Unlocking High-Efficiency Methane Oxidation with Bimetallic Pd–Ce Catalysts under Zeolite Confinement

Catalytic complete oxidation is an efficient approach to reducing methane emissions, a significant contributor to global warming. This approach requires active catalysts that are highly resistant to sintering and water vapor. In this work, we demonstrate that Pd nanoparticles confined within silicalite-1 zeolites (Pd@S-1), fabricated using a facile in situ encapsulation strategy, are highly active and stable in catalyzing methane oxidation and are superior to those supported on the S-1 surface due to a confinement effect. The activity of the confined Pd catalysts was further improved by co-confining a suitable amount of Ce within the S-1 zeolite (PdCe0.4@S-1), which is attributed to confinement-reinforced Pd–Ce interactions that promote the formation of oxygen vacancies and highly reactive oxygen species. Furthermore, the introduction of Ce improves the hydrophobicity of the S-1 zeolite and, by forming Pd–Ce mixed oxides, inhibits the transformation of the active PdO phase to inactive Pd(OH)2 species. Overall, the bimetallic PdCe0.4@S-1 catalyst delivers exceptional outstanding activity and durability in complete methane oxidation, even in the presence of water vapor. This study may provide new prospects for the rational design of high-performance and durable Pd catalysts for complete methane oxidation.


