CO₂ Driven Soot Oxidation via Reverse Boudouard Reaction over Reducible Catalyst: A Sustainable Pathway for Catalyst Regeneration

Abstract

Here, we demonstrate how CO₂ as a mild oxidant can drive soot gasification via the reverse Boudouard reaction, simultaneously regenerating deactivated catalysts and valorizing a greenhouse gas into CO  a platform chemical of industrial significance. Central to this strategy is the ability of CeO₂ to undergo reduction and regeneration: its lattice oxygen oxidizes soot, while the oxygen vacancies formed during this process help activate CO₂ and restore the catalyst. Pristine CeO₂ and aliovalently doped Ce_0.9Ni_0.1O_2−δ were synthesized via a rapid, single-step solution combustion route and benchmarked for CO₂-assisted soot oxidation in a fixed-bed reactor. A suite of techniques establishes that Ni²⁺ incorporation into the CeO₂ fluorite lattice amplifies oxygen vacancy density, enhances lattice oxygen mobility, and markedly improves reducibility. Under inert N₂ atmosphere, where lattice oxygen acts as the sole oxidant, the superior reducibility of Ce_0.9Ni_0.1O_2−δ translates directly into a 75% higher soot conversion compared to pristine CeO₂ at 800 °C, producing CO as the dominant product. Post-reaction XPS confirms a marked drop in Ce⁴⁺ content and a surge in oxygen vacancy concentration, providing unambiguous spectroscopic evidence for lattice oxygen consumption via the Mars–van Krevelen mechanism. When CO₂ is introduced as oxidant, a remarkable inversion occurs: both catalysts now achieve complete soot oxidation with near-identical light–off profiles, as gas-phase CO₂ saturation overrides differences in surface basicity and CO₂ adsorption capacity. Crucially, post-reaction XPS under CO₂ atmosphere reveals a decrease in oxygen vacancy concentration  the fingerprint of vacancy replenishment through CO₂ dissociation and catalyst self-regeneration. This work opens a tangible pathway for deploying CeO₂-based materials as catalytic additives for CO₂-driven regeneration in fluid catalytic cracking and related refinery processes, contributing to both carbon utilization and long-term process sustainability.

1. Introduction

The growing demand for energy and capital goods required for modern societal development largely relies on non-renewable resources such as crude oil, coal, and natural gas. In industrial chemical processes, more than 80% of reactions are facilitated by catalysts to enhance reaction rates and achieve selective product formation. However, catalyst deactivation owing to coke or soot deposition over the catalyst surface during operation is unavoidable, leading to a gradual decline in catalytic activity and selectivity. The accumulation of such carbonaceous deposits progressively blocks active sites, reduces accessible surface area, and ultimately necessitates process shutdown for regeneration. Therefore, efficient catalyst regeneration strategies are indispensable to sustain catalytic performance and reduce operational costs. Traditionally, regeneration is carried out by combusting deposited carbon (coke) using air or oxygen. While effective, this air-based approach generates additional CO₂ emissions and risks sintering of thermally sensitive catalysts. The use of carbon dioxide (CO₂) as a mild and sustainable regenerating agent has therefore attracted significant interest. − In processes such as fluid catalytic cracking or biomass gasification, CO₂-assisted regeneration can convert deposited coke into carbon monoxide (CO), a valuable feedstock for the synthesis of hydrogen, methanol, dimethyl ether, and hydrocarbons via the Fischer–Tropsch route. This dual benefit  simultaneous catalyst regeneration and CO₂ utilization  makes CO₂-driven soot oxidation a particularly attractive strategy from a carbon capture and utilization perspective. Global CO₂ emissions reached approximately 37.8 billion tonnes in 2025, mainly from fossil fuel combustion. Despite this staggering emission level, less than 0.1% of anthropogenic CO₂ has only been recycled or utilized. The potential to use CO₂ as a reactant in refinery processes for simultaneous catalyst regeneration and chemical valorization could significantly reduce greenhouse gas emissions and improve overall process sustainability.

A critical but underappreciated dimension of soot oxidation over reducible oxide catalysts is the direct participation of lattice oxygen as an oxidizing species. In reducible metal oxides such as CeO₂, the facile Ce⁴⁺/Ce³⁺ redox couple enables the release of reactive lattice oxygen to oxidize carbonaceous deposits via a Mars–van Krevelen (MvK) mechanism, in which the reduced catalyst surface is subsequently re-oxidized by the surrounding gas-phase oxidant. When CO₂ serves as the external oxidant, the oxygen vacancies generated upon lattice oxygen consumption act as active sites for CO₂ adsorption and dissociation, replenishing the lattice oxygen and regenerating the catalyst surface. This dynamic vacancy–replenishment cycle is central to the sustainability of CO₂-driven soot oxidation and underscores the importance of a catalyst’s intrinsic reducibility  the ease with which lattice oxygen can be mobilized and subsequently restored  as a key descriptor of its regenerative performance.

However, regenerating a catalyst using CO₂ instead of air would require a large amount of energy, due to the highly endothermic nature of the Reverse-Boudouard (RB) reaction. ,

$$\mathrm{C}(\mathrm{s})+\mathrm{C}{\mathrm{O}}_2(\mathrm{g})\to 2\mathrm{C}\mathrm{O}(\mathrm{g}),\mathrm{\Delta }\mathrm{H}=172\mathrm{k}\mathrm{J}/\mathrm{m}\mathrm{o}\mathrm{l}$$ 1

