Abstract
Density functional theory has been employed to investigate the ligand effect in the reaction of ligated NiBr+ with propane. Both initial C–H and C–C bond activation mechanisms for losses of HBr, H2, and CH4 are analyzed in terms of the topology of the potential energy surface. Losses of HBr and H2 involve three C–H activation mechanisms, that is, α,β-H, α,γ-H, and β,α-H abstractions, where the last β,α-H abstraction is the most favorable mechanism. Loss of CH4 involves initial C–C activation, but it is prevented by the high-energy barrier. When propane reacts with the open-shell ligated NiBr+, the ligand of Br in the initial C–H activation could direct abstract a H atom from propane substrate via a four-center transition state, without forming multi-σ-type bonding of Ni+, whereas the metal center in the initial C–C activation needs to experience an unfavorable three σ-type bonding (with Br, CH3, and CH2CH3), which explains why HBr and H2 are formed in the reaction of BrNi+/C3H8 and CH4 not.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Transition metal ions are of fundamental interesting in chemical and biology researches in the field of catalytic process. Because transition metal ions contain incompletely filled d or f orbits, they can supply electrons to ligands and accept electrons from ligands and thus have favorably catalytic activity. In order to deeply understand their intrinsic binding properties and reactivities, over the past three decades, a huge number of papers have been devoted to the investigation of the reactions of transition metal ions with organic molecules in the gas phase by using the experimental and theoretical methods [1–7]. In particular, the gas-phase reaction of bare transition metal ions with alkanes has undergone extensive investigation [8–31], which involves in the first three transition periods of metal. Interestingly, further study found that the introduction of ligands could change the reaction activity of metal ion greatly. Closed-shell ligand L (such as carbonyl, alkene, and phosphine) can through the electronic supply or feedback effect change electronic structure of the metal ion and may bring spatial constraints to the metal ions. Open-shell ligand X (such as halogen, hydroxyl, and cyclopentadienyl) can also change the electronic states of the metal ions via formation of the polarized covalent bonds, thereby leading to the change of thermal chemical property, ionization energy, binding energy, and even spin state of the metal ions. Generally, closed-shell ligand L only has an effect on the reaction efficiency of metal ions, whereas the influence of open-shell ligand X is according to the actual situation.
In all kinds of open-shell ligands, halogen X (X = F, Cl, Br, and I) is well known as the prototype of the class of covalent ligands [32–35]. Because of the large electronegativity and poor donor properties, halogen could form strongly polarized covalent bond to metals, and thus, it controls the reaction efficiency largely and even changes the product of the reaction. For example, bare Cr+ is one of the least reactive transition metal ions, whereas CrCl+ is significantly more reactive [36], and CrF+ is even capable of dehydrogenating propane [37]. Under thermal conditions, both PtCl+ and PtCl2 + could react with methane giving rise to HCl, while in the case of PtBr+, formation of the corresponding carbine takes place. In order to understand the influence that ligand X (X = F, Cl, Br, and I) brings to the reactivity of NiX+/RH (R = H, C2H5, nC3H7, and nC4H9) system, Schlangen et al. [38] have studied the gas-phase reaction of NiX+/RH by using the electrospray ion (ESI) mass spectrum technology. They inferred that the reaction of NiBr+/C3H8 exclusively involves initial C–H activation for losses of HBr and H2, rather than initial C–C activation (for formation of CH4). In this paper, we select the NiBr+ + C3H8 reaction as a model of our theoretical study to discuss the C–H and C–C bond activation mechanisms (see Scheme 1). We hope to determine the essential factors that ligand controls the selectivity of the NiBr+/C3H8 reaction.
1.1 Computational details
The geometry optimization and frequency calculations were carried out for all the relevant species using the B3LYP [39, 40] functional together with the DZVP(d)(opt + 3f) basis set [41] for Ni+ and the 6-311++G(2d,2p) basis set [42] for nonmetal atoms. All the energies are reported with zero-point energy (ZPE) corrections with a scaling factor of 0.961 [43]. The DZVP(d)(opt) sets built up by Chiodo et al. have presented a good reliability for the B3LYP method in predicting transition metal ion ground- and excited-state order and splitting [41, 44]. The selected method has been proven to combine reasonable computational cost with accuracy sufficient for describing open-shell metal systems [43]. The intrinsic reaction coordinate (IRC) was performed to check the pathways between transition states and their connecting minima. Besides, the natural bond orbital (NBO) theory [45, 46] was used to shed new light on the characters of some important species involved. Dispersion energy was calculated to evaluate the dispersion contribution using ωB97X-D method with the DZVP(d)(opt + 3f): 6-311++G(2d,2p) basis set [47–49]. We confirmed the minimum energy crossing point (MECP) via single-point energy calculation of both considered states at the DZVP(d)(opt + 3f): 6-311++G(2d,2p) level for the relevant IRC points along the singlet pathway until they reach an equal energy. All these calculations were performed using the Gaussian 09 package [50].
The calculation methods of spin–orbit coupling (SOC) and crossing probability at the MECP, which has been described previously [51], are presented in detail in supporting information (SI). In brief, by using the GAMESS package [52], CASSCF calculations were first performed for both states at the MECP to get the converged CASSCF wave functions. The SOC value is the matrix element that expresses the coupling of the singlet and triplet states by the operator of Eq. 1 [53],
Here, 1 Ψ 1 (3 Ψ 2) is the M S (M S′ ) component of the many-body singlet-state (triplet-state) wave function. Considering the generated spin sublevels M S , a reasonable measure of the SOC-induced triplet–quintet interaction is the root-mean-square coupling constant (SOCC) of Eq. 2.
A crude estimation of the crossing probability at the MECP can be done using the Landau–Zener formula [54–56],
where V ij is a matrix element of a diabatic operator (SOC in this case) coupling two adiabatic states i and j, Δg ij is the difference in the gradients of the two adiabatic states i and j, and v is the effective velocity with which the system is passing through the crossing point that can be calculated from the kinetic theory of gases at 298 K.
