IL NUOVO CIMENTO

V OL . 110 A, N. 9-10

Settembre-Ottobre 1997

Ionisation and fragmentation of fullerene ions by electron impact( ) ¨ FER , D. H ATHIRAMANI , K. A ICHELE , U. H ARTENFELLER V. S CH A F. S CHEUERMANN , M. S TEIDL , M. W ESTERMANN and E. S ALZBORN Institut f¨ ur Kernphysik, Justus-Liebig-Universit¨ at Giessen - D-35392 Giessen, Germany (ricevuto il 31 Luglio 1997; approvato il 15 Ottobre 1997)

Summary. — Absolute cross-sections for electron impact induced fragmentation, fragmentation-single-ionisation and fragmentation-double-ionisation were measured + 2+ for the fullerene ions C+ 60 and C58 . For the fullerene ion C60 , fragmentation-singleionisation was measured with a loss of n C2 molecules (n = 1; 2; 3). The cross-sections were determined from the respective threshold up to an electron energy of 1 keV using the animated crossed-beams technique. PACS 34.80.Gs – Molecular excitation and ionisation by electron impact. PACS 36.40.Qv – Stability and fragmentation of clusters. PACS 01.30.Cc – Conference proceedings.

1. – Introduction Many collision experiments with fullerene ions have been carried out in the past few years due to the possibility of the production of intense mass-to-charge-selected fullerene ion beams. Interesting and unexpected results were discovered in these studies. One of the most remarkable features of a singly ionised C+ 60 cluster is its exceptional stability against unimolecular decomposition. Nevertheless, the main fragmentation channel of the Buckminster fullerene ion is the sequential loss of neutral C2 molecules [1]. Collision induced fragmentation of C+ 60 by photons [2, 3], atoms [4], fullerenes [5], surfaces [6] and electrons [7-10] has been investigated. In this context we have measured absolute cross-sections for electron impact induced fragmentation, fragmentation-ionisation + 2+ and fragmentation-double-ionisation for the fullerene ions C+ 60 , C58 and C60 . The crosssections were determined from below the respective threshold up to an electron energy of 1 keV.  Paper presented at the 174. WE-Heraeus-Seminar “New Ideas on Clustering in Nuclear and Atomic Physics”, Rauischholzhausen (Germany), 9-13 June 1997. ( )

G

Societ` a Italiana di Fisica

1223

1224

¨ V. SCHAFER, D. HATHIRAMANI ETC.

+ Fig. 1. – Absolute cross-sections for the electron-impact induced fragmentation of C+ 60 ! C56 [7] + (full triangles). The error bars show (open circles) and the fragmentation of C+ C the total ! 58 54 experimental error. The arrows indicate the thresholds of the processes given by W¨org¨otter et al. [15]. 2. – Experiment The measurements were performed at the Giessen heavy ion beam facility with the electron-ion crossed-beams set-up which has been described in detail earlier [11]. With an especially developed oven a sample of C60 soot (with a purity greater than 96 %) was heated up to a temperature of about 800 K. The C60 vapour was evaporated into the plasma of a 10 GHz ECR ion source [12]. The ion source was operated at a low microwave power (less than 5 W). The so produced Cq60+ fullerene ions were extracted at energies of 6 keV for q = 1 and 20 keV for q = 2. After mass-to-charge analysis and a collimation to 2.5  2.5 mm2 , the fullerene ion beam was crossed with an intense electron beam [13] at an angle of 90 . The product ions of the observed reactions were magnetically separated from the primary fullerene ion beam and counted in a single-particle detector 1 m behind the interaction region. The flight time of the carbon cluster ions from the ECR ion source to the interaction region was approximately 122 s for q = 1 and 67 s for q = 2. The produced fragment ions were detected in a channeltron electron multiplier roughly 24 s (primary charge state q = 1) and 13 s (primary charge state q = 2) after interaction with the electron beam. The cross-sections were measured using the animated crossed-beams technique [14] where the electron beam is moved up and down through the ion beam with simultaneous registration of both actual beam currents and the signal of the observed fragment ion. The total experimental uncertainties are typically 10 % at the maximum of the cross-sections resulting from the quadrature sum of the non-statistical errors of about 8 % and the statistical error at 90 % confidence level. 3. – Results

+ + + The comparison of the cross sections C+ 60 ! C56 [7] and C58 ! C54 (fig. 1) clearly shows + the higher stability of the C60 ion against unimolecular decay. Although the thresholds

IONISATION AND FRAGMENTATION OF FULLERENE IONS BY ELECTRON IMPACT

1225

Fig. 2. – Absolute cross-sections for the electron-impact induced fragmentation-single-ionisation of 2+ + 2+ C+ 60 ! C56 [7] (open circles) and C58 ! C54 (full triangles) and the fragmentation-double-ionisation + 3+ + 3+ of C60 ! C56 (full circles) and C58 ! C54 (full diamonds). The error bars show the total experimental error. The arrows indicate the thresholds of the processes given by Scheier et al. [16].

