Applied Catalysis B: Environmental 291 (2021) 120038 Contents lists available at ScienceDirect Applied Catalysis B: Environmental journal homepage: www.elsevier.com/locate/apcatb Effects of solar irradiation on thermally driven CO2 methanation using Ni/ CeO2–based catalyst Viktoria Golovanova a, *, Maria Chiara Spadaro b, Jordi Arbiol b, c, Viacheslav Golovanov d, Tapio T. Rantala e, Teresa Andreu a, Joan Ramón Morante a, f a Catalonia Institute for Energy Research (IREC), Jardins de les Dones de Negre 1, 08930, Sant Adrià de Besòs, Spain b Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, Bellaterra, Barcelona, Catalonia, 08193, Spain c ICREA, Pg. Lluís Companys 23, 08010, Barcelona, Catalonia, Spain d South–Ukrainian National University, Staroportofrankovskaya str. 26, 65020, Odessa, Ukraine e Department of Physics, Tampere University, P.O. Box 692, FI-33014, Tampere, Finland f Department of Electronics, University of Barcelona (UB), Marti i Franquès 1, 08028, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Utilization of the renewable energy sources is one of the main challenges in the state-of-the-art technologies for CO2 methanation CO2 recycling. Here we have taken advantage of the solar light harvesting in the thermocatalytic approach to LSPR carbon dioxide methanation. The large-surface-area Ni/CeO2 catalyst produced by a scalable low-cost method Photothermal effect was characterized and tested in the dark and under solar light irradiation conditions. Light-assisted CO2 con­ Nickel-ceria catalyst DRIFTS version experiments as well as in-situ DRIFT spectrometry, performed at different illumination intensities, have revealed a dual effect of the incident photons on the catalytic properties of the two-component Ni/CeO2 catalyst. On the one hand, absorbed photons induce a localized surface plasmon resonance in the Ni nanoparticles fol­ lowed by dissipation of the heat to the oxide matrix. On the other hand, the illumination activates the photo­ catalytic properties of the CeO2 support, which leads to an increase in the concentration of the intermediates being precursor for methane production. Analysis of the methane production at different temperatures and illumination conditions has shown that the methanation reaction in our case is controlled by a photothermally- activated process. The used approach has allowed us to increase the reaction rate up to 2.4 times and conse­ quently to decrease the power consumption by 20 % under solar illumination, thus replacing the conventional thermal activation of the reaction with a green energy source. 1. Introduction the metal oxide support is used as the main photoabsorber and metallic nanoparticles act only as active catalytic sites [16,17], a rapidly Conversion of carbon dioxide to renewable hydrocarbon fuels, such emerging field in catalysis is taking advantage of the strong light as methane, is a big challenge nowadays due to the high thermodynamic extinction of the metallic nanoparticles themselves, owning to the effect stability of the CO2 molecule and, therefore, major difficulties in its called localized surface plasmon resonance (LSPR) [18,19]. The reduction [1–3]. The standard thermocatalytic approach for CO2 LSPR-driven catalysis is mostly making use of noble metals (Au and Ag) methanation requires relatively high temperatures (150− 500 ◦ C) [4], due to their strong plasmonic response in the visible and UV part of the which gives rise to a number of alternative approaches to replace or spectrum [20–24]. However, other abundant and non-expensive metals, assist thermochemical catalysis [5–8]. Plasma-assisted methanation of in particular some of the VIII group transition metals, such as Ni have CO2 was reported, being efficient even at room temperature [9,10] with also proven to exhibit plasmonic features [11,25–29], although less promising performances. Another important field of research is related pronounced due to the higher damping. Moreover, nickel nanoparticles to the utilization of solar light as an additional green energy source for in combination with metal oxides presenting high oxygen mobility, such CO2 hydrogenation to methane [11–15]. Except for the standard pho­ as CeO2, have been proposed as a candidate catalyst for the efficient tocatalytic approach, based on the metal-decorated metal oxides, where thermal conversion of CO2 to methane [30]. * Corresponding author. E-mail address: vholovanova@irec.cat (V. Golovanova). https://doi.org/10.1016/j.apcatb.2021.120038 Received 26 November 2020; Received in revised form 3 February 2021; Accepted 15 February 2021 Available online 23 February 2021 0926-3373/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Excited LSPR is followed by a strong near-field enhancement, which constant stirring. The obtained NiO/CeO2 powder was further reduced is then damped through one of the possible mechanisms: i) radiative in hydrogen atmosphere (5% H2 and 95 % Ar) for 7 h at 450 ◦ C to get the decay through re-emitting of the photons; ii) Landau damping, resulting Ni/CeO2 catalyst (Scheme 1). in a non-equilibrium charge distribution (so-called “hot electrons” or For comparison, samples without Ni were also prepared using the “hot holes”, which then can be transferred to the metal-oxide support or same template impregnation technique. 6.5 mL of 0.6 M cerium nitrate directly participate in the catalytic reaction); or iii) direct excitation of hexahydrate was added to 0.9 g of SBA-15, and the same two-step electron-hole pairs in the semiconductor through plasmon-induced procedure was applied for calcination and second impregnation of sil­ resonant energy transfer (PIRET). Another effect that has been ica template. recently introduced, consists in direct metal-to-semiconductor charge transfer [31,32]. All of the damping mechanisms stated above occur in 2.2. Physico-chemical characterization the timescale of 1–100 fs and, eventually, lead to recombination of the created non-equilibrium charge carriers (0.1–100 ps) and result in XRD spectra were recorded on a Bruker type XRD D8 diffraction- strong local heating (0.1–10 ns) [33]. The plasmon damping has found a meter (CuKα radiation, λ = 1.5418 Å, 40 kV, 40 mA) with a scanning lot of applications, and many works are dedicated to exploiting of the range of 2Θ from 20◦ to 80◦ and step size of 0.05◦ /3 s. The average “direct” improvement of catalytic properties by plasmonic hot electrons crystalline sizes of metallic nanoparticles were derived from the Scher­ [20,25,34–37]. Kλ rer’s equation D = βcosθ , where Θ is the Bragg angle, K is the shape factor, However, few works have emphasized the importance of the cata­ lytic performance enhancement of the Ni-based catalysts due to λ is the X-ray wavelength and β is the full width of the diffraction peak at plasmon-induced self-heating of nickel nanoparticles followed by a half maximum (FWHM). dissipation of heat to the oxide matrix. Due to the excellent photo­ Scanning electron microscope (SEM, Zeiss Auriga 60) was used to thermal properties of Ni nanoparticles [26,38–40], this effect in many study the morphology of the samples. The composition of the synthe­ cases can be considered as the main consequence of the plasmon exci­ tized catalyst was studied by energy dispersive X-ray spectroscopy (EDX, tation in Ni-based catalysts as compared with direct hot electrons Oxford Inca Energy). interface transfer. In this work, we have used a wet impregnation of N2-physisorbtion analysis was conducted in TriStar II 3020-Micro­ mesoporous silica SBA-15 template to prepare a high surface area metrics equipment at liquid nitrogen temperature. Before the measure­ Ni/CeO2 catalyst for CO2 methanation where a large size distribution of ments, the samples were degassed under vacuum conditions at 90 ◦ C for Ni nanoparticles allows us to perform highly selective and energetically 1 h and then at 250 ◦ C for 3 h. Brunauer-Emmett-Teller (BET) method efficient methanation of CO2 assisted by plasmonic heating of the cat­ was applied for calculation of the BET surface area for a relative pressure alysts with solar light. The use of in-situ characterization