Reactions of C+(2Pu) with CO(1Σ+) from thermal ...

0 downloads 0 Views 315KB Size Report
a quadruple mass spectrometer and finally counted by stan- dard technique. ..... 18 W. A. Lathan, W. J. Hehre, and J. A. Pople, J. Am. Chem. Soc. 93, 808. 1971.
Reactions of C+(2Pu) with CO(1Σ+) from thermal energies to 30 eV Wenyun Lu, Paolo Tosi, Mauro Filippi, and Davide Bassi Citation: J. Chem. Phys. 112, 1330 (2000); doi: 10.1063/1.480686 View online: http://dx.doi.org/10.1063/1.480686 View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v112/i3 Published by the American Institute of Physics.

Additional information on J. Chem. Phys. Journal Homepage: http://jcp.aip.org/ Journal Information: http://jcp.aip.org/about/about_the_journal Top downloads: http://jcp.aip.org/features/most_downloaded Information for Authors: http://jcp.aip.org/authors

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

JOURNAL OF CHEMICAL PHYSICS

VOLUME 112, NUMBER 3

15 JANUARY 2000

Reactions of Cⴙ „ 2 P u … with CO„ 1 ⌺ ⴙ … from thermal energies to 30 eV Wenyun Lu, Paolo Tosi,a) Mauro Filippi, and Davide Bassi INFM and Dipartimento di Fisica, Universita` degli Studi di Trento, I-38050 Povo, Italy

共Received 8 September 1999; accepted 22 October 1999兲 ⫹ The endoergic reactions of C⫹ ( 2 P u ) with CO( 1 ⌺ ⫹ ) producing CO⫹ ⫹C, C⫹ 2 ⫹O, and O ⫹C2 have been studied in a guided-ion beam apparatus. For each reaction channel, we have measured the kinetic energy dependence of the integral cross section and the reaction threshold. Analysis of the 0 reaction cross sections yield the heats of formation ⌬ f H 0 (C⫹ 2 )⫽19.79⫾0.16 and ⌬ f H (C2 ) ⫽8.37⫾0.16 eV. In addition the dissociation energy of C2 is estimated to be D 0 共C–C兲⫽6.37⫾0.16 eV. A simple state correlation diagram is used to rationalize experimental observations and to discuss reaction dynamics. © 2000 American Institute of Physics. 关S0021-9606共00兲02603-9兴

Oxygen ions O⫹ , which were not observed in previous work, have been found to be produced in reaction 共4兲, although the cross section is very small. Finally, measurements of onsets for reactions 共3兲 and 共4兲, allows us to investigate the thermochemistry of C⫹ 2 and C2 .

I. INTRODUCTION

Chemistry involving C atoms and CO molecules is important in combustion systems and in astrophysics, in outflows of carbon stars and within the expanding ejecta of type II supernova.1 At thermal energies, ground state C⫹ ( 2 P u ) ions react with CO only in an isotope exchange reaction:2–6 13 ⫹

C ⫹ 12CO→ 12C⫹ ⫹ 13CO.

II. EXPERIMENT

共1兲

Experiments were performed in a modified version of the ion-molecule reaction mass spectrometer which has been described before.8,9 Briefly, C⫹ ions produced by electron bombardment of CO are extracted from the differentially pumped ion source, mass selected by a magnet mass spectrometer, and finally injected into a radio-frequency octopole ion guide10,11 which is surrounded by the reaction cell. The collision energy is varied by changing the octopole DC potential. The axial energy distribution of the ion beam is determined by using the octopole as a retarding field energy analyzer. Typical ion energy distributions have a full width at half-maximum 共FWHM兲 of about 0.7 eV in the laboratory reference frame. The collision energy in the laboratory reference frame (E lab) is converted to that in the center-of-mass frame (E cm) by using E cm⫽E lab m/(M ⫹m), where m and M represents the mass of neutral and ionic reactants, respectively. The CO reactant gas is introduced in the reaction cell at pressures below 5⫻10⫺5 mbar to avoid multiple collisions. Reactant and product ions are collected and guided to a quadruple mass spectrometer and finally counted by standard technique. One major concern about the C⫹ reactants produced by electron impact is that part of them might be electronically excited. For the data presented here, we used a 26.0 eV electron beam. Since the appearance energy of the first excited C⫹ ( 4 P u ) from ground state CO is 27.69 eV,12 we expect the large majority of C⫹ ions to be in the ground electronic state 2 Pu . The absolute integral cross section ␴ was determined by using the expression ␴ ⫽aI p /I s , where I p is the intensity of the product ions and I s that of the primary ion beam. The constant a is determined by the well known reaction Ar⫹ ⫹CO,13 by keeping the reaction cell at the same pressure. Absolute cross sections thus determined may suffer

