Transition metals and their oxides are widely used as both catalysts and catalytic supports for C–H bond activation [1-5]. Nonetheless, these materials have not yet been fully investigated from a comprehensive mechanistic viewpoint [6]. The CH₄ + MO⁺→ M⁺+CH₃OH reaction presents one of the simplest and earliest examples of C–H bond activation by transition metal compounds. Prior experimental studies have focused on gas-phase reactions of methane with first-row transition-metal oxide ions [2-10]. Considerable theoretical studies on these reactions have been conducted by Yoshizawa’s group [11-13], who also performed the theoretical study on the reaction of FeO⁺ with benzene [14,15]. In recent years, Andrew’s group has reacted some of the group IV transition metal atoms with acetonitrile; although the observed experimental product CH₂=Zr(H)NC was assigned by matrix isolation infrared spectroscopy and isotopic substituted experiments combing with DFT frequency analysis, the detailed reaction mechanism was not took into account [16]. Further DFT theoretical calculations on the spin inversion process of the reaction pathway were reported by Jin et al. who considered two-state reactivity and spin-forbidden chemical reactions [17]. Besides the reactions of pure transition metal compounds with methane and acetonitrile, some small hydrocarbons such as C₂H₂, C₂H₄, and C₆H₆ [18-20] and halohydrocarbons such as CH₃Cl [21-24] have also been reported. However, the reactions of pure transition metal oxides with halogenated aromatic hydrocarbons have received very little attention.

Halogenated benzene compounds, one of the larger groups of anthropogenic materials, are widely used in the chemical and electronics industries. However, most of these compounds are hazardous organic pollutants because of their environmental impact and noxious effects, and are frequently found in various waste oils and other organic liquids. Recently, selective C–H activation with halo [25-27], cynao [28,29], and hydroxo [30-32] functional groups has attracted much attention in organic synthesis due to the possibility of incorporating versatile functional groups. In these reactions, the functional groups remain unreactive under suitable conditions, thereby allowing the activation of the targeted C–H bonds to form more complicated organic molecules. In addition, hydroxylation of the halogenated benzenes is biologically important, because it forms p-substituted hydroxyhalobenzenes that are mostly found in mosquito larva and water extracts. Recently, group IX metals have been extensively reported for the activation of C–H bonds in the presence of carbon halide bonds. Milstein described an exclusive activation of ortho C–H bonds in chloro- and bromobenzene via the cationic pincer complex [(PNP^*)Ir]⁺ (PNP,′-bis(di--butylphosphino)- 2,6-diaminopyridine) [33]. Kinetic preference for C–H activation and thermodynamic preference for C–Cl activation for chlorobenzene (PhCl) was observed by Ozerov [34] by using an analogous (PNP)Ir^I system and also in related DFT calculations by Hall [35].

In this paper, we report a theoretical study of the reactions of neutral MnO and FeO with chlorobenzene, taking spin multiplicities into consideration. The reaction intermediates and the energetics along the reaction pathway are computed and analyzed in detail. Our theoretical analysis on the direct hydroxylation of chlorobenzene will help the researchers in the fields of catalysis chemistry and bioinorganic chemistry.

The computations were performed using the Gaussian 09 ab initio program package [36]. The 6-311++G(d, p) all-electron basis sets were used for all atoms [37,38]. To select an appropriate functional, different functionals (including B3LYP [39-41], M06 [42], M062X [42] and M11 [43]) were tested by calculating the M–O bond lengths and the M–O stretching vibrational frequencies of all first-row transition-metal monoxides (M=Sc to Cu). The results indicated that both wB97XD and B3LYP demonstrated the best results. Since B3LYP is a frequently reported functional, it was chosen to optimize the structures of the molecules under investigation. The first structures to be optimized were the equilibrium structures of PhCl, MnO, and FeO monomers, reactant complexes (PhCl)MO, hydroxo intermediates ClPh(MOH), product complexes, and their transition states with different multiplicities. Complete optimization of the molecular geometries was done with all stationary points. The harmonic vibrational frequencies of all the species were calculated with analytic second derivatives at the same level. This confirms that each stationary point is a local minimum or is a saddle point from systematic vibrational analyses of intrinsic reaction co-ordinates (IRCs) [44,45] and that it evaluates the zero-point vibrational energies (ZPVE). Each transition state was traced from a transition state toward both reactant and product directions along the imaginary mode of vibration using the algorithm developed by Gonzalez and Schlegel [46] in the mass-weighted internal coordinate system. Each IRC was constructed from 50 to 100 steps.

