*2.3. Redox Properties (H2-Temperature Programmed Reduction (TPR))*

H2-TPR experiments took place to investigate the ceria shape effect on the redox properties of as-prepared samples. Figure 4a shows the TPR profiles of bare ceria samples, consisting of two wide-ranging peaks which are centred at 526–551 ◦C and 789–813 ◦C. These peaks are attributed to ceria surface oxygen (Os) and bulk oxygen (Ob) reduction, respectively [33,49,55]. In Table 2, the hydrogen consumption corresponding to surface oxygen as well as to bulk oxygen reduction is presented. Based on the ratio of surface-to-bulk oxygen (Os/Ob), the following order was acquired: CeO2-NR (1.13) > CeO2-NP (0.94) > CeO2-NC (0.71). This indicates the superior reducibility of the rod-shaped sample as it exhibits the highest amount of loosely bound oxygen species. The latter is expected to notably affect the deN2O process, where the desorption of adsorbed oxygen species mainly determines the reaction rate (*vide infra*).


**Table 2.** The redox properties of the bare CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

<sup>a</sup> Estimated by the area of the corresponding temperature programmed reduction (TPR) peaks, which is calibrated against a known amount of CuO standard sample.

The reduction profiles of the Co/CeO<sup>2</sup> samples as well as the one of a Co3O<sup>4</sup> reference are shown in Figure 4b. Table 2 summarizes the main TPR peaks along with the hydrogen consumption (mmol H<sup>2</sup> g −1 ). Pure Co3O<sup>4</sup> shows two reduction peaks (a and b) in much lower temperatures than those of bare ceria samples, namely 305 ◦C and 415 ◦C. They are ascribed to the stepwise reduction of Co3O<sup>4</sup> → CoO → Co, respectively [44,56–58].

On the other hand, Co/CeO<sup>2</sup> samples exhibit two main peaks at the temperature range of 318–335 ◦C (peak a) and 388–405 ◦C (peak b), ascribed to the reduction of Co3+ to Co2+ and Co2+ to Co<sup>0</sup> , respectively [33,59,60]. Obviously, the cobalt addition facilitates the reduction of ceria surface oxygen, shifting the peaks centered at 526–551 ◦C to a lower temperature (comparison of Figure 4a,b). They also exhibit a broad peak above 800 ◦C, attributed to the ceria subsurface oxygen reduction, while the capping oxygen reduction overlaps with the reduction of CoO [33,56,61]. Apparently, the reduction of the mixed oxides takes place in lower temperatures compared to the bare ceria samples, demonstrating the beneficial effect of cobalt on the surface oxygen reduction of ceria. In fact, the interaction between the two oxide phases could be considered responsible for the improved reducibility and oxygen mobility, as thoroughly discussed in previous studies [48,49,62]. According to the consumption of hydrogen in the low-temperature range (Table 2), which could be related to the cobalt species reduction along with the ceria surface oxygen reduction, the Co/CeO2-NP and Co/CeO2-NR samples exhibit a similar H<sup>2</sup> uptake (about 2.40 mmol H<sup>2</sup> g −1 ) while the sample of nanocube morphology exhibits a much lower value (2.05 mmol H<sup>2</sup> g −1 ). This trend is well-matched to the catalytic results (*vide infra*), revealing the key role of redox ability on the deN2O process.

Moreover, the Co/CeO2-NR sample exhibits the lowest reduction temperature (peak at 318 ◦C) in comparison with the other samples (peak ca. 335 ◦C), indicating the facilitation of Co3+ species reduction over ceria nanorods. Noteworthily, the theoretical amount of hydrogen for the complete reduction of Co3O<sup>4</sup> to Co (approx. 1.76 mmol H<sup>2</sup> g −1 , based on a 7.8 wt. % nominal loading of Co) is always surpassed by the hydrogen amount required for the reduction of Co/CeO<sup>2</sup> samples (Table 2). The latter reveals the facilitation of ceria capping oxygen reduction in the presence of cobalt, further corroborating the above findings. *Catalysts* **2019**, *9*, x FOR PEER REVIEW 7 of 20

**Figure 4. Figure 4.** The H The H2-TPR profiles of (**a**) bare CeO2 and (**b**) Co3O4, Co/CeO2 samples. 2 -TPR profiles of (**a**) bare CeO<sup>2</sup> and (**b**) Co3O<sup>4</sup> , Co/CeO<sup>2</sup> samples.

