Unraveling the Structural and Compositional Peculiarities in CTAB-Templated CeO2-ZrO2-MnOx Catalysts for Soot and CO Oxidation

Structure–performance relationships in functional catalysts allow for controlling their performance in a wide range of reaction conditions. Here, the structural and compositional peculiarities in CTAB-templated CeO2-ZrO2-MnOx catalysts prepared by co-precipitation of precursors and their catalytic behavior in CO oxidation and soot combustion are discussed. A complex of physical–chemical methods (low-temperature N2 sorption, XRD, TPR-H2, Raman, HR TEM, XPS) is used to elucidate the features of the formation of interphase boundaries, joint phases, and defects in multicomponent oxide systems. The addition of Mn and/or Zr dopant to ceria is shown to improve its performance in both reactions. Binary Ce-Mn catalysts demonstrate enhanced performance closely followed by the ternary oxide catalysts, which is due the formation of several types of active sites, namely, highly dispersed MnOx species, oxide–oxide interfaces, and oxygen vacancies that can act individually and/or synergistically.


Introduction
Catalytic technologies are widely implemented in both large-scale chemical manufacturing and solving environmental challenges.The latter is linked to the reduction of the impact of various hazardous substances (CO, particulate matter, VOCs) on human health and the environment [1,2].Over the last decades, a number of catalyst formulations have been introduced for the oxidation of CO and soot, and such catalysts are mostly based on noble metals and transition metal oxides.The noble metal-based supported catalysts utilizing Rh, Pt, or Pd were widely studied due to their high activity in total oxidation reactions [3][4][5][6].However, CO was found strongly adsorbed on such metals, resulting in a significant reduction in active oxygen content that suppressed the low-temperature oxidation activity and the need to apply elevated temperatures [7].Moreover, these metals feature rather high costs and limited availability.This motivated a search for alternative less expensive and abundant catalyst components that demonstrate similar or superior performance in said reactions.
A promising alternative comprises the catalysts based on oxides of transition metals (pristine and mixed oxides [8,9], spinels [10][11][12]).These materials combine a number of advantages, including availability, thermal stability, improved service period, resistance to catalytic poisons, etc.Among such materials, ceria brings about high research interest [13] since it demonstrates enhanced abilities to accumulate and release oxygen, and features rather low costs.To further boost CeO 2 performance in terms of improved oxygen mobility and capacity as well as mechanical characteristics, the interactions of ceria with other oxides, i.e., ZrO 2 , MnO x , SnO 2 , are considered in binary and ternary formulations [14][15][16][17][18][19][20].In a number of cases, the ternary mixed oxide systems exhibited rather high performance, while the reasons for such a behavior remain under debate.
Templating methods allow for creating oxide-based catalysts with improved structural and performance properties [21].A key focus to create porous oxide catalysts using template methods is a high specific surface area that ensures effective mass transfer and diffusion of reagents to the active sites of the catalyst.The size, shape, and distribution of pores in the catalyst can be controlled by choosing the type of template and synthesis conditions [22][23][24].Among the templates to synthesize the oxide catalysts [25,26], the cationic surfactant cetyltrimethylammonium bromide (CTAB) is actively used [27,28] due to the formation of micelles that help the precipitation of the oxides in the form of nanoparticles.The CTAB also allows for stabilizing the nanoparticles, preventing them from agglomeration.
Recently, several attempts were undertaken to show the effects of template-based preparation methods on the structure of ternary CeO 2 -ZrO 2 -MnO x oxides and their performance in the complete oxidation of CO and soot.Thus, the CeO 2 -ZrO 2 prepared with a sawdust template showed improved performance in CO oxidation, which was attributed to higher structural defectiveness [29].The soot combustion activity was rather low, while the CTAB-templated sample showed the opposite behavior.In Ref. [30], for the sample prepared using the evaporation-induced self-assembly method, the Mn 2+ /Mn 3+ ions were incorporated into the CeO 2 -ZrO 2 lattice to form the MnO x particles evenly distributed over the surface and bulk of the sample.The activity of such a sample was lower as compared to the one prepared by the MnO x impregnation over the mixed oxide support, where only local surface areas contained the MnO x species to ensure additional adsorption sites.With that, it is not clear how the performance of such ternary catalysts can change if the Mn-related component is introduced simultaneously with other catalyst components.
To the best of our knowledge, there have been no attempts to prepare ternary CeO 2 -ZrO 2 -MnO x catalysts using CTAB-templated synthesis when the precursor of the MnO x component is introduced not by impregnation of the CeO 2 -ZrO 2 support, but simultaneously with other oxide precursors.Despite thorough investigation, there are open questions related to the nature and function of the surface defects.A landscape of catalyst formulations that can demonstrate the synergistic action reflected in the enhanced catalyst performance can also be widened.
The present work is focused on the structural and compositional peculiarities in CTABtemplated CeO 2 -ZrO 2 -MnO x catalyst and its catalytic behavior in both CO oxidation and soot combustion.A complex of physical-chemical methods (low-temperature N 2 sorption, XRD, TPR-H 2 , Raman, XPS) is used to elucidate the features of the formation of interphase boundaries in multicomponent oxide systems and the defects formed.

Synthesis of CeO 2 -ZrO 2 -MnO x Oxide Systems
To prepare the oxide-based systems, a hydrothermal synthesis with CTAB as a structureforming additive was used [31].The oxide precursors (Ce(NO 3 ) 3 •6H 2 O, Mn(NO 3 ) 2 •nH 2 O, ZrO(NO 3 ) 2 •nH 2 O) were dissolved in water, and the CTAB was added to the obtained solution with a molar ratio of CTAB to oxides of 1:3.Then, the mixture was put into an autoclave at a temperature of 100 • C for 24 h.After filtering, the samples were dried at 80 • C for 16 h and then calcined at 600 • C for 2 h.The Ce/Mn and Ce/Zr ratios were selected to be no less than 4 and 1.5, respectively, based on the literature data [32,33].The sample designations show the nominal chemical composition (    (*) expressed as atomic ratio of metal to total of metals (sum).