2 Results and discussion
In the following sections, we will first establish the accuracy that is expected from the chosen level of theory for the NiBr+/C3H8 system. Then, we will present structures and energies of reactants, products, and encounter complexes. Last, we will examine the title reaction in detail, including geometries of various stationary points and PESs. For simplicity, optimized geometries, selected structural parameters, calculated total energies, zero-point energies as well as 〈S 2〉 values for all the species involved in the reaction are given in SI.
2.1 Calibration
To evaluate the reliability of the level of theory employed, we compare the experimental thermochemical data with the results from the B3LYP approach. Table 1 collects the theoretically predicted adiabatic bond dissociation energies (BDEs) and the most reliable experimental data for some relevant species [57–69].
As shown in Table 1, the calculated excitation energy of Ni+ (2D(d 9) → 4F(d 8 s 1)) agrees well with the experimental gap value (20.1 vs 24.9 kcal/mol [57]). Moreover, the theoretical BDEs of H–Br and H–C3H7 accord well with the experimental findings. Our calculated bonding in Ni+–X (X = H, CH2, CH4, C2H4, and C2H6) is overestimated by about 4–7 kcal/mol, which may be considered as a systematic shortcoming of the B3LYP functional. It should be noted that the calculated BDE of Ni+–Br is underestimated by 13.4 kcal/mol based on the datum derived by Allison and Ridge [69]. Thus, the relative stability of the HBr + [C3, H7, Ni]+ asymptote with respect to the remainder of the [C3, H8, Br, Ni]+ potential energy surface (PES) is probably underestimated by about 13 kcal/mol. However, the relative energies of the other parts of the PES are still described more satisfactorily. Such a situation has also been found for the PESs of the Co+/C2H6, Co+/C2H6/N2O, Fe+/C2H6, and Fe+/C2H6/N2O systems calculated using density functional theory (DFT) [70–73].
To evaluate the dispersion contribution in the NiBr+/C3H8 system, dispersion energies (DBs) were calculated using ωB97X-D method [47–49]. Calculated DB values for the species involved in the α,γ-H abstraction mechanism are given in Table S2. It is found that the DBs of NiBr+, H2, and HBr are zero nearly, while the values of C3H8 and other product ions ([NiCH2CH2CH3]+ (1 Ia) and [NiBrCH2CH2CH2]+ (Ja)) are 2.5 and 3.0–3.6 kcal/mol, respectively. For the intermediates and transition states involved, the DBs are calculated to be 3.8–4.6 kcal/mol. All these suggest that the dispersion contribution is relatively small and has a negligible effect on the relative energies of the [C3, H8, Br, Ni]+ PES.
2.2 Reactants, encounter complexes, and products
In this section, we will discuss the structures and energies of reactants, encounter complexes, and products involved in the reaction of NiBr+/C3H8. The optimized geometric configurations and structural parameters are shown in Fig. 1.
For bare Ni+, it has a doublet ground state with the 2D(d 9) → 4F(d 8 s 1) excitation energy of 20.1 (calcd) and 24.9 (expt) kcal/mol [57]. However, NiBr+ bears a high-spin ground state (triplet). NBO analysis suggests that Ni+ and Br form strong polarized covalent bonds through the 4s3d(Ni+) and 4s4p(Br) orbits. The bond in triplet comprises one σ(Ni+Br) and one β–π(Ni+Br) binding orbitals, whereas the singlet state only forms one σ(Ni+Br) binding orbital, which leads to a short Ni–Br+ equilibrium distance [2.156 (triplet) vs 2.188 (singlet) Å] and strong diabatic BDE (74.2 vs 14.8 kcal/mol) of the triplet species and a large triplet → singlet excitation energy of 39.1 kcal/mol.
Three encounter complexes (1a, 1b, and 1c) are formed in the reaction of NiBr+/C3H8. 1a is featured by a BrNi+(η 4–C3H8) structure, where Ni+ is simultaneously coordinated to 2α,2γ-H (noted as BrNi+(η 4–2α,2γ-H–C3H8)). Both 1 1b and 1 1c possess a BrNi+(η 2–α,β-H–C3H8) structure, in which the BrNi+ axis inclines to Hα and Hβ, respectively, while 3 1b and 3 1c bear a respective BrNi+(η 3–α,2β-H–C3H8) and BrNi+(η 3–2α,β-H–C3H8) structure. Energetically, the diabatic BDEs of 1a, 1b, and 1c are calculated to be in the range of 57.2–59.4 kcal/mol (singlet) and 43.5–41.3 kcal/mol (triplet), indicating the nearly identical stability as well as the coexistence for these isomers in the gas phase.
NBO analysis shows that the association of NiBr+ with C3H8 (1a, 1b and 1c) is dominated by the electrostatic interaction as well as donor–acceptor stabilization, i.e., electrons donation from the coordinated σ(CH) orbitals to unoccupied 4s3d(Ni+) and σ*(Ni+Br) orbits as well as the back-donation from the filled 4s3d(Ni+) orbit to the σ *(CH) orbit [∆E (2) = 148.6 (90.2) (1a), 147.4 (77.7) (1b), and 143.7 (73.1) (1c) kcal/mol in the singlet (triplet)]. This results in the weakening interaction of the corresponding C–H bond and in preparation for the next step of hydrogen transfer. Furthermore, the much stronger donor–acceptor interaction in the singlet explains its stronger diabatic BDE.
