Impact of Ce/Zr Ratio in the Nanostructured Ceria and Zirconia Composites on the Selective CO2 Adsorption

High surface-area, mesoporous CeO2, ZrO2, and Ce-Zr composite nanoparticles were developed using the hydrothermal template-assisted synthesis method. Samples were characterized using XRD, N2 physisorption, TEM, XPS, and FT-IR spectroscopic methods. The CO2 adsorption ability of the obtained materials was tested under dynamic and equilibrium conditions. A high CO2 adsorption capacity in CO2/N2 flow or CO2/N2/H2O was determined for all studied adsorbents depending on their composition flow. A higher CO2 adsorption was registered for Ce-Zr composite nanomaterials due to the presence of strong O2− base sites and enriched surface oxygen species. The role of the Ce/Zr ratio is the process of the formation of highly active and selective adsorption sites is discussed. The calculated heat of adsorption revealed the processes of chemisorption and physisorption. Experimental data could be appropriately described by the Yoon–Nelson kinetic model. The composites reused in five adsorption/desorption cycles showed a high stability with a slight decrease in CO2 adsorption capacities in dry flow and in the presence of water vapor.


Introduction
Today's rapid economic development contributes to a significant increase in the use of energy obtained from conventional fossil fuels (coal, oil, and natural gas) [1,2]. The massive use of fossil fuels leads to adverse effects on the environment, namely global warming, and critical climate change on earth. Global warming is the result of the increasing concentration of greenhouse gases and in particular, carbon dioxide (CO 2 ), the main anthropogenic greenhouse gas [3]. The significant increase in CO 2 concentration in the atmosphere from 340 ppm in 1980 to 408 ppm in 2019 leads to a negative impact on the environment [4]. Nowadays, the high concentration of carbon dioxide in the atmosphere can be reduced by increasing the share of renewable carbon sources or reducing CO 2 emissions with anthropogenic origin [1]. For the elimination of greenhouse gases, much attention has been focused on improving the activity of absorbents for the capture of CO 2 . Various CO 2 capture technologies are available at present including adsorption, absorption (chemical and physical absorptions), and different membrane technologies [5][6][7][8][9][10]. The development of efficient materials for these processes is of great importance. Therefore, the CO 2 adsorption efficiency can be improved by selecting an appropriate material as an The synthesis of CeO 2 and ZrO 2 was performed using the following procedure; 5.41 g cerium(III) chloride heptahydrate (CeCl 3 ·7H 2 O) or 3.75 g zirconium chloride (ZrCl 4 ), respectively, were completely dissolved in 50 mL distilled water at 50 • C and stirred for 45 min. The solution was mixed with a solution containing 10 mL 25 wt.% ammonia (NH 3 ) in 10 mL H 2 O. The homogeneous solution was transferred to a Teflon vessel jacketed in a stainless-steel autoclave and heated in an oven at 100 • C for 24 h under static conditions. The white precipitate product was filtered, washed with distilled water, and dried at 40 • C overnight. The synthesis of (1) Ce 0.5 Zr 0.5 , (2) Ce 0.67 Zr 0.33, and (3) Ce 0.33 Zr 0.67 was performed using the following procedure. A solution of (1) 1.87 g (2) 1.25 g (3) 2.5 zirconium chloride (ZrCl 4 ) in 50 mL distilled water and (1) 2.71 g, (2) 3.61 g, and (3) 1.81 g cerium(III) chloride heptahydrate (CeCl 3 ·7H 2 O) in 50 mL water were added at room temperature to a solution of 6 g CTAB in 100 mL water and stirred for 45 min. The above solution was mixed with a solution containing 10 mL ammonia 25 wt.% (NH 3 ) in 10 mL water. The homogeneous solution was transferred to a Teflon vessel jacketed in a stainless-steel autoclave. Subsequently, the temperature was increased to 100 • C and aged for 24 h under static conditions. The white precipitate product was filtered, washed with distilled water, and dried at 40 • C overnight. The template was removed by calcination at 300 • C for 6 h. An adsorbent with Ce 0.5 Zr 0.5 composition was prepared with the same procedure, but the template was removed by extraction method. The filtered synthesis product was extracted with ethanol at 80 • C for 5 h and dried under vacuum at 40 • C overnight. The latter one is denoted as ext.Ce 0.5 Zr 0.5.