Materials Characterization
A complex of physical-chemical methods was used to characterize the prepared samples to provide comprehensive information about their composition, structure, and properties.
The catalysts were characterized by low-temperature nitrogen adsorption to measure specific surface area, porosity, and pore size distribution ("3Flex", Micromeritics, Norcross, GA, USA).The samples were previously degassed under vacuum (200 • C, 2 h).The surface area and pore size distributions were determined using the BET and BJH methods, respectively.
X-ray fluorescence analysis (XRF) was used to determine the elemental composition of the obtained samples (XRF-1800, Shimadzu, Tokyo, Japan).The applied voltage, current, and diaphragm were 40 kV, 95 mA, and 10 mm, respectively.The Rh-based anode was used as a source.
X-ray phase analysis (XRD) was used to study the phase composition of samples, estimate the size of crystallites of components, and calculate the structural parameters (XRD-7000, Shimadzu, Tokyo, Japan).The range of 2θ was 10-70.CuKα radiation was used.The crystalline phase composition was established using the PDF database.The POWDER CELL 2.5 software package was used to determine the crystal cell indexing, starting from powder diffraction data with the Lorentz simulation.The crystal size and microstrains were calculated using the Williamson-Hall method.
Temperature-programmed reduction in hydrogen (TPR-H 2 ) was used to study the features of the reduction of active components in synthesized samples ("Autochem 2950", Micromeritics, Tokyo, Japan).The temperature range was 25-900 • C. The device was equipped with a thermal conductivity detector.A mixture of 10 %vol.H 2 in argon was used.The absence of water in the gas mixture was controlled by a trap with a mixture of isopropanol and liquid nitrogen with temperature of −86 • C. The heating rate and gas flow rate were 10 • C/min and 20 mL/min, respectively.The air pretreatment in the temperature-programmed oxidation mode (20 mL/min) was carried out at temperatures up to 500 • C with an exposure time of 20 min.
Raman spectroscopy was used to confirm the defective structure of as-prepared samples (InVia confocal Raman dispersive spectrometer, Renishaw, Wharton, UK).The Leica microspore (50× objective) was employed.The excitation was ensured by a solid-state Nd:YAG laser.The wavelength, radiation power, and spectral resolution were 532 nm, 100 mW, and 2 cm −1 , respectively.The powder samples were analyzed.
X-ray photoelectron spectroscopy (XPS) was used to determine the chemical state of active components (SPECS Surface Nano Analysis GmbH, Berlin, Germany).The PHOIBOS-150-MCD-9 hemispherical analyzer and the UXC 1000 X-ray source with a double Al/Mg anode were applied.Non-monochromatized Al Kα radiation was used.The charging effect was accounted for by the u ′′′ peak with a binding energy of 916.7 eV in the Ce3d spectrum.The relative concentrations of elements in the analysis zone were determined from the integral intensities of the XPS spectra to take the photoionization cross-section into account.To subtract the background, the Shirley method was used.The data were processed using CasaXPS software.The peak approximation involved the asymmetric LF function for the Mn2p spectrum and a symmetric function summing the Gauss and Lorentz functions for other elements.

Catalytic Tests in Oxidation CO and Soot
CO oxidation tests were carried out in a tubular U-shaped quartz reactor (i.d.= 6 mm) at atmospheric pressure in the temperature-programmed heating mode ("AutoChem", Micromeritics, Norcross, GA, USA).The concentrations of CO (m/z = 28), CO 2 (m/z = 44), and O 2 (m/z = 32) were controlled using the gas mass spectrometer UGA-300 (Stanford Research Systems, Sunnyvale, CA, USA).The sample pretreatment was carried out using a H 2 flow (up to 300 • C, 1 h).The particle fraction was 0.25-0.5 mm.The reaction mixture comprised 1 vol% CO, 5 vol% O 2 , and 94 vol% He.The sample mass, the flow rate, and the heating rate were 0.1 g, 100 mL/min, and 10 • C/min, respectively.
Soot combustion activity was estimated using synchronous thermal analysis (STA 449F1 thermogravimeter, NETZSCH, Selb, Germany).Carbon black (Micromeritics, Norcross, GA, USA) was used, and 5 wt% soot was mixed with the catalyst.Tight contact mode was ensured by mixing the components in an agate mortar for 10 min.The temperature range was 100-800 • C. The heating rate was 10 • C/min in an air flow.

Chemical Composition
Real catalyst compositions were determined using the XRF method.The values obtained were similar to the nominal ones.No other impurity elements were detected either by XRF or XPS.
The analysis of the Ce/Mn and Ce/Zr ratios determined based on the XRF and XPS data (Table 1) shows that the Ce/Mn (XPS) ratio for Ce 0.9 Mn 0.1 O 2 and Ce 0.5 Mn 0.2 Zr 0.3 samples are lower than those determined by XRF, indicating a surface enrichment in Mn.The contrary is observed in Ce 0.5 Mn 0.3 Zr 0.2 O 2 , whose surface is impoverished in Mn, probably due to the effect of the surface Zr.The Ce/Zr (XPS) ratio is lower for Ce 0.6 Zr 0.4 O 2 and Ce 0.5 Mn 0.2 Zr 0.3 O 2 samples, implying a surface enrichment with Zr.This can be linked to the effect of calcination temperature [34].

