3 TheoreticaI Chemistry By C. THOMSON Department of Chemistry University of St. Andrews St. Andrews KY169ST 1 Introduction A useful new textbook on modern quantum chemistry has appeared,' and recent reviews in this area include a review on hybrid orbitals,* on the axioms of quantum the~ry,~ orbital interactions through bond^,^ the application of molecular mechanics calculations in organic chemistry,' and intermolecular forces.6 A new bibliography of ub initio calculations has appeared covering the period 1978-1980,' and a recent issue of the THEOCHEM journal was denoted to an update of the latter through 1981.8 The use of large vector computers such as the CRAY-1 in quantum chemistry has been reviewed by Saunders and Guest.' This review will necessarily be rather subjective and is mainly restricted to the use of ub initio methods in chemistry.2 Advances in Theoretical Techniques Basis Sets.-Pople's group have developed several new useful basis sets suitable for applications to second-row atoms. 6-21G and 3-21G basis sets for Na to Ar,'" analcgous to their first-row counterparts," have been tested on numerous molecules containing second-row atoms and the 3-21G basis set in particular yields equili- brium geometries vibrational energies relative energies and dipole moments that are much superior to STO-3G values at relatively modest cost. Two basis sets supplemented with polarization functions for the second row have also been described the first a 3-21G* basis set (added d-functions on second-row elements only) and this gives a much improved description of hypervalent molecules,1z and the second are the large 6-31G* and 6-31G** basis These 'Modern Quantum Chemistry' ed.A. Szabo and N. S. Ostlund MacMillan Publishing Co. New York 1982. 'W. A. Bingel and W. Luettke Angew. Chem. 1981,93 944. ' P.-0. Lowdin Adv. Chem. Phys. 1982 21 275. M. N. Paddon-Row Acc. Chem. Res. 1982 15,245. ' E. Osawa and H. Musso Top. Stereochem. 1982,13 117. 'P. Schuster Angew. Chem. Int. Ed. Engl. 1981 20 546. 'K. Ohno and K. Morukuma 'Quantum Chemistry Literature Data Base' Elsevier Amsterdam 1982. 'J. Mol. Strucf. THEOCHEM. 1982,8 Parts 112. 'V. R. Saunders and M. F. Guest Comput. Phys. Commun. 1982 26 389. M. S. Gordon J. S. Binkley J. A. Pople W. J. Pietro and W.J. Hehre J. Am. Chem. SOC.,1982 '' J. 104,2797. S. Binkley J. A. Pople and W. J. Hehre J. Am. Chem. SOC.,1980 102 939. '* W. J. Pietro. M. M. Francl W. J. Hehre. D. J. DeFrees J. A Pople and J. S. Binkley J. Am. Chem. SOC.,1982 104 5039. l3 M. M. Francl W. J. Pietro. W. J. Hehre J. S. Binkley M. S.Gordon D. J. DeFrees and J. A. Pople 1. Chem. Phys. 1982.77 3654. 18 C. Thomson give equilibrium structures very close to experimental values and excellent values for several other properties. Nobes et al. have shown that it is possible to estimate MP3 energies for basis sets including polarization functions from results of HF and MP2 cal~ulations.'~ The use of 0s orbitals in SCF calculations on molecules has been advocated by Szalewicz based on a re-evaluation of these functions in calculations on He." Further studies on the uniform quality basis sets have appeared,16.17 and studies of heavily contracted basis sets for Fe Ni Pd and Mo.'* Standard basis sets with polarization functions have also been studied by Zhu et di9 Procedures for the determination of contraction coefficients have been studied by Graf and Mehler,20*21 and further new contracted basis sets have been reported by Sakai ef al.for the elements Ga through Cd.22 Finally a 6-311G** basis set suitable for use with correlated wave functions (5d-functions are used and the exponents optimized for correlated wave functions) has been reported by Nobes et af. and tested by computation of geometries and atomization energies of several small mo~ecules.~~ Self-consistent Field Theory.-There have been several new studies of the conver- gence problem in SCF calculations particularly the question of the intrinsic convergence of closed-shell SCF calculation^,^^-^^ and convergence in excited state ca~cu~ations.~~ A new quadratically convergent SCF method applicable to both closed2' and open-shell has been discussed by Backsay.Pulay has described a pro- cedure that seems to be somewhat superior to Backsay's method for difficult cases.30 Almlof3' has presented a view of the direct SCF approach which is very efficient for large systems. In this method all two-electron integrals are re-calculated in the SCF iterations and the method is particularly useful also on modern mini-computers.A detailed study by Almlof32 also concludes that the HF model is incapable of describing the metal-ring distance in ferrocene to better than 0.25 A and for transition-metal complexes one should really go beyond the HF approxi- mation. I' R. H. Nobes W. J. Bourna and L. Radorn Chem. Phys. Lerr.. 1982,89,497. *'K. Szalewicz and H. J. Monkhorst J. Chem. Phys. 1981,75 5785. I' R. A. Poirier R. Daudel P. G. Mezey and I. G. Csizrnadia Znt. J. Quanrum Chem. 1982 21 799. " R. Daudel R. A. Poirier and I. G. Csizrnadia In(. J. Quantum Chem. 1982 21 699. " L. Pettersson and U. Wahlgren Chem. Phys. Lerr. 1982. 89 26. I' Z. H. Zhu and J. N. Murrell. Chem. Phys. Letr. 1982.88 262. 2" P. Graf and E. L. Mehler Inr. J. Quanrum Chem. 1981 QB8.63. " P. Graf E. L. Mehler A.D. McLean and G. S. Chandler J. Chem. Phys. 1982,76 1593. 22 Y. Sakai H. Tatewaki and S. Huzinaga J. Compur. Chem. 1982 3 6. 2' R.H. Nobes,W. R. Rodwell and L. Radorn 1. Compur. Chem. 1982,3,561. 24 R. E. Stanton J. Chem. Phys. 1981 75 3426. 25 R. E. Stanton J. Chem. Phys. 1981,75 5416. 26 J. C. Facelli J. Chem. Phys. 1982 77 1076. 27 P. C. De Mello M. Hehenberger and M. C. Zerner Znr. J. Quantum Chem. 1982,21 251. 2R G. B. Backsay Chem. Phys. 1981 61 385. 29 G. B. Backsay. Chem. Phys. 1982.65 383. '' P. Pulay I. Compur. Chem. 1982 3 556. I' J. Almlof K. Faegri Jun. and K. Korsell J. Comput. Chem. 1982,3 385. 