Characterization
Powder X-ray diffraction patterns were collected on Bruker D8 Advance diffractometer equipped with Cu Kα radiation and LynxEye detector. Phases were identified by Diffrac.EVA v.4 using ICDD-PDF-2 (2021) database. For the estimation of the mean crystallite size, the broadening of the diffraction lines was analyzed by means of whole powder pattern profile fitting using the program Topas v.4.2. The corrections for the instrumental broadening were included by the fundamental approach technique implemented in the program (accounting for the real elements on the beam path).
Determination of the specific surface area and pore size distribution was performed by low-temperature nitrogen adsorption. The adsorption and desorption isotherms of nitrogen at −196 • C were determined in the pressure range of p/p 0 = 0.001-1, using an advanced micropore size and chemisorption analyzer "AUTOSORB iQ-MP/AG" (Anton Paar GmbH, Graz, Austria). Before every measurement, the samples were degassed at 80 • C for 16 h.
The morphology, the phase composition, and the microstructure of the samples were analyzed by Transmission Electron Microscopy (TEM) using High Resolution Transmission Electron Microscope HRTEM JEOL JEM 2100 (JEOL Ltd., Tokio, Japan). All measurements were held at 200 kV accelerating voltage. Match software (Version 3.13). Crystal Impact GbR, Bonn, Germany) with Crystallography Open Database (COD) was applied for the phase identification of the samples. The measurements of the nanoparticles diameters for the statistical analysis were implemented by means of Image J freeware (National Institutes of Health, Bethesda, MD, USA).
X-ray photoelectron spectroscopy (XPS) spectra have been obtained using a Phoibos 100 (SPECS) based X-ray photoelectron spectrometer operating in Fixed Analyzer Transmission (FAT) mode with 5-channel MCD-5 detector (SPECS). The spectrometer is equipped with a non-monochromatic X-ray source XR50 with double Al/Mg anode operated under 12 kV (200 W), the core-level spectra were measured using Al Kα radiation (hν = 1486.6 eV), and no flood gun was used. The samples were placed on double-sided carbon conductive adhesive tape. The spectra analyses were made in CasaXPS software with the use of the Shirley background model and built-in RSF for composition calculations. The binding energy was adjusted on a base of valence band XPS spectra.
In situ FT-IR spectra were measured by a Nicolet Compact 6700 spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Thin, self-supported wafers with 2 cm 2 surface area were prepared and pyridine (Py, 6 mbar) was adsorbed at 100 • C for 30 min on the formerly dehydrated samples (300 • C, 1 h) in a special, in situ FT-IR cell. Subsequently, Py was desorbed by evacuation at elevating temperatures 100-300 • C for 30 min. Spectra were recorded at the IR beam temperature with 128 scans at a resolution of 2 cm −1 . For quantitative comparison, the spectra were normalized to 5 mg/cm 2 weight.
In situ FT-IR CO 2 adsorption experiments were conducted in the same spectroscopic system. Mixed-oxide adsorbents were pretreated at 300 • C in high vacuum for 1 h before room temperature CO 2 adsorption (15 mbar) for 30 min. Spectra were recorded in CO 2 atmosphere at RT, followed by evacuation at RT, 100, 200, and 300 • C. Spectra were also measured with the same parameters as before, namely 128 scans at 2 cm −1 resolution and were normalized to 5 mg/cm 2 weight.