Ni(C3H7)+ is the product of HBr elimination in the reaction of NiBr+/C3H8. It contains three kinds of isomers, i.e., Ia and Ib via α-H abstraction and Ic via β-H abstraction, respectively. Ia is featured by cis-Ni+(η 2–α-C,γ-H–CH2CH2CH3), forming a five-membered structure, while Ib bears a trans-Ni+(η 1–α-C–CH2CH2CH3) structure. For Ic, the metal ion strongly binds to the radical Cβ atom of CH3CHCH3 (noted as Ni+(η 1–β-C–CH2CH2CH3)). Although the nearest Ni+–C distances in 3 Ia − 3 Ic (1.93–1.97 Å) are calculated to be longer than those in the singlet (1.86–1.87 Å), the triplet diabatic BDEs are larger than those of the singlet [84.0 vs 61.4 (Ia), 81.0 vs 63.3 (Ib), 80.5 vs 64.5 (Ic) kcal/mol]. This is because that the diffuse single-occupied 4s orbital in Ni+(4F(d 8 s 1)) favors forming a relative strong covalent bond with the radical-C atom of C3H7. NBO analysis shows that the Ni+–(C3H7) association is dominated by the covalent interaction via a σ(Ni+C) binding orbital, which is consisted by the 2p(radical-C) orbital and the 3d(Ni+) orbital in the singlet and the α-4s and β-3d(Ni+) orbitals in the triplet.
For the H2-loss product, two candidates can be envisaged: BrNi(CH2CH2CH2)+ complex (Ja) via α,γ-H abstraction and BrNi(CH2CHCH3)+ complex (Jb) via α,β- and β,α-H abstractions. In 1 Ja, Ni+ of NiBr+ simultaneously attacks at both terminal C atoms of CH2CH2CH2 (with Ni+–C distance of 1.883 Å), forming a quadrilateral-type structure. In the triplet, however, the C3H6 ligand forms a cyclopropane structure and NiBr+ is coordinated to one side of cyclopropane. The corresponding diabatic BDE is calculated to be 74.0 and 51.2 kcal/mol in its singlet and triplet states, respectively. NBO analysis suggests that the metal ion in the singlet state forms two strong polarized covalent bonds with two-coordinated C atoms, while the association of NiBr+ with C3H6 in the triplet state is dominated by electrostatic interaction as well as donor–acceptor stabilization (∆E (2) = 65.3 kcal/mol). The triplet BrNi+(CH2CHCH3) (Jb) is featured by an alkenyl-CC π-type bound structure (with Ni+–C distances of 2.093 and 2.402 Å) (∆E (2) = 70.4 kcal/mol), while the singlet counterpart favors a multicenter association of metal ion with Cα, Cβ, and Hγ of C3H6 (with distances of 2.053, 1.942, and 1.792 Å, respectively), suggesting the stronger interactions of metal with both the alkenyl-π and σ(CγH) orbitals (∆E (2) = 115.6 and 41.9 kcal/mol, respectively). Thus, the diabatic BDE of the singlet (81.7 kcal/mol) is stronger than that of the triplet (55.4 kcal/mol).
BrNi(CH2CH2)+ is the C–C activation product via loss of methane. In both multiplicities, the species have the C s symmetry and NiBr+ attaches the two ethylene C atoms simultaneously through the metal with the difference that the NiBr+ bond in the triplet is co-plane with the Ni+CC plane, whereas the singlet association favors an out-of-plane location of the NiBr+ bond (the dihedral angle of NiBr+ with the Ni+CC plane is 94.9°). This association results in elongation of the C–C bond as well as distortion of the C–C–H angle (especially for the singlet species). NBO analysis indicates that the complex is stabilized by the electron donor–acceptor interaction, i.e., donation of π(CC) → 4s3d(Ni+) and σ*(Ni+Br) as well as back-donation of 4s3d(Ni+) → π*(CC) (∆E (2) = 149.9 (singlet) and 27.6 (triplet) kcal/mol). Compared to the triplet species, the larger donor–acceptor interaction in the singlet leads to a stronger diabatic BDE [69.8 (singlet) vs 48.2 (triplet)] as well as a larger change of structure.
2.3 Gas-phase reaction mechanism
In this section, we will discuss all possible mechanisms for the H2, HBr, and CH4 eliminations in the gas-phase reaction of NiBr+ with C3H8.
2.3.1 Initial C–H activation
Three possible initial C–H activation pathways for losses of H2 and HBr have been surveyed, that is, α,β-H, α,γ-H, and β,α-H abstractions. PESs together with the schematic structures involved in the reaction pathways are shown in Figs. 2, 3 and 4.
α,γ-H abstraction As shown in Fig. 2, α,γ-H abstraction starts from encounter complex 1a. 1a could involve into species 2a ((HBr)Ni(CH2CH2CH3)+) via a σ-complex-assisted metathesis (σ-CAM) mechanism [74], which experiences a direct H-transfer from the metal-coordinated propane to the Br ligand via a four-center transition state (TS 1a–2a ). In this process, approaching of the metal side of NiBr+ to C3H8 induces a preferential attack of Cα–H bond and promotes the cleavage of the Cα–H bond, subsequently through direct Cα– to –Br hydrogen transfer 2a is formed. The H–Br–Ni–Cα four-centered structure of TS 1a–2a favors forming the σ(Ni+Cα) and σ(HBr) [σ(Ni+Cα), σ(HBr), and β-σ(Ni+Br)] binding orbitals in the singlet (triplet) state. Furthermore, the system is stabilized by strong electron donation of 4p(Br) and σ(HBr) → 4s*(Ni+) and σ*(Ni+Cα) and back-donation of 3d(Ni+) and σ(Ni+Cα) → σ*(HBr) (∆E (2) = 279.1 (singlet) and 69.3 (triplet) kcal/mol). It can be found that Ni+ in TS 1a–2a forms one (singlet) and two (triplet) covalent bonds with the ligand and avoids binding with multi (three)-σ-type covalent ligands (for example, (Br)Ni+(H)(C2H5)). Thus, it is an energetically favorable process (E rel(TS 1a–2a ) = −11.4 (singlet) and −6.1(triplet) kcal/mol).
