Reaction of FeCH2 D2: Probing the [FeCH4] Potential Energy

---

J. Phys. Chem. 1996, 100, 111-119

111

Reaction of FeCH2+ + D2: Probing the [FeCH4]+ Potential Energy Surface Chris L. Haynes, Yu-Min Chen,† and P. B. Armentrout* Department of Chemistry, UniVersity of Utah, Salt Lake City, Utah 84112 ReceiVed: July 26, 1995; In Final Form: September 29, 1995X

A guided-ion beam mass spectrometer is used to study the reactions FeCH2+ + D2 and Fe+(6D,4F) + CD4, thereby experimentally probing the [FeCH4]+ potential energy surface (PES). The results obtained are compared to recent theoretical results. Experiment and theory agree that dehydrogenation of methane by Fe+ is hindered by a tight, four-center transition state complex. The major discrepancy observed between experiment and theory is in the height of this barrier, which theory predicts to be 75 kJ/mol (92 kJ/mol after correction for zero-point energies) versus our experimental result of 41 ( 6 kJ/mol. Our results can be interpreted by using phase space theory to help understand how various features on the PES control branching ratios among the various channels observed, including extensive hydrogen scrambling in the FeCH2+ + D2 reaction. Lastly, results for the [FeCH4]+ PES are compared to those for the [CoCH4]+ PES obtained in a previous study in order to help assess the effect of potential energy surfaces of different spin in the former case.

Introduction Considerable research has been done to help understand how transition metals activate C-C and C-H bonds of saturated alkanes, as well as the periodic trends in this chemistry.1,2 Recently, we performed a set of experiments that provide some of the most detailed experimental data for the interaction of a transition metal with methane.3 A fairly complete potential energy surface (PES) for the [CoCH4]+ system was obtained by carefully examining the reaction of Co+ with methane, as well as a reverse reaction, CoCH2+ + D2, coupled with phase space calculations. In agreement with theoretical calculations,4 our results showed that the dehydrogenation of methane by Co+ is hindered by a tight, four-centered transition state complex. (Detailed discussions of the possible mechanisms for this dehydrogenation reaction can be found elsewhere.2,5,6-8) However, our experimental value for this barrier height is 34 ( 8 kJ/mol while the theoretical value is 96-109 kJ/mol. In their review of the periodic trends in the reactions of first row transition metal ions with alkanes, Armentrout and Beauchamp8 noted that reactions 1-3 are observed in the methane system with branching ratios that vary across the periodic table.

M+ + CH4 f MCH2+ + H2 f MH+ + CH3

(1)

reactions 2 and 3, respectively. The absence of reaction 1 is puzzling in light of our recent observation of this process for M ) Co,3 especially given that Fe+(4F) and Co+(3F) are generally observed to display similar reactivity as seen in the reactions with D2,10 C2H6,9 and CH3X (X ) Cl, Br, I).11,12 This similarity is because the Fe+(4F) and Co+(3F) states have similar electron configurations, 3d7 and 3d8, respectively. Because of this disparity, we reexamine the Fe+ + CH4 reaction here. The dehydrogenation process can also be studied further by examining its reverse, FeCH2+ + H2. Previously, Jacobson and Freiser5 found that FeCH2+ was unreactive with H2 at thermal energies even though formation of Fe+ + CH4 is exothermic by 1.18 ( 0.05 eV. Here, we study what happens at elevated kinetic energies. Combined, the Fe+ + CH4 and FeCH2+ + H2 systems provide a fairly detailed look at the [FeCH4]+ PES, which can be compared to theoretical calculations on this system.13 (In all cases, energies cited from this theoretical work are those calculated at a fairly high level of theory, MR-SDCI CASSCF+DC, but do not contain corrections for zero-point energies.) Of particular interest will be a comparison between the iron and cobalt systems, which should help elucidate the state-specific character of the PES because both sextet and quartet surfaces are involved in the iron system while only triplet surfaces (analogous to the quartet surfaces of iron) are important in the cobalt system.