Dynamic and Static CO 2 Adsorption
CO 2 adsorption experiments were performed in dynamic conditions in a flow-through system. The sample (0.40 g adsorbent) was dried at 150 • C for 1 h, and 3 vol. % CO 2 /N 2 at a flow rate of 30 mL/min was applied for the adsorption experiments at 25 • C. The gas was analyzed online by using a gas chromatograph NEXIS GC-2030 ATF (Shimadzu, Japan) with a 25 m PLOT Q capillary column. CO 2 adsorption measurements under static conditions were measured at 0 • C and 25 • C with an AUTOSORB iQ-MP-AG (Anton Paar GmbH, Graz, Austria) surface area and pore size analyzer (from Quantachrome, Anton Paar GmbH, Graz, Austria).

Textural Characterization
Nitrogen physisorption isotherms of the prepared adsorbents are shown in Figure S1. Specific surface area and total pore volume, calculated from the nitrogen adsorption/desorption isotherms are presented in Table 1. All the isotherms are of type IV, ZrO 2 and CeO 2 with H3 type, and the mixed oxides are similarly to each other, with H2 type hysteresis loops. ZrO 2 showed the highest specific surface area and pore volume, and on the contrary, CeO 2 has the lowest one. The Ce-Zr composite adsorbents showed transition values proportional to their zirconia content. The use of the extraction method for template removal (Table 1) resulted in a higher surface area than its counterpart obtained by calcination. The observed differences in the shape of the hysteresis loop indicate the presence of "slit-like" or wedge-shaped pores between flaky particles of pure ZrO 2 and CeO 2 oxides, and "ink-bottle-like" pores for all mixed-oxide materials. XRD patterns of samples are presented in Figure 1.
the contrary, CeO2 has the lowest one. The Ce-Zr composite adsorbents showed transition values proportional to their zirconia content. The use of the extraction method for template removal (Table 1) resulted in a higher surface area than its counterpart obtained by calcination. The observed differences in the shape of the hysteresis loop indicate the presence of "slit-like" or wedge-shaped pores between flaky particles of pure ZrO2 and CeO2 oxides, and "ink-bottle-like" pores for all mixed-oxide materials.  The phase composition, unit cell parameters, and average crystallite size are shown in Table 2. The XRD pattern of ZrO2 represents a broad peak typical of amorphous mate- The phase composition, unit cell parameters, and average crystallite size are shown in Table 2. The XRD pattern of ZrO 2 represents a broad peak typical of amorphous materials. The CeO 2 adsorbent exhibits a face-centered, cubic, fluorite-type structure with crystallite size of 16 nm. In the case of the mixed metal-oxide samples, the CeO 2 reflections are broader, lower in intensity, and slightly shifted to higher Bragg angles as well. The broadening of the reflections means a decrease in the crystallite size of the fluorite-like phase, whereas the intensity decrease together with the shifting of position could be assigned to the incorporation of Zr in the ceria lattice. The differences observed in the lattice parameters with the change of the Ce/Zr ratio suggest an alteration in the degree of incorporation of Zr into the fluorite lattice. Considering the smaller ionic radius of Zr 4+ (0.084 nm) compared to Ce 4+ (0.097 nm), an incorporation of the Zr into ceria lattice can be concluded [30,39,40]. A significant shift of the lattice parameter is observed for the Ce 0.33 Zr 0.67 nanomaterial most probably due to the strongest interaction between CeO 2 and ZrO 2 .