Species 2a is featured by a two-coordinated structure ((HBr)Ni+(η 2–α-C,γ-H–CH2CH2CH3)). Different with the binding situation in NiBr+ and BrNi+(C3H8) (1a–1c), the metal center in 2a involves two σ-type bonding, for which perfects pairing of electrons in preferred in the low-spin (singlet) ground state. Further NBO analysis shows that Ni+ in the singlet forms strong covalent bonds with both C3H7 and HBr groups via the σ(Ni+Cα) and σ(Ni+Br) binding orbitals, respectively, whereas in the triplet, the metal center only forms the covalent bond with C3H7 (via σ(Ni+Cα)) and the binding of Ni+–(HBr) is dominated by donor–acceptor stabilization (∆E (2) = 72.0 kcal/mol). Thus, 1 2a is 6 kcal/mol more stable than 3 2a and 24.9 kcal/mol below the reactants 3NiBr+ + C3H8.
It can be found from Fig. 2 that a triplet-to-singlet crossing occurs in the 3 1a → 1 TS 1a–2a process. For the relevant MECP 1a–2a , the activated C–H bond is stretched to 1.184 Å, and the newly formed H–Br bond length is reduced to 2.237 Å (see Fig. S1). The SOC constant (SOCC) of MECP 1a–2a is calculated to be 133.8 cm−1, and the crossing probability is estimated to be 59 % at the room temperature.
One exit of 2a is the direct dissociation into Ni(CH2CH2CH3)+ (Ia) and HBr, with the whole endothermicity of 2.8 (2.5) kcal/mol on the singlet (triplet) PES. The other is yielding 3a via a Hγ transfer to metal center. The corresponding transition state TS 2a–3a is located at −4.0 (19.9) kcal/mol on the singlet (triplet) PES. Species 3a bears a (HBr)Ni+(H)(η 2–α,γ-C–CH2CH2CH2) structure, which is located at −4.2 (singlet) and 0.6 (triplet) kcal/mol. NBO analysis shows that Ni+ in 1 3a forms strong covalent bonds with all three coordinated ligands via the σ(Ni+Br), σ(Ni+H), σ(Ni+Cα), and σ(Ni+Cγ) binding orbitals, whereas in 3 3a, metal center only binds covalently with the H and HBr groups via the σ(Ni+H) and β–σ(Ni+Br) binding orbitals, respectively, and the association of Ni+–(C3H6) is stabilized by donor–acceptor interaction (∆E (2) = 46.3 kcal/mol).
Along reaction coordination, 3a could involve into complex 4a ((Br)Ni+(H2)(η 2–α,γ-C–CH2CH2CH2)) via a direct hydrogen coupling between the HBr and Ni+H groups. The transition state TS 3a–4a in this process lies at 3.7 (6.8) kcal/mol in the singlet (triplet) state. Both two spin structures of 4a have C s symmetry, and two H atoms that absorbed on Ni+ have formed covalent bond. The difference is that the H–H bond is co-plane with the Br–Ni+–Cβ symmetry plane in the singlet and perpendicular in the triplet. Different with a low-spin ground state of 3a, 3 4a (E rel = −17.5 kcal/mol) is 14.6 kcal/mol more stable than 3 2a. Subsequently, 4a could form BrNi(CH2CH2CH2)+ (Ja) and H2 via a direct dissociation. The overall reaction is endothermic by 2.5 kcal/mol on the singlet PES and exothermic by 13.9 kcal/mol on the triplet PES.
α,β-H abstraction α,β-H abstraction begins with 1b (see Fig. 3). NiBr+ attacks Cα–H bond of C3H8 in 1b induces H shift from Cα to Br, yielding 2b. 2b is also featured by a two-coordinated structure [(HBr)Ni+(η 1–α-C–CH2CH2CH3)], which favors a low-spin (singlet) ground state (E rel = −28.6 (singlet) and −15.5 (triplet) kcal/mol). The corresponding transition state TS 1b–2b is located at −15.2 and −2.5 kcal/mol on the singlet and triplet PESs, respectively. Thus, a triplet-to-singlet crossing is expected to occur immediately before TS 1b–2b . The relevant MECP 1b–2b is located when the activated C–H bond is stretched to 1.204 Å, and the newly formed H–Br bond length is reduced to 2.181 Å. The SOCC of MECP 1b–2b is calculated to be 127.6 cm−1, and the crossing probability is found to be 64 %.
Direct decomposition of (HBr)–Ni+(CH2CH2CH3) accounts for Ni(CH2CH2CH3)+ (Ib) + HBr, but it is a slight endothermic channel with a reaction heat of 0.9 (3.0) kcal/mol on the singlet (triplet) PES. Alternatively, direct Cβ-to-Ni+ H-shift from 2b could yield species 3b ((HBr)Ni+(H)(η 2–α,β-C–CH2CHCH3)). In both multiplicities, the 3d(Ni+) and 1s(H) orbitals form a σ(Ni+H) binding orbital, while the linkage of metal center with the other two groups (HBr and CH2CHCH3) is strengthened by the strong electron donation from the 4s4p(Br) and π(CC) orbitals to 4s4p(Ni+) orbitals as well as back-donation from the 3d(Ni+) orbital to the π*(CC) orbital (∆E (2) = 256.7 (singlet) and 150.8 (triplet) kcal/mol). Energetically, species 3b is located at −25.1 (−16.2) kcal/mol on the singlet (triplet) PES, while the relevant transition state (TS 2–3b ) lies at −21.8 (−5.1) kcal/mol.
