On the kinetic energy release distribution for C 2 evaporation from fullerene ions K. Gluch a,1 , S. Matt-Leubner a , O. Echt b , R. Deng b,2 , J.U. Andersen c , P. Scheier a , T.D. M€ ark a, * a Institut f € ur Ionenphysik, Leopold Franzens Universit € at, Technikerstr. 25, Innsbruck A-6020, Austria b Department of Physics, University of New Hampshire, Durham, NH 03824, USA c Department of Physics and Astronomy, University of Aarhus, Aarhus C DK-8000, Denmark Received 19 August 2003; in final form 23 December 2003 Published online: Abstract We present a critical evaluation of the kinetic energy released in the unimolecular dissociation of C þ 60 ,C þ 58 , and C þ 56 . High-quality kinetic energy release distributions (KERD) have been recorded in a three-sector-field mass spectrometer that is equipped with two electric sectors. Systematic errors arising from assumptions about the shape of the KERD are explored by a combination of ex- periments and simulations; an alternative to the Ômodel-free approachÕ is suggested. For decay of C þ 60 , an average KER of 394  14 meV is obtained. Surprisingly, the experimental KER distributions are virtually indistinguishable from the one expected for a pure Langevin-type interaction. Ó 2004 Published by Elsevier B.V. 1. Introduction The distribution of kinetic energies that are released in spontaneous or induced unimolecular dissociation of molecules or molecular ions reveal information about numerous aspects of the reaction [1–3]. In cluster sci- ence, experimental kinetic energy release distributions (KERD) provide insight into the structure of the cluster, the nature of the reaction, and the presence of compet- ing channels [4–6]. Moreover, it is possible to determine binding energies of cluster ions from the KERD or simply from its average,  e [7–11]. Essentially,  e is a measure of the transition state temperature T y of the reaction that is observed at time t. For an evaporative ensemble, the decay rate k equals 1/t unless other channels compete with the reaction that is monitored [12,13]. From a knowledge of T y and the corresponding rate k, one can determine the activation energy E a if the pre-exponential decay factor, or some other closely re- lated quantity like the looseness of the transition state or the Gspann factor, is known [3,14]. However, the relation between  e and T y is not unique. In his Ômodel-freeÕ approach, Klots [12] suggested that the KERD for decay of fullerene ions, C þ n ! C þ n2 þ C 2 ð1Þ or for decay of other atomic clusters, may be expressed as f ðeÞ/ e ‘ e e k B T y ; ð2Þ ‘ is bound by 0 6 ‘ 6 1, hence  e ¼ð‘ þ 1Þk B T y and k B T y 6  e 6 2k B T y . A physically more transparent expression for the KERD follows from the concept of microscopic re- versibility [15–17]: f ðeÞ/ erðeÞe e k B T y ; ð3Þ * Corresponding author. Also at: Department of Plasma Physics, Comenius University, SK-84248 Bratislava, Slovak Republic. Fax: +43-512-507-2932. E-mail address: tilmann.maerk@uibk.ac.at (T.D. M€ ark). 1 Permanent address: Institute of Mathematics, Physics and Infor- matics, Maria Curie-Sklodowska University, Lublin 20-031, Poland. 2 Present address: Department of Chemistry, University of Illinois at Chicago, Chicago, IL 60607-7061, USA. 0009-2614/$ - see front matter Ó 2004 Published by Elsevier B.V. doi:10.1016/j.cplett.2003.12.113 Chemical Physics Letters 385 (2004) 449–455 www.elsevier.com/locate/cplett