XUV POSTIONIZATION OF FRAGMENTS OF PHOTODISSOCIATED (260"

Keith R. Lykke Materials Science and Chemistry Divisions

Argonne National Laboratory Argonne, IL 60439

Submitted for the Proceedings of the

Symposium on "The Fullerenes: Chemistry, Physics, and New Directions" 187th Meeting of the Electrochemical Society

Reno, Nevada

May 2 1-26, 1995

DISCLAIMER 9 1995 This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsi- bility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The Views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

_____I_

The mbmtred manuscript has been authored by a contractor of the U S. Government under coonact N o W-31-109-ENG-38 Accordingly. the U. S. Government retains a nonexddve roydty-i+ee license to publish or reproduce the: published form of this contribution, or allow others to do so. for U. S. Government p111110s~.

*Work supported by the U.S. Department of Energy, BES-Materials Sciences, under Contract W-3 1-109-ENG-38.

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

XUV POSTIONIZATION OF FRAGMENTS OF PHOTODISSOCIATED C60

Keith R. Lykke Materials Science and Chemistry Divisions

Argonne National Laboratory Argonne, IL 60439

ABSTRACT

We report experimental evidence for the loss of neutral C, C2, C3, and C4 from photoexcited c60 and C70. These species are detected by post- ionization with XUV radiation formed by four-wave mixing in Kr vapor. The fragments have been detected in the nascent distribution (i.e., before total disintegration of the fullerene-size molecular entities). A resonant autoionization transition in C atom has yielded an upper limit to the translational energy release for dissociation into C atoms. The detection of these products gives new insight into the fragmentation mechanism of fullerenes.

INTRODUCTION

There has been much experimental (1-12) and theoretical work (13-18) on the dissociation of c60 since its discovery in 1985 (19). Most of the studies have focused on the detection of only even-numbered carbon clusters (Cn, n-even) appearing in the mass spectra of fragmented c60 (1,14,16). The occurrence of these even numbers has been ascribed to predominantly C2 loss by c6() until recently (9,12,20). One alternative possibility is that c60 may fragment via loss of C atoms to a metastable C59, followed by loss of another C atom to yield C58. We have used (2+1) multiphoton ionization (MPI) to detect C atoms from c60 photodissociation previously, but were unable to determine if the C atoms resulted from the initial prompt dissociation of C6o or from smaller fragments after total disintegration of the cage (20). Unfortunately, the MPI detection technique requires intense photon fluences that resulted in dissociation of C6o down to small fragments.

We have also detected large neutral fragments (c58-c32) from UV photo- dissociation of C6o by using 118 nm (10.5 eV, vacuum ultraviolet, VUV) radiation for single-photon ionization of the photoproducts (2 1). In these measurements, the neutral fragment distribution matched the ionized fragment distribution. Unfortunately, the high ionization potentials for the small Cn species (22-24) precluded their detection with this vacuum ultraviolet light (for fragments smaller than about C6, see Fig. 1). We have recently installed an extreme-ultraviolet (XUV) light source on our mass spectrometer to photoionize all species with ionization potentials <14 eV, including all C fragments. However, the number of photons/pulse is much lower with the XUV source than with the 10.5 eV source, so we have not been able to remeasure the large (fullerene) fragment distribution in the present experiment. This is caused by the overwhelming ionization induced by the dissociation laser.

A possible complication in the detection of the nascent products from photodissociated C6o concerns the time constant for dissociation. c60 has been shown to absorb many photons before dissociating or ionizing (7). From other experiments, we have evidence that c 6 0 may accumulate up to 5 5 eV of internal energy before fragmenting. The rate of dissociation for c 6 0 at 55 eV internal energy has been calculated by RRKM theory (7) to be -IO8 sec-l (10 nsec “lifetime”), while it is -5 x lo5 sec-1 (2 psec lifetime) at 40 eV internal energy. However, we have used absolute rate theory (25) to calculate that (260 with 55 eV internal energy will dissociate down to at least C32 on our experimental time scale. It will only dissociate to -C54 with 40 eV internal energy in the initial C60 species. We have also calculated that -0.5 eV is liberated in the initial dissociation (in good agreement with the observed upper limit to the kinetic energy release mentioned above), and because of kinematic constraints (conservation of momentum and energy), almost all of this energy goes into the smallest fragment in the dissociating step. Assuming the fragment is C, the velocity of the C atom will be -2 x lo5 c d s e c . This species will travel -2 mm in 1 psec, almost out of the postionization (detection) region. Therefore, there is a small intermediate regime where the internal energy is high enough to dissociate the C6o in a reasonable time (before the fragment leaves the detection region), but not too high to dissociate the C6o species into its very small fragments (n<30) and thus preclude a measurement of the nascent distribution of fullerene fragments.

