Linear Combinations of Model Compounds

QM/MM boundary through space in such a way that the sterically bulky groups fall on the MM side and the interesting part of the molecule falls on the QM side. Finally, to avoid the question of how to deal with a cut bond, one may assume that the electronic structure of the QM region will be of similar quality with either the non-polar, bulky group as a cap, or with simply hydrogen atoms as caps. With such a philosophy, the energy of the system as a whole may be expressed as a linear combination of model compounds of different size and at different levels of theory. In simplest form 

Subsequently, Humbel, Sieber, and Morokuma (1996) generalized the IMOMM optuniza-tion scheme to the case where two different levels of QM theory were used instead of a QM/MM approach, and Svensson, Humbel, and Morokuma (1996) examined the relative efficacy of different combinations of levels for prototype problems. Corchado and Truhlar (1998) later proposed a refinement of that methodology to improve computed vibrational frequencies and Rickard et al. (2003) showed that a combination of MP2 and HF theories permits the calculation of high-quality NMR chemical shifts within the high-level system. 

---

The final states may be mixed with other orbitals and can be used to determine the coordination environment of an element in a compound, its electron density and oxidation state. The intensity can be used to accurately determine relative oxidation state ratios. Furthermore, XANES can be used to determine relative amounts of species by linear combination of individual compounds. In recent years, the codes for XANES calculation (especially John Rehr s FEFF code), have significantly improved and it can be expected that theoretical models of compounds will be accurately determined by XANES measurement and calculation in the future. 

The DD-CSA cross-correlated relaxation, namely that between 13C-1H dipole and 31P-CSA, can also be used to determine backbone a and C angles in RNA [65]. The experiment requires oligonucleotides that are 13C-labeled in the sugar moiety. First, 1H-coupled, / - DQ//Q-II CP spectra are measured. DQ and ZQ spectra are obtained by linear combinations of four subspectra recorded for each q-increment. Then, the cross-relaxation rates are calculated from the peak intensity ratios of the doublets in the DQ and ZQ spectra. The observed cross-correlation rates depend on the relative orientations of CH dipoles with respect to the components of the 31P chemical shift tensor. As the components of the 31P chemical shift tensor in RNA are not known, the barium salt of diethyl phosphate was used as a model compound with the principal components values of -76 ppm, -16 ppm and 103 ppm, respectively [106]. Since the measured cross-correlation rates are a function of the angles / and e as well, these angles need to be determined independently using 3/(H, P) and 3/(C, P) coupling constants. 

An example of how ab initio calculations may be applied to the study of fragments of polymer chains is given by Jaffe, Yoon, and McLean, who studied a series of mono- and diphenyl molecules containing up to 35 atoms. These compounds are models for a variety of important polymers such as polycarbonates (see Figure 1), polyimides, aromatic polyamides, aromatic polyesters, and polyether sulfone. A variety of basis sets, representing linear combination of Gaussian functions to approximate Slater-type orbitals (STOs) as compiled in Table 1, were employed. 

Trial wavefunctions are usually constructed by linear combination of Gaussian error functions that are convenient to integrate. The results can be of predictive value and such calculations have become everyday tools for chemists in all branches of chemistry, to guide experiments and not least to rule out untenable hypotheses. This is a remarkable achievement that seemed to be out of reach a few decades ago. Still, simple qualitative models that are amenable to perturbation theory are required to understand and predict trends in a series of related compounds. Our goal here is to describe the minimal quantum mechanical models that can still provide a useful qualitative description of electronically excited states, their electronic stmcture and their reactivity. Such models also provide a language to convey the results of state-of-the-art, but essentially black-box ab initio calculations. 

The He(I) photoelectron spectra of the cage compounds (305a), (312) and (313) <82AG(E)375, 84AG(E)805> have been recorded <85IC4020>. The assignment of the first five bands in the PE spectra was based on model calculations (MINDO/3), correlation with related species, and a band shape analysis. The highest occupied MOs of all three nortricyclanes are the E and A, linear combinations of the lone pairs centered at the As, P3, and Sbj units. 

An important qualitative description of the spectral behavior of class II compounds was presented by Robin and Day. This simple model has found applicability to the discussion of the spectra of numerous mixed valence compounds in which some delocalization occurs. In this model, it is assumed that the ground-state wave function r/fQ contains the function, a, which describes mixing of the wave function for site A with the wave function of site B. Assuming that these wave functions are linear combinations of atomic functions, 4>i, Robin and Day showed that the wave function describing the optical electron spanning sites A and B is given by equation (1). In this equation. 

---