Introduction

Upcycling of CO2 into fuels with the use of green hydrogen presents a promising route for addressing the challenges of energy crisis and climate change1,2,3. Among a variety of products from CO2 hydrogenation, CH4 is regarded as an ideal energy vector owing to its merits of high energy density, widely available infrastructure of storage, transportation, and utilization4,5. To date, a broad range of catalytic systems have been developed for CO2 hydrogenation toward CH46. However, because of the inert nature of CO2 and complex reaction network, efficient production of CH4 from CO2 hydrogenation is challenging, suffering from unsatisfactory activity, harsh reaction condition and extensive thermal input7,8,9. It is imperative to explore new and green methods for the conversion of CO2 toward CH4.

Photo-thermal-catalysis presents a synergistic configuration for mediating chemical reactions by simultaneously taking advantage of charge carriers and thermal energy10,11. Thus far, there is a growing number of researches on photo-thermal catalytic CO2 hydrogenation toward CH4 and remarkable progress has been made12,13. For example, Liu et al. reported that Co7Cu1Mn1Ox (200) was active for CH4 synthesis from CO2 hydrogenation with an production rate of 14.5 mmol g−1 h−1 and a selectivity of 85.3% at 200 °C14. Zou’s group demonstrated that Ru@Ni2V2O7 catalyst exhibited CH4 production rate of 114.9 mmol g−1 h−1 and 99.3% selectivity at 350 °C15. Overall, the performance of the catalytic systems is still far away from practical applications and the reaction mechanism remains largely unknown16. It is very desirable to explore a strategy for mediating CO2 hydrogenation toward CH4 with high efficiency and selectivity17.

Among a broad range of CO2 hydrogenation catalysts, Ru-based catalysts exhibit great potential in CO2 hydrogenation toward CH4 because of their unique catalytic properties18. Apart from metal centers, the support also plays a critical role in CO2 hydrogenation by influencing the geometric and electronic properties of active sites. Particularly, MnOx is considered a promising support for hydrogenation reactions due to some obvious advantages19. It is worth of noting that the multiple valences and reducible effect of MnOx confers flexible mediation capability on the catalysts20. The integration of Ru species with MnOx is thus highly promising for efficient CO2 hydrogenation toward CH4.

In this work, a nanostructured Ru/MnOx photo-thermal catalyst composed of well-defined Ru/MnO/Mn3O4 at reaction temperature was designed and prepared for CO2 hydrogenation toward CH4. A prominent CO2 conversion of 66.8% was achieved with a superior selectivity of 99.5% and a CH4 production rate of 166.7 mmol g−1 h−1 at relatively mild temperature of 200 °C (normalized by the amount of catalyst (~15 mg)), which is the result of a combination of external heating and irradiation. The correlative spectroscopic characterizations and theoretical calculations revealed that the structural evolution of Ru/MnOx into well-defined Ru/MnO/Mn3O4 was facilitated by Ru-mediated H-spillover in MnOx and the activity was enhanced by the synergistic effects of photon energy and thermal energy via reducing the activation energy of reaction and accelerating the key intermediate of COOH* species formation over the catalyst.

Results

Fabrication and characterization of Ru/MnOx

The synthesis process of Ru/MnOx was schematically shown in Fig. 1a. Typically, MnOx nanoparticles were first prepared via a straightforward hydrothermal method. Ru sites were then anchored onto MnOx by photo-deposition under argon atmosphere21. The content of Ru in Ru/MnOx was measured by inductively coupled plasma optical emission spectrometry (ICP-OES) (Supplementary Table 1). If not specifically noted, the content of Ru in Ru/MnOx was referred to be 7.3 wt%. The morphologies and structures of the synthesized materials were characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figs. 1b, c, MnOx displayed variable morphologies of hexagonal, octahedral, and square schistose crystals. The morphology of MnOx did not change considerably after the addition of Ru species and the average size of the deposited Ru nanoclusters is about 1.07 ± 0.26 nm (Fig. 1d and Supplementary Figs. 1, 2). The energy dispersive spectroscopy elemental mapping in Fig. 1e exhibited the even distribution of Mn, O, and Ru, which is indicative of the successful synthesis of Ru/MnOx.

Fig. 1: Structural characterization of the catalysts.
figure 1

a Schematic illustration of the synthesis of Ru/MnOx; b SEM image of MnOx; c, d TEM image of MnOx and Ru/MnOx; e STEM image and elemental mapping of Ru/MnOx.

