Methane activation by gold-doped titanium oxide cluster anions with closed-shell electronic structures

Abstract

The reactivity of closed-shell gas phase cluster anions AuTi₃O₇⁻ and AuTi₃O₈⁻ with methane under thermal collision conditions was studied by mass spectrometric experiments and quantum chemical calculations. Methane activation was observed with the formation of AuCH₃ in both cases, while the formation of formaldehyde was also identified in the reaction system of AuTi₃O₈⁻. The cooperative effect of the separated Au⁺ and O²⁻ ions on the clusters induces the cleavage of the first C–H bond of methane. Further activation of the second C–H bond by a peroxide ion O₂²⁻ leads to the formation of formaldehyde. This study shows that closed-shell species on metal oxides can be reactive enough to facilitate thermal H–CH₃ bond cleavage and the subsequent conversion.

---

Introduction

Methane, the major component of natural gas and shale gas, represents an important feedstock for the production of value-added chemicals.^1–6 However, the direct conversion of methane poses a serious challenge in contemporary catalysis owing to the significant energy required for C–H bond cleavage.¹ Many heterogeneous and homogeneous catalytic systems have been studied to transform methane, while it is challenging to uncover the elementary reactions and molecular level (ML) mechanisms associated with methane activation and conversion.^3–7 In the last decades, model investigations of the elementary reactions between methane and gas phase atomic clusters with state-of-the-art mass spectrometric experiments and quantum chemistry calculations have been serving as an important approach to discovering the ML mechanisms of methane activation and transformation.^8–16 The identified mechanisms can be very useful for catalyst design and optimization.^17–20

Many atomic clusters including oxides,^8–11 carbides,²¹ noble metals,^13–16 and so on have been identified to be able to activate methane under thermal collision conditions. Investigations of these cluster systems have revealed three types of mechanism to activate the C–H bond of methane:

[MO˙⁻] + CH₄ → [M(O–H)⁻] + CH₃˙ (1)

[M] + CH₄ → [H–M–CH₃] (2)

[ML] + CH₄ → [M–CH₃] + H–L (3)

where M is usually a metal atom and L is a ligand bonded with M. The atomic oxygen radical anions (O˙⁻) were extensively identified to activate methane through hydrogen atom abstraction (Reaction (1))^8–11,22 on many oxide clusters such as V₄O₁₀⁺,²³ Al₈O₁₂⁺,²⁴ and AuNbO₃⁺.²⁵ The coordinatively unsaturated metal atoms in naked and ligated metal species such as Au₂⁺,¹⁴ Pt_n⁺,¹⁵ PtCH₃⁺,²⁶ and TaC_n⁺ (ref. 21) can activate methane through oxidative addition (Reaction (2)).^9,22,27 The mechanism of σ-bond metathesis (Reaction (3))^9,22 was reported for a few simple mononuclear systems such as NiF⁺ (ref. 28) and HTiO⁺ (ref. 29) in which the σ-bonded ligand can be replaced through a reaction with the σ-bond of the incoming CH₄. Herein, we report a new mechanism of methane activation by gas phase atomic clusters (Reaction (4)):

[M⁺⋯O²⁻] + CH₄ → [CH₃–M⋯(O–H)⁻] (4)

where the metal cation M⁺ is separated from the oxygen anion O²⁻ on polynuclear metal oxide clusters and the cooperation of the two ions with counter polarity cleaves the C–H bond of methane.

The ability of two atoms with different polarity to promote chemical activity of atomic clusters in reactions with small molecules has been previously identified in the literature. Castleman, Khanna, and their co-workers have reported that a complementary active site composed of a pair of adjacent Al atoms that respectively act as a Lewis acid and Lewis base can activate a variety of polar molecules such as water, alcohols, aldehydes, and thiols.^30–33 Recently, we have found that a pair of non-adjacent ions Au⁺⋯O²⁻ on AuCeO₂⁺ and AuCe₂O₄⁺ cations can activate the non-polar dihydrogen.³⁴ However, these clusters are still not reactive enough to activate methane. This study reports that the cooperation of the separated Au⁺ and O²⁻ ions on gold-doped titanium oxide clusters AuTi₃O₇⁻ and AuTi₃O₈⁻ can bring about methane activation at thermal energies. Subsequent conversion of methane to a stable organic compound, formaldehyde, has also been identified. It is noteworthy that for the thermal activation of methane, most of the reactive clusters reported have open-shell electronic structures and the very few reactive species with closed-shell electronic structures are all mononuclear cations.^26,29 The cluster anions were generally found to be much less reactive than the corresponding cations in the reactions with methane.^10,13,16 For the first time, we report thermal methane activation by cluster anions with closed-shell electronic structures.

