SbF5

Calculations at B3LYP/6-31++G** + RECP (Sb) level have been performed for the protonation of C−H and C−C bonds of methane, ethane, propane, and ...
10 downloads 0 Views 137KB Size
Download PDF
J. Phys. Chem. B 2001, 105, 4331-4336

4331

DFT Calculations on the Protonation of Alkanes on HF/SbF5 Superacids Using Cluster Models Pierre M. Esteves,†,§ Alejandro Ramı´rez-Solı´s,‡ and Claudio J. A. Mota†,* Instituto de Quı´mica, Departamento de Quı´mica Orgaˆ nica, UniVersidade Federal do Rio de Janeiro, Cidade UniVersita´ ria CT Bloco A, 21949-900, Rio de Janeiro, Brazil, and Facultad de Ciencias, UniVersidad Auto´ noma del Estado de Morelos, CuernaVaca, Morelos 62210, Me´ xico ReceiVed: NoVember 28, 2000; In Final Form: February 26, 2001

Calculations at B3LYP/6-31++G** + RECP (Sb) level have been performed for the protonation of C-H and C-C bonds of methane, ethane, propane, and isobutane by models of the liquid superacid media HF/ SbF5. The antimony atoms were dealt with by relativistic effective core potentials. The species H2F+‚Sb2F11was considered as the model electrophile. The transition states for the protonation of the C-H bonds (H/H exchange) are similar to an H-carbonium ion interacting with the anion moiety. The enthalpies of activation for H/H exchange of alkanes were calculated in the range of 19 to 21 kcal/mol. For the protonation of the C-C bond, the enthalpy of activation strongly depends on the structure of the hydrocarbon being attacked, and was always higher than the enthalpy of activation for H/H exchange. This suggests the existence of steric demand for the C-C protonation.

I. Introduction Olah1

Hogeveen2

and independently studied the protonation of alkanes in liquid superacid systems and showed that alkanes behave as σ bases. The product distribution is determined by the reactivity of the C-H and C-C bonds.3 Proton transfer to a σ bond leads to the formation of a pentacoordinated carbonium ion,4 which contains a three-center, two-electron bond (3c-2e). On the basis of product distribution analysis, Olah proposed3a a qualitative scale for the reactivity order of σ bonds in superacids: tert C-H > C-C > sec C-H > prim C-H > CH4. Nevertheless, this order depends on the steric volume of the hydrocarbon and on the superacid active species (electrophile).5 Gas-phase studies of the protonation of isobutane and n-butane show6 that the attack at the C-C bonds is preferred over the protonation of the C-H bonds. This may suggest that in the absence of strong steric effects, C-C bonds are more basic than C-H bonds. Ab initio calculations7 indicate that pentacoordinated carbonium ions, supposed to be formed as discrete intermediates in this process, can easily decompose to carbenium ions plus methane or H2. An interesting reaction, parallel to the alkane ionization and observed to occur in liquid superacid media, is the H/D exchange.8 It occurs faster than ionization to carbenium ions, and is used as an evidence for the formation of carbonium ions as intermediates or transition states.8 On the other hand, there are few theoretical studies of alkane protonation using realistic models of liquid superacid media, especially if the anion is explicitly considered.9,10 Recently we carried out a study of alkane protonation by an HF/SbF5 model (HSbF6), at the DFT level of theory.9 We report here calculations with a larger and more realistic cluster of the superacid, using now the H2F+‚Sb2F11- molecule as protonating agent. * Corresponding author. † Universidade Federal do Rio de Janeiro. ‡ Universidad Auto ´ noma del Estado de Morelos. § Present address: Loker Hydrocarbon Research Institute, University of Southern California, University Park, Los Angeles, California 90089

