QUINQUEDENTATE COORDINATION OF AMINO-SUBSTITUTED TETRAAZACYCLOALKANES TO COBALT(III) .2. CRYSTAL-STRUCTURES OF TRANSISOMERS, MOLECULAR MECHANICS CALCULATIONS AND BASE-HYDROLYSIS KINETICS

Chlorocobalt(III) complexes of the pendant-arm macrocycles 1 2-methyl-1,4,7,10-tetraazacyclotridecan-12-amine (L13), 6-methyl-1,4,8,11-tetraazacyclotetradecan-6-amine (L14), 10-methyl-1,4,8,12-tetraazacyclopentadecan-10-amine (L15) and 3-methyl-1,5,9,13-tetraazacyclohexadecan-3-amine (L16) were isolated as trans isomers only for L16 and the minor isomer of L14. Of the two trans complexes, [Co(L14)Cl][ClO4]2 crystallized in the monoclinic space group P2(1)/c, a = 9.107(3), b = 16.448(4), c = 13.898(5) angstrom and beta = 99.16(3)-degrees, and [Co(L16)Cl]Cl[ClO4] crystallized in the same space group, a = 16.868(9), b = 7.531(5), c = 20.51(2) angstrom and beta = 126.61(5)-degrees. Single-crystal X-ray structure determinations were refined to residuals of 0.056 and 0.071 for 2711 and 2314 'observed' reflections respectively. In both cases the pendant primary amine and two adjacent secondary amines necessarily occupy an octahedral face, with the chloro ligand trans to the primary amine. Average macrocycle Co-N distances vary with ring size (1.96(1), 2.01(0) angstrom for L14 and L16 respectively), as does the Co-Cl distance [2.244(2), 2.222(3) angstrom respectively], yet the Co-N(pendant) distance is constant within experimental error for both structures [1.961(4), 1.956(8) angstrom respectively]. Molecular mechanics calculations have been employed to predict the isomer preferences in all cases, and define the cis isomer of the L14 complex as more stable than the trans isomer by 4.4 kJ mol-1 (corresponding to a predicted cis:trans ratio of 86:14), reasonably consistent with the experimental ratio of ca. 98:2. For the trans isomers, base hydrolyses are rapid (k(OH) 9100 and 11 100 dm3 mol-1 s-1 for L14 and L16 respectively), and not particularly sensitive to clear differences in Co-Cl distance. For the cis isomers the Co-Cl distances are minimized for L14 (2.24(5), 2.23(6), 2.27(3) angstrom for L13, L14 and L15 respectively), and this trend is reflected well in the comparative rate constants for base hydrolysis (k(OH) 4300, 76 and 6700 dm3 mol-1 s-1 respectively). Variations in rate constants are tied to variations in the activation enthalpy, but not the activation entropy. Factors influencing base-hydrolysis rate constants for the series are discussed, and the significance of ground-state effects on hydrolysis rate is examined.