has revealed (P/P0) range of 0.1− 0.3. The total pore volume and the average pore the mechanism of CO2 hydrogenation and the optimal conditions for the size were determined by applying Barrett-Joyner-Halenda (BJH) plasmon-assisted methanation reaction, which allows us to increase the method to desorption branch of the isotherms at the value P/P0 = 0.95. productivity up to 2.4 times and reduce the energy consumption by 20 % H2-TPR (hydrogen temperature programmed reduction) and CO- as compared to the conventional thermochemical approach. Further­ chemisorption were carried out using an automated chemisorption more, we have corroborated that under solar illumination conditions, in analyzer (Autochem HP-Micrometrics). In case of H2-TPR measurements spite of the significant increase of the formate intermediates due to the 100 mg of powder sample was kept under constant 5% H2/Ar flow, photocatalytic properties of CeO2, the rate of methane production is while the temperature was increased from 25 ◦ C to 800 ◦ C with a heating limited by the thermally-controlled step, related to formate hydroge­ ramp of 10 ◦ C/min. The consumed H2 was measured using a thermal nation reactions. A complete understanding of these phenomena is conductivity detector (TCD). The CO-chemisorption measurements were fundamental for the new design of efficient plasmonic held at 35 ◦ C under the flow of 10 %CO/He over the samples, which NP-semiconductor-based catalysts for thermally driven methanation were pre-reduced at 450 ◦ C under 5% H2/Ar flow for 7 h. Pulses of CO reactors. were periodically introduced until full saturation of the system. Active metal surface area and metal dispersion were calculated assuming the stoichiometric ratio CO/Ni equal to 1, Ni atomic weight to 58.71, its 2. Experimental atomic cross-sectional area to 0.0649 nm2 and density to 8.9 g/cm− 3. UV–vis light absorption spectra were measured using the Perkin 2.1. Catalyst synthesis Elmer Lambda 35 UV–vis spectrometer in the diffuse reflectance mode. Absorption of the samples was calculated according to the equation: A = SBA-15 mesoporous silica templates were synthesized by the 1 − R, as no transmission was observed. following process, based on the one described in [41]: 9 g of Pluronic® High-resolution transmission electron microscopy (HRTEM) P123 (polyethylene glycol-block-polypropylene glycol-block-poly­ together with scanning TEM (STEM) investigation was performed on a ethylene glycol, average Mn ~5,800, Sigma-Aldrich) was mixed with 90 field emission gun FEI Tecnai F20 microscope. High angle annular dark- g of diluted HCl (50 wt.%) and 245 g of Milli-Q water and stirred at 38 ◦ C field (HAADF) STEM was combined with electron energy loss spectros­ to obtain polymeric micelles. 18.7 g of TEOS (tetraethyl orthosilicate, 98 copy (EELS) in the Tecnai microscope by using a GATAN QUANTUM %, Sigma-Aldrich) was added as a silica precursor. The reaction was held filter. for another 24 h at 100 ◦ C, after which the mixture was filtered and eventually calcined at 550 ◦ C for 5 h. The Ni/CeO2 catalyst was prepared by simultaneous impregnation of 2.3. Functional characterization SBA-15 templates (1.05 g) with 6.5 mL of 0.6 M cerium (III) nitrate hexahydrate (Sigma-Aldrich) and 3 mL of 0.6 M nickel (II) nitrate A custom-made flow-type gas reactor (Scheme S1(b)) was created in hexahydrate (Sigma-Aldrich). The Ni/Ce ratio was chosen in such a way order to allow simultaneous heating and illumination of the catalyst. 50 that the estimated Ni weight concentration in the final product would be mg of the catalyst were spread uniformly over an aluminum support, 15 %. The reaction was held in two steps, to achieve a high degree of whereas its temperature was constantly measured by a K-type thermo­ impregnation [41,42]: first with impregnation of the solutions to the couple. An intimate contact was preserved throughout the reaction in mesoporous silica and annealing at 350 ◦ C for the preliminary precursor order to measure directly the changes in the catalyst temperature. decomposition, and second impregnation with half of the initial The catalyst was preliminarily reduced in hydrogen atmosphere (5% amounts, followed by annealing of the sample at 800 ◦ C. The silica H2 and 95 % Ar) for 7 h at 450 ◦ C. The methanation reaction was held template was further dissolved in 2 M NaOH for 12 h at 70 ◦ C with under a constant flow of 80 % H2 and 20 % CO2 gas mixture (99.999 %, 2 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Scheme 1. Ni/CeO2 catalyst fabrication process. Linde) at a rate of 20 mL min− 1. The outlet gas flow passed through a 3. Results and discussion cold liquid-gas separator (5 ◦ C) for water condensation, and was measured by a mass flow meter (MF, Bronkhorst). The gas products 3.1. Physical characterization of the catalyst (CH4, CO, CO2 and H2) were analyzed using a 490 microGC gas chro­ matograph (Agilent Technologies). SEM and TEM images (Fig. 1) show that the obtained catalyst con­ The methane yield was calculated according to the equation: sists of CeO2 nanofibers support (with a diameter of ~10 nm) and well- distributed Ni nanoclusters with the size distribution approximately XCO2 (%) × SCH4 (%) YCH4 (%) = , from 5 to 50 nm. The EDX analysis shows the presence of 12 %(w.t.) Ni, 100% which is in a good proximity with the initial estimations. Some residual silicon was spotted, which supposedly comes from the non-dissolved where XCO2 (%) = is the conversion of CO2 and SCH4 (%) = ΦCO2 ,in − ΦCO2 ,out ΦCO2 ,in SBA-15 template, but its amount was as low as 1% of the catalyst mass. is the selectivity to methane. ΦCO2 ,in , ΦCO2 ,out are the CO2 The ceria support, shown in Fig. 1(d), has a typical channel shape ΦCH4 ,out ΦCO2 ,in − ΦCO2 ,out molar flows in the inlet and in the outlet, respectively, and ΦCH4 ,out is the due to the SBA-15 template where each channel is 8 nm distant to the CH4 molar flow in the outlet. The catalytic activity was then evaluated closest. This configuration of the support leads to its mesoporous from the following equation: properties and allows the small Ni clusters to penetrate inside the ceria structure. k= Φ YCH4 (%) × HRTEM study has shown that the used synthesis procedure is suit­ m 100% able to create a ceria-supported catalyst with a presence of highly The reported data were the average values, taken from five mea­ distributed small Ni nanoparticles (~5 nm) which have a strong inter­ surements for each temperature point, with the measurement error less action with the ceria support (Fig. 2). They are mostly located in the than 3%. Carbon mass balance was closed within the accuracy of ± 7%. voids between the ceria rods. CO2 methanation over Ni/CeO2 is a As a source of illumination, we used a 300 W solar simulator (Xe complex reaction, requiring simultaneous H2 and CO2 activation. The lamp, AM1.5 G filter) with a variable illumination power. The power former occurs on the metallic Ni sites and the latter on the basic sites of density was measured by a laser powermeter (Gentec-EO UNO) con­ CeO2 [43]. The challenge of the catalyst design is that products of these taining a calibrated reference silicon cell. Bandpass filters (FGL400S, two reaction steps (H* and COx) are separated by a spatial constraint, FGS900S, ThorLabs) were introduced to the system to cut a part of the which has to be minimized in order to achieve better catalytic perfor­ spectrum as shown in Scheme S1(b). mance. From this point of view, highly distributed Ni nanoclusters Provided that methanation is an exothermic reaction, the tempera­ embedded in the mesoporous CeO2 (Fig. 1(d)) represent the optimal ture was allowed to reach a stable value in all of the studied conditions catalyst configuration for CO2 methanation. We assume that the high before any illumination was introduced. exposure of the boundary phase Ni/CeO2 to the gas is responsible for the significant CO2 conversion. From the low-magnification images (Figs. 1(d); S1(a), Ni panel) we 2.4. In-situ DRIFTS characterization can also see the presence of bigger Ni clusters (~20− 50 nm). They are located mostly on the