Subsequently, species 3b involves into molecular hydrogen complex 4b ((Br)Ni+(H2)(η 2–α,β-C–CH2CHCH3)). Mainly, electrostatic interaction as well as donor–acceptor stabilizations for the (H2)–BrNi+–(CH2CHCH3) association results in a triplet ground state for 4b, similar to NiBr+ (excitation energy for 3 4b → 1 4b: 17.3 kcal/mol). The situation of transition state TS 3b–4b is a compromise between 3b and 4b, where the triplet transition state lies above the singlet one by 9.2 kcal/mol. Loss of H2 from 4b accounts for BrNi+(CH2CHCH3) (Jb), exothermic by 15.9 (28.7) kcal/mol in the singlet (triplet) state.
β,α-H abstraction β,α-H abstraction pathway starts with 1c (see Fig. 4). Direct Cβ– to –Br hydrogen shift could carry 1c into 2c ((HBr)Ni+(η 1–β-C–CH3CHCH3)), which also favors a singlet ground state [E rel = −32.7 (singlet) and −22.8 (triplet)]. The relevant transition state TS 1c–2c is located at −17.8 (−3.8) kcal/mol on the singlet (triplet) PES. As shown in Fig. 4, a triplet-to-singlet crossing occurs in the 3 1c → 1 TS 1c–2c process. The activated C–H bond and newly formed H–Br bond in the relevant MECP 1c–2c are calculated to be 1.221 and 2.243 Å, respectively (see Fig. S1). The SOCC of MECP 1c–2c is calculated to be 134.3 cm−1, and the crossing probability is estimated to be 73 % at the room temperature, indicating that the β,α-H abstraction accounts for the most large crossing probability in all three H abstraction mechanisms.
Direct loss of HBr from 2c accounts for Ni(CH3CHCH3)+ (Ic), which is exothermic by 4.2 kcal/mol on the singlet PES and endothermic by 3.0 kcal/mol on the triplet PES. The other exit of 2c is α-H shift to metal center forming 3c. Similar to 3b, 3c is featured by a (HBr)Ni+(H)(η 2–α,β-C–CH2CHCH3) structure and bears singlet as ground state (E rel = −25.1 (singlet) and −14.8 (triplet) kcal/mol). The relevant transition state TS 2c–3c is located at −23.5 (−12.1) kcal/mol on the singlet (triplet) PES. Subsequently, direct H coupling between the BrH and Ni+H groups carries 3c into 4b via transition state TS 3c–4c (lying at −12.1 (singlet) and −8.9 (triplet) kcal/mol). Direct loss of H2 from 4b forming BrNi(CH2CHCH3)+ (Jb) has been discussed above.
From Figs. 2, 3, 4 and 5, we can find that all of the α,β-, α,γ-, and β,α-H abstraction mechanisms could account for losses of HBr and H2. Two spin crossing occurs in all three mechanisms, that is, initial C–H activation and hydrogen coupling. The probability of the former crossing is calculated to be 64, 59, and 73 % for the α,β-, α,γ-, and β,α-H abstraction mechanisms, respectively, indicating that the triplet pathway can easily proceed to the singlet surface near the intersection, which reduced the reaction energy barrier. In all three H abstraction mechanisms, the most favorable mechanism is the β,α-H abstraction because it is not only energetically favorable but could also take place adiabatically on both the singlet and triplet PESs. If the underestimate of the stability of the HBr + [C3, H7, Ni]+ exit channel (about 13 kcal/mol) is considered, loss of HBr is also a strong exothermic process for all three H abstraction mechanisms. Furthermore, the HBr-loss pathway is much simple compared to that of loss of H2. Therefore, the HBr/H2 branching ratio of 58:42 is determined by the ESI mass spectrum experiment [38].
2.4 Initial C–C activation
We also considered the C–C activation pathway to loss of CH4. PES together with schematic structures involved in the reaction is shown in Fig. 5. This channel involves encounter complex 1b. Starting with 1b, the nickel ion inserts into one C–C bond of propane forming species 2d ((CH3)Ni+(Br)(CH2CH3)) via transition state TS 1b–2d . In this process, metal center involves three σ-type bonding (with Br, CH3, and CH2CH3), which is unfavorable for Ni+(2D(d 9) and 4F(d 8 s 1). NBO analysis indicates that Ni+ in both 1 TS 1b–2d (3 TS 1b–2d ) and 1 2d (3 2d) has a high population of 4p orbital (3d 8.724s 0.294p 0.53 (3d 8.464s 0.484p 0.25) and 3d 8.724s 0.354p 0.39 (3d 8.594s 0.364p 0.43)). Therefore, it is a high-energy pathway [E rel(TS 1b–2d and 2d) = −2.0 (+8.6) and −20.2 (−3.2) kcal/mol in the singlet (triplet)]. PES shows that a triplet-to-singlet crossing emerges before TS 1b–2d . The relevant SOCC is calculated to be 116.7 cm−1 with the crossing probability of 45 %.
A subsequent β-H shift to methyl could form species 3d via transition state TS 2b–3d (E rel = −6.4 (singlet) and −1.6 (triplet) kcal/mol). Species 3d bears a (CH4)Ni+(Br)(η 2–C,C–CH2CH2) structure, lying at −44.3 and −31.9 kcal/mol on the singlet and triplet PES, respectively. 3d undergoes direct bond cleavage forming CH4 and BrNi(C2H4)+, with the whole exothermicity of 15.9 (33.4) kcal/mol on the singlet (triplet) PES.
As shown in Fig. 5, we can found that although loss of CH4 is an exothermic process, it needs to overcome a high-energy barrier in the initial C–C activation [E rel(TS 1b–2d ) = −2.0 (+8.6) kcal/mol in the singlet (triplet)], which is 15.8 (12.4) kcal/mol higher than that of the initial Cβ-H abstraction (TS 1c–2c ). As it is well known from the Arrhenius formula for the evaluation of kinetics, a difference of about 10 kcal/mol between two barriers implies that rates change by several orders of magnitude. Thus, it is quite clear that why HBr and H2 are formed in the BrNi+/C3H8 reaction and CH4 not.