In this paper we report the observation of C, C2, C3, and C4 in the nascent distribution of dissociated fullerenes. It is likely that C2 is the lowest energy fragmentation channel, but that C, C3, and C4 are fragments from the initial dissociation, before the n <32 channel opens. Other results have been presented elsewhere (26).

EXPERIMENTAL

The time-of-flight mass spectrometer (TOFMS) has been described in detail previously (20). The c60 or C70 vapor is produced by laser desorption (7), usually using the 337 nm output from a N2 laser. This is followed (some microseconds later, 71 in Fig. 2) by 266 nm (4th harmonic of the same Nd:YAG laser that produces the XUV light) for dissociation. The products of the dissociation are then probed with one-photon ionization from an XUV beam delayed relative to the dissociation laser (22 in Fig. 2) . The XUV generating system consists of two Nd:YAG-pumped dye-laser systems, a frequency mixing cell, and a capillary waveguide for transport of the XUV beam (27) (see Fig. 2). One of the dye lasers is tuned to 557 nm, mixed with 354.7 nm from the frequency-tripled Nd:YAG to generate 216.67 nm. This beam is then combined on a dichroic mirror with the output of the second dye laser and focused with a 20 cm fused silica lens into the four-wave-mixing medium. This medium is either Kr in a pulsed-free jet or an effusive (windowless) cell (28). The windowless cell was a 1/2-inch stainless steel tube with small holes laser drilled through the diameter using the focused, high- power 532 nm frequency-doubled output from the Nd:YAG laser. The XUV light is uenerated inside the high pressure region in the windowless cell and is then transported %to the source region of the TOFMS with an -1 mm diameter capillary waveguide that was butted up to the exit hole of the windowless cell. The waveguide is utilized both for transmission of the XUV light (transmission for XUV>98%) and as an efficient differential pumping aperture. With the proper choice of capillary size and pump to exhaust the free-jet, we see very little pressure rise in the mass spectrometer. The background pressure in the TOF is -5 x 10-10 TOK, rises to -10-8 Torr with 10 TOK of Kr

inside the cell and 3 x 10-4 Torr around the windowless cell (pumped by a small turbomolecular pump). In addition, the pressure rise is due only to Kr, which has no effect on our dissociation measurement. The delay between photodissociation and XUV postionization for the experiments reported here was -20-50 nsec.

To demonstrate the relative tunability and spectral width of the XUV light, Fig. 3 shows a scan through a resonance in Kr (absorption of 1-XUV photon followed by ionization with 1-266 nm UV photon, 1+1 ionization) with the TOF backfilled to -10-6 Torr with acetone and Kr. The linewidth is mainly determined by the 354.7 nm limht that is utilized to generate the 216.7 nm radiation. The measured linewidth is -1.3 cm- P .

RESULTS AND DISCUSSION

Shown in Fig. 4 is a mass spectrum of C6o photodissociation by 266 nm radiation, followed by single-photon-resonant excitation of the 3S 1<--3P 1 autoionization C-atom resonance at -105780 cm-1 (94.54 nm, 13.11 eV) (29). This spectrum is typical for photoionization of 0 , displaying the even-numbered fullerenes (C34-c60) and the long delayed ionization "tail" (7,30) to the high d z side of C6o [for a TOFMS, the m/z scale is proportional to (time)2]. The peak labeled prompt C60+ is caused by the desorption laser (337 nm N2 laser) only. The bottom panel in the figure shows the region around C, C2, C3. Clearly evident is the detection of C atoms before the onset of extreme dissociation (Le., no fragments smaller than -C36 are observed in the mass spectrum). This implies that C atoms do result from the prompt dissociation of C6o. However, there may be other small fragments, as observed in graphite itself (C:C2:C3 = 1.0:1.9:3.1 at 4000K) (31). In fact, C2 and C3 are indeed present in the spectrum, although at somewhat reduced intensities compared with the C-atom (resonant) peak.

