Theoretica chimica acta

, Volume 26, Issue 3, pp 211–230 | Cite as

The role of kinetic energy in chemical binding

II. Contragradience
  • William A. GoddardIII
  • C. Woodrow WilsonJr.


In a preceding paper, we examined the GI wavefunctions of small molecules and found that the nonclassical or exchange kinetic energy, Tx, dominates the changes in energy involved in chemical binding. Here we examine more closely the changes in Tx with internuclear separation and find that ΔTx is large for valence orbitals centered on different atoms because of the large region in which the orbitals are contragradient (i.e., have gradients in obtuse directions). In fact for H2 this contragradience accounts for 93% of the calculated binding energy. In addition, the behavior of Tx and ΔTx can usually be predicted from consideration of the permutational symmetry (Young tableau) involved in the wavefunction. The concepts developed here provide an alternative interpretation of the nature of the chemical bond.


Kinetic Energy Binding Energy Chemical Bond Large Region Alternative Interpretation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Die in einer vorhergehenden Arbeit durchgeführten Untersuchungen der GI-Wellenfunktionen kleiner Moleküle führten zu dem Ergebnis, daß der nichtklassische oder Austauschanteil der kinetischen Energie Txbei Änderungen der Energie chemischer Bindungen überwiegt. In dieser Arbeit werden die Änderungen von Txmit dem Kernabstand näher untersucht. Wir finden ein großes ΔTx für an verschiedenen Atomen zentrierte Valenzorbitale wegen der großen Region, in der die Orbitale „kontragradient“ (d. h. sie haben Gradienten in Richtungen, die einen stumpfen Winkel miteinander bilden) sind. Für H2 sind 93% der berechneten Bindungsenergie auf diese „Kontragradienz“ zurückzuführen. Ferner kann das Verhalten von Tx und ΔTx normalerweise aus der Permutationssymmetrie (Young Tafeln) der Wellenfunktionen vorausgesagt werden. Die hier entwickelten Methoden erlauben eine alternative Interpretation der chemischen Bindung.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Wilson,C.W., Jr., Goddard III,W.A.: Theoret. chim. Acta (Berl.), 26, 195 (1972) (Hereafter we will refer to this as paper I.)Google Scholar
  2. 2.
    Goddard III,W.A.: Physic. Rev. 157, 82 (1967).Google Scholar
  3. 3.
    —: Physic. Rev. 157, 73 (1967).Google Scholar
  4. 4.
    Wilson,C.W., Jr., Goddard III,W.A.: Chem. Physics Letters 5, 45 (1970).Google Scholar
  5. 5.
    Hellmann,H.: Z. Physik 85, 180 (1933).Google Scholar
  6. 6.
    Feinberg,M.J., Ruedenberg,K.: J. chem. Physics 54, 1495 (1971).Google Scholar
  7. 7.
    As shown by Pauli [Physics Rev. 58, 716 (1940)], Bosons must have integral spins and Fermions half-integral spin; however, here we consider spin-1/2 Bosons in order to demonstrate the effects of the Pauli principle on bonding.Google Scholar
  8. 8.
    Lieb,E., Mattis,D.: Physic. Rev. 125, 164 (1962); Courant,R., Hilbert,D.: Methods of mathematical physics, p. 451. New York: Interscience 1953.Google Scholar
  9. 9.
    A better approximation would be obtained by including a proportionality constant d jkin (19) where d jkis typically in the range 0.2 to 0.8.Google Scholar
  10. 10.
    Wilson,C.W., Jr., Goddard III,W.A.: J. chem. Physics 51, 716 (1969).Google Scholar
  11. 11.
    Ladner,R.C., Goddard III,W.A.: J. chem. Physics 51, 1073 (1969).Google Scholar
  12. 12.
    Bender,C.F., Davidson,E.R.: J. chem. Physics 49, 4222 (1968).Google Scholar
  13. 13.
    Using (33) and (20), we find that the energy separations between the states described by (30)–(32) are consistent with the interatomic potential terms being proportional to S(S + 1) (where S is the total spin). That is, at this level of approximation the interaction energies could be described by a Heisenberg Hamiltonian using an atomic spin of one on each atom.Google Scholar
  14. 14.
    Klein,D.J.: J. chem. Physics 50, 5152 (1969).Google Scholar
  15. 15.
    Feinberg,M.J., Ruedenberg,K., Mehler,E.L.: Advances in quant. Chemistry 5, 28 (1970). (We thank Dr. Ruedenberg for a copy of this paper prior to publication.)Google Scholar
  16. 16.
    Bader,R.F., Henneker,W.H., Cade,P.E.: J. chem. Physics 46, 3341 (1967); Bader,R.F., Baudraut, A.D.: J. chem. Physics 49, 1653 (1968).Google Scholar
  17. 17.
    Ruedenberg,K.: Rev. mod. Physics 34, 326 (1962).Google Scholar
  18. 18.
    See, for example, Van Vleck,J.H., Sherman,A.: Rev. mod. Physics 7, 167 (1935); Moore,W.J.: Physical chemistry (Prentice-Hall, Inc., Englewood Cliff, N.J., 1962), pp. 517ff.Google Scholar
  19. 19.
    We denote the orbital density matrices as D jk rather than D kj as in Ref. [2]. In addition, the normalization terms are now included in D jk.Google Scholar
  20. 20.
    Taking the orbitals to be real.Google Scholar

Copyright information

© Springer-Verlag 1972

Authors and Affiliations

  • William A. GoddardIII
    • 1
  • C. Woodrow WilsonJr.
    • 1
  1. 1.Arthur Arnos Noyes Laboratory of Chemical PhysicsCalifornia Institute of TechnologyPasadena

Personalised recommendations