#### *2.4. Surface Analysis (X-ray Photoelectron Spectroscopy (XPS))*

An XPS analysis was performed in order to investigate the effect of ceria morphology on the elemental chemical states and surface composition of Co/CeO<sup>2</sup> mixed oxides. Figure 5a shows the Ce3d XPS spectra of ceria nanoparticles of different morphology and the corresponding Co/CeO<sup>2</sup> samples, which can be deconvoluted into eight components [63–65], with the assignment of the characteristic peaks having been thoroughly described in our previous work [49]. In brief, the three pairs of peaks labeled as u, v; u", v"; and u"', v"' are ascribed to Ce4+, whereas the residual u' and v' peaks are ascribed to Ce3+ species.

The corresponding O 1s spectra of the samples are depicted in Figure 5b. The low binding energy peak at 529.3 eV is attributed to the lattice oxygen (O<sup>I</sup> ) of Co3O<sup>4</sup> and CeO<sup>2</sup> phases, and the high binding energy peak at 531.3 eV corresponds to the chemisorbed oxygen (OII) such as adsorbed oxygen (O−/O<sup>2</sup> <sup>2</sup>−) and water, carbonate as well as hydroxyl species [23,56].

The proportion of Ce3+ (%) as well as the OII/O<sup>I</sup> ratio for all samples is summarized in Table 3. Bare ceria supports exhibit a similar amount of Ce3+ ranging from 23.3 to 25.3%. Regarding, the cobalt-ceria samples, the population of Ce3+ species is slightly higher, varying between 26.1 and 28.5%. In particular, the Co/CeO2-NR sample exhibits the highest amount (28.5%) followed by Co/CeO2-NP (26.7%) and Co/CeO2-NC (26.1%), indicating the abundance of the nanorod samples in oxygen vacancies. Interestingly, the relative ratio of adsorbed to lattice oxygen (OII/O<sup>I</sup> ) and the Ce3+ (%) follow the same order, namely, Co/CeO2-NR (0.60) > Co/CeO2-NP (0.53) > Co/CeO2-NC (0.51), perfectly matched to the order obtained for the catalytic performance, as it will be discussed in the sequence. It should be also noted that Co addition to CeO2-NR enhances both the population of reduced Ce3+ species and the OII/O<sup>I</sup> ratio, revealing the synergistic interactions between cerium and cobalt oxides toward the formation of highly reducible composites, in agreement with the TPR results.


**Table 3.** The XPS results of bare CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

Figure 6 depicts the Co 2p XPS spectra of Co/CeO<sup>2</sup> samples along with the spectrum obtained for the Co3O<sup>4</sup> reference sample for comparison purposes. The samples exhibit two major peaks of Co2p3/2 (780 eV) and Co2p1/2 (795 eV). According to peaks' positions and shapes, the structure of the cobalt spinel is formed [23,66,67]. The Co2+/Co3+ ratio of Co/CeO<sup>2</sup> samples derived by the deconvolution of the Co2p1/2 and Co2p3/2 peaks is included in Table 3. The nanorod sample, which offers the best deN2O performance (*vide infra*), exhibits the highest Co2+/Co3+ ratio (1.32), followed by nanocubes (1.06) and nanopolyhedra (0.94). In view of this fact, it has been reported that samples with a high Co2+/Co3+ ratio exhibit better deN2O performance [3,20,22,43,59], further corroborating the present findings.