72 H. I-uthi J. H. Arnrnete J. Alrnlof and K. Faegri Jun.. J. Chem. Phys. 1982 77 2002. Theoretical Chemistry 19 Ka~fman~~ and co-workers have also emphasized the desirability of developing optimal strategies for large molecule calculations especially on molecules of biologi-cal interest The use of such strategies on vector machines’ should be very important in future work.Pople et al.34 have summarized the results of SCF calculations of vibrational frequencies and harmonic force constants for a wide variety of molecules using the 3-21G basis set. A non-conventional algorithm for SCF calculations has been described by Benzel and Dyk~tra,~’ which is essentially open ended and likely to be very cost effective on computers with relatively small core size. The doublet instability problem in radicals is well and more examples of this problem with RHF and two configuration SCF calculations on radicals (in which incorrect geometries were obtained for both singlet and triplet states of conjugated diradicals) have been reported.Multi-configuration SCF (MC-SCF) methods continue to be useful in correcting some of the deficiencies of the SCF model; there have been further studies of more efficient ways of achieving stable con~ergence.~~ Yeager and co-workers have also investigated the problems of avoided curve crossings using MC-SCF theory.38 A simplified method for determining the molecular correlation energy has been described by Lievin et al. within a MC-SCF A review of MC-SCF CI and numerical SCF calculations has been given by Hin~e.~’ Electron Correlation.-The continued search for more efficient schemes to include electron correlation continues.An important extension of the CI gradient tech- niques to MC-SCF wave functions by solving the coupled perturbed Hartree-Fock equations has been reported by Schaefer’s group.41 The (b‘B,)first excited state of CH2 was used as a test case with a DZ + P basis set. A diagrammatic formulation of the direct CI method has been reported,42 as well as renewed studies on the Unitary Group Shavitt and co-w~rkers~~ have examined various wave function optimization methods used in CI procedures. The problem of size consistency in CI calculations restricted to a limited number of excitations (such as all doubles) is well and corrections for this in CI calculations have been discussed in two paper^.^^.^^ 33 J. J. Kaufman P.C. Hariharan and H. E. Popkie fnt. J. Quantum Chem. 1981 S15 199. 34 J. A. Pople H. B. Schlegel R. Krishnan D. J. DeFrees J. S. Binkley M. S. Frisch R. A. Whiteside R. F. Hout and W. J. Hehre Znt. J. Quantum Chem. 1981 S15.269. 3s M. A. Benzel and C. E. Dykstra J. Comput. Chem. 1982,3,260. 36 W. T. Borden and E. R. Davidson Tetrahedron 1982,38,737. 37 R. C. Camp and H. F. King J. Chem. Phys. 1982 77,3056. 38 J. Olsen P. Jbrgensen and D. L. Yeager J. Chem. Phys. 1982,76 527. 39 J. Lievin J. Breulet P. Clercq and J.-Y. Metz Theor. Chim. Acta 1982 61,513. 40 J. Hinze Int. J. Quantum Chem. 1981 15 69. 41 Y. Osamura Y. Yamaguchi and H. F. Schaefer J. Chem. Phys. 1982 77,383. 42 J. M. Cullen and M. C. Zerner Chem. Phys. Lett. 1982,88 448. 43 R.D. Kent and M. Schlesinger fnt. J. Quantum Chem. 1982 22 223. 44 W. Zhen-yi Theor. Chim. Acta 1982,61 335. 45 G. Born and I. Shavitt J. Chem. Phys. 1982,76 558. “ R. Shepard I. Shavitt and J. Simons J. Chem. Phys. 1982,76,543. 47 E. R. Davidson ‘The World of Quantum Chemistry’ ed. R. Daudel and B. Pullman Reidel Dordrecht 1974. 48 S. Prime C. Rees and M. A. Robb Mol. Phys. 1981,44 173. 49 E. J. Brandas L. L. Combs and N. S. Correia fnt. J. Quantum Chem. 1982,21 259. 20 C. Thomson Integral-dependent spin couplings in CI that lead to only one spin coupling regardless of the complexity of the reference wave function yield about 98.5% of the correlation energy.5o Different ways of tackling CI calculations continue to be reported and a perturba- tive variational approach to a multireference CI which results in H and S matrices that are much reduced in size and the elements of which are obtained by diagram- matic techniques has been developed and te~ted.’~ The relationship between the CI and coupled-cluster methods (see below) has been explored by Paldus and c~-workers.~* This analysis provides a new viewpoint for the origin of Davids~n’s~’ correction for unlinked cluster contributions to CID calculations.Numerical applications to CO were reported. Nakatsuji et al.53have developed a new cluster-type approach for both closed- and open-shell systems following their earlier papers on this subject. These authors also considered the treatment of excited states. The coupled-cluster (CC) approach and its relation to many-body-perturbation (MBPT) theoretical methods has been reviewed by Bartlett.54 During 1982 there have been further developments in cluster-type theories.The formalism of the full coupled-cluster singles and doubles model has been derived by Purvis and Bartlett.” This method is a many-body theory built upon the linked-diagram theorem. Implementation of the method has been described and compared with full CI calculations on H20 and BeH,. This method should be widely applied in view of its advantages. A related approach has been taken by Banerjee and Simon~,~~ but using a multi-configuration SCF reference function. Cullen and Zerner have derived an approximation to a linked single and doubles model using diagram techniques from coupled-cluster the~ry.~’ As mentioned above MBPT methods are related to the CC methods.Redman and BartlettS8 have discussed the multidimensional many-body theory and Carsky and co-workersSg have compared CI and MBPT methods in calculations on the triplet states of CH and 0,. Carsky et af. have also compared CEPA PNO-CI and MBPT in open-shell systems.60 Further work on the formalism has appeared by Wilson,61*62 and Banerjee et al.63Several applications using Mdler-Plesset methods are dealt with below. Davidson and co-w~rkers~~ have used MBPT to ’O K. Iberle and E. R. Davidson J. Chem. Phys. 1982 76 5385. 51 G. L. Bendazolli G. Fano F. Ortolani and P. Palmieri J. Chem. Phys. 1982,76 2498. ” J. Paldus P. E. S. Wormer F.Visser and A. van der Avoird J. Chem. Phys. 1982 76 2458. 53 H. Nakatsuji K. Ohta and K. Hirao J. Chem. Phys. 1981,75 2952 and references therein. 54 R. J. Bartlett Annu. Reo. Phys. Chem. 1981 32 359. ” G. D. Purvis tert. and R. J. Bartlett J. Chem. Phys. 1982,76 1910. 56 A. Banerjee D. Mukherjee and J. Simons J. Chem. Phys. 1982,76 1972 1979 1995. ’’ J. M. Cullen and M.C. Zerner J. Chem. Phys. 1982,77,4088. ’’ L. T.Redman and R. J. Bartlett J. Chem. Phys. 1982 76 1938. 59 P. Carsky M. Svrcek I. Hubac F. Brown and I. Shavitt Chem. Phys. Lett. 1982,85 17. 6o P. Carsky I. Hubac and V. Staemmler Theor. Chim. Acta 1982 60,445. 61 S. Wilson Theor. Chim. Acta 1982,61 343. ” D. M. Silver and S. Wilson Proc. R. SOC.London Ser. A 1982 383 477. 63 A. Banerjee and J.Simons J. Chem. Phys. 1982,76,4548. “ P. Philips and E. R. Davidson J. Chem. Phys. 1982 76 516. Theoretical Chemistry 21 develop a formalism to facilitate calculation of phosphorescence lifetimes and reported applications to CH,. Carsky has given formulae for the correlation energy in triplet states using the MBPT formali~m.