**Ceria Nanoparticles' Morphological Effects on the N2O Decomposition Performance of Co3O4/CeO<sup>2</sup> Mixed Oxides**

**Maria Lykaki <sup>1</sup> , Eleni Papista <sup>2</sup> , Nikolaos Kaklidis <sup>2</sup> , Sónia A. C. Carabineiro <sup>3</sup> and Michalis Konsolakis 1,\***


Received: 15 January 2019; Accepted: 18 February 2019; Published: 3 March 2019

**Abstract:** Ceria-based oxides have been widely explored recently in the direct decomposition of N2O (deN2O) due to their unique redox/surface properties and lower cost as compared to noble metal-based catalysts. Cobalt oxide dispersed on ceria is among the most active mixed oxides with its efficiency strongly affected by counterpart features, such as particle size and morphology. In this work, the morphological effect of ceria nanostructures (nanorods (NR), nanocubes (NC), nanopolyhedra (NP)) on the solid-state properties and the deN2O performance of the Co3O4/CeO<sup>2</sup> binary system is investigated. Several characterization methods involving N<sup>2</sup> adsorption at −196 ◦C, X-ray diffraction (XRD), temperature programmed reduction (TPR), X-ray photoelectron spectroscopy (XPS) and transmission electron microscopy (TEM) were carried out to disclose structure–property relationships. The results revealed the importance of support morphology on the physicochemical properties and the N2O conversion performance of bare ceria samples, following the order nanorods (NR) > nanopolyhedra (NP) > nanocubes (NC). More importantly, Co3O<sup>4</sup> impregnation to different carriers towards the formation of Co3O4/CeO<sup>2</sup> mixed oxides greatly enhanced the deN2O performance as compared to bare ceria samples, without, however, affecting the conversion sequence, implying the pivotal role of ceria support. The Co3O4/CeO<sup>2</sup> sample with the rod-like morphology exhibited the best deN2O performance (100% N2O conversion at 500 ◦C) due to its abundance in Co2+ active sites and Ce3+ species in conjunction to its improved reducibility, oxygen kinetics and surface area.

**Keywords:** ceria nanoparticles; morphological effects; Co3O4/CeO<sup>2</sup> mixed oxides, deN2O process

### **1. Introduction**

Nitrous oxide (N2O) is one of the most significant greenhouse gases contributing to the depletion of the ozone layer. N2O has a much higher global warming potential (GWP) compared to CO<sup>2</sup> (310 times higher) and a long atmospheric lifetime (114 years). The emissions of N2O are derived by both natural and anthropogenic sources. The main anthropogenic sources for N2O emissions involve agriculture (use of fertilizers), chemical industry (adipic and nitric acid production), the combustion of fossil fuels, as well as biomass burning, etc. [1–4].

The abatement of N2O emissions is of paramount importance and the direct catalytic decomposition of nitrous oxide to molecular nitrogen and oxygen (deN2O process) is considered to be a highly efficient remediation method. Thus far, several catalytic systems, such as supported noble metals [5–7], perovskites [8–10], hexaaluminates [11–14], spinels [15–18], zeolites [19–22] and mixed oxides [23–27], have been used for N2O decomposition. Although noble metals exhibit satisfactory activity for the deN2O process, their high cost and the deterioration of their catalytic efficiency from gases present in the exhaust gas stream (e.g., O2) act as inhibiting factors for practical applications [1,28]. Hence, research efforts have focused on the development of noble metal-free mixed oxides of high activity, stability and low cost, as recently reviewed [1].

Among the different transition metal oxides, cobalt spinel shows unique physicochemical characteristics, such as thermal stability and high reducibility, making it an excellent candidate for the deN2O process [15,23,29,30]. However, the high cost of cobalt renders mandatory its dispersion to high surface area supports like ceria, magnesia, etc. [31,32]. Among the various supports investigated, ceria exhibits unique redox properties associated with its high oxygen storage capacity (OSC), rendering this material highly effective in many catalytic processes [23,33–35]. Furthermore, the synergistic effects induced by strong metal–ceria interactions, in nanoscale, can modify the surface chemistry of the materials through geometric or/and electronic perturbations, leading to improved redox properties and catalytic activity [36–40].