Shown in Fig. 5 is a wavelength scan through the C-atom autoionization resonance, with a recording of the signal from direct ionization of acetone backfilled into the chamber at -5 x 10-8 Torr (-5 x 10-10 Torr base pressure). This ionization signal is a direct measure of the XUV photon fluence; the dispersion-shaped curve is caused by the transition from positive to negative dispersion in the four-wave-mixing gas (28). This same signal dependence has been observed for other gasses (02 in particular). Thus, the curve labelled as the acetone ionization signal is due to the photon fluence alone and not on the particular ionization properties of acetone. The C-atom resonance appears to be -2.5 cm-1 wide, given by the Doppler width of the transition and partially by the XUV linewidth of -1.3 cm-1. As discussed above, the XUV linewidth was measured by 1+1 ionization of Kr through the 4p-7s 2P3/2(J=l) transition that also fixed the measured line position of the C-atom resonance. This Doppler width places an upper limit to the velocity of C atoms from dissociation of c60 at -(1 cm-1/105800 cm-l) x 3 X l O l 0 c d s e c = 2.8 x lo5 c d s e c corresponding to -0.5 eV energy release in the dissociation (or an effective C-atom temperature of -4000K). We can only place an upper limit to this because of the unknown inherent autoionization width of the resonance.

To make a quantitative comparison of the small fragments, we have employed a nonresonant postionization source. Shown in Fig. 6a are mass spectra of photo- dissociation of c60 with 266 nm dissociation, with the postionization laser delayed by -30 nsec and tuned away from the C-atom resonance (90.67 nm, 13.43 eV). In addition, we used a higher efficiency effusive source and a different tunable laser for sum- frequency generation for production of nonresonant XUV radiation (28) (- 10-fold increase in the number of photons/pulse). The relative cross section for ionization of

C:Cz:C3:C4 is unknown, but a study of XUV ionization of laser-desorbed graphite showed that the cross sections are within a factor of three of each other [the absolute cross section for C-atom ionization has been calculated (32) to be -10-17 cmz]. One could assign relative cross sections from the laser-desorbed graphite data, but the exact temperature of the laser-heated graphite is unknown and the different desorption velocities need to be taken into account. However, for the sake of this paper, we shall assume that the relative detection efficiencies are comparable for all four entities (C, C2, C3, and C4). Expanded scales of each of these species (C-C3) are displayed in Fig. 6b. The ionized fragments from the dissociation laser are labeled as positive ions, while the neutral fragments that are detected by XUV are labeled as neutrals and are shaded. The dealay of the XUV postionization laser relative to the UV photodissociation laser is evident in the figure.

We have plotted the relative signals from the four fragments that were detected as neutrals as a function of the relative photodissociation fluence levels in Fig. 7. A quick examination of this figure reveals that C2 is probably the lowest energy dissociation channel, but that the other small fragments are observed as well. In fact C3 seems to be -1/3 of the intensity of the C2 signal for most of the investigated fluence range. If C2 were the only fragment down to C32, the C2 peak should be -20 times larger than any other fragment. This is obviously not the case here, and we must try to understand the origin of these other fragments.

A similar pattern is observed in the photodissociation of C70. The top panel in Fig. 8 shows the time-of-flight spectrum for C70 for different dissociation fluence levels. The peak labeled as prompt C70+ is from the desorption laser only and contributes no neutral fragments. This was tested by blocking the dissociation laser and observing no signal in the neutral channels. The bottom panel in Fig. 8 shows an expansion of the C and C2 region. At low dissociation laser fluences, C2 rises above the background before C. Once again, this does imply that C2 is probably the lowest energy channel. However, C, C3, and C4 appear before the onset of severe dissociation (Le., before large fragments with masses lower than the fullerene stage appear).

