*3.5. CO<sup>2</sup> Hydrogenation*

The hydrogenation of carbon dioxide to value-added chemicals, such as methanol, has received considerable attention, in terms of environmental protection and sustainable energy. The significant role of copper-ceria interfacial sites in the CO<sup>2</sup> hydrogenation process has been confirmed by both theoretical and experimental studies [40,297,298]. In particular, metal-oxide interface plays a key role in CO<sup>2</sup> hydrogenation process, as it could provide the active sites for reactants adsorption, while these interfacial sites may stabilize the key intermediates [299]. In view of this fact, copper-ceria catalysts have shown higher selectivity in methanol than their zirconia-supported counterparts, as the copper-ceria interface favored the dispersion of copper and the oxygen vacancy formation, while the interaction between copper and ceria led to a decrease in copper particle size [300]. The interfaces between the defective CeO2-x and the highly dispersed Cu+/Cu<sup>0</sup> species are considered the active sites for methanol synthesis in the case of CuOx/CeO<sup>2</sup> system [301]. Furthermore, the different metal-support interactions between the two catalysts resulted in different reaction intermediates, namely, carbonates for the CuOx/CeO<sup>2</sup> catalysts and bicarbonates for the zirconia-supported ones, thus, resulting in different selectivity, with the copper-zirconia composites being highly selective in CO [300].

In view of the above aspects, the fine-tuning of the metal-support interface could lead to highly active and selective catalysts. Indicative studies towards adjusting the CO<sup>2</sup> conversion to methanol under similar reaction conditions are summarized in Table 5, and further discussed below.



42

The activity and selectivity of the CuOx/CeO<sup>2</sup> catalysts for methanol synthesis are greatly affected by the support morphology. Copper-ceria nanorods exposing {100} and {110} crystal planes exhibited the highest methanol yield (Table 5) as compared to nanocubes and nanoparticles, due to the strong interaction between the two oxide phases and the high copper dispersion [131]. Copper-ceria nanorods were also found to be more active than nanocubes, while exhibiting similar conversion performance with the nanoparticles, for carbonate (diethyl) hydrogenation [147]. The activity and selectivity of the CuOx/CeO2 catalysts for methanol synthesis are greatly affected by the support morphology. Copper-ceria nanorods exposing {100} and {110} crystal planes exhibited the highest methanol yield (Table 5) as compared to nanocubes and nanoparticles, due to the strong interaction between the two oxide phases and the high copper dispersion [131]. Copperceria nanorods were also found to be more active than nanocubes, while exhibiting similar conversion performance with the nanoparticles, for carbonate (diethyl) hydrogenation [147].

*Catalysts* **2019**, *9*, x FOR PEER REVIEW 40 of 57

In a similar manner, CuOx/CeO<sup>2</sup> catalysts led, mainly, to the production of CO at atmospheric pressures through the RWGS reaction, with the nanorod-shaped catalyst exhibiting higher CO<sup>2</sup> conversion (~50% at 450 ◦C) as compared to nanospheres (~40% at 450 ◦C), revealing the structure dependence of the RWGS [137]. The active intermediates are preferably formed on the {110} ceria exposed surface of the rod-like morphology, resulting in high catalytic performance [137]. In a similar manner, CuOx/CeO2 catalysts led, mainly, to the production of CO at atmospheric pressures through the RWGS reaction, with the nanorod-shaped catalyst exhibiting higher CO2 conversion (~50% at 450 °C) as compared to nanospheres (~40% at 450 °C), revealing the structure dependence of the RWGS [137]. The active intermediates are preferably formed on the {110} ceria exposed surface of the rod-like morphology, resulting in high catalytic performance [137].

Copper-ceria nanorods of various copper loadings were also investigated in the hydrogenation of carbonate to methanol, with the catalysts of ca. 20 wt.% Cu content exhibiting superior catalytic performance [306]. The copper content can significantly affect the mole fraction of Ce3<sup>+</sup> and Cu<sup>+</sup> species (Figure 22), and in turn, the methanol yield [306]. Copper-ceria nanorods of various copper loadings were also investigated in the hydrogenation of carbonate to methanol, with the catalysts of ca. 20 wt.% Cu content exhibiting superior catalytic performance [306]. The copper content can significantly affect the mole fraction of Ce3+ and Cu+ species (Figure 22), and in turn, the methanol yield [306].