Fig. 3. – Absolute cross-sections for the electron-impact induced fragmentation-single-ionisation of 3+ 2+ 3+ 2+ 3+ C2+ 60 ! C58 (circles), C60 ! C56 (squares) and C60 ! C54 (triangles). The error bars show the total experimental error. The arrows indicate the thresholds of the processes given by Scheier et al. [16].

1226

¨ V. SCHAFER, D. HATHIRAMANI ETC.

and the shape of the cross-sections at electron energies above 35 eV are nearly identical + for both processes, the cross-section for C+ 58 ! C54 exceeds the results for the reaction + + C60 ! C56 by almost a factor of 2 in the maximum. Figure 2 shows cross-sections for processes with a loss of two C2 molecules from + C+ 60 and C58 ions as in fig. 1. But here the reactions additionally include single- (upper cross-sections) and double-ionisation (lower cross-sections). Whereas the cross sections for the pure fragmentation (fig. 1) show a significant difference in their maximum value, the cross-sections for fragmentation-single-ionisation and fragmentation-double+ ionisation for C+ 60 and C58 ions are nearly identical (fig. 2). Different fragmentation degrees for the fragmentation-single-ionisation of primary C2+ 60 ions are shown in fig. 3. The observed cross-sections decrease by about 15 % over the whole energy range with each C2 molecule evaporated. Note that the cross-sections do not converge at higher electron energies within the range observed. + 2+ Only cross-sections for the parent fullerene ions C+ 60 , C58 and C60 could be measured, but it can be concluded that the influence of the mass of the fullerene ions seems to be less important than expected. Only the cross-sections for pure fragmentation processes show a significant dependence on the ion masses. Furthermore, the reaction channel determines the shape of the relevant cross-section. The measurement of cross-sections for electron impact induced pure evaporation of only one C2 molecule was not possible with the present experimental set-up because of the unfavourably similar m/q ratios of the parent and product ions. REFERENCES ¨ RG O ¨ TTER R., M UIGG D., M ATT S., E CHT O., F OLTIN M. ¨ NSER B., W O [1] S CHEIER P., D U ¨ RK T. D., Phys. Rev. Lett., 77 (1996) 2654. and M A [2] W URZ P., and LYKKE K. R., J. Phys. Chem., 96 (1992) 10129. [3] H OHMANN H., E HLICH R., F URRER S., K ITTELMANN O., R INGLING J. and C AMPBELL E. E. B., Z. Phys. D, 33 (1995) 143. [4] E HLICH R., W ESTERBURG M. and C AMPBELL E. E. B., J. Chem. Phys., 104 (1996) 1900. [5] R OHMUND F., G LOTOV A. V., H ANSEN K. and C AMPBELL E. E. B., J. Phys. B, 29 (1996) 5143. [6] B ECK R. D., R OCKENBERGER J., W EIS P. and K APPES M. M., J. Chem. Phys., 104 (1996) 3638. ¨ LPEL R., H OFMANN G., S TEIDL M., S TENKE M., S CHLAPP M.,T RASSL R. and [7] V O S ALZBORN E., Phys. Rev. Lett., 71 (1993) 3439. [8] M C E LVANY S. W., R OSS M. M. and C ALLAHAN J. H., Acc. Chem. Res., 25 (1992) 182. ¨ RK T. D., J. Phys. Chem., 99 (1995) 15428. ¨ NSER B. and M A [9] S CHEIER P., D U ¨ RK T. D., Phys. Rev. Lett., 74 ¨ NSER B., L EZIUS M., S CHEIER P., D EUTSCH H. and M A [10] D U (1995) 3364. ¨ LLER A., H OFMANN G., H UBER K., B ECKER R., G REGORY D. C. [11] T INSCHERT K., M U and S ALZBORN E., J. Phys. B, 22 (1989) 531. ¨ LPEL R. and [12] L IEHR M., S CHLAPP M. ,T RASSL R., H OFMANN G., S TENKE M., V O S ALZBORN E., Nucl. Instrum. Methods B, 79 (1993) 697. ¨ LLER A., A CHENBACH C., T INSCHERT K. and S ALZBORN E., Nucl. [13] B ECKER R., M U Instrum. Methods B, 9 (1985) 385. ¨ LLER A., T INSCHERT K., A CHENBACH C. and S ALZBORN E., Nucl. Instrum. Methods [14] M U B, 10/11 (1985) 204. ¨ RG O ¨ RK T. D., F OLTIN M., K LOTS C. E., ¨ TTER R., D U ¨ NSER B., S CHEIER P., M A [15] W O L ASKIN J. and L IFSHITZ C., J. Chem. Phys., 104 (1995) 1225. ¨ RG O ¨ RK T. D., Int. J. ¨ TTER R., L EZIUS M., R OBL R. and M A ¨ NSER B., W O [16] S CHEIER P., D U Mass Spec. Ion. Phys., 138 (1994) 77.