**Figure 5.** The X-ray photoelectron spectroscopy (XPS) spectra of (**a**) Ce 3d and (**b**) O 1s of bare CeO2 and Co/CeO2 samples. **Figure 5.** The X-ray photoelectron spectroscopy (XPS) spectra of (**a**) Ce 3d and (**b**) O 1s of bare CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

the present findings.

**Table 3.** The XPS results of bare CeO2 and Co/CeO2 samples.

**Sample Co2+/Co3+ Ce3+ (%) OII/OI** CeO2-NC - 23.3 0.50 CeO2-NR - 24.3 0.47 CeO2-NP - 25.3 0.49 Co/CeO2-NC 1.06 26.1 0.51 Co/CeO2-NR 1.32 28.5 0.60 Co/CeO2-NP 0.94 26.7 0.53

Figure 6 depicts the Co 2p XPS spectra of Co/CeO2 samples along with the spectrum obtained for the Co3O4 reference sample for comparison purposes. The samples exhibit two major peaks of Co2p3/2 (780 eV) and Co2p1/2 (795 eV). According to peaks' positions and shapes, the structure of the cobalt spinel is formed [23,66,67]. The Co2+/Co3+ ratio of Co/CeO2 samples derived by the deconvolution of the Co2p1/2 and Co2p3/2 peaks is included in Table 3. The nanorod sample, which offers the best deN2O performance (*vide infra*), exhibits the highest Co2+/Co3+ ratio (1.32), followed by nanocubes (1.06) and nanopolyhedra (0.94). In view of this fact, it has been reported that samples

**Figure 6.** The Co 2p XPS spectra of the Co3O4 and Co/CeO2 samples: The Co 2p XPS spectra of Co/CeO2 samples have been magnified. **Figure 6.** The Co 2p XPS spectra of the Co3O<sup>4</sup> and Co/CeO<sup>2</sup> samples: The Co 2p XPS spectra of Co/CeO<sup>2</sup> samples have been magnified.

#### *2.5. Catalytic Evaluation Studies 2.5. Catalytic Evaluation Studies*

The impact of ceria morphology on the catalytic decomposition of N2O under oxygen deficient and oxygen excess conditions was next examined. Figure 7a and b shows the N2O conversion profiles as a temperature function for bare ceria as well as Co/CeO2 samples in the absence and presence of The impact of ceria morphology on the catalytic decomposition of N2O under oxygen deficient and oxygen excess conditions was next examined. Figure 7a,b shows the N2O conversion profiles as a temperature function for bare ceria as well as Co/CeO<sup>2</sup> samples in the absence and presence of oxygen, respectively. The Co/CeO2-NR sample exhibits the best catalytic performance, both in the absence and presence of oxygen in the gas stream. Apparently, the addition of cobalt in the ceria lattice enormously enhances the catalytic efficiency without, however, affecting the catalytic order of bare ceria samples, suggesting the pivotal role of ceria morphology on the deN2O performance. In terms of the half-conversion temperature (T50), the following order is obtained for the mixed oxides in the absence of oxygen: Co/CeO2-NR (449 ◦C) > Co/CeO2-NP (458 ◦C) > Co/CeO2-NC (464 ◦C). The same trend is observed in the presence of oxygen as well, although in slightly higher temperatures, due to its competitive sorption on the catalyst surface. In this point, it should be noted that the un-catalyzed reaction shows nearly zero N2O conversion in the temperature range investigated, as previously reported [29,46,68].

previously reported [29,46,68].

oxygen, respectively. The Co/CeO2-NR sample exhibits the best catalytic performance, both in the absence and presence of oxygen in the gas stream. Apparently, the addition of cobalt in the ceria lattice enormously enhances the catalytic efficiency without, however, affecting the catalytic order of bare ceria samples, suggesting the pivotal role of ceria morphology on the deN2O performance. In terms of the half-conversion temperature (T50), the following order is obtained for the mixed oxides in the absence of oxygen: Co/CeO2-NR (449 °C) > Co/CeO2-NP (458 °C) > Co/CeO2-NC (464 °C). The same trend is observed in the presence of oxygen as well, although in slightly higher temperatures, due to its competitive sorption on the catalyst surface. In this point, it should be noted that the un-