TEM Analysis
In Figure 2, TEM micrographs at low (40,000×) and high (600,000×) magnifications, histograms of the particles size distribution, and Selected Area Electron Diffraction (SAED) patterns for the composite samples are presented.  The statistical analysis of TEM images revealed a formation of relatively uniform nanoparticles with mean sizes between 3 and 5 nm and narrow size distributions for all three materials. The calculated mean particle diameters dmean for the tree samples studied with the corresponding standard deviations (SD) are as follows: d1mean = 3.21 nm (SD = 0.92 nm) for Ce0.33Zr0.67, d2mean = 3.53 nm (SD = 0.95 nm) for Ce0.5Zr0.5, and d3mean = 4.91 nm (SD = 2.05 nm) for Ce0.67Zr0.33 sample. Тhey follow the same trend of slightly increasing with a decreasing amount of zirconia, similarly to the average crystallite diameters obtained from X-ray diffraction. The SAED patterns and HRTEM indexing and phase composition analysis clearly proved the interaction of CeO2 and ZrO2 by the established formation of a solid solution phase (Ce,Zr)O2 in the mixed-oxide samples. The phase of monometallic oxide ZrO2 was also detected in all samples (CeO2 cubic, a = 5.40730 Å, COD Entry #96-156-2990; The statistical analysis of TEM images revealed a formation of relatively uniform nanoparticles with mean sizes between 3 and 5 nm and narrow size distributions for all three materials. The calculated mean particle diameters d mean for the tree samples studied with the corresponding standard deviations (SD) are as follows: d 1mean = 3.21 nm (SD = 0.92 nm) for Ce 0.33 Zr 0.67 , d 2mean = 3.53 nm (SD = 0.95 nm) for Ce 0.5 Zr 0.5 , and d 3mean = 4.91 nm (SD = 2.05 nm) for Ce 0.67 Zr 0.33 sample. They follow the same trend of slightly increasing with a decreasing amount of zirconia, similarly to the average crystallite diameters obtained Nanomaterials 2023, 13, 2428 7 of 15 from X-ray diffraction. The SAED patterns and HRTEM indexing and phase composition analysis clearly proved the interaction of CeO 2 and ZrO 2 by the established formation of a solid solution phase (Ce,Zr)O 2 in the mixed-oxide samples. The phase of monometallic oxide ZrO 2 was also detected in all samples (CeO 2 cubic, a = 5.40730 Å, COD Entry #96-156-2990; ZrO 2 tetragonal, a = 3.61200 Å c = 5.21200 Å, COD Entry #96-152-6428).

X-ray Photoelectron Spectroscopy (XPS)
The XPS spectra (Figure 3) of the CeO 2 sample shows multiple peaks at 882.5 and 900.9 eV, due to photoemission from Ce 3d3/2 and Ce3d5/2 levels for Ce 4+ ions, respectively and the peaks at 888.7, 898.3, 907.2, and 916.7 eV are its Ce 3d Ce 4+ satellites [41][42][43][44]. Ce 3+ oxides have Ce 3d3/2 and Ce 3d5/2 spectra consisting of another two multiplets. The peaks with the highest binding energies are localized at 903.3 eV and 884.8 eV, and the energy states with low binding energy are at 899.4 eV and 880.9 eV, respectively [41]. The observed doublets in Ce 3d spectra are typical of Ce 4+ ions and Ce 3+ ions. The observed slight decrease in the atomic concentration of the Ce 3d component for the bi-component samples compared to pure CeO2 may be due to the formation of Zr-O-Ce bonds [45]. XPS spectra of the mixed-oxide samples [46,47] showed that the O 1s peak was decomposed into two sets of components. Ce 3+ oxides have Ce 3d 3/2 and Ce 3d 5/2 spectra consisting of another two multiplets. The peaks with the highest binding energies are localized at 903.3 eV and 884.8 eV, and the energy states with low binding energy are at 899.4 eV and 880.9 eV, respectively [41]. The observed doublets in Ce 3d spectra are typical of Ce 4+ ions and Ce 3+ ions. The observed slight decrease in the atomic concentration of the Ce 3d component for the bi-component samples compared to pure CeO 2 may be due to the formation of Zr-O-Ce bonds [45]. XPS spectra of the mixed-oxide samples [46,47] showed that the O 1s peak was decomposed into two sets of components. The peaks at about 529.5 and 529.7 eV correspond to oxygen from the oxide lattice in Ce 4+ and Zr 4+ . The observed slight shift to lower binding energy is probably due to the interaction between CeO 2 and ZrO 2 . The third component at 531.5 eV is usually associated with chemisorbed oxygen, such as hydroxyl groups, O 2 2− , O − oxygen centres, or molecular oxygen [46] but here it is also a component associated with oxygen bonded to Ce 3+ . This effect clearly shows that the close contact between the metals in the mixed-oxide system leads to the formation of oxygen vacancies and increased oxygen mobility. In addition, promoting the oxygen mobility leads to enhancing the basicity, and enriching the surface oxygen species, which are efficient at activating CO 2 .
