The Formation of Mn-Ce-Zr Oxide Catalysts for CO and Propane Oxidation: The Role of Element Content Ratio

Abstract

The MnO_х-ZrO₂-CeO₂ oxide catalysts were synthesized by co-precipitation method with varying (1) Zr/Zr + Ce molar ratio at constant manganese content of 0.3; (2) manganese content at constant Zr/Ce molar ratio of 1; (3) Mn/Mn + Zr molar ratio at constant Ce content of 0.5. Catalysts are characterized by XRD, N₂ adsorption, TPR, and XPS. The catalytic activity of all the series was tested in the CO and propane oxidation reactions. In contrast to the variation of the manganese content, the Zr/Zr + Ce molar ratio does not significantly affect the catalytic properties. The dependence of the catalytic activity in CO oxidation on the manganese content has a «volcano» shape, and the best catalytic performance is exhibited by the catalyst with Mn/(Zr + Ce) = 1. In the case of propane oxidation reaction, there is «sigma» like dependence, activity increases with increase of Mn/(Mn + Zr + Ce) molar ratio up to 0.3, stabilizing with a further increase in the manganese content. XRD and XPS have shown that with an increase of the Mn concentration in the MnO_x-ZrO₂-CeO₂ catalysts, the amount of crystalline manganese oxides such as Mn₂O₃ and Mn₃O₄, as well as the surface concentration of Mn cations, increases. While the content of Mn_xZr_yCe_1-x-yO₂ solid solution decreases, the concentration of manganese cations (x) in volume of Mn_xZr_yCe_1-x-yO₂ mixed oxide grows. The maximum activity in CO oxidation corresponds to the balance between the amount of the solid solution and the concentration of manganese cations in the volume of mixed oxide. The propane oxidation reaction is less sensitive to the state of manganese ion rather than to its amount. In this case, a decrease in the content of the Mn_xZr_yCe_1-x-yO₂ solid solution with increase in manganese amount in catalyst is compensated by an increase in content of crystalline manganese oxides and the surface concentration of manganese.

1. Introduction

Transition metal oxide catalysts have been widely accepted as promising catalysts for the combustion of CO, hydrocarbons, and volatile organic compounds. Among these, manganese oxide catalysts are attracting a great deal of attention since they are cheap, active, and environment-friendly. Generally, the activity of manganese catalysts correlates with their redox properties [1,2,3]. Manganese ions easily change the oxidation state, forming various types of oxides such as MnO₂, Mn₅O₈, Mn₂O₃, Mn₃O₄, and MnO, and provide highly mobile oxygen for oxidation reactions [4].

Doping the cerium oxides with manganese significantly increases catalytic activity compared to bulk manganese oxides [5]. Due to the easy transition between Ce³⁺ and Ce⁴⁺ and the high mobility of O²⁻ ions in the CeO₂ lattice, ceria modifies the redox activity of manganese, stabilizing it in higher oxidation states at low temperatures that is preferable for oxidation reactions and facilitating oxygen migration [6]. Apart from the formation of labile oxygen and promotion of the process of oxygen activation and transfer, ceria has a significant oxygen storage capacity and is itself active in oxidative reactions, for example, in soot oxidation [7]. The inclusion of zirconium in the ceria lattice improves the thermal stability of oxide and further increases the oxygen mobility [6,8]. Terribille et al. [6] found that the highest reduction degree and the lowest reduction temperatures are achieved at an equimolar ratio of cerium and zirconium Ce_0.5Zr_0.5O₂.

The properties of ternary Mn-Zr-Ce catalysts have been extensively studied in the reactions of total oxidation of O-, Cl-containing volatile organic substances and CO [9,10,11,12,13], soot oxidation [14,15], and removal of NO_x by selective catalytic reduction with ammonia [16,17,18]. Catalytic properties of Mn–Zr–Ce oxides depend on the preparation procedure and the ratio between elements. These factors determine the state of the active component: supported MnO_x nanoparticles or defective Mn-Zr-Ce solid solution. Depending on the preparation conditions, one of these states prevails. In the case of impregnation methods, manganese nanoparticles are spread uniformly on the surface of support. The impregnation approach effectiveness was shown for a low content of manganese. Kaplin et al. compared 8 wt% MnO_x-Ce_0.8Zr_0.2O₂ prepared by co-precipitation and wet impregnation and showed that in the case of impregnation no manganese ions are embedded into the crystal lattice of Ce_0.8Zr_0.2O₂ oxide, while the surface of catalyst contains MnO_x states [12]. Hou et al. utilized the Ce_0.65Zr_0.35O₂ as support for MnO_x/Ce_0.65Zr_0.35O₂ catalysts for oxidation of toluene, and the catalyst with 15 wt% MnO_x loading performed the best [13]. However, at relatively high manganese content, the formation of a solid solution has a greater impact on the catalysis properties than MnO_x nanoparticles. Solution combustion, co-precipitation, and sol–gel methods were used to prepare a predominantly Mn_xZr_yCe_1-x-yO₂ solid solution. Long et al. developed a Mn-Ce-Zr ternary mixed oxides catalyst by solution combustion method [19]. Zhu et al. synthesized Mn-Ce-Zr catalysts by the sol–gel method varying the molar ratio of Mn/(Mn + Ce + Zr), while the molar ratio of Ce/Zr was maintained as 1, the Mn_0.67Ce_0.16Zr_0.16 catalyst showed maximum activity in the series [11]. The Zr_0.4Ce_0.6−xMn_xO₂ solid solutions were synthesized by a sol–gel method and tested in the total oxidation of butanol. The presence of both Ce⁴⁺/Ce³⁺ and Mn⁴⁺/Mn³⁺ redox couples led to excellent catalytic activity for n-butanol complete oxidation with an activity maximum achieved for Zr_0.4Ce_0.24Mn_0.36O₂ [9].

