1.1 Trends and Setting the Scene

There are two well-established electrocatalytic technologies: i) to make H₂ from water electrolysis, a century-old technology but with a fast increase of interest also industrially in relation to the H₂ economy⁸ and ii) electrocatalytic organic synthesis⁹ for speciality chemicals. The latter are used in typically small-scale processes. At the same time, a few industrial examples exist for commodity chemicals, such in the adiponitrile process¹⁰ or the dimethoxylation of 4-tert-butyltoluene to produce a protected benzaldehyde, a step in the lysmeral synthesis¹¹ used as a fragrance. A pillar for electrocatalysis is the chlor-alkaline industry, which is the industrial technology to produce Cl₂ and NaOH. Even with the progress in this area, from mercury to diaphragm electrolytic cells and finally to ion exchange membrane (IEM) electrolytic cells, there are still several challenges¹². However, research effort is limited due to the scarce industrial interest in developing further this established technology. There are, however, elements of novelty to mention in this area, such as the development of novel single-atom catalysts for electrosynthesis of chlorine from seawater-like solution¹³, the coproduction of H₂ in chlor-alkali process¹⁴ or the use of chlor-alkali technology to power environmental electrochemical treatment technologies¹⁵.

Electrocatalysis is also used in organic pollutant degradation¹⁶. However, besides the scientific interest, the industrial applications are limited to niche uses because the technology is not typically cost-competitive to alternatives. It may be used as an anodic reaction alternative to oxygen evolution reaction (OER) to improve kinetics and reduce overpotential in water electrolysis¹⁷ or other reactions like CO₂ reduction (CO₂RR)¹⁸. However, the use of more valuable anodic oxidations, as described later, is preferable. Thus, there are no relevant industrial examples of using organic pollutant degradation as anodic oxidation in electrocatalytic processes, notwithstanding the many research studies.

The water electrolysis area, although still presenting active research on the development of improved electrocatalysts, is mainly focused on improving the technological aspects to decrease the cost because the H₂ produced in this way is still uncompetitive. There are plans for large industrial production under policy pressure. Still, without subsidies and/or external drivers for deploying this technology on a large scale, it may be questioned whether water electrolysis will become a pillar of future sustainable energy and chemical production¹⁹. There are, however, contrasting opinions on this indication.

One of the critical points regards the storage/transport of hydrogen to overcome temporal and geographical mismatches between locations for the use of green H₂ and for its low-cost production from RES. Using green H₂ to produce e-fuels is an option to overcome this gap²⁰, but there are strong concerns about the costs and the effective cost advantage (for society). Rather than passing through the intermediate production of H₂ by electrolysis, (photo)electrocatalytic (PEC) technologies to produce e-fuels (in one step) offer a clue to overcome the limits of e-fuels technologies²¹. We may call solar fuels those synthesised in this way to differentiate from those passing through intermediate H₂ formation^{21, 22}, e. g. e-fuels. Solar fuel processing incorporates a step of renewable energy capture to avoid dependence on the grid and become independent of external energy supplies, an important step for a resilient development that minimises market dependence and thus enables the concept of net-zero communities for a circular economy²³. Often, these technologies are called artificial leaves or trees because they mimic nature's capability to use sunlight, carbon dioxide, and water to produce chemicals²⁴.

It is not relevant here to discuss further pros/cons. Still, there is often a consensus that the large-scale deployment of H₂ production by electrolysis will not depend on the possible future improvements in the electrocatalysts. Even if minimisation in the use of critical raw materials (CRM) and the design of the electrode to improve mass and charge homogeneous distribution, minimise H₂ bubble contact, etc., will improve performances, these aspects will have a minor impact on industrial deployment of H₂ electrolysis technology, in its different technological declinations (alkaline and proton-exchange membrane cells, high-temperature solid-oxide electrolysers, etc.).

Many improvements are instead still possible in electrocatalytic organic synthesis to improve selectivity, activity, and stability. However, the (often) small-volume applications, the short-term time to market, and the high added value of the products often hinder dedicated research.