**Figure 7.** N2O conversion as a function of temperature for CeO2 and Co/CeO2 samples of different morphology (**a**) in the absence and (**b**) in the presence of oxygen: The reaction conditions are 1000 ppm N2O, 0 or 2% O2 and Gas Hour Space Velocity (GHSV) = 40,000 h<sup>−</sup>1. **Figure 7.** N2O conversion as a function of temperature for CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples of different morphology (**a**) in the absence and (**b**) in the presence of oxygen: The reaction conditions are 1000 ppm N2O, 0 or 2% O<sup>2</sup> and Gas Hour Space Velocity (GHSV) = 40,000 h−<sup>1</sup> .

The above findings can be well-interpreted by taking into account a redox-type mechanism for the decomposition of N2O over cobalt spinel oxides [4,23,24,30,59,69–73]: The above findings can be well-interpreted by taking into account a redox-type mechanism for the decomposition of N2O over cobalt spinel oxides [4,23,24,30,59,69–73]:

$$\text{Co}^{2+} + \text{N}\_2\text{O} \rightarrow \text{Co}^{3+} \text{-O}^- + \text{N}\_2\tag{1}$$

$$\text{Co}^{3+}\text{-O}^{-}+\text{N}\_{2}\text{O} \rightarrow \text{Co}^{3+}\text{-O}\_{2}^{-}+\text{N}\_{2}\tag{2}$$

$$\text{Co}^{3+}\text{-}\text{O}\_{2}\text{--}\rightarrow\text{Co}^{2+}+\text{O}\_{2}\tag{3}$$

In this mechanistic sequence, N2O is initially chemisorbed on the Co2+ sites (Equation (1)) which are considered as the active centres for initiating the N2O dissociative adsorption. Then, the regeneration of the active sites is taking place through the Co3+/Co2+ redox cycle, involving the

(Os/Ob).

reducibility, leads to a superior deN2O performance.

compromise between redox and textural characteristics.

combination of O<sup>−</sup> into O<sup>2</sup> − (Equation (2)) and the desorption of molecular oxygen (Equation (3)), which finally leads to the regeneration of those sites [69].

However, in the case of Co3O4/CeO<sup>2</sup> mixed oxides, the excellent redox characteristics of ceria, such as oxygen storage capacity and oxygen mobility, can be further accounted for the regeneration of active sites through the following steps [69]:

$$\text{Ca}^{3+}\text{-}\text{O}^{-}+\text{Ce}^{3+}\text{-O}\_{\text{vac}} \rightarrow \text{Co}^{2+}+\text{Ce}^{4+}\text{-O}^{-}\tag{4}$$

$$\text{2Ce}^{4+}\text{-O}^{-}\leftrightarrow \text{Ce}^{4+}\text{-O}\_{2}^{2-}\text{-Ce}^{4+}\tag{5}$$

$$\text{Ca}^{4+}\text{-}\text{O}^{-}+\text{N}\_{2}\text{O} \rightarrow \text{Ce}^{3+}\text{-}\text{O}\_{\text{vac}} + \text{N}\_{2} + \text{O}\_{2} \tag{6}$$

