Microwave, Ultrasound, and Mechanochemistry: Unconventional Tools that Are Used to Obtain “Smart” Catalysts for CO₂ Hydrogenation

Abstract

The most recent progress obtained through the precise use of enabling technologies, namely microwave, ultrasound, and mechanochemistry, described in the literature for obtaining improved performance catalysts (and photocatalysts) for CO₂ hydrogenation, are reviewed. In particular, the main advantages (and drawbacks) found in using the proposed methodologies will be discussed and compared by focusing on catalyst design and optimization of clean and efficient (green) synthetic processes. The role of microwaves as a possible activation tool used to improve the reaction yield will also be considered.

1. Introduction

Atmospheric CO₂ concentration is growing continuously: in 2017, a “record concentration” of 413 ppm was registered (according to https://www.co2.earth/daily-co2) and a concentration of ca. 570 ppm is expected to be reached by the end of this century [1].

The mitigation of CO₂ (the most popular among atmospheric greenhouse gases) is currently a scientific, technological, social, and economic issue [1]. Several strategies have been proposed to limit/decrease CO₂ concentration in the atmosphere: besides lowering CO₂ emissions with the help of low-C fuels (and of appropriate policies from developed countries), there is a great spur of studies and technologies aimed at optimizing several processes of CO₂ capture & storage [2,3,4,5,6,7,8,9,10] and reutilization [11,12]. The first two steps, currently fairly developed, are indeed functional to CO₂ reutilization processes, which mostly need a concentrated and rather pure CO₂ stream apt to further chemical transformations.

The chemical processes for CO₂ reutilization are catalytic [13,14,15], photocatalytic [16,17,18,19,20], electrocatalytic [21,22,23,24,25], or photoelectrocatalytic [18,26]. As a whole, the chemical reduction of CO₂ is a challenge from both the industrial and the catalytic point of view [27], but it is also a hot scientific topic, since CO₂ is an ideal C1 feedstock for producing different types of chemicals and, ultimately, fuels. Unfortunately, its high thermodynamic stability limits the number of processes in which CO₂ can be utilized as a chemical feedstock (e.g., the synthesis of urea), and usually electrocatalytic processes are needed [28].

Most of the catalytic processes (like those reported in Scheme 1) imply the use of H₂, which is per se a costly reagent due to its scarce natural abundance (from which arises the issue of its production), along with safety and transportation issues. All this notwithstanding, the products of CO₂ reduction are ultimately fuels, raw materials, and/or chemical intermediates [1].

Though some homogenous catalysts are highly active and selective in several CO₂ reduction processes [29], they are hardly applicable on an industrial scale, and therefore there is a great interest towards the development of efficient heterogeneous catalysts and/or the immobilization of homogeneous ones [30]. For the industrial application of CO₂ hydrogenation processes, heterogeneous catalysts (e.g., Fe-, Cu-, and Ni-based solid catalysts [31]) offer some practical advantages compared to homogeneous ones: solids with large specific surface area, small particles of active phase, and fair metal dispersion are usually more active and selective, and also more stable and characterized by longer life. However, they often provide low yields and poor selectivity, mainly due to fast kinetics of C–H bond formation [1].

In order to allow industrial applications of heterogeneous catalysis, green processes should be developed, which often require efficient (nano)catalysts that are able to selectively transform CO₂ into added value molecules [27]. Depending on the reaction adopted for CO₂ reduction, different active phases (mostly Ni, but also more noble metals [31]) have to be supported on different oxides (γ-Al₂O₃ being the most used one [31]) that have to be stable at high temperatures, provide a certain amount of acid/base sites, and stabilize the catalytically active metal particles, etc.

Whatever the type of CO₂ hydrogenation process, active, selective, and stable heterogeneous catalysts are preferable in view of an actual industrial application [1]. They should also be based on earth-abundant and non-toxic elements and obtained by green synthesis methods: in one word, they should be “smart”. Though the outputs of CO₂ reduction are diverse [1], “smart” catalysts have to meet some requirements that can be obtained by optimizing their preparation strategy, for instance. The physico-chemical properties of a heterogeneous catalyst are determined and controlled by its preparation method: therefore, catalysts prepared by different means may have markedly different features and, thus, different catalytic performances [32,33].

In order to allow an actual development of feasible industrial CO₂ hydrogenation processes, new catalysts should be obtained by methods that are able to control and tailor the physico-chemical properties in order to improve the catalyst performance [32]. Moreover, a deeper understanding of the reaction mechanisms with the help of both experiments and molecular simulations is needed [1], especially as far as the first step of CO₂ reduction, i.e., its adsorption at the catalyst surface, is concerned. Notwithstanding the plethora of works concerning CO₂ reduction, a thorough understanding of the mechanisms is still lacking and/or debated [18]. As far as heterogeneous catalysts are concerned, there is a general consensus about the idea that the active site is provided by the primary catalyst (e.g., the metal particle) acting synergistically with the support (and/or the promoter) [32]. Evidence of such synergy is provided in several works, but the actual mechanism is still unclear, even for the synthesis of methanoic acid (HCOOH), which is only the first step of CO₂ hydrogenation [1].

From both industrial and scientific points of view, the great interest in heterogeneous catalysts based on abundant and low cost metals like Ni, Cu, and Fe is frustrated by their poor activity in mild reaction conditions, in that an extreme dispersion of metal particles and a strong interaction with the support should be attained [31].

As mentioned before, from an environmental point of view, catalysts should be obtained by green synthesis methods [27]. In our opinion, the use of microwaves, ultrasound, or mechanochemistry represents promising and simple approaches, which are alternatives to conventionally adopted synthetic protocols, to obtain novel catalysts with improved activity and selectivity, in contrast to conventional methods, which usually encompass multi-step processes, heating, and/or the addition of expensive and/or hazardous chemicals. The development of such techniques has increased significantly in recent years, with most reports published over the past 5–10 years (vide infra).

The aim of the present review is to supply a brief (but exhaustive) description of the most recent advances in the synthesis of solid catalysts (and photocatalysts) by microwave (MW), ultrasound (US), and mechanochemistry (MC) to be compared to conventional methods.

2. Microwaves as a Tool to Improve Catalytic CO₂ Hydrogenation

2.1. General Principles of MW-Assisted Synthesis of (Nano)Materials

MW-assisted synthesis simply requires the presence of a MW reactor, simultaneously ensuring shorter reaction time, no contact between the radiation source and the reacting mixture, and a homogeneous heating of the latter [34,35].

Homogeneous heating is a crucial point, as thermal gradients affect both nucleation and growth steps, which finally determine the physico-chemical properties of the resulting nanomaterials. For this reason, MW-assisted synthesis was initially developed to prepare ceramic materials, for which it is crucial to avoid temperature gradients that may induce formation of fractures in the final product [36].

MW irradiation favors fast nucleation and crystalline growth, which can accelerate the formation of hierarchical structures, like the bismuth oxyiodide (BiOI) architectures shown in Figure 1 [37]. Hierarchical porous solids can be produced by MW-assisted synthesis in non-aqueous solvents, since organic solvents with high dielectric loss factor are required to ensure efficient MW absorption and rapid heating [37].

Additionally, MW heating can be easily coupled with several chemical methods including combustion synthesis (CS), hydrothermal (HT) synthesis, and sol-gel (SG). For instance, powders obtained by MW-assisted CS are more chemically and thermally stable than those obtained by “conventional” CS; moreover, smaller particles with a lower tendency to aggregation are obtained [36]. This occurs because MW irradiation uniformly heats the reagents mixture, whereas an electric oven heats a mixture from the exterior to the interior, finally creating a temperature gradient and leading to heterogeneous powders. Moreover, MW interacts with polar reagents, inducing dipole oscillations: the oscillating species collide more frequently and at higher energy, facilitating bond breakage and new bond formation. Consequently, the heat needed for the reaction is generated within the mixture with no need of external heat supply [34,35].