(2) Experimental Section

f MCH3 + H +

(3)

For early metals, reaction 1 dominates at low energies and reaction 2 at high energies. For late metals, such as Co+, reaction 1 is a minor process compared to reaction 2, even though it is energetically more favorable than reaction 2. In previous work on the Fe+ + CH4 system, Schultz et al.9 formed Fe+ using both surface ionization and drift cell techniques which allowed results for different electronic states of iron to be extracted. They showed that the 4F first excited state of iron is much more reactive than the 6D ground state. However, the only products observed were FeH+ and FeCH3+, formed in † Present address: Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA 02139. X Abstract published in AdVance ACS Abstracts, December 1, 1995.

0022-3654/96/20100-0111$12.00/0

The experiments were performed on a guided-ion beam tandem mass spectrometer14,15 equipped with several ion sources described below. The ions generated are extracted from the source, accelerated, and passed through a magnetic sector for mass analysis. The mass-selected ions are decelerated to the desired kinetic energy where they are focused into an octopole ion trap. This device uses radio-frequency electric fields to trap the ions in the radial direction and ensure complete collection of reactant and product ions.16 The octopole passes through a gas cell that contains the neutral reaction partner at a pressure sufficiently low that multiple ion-molecule collisions are improbable. It was verified that the results presented here exhibit no dependence on pressure and thus correspond to single ion-molecule collisions. The unreacted parent and product ions drift to the end of the octopole where they are extracted, passed © 1996 American Chemical Society

112 J. Phys. Chem., Vol. 100, No. 1, 1996 through a quadrupole mass filter for mass analysis, and detected with a secondary electron scintillation ion detector using standard pulse counting techniques. The raw ion intensities are converted to cross sections, as described previously.14 We estimate our absolute cross sections to be accurate to (20%. Laboratory (lab) energies are converted to energies in the center-of-mass (CM) frame by using the conversion ECM ) ElabM/(M + m), where m and M are the ion and neutral masses, respectively. The absolute energy scale and corresponding full width at half-maximum (fwhm) of the ion-beam kinetic energy distribution are determined by using the octopole as a retarding energy analyzer as described previously.14 The absolute uncertainty in the energy scale is (0.05 eV (lab). The energy distributions are nearly Gaussian and have a typical fwhm of 0.2-0.6 eV (lab) for the flow tube sources and 0.6-0.7 eV (lab) for the surface ionization source. CD4 gas was obtained from Cambridge Isotope Laboratories with 99% purity. The CD4 reactant gas was subjected to multiple freeze-pump-thaw cycles to remove any air impurities. Removal of residual oxygen is particularly important in this study because FeO+ has the same mass as FeCD2+, the reactions forming these ions have similar energy dependencies, and the latter product has a much smaller cross section than the former product ion, especially for Fe+(6D). Careful attention was paid to this problem, although the possibility that there may be a small contribution of FeO+ in the FeCD2+ cross sections cannot be completely eliminated. Ion Sources. FeCH2+ and Fe+ ions are made in our flow tube ion source, described in detail previously.17 Fe+ is made using a direct current discharge source18 consisting of an iron cathode held at high negative voltage (1.5-3 kV) over which a flow of approximately 90% He and 10% Ar passes. Ar+ ions created in the discharge are accelerated toward the iron cathode, sputtering off ionic and neutral metal atoms. FeCH2+ ions are formed by allowing Fe+ to react with ethylene oxide,19 which is added to the flow gases about 60 cm downstream of the source at low pressures (104 thermalizing collisions as they traverse the remaining 40 cm of the flow tube and therefore are expected to be at room temperature. Previous work on a number of systems is consistent with the production of thermalized ions under similar conditions.3,17,18,20,21 Reactant ions are extracted from the flow tube and gently focused through a 9.5 cm long differentially pumped region before entering the rest of the instrument described above. Before the reaction of FeCH2+ + D2 was run, a high-energy collision-induced dissociation (CID) spectrum was taken of FeCH2+ with Xe to ensure that Fe+ was the only product observed and that it had a threshold consistent with the previously determined thermochemistry of FeCH2+.22,23 This provides evidence that the FeCH2+ beams have no appreciable impurities or internal excitation. The dc discharge/flow tube (DC/FT) high-pressure environment creates Fe+ ions primarily in their 6D ground state, as determined in a previous study.24 There, we characterized this beam as containing about 97% 6D and 3% 4F states. Fe+ is also made by using a microwave discharge/flow tube (MW/ FT) source. In the flow tube sources, the carrier gas is pure He that is passed through a liquid nitrogen cooled molecular sieve trap to remove impurities. The gas is introduced at a flow rate of about 7000 sccm into a microwave discharge mounted at the end of the flow tube. This discharge creates He+ and He* metastable states that ionize Fe(CO)5 (freeze-pumped-