Based on the above mechanistic scheme, the superiority of the Co/CeO<sup>2</sup> sample with a rod-like morphology can receive a consistent explanation. More specifically, nanorod-shaped ceria with (110) and (100) reactive planes exhibit enhanced oxygen kinetics and reducibility as it has the highest population of loosely bound oxygen species (Table 2), which is a decisive factor in terms of deN2O activity. In other words, the high amount of weakly bound oxygen species present in the Co3O4/CeO<sup>2</sup> samples of rod-like morphology, linked directly to oxygen vacancy formation and oxygen mobility, could be considered responsible for the formation and the consequent regeneration of active sites. In this regard, a perfect interrelation between the catalytic performance (in terms of the half-conversion temperature, T50) and the redox properties (in terms of the ratio of surface oxygen to bulk oxygen, Os/Ob) is disclosed, as illustrated in Figure 8. This clearly justifies the key role of redox properties on the deN2O process. In a similar manner, Liu et al. [28] have pointed out that the synergistic interaction between the two oxide phases in a CuO–CeO<sup>2</sup> mixed oxide enhances the reducibility and consequently the deN2O efficiency as the surface-adsorbed oxygen species is easily desorbed and the active sites' regeneration is enabled. *Catalysts* **2019**, *9*, x FOR PEER REVIEW 13 of 20

**Figure 8.** The half-conversion temperature (T50) as a function of the TPR surface-to-bulk oxygen ratio **Figure 8.** The half-conversion temperature (T50) as a function of the TPR surface-to-bulk oxygen ratio (Os/O<sup>b</sup> ).

CoOX/CeO2 catalysts *via* the facilitation of oxygen transfer at the metal-support interface [74]. It should be, therefore, deduced that ceria nanorods with the exposed (110) and (100) facets show the highest surface-to-bulk oxygen ratio resulting in improved reducibility and oxygen kinetics while exhibiting the highest amount of weakly bound oxygen species which is a decisive factor in the deN2O process. Upon cobalt addition, the nanorod sample exhibits in addition the highest population in Ce3+/Co2+ redox pairs, indicative of abundant oxygen vacancies, which, along with its enhanced

Moreover, ceria nanorods facilitate the reduction of Co3+ to Co2+ active sites (Table 3), further

In this point, the enhanced textural characteristics (BET area and pore volume) of Co/CeO2-NR as compared to Co/CeO2-NC should be also mentioned, which could be further accounted for its enhanced deN2O performance. Thus, by taking into account the specific activity normalized per unit of surface area (nmol m−2 s−1) instead of mass unit (nmol g−1 s−1), an inferior performance is observed for Co/CeO2-NR compared to Co/CeO2-NC (Table 4). On the other hand, Co3O4/CeO2-NR exhibits a superior deN2O performance (both in terms of conversion and specific activity) as compared to Co3O4/CeO2-NP despite their similar structural (crystallite size) and textural (surface area) properties (Table 1). The latter clearly reveals the importance of exposed facets and redox properties on the deN2O process, as it has been similarly reported by Zabilskiy et al. [45] for CuO/CeO2 nanostructures of different morphology. Therefore, on the basis of the present findings, it can be deduced that the enhanced N2O conversion performance of Co3O4/CeO2-NR mixed oxides could be attributed to a

More interestingly, the deN2O performance of CeO<sup>2</sup> as well as the Co3O4/CeO<sup>2</sup> samples totally coincides, indicating the significance of the ceria carrier. However, the superiority of the mixed oxides in comparison to the bare ceria samples is evident, reflecting the synergistic interactions between cobalt and cerium oxides. The latter is manifested by the improved redox properties (in terms of H<sup>2</sup> consumption and TPR onset temperature) of Co3O4/CeO<sup>2</sup> mixed oxides as compared to bare ceria (Table 2). In a similar manner, the incorporation of cobalt into the ceria lattice increases both the amount of the adsorbed oxygen species (O−/O<sup>2</sup> <sup>2</sup>−) and Ce3+ (Table 3), related with the surface oxygen reduction and the abundance in oxygen vacancies (Ovac).