MW-assisted CS, for instance, allows obtaining sizeable amounts of phases (mainly oxides, nitrides, and carbides) that are stable at high temperatures and have nanometer size, highly specific surface area and defectivity, and mechanically stable pore structure [36]. MW is also known to favor the formation of given crystalline phases and can be efficiently used to drive the synthesis towards the desired products. All the above parameters are indeed important for catalytic applications.

2.2. MW-Assisted Syntheses of Heterogeneous Catalysts for CO₂ Reduction

Cai et al. [38] supported some Cu-based catalysts on ZnO and on Ga₂O₃ and tested them in the CO₂ hydrogenation to CH₃OH:

CO₂ + 3H₂ ⇌ CH₃OH + H₂O

(1)

The Cu-based catalysts were prepared by co-precipitation and impregnation, and the effect of MW on both preparation methods was studied: as a whole, the catalysts prepared by MW-assisted synthesis showed better Cu dispersion and enhanced catalytic activity in terms of both CO₂ conversion and CH₃OH yield, especially when a single-step MW-assisted method was adopted [38]. The catalysts were characterized by X-Ray Diffraction (XRD), H₂ temperature programmed reduction (TPR), N₂ adsorption/desorption isotherms at −196 °C, CO₂ temperature-programmed desorption (TPD), adsorption calorimetry, and X-ray Photoemission Spectroscopy (XPS). Since all the samples are calcined at 350 °C, no significant changes in their surface area were observed, although the sample obtained by a single-step, MW-assisted method showed slightly larger values for both surface area and pore volume. The XRD patterns of the same sample revealed the likely presence of small crystallites of Cu₂O (whereas CuO mostly formed in the other samples) and the smallest size of ZnO particles. TPR showed the occurrence of a strong Cu/support interaction in the sample obtained by single-step, MW-assisted method, which, according to CO₂ TPD and CO₂ adsorption calorimetry, adsorbed a considerable amount of CO₂ with the largest average adsorption heat value (ca. −20 kJ mol⁻¹). This latter result showed that the use of MW seemed to improve the support basicity, a feature that helps CO₂ adsorption, a crucial step during the catalytic process. The joint presence of well-dispersed Cu particles and strong basic sites at the support surface resulted in strong Cu/support synergy, leading to better catalytic performance. Besides strong Cu/support interactions, which took place with the catalysts prepared under MW irradiation, the authors observed that Cu particles were particularly active, probably because their surface was partially oxidized due to migration of Zn species from the support to the active metal (Cu) surface, which finally improved CO₂ adsorption. On the contrary, such synergistic effects were not observed with same composition catalysts that were prepared by conventional methods.

Huang et al. [39] studied the effect of MW heating during HT synthesis of CuO-ZnO-ZrO₂ catalysts for CO₂ hydrogenation to CH₃OH (Equation (1)), for which CuO-ZnO is usually considered the best catalyst and where ZrO₂ has the role of imparting high thermal and chemical stability under redox conditions. This type of catalyst is usually prepared by HT method, requiring long reaction times: the (expected) effect of MW irradiation was indeed the shortening of preparation time. Different temperatures (i.e., 80, 100, 120, and 150 °C) were adopted during the MW-assisted HT method to obtain the catalysts named CZZ80, CZZ100, CZZ120, and CZZ150 [39].

The catalyst obtained at 120 °C (CZZ120) showed the best catalytic properties in terms of both CO₂ conversion and selectivity to CH₃OH, becoming rather stable after 100 h reaction. XRD analysis allowed the calculation of the size of both CuO and Cu crystallites: the CZZ120 catalyst showed the smallest CuO and Cu crystallites before and after reduction, respectively, though Cu crystallites (ca. 18 nm) were larger than parent CuO crystallites (ca. 7.6 nm). The same catalyst was also characterized by the largest Cu area (S_Cu); the largest specific surface area and by a higher amount of CO₂ molecules adsorbed by relatively weak basic sites. CuO crystallites in CZZ120 were easily reduced, as the samples showed the lowest reduction temperature during H₂ TPR analysis. The CH₃OH yield was directly related to the surface of Cu particles (S_Cu), as shown in Figure 2. On the contrary, XPS analysis did not show significant differences in the type of surface Cu species in catalysts obtained at different temperatures. As a whole, an optimum temperature existed for the synthesis of CuO-ZnO-ZrO₂ catalysts [39]: Cu particle size, surface area, and basicity exhibited a volcano trend with a maximum at 120 °C.

CO₂ methanation (Equation (2)) is a process that allows transformation of CO₂ in a fuel through an exothermic reaction (∆H⁰ = −165 kJ mol⁻¹), for which Ni/Al₂O₃ is the commercial catalyst [40]:

CO₂ + 4H₂ ⇌ CH₄ + 2H₂O

(2)

Besides Al₂O₃, other supports can be (mostly) SiO₂ and TiO₂, ZrO₂ or CeO₂. When other (more noble) metals like Ru, Pt, Pd, or Au are used, selectivity is lower with respect to Ni, as they do not produce exclusively CH₄, but also CH₃OH and CO (by Reversed Water Gas Shift, RWGS):

CO₂ + H₂ ⇌ CO + H₂O

(3)

As a whole, Ni dispersion is crucial in the catalytic activity towards CO₂ methanation [41,42,43], and therefore the catalyst synthesis method plays a prominent role [44,45,46]: according to Stangeland et al. [40], catalysts should be developed with improved activity at lower temperatures and high activity/stability up to ca. 500 °C. Conventional co-precipitation and impregnation methods may require several hours in order to obtain a fair Ni dispersion, whereas under MW irradiation the preparation time may decrease from hours to 15 min (this is one of the greatest advantages of using MW irradiation during co-precipitation and/or impregnation).

Moreover, MW-assisted synthesis of Ni/Al₂O₃ catalysts for CO₂ methanation [47] resulted in higher Ni dispersion, as well as higher basicity of the support with respect to Ni/Al₂O₃ catalysts obtained by impregnation. Physico-chemical characterization of the catalysts showed some relevant differences: for instance, Raman spectroscopy revealed a higher amount of NiO oxide in the sample prepared by impregnation, which also showed lower surface area (≅143 m² g⁻¹ against 163 m² g⁻¹). The most important property was, however, the Ni dispersion: it was likely favored by the fact that MW helps the dispersion of inorganic salts on the support, as it can be clearly seen in Figure 3, showing the TEM analysis of two Ni/Al₂O₃ catalysts produced with and without MW assistance. The average size of Ni particles obtained by MW assisted synthesis was 10 nm and 16 nm in the sample prepared by impregnation. The sample obtained by MW-assisted synthesis also consumed ca. 25% more H₂ during TPD analysis.

The better Ni dispersion obtained by MW joint with a stronger basicity of support (as determined by CO₂ TPD) likely led to enhanced CO₂ methanation activity at lower temperature. The corresponding CH₄ selectivity was also maintained over the whole temperature range (200–400 °C), and the catalyst was rather stable under prolonged (72 h) catalytic tests without appreciable agglomeration of Ni species [47].

MW can be used during catalyst preparation in different manners: for instance, the effect of MW drying vs. conventional drying in the preparation of Ni-exchanged zeolites for CO₂ methanation (2) has been recently studied [48]. The catalysts were prepared by incipient wetness impregnation of USY (Ultra Stabilized Y) zeolite with 5 wt % Ni. The effects of drying method after impregnation, calcination temperature, and pre-reduction temperature on the catalyst performance were evaluated. The drying method deeply affected the catalyst performance: in particular, MW-assisted drying induced remarkable changes on the type, location, and reducibility of Ni species within the zeolite structure and, simultaneously, led to effects on the support structural and textural properties and on the average size of Ni particles.

