Methane has been widely used in power plants, vehicles, domestic hot-plates, gas instantaneous water heaters, and gas boilers et al. However, its flame combustion will inevitably result in heavy exhaust emissions containing HC, CO and NOx, and meanwhile the greenhouse effect of unburned methane is almost 20 times higher than that of carbon dioxide.[1] Therefore, it occurs to be very important to develop high-performance catalysts to lower down the light-off temperature below 2008C towards methane oxidation, and also to achieve long-term stability by avoiding sintering and coking. In recent years, researchers have been focusing on rational structural design for advanced catalysts.[2] For instance, self-assembled nanostructures were found to be effective to improve catalytic performance, because orderly arranged nanoparticles (NPs) could be prevented from agglomeration to ensure better exposure of active sites, resulting in significantly boosted activity.[3] The abundance of pores and channels are also conducive to promoting the flow and adsorption/desorption of reactants. In addition, the strong coupling between closely packed components could further improve their catalytic activity arising from a more facilitated reaction pathway via spillover of active species.[4] Fornasiero and co-workers reported the Pd nanospheres encapsulated by different MOx (MOx = TiO2, ZrO2, CeO2) that showed high activity on catalytic methane combustion.[5] Also we have conducted a series of preliminary works to prepare well-defined noble metal and metal oxide nanostructures encapsulated by CeO2, including Pt-CeO2, [6] PdCeO2, [7] Au-CeO2, [8] and AgxAu1x-CeO2 nanospheres,[9] and Cu2O-CeO2 nanocubes,[10] so as to improve the catalytic performance. Photocatalysis has been identified as an effective means by aid of optical energy to trigger chemical reactions under relatively mild conditions. It has achieved considerable success in photocatalytic methane selective oxidation,[11] methane dry reforming,[12] CO2 reduction,[13] and nitrogen fixation.[14] Under light irradiation, photocatalysts could produce highly active photogenerated holes (h+)/electrons (e) that were capable to oxidize or reduce substrates.[15] For instance, the quantum sized BiVO4 with a unique band structure could oxidize water via its photogenerated h+ and meanwhile reduce O2 to generate hydroxyl radical (HOC) via photogenerated e to realize CH4 activation, and finally to produce CH3OH and HCHO.[11] Besides, photothermal catalysis, an energy conversion pathway from light to heat, has been successfully applied for partial oxidation of hydrocarbons,[16] synthesis of light olefins from syngas,[17] hydrogenation of CO2, [18] and CO2 reduction.[19] Widely accepted, CH4 combustion over supported Pd catalysts often follows the classical Mars-van Krevelen (MVK) mechanism[20] that CH4 is first adsorbed on the surface of Pd species, dissociated and dehydrogenated to form CH3* radicals, and then CH3* radicals react with O in PdO to form the final products of CO2 and H2O. Synergistically, PdO with partially lost O atoms would be subsequently re-oxidized by oxidation cocatalysts like CeO2 to realize a fast and stable redox cycle of PdO!Pd!PdO.[21] Based on this mechanism, it includes two key steps to reach a higher efficiency, that is, fast CH4 activation into CH3* radicals to combine O in PdO and effective oxygen spillover from O2 to cocatalysts and then to PdO. As for the former, the rate of this step might strongly depend on surface d-orbital charge of Pd that influences the formation of Pd-O-H+ by dissociating CH4 into CH3* radicals.[22] While, the latter was determined by the migration rate of O2 in cocatalysts in order to maintain the concentration of O in PdO to balance its continuous consumption. The ultimate goal is to realize efficient energy conversion under milder conditions towards catalytic CH4 combustion by aid of stronger coupling of components. In this contribution, we have tried to address the abovementioned issues by utilizing a photoassisted thermal catalytic pathway. The classical PdO/CeO2 catalysts are hence preferably designed, which have been proved to be of good visible-light response. After introduction of Mn3O4, the oxygen storage and release capacity of CeO2 could be further improved caused by the active interface of Mn3O4/CeO2. 