Moreover, ceria nanorods facilitate the reduction of Co3+ to Co2+ active sites (Table 3), further contributing to the superior catalytic performance of the Co/CeO2-NR sample. Along the same line, it has been recently reported that ceria nanorods stabilize the partial oxidation state of Co in CoOX/CeO<sup>2</sup> catalysts *via* the facilitation of oxygen transfer at the metal-support interface [74]. It should be, therefore, deduced that ceria nanorods with the exposed (110) and (100) facets show the highest surface-to-bulk oxygen ratio resulting in improved reducibility and oxygen kinetics while exhibiting the highest amount of weakly bound oxygen species which is a decisive factor in the deN2O process. Upon cobalt addition, the nanorod sample exhibits in addition the highest population in Ce3+/Co2+ redox pairs, indicative of abundant oxygen vacancies, which, along with its enhanced reducibility, leads to a superior deN2O performance.

In this point, the enhanced textural characteristics (BET area and pore volume) of Co/CeO2-NR as compared to Co/CeO2-NC should be also mentioned, which could be further accounted for its enhanced deN2O performance. Thus, by taking into account the specific activity normalized per unit of surface area (nmol m−<sup>2</sup> s −1 ) instead of mass unit (nmol g−<sup>1</sup> s −1 ), an inferior performance is observed for Co/CeO2-NR compared to Co/CeO2-NC (Table 4). On the other hand, Co3O4/CeO2-NR exhibits a superior deN2O performance (both in terms of conversion and specific activity) as compared to Co3O4/CeO2-NP despite their similar structural (crystallite size) and textural (surface area) properties (Table 1). The latter clearly reveals the importance of exposed facets and redox properties on the deN2O process, as it has been similarly reported by Zabilskiy et al. [45] for CuO/CeO<sup>2</sup> nanostructures of different morphology. Therefore, on the basis of the present findings, it can be deduced that the enhanced N2O conversion performance of Co3O4/CeO2-NR mixed oxides could be attributed to a compromise between redox and textural characteristics.



#### **3. Materials and Methods**

#### *3.1. Materials Synthesis*

In the present work, the chemicals that were used were of analytical reagent grade. Ce(NO3)3·6H2O (Fluka, Bucharest, Romania, purity ≥99.0%) and Co(NO3)2·6H2O (Sigma-Aldrich, Taufkirchen, Germany, purity ≥98%) were employed as precursor compounds for the preparation of bare ceria as well as Co/CeO<sup>2</sup> mixed oxides. Also, NaOH (Sigma-Aldrich, Taufkirchen, Germany, purity ≥98%) and ethanol (ACROS Organics, Geel, Belgium, purity 99.8%,) were used during materials synthesis. Initially, the hydrothermal method was applied for the preparation of bare ceria nanoparticles, as described in detail in our previous work [49]. In brief, ceria nanorods (CeO2-NR) were synthesized by dissolving NaOH (36.7 M) in double deionized water and then adding an appropriate

amount of an aqueous solution of Ce(NO3)3·6H2O (0.13 M) under vigorous stirring. Next, the transfer of the final slurry into a Teflon bottle and its aging at 90 ◦C for 24 h occurred. For the synthesis of ceria nanopolyhedra (CeO2-NP), a similar procedure was followed, employing, however, a lower amount of NaOH (6 M). In order to synthesize ceria nanocubes (CeO2-NC), the same procedure as the one described above for the synthesis of ceria nanorods was followed, with the obtained slurry to be aged at 180 ◦C instead of 90 ◦C. In all cases, centrifugation was used for the recovery of the solid products that were thoroughly washed with double deionized water until a neutral pH was reached and finally washed with ethanol so as to avoid the nanoparticles' agglomeration. Afterwards, drying of the precipitate at 90 ◦C for 12 h followed by calcination at 500 ◦C for 2 h under air flow (heating ramp 5 ◦C/min) was carried out.

The Co/CeO2-NX catalysts where NX stands for NP: nanopolyhedra, NR: nanorods and NC: nanocubes were prepared by wet impregnation, employing an aqueous solution of Co(NO3)2·6H2O, in order to achieve an atomic ratio Co/(Co+Ce) of 0.2 which corresponds to 7.8 wt. % of Co loading. Heating under stirring of the obtained suspensions until complete water evaporation occurred, followed by drying at 90 ◦C for 12 h and final calcination at 500 ◦C for 2 h under air flow (heating ramp 5 ◦C/min).