DR-UV-Vis spectroscopy showed that after conventional drying, most NiO species occurred at the outer surface of zeolite particles, whereas with MW-assisted drying, octahedral Ni²⁺ species formed if drying was performed in a closed container (when MW-assisted drying was performed in an open container, similar Ni species were obtained as they were during conventional drying). According to the authors, during MW heating in a close container, higher temperature and pressure led to formation of Ni²⁺ species in exchange position within the zeolite cavities, whereas if the container was left open during drying, NiO species mostly occurred at the outer surface of zeolite particles.

MW-assisted drying also deeply affected zeolite crystallinity: in a closed container, some amorphous phase formed (as detected by X-Ray powders diffraction), likely due to the high temperature reached within the container, whereas with MW-assisted drying in an open container, the zeolite crystallinity was preserved (as for conventional drying). Finally, MW-assisted drying led to less reducible Ni species, especially when octahedral Ni²⁺ species occurred. As observed with other Ni-based catalysts, MW-assisted drying led to formation of smaller Ni particles (in the 17–27 nm range). Unfortunately, the positive effects brought about by MW-assisted drying, like better Ni dispersion, were counterbalanced by the loss in zeolite crystallinity and, as a result, no appreciable improvement in the catalytic activity was obtained in terms of both CO₂ conversion and selectivity to CH₄. This was likely due to the fact that the support was a zeolite, for which the crystallinity of porous cavities is a crucial aspect [48].

The material thermal response to MW heating is a crucial aspect, as shown by Dong et al. [49], with Re promoted Ni-Mn/Al₂O₃ catalysts for partial methanation (Scheme 2), a process in which the amount of (noxious) CO is reduced by coupling CO methanation (∆H⁰ = −206 kJ mol⁻¹):

CO + 3H₂ ⇌ CH₄ + H₂O

(4)

to water gas shift (WGS, ∆H⁰ = −41 kJ mol⁻¹)

CO + H₂O ⇌ CO₂ + H₂

(5)

Figure 4 reports heating curves under MW irradiation (1.0 kW output power) of the different reagents used to produce the catalysts: Al₂O₃ is a poor MW absorber, but when Re was firstly incorporated in γ-Al₂O₃ (Re-Al₂O₃), the materials showed heating curves with higher slope (curves 3 and 4 in Figure 4). Due to the presence of Re, both the Re-Al₂O₃ and Ni-Mn/Re-Al₂O₃ materials absorbed more MW with respect to pure γ-Al₂O₃. This had a deep impact on the reaction time, in that the synthesis temperature (ca. 400 °C) was reached shortly when Re-Al₂O₃ was used for the deposition of the bimetallic Ni-Mn catalyst starting from the respective nitrates.

Other phenomena occurring during MW heating can be inferred by a close inspection of the curves in Figure 4: when Re is absent, the temperature plateau reached at ca. 150 °C is ascribed to thermal balance between (Ni and Mn) nitrates decomposition and MW irradiation. The phenomenon does not occur in the heating curves of Re-added samples, likely because of the enhancement of thermal decomposition of nitrates due to the presence of Re. Simultaneously, MW heating favors the formation of hot spots that induce an effective surface dispersion of the metal. A side-effect of fast MW heating was observed, i.e., formation of AlO(OH) phase (as detected by X-ray diffraction). MW heating, however, favored the dispersion of the NiO phase, which showed broader X-ray diffraction peaks with respect to the catalyst obtained by conventional heating. A better dispersion of the Ni active phase was also observed after reduction, confirming the overall positive effect of MW. At variance with what was observed with Ni-USY zeolite, MW heated samples were more easily reducible, as the onset of NiO reduction shifted to lower temperatures. Another important point observed by the authors was that MW-heated catalysts led to lower formation of carbon deposits, an undesired phenomenon that can lead to catalyst deactivation [49].

The presence of Re, however, also affected some important structural and electronic properties of the catalysts, as shown by the characterization of their physico-chemical properties. For instance, H₂ TPR analysis showed a higher reducibility of β-type NiO (a type of NiO that has medium strong interaction with γ-Al₂O₃), indicating the facile formation of catalytically active Ni atoms. XPS analysis pointed out also a decrease of the binding energy of the Ni 2p line when Re was present, ascribed to the electron transfer from ReO_x species to Ni atoms. The increased electron cloud density of Ni atoms could indeed favor the breakage of the C-O bond, positively affecting the catalytic activity.

CO₂ reforming of methane (Equation (6)) is an endothermic reaction (∆H⁰ = +247 kJ mol⁻¹, also known as dry reforming) to produce syngas mixtures with low H₂:CO ratio that could be used for further Fischer-Tropsch synthesis (FTS) to produce light olefins:

CO₂ + CH₄ ⇌ 2CO + 2H₂

(6)

The reaction has not yet been industrially implemented, since there are no commercial catalysts that can operate without undergoing deactivation due to carbon deposition [50]. An alternative to Ni-based catalysts (and to noble metals like Ru, Rh, Pd, Ir, and Pt) is provided by Ni-containing oxides, i.e., either Ni perovskites with defined structure or solid solutions with La, Al, or Mg (spinel, fluorite, and perovskite) [51]. Ni perovskites with defined structure are promising catalysts, with respect to noble metals, due to their low cost, thermodynamic stability at relatively high temperatures, and catalytic activity. Barros et al. [51] prepared La₂NiO₄/α-Al₂O₃ catalysts for CO₂ reforming of CH₄ by MW-assisted CS, by using glycine or urea as fuel and metal nitrates as oxidizers. For comparison, a catalyst with similar chemical composition was prepared by impregnation of α-Al₂O₃ prepared by MW.

XRD diffraction showed the formation of La₃Ni₂O_6.92, La₂NiO₄, and LaAl_xNi_x−1O₃ crystalline structures independently of the preparation method. The sample prepared by impregnation was very active, but the crystalline structure was practically destroyed, with significant carbon deposition. The catalysts prepared by MW-assisted CS showed CH₄ conversion above 80% at 700 °C, good stability, and a low coke formation after about 60 h on stream. The catalyst prepared by impregnation showed instead a decrease of CH₄ conversion after 60 h, likely due to deactivation by coke deposition, according to TPO (Temperature Programmed Oxidation) analysis and TEM inspection, which showed the occurrence of carbon nanotubes, not observed when the catalysts were obtained by MW-assisted CS (Figure 5).

Moreover, the crystalline structure of the catalysts prepared by MW-assisted CS was preserved after CO₂ reforming of CH₄, whereas a prominent loss of crystallinity occurred in the catalyst obtained by impregnation (which was also deactivated) [51]; the authors ascribed the stability achieved by MW-assisted CS to the strong interaction of Ni active sites with the perovskite structure.

CO₂ selective reduction to light olefins is a process that recently has progressed due to the urgent need of converting CO₂ into building-block chemicals [13,14,52]. To raise selectivity towards light olefins, catalysts that are more effective are needed, since CO₂ is very stable, and FTS usually leads to mixtures of products. K-promoted Fe catalyst is very selective towards CO₂ hydrogenation to light olefins; additionally to K, Zn is also used as a promoter that accelerates FTS, increases catalyst stability, enhances light olefins selectivity, improves CO₂ adsorption, promotes H₂ dissociation, and is highly active in both WGS and RWGS reactions [52]. The effect of MW was studied on the synthesis of HT precipitation followed by incipient wet impregnation (to add K) of a series of K-promoted Fe-Zn catalysts with different Fe/Zn molar ratios. Physico-chemical characterization of the catalysts showed they were characterized by rather uniform particle size (ca. 100 nm). XRD analysis showed the Zn addition to the K-promoted Fe₂O₃ matrix led to formation of ZnFe₂O₄ spinel phase and ZnO phase, and also brought about an increase of the surface area. Moreover, the Fe-Zn interaction affected the adsorption and reduction of CO₂ adsorption.