Results

Reactivity of AuTi₃O₇⁻ and AuTi₃O₈⁻ with methane

The AuTi₃O₇⁻ and AuTi₃O₈⁻ cluster anions were prepared by a reaction of O₂ with metal plasmas generated by laser vaporization of a solid disk compressed with Au and ⁴⁸Ti powders. The clusters of interest were mass-selected by a quadrupole mass filter and entered into a linear ion trap reactor, where they were thermalized by collisions with a pulse of He gas (maximal instantaneous pressure around 2–4 Pa) and then interacted with a pulse of CH₄, CD₄, or CH₂D₂ for a period of time.³⁵ Upon the interaction of AuTi₃O₇⁻ with 50 mPa CH₄ for about 1.07 ms (Fig. 1b), a strong product peak that can be assigned to Ti₃O₇H⁻ was observed, suggesting the following reaction channel:

AuTi₃O₇⁻ + CH₄ → Ti₃O₇H⁻ + AuCH₃ (5)

Fig. 1 Time-of-flight mass spectra for the reactions of mass selected AuTi₃O₇⁻ and AuTi₃O₈⁻ with CH₄ (b and f), CD₄ (c and g), CH₂D₂ (d), and N₂ (e). The peaks marked with asterisks and hollow circles in panels (f) and (g) represent water adsorption products (Ti₃O₇HH₂O⁻ or Ti₃O₇DH₂O⁻) and Ti₃O₈⁻ from collision induced dissociation, respectively. The peaks marked with +X (X = H₂O, CH₄, CD₄, and N₂) in panels (a–c) and (e–g) denote the association products with AuTi₃O₇⁻ and AuTi₃O₈⁻, respectively. The reactant gas pressures are shown. The reaction times (t_R) are 1.07 ms for (a–d) and 9.57 ms for (e–g). The signal magnitudes below m/z 460 are amplified by 4 and 12 for (f) and (g), respectively.

The isotopic labelling experiments with CD₄ (Fig. 1c) and CH₂D₂ (Fig. 1d) confirmed the above reaction. The generation of Ti₃O₇D⁻ was observed from AuTi₃O₇⁻ + CD₄ while both Ti₃O₇H⁻ and Ti₃O₇D⁻ were produced from AuTi₃O₇⁻ + CH₂D₂. The inter- and intra-molecule isotopic effects were apparently observed. In addition to Reaction (5), a minor association reaction channel generating AuTi₃O₇CH₄⁻ (4% of the total product ions) was also observed.

The AuTi₃O₈⁻ cluster is much less reactive than AuTi₃O₇⁻ and a longer reaction time and higher methane pressures were used for AuTi₃O₈⁻ + CH₄ (Fig. 1f), in which the generation of Ti₃O₇H⁻, Ti₃O₈H⁻, AuTi₃O₆⁻, and AuTi₃O₇H₂⁻ was observed and the assignments were confirmed by the experiments with CD₄ and CH₂D₂ (Fig. 1g and S1 ESI†). In a reference experiment with N₂ (Fig. 1e), the AuTi₃O₆⁻ product was not generated, indicating that this product cluster in Fig. 1f and g was due to a chemical reaction rather than collision induced dissociation (CID, such as AuTi₃O₈⁻ + CH₄ → AuTi₃O₆⁻ + O₂ + CH₄). In addition to molecular association, the following four reaction channels are suggested by the experiments:

AuTi₃O₈⁻ + CH₄ → Ti₃O₈H⁻ + AuCH₃ (6)

AuTi₃O₈⁻ + CH₄ → AuTi₃O₇H₂⁻ + CH₂O (7)

AuTi₃O₈⁻ + CH₄ → Ti₃O₇H⁻ + AuCH₃O (CH₂O + AuH) (8)

AuTi₃O₈⁻ + CH₄ → AuTi₃O₆⁻ + CH₄O₂ (CH₂O + H₂O) (9)

The branching ratios of generating AuTi₃O₈CH₄⁻ (41%), Ti₃O₇H⁻ (35%), and AuTi₃O₆⁻ (15%) are much larger than those of AuTi₃O₇H₂⁻ (6%) and Ti₃O₈H⁻ (3%) (Fig. 1f).