All other reaction channels are endoergic, thus requiring additional energy to proceed. Given sufficient collision energy, CO⫹ and C⫹ 2 ions can be produced and their kinetic-energy distributions have been measured in the incident ion kinetic energy range 13–30 eV 共laboratory兲.7 For CO⫹ , strong evidence exists that this product ion is formed by both charge transfer and atom exchange processes. In this article, we report our own investigation of this relatively simple tri-atomic reaction system, in the energy range from thermal up to 30 eV 共center of mass兲. Since previous experiments,7 making use of a simple beam-cell configuration, suffered by uncompleted collection of the product ions, we used the octopole ion-guide technique, which provides excellent collection of the product ions. Main goals of the present work are to measure both the energy dependence of the integral cross sections and reaction thresholds for the different products. In addition, electronic correlation schemes are used for understanding the reaction dynamics. The reaction channels considered in this study are the following: C⫹ ⫹CO→CO⫹ ⫹C, ⌬H 00 ⫽2.75 eV,

共2兲

0 C⫹ ⫹CO→C⫹ 2 ⫹O, ⌬H 0 ⫽4.90 eV,

共3兲

C⫹ ⫹CO→O⫹ ⫹C2 , ⌬H 00 ⬇7.23 eV.

共4兲

CO⫹ ions in reaction 共2兲 could be formed either in the charge transfer 共i.e., 12C⫹ ⫹CO→CO⫹ ⫹ 12C兲 or in the atom exchange reaction 共i.e., 12C⫹ ⫹CO→12CO⫹ ⫹C兲. Our preliminary measurements of the respective production of 12 CO⫹ and 13CO⫹ in reaction 共2兲 give a rough indication of the relative weight for each of these two processes. a兲

Electronic mail: [email protected]

0021-9606/2000/112(3)/1330/5/$17.00

1330

© 2000 American Institute of Physics

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Reactions of C⫹ with CO

J. Chem. Phys., Vol. 112, No. 3, 15 January 2000

FIG. 2. Energy dependencies of the signals of

FIG. 1. 共a兲 Cross section (Å 2 ) for CO⫹ production; 共b兲 expanded view of the threshold region. Open circles: experimental data; dashed line: model cross section with parameters ␴ 0 ⫽0.168, E 0 ⫽2.71, and n⫽2.99; solid line: the model cross section convoluted with the energy distribution of ionic and neutral reactants.

quite large errors. We estimated the error to be about 30%, whereas uncertainties for relative cross sections are ⫾5%. III. RESULTS AND DISCUSSION A. Production COⴙ ⴙC