Haynes et al. thawed to remove noncondensable impurities) which is introduced 60 cm downstream of the discharge cavity at a pressure of 2 eV (Figure 3), even though the latter reaction is thermodynamically favored by over 200 kJ/mol. This result suggests that there is some type of restriction to reaction 13, which we would ordinarily attribute to a tight transition state in the exit channel for methane elimination. Such a restriction would mean that the simple bond cleavage reactions leading to iron-hydride and iron-methyl ions would be kinetically favored at higher energy, consistent with observation. However, the presence of a tight transition state would be at odds with the results of Musaev et al.,13 who find that the H-Fe+-CH3 intermediate is not a minimum on the potential energy surface at the highest levels of theory and that there is no energy barrier to methane elimination along the reaction coordinate. Therefore, we suggest that the restriction to reaction 13 is in the degrees of freedom other than the reaction coordinate, an idea that is explored more thoroughly by using phase space theory to model these results, as described below. Additional support for such a restriction comes from the observation of extensive isotope scrambling in this reaction (Figure 3). Without such a restriction, methane elimination from the D-Fe+-CH2D intermediate should be facile, and it is hard to imagine why isotope scrambling would occur. With such a restriction, the lifetime of the intermediate can be longer, thereby permitting rearrangement to compete with the thermodynamically preferred methane elimination channel. More detailed mechanisms for the isotope exchange have been discussed in detail previously for the analogous cobalt system.3 Briefly, they involve concerted interchange of H and D from the D-Fe+CH2D intermediate. Because a transition metal center is involved, such concerted mechanisms could be low-energy processes. The energy dependencies of the FeH+ and FeD+ cross sections are comparable (Figure 3) with FeD+ comprising 66 ( 3% of both products. The branching ratio for the methyl channels cannot be elucidated unambiguously because FeCH2D+ and FeCD2+ have the same nominal mass. If we presume that there is no FeCD2+ product present, as also indicated in the cobalt system,3 then the FeCH2D+ product comprises 83 ( 3% of the FeCH2D+ and FeCHD2+ products. The branching ratios obtained here are comparable to those obtained in the cobalt system, where CoD+ comprises 64% of the CoD+ and CoH+ products and CoCH2D+ comprises 82% of the CoCH2D+ and

J. Phys. Chem., Vol. 100, No. 1, 1996 117 CoCHD2+ products. If we presume that the FeCD2+ product cross section has a comparable magnitude to that for the FeCHD+ product, then FeCH2D+ comprises about 77% of the FeCH2D+ and FeCHD2+ products. Whether FeCD2+ is formed or not, the preferred iron-methyl ion product is FeCH2D+. On the basis of the PES shown in Figure 4, the threshold obtained for the FeCHD+ product in reaction 10 would be expected to equal the barrier height associated with TS1. This presumes that the this thermoneutral isotope exchange process occurs by forming 1, scrambling the hydrogen and deuterium atoms, and then re-forming reactants. Our analysis of the cross section for this product yields a threshold 0.6 ( 0.2 eV above this barrier height (Table 2). The reaction observed must correspond to a HD neutral product, because formation of FeCHD+ + H + D cannot occur until 4.514 eV.41 We believe that the elevated threshold is a result of competition with the thermodynamically favored formation of Fe+ + CH2D2. After formation of the D-Fe+-CH2D intermediate, elimination of CH2D2 is thermodynamically and kinetically favored compared with returning over the tight TS1. Thus, the intensity of the FeCHD+ + HD product is too small to observe at energies below 1 eV. This conclusion can be tested further by using phase space theory to model this competition, as discussed below. Phase Space Theory Calculations. The FeCH2+ + D2 reaction system can be elucidated further by using statistical phase space theory (PST) to model the results, although additional assumptions are required because of the tight transition state in the entrance channel. Phase space theory explicitly conserves energy and angular momentum while making standard statistical assumptions regarding the branching ratio among the various product channels. Phase space theory is intrinsically a loose transition state theory and therefore incapable of including the effects of a tight transition state in the entrance channel. Additional assumptions that permit treatment of such a case were originally formulated by Marcus42 and then developed further by Chesnavich and Bowers.43 This treatment assumes that the reaction is driven by translational energy, and hence we have dubbed the overall model translationally driven (TD) PST. Our formulation of this model is discussed thoroughly elsewhere.44 We have successfully used TD-PST to describe reactions of CoCH2+ + D2,3 O2+ with H2,44 and CoO+ + D2.45 Codes for PST are adapted from those of Bowers, Chesnavich, and others46 and modified to include the translationally driven assumptions. Table 3 lists the molecular constants for reactants and products, and Table 4 gives other parameters needed for the calculations.30-32,47-50 Our calculations consider only the lowest lying potential energy surface, because a model that includes excited electronic states would be prohibitively complicated (especially given that the energies of the excited states are unknown). Although this omission is certainly incorrect, the idea of exploring the application of PST in this case is to see whether a statistical model is capable of explaining the various features observed in the experimental results. Naturally, success in this endeavor is not proof that the reaction is statistically behaved (certainly not in all aspects, at all energies), but it does suggest or limit the likely possibilities. At low energies (below ∼1 eV), formation of Fe+ + CH2D2 and the return to reactants (including the hydrogen scrambling channels, FeCHD+ + HD and FeCD2+ + H2) are the only channels accessible. Using TD-PST, we can reproduce the energy dependence of the Fe+ product cross section in this threshold region with a barrier height of 0.40 ( 0.05 eV for the four-centered transition state (TS1) where D2 adds across the Fe+dCH2 π bond. This value agrees with our empirically