Low-Temperature Nitrogen Adsorption
Table 2 shows the specific surface areas (SSAs), average pore sizes, and pore volumes measured by N 2 adsorption-desorption.The isotherm for CeO 2 (Figure 1a) is characterized by a hysteresis loop with the H3 type, which indicates the formation of lamellar particles [35].A wide pore size distribution is observed.For the ZrO 2 sample, the H2 type hysteresis loop is observed (Figure 1c), which indicates a relatively wide size distribution of pore cavities compared to the one of the necks.The pore size distribution is fairly narrow.For the MnO x sample, the H3 type hysteresis loop is observed [35], with a rather low SSA and a small pore volume being observed.Due to these characteristics, the pore size is rather difficult to determine for this sample.
Binary oxide systems consisting of CeMnO x are characterized by a hysteresis loop similar in shape to the one for CeO 2 (Figure 1a,b), as well as a fairly wide distribution of pore sizes and a mean-specific surface area.The Ce 0.9 Mn 0.1 O x sample is characterized by a pore size distribution with small and large pores with sizes of 2-4 nm and 5-50 nm, respectively (Figure 1d).The Ce 0.6 Zr 0.4 O 2 sample is characterized by a hysteresis loop similar in shape to the one for zirconium oxide.
Ternary oxide systems are characterized by an adsorption-desorption isotherm with the H2 type hysteresis loop (Figure 1c).A narrow distribution of pore diameters (3-20 nm, except for the Ce 0.65 Mn 0.2 Zr 0.15 O 2 sample featuring distribution of 3-50 nm) is observed with a slight shift in the maximum of the pore size distribution (up to 9 nm) and a significant decrease in the pore volume (Figure 1d) [34].It is noteworthy that the maximum of the distribution of the above samples is similar to the one for the Ce 0.6 Zr 0.4 O 2 sample.This fact may indicate a strong interaction of oxides at the preparation stage with the preservation of the CeO 2 -ZrO 2 structure during the incorporation of Mn n+ ions into the porous space of CeO 2 -ZrO 2 .In the case of the Ce 0.65 Mn 0.2 Zr 0.15 O 2 sample (which the Mn/Ce+Zr ratio is the same as in the Ce 0.5 Mn 0.2 Zr 0.3 O 2 sample but has a lower Zr/Ce ratio), the maximum of the pore size distribution is shifted to 18 nm.This may indicate that a low Zr/Ce ratio does not allow Mn to be uniformly distributed in the CeO 2 and CeO 2 -ZrO 2 structures.Binary oxide systems consisting of CeMnOx are characterized by a hysteresis loop similar in shape to the one for CeO2 (Figure 1a,b), as well as a fairly wide distribution of pore sizes and a mean-specific surface area.The Ce0.9Mn0.1Oxsample is characterized by a pore size distribution with small and large pores with sizes of 2-4 nm and 5-50 nm, respectively (Figure 1d).The Ce0.6Zr0.4O2sample is characterized by a hysteresis loop similar

X-ray Phase Analysis (XRD)
The phase composition of the obtained samples was studied by the XRD and analyzed by Rietveld refinement using the POWDER CELL 2.5 package program.The goodness of the fit was controlled using the low R p and R wp values.The lattice parameters (a, b, and c), microstrain value (∆d/d), and crystalline size (D XRD ) were calculated using the Williamson-Hall method [36] (Table 3).Figure 2 shows the XRD patterns.
fact may indicate a strong interaction of oxides at the preparation stage with the preservation of the CeO2-ZrO2 structure during the incorporation of Mn n+ ions into the porous space of CeO2-ZrO2.In the case of the Ce0.65Mn0.2Zr0.15O2sample (which the Mn/Ce+Zr ratio is the same as in the Ce0.5Mn0.2Zr0.3O2sample but has a lower Zr/Ce ratio), the maximum of the pore size distribution is shifted to 18 nm.This may indicate that a low Zr/Ce ratio does not allow Mn to be uniformly distributed in the CeO2 and CeO2-ZrO2 structures.

X-ray Phase Analysis (XRD)
The phase composition of the obtained samples was studied by the XRD and analyzed by Rietveld refinement using the POWDER CELL 2.5 package program.The goodness of the fit was controlled using the low Rp and Rwp values.The lattice parameters (a, b, and c), microstrain value (∆d/d), and crystalline size (DXRD) were calculated using the Williamson-Hall method [36] (Table 3).Figure 2 shows the XRD patterns.The synthesized manganese oxide is characterized by the Mn 2 O 3 cubic bixbyite phase (ICSD #76087), with a particle size of 43 nm (Figure 2, Table 3).The pristine ceria is formed by the fluorite CeO 2 phase (ICSD #28753) with a crystallite size of 28 nm.The individual ZrO 2 is characterized by the reflections related to both monoclinic (ICSD #18190) (59%) and tetragonal phases (ICSD #51051) (41%), both with the crystallite size of 15 nm.
Binary systems consisting of cerium and manganese oxides are characterized by the presence of reflections of cubic cerium oxide (Figure 2a).However, the reflections shift to higher angles with an increase in the content of manganese oxide, which indicates the substitution of Ce 4+ ions (i.r.= 0.88 Å) by those of manganese Mn n+ (i.r.0.81 Å, 0.72 Å, and 0.52 Å for Mn 2+ , Mn 3+ , and Mn 4+ , respectively) in the CeO 2 network with the formation of a solid solution with a fluorite-type (F-type) structure.According to the Vegard's law, the compositions of the solid solutions were calculated as Ce 0.94 Mn 0.06 O 2−δ, Ce 0.9 Mn 0.1 O 2−δ , and Ce 0.89 Mn 0.11 O 2−δ for Ce 0.9 Mn 0.1 O 2 , Ce 0.8 Mn 0.2 O 2 , and Ce 0.67 Mn 0.33 O 2 , respectively.These calculated compositions of the solid solutions indicate that only a small part of manganese is included into the structure, and the rest is segregated in the form of highly dispersed Mn 2 O 3 and/or Mn 3 O 4 .
The Ce 0.6 Zr 0.4 O 2 sample is characterized by the presence of reflections from the CeO 2 phase and a shoulder from the cubic zirconia phase (ICSD #53998) (Figure 2b).The Rietveld refinement of the Ce 0.6 Zr 0.4 O 2 powder pattern reveals defective Ce 1−x Zr x O 2−δ (47%) and Ce x Zr 1−x O 2−δ (53%) solid solutions with a cubic structure.However, the formation of the Ce x Zr 1−x O 2 solid solution with a tetragonal structure cannot be ruled out due to the comparable R p and R wp values.The solid solution compositions evaluated according to Vegard's law are Ce 0.92 Zr 0.08 O 2−δ (F-type) and Ce 0.41 Zr 0.59 O 2−δ (F-type) or Ce 0.18 Zr 0.82 O 2−δ (tetragonal type).For Ce x Zr 1−x O 2 with 0.3 ≤ x ≤ 0.65, the formation of the most stable tetragonal phase was shown (see Ref. [37] and references within).
Ternary oxide systems are characterized by the reflections of the cubic ceria with a broad shoulder at ~30 • and a small one at 32.9 • , which can be associated with the contribution of the zirconia phase and partial segregation of the α-Mn 2 O 3 (ICSD #9090) on the surface, respectively (Figure 2b,c).The significant decrease in the lattice parameter of the CeO 2 phase for the CeO 2 -ZrO 2 -MnO x ternary systems relative to the reference samples of CeO 2 , CeO 2 -MnO x , and CeO 2 -ZrO 2 (See Figure 3a) is due to the introduction of smaller Mn 3+ and Mn 4+ ions (i.r. are 0.70 and 0.52 Å, respectively) and Zr 4+ (i.r. is 0.82 Å) compared to the Ce 4+ ion (i.r. is 0.88 Å) to form the solid solutions.The Ce 0.5 Mn 0.2 Zr 0.  The Ce 4+ substitution hinders the crystallite growth for the CeO2-ZrO2-MnOx samples (in the range of 7-21 nm), and this results in a microstrain increase (Δd/d = 0.0043-0.0094)due to the additional number of oxygen-anion vacancies formed in the Ce1−(x+y)ZrxMnyO2−δ solid solutions.When Mn dopant is added to CeO2, the crystallite size decreases slightly (see Figure 3b).Moreover, upon further increase in the Mn/Ce ratio, the crystallite size remains practically unchanged.When Zr dopant is added to CeO2, a slight decrease in the crystallite size is observed (from 28 nm to up to 26 nm).The simultaneous addition of Mn The Ce 4+ substitution hinders the crystallite growth for the CeO 2 -ZrO 2 -MnO x samples (in the range of 7-21 nm), and this results in a microstrain increase (∆d/d = 0.0043-0.0094)due to the additional number of oxygen-anion vacancies formed in the Ce 1−(x+y) Zr x Mn y O 2−δ solid solutions.When Mn dopant is added to CeO 2 , the crystallite size decreases slightly (see Figure 3b).Moreover, upon further increase in the Mn/Ce ratio, the crystallite size remains practically unchanged.When Zr dopant is added to CeO 2 , a slight decrease in the crystallite size is observed (from 28 nm to up to 26 nm).The simultaneous addition of Mn and Zr dopants to CeO 2 leads to a significant reduction in the crystallite sizes.It is worth noting the significant reduction in the crystallite size (up to 7 nm) for the Ce 0.5 Mn 0.2 Zr 0.3 O 2 sample, which is also characterized by the most significant decrease in the ceria lattice parameter.