Our data for the production of 12CO⫹ as a function of collision energy are depicted in Fig. 1共a兲. The energy dependence of the cross section is consistent with an endothermic reaction. The CO⫹ signal appears at about 2–3 eV, and peaks at about 8–9 eV. The cross section at the maximum is about 1.3 Å 2 . To determine the reaction threshold, E 0 , we used the usual function ␴ (E)⫽ ␴ 0 (E⫺E 0 ) n E ⫺1 for analyzing the data in the threshold region.14 Here E 0 , ␴ 0 , and n are adjustable parameters. After the appropriate convolution over the experimental energy distribution of both ionic and neutral reactants, the calculated cross sections are compared with the experimental data and parameters are optimized by an iterative fitting procedure. In Fig. 1共b兲 we plot the experimental data as well as the fitting results in the threshold region. For the threshold we found a value of 2.71⫾0.10, that can be compared with the known thermochemistry of reaction 共2兲. Given ⌬ f H 0 (C⫹ )⫽18.63 eV, ⌬ f H 0 (CO)⫽ ⫺1.18 eV, ⌬ f H 0 共CO⫹ )⫽12.83 eV, and ⌬ f H 0 共C兲⫽7.37 eV, the endothermicity can be easily calculated to be 2.75 eV. The good agreement between these two values corroborates the whole procedure, and indicates that reaction 共2兲 proceeds without any barrier. CO⫹ can be produced by both charge transfer reaction and atom exchange process. In their experiment, Maher et al.7 were able to separate these two channels by measuring

CO⫹ and

12

1331

CO⫹ .

13

the kinetic-energy distribution of the product ions. They found that slow CO⫹ ions are formed in a charge transfer reaction, while fast ions result from an ion-atom interchange process. It is interesting here to see which process is dominant. This could be easily done by using isotopically pure reactants, such as 12C⫹ ⫹ 13CO 共or 13C⫹ ⫹ 12CO兲, to see in which proportion the charge transfer products 13CO⫹ and the particle-exchange products 12CO⫹ are formed. While such experiment is planned in the future, at the present we use the natural mixture of 12CO/ 13CO⫽98.9/1.1. With the 12C⫹ and 12CO/ 13CO reactants, 12CO⫹ could result from both charge transfer and atomic exchange process, while 13CO⫹ should be formed solely in the charge transfer reaction. Thus a comparison of the profiles of these two products gives an indication of the relative significance of the two processes: if the two profiles are similar 共taking into account the isotopic ratio兲, then CO⫹ ions result mainly from the charge transfer reaction. On the other hand, if large differences between the two profiles exist, then the atom exchange reaction is expected to play an important role. The comparison is depicted in Fig. 2, where the count rate 共CPS兲 for 12O⫹ and 13CO⫹ as a function of collision energy are presented. It should be mentioned that the count rate for 13CO⫹ is quite small and could be affected by small amount of impurities, thus the comparison here should be considered preliminary. Nevertheless, it appears that the profiles of the two species are quite similar, and the ratio of 12CO⫹ / 13CO⫹ is close to the natural abundance of neutral reactants. Altogether these findings indicate that the charge transfer is the dominant process. We expect that future experiments using isotopic pure reactants can lead to a defined conclusion. B. Production of C2ⴙ ⴙO

Maher et al.7 observed C⫹ 2 produced in reaction 共3兲 only in the energy range 13–30 eV 共in the laboratory frame, corresponding to 9.1–21 eV in the center of mass兲. We observed C⫹ 2 in a broader energy range. The cross section of as a function of collision energy is depicted in Fig. 3共a兲. C⫹ 2 The reaction onset is located at about 5 eV, while the cross section at the maximum of 0.63 Å 2 is at 10 eV. Towards even higher energies, the signal drops down. We try to correlate the falloff with the dissociation either of CO or C⫹ 2 . However, the process C⫹ ⫹CO→C⫹ ⫹C⫹O requires 11.11

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

1332

Lu et al.

J. Chem. Phys., Vol. 112, No. 3, 15 January 2000

FIG. 4. Cross section for O⫹ production. The arrow indicates the onset of the reaction C⫹ ⫹CO→O⫹ ⫹C⫹C. The solid line is the model cross section convoluted with the energy distribution of ionic and neutral reactants. Parameters: ␴ 0 ⫽0.028, E 0 ⫽7.10, and n⫽1.25.