Full size image

The Rietveld refinement of X-ray powder diffraction (XRD) results in Supplementary Fig. 3 and Supplementary Table 2 indicated that the MnOx nanoparticles were mainly composed of Mn3O4 (JCPDS No. 80-0382), MnO2 (JCPDS 72–1806) and MnOOH (JCPDS No. 18-0804)22,23,24. Meanwhile, the XRD pattern of Ru/MnOx showed that the content of MnO2 phase decreased slightly, indicating that the process of photo-deposition of Ru had a slight reduction effect on MnO2. The structure of the samples was further characterized by Fourier transform infrared spectroscopy (FT-IR) spectroscopy (Supplementary Fig. 4). The peaks at 513 and 621 cm−1 were attributed to the distortion vibration of Mn-O in octahedral sites and Mn-O stretching modes in tetrahedral sites, respectively25. Besides, the typical peaks at 947 and 1074 cm−1 were attributed to the vibration of hydroxyl in MnOOH26. Meanwhile, Raman spectrometer was employed to study the metal-support interaction between Ru and MnOx. As illustrated in Supplementary Fig. 5, compared with the pristine MnOx, the introduction of Ru species led to a blue shift of ~7 wavenumbers, and the main peak at 637 cm−1 is assigned to A1g mode of crystalline Mn3O4, validating the strong metal–support interaction between Ru and MnOx27. Moreover, as characterized by CO2 adsorption isotherm and N2 adsorption–desorption isotherms, the addition of Ru species enhanced the CO2 adsorption capacity; and specific surface area of the catalyst was enlarged accordingly (Supplementary Figs. 6, 7). Such improvements are beneficial for the interaction between reactants and mass transfer, thus facilitating CO2 methanation. In addition, X-ray photoelectron spectroscopy (XPS) was examined to gain more insight into surface chemical state and electronic structure of the catalysts (Supplementary Fig. 8). For Ru/MnOx, Ru0 is identified by the peaks observed at ca. 462.7 and 484.8 eV, implying the reduction of Ru species under light irradiation28. Additionally, XPS analysis of the Mn species in all tested samples can be deconvolved into Mn4+, Mn3+, and Mn2+23,27. Apparently, multiple valences of MnOx indicate the reducible nature of as-prepared catalyst.

Photo-thermal catalytic CO2 hydrogenation

The catalytic performance of Ru/MnOx was evaluated at 200 °C in the batch reactor setup by feeding CO2/H2 mixed gas (the desired temperature was achieved by a combination of external heating and irradiation from the Xe lamp) and CH4 was identified as the dominant product, with no liquid products produced (Supplementary Figs. 9, 10). As shown in Fig. 2a, MnOx was hardly active for CO2 hydrogenation toward CH4. In contrast, after the addition of Ru species into MnOx, considerable activity for CH4 formation was achieved, indicating that Ru species could serve as effective active site for the reaction. Notably, the catalytic activity of Ru/MnOx gradually increased with an increasing amount of Ru. At a Ru content of 7.3 wt%, the catalyst displayed a decent CH4 activity of 103.7 mmol g−1 h−1. The CH4 activity was increased by further increasing the loading content of Ru, but the trend slowed down. The effect of CO2/H2 ratios on the yield of CH4 was then examined. It was observed that the yield of CH4 monotonically increased with the H2 proportion in the H2/CO2 mixture (Fig. 2b). A distinct CH4 production rate of 166.7 mmol g−1 h−1 was obtained for Ru/MnOx at a relatively high H2/CO2 ratio of 4/1, highlighting the importance of adequate H2 supply during the reaction to enhance its activity. Furthermore, as shown in Supplementary Fig. 11, we studied the influence of the total pressure on the reaction at high H2/CO2 ratio (4/1). The activity was enhanced markedly with the elevating total pressure, but became slowly when the pressure exceeded 1 MPa.

Fig. 2: Photo-thermal-catalytic performance.
figure 2

a Influence of the Ru content on CH4 evolution rate over Ru/MnOx; b Influence of CO2/H2 volume ratio in the feedstock on CH4 evolution rate over Ru/MnOx; c Temperature-dependent CH4 generation rate over Ru/MnOx under photo-thermal and thermal conditions; d Influence of light intensity on CH4 evolution rate over Ru/MnOx; e Corresponding Arrhenius plot with activation energies noted under photo-thermal/thermal conditions over the Ru/MnOx catalysts. f CH4 evolution as a function of reaction time over Ru/MnOx. Reaction conditions: 15 mg of catalyst, full-arc 300 W UV-xenon lamp, 2.5 W cm−2, 200 °C, irradiation time 4 h, initial pressure 1 MPa (H2/CO2 = 1/1) for Fig. 2a or initial pressure 1 MPa (H2/CO2 = 4/1) for Fig. 2c–f.