As indicated by the large positive enthalpy, the RB reaction is highly endothermic, and the equilibrium strongly favors the production of CO₂ over CO. , At temperatures exceeding 700 °C, the large positive entropic term (TΔS) dominates over the enthalpic contribution, making the Gibbs free energy (ΔG = ΔH–TΔS) negative. Recently, regenerating coked catalysts through CO₂ gasification has been investigated using group I and II elements along with vanadium supported on alumina. , The primary objective is therefore to develop a catalyst that can harness its intrinsic lattice oxygen to oxidize deposited coke at reduced temperatures, while enabling CO₂ to replenish spent lattice oxygen and sustain catalytic activity. Herein, reducible rare earth oxide CeO₂ is introduced as the model reducible catalyst. CeO₂ has been extensively studied owing to its exceptional redox ability, high oxygen storage capacity, structural stability under high-temperature conditions, and the unique ability of Ce to toggle between Ce⁴⁺ and Ce³⁺ oxidation states with concomitant generation and annihilation of oxygen vacancies. Crucially, it is this intrinsic reducibility  the capacity to supply lattice oxygen from the bulk and surface to oxidize carbonaceous deposits, and to then recover those oxygen vacancies through CO₂ dissociation  that makes CeO₂ uniquely suited for CO₂-driven catalyst regeneration and forms the conceptual foundation of the present study. CeO₂ adopts a cubic fluorite structure (space group Fm3̅m) providing a robust yet flexible framework that readily accommodates oxygen vacancy formation. The abundance of these vacancies, coupled with the high mobility of lattice oxygen within the fluorite lattice, imparts CeO₂ with its characteristic reducibility and high catalytic activity, as documented across three-way catalysis, water–gas shift, dry reforming, and soot oxidation applications. , Ni²⁺ doping introduces aliovalent substitution at the Ce Wyckoff sites, which not only generates additional oxygen vacancies through charge compensation but also modifies the local electronic environment, weakening metal–oxygen bond strength and further enhancing the lability of lattice oxygen. Moreover, Ni is well-known for its affinity toward CO₂ adsorption, and the synergistic combination of enhanced oxygen vacancy density, improved lattice oxygen mobility, and CO₂ adsorption capacity in Ni-doped CeO₂ is expected to augment soot oxidation performance. However, the potential of Ni-doped CeO₂ for CO₂-driven catalyst regeneration has not yet been systematically explored, which motivates the present study.

In the present work, the RB reaction is investigated over CeO₂ and its Ni doped variants, which are synthesized via a one-step solution combustion method, and evaluated using an in-house experimental setup. The experiments are conducted under RB conditions (10% CO₂ in N₂) in both dynamic and static conditions. The coke removal was further explored with the help of lattice oxygen from CeO₂.

$$\mathrm{C}(\mathrm{s})+\mathrm{O}(\mathrm{L}\mathrm{a}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{e})\to \mathrm{C}\mathrm{O}(\mathrm{g})$$ 2

A range of analytical techniques have been used to systematically characterize the physical, chemical, morphological, and surface electronic properties of the materials. Correlating the analytical data obtained before and after the reaction, the soot oxidation mechanism was interpreted.

2. Experimental Section

2.1. Catalyst Preparation

Pristine CeO₂ and Ni-doped Ce_0.9Ni_0.1O_2−δ were synthesized via a low-temperature-initiated, single-step solution combustion synthesis. , This self-propagating technique is rooted in propellant chemistry, wherein highly exothermic metathetical redox reactions occur between stoichiometrically mixed metal nitrate oxidizers and an organic fuel. Upon reaching the ignition temperature, the redox reaction propagates autonomously, releasing substantial energy that drives the rapid formation of nanocrystalline metal oxide products in a single step, eliminating the need for prolonged calcination. For the synthesis purpose, the precursors were water-soluble (NH₄)₂[Ce­(NO₃)₆] (Sigma-Aldrich, 99%), Ni­(NO₃)₂·6H₂O (SD Fine Chem Limited, India, 99%) as oxidizers, while oxalyldihydrazide (ODH) served as the fuel. For pristine CeO₂, 548.22 mg of (NH₄)₂[Ce­(NO₃)₆] and 283.41 mg of ODH were dissolved in 40 mL of deionized water. For the synthesis of doped Ce_0.9Ni_0.1O_2−δ, 493.39 mg of (NH₄)₂[Ce­(NO₃)₆], 29.07 mg of Ni­(NO₃)₂·6H₂O, and 401.51 mg of ODH were similarly dissolved in 40 mL of deionized water. The resulting homogeneous aqueous solutions were placed in a preheated furnace at 450 °C for 15 min. Upon heating, the precursor solutions underwent vigorous boiling and foaming, followed by complete dehydration of the reactive intermediates. At the ignition point, the exothermic redox reaction self-initiated, generating a transient flame with temperatures approaching 1000 °C. The rapid combustion propagated across the dehydrated surface, yielding a voluminous, porous solid product within minutes. The as-synthesized powders were subsequently collected and ground into a fine powder using a mortar and pestle. The overall stoichiometric chemical reaction governing the synthesis can be represented as follows

$${{({\mathrm{N}\mathrm{H}}_4)}_2\mathrm{C}\mathrm{e}({\mathrm{N}\mathrm{O}}_3)}_6+2{\mathrm{C}}_2{\mathrm{H}}_6{\mathrm{N}}_4{\mathrm{O}}_2\to \mathrm{C}\mathrm{e}{\mathrm{O}}_2+4\mathrm{C}{\mathrm{O}}_2+10{\mathrm{H}}_2\mathrm{O}+8{\mathrm{N}}_2+{\mathrm{O}}_2$$ 3

$$0.9{{({\mathrm{N}\mathrm{H}}_4)}_2\mathrm{C}\mathrm{e}({\mathrm{N}\mathrm{O}}_3)}_6+0.1{\mathrm{N}\mathrm{i}(\mathrm{N}{\mathrm{O}}_3)}_2.6{\mathrm{H}}_2\mathrm{O}+1.84{\mathrm{C}}_2{\mathrm{H}}_6{\mathrm{N}}_4{\mathrm{O}}_2\to {\mathrm{C}\mathrm{e}}_{0.9}{\mathrm{N}\mathrm{i}}_{0.1}{\mathrm{O}}_2+3.68\mathrm{C}{\mathrm{O}}_2+9.72{\mathrm{H}}_2\mathrm{O}+7.38{\mathrm{N}}_2+{\mathrm{O}}_2$$ 4