Geusic et al. (33) and Radi et al. (34,35) have shown that C3 loss is the predominant dissociation pathway for a majority of the small (1x30) carbon clusters, although C and C2 do originate from a small fraction of these clusters. However, C54-C60 were given as the only possible candidates for losing C4. Thus, our detection of C4 in the C6o photodissociation channel proves that C4 is indeed present in the initial distribution. The lack of observation of C5 implies that C15-C28 may not be present in the initial fragmentation from c60. That is, since C5 is a major channel for these small carbon clusters, and since we can photoionize it (IP-12.2 eV), we should observe it as a major species in the postionization spectra.

A recent paper addresses the shape and entropy of c 6 0 as a function of temperature (1 8). The description of the various stages of fullerene evolution as the cage is heated are the solid (1000K), floppy (3000K), pretzel (4200K), linked chains (5000K), fragments (5400K), and chain gas (10200K) stages. A possible source for the C2 and C4 molecules in the frag.ment distribution is the floppy or pretzel stage, whereas the C and C3 species may originate from the linked chains and fragment stages. However, these different stages have no clear-cut divisions, and there must be interconversions between each.

A very recent study using molecular dynamics calculations of C(j0 with a Tersoff potential (36,37) has shown a solid-to-liquid transition at -4700K. In addition, these authors have shown that C, C2, and C3 may be small fragments from the dissociation, depending on the heating rate. C2 is the lowest energy channel (in agreement with the present results) with C and C3 appearing in the distribution when the heating rate is increased.

There are a few alternative ways in which C-C4 can be detected. One of these involves the total disintegration of the fullerene to smaller (Cn, with 1x32) fragments and then further fragmentation of these species into C, C3, and C4. We have tried to limit the dissociation fluence so that only fullerene-size species remain after the dissociation has taken place. In an earlier experiment (21), at the fluences investigated in the present study, we did not observe any fragments smaller than fullerenes (nc32).

Another complication may be the dissociation of the photodissociation fragments induced by the XUV postionization source itself. This will tend to show smaller fragments than were initially in the product distribution. We do not think this is occurring because the -13 eV photon is just above the ionization potentials for the fragments (see Fig. 1). Thus, the photoion should not be left with enough energy to fragment further.

CONCLUSIONS

As opposed to some earlier work (38), it has become clear from recent studies that C6o can absorb many photons and become internally excited, at least to 50-60 eV. In fact, some very recent work shows fragmentation from c60 that has been excited to 150- 200 eV (39,40). In the present experiment, we have tried to keep the internal energy to a minimum to observe fragments in the nascent distribution (i.e., before severe frag- mentation occurs), but still high enough to observe fragmentation on an experimental time scale.

In conclusion, it seems that c60 does indeed fragment by loss of C2 as the lowest energy channel (during the so-called floppy or pretzel stage). However, C, C3, and C4 are also ejected during the fullerene-size stage (i.e., before the ne32 occurs). This "linked-chains" stage may allow the ejection of fragments other than C2 and the subsequent annealing of the chain back into a fullerene before further destruction. These findings may have some impact on the discrepancy (15,41) between the theoretical numbers for C6o dissociation energetics (-10-12 eV for C2 loss) (15,42) and the experimental numbers (-5-7 eV) (2,3,7,8,41). Basically, the theoretical results are for loss of C2 from a closed, fullerene-like structure, whereas experiments probe loss of C-containing species (C-C4) from the different, more-open structures discussed above.

ACKNOWLEDGMENTS

This work was supported by the U. S. Department of Energy, BES-iMaterials Sciences, under Contract W-3 1-109-ENG-38.

REFERENCES

1. S. C. O'Brien, J. R. Heath, R. F. Curl, and R. E. Smalley, J. Chem. Phys. 88, 220 (1988).

2. P. R. Radi, M.-T. Hsu, M. E. Rincon, P. R. Kemper, and iM. T. Bowers, Chem. Phys. Lett. 174,223 (1990).

3. P. Sandler, C. Lifshitz, and C. E. Klots, Chem. Phys. Lett. 200,445 (1992). 4. R. Voelpel, G. Hofmann, M. Steidel, M. Stenke, M. Schlapp, R. Trassl, and

E. Salzborn, Phys. Rev. Lett. 71,3439 (1993). 5. B. Walch, C. L. Cocke, R. Voelpel, and E. Salzborn, Phys. Rev. Lett. 72, 1439