However, the catalytic efficiency of transition metal oxides, involving ceria-based mixed oxides, can be considerably affected by the different counterpart characteristics, such as particle size and morphology. In this regard, engineering the particle size and shape (e.g., nanorods and nanocubes) through the employment of advanced nano-synthesis paths has lately received particular attention [33,41–43]. Interestingly, the support morphology greatly affects the redox properties, oxygen mobility and, subsequently, the catalytic activity of the mixed oxides. For instance, Lin et al. [44] prepared Co3O4/CeO<sup>2</sup> catalysts with three different support morphologies, namely polyhedra, nanorods and hexagonal shapes, with polyhedra exhibiting the highest catalytic activity for ammonia synthesis. In a similar manner, by tailoring the support morphology, CuO/CeO<sup>2</sup> nanoshaped materials of enhanced reducibility and deN2O performance can be obtained [45]. Andrade-Martínez et al. [46] investigated the catalytic reduction of N2O over CuO/SiO<sup>2</sup> catalysts, revealing the key role of the spherical ordered mesoporous support, along with its functionalization through copper addition, on the improved catalytic activity and stability, making this material comparable to noble metal-reported systems. Different support morphologies (rods, plates and cubes) have also been employed for the low temperature CH2Br<sup>2</sup> oxidation revealing the superiority of cobalt-ceria nanorods in the catalytic performance [47]. Moreover, cobalt oxide supported on ceria of different morphology (nanoparticles, nanorods and nanocubes) has been investigated for the catalytic oxidation of toluene with the nanoparticles exhibiting the highest catalytic activity due to the synergism at the interface between the two oxide phases, which leads to an improved reducibility [48]. Very recently, the influence of support morphology (nanorods, nanocubes and nanopolyhedra) on the surface and structural properties of CuO/CeO<sup>2</sup> mixed oxide has been thoroughly explored through both *in situ* and *ex situ* characterization techniques. The results disclosed the significance of the ceria morphology on the reducibility and oxygen kinetics, revealing the order nanorods > nanopolyhedra > nanocubes [49].

In this work, ceria structures of various morphologies (nanopolyhedra, nanorods and nanocubes) were hydrothermally prepared, and then cobalt was impregnated into the above ceria supports. purpose of this work was to explore the impact of support morphology on the surface chemistry The

> and the deN2O performance of Co3O4/CeO<sup>2</sup> mixed oxides. The results clearly revealed that support morphology can exert a profound influence on the N2O decomposition, paving the way toward the rational design of highly efficient deN2O catalysts.

#### **2. Results and Discussion**

#### *2.1. Textural/Structural Analysis (BET and XRD)*

The main textural and structural characteristics of bare ceria samples and Co3O4/CeO<sup>2</sup> mixed oxides (hereinafter denoted as Co/CeO2) are summarized in Table 1. According to the BET surface area, the following order is acquired: CeO2-NP (88 m<sup>2</sup> g −1 ) > CeO2-NR (79 m<sup>2</sup> g −1 ) > CeO2-NC (37 m<sup>2</sup> g −1 ). The addition of cobalt into CeO<sup>2</sup> decreases the surface area, without, however, significantly affecting the order obtained for bare ceria samples. The Co/CeO2-NR sample exhibits the highest value (72 m<sup>2</sup> g −1 ) succeeded by Co/CeO2-NP (71 m<sup>2</sup> g −1 ) and Co/CeO2-NC (28 m<sup>2</sup> g −1 ). Regarding the average pore diameter and pore volume, they both decreased upon the addition of Co to ceria nanorods and nanocubes. However, concerning ceria nanopolyhedra, the addition of cobalt leads to a small increase in the average pore diameter, whereas the pore volume is not significantly affected (Table 1).


**Table 1.** The textural and structural properties of bare CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

<sup>1</sup> Calculated applying the Williamson–Hall plot after the Rietveld refinement of diffractograms.