Raman Spectroscopy
The Raman spectrum for the CeO 2 sample taken under laser irradiation at 532 nm (Figure 4a) is characterized by a strong Raman band at 464 cm −1 , which is assigned to the F 2g vibrational active mode in the fluorite-type CeO 2 structure.A weaker band at 1150 cm −1 is an overtone of the F 2g mode [38].A band at ~594 cm −1 can be assigned to the presence of oxygen vacancies.For the MnO x sample, the vibrations are related to both Mn 3 O 4 and Mn 2 O 3 [39].When Mn is added to CeO2, the F2g band position in the spectra for CeMnOx samples slightly shifts towards the lower wavenumber from 464 to 460 cm −1 , and as the Mn content gradually increases, the F2g band decreases in intensity and becomes broader.The shifts and broadening of the F2g band are connected with the presence of surface oxygen vacancies, a loss of regularity in the crystalline network, and the ∆d/d microstrain effect [40][41][42].
For the Ce0.9Mn0.1O2sample, the appearance of a broad band at ~256 cm −1 is observed, which can be attributed to the displacement of atoms from the ideal positions of the fluorite lattice [43,44].A broad asymmetric band in the range of 540-718 cm −1 can be deconvoluted into the 590 cm −1 band attributed to defect areas involving the Mn n+ cation in an 8-fold O 2− coordination, without any vacancies [45,46], and/or O 2− interstitial oxygen defects of the Frenkel type in CeO2 [47][48][49][50].The bands at 649 and 690 cm −1 can be attributed to the presence of α-Mn2O3 [51,52].
For the Ce0.8Mn0.2O2and Ce0.67Mn0.33O2samples, the bands at 192, 308, 649, and 690 cm −1 become more intense, and they are associated with the Mn-O vibrations in the α-Mn2O3 [53].The appearance of weak bands at 310 and 366 cm −1 along with a strong band at 649 cm −1 , which is clearly overlapped with one of the bands related to α-Mn2O3, is associated with the vibrational stretching of the Mn-O bond towards Mn 2+ ions in the −1 When Mn is added to CeO 2 , the F 2g band position in the spectra for CeMnO x samples slightly shifts towards the lower wavenumber from 464 to 460 cm −1 , and as the Mn content gradually increases, the F 2g band decreases in intensity and becomes broader.The shifts and broadening of the F 2g band are connected with the presence of surface oxygen vacancies, a loss of regularity in the crystalline network, and the ∆d/d microstrain effect [40][41][42].
For the Ce 0.9 Mn 0.1 O 2 sample, the appearance of a broad band at ~256 cm −1 is observed, which can be attributed to the displacement of atoms from the ideal positions of the fluorite lattice [43,44].A broad asymmetric band in the range of 540-718 cm −1 can be deconvoluted into the 590 cm −1 band attributed to defect areas involving the Mn n+ cation in an 8-fold O 2− coordination, without any vacancies [45,46], and/or O 2− interstitial oxygen defects of the Frenkel type in CeO 2 [47][48][49][50].The bands at 649 and 690 cm −1 can be attributed to the presence of α-Mn 2 O 3 [51,52].
For the Ce 0.8 Mn 0.2 O 2 and Ce 0.67 Mn 0.33 O 2 samples, the bands at 192, 308, 649, and 690 cm −1 become more intense, and they are associated with the Mn-O vibrations in the α-Mn 2 O 3 [53].The appearance of weak bands at 310 and 366 cm −1 along with a strong band at 649 cm −1 , which is clearly overlapped with one of the bands related to α-Mn 2 O 3 , is associated with the vibrational stretching of the Mn-O bond towards Mn 2+ ions in the spinel structure of Mn 3 O 4 [53].In addition, the appearance of the band at 537 cm −1 , which is associated with defect areas including O 2− vacancies and is observed when the dopant cations in a M 3+ state are introduced into the CeO 2 lattice [45,54], and as mentioned above, the shoulder at 590 cm −1 , are associated with the defect areas in the 8-fold coordination of O 2− .Thus, for the CeMnO x samples, the introduction of Mn ions, mainly Mn 3+ , into the CeO 2 lattice is observed, while the formation of α-Mn 2 O 3 and Mn 3 O 4 cannot be excluded, which is consistent with the XRD data.
The spectrum for the binary oxide Ce 0.6 Zr 0.4 O 2 (Figure 4b) features the F 2g band and the broad weak bands at 300 and 612 cm −1 associated with the tetragonal substitution of oxygen atoms from the ideal fluorite lattice after the Zr introduction and the formation of oxygen vacancies, respectively [37,55].
For ternary oxide samples, the F 2g band is shifted towards the higher wavenumber from 464 to 476 cm −1 .For ternary samples, the bands related to Mn 3 O 4 and Mn 2 O 3 appear.The asymmetry of the band at 649 cm −1 and the presence of a band at 310 cm −1 may be a consequence of the overlapping of the bands of MnO x species and the presence of oxygen vacancy defects in CeO 2 .A shoulder at 590 cm −1 is shifted to ~610 cm −1 .According to Ref. [45], the band at ~560 cm −1 related to the presence of oxygen vacancies can be due to the different oxidation state of the dopant compared to that of Ce 4+ .The band at ~600 cm −1 is because of the different ionic radius of the dopant ions compared to that of Ce 4+ .The bands above 650 cm −1 can be assigned to the extrinsic defects in ceria as well as aliovalent defects caused by the presence of Mn-related phases where Mn exists in higher oxidation states [56,57].It is noteworthy that the increase in the Mn + Zr fraction in the series of samples Ce 0.65 Mn 0.2 Zr 0.15 -Ce 0.5 Mn 0.3 Zr 0.2 -Ce 0.5 Mn 0.2 Zr 0.3 increased the intensity of the bands at 536, 584, and 616 cm −1 .Thus, the increase in the amount of substituting metals in the CeO 2 structure results in an increase in the concentration of internal and external defects.