2.2. Catalyst Characterizations

Powder X-ray diffraction (XRD) measurements were performed on a Rigaku Ultima IV diffractometer using Cu Kα radiation (λ = 0.154 nm). Data were collected over the 2θ range of 20°–90° with a step size of 0.01° and a scan speed of 0.7° min^–1. The crystallite size (D) was estimated using the Scherrer equation, D = $\frac{0.9\lambda }{\beta \cos \theta }$ where β represents the full width half maximum (FWHM), λ is the X-ray wavelength, and θ is the diffraction angle. The precise chemical composition in the synthesized compounds was determined using energy-dispersive X-ray fluorescence (ED-XRF) analysis, performed with an Epsilon1 instrument from PAN Analytical instrument. A Microtrac BEL Corp mini-II surface area analyzer was used to examine the surface characteristics of the synthesized materials. The samples were degassed under vacuum at 200 °C for 2 h before performing N₂ sorption measurements at 77 K. The specific surface area was calculated using the BET method. The surface morphology and composition analyses were conducted using a Field Emission Scanning Electron Microscope (FE-SEM, FEI-Apreo S) equipped with an Energy Dispersive X-ray Spectroscopy (EDX) unit. The instrument was operated at an acceleration voltage of 20 kV. Before imaging, the samples were coated with a thin layer of silver (sputter coating) to enhance their electrical conductivity. The chemical surface properties of the synthesized catalysts were obtained using X-ray photoelectron spectroscopy (XPS) with a Thermo Scientific K-Alpha surface analysis spectrometer, employing Al Kα radiation (1486.6 eV). The data profiles were deconvoluted using a Gaussian–Lorentzian function under the Powell fitting algorithm with the help of Avantage software. The binding energy (B.E.) of all XPS data was calibrated with respect to the standard C1s peak at 284.85 eV. The relative surface concentrations of Ce³⁺ and Ce⁴⁺ were determined from the integrated areas of the deconvoluted peaks (A) using the following equations

$${\mathrm{A}}_{{\mathrm{C}\mathrm{e}}^{3+}}=\mathrm{A}{\mathrm{v}}^0+\mathrm{A}{\mathrm{u}}^0+\mathrm{A}{\mathrm{v}}^{\prime }+\mathrm{A}{\mathrm{v}}^{″}$$ 5

$${\mathrm{A}}_{{\mathrm{C}\mathrm{e}}^{4+}}=\mathrm{A}\mathrm{v}+\mathrm{A}\mathrm{u}+\mathrm{A}{\mathrm{v}}^{″}+\mathrm{A}{\mathrm{u}}^{″}+\mathrm{A}{\mathrm{v}}^{‴}+\mathrm{A}{\mathrm{u}}^{‴}$$ 6

$${\mathrm{C}}_{{\mathrm{C}\mathrm{e}}^{4+}}=\frac{\mathrm{A}{\mathrm{C}\mathrm{e}}^{4+}}{\mathrm{A}{\mathrm{C}\mathrm{e}}^{3+}+\mathrm{A}{\mathrm{C}\mathrm{e}}^{4+}}\times 100$$ 7

where v and u are respective deconvoluted peaks. The surface oxygen vacancy concentration (%) for each catalyst was determined from the ratio of the O_Vacancy (O_V) peak area to the total O1s core-level spectrum area (O_Total), using the following equation.

$${\mathrm{O}}_{\mathrm{V}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y}}(\%)=100\times \frac{{\mathrm{O}}_{\mathrm{V}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y}}}{{\mathrm{O}}_{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}},\mathrm{w}\mathrm{h}\mathrm{e}\mathrm{r}\mathrm{e}{\mathrm{O}}_{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}={\mathrm{O}}_{\mathrm{L}\mathrm{a}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{e}}+{\mathrm{O}}_{\mathrm{S}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e}}+{\mathrm{O}}_{\mathrm{V}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{y}}$$ 8

Raman spectra were collected using a UniRam Raman spectrometer at a wavelength of 580.77 nm with 20 accumulations. The reducibility of the catalysts was determined using Temperature-Programmed Reduction (TPR) with H₂, performed on an Altamira Instrument (AMI) connected to a Thermal Conductivity Detector (TCD), using 30 mg of the sample. In the typical analysis, the sample was pretreated for 30 min at 150 °C with inert Ar. After cooling to 45 °C, 10% H₂ in Ar was passed over the catalysts with a total flow rate of 50 mL min^–1. The gas flow then continued with a ramping rate of 10 °C min^–1 from 45 °C to 900 °C. The amount of H₂ uptake was calculated using multiple pulse calibration. The O₂ temperature-programmed desorption (O₂-TPD) experiments were done using the same instrument with 100 mg of catalyst, placed in a U-shaped quartz reactor and pretreated in He flow with 30 mL min^–1 by heating to 150 °C at 10 °C min^–1 for 30 min, and cooling to 45 °C. The samples were then exposed to O₂ with a flow rate of 30 mL min^–1 for 1 h at 45 °C, followed by He purging for 30 min to remove excess O₂. Subsequently, the temperature was raised to 900 °C at 10 °C min^–1 under He flow with 30 mL min^–1 while recording TCD signals, and finally cooled under He flow. The amount of desorbed O₂ molecules was calculated using the same procedure as followed for H₂-TPR. The calculations enable one indirect method for calculating surface oxygen vacancy density (ρ _vac ). CO₂-TPD was conducted using the same setup. In this procedure, 100 mg of catalyst was first pretreated in He flow at 150 °C for 30 min, then cooled to room temperature, followed by the sample being exposed to 10% CO₂/He mixture for 1 h, flowing at 50 mL min^–1. After stopping the CO₂ supply, the desorption profile was recorded using a TCD detector, while the temperature was increased from 45 °C to 900 °C with a ramping rate of 10 °C min^–1 under He flow of 30 mL min^–1. Desorption of CO₂ was quantified by performing pulse calibrations using the same method.