#### *3.2. Materials Characterization*

The porosity of the materials was evaluated by the N2-adsorption isotherms at −196 ◦C, using an ASAP 2010 (Micromeritics, Norcross, GA, USA) apparatus (from ReQuimTe Analyses Laboratory, Universidade Nova de Lisboa, Lisboa, Portugal). The samples were previously degassed at 300 ◦C for 6 h. The specific surface area was calculated by the Brunauer–Emmett–Teller (BET) equation [75].

Structural characterization was carried out by means of X-ray diffraction (XRD) in a PAN'alytical X'Pert MPD equipped with a X'Celerator detector and secondary monochromator (Cu Kα λ = 0.154 nm, 50 kV, 40 mA; data recorded at a 0.017º step size, 100 s/step) in the University of Trás-os-Montes e Alto Douro. The collected spectra were analyzed by Rietveld refinement using PowderCell software, allowing the determination of crystallite sizes by means of the Williamson–Hall plot.

The redox properties were assessed by Temperature Programmed Reduction (TPR) experiments in an AMI-200 Catalyst Characterization Instrument (Altamira Instruments, Pittsburgh, PA, USA), employing H<sup>2</sup> as a reducing agent. In a typical H2-TPR experiment, 50 mg of the sample (grain size 180–355 µm) was heated up to 1100 ◦C (10 ◦C/min) under H<sup>2</sup> flow (1.5 cm<sup>3</sup> ) balanced with He (29 cm<sup>3</sup> ). The amount of H<sup>2</sup> consumed (mmol g−<sup>1</sup> ) was calculated by taking into account the integrated area of TPR peaks, calibrated against a known amount of CuO standard sample [76,77].

The surface composition and the chemical state of each element were determined by X-ray photoelectron spectroscopy (XPS) analyses, performed on a VG Scientific ESCALAB 200A spectrometer using Al Kα radiation (1486.6 eV) in CEMUP. The charge effect was corrected using the C1s' peak as a reference (binding energy of 285 eV). The CASAXPS software was used for data analysis.

The samples were imaged by transmission electron microscopy (TEM). The analyses were performed on a Leo 906E apparatus (Austin, TX, USA), at 120 kV in the University of Trás-os-Montes e Alto Douro. The samples were prepared by ultrasonic dispersion, and a 400 mesh formvar/carbon copper grid (Agar Scientific, Essex, UK) was dipped into the solution for the TEM analysis.

#### *3.3. Catalytic Performance Evaluation*

The catalytic studies for the N2O decomposition took place in a quartz fixed-bed U-shaped reactor (0.8 cm i.d.) with 100 mg of catalyst loading (grain size 180–355 µm). The feed gas (1000 ppm N2O and 0 or 2 vol. % O2) total flow rate was 150 cm3/min which corresponds to a Gas Hour Space Velocity (GHSV) of 40,000 h−<sup>1</sup> . The analysis of the gases was performed by a gas chromatograph (SHIMADZU 14B). The apparatus is equipped with a TCD detector and two separation columns (Molecular Sieve 5A for O2, N<sup>2</sup> measurements and Porapack QS for N2O measurement). Prior to the catalytic activity measurements, the materials under consideration were subjected to further processing under He

flow (100 cm3/min) at 400 ◦C. In order to minimize the external and internal diffusion limitations, preliminary experiments concerning the influence of particle size and W/F ratio on deN2O catalytic performance were carried out. Based on these experiments, a catalyst particle size in the range of 180–355 µm was selected, in addition to a W/F ratio of 0.04 g s cm−<sup>3</sup> . The conversion of N2O (XN2O) was calculated from the difference in N2O concentration between the inlet and outlet gas streams, according to the equation

$$\text{X}\_{\text{N}\_2\text{O}}(\text{\textdegree\text{\textdegree\textdegree}}) = \frac{[\text{N}\_2\text{O}]\_{\text{in}} - [\text{N}\_2\text{O}]\_{\text{out}}}{[\text{N}\_2\text{O}]\_{\text{in}}} \times 100\tag{7}$$