Full size image

As shown in Supplementary Fig. 12, the control experiment showed that no carbonaceous products were detected without catalyst or reactant gas, confirming that CH4 was catalytically produced from CO2 hydrogenation. Furthermore, when the photon energy was coupled with external thermal energy, the evolution rate of CH4 was substantially enhanced compared to that achieved by thermo-catalysis in the temperature range examined (Fig. 2c). In addition, when the reaction temperature was kept at 200 °C by an external temperature-controlling system, the CH4 evolution rate of Ru/MnOx could be further enhanced by increasing the light intensity, reaching 166.7 mmol g−1 h−1 at 2.5 W cm−2 (Fig. 2d). These results provide solid support that the introduction of photon energy can significantly enhance CO2 hydrogenation reaction. Meanwhile, the activation energies of Ru/MnOx under thermal and photo-thermal conditions were estimated to be 99.9 and 68.5 kJ mol−1, respectively (Fig. 2e). The decreased activation energy and the corresponding non-parallel plots were indicative of a synergy between photon energy and thermal energy, altering the catalytic mechanism when photons were involved. Moreover, the conversion of CO2 toward CH4 as a function of reaction time is shown in Fig. 2f. It was found that the yield of CH4 continuously increased with the reaction time, suggesting the continuous generation of CH4 from CO2 hydrogenation over Ru/MnOx. Impressively, with a reaction time of 4 h at 200 °C, a considerable CO2 conversion of 66.8% was achieved with a high selectivity of 99.5% with an appreciable CH4 production rate of 166.7 mmol g−1 h−1. Furthermore, the photothermal catalytic performance of the Ru/MnOx catalyst was also assessed in a fixed-bed reactor. As shown in Supplementary Figs. 13, 14, at a gas hourly space velocity (GHSV) of 20,000 mL g−1 h−1, the catalytic activity of Ru/MnOx gradually increased with an increasing temperature, and its activities under photothermal conditions were higher than those under thermal conditions, which further proves that the involved photons were prone to promote the formation of CH4 in the fixed-bed reactor. Meanwhile, under the conditions of 200 °C and 2.5 W cm−2 irradiation, the catalytic activity of Ru/MnOx remained acceptably stable after 20 h at a high GHSV of 40,000 mL g−1 h−1 (Supplementary Fig. 15). A CO2 conversion of 29.5% was achieved with an excellent selectivity of 99.5% and a high space-time yield (STY) of \(95.8\,{{{\rm{mmol}}}_{{CH}_{4}}}\) g−1 h−1. The decrease in activity was probably caused by catalyst agglomeration and carbon deposition as confirmed by the TEM and Thermogravimetric-mass spectrometric (TG-MS) of the catalyst after the reaction (Supplementary Figs. 16, 17). In addition, as characterized by XPS, for Ru/MnOx, after the reaction, the composition of Mn4+ disappeared, and the peak appearing at 647 eV is the satellite peak of Mn2+, demonstrating that the multiphase MnOx support underwent partial reduction29. Ru species were completely converted into Ru0, indicating that the in-situ generated Ru can act as a metal active center to enhance the dissociation of H2 and spillover of hydrogen to the MnOx support (Supplementary Fig. 18). Therefore, the actual composition of the catalyst during the reaction was altered, and the observations above suggested the H-spillover effects in Ru/MnOx, which will be discussed further in the following.

Origin of the superior activity over Ru/MnOx

The optical and electronic properties of the catalyst play a vital role in photo-thermal-catalytic activity. Thereby, they were characterized by various spectroscopy techniques. Firstly, the UV–Vis–NIR diffuse reflectance spectra of different samples were measured to study the light absorption capacity. As illustrated in Fig. 3a, MnOx, as an excellent semiconducting support, exhibits suitable light absorption in the UV–Vis region. The light absorption of MnOx was further improved by the addition of Ru species, having stronger and broader light absorption from UV to NIR wavelength. Meanwhile, due to the broadening of the wavelength range of light absorption, a strong photothermal effect was expected30,31. As shown in Supplementary Fig. 19, under 2.5 W cm−2 illumination, the measured average temperature of Ru/MnOx reached 137.9 °C, higher than that of MnOx (115.4 °C), indicating that both Ru and MnOx contributed to the photothermal effect. In addition, time-resolved photoluminescence spectroscopy showed that the charge carriers lifetime of bare MnOx (τ1 = 1.05 ns, τ2 = 13.51 ns) was obviously prolonged by the incorporation of Ru species (τ1 = 1.33 ns, τ2 = 17.14 ns for Ru/MnOx) in Supplementary Fig. 2032. The behavior of charge carriers was further investigated by transient photocurrent spectra, and the results are plotted in Supplementary Fig. 21. Under light irradiation, Ru/MnOx exhibits a significantly higher photocurrent intensity than MnOx33. The results above show that both optical and electronic properties of MnOx were enhanced by the introduction of Ru species. Moreover, the flat band potentials of MnOx were investigated by Mott-Schottky plots. As shown in Supplementary Fig. 22a, MnOx is confirmed as a p-type semiconductor due to the negative slope. Meanwhile the valence band edge potential was evaluated to be c.a. 0.29 (0.49 eV vs. NHE, ENHE = EAg/AgCl + 0.197), while the flat-band potential is 0.1–0.3 eV lower than the valence band potential in the p-type semiconductor34. In addition, the band gap can be estimated to be 1.26 eV for MnOx (Supplementary Fig. 22b). Consequently, as shown in Supplementary Fig. 23, considering that the work function of Ru is 4.71 eV, the photo-excited electrons can facilely transfer from MnOx to Ru sites under light irradiation35. Together with the unique catalytic properties, the assembled photo-thermal catalyst of Ru/MnOx was therefore highly efficient and selective for catalyzing CO2 hydrogenation toward CH4.