Hydroxyl groups associated with a binding energy of about 533 eV were also observed in an amount decreasing from 9% for the CeO 2 sample to less than 3% for samples with higher Zr content. The BE of Zr 3d 5/2 and Zr 3d 3/2 is 181.9 eV and 184.2 eV, respectively [48], and the peaks are registered at the same position in the spectra of all Zr-containing samples. The surface area elemental composition is presented in Table 3. It is obvious that the Ce 3+ concentration on the surface is higher for the Ce 0.67 Zr 0.33 and Ce 0.33 Zr 0.67 mixed-oxide nanomaterials compared to pure CeO 2 (Table 3). This is an indication of a higher degree of substitution of Zr ions in the cerium-oxide lattice and enhanced oxygen mobility. It should be noted that, the small systematic shift of binding energy of O-Ce 4+ , O-Ce 3+ and O-Zr 4+ peaks (Table 3), which is in accordance with X-ray analysis ( Table 2) showing strong interaction between cerium and zirconium oxides, partial incorporation of smaller zirconium ions into the fluorite-oxide lattice and formation of cerium ions in a low oxidation state. The Ce 3+ surface concentration is higher for Ce-Zr composite materials as compared to pure CeO 2 (Table 3), which indicates a high degree of Zr incorporation into ceria lattice with the formation of oxygen vacancies [49]. XPS analysis data of high ceria containing samples (CeO 2 , Ce 0.67 Zr 0.33 ) show that the concentration of Ce is lower on the surface than theoretically calculated, whereas the oxygen content is higher than the theoretical content (73/71 at. %, instead of 66 at. %). It means that these samples are surface-rich with defect sites and OH groups due to the very small particle size or adsorbed O-containing species. The amount of surface oxygen functional groups or adsorbed O-containing species in all mixed oxides increases. The absence of zirconium oxide in a lower oxidation state is indicative of segregation of the zirconia phase over the ceria particles, which is in accordance with XRD and TEM results.

In Situ FT-IR Spectroscopy of Adsorbed Pyridine and CO 2
The surface species of oxides can also be described as acid-base pairs. Oxygen atoms of the structure serve as basic sites (Lewis) for adsorbents, whereas coordinatively unsaturated metals or oxygen vacancies act as Lewis acid centers. Bridged hydroxyl groups connecting to transition metals of different valent or oxidation states (e.g., Ce 3+ -O(H)-Ce 4+ ) show Brønsted acid character by proton donating ability. Acidity of the samples was characterized by in situ FT-IR spectroscopy using pyridine (Py) adsorption as a base probe molecule. Figure S2 shows the Py adsorption spectra of the studied samples. Py adsorption studies supported the strong Lewis acidity of the samples. It was found that only the CeO 2 sample exhibited some weak Brønsted acidity. The latter result supports the XPS data, detecting reduced Ce 3+ species on the surface. Spectra of mixed-oxide samples exhibit rather the characteristics of CeO 2 . Considering the amount of acid sites, it can be observed that ZrO 2 shows a higher acidity compared to CeO 2 . Surface basicity of oxides can be characterized by CO 2 chemisorption. When basic surface hydroxyl groups react with CO 2 , bicarbonate (hydrogencarbonate) species are formed. Through the interaction of structural O 2− ions (Lewis base) with CO 2 , carbonate species can be detected. The various surface species, formed by CO 2 adsorption, can be observed in the 1800-800 cm −1 spectral range [50]. FT-IR spectra of adsorbed CO 2 on mixed CeO 2 /ZrO 2 samples are shown in Figure 4. Spectra were collected by room-temperature adsorption for 30 min, followed by desorption at RT (Figure 4A), and at 100 • C ( Figure 4B) in high vacuum. Comparing the RT CO 2 desorption spectra of mixed oxides, similar bands with some variation in intensity ratios can be observed, however the Ce 0.67 Zr 0.33 sample shows more intensive ones. Characteristic stretching vibrations ν(CO 3 ) bands of hydrocarbonates at 1600 and 1410 cm −1 [51,52] are very intensive in the sample, and almost diminishes in the other two. This fact is in accordance with XPS investigations, showing