It is well known that Ni+ has a ground and excited electron configuration of 2D(d 9) and 4F(d 8 s 1), respectively, suggesting that it is unfavorable forming multi-σ-type bonding with other covalent ligands. When introducing an open-shell Br ligand to Ni+, the initial C–C activation of propane by BrNi+ needs to experience an unfavorable three σ-type bonding (with Br, CH3, and CH2CH3), resulting in the reaction being restrained. However, the initial C–H activation favors a σ-CAM mechanism, where the ligand of Br could direct abstract a H from the propane substrate through a four-center transition state, without forming multi-σ-type bonds of metal center, explaining the favorable C–H activation for the HBr and H2 losses in the reaction of BrNi+/C3H8.
3 Conclusions
DFT has investigated the reaction of NiBr+/C3H8, which add new insight into the experimental observations. We can make a conclusion by summarizing some of the main points in the following: Both HBr and H2 could be formed via three initial C–H activation mechanisms, that is, α,β-, α,γ-, and β,α-H abstractions, where the most favorable mechanism is the β,α-H abstraction. Losses of HBr and H2 are favored by undergoing a four-center transition state in the initial C–H activation process, avoiding forming multi-σ-type bonding of metal center. However, loss of CH4 is prevented by experience unfavorable three σ-type bonding of Ni+ (with Br, CH3, and CH2CH3) in the initial C–C activation process.
4 Supporting information
Detailed description of spin–orbit coupling calculations, geometries and selected structural parameters of MECPs, and optimized geometries, selected structural parameters, calculated energies, zero-point energies, and 〈S 2〉 for all species involved in the reaction of NiBr+ with propane.
References
Russo N, Sicilia E (2001) Reaction of Sc+(1D, 3D) with H2O, NH3, and CH4: a density functional study. J Am Chem Soc 123(11):2588–2596
Chen Q, Chen HP, Kais S, Freiser B (1997) Gas-phase reactions of Fe(CH2O)+ and Fe(CH2S)+ with small alkanes: an experimental and theoretical study. J Am Chem Soc 119(52):12879–12888
Sicilia E, Russo N (2002) Theoretical study of ammonia and methane activation by first-row transition metal cations M+ (M = Ti, V, Cr). J Am Chem Soc 124(7):1471–1480
Michelini MC, Russo N, Sicilia E (2007) Gas-phase chemistry of actinides ions: new insights into the reaction of UO+ and UO2 + with water. J Am Chem Soc 129(14):4229–4239
Irigoras A, Elizalde O, Silanes I, Fowler JE, Ugalde JM (2000) Reactivity of Co+(3F,5F), Ni+(2D,4F), and Cu+(1S,3D): reaction of Co+, Ni+, and Cu+ with water. J Am Chem Soc 122(1):114–122
Irigoras A, Fowler JE, Ugalde JM (1999) Reactivity of Cr+(6S,4D), Mn+(7S,5S), and Fe+(6D,4F): reaction of Cr+, Mn+, and Fe+ with water. J Am Chem Soc 121(37):8549–8558
Chiodo S, Rivalta I, Michelini MC, Russo N, Sicilia E, Ugalde JM (2006) Activation of methane by the iron dimer cation. A theoretical study. J Phys Chem A 110(45):12501–12511
Sunderlin LS, Armentrout PB (1988) Methane activation by Ti+: electronic and translational energy dependence. J Phys Chem 92(5):1209–1219
MacMahon TJ, Jackson TC, Freiser BS (1989) A gas-phase study of FeSn+ (n = 1–6). J Am Chem Soc 111(2):421–427
Sodupe M, Charles WBJ (1991) Theoretical study of the bonding of the first- and second-row transition-metal positive ions to acetylene. J Phys Chem 95(22):8640–8645
Blomberg MRA, Siegbahn PEM, Nagashima U, Wennerberg J (1991) Theoretical study of the activation of alkane carbon–hydrogen and carbon–carbon bonds by different transition metals. J Am Chem Soc 113(2):424–433
Perry JK, Ohanessian G, Goddard WA (1993) Molecular complexes of small alkanes with Co+. J Phys Chem 97(20):5238–5245
Chen YM, Armentrout PB (1995) Guided ion beam studies of the reactions of Ag+ with C2H6, C3H8, HC(CH3)3 and c-C3H6. J Phys Chem 99(29):11424–11431
Hornung G, Schrijder D, Schwarz H (1995) Diastereoselective gas-phase carbon–carbon bond activation mediated by “bare” Co+ cations. J Am Chem Soc 117(31):8192–8196
Cornehl HH, Heinemann C, Schroder D, Schwarz H (1995) Gas-phase reactivity of lanthanide cations with hydrocarbons. Organometallics 14(2):992–999
Chen YM, Armentrout PB (1995) Activation of C2H6, C3H8, and c-C3H6 by gas-phase Rh+ and the thermochemistry of Rh-ligand complexes. J Am Chem Soc 117(36):9291–9304
Holthausen MC, Koch WA (1996) Theoretical view on Co+-mediated C–C and C–H bond activations in ethane. J Am Chem Soc 118(41):9932–9940
Noll RJ, Yi SS, Weisshaar JC (1998) Bimolecular Ni+(2D5/2) + C3H8 reaction dynamics in real time. J Phys Chem A 102(2):386–394
Yi SS, Blomberg MRA, Siegbahn PEM, Weisshaar JC (1998) Statistical modeling of gas-phase organometallic reactions based on density functional theory: Ni++ C3H8. J Phys Chem A 102(2):395–411
Blomberg M, Yi SS, Noll RJ, Weisshaar JC (1999) Gas-phase Ni+ (2D5/2)+ n-C4H10 reaction dynamics in real time: experiment and statistical modeling based on density functional theory. J Phys Chem A 103(36):7254–7267