FIG. 3. a兲 Cross section (Å 2 ) for C⫹ 2 production. The arrow indicates the onset of the reaction C⫹ ⫹CO→C⫹ ⫹C⫹O; 共b兲 expanded view of the threshold region. Open circle: experimental data; dashed line: model cross section with parameters ␴ 0 ⫽0.294, E 0 ⫽4.90, and n⫽2.32; solid line: the model cross section convoluted with the energy distribution of ionic and neutral reactants.

eV, while the dissociation of C⫹ 2 probably requires even a larger value, since only a fraction of reactant energy can be disposed in product vibration. Thus the cross section decline appears to be not related to these processes and its rationale is still an open question. The C⫹ 2 ion is of considerable importance in the area of combustion chemistry and astrophysics.15,16 A number of theoretical17–26 and experimental27–36 investigations are available about its thermodynamic properties, such as the heat of formation and the dissociation energy. In our recent paper9 on the reactive collisions between CO⫹ and CO, we have discussed the thermochemistry of C⫹ 2 , and estimated to be 19.8⫾0.2 eV and the the heat of formation of C⫹ 2 dissociation energy D 0 (C⫹ – C)⫽6.2⫾0.2 eV. In the present work, the appearance energy of C⫹ 2 in reaction 共3兲 can also be used to derive its thermodynamic properties. By using the fitting procedure described in the previous section, we found the threshold of reaction 共3兲 to be E 0 ⫽4.90⫾0.16 eV. The fitting results as well as the experimental data are depicted in Fig. 3共b兲. Given ⌬ f H 0 共O兲⫽2.56 eV, and by 0 0 ⫹ 0 using E 0 ⫽⌬ f H 0 (C⫹ 2 )⫹⌬ f H (O)⫺⌬ f H (C )⫺⌬ f H 共CO兲, the heat of formation is easily calculated to be ⌬H 0f (C⫹ 2 ) ⫽19.79⫾0.16 eV. This value is in agreement with our recent estimate, 19.8⫾0.2 eV and thus confirms also our re⫹ evaluation of the dissociation energy of C⫹ 2 , D 0 (C – C) 9 ⫽6.2⫾0.2 eV. C. Production of Oⴙ ⴙC2

In Fig. 4 we plot the cross section of O⫹ as a function of collision energy up to 16 eV. The signal at even higher en-

ergies is close to the noise level. The cross section of O⫹ is found to be quite small and that’s why this product was not observed previously. Even at the maximum, the cross section is only 0.0075 Å 2 . The O⫹ signal starts to appear at about 7 eV which indicates that the onset is that of reaction 共4兲 from known thermochemistry of C2 . As a good practice, here we also fit the experimental data with the usual model function convoluted with the ion and neutral distributions. Results are shown in Fig. 4. For reaction 共4兲, we obtained E 0 ⫽7.10⫾0.16 eV. By using E 0 ⫽⌬ f H 0 (C2 )⫹⌬ f H 0 (O⫹ ) ⫺⌬ f H 0 (C⫹ )⫺⌬ f H 0 共CO兲, the heat of formation of C2 can be derived to be 8.37⫾0.16 eV. This value appears a little smaller than the recent determinations 8.45⫾0.02,37 and 8.47 ⫾0.08 eV.38 By considering that ⌬ f H 0 共C兲⫽7.37 eV, our estimate for the dissociation energy of C2 is D 0 共C–C兲⫽6.37 ⫾0.16 eV. This value can be compared with that of 6.30 ⫾0.02 eV, obtained by spectroscopic methods.37 The signal of O⫹ reaches a maximum at about 10 eV, then it drop towards even higher energy. We have no evidence for the production of O⫹ ⫹C⫹C. The reason may be that the dissociative charge transfer reaction favors instead C⫹ ⫹O⫹C, since in the CO⫹ ion the positive charge is on the carbon atom.39 The failure to observe the products O⫹ ⫹C⫹C may also indicate that the reaction proceeds through a 关C•••C•••O兴⫹ intermediate, and not 关C•••O•••C兴⫹ 共the latter one was thought to be responsible for the isotopic exchange reaction40兲. D. Electronic correlation diagram

