Oxidative Reactions of Silicon Atoms and Clusters at Ultralow

Nov 19, 2010 - ... that the abundance of SiO molecules in the universe is governed by the ..... The mass spectra of He droplets doped with O2 alone an...
0 downloads 0 Views 819KB Size
Download PDF
J. Phys. Chem. A 2010, 114, 13045–13049

13045

Oxidative Reactions of Silicon Atoms and Clusters at Ultralow Temperature in Helium Droplets Serge A. Krasnokutski and Friedrich Huisken* Laboratory Astrophysics Group of the Max Planck Institute for Astronomy at the Friedrich Schiller UniVersity Jena, Institute of Solid State Physics, Helmholtzweg 3, D-07743 Jena, Germany ReceiVed: October 28, 2010

The reaction between Si and O2 was studied in liquid He droplets at low temperature (T ) 0.37 K) by monitoring the energy release during the reaction. Additionally, the reactions of Si atoms and clusters with the oxidation agents H2O and O2 have been studied by mass spectrometry. It was found that Si atoms react fast with O2 molecules. On the other hand, Si atoms and clusters do not react with H2O molecules. The energy released during the chemical reaction leads to the ejection of the products from small He droplets. In contrast, large He droplets (NHe > 20000) are capable of keeping part of the reaction products in their interior. The observation of SiO2 products with the mass spectrometer reveals that the He droplet can stabilize intermediate products in the exit channel. Introduction Oxidative reactions of Si atoms at low temperatures have been studied mainly because of their astrochemical importance.1 As compared to low-temperature environments, the high-temperature regions of the universe reveal much higher concentrations of SiO.2 As high-temperature kinetic studies of the reaction Si + O2 f SiO + O predicted that this reaction should be extremely slow at low temperature,3 it was concluded that the abundance of SiO molecules in the universe is governed by the Si + O2 f SiO + O reaction.4 However, later studies of this reaction performed at low temperatures showed that, for T < 300 K, the reaction rate increases when the temperature is reduced.5,6 These results support a model that explains the lower SiO concentration in quiescent low-temperature regions by the depletion of silicon-containing molecules on cold grains, whereas in photodominated or dynamically active regions, the abundance of SiO increases due to their photodesorption from grains.1,7 Ab initio calculations predict that all possible reaction pathways of the reaction Si + O2 f SiO + O have no energy barrier and yield three possible intermediate products associated with a well in the exit channel potential energy surface.8 Crossed molecular beam experiments and theoretical studies applied to this reaction show that, at low temperature, the reaction goes through the SiOO intermediate. With increasing temperature, the probability of an insertion reaction yielding an intermediate with OSiO geometry also increases.9,10 Helium droplets provide an ideal medium to study chemical reactions at ultralow temperature. The first reaction studied inside He droplets by observing the chemiluminescence light was the reaction Ba + N2O f BaO + N2.11 Recently, the reaction of alkali metal clusters with water molecules inside He droplets was investigated by mass spectrometry.12 It was found that Cs clusters react with water clusters, while Na clusters predominantly form van der Waals complexes. Very recently, we have studied the reaction of Mg atoms and clusters with O2 molecules in He droplets by observing the chemiluminescent light and using mass spectrometric detection.13 We could show * To whom correspondence should be addressed. E-mail: friedrich.huisken@ uni-jena.de.