( 1994). 6. 2. Wan, J. F. Christian, and S. L. Anderson, Phys. Rev. Lett. 69, 1352 (1992). 7. P. Wurz and K. R. Lykke, J. Phys. Chem. 96, 10129 (1992). 8. M. Foltin, M. Lezius, P. Scheier, and T. D. Mark, J. Chem. Phys. 98, 9624 (1993). 9. P. Hvelplund, L. H. Andersen, H. K. Haugen, J. Lindhard, D. C. Lorents,

R. Malhotra, and R. Ruoff, Phys. Rev. Lett. 69, 1915 (1992). 10. T. Lill, H.-G. Busmann, B. Reif, and 1. V. Hertel, Appl. Phys. A 55,461 (1992). 11. D. M. Gruen, S . Liu, A. R. Krauss, and X . Pan, J. Appl. Phys. 75, 1758 (1994). 12. D. Ding, R. N. Compton, R. E. Haufler, and C. E. Klots, J. Phys. Chem. 97, 2500

(1993). 13. C. Xu and G. E. Scuseria, Phys. Rev. Lett. 72,669 (1994). 14. C. 2 . Wang, C. H. Xu, C. T. Chan, and K. M. Ho, J. Phys. Chem. 96,3563 (1992). 15. W. C. Eckhoff and G. E. Scuseria, Chem. Phys. Lett., submitted (1993). 16. R. L. DeMuro, D. A. Jelski, and T. F. George, J. Phys. Chem. 96, 10603 (1992). 17. E. Kim, Y. H. Lee, and J. Y. Lee, Phys. Rev. B 48, 18230 (1993). 18. S. G. Kim and D. Tomanek, Phys. Rev. Lett. 72,2418 (1994). 19. H. W. Kroto, J. R. Heath, S. C. O'Brien, R. F. Curl, and R. E. Smalley, Nature 318,

162 (1985). 20. K. R. Lykke, P. Wurz, D. H. Parker, and M. J. Pellin, Appl. Opt. 32, 857 (1993). 21. K. R. Lykke and P. Wurz, J. Phys. Chem. 95,3191 (1992). 22. R. Ramanathan, J. A. Zimmerman, and J. R. Eyler, J. Chem. Phys. 98,7838 (1993). 23. S . B. H. Bach and J. R. Eyler, J. Chem. Phys. 92,358 (1990). 24. J. A. Zimmerman, J. R. Eyler, S. B. H. Bach, and S. W. McElvany, J. Chem. Phys.

94, 3556 (1991). 25. P. Wurz and K. R. Lykke, Chem. Phys. 184,335 (1994). 26. K. R. Lykke, Phys. Rev. A (accepted). 27. R. G. Tonkyn and M. G. White, Rev. Sci. Instrum. 60,1245 (1989). 28. G. Hilber, A. Lago, and R. Wallenstein, J. Opt. SOC. Am. B 4, 1753 (1987). 29. L. Johansson, Arkiv fur Fysik 31,201 (1965/1966). 30. E. E. B. Campbell, G. Ulmer, and I. V. Hertel, Phys. Rev. Lett. 67, 1986 (1991). 31. H. B. Palmer and M. Shelef, Vaporization of Carbon, in Chemistry and Physics of

Carbon, Marcel Dekker, New York, 1968,85. 32. S . L. Carter and H. P. Kelly, Phys. Rev. A 13, 1388 (1976). 33. M. E. Geusic, M. F. Jarrold, T. J. McIlrath, R. R. Freeman, and W. L. Brown,

J. Chem. Phys. 86,3862 (1987). 34. P. P. Radi, T.L. Bunn, P. R. Kemper, M. E. Molchan, and M. T. Bowers, J. Chem.

Phys. 88,2809 (1988). 35. P. P. Radi, M. T. Hsu, J. Brodbelt-lustig, M. Rincon, and M. T. Bowers, J. Chem.

Phys. 92,4817 (1990). 36. S. Serra, S. Sanguinetti, and L. Colombo, Chem. Phys. Lett. 225, 191 (1994). 37. S . Serra, S . Sanguinetti, and L. Colombo, J. Chem. Phys. 102,2151 (1995).