surface of the catalyst and therefore play the major In-situ diffuse reflectance infrared Fourier transform spectroscopy role in the solar light absorption and scattering. Relatively high distri­ (DRIFTS) was used in order to study the nature of intermediate species bution of the nanoclusters’ sizes is responsible for broadening of the formed during the methanation reaction over the Ni/CeO2 catalyst plasmonic extinction peak and black coloring of the reduced powder. under dark and light-assisted reaction conditions. The spectra were The N2-physisorption analysis of the silica SBA-15 template, CeO2 recorded using Bruker-Vertex70 spectrophotometer equipped with a and Ni/CeO2 replica has shown that the obtained powders possess a high-temperature reaction control cell with two ZnSe windows, one mesoporous structure with a high BET surface area. The surface areas quartz window and a catalytic bed. The catalyst was pre-reduced at 450 and pore dimensions of pure CeO2 support and Ni/CeO2 catalyst used in ◦ C under 5% H2/He flow of 50 mL/min for 1 h, then purged in He gas for this study are quite similar, which indicates that Ni impregnation has not 40 min to remove the residual H2 from the cell and finally cooled down influenced the mesoporous morphology of the ceria support. Moreover, to desired temperature. The background spectrum was recorded in He at the Ni/CeO2 catalyst exhibits a good dispersion of Ni nanoparticles with room temperature. CO2 methanation reaction was performed in 20 mL/ high metal surface area, which was derived from CO-chemisorption min of gas mixture (75 % of stoichiometric H2/CO2 ratio = 4 and 25 % of analysis (Table 1). He balance) at 100 ◦ C, 150 ◦ C, 200 ◦ C, 225 ◦ C and 250 ◦ C. Each mea­ H2-TPR analysis was used to study reduction properties of the ob­ surement was recorded after 30 min in isothermal conditions, and fol­ tained catalyst (Fig. 3). The Ni/CeO2 sample presents a strong peak at T lowed by purging in He for 30 min to remove the excess water =328 ◦ C and two smaller ones at T =213 ◦ C and T =271 ◦ C, attributed to molecules. Spectra were recorded from 4200 cm− 1 to 1000 cm− 1 with a the reduction of the weakly bound NiO species on the CeO2 surface [44]. step of 1 cm− 1. Thermal analyzer Thermostat was used to control the The more intensive one is related to the reduction of smaller nano­ product stream by their molecular weights (m/z ratios). The illumina­ particles, whereas the low-temperature peaks are due to the large NiO tion (5 suns) during the DRIFTS measurement was applied through the particles on the CeO2 support. The peak at T =442 ◦ C was assigned to the quartz window from the solar simulator used during the functional reduction of strongly bound NiO [45], and the high-temperature process characterization. 3 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Fig. 1. SEM image of the Ni/CeO2 catalyst (a) and the corresponding EDX spectrum (b); HAADF STEM (c) and TEM (d) general view images of the sample at low magnification. at T =697 ◦ C was attributed to the Ce4+ to Ce3+ reduction in CeO2 [46]. corroborates that the CeO2 {111} planes suffer a 4–5 % compression The H2-TPR analysis confirmed that the reduction temperature of close to the Ni nanoparticle, which is in line with the previous XRD 450 ◦ C (dashed line on Fig. 3), chosen for NiO/CeO2 catalyst is sufficient analysis. to reduce all the types of NiO clusters. EELS chemical composition maps prove that the Ni present in the reduced sample is mostly metallic 3.2. Catalytic performance of the catalyst in the CO2 methanation (Fig. S1). reaction Fig. 4 shows XRD spectra of the as-prepared samples and reduced under H2 atmosphere (5% H2 in Ar, 7 h, 450 ◦ C). As we can see from the We performed functional characterization of the synthetized catalyst figure, NiO phase, that was present in the as-prepared catalyst, disap­ in the tubular-type reactor in dark conditions (Scheme S1(a)). The pears completely in favor of metallic Ni. The peaks attributed to CeO2 measured catalytic activity at different reaction temperatures is shown are shifted with respect to the clear CeO2 phase in case of the non- on the Fig. 5(a). We found that CO2 conversion reaches 80 % at 300 ◦ C, reduced samples, which indicates the incorporation of Ni atoms and and the selectivity to methane was as high as 95 %. nanoclusters to CeO2, causing stress onto the lattice [47]. This shift can Due to its high surface area the catalytic activity of Ni/CeO2 catalyst be clearly seen on the inset of Fig. 4, where the CeO2 diffraction peak used in this work is comparable to the activity of the other Ni/CeO2 (56.34◦ ) present in both pure CeO2 and reduced Ni/CeO2 samples, is catalysts reported previously [49,50]. shifted to 56.41◦ in the case of non-reduced NiO/CeO2. After reducing, We have also carried out a stability test which has revealed a high all Ni is removed from the CeO2 lattice, which results in shifting of the stability of the catalyst under constant reaction conditions at T =270 ◦ C peaks back to the positions of the CeO2 phase. It is also worth noticing, (Fig. 5(b)). that the peak is broadened with respect to the pure CeO2 support, which indicates the stress and the lattice parameter inhomogeneity due to the presence of the Ni atoms and nanoclusters inside of the CeO2 rods. The 3.3. Photothermal effects crystallite sizes of Ni particles were estimated by Sherrer’s equation using the peak of 2Θ = 44.51◦ . The average size was found to be 27.9 UV–vis spectroscopy was used to measure the reflectance of bare nm, which is in accordance with the nanoparticle dimensions extracted CeO2 and Ni/CeO2 samples. The obtained data were transformed into from the TEM data. absorption as no transmittance was observed. As it can be seen from the In order to analyze the possible strain influence of the Ni nano­ Fig. 6(a), bare cerium oxide powder does not absorb under IR and visible particles embedded in the CeO2 mesoporous channels, we used the illumination, but starts absorbing strongly starting from 430 nm, which Geometrical Phase Analysis (GPA) routines in the Digital Micrograph corresponds to the reported band gap of oxygen vacancy-enriched CeO2 software [48]. Our GPA analyses showed a relative compression of the nanoparticles (2.87 eV) [51]. In contrast, samples with Ni nanoparticles CeO2 lattice planes when they are close to a Ni nanoparticle (Fig. S2). exhibit an extraordinary high absorption over all of the measured illu­ The profile obtained along the dark blue arrow pointed out in the figure mination range. We attribute this significant rise in the visible and IR light absorption to the localized surface plasmon resonance of the Ni 4 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Fig. 2. HRTEM micrographs of a detail of the non-reduced NiO/CeO2 (left) and reduced Ni/CeO2 (right) nanostructures. In the top panel we report the magnification in the squared area and the corresponding indexed power spectra. In the bottom panel, we report the frequency filtered maps evidencing NiO, Ni and CeO2 planes with different colors. Although the plasmonic Ni nanoparticles have a higher damping Table 1 factor comparing to frequently used Au or Ag nanoparticles, the overlap N2-physisorption and CO-chemisorption analysis data of the Ni/CeO2 catalyst, of their absorption spectrum with the solar one is much better [55]. This pure CeO2 support and SBA-15 mesoporous silica template. ensures an effective absorption of the solar illumination and trans­ Sample Surface Micropore Pore Ni Active metal forming it to heat. area, m2/ volume, cm3/ size, dispersion, surface area, g g nm % m2/gNi We have used three different light intensities and operated at three different starting temperatures in order to study the effects related to the SBA-15 866.2 0.99 6.6 – – photoinduced catalytic activity. Also, in order to investigate the nature CeO2 121.1 0.21 3.3 – – Ni/ 103.5 0.20 2.9 1.3 8.6 of the activity enhancement, we performed wavelength-dependent tests CeO2 by introducing bandpass filters into the experimental setup. UV filter (λ > 400 nm) was used to cut off the UV part of the solar spectrum, while