2.3. Catalytic Activity Evaluation

The catalytic activity of each sample for the CO₂-assisted soot oxidation was carried out using a fixed-bed quartz tube reactor with a 4 mm diameter and 1 cm length, and a total of 200 mg admixture of 160 mg of catalyst and 40 mg of soot (Synco Industries Limited, India). The catalytic CO₂-assisted soot oxidation proceeds mainly as a heterogeneous reaction occurring at a three-phase interface between the solid catalyst, solid soot particles, and the CO₂. The catalytic efficiency largely depends on the degree of contact between the catalyst surface and soot particles. Therefore, to ensure intimate contact, the soot and catalyst were thoroughly homogenized together in an agate mortar for 30 min, maintaining a weight ratio of 1:4 (soot: catalyst). Before loading, the admixtures were transformed into pellets, passed through a 40–70 mesh sieve to ensure uniform particle size, and placed inside a quartz tube. The assembled reactor was inserted into a tubular furnace where thermal conditions were monitored via a K-type thermocouple interfaced with a temperature control unit. The temperature was gradually increased from room temperature to 800 °C at a rate of 10 °C min^–1, with a total gas flow of 100 mL min^–1, composed of 10% CO₂ and balanced with pure (99.999%) N₂ with a Gas Hourly Space Velocity (GHSV) of 47,770 h^–1. High-purity gases (CHEMIX, Bangalore) were used, controlled by a combination of pressure regulators, needle valves, and a gas mixing manifold. The flow rates were monitored using an in-house bubble-type flow meter. At every 50 °C interval, the gaseous products were taken from the reactor outlet using a Hamilton syringe (capacity: 500 μL) and introduced manually into a 5765 Nucon Gas Chromatograph (India) configured with a Flame Ionization Detector (FID) passed through a Porapak Q column (80–100), coupled with a methanizer unit, − for the analysis of the evolved species. Argon gas (99.99% pure) was used to carry the sample through the system. Each gas was checked using a standard mixture, with different concentration levels of CO and CO₂. The conversion factors were calculated from the slope of the graph showing peak area versus concentration. The conversion of soot (%) was calculated according to (9) in case of N₂

$$\mathrm{S}\mathrm{o}\mathrm{o}\mathrm{t}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}(\%)=\frac{[\mathrm{C}\mathrm{O}]+[{\mathrm{C}\mathrm{O}}_2]}{{[\mathrm{C}]}_{initial}}\times 100$$ 9

The conversion of soot (%) was calculated according to (10) in case of CO₂

$$\mathrm{S}\mathrm{o}\mathrm{o}\mathrm{t}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}(\%)=\frac{{\int }_0^t[\mathrm{C}\mathrm{O}+{\mathrm{C}\mathrm{O}}_2]\mathrm{d}\mathrm{t}}{{\int }_0^{total}[\mathrm{C}\mathrm{O}+{\mathrm{C}\mathrm{O}}_2]\mathrm{d}\mathrm{t}}\times 100$$ 10

The calculation of CO selectivity (%) was calculated using the following expression

$$\mathrm{C}\mathrm{O}\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{i}\mathrm{t}\mathrm{y}(\%)=\frac{[\mathrm{C}\mathrm{O}]}{[\mathrm{C}\mathrm{O}]+[{\mathrm{C}\mathrm{O}}_2]}\times 100$$ 11

The surface vacancy density, ρ _vac (in sites m^–2), was calculated using the following equation

$${\rho }_{vac}=\frac{{{\mathrm{N}}_{\mathrm{A}}\mathrm{n}}_{\mathrm{v}\mathrm{a}\mathrm{c}}}{{\mathrm{S}\mathrm{A}}_{\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{t}}{\mathrm{m}}_{\mathrm{T}\mathrm{P}\mathrm{D}}}$$ 12

Where: N_A = Avogadro’s number, n_vac = The moles of O₂ desorbed from the vacancies, SA_catalyst = Surface area of the catalyst determined from BET analysis, m_TPD = mass of sample used for O₂-TPD analysis.

Figure S1 depicts a schematic outline of the experimental setup. The integrated design ensured stable and accurate control of reaction parameters, including temperature, gas composition, and flow rates. The catalytic soot oxidation under N₂ atmosphere was also done in a similar manner. However, instead of gaseous CO₂, ultra-pure N₂ was passed with a total flow of 100 mL min^–1.

3. Results and Discussion

An energy dispersive XRF spectrometer was used to determine the chemical composition of the synthesized catalysts. As shown in Table S1, the experimentally determined atomic percentages are in good agreement with the theoretical (nominal) values, confirming the successful composition at the targeted stoichiometry. XRD analysis was performed to investigate the crystal structures of the synthesized materials. As shown in Figure a, the diffraction pattern of pristine CeO₂ exhibits four characteristic peaks at 2θ = 28.6°, 33.2°, 47.4°, and 56.7°, which are attributed to the (111), (200), (220), and (311) crystal planes, respectively consistent with the formation of a crystalline cubic fluorite phase belonging to the space group Fm3̅m (JCPDS 34-0394). , The XRD pattern of Ce_0.9Ni_0.1O_2−δ (Figure a) confirms that the fluorite structure is well retained upon Ni incorporation. No additional diffraction peaks corresponding to metallic Ni or NiO phases were detected. However, a careful comparison of the diffraction patterns reveals that the peaks of Ce_0.9Ni_0.1O_2−δ are systematically shifted to higher 2θ values relative to those of pristine CeO₂ (Figure b). According to Shannon’s ionic radii, six-coordinated Ni²⁺ has an ionic radius of 0.69 Å, which is considerably smaller than that of six-coordinated Ce⁴⁺ (0.87 Å) in CeO₂. Based on Bragg’s law, substitution of the larger Ce⁴⁺ by the smaller Ni²⁺ would be expected to contract the unit cell and reduce the d₁₁₁ spacing, thereby shifting the diffraction peaks to higher 2θ angles. , This observation is indicative of the lattice substitution of Ni at the Ce Wyckoff position (4a), resulting in the formation of a phase-pure solid solution. The pristine CeO₂ sample exhibited a crystallite size of 20 nm, whereas the Ni-doped Ce_0.9Ni_0.1O_2−δ showed an increased crystallite size of 44 nm, as determined from the Scherrer eq (Table S1).

1.

(a) Powder X-ray diffraction patterns of the synthesized materials, (b) enlarged XRD patterns, (c) Rietveld refinement, and (d) Raman spectra of the as-prepared materials.