~’ More applications of the CEPA methods have appeared including studies on the Rydberg states of H20,66and the vibration frequencies of SF and SCl.67 One method that has received more attention recently after a relatively long gap is the theory of two electron functions or geminals. Rgeggen has developed an extended geminal and reported test calculations on H,06* and Li2.69 It is suggested that this model should recover up to 97% of the correlation energy obtainable by a full CI calculation.There have been other studies on geminal the~ry~’-~~ and applications reported on Be and LiH. A method related to electron pair theory and to CI is the theory of self-consistent electron pairs (SCEP) proposed by Me~er’~ and further extended by Dykstra et .~~ ~1 Applications were reported on the stability of glycine conformers. A recent development has been the extension of the formalism to the multi-reference a move which reduces the computational effort. Applications to CH2(3B1)+ HZ+CH3 + H have been reported. The barrier to this reaction was calculated to be 43.7 kJ mol-’ using a large basis set. An interesting recent paper by Jasien and Dyk~tra~~ has pointed out the advantages of using a common one-electron basis representation in seemingly different methods of handling electron correlation such as SCEP and CI.These methods use coefficient and associated operator matrices to represent the wave function directly. Illustrative calculations were presented on N2H2,.HCN and HNC. The energy difference between the two isomers was predicted to be 63.0 f 4 kJ mol-’ with the largest basis set in good agreement with the experimental value. Valence Bond The~ry.-Simonetta~~ and co-workers have continued their studies of ab initio VB theory and K~wajirna~~’~~ has investigated further a new self- consistent VB method and the use of an effective spin Hamiltonian in such a context. The aim was to develop a more practical method applicable to conjugated systems.Pseudopotential Methods.-Developments in this area continue since it is hoped that ab initio methods of this type will enable accurate valence electron only calculations to be made on quite complex molecules. Methods using this formalism 65 P. Carsky and J. Hubac Collect. Czech. Chem. Commun. 1981,46 1324. 66 V. Staemmler R. Jaquet and M. Jungen J. Chem. Phys. 1981 74 1285. 67 V. Staemmler Theor. Chim. Acta 1982,62 69. 68 I. Reeggen Znt. J. Quantum Chem. 1982,22 149. 69 I. Reeggen Chem. Phys. Lett. 1982,92 398. 70 A. Kurtz N. Elander 0.Goskinski and E. Sangfelt Znt. J. Quantum Chem. 1981 S15 143. 71 J. V. Ortiz B. Weiner and Y. O.hcn Znt. J. Quantum Chem. 1981 S15,113. 72 R. L. Lozes B.Weiner and Y. Ohm Znt. J. Quantum Chem. 1981 S15 129. 73 W. Meyer J. Chem. Phys. 1976,64,2901. 74 C. E. Dykstra R. A. Chiles and M. D. Garrett J. Comput. Chem. 1981 2 266. 75 H.-J. Werner and E.-A. Reinsch J. Chem. Phys. 1982,76,3144. 76 P. G. Jasien and C. E. Dykstra J. Chem. Phys. 1982 76,4564. 77 M. Simonetta M. Raimondi and G. F. Tandardini Int. J. Quantum Chem. 1981 S15,225. 78 S. Kuwajima J. Chem. Phys. 1981 74 6342. 79 S. Kuwajima J. Chem. Phys. 1982.77 1930. 22 C. Thomson have been reviewed by Topiol et u1.,80and its use in molecular structure calculations of geometries and transition states using gradient-optimization methods has been tested by Morukuma’s group” and the results found to be in excellent agreement with full ub initio calculations.The extension of such calculations to transition-metal atoms has been reported by several groups. One calculation of the barrier to internal rotation” in Ni&*- gave rather poor results and reasons for this were discussed. However some successful results have been obtained using minimum basis adapted pseudo- potentials for Cu in calculations on Cu Cu2 and CUF.~~ Gaspar’s group using pseudopotentials within the FSGO method have deter- mined improved parameters for these pseudo potentials and further applications to the structure of ethylene acetylene and H20 were rep~rted.’~ In a later paper,” the pseudopotential parameters of DZ quality were developed and used to study various properties of Liz LiH and BeH2. Generalized valence bond (GVB)-effective potentials have been used in studies of the stabilities of HSiSiH and five isomeric structures.86 The acetylene analogue was the least stable structure with the analogue of vinylidene carbene H2SiSi the most stable.In the above calculations a DZ-quality valence basis set was used with the Si pseudopotential. A scaled single zeta basis of wider applicability has been developed by the same group and used in an investigation of the structure of Si2H6.87 The structure is explicable in terms of localized u and SiH bonds analogous to C2H6 unlike the case of disilene and disilyne. 3 Potential Energy Surfaces and Reaction Mechanisms Potential Energy Surfaces.-The fundamental importance of the potential energy hypersurface in the analysis of chemical reactions is unquestioned and Mezey in particular has investigated several important fundamental questions concerning the general characteristics of these s~rfaces.~~*~~ Fukui” has reviewed the ideas of the Intrinsic Reaction Co-ordinate approach and Nauts and Chapuisat have given criteria for the definition of an Intrinsic Reaction Path.” The problem of locating transition states has been discussed by S~hlegel,~’ and a practical implementation and results described.Alternative procedures have been ‘O S. Topiol R. Osman and H. Weinstein Ann. N.Y. Acad. Sci. 1981 367 17. ” S. Obara K. Kitaura and K. Morukuma Theor. Chim. Acta 1981,60 227. 82 0.Gropen U. Wahlgren and L. Pettersson Chem. Phys. 1982 66,453 459. 83 G. H.Jeung J. C. Barthelat and M. Pelissier Chem. Phys. Lett. 1982,91 81. 84 R. Gaspar Jun. and R. Gaspar Int. J. Quantum Chem. 1982 22,415. ’’ R. Gaspar Jun. and R. Gaspar Int. J. Quantum Chem. 1982 22,421. 86 L. C. Snyder Z. R. Wasserman and J. W. Moskowitz Int. J. Quantum Chem. 1982,21,565. 87 J. W. Moskowitz S. Topiol L. C. Snyder and M. Ratner Int. J. Quantum Chem. 1981,19 131. ’’ P. G. Mezey Theor. Chim. Acta 1982,60,409. 89 P. G. Mezey Theor. Chim. Acta 1981 S15 279. 90 K. Fukui Acc. Chem. Res. 1981 14 363. 91 A. Nauts and X. Chapuisat Chem. Phys. Lett. 1982,85 212. 92 H. B. Schlegel J. Comput. Chem. 1982 3 214. Theoretical Chemistry 23 developed by Bell93 and S~harfenberg.'~ Almlof has derived a compact formalism for the calculation of MC-SCF gradients a problem also tackled by Schaefer et ~1.~' An algorithm for efficiently evaluating energy gradients using s and p functions has been described by S~hlegel.~' Turning now to particular potential energy (P.E.) surfaces computations corre- lated P.E.curves have been computed for several diatomic molecules among them a detailed study of various states of N2.96MC-SCF studies on the reaction C' + H2 -+ CH' + H have been reported,97 and a detailed MRD-CI study of the photo- dissociation of H20 -+ O('D,) + H2(1Xi).98At the SCF level a study of the NSF surface reveals NSF to be more stable than FNS by -85 kJ m01-l.~~ An interesting study has been made of various 22 electron molecules and the dissociation process leading to the isoelectronic products CO or N2 or BH3 CH2 NH and O.looCalculated values of the dissociation energies are in good agreement with experiment where known lending some confidence to predicted values for the dissociation of HNCO BH,CO CH2N2 and BH3N2.An investigation of the insertion reactions of CH and SiH2 into H has been reported by Gordon.'" A similar study of SiH + H,O has been reported using an extended basis set and correlation corrections.102 An interesting study of the oxidation mechanism of NH2 + O2 $ NH202 using calculated parameters from a 4-31G basis set calculation determined the transition- state geometry and calculated the thermodynamic parameter^."^ SN2transition states have been studied in detail by Wolfe et aZ.'04~105 following earlier ~~rk.'~~~~~~ It was shown that 4-3 1G calculations are adequate to characterize the surface and the results of the calculation of AH and for ten transition states in process (1)gives theoretical backing for Hammond's Rule.A second paper dealt with intrinsic barriers,106 rate equilibria relationships and the Marcus equation. X-+ CH3F + [X-CH3-F]-+ CH3X + F-(1) Two papers have dealt with the CH20 system important in the ozonolysis of C2H4.The minimum energy pathway diradical state and saddle point were deter- mined.108.109 Bouma el al. have explored in detail the C,H20 surface at the 4-31G level. Relative energetics were then more accurately calculated using the MP3 93 S. Bell J. S. Crighton and R. Fletcher Chem. Phys. Lett. 1.981,82 122. 94 P. Scharfenberg J.Compuf. Chem. 1982 3 277. 95 H. B. Schlegel I. Chem. Phys. 1982,77 3676. 96 W. C. Ermler A. D. McLean and R. S. Mulliken J. Phys. Chem. 1982,86 1305. 97 S. Sakai S. Kato K. Morukuma and I. Kusunoki J. Chem. Phys. 1981,75 5398. 98 G. Theodorakopoulos C. A. Nicolaides R. J. Buenker and S. D. Peyerimhoff Chem. Phys. Lett. 1982,89 164. 99 R. Seeger U. Seeger R. Bartetzko and R. Gleiter Inorg. Chem. 1982 21,3473. IonJ. Breulet and J. Lievin Theor. Chim. Actu 1982,61 59. M. S. Gordon J. Chem. SOC.,Chem. Commun. 1981,890. K. Raghavachari J. Chandrasekhar and M. J. Frisch J. Am. Chem. SOC..1982,104,3779. C. Pouchan and M. Chaillet Chern. Phys. Lett. 1982 90 310. Io4 S. Wolfe D. J. Mitchell and H. B. Schlegel Can. J. Chem. 1982,60 1291. '" S.Wolfe D. J. Mitchell H. B. Schlegel C. Minot and 0.Eisenstein Terruhedron Lett. 1982 23 615. S. Wolfe D. J. Mitchell and H. B. Schlegel J. Am. Chem. Soc. 1981 103 7692. S. Wolfe. D. J. Mitchell and H. B. Schlegel J. Am. Chem. SOC.,1981 103,7694. R. Cimiraglia T.-K. Ha and Hs. H. Gunthard Chem. Phys. Lett. 1982.85 262. R. Cimiraglia T.-K. Ha R. Meyer and Hs. H. Gunthard Chem. Phys. 1982,66 209. C. Thomson method. Ketene is the most stable isomer but hydroxyacetylene is a potentially observable isomer.' lo Several P.E. surfaces of the important combustion intermedi- ate CH30 have been studied. '" Pop1e's'l2 group have carried out a similar study of the C2H4 surface using an extended polarized basis set (6-311G**) with up to MP4 corrections. After inclusion of zero-point vibrational corrections it was found that singlet CH3CH is not a local minimum rearrangement to CzH4 should occur without activation.The structures and rearrangements of various cyclic and acyclic C4HsNO+ cations have been studied by MNDO and ab initio (4-31G) method^."^ The cyclic cations have unusual stability in this system the most stable being (1). (1) K~tzelnigg"~ has carried out a very detailed study of the reaction PH5 +PH + H2.Of the numerous other studies in this area we draw attention to the MC-SCF study of the O(3P)+ C& -+ CH,CH,O-reaction where a stable diradical is produ~ed."~ The transition state in this case has C symmetry i.e. the addition is asymmetric. The decomposition of glyoxal continues to fascinate theoreticians and the transi- tion state of C2"symmetry has been located by Schaefer's group using several different basis sets.116 Considerably different results are obtained in the vibrational analysis of the four stationary point structures and a detailed discussion of the implications of this is given.The pathway for the interconversion of the cyclobutene cation (2) to the buta-1,3- diene cation (3) has been studied by both MIND0/3 and ab initio methods with an STO-3G basis and also with a 6-31G* basis. The activation energy is in good agreement with the experimental value. A model for the Grignard reaction has been studied by Nagase et al.,"' and by Sakai and Jordan."' 1,2-Hydrogen shifts continued to be studied,12' including the barrier for reaction (2).Addition of polarization functions however does not change the predicted H H H \/ / H/sI'c\H -b H H /Si-Y-H W. J. Bouma R. H. Nobes L. Radom and C. E. Woodward J. Org. Chem. 1982,47 1869. 'I' C. F. Jackels J. Chem. Phys. 1982,76,505. 'I2 R. Krishnan M. J. Frisch J. A. Pople and P. von R. Schleyer Chem. Phys. Left. 1982.85 145. G. Frenking and H. Schwarz J. Comput. Chem. 1982 3 251. '14 W. Kutzelnigg and J. Wasilewski J. Am. Chem. SOC.,1982,104,953. I" M. Dupuis J. J. Wendoloski T. Takada and W. A. Lester Jun. J. Chem. Phys. 1982 76 481. 'I6 Y. Osamura H. F. Schaefer tert. M. Dupuis and W. A. Lester Jun. J. Chem. Phys. 1981.75 5828. '" D. J. Bellville R. Chelsky and N. L. Bauld J. Comput. Chem. 1982 3 548. '" S. Nagase and Y.Uchibori Tetrahedron Left. 1982,23,2585. 'I9 S. Sakai and K. D. Jordan J. Am. Chem. SOC.,1982,104,4019. Y. Yoshioka and H. F. Schaefer tert. J. Am. Chem. SOC.,1981,103,7366. Theoretical Chemistry 25 barrier substantially even after extensive CI. There is still a continuing discrepancy with experiment. 1,2-Shifts in open-shell systems have been studied at the POL-CI level.'