Potential energy diagram of the reaction between MO and C₆H₅Cl

DFT/B3LYP calculations were performed on the potential reaction products. Figures 1 and 2 shows the computed profiles of the potential energy surfaces for the reaction of transition metal monoxides with chlorobenzene in both the quartet and sextet states for MnO and in both the triplet and quintet states for FeO. Every possible computed structure of all intermediates and transition states is also displayed in Figures S1 and S2 (Provided in supplementary information). From the figures, we conclude that the chlorobenzene–p-chlorophenol reaction is a two-step reaction: in the first step, the compound passes through a transition state (TS1) to form a hydroxyl intermediate (HOM- C₆H₄Cl); in the second step, another transition state (TS2) forms that leads to the product. Figure 1 shows that the computed potential energy profiles of the quintet states have lower energy than the triplet states for FeO. Figure 2 indicates that the energies of the sextet states are below that of the quartet states during the entire reaction process. All the triplet states lie above the quintet ones; therefore, we discuss only the quintet states here. The activation energies required to form TS1 from the reactant are 15.7 kcal/mol for MnO and 20.2 kcal/mol for FeO along the sextet and quintet states of the reaction coordinates. Two stable hydroxyl intermediates are also optimized. Notably, no spin crossing was detected for the entire reaction. With the help of B3LYP computation, the overall reaction is predicted to be exothermic for MnO, with the release of 4.4 kcal/mol of energy. In contrast, the reaction is endothermic for FeO, with the required energy being 14.4 kcal/mol.

Figure 1: Potential energy diagram (including zero-point energy) along the reaction pathway, FeO+C₆H₅Cl→Fe+HOC₆H₄Cl, in the quintet states. Relative energies are in kcal/mol.

Figure 2: Potential energy diagram (including zero-point energy) along the reaction pathway, MnO+C₆H₅Cl→Mn+HOC₆H₄Cl, in the sextet states. Relative energies are in kcal/mol.

IRC analyses

As shown in Figures 3-5, the quintet-state IRC analysis can be used to investigate the reaction intermediates and transition states for the FeO system. We begin with by investigating the first step-reaction using IRC analysis in which the reactant complex forms the intermediate through TS1. This first step-reaction can be viewed as a 1,3-hydrogen migration (Figure 1). This is the most important step for cleaving p-hydrogen that migrates to oxygen and finally forms a hydroxy intermediate, which combines with OH and C₆H₄Cl ligands. The first step-reaction in the quintet state has been discussed in detail in Figure 3, in which we present the change in the geometrical parameters during the reaction. The IRC was started from TS1 (s=0), which exhibits a C_s structure with an imaginary vibrational mode of 1678i cm^-1, toward both the reactant (s<0) and product (s>0) directions. In principle, both directions would lead to an energy minimum in the reactant or product “valleys.” Unfortunately, the IRC ended before the true structure of the reactant complex could be acquired, although the terminal energy is very close to that of the reactant complex. However, this IRC analysis is good enough to increase our understanding of the reaction pathway.

The most important first step-reaction along the reaction pathway is discussed in detail in Figure 3a, in which the migrating hydrogen atom interacts not only with the C and O atoms, but also with the Fe atom to some extent. The Fe atom can significantly contribute to the migration of the H atom. It is obvious that the Fe–H distance exhibits an interesting feature; the distance reaches the minimum value (1.78 Å) in TS1. This means that there is an orbital overlap between the migrating hydrogen atom and the Fe atom in the vicinity of TS1. Therefore, the Fe atom must play an essential role in C–H bond dissociation. In addition, the Fe–O bond length does not change very much through the entire process.

Figure 3: Fe compound potential energies along the quintet IRC (a) from the reactant complex to the hydroxo intermediate and (b) from the hydroxo intermediate to the product complex.

Next, let us look at the change in the bond angle for the first stepreaction. In Figure 3b, the change in the C–Fe–O angle exhibits an interesting feature: the angle keeps changing and reaches the minimum value (75°) in TS1. The C–H bond begins to dissociate at s=-0.6 and the O–H bond distance is nearly constant (0.96 Å) after passing s=1.5. Therefore, we consider that H-atom migration happens in the range -0.61.5. Bending of the C– Fe–O angle would play a supportive part in the C–H bond dissociation and O–H bond formation.

After discussing the first step-reaction of the reaction pathway, let us now look at the second step-reaction. Here, the intermediate forms the product through TS2, which can be viewed as 1,2-p-chlorophenyl migration on the hydroxy intermediate (Figure 4). The IRC was started from TS2 (s=0), which exhibits a C₁ structure with an imaginary vibrational mode of 416i cm^-1, toward both the intermediate (s<0) and product complex (s>0) directions. Fortunately, this IRC analysis was successful. The most significant aspect in the second step-reaction is the dissociation of the Fe–C bond and formation of the C–O bond.

Figure 4: Change of internal coordinates for concerted C–H bond dissociation along the quintet IRC from the reactant complex to the hydroxo intermediate. Atomic distances are in Å and bond angles are in degrees.

We similarly investigated the details of the MnO and C₆H₅Cl reaction process by using the IRC analysis method. Because all the quartet states lie above the sextet ones, we discuss only the sextet states here.