Fig. 3: The mechanism analysis.
figure 3

a UV–Vis–IR absorption spectra of MnOx and Ru/MnOx; b H2−TPR and c CO2-TPD characterization for MnOx and Ru/MnOx; d, e The variable-temperature XRD patterns of MnOx and Ru/MnOx recorded under 20% CO2/H2 atmosphere at different temperatures; f Photographs of WO3 and the mixture of Ru/MnOx and WO3 samples after treatment with H2 at 80 °C with a light intensity of 0.3 W cm−2 for 20 min; g, h XPS spectra of Ru/MnOx in 20% CO2/H2 atmosphere under variable temperature; i Spectra of FT-IR study of Ru/MnOx at different conditions.

Full size image

The characterizations of H2 temperature programmed reduction (H2-TPR) and CO2 temperature programmed desorption (CO2-TPD) were performed to figure out the important role of Ru and MnOx in CO2 methanation. Compared to pristine MnOx, Ru/MnOx exhibited remarkably lowered H2 reduction temperature, which was as low as 140 °C (Fig. 3b), suggesting the superior H2 activation capacity of Ru species. The temperature of H2 activation over Ru/MnOx was even much lower than that for CO2 hydrogenation (200 °C). Hence, CO2 hydrogenation could benefit from low-temperature H2 activation by Ru species under experimental conditions. Moreover, upon the introduction of Ru species, the reduction temperature peaks of MnOx were observed to emerge at low temperature range of 180–250 °C. Herein, the partial reduction of the support MnOx was probably associated with the hydrogen dissociation on Ru and its subsequent spillover to MnOx36. The CO2 adsorption properties were then evaluated by CO2-TPD. As shown in Fig. 3c, compared to pristine MnOx, Ru/MnOx showed remarkable desorption peaks between 250 and 350 °C, indicating that Ru species obviously enhanced the adsorption of CO237. Therefore, the dissociated H can further modulate the hydrogenation activity by directly reacting with CO2 adsorbed on Ru sites or MnOx support, thereby exerting a positive impact on the methanation.

To further explore the Ru-mediated H-spillover effect on the structural composition of Ru/MnOx, qualitative analysis of the crystal phases of Ru/MnOx during the CO2 methanation process was conducted by variable temperature XRD measurements, with a focus on revealing the structural evolution of MnOx at different stages. As illustrated in Fig. 3d, e, all the tested samples are primarily composed of Mn3O4, MnO2, and MnOOH. For MnOx, MnO2 phase can be reduced when the temperature reaches 100 °C. Meanwhile, the transformation of MnOOH to Mn3O4 phase is observed between 100 and 200 °C. In contrast, the Mn3O4 phase remains stable at least 4 h at 200 °C (Supplementary Fig. 24). Interestingly, upon the introduction of Ru species, the variable temperature XRD patterns of Ru/MnOx showed notable changes. The terminal transforming temperature of MnOOH phase markedly decreases from 200 °C to 140 °C. Then, MnO (JCPDS No. 80-0382) with diffraction peaks at 34.7° (111), 40.5° (200), 70.1° (311) and 73.8° (222) were formed when the temperature rises from 140 °C to 200 °C. The observation above is well consistent with the H2-TPR characterization38. As a result, the assembled photo-thermal catalyst of Ru/MnOx was composed of MnO phase and Mn3O4 phase at 200 °C. The phases of materials could remain stable at least 4 h at 200 °C, implying that the Ru/MnOx photo-thermal catalyst composed of well-defined Ru/MnO/Mn3O4 can efficiently and stably catalyze CO2 hydrogenation to CH4 at reaction temperature. Therefore, the phase transformation of multiple-phase MnOx is clearly indicative of Ru-mediated H-spillover effect. Furthermore, as shown in Supplementary Fig. 25, the catalytic activity of MnOx supports surpasses that of the other specific manganese oxide alone. It indicates that the H-spillover effect in Ru/MnOx can effectively transfer dissociated H to the support due to the multivalent states (Mn2+/Mn3+/Mn4+) with varied reducibility, thereby promoting the hydrogenation reaction.

In addition, to investigate the impact of photons on the H-spillover effect under photothermal conditions, we employed WO3 as a means to quantify the extent of H-spillover effect, by which the spillover hydrogen can migrate and readily react with yellow WO3, resulting in a dark coloration39,40. The experiment was conducted at 80 °C with a light intensity of 0.3 W cm−2 under 1 MPa H2 to ensure that the temperature induced by the photothermal effect remained below the designated temperature (Supplementary Fig. 26). As shown in Fig. 3f, it was revealed that the color of WO3 remained unchanged under both photothermal and thermal conditions. In contrast, the mixture of Ru/MnOx and WO3 exhibited a darker color under photothermal conditions compared to thermal conditions. This observation suggests that under photothermal catalysis, the irradiation can enhance the H-spillover effect, thereby promoting the subsequent CO2 hydrogenation reaction.