An atomic cluster often has different structural isomers with very different reactivities.²⁹ The analysis of the methane-pressure dependent reactivity indicated that (87 ± 1)% of the experimentally generated AuTi₃O₇⁻ ions (Fig. S2†) and only (37 ± 4)% of the AuTi₃O₈⁻ ions (Fig. S3†) were reactive with CH₄. For the reactive component of AuTi₃O₇⁻, the pseudo first-order rate constant (k₁) of Reaction (5) is (7.7 ± 2.3) × 10⁻¹¹ cm³ per molecule per second, corresponding to a reaction efficiency (Φ)³⁶ of (7.7 ± 2.3)%. The intra and inter-molecular kinetic isotope effects (KIEs) amount to 5.0 ± 1.1 and 4.5 ± 1.3, respectively. For the reactive component of AuTi₃O₈⁻, the summed k₁ value of Reactions (6)–(9) is (1.0 ± 0.3) × 10⁻¹² cm³ per molecule per second [Φ = (0.1 ± 0.03)%]. The inter-molecular KIE amounts to 4.0 ± 1.3.

Reaction mechanisms of AuTi₃O₇⁻ and AuTi₃O₈⁻ with CH₄

Density functional theory (DFT) calculations at the TPSS level, which shows the overall best performance in calculating several critical bond energies among the 18 tested methods (Table S1, ESI†), have been conducted to explore the detailed reaction mechanisms. A Fortran code based on the genetic algorithm was used to search the global minimum structures of AuTi₃O₇⁻ as well as AuTi₃O₈⁻ clusters with different spin multiplicities.³⁷ To determine the lowest-energy isomers of AuTi₃O₇⁻ and AuTi₃O₈⁻, further single-point energy calculations with a high-level quantum chemistry method of a restricted coupled-cluster method with single, double, and perturbative triple excitations [RCCSD(T)] have also been performed for the TPSS optimized structures. In the reaction pathway of AuTi₃O₈⁻ + CH₄, concomitant cleavage and formation of several chemical bonds are involved in important transition states. It is known that commonly used density functionals do not correctly describe the long- and mid-range dispersion interactions, which can influence the chemical reaction energies.³⁸ Thus, TPSS functional calculated energies with dispersion corrections³⁸ are given throughout the reaction pathways.

The lowest-lying isomer of AuTi₃O₇⁻ (Fig. 2 and S5†) has two terminally-bonded oxygen anions (O_t²⁻, −0.62e) and a one-fold coordinated gold cation (Au_1f⁺, +0.42e). It is noteworthy that the superscripts “2−” of O²⁻ and “+” of Au⁺ denote the formal oxidation states rather than the net charges on the atoms. As marked in Fig. 2, the Au_1f⁺ cation is separated from the O_t²⁻ anions. The Au-side of AuTi₃O₇⁻ is the least negatively charged, so it is expected that the Au_1f⁺ atom traps CH₄ to form the encounter complex I1 with a significant binding energy (54 kJ mol⁻¹). Then the CH₄ is delivered to be close to a O_t²⁻ ion (TS1) so that the cleavage of one C–H bond and the concomitant formation of one Au–CH₃ bond and one O–H bond can take place. Such a process (I1 → TS1 → I2) is subject to an energy barrier of 13 kJ mol⁻¹ which is surmountable by the binding energy of I1. The formation of I2 can release a high energy of 198 kJ mol⁻¹, which is enough to evaporate the AuCH₃ species rather than Au + CH₃ (endothermic by 244 kJ mol⁻¹) from the reaction complex to form the experimentally observed Ti₃O₇H⁻ cluster (Fig. 1b).