To begin with, we first take a look at the detailed IRC analyses. The sextet-state IRC analysis is correctly connected by the reaction intermediates and transition states, as can be seen in Figure 6. Firstly, we consider the IRC analysis for the first step-reaction, which involves the transition of the reactant complex to the intermediate through TS1. This step-reaction can be viewed as 1,3-hydrogen migration. This is the most important step for cleaving p-hydrogen that migrates to oxygen and finally forms the hydroxy intermediate made of OH and C₆H₄Cl ligands (Figure 7). The IRC was started from TS1 (s=0), which exhibits a C_s structure with an imaginary vibrational mode of 1733i cm^-1, towards both the reactant (s<0) and product (s>0) directions. In principle, both directions would lead to an energy minimum in the reactant or product “valleys.” Unfortunately, similar to the case with FeO, our IRC ended before the true structure of the reactant complex could be achieved, although the terminal energy is very close to that of the reactant complex. Despite this, the IRC analysis considered in this study is sufficient to increase our understanding of the reaction pathway.

Figure 5: Change of internal coordinates for the concerted p-chlorophenyl migration along the quintet IRC from the hydroxo intermediate to the product complex. Atomic distances are in Å and bond angles are in degrees.

Figure 6: Mn compound potential energies along the sextet IRC (a) from the reactant complex to the hydroxo intermediate and (b) from the hydroxo intermediate to the product complex.

Figure 7: Change of internal coordinates for concerted C–H bond dissociation along the sextet IRC from the reactant complex to the hydroxo intermediate. Atomic distances are in Å and bond angles are in degrees.

The most important first step-reaction along the reaction pathway is discussed in detail in Figure 7a. The migrating hydrogen atom interacts not only with the C and O atoms but also with the Mn atom to some extent. The Mn atom is not a spectator in the first step-reaction process; it also significantly contributes to the migration of the H atom. It is obvious that the Mn–H bond distance exhibits an interesting feature: the distance reaches the minimum value (1.85 Å) in TS1. This means that there is an orbital overlap between the migrating hydrogen atom and the Mn atom in the vicinity of TS1. Therefore, the Mn atom must play an essential role in this C–H bond dissociation. In addition, the Mn–O distance remains unchanged during the entire process.

Next, let us look at the change in the bond angle for the first stepreaction. In Figure 7b, the change in the C–Mn–O angle exhibits an interesting feature; the angle keeps changing and reaches the minimum value (79°) in TS1. The C–H bond starts to dissociate at s=-0.6 and the O–H distance remains nearly constant (0.96 Å) after passing s=1.0. Therefore, we consider that H-atom migration happens in the range -0.61.0. Bending of the C–Mn–O angle would play a supporting role in C–H bond dissociation and O–H bond formation.

After discussing the first step-reaction of the reaction pathway, let us now look at the second step-reaction. In this section, the intermediate forms the product through TS2, which can be viewed as a 1,2-p-chlorophenyl migration on the hydroxy intermediate (Figure 8). The IRC was started from TS2 (s=0), which exhibits a C₁ structure with an imaginary vibrational mode of 368i cm^-1, toward both the intermediate (s<0) and product complex (s>0) directions. Fortunately, this IRC analysis was successful. The most significant aspect in the second step-reaction is the dissociation of the Mn–C bond and the formation of the C–O bond.

Figure 8: Change of internal coordinates for the concerted p-chlorophenyl migration along the sextet IRC from the hydroxo intermediate to the product complex. Atomic distances are in Å and bond angles are in degrees.

In this study, we have described the IRC analyses of the chlorobenzene→p-chlorophenol reaction by FeO and MnO. The geometries of the reactants, products, intermediates, and transition states along the reaction pathway via the important insertion intermediate, HO-M-C₆H₄Cl (M=FeO, MnO), have been described in detail for the quintet-state FeO and sextet-state MnO. A possible cross-point has been assumed to exist between the high-spin and low spin potential energy surfaces. Nonetheless, no spin crossing between that two states’ potential energy surfaces was found after computing both the triplet and quintet states for FeO and both the quartet and sextet states for MnO. The low-spin potential energy diagram lies above the high-spin one for the entire reaction pathway. We believe that our theoretical study on the possible reaction pathways for the conversion of chlorobenzene to p-chlorophenol will help in the analysis of catalytic and enzymatic functions of C-H and C-C bond activation by transition metal complexes.

We gratefully acknowledge financial support from National Natural Science Foundation of China (Grants No. 21273202 and 21473162) and National Basic Research Program of China (2013CB834604). Zhao YY is grateful to the Project Grants 521 Talents Cultivation of Zhejiang Sci-Tech University. This work is also supported by Zhejiang Provincial Top Key Academic Discipline of Chemical Engineering and Technology. We also thank editage (http://www.editage.cn) assistance during the preparation of this manuscript.