To investigate the crystal structure in greater detail, Rietveld refinement was carried out on the diffraction patterns of the synthesized materials using FullProf Suite. The diffraction data were refined by assigning Ni to the Ce Wyckoff 4a sites and oxygen to the 8c sites. The refinement was performed by optimizing the scale factor, zero shift, background function, lattice parameters, and peak shape using a pseudo-Voigt profile function. After achieving a stable fit, occupancy parameters and atomic positions were also refined. The corresponding refinement profiles are presented in Figure c. The refined structural parameters, including unit cell length, cell volume, oxygen occupancy, and fitting reliability factors such as R_Bragg and χ², are listed in the Table and confirm the goodness of fit. In the Figure c, the black line represents the experimentally obtained diffraction data, while the superimposed red line corresponds to the fitted profile based on the calculated intensities. The blue line at the bottom indicates the difference between the experimental and calculated patterns. The cubic unit cell parameter of pristine CeO₂ (a = 5.411 Å) decreased slightly upon Ni doping to 5.409 Å in Ce_0.9Ni_0.1O_2−δ. This reduction in unit cell parameter, along with the corresponding shrinkage in cell volume, confirms the formation of a solid solution through the successful incorporation of Ni into the cubic fluorite lattice. Although oxygen has a lower X-ray scattering factor than cerium, the refined oxygen occupancy obtained from the Rietveld refinement can still be used to examine trends in oxygen content variation. Aliovalent substitution of divalent Ni²⁺ for Ce⁴⁺ is expected to reduce the oxygen content by generating oxygen vacancies. Accordingly, a decrease in oxygen content from 2.00 in pure CeO₂ to 1.97 in Ce_0.9Ni_0.1O_2−δ is apparently observed in the Table . The Raman spectra of pristine CeO₂ and Ce_0.9Ni_0.1O_2−δ both exhibit a strong band at 465–470 cm^–1, assigned to the F_2g vibrational mode of the cubic fluorite structure, which arises from the symmetric stretching vibration of Ce–O bonds (Figure d). , The decrease in intensity of this F_2g band reflects a weakening of the symmetric Ce–O stretching mode and indicates the presence of structural defects, such as oxygen vacancies, in the lattice. In addition, a weak band appears at approximately 600 cm^–1, attributed to a defect-induced (A_1g-type) mode associated with oxygen vacancies. The intensity of this band increases upon Ni doping, providing further evidence for the formation of oxygen vacancies in Ce_0.9Ni_0.1O_2−δ.

1. Structural Parameters of the Synthesized Catalysts Derived From Rietveld Refinement.

catalysts	cell parameters (Å)	cell volume (Å3)	χ2	RBragg (%)	Wycoff site			oxygen occupancy
					Ce	Ni	O
CeO2	5.4110	158.43	2.1	6.1	4a	–	8c	2
Ce0.9Ni0.1O2−δ	5.4099	158.33	4.7	4.8	4a	4a	8c	1.97

The surface area and pore size distribution of pristine CeO₂ and doped Ce_0.9Ni_0.1O_2‑δ were investigated using nitrogen adsorption–desorption (BET) analysis. According to the IUPAC classification, the adsorption isotherms presented in Figure S2a for both samples exhibit a Type IV profile with H4 hysteresis loops, indicating a mesoporous nature. Pristine CeO₂ exhibited a higher specific surface area (33.4 m² g^–1) than Ce_0.9Ni_0.1O_2−δ (18.6 m² g^–1). The corresponding pore size distributions obtained from Figure S2b are summarized in Table S2; BJH analysis yielded an average pore diameter of approximately 10 nm for both the samples, consistent with a mesoporous structure. The morphology of the catalysts was examined by FE-SEM, which revealed that both CeO₂ and Ce_0.9Ni_0.1O_2−δ consist of highly agglomerated particles with irregular grain shapes (Figure S3). EDX analysis was also performed in conjunction with FE-SEM to verify the elemental composition (Figure S3), and the results were consistent with those obtained from ED-XRF (Table S1). Elemental mapping (Figure S3) further confirmed the uniform distribution of Ni throughout the sample.

Ceria exhibits unique redox behavior due to the ability of cerium to switch between Ce⁴⁺ and Ce³⁺ oxidation states, accompanied by formation of oxygen vacancies vis-à-vis characteristic oxygen storage capacity. H₂-TPR studies were conducted to probe the redox chemistry and oxygen vacancy formation of pristine CeO₂ and doped Ce_0.9Ni_0.1O_2−δ. In a typical H₂-TPR experiment, low-temperature peaks generally indicate the reduction of weakly bound surface oxygen species, while peaks appearing at intermediate temperatures are associated with subsurface oxygen. High-temperature peaks, in contrast, arise from the extraction of strongly bonded lattice oxygen, revealing information about the catalyst’s intrinsic oxygen mobility and the strength of its metal–oxygen framework. As shown in Figure a, the H₂-TPR profile of pristine CeO₂ exhibits two distinct reduction peaks: the low-temperature peak (v), appearing below 600 °C, corresponds to the removal of surface or “capping” oxygen species, while the high-temperature peak (w), centered around 800 °C, is attributed to the reduction of bulk lattice oxygen. , Upon Ni incorporation into the CeO₂ lattice, the H₂-TPR profile shows a marked shift toward lower reduction temperatures (below 500 ^oC), with weak peaks appearing even below 200 °C, reflecting the more facile reduction of adsorbed surface oxygen species in Ce_0.9Ni_0.1O_2−δ. The mid-temperature range peaks (u and u₁) may represent the reduction of the different ionic Ni species in the lattice, whereas the high temperature peak at around 800 °C corresponds to the reduction of bulk lattice oxygen in Ce_0.9Ni_0.1O_2−δ similar to the pristine material. ,− As shown by the total H₂ consumption values in Table S3, Ni doping markedly increases the reducible oxygen content (and thus the oxygen storage capacity) of Ce_0.9Ni_0.1O_2−δ compared to pure CeO₂, which may improve its soot oxidation activity. Soot oxidation on catalysts occurs primarily through surface-adsorbed oxygen and lattice oxygen. To assess the lability of lattice oxygen in the synthesized catalyst, an O₂-TPD analysis was performed. The O₂-TPD profiles (Figure b) comparing CeO₂ and Ce_0.9Ni_0.1O_2−δ show that both contain a low-temperature peak (O_α at 116–118 °C) from weakly adsorbed surface oxygen. The Peaks around 200–400 °C illustrate oxygen molecules are chemically adsorbed on oxygen vacancies (O_β). , Ni doping shifts the vacancy adsorbed oxygen species peaks at lower temperature of 271 °C from 316 °C, which is accompanied by pristine ceria. A prominent peak near 500 °C in Ce_0.9Ni_0.1O_2−δ reflects enhanced lattice oxygen activity (O_γ), occurring at a much lower temperature than the broad peak near 793 °C observed for pristine CeO₂, indicating significantly improved lattice oxygen mobility upon Ni doping. The calculated total amount of desorbed O₂ from the O₂-TPD profiles shown in Table S3 apparently highlights the improved lattice oxygen availability in doped catalyst. However, we also calculated the amount of oxygen released from the vacancies, which corresponds to the number of surface vacancy sites present in each sample. Figure c demonstrates that substitution of Ce with Ni significantly enhances the surface oxygen vacancy concentration (ρ _vac ), a parameter that is directly correlated with the catalytic activity of the material. Since this study also focuses on coke removal in the presence of CO₂ (RB reaction), CO₂-TPD was carried out to analyze the surface basicity, adsorption strength, and associated surface properties of the prepared materials. From Figure d, it is evident that Ce_0.9Ni_0.1O_2−δ generates additional moderate (200–500 °C) and strong (>500 °C) basic sites that are absent in pristine CeO₂. − The high-temperature CO₂ desorption profile for Ce_0.9Ni_0.1O_2−δ is related to the decomposition of stable carbonate species formed at the strong basic sites generated by oxygen-vacancy-rich Ce–O–Ni interfacial sites. The introduction of Ni promotes the formation of oxygen vacancies and strengthens the Lewis basic character, resulting in stronger CO₂ adsorption and greater stability of the surface carbonates. Consequently, these species require higher temperature to decompose, leading to their desorption at high temperatures. The total CO₂ uptake on the basic sites over the two catalysts are provided in Table S3. The increase in surface basicity observed for the Ce_0.9Ni_0.1O₂ sample can promote a higher availability of reactive oxygen species during CO₂-assisted reactions, which can play a key role in enhancing soot oxidation in our system.