118 J. Phys. Chem., Vol. 100, No. 1, 1996

Haynes et al.

TABLE 4: Parameters Used in Phase Space Calculations

reaction channel FeCH2+(4B1) + D2(1Σg) Fe+(6D) + CH2D2(1A1) Fe+(4F) + CH2D2(1A1) FeCH2D+(5A′) + D(2Sg) FeCHD2+(5A′) + H(2Sg) FeD+(5∆) + CH2D(2B1) FeH+(5∆) + CHD2(2B1)

reduced polarizability mass of neutral symmetry no. of no.a surfacesb (amu) (Å3) 3.81 13.64 13.64 1.96 0.99 12.56 13.11

0.775c 2.56d 2.56d 0.67e 0.667f 2.22g 2.22g

4 2 2 1 1 2 2

4(4) 30(4) 28(4) 20(4) 20(4) 20(4) 20(4)

a Product of the symmetry numbers for the ionic and neutral species. Electronic degeneracy (assumed number of reactive surfaces). c Hirschfelder, J. O.; Curtiss, C. R.; Bird, R. B. Molecular Theory of Gases and Liquids; Wiley: New York, 1954; p 947. d R(CH2D2) assumed to equal R(CH4) given in: Rothe, R. W.; Bernstein, R. B. J. Chem. Phys. 1959, 31, 1619. e R(D) assumed to equal R(H). f Miller, M. T.; Bederson, B. AdV. At. Mol. Phys. 1977, 13, 1. g R(CH2D) and R(CHD2) are assumed to equal R(CH3), which is estimated by using the empirical method of: Miller, K. J.; Savchik, J. A. J. Am. Chem. Soc. 1979, 101, 7206. b

Figure 5. Comparison of experimental results for FeCH2+ + D2 with translationally driven phase space theory (TD-PST) calculations as a function of kinetic energy in the center-of-mass frame (lower x axis) and laboratory frame (upper x axis). Symbols show the experimental results, which are the same as in Figure 3, although only the sums of the cross sections for FeH+ and FeD+ (open triangles) and for FeCH2D+ and FeCHD2+ (open circles) are shown for clarity. The TD-PST theoretical calculations detailed in the text (after convolution over the experimental energy distributions) are shown as solid lines for Fe+, Fe+-methyls, and Fe+-hydrides and as a dashed line for FeCHD+.