Figure 1a shows the Barret-Joyner-Halenda (BJH) desorption pore size distributions (PSD) of bare ceria and Co/CeO<sup>2</sup> catalysts. According to the pore size distribution, all the samples have their maxima at a pore diameter more than 3 nm, designating the presence of mesopores [50]. It is obvious that bare ceria samples with the nanocube (CeO2-NC) and nanorod morphology (CeO2-NR) exhibit similar particle size distributions, whereas in ceria nanopolyhedra (CeO2-NP), a narrower PSD is observed. Noteworthily, PSD remains practically unaffected by the addition of cobalt in all cases. As it can be observed in Figure 1b which shows the adsorption–desorption isotherms, all samples demonstrate type IV isotherms with a hysteresis loop at a relative pressure > 0.5, further corroborating the mesoporous structure of the materials [51,52].

The XRD patterns of the samples are shown in Figure 2. The main peaks can be indexed to (111), (200), (220), (311), (222), (400), (331), (420), (422), (511) and (440) planes which are attributed to ceria face-centered cubic fluorite structure (Fm3m symmetry, no. 225) [53,54]. There are three small peaks at 2θ values of approx. 36, 44 and 64<sup>o</sup> which are typical of Co3O<sup>4</sup> [33]. These three diffraction peaks correspond to the (311), (400) and (440) planes of Co3O4, respectively. The average crystallite diameter of the oxide phases (CeO<sup>2</sup> and Co3O4) was assessed by an XRD analysis by means of the Williamson–Hall plot (Table 1). The CeO<sup>2</sup> crystallite size measurements showed 24, 14 and 11 nm for Co/CeO2-NC, Co/CeO2-NR and Co/CeO2-NP, respectively. As it is obvious from Table 1, there is a small decrease in the ceria crystallite size for nanocubes and nanorods, whereas no changes were observed for nanopolyhedra, indicating that the structural characteristics of ceria supports do not get significantly affected upon cobalt addition, as it will be further corroborated by a TEM analysis (see below). In a similar manner, the BET analysis (Table 1) indicates no significant modifications on the pore characteristics of ceria nanopolyhedra upon cobalt addition, which could be ascribed to their irregular morphology. It should be also noted that the samples with nanocube morphology exhibit the smallest BET surface area and the largest CeO<sup>2</sup> and Co3O<sup>4</sup> crystallite sizes in comparison to nanorods and nanopolyhedra. As for the crystallite size of cobalt oxide phase, the following sequence was obtained: Co/CeO2-NC (19 nm) > Co/CeO2-NR (16 nm) > Co/CeO2-NP (15 nm), which perfectly matches the order obtained for CeO2.

**Figure 1.** (**a**) The BJH (Barret-Joyner-Halenda) desorption pore size distribution (PSD) and (**b**) the adsorption–desorption isotherms of CeO2 and Co/CeO2 samples. **Figure 1.** (**a**) The BJH (Barret-Joyner-Halenda) desorption pore size distribution (PSD) and (**b**) the adsorption–desorption isotherms of CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

The XRD patterns of the samples are shown in Figure 2. The main peaks can be indexed to (111), (200), (220), (311), (222), (400), (331), (420), (422), (511) and (440) planes which are attributed to ceria face-centered cubic fluorite structure (Fm3m symmetry, no. 225) [53,54]. There are three small peaks at 2θ values of approx. 36, 44 and 64o which are typical of Co3O4 [33]. These three diffraction peaks correspond to the (311), (400) and (440) planes of Co3O4, respectively. The average crystallite diameter of the oxide phases (CeO2 and Co3O4) was assessed by an XRD analysis by means of the Williamson– Hall plot (Table 1). The CeO2 crystallite size measurements showed 24, 14 and 11 nm for Co/CeO2- NC, Co/CeO2-NR and Co/CeO2-NP, respectively. As it is obvious from Table 1, there is a small decrease in the ceria crystallite size for nanocubes and nanorods, whereas no changes were observed for nanopolyhedra, indicating that the structural characteristics of ceria supports do not get significantly affected upon cobalt addition, as it will be further corroborated by a TEM analysis (see below). In a similar manner, the BET analysis (Table 1) indicates no significant modifications on the pore characteristics of ceria nanopolyhedra upon cobalt addition, which could be ascribed to their irregular morphology. It should be also noted that the samples with nanocube morphology exhibit the smallest BET surface area and the largest CeO2 and Co3O4 crystallite sizes in comparison to nanorods and nanopolyhedra. As for the crystallite size of cobalt oxide phase, the following sequence was obtained: Co/CeO2-NC (19 nm) > Co/CeO2-NR (16 nm) > Co/CeO2-NP (15 nm), which perfectly

**Figure 2.** The XRD patterns of the CeO2 and Co/CeO2 samples. **Figure 2.** The XRD patterns of the CeO<sup>2</sup> and Co/CeO<sup>2</sup> samples.

