Catalytic Co‐Conversion of CH4 and CO2 Mediated by Rhodium–Titanium Oxide Anions RhTiO2 −

Abstract Catalytic co‐conversion of methane with carbon dioxide to produce syngas (2 H2+2 CO) involves complicated elementary steps and almost all the elementary reactions are performed at the same high temperature conditions in practical thermocatalysis. Here, we demonstrate by mass spectrometric experiments that RhTiO2 − promotes the co‐conversion of CH4 and CO2 to free 2 H2+CO and an adsorbed CO (COads) at room temperature; the only elementary step that requires the input of external energy is desorption of COads from the RhTiO2CO− to reform RhTiO2 −. This study not only identifies a promising active species for dry (CO2) reforming of methane to syngas, but also emphasizes the importance of temperature control over elementary steps in practical catalysis, which may significantly alleviate the carbon deposition originating from the pyrolysis of methane.


Theoretical method. (Page S3)
3. Additional mass spectra for cluster reactions. Figure S1. Assignment of the grey peaks in Figure 1   The negatively charged rhodium-titanium bimetallic oxide clusters (RhxTiyOz -) were generated by laser ablation of a rotating and translating metal disk compressed with Rh and 48 Ti powders (molar ration of 2:1) in the presence of 0.4% O2 seeded in a helium carrier gas with the backing pressure of 6.0 standard atmospheres. Among many tested RhxTiyOzclusters, the RhTiO2 − cluster ions of interest were mass-selected using a quadrupole mass filter (QMF) 1 and entered into a linear ion trap (LIT) reactor 2 , where they were confined and thermalized by collisions with a pulse of buffer gas He and then reacted with 12 CH4, 13 CH4, CD4, C 16 O2 or C 18 O2. The RhTiO2CH4 − , RhTiO2 13 CH4 − , and RhTiO2CD4 − ions were pre-produced by seeding 20% CH4, 13 CH4 and CD4 in the He buffer gas to react with the RhTiO2 − cluster ions in the LIT, respectively. The generated RhTiO2CH4 − , RhTiO2 13 CH4 − or RhTiO2CD4 − ions then reacted with 5% C 16 O2 or 20% C 18 O2 seeded in He gas for about 2.5 ms in the LIT. A reflection time-of-flight mass spectrometer 3 (TOF-MS) was used to detect the cluster ions ejected from the LIT reactor.
The rate constants of the reactions between RhTiO2 − cluster and CH4, CD4, or CO2 were determined by using the following equations: in which IR and Ip are the relative intensities of the reactant and product cluster ions, respectively; k1 is the (second order) rate constant of a pseudo-first-order reaction, ρ is the molecular density of CH4, CD4, or CO2 in the ion trap reactor, and tR is the reaction time.
The collision-induced dissociation (CID) experiments of RhTiO2CO − were performed by introducing xenon into the LIT reactor (run at the collision cell mode) for collisions with RhTiO2CO − clusters ions, of which the translational energies could be fixed at different values. The RhTiO2CO − ions were generated by laser ablation in the presence of an O2/CO gas mixture seeded in a helium carrier gas. The pressure of Xe in the LIT was low (60 mPa) for single collision conditions (the average number of collisions was estimated to be 0.08). The collision energies between a cluster ion and an Xe atom were converted from the laboratory frame (Elab) to center-of-mass frame (Ecm) by using Ecm = Elab × m/(m + M), in which M and m are the masses of the cluster ion and Xe, respectively.
In addition to the above reactivity and CID experiments in which a single ion trap system was used, the experiments by using a newly-developed double ion trap system were carried out. The double ion trap system includes two QMFs and two LITs. The first set of QMF/LIF was used for the reaction of mass-selected RhTiO2 − with CH4 or gas mixtures of CH4/CO2 and the second QMF can mass-select the product ions from the first LIT to inject into the second LIT for further reaction with CO2 molecule or collision with Xe atom, respectively. In such experiments, the reactant cluster ions of interest (RhTiO2 − ) were generated in the same way as in the single ion trap experiments. The conditions to run the second LIT for the reaction (with CO2) and the collision (with Xe) are similar to those used in the corresponding single ion trap experiment.
The photoelectron imaging spectroscopy (PEIS) of RhTiO2 − was studied with a separated vacuum system. The generated RhTiO2 − was selected by a mass gate and crossed with a 355 nm or 425 nm laser beam. The electrons from photo-detachment were energy-analyzed by the photoelectron imaging spectrometer. 4 The PEIS spectrum was calibrated using the spectrum of Au − taken at the similar conditions. The resolution of the photoelectron imaging spectrometer was approximately 30 meV at electron kinetic energy of 1 eV. S3