The specific reaction rate (r, mol m−<sup>2</sup> s −1 ) of the N2O decomposition was also estimated using the following formula:

$$\mathrm{tr}\left(\mathrm{mol}\,\mathrm{m}^{-2}\,\mathrm{s}^{-1}\right) = \frac{\mathrm{X}\_{\mathrm{N\_2O}} \times [\mathrm{N\_2O}]\_{\mathrm{in}} \times \mathrm{F}\left(\frac{\mathrm{cm}^3}{\mathrm{min}}\right)}{100 \times 60\left(\frac{\mathrm{s}}{\mathrm{min}}\right) \times \mathrm{V\_m}\left(\frac{\mathrm{cm}^3}{\mathrm{mol}}\right) \times \mathrm{m\_{\mathrm{cal}}}(\mathrm{g}) \times \mathrm{S\_{\mathrm{BET}}}\left(\frac{\mathrm{m}^2}{\mathrm{g}}\right)} \tag{8}$$

where F and V<sup>m</sup> are the total flow rate and gas molar volume, respectively, at standard ambient temperature and pressure conditions (298 K and 1 bar), mcat is the catalyst's mass and SBET is the surface area.

#### **4. Conclusions**

In this work, three different ceria nanoshaped materials (nanorods, nanocubes and nanopolyhedra) were hydrothermally synthesized and used as supports for the cobalt oxide phase. Both single CeO<sup>2</sup> and Co/CeO<sup>2</sup> mixed oxides were catalytically assessed during the decomposition of N2O in the presence and absence of oxygen. For bare ceria samples, the following deN2O order was obtained: CeO2-NR (nanorods) > CeO2-NP (nanopolyhedra) > CeO2-NC (nanocubes). Most importantly, cobalt addition to the CeO<sup>2</sup> carriers greatly enhances the N2O decomposition, not affecting at all the order obtained for the bare ceria supports and clearly reflecting the key role of support. The present results clearly reveal the key role of support morphology on the textural, structural and redox properties, reflected then on the catalytic performance of Co3O4/CeO<sup>2</sup> mixed oxides. Among the different samples investigated, the cobalt-ceria nanorods (Co/CeO2-NR) exposing {100} and {110} facets showed the best deN2O performance, ascribed mainly to their abundance in Co2+ active sites in conjunction to their enhanced redox and textural properties.

**Author Contributions:** M.L. contributed to materials synthesis, results interpretation and paper writing; E.P. and N.K. contributed to catalytic evaluation studies; S.A.C.C. contributed to the materials characterization; M.K. contributed to the conception, design, results interpretation and writing of the paper; all authors contributed to the discussion and read and approved the final version of the manuscript.

**Acknowledgments:** The research work was supported by the Hellenic Foundation for Research and Innovation (HFRI) and the General Secretariat for Research and Technology (GSRT), under the HFRI PhD Fellowship grant (GA. no. 34252). This research has been cofinanced by the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH—CREATE—INNOVATE (project code: T1EDK-00094). This work was also financially supported by Associate Laboratory LSRE-LCM—UID/EQU/50020/2019—funded by national funds through FCT/MCTES (PIDDAC). S.A.C.C. acknowledges Fundação para a Ciência e a Tecnologia (Portugal) for Investigador FCT program (IF/01381/2013/CP1160/CT0007), with financing from the European Social Fund and the Human Potential Operational Program. We are grateful to Carlos Sá (CEMUP) for the assistance with the XPS measurements, Pedro Tavares (UTAD) for the TEM and XRD analyses and Nuno Costa (ReQuimTe) for the N<sup>2</sup> adsorption results.

**Conflicts of Interest:** The authors declare no conflict of interest.