A systematic study of the protonation of C-H and C-C bonds of methane, ethane, propane, and isobutane was performed using the H2F+‚Sb2F11- cluster to mimic the liquid superacid media HF/SbF5. This model was chosen because experimental results indicate that it is the major electrophilic species observed in the superacid media with high contents of SbF5,11 namely the HF/SbF5 with 1:1 molar ratio. II. Computational Details Quantum chemical calculations were performed using the DFT approximation at the B3LYP level of theory and using a 6-31++G** basis set for all atoms, except for the antimony atoms, where a relativistic effective core potential (RECP) was used to represent the core electrons.12 It is common knowledge nowadays that the use of well-extracted RECPs leads to excellent results in ab initio studies, as compared with allelectron (AE) calculations. It should be stressed that the use of RECP for heavy elements is a must since, for these atoms, all the scalar relativistic effects are crucial (especially the massvelocity and the Darwin relativistic corrections) and still cannot be accounted for using AE methods for large molecules. In this case, for the fluorine and antimony atoms, we have made some benchmark calculations using the present RECPs and compared our results with Car-Parrinello ab initio molecular dynamics10 (CPMD) and with available experimental results13 for some smaller subsystems of our model superacid clusters. The RECP results are, as expected, excellent and the interested reader can find the corresponding table as Supporting Information. This level of calculation will be represented as B3LYP/6-31++G** + RECP(Sb) from now on. We used the Gaussian 94 and Gaussian 98 packages14,15 for performing these calculations. The transition states (TS) for the H/H exchange (C-H protonation) and cracking (C-C protonation) were fully optimized using the Berny algorithm16 and were characterized, after analysis of the vibrational modes, computed at the same level of theory, as possessing only one imaginary frequency. The animation of the normal modes associated with this imaginary

10.1021/jp004310f CCC: $20.00 © 2001 American Chemical Society Published on Web 04/17/2001

4332 J. Phys. Chem. B, Vol. 105, No. 19, 2001

Esteves et al.

Figure 1. Optimized structure of the H2F+‚Sb2F11- model. Number of imaginary frequencies in parentheses.

frequency showed that the transition state found corresponds to the reaction of interest (reaction mode). All of the energies corresponding to the optimized structures were corrected for zero point energy (ZPE) and to 298.15 K. If not otherwise stated, all of the energies referred in this paper correspond to enthalpies. The activation enthalpies were calculated taking the differences between the transition states and the isolated alkanes and the superacid model, at 298.15 K and 1 atm. III. Results and Discussion Figure 1 shows the optimized structure of the superacid model. It can be seen that the geometry corresponds to the species H2F+ asymmetrically interacting with the Sb2F11- anion. This species will be referred to as H2F+‚Sb2F11- from now on. Figure 2 shows the transition states for the C-H protonation of methane (structure 2), ethane (structure 3), propane primary C-H bonds (structure 4) and secondary C-H bonds (structure 5), and isobutane primary (structure 6) and tertiary (structure 7) C-H bonds by the H2F+‚Sb2F11- cluster model. Figure 3 shows the transition states for the C-C protonation of ethane (structure 8), propane (structure 9), and isobutane (structure 10). Table 1 reports the absolute energy, the corrections for zero point energy (ZPE) and finite temperature (298.15 K), and absolute entropies, computed at B3LYP/6-31++G**//B3LYP/ 6-31++G** + RECP(Sb), as well as the imaginary frequencies (νimag) for all transition states, the isolated superacid model, and the alkane molecules studied in this work. Table 2 shows the activation parameters for the C-H protonation (H/H exchange) and Table 3 reports the activation parameters for C-C protonation (cracking) of the selected alkanes. Data on the activation energy for protonation of C-H and C-C bonds, computed at the same level of theory but using a simpler superacid model comprising an HF molecule complexed with a SbF5 (called the H+‚SbF6- model),9 were also included in Tables 2 and 3 for comparison. One can observe that the activation enthalpy for H/H exchange presents a relatively small range of variation, regardless of the C-H bond being protonated throughout the series, from methane to isobutane. The activation enthalpy ranged from 18.7 to 21.0 kcal/mol for the H2F+‚Sb2F11- system. Using the simpler superacid model (H+‚SbF6-), the activation enthalpy was found to be insensitive to the nature of the alkane, with

Figure 2. Optimized structures of the TS for H/H exchange reaction with methane, ethane, propane, and isobutene with the H2F+‚Sb2F11model. Number of imaginary frequencies in parentheses.

