Synthesis of Ceria-Based Mixed Oxides with Copper, Manganese, and Molybdenum for Diesel Soot Catalytic Combustion

Abstract

Emission control of diesel particulate matter (soot) combustion is important for environmental reasons. Catalysts are indispensable for optimizing these processes, as they significantly reduce the combustion temperature. In this work, mixed oxides (cerium–copper, cerium–manganese, and cerium–molybdenum) were prepared by co-precipitation under reasonably similar synthesis conditions, and the effects of their chemical composition on diesel soot combustion were evaluated using the Printex U model particulate. Thermogravimetric analysis (TG/DTG) and temperature-programmed oxidation coupled with mass spectrometry (TPO/MS) were employed for activity characterization. Structural analyses revealed the presence of nanocrystalline phases containing CeO₂ (fluorite), CuO (monoclinic), Mn₂O₃ (cubic), and MoO₃ (orthorhombic), depending on the catalyst composition. The most effective catalysts exhibited an equimolar oxide composition (CeO₂–MO_x). Tests performed at optimized calcination temperatures and with the addition of promoters led to the identification of optimal combustion conditions. The highest activity, corresponding to the lowest combustion temperature, was observed in the following order: CeO₂–Mn₂O₃ > CeO₂–CuO > CeO₂–MoO₃, with values of 382, 409, and 425 °C, respectively, under tight-contact conditions at a Printex U:catalyst ratio of 1:20. With the addition of a 10% Ag₂O promoter, the CeO₂–Mn₂O₃ catalyst further reduced the oxidation temperature to 376 °C. Reusability tests generally indicated a 10–20% decrease in catalytic activity by the third reaction cycle.

1. Introduction

Particulate matter (PM) emissions resulting from diesel combustion are one of the main environmental and health problems associated with road transport and stationary engines. Diesel particulate matter (soot/black carbon) contains a complex mixture of elemental carbon, organic carbon, trace metals, and adsorbed hydrocarbons that, when inhaled, are associated with acute and chronic effects on the respiratory and cardiovascular systems and are also implicated in carcinogenic risks [1,2]. In addition to the direct impact on human health, diesel particles also play an important climatic role: black carbon absorbs solar radiation and contributes to regional and global warming over short timescales, which is why mitigating particulate emissions provides climatic benefits in addition to improving air quality [3,4,5]. From a technological and regulatory standpoint, particulate reduction involves several complementary actions: improved combustion, the use of fuels and blends that generate less soot, the adoption of post-combustion treatments such as diesel particulate filters (DPFs), and strategies for controlling the regeneration/oxidation of the material accumulated in the filter. DPFs have demonstrated significant reductions in the mass of PM emissions, but their effectiveness depends on the composition of the particulate matter, engine operating conditions, regeneration frequency, and regeneration temperature [6,7].

In the experimental and research field, commercial carbon black (e.g., Printex U) is frequently used as a surrogate material to study oxidation kinetics, accumulation behavior, and DPF regeneration because its physicochemical properties allow reproducible experiments, which preceded complementary tests with real engine soot. These studies of soot and Printex U kinetics and reactivity are crucial for designing efficient regeneration strategies and predicting the durability of emission control systems [8,9]. Finally, the sustained reduction in diesel particulate emissions through fuel policies, stricter emission standards, fleet modernization, and the application of after-treatment technologies results in measurable gains for public health, reduces the burden on healthcare systems, and provides long-term climate and environmental benefits. The recent literature highlights both the benefits achieved by measures already implemented and the remaining technical challenges, such as variations in PM composition and the need to validate technologies under real-world operating conditions [10,11].

One of the most effective strategies to mitigate these impacts is the catalytic oxidation of particulate matter, which allows for the regeneration of diesel particulate filters (DPFs), emission reduction, and improved energy efficiency. A widely used model in laboratory studies to represent engine soot is Printex U, a carbon black that has a high elemental carbon content and a porous structure, which facilitates the controlled analysis of oxidation kinetics. In classic investigations, the oxidation kinetics of diesel soot and Printex U by O₂ have been shown to occur in two distinct combustion stages: a rapid initial phase (consuming ~20% of the carbon) followed by a second phase in which the porous spheres of Printex U burn internally. In this second stage, the kinetic reaction order with respect to O₂ is approximately 0.5, and the activation energy obtained is about 150 kJ/mol [12,13].

The presence of catalysts can significantly reduce the ignition temperature of soot, facilitate oxygen transfer, and accelerate oxidation. For example, in the study by Zhang et al. [14], it was found that catalysts such as ZrO₂, CeO₂, and especially those containing Pt, when mixed with Printex U, decrease the peak oxidation temperature and improve combustion indices and oxidation stability. Furthermore, ash materials from accumulated diesel particulate matter (ash) can exert an additional catalytic effect on soot oxidation [14,15]. Recent work has demonstrated that ash deposited on the substrate of diesel particulate filters (DPFs) contributes to lowering the burnout temperature (the temperature at which soot combustion is complete) of diesel particulate matter, although the effect for Printex U is smaller, indicating that its more resistant structure slightly delays this reduction in temperature [16].

It is also important to understand the effects of the internal structure of the soot, its particle size, porosity, the presence of soluble organic fractions (SOF, soluble organic fraction), and its mixture with inert or catalytic materials. The presence of soluble organic fractions and ash alters both the reactivity and the oxidation mechanisms (for example, by facilitating or hindering oxygen diffusion). These factors directly affect kinetic parameters such as activation energy, reaction order, and oxidation rates [12,17]. Noble-metal or mixed-oxide-based catalysts, transition metals, and alkaline promoters (such as potassium and silver) have been extensively researched for this purpose. For example, M. Gross et al. [18] explored how potassium (K) impregnation on CeO₂ affects the diesel soot oxidation mechanism, revealing that multiple intermediate steps occur and that intimate contact between soot and catalyst is essential to maximize efficiency. It has been demonstrated that a spillover mechanism, together with oxidation–reduction processes, acts synergistically during soot catalytic combustion [19].