a high concentration of OH groups on the CeO 2 sample and a decreasing tendency with increasing Zr content. The surface species of oxides can also be described as acid-base pairs. Oxygen atoms of the structure serve as basic sites (Lewis) for adsorbents, whereas coordinatively unsaturated metals or oxygen vacancies act as Lewis acid centers. Bridged hydroxyl groups connecting to transition metals of different valent or oxidation states (e.g., Ce 3+ -O(H)-Ce 4+ ) show Brønsted acid character by proton donating ability. Acidity of the samples was characterized by in situ FT-IR spectroscopy using pyridine (Py) adsorption as a base probe molecule. Figure S2 shows the Py adsorption spectra of the studied samples. Py adsorption studies supported the strong Lewis acidity of the samples. It was found that only the CeO2 sample exhibited some weak Brønsted acidity. The latter result supports the XPS data, detecting reduced Ce 3+ species on the surface. Spectra of mixed-oxide samples exhibit rather the characteristics of CeO2. Considering the amount of acid sites, it can be observed that ZrO2 shows a higher acidity compared to CeO2. Surface basicity of oxides can be characterized by CO2 chemisorption. When basic surface hydroxyl groups react with CO2, bicarbonate (hydrogencarbonate) species are formed. Through the interaction of structural O 2− ions (Lewis base) with CO2, carbonate species can be detected. The various surface species, formed by CO2 adsorption, can be observed in the 1800-800 cm −1 spectral range [50]. FT-IR spectra of adsorbed CO2 on mixed CeO2/ZrO2 samples are shown in Figure 4. Spectra were collected by room-temperature adsorption for 30 min, followed by desorption at RT (Figure 4A), and at 100 °C ( Figure 4B) in high vacuum. Comparing the RT CO2 desorption spectra of mixed oxides, similar bands with some variation in intensity ratios can be observed, however the Ce0.67Zr0.33 sample shows more intensive ones. Characteristic stretching vibrations ν(CO3) bands of hydrocarbonates at 1600 and 1410 cm −1 [51,52] are very intensive in the sample, and almost diminishes in the other two. This fact is in accordance with XPS investigations, showing a high concentration of OH groups on the CeO2 sample and a decreasing tendency with increasing Zr content.  By increasing the desorption temperature to 100 °C, the weakly bound hydrogencarbonate bands disappear and several, better resolved ones appear ( Figure 4B). According to Datturi et al. [52], the bands can be associated with surface monodentate (1518 cm −1 ), bidentate (1587, 1346 cm −1 ), and polydentate (1439, 1402 cm −1 ) carbonate species. The other two mixed-oxides with a lower Ce ratio show only monodentate and bidentate carbonate species with much lower intensity. Formation of thermally stable, polydentate, or core- By increasing the desorption temperature to 100 • C, the weakly bound hydrogencarbonate bands disappear and several, better resolved ones appear ( Figure 4B). According to Datturi et al. [52], the bands can be associated with surface monodentate (1518 cm −1 ), bidentate (1587, 1346 cm −1 ), and polydentate (1439, 1402 cm −1 ) carbonate species. The other two mixed-oxides with a lower Ce ratio show only monodentate and bidentate carbonate species with much lower intensity. Formation of thermally stable, polydentate, or core-carbonate species indicates the incorporation of carbonate ions into the surface layer, thus its restructuration. It seems that the high mobility of oxygen atoms in pure ceria is stabilized by the presence of zirconium ions and an optimal composition makes the surface more appropriate for CO 2 adsorption. Daturi et al. [52] draw the conclusion from their similar FT-IR spectroscopic results that the incorporation of Zr 4+ ions into the ceria framework enhances the surface oxygen relaxation over 50% Ce content and has the opposite effect on the relaxion of Ce 4+ with already low Zr level, thus stabilizing the surface Ce 4+ framework. Based on the FT-IR spectroscopic results, it can be expected that Ce 0.67 Zr 0.33 sample will show more favorable CO 2 adsorption properties compared to other mixed and pure oxides.

