Geometries, calculated accurate

It is quite common to do the conformation search with a very fast method and to then optimize a collection of the lowest-energy conformers with a more accurate method. In some cases, single geometry calculations with more accurate methods are also performed. Solvent effects may also be important as discussed in Chapter 24. 

The ability of the MM method to calculate accurately potential energies and geometries of molecules is best suited for quantitative conformational analysis (44). Numerical values can readily be given to such terms as strain and stability (45). The topic is extensively covered in Allinger s review (lOd), therefore only some recent developments are mentioned. 

We now turn to the issue of physical accuracy. Just as accuracy should not be considered as a single topic, physical accuracy is too vague of an idea to be coherently considered. It is much better to ask how accurate the predictions of DFT are for a specific property of interest. This approach recognizes that DFT results may be quite accurate for some physical properties but relatively inaccurate for other physical properties of the same material. To give just one example, plane-wave DFT calculations accurately predict the geometry of crystalline silicon and correctly classify this material as a semiconductor,... 

Fig. 12. Comparison between the interaction energies derived from KSCED calculations applying local density approximation for 22 intermolecular complexes at equilibrium geometry with accurate ab initio (CCSD(T)) reference data. Details of calculations can be found in [Wesolowski and Tran,...

The Heisler [36] cooling charts and the Grober [29] heat loss fraction charts for the three geometries can be calculated accurately by the single-term approximations [28,56]... 

CISD calculations of molecular properties often do not give results of high accuracy. For example, a comparison of molecular geometries calculated by several correlation methods found that the CISD method gave the poorest results of those correlation methods studied [T. Helgaker et al., J. Chem. Phys., 106,6430 (1997)]. Highly accurate Cl results require a CISDTQ calculation, which is generally impractical. 

Beyond a much higher speed, the major distinction of IMS from condensed-phase separations is that it is also a stmctural characterization tool of broad utility. This capability, central to the analytical profile and potential of IMS, arises from the possibility to compute the mobility (under some conditions) for any hypothetical geometry reasonably accurately. In this section, we review the approaches to calculation of ion mobilities in gases and point out the challenges of extending those methods to differential IMS. 

A useful alternative approach is to isolate the components of the perturbation expansion, namely the repulsion, electrostatic interaction, induction, and dispersion terms, and to calculate each of them independently by the most appropriate technique. Thus the electrostatic interaction can be calculated accurately from distributed multipole descriptions of the individual molecules, while the induction and dispersion contributions may be derived from molecular polarizabilities. This approach has the advantage that the properties of the monomers have to be calculated only once, after which the interactions may be evaluated easily and efficiently at as many dimer geometries as required. The repulsion is not so amenable, but it can be fitted by suitable analytic functions much more satisfactorily than the complete potential. The result is a model of the intermolecular potential that is capable of describing properties to a high level of accuracy. 

---