In this article, a series of CeO₂-based nanocatalysts doped with copper, manganese, or molybdenum were synthesized using reasonably similar co-precipitation method and characterized to evaluate the effect of metal incorporation into the ceria matrix on improving catalytic performance in diesel soot combustion. Printex U was used as a surrogate for soot, representing the least reactive fraction of particulate matter produced during engine combustion. Various characterization techniques, such as X-ray diffraction (XRD), Brunauer–Emmett–Teller (BET) specific surface area analysis, thermal analysis (TG/DTG), and temperature-programmed oxidation coupled with mass spectrometry (TPO/MS), were used to evaluate the materials and their catalytic activity. Additionally, some catalysts were evaluated in the presence of promoters, and the most active catalyst was reused to perform a preliminary assessment of its robustness and catalytic stability.

2. Experimental

2.1. Preparation of Cerium-Copper Oxide

All reagents employed in the syntheses were analytical grade (≥99%), obtained from Sigma-Aldrich (St. Louis, MO, USA) or Merck (São Paulo, Brazil), and used without further purification.

Mixed oxides of the type Ce_xCu_1-xO₂ were prepared in different proportions ranging from 10 to 80 mol% CuO. For this purpose, mass calculations were performed to obtain the corresponding proportions for a total mixed-oxide mass of 5 g. These precursors were dissolved in distilled water in a 100 mL volumetric flask and subsequently transferred to a 250 mL round-bottom flask under magnetic stirring; 5 mol·L⁻¹ NaOH solution was then added to adjust the pH to 10 (measured using a pH meter). The solution was allowed to react for 4 h, after which it was filtered and washed until free of chloride ions.

After filtration, the resulting solid was dried at 100 °C in an oven and subsequently calcined in a muffle furnace (EDG, model EDG3P-S, São Paulo, Brazil) for 2 h at temperatures of 300, 400, 500, 600, and 700 °C, under a static air atmosphere with a heating rate of 10 °C·min⁻¹. The calcination step promoted the decomposition of the precursors, the formation and stabilization of oxygen-mediated bonds between the metallic species, and the homogenization of the metal oxides. The same procedure was applied to all compositions.

A promoter was introduced into some samples after calcination using the aqueous impregnation method. The third metal incorporated was Cr, Fe, or Ag, followed by complete evaporation of the solvent, using Cr(NO₃)₃·9H₂O, Fe(NO₃)₃·9H₂O, and AgCl as precursor salts, with metal loadings of 5 and 10 wt.% relative to the total mass of the catalyst. The promoted catalysts were subsequently calcined at different temperatures.

2.2. Preparation of Cerium-Manganese Oxide

The samples Ce_xMn_1-xO₂ were obtained by the co-precipitation method. Mixtures in the range of 20 to 75 mol% manganese(III) oxide (Mn₂O₃) were prepared using CeCl₃·7H₂O and MnCl₂·4H₂O as metal precursors. These precursors were dissolved in 100 mL of distilled water and transferred to a 250 mL flask, and 23 mL of ammonium hydroxide solution (NH₄OH) was added dropwise to adjust the pH to 10 (measured using a pH meter). The solution was maintained under constant stirring for 4 h at room temperature. The resulting material was vacuum-filtered and washed with distilled water to eliminate chloride ions and any remaining substances from the precipitating agent, and then it was dried in an oven at 100 °C. It was subsequently calcined for 2 h at 600 °C in a muffle furnace (EDG, model EDG3P-S, Brazil) under a static air atmosphere with a heating rate of 10 °C·min⁻¹, resulting in the complete removal of volatile species and the formation of the mixed oxide.

For the incorporation of a promoter, after calcination an aqueous impregnation step with a third metal (K, Cs, Fe, Cr, or Ag) was performed, followed by complete evaporation of the solvent. For this purpose, potassium chloride, cesium chloride, iron(III) nitrate nonahydrate, chromium(III) nitrate nonahydrate, and silver nitrate were used as precursor salts. Metal loadings of 10 wt.% relative to the total mass of the catalyst were applied. The solids obtained after impregnation and evaporation were calcined under the same conditions as the unpromoted mixed oxide (600 °C for 2 h).

2.3. Preparation of Cerium-Molybdenum Oxide

This system was investigated only at the 50 mol% composition because this stoichiometry showed the highest catalytic activity in the Ce–Cu and Ce–Mn systems (see Results and Discussion). Therefore, in this case, the effect of the preparation method was investigated instead of varying the composition.

Three preparation methods of CeO₂–MoO₃ mixed oxides with a 50 mol% proportion of each metal were carried out to obtain compositions of the type Ce_0.5Mo_0.5O₂: (1) co-precipitation with the Ce precursor added first, followed by the Mo precursor; (2) co-precipitation with the Mo precursor added first, followed by the Ce precursor; and (3) solid-state synthesis. In all cases, it was used CeCl₃·7H₂O and (NH₄)₆Mo₇O₂₄·4H₂O as precursors.

Each precursor was separately dissolved and diluted in a 100 mL volumetric flask and subsequently transferred to a 1 L beaker, with the cerium salt added first followed by the molybdenum salt (and vice versa for method 2). The precursor solutions were stirred with a magnetic stir bar, and after homogenization of the solution, a 5 mol·L⁻¹ NaOH solution (approximately 17 mL) was added until the pH reached 10 (measured using a pH meter), after which the mixture remained under stirring for an additional 4 h. The material was then vacuum-filtered and washed with at least 1 L of deionized water until the wash water reached neutral pH. The resulting solid was left to stand for 24 h to ensure complete drying. Subsequently, the material was transferred to a porcelain crucible, ground, and calcined at 400 °C for 2 h in a muffle furnace (EDG, model EDG3P-S, Brazil) using a heating rate of 10 °C·min⁻¹. For the solid-state synthesis, the precursors were weighed in the same proportions and mixed in a mortar with a pestle for 5 min, and the resulting material was calcined under the same conditions described above, but for 6 h.