#### *2.2. Morphological Characterization (TEM) 2.2. Morphological Characterization (TEM)*

Transmission electron microscopy (TEM) has been applied so as to examine the morphological differences among the materials. Figure 3a–c shows the TEM images of ceria supports. The CeO2-NR sample (Figure 3a) exhibits a rod-shaped morphology with the length varying between 25 and 200 nm. Figure 3b and c demonstrates mainly irregular-shaped nanopolyhedra and cubes, respectively. Figure 3d–f illustrates the images derived by TEM analysis for the Co/CeO2 mixed oxides. Evidently, the morphology is not affected by cobalt addition to the ceria carrier. Transmission electron microscopy (TEM) has been applied so as to examine the morphological differences among the materials. Figure 3a–c shows the TEM images of ceria supports. The CeO2-NR sample (Figure 3a) exhibits a rod-shaped morphology with the length varying between 25 and 200 nm. Figure 3b and c demonstrates mainly irregular-shaped nanopolyhedra and cubes, respectively. Figure 3d–f illustrates the images derived by TEM analysis for the Co/CeO<sup>2</sup> mixed oxides. Evidently, the morphology is not affected by cobalt addition to the ceria carrier.

*Catalysts* **2019**, *9*, x FOR PEER REVIEW 6 of 20

**Figure 3.** The transmission electron microscopy images of CeO2 (**a**–**c**) and Co/CeO2 (**d**–**f**) samples: (**a**) CeO2-NR, (**b**) CeO2-NP, (**c**) CeO2-NC, (**d**) Co/CeO2-NR, (**e**) Co/CeO2-NP and (**f**) Co/CeO2-NC. **Figure 3.** The transmission electron microscopy images of CeO<sup>2</sup> (**a**–**c**) and Co/CeO<sup>2</sup> (**d**–**f**) samples: (**a**) CeO<sup>2</sup> -NR, (**b**) CeO<sup>2</sup> -NP, (**c**) CeO<sup>2</sup> -NC, (**d**) Co/CeO<sup>2</sup> -NR, (**e**) Co/CeO<sup>2</sup> -NP and (**f**) Co/CeO<sup>2</sup> -NC.

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-

*2.3. Redox Properties (H2-temperature programmed reduction (TPR))*

determines the reaction rate (*vide infra*).

→ CoO → Co, respectively [44,56–58].

calibrated against a known amount of CuO standard sample.

**Sample** 

The reduction profiles of the Co/CeO2 samples as well as the one of a Co3O4 reference are shown in Figure 4b. Table 2 summarizes the main TPR peaks along with the hydrogen consumption (mmol H2 g−1). Pure Co3O4 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 Co3O4

**Table 2.** The redox properties of the bare CeO2 and Co/CeO2 samples.

**Ratio** 

**Os Peak Ob Peak Total Os Peak Ob Peak** 

**Peak Temperature (°C)** 

**H2 consumption (mmol H2 g−1) a Os/Ob**

CeO2-NP 0.48 0.51 0.99 0.94 555 804 CeO2-NR 0.59 0.52 1.11 1.13 545 788 CeO2-NC 0.41 0.58 0.99 0.71 589 809 Peaks a+b CeO2 Peak Total Peak a Peak b Co/CeO2-NP 2.40 0.61 3.01 333 388 Co/CeO2-NR 2.37 0.62 2.99 318 388 Co/CeO2-NC 2.05 0.32 2.37 335 405 a Estimated by the area of the corresponding temperature programmed reduction (TPR) peaks, which is

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
