Semi-Ab-initio Model 1

The newest semiempirical method is Semi-Ab initio Model 1 (SAM 1) by Dewar et al. [83]. As the acronym suggests, it is based on AML The two-electron integrals (jiv (tt), however, are calculated analytically over Gaussian-type functions and scaled empirically. Up to the present, no comprehensive list of SAMI parameters and error... 

In 1993, Dewar and co-workers modified AMI to give the SAMI (semi-ab initio model 1) method [M. J. S. Dewar, C. lie, and G. Yu, Tetrahedron, 23,5003 (1993) A. J. Holder and E. M. Evleth in D. A. Smith (ed.). Modeling the Hydrogen Bond, American Chemical Society, 1994, p. 113]. A major difference between SAMI and AMI is that SAMI evaluates the two-center ERIs as (/ii Acr)sAMi = s( ab)(m Ko-)stc)-3g> where ( v Ao-)sto-3g is the accurate value of the ERI calculated using a STO-3G basis set, and the function (Rab) is a certain function of the intemuclear distance that reduces the magnitudes of the ERIs so as to allow for electron correlation and use of a minimal basis set. The function (Rab) contains parameters whose values have been adjusted to maximize the performance of the method. Because of the need to calculate two-center ERIs accurately, SAMI is slower than AMI, but is still far faster than ab initio methods, due to the NDDO approximation. 

The semi ab initio model 1 (SAMI) is another modified NDDO method, but it does not replace integrals by parameters. The one- and two-center electron repulsion integrals are explicitly calculated from the basis functions [employing a standard STO-3G (Slater-type orbital from three Gaussian functions) Gaussian basis set] and scaled by a function which has to be parametrized. SAM 1 calculations take about twice as long as AMI or PM3 calculations do. 

At the first stage quantum-chemical investigation of ethyl alcohol molecule interaction with one and a few water molecules was carried out. The calculations were carried out in the frameworks of different semi-empiri-cal and ab initio models. Thus based on the data obtained we can make a conclusion on the error value that can occur due to the use of simplified models with small basis sets. The calculation results are given in Table 8.1. 

There are basically two current approaches in molecular applications for approximating (1) the model core potential (MCP) one, also its extension to the ab initio model potential (AIMP) [72, 73] and (2) the semi-local pseudopotential (PP) approximation [74, 75]. 

Results. In Table 5.1 we compare a few results of classical, semi-classical and quantum moment calculations. An accurate ab initio dipole surface of He-Ar is employed (from Table 4.3 [278]), along with a refined model of the interaction potential [12]. A temperature of 295 K is assumed. The second line, Table 5.1, gives the lowest three quantum moments, computed from Eqs. 5.37, 5.38, 5.39 the numerical precision is believed to be at the 1% level. For comparison, the third line shows the same three moments, obtained from semi-classical formulae, Eqs. 5.47 along with 5.37 with the semi-classical pair distribution function inserted. We find satisfactory agreement. We note that at much lower temperatures, and also for less massive systems, the semi-classical and quantal results have often been found to differ significantly. The agreement seen in Table 5.1 is good because He-Ar at 295 K is a near-classical system. 

The basis of molecular modeling is that all important molecular properties, i. e., stabilities, reactivities and electronic properties, are related to the molecular structure (Fig. 1.1). Therefore, if it is possible to develop algorithms that are able to calculate a structure with a given stoichiometry and connectivity, it must be possible to compute the molecular properties based on the calculated structure, and vice versa. There are many different approaches and related computer programs, including ab-initio calculations, various semi-empirical molecular orbital (MO) methods, ligand field calculations, molecular mechanics, purely geometrical approaches, and neural networks, that can calculate structures and one or more additional molecular properties. 

The ground-state vibrational normal modes of uracil have also been extensively studied, both experimentally and computationally. The IR and Raman spectra in Ar matrix have been measured for the 5-d, 6-d, 5,6-d2, 1,3-d2, l,3,5-<73, 1,3,6-d3, and d4 isotopomers [120-122], Vibrational spectra in the crystalline phase have been reported for the 5,6-d2, 1,3-d2, and d4 isotopomers of uracil [123] and of the 2-1S0, 4-lsO, 3-d, 5-d, 6-d, 5,6-d2 and l-methyl-<73 isotopomers of 1-methyluracil [124], UV Resonance Raman spectra have been reported for natural abundance, 2-lsO, 4-lsO, and 2,4-1802 uracil in neutral aqueous solution [125]. These data have been modeled successfully by both ab initio [94, 117, 126-132] and semi-empirical [133, 134] calculations. However, most of these caculations ignore electron correlation effects on the vibrational properties of uracil, particularly the Raman and resonance Raman spectra. However, the most robust reconciliation of experiment and computation is a recent attempt to computationally reproduce the experimentally observed isotopic shifts in 4 different uracil isotopomers [116], The success of that attempt is an indication of the reliability of the resulting force field and normal modes for uracil. The resonance Raman vibrations of uracil, and their vibrational assignments, are given in Table 9-2. 

---