2.4. Methods for Characterization

Elemental analysis of the metal content was performed using energy-dispersive X-ray fluorescence (EDXRF, Shimadzu, model EDX-720, Kyoto, Japan). The reported sample compositions are referred to in the text as nominal loadings.

Powder diffraction patterns of the materials were acquired using a Bruker diffractometer (D8 Focus, θ-2θ, Bremen, Germany) with CuK_α radiation (λ = 0.15418 nm), operated at a tube power of 40 kV and 30 mA. The 2θ diffraction angle was scanned in the range of 2° to 80° at an angular velocity of 1° min⁻¹.

The thermal stability of the catalysts was evaluated using TG/DTG curves recorded on a thermal analyzer system (TA Instruments, model SDT 2960, New Castle, DE, USA). The measurements were performed under a synthetic air (99.999%) flow rate of 110 mL·min⁻¹, at a heating rate of 10 °C·min⁻¹, and over a temperature range from 25 to 800 °C.

The specific surface area, pore diameter, and pore volume of the solids were determined using a gas sorption system (Micromeritics, model ASAP 2020C, Norcross, GA, USA) under nitrogen adsorption–desorption at −196 °C. The specific surface area was calculated using the Brunauer–Emmett–Teller (BET) desorption isotherm, the total pore volume (V_p) was determined at a relative pressure (P/P₀) of 0.98, and the average pore diameter was calculated using the Barrett–Joyner–Halenda (BJH) model. The samples (approximately 0.5 g) were pretreated under vacuum at 300 °C for 6 h to remove adsorbed gases from the solids.

2.5. Analytical Techniques for Catalytic Diesel Soot Oxidation

2.5.1. Mixture of Catalyst with Diesel Soot

Prior to evaluating the catalytic activity of the mixed oxides, each catalyst was combined with a standard diesel soot (Printex U, Evonik, Essen, Germany). This model soot was previously characterized and corresponds to compositions reported in the literature [20]. Printex U was mixed with the oxide at a variable mass ratio (e.g., 1:20), and tight-contact conditions were employed, since the degree of contact between the catalyst and soot plays a crucial role in the reaction. This method was chosen because of its high reproducibility in the laboratory compared to alternative approaches, as reported in the literature [15]. The mixing process was performed using an agate mortar and pestle for 5 min to ensure intimate contact between the components. The resulting catalyst–soot mixture was stored in a dark glass flask for subsequent use in the reactor. This procedure was carried out in triplicate for oxidation studies and exhibited excellent reproducibility.

2.5.2. Thermal Analysis (TG/DTG)

The catalytic activity was evaluated for the oxidation of Printex U, which was mixed with the catalyst as described above. The resulting mixture (~15 mg) was placed in an alumina crucible and subjected to a temperature ramp from room temperature (~25 °C) to 800 °C, using compressed air (99.999%; (O₂ + N₂) with 20 ± 0.5% O₂) at a flow rate of 110 mL·min⁻¹ and a heating rate of 10 °C·min⁻¹. The TG/DTG curves were recorded using a simultaneous TG–DSC analyzer (TA, model SDT 2960, New Castle, DE, USA).

2.5.3. Temperature-Programmed Oxidation Coupled with Mass Spectrometry (TPO/MS)

The combustion of Printex U was evaluated by temperature-programmed oxidation coupled with mass spectrometry (TPO/MS), performed on a reaction system (Altamira Instruments, model AMI-90R, Pittsburgh, PA, USA). This equipment was coupled to a mass spectrometer (Ametek, Dycor Dymaxion System, Berwyn, PA, USA) with a mass-to-charge (m/e) detection range of 0–100, allowing continuous and simultaneous monitoring of up to eight channels (compounds). The catalyst–Printex U mixture (100 mg, mass ratio of 1:20) was loaded into a U-shaped quartz reactor tube and placed inside the furnace. The experiments were conducted in two steps: (1) the catalyst was dried at 150 °C (from 25 to 150 °C, with a heating rate of 5 °C·min⁻¹) under an argon (Ar) flow of 20 cm³·min⁻¹; (2) the temperature was increased from 150 to 750 °C at a heating rate of 10 °C·min⁻¹, while switching the gas feed to 10% O₂ in Ar at a flow rate of 10 cm³·min⁻¹. The reaction products were monitored via their characteristic mass fragments (m/e), according to the NIST webbook [21]: 18 (base peak of H₂O); 28 and 29 (base and secondary peaks of CO); 32 (base peak of O₂); and 44 and 45 (base and secondary peaks of CO₂). The oxygen storage capacity (OSC) was measured following the methodology described in a previous study [20].

3. Results and Discussion

3.1. Cerium–Copper Oxide Characterization

In this report, the term “mixed oxide” will be used in its most general sense, i.e., an oxide containing cations of more than one chemical element. The results showed that the obtained values were close to the expected (theoretical) compositions, as shown in

Table 1. Elemental analysis of Cu and Ce by EDXRF, showing theoretical (T) and experimental (E) values. The metals were considered in the form of their oxides. All catalysts were calcined at 700 °C.

Catalyst	CuO—T (%)	CuO—E (%)	CeO2—T (%)	CeO2—E (%)
Ce–Cu–10	10	9.2	90	90.8
Ce–Cu–20	20	22.1	80	77.9
Ce–Cu–30	30	32.1	80	67.9
Ce–Cu–50	50	51.3	50	48.7
Ce–Cu–80	80	79.4	20	20.6

. Therefore, the catalysts are hereafter abbreviated according to their nominal compositions for ease of notation. The small deviations between nominal and measured compositions confirm the reliability of the synthesis procedure.