the IR filter (λ < 800 nm) removed the near-IR and IR photons. Trans­ nanoparticles, which is known to have a broad band in the IR-vis range, mission spectra of the used filters are shown on the Fig. 6(b). We have as was previously reported [52–54]. used the combination of two bandpass filters in order to study the pure To take advantage of the extra light absorption of the catalyst used in effect of visible light on the CO2 conversion. It is worth noting that this work, we performed a series of CO2 methanation tests under solar introducing of the UV filter solely does not affect the catalytic perfor­ illumination. We observed a significant increase in the CH4 yield under mance of the Ni/CeO2 catalyst, indicating that the absorption of light by illumination, which was concomitant with a temperature increase in the CeO2 is not responsible for the increase in the catalytic activity. reactor. Fig. 7(a) sums up the changes in catalytic activity that occur in the 5 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 methane production enhancement. Eventually, the illumination of the catalyst may increase the rate of the methanation reaction as high as 2.4 times under suitable reaction conditions (Fig. 7(b)). 3.4. DRIFTS characterization To study the effect of illumination on the intermediate steps of the reaction, we have performed in-situ DRIFTS experiments at different starting temperatures. The measurements were held both in dark con­ ditions and under solar illumination. The corresponding spectra are shown on the Fig. 8 except the band at 2350 cm− 1 corresponding to gas phase CO2. The pure CeO2 samples without Ni nanoclusters show presence of hydroxyl groups with vibrations υ(OH) identified at 3725-3560 cm− 1 and carbonate species only (Fig. S3). At 100 ◦ C the surface is covered by bidentate carbonates (1586, 1300 cm− 1), bridged carbonates (1646, 1120 cm− 1) and hydrogen carbonates (1681, 1630, 1422, 1392 and 1212 cm− 1), presumably formed over OH groups on the CeO2 surface. Fig. 3. H2-TPR profile of the CeO2 support (a) and Ni/CeO2 catalyst (b). The temperature increase stimulates transformation of these carbonates into more stable mono- and polydentate carbonates (1483–1500, 1357 and 1150 cm− 1). In the absence of Ni◦ the dissociation of the H2 mole­ cule is not favored, and, therefore, formates are not formed on pure CeO2 surface. For Ni/CeO2 catalyst at 100 ◦ C no methane is produced, and CO2 is mainly adsorbed on the basic Ce3+ sites [56,57], forming carbonate and hydrogen carbonates species, which are known as precursor for creation of formates. The bands at 1680 cm− 1, 1417 cm− 1 and 1209 cm− 1 were attributed to hydrogen carbonates vibrations, and the peak at 1642 cm− 1 was assigned to bridged carbonates. The broad band (1472− 1530 cm− 1) and a peak at 1150 cm− 1 were attributed to mono- and poly­ dentate carbonates. At 150 ◦ C the weakly bound bridged carbonates are being transformed into mono- and polydentate carbonates and a new bands (1583, 1371 and 1343 cm− 1) attributed to formates begin to grow [58]. Starting from 200 ◦ C a band attributed to production of methane appears at 3016 cm− 1. It grows further with the temperature increase, Fig. 4. XRD spectra of the CeO2 support (a), non-reduced (b) and reduced (c) while the intensity of the formate peaks is being diminished. On the Ni/CeO2 catalyst. The inset shows the shift of the 56.335 ◦ CeO2 diffraction peak other hand, the band, attributed to mono- and polydentate carbonates, in the non-reduced Ni/CeO2 catalyst with respect to the reduced one and the pure CeO2 support. continuously grows with the temperature increase. This testifies that formates are the precursor in the methane formation, while the mono- and polydentate carbonates species do not take significant part in the catalyst under solar light illumination at different temperatures, light reaction. intensities and illumination ranges. As it can be seen from the graph, Under illumination, the DRIFTS spectra undergo some qualitative introducing of visible light has a notable effect on the reaction rate, changes. At 100 ◦ C the peaks at 1583 and 1371 cm− 1, attributed to vi­ while the full solar spectrum induces an even higher increase to the bration of monodentate formate (υasym(COO)) become much more catalytic activity. As it was stated previously, cutting off the UV part of pronounced, while the peaks related to hydrogen carbonate species the solar spectrum (hυ > 3.1 eV) does not affect the reaction rate, thus (1680, 1417 or 1209 cm− 1) are distinctly decreased. neglecting the role of band-band electron transitions in ceria in the Another significant feature of the DRIFTS spectra, recorded under Fig. 5. (a) Temperature dependence of the Ni/CeO2 catalytic activity in dark conditions; (b) Stability test of the Ni/CeO2 catalyst. 6 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Fig. 6. (a) Reflectance absorption spectra of the bare CeO2 and Ni/CeO2 samples. Backround yellow color represents the visible irradiation band used in this work. (b) Transmission spectra of the bandpass filters used in this work. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). Fig. 7. Catalytic activity of the Ni/CeO2 catalyst under visible light and full solar illumination of 3 to 5 suns at three different starting temperature points (200 ◦ C, 225 ◦ C and 250 ◦ C) (a); Relative increase of the catalytic activity under illumination with respect to the dark reaction at the corresponding conditions (b). Solid bars correspond to the visible illumination, whereas the full bar corresponds to the full solar illumination. The light intensities correspond to the ones measured before introducing of the filters. illumination is appearance of the carbonyl species on the Ni surface, carbonates on the reduced active sites of CeO2, mostly present at the identified at 2041 cm− 1 (monocarbonyl), 1927 cm− 1 (bridged carbonyl) Ni/CeO2 interface. At the same time, H2 molecules dissociate on the and 1830 cm− 1 (two-fold bridged carbonyl). These bands are growing metallic Ni clusters, facilitate C– – O double-bond breaking and creation with temperature increase up to 200 ◦ C and afterwards turn to decrease of the formates followed by their step-by-step hydrogenation to CH4. along with formation of methane. This evidently indicates that car­ This mechanism of CO2 activation is energetically more favorable than bonyls take part in the process of CO2 hydrogenation to methane. The the route through CO formation, and is ensured by high mobility of the formation of carbonyls on the Ni surface may occur either through the oxygen vacancies in ceria. The latter allows partial reduction of the CeO2 direct dissociation of CO2 or through decomposition of formate species. material by H2 at the Ni/CeO2 interface with subsequent formation of water molecules, while keeping the active sites reduced. In dark conditions, the DRIFTS data recorded from our samples are in 3.5. Mechanism of photothermal CO2 methanation agreement with described above associative mechanism, indicating that stable bridged formate species serve as a main precursor for CO2 hy­ Typically, CO2 hydrogenation on the Ni-based catalysts with inactive drogenation to methane, probably through formation of formaldehyde support, such as Al2O3 or SiO2 occurs through the dissociation mecha­ and methoxy adsorbates. However, when the samples are illuminated, nism, i.e. formation of carbonyl or formate intermediates on the Ni0 we observe immediate changes in the DRIFTS spectra. In particular, active sites [43]. On the other hand, Ni/CeO2 catalysts were found to variations of the peaks suggest that hydrogen carbonates are more likely favor the associative mechanism of CO2 methanation [59–61]. It means transformed to formates. Besides, we detect the appearance of carbonyl that the CO2 molecules are chemisorbed as monodentate or bidentate 7 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Fig. 8. In-situ DRIFTS spectra measured during CO2 methanation over the Ni/CeO2 catalyst (a) in dark conditions; (b) under solar illumination (5 suns). species, which may be formed through