ORGANIC MASS SPECTROMETRY, VOL. 26, 117-118 zyxwvutsr (1991) OMS Letters zyxw Dear Sir New Basis for a Method for the Estimation of Secondary Ion Internal Energy Distribution in 'Soft' Ionization Techniques The advent of soft ionization methods such as solid secondary ion mass spectrometry (SSIMS), fast atom bombardment (FAB), plasma desorption and laser desorption has permitted the gas-phase production of ions from organic thermally labile and/or non-volatile molecules. Such ion sources have greatly extended the applications of mass spectrometry. The crucial process involved in these techniques is the transformation of the primary energy into kinetic and internal energies of the target molecules. Kinetic energy is required for particle desorption. The amount of internal energy deposited governs the analyte decomposition processes (fragmentation, irradiation damage). Our laboratory has developed a method for estimating the analyte internal energy distribution. This method is based on the fragmentation of a family of target molecules that quanti- tatively reflects the internal energy contents of analyte ions. Benzylpyridinium salts constitute this target system. The reac- tion selected as a probe for the internal energy is the follow- ing: + Experimental and results Mass spectra were obtained on a VG Autospec-Q mass spec- trometer. The ion gun was operated with caesium ions at 8 keV. Samples were dissolved in glycerol (1-10 pg p1-l). In a first approach,'.* we considered that the fraction of ions decomposing is essentially fixed by the ionization energy of the corresponding benzyl radical. We then estimated the internal energy distribution of the target molecules' secondary ions by correlating their fragmentation extents, zyxwvutsr T, with the differences between the ionization energies of the correspond- ing substituted and unsubstituted benzyl radicals, 6(IE). The fragmentation extents are defined in the following manner: T = I[R-benzyl]' / (I[R-benzyl]+ + zyxwvut I[ R-benzylpyridinium] +) where I represents ionic abundance. The approximation of the correlation between fragmenta- tion extents and zyxwvutsrqpo 6(IE) is correct in the absence of strong elec- tronic effects of substituents (these effects can be evaluated from the substituent Hammett constants, zyxwvuts 0). However, we can infer from the fragmentation reaction itself (Eqn (1)) that sub- stituents with strong electronic effects could influence differ- ently the stabilities of benzyl radicals and molecular ions. Rigorous data handling requires a correlation between frag- mentation extents and dissociation reaction energy barriers, E, , defined by E, = AHf[RC6H4CH2] + + AH,.(C,H,N) - AH,[RC,jH4CH,NC,H,IC (2) In this expression, we have neglected the entropy term with respect to the enthalpy term, considering their relative values. The AHf values were taken from Ref. 3. This evaluation of the internal energy distribution is based on the following assumptions : (i) In our experiments, the analyte exists in solution as a preionized species. (ii) The primary energy dissipation results in an energy release along the primary particle track. The global process for ion ejection takes 10- zyxw 'o-lO- l1 s.~ The system is then likely to approach a thermal equilibrium state around the primary particle track.5 The same amount of energy is thus imparted to target molecules. (iii) Primary particle impacts on the sample droplet result in gas-phase clusters of solute and matrix molecules. Colli- sions between clusters and their desolvation in the gas phase release their excess energy and thus decrease the residual internal energy of the analyte molecules.6 We assume that molecular ions fragmentation occurs prin- cipally from molecular ions that are completely desolvat- ed. The energy barrier values for some benzylpyridinium frag- mentation reactions, E,, calculated with Eqn (2), are given in Table 1. The variations of this parameter with substitution, a@,), have been calculated and compared with 6(IE). It can be seen that the 6(IE) values, the energetic parameter adopted in a first approach, agree reasonably well with the 6(E,) values for molecules for which the substituents do not exhibit too strong an electronic effect. For these molecules we can estimate the energy barriers when these values are unknown, from the 6(IE) values. By doing so we have a good approximation for extending the number of studied target molecules. The fragmentation extents of these molecules, T, and the corresponding energy barriers are given in Table 2. Values of E, in parentheses are estimated values. Table 1. Energy barriers (Eb), varia- tions with substitution (6(Eb)) and comparison with 6(ZE) R E, (eV) 6W,) 6UE) p-OCH, P-CH, O-CH, m-CH, m-OCH, H p- Br m-F p-61 p-CN P-NO, m-CN zyxwv 1.34 1.60 - - - 1.73 1.78 1.78 - - 2.1 2 - -0.43 -0.1 7 - - -0.04 0 0 - - 0.35 - -0.94 -0.19 -0.1 4 -0.1 3 -0.1 0.03 0 0.04 0.42 0.58 0.64 0.82 Table 2. Fragmentation extents (T) and energy barriers ( Eb) R r (FAB) E, (eV) P-CH, 0.41 1.60 0-CH, 0.33 (1.64) m-CH, 0.33 (1.65) m-OCH, 0.34 (1.68) p-CI 0.37 1.73 H 0.35 1.77 p-Br 0.39 1.77 m-F 0.24 (2.20) P-NO, 0.24 2.1 2 p-OCH3 0.53 1.34 0030-493X/91/02011742 $05.00 zyxwvut 0 1991 by John Wiley & Sons, Ltd. Received 9 July 1990 Revised manuscript received 25 September 1990 Accepted 30 September 1990