CO 2 Adsorption
CO 2 adsorption breakthrough curves under dynamic conditions of the pure and the mixed-oxide materials are presented in Figure 5.
carbonate species indicates the incorporation of carbonate ions into the surface layer, thus its restructuration. It seems that the high mobility of oxygen atoms in pure ceria is stabilized by the presence of zirconium ions and an optimal composition makes the surface more appropriate for CO2 adsorption. Daturi et al. [52] draw the conclusion from their similar FT-IR spectroscopic results that the incorporation of Zr 4+ ions into the ceria framework enhances the surface oxygen relaxation over 50% Ce content and has the opposite effect on the relaxion of Ce 4+ with already low Zr level, thus stabilizing the surface Ce 4+ framework. Based on the FT-IR spectroscopic results, it can be expected that Ce0.67Zr0.33 sample will show more favorable CO2 adsorption properties compared to other mixed and pure oxides.

CO2 Adsorption
CO2 adsorption breakthrough curves under dynamic conditions of the pure and the mixed-oxide materials are presented in Figure 5. CeO2 and ZrO2 show low adsorption capacity, whereas a significantly higher capacity is detected for the mixed-oxide materials. The shape of the breakthrough curves is similar, which can be due to their similar porosity. The ratio between Ce/Zr influences the adsorption performance of the materials. As predicted by the FT-IR results, the highest capacity for CO2 adsorption is detected for Ce0.67Zr0.33 (3.5 mmol/g) ( Table 4). A large quantity of surface oxygen groups or adsorbed O-containing species in Ce-Zr composite nanomaterials could be a reason for the higher CO2 capacity. The Ce0.5Zr0.5 and ext. Ce0.5Zr0.5 show similar CO2 adsorption despite the difference in the surface area.  CeO 2 and ZrO 2 show low adsorption capacity, whereas a significantly higher capacity is detected for the mixed-oxide materials. The shape of the breakthrough curves is similar, which can be due to their similar porosity. The ratio between Ce/Zr influences the adsorption performance of the materials. As predicted by the FT-IR results, the highest capacity for CO 2 adsorption is detected for Ce 0.67 Zr 0.33 (3.5 mmol/g) ( Table 4). A large quantity of surface oxygen groups or adsorbed O-containing species in Ce-Zr composite nanomaterials could be a reason for the higher CO 2 capacity. The Ce 0.5 Zr 0.5 and ext. Ce 0.5 Zr 0.5 show similar CO 2 adsorption despite the difference in the surface area.
The observed effect indicates that the surface composition is more important than other structure peculiarities. Moreover, the time required to reach the total adsorption for CeO 2 and ZrO 2 materials (T = 17 min) is shorter than that of the mixed-metal oxides (T = 20-26 min). Additional adsorption experiments were performed with the addition of 1 vol.% water vapor to the CO 2 /N 2 flow at a rate of 30 mL/min to determine the CO 2 selectivity. In the latter case, higher CO 2 adsorption capacity was determined for all the adsorbents ( Table 4). The highest capacity in the humid environment was detected for Ce 0.67 Zr 0.33 (3.7 mmol/g). Under dry conditions, surface oxygen reacts with CO 2 generating carbonate species [14,53]. Under wet conditions, it is assumed that the surface oxygen reacts with water molecules in the gas leading to the formation of hydroxyl groups. The reaction of the hydroxyl groups with CO 2 results in the generation of hydrogen carbonate species and thereby accelerating the carbon dioxide adsorption.