In the XRD powder pattern (Figure 1), an amorphous peak is present between 2θ = 10–20°. The cause of this behavior has been attributed to the formation of small, disordered, or poorly crystalline intermediate hydroxides/hydrous oxides that do not undergo complete transformation into well-ordered crystalline oxides (either a fluorite-type lattice or a metal oxide lattice). Fast precipitation using NaOH might have accelerated this condition due to local supersaturation phenomenon. However, a predominant diffraction pattern with peaks attributed to a cubic fluorite crystalline structure (Fm-3m [225]) is observed. The characteristic diffraction peaks at 2θ values of 28.6, 33.3, 47.6, 56.3, 76.6, and 79.0° were identified and indexed as the (111), (200), (202), (311), (313), and (402) planes, respectively (PDF # 96-900-9009), which correspond to most relevant detected peaks of ceria. Nonetheless, nascent CuO peaks could be detected even when the Cu content was as low as 10%. Thus, additional peaks related to the monoclinic CuO phase (C2/c [15]) were assigned at 32.9, 35.6, 38.6, 48.9, 53.6, 59.0, 61.5, 66.1, and 69.4° (PDF # 96-157-1908), as well as 75.1, and 88.3° (PDF # 96-901-4935), indexed as (110), (11-1), (111), (20-2), (020), (202), (-113), (022), (220), (22-2), and (-131) planes, respectively. In addition, peaks relative to Cu₂O and CuOH, involving Cu(I) species were detected, probably due to partial thermal decomposition of copper(II) salts above 600 °C. For cubic Cu₂O, it was marked the peak (+) at 29.5°, plane (101) (PDF # 96-900-5770), whereas for triclinic CuOH it was marked (*) the peaks at 22.1° (002), 31.9° (11-2), and 45.5° (203) (PDF # 96-434-5989).

The presence of the CuO phase together with CeO₂ at low Cu contents indicates some oxide segregation rather than the formation of only a solid solution of metal oxide, as observed for the CeO₂–ZrO₂ system [22,23]. This crystalline CuO phase has also been reported in the literature even at low copper loadings. Furthermore, it has been shown that the amounts of crystalline and non-crystalline CuO depend on the calcination temperature, which can significantly influence the activity of these mixed oxides toward CO oxidation. The nature of the synthesized material is therefore a determining factor for its application [24,25,26].

The CeO₂–CuO mixed oxides were applied in the oxidation of Printex (a soot model) under various calcination conditions and Printex-to-catalyst ratios. Initially, all catalysts were calcined at 500 °C, and the maximum oxidation temperature (T_MAX) was measured using TG/DTG under synthetic air atmosphere. The results are presented in

Table 2. Maximum oxidation temperature (T_MAX) of soot over the catalysts (all calcined at 500 °C) and tested at a Printex-to-catalyst ratio of 1:20.

Catalyst	TMAX (°C)
Ce–Cu–10	452
Ce–Cu–20	445
Ce–Cu–30	438
Ce–Cu–50	431
Ce–Cu–80	435

and clearly demonstrate that the Ce–Cu–50 sample was the most active (T_MAX = 431 °C), since lower oxidation temperatures indicate higher catalytic activity.

The Ce–Cu–50 catalyst was calcined at different temperatures to evaluate the effect of calcination temperature on catalytic activity. The results showed that calcination at 700 °C provided the highest activity (

Table 3. Maximum oxidation temperature (T_MAX) of soot over the Ce–Cu–50 catalyst (50 mol % Cu) at different calcination temperatures, evaluated at a Printex-to-catalyst ratio of 1:20.

Temperature (°C)	300	400	500	600	700
TMAX (°C)	474	440	431	417	409

). Subsequently, soot oxidation reactions were carried out using different Printex-to-catalyst ratios with the Ce–Cu–50 catalyst calcined at 700 °C (

Table 4. Maximum oxidation temperature (T_MAX) of soot over the Ce–Cu–50 catalyst (calcined at 700 °C) as a function of the Printex-to-catalyst ratio.

Calcination Temperature (°C)	Ratio Printex: Catalyst	TMAX (°C)
700	1:5	426
700	1:10	428
700	1:20	409

).

A representative TG/DTG/DTA profile for soot oxidation is shown in Figure 2. The results indicate that the optimal soot oxidation was achieved using a Printex-to-catalyst ratio of 1:20 with the Ce–Cu–50 catalyst calcined at 700 °C, which exhibited a maximum oxidation temperature (T_MAX) of 409 °C. This ratio has previously been established for another catalytic system studied in our laboratory [27]. The relatively low T_MAX value indicates the high catalytic activity of the Ce–Cu–50 system under tight-contact conditions.

In order to enhance the stability and activity of these catalysts, three distinct promoters (Cr, Ag, and Fe) were added at a 10 wt.% loading to the most active catalyst in the series, namely Ce–Cu–50, and the resulting materials were calcined at 700 °C. X-ray diffraction (XRD) patterns of the promoted catalysts are presented in Figure 3. As observed in Figure 1, species of CuOH were detected and marked (*) as peaks at 22.1° (002), 31.9° (11-2), and 45.5° (203) (PDF # 96-434-5989). In addition, it was detected the presence of all promoters in the form of: (a) trigonal Cr₂O₃ (marked as ■) at 37.5°, plane (110) (PDF # 96-901-4204); (b) cubic Ag₂O (marked as ▲) at: 38° (200), 44.5° (211), 64.5° (311), and 77° (321) (PDF # 96-431-8189); (c) tetragonal Fe₂O₃ (marked as ●) at 33.9°, plane (013) (PDF # 96-901-2693). Elemental analysis confirmed the presence of the promoter species in the catalysts. Thus, at this promoter loading, a third crystalline oxide phase is present in the materials in addition to CeO₂ and CuO.

