1786
J. Phys. Chem. 1995, 99, 1786-1791
Laser-Induced Polymerization within Carbon Disulfide Clusters Sunil R. Desai,? C. S. Feigerle,? and John C. Miller* Chemical Physics Section, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831 -6125 Received: August 8, 1994; In Final Form: October 3, 1994@
"Laser snow" or laser-induced clustering describes a process whereby laser irradiation of a moderate- to high-pressure molecular gas results in the visible precipitation of particulates. We report a molecular beam analog of laser snow where photochemical polymerization occurs within molecular clusters rather than via a collisional mechanism. Clusters of CS2 are formed in a supersonic expansion with argon and photoionized by two-photon absorption of 239.53 nm photons. Various ionic photoproducts such as Sm+, Smf(CS2),, (CS),', and (CSz),+ clusters are detected by mass spectrometry. A cycle of ion-molecule reactions occurring entirely within an isolated molecular cluster is proposed to explain the observed ions.
the wall and formed a brownish coating. The mechanism given for the production of S, is highly dependent on the SZmolecule Carbon disulfide has been studied extensively by electron and which is produced by the collision of an excited CS2 molecule photon excitation techniques to investigate its fragmentation with a molecule in the ground state. The S2 molecule serves dynamics,'S2 spectro~copy,~-~ photochemistry?x8 and nucleas a seed for the formation of an S, cluster of critical size, and a t i ~ n . ~Recently, ~'~ the phenomena of nucleation and laseronce this size is reached the cluster itself becomes the nucleus induced polymerization of small molecules such as CS2 have for the formation of S, droplets that are observed in the TDCC. generated great interest. These experiments fall into three main That is, the S2 molecule reacts with another S2 molecule to categories which are described separately below: those that are produce S4, which in turn reacts with an S2 molecule to produce performed at low to moderate pressures, those that are performed s6, and so on. The same group also investigated photoinduced at high pressures, and those in molecular beams. nucleation involving small amounts of carbon disulfide in At moderate pressures (1-75 Torr) laser-induced particle supersaturated host vapors, such as ethanol and helium. formation or "laser snow" can be observed in CS2 v a p ~ r . ' ~ - ' ~ In contrast to the phenomena described above, reactions This phenomenon was first seen in a mixture of cesium and within van der Waals (vdW) clusters in molecular beams at very hydrogen after excitation of the 7P state of cesium.14 The laserlow pressures are a relatively new area of study. There have induced reaction was extensive enough to lead to the production been many reported instances of intracluster which of macroscopic particles of cesium hydride which were visible are initiated by the ionization of a cluster to form the cation. to the naked eye, hence the term laser snow. Laser snow studies The excess energy that is deposited into the cluster can be in CS2 vapor by Ernst et al.I1-l3 used 337 nm light from a dissipated by the successive evaporation of species that are nitrogen laser and the 351 and 357 nm lines of a krypton laser weakly bound to the ion, by the fragmentation of the cluster to induce clustering. A more detailed wavelength dependence ion where the fragment also carries off the excess energy, or study of clustering in the 335-342 nm regionla indicated that by the initiation of reactions within the cluster to form new the process was initiated following the absorption of photons cationic species. A subset of this class of reactions involves to produce electronically excited CS2 molecules, which, in turn, polymerization processes within vdW clusters. They have been collided with ground state molecules and underwent nucleation reported by two g r o ~ p swhich ~ ~ , have ~ ~ studied species that are and dissociation. The entire reaction sequence was thought to also known to polymerize in bulk samples. In these reactions proceed through the neutral manifold of electronic states. The the fragmentation of the parent cluster is thought to be followed rate of growth of (CS), and S, was measured by collecting the by a series of addition and elimination reactions that produce light from a helium-neon laser that scattered from macroscopic polymeric cations. As an example, El-Shall et al.24 have particles. It is interesting to note that the scattered light signal reported cationic polymerization in isoprene clusters. Clusters exhibited oscillations as a function of time.I0 The scattering were generated by expanding isoprene and helium through a increased as a function of particle size, but after a critical size conical nozzle, ionized by electron impact and detected by a was reached larger particles were removed from the vapor phase quadrupole mass spectrometer. The mass spectrum showed pure in the form of particulates and, consequently, a decrease in the isoprene clusters and numerous fragment ions that were a result scattering was observed. This was followed by the regrowth of a series of addition-elimination reactions. of the smaller particles that remained in the vapor and the cycle The isoprene intracluster processes have a well-known analog resumed. in ion-molecule chemistry as demonstrated by Kascheres et At higher total pressures (100-500 Torr), nucleation can be a1.26 In Kascheres' work an isoprene radical cation and its photoinduced in supersaturated vapor^.^,]^^'^ In experiments by neutral counterpart were guided into an ion trap where they were Kalisky et aL9 a thermal diffusion cloud chamber (TDCC) was allowed to react for 300 ms. The resulting fragment ion used to contain a supersaturated vapor of CS;?,and a xenon arc distribution was very similar in extent and abundance when lamp was used to induce nucleation. S, particles were deposited compared to the intracluster reactions. on the bottom of the chamber while CS molecules diffused to In the present paper, we report the observation of S+, and (CS),+ ions and provide unambiguous evidence of their ' Also Department of Chemistry, University of Tennessee, Knoxville, formation via a unique cycle of consecutive intracluster reactions TN 37996. within parent CS2 clusters. The proposed reactions are supAbstract published in Advance ACS Ahstracrs, January 15, 1995.