Adsorption capacity of synthesized nanomaterials were determined using a laboratory scale fixed-bed reactor. The Yoon-Nelson model [54] was applied for adsorption kinetics in a fixed-bed column. The linear form of the model is represented by Equation (1): where k YN is the Yoon-Nelson rate constant (min −1 ), τ is the time required for 50% of adsorbate breakthrough (min), t is the sampling time (min), C 0 is the initial concentration of CO 2 , and C is the concentration of CO 2 at any time during evaluation. It was found that the model and experimental data have a strong correlation (R 2 > 0.99) ( Figure S3). Samples were also tested for CO 2 capture under equilibrium conditions and the results are presented in Figures 6-8. CO 2 adsorption under static saturation mode without nitrogen stream shows a lower adsorption capacity than those obtained in the dynamic conditions. The heat of adsorption was calculated from CO 2 adsorption isotherms at 0 • C and 25 • C using the Clausius-Clapeyron equation. They ranged between 25 and 130 kJ/mol. The mixed cerium-zirconium materials exhibited higher values in comparison to CeO 2 and ZrO 2 materials . The Ce 0.33 Zr 0.67 and Ce 0.67 Zr 0.33 (Figure 7) adsorbents demonstrated the highest ones (115−128 kJ/mol) due to the formation of Ce-O-Zr species. They are around three times higher than that of CeO 2 and ZrO 2 (40-55 kJ/mol) ( Figure 6). Ce 0.5 Zr 0.5 and ext.Ce 0.5 Zr 0.5 (Figure 8) have very similar values, which are two times higher than pure oxides ( Figure 6).   Based on the calculated adsorption heats, CO2 is physisorbed on CeO2 and ZrO2 materials whereas chemisorption of CO2 is assumed on the mixed cerium-zirconium nanomaterials. According to the literature [14], under dry conditions, the surface oxygen of CeO2 reacts with CO2, generating carbonate species when the Ce-Zr composites are used Based on the calculated adsorption heats, CO 2 is physisorbed on CeO 2 and ZrO 2 materials whereas chemisorption of CO 2 is assumed on the mixed cerium-zirconium nanomaterials. According to the literature [14], under dry conditions, the surface oxygen of CeO 2 reacts with CO 2 , generating carbonate species when the Ce-Zr composites are used as adsorbents. When the mixed Ce/Zr oxides are used as adsorbents, we assume that the stronger interaction of CO 2 molecules with oxygen of the adsorbent surface is due to the stronger basicity of O 2− in them than that of O 2− in the pure CeO 2 and ZrO 2 nanoparticles.
A stronger interaction between CO 2 molecules and the Ce-Zr composite materials is also proved by the higher CO 2 desorption temperature (100 • C), which is needed for total regeneration of the adsorbents.

Conclusions
CeO 2 , ZrO 2 , and the Ce-Zr composite nanoparticles with a high specific surface area were successfully synthesized using the hydrothermal synthesis procedure. A high CO 2 adsorption capacity was determined for all the adsorbents depending on their composition and structural peculiarities. Additionally, CO 2 chemisorption enhanced the CO 2 capture on Ce-Zr composites due to the presence of strong O 2− base sites and enriched surface oxygen species. Materials reused in five adsorption/desorption cycles revealed a high stability with only a slight decrease in adsorption capacity. Among the studied materials, the Ce 0.67 Zr 0.33 material showed the highest adsorption capacity (3.7 mmol/g). CO 2 chemisorption is assumed based on the calculated adsorption heat. Enhanced CO 2 adsorption capacities were detected in experiments with 3 vol.% CO 2 plus 1 vol.% water vapor due to the additional chemisorption of CO 2 . Total CO 2 desorption from the Ce-Zr composites was achieved at 100 • C. Experimental data can be appropriately described by the Yoon-Nelson kinetic model. For the first time, it is described that Ce-Zr composite nanomaterials are promising materials for CO 2 adsorption in dry and humid media.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nano13172428/s1, Figure S1. Nitrogen physisorption isotherms of the studied CeO 2 , ZrO 2 and CeO 2 /ZrO 2 mixed oxide materials. Figure S2. FT-IR spectra of adsorbed pyridine on the studied adsorbents. Py (6 mbar) was adsorbed on 300 • C dehydrated samples at 100 • C and desorbed at 100 • C in high vacuum. Figure S3. Fitting of experimental data on kinetic model at 0 • C.