The catalysts exhibited excellent activity for CO₂ hydrogenation and high selectivity toward C₂–C₄ olefins owing to the promoting effect of Zn, due to the fact that particles with uniform shape were obtained, and a fair dispersion of active components occurred with respect to conventional co-precipitation. Additionally, the formed ZnFe₂O₄ phase imparted stability to the catalyst, although lowering its activity due to segregation of a fraction of Fe (the active phase).

2.3. MW-Assisted Synthesis of Photocatalysts for CO₂ Reduction

Photocatalytic CO₂ reduction processes (i.e., biomimetic artificial photosynthesis) are very challenging due to intrinsic limitations of the used semiconductors, like poor absorption in the visible range and fast electron/holes recombination rate (e.g., for TiO₂) and stability (e.g., for CdS). “Plasmonic” photocatalysis is a plasmonic metal-enhanced photocatalytic process occurring when noble metal nanoparticles (NPs) decorate the surface of a semiconductor (e.g., TiO₂) to improve its photocatalytic efficiency through the localized surface plasmon resonance (LSPR) effect of NPs [53].

Most of the studies on the subject concern either Au or Ag NPs supported on TiO₂, but their actual use is hampered by the cost and the limited abundance of the noble metals. Recently, it has been shown that also Cu NPs can have LSPR effect, indicating that Cu could replace noble metals.

Cu NPs-decorated TiO₂ nanoflower films (Cu/TiO₂ NFFs) were prepared using both HT method and MW-assisted reduction process. Polyvinylpyrrolidone was used as capping agent and NaH₂PO₂*H₂O as reducing agent in diethylglycole solution under MW irradiation. The catalysts were tested in the photocatalytic reduction of CO₂ to CH₃OH [54].

Physico-chemical characterization of the films showed that both Cu NPs and TiO₂ films were characterized by a fair dispersion of Cu NPs on the TiO₂ NFFs (Figure 6) and by visible light harvesting properties, due to LSPR of Cu NPs and the peculiar nanostructure of the TiO₂ film. Moreover, the presence of Cu NPs allowed fluorescence quenching due to the suppression of charge carriers. The CH₃OH production rate reached 1.8 mmol cm⁻² h⁻¹ (with an energy efficiency of 0.8%) with a 0.5 wt % Cu/TiO₂ NFFs under UV and visible light irradiation: with Cu NPs, the production rate was ca. 6.0 times higher than with pure TiO₂ film (Figure 7).

The positive photocatalytic performance towards CO₂ reduction to CH₃OH observed when Cu NPs were present (Figure 7) was ascribed to a synergistic mechanism of Cu NPs at the surface of TiO₂ (Scheme 3): Cu NPs obtained by MW-assisted reduction method likely suppressed the recombination of charge carriers and improved the charge transfer efficiency.

In another study, Cu-Graphene Oxide (Cu/GO) composite photocatalysts were obtained using a simple and rapid one-pot MW process [55], shown in Figure 8a.

By using Cu(NO₃)₂ as metal precursor, short MW heating (ca. 3 min at 400 W) was performed in ethylene glycol solution (Figure 8a), leading to a fair dispersion of crystalline Cu NPs in the GO.

The obtained Cu NPs (with a size of ca. 4–5 nm, Figure 8) were strongly interacting with GO: the obtained composite was particularly active under solar irradiation in the production of both CH₃OH and acetaldehyde. Cu NPs were shown to enhance the GO photocatalytic activity of GO, essentially by suppressing electron-hole recombination, and also by reducing the GO bandgap and by modifying its work function (Scheme 4). The presence of Cu NPs greatly enhanced absorption of GO in the visible range, though the shift in the band gap by effect of Cu NPs was not well understood. XPS analysis also showed that oxygenated groups of GO (i.e., -COOH and C-O-C groups) were almost completely removed by the MW treatment. No other Cu oxides were detected, indicating that the MW heating led to the reduction of all the Cu ions.

XPS analysis [55] showed the likely occurrence of a charge transfer from GO to Cu NPs, and a correlation was found between the Cu content and the rate of formation and selectivity of the products. The Cu/GO catalyst with 10 wt % Cu was more than 60 times more active than pure GO and ca. 240 times more active than commercial TiO₂ (Degussa P25) photocatalyst under visible light.

Also, Cu₂O can be used for CO₂ photoconversion: An et al. [56] produced Cu₂O/RGO (Reduced Graphene Oxide) composites using a facile, one-step, MW-assisted HT synthesis. X-ray diffraction showed that the Cu₂O was likely the only Cu-containing phase in the composites, and MW-assisted synthesis allowed the formation of composites where intimate contact between RGO sheets and Cu₂O microspheres (a result of the strong affinity between the metal oxide and the abundant functional groups of GO) was enhanced. Such strong interaction between Cu₂O and RGO was attributed to the adopted MW-assisted synthesis: such interaction was very effective in that RGO positively affected the photocatalytic activity, improving both activity and stability of Cu₂O, due to an efficient charge separation and transfer to RGO (Scheme 5).

Cu-decorated graphitic carbon nitride (g-C₃N₄) nanosheets were obtained by a procedure depicted in Scheme 6, in which the final step involved the MW-assisted HT synthesis of Cu NPs [57].

The composites were tested for the photocatalytic reduction of CO₂ to CO. During MW-assisted HT synthesis, crystalline Cu NPs were obtained: some of them were deposited at the outer surface of nanosheets, and other were wrapped within them, as shown by TEM pictures in Figure 9. Such an intimate contact between the two phases likely enhanced the transfer of electrons and positively affected the photocatalytic activity.

The presence of Cu NPs also affected the ability of absorbing visible light, as shown by DR UV- Vis spectroscopy: Cu NPs likely facilitate the light harvesting process; moreover, the absorption edge of the composites shifted towards longer wavelength, indicating that a higher number of electrons/holes could be generated.

Photocatalytic tests showed that Cu had a positive effect of the CO yield. An optimal amount of Cu was found, corresponding to 6 wt % Cu. That content was a kind of tradeoff between 3 wt % (a too low Cu content) and 9 wt %, in which a higher number of Cu NPs can became centers for electrons/holes recombination. Scheme 7 reports the proposed mechanism, in which photogenerated electrons excited in g-C₃N₄ migrate to the surface of Cu NPs, acting both as trapping centers and catalytic sites, for the photocatalytic CO₂ reduction to CO. Simultaneously, photogenerated holes migrated to the surface of the photocatalyst for water oxidation to O₂.

CdS is one of the most promising candidate semiconductors as photocatalysts active in the Vis range: Yu et al. [58] prepared RGO–CdS nanorods composites in ethanolamine–water solution using a one-step MW-assisted HT method. The composites exhibited a high activity in the photocatalytic reduction of CO₂ to CH₄, even without noble metal catalyst (i.e., Pt, which is usually adopted as co-catalyst). MW-assisted HT synthesis requires shorter time and less energy than conventional HT synthesis. Furthermore, particles with uniform size are usually obtained. Raman and XPS spectroscopies confirmed the reduction of parent GO using the MW-assisted method. TEM analysis (Figure 10) showed an intimate connection between the two materials.

An optimal RGO content of 0.5 wt % was found, leading to the highest CH₄ yield. Higher RGO content led to lower CH₄ yields; the authors ascribed this result to RGO shielding light absorption by CdS nanorods. RGO acted as co-catalyst, enhancing the photocatalytic reduction of CO₂ to CH₄, with the best performance attained with a RGO content of 0.5 wt %; the corresponding CH₄ production rate was 2.51 mmol h⁻¹ g⁻¹, which exceeded that of pure CdS nanorods by more than 10 times. According to the authors, RGO was able to adsorb more CO₂ molecules and to activate them very efficiently, simultaneously generating a higher number of photogenerated electrons on its surface, finally leading to an enhancement of CO₂ reduction performance of the RGO–CdS composite (Scheme 8).