Introduction
@
0022-365419512099-1786$09.0010
0 1995 American Chemical Society
Polymerization within Carbon Disulfide Clusters
J. Phys. Chem., Vol. 99, No. 6,1995 1787
ported not only by the observation of the cluster ions that initiate the cycles but also by the intermediates that are involved. Furthermore, the reactions are based on previously observed ion-molecule reactions involving excited CS2+. The excited states of the carbon disulfide ion have been studied, and their respective dissociation limits and dissociative products have been characterized.' The photoelectron spectrum of CS2 shows five prominent bands with four of the bands having resolved vibrational ~tructure.~The spectroscopy of neutral CS2 has also been well studied and shows an extensive and complex absorption spectrum. Some of the stronger bands include the g3A2 8'Z: transition between 330 and 430 nm, the A'B2 8'Z: transition between 180 and 220 nm, the % transition between 162 and 172 nm, and the E 2 transition between 136 and 153.5 nm.6,20327The spectrumbelow 136 nm is dominated by a number of Rydberg transitions. The present work is a continuation of previous cluster studies using picosecond laser ionization coupled with mass spectrometry. Earlier work by this group has involved the study of "magic numbers" in both homonuclear and heteronuclear cluster
-
-
L
4
0 -
'I
S'
*E
h
v
h *
'IQ
-1-
I
cs;
-e
0
1
1
I
I
10
20
30
40
50
,
I
60
70
80
90
100
Mass (amu)
Figure 1. Mass spectrum of a 125 Torr expansion of CS2 using 239.53 nm light showing the S+ and CS2+ photoions. The small peak that is separated by 1 m u from the CS2+ peak is due to HCS2+ while the peak that is separated by 2 amu is due to one 34S isotope on C&+.
distribution^.^^-^^ Experimental Section The apparatus is described in more detail elsewhere2*but will be briefly described here. Various concentrations of CS2 in argon were expanded in a pulsed supersonic jet. In some experiments nitric oxide or methane was also added. The isentropic core of this expansion was skimmed and subsequently intersected at 90" with a focused 30 ps Nd:YAG laser beam. The resulting photoions were extracted into a linear time-offlight (TOF) mass spectrometer where a dual channel plate electron multiplier was used for detection. The wavelengths used included the fourth harmonic of the YAG laser (266 nm) and the first Stokes and anti-Stokes lines at 299.06 and 239.53 nm, respectively, produced by Raman scattering in molecular hydrogen. The Raman shifter (Quanta Ray RS-1) was used with hydrogen because of its large frequency shift. The optimum hydrogen pressure for the fiist Stokes line was found to be approximately 200 psi whereas the optimum pressure for the fiist anti-Stokes line was approximately 150 psi. Pulse energies for the first Stokes and anti-Stokes lines were measured to be 60 and 20 pJ, respectively. Raman shifting of nanosecond3' and picosecond p ~ l s e shas ~ ~been , ~ ~previously demonstrated, and in both cases up to the fifth-order Stokes and third-order anti-Stokes lines have been observed. The various orders of Raman shifted light are separated in a Pellin Broca prism. The laser was operated at a 10 Hz repetition rate, and the opening of the pulsed valve (R. M. Jordan Co.) was triggered by a signal from the laser. The laser pulse was incident on the early edge of the gas jet in order to optimize cluster detection. After each laser shot a distribution of masses was detected by the TOF mass spectrometer, and the mass spectrum was recorded with a digitizing oscilloscope (Tektronix 11402). The signal-to-noise ratio was improved by averaging spectra from about 4000 laser shots. There are several advantages to the use of picosecond lasers for multiphoton ionization (MPI) of clusters. First, at 239.53 and 299.06 nm, MPI in CS2 is nonresonant, and so the higher peak power of the picosecond laser enhances the ion signal. Second, a picosecond pulse allows ionization t o compete favorably with dissociation if short-lived states of the cluster are resonantly populated.33
p - 1 J
12
t
i3A2
I
1
Figure 2. Schematic of the energy level diagram for CS2 and CS2+. Also shown, in the CS2+ ionic manifold, are the dissociation products and their respective dissociation limits. The arrows represent the number of photons needed to ionize CS2 at 239.53 nm and the energy corresponding to one- and two-photon absorption in the neutral and ionic manifold.