Temperature-Programmed Reduction in Hydrogen (TPR-H 2 )
The reduction profile for the MnO x sample is represented by a peak centered at 348 • C (Figure 5), which corresponds to the Mn 2 O 3 reduction to Mn 3 O 4 , and a peak at 463 • C that corresponds to the Mn 3 O 4 reduction to MnO [58,59].The reduction profile for CeO 2 is characterized by the presence of two temperature regions of hydrogen consumption, namely, the consumption peak in the temperature region of 300-600 • C, at which the surface of ceria nanoparticles is reduced, and at temperatures above 700 • C, at which the bulk of ceria nanoparticles is reduced [60,61].The main reduction process for ZrO 2 occurs above 1000 • C (not shown for clarity) [62].
The profiles for the CeMnO x composites are characterized by two regions associated with the reduction of the highly dispersed Mn 2 O 3 species (200-350 • C) and the reduction of Mn 3 O 4 particles along with the surface CeO 2 (350-650 • C) [63,64].Both peak maxima are shifted progressively to lower temperatures with the increase in the Ce content and a decrease in the Mn content, indicating a progressively stronger interaction between the two oxides.It is noteworthy that the intensity ratio between the two peaks differs when the manganese to cerium content varies.This is due to both different phase ratios of Mn 2 O 3 /Mn 3 O 4 and the strength of the MnO x interaction with CeO 2 .
The Zr-containing catalysts (Ce 0.6 Zr 0.4 O 2 and CeMnZrO x ) are reduced more easily due to the simultaneous reduction of the surface and bulk ceria caused by the introduction of Zr and/or Mn.The progressive substitution of Ce by Zr shifts the maxima towards higher temperatures, i.e., decreases reducibility.This can be because the CeO 2 -ZrO 2 interaction reduces the CeO 2 -MnO x interactions.
that corresponds to the Mn3O4 reduction to MnO [58,59].The reduction profile for CeO2 is characterized by the presence of two temperature regions of hydrogen consumption, namely, the consumption peak in the temperature region of 300-600 °C, at which the surface of ceria nanoparticles is reduced, and at temperatures above 700 °C, at which the bulk of ceria nanoparticles is reduced [60,61].The main reduction process for ZrO2 occurs above 1000 °C (not shown for clarity) [62].The profiles for the CeMnOx composites are characterized by two regions associated with the reduction of the highly dispersed Mn2O3 species (200-350 °C) and the reduction of Mn3O4 particles along with the surface CeO2 (350-650 °C) [63,64].Both peak maxima are shifted progressively to lower temperatures with the increase in the Ce content and a decrease in the Mn content, indicating a progressively stronger interaction between the two oxides.It is noteworthy that the intensity ratio between the two peaks differs when the manganese to cerium content varies.This is due to both different phase ratios of Mn2O3/Mn3O4 and the strength of the MnOx interaction with CeO2.Table 4 presents data on the amount of H 2 consumed.For binary and ternary oxide systems, there is an increase in the amount of consumed hydrogen, which indicates an increase in the reactivity for these systems.The increase in both total H 2 consumption (column Σ) and hydrogen consumption at low temperatures (150-500 • C) due to the simultaneous reduction of highly dispersed MnO x particles and the surface of ceria particles indicates a positive effect of both Mn and Zr on reducibility of the CeO 2 phase.

XPS
To study the states of elements on the sample surfaces, the XPS method was employed.Figure 6 and Table 5 show the XPS Ce3d, Mn2p, Zr3d, and O1s spectra and the relative concentrations (atomic ratios) of elements in the subsurface layer, respectively.The normalization of spectra to the total integral intensity (for the Ce3d, Mn2p, and Zr3d spectra) of the corresponding spectra for supports was carried out.The analysis of the spectra is based on their deconvolution.
The Ce 3+ fraction increases as the Mn content decreases in the series of samples MnO x -Ce 0.67 Mn 0.33 O 2 -Ce 0.5 Mn 0.3 Zr 0.2 O 2 -Ce 0.5 Mn 0.2 Zr 0.3 O 2 -Ce 0.9 Mn 0.1 O 2 -CeO 2 .The Ce 0.67 Mn 0.33 O 2 sample features the lowest value.The samples feature rather low fractions of the Mn 4+ state, the fraction of the Mn 3+ state increases and the one of the Mn 2+ state decreases, with the highest Mn 3+ and lowest Mn 2+ fractions being reached over Ce 0.9 Mn 0.1 O 2 , which also features the highest fraction of active oxygen.The lowest fraction of active oxygen is shown by the Ce 0.5 Mn 0.3 Zr 0.2 O 2 sample.A higher fraction of the Mn 3+ state in the samples is usually related to a lower fraction of the O 2− state.

