Natural resonance theory: II. Natural bond order and valency
E. D. Glendening,
F. Weinhold,
E. D. Glendening
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706
Search for more papers by this authorCorresponding Author
F. Weinhold
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706Search for more papers by this authorE. D. Glendening,
F. Weinhold,
E. D. Glendening
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706
Search for more papers by this authorCorresponding Author
F. Weinhold
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706
Theoretical Chemistry Institute and Department of Chemistry, University of Wisconsin, Madison, Wisconsin 53706Search for more papers by this authorFirst published: 07 December 1998
Citations: 470
Abstract
Resonance weights derived from the Natural Resonance Theory (NRT), introduced in the preceding paper are used to calculate “natural bond order,” “natural atomic valency,” and other atomic and bond indices reflecting the resonance composition of the wave function. These indices are found to give significantly better agreement with observed properties (empirical valency, bond lengths) than do corresponding MO-based indices. A characteristic feature of the NRT treatment is the description of bond polarity by a “bond ionicity” index (resonance-averaged NBO polarization ratio), which replaces the “covalent-ionic resonance” of Pauling-Wheland theory and explicity exhibits the complementary relationship of covalency and electrovalency that underlies empirical assignments of atomic valency. We present ab initio NRT applications to prototype saturated and unsaturated molecules methylamine, butadiene), polar compounds (fluoromethanes), and open-shell species: (hydroxymethyl radical) to demonstrate the numerical stability, convergence, and chemical reasonableness of the NRT bond indices in comparison to other measures of valency and bond order in current usage. © 1998 John Wiley & Sons, Inc. J Comput Chem 19: 610–627, 1998
References
- 1(a) L. Pauling and G. W. Wheland, J. Chem Phys., 1, 362 (1933). G. W. Wheland and L. Pauling, J. Am. Chem. Soc., 57 2086, (1935). G. W. Wheland, The Theory of Resonance and Its Application to Organic Chemistry, John Wiley, New York, 1955. (b) L. Pauling, Nature of the Chemical Bond, 3rd ed., Cornell University Press, Ithaca, NY, 1960.
- 2 P.-O Löwdin, Phys. Rev., 97, 1474, (1955). E. R. Davidson, Density Matrices in Quantum Chemistry, Academic, New York, 1976.
- 3 J. P. Foster and F. Weinhold, J. Am. Chem. Soc., 102, 7211 (1980). A. E. Reed and F. Weinhold, J. Chem. Phys., 78, 4066 (1983). A. E. Reed, R. B. Weinstock, and F. Weinhold, J. Chem. Phys., 83, 735 (1985). A. E. Reed and F. Weinhold, J. Chem. Phys., 83, 1736 (1985).
- 4
NBO Program:
E. D. Glendening,
A. E. Reed,
J. E. Carpenter, and
F. Weinhold,
QCPE Bull.,
10,
58
(1990).
A. E. Reed and
F. Weinhold
QCPE Bull.,
5,
141
(1985).
Reviews of NBO Method and Applications:
A. E. Reed,
L. A. Curtiss, and
F. Weinhold,
Chem. Rev.,
88,
899
(1988).
F. Weinhold and
J. E. Carpenter,
in R. Naaman, and
Z. Vager, (eds).
The Structure of Small Molecules and Ions,
Plenum, New York,
1988.
10.1007/978-1-4684-7424-4_24 Google Scholar
- 5(a) Bond length: L. Pauling, J. Am. Chem. Soc., 69, 542 (1947). Reference 1 (b), p. 255; C. A. Coulson, Proc. R. Soc. London, A207, 91 (1951). C. A. Coulson and A. Golebiewski, Proc. Phys. Soc., 78, 1310 (1961). G. Grampp, M. Cebe, and E. Cebe, Z. Phys. Chem., 166, 93 (1990). (b) Bond energy: H. S. Johnson, Adv. Chem. Phys., 3, 131 (1960). H. S. Johnston and A. C. Parr, J. Am. Chem. Soc., 85, 2544 (1963). N. Agmon and R. D. Levine, J. Chem. Phys., 71, 3034 (1979). P. Politzer and S. Ranganathan, Chem. Phys. Lett., 124, 527 (1986). (c) Other: W. Gordy, J. Chem. Phys., 14, 305 (1946). L. Peter, J. Chem. Ed., 63, 123 (1986). M. Barfield, M. J. Collins, J. E. Gready, S. Sternhell, and C. W. Tansey, J. Am. Chem. Soc., 111, 4285 (1989). D. K. Maity, and S. P. Bhattacharyya, J. Am. Chem. Soc., 112, 3223 (1990).