Results and Discussion The TOF mass spectrum produced by two-photon ionization (239.53 nm) of a low-pressure (125 Torr) expansion of a CS2/ Ar mixture is shown in Figure 1. This expansion produced no clusters, and the only ions observed are S+ and CS2+. In the mass spectrum the small peak that is separated by 1 amu from CS2+ is due to HCS2+, which is a result of reactions with trace quantities of water, while the peak that is separated by 2 amu is due to CS2+ with one 34S isotope. Figure 2 presents an energy level diagram of CS2 showing the excited states of both the neutral and the ion. The excited states of the ion are shown with their respective dissociation limits and products while the arrows indicate the number of photons needed to ionize CS2 at 239.53 nm and the energy corresponding to one- and two-photon absorption in the neutral and ionic manifolds. The CS2+ that is observed in Figure 1 arises from direct two-photon MPI of the neutral. The formation of the sulfur ion occurs by twophoton ionization of CS2 with subsequent absorption of a third 239.53 nm (5.17 eV) photon by CS2+ into the B 2 c state, which can predissociate via the 4Z- repulsive state: The threshold for this process is 14.81 eV with respect to the ground
1788 .I. Phys. Chem., Vol. 99, No. 6, 1995
Desai et al. I 1 0
-1
cE
-2
Y
.3 -80
1
-4
I (cs; I
L
-100 ,
I
-120 1 0
I 20
40
60
80
100
120
140
160
180
Mass (amu)
Figure 3. Low mass region of a 5% CS2 in argon mixture showing the fragments resulting from the parent and cluster ions.
state of CS2. The last step in the formation of S+ can therefore be described as follows:
-
+
C S ~ + ( ~ Z - > s + ( ~ s >CS('Z+)
(1)
This process has been previously observed at 240 nm by Fischer et who also see only CS2+ and S+ in their mass spectrum. Figure 3 shows the low mass region of the time-of-flight mass spectrum obtained by ionizing a 5% CSdAr mixture using focused 239.53 nm light. In this case the backing pressure has been increased to 100 psi in order to enhance cluster formation. In contrast to the mass spectrum obtained at the low backing pressure (Figure 1) the spectrum in Figure 3 shows numerous ions with masses up to 180 amu. In addition to the S+ and CS2' ions seen previously in Figure 1, we observe two main classes of cluster ions in the mass spectrum. The first includes homogeneous clusters of the form (CS,),+, (S),+, or (CS),+ while the second category includes sulfur and carbon monosulfide cluster ions solvated by additional CS, molecules [Le., Smf(CS,), or CS,+(CS2),]. Again, the small peaks that are separated by 1 amu from the S+ and CS+ ions in Figure 4 are due to HS+ and HCS+ ions, respectively, and clusters with one 34S are observed 2 amu from most of the peaks. The presence of CS+ and S2+ in this mass spectrum (Figure 3) and their absence in Figure 1 indicates that these ions are produced from the dissociation of carbon disulfide cluster ions rather than from the monomer. Although CS, has a number of neutral dissociation channels which could give rise to both CS and S species which could, in turn, be ionized, we believe that the formation of the sulfur and carbon monosulfide ions occurs from the dissociation of CS,+ and (CS2),+, respectively. The dissociation channel producing CS S in the 'Z+and 3Pstates, respectively, has a threshold of 4.46 eV (278 nm).'O Thus, absorption into '2;transition which extends from 189 to 220 nm the 'B2 can result in neutral fragments. Another broad absorption band extends from 290 to approximately 350 nm which cannot lead to dissociation with one photon.6 However, the absorption cross section at 239.53 nm appears to be very smalL6 Consequently, we expect two-photon ionization to dominate over any possible one-photon dissociation processes. Figure 4 shows the higher mass region of Figure 3 on an expanded and amplified scale, and we observe pure sulfur clusters up to the hexamer, carbon monosulfide clusters up to the trimer, and various heterogeneous cluster ions. The observed ions in Figures 3 and 4 are summarized in Table 1. The largest cluster of CS2 observed (but not shown) under the conditions of Figures 3 and 4 is the (CS2)5+ species. However, experi-
+
-
-7
1
I 100
120
140
160
180
200
Mass (amu) Figure 4. Extension of Figure 3 showing a higher mass region. The (S),' and (CS)m+peaks are a result of the intracluster polymerization reactions while the other ions come from purely fragmentation and dissociation processes.
TABLE 1: Summary of Ions Observed in Figures 3 and 4 (a = 239.53 ion (S)m+
(CS),' (CS2)m+
m
ion
n
ccs+
1-11 1-5 1-5
Sm+(CS2)n (CS),+(CS2),
m
n
1 1-9 1-3
1-4 1-3
mental conditions of backing pressure and laser power can be varied such that clusters up to (CS2)13+ are observed. The formation of the fragment ions in Figures 3 and 4 (except for S+ and CS2+) is attributed to photochemistry occurring entirely within individual clusters. Since the CS2 clusters are ionized approximately 6 in. downstream from the supersonic nozzle, it can be safely assumed that all collisions have ceased. It is wellknown that species such as (CS),+, CS+(CS2)m9and (S),+(CS,) can be created as a result of ion-molecule reactions which are given below and were proposed by Henglein35to explain his observations in electron impact high-pressure mass spectrometry studies:
-
+ cs, cs;* + cs, cs;* + cs, CS+ + cs, cs;*
S+(CS,)
+ cs
+ s, CS+(CS,) + s (a),+
(CS),+
+s
(2)
(3) (4)
(5) Ionization efficiency (E) curves were also reported, and all the secondary ions were observed to have the same appearance potential (13.3 eV). The IE curve for (CS)2+ did not rise as steeply as the other ions at threshold while above 16 eV it increased dramatically. Henglein concluded that (CS)2+ is formed by the two separate processes shown above. Reaction 3 accounted for the (CS),+ ions that were observed up to 16 eV while those (CS)2+ ions observed at higher energies had contributions from both reactions 3 and 5 since the threshold for reaction 5 is 16 eV. The initial steps in the mechanisms that we propose for the formation of (S),+, (CS),+, and their respective intermediates are given below:
J. Phys. Chem., Vol. 99, No. 6, I995 1789
Polymerization within Carbon Disulfide Clusters
Enthalpies for Cluster Reaction& cluster reactions AHr (kcaVmo1)
TABLE 2
--
+ cs +s
-26.6 1.9 18.8
CSZ+*(CS2) S+(CSZ) cs2+* (CS2) CS+(CSZ) S’(CS2) s2+ +
-
a
cs
Enthalpies of formation obtained from refs 36-39.