2.

(a) H₂-TPR profile, (b) O₂-TPD profile, (c) surface oxygen vacancy density plot, and (d) CO₂-TPD profile of the as-prepared catalysts.

We evaluated soot oxidation under two different reaction environments. First, we focused on utilizing lattice oxygen as the active oxidizing species. These experiments were carried out under dynamic reaction conditions, in which soot oxidation was monitored with the ramping of temperature from room temperature to 800 °C at 10 °C min^–1 under a flow of 100 mL min^–1 ultrapure N₂. As the temperature increased, the outlet gas composition was monitored by GC, revealing two reaction products, predominantly CO with a minor amount of CO₂. The corresponding soot conversion (%) derived from eq is plotted in Figure a, while the concentrations of the reaction products over the two catalysts are presented in Figure b. Soot conversion increased progressively with temperature, reaching a maximum at 800 °C, and was consistently higher over Ce_0.9Ni_0.1O_2−δ than pristine CeO₂. A maximum soot conversion of 35% was achieved over Ce_0.9Ni_0.1O_2−δ at 800 °C, whereas pristine CeO₂ exhibited only 20% soot conversion under the same conditions. Figure b shows that CO is the dominant product upon soot oxidation, with CO₂ formed only in smaller amounts; the corresponding CO selectivity is plotted in Figure c. This predominance of CO over CO₂ is consistent with partial oxidation of soot, likely arising from the limited availability of lattice oxygen under these conditions. Notably, Ce_0.9Ni_0.1O_2−δ produced markedly higher amounts of both CO and CO₂ than pristine CeO₂, reflecting a faster reaction rate and superior catalytic efficacy. At 800 °C, the CO yield over Ce_0.9Ni_0.1O_2−δ was ∼7 mmol g^–1, whereas pristine CeO₂ exhibited a lower CO yield less than 4 mmol g^–1 at the same temperature. This enhanced performance is consistent with the H₂-TPR and O₂-TPD results, which indicated a higher reducibility, oxygen storage capacity, and greater availability of reactive oxygen species in Ce_0.9Ni_0.1O_2−δ relative to pristine CeO₂, thereby accounting for its improved soot oxidation activity.

3.

(a) Percentage soot conversion in the presence of N₂, (b) corresponding yield of product CO and CO₂, and (c) CO selectivity over CeO₂ and Ce_0.9Ni_0.1O_2−δ.

Soot oxidation was subsequently carried out over the two catalysts in the presence of CO₂ as a mild oxidant. The outlet gas composition, monitored by GC, revealed the presence of both CO and CO₂; however, since CO₂ was also present in the feed, its concentration could not be quantified as a reaction product. The percentage soot conversion was calculated as described in the experimental section (eq ) and is plotted in Figure a. The light–off profiles for both catalysts indicated complete soot oxidation, and the nearly identical profiles point to a comparable soot oxidation performance for CeO₂ and Ce_0.9Ni_0.1O_2−δ under CO₂ flow. This was further corroborated by the colour of the spent CeO₂ and Ce_0.9Ni_0.1O_2−δ samples, which closely resembled that of the as-prepared materials (Figure S4a), confirming complete soot oxidation. The complete soot oxidation was further substantiated from the TGA studies of the exhausted catalysts (Figure S4b). The CO yield was also similar for both materials, reaching a maximum of ∼22 mmol/g at 800 °C (Figure b). To further validate these observations, soot oxidation was carried out under steady-state conditions at 800 °C, the maximum temperature identified from the dynamic-condition experiments (Figure c). This enabled a direct comparison of the catalysts under conditions optimized for soot conversion. Both catalysts required nearly the same time to achieve complete soot oxidation (t_100%), confirming their comparable performance under these conditions. Notably, the difference in CO₂ uptake observed for CeO₂ and Ce_0.9Ni_0.1O_2−δ via CO₂-TPD did not appear to influence soot oxidation under CO₂ flow.

4.

(a) Percentage soot conversion in the presence of CO₂, (b) corresponding yield of product CO, and (c) isotherm reaction over CeO₂ and Ce_0.9Ni_0.1O_2−δ.