When the promoted catalysts were applied to soot oxidation, the maximum oxidation temperature of Printex (T_MAX) decreased in all cases, indicating an effective promoting effect; however, the best performance was observed for the Ag₂O-promoted catalyst (

Table 5. Maximum oxidation temperature (T_MAX) values for catalysts calcined at 700 °C and tested at a Printex-to-catalyst ratio of 1:20, containing different promoters (10 wt.% of Cr₂O₃, Ag₂O, and Fe₂O₃) in the Ce–Cu–50 system, together with the actual content of each promoter (P) determined by EDXRF analysis.

Promoter/Catalyst	TMAX (°C)	P (%)
Ce–Cu–50	409	0
Cr2O3/Ce–Cu–50	400	9.8
Ag2O/Ce–Cu–50	390	8.9
Fe2O3/Ce–Cu–50	398	9.5

). Thus, the synergistic interaction among the three metals present in the material is likely fundamental for enhancing the oxidation capacity and promoting the combustion mechanism, thereby lowering T_MAX [25,26].

Catalysts based on CeO₂ and CuO, and in some cases incorporating a third metal, have been extensively prepared using different methods for application in CO oxidation, ethyl acetate oxidation, and diesel soot combustion, to cite only a few examples [28,29,30,31,32]. The preparation method and stoichiometric composition are crucial for increasing the catalytic activity of these materials, which typically consist of nanometric CuO crystallites dispersed on ceria. Accordingly, the preparation method developed in this work showed effective activation of soot, which is considered the most challenging process among the reactions mentioned above. The effectiveness of soot oxidation depends on the redox cycling capacity of both the Ce(IV)/Ce(III) and Cu(II)/Cu(I) couples, which is associated with improved oxygen mobility and enhanced oxygen transfer between the catalyst surface and the soot particles. This behavior has been investigated through temperature-programmed reduction (TPR) profiles of these composites, demonstrating that higher reductive capacity at lower temperatures improves TMAX, i.e., shifts the temperature corresponding to the maximum oxidation rate to lower values [32,33].

3.2. Cerium–Manganese Oxide Characterization

Considering the results obtained for cerium–copper oxide materials, the catalysts containing manganese were prepared with a similar stoichiometric composition. The manganese and cerium contents were determined by EDXRF analysis using the Quali-Quant method with primary standards. The experimental values were close to the theoretical values calculated during the sample preparation (

Table 6. Elemental analysis of Mn and Ce by EDXRF, showing theoretical (T) and experimental (E) values. The metals were considered in the form of their oxides. All catalysts were calcined at 600 °C.

Catalyst	Mn2O3—T (%)	Mn2O3—E (%)	CeO2—T (%)	CeO2—E (%)
Ce–Mn–20	20	20.2	80	79.8
Ce–Mn–35	35	36.8	65	63.2
Ce–Mn–50	50	53.3	50	46.7
Ce–Mn–75	75	75.9	25	24.1

). The reported compositions are therefore based on the nominal quantities. The small deviations between nominal and experimental compositions confirm the reliability of the synthesis procedure.

The XRD patterns of the oxides were obtained, and the corresponding diffraction patterns are shown in Figure 4. At 20 mol% Mn₂O₃, only the diffraction peaks related to the ceria phase were detected, at 2θ values of 29.3, 33.7, 48.2, and 55.5°, indexed to the (111), (200), (220), and (311) planes, respectively. At Mn contents above 20%, additional peaks corresponding to cubic Mn₂O₃ were observed at the planes (200), (222), and (321). For comparative purposes, pure Mn₂O₃ is also shown, with the main assigned peaks at 18.4, 33.3, 36.5, 38.6, 55.6, 60.9, and 66.7°, indexed to the (200), (222), (321), (400), (530), (620), and (444) planes, respectively (PDF #76-0150). The presence of the Mn₂O₃ phase together with CeO₂ at higher Mn contents indicates oxide segregation rather than the formation of a single solid solution, as reported in the literature [33,34,35]. This oxide segregation may influence the redox properties and catalytic performance of the Ce–Mn oxide system.

It was possible to obtain textural data for these oxides and to calculate the average crystalline domain size, based on the average of the four ceria 2θ diffraction peaks (Figure 4), except for pure Mn₂O₃, for which the peak at the (222) plane was used (

Table 7. Total specific surface area (S_BET, calculated by BET method), total pore volume (V_p, obtained at P/P₀ = 0.98), average crystalline domain size (D, calculated using the Scherrer equation), and average pore diameter (P_D, calculated by BJH method) of the CeO₂–Mn₂O₃ catalysts.

Catalyst	SBET (m2/g)	Vp (cm3/g)	D (nm)	PD (nm)
CeO2	30.2	0.14	19.6	19.8
Ce–Mn–20	12.7	0.24	58.9	17.0
Ce–Mn–35	50.0	0.26	55.2	16.5
Ce–Mn–50	53.3	0.30	40.4	18.2
Ce–Mn–75	29.0	0.16	44.8	17.3
Mn2O3	6.1	0.03	228.8	11.4

). The Ce–Mn–50 catalyst exhibits the highest specific surface area and pore volume. Moreover, the increased domain size of the particles compared to pure CeO₂ indicates aggregation of the primary oxide particles.