*' In simple alkyl radicals and triplet carbenes the lowest energy pathway for 1,2-hydrogen migration will be CH bond cleavage in most cases followed by reattachment. More complicated systems like that of the cycobutadiene radical cation,'22 and me thylene cy~lopropane'~~ and cyclopropane cation'24 surfaces are necessarily restricted to smaller basis sets such as STO-3G or to semi-empirical studies i.e.OCS,.'~~ A reaction that is of some importance is the H20 + CO -D H2CO3 reaction. A detailed (4-31G basis) study126 of various conformers of the acid and also of the bicarbonate ion has led to an estimate of the AHo for the hydration reaction. Discrepancies between theory and experiment are ascribed to basis set effects not to correlation effects. Harding and co-w~rkers'~~~'~~ have computed the rate constants for the reactions H2 + C2H -+ H + C2H2 and also located the transition state. The method used was the GVB-CI SOGVB or POL-CI. Finally the reaction surface for the H20-Li20 system has been studied at the MP3 level of The reaction proceeds with activation to yield the cyclic dimer of LiOH. The results agree well with experiment.4 Electronic Structure and Molecular Geometries Large numbers of papers have appeared this year indeed the pace of the application of theoretical methods of ab initiu quality to organic systems continues to accelerate. The papers referred to in this review will necessarily be selected according to this reviewer's prejudices. A thorough study of 30 organic corn pound^'^^ using complete geometry optimiz- ation at the 4-21G SCF level has led the authors to suggest a series of empirical corrections to the computed geometry to give geometries in agreement with experiment. An important paper by De Frees et al.'31compares theoretical geometries calculated with the MP3 or CID levels of theory and a 6-31G* basis set for a large set of one- and two-heavy-atom molecules.The results are an improvement over the MP2 results and are in good agreement with experiment. High quality CASSCF calculations on HNOi32 have been reported but CI is necessary to obtain results in agreement with experiment. 12' L. B. Harding J. Am. Chem. SOC.,1981 103 7469. 122 W. T. Borden E. R. Davidson and D. Feller J. Am. Chem. Soc. 1981,103,5725. 123 D. Feller K. Tanaka E. R. Davidson and W. T. Borden J. Am. Chem. SOC.,1982 104. 967. 124 J. R. Collins and G. A. Gallup J. Am. Chem. SOC.,1982,104 1530. 12' L. Carlsen J. Cornput. Chem. 1982 3 123. 126 P. George C. W. Bock and M. Trachtman J. Comput. Chem. 1982,3,283. '*' L. B. Harding G. C. Schatz and R. A. Chiles J. Chem. Phys. 1982,76,5172. 12* L. B.Harding A. F. Wagner J. M. Bowman G. C. Schatz and K. Christoffel J. Phys. Chem.. 1982 86,4312. 129 K. Raghavachari J. Chem. Phys. 1982,76 5421. L. Schafer C. Van Alsenoy and J. N. Scarsdale J. Mol. Srruct. 1982,86 349. 131 D. J. De Frees K. Raghavachari H. B. Schlegel and J. A. Pople J. Am. Chem. SOC.,1982,104,5576. 13' A. Heibern and J. Almlof. Chem. Phys. Lett. 1982.85 542. C. Thomson The remarkable stabilities of Li-substituted carbenium ions CLi3- H,' (n = 0-3) have been discovered by Pople and co-worker~.~~~ CLi3+ in particular is very stable but in the triplet state. The ionization potential of CLi is predicted to be 6 eV less than that of CH,. The stabilizing influence of Li is such as to make CLi3+ one of the most stable carbenium ions known.Recently it was also shown that the CH3+ triplet state is n0np1anar.l~~ The same group have shown that in neutral AX molecules when X = Li the octet rule is violated CLi3 and OLi are predicted to be ~tab1e.l~~ The cation CLi5+ has recently been observed in the gas phase following earlier theoretical predictions and a more accurate recent study has reconsidered this species together with CLi6'+ CLi6 and CLi8 all of which are predicted to be stable.136 The structures of other carbenium ions with electropositive ~ubstituents'~~ have been studied as well as those of CH2" and CH42+.138 The latter have relatively high barriers to proton loss in contrast to the case of CH3'+ and CH'+. Various C2H6'+ structures have been studied by Olah139 and also by Pople et aI.14' The methanol cation (4) has been to be higher in energy on the CH40+ surface than the alternative structure (5).Experimental evidence for this species has been pre~ented.'~' A similar study of the pathways'43 for the decomposition of CH30H2+ has been reported. H H 1' 1' 1 ,c =c C. H H/y..H I H 0' 0 H\ ..I= H. 1; F--C=C*-F 1' I "H F' H (6) (7) (4) (5 ) Alternative structures for the ions of C2H2- have been Diffuse d-orbitals are essential to describe anions correctly and these studies utilized the 4-31G+ basis set. The conclusion was the structure is the vinylidene species (6). The entire set of anions XCH2- where X = Li BeH BH2 CH3 NH2 OH and F were studied with 4-31G+ and 6-31G+* basis sets and also at the MP2 level by Spitznagel et 133 J.Chrandresekhar J. A. Pople R. Seeger U. Seeger and P. von R. Schleyer J. Am. Chem. SOC. 1982,104,3651. 134 J. A. Pople and P. von R. Schleyer Chem. Phys. Lerr. 1982.91 9. P. von R. Schleyer E.-U. Wurthwein and J. A. Pople J. Am. Chem. SOC.,1982,104,5839. E. D. Jemmis J. Chandresekhar E.-U. Wurthwein P. von R. Schleyer J. W. Chinn Jun. F. J. Landro R. J. Lagow B. Luke and J. A. Pople. J. Am. Chem. Soc. 1982,104,4275. 13' J. A. Pople. Y. Apeloig and P. von R. Schleyer Chem. Phys. Lett. 1982 85,489. 13* J. A. Pople B. Tidor and P. von R. Schleyer Chem. Phys. Lett. 1982,88,533. G. A. Oiah and M. Simonetta J. Am. Chem. SOC.,1982 104. 330. I4O P. von R. Schleyer A. J. Kos J. A. Pople and A. T. Balaban J.Am. Chem. Soc. 1982 104 3771. 141 W. J. Bouma R. H. Nobes and L. Radom J. Am. Chem. Soc. 1982,104,2929. W. J. Bouma J. K.MacLeod and L. Radom J. Am. Chem. Soc. 1982,104,2930. 143 R. H. Nobes and L. Radom Org. Mass.Spectrom. 1982,17 340. 144 J. Chandresekhar R. A. Kahn and P. von R. Schleyer Chem. Phys. Lett. 1982,85,493. "Ii G. W. Spitznagel T. Clark J. Chandresekhar and P. von R. Schleyer J. Comput. Chem. 1982,3,363. TheoreticaI Chemistry 27 The geometry of the fluoro-substituted ethylene radical ions has been established by Merry and Th~mson,'~~ and independently by Paddon-Row et uf.14' Although the radical cations are planar substitution of only one F-atom is sufficient to give a markedly non-planar radical centre. The extreme case of C2F4- is that of CZh symmetry (7).There have been various other studies of the structure of radicals including an ab initio SCF-CI of a stable nitroxide of known geometry and in particular the calculation of the hyperfine tensors from the wave function. 