Fig. 2 TPSS functional calculated potential energy profile for AuTi₃O₇⁻ + CH₄ → Ti₃O₇H⁻ + AuCH₃. All of the species are in a singlet spin state. The dispersion corrected energies of the reaction intermediates (I1 and I2), transition state (TS1), and products (P1) with respect to the separated reactants (R1) are given in kJ mol⁻¹. Bond lengths are given in pm.

The gold atom in the lowest-lying isomer of AuTi₃O₇⁻ is one-fold coordinated (Fig. 2). In a low-lying isomer of AuTi₃O₇⁻ (Fig. S5†), the Au⁺ ion can be two-fold coordinated (Au_2f⁺). Such a cluster anion with Au_2f⁺ can hardly trap (binding energy is only 8 kJ mol⁻¹) the reactant molecule CH₄. The activation of CH₄ by AuTi₃O₇⁻ with the Au_2f⁺ is subject to an additional transformation so that the Au_2f⁺ ion becomes Au_1f⁺ (as in I1 of Fig. 2). The Au_2f⁺ → Au_1f⁺ transformation has an overall positive barrier (3 kJ mol⁻¹) which hinders subsequent methane activation, suggesting that the AuTi₃O₇⁻ isomer with Au_2f⁺ corresponds to the un-reactive component (13%) of AuTi₃O₇⁻ in the experiments.

When one of the O_t²⁻ ions in the lowest-lying isomer of AuTi₃O₇⁻ is replaced by a peroxide unit (O₂²⁻), a low-lying isomer of AuTi₃O₈⁻ with Au_1f⁺ and O_t²⁻ ions can be formed (see I3 of Fig. 3). However, this isomer with Au_1f⁺ is less stable by 24 kJ mol⁻¹ than the lowest-lying isomer of AuTi₃O₈⁻ that contains an Au_2f⁺ ion (Fig. S6†). The activation of CH₄ by the lowest-lying isomer of AuTi₃O₈⁻ also involves the Au_2f⁺ → Au_1f⁺ conversion which is hindered by an overall positive reaction barrier (1 kJ mol⁻¹). As a result, this lowest-lying isomer with Au_2f⁺ accounts for the 63% un-reactive component of the experimentally generated AuTi₃O₈⁻ ions. The 37% reactive ions can then be assigned to the low-lying isomer with Au_1f⁺ and the reaction mechanism is shown in Fig. 3.

Fig. 3 TPSS functional calculated potential energy profile for AuTi₃O₈⁻ + CH₄ to generate the products Ti₃O₈H⁻ + AuCH₃ (P2), AuTi₃O₇H₂⁻ + CH₂O (P3), Ti₃O₇H⁻ + CH₂O + AuH (P4), and AuTi₃O₆⁻ + CH₂O + H₂O (P5). All of the species are in a singlet spin state. The dispersion corrected energies of the reaction intermediates (I3–I8), transition states (TS2–TS6), and products (P2–P5) with respect to the separated reactants (R2) are given in kJ mol⁻¹. Bond lengths are given in pm.

Similarly to AuTi₃O₇⁻ + CH₄ in Fig. 2, the Au_1f⁺ on AuTi₃O₈⁻ traps CH₄ and delivers CH₄ to be close to the O_t²⁻ ion for C–H bond cleavage (I4 → TS3 → I5, Fig. 3). The large amount of exothermic energy released (191 kJ mol⁻¹) can evaporate the AuCH₃ unit to produce Ti₃O₈H⁻ (I5 → P2, Reaction (6)). Alternatively, I5 can overcome the energy barrier (122 kJ mol⁻¹, I5 → TS4) involving the activation of a second C–H bond by the O₂²⁻ unit (I5 → TS4 → I6) to produce a neutral CH₂O molecule and AuTi₃O₇H₂⁻ ions (I6 → TS5 → I7 → TS6 → I8 → P3, Reaction (7)). The formation of AuTi₃O₇H₂⁻ and CH₂O is highly exothermic (ΔH₀ = −261 kJ mol⁻¹), so the resulting AuTi₃O₇H₂⁻ has enough internal energy to evaporate AuH and H₂O (Fig. S7†) to form the product ions Ti₃O₇H⁻ (Reaction (8)) and AuTi₃O₆⁻ (Reaction (9)), respectively. It can be seen that Reactions (7)–(9) all involve the generation of formaldehyde (CH₂O) and the DFT calculations correctly predict that Reactions (6)–(9) are all kinetically and thermodynamically favorable. Furthermore, the lower energy of TS4 (−69 kJ mol⁻¹) than that of P2 (−46 kJ mol⁻¹) can well rationalize the experimental branching ratios that show Reaction (6) as a minor channel.

In the reaction of AuTi₃O₈⁻ with CH₄, the Au_1f⁺ cation can also deliver CH₄ to be close to the O₂²⁻ anion to activate the first C–H bond of methane (Fig. S9†). Subsequent transformation to form the intermediate I5 is kinetically less favorable than the reaction path of Fig. 3. A reaction path to form I7 (Fig. S10†) is slightly more favorable than the path of Fig. 3 kinetically (−45 kJ mol⁻¹versus −38 kJ mol⁻¹ for the critical transition states). This alternative path has a very deep potential well (209 kJ mol⁻¹) which can hinder the further transformation of the reaction complex into separate products and leads to the formation of the association species AuTi₃O₈CH₄⁻. This result is consistent with the experimental observation that molecular association is a major reaction channel (41%) for AuTi₃O₈⁻ + CH₄ (Fig. 1f).

Discussion

Many metal oxide clusters including homo-nuclear (M_xO_y^±) and hetero-nuclear oxide clusters (M¹_x1M²_x2O_y^±) have been found to react with methane under thermal collision conditions.^{8–11,22–25} All of the reactive oxide clusters were open-shell systems with oxygen radical centers and methane activation primarily followed Reaction (1). Recently, it has been demonstrated that the PtAl₂O₄⁻ cluster can activate CH₄ through Reaction (2) and then the oxygen radical accepts the transferred H atom.³⁹ Herein, the AuTi₃O₇⁻ and AuTi₃O₈⁻ clusters are closed-shell systems (without radical oxygen species) and they activate methane through the new mechanism, Reaction (4). Previously, a few positively charged mononuclear species^26,29 with closed-shell electronic structures were shown to activate methane through Reactions (2) and (3).

The observed reactivity of AuTi₃O₇⁻ and AuTi₃O₈⁻ with methane (Reactions (5)–(9)) can be closely related to the extraordinary properties of gold, and results from the strong relativistic effect on this element.⁴⁰ The high electro-negativity of gold leads to a rather weak Au–O chemical bond (bond energy of 219 kJ mol⁻¹)⁴¹ so that the O–Au_1f⁺ bond can be flexible for the delivery of CH₄ to be close to O²⁻ (Fig. 2 and 3) or O₂²⁻ (Fig. S9†) for C–H activation. Moreover, the analogous Au/H⁴² results in relatively strong Au–CH₃ (232 kJ mol⁻¹ by the TPSS functional) and Au–H (292 kJ mol⁻¹ by the TPSS functional) bonds so that the evaporation of AuCH₃ (Reactions (5) and (6)) and AuH (Reaction (8)) from the reaction complex is possible. Upon generation of the CH₂O moiety in AuTi₃O₈⁻ + CH₄, the Au atom is bonded with the metal atom Ti (I8 of Fig. 3) and the gold atom becomes negatively charged (−0.20e), which is also a result of the relativistic effect.^40,43,44

In addition to the extraordinary properties of gold, the co-participation of both Au_1f⁺ and O_t²⁻ ions is very important for the activation of methane (Fig. 2 and 3). The reaction paths for the C–H bond cleavage of CH₄ by the single Au_1f⁺ cation and a single O_t²⁻ anion of AuTi₃O₇⁻ have been also followed. These processes are subject to very high energy barriers (Fig. S11†). In addition, the reaction of AuTi₃O₇⁻ with CH₄ on the triplet potential energy surface has also been calculated. It turned out that all of the triplet reactants, intermediates, and transition states are much higher (>100 kJ mol⁻¹) in energy than the corresponding singlet counterparts. As a result, the cooperative activation by the Au_1f⁺ and O_t²⁻ ions, as shown in Fig. 2, is the only mechanism of methane activation by AuTi₃O₇⁻.

To explore the excellent ability of the cooperative Au_1f⁺ and O_t²⁻ ions to promote methane activation by the closed-shell cluster anions of AuTi₃O₇⁻ and AuTi₃O₈⁻ under thermal collision conditions, variation of the geometrical structures of the reaction intermediates and the change of natural charge (Table S2†) as well as the Wiberg bond order (Table S3†) of critical atoms and chemical bonds have been analyzed for the reaction system of AuTi₃O₇⁻ + CH₄ (Fig. 2). During the course of C–H bond cleavage (I1 → TS1 → I2), the electron population on the CH₃ group and Au atom increases (−0.11e → −0.41e for CH₃ and +0.44e → +0.28e for Au) and that on the transferring H decreases (+0.18e → +0.47e), indicating that the C–H bond may be cleaved in a heterolytic manner and the reaction may follow a Lewis acid–base pair mechanism taking into account that the separated Au_1f⁺ and O_t²⁻ ions in AuTi₃O₇⁻ can be considered as a Lewis acid and Lewis base, respectively. However, the change in the natural charges of the CH₃ moiety and H is not very large (around 0.3e), additionally, the charge increase on the H is normal for a bond conversion of C–H → O–H. In contrast to the Lewis acid–base mechanism, another mechanism of the flexible switch of the roles of the two O_t atoms in [Ti₃O₇]²⁻ which enables the favorable C–H activation is proposed from a bonding point of view. The reactant of AuTi₃O₇⁻ can be viewed as Au⁺[Ti₃O₇]²⁻, in which Au⁺ (+0.42e) is attached to one of the O_t in [Ti₃O₇]²⁻ (−1.42e) with a bond strength of 1058 kJ mol⁻¹ and the O_t becomes a bridging-bonded oxygen (O_b, as marked in Fig. 2). During the C–H activation, such O_b releases gold to bond with the CH₃ moiety and the O_b itself switches to O_t after AuCH₃ evaporation (Fig. 2). At the same time, a different O_t in [Ti₃O₇]²⁻ switches to O_b after attaching to H⁺ (+0.46e) with a much stronger bond strength of 1710 kJ mol⁻¹ (Fig. 2). Thus, the overall increased bond strengths in the products drive the C–H activation thermodynamically. The switch of the roles of the two O_t in [Ti₃O₇]²⁻ can also be evidenced by the change of the Wiberg bond order of Ti–O_t (Table S3†). The Ti1–O_t and Ti2–O_t bonds (Fig. 2) gradually switch from single/double bonds (187/167 pm) in the reactant to double/single bonds (166/187 pm) in the product. This mechanism suggests that the C–H bond may be cleaved in a homolytic manner, namely, hydrogen atom transfer (HAT).

The identification of the Lewis acid–base mechanism or HAT for AuTi₃O₇⁻ + CH₄ relies on the transfer mode of one electron (e⁻) and one proton (H⁺) of a H atom. HAT is characterized by the transfer of the electron and proton to a single site. In contrast, a Lewis acid–base mechanism corresponds to the transfer of the electron and proton to different acceptor sites. Such a transfer mode can be called an electron–proton transfer (EPT).⁴⁵ It was proposed that a key element in the theoretical characterization of the mechanisms of proton and electron transfer is the formulation of their localized diabatic states.⁴⁶ However, the electron and proton described by standard quantum mechanical methods tend to be delocalized, and the analysis of the electron or proton acceptors such as a molecular orbital or a chemical bond depends on the adopted computational level of theory.⁴⁵ Consequently, it is hard to distinguish exactly the two proposed mechanisms of AuTi₃O₇⁻ + CH₄ at the present level of theory. More advanced methods such as multistate DFT, in which the electron and proton localized diabatic configurations can be constructed through block-localization of Kohn–Sham orbitals, should be employed to study the potential energy surfaces of the HAT and EPT, which may provide clues to recognize the mechanistic details of our reaction systems.

Conclusions

The reactions of methane with negatively charged titanium oxide clusters doped with single gold atoms, AuTi₃O₇⁻ and AuTi₃O₈⁻, have been identified by mass spectrometry and quantum chemistry calculations. To the best of our knowledge, this is the first example of the thermal activation and transformation of methane by atomic cluster anions with closed-shell electronic structures. Unlike the previously reported three general mechanisms including hydrogen atom abstraction, oxidative addition, and σ-bond metathesis for methane activation by atomic clusters, the cooperation of the separated Au⁺ and O²⁻ induces the cleavage of the first C–H bond of methane for AuTi₃O₇⁻ and AuTi₃O₈⁻ clusters. Further activation of the second C–H bond of methane on the oxygen-rich system of AuTi₃O₈⁻ leads to the formation of the stable organic compound formaldehyde (CH₂O). The observed unique reactivity of AuTi₃O₇⁻ and AuTi₃O₈⁻ toward methane results from the strong relativistic effect on the gold element. This study not only serves as an important step in understanding that closed-shell species on metal oxides can be reactive enough to facilitate thermal H–CH₃ bond cleavage, but also provides molecular-level insights into the design of active sites on metal oxide supported gold catalysts to activate and transform methane.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (No. 21273247, 21325314, and 21573247), the Major Research Plan of China (No. 2013CB834603), and the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDA09030101).