Meanwhile, XPS was performed to gain insight into the composition and chemical valence. As can be observed in Fig. 3g, h and Supplementary Fig. 27, Ru 3p and Mn 2p peaks of Ru/MnOx displayed a negative shift from 100 °C to 200 °C. Then, the Mn 2p and Ru 3p spectra remain unvaried within 4 h at 200 °C, which is in good agreement with the variable XRD measurements. Supplementary Fig. 28 shows that the binding energies of Ru shift to 462.7 eV, corresponding to Ru0 and the peaks of Mn4+ disappeared. The peak at 647 eV is the satellite peak of Mn2+. Such binding energy shifts indicate that Ru3+ and Mn4+ were reduced under the CO2/H2 mixed gas (H2:CO2 = 4:1), which is consistent with the H2-TPR characterization.

To better understand the reaction at molecular level and explore the impact of the involved photon on the reaction, CO2 hydrogenation was studied by FT-IR under different conditions (Fig. 3i and Supplementary Figs. 29, 30). For thermocatalysis, the typical peaks of monodentate carbonates (m-CO32−, 1509 cm−1) and ν(C-H) vibration of CH4 (1305 cm−1) were apparently strengthened by increasing the reaction temperature41,42. Notably, the intermediate of formate species was observed at 1541 cm−1 (COOH*, ν(OCO)as) and 1373 cm−1 (COOH*, ν(OCO)s) when the reaction temperature increased up to 200 °C43. In contrast, upon light irradiation, the typical peaks of COOH* species appeared at a lower reaction temperature of 170 °C44. This finding further validated the synergy between photon energy and thermal energy on CO2 hydrogenation toward CH4, and COOH* is the most likely key intermediate during either thermo-catalysis or photo-thermal-catalysis. Hence, the involved photons were prone to accelerate the formation of intermediate species, thus can significantly reduce the activation energy of CO2 hydrogenation reaction and promote the formation of CH4.

To better understand the mechanism of the superior performance, first-principles density functional theory calculations were carried out on the basis of the models of Ru/Mn3O4 (321) slabs, Ru/MnO (200) slabs and Ru/Mn3O4−x (321) slabs that simulated the partial reduction of Mn3O4 by the H-spillover effect (Supplementary Fig. 31)45. As shown in Fig. 4 and Supplementary Fig. 32–34, Ru/Mn3O4−x has a more negative Gibbs free energy (ΔG) than both Ru/Mn3O4 and Ru/MnO during the adsorption of CO2, indicating a strong adsorption capacity for CO2, which is beneficial for CO2 hydrogenation (ΔG = −0.914 eV, Ru/MnO; ΔG = −1.475 eV, Ru/Mn3O4; ΔG = −1.651 eV, Ru/Mn3O4−x). Afterwards, notable variations for the subsequent CO2 hydrogenation were observed among Ru/MnO, Ru/Mn3O4, and Ru/Mn3O4-x. The formation of COOH* from CO2* is a rate determining step (RDS) for CO2 hydrogenation over Ru/Mn3O4−x and Ru/MnO, which requires 1.232 and 1.544 eV, respectively. The protonation and subsequent dehydration of COOH* results in the generation of the intermediate of CO*, which is the RDS for the Ru/Mn3O4, (ΔG = 1.918 eV for Ru/Mn3O4). Notably, compared to HCO* formation, the CO* desorption from the catalytic surface as CO is relatively difficult for all the samples46. As a result, it is favorable to yield CH4 via further hydrogenation. It is worth mentioning that in the process of CO2 hydrogenation, ΔG of RDS over Ru/Mn3O4−x (1.232 eV) is obviously lower than that on Ru/Mn3O4 (ΔG = 1.918 eV) and Ru/MnO (ΔG = 1.544 eV), thus facilitating the subsequent hydrogenation steps toward CH4. Moreover, as shown in Supplementary Fig. 35, compared with Ru/Mn3O4−x (321) slabs that simulated dark state, the conduction band that simulated the light state moved to the low energy region, indicating that the involved photons were conducive to electron transfer, which is favorable to CO2 hydrogenation toward CH447,48,49. Together with the FT-IR spectroscopic characterization above, it was rationalized that the synergy between photon energy and thermal energy favored the formation of COOH*, thus exerting a positive impact on the CO2 methanation over Ru/MnOx, which is generated by the partial reduction of Mn3O4 by Ru-mediated H-spillover effect in CO2 hydrogenation50.

Fig. 4: Gibbs free energy pathway for the formation of HCO* and CO from CO2 over Ru/Mn3O4 (321), Ru/Mn3O4-x (321), and Ru/MnO (200).
figure 4

The blue, red, purple, yellow, and green spheres represent the Mn, O, Ru, C, and H atoms, respectively, in the calculation model.