Most studies of Mn-Zr-Ce catalysts are aimed at determining the effect of manganese content, method of preparation, and support composition on the catalytic activity. To the best of our knowledge, no systematic studies on the role of element ratio have been reported for Mn-Zr-Ce catalysts synthetized by co-precipitation method, although the composition of the catalyst is an important factor determining catalytic activity. Understanding how the composition of the catalyst affects the state of the active component and its properties in the target reaction is the key to successfully achieving the best performance of the catalytic process [20,21]. In this work, we studied the effect of composition on the catalytic, structural, microstructural, and redox properties of the Mn-Ce-Zr oxide catalysts. Three series of Mn-Ce-Zr catalysts were synthesized by co-precipitation with varying (1) Zr/Zr + Ce molar ratio at constant manganese content of 0.3; (2) manganese content at constant Zr/Ce molar ratio of 1; (3) Mn/Mn + Zr molar ratio at constant Ce content of 0.5. The physicochemical properties were studied by X-ray diffraction (XRD), temperature-programmed reduction (TPR-H₂), and X-ray photoelectron spectroscopy (XPS), the catalytic activity was tested in the reaction of CO and propane oxidation.

2. Results

2.1. Variation of Mn/(Mn + Zr) Molar Ratio

The influence of the Mn/(Mn + Zr) molar ratio on the catalytic properties was studied for the series of (Mn,Zr)0.5Ce0.5 catalysts (Mn/(Mn + Zr) molar ratio was in the range from 0 to 1 at constant Ce content of 0.5). As reference, the Zr0.5Ce0.5 sample was used. Figure 1a,c show the temperature dependence of CO and C₃H₈ conversion a for (Mn,Zr)0.5Ce0.5 catalysts. Figure 1b,d illustrate the temperature of 50% conversion of CO and C₃H₈ depending on the Mn/(Mn + Zr) molar ratio. For the Zr0.5Ce0.5 catalyst, CO oxidation begins at a reaction temperature of 250 °C, and 50% CO conversion is achieved at 389 °C. For (Mn,Zr)0.5Ce0.5 samples, CO conversion curves drastically shift to lower temperatures with increasing Mn content. In the range of Mn/(Mn + Zr) from 0.2 to 0.8, a noticeable shift of the CO conversion curve to a lower temperature region was observed. The Т₅₀ gradually decreases from 179 to 157 °С for the Mn/(Mn + Zr) ratio of 0.2 and 0.8, correspondingly. Sample Mn0.5Ce0.5 activity does not change significantly, and T₅₀ is 160 °С, which is slightly higher than for Mn0.4Zr0.1Ce0.5.

In the reaction of propane oxidation, a smoother change in the behavior of the catalyst is observed. The curves of the propane conversion gradually shift to lower temperatures with increasing Mn content, Т₅₀ for the Zr0.5Ce0.5 sample is 470 °С and drops to 260 °С for the sample Mn0.5Ce0.5 (Mn/(Mn + Zr) = 1).

Figure 2 shows XRD patterns for (Mn,Zr)0.5Ce0.5 catalysts. All XRD patterns contain wide peaks located at 2θ = 13.1, 15.1, 21.4, 25.1, 26.3, 30.4, 33.3, and 34.2. These peaks correspond to the CeO₂ fluorite-type structure (sp.gr. Fm(-)3m, PDF No. 43-1002). The lattice parameter a of fluorite-type phase in the sample Ce0.5Mn0.5 is equal to 5.397 Å (

Table 1. Phase composition, lattice parameter of solid solution (Ce(Mn,Zr)O₂, and t-Zr(Ce,Mn)O₂CSR and surface area of (Mn,Zr)0.5Ce0.5 catalysts.