**Figure 22.** Mole ratio of Ce3+ and Cu+ over the reduced Cu/CeO2 catalysts as a function of Cu content. Adapted from Reference [306]. Copyright© 2018, John Wiley and Sons. **Figure 22.** Mole ratio of Ce3<sup>+</sup> and Cu<sup>+</sup> over the reduced Cu/CeO<sup>2</sup> catalysts as a function of Cu content. Adapted from Reference [306]. Copyright© 2018, John Wiley and Sons.

In addition, a space-confined synthetic approach was applied for the synthesis of highly dispersed copper-ceria catalysts for RWGS, offering 100% CO selectivity at 300 °C and ambient pressure [307]. The enhanced catalytic performance was ascribed to the abundance in interfaces formed among the highly dispersed copper nanoparticles and the Ce3+ species, thus, favoring H2 In addition, a space-confined synthetic approach was applied for the synthesis of highly dispersed copper-ceria catalysts for RWGS, offering 100% CO selectivity at 300 ◦C and ambient pressure [307]. The enhanced catalytic performance was ascribed to the abundance in interfaces formed among the highly dispersed copper nanoparticles and the Ce3<sup>+</sup> species, thus, favoring H<sup>2</sup> spillover [307].

spillover [307]. The controlled synthesis of multicomponent systems could also be an effective approach for highly active and selective hydrogenation catalysts (e.g., [302–305,308]). For instance, the ternary composite consisting of Cu, ZnO, CeOx, supported on SBA-15 exhibited high catalytic activity for methanol synthesis (Table 5), due to the peculiar synergistic effects between the different counterparts [302]. Copper was considered to be the active site for hydrogen activation, while ZnO and CeOx oxides facilitated the CO2 adsorption and hydrogen spillover on the interfacial sites [302]. In a similar manner, the introduction of alumina to ceria carrier (Cu/AlCe) led to highly active composites (Table 5), mainly ascribed to the high copper dispersion [303]. Interestingly, highly active and selective CO2 hydrogenation catalysts can be obtained by co-adjusting the composition, structure The controlled synthesis of multicomponent systems could also be an effective approach for highly active and selective hydrogenation catalysts (e.g., [302–305,308]). For instance, the ternary composite consisting of Cu, ZnO, CeOx, supported on SBA-15 exhibited high catalytic activity for methanol synthesis (Table 5), due to the peculiar synergistic effects between the different counterparts [302]. Copper was considered to be the active site for hydrogen activation, while ZnO and CeO<sup>x</sup> oxides facilitated the CO<sup>2</sup> adsorption and hydrogen spillover on the interfacial sites [302]. In a similar manner, the introduction of alumina to ceria carrier (Cu/AlCe) led to highly active composites (Table 5), mainly ascribed to the high copper dispersion [303]. Interestingly, highly active and selective CO2hydrogenation catalysts can be obtained by co-adjusting the composition, structure and shape, in line with the fine-tuning methodology herein proposed (Figures 2 and 3). In this perspective, bimetallic

and shape, in line with the fine-tuning methodology herein proposed (Figures 2 and 3). In this perspective, bimetallic composites (Cu-Ni) incorporated into ceria nanoparticles of specific composites (Cu-Ni) incorporated into ceria nanoparticles of specific morphology (e.g., nanotubes, nanorods) could lead to highly active composites for the CO<sup>2</sup> hydrogenation to methanol (Table 5). The enhanced catalytic performance of Cu-Ni/CeO<sup>2</sup> nanoshaped catalysts was mainly interpreted on the basis of the synergistic interaction between Cu and Ni as well as of that between the ceria carrier and Cu-Ni alloy [304,305]. Similarly, the bimetallic Cu-Fe/CeO<sup>2</sup> catalyst has shown enhanced stability in the high-temperature RWGS due mainly to the fact that iron oxide clusters (FeOx) highly dispersed over ceria act as textural promoters [308]. Finally, copper nanocrystals encapsulated in Zr-based MOFs demonstrated high activity and selectivity for CO<sup>2</sup> hydrogenation to methanol, outperforming the benchmark Cu/ZnO/Al2O<sup>3</sup> catalyst [197].