cluster reactions
AHr (kcaVmo1)
--cs+++cscs2
77.4 37
SI,+(CSI,) S3+ CS+(CS*)
CS2+* denotes an excited ion following absorption of a 5.17 eV photon by the ground
state ion. TABLE 3: Enthalpies for Known Ion-Molecule Reaction@ ion -molecule reactions AHr (kcdmol)
---
csz+*+ cs2 S+(CSZ) + cs cs2+*+ CSI, CS+(CSZ)+ s s++ cs2 s*+ + cs s2+ + cs2 s3++ cs s3++ cs2 S4+ + cs
-51.5 -23.0 -22.5 46.5 49.0
ion-molecule reactions
AHr (kdmol)
+ cs2 - ss++ cs ss++ cs2 - + cs ss+ + cs2 - S,+ + cs + cs2 ss++ cs
-10.0
s4+
s7+
-
44.0
sg+
31.0 41.0
Enthalpies of formation obtained from refs 36-39 (also see text). CS2+* denotes an excited ion following absorption of a 5.17 eV photon by the ground state ion. The formation of carbon disulfide clusters is followed by their two-photon ionization and subsequent absorption of an additional 5.17 eV photon to an excited state. 411 clusters of carbon disulfide can be ionized by two photons of 5.17 eV energy since their IPS decrease from that of the monomer (10.068 eV) as a function of increasing cluster size.36 The extra photon is necessary to reach the fragmentation appearance potentials measured by H e n g l e i ~ ~ ~ The following two reactions are proposed for the creation of sulfur and carbon monosulfide ions that are solvated by carbon disulfide molecules:
-
+
+
(cs2)+*(cs2),-1 hv CS+(CS2)n-l s (9) Note that reactions 8 and 9 are cluster analogs of the ionmolecule reactions 2 and 4. Reaction 9 requires an additional 5.17 eV photon because the reaction is slightly endothermic (see Table 2). The enthalpies of reaction for a few of the cluster reactions and for the analogous ion-molecule reactions are given in Tables 2 and 3 and are discussed later. We propose the following cycle of intracluster ion-molecule reactions to account for the creation of pure sulfur cluster ions:
-
s+(Cs,),-,
S,+(CS,),_,
+ cs
(10)
S,+(CS2),-,
S,+(CS2)n-3
+ cs
(11)
~ m + l + ( c ~ ~ n - ( m++ cs l)
(12)
Likewise, the following reactions are proposed for the formation of carbon monosulfide cluster ions: CS+(CS2)n-l (cs)2+(cs2>,-2
(cs),+(cs2),-2 (cs)3f(cs2>n-3
+s
(13)
+
(14)
..
( c s > ~ + ( c s ~ > , -e ~ (cs>m+1+(~~2),-,m+1, + s (15) Again, note that reaction 15 is analogous to the ion-molecule reaction given previously. The ion-molecule analog of reaction 10 has been previously reported by Hiraoka et u Z . , ~ ~who observed the following addition reaction:
S+
+ CS,
-
S,’
+ CS
AHO = -22.5 kcal/mol
-
ocs++ ocs s,++ 2 c o s++ OCS“ s;
+ co s; + ocs S3++ co
(16)
(17) (18) (19)
In both of our sulfur and carbon monosulfide polymerization mechanisms, the respective intermediates cycle through intracluster.ion-molecule reactions and terminate with the formation of a homogeneous cluster ion. For example, the production of S5+ proceeds as follows:
-
+ cs S2+(CS2), + cs
CS2+*(CS2)4 S+(CS,), S+(CS,), S,+(CS,), S,+(CS,),
-
S3+(CS,>,
+ cs
S,t(CS,)
+ cs
s,++ cs
(20) (21)
(23)
(24) Liewise, Sm+and (CS),+ arise from (CS2>,+ or larger clusters. The reactions above show that the production of a given Sm+ ion needs a CS~+*(CS~),-Icluster as the initiating species. Therefore, the termination step and, hence, the formation of the resulting pure sulfur ion are dependent upon the size of the carbon disulfide cluster that initiates the cycle of reactions. Since the molecular beam has a distribution of CS2 cluster sizes, a similar distribution of pure sulfur cluster ion results. In the case above, reaction 24 is the terminating step since there are no more CS2 “reactant” molecules in the cluster. Alternatively, if the clusters have insufficient internal energy to proceed to completion, then the intermediate heteronuclear ions are detected. The formation of S+(CS2) and (CS)2+ has been previously investigated by Patsilinakou et aL4I in a similar mass-analyzed multiphoton ionization experiment. They observed CS2+, CS+, (CS)2+, and S+(CS2) ions. They concluded that both the S+(CS2) and (CS)2+ fragment ions were created either by a dissociation of the carbon disulfide dimer ion or by the fragmentation of the neutral dimer and its subsequent ionization. The absence of carbon disulfide cluster ions in their mass spectra S,+(CS,)
.* .. ..
sm+(cs~n-m
Related ion-molecule reactions that cycle to produce sulfur cluster ions have also been observed in the isoelectronic OCS molecule by Dzidic et al.@
1790 J. Phys. Chem., Vol. 99, No. 6, 1995 was attributed to the high effiency of fragmentation processes. Their mechanism for the production of the S+(CS2) and (CS)2+ clusters of CS2 was also supported by their pressure dependence studies. They observed a quadratic dependence of the (CS)2+ and S+(CS2) signal on the CS2 pressure and, therefore, concluded that both these fragment ions were products of carbon disulfide clusters and not the monomer. In our mechanisms the elimination reactions and the processes by which sulfur and carbon monosulfide are solvated by CS2 are very similar to those proposed by El-Shall et aL2, and by the Garvey group.25 Both suggest processes where the final size of an ion is dependent on the size of the parent ion and internal barriers to intracluster reactions coupled with the thermodynamics and rate kinetics of particular elimination reactions. However, we believe that we have provided unambiguous evidence for a unique cycle of consecutive intracluster reactions within CS2 clusters that produces pure sulfur and carbon monosulfide cluster ions. Our proposed reactions are supported by the observation of the cluster ions that initiate the cycles and the respective intermediates that are involved. Our mechanisms are based on known ion-molecule reactions and CS2 spectroscopy and are similar in this respect to the isoprene studies by El-Shall et ~ 1and. methyl ~ ~halide reactions studied by Garvey et ~ 1 . ~ ~ The heats of formation that were used to determine the values in Tables 2 and 3 come from a variety of sources. Heats of formation for Sf(CS2) and CS+(CS2) were obtained by Ono et who used a helium light source to ionize a CS2/Ar mixture and subsequently detected the ions with a quadrupole mass spectrometer. They determined the appearance potentials for these fragment ions which, combined with the binding energies, gave heats of formation for S+(CS2) and CS+(CS2) as 291.7 and 318 kcal/mol, re~pectively.~~ The bond dissociation energies for (CS#(CSz) and S2+(CS2) were obtained by Hiraoka et ~ 2 1 .using ~ ~ a pulsed electron beam high-pressure mass spectrometer and are given as 30.9 and 24.9 kcal/mol, respect i ~ e l y .Using ~ ~ these values and knowing the heats of formation for CS2, CS2+ and S2+, we can obtain the heat of formation for (CS2)+(CS2) and Sz+(CS2) as 263.1 and 243.6 kcallmol, respectively. The heats of formation for S,, S+,, CS, CS+, CS2, and CS2+ were obtained from tabulated values given by Structures for Sz+(CS2), and CS2+(CS2), have been Lias et calculated by Hiraoka et and energy differences between geometrical isomers are given. The discrepancies that arise between the different values may be due to the fact that some of the heats of formation have been obtained by ionization methods where unfavorable Franck-Condon factors near the ionization limit could make the determination of the true adiabatic energies difficult. In these cases, the appearance potential is the only value that can be extracted from the data.