Experimental details
The metal loading was determined using inductively coupled plasma -mass spectrometry (ICP-MS) on a Horiba Ultra 2 instrument equipped with a photomultiplier tube detector.
Powder X-ray diffraction (PXRD) patterns were obtained using a Bruker D8 Advance X diffractometer equipped with a Cu Kα source at ambient conditions. The data were collected at a scan rate of 10 ° min -1 over a scanning range of scattering angle 2θ from 10° to 80°. Nitrogen physisorption isotherms were measured on an ASAP 2020 automatic surface analyzer (Micromeritics, USA). Prior to the measurements, all samples were degassed at 250 °C under vacuum for 10 h. The specific surface area was calculated from the Brunauer−Emmett−Teller (BET) model, and the t-plots model was used to determine the external surface area and micropore volume. Thermal gravimetric analysis was conducted on thermogravimetric analyzer STA 449 system (Netzsc, Germany). The sample was heated from room temperature to 800 °C in air at a heating rate of 10 °C min -1 .
Transmission electron microscopy (TEM) images were obtained using a JEOL JEM-2100HR. The sample was dispersed in ethanol by ultrasound, and a few drops of the suspension were tiled on a copper grid. High resolution Transmission electron microscopy (HRTEM) images were obtained using a FEI Tecnai G2 f20 microscope operating at an accelerating voltage of 200 kV. In addition, to confirm the exact location of Pd and Ce within the S-1 zeolite, the resin-embedded catalysts were ultra-microtomed and sliced into 80-nm-thick sections using a diamond knife, following the procedures described in previous studies. 1,2 To further prove the spatial distribution of Pd and Ce in the PdCe0.4@S-1 catalyst, high-angle annular dark-field scanning TEM (HAADF-STEM), corresponding energy-dispersive X-ray (EDX) mapping images (EX-37001), and line scans were performed. X-ray photoelectron spectroscopy (XPS) analysis was conducted on a Physical Electronics Quantum 2000 equipped with monochromatic Al Kα source (Kα = 1486.6 eV) and a charge neutralizer. The binding energies of all the elements were calibrated using the C 1s line at 284.8 eV. Moreover, to determine the valence of Pd and Ce atoms within the zeolite, depth-profile analysis was performed using an XPS system with an argon ion etch gun. The test was conducted using 3000 eV of ion energy and 180 s of etch time.
The Raman test was performed using a LabRAM Aramis Raman spectrometer (HYJ, France). The excitation light source has a wavelength of 325 nm (ultraviolet) and the scanning range is 200-1400 cm -1 . Low-temperature electron paramagnetic resonance (EPR) was 4 measured using a JEOL electron spin resonance spectrometer.
The O 2 temperature-programmed desorption (O 2 -TPD) experiment was carried out using a Micromeritics Autochem II 2920. First, the catalysts were pretreated in a He flow (50 mL min -1 ) at 300 °C for 1 h. After cooling to 60 °C, the flow was switched to 5.0 vol % O 2 /He and kept for 1 h to adsorb O 2 . Finally, the temperature was increased to 800 °C in He (50 mL min -1 ) at a heating rate of 10 °C min -1 , while recording the thermal conductivity detector (TCD) signal. CO chemisorption was performed using the Micromeritics AutoChem II 2920. Prior to adsorption, a 50 mg sample was pretreated at 300 °C for 1 h in 10 vol % H 2 /Ar and cooled down to 30 °C in Ar. Pulse chemisorption was carried out with pulses of 0.5243 mL of 10 vol % CO/He until saturation, which corresponds to 2.14×10 -6 mol of saturated CO amount (CO S ).
The Pd dispersion was estimated using the following equation: Pd dispersion (%) = (mole of exposed Pd atom) / (mole of total Pd atom of catalyst) = (mole of adsorption of CO) / (mole of total Pd atom of catalyst).
The adsorption of CO was estimated using the following equation: adsorption of CO (mol) = (peak area of CO adsorption) / (peak area of saturated CO) × CO S (peak area of CO was obtained by integrating the data in Figure S5).
The CH 4 temperature-programmed reduction (CH 4 -TPR) experiment was carried out using the same apparatus that was equipped with a HIDEN HPR-20 mass spectrometer (MS). First, 50 mg of the catalyst was pretreated at 300 °C for 40 min in Ar (50 mL min -1 ), then cooled to room temperature. The TPR test was conducted by increasing the temperature of the samples to 800 °C at a heating rate of 10 °C min -1 in 1.0 vol % CH 4 /Ar (30 mL min -1 ). The signal of CH 4 (m/z=16), as well as CO 2 , CO and H 2 (m/z=44, 28, 2) were recorded online by the MS.
In situ diffuse reflectance infrared Fourier transform (DRIFT) spectra were obtained using a Nicolet Nexus spectrometer equipped with a liquid nitrogen-cooled mercury cadmium telluride (MCT) detector. The sample cell featured KBr windows. DRIFT spectra of the catalysts were collected with a resolution of 4 cm min -1 and 64 scans in absorbance units. For the DRIFT spectra of CO adsorption, the samples were pretreated with 10 vol % H 2 /Ar (30 mL min -1 ) at 300 °C for 1 h. Then, 1.0 vol % CO/Ar (30 mL min -1 ) was passed through the cell at 25 °C, and the IR spectra were monitored. Similarly, DRIFT spectra of CH 4 adsorption on the catalysts were measured. For the CH 4 temperature-programmed oxidation with O 2 and without 5 O 2 experiments, prior to each experiment, the catalyst was heating to 550 °C for 1 h under a He flow to remove surface impurities and then cooled to 30 °C to obtain a background spectrum.
Afterward, 1.0 vol % CH 4 /He (30 mL min -1 ) was flowed to the IR cell, and the temperature was increased from 30 °C to 450 ℃ at various intervals while allowing for adsorption for 30 min.
The IR spectra were monitored in real-time. Note: As shown in Figure S4b, a stronger intensity of Ce 3d is observed over PdCe 2.2 @S-1 compared to PdCe 0.4 @S-1. This suggests that although the Pd species were encapsulated within S-1 zeolites of PdCe 2.2 @S-1 catalyst, partial CeO 2 aggregated on the outer surface of S-1 due to the excessive addition of Ce. This result is consistent with the slightly decreasing specific surface area. After Ar + sputtering, Figure S4c   Note: It is generally well-documented that higher metal dispersion signifies smaller metal particle size 6,7 . However, in comparison to Pd@S-1, catalysts containing Ce with the same confined structure exhibit a larger amount of exposed Pd atoms, but with a larger particle size. Therefore, it is reasonable to conclude that the particle size of PdCe x @S-1 catalysts gradually increases with increasing Ce loading due to the formation of Pd-CeO 2 mixed oxide particles within PdCe x @S-1, rather than purely Pd NPs. Additionally, the higher Pd dispersion of PdCe x @S-1 catalysts indicates that the form of Pd-CeO 2 interaction further enhances the number of accessible Pd sites. However, as a result of the higher Ce loading, PdCe 2.2 @S-1 displays a larger particle size (3.3 nm) (Figures S4e and S4f), but less Pd dispersion compared to PdCe 0.4 @S-1, which is likely due to the over-accumulation of excessive CeO 2 surrounding the Pd NPs.    Note: Figure S11 shows the CH 4 conversions at various temperatures within the temperature tolerance range of PdCe 0.4 @S-1, Pd@S-1 and Pd/S-1, starting from 300 °C and increasing up to 600 °C and then decreasing back to 300 °C. As shown earlier in Figure 3a, PdCe 0.4 @S-1 and Pd@S-1 exhibit initially higher activity compared to Pd/S-1. Notably, after cooling down to 300 °C, the methane conversion over PdCe 0.4 @S-1 and Pd@S-1 only slightly decreased from 91% to 88% and 82% to 78%, respectively. In contrast, the conversion over Pd/S-1 decreased from 28% to 17%, indicating the exceptional temperature tolerance of PdCe 0.4 @S-1 and Pd@S-1 with the help of their confined structures.    Note: The absorption bands at 3016 and 1300 cm -1 correspond to the asymmetric stretching 15 vibration ν as (C-H) and deformation vibration δ(C-H) of CH 4 , respectively. 8 Additionally, the signals around 1750 and 3500-3800 cm -1 are attributed to the hydroxyl groups (OH − ) adsorbed on the surface, which originate from CH 4 oxidation.

Additional information of catalysts
Pd and Ce content of catalysts