Full size image

Based on the spectroscopic and theoretical investigations above, a possible mechanism for photo-thermal-catalytic CO2 hydrogenation over Ru/MnOx was proposed. The deposited Ru species is highly efficient for H2 dissociation. Benefitting from the strong interaction between Ru and MnOx, Ru-mediated H-spillover effect facilitated the structural evolution of Ru/MnOx into well-defined Ru/MnO/Mn3O4 at low temperature, and the synergy between photon energy and thermal energy could facilitate further hydrogenation of the adsorbed CO2 molecules over Ru and MnOx via accelerating the formation of the key intermediate COOH*. Critically, compared to the process of CO evolution from *CO, Ru/Mn3O4−x is energetically favored for catalyzing *CO hydrogenation toward *HCO, thus promoting the subsequent hydrogenation toward the eventual product of CH4 with high activity and selectivity.

Discussion

To summarize, a nanostructured Ru/MnOx photo-thermal catalyst composed of well-defined Ru/MnO/Mn3O4 at reaction temperature was assembled for CO2 hydrogenation toward CH4. The catalyst illustrated a considerable CO2 conversion of 66.8% with a superior selectivity of 99.5% and a CH4 production rate of 166.7 mmol g−1 h−1 at 200 °C. A series of spectroscopic characterizations and theoretical investigations revealed that benefitting from the strong metal-support interaction between Ru and MnOx, Ru-mediated H-spillover facilitated the structural evolution of Ru/MnOx into Ru/MnO/Mn3O4 at low temperature, and the synergy between photon energy and thermal energy promoted the yield of CH4 via reducing the activation energy of reaction and facilitating the formation of COOH* species. This work opens up a new way for photo-thermal-enhanced CO2 hydrogenation toward CH4.

Methods

Chemicals

MnSO4·H2O (99.99%), MnO (99.9%), MnO2 (99.95%), Mn2O3 (99.9%), WO3 (99.99%) and Ammonium acetate (99.99%) were purchased from Aladdin Chemical Reagent, Ltd. Acetylacetone (99.5%) was provided by Alfa Aesar Chemical Co. Ltd. H2SO4 (AR), CH3COOH (AR) and CH3OH (99.9%) was purchased from Sinopharm Chemical Reagent Co., Ltd. Ruodium (III) chloride hydrate (99%, Ru 37-40%) and NaOH (96%) were obtained from Beijing InnoChem Science & Technology Co., Ltd. Mn3O4 (99.95%) was obtained from Shanghai Macklin Biochemical Technology Co., Ltd. CO2 (99.995%) and H2(99.9995%) were provided by Air Liquid Houlding Co., Ltd., China. Deionized water was used in all the experiments.

Materials synthesis

For the synthesis of MnOx, 1 mmol of MnSO4·H2O and 120 mmol of NaOH were dissolved in 30 mL of deionized water with stirring for 0.5 h. The suspension was subsequently transferred into a 100 mL Teflon liner autoclave. The autoclave was heated at 393 K for 12 h, and then cooled to room temperature. The precipitate was washed with distilled water several times until pH = 7, followed by drying in a vacuum at 333 K overnight to obtain the final product.

For the synthesis of Ru/MnOx, 0.1 g MnOx, 10 mL of CH3OH and suitable amount of RuCl3·3H2O (0.02, 0.04, 0.06, 0.08, 0.1, 0.12 mmol) were added in 50 mL of deionized water with stirring in a glass reactor (250 mL) with a quartz window. The chamber was evacuated and then filled with Ar of 1 atm and irradiated with 300 W UV–Xe lamp for 1 h in 3 W cm−2. The precipitate was washed with distilled water several times, followed by drying in a vacuum at 333 K overnight to obtain the final product. Unless otherwise specified, Ru/MnOx indicates that the added amount of RuCl3·3H2O is 0.1 mmol.

Ru/MnO, Ru/Mn2O3, Ru/Mn3O4, and Ru/MnO2 were synthesized through an identical procedure. The major difference was that the corresponding commercial support was added to the reaction. 0.1 g MnO, Mn2O3, Mn3O4 or MnO2, 10 mL of CH3OH, and 0.1 mmol RuCl3·3H2O were added in 50 mL of deionized water with stirring in a glass reactor (250 mL) with a quartz window. The chamber was evacuated and then filled with Ar of 1 atm and irradiated with 300 W UV-Xe lamp for 1 h in 3 W cm−2. The precipitate was washed with distilled water several times, followed by drying in a vacuum at 333 K overnight to obtain the final product.