determined threshold for reaction 13 within experimental error. Lower and higher barrier heights were explicitly tested and could not reproduce the Fe+ cross section. In order to reproduce the absolute magnitude of the Fe+ cross section in the threshold region, the rotational constant for TS1 was taken as B ) 0.40 cm-1, slightly lower than a value of 0.53 cm-1 calculated from the theoretical geometry of Musaev et al.13 (This result suggests a slightly different geometry for TS1, consistent with our finding that the experimental energy of TS1 also differs from the theoretical value; however, the absolute magnitudes of the TDPST calculations are not sufficiently reliable that such a conclusion is definitive. In this regard, it is worth noting that the geometry calculated for FeCH2+ by Musaev et al.13 differs from that calculated by Bauschlicher et al.30) After convoluting over the experimental energy distributions, the TD-PST result is shown along with the experimental data in Figure 5. The TD-PST results also demonstrate that a return to reactants (including hydrogen scrambling) is very inefficient. Indeed, the calculations show that the thermoneutral ligand

exchange reactions to form FeCHD+ + HD and FeCD2+ + H2 have cross sections that rise slowly from thresholds of 0.40 ( 0.05 eV but do not reach magnitudes of 10-3 Å2 until about 1 eV, consistent with experiment (Figure 3). This result demonstrates a competitive shift that delays the appearance of FeCHD+ to higher energies. Above about 1 eV, reproduction of the experimental data requires that the exit channel for loss of methane be carefully considered. If the transition state for this channel is treated as loose, i.e., parameters associated with Fe+ + CH2D2 are used, TD-PST predicts magnitudes for the Fe+-methyl, Fe+-hydride, and FeCHD+ cross sections that are too low by about a factor of 3. In this calculation, we assumed that the reaction of FeCH2+ + D2 forms Fe+(4F) + CH2D2, the spin-allowed pathway. If the production of Fe+ in its 6D state is assumed, TD-PST calculations deviate even more from our experimental results. Consideration of nonadiabatic effects would decrease the efficiency even further. We then performed TD-PST calculations that included a transition state for methane elimination from 1, TS2, set at an energy ranging from 0 to 1 eV below the FeCH2+ + D2 asymptotic limit. The best results are obtained when the energy is set to 0.8 eV below the FeCH2+ + D2 asymptotic limit (0.45 eV above Fe+(6D) + CH2D2). This calculation now reproduces all the data nicely (Figure 5). Note that this energy is comparable to the bond additivity estimates and theoretical calculations for the energy of intermediate 1, consistent with the theoretical result that there is no energetic barrier to methane elimination from 1.13 Whether TS2 is included in the calculation or not, our TDPST results find that FeD+ comprises 75% of the FeH+ and FeD+ cross sections, in reasonable agreement with the experimental value of 66%. The TD-PST results show that FeCH2D+ comprises 82% of the FeCH2D+ and FeCHD2+ cross sections, also consistent with the experimental value of 83% obtained if no FeCD2+ is formed. This good agreement is consistent with little or no contribution of the FeCD2+ product to the FeCH2D+ cross section. It also means that the observed predominance of the FeD+ + CH2D and FeCH2D+ + D channels over FeH+ + CHD2 and FeCHD2+ + H, respectively, is a statistical phenomenon rather than a dynamic one having to do with the initially formed intermediate. Conclusions In these experiments, we probe the potential energy surface (PES) for activation of methane by atomic Fe+ ions using guided-ion beam mass spectrometry. We are able to map the [FeCH4]+ PES in detail by looking at the reaction of FeCH2+ + D2 and the state-specific reactions of Fe+(6D, 4F) + CD4. We directly measure a barrier in excess of the endothermicity for dehydrogenation of methane by Fe+ as 41 ( 6 kJ/mol. This value is substantially lower than that calculated by theory,13 which attributes the barrier to a tight four-centered transition state. We use phase space theory (PST) modified to include assumptions appropriate for a tight transition state in the entrance channel (TD-PST) to model our results for the reaction of FeCH2+ + D2. This permits a further refinement of the quantitative details of the [FeCH4]+ PES. TD-PST is able to reproduce the cross section for formation of Fe+ + CH2D2, the elevated threshold observed for the thermoneutral FeCHD+ + HD reaction, and the branching ratios for hydrogen scrambling in the iron-hydride and iron-methyl ion channels. In addition, the TD-PST calculations are consistent with a tight transition state for elimination of methane from the D-Fe+-CH2D intermediate.