Previous investigations on similar tri-atomic reaction systems, such as N⫹ ⫹CO,41 C⫹ ⫹N2 , 42 and C⫹ ⫹O2 , 43show that simple electronic correlation schemes44 can be used to help the understanding of the reaction dynamics. Unfortunately, only very limited information on the potential energy surfaces of the intermediate CCO⫹ are available. The ground state is calculated to be a linear 2 ⌸ state,40,45–47 located 2.60 eV below the ground state C⫹ ⫹CO. 40 The ground state structure clearly represents an adduct of C⫹ and an only slightly perturbed CO.40 Regarding the excited states, very little is known. As a close approximation, here the relative

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

Reactions of C⫹ with CO

J. Chem. Phys., Vol. 112, No. 3, 15 January 2000

1333

plain the very small cross section observed for O⫹ production. IV. SUMMARY

FIG. 5. Electronic correlation diagram for the C⫹ ⫹CO reaction in C⬁ v (Cs ) symmetry. The sequence of levels of the intermediate CCO⫹ are considered similar to those of CCN and CNN⫹ .

energies of the excited states of CCO⫹ are considered to be similar to the isoelectronic CCN48 and CNN⫹ . 42 To be consistent with the theoretical calculation of Cao et al.,49 the absolute energies of the excited states are shifted down by about 1 eV, so that the 2 ⌺ ⫺ state lies at 16.43 eV. The resulted correlation diagram 共considering only the doublet surfaces兲 depicted in Fig. 5 should be considered only qualitative. Ground state reactants C⫹ ( 2 P u )⫹CO( 1 ⌺ ⫹ ) lead to 2 ⌺ ⫹ and 2 ⌸ terms. The next higher energy asymptote is given by the charge transfer products C( 3 P g )⫹CO⫹ (X 2 ⌺ ⫹ ), and it leads to 2 ⌺ ⫺ and 2 ⌸ terms. The following higher asymptote is C( 1 D g )⫹CO⫹ (X 2 ⌺ ⫹ ) and results in 2 ⌺ ⫹ , 2 ⌸, 2 ⌬ terms. The nearly degenerate C( 3 P g )⫹CO⫹ (A 2 ⌸) and C⫹ ( 4 P u )⫹CO( 1 ⌺ ⫹ ) levels lie even higher. For the atom transfer products, the lowest level corresponds to ground 4 ⫺ 3 2 ⫹ state C⫹ and 2 ⌸ 2 (X ⌺ g )⫹O( P g ), which lead to ⌺ ⫹ 2 3 terms. The first excited C2 (a ⌸ u ) with O( P g ) lies about 0.7 eV higher and they lead to 2 ⌺ ⫹ , 2 ⌺ ⫺ , 2 ⌸, and 2 ⌬ terms. The next higher asymptote is given by C2 ( 1 ⌺ ⫹ g ) ⫹O⫹ ( 4 S u ), which leads to 2 ⌺ ⫺ term. From the diagram, it is clear that in a collinear configuration ground state reactants C⫹ ( 2 P u )⫹CO( 1 ⌺ ⫹ ) correlate 4 ⫺ 3 directly with ground state C⫹ 2 (X ⌺ g )⫹O( P g ) products, 2 ⫹ 2 via either ⌺ or ⌸ surfaces. This fact can explain the relatively efficient formation of C⫹ 2 and is consistent with the experimental value of the reaction threshold, that suggests ⫹ the formation of ground state C⫹ 2 ions. The production of C2 in the first excited state is also possible, through a transition at the avoided crossing between 2 ⌺ ⫹ (A ⬘ ) and 2 ⌬(A ⬘ ), under Cs symmetry, since the latter surface correlates directly 2 3 with C⫹ 2 (a ⌸ u )⫹O( P g ). Regarding the charge transfer products, the reaction path goes through some avoided crossing under Cs symmetry. Reactants start on the 2 ⌺ ⫹ (A ⬘ ) surface and cross the 2 ⌬(A ⬘ ) surface under Cs symmetry. The 2 ⌬(A ⬘ ) surface then crosses the 2 ⌸(A ⬘ ) surface, which correlates with C( 3 P g ) ⫹CO⫹ (X 2 ⌺ ⫹ ). This analysis shows that the charge transfer reaction proceeds through a bending 关C–C–O兴⫹ intermediate. From the diagram, it is also clear that ground state reac⫹ 4 tants do not correlate with C2 ( 1 ⌺ ⫹ g )⫹O ( S u ). This can ex-

