Excitation of gas-phase fullerenes with laser pulses at 1064 nm yields large amounts of higher fullerene ions C,+, 72 5 n I 96, even though their neutral precursors are only minor contaminants in the vapor over the toluene extract of soot. The ions are particularly prominent in the size distribution of delayed ions recorded at low fluence. Comparison with data obtained a t 532 nm suggests that the size-dependent variations in the onset of photoabsorption are responsible for the observed phenomena. We propose that delayed ionization may be utilized to distinguish between higher fullerenes and other, more loosely bound, carbon aggregates. quartz microbalance to determine the flux of particles emerging from the oven. As a result, we find that the number density of particles at the point of photoexcitation is 2 X 1O1O cm-3, for a temperatureof 600 O C . Taken together with the fact that similar findings are obtained at reduced oven temperature, we conclude that ion-molecule reactions do not play a significant role in the present study.The molecular beam emerging from this effusive source is collimated and intersected by light from a mildly focused (beam diameter 1.1 mm) Q-switched Nd:YAG laser in the first gap of a Wiley-McLaren ion source.l* The laser operates at any one of its harmonics (i.e., at 1064, 532,355, or 266 nm), and the pulse duration is 5-8 ns, with a repetition rate of 50 Hz. The energy per laser pulse is controlled by varying the time delay between the flash-lamp pulse and the Q-switch trigger. The time-offlight mass spectrometer, whose axis is collinear with the molecular beam, may be operated in the conventional mode, with static extraction fields. In this case, the tail of a mass peak immediately reveals the presence of delayed i0ns.3.~ However, these "prompt" spectra are difficult to analyze if more than one delayed ion species is present.Alternatively, the potentials applied to the two plates which define the extraction region may be switched rapidly by up to 8 kV, such that only those ions that are formed within a welldefined time window contribute to the mass peaks in the "delayed" mass spectra. This technique is a modification of the doublepulse technique introduced by Leisner et al.2 A detailed description of our version has been published elsewhere.6 In short, all delayed ions of a given mass, formed during the sampling time [tbl, rex], will arrive simultaneously at the detector. Ions formed earlier are totally rejected. Ions formed after t,, (measured with respect to the laser pulse) will still arrive at the detector, but with a corresponding delay.The strength of the method stems from the fact that delayed ions of different masses are no longer superimposed in the timeof-flight spectra. We have succeeded in detecting all even-sized fragments of Cm and c 7 0 as small as c 3 6 in delayed spectra, and delayed formation of C a + after photoexcitation at 266 nm has been traced over 100 p s . 6~~