#### **4. Outlook and Challenges**

In the present review, the copper-ceria binary system has been employed as a reference system to reveal the different approaches that can be followed to modulate the local surface chemistry, and in turn, the catalytic performance of metal oxides (MOs), by means of size, shape and electronic/chemical functionalization. More importantly, the fine-tuning of the above-mentioned parameters can affect not only the reactivity of metal sites in its own right, but also the interfacial activity (e.g., through the formation of oxygen vacancies, facilitation of redox interplay between the different counterparts, etc.) offering a synergistic contribution towards the development of NMs-free highly active and selective composites for several energy and environmental applications.

For instance, the employment of appropriate synthetic routes, such as the hydrothermal method, leads to the development of nanoparticles with specific morphologies, exposing distinct crystal facets of different coordination environments, with great implications in catalysis. Moreover, particles size/shape engineering strongly affects the interfacial reactivity through both geometric and electronic interactions, offering metal oxide systems with the desired properties. In addition, special pretreatment protocols or activation procedures can notably affect the metal dispersion and the population of oxygen defects, with great consequences in the catalytic efficiency. In view of the above, the fine tuning of metal oxides by combining bulk and nano effects or by adjusting the coordination environment could lead to highly efficient catalysts.

Besides the modulation of local surface chemistry by means of size and shape engineering, the electronic/chemical modification (e.g., alkali promotion, incorporation of rGO or g-C3N4, employment of MOFs) can be adopted as an additional functionalization tool to regulate the electronic environment and the oxygen exchange kinetics of MOs.

In view of the above aspects, highly active composites, with a comparable or even better performance than that of NMs, have been developed for various processes, such as CO oxidation, N2O decomposition, preferential oxidation of CO, among others. As for example, the combination of precipitation method with alkali promotion can lead to highly active and oxygen-tolerant CuOx/CeO<sup>2</sup> catalysts for N2O decomposition. On the other hand, the modulation of ceria support morphology (nanorods) by the hydrothermal method resulted in CuOx/CeO<sup>2</sup> composites with superior CO oxidation performance, even better to that of Pt-based catalysts. More importantly, the co-adjustment of different parameters (e.g., the shape of individual counterparts along with the electronic state of metal entities) could lead not only to distinct reactivity of each counterpart, but also to different synergistic interactions, offering mixed metal oxides of unique features. Hence, metal oxides appropriately adjusted by means of suitable synthetic and electronic/chemical modification routes could provide the materials platform for real-life energy and environmental applications.

Another approach towards the fine-tuning of metal oxides could be the employment of computational studies (e.g., DFT calculations) prior to the synthesis of the catalysts, providing, thus, the required feedback that would lead to the focused functionalization of specific parameters. This combinatorial theoretical and experimental approach could result in specific composites with predefined characteristics, while it would save precious time during experimental trials.

Last, but not least, the conclusions drawn from the present survey can provide the design principles for the development of low-loading NMs-based catalysts, paving the way for the decrease of noble metals content in energy and environmental applications in which their use is inevitable. In any case, the fundamental understanding of structure-property relationships is a prerequisite factor towards the rational design of efficient and inexpensive catalytic composites.

**Author Contributions:** M.L. contributed to paper writing; M.K. contributed to the conception, design and writing of the paper; All authors contributed to the discussion, read and approved the final version of the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research has been co-financed 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).

**Acknowledgments:** The authors would like to express their sincere gratitude to the anonymous reviewers for their constructive remarks that greatly contributed to improving the content and the scientific impact of this review article. The authors also would like to thank the editors for their efforts during the review process.

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