∆Hq of about 16 kcal/mol for all C-H bonds being examinated.9 We found that the use of a more realistic superacid model, in this case the H2F+‚Sb2F11-, leads to transition states with energy barriers more sensitive to the type of C-H bond being protonated than those found for the H+‚SbF6- complex, in agreement with what is observed experimentally.8 It is important to mention here that all of the ∆Hq discussion is relative to the isolated reactants, as ground state. Although there will be a small interaction energy between the alkanes and the superacid cluster, we did not take this into account in this study. There are three main reasons for this. First, alkanes are barely soluble in liquid superacid, showing that interactions among superacid molecules are dominant (probably hydrogen bonding, much stronger than van der Waals interactions) in liquid phase, where reactions are normally carried out. Second, as the main interaction forces are

DFT Calculations on the Protonation of Alkanes

J. Phys. Chem. B, Vol. 105, No. 19, 2001 4333

Figure 3. Optimized structures of the TS for C-C protonation (cracking) of ethane, propane, and isobutane with the H2F+‚Sb2F11- model. Number of imaginary frequencies in parentheses.

of van der Waals types, they should be of similar magnitude for all of the alkanes studied, thus adding an almost constant value in all calculated activation enthalpies. Last, the level of theory we used in this work (DFT) is not reliable enough for performing such calculations involving weak interactions. All transition states are similar to carbonium ions interacting with the Sb2F11- moiety. For the H/H exchange we can consider, as a rough view, the H-carbonium and, for the cracking reaction, the C-carbonium ion as the forming species. In the studies with the H+‚SbF6- model,9 analysis of the vibrational mode related to the imaginary frequency for the H/H exchange (the reaction coordinate), indicated a concerted mechanism. While the superacid proton is transferred toward the alkane, the other proton, from the alkane moiety, is being transferred back to the anion (Scheme 1a). Nevertheless, for the H/H exchange with the H2F+‚Sb2F11- model, a late transition state takes place. Nevertheless, the vibrational analysis of the reaction mode still indicates a concerted mechanism. The main movement associated with the reaction coordinate in these transition states (Scheme 1b) is the proton from the H2F+‚Sb2F11- basically completely transferring to the alkane molecule, and another

proton transferring back from the alkane to the Sb2F11- anion. Thus, this more realistic study reveals that pentacoordinated carbonium ions are not formed as discrete intermediates but as transition states, at least for the systems studied. It is of note to point out the similarity of the protonated alkane moiety in the transition states with the calculated structures of the methonium, ethonium, proponium, and isobutonium cations.17 Despite being theoretically characterized as minima on the potential energy surface, these species become transition states when the interaction with the anion is explicitly considered, showing that solvation effects could play a crucial role on the nature and reactivity of these species in the condensed phase. The same behavior is predicted to occur in the protonation of these alkanes by models of acid zeolite catalysts18 used in petroleum refining. The H/D exchange with methane and ethane in DF/SbF5 was studied experimentally and involves activation energies of 18 and 16.6 kcal/mol,19 respectively. The activation enthalpy computed for the H/H exchange with methane (16.8 kcal/mol) and ethane (15.8 kcal/mol) with the H+‚SbF6- model do agree well with the experimental values of ∆Hq reported for the H/D exchange in DF/SbF5 solutions. The activation enthalpies for

4334 J. Phys. Chem. B, Vol. 105, No. 19, 2001

Esteves et al.

TABLE 1: Absolute Energies, Zero Point Energies (ZPE), Thermal Corrections (298.15 K and 1 atm) and Absolute Entropies, and Imaginary Frequency (νimag), Calculated at B3LYP/6-31++G**// B3LYP/6-31++G** RECP(Sb) Level to the Transition State for C-H and C-C Protonation for Methane, Ethane, Propane, and Isobutane by the H2F+‚Sb2F11- Model of Liquid Superacids. species CH4 C2H6 C3H8 i-C4H10 H2F+‚Sb2F11- (1) 2 3 4 5 6 7 8 9 10 a

U (hartrees) B3LYP/basis seta

ZPE (kcal/mol)

H° (298)-H° (0) (kcal/mol)

S° (298) (cal/mol.K)