Photo-thermal CO2 hydrogenation

Photo-thermal CO2 hydrogenation experiments were carried out in stainless steel reactor of 180 mL (CEL-MPR, Beijing China Education Au-Light Co., Ltd.). In a typical experiment, 15 mg of the catalyst powder was dispersed in 10 mL water in reactor, and the reactor was heated at 333 K overnight to volatilize the solvent and the thin film of catalyst was formed for further experiment (the thickness of the catalyst bed was 25.3 ± 4 µm). Prior to photo-thermal reaction, the reactor was sealed and the air in it was substituted with CO2 of 1 MPa three times and then CO2/H2 mixed gas of 1 MPa with desired ratio was charged at room temperature. Then, the external heating and the 300 W UV-Xe lamp (Beijing China Education Au-Light Co., Ltd) with an intensity of 2.5 W cm−2 both contributed to maintaining the reactor temperature at 200 °C. After the desired reaction time, the gas products were detected by a gas chromatograph (Agilent GC-8860) and calibrated with a standard gas mixture. To detect liquid products, 10 mL of water was injected into the system after the reaction. Possible liquid products such as methanol, ethanol, acetic acid, and acetaldehyde were detected with an Agilent Technology 7890B gas chromatography system with a flame ionization detector using a DB-WAX-UI column. The possible product formic acid was analyzed by HPLC (Waters 2695) equipped with Aminex HPX-87H column, UV/visible detector (WATER2489), and the mobile phase was 5 mM sulfuric acid and the flow rate of 0.7 mL/min. The amount of HCHO was analyzed by using the acetylacetone color-development method. Specifically, 1 mL of the as-prepared acetylacetone solution was mixed with 4 mL of the liquid product in a glass bottle, and heated for 5 min in boiling water. The yellow color of the mixed solution was then investigated. Afterwards, a specific amount of solution was taken out, and examined the UV–vis absorption spectrum by using a Shimadzu UV-2700 spectrophotometer. Through the absorbance intensity at 413 nm, the HCHO concentration was obtained. Typically, 100 mL of acetylacetone solution was first prepared by dissolving 15 g of ammonium acetate, 0.3 mL of acetic acid, and 0.2 mL of acetylacetone in water, and was stored in refrigerator with 2–6 °C.

The photothermal CO2 conversion is also performed in the fixed-bed reactor (CEL-GPPCM, Beijing China Education Au-Light Co., Ltd.) at 200 °C. 150 mg of catalyst and CO2/H2 mixed flow (20 mL min−1/80 mL min−1) were used. A 300 W UV-Xe lamp (Beijing China Education Au-Light Co., Ltd) was used as the light source for the reaction (light intensity: 2.5 W cm-2). The products in the effluent gas were periodically analyzed by using a gas chromatograph (GC-7920, Beijing China Education Au-Light Co., Ltd.). STY of CH4 (\({{{\rm{mol}}}_{{CH}_{4}}}\) g−1 h−1), was calculated according to the following equation

$${{{{{{\rm{CH}}}}}}}_{4}{{{{{\rm{STY}}}}}}=\frac{{F}_{{CO}_{2},{in}}\times {X}_{{CO}_{2}}\times {S}_{{CH}_{4}}}{{W}_{{cat}}\times {V}_{m}}$$
(1)

where \({{F}_{{{\rm{CO}}}_{2}}}\), it is the volumetric flow rate of CO2, \({{X}_{{{\rm{CO}}}_{2}}}\) is the CO2 conversion, \({{S}_{{{\rm{CH}}}_{4}}}\) is the CH4 selectivity, Wcat is the overall mass of catalyst (g), and Vm is the ideal molar volume of CO2 at standard temperature and pressure.