dissociation of excessive for­ chemistry of the catalyst as a result of photoelectronic processes. The mates [62]. obtained changes in the DRIFTS spectra indicate that illumination is Based on the obtained results, we suggest that illumination has a promoting the stabilization of formate species from less stable hydrogen two-fold effect on the operational performance of the Ni/CeO2 catalyst. carbonates. First, visible and IR irradiations excite LSPR in Ni nanoparticles. As the illumination of the catalyst is followed by the temperature Although plasmons in Ni are effectively damped, they result in strong increase (summarized in the Table S1), we have further analyzed the local heating of the nickel nanoclusters with following heat dissipation data from Fig. 7 in order to find out the nature of the improved catalytic to the surrounding oxide matrix. Such a plasmon-induced increase of the activity. A thermally activated methanation reaction follows an expo­ temperature facilitates the principal reaction channel at the Ni/CeO2 Ea nential trend, i.e. k ∼ e− kT , where Ea is the activation energy of the interface. reaction, or the rate-limiting step [63]. Therefore, ln(k) ∼ − T1 for the On the other hand, the illumination is changing the surface Fig. 9. Arrhenius slope of the temperature activated methanation reaction. The y axis indicates the natural logarithm of catalytic activity under different reac­ tion conditions. 8 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 process that is fully controlled by the temperature. We have plotted a natural logarithm of the measured catalytic activities in all of the applied conditions versus the inverse temperature (Fig. 9). Remarkably, all the light-assisted measurements of the catalytic activity at low con­ version rates were followed the same linear trend, indicating that the solar light illumination of the Ni/CeO2 catalyst caused an increase of the catalytic activity due to the local temperature changes induced by plasmonic heating of the Ni nanoparticles. From the Fig. 9 we have evaluated the Arrhenius slope for the methanation reaction (dashed line), with the activation energy of 0.96 eV. As discussed above, we attribute this energy to the methanation pathway that takes place through hydrogenation of the formates on ceria, presumably to the dissociation of formalin to formaldehyde, as it was shown in [64]. The light-induced increase in the concentration of formates, seen from the DRIFTS data, does not affect the main reaction channel, the latter being totally controlled by temperature. However, we have found that with increase of the illumination intensity the formate and carbonyl peaks on the DRIFTS spectra become more pronounced. We applied Gaussian multi-peak fitting to the spectra measured under dark conditions, 3 and 5 suns illumination in order to quantify the contribution of the formate intermediates depending on the intensity of the photon flux (Fig. S4). We found that the amount of adsorbed species is proportional to the power of the applied illumination, which indicates that the electronic effect of the solar illumination can be much more Fig. 10. Differential DRIFTS spectra of the Ni/CeO2 catalyst measured at starting temperature of 200 ◦ C under different wavelength illumination. noticeable when higher light intensities are applied. However, when reaching higher temperatures, one can observe a strong deviation from the linear trend towards lower catalytic activities, compared DRIFTS spectra of the catalyst under illumination with which points out on a new rate-limiting process that occurs at higher spectra recorded at elevated temperature in dark conditions (Fig. S5). temperatures. As the selectivity to methane was not altered upon the The differential graph exhibits the same features as in the Fig. 10, which temperature increase (see Table S2), this deviation cannot be explained proves the non-thermal nature of the light-induced changes on the by the activation of a side reaction, i.e. producing of carbon monoxide catalyst surface. (reverse water gas-shift reaction), which is enabled at higher tempera­ According to DRIFTS data, solar light promotes stabilization of the tures [65]. It is well known that the main deactivation pathways for the formate species on ceria surface. The increased concentration of for­ Ni-based catalysts are sintering of the Ni nanoparticles and coke for­ mates, however, does not facilitate the overall methane production at mation [66]. However, these processes are also taking place at higher the Ni/CeO2 interface, which follows the Arrhenius slope (Fig. 9). operation temperatures. On the other hand, Cárdenas-Arenas et al. The surface of cerium oxide is known to be enriched with oxygen conducted isotopic experiments on Ni/CeO2 catalyst and showed that vacancies, which can trap photogenerated electrons under UV light water desorption is the slowest step of the methanation mechanism [61]. excitation [67,68]. Furthermore, it was reported that chemisorption of As it was mentioned before, the light-assisted reaction was held in a CO2 becomes favorable on the reduced CeO2 surface with abundance of custom-designed planar-type reactor under low gas flows. Apparently, at additional charge [57,69]. Therefore, it is plausible that UV and blue higher CO2 conversion values the physisorbed water molecules, formed light illumination, present in the solar spectrum, can generate in big amounts during the reaction, are not efficiently removed from the electron-hole pairs in ceria that activate surface vacancy sites and reactor and block the active sites of the catalyst. Thus, in our case, enhance sorption processes at CeO2 surface. Photogenerated carriers desorption of water becomes a new rate-limiting step in the methanation facilitate transformation of the carbonate species to the more energeti­ process at high conversion rates. We assume that changing of the reactor cally stable formates, which is reflected in the corresponding changes of design to being suitable for high-flow operation conditions will allow to DRIFTS spectra. The concentration of formates is increased throughout examine the effect of strong illumination on methanation reaction rate the ceria support, including the sites apart from the Ni/CeO2 interface. with no limitations due to poisoning of the active sites by water On the other hand, hydrogenation of formates, being the rate-limiting molecules. step of the methanation reaction, requires active hydrogen atoms, To understand the effect of different wavelengths on the catalyst we which are abundant at the interface with nickel nanoparticles. Thus, the performed a series of DRIFTS experiments under solar illumination with formates adsorbed on ceria surface apart from the interface are lacking different optical filters. The dark spectra were subtracted from the ones hydrogens and do not contribute effectively to the overall methane recorded under illumination in order to obtain differential data production. (Fig. 10). The results show that the UV part of the solar spectrum has the The suggested mechanism of CO2 methanation on the Ni/CeO2 largest effect on the growth of formate and carbonyl peaks, while the catalyst is schematically shown on Fig. 11 and can be described as fol­ visible light has a much weaker influence on the formate concentration. lows. In dark conditions, CO2 is adsorbed on the active sites at Ni/CeO2 In the table S3 we have summarized the relative amounts of incident interface as bridged carbonates or hydrogen carbonates, which are light intensities, used in this work. The intensity of transmitted light further transformed to formates and monodentate carbonates. Under with the energy sufficient to excite band-to-band transitions in CeO2 constant supply of hydrogen atoms from Ni nanoclusters, the formates at (250− 400 nm) sums up to 10.5 % of the total solar spectrum, which the interface are hydrogenated to methane. Solar illumination has a two- results in a non-negligible amount of photogenerated carriers in the fold effect on the reaction. First, Ni nanoparticles exhibit a rapid local ceria support. heating due to the strongly damped localized surface plasmons gener­ It should be noted that the real temperature of the illuminated ated on Ni. This leads to an increase of the average temperature of the catalyst is higher than the nominal value of the starting point. To ensure catalyst and to the overall increase of the methane production. Secondly, that the DRIFTS changes do not occur due to temperature-induced ef­ UV and visible illumination is enhancing photocatalytic properties of fects, caused by the heating of the catalyst under solar light, we have ceria. Photoinduced charge carriers facilitate transformation of 9 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 Fig. 11. Proposed mechanism of CO2 methanation under solar illumination. hydrogen carbonates to formates, giving rise to the corresponding peak on the DRIFTS spectra. However, the newly formed species adsorbed apart from the Ni/CeO2 interface do not contribute to the methanation due to the limited hydrogen supply. 