Apparently, the catalytic soot oxidation over CeO₂ and Ce_0.9Ni_0.1O_2−δ proceeded via two distinct mechanistic pathways, depending on whether lattice oxygen or CO₂ served as the active oxidizing species. When lattice oxygen acted as the sole oxidant under N₂ atmosphere, the higher reducibility of Ce_0.9Ni_0.1O_2−δ translated directly into superior soot oxidation activity, as the reaction rate under these conditions was governed primarily by the availability and mobility of lattice oxygen species. In contrast, upon introduction of CO₂ as an external soft oxidant, the contribution of lattice oxygen became less distinguishable. Notably, under the experimental conditions employed, the excess CO₂ present in the gas feed rendered the reaction largely independent of the CO₂ adsorption characteristics of the catalyst surface. To further elucidate the role of lattice oxygen in these reducible oxide catalysts during soot oxidation, XPS analysis was performed to probe the surface elemental composition and chemical states of both fresh and spent samples. Survey spectra of the as-prepared catalysts confirmed the presence of all expected elements (Figure S5). The deconvoluted Ce 3d spectra of the fresh samples, presented in Figure a, depict 10 different peaks resulting from different electron occupancies and reveal the characteristic signature of CeO₂. The first set of five peaks, denoted as ‘v’, is assigned to Ce 3d _5/2 term, while the remaining five peaks, labeled as ‘u’, correspond to the spin–orbit splitting of Ce 3d _3/2 level. Three Ce⁴⁺ spin–orbit doublet pairs are observed: v-u at 882.9 and 900.6 eV, v″-u″ at 889.6 and 907.1 eV, and v‴-u‴ at 898.8 and 916.3 eV, originating from Ce 3d ⁹4f ² O 2p,⁴ Ce 3d ⁹4f ¹ O 2p,⁵ and Ce 3d ⁹4f ⁰ O 2p ⁶ final states, respectively. ,,, The v‴-u‴ doublet, which appears distinctly resolved from the spectral features, confirms the exclusive presence of Ce⁴⁺. Two additional doublet pairs corresponds to Ce³⁺ are identified at v⁰-u⁰ (880.2–898 eV) and v′-u′ (885–903.4 eV), corresponding to the Ce 3d ⁹4f ² O 2p ⁵ and Ce 3d ⁹4f ¹ O 2p ⁶ final states, respectively. , The Ce⁴⁺ content in CeO₂ and Ce_0.9Ni_0.1O_2−δ was determined to be 66.4% and 57.5%, respectively (Table S4). The incorporation of Ni into the CeO₂ lattice resulted in a reduction of Ce⁴⁺ concentration, attributable to electronic interactions between the two metals; specifically, Ni donates electrons to cerium, thereby facilitating the partial reduction of Ce⁴⁺ to Ce³⁺. This synergistic interaction between Ni and Ce significantly modulates the redox properties of the mixed oxide system, promoting the formation of a higher concentration of oxygen vacancies, which likely accounts for the observed improvement in catalytic performance. The spent catalyst recovered after soot oxidation under N₂ atmosphere exhibited a significantly reduced Ce⁴⁺ content, with values of 60.7% and 52.3% measured in CeO₂ and Ce_0.9Ni_0.1O_2−δ, respectively (Figure b). This reduction is indicative of the facile Ce⁴⁺/Ce³⁺ redox couple, which facilitates the participation of lattice oxygen in soot oxidation under inert conditions. Similarly, the spent catalyst recovered after soot oxidation under CO₂ atmosphere also displayed a decreased Ce⁴⁺ content, reaching 61.1% in CeO₂ and 55.3% in Ce_0.9Ni_0.1O_2−δ (Figure c).

5.

Deconvoluted core-level spectra of Ce 3d from (a) as prepared, (b) exhausted catalysts under N₂ atmosphere, and (c) exhausted catalysts under CO₂ atmosphere.

The Ni 2p core-level spectra of Ce_0.9Ni_0.1O_2−δ in its as-prepared and exhausted forms presented in Figure show the distinct Ni²⁺ 2p _3/2 and Ni 2p _1/2 peaks at ∼855.1 and ∼872.3 eV, respectively, with their associated satellite feature located at 861.2 eV. ,− In the exhausted Ce_0.9Ni_0.1O_2−δ catalyst under CO₂ atmosphere, the Ni 2p core-level spectrum reveals a subtle shoulder emerging at approximately ∼853.4 eV, indicative of nickel migrating out of the lattice along with a minor presence of NiO. ,, This observation is further validated by the XRD and FE-SEM analyses of the exhausted catalyst (Figure S6), which shows a weak NiO diffraction peak, collectively confirming the active participation of nickel in the catalytic reaction.

6.

Ni 2p XPS core level spectra of (from bottom to top) as-prepared, exhausted catalysts under N₂ atmosphere, and exhausted catalysts under CO₂ atmosphere.