Other Carbon Disulfide Reactions Having a ready source of excited CS2+ ions, it was of interest to identify other intracluster reactions suggested by known ionmolecule reactions. The first system to be studied was a mixture of methane and carbon disulfide. Here, 260 Torr of CS2 was coexpanded with an equal amount of methane and 100 psi of argon, and the resulting supersonic beam was photoionized with 239.53 nm light. Although extensive fragmentation of CS2 was observed, the new ions of interest in the mass spectrum were CH3Sf, CHS+, and CSfC&. The formation of these ions is attributed to intracluster chemistry, but they also have wellknown ion-molecule analogs. In the case of CHS+, a bimolecular reaction between the CS+ ion and a neutral methane molecule produces the product ion e x c l ~ s i v e l y : ~ ~
CS'
+ CH,
-
Desai et al. HCS'
+ CH,
(25)
The bimolecular reaction between the sulfur ion and neutral methane produces the CH2SH+ and HCS+ ions.,*
+ CH, S+ + CH, S+
+ CH, CH2SHf + H
==)
+
HCS'
5% 95%
(27)
Experiments were also performed on mixtures of NO/Ar and carbon disulfide to investigate the possibility of intracluster chemistry. A 100 psi mixture of 5% N0/5% CSd90% Ar was irradiated with focused 239.53 nm light. Three new ions were observed that were of particular interest, NO%, NO+&, and (N0)2+S. The corresponding ion-molecule reactions have not been studied, but the neutral species have been observed by Hawkins et ~ 1in photolysis . ~ experiments of carbonyl sulfide with NO monomer and dimer.44 Calculations have been performed on SN202 where the geometric, electronic, and vibrational characteristics were s t ~ d i e d . ~ , , ~
Conclusions Multiphoton-induced cationic polymerization in carbon disulfide clusters has been observed by photoionizing a supersonic expansion containing carbon disulfide in argon using 239.53 nm light. The resulting polymerization was detected by mass spectrometry which indicated the presence of (CS),+, (S),+, (S),+(CS2),, and (CS),+(CS,),. Possible mechanisms have been proposed that attribute the formation of sulfur cluster ions and carbon monosulfide cluster ions to a cycle of reactions that occur entirely within clusters. These mechanisms are based on ionmolecule reactions which have been previously observed in collisional environments. The formation of Sf is attributed to the absorption of a photon by the CS2+ ion into the BIZ: excited state which predissociates via the 4Z- repulsive state to form S+(,S) CS('Z+). The intracluster mechanisms that have been proposed account for all the ions that we observe and involve reactions where neutrals are ejected.
+
Acknowledgment. Research is sponsored by the Office of Health and Environmental Research, U.S. Department of Energy, under Contract DE-AC05-850R21400 with Martin Marietta Energy Systems, Inc. J.C.M. acknowledges the use of a NATO International Collaboration Grant (No. 870474) during the course of this work. References and Notes (1) Brehm, B.; Eland, J. H. D.; Frey, R.; Kiistler, A. Int. J . Muss Spectrom. Ion. Phys. 1973, 12, 213. (2) Waller, I. M.; Hepbum, J. W. J . Chem. Phys. 1987, 87, 3261. (3) Greening, F.; King, G. W. J . Mol. Spectrosc. 1976, 59, 312. (4) Maier, J. P. J. Electron Spectrosc. Relat. Phenom. 1993, 66, 15. ( 5 ) Brundle, C. R.; Turner, D. W. Int. J . Mass Spectrom. Ion Phys. 1969, 2, 195. (6) Rabalais, J. W.; McDonald, J. M.; Scheer, V.; McGlynn, S. P. Chem. Rev. 1971, 71, 73 and references therein. (7) Eland, J. H. D.; Berkowitz, J. J . Chem. Phys. 1979, 70, 5151. (8) Tzeng, W.-B.; Yin, H.-M.; Leung, W.-Y.; Luo, J.-Y.; Nourbakhsh, S.; Flesch, G. D.; Ng, C. Y . J . Chem. Phys. 1988, 88, 1658. (9) Kalisky, 0.;Heist, R. J . Chem. Phys. 1985, 83, 3668. (10) Vlahoyannis, Y. P.; Patsilinakou, E.; Fotakis, C.; Stockale, J. A. D. Radiut. Phys. Chem. 1990, 36, 523. Patsilinakou, E. Ph.D. Thesis. (1 1) Emst, K.; Hoffman, J. J. Chem. Phys. Lett. 1979, 68, 40. (12) Emst, K.; Hoffman, J. J. Chem. Phys. Lett . 1980, 75, 388. (13) Emst, K.; Hoffman, J . J. Phys. Lett. 1981, 87A, 133. (14) Tam, A.; Moe, G.; Happer, W. Phys. Rev. Leu. 1975, 35, 1630. (15) Wen, F. C.; McLaughlin, T.; Katz, J. L. Phys. Rev. A 1982, 26, 223.5. (16) Katz, J. L.; Wen, F. C.; Mclaughlin, T.; Reusch, R. J. Science 1977, 196, 1203.
Polymerization within Carbon Disulfide Clusters (17) Garvey, J.; Peifer, W. R.; Coolbaugh, M. T. Acc. Chem. Res. 1991, 24, 48.