-40.52615 -79.84168 -119.15922 -158.47764

28.10 46.80 64.79 82.39

2.39 2.78 3.44 4.17

47.24 54.43 65.56 72.16

-1210.45963 -1250.95158 -1290.27012 -1329.58764 -1329.58859 -1368.90528 -1368.90647 -1290.24675 -1329.57054 -1368.89049

28.98 57.15 75.44 93.30 93.06 110.91 110.51 75.65 92.95 110.70

12.84 14.73 16.65 16.44 16.71 17.26 17.63 16.56 16.75 17.06

151.07 161.85 172.29 179.00 180.42 186.20 185.37 177.82 182.32 181.22

νimag (cm-1)

-168 -221 -217 -503 -236 -536 -287 -695 -754

Basis set 6-31++G** for C, H and F, and relativistic semilocal pseudopotential for Sb.

TABLE 2: Activation Barriers for the H/H Exchange Reaction at B3LYP/6-31++G**// B3LYP/6-31++G** + RECP (Sb) Level

SCHEME 1: Description of the “reaction mode” (animation of the vibration frequency associated to the imaginary constant) for the H/H Exchange with the Models H+SbF6- and H2F+‚Sb2F11- a

a Arrows sizes ares related with the amplitude of the movement of the hydrogen atoms in the “reaction mode”.

TABLE 3: Activation Barriers for Cracking Reactions (C-C Protonation), Calculated at B3LYP/6-31++G**// B3LYP/6-31++G** + RECP(Sb) Level ∆H°q (298.15 K) (kcal/mol) alkane/superacid model

H+‚SbF6-

H2F+‚Sb2F11-

C2H6 C3H8 i-C4H10

28.6 25.1 22.7

35.0 30.0 28.7

the H/H exchange with the H2F+‚Sb2F11- cluster, despite predicting a larger ∆Hq for methane relative to the reported experimental value, indicated that exchange of the tertiary C-H bond of isobutane is favored over the exchange of the primary C-H bond, a fact that is observed experimentally.8 The H/D exchange is generally accompanied by ionization, especially with propane and isobutane. Thus, one does not dispose of accurate measurements of the activation enthalpy for those hydrocarbons. An activation enthalpy of 18.3 kcal/mol for the ionization of isobutane was reported,20 being in good agreement with the calculated value for the H/H exchange at the methine position.9 This suggests that the H/D exchange and ionization to carbenium ions are related processes.

The protonation of the C-C bonds shows different features. The computed activation enthalpies are higher than for C-H protonation, ranging from 28.7 kcal/mol for isobutane to 35.0 kcal/mol for ethane. This agrees with experimental data in superacids, where the H/D exchange occurs much faster than ionization and cracking.5,8 Steric effects are usually attributed to explain this behavior, because the protonation of the C-C bond is sterically more demanding compared to protonation of C-H bonds. This implies that C-C protonation in liquid superacid systems presents bulkier transition states, which is reflected in the higher activation enthalpies compared to protonation of the C-H bonds (H/H exchange). In the transition state of C-C protonation of isobutane in H+‚SbF6-, the C-H bond distances are 1.253 Å and 1.515 Å, whereas the C-H bond distances for the H/H exchange of the tertiary C-H bond of isobutane using the same cluster were calculated as 1.178 and 1.419 Å. The transition states for cracking (C-C protonation) present longer bond lengths than do the transition states for H/H exchange, indicating a bulkier nature. The low imaginary vibrational frequencies observed for C-H and C-C protonation indicate that the carbonium ion and the anion moieties are weakly associated in a loosely bounded transition state. Therefore, explaining the reaction in terms of

DFT Calculations on the Protonation of Alkanes SCHEME 2: Interconversion, Bond-to-Bond Rearrangement, of H-Carbonium to C-Carbonium Ion as a Possible Pathway to Explain Products of C-C Protonation in Superacid Catalyzed Alkane Reactions