3.6. Energy consumption One of the effective ways of exploiting the plasmon-induced photo­ thermal effect of the Ni/CeO2 catalyst for CO2 methanation is to main­ tain the constant temperature of the reaction, by controlling the power supplied to the reactor. After compensation of the increased temperature to the initial value by reducing the power consumption, no improvement in the reaction rate was observed, which confirms that the photothermal effect of the illumination on the main reaction pathway is dominant at low temperatures and light intensities. However, we have measured the power, supplied to the heater of the reactor in the dark conditions (Pdark ), and the corresponding power under illumination (Plight ), reduced in order to maintain the working temperature at its initial value. Then, the reduced power consumption (RPC) for all of the experimental con­ ditions used in this work was calculated according to the following formula: Pdark − Plight RPC = ∙100% Pdark The results are presented in the Fig. 12. We can see that using concentrated sun illumination we are able to Fig. 12. Power consumption savings under visible light and full solar illumi­ save up to 20 % of power applied for the conventional thermocatalytical nation of 3 to 5 suns at three different temperatures: 200 ◦ C, 225 ◦ C and 250 ◦ C. approach to CO2 methanation. Utilization of renewable solar energy Solid bars correspond to the visible illumination, whereas the full bar corre­ sponds to the full solar illumination. source and an earth-abundant material with plasmonic properties for facilitation of effective thermal-driven CO2 methanation allowed us to gain energy for further lowering of the cost, scalability and industriali­ rate under solar light illumination, which allowed us to decrease the zation of the process. power consumption by 20 % with respect to the dark conditions. In-situ DRIFTS analysis revealed the mechanism of the CO2 metha­ 4. Conclusions nation in dark and light-assisted conditions. In dark conditions the CO2 methanation is following the associative route through formate hydro­ In summary, we have demonstrated that the catalytic activity of the genation on CeO2, using Ni0 active sites only for H2 dissociation and methanation reaction can be improved by the photothermal effect due to spillover. When illuminating the catalyst with solar light, two main ef­ localized surface plasmonic properties of the Ni nanoparticles in the Ni/ fects were observed. On the one hand, the visible and IR illumination CeO2 catalyst under visible and infrared light illumination. Using of the results in a strong photothermal plasmon-induced effect, followed by mesoporous SBA-15 silica template allowed us to synthetize large- local heating and enhanced methane production. On the other hand, UV surface-area Ni/CeO2 catalyst, which ensures highly stable and effi­ and blue light induced photogenerated free carriers in CeO2 can improve cient conversion of CO2 (80 %) with high selectivity to methane (95 %). the CO2 coordination and promote the creation of formates. However, The obtained catalyst exhibits up to a 2.4-fold increase in the reaction their further hydrogenation remains fully controlled by the 10 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 photothermally activated process due to the limited hydrogen supply to [2] G. Centi, E.A. Quadrelli, S. Perathoner, Catalysis for CO2 conversion: a key technology for rapid introduction of renewable energy in the value chain of the formate species. Investigation of the rate-limiting step by means of chemical industries, Energy Environ. Sci. 6 (2013) 1711–1731, https://doi.org/ the first-principles calculations will be the subject of the future work. 10.1039/c3ee00056g. At high conversion rates, the activity of the Ni/CeO2 catalyst in the [3] A. Álvarez, M. Borges, J.J. Corral-Pérez, J.G. Olcina, L. Hu, D. Cornu, R. Huang, current reactor design is damped by the slow water desorption process D. Stoian, A. Urakawa, CO2 activation over catalytic surfaces, Chem. Phys. Chem. 18 (22) (2017) 3135–3141, https://doi.org/10.1002/cphc.201700782. from the active Ni0 sites. Therefore, we have utilized the plasmon- [4] M. Younas, L. Loong Kong, M.J.K. Bashir, H. Nadeem, A. Shehzad, S. Sethupathi, induced photothermal effect of the Ni/CeO2 catalyst for CO2 methana­ Recent advancements, fundamental challenges, and opportunities in catalytic tion by operating at low reaction temperatures, reducing the power methanation of CO2, Energy Fuels 30 (11) (2016) 8815–8831, https://doi.org/ 10.1021/acs.energyfuels.6b01723. supplied to the reactor. Nevertheless, we believe that new reactor design [5] M.A.A. Aziz, A.A. Jalil, S. Triwahyono, A. Ahmad, CO2 methanation over may open a possibility of a thorough study and exploitation of the new heterogeneous catalysts: recent progress and future prospects, Green Chem. 17 light-induced methanation route at higher illumination intensities and (2015) 2647–2663, https://doi.org/10.1039/c5gc00119f. [6] J. Wu, Y. Huang, W. Ye, Y. Li, CO2 reduction: from the electrochemical to conversion rates, thus further reducing the power consumed by the photochemical approach, Adv. Sci. 14 (11) (2017) 1700194, https://doi.org/ reactor. 10.1002/advs.201700194. Our results emphasize the two-fold effect of solar illumination on the [7] R.P. Ye, J. Ding, W. Gong, M.D. Argyle, Q. Zhong, Y. Wang, C.K. Russell, Z. Xu, A. G. Russell, Q. Li, M. Fan, Y.G. Yao, CO2 hydrogenation to high-value products via plasmonic Ni-based catalysts: photothermal and electronic. Hence, we heterogeneous catalysis, Nat. Commun. 10 (2019) 5698, https://doi.org/10.1038/ expect the combined photo- and thermocatalytical approach to provide s41467-019-13638-9. a promising pathway for Ni-based heterogeneous catalysts for green CO2 [8] K. Manthiram, B.J. Beberwyck, A.P. Alivisatos, Enhanced electrochemical methanation of carbon dioxide with a dispersible nanoscale copper catalyst, J. Am. hydrogenation. Chem. Soc. 136 (38) (2014) 13319–13325, https://doi.org/10.1021/ja5065284. [9] M.C. Bacariza, M. Biset-Peiró, I. Graça, J. Guilera, J. Morante, J.M. Lopes, CRediT authorship contribution statement T. Andreu, C. Henriques, DBD plasma-assisted CO2 methanation using zeolite- based catalysts: Structure composition-reactivity approach and effect of Ce as promoter, J. CO2 Util. 26 (2018) 202–211, https://doi.org/10.1016/j. Viktoria Golovanova: Writing - original draft, Conceptualization, jcou.2018.05.013. Investigation. Maria Chiara Spadaro: Investigation, Writing - review & [10] Y. Ge, T. He, D. Han, G. Li, R. Zhao, J. Wu, Plasma-assisted CO2 methanation: effects on the low-temperature activity of an Ni–Ce catalyst and reaction editing. Jordi Arbiol: Investigation, Writing - review & editing. Via­ performance, R. Soc. Open Sci. 6 (2020), 190750, https://doi.org/10.1098/ cheslav Golovanov: Writing - review & editing, Conceptualization. rsos.190750. Tapio T. Rantala: Writing - review & editing. Teresa Andreu: Writing - [11] X. Dai, Y. Sun, Reduction of carbon dioxide on photoexcited nanoparticles of VIII review & editing, Supervision, Conceptualization. Joan Ramón Mor­ group metals, Nanoscale 11 (2019) 16723–16732, https://doi.org/10.1039/ c9nr05971g. ante: Writing - review & editing, Supervision, Conceptualization. [12] U. Ulmer, T. Dingle, P.N. Duchesne, R.H. Morris, A. Tavasoli, T. Wood, G.A. Ozin, Fundamentals and applications of photocatalytic CO2 methanation, Nat. Commun. 