We used a guided-ion beam tandem mass spectrometer to investigate the endoergic reactions of C⫹ ( 2 P u ) with CO ( 1 ⌺ ⫹ ). For the reaction channels producing CO⫹ ⫹C, ⫹ C⫹ 2 ⫹O, and O ⫹C2 we have measured the kinetic energy dependence of the integral cross section and the reaction threshold. Analysis of the reaction cross sections in the threshold region yield the heats of formation ⌬ f H 0 0 共C⫹ 2 )⫽19.79⫾0.16 and ⌬ f H (C2 )⫽8.37⫾0.16 eV. In addition the dissociation energy of C2 is estimated to be D 0 共C–C兲⫽6.37⫾0.16 eV. A simple state correlation diagram is used to rationalize experimental observations and to discuss reaction dynamics. Ground state reactants C⫹ ( 2 P u )⫹CO( 1 ⌺ ⫹ ) correlate directly with ground state 4 ⫺ 3 2 ⫹ and 2 ⌸ surC⫹ 2 (X ⌺ g )⫹O( P g ) products, both on ⌺ faces. The charge transfer reaction producing CO⫹ involves a complicated sequence of avoided crossings, through a bending 关C–C–O兴⫹ intermediate. ACKNOWLEDGMENT

The authors thank F. Tonelli for his assistance in the experiment. D. D. Clayton, W. Liu, and A. Dalgarno, Science 283, 1290 共1999兲. D. Smith and N. G. Adams, Int. J. Mass Spectrom. Ion Processes 23, 123 共1977兲. 3 M. Tichy, A. B. Bakshit, D. G. Lister, and N. D. Twiddy, Int. J. Mass Spectrom. Ion Processes 29, 231 共1979兲. 4 D. K. Bohme et al., J. Chem. Phys. 87, 6934 共1987兲. 5 W. D. Watson, V. G. Anicich, and W. T. Huntress, Jr., Astrophys. J. 205, L165 共1976兲. 6 D. Smith and N. G. Adams, Astrophys. J. 242, 424 共1980兲. 7 K. E. Maher, G. J. Mcclure, and J. J. Leventhal, J. Chem. Phys. 55, 5549 共1971兲. 8 D. Bassi, P. Tosi, and R. Schlo¨gl, J. Vac. Sci. Technol. A 16, 114 共1998兲. 9 W. Lu, P. Tosi, and D. Bassi, J. Chem. Phys. 111, 8852 共1999兲. 10 P. Tosi, G. Fontana, S. Longano, and D. Bassi, Int. J. Mass Spectrom. Ion Processes 93, 95 共1989兲. 11 P. Tosi, Chem. Rev. 92, 1667 共1992兲. 12 A. A. Radzig and B. M. Smirnov, Reference Data on Atoms, Molecules and Ions 共Springer-Verlag, Berlin, 1985兲. 13 G. D. Flesch, S. Nourbakhsh, and C. Y. Ng, J. Chem. Phys. 95, 3381 共1991兲. 14 P. B. Armentrout, in Advances in Gas Phase Ion Chemistry, edited by N. G. Adams and L. M. Babcock 共JAI, Greenwich, 1992兲, p. 83. 15 W. Weltner, Jr. and R. J. Van Zee, Chem. Rev. 89, 1713 共1989兲. 16 A. Van Orden and R. J. Saykally, Chem. Rev. 98, 2313 共1998兲. 17 G. Verhaegen, J. Chem. Phys. 49, 4696 共1968兲. 18 W. A. Lathan, W. J. Hehre, and J. A. Pople, J. Am. Chem. Soc. 93, 808 共1971兲. 19 P. E. Cade et al., At. Data Nucl. Data Tables 13, 339 共1974兲. 20 P. Carsky, J. Kuhn, and R. Zahradnik, J. Mol. Spectrosc. 55, 120 共1975兲. 21 C. Petrongolo, P. J. Bruna, S. D. Peyerimhoff, and R. J. Buenker, J. Chem. Phys. 74, 4594 共1981兲. 22 P. Rosmus, H. J. Werner, E. A. Reinsch, and M. Larsson, J. Electron Spectrosc. Relat. Phenom. 41, 289 共1986兲. 23 W. P. Kraemer and B. O. Roos, Chem. Phys. 118, 345 共1987兲. 24 P. J. Bruna and J. S. Wright, J. Phys. Chem. 96, 1630 共1992兲. 25 A. D. Mclean, B. Liu, and G. S. Chandler, J. Chem. Phys. 97, 8459 共1992兲. 26 R. Merkle, A. Savin, and H. Preuss, J. Chem. Phys. 97, 9216 共1992兲. 27 J. Drowart, R. P. Burns, G. De Maria, and M. G. Inghram, J. Chem. Phys. 31, 1131 共1959兲. 1 2