14' Pacansky and co-workers have continued their studies of simple alkyl radicals. Barriers to internal rotation were found to be small in the neopentyl and isobutyl radical^.'^' The same author15' has used a very large basis set to study CH3 and shown that only if inadequate bases are used is the geometry predicted to be non-planar. Bair and Goddard have reported GVB calculations of the structures of several peroxides and peroxy radicals. There have been fewer studies of carbenes but an SCF-CI of CH3CH has suggested that this species should not exist as a stable intermediate although there is a local minimum on the P.E.surface. The singlet-triplet gap is -40 kJ mol-'. Kim So and Schaefer have looked at three low-lying states of HOCOC using SCF and CI methods (including up to 31 532 configurations). The ground state is 'A" lying 40 kJ mol-' lower than the 'A' The calculation of the acid-base behaviour of several systems has been pursued by several groups. A comprehensive study of the methyl and silyl amines and their ions using the 3-21G basis set and full geometry optimizations was reported by Thornson and Glide~el1.I~~ With the exception of (SiH3)'N- [and the isoelectronic molecules (SiH3)*0 and (SiH3)2C2-] the observed data can be rationalized without the inclusion of 3d functions in the basis set.A related study by del Bene et ~1.'~~ examined the protonation of NH3 H20 and HF and the closed-shell bases H,ABH, where A B may be C N 0,and F. The inclusion of correlation (up to MP4 level) did not change the results much from the SCF values however the barriers to proton transfer are more sensitive to correlation. Similar FSGO basis set calculations lead to similar conclusions. 156 Zielinski et af.15' have examined the structures of protonated neutral and deprotonated amidine HC(NH,)(=NH) at the 3-21G level and the structures of several transi- tion states in this system. 146 S. Merry and C. Thornson Chem. Phys. Lett. 1981,82 373. 147 M. N. Paddon-Row N. G. Rondan K. N. Hauk and K.D. Jordan J. Am. Chem.SOC.,1982,104,1143. 14* R. Briere T. A. Claxton Y. Ellinger P. Rey and J. Langier J. Am. Chem. Soc. 1982,104.34. 149 J. Pacansky and W. Schubert J. Chem. Phys. 1982.76 1459. J. Pacansky J. Phys. Chem. 1982.86,485. lS1 R. A. Bair and W. A. Goddard tert. I. Am. Chem. Soc. 1982 104 2719. 15' T.-K. Ha M. T. Nguyen and L. G. Vanquickenborne Chem. Phys. Lett. 1982,92,459. K. S. Kim S. P. So and H. F. Schaefer tert. J. Am. Chem. Soc. 1982 104 1457. C. Glidewell and C. Thornson J. Comput. Chem. 1982,3,495. 155 J. del Bene M. J. Frisch K. Raghavachari. and J. A. Pople J. Phys. Chem. 1982,86 1529. H. Huber and J. Vogt Chem. Phys. 1982,64 399. T. J. Zielinski M. R. Peterson I. G. Csizmadia and R. Rein J. Comput. Chem. 1982 3,62. C.Thomson The destabilizing effects of CN and CF3 groups on a-heteroatom stabilized cations has been rep~rted,'~~ but on the other hand CN and NC stabilize car- bani on^.'^^*'^^ Calculations on several different aldoximes RCH=NOH have been reported,161 including the low-lying states of CH2=NOH.'62 There have been two further sets of calculations on ketenimine H2C=C=NH. 163~164 A complete yo structure has been computed for pr~padienone,'~~ including a study of the effect of electron correlation. The molecule (8)is planar but with a CCC angle of 145". A long standing problem of the cis-trans energy difference between the isomers of 1,2-difluoroethylene has been studied'66 with SCF method the SCEP method and also MP2 and coupled-cluster methods.Correlation plays a crucial role in determining the structure here. Among several miscellaneous molecules studied this year are HCCNO 167 as yet not synthesized the triplet species on the C3H60 surface,'68 the mechanism of the reduction of R2C0 by LiBH4,'69 and a study of the (CO)nm- isomers.'7o HF instabilities occur in these systems and hence correlation effects are expected to be important. Several conformational studies on glycine and substituted glycines have a~peared,'~72 'J as well as studies on several alkanes:'73 n-propane n-butane iso-butane n-pentane and n-r~onane'~~ have all been studied by complete geometry optimization at the 4-21G basis set level. Turning now to the results for cyclic systems there has been a recent detailed study of C6H6 and C6F6 with 20 different basis Minimal basis contractions M.N. Paddon-Row K. N. Houk and T. T. Tidwell Tetrahedron Lett. 1982,23,383. J. B. Moffat Int. J. Quantum Chem. 1982 22 299. 160 J. B. Moffat J. Am. Chem. SOC.,1982,104 3949. 16' M. T. Nguyen and T.-K. Ha J. Mol. Struct. 1982 88 127. "* T.-K. Ha Chem. Phys. Lett. 1982,86 477. 163 J. Kaneti and M. T. Nguyen J. Mol. Struct. 1982 87 205. T.-K. Ha and M. T. Nguyen J. Mol. Struct. 1982,87,355. L. Farnell and L. Radom Chem. Phys. Lett. 1982 91 373. 166 S. R. Gandhi M. A. Benzel C. E. Dykstra and T. Fukunaga J. Phys. Chem. 1982,86 3121. S. Bell J. Mol. Strucf. 1982 87 25. M. R. Peterson and G. R. De Mare J. Mol. Struct. 1982,88 1. 169 R. Bonaccorsi P.Palla and J. Tomasi J.Mol. Srrucr. 1982,87 181. D. P. Vercanteren J. G. Fripiat and J. M. Andre Int. J. Quantum Chem. 1981 SlS,219. 17' L. Schafer C.Van Alsenoy and J. N. Scarsdale J. Chem. Phys. 1982,76 1439. 17* V. J. Klimkowski J. D. Ewbank C. Van Alsenoy J. N. Scarsdale and L. Schafer J. Am. Chem. Soc. 1982,104 1476. 173 J. N.Scarsdale C. Van Alsenoy and L. Schafer J. Mol. Strucr. 1982,86,277. 174 J. N. Scarsdale H. L. Sellers L. Schafer and N. L. Allinger J. Comput. Chem. 1982 3 269. 17' J. Almlof and K. Faegri Jun. J. Chem. Phys. 1982 xxx. Theoretical Chemistry 29 of atom-optimized basis sets are quite unreliable particularly for such basis sets on H; d-orbitals are essential for high accuracy in the case of C6F6. Boggs and co-w~rkers'~~ have also investigated various fluorobenzenes (including C6F6) using the 4-21G basis set.Several nitrogen-containing heterocycles have been studied by Boggs' group,'77 and the dihydroderivatives of benzene naphthalene and anthracene have been calculated to be ~1anar.l~~ The lowest triplet state of pyrazine has been studied at the MC-SCF particularly with regard to the e.s.r. parameters of the 3(nn") state. However the MC-SCF wave function was shown to break symmetry at D, geometries. Several cyclobutanes18' and ions derived from them have been studied,18' also bicyclobutanes,'82 a tricy~lohexane,'~~ and ~ubane.'