Early Examples of Surface Calculations for Closed-shell Systems

Early Examples of Surface Calculations for Closed-shell Systems.—The first LCAO-MO SCF calculation of a reaction pathway on a potential surface for a reaction between two closed-shell species was performed by Clementi for the system NHa + HC1 to form NH4CI. The calculation employed contracted sets of GTO s (llf,7p)- [5j,3p] for N, (6s,p) [3s,p for H and (llr,7p)— [5r,4p] for Cl. Since the basis set does not contain any polarizing functions on the N and Q centres, it will not yield results at the Hartree-Fock limit. For example, this basis set predicts a barrier height for the inversion motion in NH3 which is almost twice as large as the experimental value, while as pointed out before, this same barrier is quantitatively accounted for at the Hartree-Fock limit. The HQ molecule was constrained to 

It had been anticipated that the reaction pathway for a proton transfer reaction should be characterized by a double minima in the potential surface.100 The failure of Qementi s calculation to predict a barrier along the reaction pathway for the transfer of H + from HQ to NH3 to form NHJC1-, must be accepted with some caution in view of the lack of polarizing functions on N and Cl in the basis set. This work, however, broke new ground and dementi99 includes in his paper a detailed discussion of the merits and faults to be expected for the Hartree-Fock method in the different regions of the potential surface. 

Mulliken and W. B. Pearson, Molecular Complexes A Lecture and Reprint Volume , John Wiley and Sons, New York, 1969. 

Gordon and Secrest110 in calculating the potential surface for the He-Hg system also find that the near-Hartree-Fock SCF results parallel those obtained using correlated wavefunctions for those portions of the surface for which the H—H separation is held fixed at its equilibrium value. 

The portion of the surface for the HeH+-H2 system calculated by Benson and McLaughlin,104 and referred to above, lies within the range of H-H separations found by Kutzelnigg et a/.105 to be adequately described at the Hartree-Fock limit. Their Hartree-Fock SCF results should, therefore, be accurate enough to warrant their use in classical trajectory calculations. Such calculations have indeed been performed by McLaughlin and Thompson.111 Spline interpolation functions were used to express the tabular near-Hartree-Fock energies in analytical form. The trajectory calculations for the reaction 

---