10 (2019) 3169, https://doi.org/10.1038/s41467-019-10996-2. Declaration of Competing Interest [13] P.G. O’Brien, A. Sandhel, T.E. Wood, A.A. Jelle, L.B. Hoch, D.D. Perovic, C. A. Mims, G.A. Ozin, Photomethanation of gaseous CO2 over Ru/silicon nanowire The authors declare that they have no known competing financial catalysts with visible and near-infrared photons, Adv. Sci. 1 (1) (2014) 1400001, https://doi.org/10.1002/advs.201400001. interests or personal relationships that could have appeared to influence [14] Ş. Neat ̧u, J.A. Maciá-Agulló, P. Concepciòn, H. Garcia, Gold-copper nanoalloys the work reported in this paper. supported on TiO2 as photocatalysts for CO2 reduction by water, J. Am. Chem. Soc. 136 (45) (2014) 15969–15976, https://doi.org/10.1021/ja506433k. [15] F. Sastre, A.V. Puga, L. Liu, A. Corma, H. García, Complete photocatalytic Acknowledgments reduction of CO2 to methane by H2 under solar light irradiation, J. Am. Chem. Soc. 136 (19) (2014) 6798–6801, https://doi.org/10.1021/ja500924t. This work has been done in the framework of the doctorate pro­ [16] W.N. Wang, W.J. An, B. Ramalingam, S. Mukherjee, D.M. Niedzwiedzki, S. Gangopadhyay, P. Biswas, Size and structure matter: enhanced CO2 gramme in Materials Science of the Autonomous University of Barce­ photoreduction efficiency by size-resolved ultrafine Pt nanoparticles on TiO2 single lona. V. Golovanova has received funding from the European Union’s crystals, J. Am. Chem. Soc. 134 (27) (2012) 11276–11281, https://doi.org/ Horizon 2020 research and innovation programme under the Marie 10.1021/ja304075b. [17] O. Ola, M. Mercedes Maroto-Valer, Role of catalyst carriers in CO2 photoreduction Skłodowska-Curie grant agreement No 754397. Authors acknowledge over nanocrystalline nickel loaded TiO2-based photocatalysts, J. Catal. 309 (2014) Generalitat de Catalunya for financial support through the CERCA 300–308, https://doi.org/10.1016/j.jcat.2013.10.016. Programme, M2E (2017 SGR 1246) and the CSC – IT Center for Science, [18] S.C. Warren, E. Thimsen, Plasmonic solar water splitting, Energy Environ. Sci. 5 (1) (2012) 5133–5146, https://doi.org/10.1039/c1ee02875h. Espoo, Finland for provided resources. IREC also acknowledges addi­ [19] P. Wang, B. Huang, Y. Dai, M.H. Whangbo, Plasmonic photocatalysts: harvesting tional support by the European Regional Development Funds (ERDF, visible light with noble metal nanoparticles, Phys. Chem. Chem. Phys. 14 (2012) FEDER) and by MINECO coordinated projects MAT2014-59961-C2, 9813–9825, https://doi.org/10.1039/c2cp40823f. ENE2016-80788-C5-5-R and ENE2017-85087-C3. [20] H. Song, X. Meng, T.D. Dao, W. Zhou, H. Liu, L. Shi, H. Zhang, T. Nagao, T. Kako, J. Ye, Light-enhanced carbon dioxide activation and conversion by effective M. C. Spadaro and J. Arbiol acknowledge funding from Generalitat plasmonic coupling effect of Pt and Au nanoparticles, ACS Appl. Mater. Interfaces de Catalunya 2017 SGR 327. ICN2 is supported by the Severo Ochoa 10 (1) (2018) 408–416, https://doi.org/10.1021/acsami.7b13043. programme from Spanish MINECO (Grant No. SEV-2017-0706) and is [21] H.M. Gong, L. Zhou, X.R. Su, S. Xiao, S.D. Liu, Q.Q. Wang, Illuminating dark plasmons of silver nanoantenna rings to enhance exciton-plasmon interactions, funded by the CERCA Programme / Generalitat de Catalunya. M. C. Adv. Funct. Mater. 19 (2) (2009) 298–303, https://doi.org/10.1002/ Spadaro has received funding from the European Union’s Horizon 2020 adfm.200801151. research and innovation programme under the Marie Skłodowska-Curie [22] G. Kumari, X. Zhang, D. Devasia, J. Heo, P.K. Jain, Watching visible light-driven CO2 reduction on a plasmonic nanoparticle catalyst, ACS Nano 12 (8) (2018) grant agreement No. 754510 (PROBIST) and the Severo Ochoa 8330–8340, https://doi.org/10.1021/acsnano.8b03617. programme. [23] W. Hou, W.H. Hung, P. Pavaskar, A. Goeppert, M. Aykol, S.B. Cronin, Photocatalytic conversion of CO2 to hydrocarbon fuels via plasmon-enhanced absorption and metallic interband transitions, ACS Catal. 1 (8) (2011) 929–936, Appendix A. Supplementary data https://doi.org/10.1021/cs2001434. [24] C.G. Silva, R. Juá rez, T. Marino, R. Molinari, H. García, Influence of excitation Supplementary material related to this article can be found, in the wavelength (UV or visible light) on the photocatalytic activity of titania containing gold nanoparticles for the generation of hydrogen or oxygen from water, J. Am. online version, at doi:https://doi.org/10.1016/j.apcatb.2021.120038. Chem. Soc. 133 (3) (2011) 595–602, https://doi.org/10.1021/ja1086358. [25] H. Liu, X. Meng, T.D. Dao, L. Liu, P. Li, G. Zhao, T. Nagao, L. Yang, J. Ye, Light References assisted CO2 reduction with methane over SiO2 encapsulated Ni nanocatalysts for boosted activity and stability, J. Mater. Chem. A 5 (21) (2017) 10567–10573, https://doi.org/10.1039/c7ta00704c. [1] D.R. Lide, CRC Handbook of Chemistry and Physics, 88th ed., Taylor Fr., 2007 https://doi.org/10.1021/ja077011d. 11 V. Golovanova et al. Applied Catalysis B: Environmental 291 (2021) 120038 [26] X. Meng, T. Wang, L. Liu, S. Ouyang, P. Li, H. Hu, T. Kako, H. Iwai, A. Tanaka, [48] M.J. Hÿtch, E. Snoeck, R. Kilaas, Quantitative measurement of displacement and J. Ye, Photothermal Conversion of CO2 into CH4 with H2 over Group VIII strain fields from HREM micrographs, Ultramicroscopy 74 (3) (1998) 131–146, Nanocatalysts: An Alternative Approach for Solar Fuel Production, Angew. Chem. https://doi.org/10.1016/S0304-3991(98)00035-7. 126 (43) (2014) 11662–11666, https://doi.org/10.1002/ange.201404953. [49] S. Tada, T. Shimizu, H. Kameyama, T. Haneda, R. Kikuchi, Ni/CeO2 catalysts with [27] A.M. Watson, X. Zhang, R. Alcaraz De La Osa, J.M. Sanz, F. González, F. Moreno, high CO2 methanation activity and high CH4 selectivity at low temperatures, Int. J. G. Finkelstein, J. Liu, H.O. Everitt, Rhodium nanoparticles for ultraviolet Hydrogen Energy 37 (7) (2012) 5527–5531, https://doi.org/10.1016/j. plasmonics, Nano Lett. 15 (2) (2015) 1095–1100, https://doi.org/10.1021/ ijhydene.2011.12.122. nl5040623. [50] G. Zhou, H. Liu, K. Cui, A. Jia, G. Hu, Z. Jiao, Y. Liu, X. Zhang, Role of surface Ni [28] A. Picciotto, G. Pucker, L. Torrisi, P. Bellutti, F. Caridi, A. Bagolini, Evidence of and Ce species of Ni/CeO2 catalyst in CO2 methanation, Appl. Surf. Sci. 383 (2016) plasmon resonances of nickel particles deposited by pulsed laser ablation, Radiat. 248–252, https://doi.org/10.1016/j.apsusc.2016.04.180. Eff. Defects Solids 163 (4) (2008) 513–518, https://doi.org/10.1080/ [51] S. Soni, S. Kumar, B. Dalela, S. Kumar, P.A. Alvi, S. Dalela, Defects and oxygen 10420150701780656. vacancies tailored structural and optical properties in CeO2 nanoparticles doped [29] Z. Xiong, X. Chen, X. Wang, L. Peng, D. Yan, H. Lei, Y. Fu, J. Wu, Z. Li, X. An, with Sm3+ cation, J. Alloys Compd. 752 (2018) 520–531, https://doi.org/ W. Wu, Size dependence of plasmon absorption of Ni nanoparticles embedded in 10.1016/j.jallcom.2018.04.157. BaTiO3 /SrTiO3 superlattices, Appl. Surf. Sci. 268 (2013) 524–528, https://doi. [52] H. Liu, T.D. Dao, L. Liu, X. Meng, T. Nagao, J. Ye, Light assisted CO 2 reduction org/10.1016/j.apsusc.2012.12.161. with methane over group VIII metals: universality of metal localized surface [30] A. Alarcón, J. Guilera, J.A. Díaz, T. Andreu, Optimization of nickel and ceria plasmon resonance in reactant activation, Appl. Catal. B Environ. 209 (2017) catalyst content for synthetic natural gas production through CO2 methanation, 183–189, https://doi.org/10.1016/j.apcatb.2017.02.080. Fuel Process. Technol. 193 (2019) 114–122, https://doi.org/10.1016/j. [53] Z. Xin, L. Lu, B. Wang, X. Wang, K. Zhu, Z. Xu, Z. Yu, S. Yan, Z. Zou, Lewis acid fuproc.2019.05.008. activated CO2 reduction over a Ni modified Ni-Ge hydroxide driven by visible- [31] K. Wu, J. Chen, J.R. McBride, T. Lian, Efficient hot-electron transfer by a plasmon- infrared light, Dalton Trans. 