In another work, MW were found to positively affect the immobilization of homogeneous Co(III) cyclam complex (cyclam is 1,4,8,11-tetraazacyclotetradecane, namely: [Co(cyclam)Cl₂]Cl) at the surface of SBA-15 silica. The photocatalytic activity towards CO₂ reduction to CO was studied in the presence of p-terphenyl as a photosensitizer [59]. Usually, Co(III) cyclam immobilization is performed by reflux method, but MW-assisted immobilization of the Co(III) cyclam complex occurred more rapidly and led to more uniform Co(III) surface species. Moreover, a smaller amount of Co(III) complex was needed with respect to reflux method. This latter aspect led to well-defined, molecularly dispersed Co(III) sites within SBA-15 mesopores. IR spectroscopy showed that the complex interacts with the silica surface mainly by chemisorption on silanol (≡SiOH) groups. The photocatalyst prepared by MW-assisted synthesis was more active and selective than that prepared by reflux method, in the presence of a molecular photosensitizer and sacrificial electron donors (Figure 11).

3. Ultrasound as Unconventional Activation Method

3.1. General Principles of US-Assisted Synthesis of (Nano)Materials

US is essentially a mechanical longitudinal wave with frequency equal to or higher than 20 kHz, which is the upper limit of human hearing range. US propagation gives sinusoidal variation in the static pressure in the liquid medium inducing cavitation, a phenomenon controlled by external parameters, i.e., US frequency and intensity and type solvent. Cavitation is based on radical and mechanical effects. The radical effect occurs either at the bubble interface or in the interior cavity and is ascribed to the occurrence of the sonolysis of molecules. The mechanical effect, occurring after the cavity collapse, arises from shear forces, microjets, and shock waves outside the bubble, therefore producing severe physical changes in the presence of solids or metals. The transient implosive collapse of the bubble is an adiabatic phenomenon, during which inside the bubble very high temperature and pressure occur (~4600 °C and ~50 MPa). The phenomenon affects the reaction system [60] by generating both an intense local convection in the reaction medium (finally enhancing the mass transfer processes) and highly reactive radical species within the bubbles, due to high local temperature and pressure. Upon fragmentation of the bubble, the radicals generated therein get released into the medium, where they can induce/accelerate chemical reactions [61].

On the one hand, US has emerged as a potential green technique for the intensification of different physical, chemical, and biological processes. Compared with conventional heating, the use of US irradiation reduced the reaction time from hours to minutes. On the other hand, US can be effectively employed in the synthesis of NPs with a high production yield, without using expensive instruments and/or extreme conditions [62,63], as, for example, carbon encapsulated Fe NPs with superior catalytic activity and reusability for hydrogenation reaction in the liquid phase [64]. US allow obtaining of highly reactive surfaces, enhancing mixing and increasing mass transport. Therefore, such unconventional method can be particularly helpful for preparing and activating catalytic materials [65].

3.2. US-Assisted Synthesis of Cu–ZnO Catalysts and of Their Improved Formulations

Under typical industrial conditions, CH₃OH is formed mostly via CO₂ hydrogenation, with CO acting as CO₂ source and as oxygen scavenger from water, which inhibits the active metal sites [66]. A synergism between Cu⁰ and Cu⁺ sites is the most adopted explanation for the activity of Cu/ZnO-promoted catalysts [38,61,67,68]; however, Cu⁰ sites [69,70] and/or Cu–Zn alloy [71,72,73] have been proposed to have a catalytic role as well.

Many unconventional preparation routes have been adopted to obtain Cu–ZnO catalysts with improved activity, selectivity, resistance to poisons, and enhanced lifetime [38,74,75,76,77,78]. In this frame, the SG method was largely adopted in order to increase the metal surface area along with the total surface area [79,80]. Moreover, the excellent activity of oxalate co-precipitated catalysts was also demonstrated [81], and Cu–Zn catalysts prepared by “spark-erosion” showed superior performance in terms of selectivity [82], with noticeable TOF in the CO₂ hydrogenation reaction that was ascribed to a close contact among the phases in Cu/ZnO/ZrO₂ catalyst produced by combustion.

To further improve the physico-chemical properties of a Cu–ZnO/ZrO₂ catalyst (Zn_at/Cu_at, 0–3; ZrO₂, 42–44 wt %), a new synthetic procedure consisting in the US-assisted reverse co-precipitation was proposed [83]. The US irradiation guaranteed highly increased total surface area (BET, 120–180 m² g⁻¹), although the Zn/Cu atomic ratio strongly affected either the specific surface area or the volume of the pores. Indeed, the materials with low (≤0.1) and high (>0.7) Zn/Cu ratios have lower surface area than those with a Zn/Cu ratio of 0.3–0.4, corresponding to a maximum of 175 m²/g. An analogous volcano trend was observed for the pore volume and pore diameter, in which the catalyst with Zn/Cu ratio equal to 0.4 contains mainly mesopores with diameter in the 70–120 Å range and, as a consequence, an enhanced cumulative volume (0.6 cm³ g⁻¹). Moreover, the authors found that the US irradiation led to very high dispersion (3–58%) and exposure of Cu (S_Cu = 9–63 m² g⁻¹). The catalytic activity in the CH₃OH production by CO₂ hydrogenation was explored in a range of temperatures (160–260 °C) and pressures (1.0–3.0 MPa), and compared with a commercial Cu–ZnO/Al₂O₃ catalyst with optimum results ascribed to the presence of highly dispersed Cu particles.

The direct synthesis of dimethyl ether starting from syngas involves two consecutive reactions, i.e., CH₃OH synthesis and CH₃OH dehydration [84]. Alternatively, dimethyl ether can be directly obtained from syngas or CO₂ hydrogenation.

The use of novel bi-functional catalysts based on metal oxides (namely, CuO, ZnO, Al₂O₃, and possibly some precious metal oxides, such as Cr₂O₃) to synthesize CH₃OH and on acid solids (namely, HZSM-5 and HY zeolites or SAPOs) to convert it into dimethyl ether can significantly enhance the yield with respect to the two-step process, because CH₃OH dehydration can interfere with the equilibrium of the synthesis reaction [85].

In this contest, Cu-ZnO-Al₂O₃, as well as Al₂O₃ and HZSM-5 catalysts, have been widely investigated [86,87,88]. In particular, a previous study reported on the preparation of CuO-ZnO-Al₂O₃/HZSM-5 hybrid catalysts (CZA/HZSM-5) using different techniques, such as impregnation, co-precipitation-physical mixing method, and a US-assisted co-precipitation procedure [89]. It was observed that the CZA/HZSM-5 catalysts prepared by US-assisted co-precipitation displayed higher activity than the systems obtained by conventional physical mixing and impregnation. Further, various catalysts with different CZA and HZSM-5 wt. ratio (namely, 2, 3, and 4) were synthesized by the US-assisted co-precipitation procedure summarized in Figure 12 in order to establish a correlation between the catalyst composition and the displayed activity [90]. The experimental conditions for US irradiation in Ar atmosphere were 90 W for 45 min, with 10 min current pulse time and 0.5 min rest time, by means of the experimental apparatus shown in Figure 13.

It is worth noting that the zeolite preserved its crystal structure during the preparation. Moreover, the crystallinity was significantly influenced by varying of the CZA content. XRD peaks related to monoclinic CuO, and hexagonal ZnO phases were observed. FESEM analyses indicated the presence of a uniform coating of the HZSM-5 surface by CZA upon US irradiation during the synthesis. In addition, the catalysts consisted of many small particles with almost spherical shape. At variance with reducibility, the surface area was not dependent on the CZA content. The HZSM-5 structure was not damaged upon loading CZA, as demonstrated by FTIR spectroscopy results. The optimal composition of the CZA/HZSM-5 catalysts for the conversion of syngas to dimethylether was 4:1 at 275 °C and 40 bar.