TEM
Figure 7 shows the TEM images for the synthesized CeO 2 , binary, and ternary oxides.The CeO 2 sample is characterized by the presence of well-crystallized particles with a size of 10-40 nm, which is consistent with the XRD data.Such a structure is caused by the hydrothermal treatment of the oxide during the preparation.The structure of binary Ce 0.9 Mn 0.1 O 2 oxide is significantly different from the one of the CeO 2 sample.Loose aggregates of particles with the size of a few nanometers are observed.Thus, the addition of even small amounts of Mn into the CeO 2 leads to significant changes in the sample structure.The ternary Ce 0.5 Mn 0.3 Zr 0.2 O 2 oxide is also characterized by the presence of loose aggregates of small particles.The TEM images show that the surface of the samples is determined by the surface of the primary particles, and the growth of the specific surface area for binary and ternary oxides is a result of the decrease in the particle size.Additionally, the porous structure of the sample is represented by the interparticle space and the mesoporous structure of the sample can be well observed in the presented TEM images.To determine the distribution of elements in the binary and ternary oxides, HR TEM and EDX analysis were applied.For both Ce0.9Mn0.1O2and Ce0.5Mn0.3Zr0.2O2,only the CeO2 phase is determined, while the interlayer distances for the ZrO2 or MnOx phases are not identified.Figure 8a shows the HR-STEM image for the Ce0.9Mn0.1O2sample, and wellcrystallized particles with sizes of 4-10 nm and the interlayer distance of the CeO2 phase may be observed.The maps of Ce and Mn distribution (Figure 8b) show the uniform Mn distribution in the sample.Mn can be stabilized as highly dispersed species on the CeO2 surface and/or as Mn ions incorporated into the CeO2 structure.Separated MnOx particles are not observed, which is consistent with the XRD data.Figure 8c shows the distribution of metals for ternary Ce0.5Mn0.3Zr0.2O2oxide.The Ce and Zr distributions demonstrate that the Ce-rich and Zr-rich areas can be highlighted.This is consistent with the XRD results because two Ce1−xZrxO2 and CexZr1−xO2 phases are observed for the Ce0.6Zr0.4O2sample and ternary CeO2-ZrO2-MnOx samples.The Ce-rich phase is predominant, and it is also consistent with the XRD data.The Mn distribution is relatively uniform, and the stabilization of the MnOx clusters on the surface and its incorporation into both Ce-rich and Zr-rich mixed oxide cannot be excluded.The Mnrelated species seem to be localized mostly near the areas rich in ceria and less near the Zr-rich areas.To determine the distribution of elements in the binary and ternary oxides, HR TEM and EDX analysis were applied.For both Ce 0.9 Mn 0.1 O 2 and Ce 0.5 Mn 0.3 Zr 0.2 O 2 , only the CeO 2 phase is determined, while the interlayer distances for the ZrO 2 or MnO x phases are not identified.Figure 8a shows the HR-STEM image for the Ce 0.9 Mn 0.1 O 2 sample, and well-crystallized particles with sizes of 4-10 nm and the interlayer distance of the CeO 2 phase may be observed.The maps of Ce and Mn distribution (Figure 8b) show the uniform Mn distribution in the sample.Mn can be stabilized as highly dispersed species on the CeO 2 surface and/or as Mn ions incorporated into the CeO 2 structure.Separated MnO x particles are not observed, which is consistent with the XRD data.To determine the distribution of elements in the binary and ternary oxides, HR TEM and EDX analysis were applied.For both Ce0.9Mn0.1O2and Ce0.5Mn0.3Zr0.2O2,only the CeO2 phase is determined, while the interlayer distances for the ZrO2 or MnOx phases are not identified.Figure 8a shows the HR-STEM image for the Ce0.9Mn0.1O2sample, and wellcrystallized particles with sizes of 4-10 nm and the interlayer distance of the CeO2 phase may be observed.The maps of Ce and Mn distribution (Figure 8b) show the uniform Mn distribution in the sample.Mn can be stabilized as highly dispersed species on the CeO2 surface and/or as Mn ions incorporated into the CeO2 structure.Separated MnOx particles are not observed, which is consistent with the XRD data.Figure 8c shows the distribution of metals for ternary Ce0.5Mn0.3Zr0.2O2oxide.The Ce and Zr distributions demonstrate that the Ce-rich and Zr-rich areas can be highlighted.This is consistent with the XRD results because two Ce1−xZrxO2 and CexZr1−xO2 phases are observed for the Ce0.6Zr0.4O2sample and ternary CeO2-ZrO2-MnOx samples.The Ce-rich phase is predominant, and it is also consistent with the XRD data.The Mn distribution is relatively uniform, and the stabilization of the MnOx clusters on the surface and its incorporation into both Ce-rich and Zr-rich mixed oxide cannot be excluded.The Mnrelated species seem to be localized mostly near the areas rich in ceria and less near the Zr-rich areas.The Ce-rich phase is predominant, and it is also consistent with the XRD data.The Mn distribution is relatively uniform, and the stabilization of the MnO x clusters on the surface and its incorporation into both Ce-rich and Zr-rich mixed oxide cannot be excluded.The Mnrelated species seem to be localized mostly near the areas rich in ceria and less near the Zr-rich areas.

Catalytic Activity in CO and Soot Oxidation
The catalytic activity of the obtained samples in CO oxidation was studied (Figure 9).For all samples, with the exception of cerium and zirconium oxides, 100% conversion was achieved at temperatures below 450 • C. Binary systems comprising cerium and manganese oxides are characterized by the onset of conversion at ~100 • C. As the manganese content increases in these samples, a decrease in the catalytic activity is observed.The most active is Ce 0.9 Mn 0.1 O 2 , for which 50% and 98% conversion is achieved at temperatures of 174 • C and 298 • C, respectively, with an activation energy of 49.0 kJ/mol (Figure 9c).Ternary oxide systems typically begin conversion also at ~100 • C. The Ce 0.55 Mn 0.08 Zr 0.37 O 2 sample is the least active: 50% and 98% conversions are achieved at 261 • C and 414 • C, respectively.The Ce 0.5 Mn 0.2 Zr 0.3 O 2 sample shows the highest activity: T 50% and T 98% are 184 • C and 260 • C, respectively, and the activation energy is 48.2 kJ/mol (Figure 9d).

