metal cations which is known to decrease the reactivity of the oxygen site relative to the carbon site.'" Arylcarbenes readily undergo intramolecular reactions with nucleophilic groups in ortho substituents whereas simple benzyl cations do not in spite of similar intermolecular reactivities of carbenes and cations. This dramatic difference has been attributed to considerably lower barriers to rotation about the bond connecting the sp2 carbon atom to the ring in the arylcarbenes.'06 96 C.Clavero M. Duran A. Lledos 0. N. Ventura and J. Bertran J. Am. Chem. Soc. 1986 108 923. 97 D. J. Bellville and N. L. Bauld Tetrahedron 1986 42 6167. 98 G. J. M. Dormans and H. M. Buck 1.Am. Chem. Soc. 1986 108 3253. 99 M. T. Nguyen A. F. Hegarty T.-K. Ha and G. R. De Mare J. Chem. Soc. Perkin Trans. 2 1986 147. 100 E. A. Hillenbrand and S. Scheiner J. Am. Chem. Soc. 1986 108 7178 101 P. Ruelle U. W. Kesselring and H. Nam-Tran J. Am. Chem. SOL 1986 108 371. 102 N. Heinrich W. Koch G. Frenking and H. Schwarz J. Am. Chem. SOC.,1986 108 593. 103 R. D. Bach B. A. Coddens and G. J. Wolber J. Org. Chem. 1986 51 1030. 104 D. Kost and K. Aviram J.Am. Chem. SOC.,1986 108 2006. K. N. Houk and M. N. Paddon-Row J. Am. Chem. Soc. 1986 108 2659. 106 W. Kirmse. K. Kund E. Ritzer A. E. Dorigo and K. N. Houk J. Am. Gem. SOC. 1986 108 6045. 26 M. Godfrey Studies of potential energy surfaces indicate that the dication of formic acid is not stable and that C(OH):+ is the global minimum;'07 that H2NCH2+ and H3NC2+ are the stable species on the NCH3'+ surface;lo8 and that CH=CH-C=O' is a stable species on the C3H20t s~rface.''~ Calculations with biological significance include the reaction of hydroxyl radical with cytosine;"' and the oxidation of triose reductone which has the same functional group as L-ascorbic acid."' Semi-empirical.-The emphasis in reported semi-empirical MO calculations has been on mechanism rather than structure.The main methods used have been MIND0/3 MNDO MNDOC and AM1. Cycloadditions have been studied with the following results. The Diels- Alder reaction of butadiene cannot be synchronous with cyanoethylenes but the mechanism remains uncertain with ethylene itself."' The transition states for the Diels- Alder reaction of acrolein with cyclopentadiene and several monosubstituted derivatives of butadiene have been located there is a possibility of a two-step mechanism where charge-transfer is great. The results are in good agreement with experimental observa- tions regarding selectivity and substituent effects on reaction rate.' l3 The thermal 2,5-cycloaddition of alkenes to phenols can occur only when strained cyclic alkenes are empl~yed."~ An asynchronous mechanism has been proposed for the 1,3-dipolar cycloaddition of substituted cyclic azomethine ylides to chloroacrylonitrile the minimum potential energy surface explains the observed regioselectivity.' l5 The cycloaddition of cyclopentyne to ethylene proceeds via a very short-lived diradical intermediate.lI6 In the tetrahedral intermediate in the BAc2 hydrolysis of an ester acyl-oxygen bond dissociation preceeds hydroxylic hydrogen tran~fer."~ In the acid-catalysed hydrolysis of acetamide rate-determining nucleophilic attack of water at the carbonyl group occurs in the N-protonated species rather than in the 0-protonated tautomer.The resulting tetrahedral species are not at the energy minima but at or near the saddle points."* The mechanism for ips0 attack in the nitration of certain aromatic molecules involves radical pair recombination rather than classical electrophilic substitution.' l9 Calculations of potential energy surfaces for the interaction between peroxyformic acid and ethylene or substituted ethylenes do not support the electrophilic mechan- isms that are currently favoured.Instead they suggest that peroxyformate anions should add to olefins. With electron-withdrawing substituents in the olefins there is lo' W. Koch and H. Schwarz Chem. Phys. Lett. 1986 125 443. W. Koch N. Heinrich and H. Schwarz J. Am. Chem. SOC.,1986 108 5400. 109 G. Bouchoux Y. Hoppilliard J.-P. Flament J. K. Terlouw and F. van der Valk J. Phys. Chem. 1986 90 1582.110 B. G. Eatock W. L. Waltz and P. G. Mezey Can. J. Chem. 1986 64 914. 111 Y. Abe H. Horii S. Taniguchi S. Yamabe and T. Minato Can. J. Chem. 1986 64,360. 112 M. J. S. Dewar S. Olivella and J. J. P. Stewart 1.Am. Chem. SOC.,1986 108 5771. 113 V. Branchadell A. Oliva and J. Bertran THEOCHEM 1986 29 25. 114 J. Amau J. Fernandez and P. Yianni J. Chem. SOC.,Perkin Trans. 2 1986 2013. M. C. Cardozo M. T. Pizzorno S. M. Albonico and A. B. Pierini Tetrahedron 1986 42 5857. 116 S. Olivella M. A. Pericas A. Riera and A. SolC J. Chem. SOC.,Perkin Trans. 2 1986 613. 117 J. J. Maraver E. S. Marcos and J. Bertran J. Chem. SOC.,Perkin Trans. 2 1986 1323. 118 I. Lee C. K. Kim and H. S. Seo Tetrahedron 1986,42 6627. J. Feng X. Zheng and M.C. Zerner J. Org. Chem. 1986 51 4531. Theoretical Chemistry no activation energy. With electron-donating substituents a two-step mechanism is predicted involving the formation of a metastable intermediate and rate-determining epoxide ring closure.'2o The MIND0/3 potential energy surface'21 for the ethylene dication is not similar to the ab initio surface.'22 Other Computational Procedure~.-Jorgensen'~~has used a novel combination of ab initio quantum mechanics and statistical perturbation theory to study the effect of hydration on the structure of the transition state for the SN2exchange reaction of C1- and CH,Cl. The effect is found to be very small. In earlier work on this reaction he had found that the activation barrier induced by hydration results primarily from reduction in strength rather than reduction in the number of solute- water hydrogen bonds upon proceeding to the transition state.This year he has obtained a similar result for nucleophilic attack of hydroxide ion on the carbonyl carbon of f0rma1dehyde.l~~ Stimulated by the importance of -NH,+ and -COO- groups in molecules of biological interest and the role of solvation by water in influencing the interactions of these groups K01lman'~~ has initiated a Monte Carlo simulation study of the solvation of these groups using CH3NH3' and CH3COO- as models. The nature of the water structure around the methyl group of CH,NH,+ is significantly different from that around the methyl group in CH,COO-. The calculations reported in this sub-section involve considerable amounts of computer time and approach the limit of what can reasonably be attempted at the present time.I20 J. Kaneti Tetrahedron 1986 42 4017. 121 M. J. S. Dewar and C. H. Reynolds THEOCHEM 1986 29 209. 122 K. Lammertsma M. Barzaghi G. A. Olah J. A. Pople A. J. Kos and P. v. R. Schleyer J. Am. Chem. Soc. 1983 105 5252. 123 W. L. Jorgensen and J. K. Buckner J. Phys. Chem. 1986,90,4651. 124 J. D. Madura and W. L. Jorgensen J. Am. Chem. SOC.,1986 108 2517. 125 G. Alhgona C. Ghio and P. Kollman J. Am. Chem. SOC.,1986 108 185.