Reaction of FeCH2+ + D2 Acknowledgment. This work is supported by the National Science Foundation, Grant CHE-9221241. We thank Dr. D. G. Musaev and Prof. K. Morokuma for providing us with calculated frequencies for the TS1 and TS2 complexes. References and Notes (1) Allison, J. Prog. Inorg. Chem. 1986, 34, 627. Squires, R. R. Chem. ReV. 1987, 87, 623. Gas Phase Inorganic Chemistry; Russell, D. H., Ed.; Plenum: New York, 1989. Eller, K.; Schwarz, H. Chem. ReV. 1991, 91, 1121. Weisshaar, J. C. AdV. Chem. Phys. 1992, 82, 213. van Koppen, P. A. M.; Kemper, P. R.; Bowers, M. T. Organometallic Ion Chemistry; Freiser, B. S., Ed.; Kluwer: Dordrecht, in press. (2) Armentrout, P. B. In SelectiVe Hydrocarbon ActiVation: Principles and Progress; Davies, J. A., Watson, P. L., Liebman, J. F., Greenberg, A., Eds.; VCH: New York, 1990; pp 467-533. Armentrout, P. B.; Beauchamp, J. L. Acc. Chem. Res. 1989, 22, 315. (3) Haynes, C. L.; Chen, Y.-M.; Armentrout, P. B. J. Phys. Chem. 1995, 99, 9110. (4) Musaev, D. G.; Morokuma, K.; Koga, N.; Nguyen, K. A.; Gordon, M. S.; Cundari, T. R. J. Phys. Chem. 1993, 97, 11435. (5) Jacobson, D. B.; Freiser, B. S. J. Am. Chem. Soc. 1985, 107, 5870. (6) Aristov, N.; Armentrout, P. B. J. Phys. Chem. 1987, 91, 6178. (7) Sunderlin, L. S.; Armentrout, P. B. J. Phys. Chem. 1988, 92, 1209. (8) Armentrout, P. B.; Beauchamp, J. L. Acc. Chem. Res. 1989, 22, 315. (9) Schultz, R. H.; Elkind, J. L.; Armentrout, P. B. J. Am. Chem. Soc. 1988, 110, 411. (10) Elkind, J. L.; Armentrout, P. B. J. Phys. Chem. 1987, 91, 2037. (11) Fisher, E. R.; Sunderlin, L. S.; Armentrout, P. B. J. Phys. Chem. 1989, 93, 7375. (12) Fisher, E. R.; Schultz, R. H.; Armentrout, P. B. J. Phys. Chem. 1989, 93, 7382. (13) Musaev, D. G.; Morokuma, K. J. Chem. Phys. 1994, 101, 10697. (14) Ervin, K. M.; Armentrout, P. B. J. Chem. Phys. 1985, 83, 166. (15) Sunderlin, L. S.; Armentrout, P. B. Chem. Phys. Lett. 1990, 167, 188. (16) Teloy, E.; Gerlich, D. Chem. Phys. 1974, 4, 417. Gerlich, D. Diplomarbeit, University of Freiburg, Federal Republic of Germany, 1971. (17) Schultz, R. H.; Armentrout, P. B. Int. J. Mass Spectrom. Ion Processes 1991, 107, 29. (18) Schultz, R. H.; Crellin, K. C.; Armentrout, P. B. J. Am. Chem. Soc. 1991, 113, 8590. (19) Loh, S. K.; Fisher, E. R.; Lian, L.; Schultz, R. H.; Armentrout, P. B. J. Phys. Chem. 1989, 93, 3159. (20) Fisher, E. R.; Armentrout, P. B. J. Chem. Phys. 1991, 94, 1150. (21) Fisher, E. R.; Kickel, B. L.; Armentrout, P. B. J. Chem. Phys. 1992, 97, 4859. (22) Schultz, R. H.; Armentrout, P. B. Organometallics 1992, 11, 828. (23) Our previously published transition metal thermochemistry has been reevaluated in: Armentrout, P. B.; Kickel, B. L. Organometallic Ion Chemistry; Freiser, B. S., Ed.; Kluwer: Dordrecht, 1995; pp 1-45. (24) Clemmer, D. E.; Chen, Y.-M., Khan, F. A.; Armentrout, P. B. J. Phys. Chem. 1994, 98, 6522.

