Collision-induced dissociation and theoretical studies of K + complexes with ammonia: a test of theory for potassium ions

Abstract

The sequential bond energies of K + with ammonia are determined using collision-induced dissociation (CID) with xenon gas in a guided ion beam tandem mass spectrometer. The kinetic energy dependences of the CID cross-sections are analyzed to give 0 and 298 K bond energies for the successive loss of ammonia after accounting for multiple collisions, internal energy, and dissociation lifetimes. We also consider whether to treat torsional motions as vibrations or rotations and whether to include the ion–dipole potential in the treatment of the transition state for dissociation. Bond energies at 0 K (in kJ/mol) of K +(NH 3) x are determined for x=1–5 as 79±7, 69±6, 59±5, 46±6, 31±11, respectively, decreasing with increasing x as expected for electrostatic bonds. These dissociation energies agree quantitatively with literature values available from equilibrium studies for x=1–4 (mean absolute deviation=2±1 kJ/mol). This agreement suggests that these systems can be used to test theory for K + complexes, and therefore several levels of theory are explored here. The x=1–5 complexes are calculated to adopt high symmetry structures where the ligands bind directly to the metal. We find that trends in the experimental bond energies are accurately reproduced at the MP2(full)/6-311+G(2d,2p)//MP2(full)/6-31G(d) and B3LYP/6-311+G(2d,2p)//B3LYP/6-31G levels, although the values are generally low, by an average of 4±3 kJ/mol (8±3%). Values obtained using the effective core potentials of Hay–Wadt and Stuttgart–Dresden on potassium are less satisfactory and underestimate experiment by an average of 9±6 kJ/mol (14±7%).