formation of discrete intermediates is a good approximation, although they do not seem to be really formed. As recently shown,7 the experimental scale of σ bond reactivity in liquid superacids, proposed by Olah, reflects the kinetics of C-C and C-H protonation in superacids, rather than their intrinsic basicity. This is reflected in the activation barriers. The C-H bond, being more external and accessible than the C-C bond, involves lower activation energies, especially for bulky electrophiles. Nevertheless, one should note that rearrangement of carbonium ions, also known as “bond-to-bond” rearrangement,4 is an easy process. Calculations indicate17g,h that protonation of a primary C-H bond of isobutane could yield methane and the isopropyl cation, because the intramolecular (bond-to-bond) rearrangement of the 1-H-carbonium ion to the C-isobutonium ion and its subsequent decomposition are barrierless processes (Scheme 1). Carbonium ions have been quoted to have a very “fluxional” structure,21-23 and even as not having an assigned structure. This is specially valid for some isolated species in the gas phase, such as the isolated methonium cation.23b Nevertheless, one could define structure as an average nuclear configuration (geometry) in which the system spend most of the time (ca. 90%). When some solvation effects,21-23 even with H2 molecules,22 are considered, these species have their bond-to-bond rearrangements slowed, and the system seems to assume the geometry with Cs symmetry.17a-e This seems to be a useful discussion for carbonium ions, when anions are not present or if they are far from the cation. In fact, ion pairing and coordination in the transition state limits large fluxional movements of the nuclei. Thus, the description of the “fluxional” nature of carbonium ions in condensed phase or when some solvation takes place may be of secondary relevance. IV. Conclusions The transition states found at B3LYP/6-31++G** + RECP(Sb) for the C-H and C-C protonation of some alkanes by cluster models of HF/SbF5 superacid system were quite similar to the respective carbonium ions interacting with the anion moiety. The transition states for the H/H exchange resemble the H-carbonium ions interacting with the Sb2F11- anions, whereas the transition states for C-C protonation are similar to C-carbonium ions interacting with this anion. Nevertheless, these calculations suggest a late transition state, where the proton was basically transferred to the alkane and another one is being donated back to the Sb2F11- for the C-H protonation. The activation barriers for the C-H protonation were calculated in the range of 19-21 kcal/mol, depending on the alkane, whereas the C-C protonation requires higher energy (activation barrier ranging from 28.7 kcal/mol for isobutane, to 35 kcal/mol for ethane). The transition states were found to be carbonium ionlike, which means that thinking of carbonium ion structures for the transition states is a good approximation for guessing about reactivity of alkane σ-bonds. Acknowledgment. C.J.A.M. thanks support from FAPERJ, CNPq and FINEP/PRONEX. A.R.-S. thanks support from