Experimental Procedures Synthesis of HPMC: HPMC was synthesized using the classic auto-redox strategy. Typically, 1 g HNTs were dissolved in 80 mL deionized water under stirring for about 5 min in Ar. Then 1 mL of 0.4 M Ce(NO3)3•6H2O aqueous solution (5.6 wt % of Ce) were added, followed by addition of 0.2 mL of 6 M NaOH. Then, 0.2mL of 0.2 M KMnO4 aqueous and 24.35mg of Pd(NO3)2 (the content of palladium is 38.84%) which dissolved in 0.5 mL deionized water were added at 30 oC and stirred for another 30 min. The precipitate was collected and washed with deionized water several times, followed by vacuum-drying at 60 oC overnight. Finally, all the samples were calcined at 400 oC for 2 h at a heating rate of 10 oC min−1 under air. After calcination, the as-obtained powders were collected and named as HPMC. Synthesis of HMC and HPC: Mn3O4/CeO2 nanocomposites coated on halloysite nanotubes named as HMC, the process was similar to the synthesis of HPMC nanotubes as mentioned above except for addition of Pd(NO3)2. PdO/CeO2 nanocomposites coated on halloysite nanotubes named as HPC, the process was similar to the synthesis of HPMC nanotubes as mentioned above except for addition of Pd(NO3)2. Characterization: The powder X-ray diffraction (XRD) patterns were collected on a Shimadzu XRD-6000 with Cu-Kα radiation (λ = 1.5418 Å), with an operating voltage and current maintained at 40 kV and 30 mA. Transmission electron microscopic (TEM) images were obtained with a JEM-2100F field emission gun transmission electron microscope operating at 200 kV. XPS measurements were performed on an ESCALab250Xi photoelectron spectrometer with Al Kα X-ray radiation as the X-ray source for excitation. UV–Vis diffuse reflectance absorption was detected on Cary 7000, Agilent. N2 sorption was carried out on a NOVA 2200e gas sorption analyzer. The samples were first degassed in vacuum at 300 oC for 4 h prior to N2 adsorption at liquid nitrogen temperature. H2-Temperature programmed reduction (H2-TPR) and CH4-Temperature programmed reduction (CH4-TPR) analyses were carried out in a quartz tube reactor. 30 mg of the samples were balanced at the flow of H2/Ar (5 vol. % of H2, 25 mL min−1 ) or CH4/Ar (1.5 vol. % of CH4, 25 mL min−1 ) for a while, and the quartz tube reactor was heated to 400 °C with a linear heating rate of 10 oC min−1 . O2-Temperature programmed desorption (O2-TPD) analyses were using the same quartz tube reactor. 30 mg of the samples were treated with 20 % O2/He at 400 °C for 30 min (25 mL min−1 , 10 °C min−1 ), then cooled in a 20 % O2/He mixture to 25 °C. The weakly adsorbed oxygen was removed by passing He (25 mL min−1 ) through HPMC for 1 h at 25 °C. The TPD was performed from 25 °C to 400 °C with a ramping rate of 10 °C min−1 in He (25 mL min−1 ). The H2, CH4 and O2 signals were monitored online by Hiden Analytical DECRA mass spectrometer. The electron paramagnetic resonance (EPR) signals were obtained with Bruker-A300-10/12. The visible light irradiation source was a 300 W Xe arc lamp system equipped with a UV cutoff filter (λ > 420 nm), the same light source as that for CH4 combustion test in the following. Catalytic test: 60 mg of catalysts were filled into a quartz reaction tube, the micro photothermal catalytic micro reaction system (CEL-GPPCM, Beijing China Education Au-Light Co., Ltd.) was used for CH4 combustion. The experiment was carried out under a flow of reactant gas mixture (0.5 % CH4, 2 % O2, balance Ar) at a rate of 42 mL min−1 . The samples were irradiated by a 300 W Xe arc lamp with a UV-CUT filter (λ > 420 nm). The composition of the tail gas was monitored online by Hiden Analytical DECRA mass spectrometer.

The X-ray photoelectron spectroscopy (XPS) spectra of Pd 3d, Ce 3d, and O 1s have been presented in Figure S3. The Pd 3d spectrum of HPMC (Figure S3a) could be fitted into four peaks. The two peaks at 337.2 and 342.9 eV correspond to Pd2+ species, which should be assigned to the binding energies of PdO, and another two peaks at 338.0 and 343.9 eV should be related to some highly oxidized Pd4+ species, proving the strong coupling between PdO and CeO2. [1] The characteristic peaks of Ce3+ and Ce4+ both appeared in the Ce 3d spectrum in Figure S3b. As reported, the stable presence of Ce3+ ions in the CeO2 lattice could lead to charge imbalance, creation of oxygen vacancies, and unsaturated chemical bonds.[2] Furthermore, these oxygen vacancies ware able to be used as nucleation centres to promote dispersion of the active phase in catalysts.[3] As for the spectrum of O 1s (Figure S3c), one of the two fitted peaks at 529.8 eV should be assigned to surface lattice oxygen, and the other at 532.4 eV to adsorbed oxygen species. It is also reported that the surface oxygen species do facilitate hydrogen abstraction from CH4 and/or adsorbed CHx species.[4] However, the characteristic signal of Mn 2p was still unable to be monitored as seen in Figure S3d.