Catalyst	Phase Content, % Wt.	Lattice Parameter *, Å	CSR, Å	SBET, m2/g
Mn0.1Ce0.5Zr0.4	60% Ce(Mn,Zr)O2	5.371	100	87
	40% t-Zr(Ce,Mn)O2	5.213	70
Mn0.2Ce0.5Zr0.3	46% Ce(Mn,Zr)O2	5.391	70	103
	52% t-Zr(Ce,Mn)O2	5.297	40
	2% Mn2O3	-	230
Mn0.3Ce0.5Zr0.2	53% Ce(Mn,Zr)O2	5.383	80	91
	43% t-Zr(Ce,Mn)O2	5.307	30
	4% Mn2O3	-	270
Mn0.4Ce0.5Zr0.1	36% Ce(Mn,Zr)O2	5.414	50	77
	56% t-Zr(Ce,Mn)O2	5.354	130
	6% Mn2O3	-	110
	2% Mn3O4	-	80
Ce0.5Zr0.5	40% (Ce,Zr)O2	5.396	120	66.5
	60% t-(Zr,Ce)O2	5.241	60
Mn0.5Ce0.5	87% Ce(Mn)O2	5.397	80	55.5
	8% Mn2O3	-	110
	5% Mn3O4	-	150
Mn0.1Ce0.5Zr0.4	60% Ce(Mn,Zr)O2	5.371	100	87
	40% t-Zr(Ce,Mn)O2	5.213	70
Mn0.2Ce0.5Zr0.3	46% Ce(Mn,Zr)O2	5.391	70	103
	52% t-Zr(Ce,Mn)O2	5.297	40
	2% Mn2O3	-	230

* for t-ZrO₂ lattice parameter was calculated as a * = (2 × a × a × c)^1/3.

). This value slightly differs from the lattice parameter of pure CeO₂ (5.411 Å), indicating the formation of Ce(Mn)O₂ solid solution, since the ionic radii of Ce⁴⁺ and Mn³⁺ are 0.97 Å and 0.66 Å, respectively [22]. For the samples with high Mn content, additional reflections were observed belonging to Mn₃O₄ (sp.gr. I4₁/amd, PDF No. 24-0734) and Mn₂O₃ (sp.gr. Ia(-)3, PDF No. 41-1442). An increase in the content of Zr results in a decrease of the amount of crystalline Mn oxides and appearance of the oxide based on the structure of tetragonal ZrO₂, as evidenced by the appearance of additional reflections at 2θ = 13.4, 15.4, 15.6, 22.0, and 22.1, corresponding to 011, 002, 110, 112, and 020 reflections (ZrO₂, sp.gr. P4₂/nmc, PDF No. 42-1164). For a convenient comparison of the lattice parameters of mixed oxides, we used the normalized lattice parameter of the unit cell in the case of tetragonal oxide (Table 1, a * = (a × a × c)^1/3). The normalized lattice parameter of oxide with tetragonal structure varies in the range of 5.213–5.354 Å, which significantly differs from the value of pure ZrO₂ (a * = 4.110 Å, PDF No. 42-1164), indicating the formation of the second solid solution t-Zr(Ce,Mn)O₂. Unfortunately, for both solid solutions it is impossible to estimate its composition from the cell parameters. The variation of Mn/Mn + Zr molar ratio leads to a change in the lattice parameters of mixed oxides and their content ratio, indicating the redistribution of cations between two oxides Ce(Mn,Zr)O₂ and t-Zr(Ce,Mn)O₂ (Table 1). CSR of solid solutions changes in the range of 50–130 Å.

2.2. Variation of Zr/(Ce + Zr) Molar Ratio

The influence of the Zr/(Ce + Zr) molar ratio on the catalytic properties was studied for the series of Mn0.3(Zr,Ce)0.7 catalysts (Zr/(Ce + Zr) molar ratio was in the range from 0 to 1 at constant Mn content of 0.3). Figure 3a,c show the temperature dependence of CO and C₃H₈ conversion a for Mn0.3(Zr,Ce)0.7 catalysts. Figure 3b,d illustrate the temperature of 50% conversion of CO and C₃H₈ depending on the Zr/(Ce + Zr) molar ratio. For ternary Mn0.3(Zr,Ce)0.7 catalysts, similar catalytic properties in a wide range of cerium and zirconium contents are observed. Mn0.3(Zr,Ce)0.7 catalysts have a slightly higher catalytic activity in the CO oxidation reaction compared to the double oxides Mn0.3Zr0.7 and Mn0.3Ce0.7.

The value of Т₅₀ is in the range of 159–164 °C for catalysts with molar ratio Zr/(Zr + Ce) = 0.14–0.71 and 178 and 172 °С for Mn0.3Ce0.7 and Mn0.3Zr0.7, respectively. Catalytic tests of the studied series of samples in the propane oxidation reaction showed similar results. Mn0.3(Zr,Ce)0.7 catalysts with the molar ratio Zr/(Zr + Ce) of 0.14–0.71 are more active than double oxides Mn0.3Zr0.7 and Mn0.3Ce0.7. The value of Т₅₀ is in the range of 258–284 °C for catalysts with molar ratio Zr/(Zr + Ce) = 0.14–0.71 and 309 and 314 °C for Mn0.3Ce0.7 and Mn0.3Zr0.7, respectively. The Mn0.3Zr0.6Ce0.1 catalyst is characterized by the lowest activity in the propane oxidation reaction and similar activity to Mn0.3Ce0.7 and Mn0.3Zr0.7 in the CO oxidation reaction; T₅₀ of this sample is 355 °C and 176 °С, respectively. The reasons for this behavior will be discussed further.