Catalytic Activity in CO and Soot Oxidation
The catalytic activity of the obtained samples in CO oxidation was studied (Figure 9).For all samples, with the exception of cerium and zirconium oxides, 100% conversion was achieved at temperatures below 450 °C.Binary systems comprising cerium and manganese oxides are characterized by the onset of conversion at ~100 °C.As the manganese content increases in these samples, a decrease in the catalytic activity is observed.The most active is Ce0.9Mn0.1O2,for which 50% and 98% conversion is achieved at temperatures of 174 °C and 298 °C, respectively, with an activation energy of 49.0 kJ/mol (Figure 9c).Ternary oxide systems typically begin conversion also at ~100 °C.The Ce0.55Mn0.08Zr0.37O2sample is the least active: 50% and 98% conversions are achieved at 261 °C and 414 °C, respectively.The Ce0.5Mn0.2Zr0.3O2sample shows the highest activity: T50% and T98% are 184 °C and 260 °C, respectively, and the activation energy is 48.2 kJ/mol (Figure 9d).Simultaneous thermal analysis was used to determine the catalytic activity of oxide catalysts in soot oxidation.Figure 10 shows the TG and DSC curves obtained from soot oxidation simulations using the resulting catalysts.Non-catalytic soot oxidation is observed in the temperature range 550-700 °C with Tmax at 661 °C.The presence of ZrO2 leads to a slight increase in soot oxidation activity (Tmax = 613 °C) (not shown).The presence of cerium and manganese oxides decreases the maximum temperature to 512 °C Simultaneous thermal analysis was used to determine the catalytic activity of oxide catalysts in soot oxidation.Figure 10 shows the TG and DSC curves obtained from soot oxidation simulations using the resulting catalysts.Non-catalytic soot oxidation is observed in the temperature range 550-700 • C with T max at 661 • C. The presence of ZrO 2 leads to a slight increase in soot oxidation activity (T max = 613 • C) (not shown).The presence of cerium and manganese oxides decreases the maximum temperature to 512 • C and 515 • C, respectively.The presence of binary and ternary oxide systems shifts the reaction temperature range to 350-600 • C (i.e., it decreases the reaction temperature range by 100-170 • C).Samples Ce 0.8 Mn 0.2 O 2 and Ce 0.65 Mn 0.2 Zr 0.15 O 2 showed the highest activity: the maximum conversion temperatures for these samples were 422 and 439 • C, respectively.Comparisons of soot combustion and CO oxidation performances of the prepared samples and counterparts presented in the literature are presented in Tables S1 and S2 (Supplementary materials).The materials obtained show comparable activity.and 515 °C, respectively.The presence of binary and ternary oxide systems shifts the reaction temperature range to 350-600 °C (i.e., it decreases the reaction temperature range by 100-170 °C).Samples Ce0.8Mn0.2O2and Ce0.65Mn0.2Zr0.15O2showed the highest activity: the maximum conversion temperatures for these samples were 422 and 439 °C, respectively.Comparisons of soot combustion and CO oxidation performances of the prepared samples and counterparts presented in the literature are presented in Tables S1  and S2 (Supplementary materials).The materials obtained show comparable activity.

Discussion
The results of the catalytic studies show that the binary Ce-Mn catalysts demonstrate higher performance in both CO oxidation (Ce0.9Mn0.1O2)and soot combustion (Ce0.8Mn0.2O2)reactions, which are closely followed by the ternary oxide catalysts.The key features of the Ce0.9Mn0.1O2sample, which also shows a rather low Tmax value in soot combustion, include the following: (1) the presence of both small (2-4 nm) and large (5-50 nm) pores; (2) the formation of Ce0.94Mn0.06O2−δsolid solution where the Mn component can both be included into the ceria structure and form several highly dispersed surface species related to Mn oxides; (3) the lowest value of H2 consumption in the temperature range of 150-500 °C (related to the reduction of dispersed MnOx and surface CeO2) and the highest one at temperatures above 650 °C (related to bulk CeO2); (4) the highest fractions of Mn 3+ state and active oxygen; and (5) the predominant presence of intrinsic and Frenkel-type oxygen vacancies.Sample Ce0.8Mn0.2O2features include the following: (1) high pore volume (0.290 cm 3 /g, the largest in the whole series) and large pore sizes (24.4 nm), with the latter being predominantly formed; (2) the presence of the solid solution with the estimated composition of Ce0.9Mn0.1O2−δwhere the content of highly dispersed surface Mn oxides species is even higher as compared to Ce0.9Mn0.1O2;(3) higher impact of microstrains as compared to Ce0.9Mn0.1O2;and (4) higher concentration of extrinsic MnOx-related defects (Raman spectrum).The interplays of the mentioned features can be the reasons for the observed activity.
The ternary Ce0.5Mn0.2Zr0.3O2features a higher specific surface and narrower particle size distribution as compared to the mentioned binary samples.The sample shows high H2 consumption in the temperature ranges below 500 °C and above 650 °C.The sample is characterized by a higher fraction of high oxidation states of Mn (i.e., 3+, 4+).The formation of solid solutions with different compositions, say, Ce1−xMnxO2 and Ce1−xZrxO2, and their interplay on the surface can also be the reason for the observed performance.