- 6 L. Pauling, L. O. Brockway, and J. Y. Beach, J. Am. Chem. Soc., 57, 2705 (1935).
- 7(b) Reference 1 p. 282.
- 8(b) Reference 1, p. 499; (b) Pauling, L. Proc. Roy. Soc. London, A356, 433 (1977).
- 9 Useful discussion of the nature of valency, electrovalency, and covalency from a historical perspective are given by N. V. Sidgwick, The Electronic Theory of Valency, Oxford University Press, Longon, 1929, particularly Chapter VI; C. A. Russell, History of Valency, University Press, Leicester, 1971.
- 10 In principle, one might have (e.g), for an excited state. higher occupancy of an antibond than of the corresponding bond, which would lead to a net negative contribution to formal bond order. However, this situation is not provided for in the present version of the NRT program.
- 11 See, e.g., Reference 1 pp. 88–95, for discussion of the relationship of bond energies to electronegativity differences.
- 12
C. A. Coulson,
Proc. R. Soc. London,
A169,
413
(1939),
10.1098/rspa.1939.0006 Google ScholarProc. R. Soc. London, A207, 91 (1951). C. A. Coulson, Valence, 2nd ed., Oxford University Press, London, 1961, pp. 266–70.
- 13 K. B. Wiberg, Tetrahedron, 24, 1083 (1968).
- 14 D. R. Armstrong, P. G. Perkins, and J. J. P. Stewart, J. Chem Soc., Dalton Trans., 1973, 838 (1973). N. P. Boresova and S. G. Semenov, Vestn. Leningrad Univ., 16, 119 (1973). M. S. Gopinathan and K. Jug, Theor. Chim. Acta, 63, 497, 511 (1983). I. Mayer and M. Revesz, Inorg. Chim. Acta, 77, L205 (1983). I. Mayer, Chem. Phys. Lett., 97, 270 (1983). I. Mayer, Int. J. Quantum Chem., 26, 151 (1984). M. S. De Giambiagi, M. Giambiagi, and F. E. Jorge, Z. Naturforsch., A39, 1259 (1984). M. A. Natiello and J. A. Medrano, Chem. Phys. Lett., 105, 180 (1984). K. Jug, J. Comput. Chem., 5, 555 (1984). I. Mayer, Int. J. Quantum Chem., 29, 477 (1986). I. Mayer, J. Mol. Struct. Theochem., 149, 81 (1957). K. Jug, E. Fasold, and M. S. Gopinathan, J. Comput. Chem., 10, 965 (1989). G. Lendvay, J. Phys. Chem., 93, 4422 (1989).
- 15 For critical discussion of the terminology employed by Jug and coworkers,14 see I. Mayer, Theor. Chim. Acta, 67, 315 (1985). Note that the orbitals called “natural hybrid orbitals” and “natural bond orbitals” by Jug and coworkers in 1983 are unrelated to the corresponding quantities defined previously in references 3 and 4.
- 16 The quantities introduced by I. Mayer14 are closely associated with Mulliken population analysis and its well-known pathologies in larger basis sets. For example, Baker “J. Baker, Theor. Chim. Acta, 68, 221 (1985)” cites an example in which the calculated Mayer valency of carbon changes from +3.39 to y- (!)4.86 when diffuse functions are added.
- 17 Jug and coworkers (see citations in reference 14) have variously suggested Wiberg's formula with no overlap, with Löwdin orthogonalization, and with a modified scheme of occupancy-weighted symmetric orthogonalization.
