Partially bridge-fluorinated dimethyl bicyclo[1.1.1]pentane-1,3-dicarboxylates: preparation and NMR spectra.

Direct fluorination of dimethyl bicyclo[1.1.1]pentane-1,3-dicarboxylate, obtained from [1.1.1]propellane prepared by an improved synthetic procedure, furnished esters of 14 of the 15 possible bridge-fluorinated bicyclo[1.1.1]pentane-1,3-dicarboxylic acids, isolated by preparative GC. Calculated geometries reflect the substitution pattern in a regular fashion compatible with Bent's rules. Considerable additional strain is introduced into the bicyclo[1.1.1]pentane cage by polyfluorination; it is calculated to be as high as 33-35 kcal/mol for hexasubstitution. Three arrangements of the fluorine substituents are especially strain-rich: geminal, proximate, and W-related. The (1)H, (13)C, and (19)F NMR spectra exhibit a striking variety of chemical shifts and long-range coupling constants. These are in good agreement with results calculated with neglect of the bridgehead substituents for all of the chemical shifts by the GIAO-RHF/6-31G//RHF/6-31G and GIAO-RHF/6-31G//MP2/6-31G methods and for many of the coupling constants by the EOM-CCSD/6-311G//MP2/6-311G method. The proximate (4)J(FF) constants are particularly large (50-100 Hz) and show an inverse linear dependence on the calculated F-F distance in the range 2.43-2.58 A.

Stability of complexes of aromatic amides with bromide anion: quantitative structure--property relationships

Most of the theoretical studies published to-date on the structural and electronic properties of supramolecules have been devoted to the neutral or cationic complexes, while little is known about anionic systems. A detailed theoretical study of the interaction between simple aromatic amides and the bromide anion has recently been published (Cajan, M.; Stibor, I.; Koca, J. J. Phys. Chem. A 1999, 103, 3778). The present work focuses on the structural and physicochemical parameters of simple aromatic amides related to their ability to form the 1:1 complex with a bromide anion. A quantitative structure-property relationships (QSPR) model for the prediction of association constants is proposed. The model based on 22 complexes and nine molecular descriptors explained 96% (84% cross-validated) of the variance in association constants. The descriptors employed in this model included parameters for the characterization of conformational behavior and the 3D structure of amide molecules, distribution of electron density on the amidic functional group, and parameters for substitution on aromatic units. The quantitative structure-property relationship approach predicted the association constants with comparable quality, but significantly lower computational demand, than molecular modeling or standard quantum chemistry calculations.