48 (2019) 1672–1679, https://doi.org/10.1039/ induced interfacial charge-transfer transition, Science 349 (6248) (2015) 632–635, c8dt04408b. https://doi.org/10.1126/science.aac5443. [54] H. Liu, H. Song, X. Meng, L. Yang, J. Ye, Light irradiation enhanced CO2 reduction [32] S. Tan, A. Argondizzo, J. Ren, L. Liu, J. Zhao, H. Petek, Plasmonic coupling at a with methane: a case study in size-dependent optical property of Ni nanoparticles, metal/semiconductor interface, Nat. Photonics 11 (2017) 806–812, https://doi. Catal. Today 335 (2019) 187–192, https://doi.org/10.1016/j.cattod.2018.11.005. org/10.1038/s41566-017-0049-4. [55] G.E. Jonsson, V. Miljkovic, A. Dmitriev, Nanoplasmon-enabled macroscopic [33] L. Mascaretti, A. Naldoni, Hot electron and thermal effects in plasmonic thermal management, Sci. Rep. 4 (2014), 5111, https://doi.org/10.1038/ photocatalysis, J. Appl. Phys. 128 (2020) 041101, https://doi.org/10.1063/ srep05111. 5.0013945. [56] N.M. Martin, F. Hemmingsson, A. Schaefer, M. Ek, L.R. Merte, U. Hejral, [34] S. He, J. Huang, J.L. Goodsell, A. Angerhofer, W.D. Wei, Plasmonic nickel–TiO2 J. Gustafson, M. Skoglundh, A.C. Dippel, O. Gutowski, M. Bauer, P.A. Carlsson, heterostructures for visible-light-driven photochemical reactions, Angew. Chem. Structure-function relationship for CO2 methanation over ceria supported Rh and Int. Ed. 58 (18) (2019) 6038–6041, https://doi.org/10.1002/anie.201901987. Ni catalysts under atmospheric pressure conditions, Catal. Sci. Technol. 9 (2019) [35] X. Zhang, X. Li, D. Zhang, N.Q. Su, W. Yang, H.O. Everitt, J. Liu, Product selectivity 1644–1653, https://doi.org/10.1039/c8cy02097c. in plasmonic photocatalysis for carbon dioxide hydrogenation, Nat. Commun. 8 [57] F. Wang, S. He, H. Chen, B. Wang, L. Zheng, M. Wei, D.G. Evans, X. Duan, Active (2017), 14542, https://doi.org/10.1038/ncomms14542. site dependent reaction mechanism over Ru/CeO2 catalyst toward CO2 [36] C. Kim, S. Hyeon, J. Lee, W.D. Kim, D.C. Lee, J. Kim, H. Lee, Energy-efficient CO2 methanation, J. Am. Chem. Soc. 138 (19) (2016) 6298–6305, https://doi.org/ hydrogenation with fast response using photoexcitation of CO2 adsorbed on metal 10.1021/jacs.6b02762. catalysts, Nat. Commun. 9 (2018), 3027, https://doi.org/10.1038/s41467-018- [58] Y. Yang, C.A. Mims, R.S. Disselkamp, J.H. Kwak, C.H.F. Peden, C.T. Campbell, 05542-5. (Non)formation of methanol by direct hydrogenation of formate on copper [37] J.S. Pelli Cresi, M.C. Spadaro, S. D’Addato, S. Valeri, S. Benedetti, A. Di Bona, catalysts, J. Phys. Chem. C 114 (40) (2010) 17205–17211, https://doi.org/ D. Catone, L. Di Mario, P. O’Keeffe, A. Paladini, G. Bertoni, P. Luches, Highly 10.1021/jp104068k. efficient plasmon-mediated electron injection into cerium oxide from embedded [59] P.A.U. Aldana, F. Ocampo, K. Kobl, B. Louis, F. Thibault-Starzyk, M. Daturi, silver nanoparticles, Nanoscale 11 (21) (2019) 10282–10291, https://doi.org/ P. Bazin, S. Thomas, A.C. Roger, Catalytic CO2 valorization into CH4 on Ni-based 10.1039/c9nr01390c. ceria-zirconia. Reaction mechanism by operando IR spectroscopy, Catal. Today [38] J. Chen, Y. Zhou, R. Li, X. Wang, G.Z. Chen, Highly-dispersed nickel nanoparticles 215 (2013) 201–207, https://doi.org/10.1016/j.cattod.2013.02.019. decorated titanium dioxide nanotube array for enhanced solar light absorption, [60] S.M. Lee, Y.H. Lee, D.H. Moon, J.Y. Ahn, D.D. Nguyen, S.W. Chang, S.S. Kim, Appl. Surf. Sci. 464 (2019) 716–724, https://doi.org/10.1016/j. Reaction mechanism and catalytic impact of Ni/CeO2–x catalyst for low- apsusc.2018.09.091. temperature CO2 methanation, Ind. Eng. Chem. Res. 58 (20) (2019) 8656–8662, [39] Q. Zhang, M. Mao, Y. Li, Y. Yang, H. Huang, Z. Jiang, Q. Hu, S. Wu, X. Zhao, Novel https://doi.org/10.1021/acs.iecr.9b00983. photoactivation promoted light-driven CO2 reduction by CH4 on Ni/CeO2 [61] A. Cárdenas-Arenas, A. Quindimil, A. Davó-Quiñonero, E. Bailón-García, nanocomposite with high light-to-fuel efficiency and enhanced stability, Appl. D. Lozano-Castelló, U. De-La-Torre, B. Pereda-Ayo, J.A. González-Marcos, J. Catal. B Environ. 239 (30) (2018) 555–564, https://doi.org/10.1016/j. R. González-Velasco, A. Bueno-López, Isotopic and in situ DRIFTS study of the CO2 apcatb.2018.08.052. methanation mechanism using Ni/CeO2 and Ni/Al2O3 catalysts, Appl. Catal. B [40] J. Albero, H. Garcia, A. Corma, Temperature dependence of solar light assisted CO2 Environ. 265 (2020), 118538, https://doi.org/10.1016/j.apcatb.2019.118538. reduction on Ni based photocatalyst, Top. Catal. 59 (2016) 787–791, https://doi. [62] A. Westermann, B. Azambre, M.C. Bacariza, I. Graça, M.F. Ribeiro, J.M. Lopes, org/10.1007/s11244-016-0550-x. C. Henriques, The promoting effect of Ce in the CO2 methanation performances on [41] E. Rossinyol, J. Arbiol, F. Peiró, A. Cornet, J.R. Morante, B. Tian, T. Bo, D. Zhao, NiUSY zeolite: a FTIR in situ/operando study, Catal. Today 283 (2017) 74–81, Nanostructured metal oxides synthesized by hard template method for gas sensing https://doi.org/10.1016/j.cattod.2016.02.031. applications, Sens. Actuators B Chem. 109 (1) (2005) 57–63, https://doi.org/ [63] C.E. Verostko, R.K. Forsythe, A Study of the Sebatier-Methanation Reaction, 1974, 10.1016/j.snb.2005.03.016. https://doi.org/10.4271/740933. [42] L. Almar, T. Andreu, A. Morata, M. Torrell, L. Yedra, S. Estradé, F. Peiró, [64] Z. Cheng, C.S. Lo, Mechanistic and microkinetic analysis of CO2 hydrogenation on A. Tarancón, High-surface-area ordered mesoporous oxides for continuous ceria, Phys. Chem. Chem. Phys. 18 (2016) 7987–7996, https://doi.org/10.1039/ operation in high temperature energy applications, J. Mater. Chem. A 2 (2014) c5cp07469j. 3134–3141, https://doi.org/10.1039/c3ta13951d. [65] L. Pastor-Pérez, F. Baibars, E. Le Sache, H. Arellano-García, S. Gu, T.R. Reina, CO2 [43] B. Miao, S.S.K. Ma, X. Wang, H. Su, S.H. Chan, Catalysis mechanisms of CO2 and valorisation via Reverse Water-Gas Shift reaction using advanced Cs doped Fe-Cu/ CO methanation, Catal. Sci. Technol. 6 (2016) 4048–4058, https://doi.org/ Al2O3 catalysts, J. CO2 Util. 21 (2017) 423–428, https://doi.org/10.1016/j. 10.1039/c6cy00478d. jcou.2017.08.009. [44] V. La Parola, G. Pantaleo, A. Venezia, Effects of synthesis on the structural [66] S. Abelló, C. Berrueco, D. Montané, High-loaded nickel-alumina catalyst for direct properties and methane partial oxidation activity of Ni/CeO2 catalyst, Catalysts 8 CO2 hydrogenation into synthetic natural gas (SNG), Fuel 113 (2013) 598–609, (5) (2018) 220, https://doi.org/10.3390/catal8050220. https://doi.org/10.1016/j.fuel.2013.06.012. [45] S. Xu, X. Wang, Highly active and coking resistant Ni/CeO2-ZrO2 catalyst for [67] B. Choudhury, P. Chetri, A. Choudhury, Oxygen defects and formation of Ce3+ partial oxidation of methane, Fuel 84 (5) (2005) 563–567, https://doi.org/ affecting the photocatalytic performance of CeO2 nanoparticles, RSC Adv. 4 (2014) 10.1016/j.fuel.2004.10.008. 4663–4671, https://doi.org/10.1039/c3ra44603d. [46] D. Chen, H. Hao, J. Yu, J. Liu, J. Lu, F. Liu, G. Wan, S. He, Y. Luo, Structural/ [68] C. Yang, J. Yang, X. Duan, G. Hu, Q. Liu, S. Ren, J. Li, M. Kong, Roles of photo- surface characterization and catalytic evaluation of rare-earth (Y, Sm and La) generated holes and oxygen vacancies in enhancing photocatalytic performance doped ceria composite oxides for CH3SH catalytic decomposition, Appl. Surf. Sci. over CeO2 prepared by molten salt method, Adv. Powder Technol. 31 (9) (2020) 390 (2016) 959–967, https://doi.org/10.1016/j.apsusc.2016.08.129. 4072–4081, https://doi.org/10.1016/j.apt.2020.08.017. [47] R. Murugan, G. Vijayaprasath, T. Mahalingam, G. Ravi, Enhancement of room [69] K.R. Hahn, M. Iannuzzi, A.P. Seitsonen, J. Hutter, Coverage effect of the CO2 temperature ferromagnetic behavior of rf sputtered Ni-CeO2 thin films, Appl. Surf. adsorption mechanisms on CeO2(111) by first principles analysis, J. Phys. Chem. C Sci. 390 (2016) 583–590, https://doi.org/10.1016/j.apsusc.2016.08.166. 117 (4) (2013) 1701–1711, https://doi.org/10.1021/jp309565u. 12