Interestingly, it was reported that the sonication time significantly affected both morphology and structure of the CZA/HZSM-5 catalyst [91]. In particular, larger CuO crystals formed at longer irradiation times, and at the same time the particle aggregates changed into smaller size and more uniform shape, resulting in an increase of the surface area. The catalyst synthesized under long US irradiation showed very good activity during time on stream test, whilst the non-sonicated sample lost its activity.

The effect of US irradiation power was evaluated on different CZA/HZSM-5 catalysts prepared by US-assisted co-precipitation method [92]. The results of a thorough physico-chemical characterization evidenced that the nature of the precursors and the time of irradiation strongly affected the morphology, the surface area, the functional groups, and the structure of the final materials. In particular, as shown in Figure 14, highly irradiated acetate-based catalysts had uniform morphology, as well as smaller CuO particles with higher surface area, due to high nucleation rate and high cooling rates, resulting in a stronger interaction among the components in comparison to the catalysts prepared by starting from nitrate precursors.

It is worth noting that an increase of the irradiation power gave rise to smaller particles, resulting in an improved dispersion and surface area of the catalysts. Moreover, CuO particles with lower crystallinity were generated at high power, resulting in an enhanced interaction among the metal oxide particles. In addition, such ultrasound-assisted co-precipitation method allowed to produce CZA/HZSM-5 catalysts that were more stable with respect to those prepared by conventional precipitation, for which 18% CO conversion and 58% dimethyl ether yield decreases in 24 h time on stream test were observed.

The authors proposed a pathway for the ZnO synthesis under ultrasound irradiation in CZA that is illustrated in Figure 15. Such pathway can be extended to other precursors, as well as to CuO and Al₂O₃. More specifically, the metallic precursors underwent decomposition and hydratation upon ultrasound irradiation, and H· and OH· radicals formed by water sonolysis [93]. The free OH· radicals formed in the cavities or at interface can react with the metal cations forming metal hydroxides that are converted to the corresponding metal oxides upon thermal treatment. Simultaneously, by co-precipitation of the metal precursor, the metal carbonates (that will then be converted into the metal oxides) are produced. Therefore, a strong synergistic action of US during co-precipitation was advised. In particular, the OH· radicals coming from the water sonolysis are able to convert a fraction of the metal salt to the corresponding metal oxide without any help from the precipitant. Moreover, a significant increase of the nucleation rate under peculiar temperature and pressure conditions upon cavity collapse gives rise to the formation of small and well-dispersed NPs. In particular, the irradiation cycles impede the growth of the nucleated particles, resulting in the production of small catalyst particles. Finally, particle agglomeration is prevented, indeed, the populations of OH· radicals and of the nucleation sites are enhanced at high US power, and the conversion to metal oxides is therefore improved.

An US-assisted co-precipitation method was employed for the preparation of CZA/HZSM-5 catalysts supported on multi-walled carbon nanotubes pretreated with H₂SO₄/HNO₃ mixture [94], in which the acid pretreatment had the aim of activating and functionalizing the surface of the nanotubes. The acid-pretreated carbon nanotubes were dispersed in deionized water by sonication more easily than the pristine ones, because the presence of the surface functional groups increased the capability to form hydrogen bonds, converting the surface of the multi-walled carbon nanotubes from hydrophobic to hydrophilic, therefore resulting in improved dispersion and suspension stability. The CZA NPs of about 10–20 nm were quite homogeneously distributed and densely deposited at the surface of the acid-pretreated, multi-walled carbon nanotubes (as shown in Figure 16).

The catalysts were tested in the hydrogenation of CO₂ to produce dimethyl ether in a fixed-bed reactor. It was shown that the presence of the multi-walled carbon nanotubes has a promoting effect on the catalytic activity and on the dimethyl ether. The CO₂ conversion was 46.2%, with dimethyl ether yield and selectivity of 20.9% and 45.2%, respectively, at 262 °C, 3.0 MPa, volume ratio H₂/CO₂ = 3, and space velocity = 1800 mL g_cat⁻¹h⁻¹.

3.3. US-Assisted Synthesis of Catalysts with Ferromagnetic Properties

The synthesis of nanostructured materials based on ferromagnetic elements is of great interest due to the possibility to tune their magnetic properties [95,96,97], as well as to promote physico-chemical surface processes [98,99] by controlling the shape and size of the obtained NPs. In addition, magnetic catalysts can be easily positioned and recovered by using an external magnetic field.

In a recent paper, Vargas et al. reported a facile two-step UP procedure to synthesize Ni NPs and then support them on m-ZrO₂ to obtain low-cost and active catalysts for CO₂ methanation (Equation (2)) [100]. More in detail, Ni NPs were firstly produced by chemical reduction of a metal precursor (nickel chloride) with either hydrazine hydrate (N₂H₄·H₂O) or sodium borohydride (NaBH₄) in water, in the presence of citric acid and sodium citrate as protecting agents. The chemical reduction was carried out under US irradiation (40 kHz, 300 W) without any additional stirring sources. Then, the black colored solution, indicating the presence of Ni⁰ NPs, was added to other aqueous suspensions of commercial monoclinic ZrO₂. In order to obtain homogenous materials, the obtained mixtures were maintained overnight in a conventional ultrasonic bath at ambient conditions. The two samples were centrifuged, washed several times with a mixture of deionized water and ethanol, and dried under vacuum. The catalysts had the same metal loading (1 wt % Ni), but different metal particle sizes. It was found that the final particle size was strongly related to the employed reducing agent. Particularly, homogeneous Ni particles with globular shape and average size of 40 ± 1 nm formed in the presence of hydrazine, as shown in Figure 17A.

On the contrary, when NaBH₄ was employed, smaller and well isolated Ni NPs with average diameter equal to 2.7 ± 0.4 nm were observed (Figure 17B). Such differences can be explained by considering that in the case of N₂H₄*H₂O, the formation of the [Ni(N₂H₄)₃]²⁺ complex implied a slower kinetics, hence inhibiting the formation of nuclei and favoring particle growth. Instead, the use of sodium borohydride gave rise to a larger number of Ni nuclei and, consequently, to smaller particles. Hydrodynamic diameters larger than those measured by TEM were detected in both samples (Figure 17C) in agreement with the presence of citrate at the surface of NPs and of the occurrence of dynamics dipolar magnetic interactions. Peaks at 44.7°, 52.1°, and 76.7° 2 Theta degrees were observed in the diffraction pattern of the catalyst with large Ni particles (Ni(40) in Figure 18c), assigned to the (111), (200), and (220) reflections of the cubic Ni phase (Figure 17D). As for the catalyst with small NPs, (Ni(3) in Figure 18c), only the (111) peak was detected. The catalysts were tested in the CO₂ methanation reaction, revealing high activity (70%) and excellent selectivity to CH₄ (90%) at 720 K.

Ni/ZrO₂ catalysts modified with rare earth elements, in particular Ni/(Zr-Sm)Ox catalysts, showed improved activity in the CO₂ methanation (Equation (2)) [101,102]. Conventional wet impregnation method or preparation starting from amorphous Ni-Sm alloys were the preparation procedures usually adopted for these catalysts. In particular, the activity of these systems was correlated with the presence of (Zr–Sm)Ox units with the tetragonal ZrO₂ structure. Moreover, the catalytic performance was also influenced by preparation method, surface area, and pore size distribution of the catalysts [101,103].