XRD patterns of Mn0.3(Zr,Ce)0.7 catalysts are shown in Figure 4. Similarly to the (Mn,Zr)0.5Ce0.5 catalysts series, phases based on cubic CeO₂ and tetragonal ZrO₂ are observed. All the catalysts also contained some amount of Mn₂O₃. The cubic lattice parameter gradually shifts from 5.383 to 5.100 for Mn0.3Zr0.1Ce0.6 and Mn0.3Zr0.7, respectively (

Table 2. Phase composition, lattice parameter of solid solution, CSR, and surface area of Mn0.3(Zr,Ce)0.7 catalysts.

Catalyst	Phase Content, % Wt.	Lattice Parameter *, Å	CSR, Å	SBET, m2/g
Mn0.3Ce0.6Zr0.1	20% amorphous ZrO2	-	-	57.5
	74% Ce(Mn,Zr)O2	5.383	90
	6% Mn2O3	-	110
Mn0.3Ce0.5Zr0.2	53% Ce(Mn,Zr)O2	5.383	80	91
	43% t-Zr(Ce,Mn)O2	5.307	30
	4% Mn2O3	-	270
Mn0.3Ce0.35Zr0.35	66% Ce(Mn,Zr)O2	5.338	70	132
	29% c-Zr(Ce,Mn)O2	5.162	50
	5% Mn2O3	-	310
Mn0.3Ce0.2Zr0.5	Amorphous	-	-	162
	t-Zr(Ce,Mn)O2	5.234	40
	Mn2O3	-	130
Mn0.3Ce0.1Zr0.6	Amorphous	- - -	- - -	223
	CeO2
	Mn2O3
Mn0.3Zr0.7	54% amorphous	-	-	176
	40% t-Zr(Mn)O2	5.100	80
	6% Mn2O3	-	90

* for t-ZrO₂ lattice parameter was calculated as a * = (2 × a × a × c)^1/3.

). Moreover, an increase in the zirconium content leads to a partial amorphization of phases. The most pronounced effect is for the Mn0.3Zr0.6Ce0.1 catalyst, which could be a reason for its low catalytic activity. In general, the Mn0.3(Zr,Ce)0.7 catalysts have a higher catalytic activity in the CO and propane oxidation reaction compared to the double oxide catalyst, such as Mn0.3Zr0.7 and Mn0.3Ce0.7. It is likely that the increased activity of ternary systems compared to binary systems is associated with additional restructuration of the solid solution (formation of ternary Zr(Ce,Mn)O₂ oxides, a decrease in crystallite size, appearance of second phase).

2.3. Variation of Mn/(Mn + Ce + Zr) Ratio

The influence of the Mn/(Mn + Ce + Zr) molar ratio on the catalytic properties was studied for the series of Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts (Mn/(Mn + Ce + Zr) molar ratio was in the range from 0.1 to 0.8 at constant Zr/Ce molar ratio of 1). Figure 5a,c show the temperature dependence of CO and C₃H₈ conversion a for Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts. Figure 5b,d illustrate the temperature of 50% conversion of CO and C₃H₈ depending on the Mn/(Mn + Ce + Zr) molar ratio.

Catalyst tests show an increase in catalytic activity with increasing Mn content from 0.1 to 0.5, as evidenced by the shift of the CO conversion curve to the low-temperature region. The dependence of T₅₀ on the manganese content in the CO oxidation reaction has a U-shape with a minimum for the Mn0.5Zr0.25Ce0.25 catalyst; T₅₀ varies in the range of 195 to 140 °C (Figure 5b). For the reference sample of pure manganese oxide, Т₅₀ is 191 °С. For the reference sample Zr0.5Ce0.5, Т₅₀ is more than 375 °С.

In the reaction of propane oxidation, similarly to the (Mn,Zr)0.5Ce0.5 catalysts series the introduction of manganese into the Zr-Ce system is accompanied by a significant increase in activity with a plateau for catalysts with Mn/(Mn + Zr + Ce) > 0.3. The increase in the Mn/(Mn + Ce + Zr) ratio up to 0.3 leads to a rapid decrease in the Т₅₀ from 471 to 268 °С (Figure 5d). For other compositions, T₅₀ fluctuates in the range of 267–276 °C. For the reference sample MnO_x, a slight decrease in the degree of propane conversion is observed (T₅₀ = 285 °C). For the reference sample Zr0.5Ce0.5, Т₅₀ is more than 450 °С.

Therefore, the catalytic activity (Т₅₀) behavior for the Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts in the propane and in the CO oxidation reaction differs significantly. It is worth mentioning that activity in propane oxidation of pure manganese oxide is close to those for Mn-Ce-Zr catalysts with Mn/(Mn + Ce + Zr) = 0.3–0.8. Apparently, this is due to the lower sensitivity of the propane oxidation reaction to the environment of manganese than to its amount.

The crystal phases of Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts were determined by XRD, and the results are shown in Figure 6 and

Table 3. Phase composition, lattice parameter of solid solution, CSR, and surface area for Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2)catalysts.