Discussion
The results of the catalytic studies show that the binary Ce-Mn catalysts demonstrate higher performance in both CO oxidation (Ce 0.9 Mn 0.1 O 2 ) and soot combustion (Ce 0.8 Mn 0.2 O 2 ) reactions, which are closely followed by the ternary oxide catalysts.The key features of the Ce 0.9 Mn 0.1 O 2 sample, which also shows a rather low T max value in soot combustion, include the following: (1) the presence of both small (2-4 nm) and large (5-50 nm) pores; (2) the formation of Ce 0.94 Mn 0.06 O 2−δ solid solution where the Mn component can both be included into the ceria structure and form several highly dispersed surface species related to Mn oxides; (3) the lowest value of H 2 consumption in the temperature range of 150-500 • C (related to the reduction of dispersed MnO x and surface CeO 2 ) and the highest one at temperatures above 650 • C (related to bulk CeO 2 ); (4) the highest fractions of Mn 3+ state and active oxygen; and (5) the predominant presence of intrinsic and Frenkel-type oxygen vacancies.Sample Ce 0.8 Mn 0.2 O 2 features include the following: (1) high pore volume (0.290 cm 3 /g, the largest in the whole series) and large pore sizes (24.4 nm), with the latter being predominantly formed; (2) the presence of the solid solution with the estimated composition of Ce 0.9 Mn 0.1 O 2−δ where the content of highly dispersed surface Mn oxides species is even higher as compared to Ce 0.9 Mn 0.1 O 2 ; (3) higher impact of microstrains as compared to Ce 0.9 Mn 0.1 O 2 ; and (4) higher concentration of extrinsic MnO x -related defects (Raman spectrum).The interplays of the mentioned features can be the reasons for the observed activity.
The ternary Ce 0.5 Mn 0.2 Zr 0.3 O 2 features a higher specific surface and narrower particle size distribution as compared to the mentioned binary samples.The sample shows high H 2 consumption in the temperature ranges below 500 • C and above 650 • C. The sample is characterized by a higher fraction of high oxidation states of Mn (i.e., 3+, 4+).The formation of solid solutions with different compositions, say, Ce 1−x Mn x O 2 and Ce 1−x Zr x O 2 , and their interplay on the surface can also be the reason for the observed performance.
The mentioned features imply that the surface of the binary Ce-Mn samples with low Mn content is represented by a "cocktail" of active sites (e.g., oxygen vacancies, highly dispersed MnO x species with various compositions, CeO 2 -MnO x interfaces, etc.) that coexist, acting separately and/or synergistically and allowing the samples to exhibit high activity in conversion of both CO and soot.The observed differences in performance can be attributed to the effect of Mn loading, where lower Mn content ensures the formation of a balanced mixture of MnO x species, while higher Mn content probably results in the formation of the Mn-related species that partially block other active sites to reduce overall performance.The formation of interfaces between various Mn-related species that are less catalytically active in said reactions cannot be excluded.

Conclusions
In the present work, a series of CTAB-templated CeO 2 -ZrO 2 -MnO x catalysts was prepared by coprecipitation of the corresponding precursors.The addition of Mn and/or Zr dopant to ceria improved its performance in both CO oxidation and soot combustion.A complex of physical-chemical methods (low-temperature N 2 sorption, XRD, TPR-H 2 , Raman, HR TEM, XPS) allowed revealing the structural and compositional peculiarities of the catalysts that caused the observed performance.Binary Ce-Mn catalysts demonstrated higher performance, and were closely followed by the ternary oxide catalysts.The reason was linked with the formation of several types of active sites, namely, highly dispersed MnO x species, interfaces, and oxygen vacancies that can act individually and/or synergistically to allow said samples to be active in both reactions.Future work can include revealing the intrinsic activity and mechanisms of action of such species in environmental catalytic reactions.

Figure 2 .Table 3 .
Figure 2. XRD patterns for the prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides, (c) excerpt from patterns (b) Table3.Phase composition of samples and structural parameters, microstrains, and crystallite sizes of the crystalline phases identified from XRD data.

Figure 2 .
Figure 2. XRD patterns for the prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides, (c) excerpt from patterns (b).
3 O 2 sample is characterized by the lowest lattice parameter (5.331 Å), which can indicate the highest amount of Mn and Zr ions incorporated into the CeO 2 structure.Nanomaterials 2023, 13, x FOR PEER REVIEW 8 of 21

Figure 3 .
Figure 3. Relationships between cell parameter of CeO2 (a) and crystallite size (b) and fraction of substitution metals.

Figure 3 .
Figure 3. Relationships between cell parameter of CeO 2 (a) and crystallite size (b) and fraction of substitution metals.

Figure 4 .
Figure 4. Raman spectra for the prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Figure 4 .
Figure 4. Raman spectra for the prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Figure 5 .
Figure 5. TPR profiles for the obtained samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Figure 5 .
Figure 5. TPR profiles for the obtained samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Figure
Figure 8c shows the distribution of metals for ternary Ce 0.5 Mn 0.3 Zr 0.2 O 2 oxide.The Ce and Zr distributions demonstrate that the Ce-rich and Zr-rich areas can be highlighted.This is consistent with the XRD results because two Ce 1−x Zr x O 2 and Ce x Zr 1−x O 2 phases are observed for the Ce 0.6 Zr 0.4 O 2 sample and ternary CeO 2 -ZrO 2 -MnO x samples.The Ce-rich phase is predominant, and it is also consistent with the XRD data.The Mn distribution is relatively uniform, and the stabilization of the MnO x clusters on the surface and its incorporation into both Ce-rich and Zr-rich mixed oxide cannot be excluded.The Mnrelated species seem to be localized mostly near the areas rich in ceria and less near the Zr-rich areas.

Figure 9 .
Figure 9. Catalytic properties of prepared systems in CO oxidation: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides; (c, d) corresponding activation energies.

Figure 9 .
Figure 9. Catalytic properties of prepared systems in CO oxidation: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides; (c,d) corresponding activation energies.

Figure 10 .
Figure 10.Soot combustion activity of prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Figure 10 .
Figure 10.Soot combustion activity of prepared samples: (a) individual and Ce-Mn oxides, (b) Ce-Zr and ternary oxides.

Table 2 .
Textural properties of samples according to low-temperature nitrogen adsorption data.

Table 3 .
Phase composition of samples and structural parameters, microstrains, and crystallite sizes of the crystalline phases identified from XRD data.Sample Phase Structural Parameters a, Å b, Å c, Å ∆d/d × 10 −3

Table 4 .
The H 2 consumption in TPR for the prepared catalysts.

Table 5 .
Relative concentrations of elements in the near-surface layer of the samples.

Table 5 .
Relative concentrations of elements in the near-surface layer of the samples.