Materials characterization

The morphology of the samples was characterized by a Zeiss Sigma HD SEM and a JEOL JEM−2100F TEM. The high-angle annular dark-field scanning transmission electron microscope was operated by EM-ARM300F. A Rigaku Ultima VI XRD was employed to record the X-ray diffraction patterns with a scanning speed of 5°/min between 10° and 90°, which was operated at 25 kV and 35 mA with Cu Kα radiation. XPS measurements were operated on AXIS Supraelectron spectrometer with Al Kα radiation. BET measurements were carried out by N2 at −196 °C in a Quadrasorb evo. Fourier transform-infrared spectroscopy was performed using Nicolet NEXUS670. UV–VIS–NIR diffuse reflectance spectra were obtained by a UV–VIS–NIR spectrophotometer (UV-3600 Plus, Shimadzu, Japan). Raman analysis was conducted on a Thermo Scientific DXR2 Smart Raman Spectrometer with a 532 nm laser. The adsorption isotherms of CO2 were determined at 273 K on a BELSORP-max II equipment. A liquid nitrogen-cooled charge-coupled device spectrometer (Princeton Instruments) and a microchannel plate photomultiplier tube (Hamamatsu) combined with time-correlated single photon counting technique (Edinburgh Instruments) were used for photon counting and lifetimes measurements under 375 nm excitation. The Ru contents were quantified by an inductively coupled plasma emission spectrometer (ICP-OES) on an Optima 8300. The H2-TPR were measured on Micromeritics AutoChem II chemisorption analyzer with a TCD detector, the sample was heated to 200 °C at 10 °C min−1 in an He flow (50 mL min−1) and then cooled to 80 °C. Next, the sample was heated to 700 °C at 20 °C min−1 in a 10% H2/He mixed flow (50 mL min−1) atmosphere and the outlet gas was detected by TCD. The CO2-TPD was measured on Micromeritics AutoChem II chemisorption analyzer with a TCD detector. The sample was heated to 200 °C at 10 °C min−1 in an He flow (50 mL min−1) and then cooled to 80 °C. Next, a 10% CO2/He mixed flow (50 mL min−1) was introduced to the catalyst bed for 0.5 h. The sample was then exposed to He (50 mL min−1) for 0.5 h to remove the weakly adsorbed CO2 from the surface. Finally, the sample was heated to 700 °C at 10 °C min−1 in a He atmosphere and the outlet gas was detected by TCD. The temperature of samples was recorded by an infrared thermal imaging camera (Fotrfic 315, Shanghai Thermal Imaging Technology Co., Ltd.). Considering the limited ability of the reactor window composed of aluminum oxide to penetrate 7–15 µm of infrared light, the image captures the internal temperature of the reactor by quickly removing the reactor window. Variable temperature XRD measurements were collected by an X-ray diffraction patterns (D8 Advance). The samples were heated in CO2/H2 mixed flow (10 mL min−1/40 mL min−1) from 25 °C to 200 °C. Data collection after maintaining the specified temperature for 5 min. XPS measurements under variable temperature were operated on Thermo EXCALAB 250Xi electron spectrometer with Al Kα radiation. The catalysts were held on the sample holder and activated with illumination in the pretreatment chamber under CO2/H2 mixed flow (10 mL min−1/40 mL min−1) from 25 °C to 200 °C. The sample was then introduced into the ultrahigh-vacuum chamber for XPS measurement at room temperature after maintaining the specified temperature for 5 min. FT-IR spectra under variable temperature were recorded with a NICOLET iS50 FTIR spectrometer (Thermo SCIENTIFIC, USA) equipped with a high-temperature reaction chamber and a mercury cadmium telluride (MCT) detector at a resolution of 4 cm−1 and 32 scans per spectrum. The background spectrum was scanned after mixture gas (CO2:H2 = 1:4) was introduced. TG-MS analyses were performed on a thermogravimetric analyser (NETZSCH STA449 F3-QMS403D) instrument under air. The catalysts were held on the sample holder and the reactor was sealed until the air in it was substituted with CO2/H2 mixed flow (10 mL min−1/40 mL min−1). After that the background spectrum was recorded. Upon reaching the desired temperature through simultaneous external heating and irradiation, the system was maintained for 5 min. Subsequently, the light source was removed for FT-IR measurement. The thickness of the catalyst bed in the batch reactor was measured by laser scanning confocal microscopy LEXT OLS5100.

H-spillover effect detection by WO3

In a typical experiment, a mixture containing 1 g of WO3 and 0.015 g of catalyst was placed in a quartz glass culture dish. Then the quartz glass culture dish was placed in stainless steel reactor of 180 mL (CEL-MPR, Beijing China Education Au-Light Co., Ltd.). Prior to photo-thermal reaction, the reactor was sealed, and the air was replaced by H2 for three times, followed by filling with H2 (1 MPa). Then, the external heating and the 300 W UV-Xe lamp (Beijing China Education Au-Light Co., Ltd) with an intensity of 0.3 W cm−2 both contributed to maintain the reactor temperature at 80 °C. After the desired reaction time, the color change of the powder samples was recorded.

The photoelectrochemical (PEC) tests

The photoelectrochemical tests of the samples were carried out on an electrochemical workstation (CHI660e, Chenhua Instrument, Shanghai, China) by using a three- electrode system. The catalyst was drop-coated on clean FTO glass, which was used as a working electrode, while Pt and Ag/AgCl electrodes acted as counter and reference electrodes, respectively. A 300 W Xe lamp (Aulight, Beijing) acted as the light source and all of the electrochemical tests were carried out in 0.1 mol L−1 sodium sulfate solution.

Computational methods

We have employed the Vienna Ab initio Simulation Package to perform all density functional theory (DFT) calculations. The elemental core and valence electrons were represented by the projector augmented wave method and plane-wave basis functions with a cutoff energy of 400 eV. Generalized gradient approximation with the Perdew–Burke–Ernzerh of (GGA-PBE) exchange-correlation functional was employed in all the calculations. Geometry optimizations were performed with the force convergency smaller than 0.05 eV/Å. The spin-polarization effect was also considered. A climbing image nudged elastic band method was used to locate the transition states with the same convergence standard. The spin-polarization effect was also considered. The DFT-D3 empirical correction method was employed to describe van der Waals interactions. The DFT + U approach was introduced to treat the highly localized Mn 2p states, using parameters of U–J = 4. For FM of Mn3O4 and MnO, the initial magnetic moments of Mn atoms were set to be +5 μB. Monkhorst-Pack k-points of 1 × 1 × 1 and 2 × 2 × 1 were applied for all the surface calculations of Ru-Mn3O4 and Ru-MnO. Half atoms at bottom were fixed in all the calculation. The Gibbs free energy was calculated by the following equation: ΔG = ΔE + ΔEZPE − TΔS, where the value of ΔE, ΔEZPE, and ΔS denotes the changes of DFT energy, the zero-point energy and the entropy at 473.15 K, respectively.