Catalyst	Phase Content, % Wt.	Lattice Parameter *, Å	CSR, Å	SBET, m2/g
Mn0.1Ce0.45Zr0.45	47% Ce(Mn,Zr)O2	5.351	70	106
	53% t-Zr(Ce,Mn)O2	5.216	50
Mn0.2Ce0.4Zr0.4	64% Ce(Mn,Zr)O2	5.328	60	123
	36% t-Zr(Ce,Mn)O2	5.210	50
Mn0.3Ce0.35Zr0.35	47% Ce(Mn,Zr)O2	5.352 5.220 -	60	132
	47% t-Zr(Ce,Mn)O2		40
	6% Mn2O3		170
Mn0.4Ce0.3Zr0.3	37% Ce(Mn,Zr)O2	5.314	75	138.5
	55% t-Zr(Ce,Mn)O2	5.258	20
	4% Mn2O3	-	220
	4% Mn3O4	-	170
Mn0.5Ce0.25Zr0.25	34% Ce(Mn,Zr)O2	5.330	60	141.3
	48% t-Zr(Ce,Mn)O2	5.234	20
	7% Mn2O3	-	225
	11% Mn3O4	-	140
Mn0.6Ce0.2Zr0.2	29% Ce(Mn,Zr)O2	5.345	60	150.3
	43% Zr(Ce,Mn)O2	5.189	20
	12% Mn2O3	-	220
	15% Mn3O4	-	125
Mn0.7Ce0.15Zr0.15	36% Ce(Mn,Zr)O2	5.364	50	118
	14% Zr(Ce,Mn)O2	4.922	230
	38% Mn2O3	-	160
	12% Mn3O4	-	40
Mn0.8Ce0.1Zr0.1	22% Ce(Mn,Zr)O2	5.371	50	92.7
	11% Zr(Ce,Mn)O2	4.961	50
	30% Mn2O3	-	245
	22% Mn3O4	-	160
	15% MnO2	-	55
MnOx	Mn2O3	-	210

* for t-ZrO₂, lattice parameter was calculated as a * = (2 × a × a × c)^1/3.

. According to XRD data, with an increase in the Mn content in the catalyst, the content of crystalline manganese oxides such as Mn₂O₃ and Mn₃O₄ increases and the amount of the solid solution decreases. For Mn0.3Ce0.35Zr0.35, the content of manganese oxide was 6 wt%, while for Mn0.8Ce0.1Zr0.1 it reaches 67 wt%. Simultaneously, the concentration of manganese and, accordingly, oxygen vacancies in the composition of mixed oxides increases (details of estimation of solid solution composition is given in Supplementary Materials).

Figure 7 shows TPR-H₂ curves and total amount of hydrogen consumed for the catalysts Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) by varying the Mn/(Ce + Zr + Mn) ratio from 0.1 to 0.6. For samples with a low manganese content (Mn0.1Zr0.45Ce0.45 and Mn0.2Zr0.4Ce0.4), a wide halo was observed in the temperature range of 160–520 °C. In this range, the TPR profiles contain a broad signal of partially overlapped peaks with the main maxima at ~260–270 °C and 420 °C. These signals are associated with the consumption of hydrogen due to the stepwise partial reduction of manganese cations present in the Zr-Ce-containing solid solution [23,24,25]. The signal at ~260–270 °C corresponds to the reduction of Mn⁴⁺/Mn³⁺ ions to Mn³⁺/Mn²⁺, while the peak at ~350–360 °C shows Mn³⁺/Mn²⁺ → Mn²⁺ transition [26,27,28]. On the other hand, the consumption of hydrogen in the region at 150–300 °C can be explained by the reduction of finely dispersed MnO_x particles, which can be present on the surface of the solid solution [23,24]. As Mn content increases, more pronounced peaks with maxima at 304 and 388 °C appear, indicating the two-stage reduction of Mn₂O₃ to Mn₃O₄ and then to MnO [28], which is present according XRD data. There is also an almost linear increase in the total amount of hydrogen consumption with an increase in the manganese content (Figure 7b). The value grows from 1.44 to 3.51 mmol (H₂)/g. It is worth mentioning that the amount of hydrogen consumption before 300 °C gradually grows from 0.35 to 1.37 mmol (H₂)/g with an increase in the Mn/(Ce + Zr + Mn) ratio from 0.1 to 0.5. Further increase in the Mn/(Ce + Zr + Mn) ratio to 0.6 leads to a decrease of the amount of adsorbed H₂ before 300 °C to the value of 1.21 mmol (H₂)/g. The amount of low-temperature hydrogen consumption correlates well with catalytic activity (Figure 5 and Figure S1).

The XPS was used to evaluate changes in the electronic properties and ratios between the elements on the surface of the catalysts.

Table 4. Atomic ratios of the elements on the surface according to XPS results.