Considering all the diffractograms presented in Figure 4, the pattern corresponding to the oxide with a 50 mol % of CeO₂–Mn₂O₃ showed the most asymmetric diffraction peaks. This behavior may be associated with a loss of crystallinity introduced by distortions in the crystal lattice, resulting in the formation of more defects or vacancies in the CeO₂ crystal structure. This effect may explain why this sample exhibits a slightly higher surface area [36]. Analysis of the average crystalline domain size of the different mixed oxides (Table 7) shows that the oxide containing a 50 mol % of Mn₂O₃ exhibited the smallest crystallite size. In addition, the specific surface area of the catalysts gradually increases with increasing manganese content, reaching 53.3 m² g⁻¹ for Ce–Mn–50; beyond this composition, the surface area decreases, reaching 6.1 m² g⁻¹ for pure Mn₂O₃. The pore volume follows the same trend as the specific surface area. The average pore diameter ranged from 11.4 to 18.2 nm, characterizing the obtained oxides as mesoporous solids, with the maximum value observed for Ce–Mn–50. Combined with the XRD results, these findings support greater structural vacancy formation due to decreased crystallite size.

The CeO₂–Mn₂O₃ oxides, calcined at 600 °C, were applied in the oxidation of Printex at a Printex-to-catalyst ratio of 1:20. The maximum oxidation temperature (T_MAX) was measured using TG/DTG under a synthetic air atmosphere, and the results are presented in Figure 5. Clearly, the Ce–Mn–50 sample exhibited the highest catalytic activity, with a T_MAX of 382 °C. In all cases, it can be observed that the oxidation temperature was significantly lower than the combustion temperature of pure Printex (622 °C), which corresponds to non-catalytic oxidation. The higher surface area and smaller crystallite size of Ce–Mn–50 likely facilitate oxygen adsorption and diffusion, contributing to the observed lower T_MAX.

The Ce–Mn–50 oxide was calcined at 500, 600, and 700 °C and tested for the oxidation of Printex, but the best result was obtained for the sample calcined at 600 °C (T_MAX = 400, 382, and 387 °C, respectively). Thus, this sample was used in the subsequent optimization step. In a search for improved catalytic performance, the best catalyst (Ce–Mn–50) was tested with different impregnated promoters (K₂O, Cs₂O, CuO, Cr₂O₃, Ag₂O, and Fe₂O₃) at loadings of 5 and 10 wt.%. As previously observed for the Ce–Cu–50 catalysts, the best results were obtained at 10 wt.% loading, and the corresponding results are shown in

Table 8. Maximum oxidation temperature (TMAX) of soot obtained from TG/DTG profiles for the Ce–Mn–50 mixed oxide with different promoters (10 wt.%), and the corresponding actual promoter content (P) determined by EDXRF analysis.

Promoter/Catalyst	TMAX (°C)	P (%)
Ce–Mn–50	382	0
K2O/Ce–Mn–50	467	11.1
Cs2O/Ce–Mn–50	493	10.4
CuO/Ce–Mn–50	387	10.5
Fe2O3/Ce–Mn–50	445	9.7
Cr2O3/Ce–Mn–50	469	9.9
Ag2O/Ce–Mn–50	376	9.3

. As observed, several promoters led to a decrease in catalytic activity, evidenced by an increase in the maximum oxidation temperature (T_MAX). A possible reason is that the additional calcination step at 500 °C required for promoter activation may have caused sintering. In this context, alternative promoter incorporation methods could potentially improve the catalytic activity of these materials. Nonetheless, the maximum oxidation temperature (T_MAX) decreased from 382 °C to approximately 376 °C when Ag₂O was used, confirming it as the most effective promoter.

The XRD pattern of the 10 wt.% Ag₂O/Ce–Mn–50 catalyst (Figure 6) confirmed the presence of the Ag₂O phase, with a diffraction peak at 2θ = 38.2°, indexed as the (200) plane. Using the Scherrer equation, the average crystallite size of the mixed oxide was recalculated, and it was observed that it remained close to its original value (approximately 40 nm) even in the presence of Ag₂O nanocrystallites.

Based on the analysis of the TG/DTG curves and the maximum oxidation temperatures of the most active catalyst with the most effective promoter, namely 10 wt.% Ag₂O/Ce–Mn–50, temperature-programmed oxidation coupled with mass spectrometry (TPO/MS) experiments were conducted to evaluate the redox properties of the materials. Figure 7 highlights the significant effect of promoter addition to the catalyst. The oxidation temperature range of Printex in the presence of Ag₂O is considerably narrower, with a marked decrease in the maximum CO₂ formation temperature and a minor release of CO. In comparison with the oxidative process in the absence of a catalyst (Figure 8), the reaction was carried out with pure Printex and with Printex mixed with silica, which is considered an inert oxide with no oxidizing activity. In these cases, the oxidation temperature is significantly higher, and a substantial amount of CO is formed during the combustion process. It should be noted that the peaks marked with (*) correspond to the first temperature maximum related to the most volatile compounds adsorbed on the particulate matter, which has a predominantly carbonaceous composition. In the catalyzed process, this peak duplication is not observed, due to the higher efficiency of the catalytic combustion.

Therefore, the enhanced activity of CeO₂–Mn₂O₃ materials can be attributed to an increase in oxygen vacancies, which play an important role in the oxidation of soot to CO₂. The insertion of Mn modifies both the redox and structural properties of pure ceria, leading to an increase in the activity of these catalysts compared with CeO₂.

3.3. Cerium–Molybdenum Oxide Characterization

The third mixed oxide system studied consisted of cerium and molybdenum. This system was prepared only at a 1:1 ratio (i.e., 50 mol % of each oxide), since this composition showed the highest catalytic activity in diesel soot oxidation among the previously studied mixed oxides. Subsequently, for the Ce–Mo catalyst, variations in the preparation method were investigated to evaluate the influence of the synthesis route. As in previous sections, elemental analysis was used to determine the actual oxide composition (

Table 9. Elemental analysis of Mo and Ce by EDXRF, showing theoretical (T) and experimental (E) values, for Ce–Mo–50, prepared using three different methods: (1) co-precipitation with the Ce precursor added first, followed by the Mo precursor; (2) co-precipitation with the Mo precursor added first, followed by the Ce precursor; and (3) solid-state synthesis. The metals were considered in the form of their oxides. All catalysts were calcined at 400 °C.