- 18 F. Weinhold and J. E. Carpenter, J. Mol. Struct. Theochem., 165, 189 (1980).
- 19 A. E. Reed and P. v. R. Schleyer, Inorg Chem., 27, 3969 (1988). J. Am. Chem. Soc., 112, 1434 (1990).
- 20 J. Cioslowski and S. T. Mixon, J. Am. Chem. Soc., 113, 4142 (1991).
- 21 R. F. W. Bader and T. T. Nguyen-Dang, Adv. Quantum Chem., 14, 63 (1981).
- 22 K. B. Wiberg and P. R. Rablen, J. Am. Chem. Soc., 115, 614 (1943).
- 23 I. D. Brown and R. D. Shannon, Acta Crystall., A29, 266 (1973). I. D. Brown and D. Altermatt, Acta Crystall., B41, 244 (1985). For applications, see I. D. Brown, J. Solid State Chem., 82, 122 (1989). M.-H. Whangbo and C. C. Torardi, Science, 249, 1143 (1990).
- 24 W. G. Penny, Proc. R. Soc. London, A158, 306 (1937). W. C. Herndon, J. Am. Chem. Soc., 95, 2404 (1973). I. Gutman and N. Trinajstic, Top. Curr. Chem., 42, 49 (1973). I. Gutman and H. Sachs, Z. Phys. Chem., 268, 257 (1987), and references therein.
- 25 J. A. Pople and M. Gordon, J. Am. Chem. Soc., 87, 4253 (1967).
- 26 For a comprehensive description of computational methods and basis set designations used herein, see W. J. Hehre, L. Radom, P. v. R. Schleyer, and J. A. Pople, Ab Initio Molecular Orbital Theory, John Wiley, New York, 1986. All wave function calculations were performed with the Gaussian 92 program system: M. J. Frisch, G. W. Trucks, M. Head-Gordon, P. M. W. Gill, M. W. Wong, J. B. Foresman, B. G. Johnson, H. B. Schlegel, M. A. Robb, E. S. Repogle, R. Gomperts, J. L. Andres, K. Raghavachari, J. S. Binkley, C. Gonzalez, R. L. Martin, D. J. Fox, D. J. Defrees, J. Baker, J. J. P. Stewart, and J. A. Pople, Gaussian 92, Revision A, Gaussian, Inc., Pittsburgh, 1992.
- 27 Note that bond order-bond length correlations are not necessarily expected to be linear. For example, Coulson12originally suggested a linear relation between bond length and PAB, but the same cannot be expected for MM, WBI, and Jug indices, which are quadratic in PAB.
- 28 Of course, it is quite artificial to consider such a “transition” in the present case; the default NRTWGT criterion for acceptance as a reference structure (wα ≥ 0.35*wmax) en- sures that only the most strongly delocalized systems will have more than a single reference structure.
- 29 The default NRT expansions for these wave functions include about 20 additional structures (not shown) with smaller weightings ≥ 0.01%.
- 30 The superficially high weight of “long bond” structures in earlier RT treatments can be attributed to the “overcorrelation” artifact of Heitler-London valencebond wave functions; see R. McWeeny, Proc. R. Soc. London, A223, 63, 306 (1954).
- 31 This differs slightly from the geometries employed by Cioslowki and Mixon20 (optimized HF/6-31G*. and Wiberg and Rablen22 (partially optimized MP2/6-31G*, with C-F bonds constrained to equal the value for CH3 F). These 3 geometry differences are likely to be unimportant on the scale of the comparisons considered here. The large difference between CHF3 values of bCF reported in references 20 (0.572, RHF/6-31+ + G* level) and 22 (0.616 RHF/6-311 + +. G* level) is therefore puzzling.
- 32 The CM VC is rather close to the NRT covalency (2.427) in this case, but the corresponding valencies for fluorine differ by almost a factor of two.
- 33 J. E. Carpeter and F. Weinhold, J. Mol. Struct. Theochem., 169, 41 (1988).