A Ni-based (40 at % Ni) catalyst supported on mesoporous yttria-stabilized-zirconia composite was prepared by a US method using sodium dodecyl sulfate as a templating agent to obtain high surface area (193 m²/g) [104]. The application of the US method resulted in short reaction time (6 h) and no requirement for the glycolation of nickel, yttrium, and zirconium ions. In another paper by the same research group, a simple one-step US-assisted synthesis (in which the corresponding inorganic salts were used as precursors) was applied to prepare 20–40 at % Ni supported on mesoporous ZrO₂ [105]. In these catalysts, Ce and Sm rare earth ions were introduced in the support to further enhance its activity towards CO₂ methanation. The obtained catalysts possessed in situ generated mesopores, and the Ni oxide (or hydroxide) that were about 10 nm were incorporated into tetragonal ZrO₂ modified by the rare earth metals. In particular, the catalyst containing 30 at % Ni loading had maximum porous volume and size, and at the same time displayed the highest activity, which was further enhanced upon proper oxidation–reduction activation treatment, during which additional active centers were produced.

CO₂ methanation was also performed over Ni/Al₂O₃ catalysts prepared by a US-assisted co-precipitation [106]. The authors observed that an increase in the Ni loading from 5 wt % up to 25 wt % produced a surface area enhancement, resulting in better catalytic performances, and simultaneously decreased the crystallinity and improved the catalyst reducibility. A further increase of the metal loading had a negative effect on the surface area, as well as on the catalytic activity, due to the drop of the Ni dispersion. The 25 wt % Ni/Al₂O₃ catalyst attained high CO₂ conversion (74%) and almost complete CH₄ selectivity (99%) at 350 °C. The catalyst demonstrated also high stability at the same temperature after 10 h time-on-stream. The catalytic performance was observed to decrease at higher Gas Hourly Space Velocity, due to both reduced contact time between the reactants and the catalyst surface, as well as the amount of adsorbed reactants. Conversely, a positive effect on the catalyst activity was observed when higher H₂/CO molar ratio was employed. Furthermore, the performance of catalysts calcined at higher temperature was observed to decrease.

In another recent study, Co/SBA-15 and Co/SBA-16 catalysts with 20 wt % Co loading were prepared by conventional incipient wetness impregnation method [107]. Additional vacuum and US treatments were performed after the synthesis. N₂ physisorption, small angle XRD (SAXRD), wide angle XRD, SEM and energy dispersive X-ray spectroscopy (SEM/EDX), TEM, and TPR were employed to characterize the catalysts. It was observed that Co dispersion on the SBA-15 support was enhanced upon such post-synthesis treatments, due to the decrease of the metal crystallite size. The presence of Co₃O₄ and of different various Co_xO_y species was detected. Both post-impregnation treatments affected the metal-support interaction, increasing the activity in the CO₂ hydrogenation. It was observed that the vacuum treatment gave rise to a material in which crystalline cobalt oxide was located in the inner pores, whilst cobalt oxide agglomeration at the outer surface took place upon the US treatment.

A series of 0.5–5.0% Ru catalysts supported on Al₂O₃ for the selective methanation of CO in H₂-containing streams were investigated with the aim to clarify the influence of process parameters, such as the metal loading, the temperature of calcination, the chlorine ions content, and the space velocity, on the catalytic performances [108]. Careful characterization by XRD, Thermal Gravim- etry/Differential Thermal Gravimetry (TG/DTG), and SEM was performed. A number of laboratory experiments were performed to clarify the influence of process parameters (Ru loading, calcinations temperature, space velocity, and chlorine ions content) on the activities of catalysts. It was observed that by carrying out the synthesis of the catalysts by US-assisted impregnation, the catalytic activity was significantly improved, and the impregnation time was significantly reduced. In particular, agglomeration accompanied by non-homogeneous distribution of Ru deposits on Al₂O₃ were observed when the catalysts were prepared by wetness impregnation method. Conversely, US impregnation strongly decreased the particle size and improved the distribution of Ru during the impregnation. The best results were achieved by the catalyst containing 2 wt % Ru on Al₂O₃ calcined at 400 °C. Long-term tests demonstrated high activity and stability under realistic reaction conditions.

4. The Role of Mechanochemical Activation in the Synthesis of Nanostructured Catalysts

4.1. General Principles of MC

MC induces chemical transformations by mechanical action (compression, shear or friction) that allows breaking intramolecular bonds and further chemical reactions. More in detail, radical species are mechanochemically produced due to the breaking of weak bonds and to the extreme surface plasma conditions generated by the mechanical impact. Mechanochemical activation processes have a long history, and their importance does not decrease with time, because many solid-phase reactions (with stoichiometric amount of reactants) can be quantitatively and rapidly promoted under solvent-free conditions [109]. Therefore, the mechanochemical mixing process can be an environmentally advantageous alternative to conventional chemical syntheses, in which solvents often play a key role in energy dispersion and chemicals solvation and transportation [110].

The preparation of nanomaterials for advanced applications by mechanochemical synthesis is among the most suitable alternative routes because of its simplicity and high reproducibility. Indeed, such methodology is also environmentally advantageous in comparison to conventional preparation procedures, since it is often solvent-free, and it is carried out under mild reaction conditions and involving short reaction times.

4.2. MC as a Tool for the Synthesis of Catalysts

The MC synthesis of heterogeneous catalysts strongly affects their catalytic properties. Indeed, the excess of potential energy produced during grinding, together with shear and friction forces, can deeply modify the materials by producing a large variety of defects, hence improving or modifying the reactivity [109,111].

A variety of heterogeneous catalysts, ranging from supported NPs to nanocomposites has been obtained by using MC synthesis, as summarised in Scheme 9 [109].

MC catalyst activation has been successfully applied for many years. Mori et al. [112] investigated the influence of the mechanical milling of MgO-mixed catalysts of Ru, Ni, Fe, or their combinations at temperatures ranging from 80 to 150 °C under initial pressures of 100 Torr CO₂ and 500 Torr H₂ on the hydrogenation of CO₂ to CH₄. They found that the CH₄ yield was related to the nature of the catalyst (Ru/MgO gave the best yield, whilst Ni/Fe showed the worst activity). Interestingly, the MC activation produced catalysts with improved CH₄ production rate and significantly lowered the reaction activation energy. Indeed, an apparent activation energy of 39 kJ mol⁻¹ and kJ mol⁻¹ was accomplished for Ni-Fe-MgO and Ru-MgO catalysts, respectively, upon the milling activation. Such values were almost halved with respect to the same systems produced without mechanical activation. Interestingly, the authors found that the catalysts submitted to further milling activation pre-treatment did not shown any improvement in the activity, suggesting that the milling during reaction played the major role. SEM revealed an increased number of larger particles at increasing milling time, due to small particle agglomeration during milling. However, no significant increase of surface area was observed. Electron probe microanalysis (EPMA) results revealed uniform distribution and aggregation of Ni, Fe, and MgO particles during milling, indicating that the activity had to be ascribed to changes in the surface structure rather than to increased surface area.

In the same years, another study reported on the influence of the MC activation of several traditional catalysts such as Ni, Zr, Zr-Ni-containing materials, Zr hydride, and different Zr-Ni hydrides employed in the hydrogenation of CO and of CO₂ [113]. Inactive NiO and ZrO₂ were also investigated for comparison purposes. In particular, the study underlined the difference between the traditional catalytic hydrogenation and the mechanically induced hydrogenation to CH₄. The study was carried out by employing a flow-milling vial as a catalytic reactor. The authors found that Zr and Zr-Ni hydrides were the most active in CH₄ production, despite the fact the hydrogen present in the mixture had a negative effect, since it overcame the CH₄ formation. On the contrary, Ni-based catalysts were unable to produce CH₄, but were active in CO disproportionation. Zirconium hydride was the only one that showed activity for CO₂ hydrogenation. Unexpectedly, the production of CH₄ was observed on NiO and ZrO₂. The differences in activity were ascribed to the structural transformations of the metal and hydride catalysts under milling, i.e., to the presence of low-coordinate centers upon MC treatment that were able to efficiently activate CO.