Method (X)/Catalyst	MoO3—T (%)	MoO3—E (%)	CeO2—T (%)	CeO2—E (%)
(1)/Ce–Mo–50	50	50.7	50	49.3
(2)/Ce–Mo–50	50	51.2	50	48.8
(3)/Ce–Mo–50	50	54.1	50	45.9

). The experimental values were reasonably close to the theoretical ones; therefore, the nominal composition notation was maintained throughout the article.

XRD analysis was the primary technique used to identify differences among the materials after calcination at 400 °C. Figure 9 presents the XRD patterns obtained for the different preparation methods. At first glance, it is possible to observe that methods (1) and (2) produced similar materials, whereas method (3) resulted in a markedly different XRD pattern. For methods (1) and (2), an amorphous halo was detected between 2θ = 10–20°, along with the main diffraction peaks corresponding to the cubic fluorite CeO₂ phase. The characteristic peaks at 2θ = 28.6, 33.1, and 47.5° were identified and indexed as the (111), (200), and (220) planes, respectively. In addition, diffraction peaks related to α-MoO₃ were observed at 2θ = 45.3, 56.3, and 59.8°, indexed as the (004), (103), and (081) planes, respectively, according to JCPDS file # 05-0508 [37,38]. This result indicates that mixed crystalline phases were formed. In contrast, the sample synthesized by the solid-state method (3) exhibited only two small diffraction peaks associated with CeO₂, located at 2θ = 47.5 and 55.6°, corresponding to the (220) and (311) planes, respectively. All remaining diffraction peaks matched those of α-MoO₃, appearing at 2θ = 12.5, 22.2, 25.8, 27.7, 32.9, 39.7, 45.3, 49.2, and 59.8°, indexed as the (020), (110), (040), (021), (111), (060), (004), (002), and (081) planes, respectively. These results clearly indicate that solid-state synthesis led to the formation of a material with a higher content of crystalline MoO₃. In addition, it was also observed that there is an amorphous halo between 2θ = 10–20°, as in the case of the Ce–Cu system, using methods (1) and (2). Probably due to the use of NaOH as a precipitating agent, it was possible to detect some poly-molybdate species such as Mo₅O₁₄ (marked as ▼) (PDF # 96-153-7519).

It was possible to obtain textural properties for these oxides, as well as to calculate the crystalline domain size based on the CeO₂ diffraction peak at 2θ = 28.6°, except for sample (3)/Ce–Mo–50, for which the crystalline domain size was calculated using the MoO₃ phase peak at 2θ = 27.7° (

Table 10. Total specific surface area (S_BET, by BET method), total pore volume (V_p, obtained at P/P₀ = 0.98), average pore size (P_s, BJH desorption average pore width, 4V/A), and average crystalline domain size (D, calculated using the Scherrer equation). – not available datum.

Method (X)/Catalyst	SBET (m2/g)	Vp (cm3/g)	Ps (nm)	D (nm)
CeO2	30.2	–	–	–
(1)/Ce–Mo–50	9.9	0.13	30.5	55
(2)/Ce–Mo–50	10.4	0.12	35.1	56
(3)/Ce–Mo–50	12.7	0.11	38.3	55
Ag2O/(1)/Ce–Mo–50	20.1	0.10	18.2	57
MoO3	27.6	–	–	–

). It can be observed that the Ce–Mo–50 catalysts, regardless of the synthesis method, showed low specific surface areas. This behavior may be related to the relatively low calcination temperature (400 °C), which may be insufficient to promote the full development of crystalline domains in these mixed oxides. In addition, the solid-state synthesis method was also ineffective in promoting the development of the ceria crystalline phase, since pure MoO₃ usually crystallizes at lower temperatures [39]. Therefore, higher temperatures may be required, either to calcine the products of coprecipitation (methods 1 and 2) or to heat the precursor mixture more intensely, thereby promoting more effective diffusion of the metal elements and the formation of better-defined crystalline phases in the mixed oxides [40]. It should be mentioned that the material obtained by this solid-state synthesis method exhibited poor homogeneity, with a mixture of colors, including green, gray, and yellow. In addition, two important observations should be noted. First, during the preparation of sample (1)/Ce–Mo–50, the XRD pattern showed a higher number of peaks related to the MoO₃ phase, which was attributed to the condition of the Mo precursor, (NH₄)₆Mo₇O₂₄·4H₂O. This reagent tends to form MoO₃ over time during storage. Thus, when the Mo(VI) solution was prepared, a turbid solution was obtained, which was attributed to the presence of MoO₃, since this oxide is not soluble in water. Second, another key step in syntheses (1) and (2) was the control of pH. The use of a pH meter proved to be more reliable than pH paper, as reflected in the XRD patterns. Therefore, careful control of the initial precursors and accurate pH measurements are essential to obtain reproducible materials.

Preliminary test using TG/DTG was conducted on the Ce–Mo–50 sample (prepared by method 1) calcined at 400 and 500 °C for the oxidation of Printex. The T_MAX obtained was 435 and 449 °C, respectively. Therefore, calcination at 400 °C was selected for the subsequent performance tests.

The materials were first tested in the oxidation of Printex using TG/DTG analysis (Figure 10). The maximum oxidation temperature (T_MAX) indicated that the catalysts prepared by methods (1), (2), and (3) exhibited values of 435, 456, and 465 °C, respectively. A complementary analysis using TPO showed the corresponding profile curves (Figure 11) and the relative maximum of the main products (CO₂, CO, and H₂O). Considering CO₂ production, which showed a sharper peak, T_MAX values of 425, 445, and 475 °C were obtained for methods (1), (2), and (3), respectively. At the end of the combustion, all the Printex carbon was oxidized, as indicated by the color change in the mixture from black to pale yellow, corresponding to the original catalysts from methods (1) and (2). However, the catalyst prepared by method (3) could not be recovered due to apparent decomposition; the color remained gray and yellowish, whereas the first two catalysts were fully recovered and could be reused. Therefore, under the synthesis conditions employed, the catalyst prepared by the solid-state reaction was less efficient than those prepared by co-precipitation.