J. Phys. Chem. B, Vol. 105, No. 19, 2001 4335 CONACYT (Me´xico) through project 34673-E. The authors also thank the Ibero-American Program of Science and Technology for the Development - CYTED (V-B), for partial support. P.M.E. thanks FAPERJ for a fellowship. Supporting Information Available: Table showing the comparison of bond lengths and bond angles obtained at the B3LYP/6-31++G** + RECP(Sb) level with Car-Parrinello ab initio molecular dynamics and with experimental results. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Olah, G. A.; Schlosberg, R. H J. Am. Chem. Soc. 1968, 90, 2726. (2) Hogeveen, H.; Bickel A. F. Rec. TraV. Chim. Pays-Bas 1967, 86, 1313. (3) (a) Olah, G. A. Angew. Chem., Int. Ed. Engl. 1973, 12, 173. (b) Olah, G. A.; Prakash, G. K. S.; Sommer, J. Science 1979, 206, 13. (4) Olah, G. A. J. Am. Chem. Soc. 1972, 94, 807. (5) Olah, G. A.; Halpern, Y.; Shen, J.; Mo, Y. K. J. Am. Chem. Soc. 1973, 95, 4960. (6) Aquilanti, V.; Galli, A.; Giardini-Guidoni, A.; Volpi, G. G. J. Chem. Phys. 1968, 48, 4310. (7) Esteves, P. M.; Alberto, G. G. P.; Ramirez, A. Mota, C. J. A. J. Am. Chem. Soc. 1999, 121, 7345. (8) (a) Sommer, J.; Hauchomy, M.; Garin, F.; Barthomeuf, D. J. Am. Chem. Soc. 1994, 116, 5491. (b) Sommer, J.; Bukala, J.; Hauchomy, M.; Jost, R. J. Am. Chem. Soc. 1997, 119, 3274. (9) Esteves, P. M.; Ramı´rez-Solı´s, A.; Mota, C. J. A. J. Braz. Chem. Soc. 2000, 11, 345-348. (10) Kim, D.; Klein, M. L. J. Phys. Chem. B 2000, 104, 10074. (11) Mootz, D.; Bartmann, K. Angew. Chem., Int. Ed. Engl. 1988, 27, 391. (12) Bergner, A.; Dolg, M.; Kueche, W.; Stoll, H.; Preuss, H. Mol. Phys. 1993, 80, 1431. (13) (a)Latajka, Z.; Bouteiller, Y. J. Chem. Phys. 1994, 101, 9793. (b) Sondag, H.; Weber, R. J. Mol. Spectrosc. 1982, 91, 91. (c) Helminger, P. Phys. ReV. A 1971, 3, 122. (d) Konaka, S.; Kimura, M. Bull. Chem. Soc. Jpn. 1970, 43, 1693. (e) Brunvoll, J.; Ischenko, A. A.; Miakshin, I. N.; Romanov, G. V.; Spiridonov, V. P.; Strand, T. G.; Sukhoverkhov, V. F. Acta Chem. Scand., Ser. A 1980, 34, 733. (f) Minkwitz, R.; Schneider, S.; Kornath, A. Inorg. Chem. 1998, 37, 4662. (g) Ohlberg, S. M. J. Am. Chem. Soc. 1959, 81, 811. (14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Headgordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, ReVision C.3, 1995, Gaussian Inc., Pittsburgh, PA. (15) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998. (16) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput. Chem. 1996, 17, 49. (17) (a) Muller, H.; Kutzelnigg, W.; Noga, J.; Klopper, W. J. Chem. Phys. 1997, 106, 1863. (b) Schleyer, P. v. R.; Carneiro, J. W. M. J. Comput. Chem. 1992, 13 (8), 997. (c) Olah, G. A.; Rasul, G. Acc. Chem. Res. 1997, 30, 245 and references therein. (d) Carneiro, J. W. M.; Schleyer, P. v. R.; Saunders, M.; Remington, R.; Schaefer, H. F., III.; Rauk, A.; Sorensen, T. S. J. Am. Chem. Soc. 1994, 11, 6, 3483. (e) Schreiner, P. R.; Kim, S. J.; Schaefer, H. F. III; Schleyer, P. v. R. J. Chem. Phys. 1993, 99, 3716. (f) Esteves, P. M.; Mota, C. J. A.; Ramı´rez-Solı´s, A.; Hernandez-Lamoneda, R. J. Am. Chem. Soc. 1998, 120, 3213. (g) Mota, C. J. A.; Esteves, P. M.; Ramı´rez-Solı´s, A.; Hernandez-Lamoneda, R. J. Am. Chem. Soc. 1997, 119, 5193. (h) Esteves, P. M.; Mota, C. J. A.; Ramı´rez-Solı´s, A.; HernandezLamoneda, R. Top. Catal. 1998, 6, 163.

4336 J. Phys. Chem. B, Vol. 105, No. 19, 2001 (18) Esteves, P. M.; Nascimento, M. A. C.; Mota, C. J. A. J. Phys. Chem. B 1999, 103, 10417. (19) Hogeveen H.; Bickel, A. F. Rec. TraV. Chim. Pay-Bas 1969, 88, 371. (20) Hogeveen, H.; Gaasbeck, C. J.; Bickel, A. F. Rec. TraV. Chim. Pay-Bas 1969, 88, 703. (21) Schreiner, P. R. Angew. Chem., Int. Ed. Engl. 2000, 39, 3239.

Esteves et al. (22) (a) Hiraoka, K.; Kudaka, I.; Yamabe, S. Chem. Phys. Lett. 1991, 184, 271. (b) Boo, D. W.; Lee, Y. T. Chem. Phys. Lett. 1993, 211, 358. (c) Boo, D. W.; Lee, Y. T. J. Chem. Phys. 1995, 103, 520. (23) (a)Wan B., D.; Lee, Y. T. Int. J. Mass Spectrom., Ion Processes 1996, 159, 209. (b) White, E. T.; Tang, J.; Oka, T. Science 1999, 284(5411), 135.