Catalyst	[Mn]/[Mn + Ce + Zr]				[O]/[Me]	Ce3+, %
	Total	%, Mn2+	%, Mn3+	%, Mn4+
Mn0.1Ce0.45Zr0.45	0.13	92	8	0	2.31	22
Mn0.2Ce0.4Zr0.4	0.27	51	49	0	2.35	20
Mn0.3Ce0.35Zr0.35	0.49	22	77	1	1.86	18
Mn0.4Ce0.3Zr0.3	0.51	29	64	7	1.98	18
Mn0.5Ce0.25Zr0.25	0.58	14	80	6	1.80	16
Mn0.6Ce0.2Zr0.2	0.65	20	70	10	1.76	20

summarizes the relative concentrations (atomic ratios) of the elements in the surface of the catalysts. Figure 8 shows the Zr3d, Ce3d, and Mn2p XPS core-level spectra of Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts. The Zr3d spectra are fitted by Zr3d_5/2-Zr3d_3/2 doublet with the Zr3d_5/2 binding energy of 181.9 eV. The binding energy of Zr3d_5/2 peak corresponds to zirconium in the Zr⁴⁺ state [29,30,31]. The Ce3d spectra of the catalysts have a complex structure of peaks corresponding to Ce³⁺ and Ce⁴⁺ states [32,33]. The fitting into individual components allows one to evaluate the fraction of Ce³⁺ cations (Table 4). It was found that cerium predominantly exists in the form of Ce⁴⁺, while the fraction of Ce³⁺ ions vary in the range of 16–22%.

To identify the chemical state of manganese, the binding energy of the Mn2p_3/2 peak and the intensity and position of satellites are used [34,35,36]. The asymmetry of main peaks is caused by the multielectron process. The fitting of the Mn2p spectra allows us to evaluate the fraction of manganese cations in the different oxidation states: Mn²⁺, Mn³⁺, and Mn⁴⁺. The Mn2p spectra are fitted by three Mn2p_3/2–Mn2p_1/2 doublets corresponding to manganese in the Mn²⁺, Mn³⁺, and Mn⁴⁺ states, and by the satellites located, respectively, at a distance of 6.7, 10.6, and 11.8 eV from the main peaks. The binding energies of Mn2p_3/2 peaks equal to 639.0, 640.7, and 641.4 eV correspond to the Mn²⁺, Mn³⁺, and Mn⁴⁺ states, respectively. The Mn2p_3/2 binding energies reported for MnO, Mn₂O₃, and MnO₂ oxides are in the ranges of 640.4–641.7, 641.5–641.9, and 642.2–642.6 eV, respectively [34,35,37,38,39]. The XPS analysis allows us to estimate the fraction of manganese in different oxidation states (Table 4). The relative surface concentrations (atomic ratios) of elements are also listed in Table 4,. With an increase in the Mn/(Mn + Ce + Zr) ratio from 0.1 to 0.6, the surface [Mn]/[Mn + Ce + Zr] ratio linearly increases from 0.13 to 0.65. Simultaneously, the average oxidation state of Mn grows from 2.1 to 2.9. With an increase in the surface concentration of Mn, a redistribution of the charge state of cations from 2+ to 4+ is observed.

3. Discussion

The ternary oxides Mn0.3(Zr,Ce)0.7 demonstrate similar catalytic properties in a wide range of cerium and zirconium contents without a clearly pronounced dependence. In general, Mn0.3(Zr,Ce)0.7 catalysts have a higher catalytic activity in the CO and propane oxidation reaction compared to the binary oxides Mn0.3Zr0.7 and Mn0.3Ce0.7. As the manganese content increases and the zirconium content decreases accordingly, the catalytic activity of (Mn,Zr)0.5Ce0.5 ternary oxides gradually increases. The highest catalytic activity in the CO oxidation reaction is achieved for the composition Mn0.4Zr0.1Ce0.5. Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) shows different behavior in the reaction of CO and propane oxidation. The dependence of the catalytic activity in CO oxidation for Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) on the manganese content has a pronounced extremum, the highest activity is observed for the composition Mn0.5Zr0.25Ce0.25. In the propane oxidation reaction, the activity of a series of catalysts Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) increases with an increase in manganese concentration to 0.3, and with further introduction of manganese remains at a constant level. Catalytic activity in propane oxidation is comparable for pure Mn₂O₃ and ternary Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) catalysts (Figure 5d). Differences in the properties of Mn(x)Zr(0.5 − x/2)Ce(0.5 − x/2) samples with varying manganese content in CO and propane oxidation reactions can be associated with different mechanisms of these two processes. The reaction of CO oxidation to CO₂ is a simpler process, the activity of which is primarily determined by the ability of the catalyst to activate oxygen [40]. In this case, the catalytic activity is determined by the content of manganese in the composition of the Mn-Zr-Ce solid solution, since the incorporation of manganese cations into the lattice of cerium and zirconium oxides is accompanied by the formation of lattice defects and oxygen vacancies, which increases the ability of the compound to accumulate oxygen and increases the oxygen mobility on the catalyst surface [41]. The propane oxidation reaction is a more complex process. Generally, propane oxidation on Mn-based catalysts follows two major steps: dehydrogenation of propane and deep oxidation of oxygenated intermediate species (aldehydes, carboxylic acid, carbonate) to CO₂ and H₂O [42]. Therefore, the catalyst must be able to activate C–H bonds and promote deep oxidation. C–H cracking is the rate-determining step for propane oxidation, so propane adsorption on the catalyst surface and its activation play a crucial role in achieving high activity [43]. In addition, according to Hu et al. [43], weakly bound oxygen and chemically adsorbed oxygen play a key role in the propane oxidation reaction, in contrast to the process of carbon monoxide oxidation.