3.4. General Comparison of the Performance of the Mixed Oxides

CeO₂ is known to be an excellent heterogeneous catalyst for many reactions, such as diesel soot oxidation [41], the dehydration of alcohols [42], and organic synthesis [43]. Its crystalline and electronic structure facilitate a reversible redox cycle between Ce(IV)/Ce(III), which results in the presence of oxygen vacancies both in the bulk and on the surface. However, limitations in the oxygen storage capacity (OSC) of ceria at high temperatures require its modification in order to enhance its thermal, redox, and acid–base properties [35]. In this context, CeO₂ was modified with different metal oxides (CuO, Mn₂O₃, and MoO₃) to improve its performance in soot oxidation. The resulting materials exhibited crystalline phases of these oxides, which are known to affect catalytic activity.

Table 11. Comparative T_MAX and T_RECY (temperature of the third reutilization cycle), crystalline domain size (D), and oxygen storage capacity (OSC) of the best catalysts for Printex oxidation.

Catalyst	TMAX (°C)	TRECY (°C)	D (nm)	OSC (µmol/g)
Ce–Cu–50	409	480	58	215
Ag2O/Ce–Cu–50	390	468	60	245
Ce–Mn–50	382	432	40	326
Ag2O/Ce–Mn–50	376	418	41	351
(1)/Ce–Mo–50	425	468	55	197
Ag2O/(1)/Ce–Mo–50	410	463	57	212

summarizes the relevant parameters and properties that best describe the performance of the most effective catalysts prepared in this study.

It can be observed that the system Ce–Mn–50 showed the best performance, i.e., the lowest temperature for soot oxidation, which was slightly improved with the addition of an Ag₂O promoter. In comparison with data reported in the literature, this performance can be attributed to the increase in oxygen vacancies, which enhances the OSC of the materials [31,35,40,44], depending on the additional metal included in the ceria composition. This trend corresponds with the observed OSC values for the catalysts examined in this study (see Table 11). The insertion of copper, manganese, or molybdenum into the ceria structure decreases the lattice parameter, increases the degree of disorder in the fluorite lattice, and promotes the formation of oxygen vacancies. The differences in cationic radii (Ce(IV) = 97 pm; Cu(II) = 84 pm; Mn(III) ≈ 90 pm; Mo(VI) = 73 pm, assuming cubic coordination) allow these ions to be introduced into the lattice. This enhances the synergistic interaction between CeO₂ and MO_x, contributing to improved redox properties during soot oxidation cycles [35,41,44]. An increase in the amount of M incorporated into the CeO₂ lattice distorts the original fluorite structure and likely leads to the formation of both solid solutions and segregated oxide phases, which can enhance the combustion capacity of these mixed oxides. It should be noted that the nature of the resulting material strongly depends on the preparation method, as widely reported in the literature [45,46,47,48].

Thus, the best results were clearly obtained for the CeO₂–Mn₂O₃ system, and the addition of 10% Ag₂O nanoparticles, localized at the interface of the mixed oxide phase, reduced T_MAX to the lowest soot oxidation temperature. Another important parameter is the reusability of the catalyst. Accordingly, three reuse cycles were tested for each catalyst, and the loss of activity was generally around 10–20% after the third cycle. This decrease is probably because of sintering of the nanocatalysts, a trend also reported for other systems in the literature. It should be noted that during each cycle, the catalyst–soot system was repeatedly heated to approximately 750 °C, which also affects particle size and the exposure of the most active ceria planes, resulting in the observed loss of catalytic activity [20,22,35]. Further studies in this reuse process are necessary to minimize this activity loss and thus improve the number of cycles in which this material can be used.

4. Conclusions

In this work, several ceria-based catalysts forming mixed oxides (CeO₂–CuO, CeO₂–Mn₂O₃, and CeO₂–MoO₃) were prepared using the reasonable similar co-precipitation method under controlled conditions. Variations in their compositions and calcination temperatures were investigated to evaluate structural effects and performance improvements in the catalytic combustion of diesel particulates. For each system, the best catalyst was calcined at optimized temperatures for diesel soot oxidation (700, 600, and 400 °C, for Ce–Cu–50, Ce–Mn–50, and Ce–Mo–50, respectively). The CeO₂ (fluorite), CuO (monoclinic), Mn₂O₃ (cubic), and MoO₃ (orthorhombic) crystalline phases were readily identified by XRD, depending on the MO_x content. The formation of both solid solutions and segregated oxide phases was found to influence the catalytic activity of these materials. The best composition was an equimolar mixture of binary oxides (Ce–M–50), which showed the lowest oxidation temperatures (T_MAX). Under tight-contact conditions and a Printex U-to-catalyst ratio of 1:20, the following activity order was observed: CeO₂–Mn₂O₃ > CeO₂–CuO > CeO₂–MoO₃ with T_MAX values of 382, 409, and 425 °C, respectively. With the addition of 10% Ag₂O (the most effective promoter), the order remained the same, but at lower temperatures, namely 376, 390, and 410 °C, respectively. TPO/MS results demonstrated that the catalytic process is much more effective both in lowering the combustion temperature of Printex U and in enhancing selectivity toward the formation of CO₂ and H₂O. Studies on the reuse of the top-performing catalysts showed a decrease in activity, ranging from 10% to 20% after the third catalytic cycle. Therefore, the results presented highlight pathways for structural optimization of these catalysts and indicate the need for further detailed studies to better understand their structure and the mechanisms responsible for deactivation in the most active materials.