ORGANIC MASS SPECTROMETRY, VOL. 27, 1203-1209 (1992) Calculation of the Kinetic Energy Release of Charge Separation Processes? Karoly Vekey Central Research Institute for Chemistry zyxwvutsrq of the Hungarian Academy of Sciences, Pusztaszeri ut 59-67, H-1025 Budapest, Hungary Gabriella Pksfalvi Department of Organic Chemistry, Kossuth Lajos University, P.O. Box 20, H-4010 Debrecen, Hungary Kinetic energy release (KER) was studied by experimental methods and semiempirical (MNDO and AMl) molec- ular orbital calculations in the case of various charge separation processes: loss of a methyl ion from [CH,-C,-CH,] zyxwvutsrq *+, ICH,-C,-CH3]*+ and [N,Ndimethyl-p-phenylenediamine] ’+. It was found that the KER corresponding to the width of a dish-topped peak at half-height is very close to the mean KER of the process. The calculated potential energy curves of these reactions show significant reverse critical energies, a large part of which was found to be due, in agreement with conventional assumptions, to electric repulsion between the two separating singly charged products. The bond order between the two separating ions is nearly zero in the transition state, so exchange of internal energy between them is unlikely. These explain the good agreement between the (calculated) reverse critical energy and the measured kinetic energy release. INTRODUCTION Kinetic energy release (KER) is one of the most impor- tant characteristics of a fragmentation process, giving valuable experimental information about the potential surface. Part of the KER originates from the excess energy of a reaction. This is distributed among vibra- tional degrees of freedom and the reaction coordinate. Some time ago’ a simple empirical equation was sug- gested : where zyxwvutsrqp (T) indicates the average KER, E$ is the excess energy and (3N - 6) is the number of vibrational degrees of freedom. Recent work indicates* that the empirical ‘constant’ 2.27 is not a constant, but depends on the excess energy. This ‘statistical’ part of the KER may be calculated by RRKM calculations based on the unimolecular reaction rate theory or the quasi- equilibrium the0ry.j KER originating from the excess energy is usually relatively small, rarely exceeding - 50 meV. Higher KER values, especially those exceeding 100 meV, usually have a different origin. When the reverse reaction has a critical energy (reverse critical energy, RCE), a significant part of it may be converted into KER in addition to that discussed above. The combined application of detailed molecular orbital and molecular dynamic calculations can accurately determine the par- titioning ratio: but this is possible only for very small systems. Generally it is little known how much of the RCE is release as KER and how much is converted into internal energy. trometry, Padova, Italy, 1992. (T) = 2.27E$/(3N - 6) t Presented in part at the 10th Informal Meeting on Mass Spec- A special case of large KER is observed in the charge separation reactions of doubly charged ions.’ These give a large KER (a few eV). It is assumed that the kinetic energy which is released as the two positively charged ions separate is mainly due to the coulombic portion of the RCE, while other contributions of the RCE are negligible. This feature of charge separation processes was used to probe the structure of doubly charged ions. The measured KER was assumed to be equal to the potential energy necessary to bring two point charges to a certain distance from each other. This distance gave an estimate (or more precisely an estimated lower limit) of the diameter (length) of the ion in its reactive configuration. To give an example,6 the KER of the [benzene]” -+ CH3+ + C,H,+ process is about 2.6 eV and the intercharge distance calculated from it is 5.5 A. This is significantly larger than the diameter of benzene. This suggests that in its reactive configuration the [benzeneI2+ ion has isomerized to an open-chain structure. The limitations of this, otherwise very successful, method are that it gives only lower limits and not actual values for the molecular size, uses the idea of completely localized charges and it is not always clear, where those charges should be located. In this work we have performed semiempirical (MNDO and AM1) molecular orbital calculations to examine this problem in a quantitative way, to study the basic assumptions discussed above, to determine the potential curves of some selected charge separation reactions and to compare the calculated RCE with the observed KER. In addition to energetic values, other important structural parameters such as bond orders and charge densities were also considered. The experimental problem of determining the average KER of dish-topped peaks is discussed. 0030-493X/92/111203-07 $08.50 zyxwvut 0 1992 by John Wiley & Sons, Ltd. Received 26 May 1992 Revised manuscript received 23 July 1992 Accepted 24 July 1992