that the reaction between magnesium clusters and oxygen molecules is rather fast. In the present study, we describe a new method to monitor the chemical reaction between two neutral molecules in He droplets by observing the reduction of the diameter of the He droplets resulting from the energy released by the reaction. Additionally, mass spectroscopic measurements were applied for the characterization of the reaction products. Experimental Section The experiments have been carried out in a He droplet apparatus reported earlier.13,14 Figure 1 shows the experimental setup used in the present study. Large helium clusters were produced by supersonic expansion of pure helium gas at high pressure (p0 ) 20 bar) through a 5 µm diameter pinhole nozzle cooled by liquid helium. The temperature of the nozzle could be varied between 5 and 20 K, thus providing different sizes of He droplets. The average diameter (dHe) and the average number of He atoms (NHe) are evaluated according to the correlation between He droplet size and nozzle temperature published by Toennies and Vilesov.15 The values relevant for this study are given in the Supporting Information. For the present experiments, commercially available silicon (CrysTec wafer, boron doped, F < 0.05 Ω cm) and oxygen (Air Liquide 99.999%) were used without further purification. For the incorporation of Si atoms, a silicon slab (50 × 3 × 0.5 mm3) was cut from the wafer and mounted parallel to the helium droplet beam with a separation of 3 mm from the beam axis. The slab was resistively heated to a temperature close to the melting point of the silicon. The silicon slab and the helium droplet beam were surrounded by a quartz tube, which prevented the Si atoms from being spread into the main chamber. Control of the silicon incorporation into the He droplets was obtained in the following way. We slowly increased the current through the silicon slab until the incorporation of silicon into the He droplets could be first detected with the mass spectrometer. At this point, we kept the temperature of the silicon slab constant and changed the He droplet size, thus varying the pickup cross-section of the He droplets. The probabilities that the

10.1021/jp110323n  2010 American Chemical Society Published on Web 11/19/2010

13046

J. Phys. Chem. A, Vol. 114, No. 50, 2010

Krasnokutski and Huisken

Figure 1. Schematic diagram of the experimental setup used for mass spectroscopic and depletion measurements. For the case of kinetic measurements, the O2 pick-up cell was replaced by a multichannel effusive source.

helium droplets have picked up n Si atoms are given by Poisson distributions

Pn )

(Nσl)n exp(-Nσl) n!

(1)

where N is the number density of Si atoms in the volume crossed by the helium droplet beam, σ is the cross-section of the helium droplet for picking up one Si atom, and l is the length of the interaction path.16 The pick-up cross-section σ is proportional to the geometric cross-section of the He droplet, which is πd2/4 (d ) diameter). To calibrate the set of Poisson distributions with respect to the experimental conditions, we have measured the intensity of the Si peak in the mass spectrum (m ) 28 amu) as a function of the diameter of the He droplets in the regime where dimer formation is negligible. The theoretical curve for the Si monomer (n ) 1 in eq 1) was matched with the experimental monomer curve by adjusting the parameter N, which was then used for calculating the distributions of Si clusters in larger He droplets. To account for the increase of the ionization cross-section with increasing He droplet size, the Si mass peak intensity was normalized to the intensities of the He droplet fragments (Hen, n g 17). The ratios of the ion intensities of Si, Si2, and Hen peaks were measured for several He droplet sizes and calibrated against the known Poisson distributions. In this way, we can correlate the measured ion intensities of the Sin peaks with the probabilities that the He droplets have picked up n Si atoms. The oxygen was introduced through a leak valve (Vacuum Generators, LVM series). To monitor the incorporation of oxygen, the following procedure was applied. In a preparatory experiment, the oxygen was supplied to the main chamber through a tube whose exit was located far away from the He droplet beam, thus providing an equilibrium pressure of oxygen along the complete path of the He droplet beam. The pressure of oxygen was monitored by an ionization gauge, thus providing the input data (N) for deriving the Poisson distribution of oxygen. At the same time, the mass spectra were recorded, and the ratios between the ion intensities of O2, O4, and Hen peaks were determined. When studying the reactions, a direct measurement of the oxygen pressure in the small pick-up cell was not possible, and the previously determined correlation between the product Nl in eq 1 and the ion signal was used to monitor the oxygen incorporation. The described calibration procedures allow us to correlate the intensity of an ion signal [(O2)m+ or Sin+] with the percentage of He droplets having picked up m oxygen molecules or n Si atoms, respectively. In this way, during the chemical reaction study, the incorporation of all reactants was quantified by monitoring the ion signal at the respective masses.

For depletion and mass spectroscopic measurements, the oxygen was supplied to a small pick-up cell as indicated in Figure 1. In contrast, to carry out kinetics measurements, the oxygen was provided through a multichannel effusive source, which has been used in earlier pick-up experiments.13,17 In both cases, under the condition of single O2 molecule incorporation, the gas load was so small that the background pressure of 5 × 10-7 mbar in the main chamber remained unaffected. Therefore, the partial pressure of oxygen in the main chamber is estimated to be lower than 5 × 10-8 mbar. Moreover, the partial pressure of the products, which could possibly be formed if oxygen molecules managed to reach the Si vaporization zone, should be even lower, considering the high depletion rate of silica molecules at any surface except the surface of the hot Si slab. Hence, they cannot be incorporated into He droplets and detected by the mass spectrometer. The position of the multichannel source, defining the point of oxygen incorporation, was translated along the He droplet beam, thus varying the time given to the reaction partners to react before the detection. The pressure in the ultrahigh vacuum (UHV) detector chamber was recorded with a precision ion gauge (Varian UHV-24), while the pressure in the main chamber was measured with a simpler ion gauge (Leybold Heraeus IE-20). Results and Discussion The liquid He droplet is an ideal nanocalorimeter, which is capable of detecting even small amounts of heat. In the case of a chemical reaction, the energy released during this reaction is partially transferred to the He droplet, causing the evaporation of He atoms. This results in a reduction of the He droplet size, which can be detected in the experiment. For this purpose, we have tested several methods. The pressure-based method was found to be the most sensitive and reliable technique. In this method, we monitored the partial pressure of helium in the detector chamber. This chamber is separated from the main chamber by a pinhole of 0.8 mm diameter and two additional differentially pumped chambers. This allows us to keep the background pressure in the detector chamber around 2 × 10-9 mbar, in the case where the He droplet beam is blocked. At normal operating conditions, the He droplets pass through the pinhole and evaporate completely in the detector chamber after collision with the wall. This can be measured by an increase of the helium partial pressure in the detector chamber. The general expectation is that we should have a linear correlation between the average number of He atoms per droplet and the pressure in the detector chamber, as the pumping speed does not vary much in the small pressure range. This was also found in an earlier study of Mozhayskiy et al.,18 where the depletion of the He droplet beam as a result of the pick-up process was monitored by a pressure measurement. In our

Oxidative Reactions of Silicon Atoms and Clusters

Figure 2. (a) Correlation between the average He droplet size and the pressure in the detector chamber. Dots represent measured pressure values while the solid line is a linear fit to this data in the range NHe ) 2000-8500. The same linear fit is shown in panels b and c. The horizontal solid lines in the yellow boxes give the pressure values measured upon incorporation of the reactants. The two yellow boxes in panel b correspond to experiments with different He droplet sizes obtained with the nozzle temperatures given in the boxes. The experiments in panels a-c had to be carried out on different days. Thus, the absolute pressure values are slightly different.

experiment, we changed the temperature of the nozzle, varying the average size of the He droplets, and measured the pressure in the UHV chamber. The measured pressure values are given in Figure 2a by the dots. For He droplet sizes 2000 < NHe < 8500 (13 K < T < 16 K), we found a correlation between the average number of He atoms per droplet and the pressure in the detector chamber, which was close to linear. The deviation from the linear dependence in the large He droplet size range is explained by a reduction of the number of He droplets produced by the source. Under the conditions where very small He droplets are produced, a supersonic well-collimated jet of He atoms overlaps with the cluster beam giving rise to slightly enhanced pressure values. From these results, we can conclude that the number of He droplets produced by the source in the temperature range between 13 and 16 K does not change considerably. The linear fit to this data is shown by the solid line in all three panels a-c of Figure 2. After having established the dependence of the pressure in the main chamber on the size of the He droplets, we monitored the depletion of the He droplet beam after incorporation of each reactant. The pick-up process leads to a moderate reduction of the He droplet size because of the need to dissipate the translational and internal energy. If no reaction between the reactants occurs, the total depletion after incorporation of both reactants should be roughly equal to the sum of the depletions caused by the incorporation of each individual reactant. However, in the case of a reaction, additional energy will be released. As a result, the total depletion will be larger than the sum of the individual depletions.

J. Phys. Chem. A, Vol. 114, No. 50, 2010 13047 As can be seen in Figure 2b, the depletion of the He droplet beam, caused by the incorporation of Si and O2, is considerably larger than the sum of the depletions due to the incorporation of the individual reactants. In a test experiment, in which silicon was replaced by nitrogen (see Figure 2c), we found that the pressure changes were additive, indicating that no additional heat was dissipated. Careful inspection of the data shows that the depletion caused by the O2 incorporation into N2-doped He droplets is even slightly smaller as compared to the case where N2 was not present. This can be understood if one considers that the incorporation of N2 leads to a size reduction of the He droplets and, consequently, to a smaller probability for the O2 molecule being picked up. On the basis of these results, we can conclude that the reaction between Si atoms and O2 molecules definitely takes place in the He droplets at the extremely low temperature of T ) 0.37 K. To estimate the rate of this chemical reaction, we translated the multichannel source along the He droplet beam, in this way shifting the incorporation point of the second reactant (O2), and monitored the depletion of the He droplet beam, which was caused by the incorporation of O2. This allows us to vary the time given for the reaction to occur. This time is defined by the time-of-flight of the He droplets from the point where the second reactant (O2) is incorporated to the pinhole in front of the detector chamber. In our experiment, we found no change in the depletion of the He droplet beam as a function of the position of the O2 incorporation. The minimum distance between the point of O2 incorporation and the pinhole, which could be achieved in our experiment, was 30 mm. Thus, we conclude that the reaction between at least 90% of the reactants has been completed during the time needed for the He droplets to travel this distance. This time is approximately 100 µs. On the basis of these data, we can evaluate a lower limit of the reaction rate to be 5 × 10-14 cm3 mol-1 s-1. The temperature dependence of the reaction rate for the Si + O2 reaction was measured by Le Picard et al.5 Fitting their experimental data in the temperature range between 15 and 300 K and extrapolating it to T ) 0.37 K, we obtain a more than 10 orders of magnitude slower reaction rate. Consequently, our observation suggests a dramatic change of the temperature dependence of the reaction rate somewhere below T ) 15 K. With the finding that the reaction between Si and O2 is fast, we can completely rule out the possibility of an ion-molecule reaction in the ionizer. If two reactants are in the same droplet, they will react before reaching the ionizer, as the minimum flight time of the He droplets from the main chamber to the ionizer is 1 ms. Therefore, we can apply mass spectroscopy to characterize the products of the Si + O2 reaction. The mass spectra of He droplets doped with O2 alone and with Si and O2 together are shown in Figure 3. The experiment was performed under the higher temperature conditions used in the previous depletion experiment, that is, with a nozzle temperature of 15 K yielding dHe ) 7 nm. As can be seen in the figure, the intensity of the ion peak at 32 amu (O2+) is reduced due to the incorporation of Si atoms. At the same time, no change in the intensity of any other ion was observed. The depletion of the O2 peak demonstrates the occurrence of reaction between Si and O2, which confirms the results of our depletion measurements. However, under the conditions of this measurement, we could not observe any indication of product molecules (SiO or SiO2) in the mass spectra. The expected exothermicity for the Si + O2 f SiO + O and Si + O2 f SiO2 reactions are 52.9 and 147.2 kcal mol-1, respectively.19 Even the smaller reaction energy (52.9 kcal