When considering the surface area that determines quantity of active sites, one would expect that the activity will be governed by the surface area of the catalysts. Therefore, samples with the highest surface area should be the most active ones. However, although the Mn0.3Ce0.1Zr0.6 catalyst possesses the highest surface area of 223 m²/g (Table 2), its activity is not the best (T₅₀ (C₃H₈) is 268 °С, T₅₀ (CO) is 175 °С). Figures S2 and S3 show that there is no dependence between surface area and T₅₀ (C₃H₈) and T₅₀ (CO). Since our catalyst consists of different phases, we also tried to find a correlation between a certain phase and catalytic activity. Figures S4 and S5 show T₅₀ in CO and propane oxidation reactions versus content of Mn_xZr_yCe_1-x-yO₂ solid solution and crystalline manganese oxides. In the case of CO, there are no clear dependencies. When the content of solid solutions is 90–100%, both the largest and smallest values of T₅₀ are observed (for Mn0.4Ce0.3Zr0.3 and Mn0.1Ce0.45Zr0.45 catalyst T₅₀ is 147 and 196 °С, respectively). However, in the case of propane oxidation, an increase in the content of Mn oxide to 15–20 wt% leads to an increase in the catalytic activity. With further growth of Mn oxide content, T₅₀ changes in the narrow range of temperatures from 267 to 290 °С. The content of manganese (x) in the composition of the Mn_xZr_yCe_1-x-yO₂ solid solution was also estimated (the procedure of x estimation is given in Supporting Information). Figures S6 and S7 show the correlation between activity and x in Mn_xZr_yCe_1-x-yO₂ solid solution. In the case of CO oxidation, with an increase in the x content in Mn_xZr_yCe_1-x-yO₂ solid solution, a decrease in T₅₀ is observed. As x increases, the concentration of oxygen vacancies is expected to increase accordingly. According to XPS analysis, catalysts contain Mn²⁺, Mn³⁺, Mn⁴⁺, Ce³⁺, Ce⁴⁺, and Zr⁴⁺, and the average oxidation state of Mn is 2.1–2.9 (Table 4). The presence of cations in oxidation state less than 4+ implies the creation of an oxygen vacancy, based on the electroneutrality principle, which, in turn, enhances the adsorption capacity of reactive oxygen species on the catalyst surface [44]. According to TPR data, low-temperature reducibility of catalyst (before 300 °C) correlates well with catalytic activity (Figure S1). The maximum activity in CO oxidation likely corresponds to the balance between the fraction of the solid solution and the concentration of manganese in it. Crystalline manganese oxides seem to make a smaller contribution to the activity of the catalyst in the reaction of CO oxidation. At the same time, the propane oxidation reaction is less sensitive to the environment of manganese cations than to its amount. A decrease in the content of the solid solution is compensated for by an increase in the content of crystalline phases of manganese oxides and an increase in the surface content of manganese, thus maintaining high activity values. It should be noted that the catalysts obtained in this work are comparable to catalytic activity in CO and propane oxidation reactions with oxide ternary Mn-Zr-Ce systems described by other researchers (

Table 5. Conversion of propane and CO on oxide ternary Mn-Zr-Ce catalysts reported in recent publications.

Catalyst	Preparation Method	Calcination Temperature, °C	Concentration of C3H8 (CO)	GHSV, mL/(g*h)	T50, °C	Reference
Propane Oxidation
Ce0.76Zr0.19Mn0.05O(2-x)	co-precipitation	650	1 mol.%	50,000	408	[6]
Mn0.3Ce0.2Zr0.5	co-precipitation	600	1 vol.%	37,200	258	This work
CO Oxidation
Mn0.45Ce0.45Zr0.1	co-precipitation	550	4 vol.%	1800	105	[45]
MnOx–Ce0.8Zr0.2O2	co-precipitation	500	2 vol.%	36,000	160	[12]
Mn0.5Ce0.25Zr0.25	co-precipitation	600	1 vol.%	58,440	140	This work

).

4. Experimental

4.1. Catalyst Preparation

The samples were prepared by the co-precipitation method. A joint solution of ZrO(NO₃)₂, Ce(NO₃)₃, and Mn(NO₃)₂ salts was obtained, and the NH₄OH solution was gradually added into the mixture with continual stirring until the pH reached 10. The precipitation process was carried out at 80 °C. Stirring of the suspension was continued for 1 h, then H₂O₂ was added dropwise to ensure the completeness of precipitation. The amount of H₂O₂ corresponded to the molar ratio of H₂O₂:(Mn + Zr + Ce) = 1. The suspension was finally kept without stirring for 2 h. The obtained precipitate was filtered off and washed with distilled water on the filter to pH = 6–7. The samples were dried at 120 °C for 2 h and calcined in a muffle furnace at 600 °C for 4 h. Three series of Mn-Ce-Zr catalysts were synthesized by co-precipitation with varying (1) Zr/Zr + Ce molar ratio at constant manganese content of 0.3; (2) manganese content at constant Zr/Ce molar ratio of 1; (3) Mn/Mn + Zr molar ratio at constant Ce content of 0.5.

The samples are designated as MnxZryCez, where x, y, z are the molar fraction of Mn, Zr, Ce, and x + y + z = 1.