~~*l~' Borepin (9) and cycloheptatriene'86 have also been examined and Goddard et have fully optimized structures for the classical and non-classical forms of the 2-norbornyl cation and edge-protonated nortricyclene.The calculated geometry suggests that the classical isomer is best viewed as an unsymmetrically bridged species. The tautomeric equilibria of 4-hydroxy-2-pyridinone and its other forms has been studied by Schlegel'88 using the 3-21G basis set. The relative energy differences were in reasonable agreement with the available experimental data. The same author'89 has also investigated 2- and 4-pyridone tautomers and the tautomers of uracil have been studied by Mondragon and Blake.'" Turning now to calculations on molecules containing elements of the second row there have been many interesting studies. The paperlS4 on the basicities of sub- stituted phosphines has been referred to above and Schaefer has reviewed the evidence for Si=C double bonds.'" The interest in organosilicon compounds continues C2H2 and Si2H2 have been compared,19* the disilyne being twisted in the ground-state configuration.The geometries and relative energies of singlet silylsilylene and disilene have been and it was found that the planar disilene is also not a minimum on the surface. However the energy difference between the two isomers is very small. 176 J. E. Boggs F. Pang and P. Pulay J. Comput. Chem. 1982,3 344. 177 F. Pang P. Pulay and J. E. Boggs J. Mol. Struct. 1982,88,79. 17' K. B. Lipkowitz P. W. Rabideau D. J. Raber L. E. Hardee P. von R. Schleyer A. J. Kos,and R. A. Kahn J. Org. Chem. 1982 47 1002. J. C. Ellenbogen D. Feller and E.R. Davidson J. Phys. Chem. 1982,86 1583. D. Feller E. R. Davidson and W. T. Borden J. Am. Chem. SOC., 1982,104 1216. T. Jonvik and J. E. Boggs J. Mol. Struct. 1981,85 293. P. N. Skancke J. Mol. Strucr. 1982 86 255. P. N. Skancke Acta Chem. Scand. Ser. A 1982 36 513. J. Almlof and T. Jonvik Chem. Phys. Lett. 1982 92. 267. C. Van Alsenoy J. N. Scarsdale and L. Schafer J. Comput. Chem. 1982 3 53. J. M. Schueman R. L. Disch and M. L. Sabio J. Am. Chem. SOC.,1982 104 3785. 187 J. D.Goddard Y. Osamura and H. F. Schaefer tert. J. Am. Chem. SOC.,1982,104 3258. 188 H. B. Schlegel Int. J. Quantum Chem. 1982 22 1041. H. B. Schlegel P. Gund and E. M. Fluder J. Am. Chem. SOC.,1982,104,5347. 190 A. Mondragon and I. Ortega Blake Int. J. Quantum Chem. 1982 22,89.H. F. Schaefer tert. Acc. Chem. Res. 1982 15 283. 'q2 F. Kawai T. Noro A. Murakami and K.Ohno Chem. Phys. Lett. 1982 92,479. 193 K. Krogh-Jespersen J. Phys. Chem. 1982,86 1492. C. Thornson has examined various carbon-silicon-bonded isomers with F and Li substituents and concluded that fluorine attached to Si has a strong stabilizing effect. Similar species containing Ge have also been studied i.e. H2Ge=CH2.195 Pseudopotential calculations were used in calculations on digermene.'96 Its geometry is trans-bent (10)and it is more stable than the HGe-GeH3 isomer. (10) (11) The inversion barrier in PH3 has been computed with a near-HF basis of STO's,19' and extensive CI in the valence space. The dependence of the barrier on basis set size the method of computation and geometry were studied.A moderate effect of correlation on this barrier was found. Comparisons of the results of SCF calculations and correlation calculations on the structures of HCP and FCP have been reported by Thomson and Ellam.'98 STO-3G basis sets for these molecules lead to unacceptable errors in the geometries. The structures and inversion barriers in CF3- and SiF3 have been computed.'99 The HOMO of planar SiF3- is of al' symmetry rather than the expected a,". A variety of calculations deal with the geometry of other sulphur-containing compounds such as thionylimide2" (HNSO) thioacrolein and thioglyoxal,20' and H2CS for which harmonic vibrational frequencies were calculated.20z Calculations on various sulphanes at the STO-3G" level were reported by Boyd.203 This basis is however inferior to those developed for the second row by Huzinaga's group.204 Several thiocarbonyls have been studied by Ha and co- worker~,~~~ as well as sulphido-borons206 R-B=S studied earlier by Thom~on.~~' Finally the simplest metallocyclobutane (11)has been studied at the SCF level and also by GVB-CI.208 5 Excited States and Ionization Processes There have been computations of the vertical ionization potential of OF2 HNFz and CH2F2 by a perturbation method developed by Ch~ng,~'~ but with a DZ-quality 194 M.S. Gordon J. Am. Chem. SOC.,1982 104,4352. 195 T. Kudo and S. Nagase Chem. Phys. Lett. 1981,84 375. 196 C. Trinquier J.-P. Malrieu and P. Riviere J. Am. Chem. SOC.,1982 104,4529.D. S. Marynick and D. A. Dixon J. Phys. Chem. 1982,86,914. 198 C. Thomson and P. Ellam Theor. Chim. Actu 1982 62 81. 199 D. S. Marynick J. Mol. Struct. 1982,87 161. 2oo K. Raghavachari J. Chem. Phys. 1982,76 3668. 201 T.-K. Ha M.-T. Nguyen and L. Vanquickenborne Z. Nuturforsch. Teil A 1982,37 125. '02 J. D. Goddard and D. J. Clouthier J. Chem. Phys. 1982 76 5039. 203 R. J. Boyd and J. F. Szabo Can. J. Chem. 1982,60,730. 204 Y. Sakai H. Tatewaki and S. Huzinaga J. Comput. Chem. 1981 2 108. 205 T.-K. Ha M.-T. Nguyen and L. G. Vanquickenborne J. Mol. Struct. 1982,90 107. 206 T.-K. Ha M.-T. Nguyen and L. G. Vanquickenborne J. Mol. Strwct. 1982 90 99. 207 C. Thomson Chem. Phys. Lett. 1974,25 59. '08 A. K. Rappe and W. A. Goddard tert. J. Am Chem.SOC.,1982 104,297. '09 S. R. Langhoff and D. P. Chong Chem. Phys. Lett. 1982,86,487. 19' Theoretica1 Chemistry 31 basis set. Extensive CI calculations continue to be made on the states of HC0,210 of acrolein,211 and of the ionic states of vinylidene and acetylene.212 The lowest excited state of the latter is 4A2with a cis-bent configuration. The geometry of the nII* states of p-benzoquinone have been computed with both a CI technique and a VB Calculations of the excitation energies in trimethylenemethane derivatives214 using better than DZ-quality basis sets were made to aid the understanding of substituent effects in these systems. The 'n-ll state of pyrazine was also investigated by Kleier and co-worker~.~'~ Using an STO-3G basis set and CI the triplet-triplet absorption spectrum of naphthalene has been studied,216 and the authors conclude that T-T transitions can be treated at this level of theory.6 Molecular Interactions Solvation Phenomena and the Hydrogen Bond.