MC synthesis was also proven to be effective for obtaining nanophase carbides by facile ball milling of elemental iron and carbon powders at room temperature. Such catalysts were tested in the hydrogenation of CO₂ and showed superior activity and selectivity with respect to Fe/C mixtures and coarse-grained conventionally synthesized carbides [114]. It was proposed that nanocrystalline iron carbides with sizes ranging between 8 and 16 nm were the catalytically active species. The presence of defects and/or highly active sites, strained regions, and grain boundaries formed during the grinding guaranteed effective hydrocarbons chain propagation. Interestingly, also a commercial cementite powder displayed some activity upon the ball-milling activation.

Re–Co/Al₂O₃ catalysts were synthesized via ball milling by using tungsten carbide balls in a tungsten carbide container [115]. In contrast to the preparation of catalysts containing noble metal NPs, in which an oxide precursor or a metal salt are usually employed, the preparation of catalysts with transition metal involves the direct milling of the metal precursors. The obtained bimetallic Re–Co-containing catalysts were more active in the CH₄ conversion than in the CO hydrogenation, whilst the activity of the catalysts prepared by incipient wetness impregnation was exactly the opposite. Such difference was explained by assuming that crystallinity in MC prepared material was introduced upon the activation heat treatment at 650 °C. Indeed, X-ray powder diffraction indicated the presence of disordered Re–Co phases before the heat treatment.

Similar systems, such as Co–Fe and Ni catalysts supported on ZrO₂ and TiO₂, were also prepared by ball milling and showed enhanced catalytic activity in CO hydrogenation [116]. It is worth noting that upon mechanical treatment of TiO₂ in the anatase form, the authors observed the transition to brookite and rutile already at room temperature. The activity was ascribed to Co50Fe50 and Ni domains with average size of about 30 and 50 nm, distributed on the supports. XRD measurements confirmed that the mechanochemical dispersion of the Co50Fe50 and Ni phases on TiO₂ and ZrO₂ prevented the Co, Fe, and Ni oxidation by reaction with the supports. Oxidation of such species usually takes place during calcination during preparation of the catalysts.

Among the most effective and economic materials able to hydrogenate CO₂ to CH₃OH, Cu–Zn-based catalysts have been extensively studied [117,118,119,120,121]. Usually, commercial Cu–Zn catalysts are prepared by co-precipitation on Al₂O₃ support: such method implies several steps (precipitation, aging, filtration, drying, calcination, and reduction). Along with specific surface area and CuO and ZnO distribution, such steps have a strong influence on the activity of the final catalyst [122,123,124,125].

In particular, the distribution of CuO and ZnO has pivotal relevance during the aging, because the ion exchange between Cu²⁺ in malachite (Cu₂CO₃(OH)₂) and Zn²⁺ in hydrozincite (Zn₅(OH)₆(CO₃)₂) occurs, and the formation of zincian malachite ((Cu,Zn)₂CO₃(OH)₂) and/or aurichalcite ((Cu,Zn)₅(OH)₆(CO₃)₂) hydroxy carbonate precursors are extremely sensitive to pH, duration, temperature, and stirring [126,127,128].

To overcome such problem, very recently a new mechanical-force-driven, solid-state ion-exchange procedure has been refined [129]. It was found that the mutual substitution between Cu²⁺ and Zn²⁺ was improved with increasing mechanical ball-milling speed from 200 to 400 rpm, leading to significant enhancement of the catalytic activity. The XRD characterization shown in Figure 18 provided evidence of the presence of peaks related either to the CuO phase or the ZnO phase. Such peaks become broader and at the same time decrease in intensity upon the increase of the milling speed, indicating a decrease in crystallinity and crystal size. The absence of the peaks related to the presence of the Al₂O₃ phase was explained by the alumina change into an amorphous state after ball-milling [126,129].

The results of TEM measurements performed on the CZA400 catalyst reported in Figure 19 indicated a close cross-distribution of CuO and ZnO due to the occurrence of the solid-state ion exchange, in which ZnO NPs act as spacers, thereby avoiding the CuO agglomeration (as shown in Figure 19b).

Such alternation of CuO and ZnO NPs represents the result of the complete ion exchange between Cu²⁺ and Zn²⁺ in the precursors. For the CZA400 catalyst, impressive 59.5% CO₂ conversion and 73.4% CH₃OH selectivity were obtained at 240 °C and 4 MPa, resulting in a noticeable yield to CH₃OH of 43.7%. Therefore, this mechanical-force-driven, solid-state ion-exchange method proved to be a valuable alternative to synthesizing CZA catalysts with improved practical application potential for chemical hydrogenation of CO₂ to CH₃OH.

In another very recent study, Au clusters were directly deposited onto a functionalized support by newly developed, milling-mediated solid reduction procedure [130]. The proposed method involved the use of HAuCl₄ as Au precursor, which was pre-adsorbed on the Schiff base modified silica surface by electrostatic interaction in water. After grinding the dried carrier with solid NaBH₄, according to Scheme 10, the Au precursor can be reduced in situ by atomic H coming from the NaBH₄ reducing agent in the solid state. Depending on the gold loading, highly dispersed clusters and isolated Au atoms can be obtained by solid grinding.

Following such activation, it was found that the Au species were dispersed as 1.0 nm nanoclusters (Figure 20a), as determined by the aberration-corrected, HAADF-STEM (high-angle annular dark-field scanning transmission electron microscopy) analysis. At the same time, a lot of isolated Au atoms were detected, even if as a minor fraction of the total metal content. It was proposed that such Au subnanoclusters were effectively stabilized by the Schiff base groups present on the silica surface by providing strong electronic interaction. Differently, a variety of Au shapes and sizes were observed on the support (Figure 20b) when an Au-containing solution was reduced by NaBH₄. The obtained statistical particle size distributions were in agreement with the L₃-edge extended X-ray absorption fine structure (EXAFS, Figure 20c), revealing an increase in intensity of the peak related to the first shell, as well as in the Au coordination number when the conventional wet method was employed. In addition, a decrease in the white line intensity compared with the Au foil in the X-ray absorption near-edge structure (XANES) spectra obtained for the Au NPs (Figure 20d) is an indication that the functionalization of the silica surface with nitrogen allowed one to optimize the Au d-electron density [131].

These features put in evidence that the opportunity to use the solvent-free solid reduction synthesis allows the prevention of Au growth; indeed, crystal nucleation was hampered, giving rise to either subnanoclusters or single-atoms, depending on the Au content. The catalysts were tested in the hydrogenation of CO₂ to formate, and a marked dependence on the Au size, due to the different electronic structure of the metal species, was observed. The authors claimed that the milling-mediated solid reduction procedure could be considered as a general methodology to achieve basic knowledge ion the catalysis by Au, providing new insights for supported catalyst design for reactions involving the CO₂ transformation.

5. Final Remarks

The main purpose of this review was to point out the possibilities offered by microwave, ultrasound, and mechanochemical protocols in materials engineering at a nanoscale level for catalytic applications.

Summarising, concerning the synthesis of CO₂ reduction catalysts, assistance of microwave, ultrasound, and mechanochemistry allows one to obtain highly dispersed active sites and controlled particle dimension, and in most cases reinforces the active site/support interaction, giving rise to synergistic effects that help CO₂ reduction and improve catalyst stability. The procedures reported are usually simple, and require relatively short reaction times (minutes) without implying many steps. Moreover, stronger and/or more abundant basic sites occurring at the support surface favour CO₂ adsorption, a key step for its further reduction. Complex systems (hierarchical structures, composites, bi-functional catalysts, etc.) may be obtained by relatively simple methods.

Finally, a further crucial aspect concerns the effect of microwave irradiation during reaction, whilst carbon deposition usually leads to catalyst deactivation, under microwave the same carbon deposits may act as hot spots per se favouring the reaction.