Notes and references

1. R. H. Crabtree, Chem. Rev., 1995, 95, 987–1007.
2. J. H. Lunsford, Catal. Today, 2000, 63, 165–174.
3. R. A. Periana, O. Mironov, D. Taube, G. Bhalla and C. Jones, Science, 2003, 301, 814–818.
4. Q. Zhu, S. L. Wegener, C. Xie, O. Uche, M. Neurock and T. J. Marks, Nat. Chem., 2013, 5, 104–109.
5. X. Guo, G. Fang, G. Li, H. Ma, H. Fan, L. Yu, C. Ma, X. Wu, D. Deng, M. Wei, D. Tan, R. Si, S. Zhang, J. Li, L. Sun, Z. Tang, X. Pan and X. Bao, Science, 2014, 344, 616–619.
6. J. Gao, Y. Zheng, J.-M. Jehng, Y. Tang, I. E. Wachs and S. G. Podkolzin, Science, 2015, 348, 686–690.
7. S. Arndt, G. Laugel, S. Levchenko, R. Horn, M. Baerns, M. Scheffler, R. Schlögl and R. Schomäcker, Catal. Rev.: Sci. Eng., 2011, 53, 424–514.
8. N. Dietl, M. Schlangen and H. Schwarz, Angew. Chem., Int. Ed., 2012, 51, 5544–5555.
9. J. Roithová and D. Schröder, Chem. Rev., 2010, 110, 1170–1211.
10. X.-L. Ding, X.-N. Wu, Y.-X. Zhao and S.-G. He, Acc. Chem. Res., 2012, 45, 382–390.
11. G. de Petris, A. Troiani, M. Rosi, G. Angelini and O. Ursini, Chem.–Eur. J., 2009, 15, 4248–4252.
12. R. Liyanage, X.-G. Zhang and P. B. Armentrout, J. Chem. Phys., 2001, 115, 9747–9763.
13. U. Achatz, C. Berg, S. Joos, B. S. Fox, M. K. Beyer, G. Niedner-Schatteburg and V. E. Bondybey, Chem. Phys. Lett., 2000, 320, 53–58.
14. S. M. Lang, T. M. Bernhardt, R. Barnett and N. U. Landman, Angew. Chem., Int. Ed., 2010, 49, 980–983.
15. D. J. Harding, C. Kerpal, G. Meijer and A. Fielicke, Angew. Chem., Int. Ed., 2012, 51, 817–819.
16. C. Adlhart and E. Uggerud, Chem. Commun., 2006, 2581–2582.
17. A. W. Castleman Jr, Catal. Lett., 2011, 141, 1243–1253.
18. S. Yin and E. R. Bernstein, Int. J. Mass Spectrom., 2012, 321–322, 49–65.
19. R. A. J. O'Hair, Int. J. Mass Spectrom., 2015, 377, 121–129.
20. M. Diefenbach, M. Brönstrup, M. Aschi, D. Schröder and H. Schwarz, J. Am. Chem. Soc., 1999, 121, 10614–10625.
21. C. J. Cassadyt and S. W. McElvany, J. Am. Chem. Soc., 1990, 112, 4188–4191.
22. H. Schwarz, Isr. J. Chem., 2014, 54, 1413–1431.
23. S. Feyel, J. Döbler, D. Schröder, J. Sauer and H. Schwarz, Angew. Chem., Int. Ed., 2006, 45, 4681–4685.
24. S. Feyel, J. Döbler, R. Höckendorf, M. K. Beyer, J. Sauer and H. Schwarz, Angew. Chem., Int. Ed., 2008, 47, 1946–1950.
25. X.-N. Wu, X.-N. Li, X.-L. Ding and S.-G. He, Angew. Chem., Int. Ed., 2013, 52, 2444–2448.
26. M. Armélin, M. Schlangen and H. Schwarz, Chem.–Eur. J., 2008, 14, 5229–5236.
27. P. B. Armentrout and J. L. Beauchamp, Acc. Chem. Res., 1989, 22, 315–321.
28. M. Schlangen and H. Schwarz, Helv. Chim. Acta, 2008, 91, 2203–2210.
29. R. Kretschmer, M. Schlangen and H. Schwarz, Angew. Chem., Int. Ed., 2013, 52, 6097–6101.
30. P. J. Roach, W. H. Woodward, A. W. Castleman Jr, A. C. Reber and S. N. Khanna, Science, 2009, 323, 492–495.
31. A. C. Reber, P. J. Roach, W. H. Woodward, S. N. Khanna and A. W. Castleman Jr, J. Phys. Chem. A, 2012, 116, 8085–8091.
32. W. H. Woodward, A. C. Reber, J. C. Smith, S. N. Khanna and A. W. Castleman Jr, J. Phys. Chem. C, 2013, 117, 7445–7450.
33. Z. Luo, J. C. Smith, C. Berkdemir and A. W. Castleman Jr, Chem. Phys. Lett., 2013, 590, 63–68.
34. J.-H. Meng and S.-G. He, J. Phys. Chem. Lett., 2014, 5, 3890–3894.
35. Z. Yuan, Z.-Y. Li, Z.-X. Zhou, Q.-Y. Liu, Y.-X. Zhao and S.-G. He, J. Phys. Chem. C, 2014, 118, 14967–14976.
36. G. Kummerlöwe and M. K. Beyer, Int. J. Mass Spectrom., 2005, 244, 84–90.
37. J.-B. Ma, B. Xu, J.-H. Meng, X.-N. Wu, X.-L. Ding, X.-N. Li and S.-G. He, J. Am. Chem. Soc., 2013, 135, 2991–2998.
38. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
39. Y.-X. Zhao, Z.-Y. Li, Z. Yuan, X.-N. Li and S.-G. He, Angew. Chem., Int. Ed., 2014, 53, 9482–9486.
40. P. Pyykkö, Angew. Chem., Int. Ed., 2004, 43, 4412–4456.
41. J. B. Pedley and E. M. Marshall, J. Phys. Chem. Ref. Data, 1983, 12, 967–1031.
42. L. S. Wang, Phys. Chem. Chem. Phys., 2010, 12, 8694–8705.
43. X.-N. Li, Z. Yuan and S.-G. He, J. Am. Chem. Soc., 2014, 136, 3617–3623.
44. M. Jansen, Chem. Soc. Rev., 2008, 37, 1826–1835.
45. A. K. Harshan, T. Yu, A. V. Soudackov and S. Hammes-Schiffer, J. Am. Chem. Soc., 2015, 137, 13545–13555.
46. A. Cembran, M. R. Provorse, C. Wang, W. Wu and J. Gao, J. Chem. Theory Comput., 2012, 8, 4347–4358.

---

Footnote

† Electronic supplementary information (ESI) available: The detailed experimental and computational methods as well as additional experimental and computational results. See DOI: 10.1039/c6sc00539j

---