J. Phys. Chem., Vol. 100, No. 1, 1996 119 (25) van Koppen, P. A. M.; Kemper, P. R.; Bowers, M. T. J. Am. Chem. Soc. 1992, 114, 10941. (26) Armentrout, P. B. In AdVances in Gas Phase Ion Chemistry; Adams, N. G., Babcock, L. M., Eds.; JAI: Greenwich, 1992; Vol. 1, pp 83-119. (27) Beyer, T.; Swinehart, D. F. Commun. ACM 1973, 16, 379. Stein, S. E.; Rabinovitch, B. S. J. Chem. Phys. 1973, 58, 2438; Chem. Phys. Lett. 1977, 49, 183. Gilbert, R. G.; Smith, S. C. Theory of Unimolecular and Recombination Reactions; Blackwell Scientific: Oxford, UK, 1990. (28) Hague, R. H.; Margrave, J. L.; Kafafi, Z. H. In Chemistry and Physics of Matrix-Isolated Species; Andrews, L., Moskovits, M., Eds.; North-Holland: Amsterdam, 1989; Chapter 10. (29) Sugar, J.; Corliss, C. J. J. Phys. Chem. Ref. Data 1985, 14, Suppl. 2. (30) Bauschlicher, Jr., C. W.; Partridge, H.; Sheehy, J. A.; Langhoff, S. R.; Rosi, M. J. Phys. Chem. 1992, 96, 6969. (31) Bauschlicher, C. W., Jr.; Langhoff, S. R.; Partridge, H.; Barnes, L. A. J. Chem. Phys. 1989, 91, 2399. (32) Schilling, J. B.; Goddard, III, W. A.; Beauchamp, J. L. J. Am. Chem. Soc. 1986, 108, 582. (33) Armentrout, P. B. Annu. ReV. Phys. Chem. 1990, 41, 313; Science 1991, 251, 175. (34) Elkind, J. L.; Armentrout, P. B. J. Phys. Chem. 1986, 90, 5736. (35) Elkind, J. L.; Armentrout, P. B. J. Am. Chem. Soc. 1986, 108, 2765. (36) Elkind, J. L.; Armentrout, P. B. J. Chem. Phys. 1986, 84, 4862. (37) Schultz, R. H.; Armentrout, P. B. J. Phys. Chem. 1993, 97, 596. (38) Perry, J. K. Personal communication. These calculations are done at the MCPF level with a relativistic ECP replacing the Ne core. The basis set includes a diffuse f function on the metal. (39) Musaev, D. G.; Koga, N.; Morokuma, K. J. Phys. Chem. 1993, 97, 4064. (40) Chen, Y.-M.; Armentrout, P. B. J. Phys. Chem. 1995, 99, 10775. (41) Chase, Jr., M. W; Davies, C. A.; Downey, Jr., J. R.; Frurip, D. J.; McDonald, R. A.; Syverud, A. N. J. Phys. Chem. Ref. Data 1985, 14, Suppl. No. 1 (JANAF Tables). (42) Marcus, R. A. J. Chem. Phys. 1975, 62, 1372. (43) Chesnavich, W. J.; Bowers, M. T. J. Phys. Chem. 1979, 83, 900. (44) Weber, M. E.; Dalleska, N. F.; Tjelta, B. L.; Fisher, E. R.; Armentrout, P. B. J. Chem. Phys. 1993, 98, 7855. (45) Chen, Y.-M; Clemmer, D. E.; Armentrout, P. B. J. Am. Chem. Soc. 1994, 116, 7815. (46) Programs are available from the Quantum Chemistry Program Exchange, Indiana University, Program No. 557. Contributors to the original and revised programs include M. T. Bowers, W. J. Chesnavich, M. F. Jarrold, L. Bass, M. E. Grice, K. Song, D. A. Webb. (47) Pamidimukkala, K. M.; Rogers, D.; Skinner, G. B. J. Phys. Chem. Ref. Data 1982, 11, 85. (48) Shimanouchi, T. Tables of Molecular Vibrational Frequencies, Consolidated Vol. I; National Bureau of Standards: Washington, DC, 1972. (49) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular Structure Constants of Diatomic Molecules; Van Nostrand Reinhold: New York, 1979. (50) Pettersson, L. G. M.; Bauschlicher, C. W., Jr.; Langhoff, S. R. J. Chem. Phys. 1987, 87, 481.

JP9521215