13048

J. Phys. Chem. A, Vol. 114, No. 50, 2010

Figure 3. Mass spectra of He droplets doped with O2 (black curve) and with Si and O2 (red curve). The inset shows a detail of the same spectrum with enhanced mass resolution. The spectra were recorded at a nozzle temperature of 15 K yielding an average He droplet diameter of dHe ) 7 nm (NHe ) 3500). Under the conditions of this experiment, about 21% of the He droplets contained one Si atom, while the number of He droplets hosting more Si atoms was below 4%.

mol-1) would result in the evaporation of approximately 3500 He atoms. Thus, the reaction energy is capable of completely evaporating a small He droplet. However, even for much larger He droplets (NHe e 15000), which cannot be completely vaporized by the reaction energy, no products were observed in the mass spectrum. The situation seems to be very similar to the case of the Mg + O2 reaction, where the ejection of the chemiluminescent product molecules was observed even for very large He droplets containing ∼106 He atoms.13 Hence, we suggest the ejection of the reaction products according to the following scenario rather than the complete vaporization of the He droplets. Part of the reaction energy is transferred to the He droplet resulting in the evaporation of He atoms and the sudden formation of a bubble of gaseous helium surrounding the product molecule. If the He droplet is small, the bubble can tear apart the He droplet into two or more parts, thus releasing the incorporated molecule. As a result, most of the product molecules will not reach the detector. The product molecules, which are expelled in forward direction so that they can reach the ionizer, are not observed due to their much smaller ionization cross-section as compared to the one of the He droplets. Obviously, with larger He droplets, we should have a higher probability of trapping the product molecules inside the He droplets. Reducing the source temperature to values below 11 K (NHe g 15000), we start to observe the appearance of ion peaks, which can be attributed to the products of the chemical reaction between Si and O2. In Figure 4, we display the differential mass spectrum obtained by subtracting the mass spectrum of He droplets doped with Si atoms alone from the mass spectrum of He droplets doped with Si and O2. Both spectra were recorded at a nozzle temperature of 10 K. The differential spectrum demonstrates the effect of O2 incorporation when the He droplets already contain Si. Positive peaks belong to oxygen and also point to new species formed as a result of the reaction, whereas negative peaks reveal the species that were consumed. The product molecules SiO, SiO2, and SiO3 (the corresponding mass peaks are displayed in green) as well as Si2O, Si2O2, and Si2O3 (dark blue) were identified in the mass spectrum. The intensity of the negative peaks to be assigned to silicon (Si, Si2, Si3, and Si4) is much smaller than the intensity of product peaks, which appear as positive peaks in the mass spectrum. From this, it is concluded that, similar to some other

Krasnokutski and Huisken

Figure 4. Differential mass spectrum obtained by subtracting the mass spectrum of He droplets doped with Si alone from the mass spectrum of He droplets doped with Si and O2. Both spectra were recorded at a nozzle temperature of 10 K (dHe ) 13 nm, NHe ) 23000). Prior to the subtraction, the mass spectra were normalized so as to obtain equal heights of the Hen peaks at large n (n g 17). Under the conditions of this experiment, 37, 17, 5, and 3% of the helium droplets had picked up one, two, three, and four Si atoms, respectively. The original mass spectra used for the calculation of the present spectrum are available in the Supporting Information.