4.2. Catalyst Characterization

The phase composition was determined by XRD using a STOE STADI MP diffractometer equipped with a Mythen2 1K (Dectris, Baden, Switzerland) linear detector. The diffraction patterns were obtained in transmission θ/2θ geometry in the 2θ range from 3 to 50° with a step of 0.015° using monochromatic MoK_α1 radiation (λ = 0.7093 Å, Ge crystal monochromator). Rietveld refinement for quantitative analysis was carried out using the software package Topas V.4.2.

The specific surface area was calculated with the Brunauer–Emmett–Teller (BET) method using nitrogen adsorption isotherms measured at liquid nitrogen temperatures on an automatic Micromeritics ASAP 2400 sorptometer (Norcross, GA, USA).

The temperature-programmed reduction in hydrogen (TPR-H₂) was performed with 40–60 mg of sample in a quartz reactor using a flow setup with a thermal conductivity detector. The reducing mixture (10 vol.% of H₂ in Ar) flow was 40 mL/min. The heating rate from room temperature to 900 °C was 10 °C/min. The TPR curves were normalized per catalyst mass.

The XPS measurements were performed on a photoelectron spectrometer (SPECS Surface Nano Analysis GmbH, Berlin, Germany) equipped with a PHOIBOS-150 hemispherical electron energy analyzer and an XR-50 X-ray source with a double Al/Mg anode. The core-level spectra were obtained using AlKα radiation (hν = 1486.6 eV) under ultrahigh vacuum conditions. The binding energy (E_b) of photoemission peaks was corrected for the Ce3d_3/2-U‴ peak (E_b = 916.7 eV) of cerium oxide. The curve fitting was done by the CasaXPS software [46]. The line shape used in the fit was the sum of Lorentzian and Gaussian functions. A Shirley-type background was subtracted from each spectrum [47]. Relative element concentrations were determined from the integral intensities of the core-level spectra using the theoretical photoionization cross-sections according to Scofield [48].

4.3. Catalyst Tests

Catalyst samples were tested in CO and propane oxidation reactions in a glass reactor (170 mm × Ø 10 mm) at atmospheric pressure in a flow mode. Following that, 0.5 g of the catalyst (fraction 0.4–0.8 mm) was diluted with an inert SiO₂ sand (up to 3.0 mL). The temperature in the catalyst bed was controlled and regulated using a chromel-alumel thermocouple connected to a heat controller.

In the case of CO oxidation reaction, the gas mixture contained 1 vol.% of CO and 99 vol.% of air. The total flow rate was 487 mL/min. Catalytic tests in the CO oxidation reaction were carried out in the temperature range from 30 to 500 °C. The analysis of the inlet and outlet reaction flows were carried out using a gas chromatograph. The mixture was separated on a packed column filled with CaA zeolite (3 m). The unreacted amount of CO was determined using a thermal conductivity detector. CO conversion (X_СО) was calculated by the equation:

X_CO = |(P_CO/P_N2)ⁱⁿ − (P_CO/P_N2)^out|/(P_CO/P_N2)ⁱⁿ

(1)

where Р_СО and Р_N2 are the peak areas on the chromatogram corresponding to the CO and N₂ concentrations in the inlet (ⁱⁿ) and outlet (^out) flows. P_N2 was used as an internal standard.

Catalytic tests in the propane oxidation reaction were carried out in the temperature range from 150 to 500 °C. The reaction gas mixture contained 1 vol.% of C₃H₈ and 99 vol.% of air and was fed into the reactor at a rate of 310 mL/min. The inlet and outlet reaction flows were analyzed using a Tsvet-800 gas chromatograph supplied with a capillary column (filled with SiO₂) and a flame ionization detector. Nitrogen was used as the carrier gas. The conversion of propan (X_C3H8) was calculated from the chromatographic data using the following equation:

X_C3H8 = (Pⁱⁿ − P^out)/Pⁱⁿ

(2)

where Pⁱⁿ and P^out are the peak areas on the chromatogram corresponding to the propane concentration in the inlet (ⁱⁿ) and outlet (^out) flows.

5. Conclusions

The influence of the cation ratio in the MnO_x-ZrO₂-CeO₂ catalysts on their structural properties and catalytic activity in the CO and С₃H₈ oxidation reactions has been studied. It was shown that ternary Mn0.3(Zr,Ce)0.7 catalysts have a higher catalytic activity in the CO and propane oxidation reaction compared to the double oxide catalyst such as Mn0.3Zr0.7 and Mn0.3Ce0.7. The change in Zr/(Zr + Ce) mole ratio does not significantly affect the catalytic properties. The maximum activity in CO oxidation corresponds to the balance between the amount of the solid solution and the concentration of manganese in it. The best catalytic performance is exhibited by the catalyst with Mn/(Mn + Zr + Ce) = 0.5. In the case of propane oxidation reaction, there is «sigma» like dependence. Activity increases with manganese concentration to Mn/(Mn + Zr + Ce) = 0.3, and the activity remains constant with further introduction of manganese. The propane oxidation reaction is less sensitive to the environment of manganese cations than to its amount.