Fedorov DG, Gordon MS (2000) A theoretical study of the reaction paths for cobalt cation + propane. J Phys Chem A 104(11):2253–2260
Moc J, Fedorov DG, Gordon MS (2000) A theoretical study of the reaction of Ti+ with ethane. J Chem Phys 112(23):10247–10258
Li FX, Zhang XG, Armentrout PB (2006) The most reactive third-row transition metal: guided ion beam and theoretical studies of the activation of methane by Ir+. Int J Mass Spectrom 255–256:279–300
Armentrout PB (2006) Activation of CH4 by gas-phase Mo+, and the thermochemistry of Mo-ligand complexes. J Phys Chem A 110(27):8327–8338
Armentrout PB, Shin S, Liyanage R (2006) Guided-ion beam and theoretical study of the potential energy surface for activation of methane by W+. J Phys Chem A 110(4):1242–1260
Armentrout PB (2007) Activation of C2H6 and C3H8 by gas-phase Mo+: thermochemistry of Mo-ligand complexes. Organometallics 26(23):5473–5485
Armentrout PB (2007) Activation of C2H6 and C3H8 by gas-phase Mo+: potential energy surfaces and reaction mechanisms. Organometallics 26(23):5486–5500
Parke LG, Hinton CS, Armentrout PB (2007) Experimental and theoretical studies of the activation of methane by Ta+. J Phys Chem C 111(48):17773–17787
Shayesteh A, Lavrov VV, Koyanagi GK, Bohme DK (2009) Reactions of atomic cations with methane: gas phase room-temperature kinetics and periodicities in reactivity. J Phys Chem A 113(19):5602–5611
Lv LL, Wang YC, Geng ZY, Si YB, Wang Q, Liu HW (2009) Activation of C2H6 by gas-phase Ta+: potential energy surfaces, spin-orbit coupling, spin-inversion probabilities, and reaction mechanisms. Organometallics 28(21):6160–6170
de Almeida KJ, Duarte HA (2010) Dehydrogenation of methane by gas-phase Th, Th+, and Th2+: theoretical insights into actinide chemistry. Organometallics 29(17):3735–3745
Mandich ML, Steigerwald ML, Reents WD (1986) The effects of chloro substitution on the electronic-structure of ClCr+, ClMn+, and ClFe+ and their reactivity with small alkanes. J Am Chem Soc 108(20):6197–6202
Mazurek U, Schroder D, Schwarz H (1998) Generation and reactivity of chromium fluoride cations (CrF +n , n = 0–4) in the gas phase. Collect Czech Chem C 63(9):1498–1512
Roithova J, Schroder D (2007) Bimolecular reactions of molecular dications: new reactivity paradigms and bond-forming processes. Phys Chem Chem Phys 9(19):2341–2349
Schroder D, Hrusak J, Schwarz H (1993) Ligand effects on the reactivity of iron (II) cations FeX+ in the gas phase. Ber Bunsenges Phys Chem 97(9):1085–1090
Mandich ML, Steigerwald ML, Reents WD (1986) The effects of chloro substitution on the electronic-structure of ClCr, ClMn, and ClFe and their reactivity with small alkanes. J Am Chem Soc 108(20):6197–6202
Mazurek U, Schroder D, Schwarz H (1998) Generation and reactivity of chromium fluoride cations (CrFn, n = 0–4) in the gas phase. Coll Czech Chem Comm 63:1498–1512
Schlangen M, Schwarz H, Schroder D (2007) Ligand and substrate effects in gas-phase reactions of NiX+/RH Couples (X = F, Cl, Br, I; R = CH3, C2H5, n-C3H7, n-C4H9). Chem Eur J 13(24):6810–6816
Becke AD (1993) Density-functional thermochemistry. III. The role of exact exchange. J Chem Phys 98(7):5648–5652
Lee C, Yang W, Parr RG (1988) Development of the Colle–Salvetti correlation-energy formula into a functional of the electron density. Phys Rev B 37:785–789
Chiodo S, Russo N, Sicilia E (2005) Newly developed basis sets for density functional calculations. J Comput Chem 26:175–183
Frisch MJ, Pople JA, Binkley JS (1984) Self-consistent molecular orbital methods 25. Supplementary functions for Gaussian basis sets. J Chem Phys 80:3265–3269
Zhao LM, Liu ZC, Guo WY, Lu XQ, Lin XQ, Shan HH (2008) Mechanisms for the Ni+-mediated oxidation of benzene to phenol by N2O. Chem Phys Lett 463:54–59
Rondinelli F, Russo N, Toscano M (2007) On the origin of the different performance of iron and manganese monocations in catalyzing the nitrous oxide reduction by carbon oxide. Inorg Chem 46:7489–7493
Glendening ED, Reed AE, Carpenter JE, Weinhold F NBO Version 3.1
Reed AE, Curtiss LA, Weinhold F (1988) Intermolecular interactions from a natural bond orbital, donor–acceptor viewpoint. Chem Rev 88(6):899–926
Grimme S (2011) Density functional theory with London dispersion corrections. WIREs Comput Mol Sci 1(2):211–228
Chai JD, Head-Gordon M (2008) Long-range corrected hybrid density functionals with damped atom–atom dispersion corrections. Phys Chem Chem Phys 10:6615–6620
Stefan G (2006) Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J Comput Chem 27(15):1787–1799
Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Scalmani G, Barone V, Mennucci B, Petersson GA, Nakatsuji H, Caricato M, Li X, Hratchian HP, Izmaylov AF, Bloino J, Zheng G, Sonnenberg JL, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Vreven T, Montgomery JJA, Peralta JE, Ogliaro F, Bearpark MJ, Heyd J, Brothers EN, Kudin KN, Staroverov VN, Kobayashi R, Normand J, Raghavachari K, Rendell AP, Burant JC, Iyengar SS, Tomasi J, Cossi M, Rega N, Millam NJ, Klene M, Knox JE, Cross JB, Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW, Martin RL, Morokuma K, Zakrzewski VG, Voth GA, Salvador P, Dannenberg JJ, Dapprich S, Daniels AD, Farkas Ö, Foresman JB, Ortiz JV, Cioslowski J, Fox DJ (2009) Gaussian 09. Gaussian, Wallingford