38. A. C. Jones, M. J. Dale, M. R. Banks, I. Gosney, and P. R. R. Langridge-Smith, Mol.

39. H. Hohmann, C. Callegari, S. Furrer, D. Grosenick, E. E. B. Campbell, and

40. H. Hohmann, R. Ehlich, S. Furrer, 0. Kittelmann, J. Ringling, and E.E.B. Campbell,

4 1. C. Lifshitz, Mass Spectrometry Reviews 12,26 1 (1993). 42. R. E. Stanton, J. Phys. Chem. 96, 11 1 (1992).

Phys. 80,583 (1993).

I. V. Hertel, Phys. Rev. Lett. 73, 1919 (1994).

2. Phys. D 33, 143 (1995).

Figure 1 -- Ionization potentials for carbon molecules from C to C1m.

Tunable XUV Setup

2 1 6 6 7 ~ ~

t

cw vapor

Figure 2 -- Tunable XUV setup to detect the products from photodissociation of fullerenes. The left part of the figure displays the four-wave-mixing process involved in generation of the tunable X W light. The lower right part of the figure is a blow-up of the interaction region, where C6o vapor is generated by laser desorption, dissociated, then probed by the X W light. A timing diagram describing the experiment is displayed in the upper right hand comer.

5000

4000

~ 3000 (I) c Q, C

.-

- 2000

1000

0 105764 105768 105772 105776 105780

frequency (crn-')

Figure 3 -- Scan of the region of interest near the autoionization resonance for C. Acetone is detected by single-photon ionization at these wavelengths and the part of the spectrum where the tunable dye laser is blocked shows that all of the signal is from the XUV radiation. Also shown is the (1+1) resonant ionization of Kr in the TOF.

1200

1000

800

2.

E e 5 600

400 c

200

0

-200

600

500

400 2. 0) - 5 300

200

G -

100

0

1 200 400 600 800 1 I . . . , , , . I . . . , , , .

m h IO

I . . . . I . . . . I . . . . I . . . . 1 . . . , I . . . . I . . . . 5 10 15 20 25 30 35

mlz

Figure 4 --Photodissociation spectrum of CfjO followed by postionization of small C fragments in the nascent distribution. The C-atom resonant 94.5 nm transition yields an -100-fold increase in C-atom detection sensitivity.

% C-atom resonant autoionization Acetone

ionization Acetone ionization r""r n.

3.5, <--3P2

\ I

1s22s2p3 3s

FC+- hv

I I i I ; i L C atom

105760 105770 105780 105790 105800 10581 0 j.-94.5am Photon Energy (cm-') E-13.leV

Figure 5 -- The 3S 1 <-3P 1 autoionization resonance in C atoms from photodissociation of Cm. Also shown is the ionization signal from acetone (proportional to XUV radiation).

I

h Y I I . 'A (I)

B +J I 4 c

(I

0 5 10 15 20 25 30 35 4 0

time (psec) Figure 6a -- Nonresonant postionization of fragments from photodissociation of C6o. The 337 nm desorption laser precedes the 266 nm dissociation laser by -7 psec, and the 92.32 nm laser is delayed by 30 nsec. The peak at 24 psec is caused by the desorption laser alone.

> c .L

2 c C b

i . 1 . 1 . 1 3 I ti I I I I I.LI I.I.LLILI.L

A

4 4.05 4.1 4.15 4.2 4.25 5.65 5.7 5.75 5.8 5.85 5.9 6.85 6.9 6.95 7 7.05 7.1

time ( w e d time ( p e d time ( w e d

Figure 6b -- Display of the the small !hagnent distribution for C, C2, and C3.

13.43 eV postionization of 266 nm photodissociated products from C,,

0 .

Relative Fluence -> Figure 7 -- Relative intensities for the fiagnents detected in Fig. 6.

1 mJ

0 mJ

A .

3000

2500

2000 ). - .- ; 1500 - C -

1000

500

4 4.5 5 5.5 6 time (pec)

Figure 8 -- Nonresonant postionization of fragnents from dissociation of C70. The 337.1 nm output from a N2 laser is used for desorption, 266.0 nm is used for dissociation, and 92.86 nm for postionization. The spot size for dissociation is -2 mm diameter and counter-rotating mirrors are used as a variable attenuator. The neutral species are shaded in the bottom panel.