Theoretical Method.
The density functional theory (DFT) calculations using Gaussian 09 program 5 were carried out to investigate the structures of reactant cluster RhTiO2 − , reaction complex RhTiO2X − (X= CH4, CH4CO2, CO), and products as well as the reaction pathways of RhTiO2 − with CH4 and RhTiO2CH4 − with CO2. The TPSS functional 6 has been proved to perform well for many Rh-doped metal oxide systems, 7,8 so the results by TPSS method are given throughout this work. The TZVP basis sets 9 for C, H, O, and Ti atoms and the D95V basis set combined with the Stuttgart/Dresden relativistic effective core potentials (denoted as SDD in Gaussian software) 10 for Rh atom were used. The reaction pathways calculations involved geometry optimization of reaction intermediates (IMs) and transition states (TSs) through which the IMs transfer to each other. The initial guess structures of the TS species were obtained through relaxed potential energy surface scans using single or multiple internal coordinates. 11 Vibrational frequency calculations were performed to check that the IMs or TSs have zero and only one imaginary frequency, respectively. Intrinsic reaction coordinate calculations were performed so that a transition state connects two appropriate local minima. The zero-point vibration corrected energies (ΔH0) in unit of eV are reported in this work. S4 3. Additional mass spectra for cluster reactions. Figure S1. Assignment of the grey peaks in Figure 1. These minor peaks are due to the reactions with the second CH4 or second CO2 molecule, and with the residual H2O in the LIT. The RhTiO2X − (X= O, H2O, etc.) species are labeled as +X (see Figure 1 in the main text for the assignment of unlabeled peaks). S5 The minor peaks can be formed by the following reactions.  Figure S2. TOF mass spectra for the reactions of RhTiO2 − cluster with CH4 (a1-a4) and CD4 (b1-b4) at 298 K. The reactant gas pressures are shown. The RhTiO2X − (X = CH4, CD4. etc.) species are labeled as +X. The pannels of (a5) and (b5) plot the signal variation of the reactant and product ions with respect to the CH4 and CD4 pressure, respectively. The solid lines were fitted to the experimental data by using equations (S1a) and (S1b).  The pannel of (a5) plots the signal variation of the reactant and product ions with respect to the CO2 pressure. The solid lines were fitted to the experimental data by using equations (S1a) and (S1b).       Figure S13 shows the thermodynamic data, obtained from DFT calculations, for the elementary steps of DRM to syngas catalyzed by RhTiO2 − . The formation of RhTiO2CO -+ 2H2 + CO (i−v) from the reactants is exothermic, while the final reaction step, CO desorption from RhTiO2COto reform RhTiO2 -(vi), is markedly endothermic. S17 Figure S14. A comparison of the DFT calculated potential energy profiles for the formations of HCHO versus H2 + CO from RhTiO2 − + CH4 + CO2. The relative energies (ΔH0, eV) of the reaction intermediates, transition states, and products with respect to the separated reactants (RhTiO2 − + CH4 + CO2) are given. S18 Figure S15. A comparison of the DFT calculated potential energy profiles for the formations of CH3OH versus 2H2 + CO from RhTiO2 − + CH4 + CO2. The relative energies (ΔH0, eV) of the reaction intermediates, transition states, and products with respect to the separated reactants (RhTiO2 − + CH4 + CO2) are given. It is very important to note that once the reaction complex RhTiO2CH4CO2 − desorbs the simple molecule H2 (RhTiO2CH4CO2 − → RhTiO2CH2CO2 − + H2), the remaining system RhTiO2CH2CO2 − will never have the chance to form CH3OH. The reaction mechanisms of DRM to syngas catalyzed by oxide-supported metal catalysts have attracted considerable attention. Two different proposals, either involving the Langmuir-Hinshelwood (L−H) or the Mars-van Krevelen (M−vK) mechanism, have been discussed to date. [12] Detailed mechanistic studies by isotopic labeling experiments and theoretical calculations reveal that these two mechanisms operate simultaneously in our reaction system of DRM to syngas over RhTiO2 − anions ( Figure S16). After dissociative co-adsorption of CH4 and CO2 onto RhTiO2 − , the intermediate species [CH2] can form [CH2O] moiety and then dehydrogenates to generate CCH4OCO2 (Figures S16a and S17). [13] Alternatively, the CH2 fragment can be directly oxidized by the [TiO2] cluster-support to become CCH4Ocluster, the process of which corresponds to the M−vK mechanism ( Figure S16b). [14] Figure S17. DFT calculated potential energy profile for the coupling of CCH4 and OCO2. The zero−point vibration−corrected energies (ΔH0, in unit of eV) of the reaction intermediates, transition states, and products with respect to the separated reactants (RhTiO2 − + CH4 + CO2) are given.