elements like, for example, Ne,20 the probability of charge transfer from initially formed He+ to Si atoms or Si clusters is rather low. This conclusion becomes even more evident if one considers that it is very likely that many products are ejected from the He droplets. This low ionization probability could possibly affect the accuracy of the method, which was used to calibrate the incorporation of Si atoms. To check this, we can resort to the mass spectra recorded after reaction has occurred. Consider the situation reflected by Figure 3 where single O2 molecules are incorporated into Si-doped He droplets and, for comparison, into pure He droplets. If the O2 signal is depleted by a certain percentage p, it can be concluded that at least p% of the He droplets must have contained a Si atom. This analysis yields slightly higher Si doping values than the first calibration against the Poisson distributions. All concentrations of Si atoms and clusters given in this article and the Supporting Information have been subjected to this correction. Both SiO and SiO2 product molecules are observed in the mass spectrum. Therefore, we suppose that the reaction between Si atoms and O2 molecules in He droplets follows the two pathways represented by the equations Si + O2 f SiO + O and Si + O2 + M f SiO2 + M, where M is a third body to which the energy is transferred. The role of M can also be taken over by the He bath. The intensity of the SiO2 peak is much larger than that of the SiO peak. This finding is comparable to the results of a matrix isolation study21 where Si and O2 were codeposited into an argon matrix at T ) 11 K and where SiO2 was found to be the most abundant product as revealed by IR spectroscopy. The dissociation of the SiO2 molecule to SiO and O requires an energy of about 94.3 kcal mol-1.19 Thus, in the case of a fast dissipation of the reaction energy, the dissociation will not occur. In this case, the liquid helium environment stabilizes the intermediate reaction product, SiOO, and prevents this molecule from fragmentation. The mass spectrum also reveals several weak peaks that are assigned to van der Waals complexes of water molecules with reactant or product species (O2 · H2O, Si · H2O, SiO · H2O, and SiO2 · H2O). The lowest achievable background pressure in the main chamber (5 × 10-7 mbar) is mainly caused by the presence of water molecules. Because of this fact, water molecules are

Oxidative Reactions of Silicon Atoms and Clusters

J. Phys. Chem. A, Vol. 114, No. 50, 2010 13049 This method was used to study the reaction between silicon and oxygen. A lower limit of the reaction rate was evaluated to be k ) 5 × 10-14 cm3 mol-1 s-1. The relatively fast reaction at T ) 0.37 K indicates that there is no barrier in the entrance channel of this reaction. The present results suggest that the Si + O2 reaction may play an important role in interstellar environments. The liquid helium is capable of stabilizing the intermediate reaction products, thus affecting the exit channel of the reaction. The stabilization of the intermediate product (SiO2) may also be relevant if the reaction occurs on a cold surface provided, for example, by a dust grain. In the gas phase, we would expect the formation of SiO. However, the entrance channel should only be little affected by the liquid helium environment.

Figure 5. Mass spectra of He droplets with different average sizes doped with Si and H2O together. The incorporation of H2O molecules has been facilitated by a higher residual gas pressure in the main chamber as compared to the other experiments. The number of He droplets containing more than a single water molecule can be monitored by the intensity of the m ) 19 amu peak.

always incorporated into a few He droplets. Therefore, the possibility that the Si + H2O reaction occurs in the He droplets and interferes with the Si + O2 reaction should be considered. Figure 5 shows mass spectra of He droplets doped with Si and H2O. The spectra were recorded at high residual gas (mainly water) pressure (p ) 2 × 10-6 mbar) in the main chamber employing He droplets of different sizes. The amount of Si atoms and water molecules incorporated into the He droplets increases with the He droplet diameter. With the largest He droplets, the incorporation of up to four Si atoms per droplet was detected. The change in the intensity of the Si-related peaks is much weaker as compared to the change in the intensity of water-related peaks. This confirms the previous conclusion about the low probability of charge transfer from initially formed He+ to Si. While observing the expected intensity rise of the reactant ion peaks, we also find an increase in intensity of the ion peak at 45 amu, which can be assigned to SiOH+. This ion could be produced by ionization of the product of the chemical reaction Si + H2O f SiOH + H as well as by ionization of the van der Waals complex Si · OH2. Here, we should consider two facts. First, similar mass spectra were observed in our previous study where we incorporated Mg atoms together with H2O molecules.13 However, the reaction Mg + H2O f MgOH + H is endothermic and cannot occur at 0.37 K. Second, in the present study, the SiOH peak is also observed for small He droplets while, under the same conditions, the products of the Si + O2 reaction are ejected from the He droplets and not detected with the mass spectrometer. Thus, it is evident that not much energy is released when Si and H2O interact in the He droplet. Unfortunately, it was not possible to carry out depletion studies for this reaction as the concentration of water molecules in the main chamber cannot be easily changed. On the basis of these observations, we conclude that the SiOH+ peak observed in our mass spectra originates from the ionization of the Si · OH2 van der Waals complex. Conclusions We have developed a new method that allows us to monitor the chemical reaction between neutral species inside He droplets.