-The incorporation of solvent effects in quantum chemical calculations is of great importance and especially if changes in reaction mechanisms occur between the gas phase and solution reactions. More progress has been made in this area and the results of the ab initio studies are described in this Section. Some of these follow Clementi's approach utilizing intermolecular potential functions derived from high quality SCF wave functions and the properties of H20 have been investigated in detail by Reimer et al.217A further example of this approach is the study of the solvation in a dilute methanol An alternative is the use of a polarizable continuum in which the solvent is treated as a structureless polarizable continuum having a dielectric constant of the pure bulk material.Solvent effects on the internal rotation barrier of formamide were reproduced using this model and a 4-31G basis set ab initio wave function.220 Donor-acceptor complexes of the form A -OH2 (A = Li,Be,C,Na,Mg,Al) have been studied at the SCF level with a 6-31G** basis set.221 The gas-phase solvation of H' by clusters of 1-4 molecules of methanol or 1-3 molecules of (CH,),O have been studied by Hirao et a1.,222 who also studied the interaction of C1- with several RH Studies of the hydrogen bond continue.del Bene has continued her studies at both SCF and SCF-CI levels using a minimal basis set of H-bonding in H20-5-fluorouracil complexes224 and H20-thymine complexes.225 210 K. Tanaka and K. Takeshita Chem. Phys. Lett. 1982 87 373. 211 K. Valenta and F. Grein Can.J. Chem. 1982.60 601. 212 P. Rosmus P. Botschwina and J. P. Maier Chem. Phys. Lett. 1981 84 71. 213 R. L. Martin and W. R. Wadt J. Phys. Chem. 1982,86 2382. 214 S. B. Auster R. M. Petzer and M. S. Platz J. Am. Chem. SOC.,1982 104,3812. 215 D. A. Kleier R. L. Martin W. R. Wadt and W. R. Moomaw J. Am. Chem. SOC.,1982,104,60. 'I6 T.-K. Ha and U.P. Wild J. Compuf. Chem. 1982 3 1. 217 J. R. Reimer R. 0.Watts and M. L. Klein Chem. Phys. 1982,64,95. 218 K. Nakanishi S. Okazaki K. Ikar and H.Touhara Chem. Phys. 1981,84,428. 219 S. Miertus E. Scrocco and J. Tomasi Chem. Phys. 1981,55 117. 220 A. J. Duben and S. Miertus Chem. Phys. Lett. 1982 88 395. 221 J. Bentley J. Am. Chem. SOC.,1982 104 2754. 222 K. Hirao M. Sano and S. Yamabe Chem. Phys. Lett. 1982 87 181. 223 S. Yamabe N. Ihira and K. Hirao Chem. Phys. Lett. 1982.92 172. 224 J. E. del Bene J. Phys. Chem. 1982,86 1341. 225 J. E. del Bene J. Chem.Phys.. 1982 76 1058. C.Thornson The intermolecular H-bond in singlet and triplet excited states of the adenine- thymine pair has been studied,226 as well as the adenine-adenine interaction energy.227 Proton transfer in H-bonded systems has been extensively studied by Scheiner and further work has appeared on the cation dimers of NH and OHz,z28 (NzH7)' and (OzH5)',229 and a detailed studyz3' of correlation effects up to the MP3 level in (H0HOH)- with a range of basis sets up to 6-311G**.Several configurations of the complex between CzH4 and HF have been as has the barrier to proton transfer in the H-bonded dimers of benzoic and acetic Geometry optimization provides better values of the barrier heights in these cases. SCF calculations233 on the complexes of NH with I2 or HI at valence shell DZ level lead to the prediction of complete proton transfer in the case of H3N - - HI i.e. leading to a structure H4N'...I- . Studies of several other complexes have appeared but space limitations precludes any discussion of these. Metal or Metal Ion Interactions.-The interaction of Na' with ethene 2-methyl- propene and benzene has been studied and the optimum structures There have been further studiesz3* of models for cation binding to nucleic acids using HZPO4- and the ions Li' Na+ Be2' and Mg".The interactions are not totally electrostatic apart from those involving Na'. All cations affect the phosphate group substantially. The binding of Ca2' to malonate and formate has been inve~tigated,~,~ both with and without &orbitals in the basis set. d-Orbitals make little difference to the conformational minimum. The interaction between Oz NZ and COz and hydrocarbons and fluorocarbons has been investigated. The electrostatic interaction energy is greater in the former case and electrostatic potential maps were used to predict the structure of the complexes.237 These maps have continued to be of great use in the study of the behaviour of molecules.For instance the most favoured sites for hydration for the tautomeric forms of 2-and 4-hydroxypyridine have been investigated in this way,238 and the charge alternation in substituted alkanes has been investigated by Politzer et aZ.239The electrostatic potentials do not however alternate in the same way. 226 J. Maranon H. Grinberg and N. S. Nudelrnan Int. J. Quantum Chem. 1982 22 69. 227 M. Aida and C. Nagata Chem. Phys. Lett. 1982,86,44. S. Scheiner J. Phys. Chem. 1982 86. 376. 229 S. Scheiner J. Chem. Phys. 1982,75 5791. 230 M. M. Szczesniak and S. Scheiner J. Chem. Phys. 1982 77,4586. 231 D. Volkmann B. Zurawski and D. Heidrich Int.J. Quantum Chem. 1982,22,631. 232 S. Nagaoka N. Hirota T. Matsushita and K. Nishirnoto Chem. Phys. Lett. 1982 92,498. 233 P. Kollrnan A. Dearing and E. Kochanski J. Phys. Chem. 1982,86 1607. 234 J. Sauer and D. Delniger I. Phys. Chem. 1982,86 1327. 235 P. Liebman C. Loew A. D. MacLean and G. R. Pack J. Am Chem. SOC.,1982,104,691. 236 K. E. Gottschalk R. G. Hiskey L. G. Pedersen and K. A. Koehler J. Mol. Struct. 1982 87 155. 237 P. Ruelle and C. Sandorfy Theor. Chim. Acta 1982,61 11. 238 H.-J. Hofmann C. Peinel C. Krebs and C. Weiss Inr. J. Quantum Chem. 1981 20 785. 239 P. Politzer S. L. Whittenberg andT. Warnheirn J. Phys. Chem. 1982,86 2609. Theoretical Chemistry 7 Miscellaneous The effect of correlation on theoretical vibrational frequencies has been investigated in detail by Hehre and co-~orkers.~~~ The errors at the MP3/6-31G* level were found to be an average of -7% too high in some 36 molecules.Cremer has investigated the use of ab initio calculations to calculated thermochemical data such as heats of formation.241 Finally the number of topics treated and the accuracy of the calculations continues to increase and it is clear that ab initio computations will be an increas- ingly important tool in the investigation of chemical processes. Acknowledgement.-The author wishes to acknowledge the invaluable help of Mrs. Maureen Thomson in the mammoth literature search involved in preparing this Report. 240 R. F. Hout Jun. B. A. Levi and W. J. Hehre J. Comput. Chem. 1982 3 234.24 1 D. Cremer J. Cornput. Chem. 1982,3 154 165.