A conformational search at the MM3 level was performed for 12-crown-4 (12c4) whereby 180 conformations were predicted. To determine the lowest energy conformations and to get a more accurate energy order of the predicted conformations, geometry optimization was performed for the 180 conformations at the HF/STO-3G level and for the 100 lowest energy conformations, according to HF/STO-3G energy order, at the HF/4-31G and HF/6-31+G* levels. Some of the 100 conformations had equal energies at the three abovementioned levels and consequently 37 conformations were excluded. Further computations were performed for the 20 lowest energy unique conformations, according to the MP2/6-31+G*//HF/6-31+G* energy order, at the B3LYP/6-31+G*, MP2/6-31+G*//B3LYP/6-31+G*, and MP2/6-31+G* levels. Good agreement was found between the energy order of the conformations at the MP2/6-31+G*//HF/6-31+G* and MP2/6-31+G*// B3LYP/6-31+G* levels and that at the MP2/6-31+G* level, the most accurate level considered in this work. The relative energies of the predicted conformations at the MP2/6-31+G*//B3LYP/6-31+G* level are close to those at the MP2/6-31+G* level, to within 0.1 kcal/mol at most. This is with the exception of only two conformations. This suggests that the cheaper MP2/6-31+G*//B3LYP/6-31+G* level may be used to determine the relative energy order of conformations of larger molecules where the MP2/6-31+G* computations are prohibitively expensive. The closeness of the MP2/6-31+G*//B3LYP/6-31+G* and MP2/6-31+G* relative energies is shown to be a reflection of the closeness of the B3LYP6-31+G* and MP26-31+G* optimized geometries. For the two conformations where the difference of the relative energies was larger than 0.1 kcal/mol, large differences between some of the B3LYP/6-31+G* and MP2/6-31+G* ring dihedral angles were found. The calculated results show that the correlation energy is necessary to obtain an accurate energy order of the predicted conformations. A rationalization of the energy order of some of the predicted conformations in terms of the CH‚‚‚O interactions is given.

The vibrational, Raman, and IR, spectra of the five 12-crown-4 (12c4) complexes with Li+, Na+, K+, Rb+, and Cs+ alkali metal cations were measured. Except for a small shift of the position of some bands in the vibrational spectra of the Li+ complex, the vibrational spectra of the five complexes are so similar that it is concluded that the five complexes exist in the same conformation. B3LYP/6-31+G* force fields were calculated for six of the eight predicted conformations in a previous report (J. Phys. Chem. A 2005, 109, 8041) of the 12c4-Li+, Na+, and K+ complexes that are of symmetries higher than the C1 symmetry. These six conformations, in energy order, are of C4, Cs, Cs, C(2v), C(2v), and Cs symmetries. Comparison between the experimental and calculated vibrational frequencies assuming any of the above-mentioned six conformations shows that the five complexes exist in the C4 conformation. This agrees with the fact that the five alkali metal cations are larger than the 12c4 ring cavity. The B3LYP/6-31+G* force fields of the C4 conformation of the Li+, Na+ and K+ complexes were scaled using a set of eight scale factors and the scale factors were varied so as to minimize the difference between the calculated and experimental vibrational frequencies. The root-mean-square (rms) deviations of the calculated frequencies from the experimental frequencies were 7.7, 5.6, and 5.1 cm(-1) for the Li+, Na+, and K+ complexes, respectively. To account for the earlier results of the Li+ complex that the Cs conformation is more stable than the C4 conformation by 0.16 kcal/mol at the MP2/6-31+G* level, optimized geometries of the complex were calculated for the C4 and Cs conformations at the MP2/6-311++G** level. The C4 conformation was calculated to be more stable than the Cs conformation by 0.13 kcal/mol.

A conformational search was performed for 18-crown-6 using the CONLEX method at the MM3 level. To have a more accurate energy order of the predicted conformations, the predicted conformations were geometry optimized at the HF/STO-3G level and the 198 lowest energy conformations, according to the HF/STO-3G energy order, were geometry optimized at the HF/6-31+G level. In addition, the 47 nonredundant lowest energy conformations, according to the MP2/6-31+G energy order at the HF/6-31+G optimized geometry, hereafter the MP2/6-31+G//HF/6-31+G energy order, were geometry optimized at the B3LYP/6-31+G level. According to the MP2/6-31+G//B3LYP/6-31+G energy order, three conformations had energies lower than the experimentally known Ci conformation of 18c6. At the MP2/6-31+G//B3LYP/6-31+G level, the S6 lowest energy conformation is more stable by 1.96 kcal/mol than this Ci conformation. This was confirmed by results at the MP2/6-31+G level with an energy difference of 1.84 kcal/mol. Comparison between the structure of the S6 conformation of 18c6 and the S4 lowest energy conformation of 12-crown-4, as well as other important conformations of both molecules, is made. It is concluded that the correlation energy is necessary to have an accurate energy order of the predicted conformations. A rationalization of the conformational energy order in terms of the hydrogen bonding and conformational dihedral angles is given. It is also suggested that to have a better energy order of the predicted conformations at the MM3 level, better empirical force fields corresponding to the hydrogen bond interactions are needed.

A conformational search was performed for the 12-crown-4 (12c4)-alkali metal cation complexes using two different methods, one of them is the CONFLEX method, whereby eight conformations were predicted. Computations were performed for the eight predicted conformations at the HF/6-31+G*, MP2/6-31+G*//HF/6-31+G*, B3LYP/6-31+G*, MP2/6-31+G*//B3LYP/6-31+G*, and MP2/6-31+G* levels. The calculated energies predict a C4 conformation for the 12c4-Na+, -K+, -Rb+, and -Cs+ complexes and a C(s) conformation for the 12c4-Li+ complex to be the lowest energy conformations. For most of the conformations considered, the relative energies, with respect to the C4 conformation, at the MP2/6-31+G*//B3LYP/6-31+G* are overestimated, compared to those at the MP2/6-31+G* level, the highest level of theory considerd in this report, by 0.2 kcal/mol. Larger relative energy differences are attributed to larger differences between the B3LYP and MP2 optimized geomtries. Binding enthalpies (BEs) were calculated at the above-mentioned levels for the eight conformations. The agreement between the calculated and experimental BEs is discussed.