(18) Ono, Y.; Ng, C. Y. J . Am. Chem. SOC.1982,104, 4752. (19) Stace, A. J. J. Am. Chem. SOC.1985, 107, 755. (20) Kenny, J. E.; Brumbaugh, D. V.; Levy, D. H. J. Chem. Phys. 1979, 71, 4757. (21) Stace, A. J.; Shukla, A. K. J . Phys. Chem. 1982, 86, 865. (22) Nishi, N.; Yamamoto, K.; Shinohara, H.; Nagashima, U.; Okuyama. Chem. Phys. Lett. 1985, 122, 599. (23) Wei, S.; Tzeng, W. B.; Castleman, A. W. J. Phys. Chem. 1991, 95, 5080 and references therein. Castleman, A. W.; Tzeng, W. B.; Wei, S.; Morgan, S. J . Chem. SOC.,Faraday Trans. 1990,86,2417 and references therein. (24) El-Shall, M. S.; Marks, C. J . Phys. Chem. 1991, 95, 4932. ElShall, M. S.; Shriver, K. E. J . Chem. Phys. 1991, 95, 3001. (25) Whitney, S. G.; Coolbaugh, M. T.; Vaidyanathan, G.; Garvey, J. F. J. Phys. Chem. 1991, 95, 9625. Coolbaugh, M. T.; Vaidyanathan, G.; Peifer, W. R.; Garvey, J. F. J . Phys. Chem. 1991, 95, 8337. (26) Kascheres, C.; Cooks, R. G. Anal. Chim. Acta 1988, 215, 233. (27) Herzberg, G. In Molecular Spectra and Molecular Structure III; Van Nostrand Rheinhold: New York, 1966. (28) Desai. S. R.: Feiwrle. C. S.: Miller. J. C. J. Chem. Phvs. 1992. 97. 1793; Z. Phys. D 1993, 26, 220. (29) Desai, S . R.; Feigerle, C. S.; Miller, J. C. Z. Phys. D 1993, 26, S183. (30) Desai, S. R.; Feigerle, C. S.; Miller, J. C. J. Chem. Phys. 1994, 101, 4526. (31) Sato, K.; Achiba, Y.; Kimura, K. Chem. Phys. Lett. 1986, 126, 306. ~~~~
J. Phys. Chem., Vol. 99, No. 6, 1995 1791 (32) Wilkerson, C. W., Jr.; Sekreta, E.; Reilly, J. Appl. Opt. 1991, 30, 3855. (33) Smith, D. B.; Miller, J. C. J . Chem., SOC.Faraday Tran. 1990,86, 2441; J. Chem. Phys. 1989, 90, 5203. (34) Fischer, I.; Lochschmidt, A.; Strobel, A,; Neidner-Schatteburg, G.; Muller-Dethlefs, K.; Bondybey, V. E. Chem. Phys. Left. 1993, 202, 542. (35) Heinglein, A. Z. Nutwforsch. 1962, 17A, 37. (36) Ono, Y.; Linn, S.H.; Prest, H. F.; Gress, M. E.; Ng, C. Y. J . Chem. Phys. 1980, 73, 2523. (37) Ono, Y.;Linn, S. H.; Prest, H. F.; Gress, M. E.; Ng, C. Y. J. Chem. Phys. 1981, 74, 1125. (38) Hiraoka, K.; Fujimaki, S.; Aruga, K.; Yamabe, S. J. Phys. Chem. 1994, 98, 1802. (39) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D.; Mallard, W. G. J. Phys. Chem. Re$ Data 1988, 17 (Suppl. No. 1). (40) Dzidic, I.; Good, A.; Kebarle, P. Can. J. Chem. 1969, 48, 664. (41) Patsilinakou, E.; Proch, D.; Fotakis, C. Chem. Phys. 1991, 153, 503. Prinslow, D. A.; Vaida, V. J. Phys. Chem. 1989, 93, 1836. (42) McAllister, T. Int. J. Mass Spectrom. lon Phys. 1974, 13, 63. (43) Hawkins, M.; Downs, A. J. J . Phys. Chem. 1984, 88, 3042. (44) Stradella, 0. G.; Maluendes, S. A.; Castro, E. A.; Jubert, A. H. J . Mol. Struct. (THEOCHEM) 1990, 210, 169. (45) Maluendes, S. A.; Jubert, A. H.; Castro, E. A. J. Mol. Struct. (THEOCHEM) 1990, 204, 145.
JP942086M