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

1334

W. H. Dibeler and S. K. Liston, J. Chem. Phys. 47, 4548 共1967兲. D. L. Cocke and K. A. Gingerich, J. Chem. Phys. 60, 1958 共1974兲. 30 S. K. Gupta and K. A. Gingerich, J. Chem. Phys. 71, 3072 共1979兲. 31 E. A. Rohlfing, D. M. Cox, and A. Kaldor, J. Chem. Phys. 81, 3322 共1984兲. 32 P. Plessis and P. Marmet, Int. J. Mass Spectrom. Ion Processes 70, 23 共1986兲. 33 J. P. Meier and M. Ro¨sslein, J. Chem. Phys. 88, 4614 共1988兲. 34 A. N. Pargellis, J. Chem. Phys. 93, 2099 共1990兲. 35 M. B. Sowa, P. A. Hintz, and S. L. Anderson, J. Chem. Phys. 95, 4719 共1991兲. 36 C. J. Reid, J. A. Ballantine, S. R. Andrews, and F. M. Harris, Chem. Phys. 190, 113 共1995兲. 37 R. S. Urdahl, Y. Bao, and W. M. Jackson, Chem. Phys. Lett. 178, 425 共1991兲. 38 K. A. Gingerich, H. C. Finkbeiner, and R. W. Schmude, Jr., J. Am. Chem. Soc. 116, 3884 共1994兲. 28 29

Lu et al.

J. Chem. Phys., Vol. 112, No. 3, 15 January 2000 39

H. Lavendy, J. M. Robbe, and J. P. Flament, Chem. Phys. Lett. 205, 456 共1993兲. 40 L. L. Lohr, J. Chem. Phys. 108, 8012 共1998兲. 41 W. Frobin, Ch. Schlier, K. Strein, and E. Teloy, J. Chem. Phys. 67, 5505 共1977兲. 42 J. D. Burley, L. S. Sunderlin, and P. B. Armentrout, J. Chem. Phys. 94, 1939 共1991兲. 43 J. D. Burley and P. B. Armentrout, Int. J. Mass Spectrom. Ion Processes 84, 157 共1988兲. 44 K. E. Shuler, J. Chem. Phys. 21, 624 共1953兲. 45 N. N. Haese and R. C. Woods, Chem. Phys. Lett. 91, 190 共1982兲. 46 R. G. A. R. Maclagan and P. Sudkeaw, J. Chem. Soc., Faraday Trans. 89, 3325 共1993兲. 47 R. Flammang et al., J. Phys. Chem. 98, 4814 共1994兲. 48 G. Herzberg, Molecular Spectra and Molecular Structure: III. Electronic Spectra and Electronic Structure of Polyatomic Molecules 共Van Nostrand Reinhold, New York, 1966兲. 49 Z. X. Cao and A. M. Tian, J. Mol. Struct.: THEOCHEM 334, 45 共1995兲.

Downloaded 11 Oct 2012 to 128.112.70.45. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions