Mechanistic Insight into the Gas-Phase Reactions of Methylamine with

Nov 4, 2010 - College of Physics Science and Technology, China University of Petroleum, Dongying, Shandong 257061, ... City University of Hong Kong...
1 downloads 0 Views 1MB Size
Download PDF
12490

J. Phys. Chem. A 2010, 114, 12490–12497

Mechanistic Insight into the Gas-Phase Reactions of Methylamine with Ground State Co+(3F) and Ni+(2D) Xiaoqing Lu,†,‡ Shuxian Wei,† Wenyue Guo,*,† and Chi-Man Lawrence Wu*,‡ College of Physics Science and Technology, China UniVersity of Petroleum, Dongying, Shandong 257061, People’s Republic of China, and Department of Physics and Materials Science, City UniVersity of Hong Kong, Hong Kong SAR, People’s Republic of China ReceiVed: July 10, 2010; ReVised Manuscript ReceiVed: October 5, 2010

The gas-phase reaction mechanisms of methylamine (MA) with the ground-state Co+(3F) and Ni+(2D) are theoretically investigated using density functional theory at both the B3LYP/6-311++G(d,p) and B3LYP/ 6-311++G(3df,2p) levels. The reactions for hydride abstraction and dehydrogenation are analyzed in terms of the topology of potential energy surfaces (PESs). Co+ and Ni+ perform similar roles along the isomerization processes to the final products. Hydride abstraction takes place via the key species of metal cation-methyl-H intermediate, followed by a charge transfer process before the direct dissociation of CH2NH2+ · · · MH (M ) Co, Ni). The enthalpies of reaction, stability of metal cation-methyl-H species, and competition between different channels account for the sequence of the hydride abstraction products: CoH < NiH < CuH. The most competitive dehydrogenation route occurs through a stepwise reaction, consisting of initial C-H activation, amino-H shift, and direct dissociation of the precursor CH2NHM+ · · · H2. This theoretical work sheds new light on the experimental observations and provides fundamental understanding of the reaction mechanisms of amine prototype with late first-row transition metal cations. 1. Introduction In the past few decades, the chemistry of transition metals with nitrogen-containing species has been investigated in nitrogen fixation processes, biological processes, and organic syntheses.1-4 Earlier experiments on the gas-phase reactions of primary amines with first-row transition metal cations were elementary and substantial for unveiling mechanisms of the metal cations-mediated reactions as well as providing fundamental knowledge of the catalytic processes.5-14 However, experimental descriptions are significantly limited to certain experimental methods and apparatuses used in operation. For instance, in the Co+ + methylamine (MA) reaction, the final products were 53% CoH and 47% H2 in the ion cyclotron resonance (ICR) experiment,15 but 26% CoH and 74% H2 using the ion beam apparatus.16 On the other hand, owing to the different characteristics of transition metal cations involved, diversity of the final product contributions was detected. For example, the final products via reaction with MA are metal hydride, hydrogen, and methane for Ti+, V+, and Cr+;8,10 metal hydride and hydrogen for Co+ and Ni+;15,16 and only metal hydride for Cu+.5 Furthermore, although the kinds of final products are the same for some analogous reactions with different metal ions, relative intensities between the different products might be quite different. In the ion beam experiment, the products of 60% NiH and 40% H2 were observed in the Ni+/MA reaction, compared to 26% CoH and 74% H2 in the Co+/MA reaction.16 In a word, in spite of the extensive experimental studies, the mechanistic details of the reactions concerning transition metal cations with amine prototype are not clear enough. * To whom correspondence should be addressed. E-mail: wyguo@ upc.edu.cn (W.G.), [email protected] (C.-M.L.W.). † China University of Petroleum. ‡ City University of Hong Kong.

In light of the successful use of the density functional theory (DFT) in predicting molecular geometries and energies of smallsized organometallics,17-30 in this paper we perform a DFT investigation on the reactions of MA with low-spin ground state Co+(3F) and Ni+(2D). The structures and energies for all the reactants, intermediates, and products involved are discussed in detail. We aim to elucidate the effect of electronic configuration of metal cations on the reactivity, generalize the mechanisms of amine prototype with late first-row transition metal cations, and bridge the gap between experiment and theory. 2. Computational Methods The geometires and energies of intermediates, transition states, and products involved in the gas-phase reactions of methylamine with the ground-state Co+(3F) and Ni+(2D) were investigated by using the B3LYP functional31 in conjunction with both the B3LYP/6-311++G(d,p) and the B3LYP/6-311++G(3df,2p) basis sets. Vibrational frequencies of all the optimized species were calculated at the same levels to identify the stationary points (minima or transition states) and to estimate the zeropoint-energy (ZPE) corrections that are applied to all reported energies. Scaling factor for the ZPE was selected as 0.961.29,32 To confirm the connection of transition states with the corresponding reactants and products for some key steps, intrinsic reaction coordinate (IRC) calculations33,34 were performed to follow the reaction pathways. The values of 〈S2〉 for all the calculated species were detected to evaluate if spin contamination can influence the quality of the results and found it is less than 5% (see Supporting Information Tables S1 and S2), suggesting that spin contamination is negligibly small in the calculated results. All the calculations were carried out using the Gaussian 03 program package.35 Natural bond orbital (NBO) theory was used to characterize interactions between different groups for some key species

10.1021/jp106397g  2010 American Chemical Society Published on Web 11/04/2010

Gas-Phase Reactions of Methylamine

J. Phys. Chem. A, Vol. 114, No. 47, 2010 12491

Figure 1. Geometries and selected structural parameters optimized at both the B3LYP/6-311++G(d,p) and B3LYP/6-311++G(3df,2p) levels for reactants, products, intermediate minima, and transition states involved in the reaction of methylamine with Co+. Bond lengths are in Å, and angles are in °.

involved. These calculations were performed by using the program package of NBO5.0.36 3. Results and Discussion Geometries together with structural parameters for the reactants, intermediates, and products involved in the Co+/MA and Ni+/MA reactions are shown in Figures 1 and 2. PESs for the hydride abstraction and dehydrogenation processes are shown in Figures 3 and 4. Total energies (E) together with ZPE and 〈S2〉 for these calculated species at both the B3LYP/6311++G(d,p) and B3LYP/6-311++G(3df,2p) levels are listed in the Supporting Information Tables S1 and S2. According to the calculated geometries and PESs, we find that Co+ and Ni+ perform similar roles along the isomerization processes to the final products of hydride abstraction and dehydrogenation. Since the results obtained from the two basis sets are very close to each other (see Figures 3 and 4), only the B3LYP/6311++G(3df,2p) results will be used in the discussion unless otherwise indicated. 3.1. Encounter Complexes and Conversions. All possible encounter complexes of metal cations M+ (M ) Co, Ni) with MA are considered, including the N-attached, η1-methyl-Hattached, η2-methyl-amino-H-attached, and η3-methyl-H-attached complexes in reference to the reaction of Co+ with ethylamine (see Figure S1).30 However, only two isomers are found to exist stably in the reaction of MA with M+ (M ) Co, Ni), namely N-attached complex 1 and η1-methyl-H-attached complex 3, as shown in Figures 1 and 2. Complex 1 corresponds to the global minimum, and the existence of a lone pair on the N atom makes it the most suitable position for the attack of M+, characteristic of association involving amines and M+, for

example, Na+,13 Mg+,24-26 Co+,28,30 Ni+,21 Cu+,17-19,29 and Ag+.13 This fact can be understood by considering the electron transfer that occurs from the N lone pair into the empty metal 4s orbital, and taking into account the reduction of repulsion caused by the 3d/4s hybridization at the metal center that moves electron density away from the M+-N axis. Correspondingly, the binding energies are calculated to be as large as 55.9 and 62.7 kcal/mol for the Co+-MA and Ni+-MA associations. The binding energy of 59.7 kcal/mol for Ni+-MA obtained at the B3LYP/6-311++G(d,p) level (see Figure 3) is in good agreement with the previous result at the B3LYP/6-311G(d,p) level (58.8 kcal/mol).21 The calculated binding energies of the Co+-MA and Ni+-MA associations are also very close to the binding energy of 60.7 kcal/mol for the Cu+-MA association at the B3LYP/6-311++G(3df,2p) level.29 However, NBO analysis shows, quite different from the bonding situation of σ(Cu-N) (8.05% Cu(4s) + 91.95% N(2s12p5)) in the Cu+-MA association,29 the Co+ and Ni+ associations are stabilized by the electron donation of the lone pair on the N atom into the metal 4s orbital (∆E(2): 43.25 and 47.45 kcal/mol for Co+-MA and Ni+-MA; see Table 1). This difference is largely due to the different characteristics between the fulfilled 3d orbitals of Cu+ and unfulfilled ones of Co+ and Ni+. In the latter two cases, such donor-acceptor interactions result in strong combinations of M+ with MA and weaken the skeleton bonds of MA as well. Seen from Figures 1 and 2, the C-N bond length indeed stretches to 1.503 and 1.500 Å in the Co+-MA and Ni+-MA associations, with respect to 1.465 Å in the free MA. Aside from the most favorable base site of N, there exists another base site of η1-methyl H that could interact with M+ to construct encounter complex 3. This type of encounter complex

12492

J. Phys. Chem. A, Vol. 114, No. 47, 2010

Lu et al.

Figure 2. Geometries and selected structural parameters involved in the reaction of methylamine with Ni+. Parameters follow the same notations as in Figure 1.

Figure 3. Potential energy surface associated with the hydride abstraction of methylamine with M+ (M ) Co, Ni) calculated at both the B3LYP/6-311++G(d,p) and B3LYP/6-311++G(3df,2p) levels. Numbers refer to the relative energies with respect to the entrance channel including ZPE corrections. Energies are in kcal/mol.

Figure 4. Potential energy surface associated with the dehydrogenation of methylamine with M+ (M ) Co, Ni). Parameters follow the same notations as in Figure 3.

was previously located in the Cu+-MA29 and Co+-ethylamine associations.30 An inspection of Figures 1 and 2 shows that the coordinated C-H bond is largely stretched (1.247 Å in Co+-H-CH2NH2 and 1.286 Å in Ni+-H-CH2NH2), and the M+-H bond is shortened to the regular range for the M+-H-C agostic interaction.22 NBO analysis shows that this agostic

structure is stabilized not only by direct donation from σ*(M-N) to σ(M-N), but also by a circuitous route donating electrons from the filled σ*(M-N) antibonding orbital to σ*(C-H1), back-donation from σ(C-H1) to the σ(M-N) bonding orbital, as well as σ*(M-N) to σ(M-N) electron transfer (see Table 1). These electron transfers result in the strengthening interaction between M+ and N as well as the binding of M+ to H1.

Gas-Phase Reactions of Methylamine

J. Phys. Chem. A, Vol. 114, No. 47, 2010 12493

TABLE 1: Second-Order Perturbation Theory Analysis of the Fock Matrix in the NBO Basis for the Selected Species in the Co+/MA and Ni+/MA Reactionsa Co+/MA species

orbitals involved

E(2)b

orbitals involved

E(2)b

1 2

LP(N) f 4s*(Co) LP(N) f σ*(C-H1) LP(N) f σ*(Co-C) σ(C-H1) f 4s*(Co) σ(Co-C) f σ*(Co-H1) σ(Co-H1) f σ*(Co-C) LP(N) f 3d*(Co) 3d(Co) f σ*(C-H1) σ*(Co-N) f σ*(C-H1) σ(C-H1) f σ(Co-N) σ*(Co-N) f σ(Co-N) σ(Co-H1) f σ*(N-H2) LP(N) f 4s*(Co) LP(N) f σ*(Co-H2) σ*(Co-H2) f 4s*(Co) σ(Co-H2) f σ*(Co-N) σ(Co-N) f σ*(Co-H2) LP(N) f 4s*(Co) σ(H1-H2) f 4s*(Co)

43.25 21.05 16.25 54.09 18.69 16.60 21.20 34.88 40.99 19.22 10.57 31.74 48.19 29.92 68.34 9.65 6.17 60.69 41.46

3d(Co) f σ*(H1-H2)

13.10

LP(N) f 4s*(Ni) LP(N) f σ*(C-H1) LP(N) f σ*(Ni-C) σ(C-H1) f 4s*(Ni) σ(Ni-C) f σ*(Ni-H1) σ(Ni-H1) f σ*(Ni-C) LP(N) f 4s*(Ni) 3d(Ni) f σ*(C-H1) σ*(Ni-N) f σ*(C-H1) σ(C-H1) f σ(Ni-N) σ*(Ni-N) f σ(Ni-N) σ(Ni-H1) f σ*(N-H2) σ(Ni-H2) f σ*(Ni-N) LP(H2) f 4s*(Ni) LP(H2) f σ*(Ni-N) σ(Ni-N) f σ*(Ni-N) σ*(Ni-N) f 3d*(Ni) LP(N) f 4s*(Ni) σ(H1-H2) f 4s*(Ni) σ(H1-H2) f σ*(Ni-N) 3d(Co) f σ*(H1-H2)

47.45 24.89 22.82 68.38 15.06 12.55 17.78 40.48 48.94 24.00 12.89 33.07 25.83 86.54 56.75 43.57 37.72 26.40 18.94 22.24 11.63

3 4 6

7

a

Ni+/MA

Orbital interaction energies E

(2)

b

are in kcal/mol. Values calculated at the NBO//B3LYP/6-311++G(3df,2p) level.

The attack of M+ at different base sites of MA leads to encounter complexes 1 and 3. Once formed, they could transfer into each other through two possible pathways, that is, direct metal movement and/or stepwise C-H activation-rearrangement. Along the first route, M+ moves directly from the NH2 end to the CH3 side. This process is connected by transition state TS1-3, which lies 24.3 (M ) Co) and 27.7 (M ) Ni) kcal/mol below their respective entrance channel, as shown in Figure 3. Along the second route, metal cation insertion into the closer C-H bond of the adjacent CH3 group accounts for species 2 via transition state TS1-2. The new species is feathered by a Cs symmetry defined by H1-M+-C-N (see Figures 1 and 2). 1 f TS1-2 f 2 provides a complete description of the bond breaking process of C-H1 and bond forming process of M+-C and M+-H. The activation barriers of the C-H bond is 28.6 kcal/mol for the Co+ insertion and 32.5 kcal/mol for the Ni+ insertion, which are both smaller than that of 43.1 kcal/mol for the Cu+ insertion.29 The following conversion of 2 f TS2-3 f 3 takes place via an intrarotation of the MH+ entity. In this conversion the energy of the involved transition state TS2-3 lies 18.8 kcal/mol above 2 or 14.1 kcal/mol above 3 for Co+/MA. The energy barrier in the Ni+/MA reaction is relatively low, lying 9.3 kcal/mol above 2 or 8.3 kcal/mol above 3. Because transition states for both routes are at least 10.5 (M ) Co) and 20.9 (M ) Ni) kcal/mol lower in energy than their respective reaction entrances, the two encounter complexes could convert into each other readily, and the direct metal movement pathway might be more favorable since it is not only simple but is also lower in energy. 3.2. Hydride Abstraction Process. The mechanism of hydride abstraction from MA has been established for late firstrow transiton metal cation Cu+.29 Seen from Figure 3, we find that Co+ and Ni+ play similar roles as Cu+ along the reaction coordinate to the final products of hydride abstraction. That is, reactions take place via the key complex 3, followed by a charge transfer process to 4 and a nonreactive dissociation process to 5. All species involved in 3 f TS3-4 f 4 are rather stable for both the Co+/MA and Ni+/MA reactions, thus the hydride abstraction processes could take place easily.

Here, we focus on the consistency of their PESs as well as on the particularity and the impact to the final products arising therefrom. For the MA + M+ (M ) Co, Ni, and Cu) reactions, the experimental result of the hydride abstraction products is 5, CH2NH2+ + MH; and the products in the increasingabundance sequence are CoH (26% in ion beam; 53% in ICR experiment), NiH (60%), and CuH (100%).5,15,16 According to the calculated PESs, the overall reaction is endothermic by 2.3 kcal/mol for Co+ + CH3NH2 f CH2NH2+ + CoH, exothermic by 4.4 kcal/mol for Ni+ + CH3NH2 f CH2NH2+ + NiH, and exothermic by 8.8 kcal/mol for Cu+ + CH3NH2 f CH2NH2+ + CuH.29 As is well-known from the Arrhenius formula for the evaluation of kinetics, a difference of about 10 kcal/mol between two barriers implies that reaction rates change by several orders of magnitude. Thus, the calculated enthalpy difference indicates that reactions for the hydride abstraction products are inclined to the sequence of CoH < NiH < CuH, in line with the experimental findings.5,15,16 In addition to the enthalpies of reaction, two factors including the stability of the key complex (3) and competition between different channels in the conversion of 1 f 3 would have some effect on the hydride abstraction products. Seen from their PESs, complex 3 appearing in the Cu+/MA reaction is more stable (-30.7 kcal/ mol) than those appearing in both Ni+/MA (-29.2 kcal/mol) and Co+/MA (-24.6 kcal/mol) reactions, originated from the difference of electron donor ability caused by the fulfilled Cu+ and unfulfilled Co+ and Ni+. Therefore, the reaction proceeding from the relative stable Cu+-MA association is more competitive than those from Ni+-MA and Co+-MA associations. Besides, in the 1 f 3 conversion processes, only transition state TS1-3 appearing in the Cu+/MA reaction is lower in energy than that of the C-H activation transition state TS1-2, whereas the route via direct metal movement to 3 is less competitive compared to that via the C-H activation to 2 for the Co+/MA and Ni+/MA reactions. Note that complex 2 is the key intermediate for branching to the dehydrogenation products. In a word, the enthalpies of reaction, stability of 3, and competition between different channels are the factors that account for the

12494

J. Phys. Chem. A, Vol. 114, No. 47, 2010

Lu et al.

TABLE 2: Orbitals Responsible for Charge Distributions Together with Their Characters for the Selected Species Calculated at the NBO//B3LYP/6-311++(3df,2p) Level Co+/MA species 2

orbitalsa σ(Co-C) σ(Co-H1)

3

σ*(Co-N) σ(C-H1)

4

σ(Co-H1) π(C-N)

6

σ(Co-H2) σ(Co-H2) σ(Co-N) σ(Co-N) σ(C-H1)

TS2-3

σ(Co-H1) π(C-N)

TS3-4

σ(Co-H1) π(C-N)

TS6-7

σ(Co-N) σ(Co-H2) σ(C-H1)

relative contributions 46% Co(4s3d5.95) 54% C(2s2p8.48) 42% Co(4s3d2.35) 58% H(1s) 7% Co(4s) 93% N(2p) 46% C(2s2p4.32) 54% H(1s) 17% Co(4s) 83% H(1s) 31% C(2p) 69% N(2p) 67% Co(4s3d7.65) 33% H(1s) 40% Co(4s3d1.82) 60% H(1s) 43% Co(3d) 57% N(2p) 23% Co(4s3d2.83) 77% N(2s2p2.68) 61% C(2s2p2.95) 39% H(1s) 31% Co(4s) 69% H(1s) 26% C(2p) 74% N(2p) 22% Co(4s) 78% H(1s) 28% C(2p) 72% N(2p) 61% Co(3d) 39% N(2p) 40% Co(4s3d4.84) 60% H(1s) 55% C(2s2p3.82) 45% H(1s)

Ni+/MA occupancies

orbitalsa

0.921

σ(Ni-C)

0.935

σ(Ni-H1)

1.822

σ*(Ni-N)

1.900

σ(C-H1)

1.894

σ(Ni-H1)

1.999

π(C-N)

0.998

σ(Ni-H2)

0.939

σ(Ni-N)

0.949

σ(Ni-N)

0.953

σ(Ni-N)

1.986

σ(C-H1)

1.799

σ(Ni-H1)

1.994

π(C-N)

1.962

σ(Ni-H1)

1.999

π(C-N)

1.911

σ(Ni-N)

0.955

σ(Ni-H2)

1.640

σ(C-H1)

relative contributions 54% Ni(4s3d6.40) 46% C(2s2p10.18) 48% Co(4s3d2.35) 52% H(1s) 6% Ni(4s) 94% N(2p) 45% C(2s2p4.58) 55% H(1s) 19% Ni(4s) 81% H(1s) 31% C(2p) 69% N(2p) 51% Ni(4s3d0.77) 49% H(1s) 62% Ni(4s3d21.22) 38% N(2s2p6.38) 40% Ni(4s3d8.78) 60% N(2s2p4.86) 21% Ni(4s3d1.41) 79% N(2s2p3.73) 61% C(2s2p2.89) 39% H(1s) 30% Ni(4s) 70% H(1s) 28% C(2p) 72% N(2p) 20% Ni(4s) 80% H(1s) 28% C(2p) 72% N(2p) 73% Ni(3d) 27% N(2p) 50% Ni(4s3d5.41) 50% H(1s) 53% C(2s2p3.84) 47% H(1s)

occupancies 0.934 0.954 1.803 1.875 1.891 1.999 0.885 0.925 0.937 0.955 1.988 1.818 1.998 1.960 1.998 0.960 1.952 0.843

a Not included are one σ(C-N), one σ(N-H), and two σ(C-H) values in all species; another σ(N-H) is also not included in 2, 3, 4, TS2-3, and TS3-4.

sequence of the hydride abstraction products for these three late first-row transition metal cations. In addition to 5, another possible product 5′ (CH2NH2 + MH+) may arise from the direct decomposition of intermediates 2 and/or 3. Seen from Figure 3, 5′ is calculated to be 45.0 kcal/ mol higher in energy than 5 in the Co+/MA reaction, in accord with the energy difference of 42.5 kcal/mol between 5′ and 5 estimated with the ionization potentials of CoH (∼8.14 eV)16 and CH2NH2 (∼6.3 eV).24,37 In the Ni+/MA reaction, this energy difference is 50.6 kcal/mol compared to the experimental results (50.8 kcal/mol) estimated with the ionization potentials of NiH (∼8.50 eV)38 and CH2NH2 (∼6.3 eV).24,37 This point is consistent with the previous reported energy difference of 72.2 kcal/mol in the Cu+/MA reaction.29 It is this energy difference that dictates 5 rather than 5′ to be the hydride abstraction products. Our present results provide good support to the relation between charge distributions and geometries as observed in the Cu+/MA reaction,29 that is, the distribution of positive charge is strictly determined by their geometries. The positive charge would localize primarily on the MH entity when one atom of MH could form a σ covalent bond with the fourth sp3 hybridized orbital at the C atom. For example, in species 2 and 3 for both Co+/MA and Ni+/MA reactions, the C atom interacts with M

(as in 2) or H (as in 3) to form the fourth covalent bond (see Table 2). In such cases, the C atom can not form further a π bond with the N atom, thus the C, N-containing species is CH2NH2 rather than CH2NH2+. Otherwise, if there is no the fourth covalent bond forming between the C atom and either atom of the MH entity, the C atom would form a π bond with the N atom, and the uncoupled electron of the CH2NH2 radical would move into the MH+ entity (see Table 2). Thus, the positive charge localizes primarily on the CH2NH2 entity, such as in 4, TS2-3, and TS3-4. From the point of view of ionization potential, the relation between charge distributions and geometries can be explained in a much simpler way. According to the experimental values, the ionization potential of MA (8.9 eV) is higher than that of Co (7.8 eV) or Ni (7.6 eV), while the ionization potential (6.3 eV) of CH2NH2 is lower than that of CoH (8.1 eV) or NiH (8.5 eV). Therefore, the positive charge tends to locate in CH2NH2 in [CH2NH2-CoH]+, while in the case of MA associations remains in the metal. That is, the positive charge locates in the entity with lower ionization potential.5,9,15,16 In fact, the relation between charge distributions and geometries is not only suitable for the species occurred in hydride abstraction processes, but also for all species involved in dehydrogenation processes.

Gas-Phase Reactions of Methylamine 3.3. Dehydrogenation Process. All possible reaction mechanisms for dehydrogenation have been surveyed, and we found three reaction pathways, involving direct elimination from adjacent C and N groups, initial N-H activation, and initial C-H activation. PESs for the dehydrogenation processes are shown in Figure 4. The dehydrogenation processes via the direct elimination mean that the precursor of CH2NHM+ · · · H2 (M ) Co, Ni) forms without involving the metal cations-mediated intermediates. The “direct” mechanism has been established for alkane dehydrogenation,39 ethylenediamine + Cu+,40 and ethylamine + Co+ reactions.30 For the MA + M+ (M ) Co, Ni) reactions, the mechanisms can be envisaged that the two eliminated H atoms come from the adjacent C and N groups starting with 1. Seen from Figure 4, although the final product 8 is quite stable, the relevant energy barrier (TS1-8) locates above the entrance channel by as high as 30.9 and 23.1 kcal/ mol in the Co+/MA and Ni+/MA reactions; thus, forming the dehydrogenation products via direct mechanism is not favorable under the reaction energy of 0.5 eV. The dehydrogenation processes via the “carriers” mechanism mean that M+ acts as “carriers” to ferry hydrogen atoms in isomerization processes.28,30,40 Starting with initial N-H activation, species 6 could be formed via M+ insertion into one of the N-H bonds of complex 1. This probability involves TS1-6, which lies only 0.5 and 0.1 kcal/mol above 6 in the Co+/MA and Ni+/MA reactions, as shown in Figure 4. This transition state is largely similar to species 6 in both geometry and energy, indicating that it is indeed a late transition state. Once species 6 is reached, a subsequent methyl-H shift would be followed along the reaction coordinate to form molecular-hydrogen complex 7, CH2NHM+ · · · H2. Seen from Table 1, the newly formed structure of CH2NHM+ · · · H2 is stabilized by the strong donation from the lone pair of N and σ(H1-H2) bonding orbital to the metal 4s orbital, and the donation from the metal 3d orbital to the σ*(H1-H2) antibonding orbital. Species 7 is found to be 47.1 and 52.0 kcal/mol below the respective entrance channel in the Co+/MA and Ni+/MA reactions. The involved TS6-7 in the methyl-H shift process lies 12.0 and 16.0 kcal/mol on their respective PESs. Direct dissociation of 7 would account for dehydrogenation product 8, CH2NHM+ + H2. The overall reaction of M+ + CH3NH2 f CH2NHM+ + H2 is exothermic by 32.4 (M ) Co) and 30.1 (M ) Ni) kcal/mol. Starting with initial C-H activation, complex 1 would convert to C-H insertion species 2 via TS1-2 along the reaction pathway with smooth PES as discussed above. Once species 2 is formed, a subsequent amino-H shift would be followed to form the molecular-hydrogen complex 7. Complexes 2 and 7 are connected by TS2-7, which is 7.7 and 4.7 kcal/mol above the entrance channel in the Co+/MA and Ni+/MA reactions. Direct dissociation of 7 would account for dehydrogenation product 8, as mentioned above. Seen from Figure 4, the PESs of the initial C-H activation pathway are lower and smoother than those for the initial N-H activation pathway, indicating the pathway of initial C-H activation is energetically favorable for the loss of H2, according well with our previous studies in the Co+/ethylamine reaction.30 Now we consider the relation between charge distributions and geometries appearing in the N-H activation process. Because of the saturation of covalent bond, such as in species 6 and TS6-7 shown in Table 2, the C atom is inhibited to further form a π bond with the N atom. This means that the CT process from the CH3NH entity to the MH+ entity would be impossible to take place, and the positive charge would be localized mainly on the MH (M ) Co, Ni) entity, as shown in Table 3.

J. Phys. Chem. A, Vol. 114, No. 47, 2010 12495 TABLE 3: Calculated Natural Charges (e) for All Species Involved in the Co+/MA and Ni+/MA Reactions at the NBO//B3LYP/6-311++(3df,2p) Levela charges +

charges +

+

species

Co /MA

Ni /MA

species

Co /MA

Ni+/MA

1 2 3 4 5 5′ 6 7 8

0.85 0.77 0.71 0.11 0.00 1.00 0.96 0.80 0.88

0.83 0.67 0.65 0.11 0.00 1.00 0.84 0.80 0.88

TS1-2 TS1-3 TS1-6 TS1-8 TS2-3 TS2-7 TS3-4 TS6-7

0.83 0.86 0.91 0.82 0.35 0.76 0.06 0.94

0.70 0.85 0.89 0.81 0.36 0.72 0.04 0.83

a Charges on M for 1, 7, 8, and TS1-8, and on MH for the others (M ) Co and Ni).

3.4. Comparison with Experimental Results. To bridge the gap between theory and experiment for the gas-phase chemistry of amine prototype with late first-row transition metal cations, we briefly compare our results with experimental observations in the following. In the hydride abstraction processes, the abstracted H atom would be expected to originate from the methyl and/or amino groups of amines. However, the facts that, in the ICR experiment, hydride abstraction with Co+ occurred in the case of triethylamine (which contains no amine-hydrogen atom) and methylamine (which contains only CR-H atom), but not for tertbutylamine (which has no CR-H atom),15 indicate that the hydrogen atom originated from the R-methylene group. Metastable ion decompositions of the D-labeled n-propylamine complexes with Co+ and Ni+ confirmed this point.41 Our results, from the theoretical viewpoint, give good support to the experimental findings and unravel the reasons why methyl-H abstraction is favorable. On one hand, the activation energy of the C-H bond is much lower than that of the N-H bond, indicating that M+ (M ) Co and Ni) insertion into the former bond is kinetically favorable. On the other hand, according to the relation between charge distributions and geometries, it is unable to further form a precursor of CH3NH+ · · · MH after the N-H activation (as 6; see Figures 1 and 2) because the CT process from CH3NH to MH+ is forbidden. This means that the positive charge is mainly localized on the MH+ entity and thus the overall reaction of M+ + CH3NH2 f CH3NH + MH+ is highly endothermic. In other words, the abstraction of amino-H atom is thermodynamically unfavorable. In the dehydrogenation processes, labeling experiments by virtue of Co+ reacting with C2H5ND2 (eliminated HD predominately) in the ICR studies let the authors propose a dehydrogenation mechanism of initial N-H insertion followed by R-methylene-H shift.15 The resulting product was thus conjectured to be Co+-imine complex.15 Our theoretical studies elaborate three possible mechanisms of the dehydrogenation from MA, that is, direct elimination from adjacent C and N group, initial N-H activation, and initial C-H activation. Among them, the most competitive channel is confirmed through a consecutive reaction of initial C-H activation, subsequent amino-H shift, and direct dissociation of precursor CH2NHM+ · · · H2. More interestingly, as shown in Figure 4, energy barrier for the initial N-H activation process appears at the amino-H shift transition state TS6-7, reaching 12.0 kcal/mol in the Co+/MA reaction and 16.0 kcal/mol in the Ni+/MA reaction. These energy barriers are higher than the relative kinetic energy of 0.5 eV (∼11.53 kcal/mol) used in the ion beam experiment.16 That is, both direct elimination from adjacent C and N groups and initial

12496

J. Phys. Chem. A, Vol. 114, No. 47, 2010

N-H activation mechanisms are energetically unfavorable for dehydrogenation, and the reaction starting with initial C-H activation would be the most competitive channel for the H2 loss from MA under reaction energy of 0.5 eV. Analyses based on the calculations not only provide detailed information to complement the existing experiments but also help to understand the mechanistic properties that facilitate new experiments with improved performances. As discussed above, our calculations show these factors, including the enthalpies of reaction, stability of 3, and competition between different channels, would account for the experimental results that the product yield of hydride abstraction in sequence is CoH < NiH < CuH.15,16,5 PESs analyses show that reaction channels with smoother pathways of NiH are more competitive than that of H2, identifying the experimental observations in the Ni+/MA reaction.16 In the Co+/MA reaction, the endothermic characteristics determines that the hydride abstraction is more unfavorable than that of dehydrogenation under lowenergy reaction condition.16 The relatively simpler and smoother pathways of CoH are more competitive than those of H2 if without considering the enthalpies of reaction, explaining the experimental results of 53% CoH and 47% H2 via the ICR experiment.15 On the other hand, the fact that complex 3 is the key species for the hydride abstraction, but it could readily convert to the global minimum 1, determines that using experimental apparatus with a rapid response, such as IRC,15 is preferable for the hydride abstraction compared to those with slow response like the ion beam16 or metastable ion spectrum.42 4. Conclusions The present DFT calculations shed new light on the experimental observations and provide a basis for understanding the gas-phase reaction mechanisms of methylamine with low-spin ground state Co+(3F) and Ni+(2D) in terms of the topology analysis of potential energy surfaces. Co+ and Ni+ perform similar roles along the isomerization processes to the final products. Hydride abstraction from methylamine takes place via the metal cation-methyl-H intermediate, followed by a charge transfer process, and a direct dissociation of CH2NH2+ · · · MH (M ) Co, Ni). Lower C-H activation energy and chargetransfer favored structure of CH2NH2+ · · · MH determine the hydride abstraction from methyl-H. The factors including the enthalpies of reaction, stability of metal cation-methyl-H species, and competition between different channels would account for the sequence of the hydride abstraction products, CoH < NiH < CuH. The most competitive channel for dehydrogenation reaction occurs through a stepwise reaction of initial C-H activation, a subsequent amino-H shift, and a direct dissociation of precursor CH2NHM+ · · · H2. Both direct elimination from adjacent C and N group and initial N-H activation mechanisms are energetically unfavorable for dehydrogenation. From a methodological viewpoint, the DFT calculations provide comprehensive description of reaction mechanisms, reliable analysis of deviations between different reactions, and effectively screening the experiments with improved performances. Acknowledgment. This work is supported by National Natural Science Foundation of China (21003158 and 10979077). We would also like to thank the City University of Hong Kong for the financial support. Supporting Information Available: Calculated total energies Es, ZPE (hartrees), and 〈S2〉 for all the species involved in the reaction of methylamine with M+ (M ) Co, Ni) at both the B3LYP/6-311++G(d,p) and B3LYP/6-311++G(3df,2p) levels,

Lu et al. and all possible encounter complexes obtained in the Co+/ ethylamine system as a reference for the reaction of methylamine with M+ (M ) Co, Ni). This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Wagenknecht, P. S.; Norton, J. R. J. Am. Chem. Soc. 1995, 117, 1841. (2) Richards, R. L. Coord. Chem. ReV. 1996, 154, 83. (3) Corbin, D. R.; Schwarz, S.; Sonnichsen, G. C. Catal. Today 1997, 37, 71. (4) Hoyau, S.; Norrman, K.; McMahon, T. B.; Ohanessian, G. J. Am. Chem. Soc. 1999, 121, 8864. (5) Sigsworth, S. W.; Castleman, J. A. W. J. Am. Chem. Soc. 1989, 111, 3566. (6) Eller, K.; Schwarz, H. Chem. ReV. 1991, 91, 1121. (7) Clemmer, D. E.; Armentrout, P. B. J. Phys. Chem. 1991, 95, 3084. (8) Chen, Y.-M.; Clemmer, D. E.; Armentrout, P. B. J. Chem. Phys. 1991, 95, 1228. (9) Chen, L. Z.; Miller, J. M. J. Am. Soc. Mass Spectrom. 1992, 3, 451. (10) Chen, Y.-M.; Clemmer, D. E.; Armentrout, P. B. J. Chem. Phys. 1993, 98, 4929. (11) Luna, A.; Amekraz, B.; Morizur, J.-P.; Tortajada, J.; Mo´, O.; Ya´n˜ez, M. J. Phys. Chem. A 1997, 101, 5931. (12) Alcamı´, M.; Mo´, O.; Ya´n˜ez, M. Mass Spectrom. ReV. 2001, 20, 195. (13) Aribi, H. E.; Rodriquez, C. F.; Shoeib, T.; Ling, Y.; Hopkinson, A. C.; Siu, K. W. M. J. Phys. Chem. A 2002, 106, 8798. (14) Alcamı´, M.; Luna, A.; Mo´, O.; Ya´n˜ez, M.; Tortajada, J.; Amekraz, B. Chem.sEur. J. 2004, 10, 2927. (15) Radecki, B. D.; Allison, J. J. Am. Chem. Soc. 1984, 106, 946. (16) Tolbert, M. A.; Beauchamp, J. L. J. Phys. Chem. 1986, 90, 5015. (17) Hoyau, S.; Ohanessian, G. Chem. Phys. Lett. 1997, 280, 266. (18) Luna, A.; Amekraz, B.; Tortajada, J. Chem. Phys. Lett. 1997, 266, 31. (19) Luna, A.; Alcamı´, M.; Mo´, O.; Ya´n˜ez, M. Int. J. Mass. Spectrom. 2000, 201, 215. (20) Shiota, Y.; Yoshizawa, K. J. Am. Chem. Soc. 2000, 122, 12317. (21) Luna, A.; Alcamı´, M.; Mo´, O.; Ya´n˜ez, M.; Tortajada, J. Int. J. Mass. Spectrom. 2002, 217, 119. (22) Galiano, L.; Alcamı´, M.; Mo´, O.; Ya´n˜ez, M. J. Phys. Chem. A 2002, 106, 9306. (23) Zhang, D. J.; Liu, C. B.; Bi, S. W.; Yuan, S. L. Chem.sEur. J. 2003, 9, 484. (24) Guo, W. Y.; Lu, X. Q.; Hu, S. Q.; Yang, S. H. Chem. Phys.Lett. 2003, 381, 109. (25) Liu, H. C.; Hu, Y. H.; Yang, S. H.; Guo, W. Y.; Lu, X. Q.; Zhao, L. M. Chem.sEur. J. 2005, 11, 6392. (26) Guo, W. Y.; Yuan, T.; Chen, X. F.; Zhao, L. M.; Lu, X. Q.; Wu, S. J. J. Mol. Struc (THEOCHEM) 2006, 764, 177. (27) Zhao, L. M.; Zhang, R. R.; Guo, W. Y.; Wu, S. J.; Lu, X. Q. ChemPhysChem 2006, 7, 1345. (28) Ma, Y.; Guo, W. Y.; Zhao, L. M.; Hu, S. Q.; Zhang, J.; Fu, Q. T.; Chen, X. F. J. Phys. Chem. A 2007, 111, 6208. (29) Lu, X. Q.; Guo, W. Y.; Zhao, L. M.; Chen, X. F.; Fu, Q. T.; Ma, Y. J. Organomet. Chem. 2007, 692, 3796. (30) Lu, X. Q.; Guo, W. Y.; Yang, T. F.; Zhao, L. M.; Du, S. C.; Wang, L.; Shan, H. H. J. Phys. Chem. A 2008, 112, 5312. (31) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (32) Zhao, L. M.; Zhang, R. R.; Guo, W. Y.; Wu, S. J.; Lu, X. Q. Chem. Phys. Lett. 2005, 414, 28. (33) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154. (34) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523. (35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision D.01; Gaussian, Inc.; Pittsburgh, PA, 2003.

Gas-Phase Reactions of Methylamine (36) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. NBO50; Theoretical Chemistry Institute, University of Wisconsin: Madison, WI, 2001. (37) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D. in: Linstrom, P. J. Mallard, W. G. Eds. NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Ion Energetics Data; National Institute of Standards and Technology: Gaithersburg, MD 20899, March 2003. (38) Kickel, B. L.; Armentrout, P. B. J. Am. Chem. Soc. 1995, 117, 764.

J. Phys. Chem. A, Vol. 114, No. 47, 2010 12497 (39) Yi, S. S.; Reichert, E. L.; Holthausen, M. C.; Koch, W.; Weisshaar, J. C. Chem.sEur. J. 2000, 6, 2232. (40) Alcamı´, M.; Luna, A.; Mo´, O.; Ya´n˜ez, M. J. Phys. Chem. A 2004, 108, 8367. (41) Karrass, S.; Priisse, T.; Eller, K.; Scbwarz, H. J. Am. Chem. Soc. 1989, 111, 9018. (42) Chen, L. Z.; Miller, J. M. Inorg. Chem. 1992, 31, 4029.

JP106397G