Since soot oxidation under N₂ atmosphere is governed primarily by surface-active oxygen species, the O 1s core-level spectra of both fresh and spent catalysts were examined. Deconvolution of the O 1s spectra revealed three distinct components: a peak at 529.2 eV assigned to surface lattice oxygen (O_I), a peak at 531.4 eV attributed to surface-adsorbed hydroxyl groups in the oxygen vacancies (O_II), and a peak at 533.2 eV corresponding to weakly bound surface oxygen (O_III) (Figure ). ,, The relative concentrations of O_II in the as-prepared and spent catalysts are summarized in Table S4. While it is acknowledged that O 1s deconvolution carries inherent limitations  particularly the non-uniqueness of peak fitting and the potential overlap of vacancy-associated oxygen with surface hydroxyl and carbonate species  the data nonetheless reveal meaningful trends. The as-prepared Ni-doped Ce_0.9Ni_0.1O_2−δ exhibited a higher O_II concentration relative to pristine CeO₂, consistent with the charge compensation requirement arising from the incorporation of Ni²⁺ into the CeO₂ lattice. It should be noted that the increase in oxygen vacancy was also observed from the Rietveld refinement and Raman data. Following soot oxidation under N₂ atmosphere, a significant increase in O_II was observed over both catalysts, consistent with the consumption of lattice oxygen during the reaction. In contrast, soot oxidation conducted under a CO₂ atmosphere resulted in a marked decrease in O_II concentration over both catalysts, indicating that the vacancies were replenished by oxygen derived from CO₂ activation. This is consistent with a mechanism in which surface oxygen vacancies act as adsorption and activation sites for CO₂, facilitating its dissociation and the reincorporation of oxygen into the lattice  thereby restoring the oxidized state of the catalyst surface and reducing the vacancy-associated signal observed in the XPS deconvolution. To provide more validation, EPR and Raman spectra were collected. Figure S7 displays the EPR spectra of both fresh and exhausted catalysts. The calculated g value was about 2.013, confirming the presence of oxygen-vacancy-related paramagnetic centers. The generation of oxygen vacancies leaves unpaired electrons associated with the O 2p orbitals, which become localized near the vacancy sites and give the characteristic EPR signal. The resonance peak at around 336 mT is attributed to surface oxygen vacancies. An increase in the peak intensity indicates a higher concentration of vacancies in the as-prepared Ce_0.9Ni_0.1O₂ compared to pristine CeO₂. After soot oxidation under N₂ atmosphere, a resonance peak corresponding to Ce³⁺ was observed at approximately 342 mT with a g value of 1.96, suggesting the formation of Ce³⁺ species within the catalysts. − The intensity of the Ce³⁺ EPR signal is significantly higher in Ce_0.9Ni_0.1O₂ than for pristine CeO₂. Consequently, for CO₂-exhausted samples, the peak intensity related to vacancies is reduced, indicating partial replenishment of oxygen vacancies by CO₂. This observation correlates with our XPS analysis. For further clarification, we collected Raman data for the exhausted catalysts. The Raman data are displayed in Figure S7. A sharp peak around 465–470 cm^–1 arising due to the symmetric vibration of Ce–O bonds in a fluorite structure confirms the structural integrity maintained for all the spent catalysts after both experiments. The exhausted catalysts under N₂ atmosphere exhibited the defect band D ∼ 1330 cm^–1 and G band ∼ 1580 cm^–1 arising from carbon materials, − and their intensity decreased in Ce_0.9Ni_0.1O₂ compared to CeO₂. These D and G bands arising from carbon materials are not detectable for CO₂-exhausted samples. This leads to the conclusion that the coke is removed completely. EPR and Raman spectra are strongly related to our mechanistic conclusion.

7.

Core level analysis of O 1s (a) as prepared, (b) exhausted catalysts under N₂ atmosphere, and (c) exhausted catalysts under CO₂ atmosphere.

The mechanistic interpretation of catalytic soot oxidation over Ce_0.9Ni_0.1O_2−δ can be delineated into two distinct pathways based on the prevailing oxidant conditions. The formation of heteroatomic Ce⁴⁺–O–Ni²⁺ linkages within the solid solution modifies the local electronic structure and weakens metal–oxygen bonding in the lattice. This structural perturbation enhances the reducibility of the oxide and increases the lability of lattice oxygen, thereby establishing the foundation for an active Mars–van Krevelen-type (MvK) redox cycle. In the absence of an external oxidant, soot oxidation proceeds exclusively via the MvK mechanism at the soot–catalyst interface. The facile Ce⁴⁺/Ce³⁺ redox couple, as evidenced by the concomitant decrease in Ce⁴⁺ content, facilitates the release of reactive lattice oxygen species, which oxidize the carbonaceous soot predominantly to CO, with CO₂ as a minor product. This process is accompanied by the progressive generation of oxygen vacancies (O_V) within the Ce_0.9Ni_0.1O_2−δ lattice, and the reaction rate is therefore governed by the availability and mobility of lattice oxygen. The superior reducibility of Ce_0.9Ni_0.1O_2−δ relative to pristine CeO₂ accounts for its enhanced soot oxidation activity in the absence of an external oxidant. Under a CO₂ atmosphere, both lattice oxygen and CO₂-derived oxygen contribute simultaneously to the soot oxidation process. The oxygen vacancies generated through lattice oxygen consumption serve as active sites for the adsorption and dissociative activation of CO₂, enabling the reincorporation of oxygen into the lattice and the regeneration of the catalyst surface. This dynamic vacancy–replenishment cycle sustains the redox activity of the catalyst throughout the reaction, as schematically illustrated in Figure .

8.

Schematic representation of redox mechanism during the reaction as observed from the core-level XPS data.

4. Conclusion

CeO₂ and Ce_0.9Ni_0.1O_2−δ were synthesized via a one-step solution combustion method and characterized before and after catalytic testing. Rietveld refinement confirmed the formation of a phase-pure cubic fluorite solid solution, with Ni occupying Ce Wyckoff sites and a decrease in oxygen occupancy consistent with vacancy formation through aliovalent substitution  corroborated by an enhanced defect-induced Raman band at ∼600 cm^–1. H₂-TPR and O₂-TPD demonstrated that Ni doping markedly increases reducibility and lattice oxygen mobility, while CO₂-TPD revealed additional moderate and strong basic sites absent in pristine CeO₂. Soot oxidation under N₂ atmosphere, where lattice oxygen serves as the sole oxidant, yielded ∼35% conversion over Ce_0.9Ni_0.1O_2−δ versus ∼20% over CeO₂ at 800 °C, with CO as the dominant product consistent with a MvK mechanism. Post-reaction XPS confirmed a decrease in Ce⁴⁺ content and an increase in oxygen vacancy concentration, providing direct spectroscopic evidence for lattice oxygen participation. Under CO₂ flow, both catalysts achieved complete soot oxidation with comparable light–off profiles and CO yields (∼22 mmol g^–1 at 800 °C), as excess CO₂ in the gas phase saturated adsorption capacity and masked differences in surface basicity. Post-reaction XPS revealed a decrease in vacancy concentration under CO₂ conditions, consistent with vacancy replenishment through CO₂ dissociation, while partial Ni exsolution in exhausted Ce_0.9Ni_0.1O_2−δ confirmed its active role in the catalytic cycle. These results establish that the superior activity of Ce_0.9Ni_0.1O_2−δ under inert conditions stems from its higher oxygen vacancy concentration, improved lattice oxygen lability, and facile Ce⁴⁺/Ce³⁺ redox cycling, while under CO₂, both catalysts sustain activity through a dynamic vacancy–replenishment cycle. This work demonstrates the feasibility of CeO₂-based materials for CO₂-driven catalyst regeneration in industrial processes, contributing to both carbon utilization and greenhouse gas mitigation.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.6c07523.

• ED- XRF, BET isotherms, FE-SEM, XPS survey spectra and XRD patterns of the catalysts (PDF)

The authors declare no competing financial interest.