Li YY, Guo WY, Zhao LM, Liu ZC, Lu XQ, Shan HH (2012) Theoretical investigation of the reaction of Mn+ with ethylene oxide. J Phys Chem A 116(1):512–519
Schmidt MW, Baldridge KK, Boatz JA, Elbert ST, Gordon MS, Jensen JH, Koseki S, Matsunaga N, Nguyen KA, Su SJ, Windus TL, Dupuis M, Montgomery JA (1993) General atomic and molecular electronic structure system. J Comput Chem 14:1347–1363
Matsunaga N, Koseki S, Gordon MS (1996) Relativistic potential energy surfaces of XH2(X = C, Si, Ge, Sn, and Pb) molecules: coupling of 1A1 and 3B1 states. J Chem Phys 104:7988–7996
Zener C (1932) Non-adiabatic crossing of energy levels. Proc R Soc Lond Ser A 137:696–702
Zener C (1933) Dissociation of excited diatomic molecules by external perturbations. Proc R Soc Lond Ser A 140:660–668
Fedorov DG, Koseki S, Schmidt MW, Gordon MS (2003) Spin–orbit coupling in molecules: chemistry beyond the adiabatic approximation. Int Rev Phys Chem 22:551–592
Moore CE (1952) Atomic energy levels, national bureau of standards. US Government Printing Office, Washington
Luo YR (2007) Comprehensive handbook of chemical bond energies. CRC Press, Boca Raton
Rodríguez-Santiago L, Sodupe M, Tortajada J (2001) Gas-phase reactivity of Ni+ with glycine. J Phys Chem A 105(22):5340–5347
Carpenter CJ, van Koppen PAM, Bowers MT (1995) Details of potential energy surfaces involving C–C bond activation: reactions of Fe+, Co+, and Ni+ with acetone. J Am Chem Soc 117(44):10976–10985
Rodríguez-Santiago L, Noguera M, Sodupe M, Salpin JY, Tortajada J (2003) Gas phase reactivity of Ni+ with urea. Mass spectrometry and theoretical studies. J Phys Chem A 107(46):9865–9874
Rodríguez-Santiago L, Tortajada J (2002) Experimental and theoretical studies on the gas phase reactivity of formamide-Ni+ complexes generated by FAB and electrosprayionization. Int J Mass Spectrom 219:429–443
Armentrout PB, Beauchamp JL (1980) Endothermic reactions of Ni+ with H2, HD and D2. Chem Phys 50(1):37–43
Liu F, Zhang XG, Armentrout PB (2005) Activation of CH4 by gas-phase Ni+ and the thermochemistry of Ni-ligand complexes. Phys Chem Chem Phys 7:1054–1064
Zhang Q, Kemper PR, Bowers MT (2001) Fe(CH4) +n and Ni(CH4) +n clusters: experimental and theoretical bond energies for n = 1–6. Int J Mass Spectrom 210–211:265–281
Sievers MR, Jarvis LM, Armentrout PB (1998) Transition-metal ethene bonds: thermochemistry of M+ (C2H4)n (M = Ti–Cu, n = 1 and 2) complexes. J Am Chem Soc 120:1891–1899
Hanratty MA, Beauchamo JL, Illies AJ, Koppen PV, Bowers MT (1988) Kinetic energy release distributions as a probe of transition-metal-mediated hydrogen–hydrogen, carbon–hydrogen, and carbon–carbon bond formation processes: reactions of cobalt and nickel ions with alkanes. J Am Chem Soc 110(1):1–14
Fisher ER, Armentrout PB (1990) Reactions of Co+, Ni+, and Cu+ with cyclopropane and ethylene oxide. Metal–methylidene ion bond energies. J Phys Chem 94(4):1674–1683
Allison J, Ridge DP (1979) Reactions of atomic metal ions with alkyl halides and alcohols in the gas phase. J Am Chem Soc 101:4998–5009
Zhao LM, Lu XQ, Li YY, Chen J, Guo WY (2012) On the gas-phase Co+-mediated oxidation of ethane by N2O: a mechanistic study. J Phys Chem A 116(12):3282–3289
Holthausen MC, Koch WA (1996) Theoretical view on Co-mediated C–C and C–H bond activations in ethane. J Am Chem Soc 118:9932–9940
Holthausen MC, Fiedler A, Schwarz H, Koch W (1996) How does Fe+ activate C–C and C–H Bonds in ethane? A theoretical investigation using density functional theory. J Phys Chem 100:6236–6242
Zhao LM, Guo WY, Liu ZC, Li YY, Lu XQ (2011) Theoretical study of the gas-phase Fe+-mediated oxidation of ethane by N2O. Theor Chem Acc 128:349–358
Perutz RN, Sabo-Etienne S (2007) The σ-CAM mechanism: σ-complexes as the basis of σ-bond metathesis at late-transition-metal centers. Angew Chem Int Ed 46:2578–2592
Acknowledgments
This work was supported by the Program for NSFC (21003158 and 21303266), Shandong Province Natural Science Foundation (ZR2011EMZ002), Taishan Scholar Foundation (ts20130929), Promotive Research Fund for Excellent Young and Middle-aged Scientists of Shandong Province (BS2012NJ015), and the Fundamental Research Funds for the Central Universities (12CX02014A, 14CX02004A, and 14CX06001A).
Author information
Authors and Affiliations
Corresponding authors
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Zhao, L., Ding, Q., Xu, W. et al. The ligand effect on the selective C–H versus C–C bond activation of propane by NiBr+: a theoretical study. Theor Chem Acc 134, 27 (2015). https://doi.org/10.1007/s00214-015-1628-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00214-015-1628-4