Acknowledgment. We are grateful for the support by the Max Planck Institute for Astronomy (MPIA) and the Deutsche Forschungsgemeinschaft DFG (Contract No. Hu 474/22-1). Supporting Information Available: Original mass spectra used to obtain the differential mass spectrum displayed in Figure 4; pressure response on doping pure and Si-containing He droplets with O2 molecules, measured in the detector chamber for the condition of low reactant concentrations; and table of the experimental conditions used. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Walmsley, C. M.; des Foreˆts, G. P.; Flower, D. R. Astron. Astrophys. 1999, 342, 542. (2) Ziurys, L. M.; Friberg, P.; Irvine, W. M. Astrophys. J. 1989, 343, 201. (3) Swearengen, P. M.; Davis, S. J.; Niemczyk, T. M. Chem. Phys. Lett. 1978, 55, 274. (4) Langer, W. D.; Glassgold, A. E. Astrophys. J. 1990, 352, 123. (5) Le Picard, S. D.; Canosa, A.; des Foreˆts, G. P.; Rebrion-Rowe, C.; Rowe, B. R. Astron. Astrophys. 2001, 372, 1064. (6) Le Picard, S. D.; Canosa, A.; Reignier, D.; Stoecklin, T. Phys. Chem. Chem. Phys. 2002, 4, 3659. (7) Turner, B. E. Astrophys. J. 1998, 495, 804. (8) Dayou, F.; Spielfiedel, A. J. Chem. Phys. 2003, 119, 4237. (9) Yamashiro, R.; Matsumoto, Y.; Honma, K. J. Chem. Phys. 2008, 128, 084308. (10) Dayou, F.; Larre´garay, P.; Bonnet, L.; Rayez, J. C.; Arenas, P. N.; Gonza´lez-Lezana, T. J. Chem. Phys. 2008, 128, 174307. (11) Lugovoj, E.; Toennies, J. P.; Vilesov, A. J. Chem. Phys. 2000, 112, 8217. (12) Mu¨ller, S.; Krapf, S.; Koslowski, T.; Mudrich, M.; Stienkemeier, F. Phys. ReV. Lett. 2009, 102, 183401. (13) Krasnokutski, S. A.; Huisken, F. J. Phys. Chem. A 2010, 114, 7292. (14) Krasnokutski, S.; Rouille´, G.; Huisken, F. Chem. Phys. Lett. 2005, 406, 386. (15) Toennies, J. P.; Vilesov, A. F. Angew. Chem. 2004, 43, 2622. (16) Hartmann, M.; Miller, R. E.; Toennies, J. P.; Vilesov, A. F. Science 1996, 272, 1631. (17) Huisken, F.; Stemmler, M. J. Chem. Phys. 1993, 98, 7680. (18) Mozhayskiy, V.; Slipchenko, M. N.; Adamchuk, V. K.; Vilesov, A. F. J. Chem. Phys. 2007, 127, 094701. (19) Agrawal, P. M.; Raff, L. M.; Hagan, M. T.; Komanduri, R. J. Chem. Phys. 2006, 124, 134306. (20) Ruchti, T.; Fo¨rde, K.; Callicoatt, B. E.; Ludwigs, H.; Janda, K. C. J. Chem. Phys. 1998, 109, 10679. (21) Zhou, M. F.; Chen, M. H. Chem. Phys. Lett. 2001, 349, 64.

JP110323N