There are instead emerging directions in electrocatalysis, which combine a large potential for application and a relevant need to improve electrocatalysts/electrodes in terms of Faradaic efficiency, selectivity, stability and productivity, e. g. current density^{2a, 3c, 5a}. These areas are related to the electrocatalytic i) CO₂RR²⁵, ii) N₂ fixation²⁶ and iii) electrocatalytic conversion of biomass-derived chemicals to produce large-volume chemicals²⁷. We should also add the electrocatalytic production of H₂O₂²⁸, a valuable clean oxidant in industrial-relevant processes (propylene oxide, caprolactam), besides relevant uses as bleaching agents (particularly, but not only pulp and paper industry), water treatment, Na-perborate and -percarbonate production, etc. However, the industrial potential of the three above areas is higher.

As reported by Liu et al²⁹ in their roadmap of electrocatalysts for green catalytic processes, the research on electrocatalysis (in terms of the number of papers up to March 2020) is largely dominated by water electrolysis and related aspects (OER and oxygen reduction – ORR: hydrogen evolution – HER). CO₂RR and N₂ reduction (NRR) account for about 4 % and 1 %, respectively, with respect to the sum of papers on water electrolysis and related aspects. This roadmap does not consider the electrocatalytic production of biobased chemicals and H₂O₂.

The aim here is not to overview the recent progress in these areas, being already several additional reviews of the few cited above. As an example, limiting to CO₂RR, about 500 reviews (in English using the keywords “CO₂ and electro” in SciFinder) have been published in the last decade. The number of papers is impressively higher. The question is whether there is a correspondence between this massive research effort and technological improvements towards application.

For example, analysing selected recent reviews on the rational design of CO₂RR electrocatalysts,³⁰ it may be concluded that despite the scientific advances, a generalised approach, e. g. not valid for the specific set of results, is missing³¹. There is a lack of effective progress towards industrialisation, i. e. overcoming current limits for reliable electrode sizes (above at least 10 cm²) and effective practical operative conditions. There is a gap between scientific advances and progress in the industrial application of electrocatalytic technologies, which is not only related to the usual effort necessary in scaling chemical processes.

In terms of practical challenges and limitations associated with making large-volume chemicals from CO₂ through electrocatalytic processes, there are other aspects to consider besides the scientific and technological challenges, including scalability, mentioned above. The first is the availability and cost of RES to perform the process on a large scale. It is necessary to avoid or minimize the use of external RES, which are critical elements in costs and dependence on externalities. It is necessary to develop approaches which directly integrate the use of solar energy, as the PEC approaches discussed in the next section. A main issue discussed later in more detail is that this implies the change to a distributed model of production, which is, thus, a radical change with respect to the current modalities of chemical production. Avoiding the cost of separate units for capturing and purifying CO₂ is another main issue. The current approaches in this direction are unsatisfactory in terms of cost and energy use. Fully novel approaches are needed. Downstream operations are often not considered, but the key issue is applicability.

A general question is that the fixed and operative costs related to electrocatalytic technology are still too high. While the focus of research is often on performance, cost-effectiveness (considering the whole technology, including ancillary elements) has to be the goal. The design of both electrocatalysts/electrodes and electrocatalytic cells has to be revised to minimize these costs. In addition, operative conditions have often been unsuitable in terms of exploitability. While improving the electrocatalysts/electrodes is a worthy exercise as well as a better mechanistic understanding, the critical issue in accelerating the applicability of electrocatalytic technologies is related to the engineering of the systems. However, the largest majority of the publications are on the mechanistic aspects and design of the electrocatalysts, with limited studies and a lack of effective new ideas on system engineering. In other words, the requirements for introducing electrocatalytic technologies on a large scale are the opposite.

Many will disagree with these statements and claim a higher perspective for electrocatalytic applications. There are examples, although still limited, of pilot units and efforts in scaling up electrocatalytic technologies.³² In addition, some research projects aim to construct and test electrocatalytic bench/pilot scale units. For example, the EU projects i) RECODE (grant 768583) for CO₂ to formic acid as a part of recycling carbon dioxide in the cement industry, ii) OCEAN (grant 767798) also for CO₂ to formic acid but as a part of a value chain to produce C2 monomers³³, iii) PERFORM (grant 820723) for a platform pilot unit for biobased chemicalsì electrochemical production, with two showcases of maleic acid from furfural paired with the production of valeric acid from levulinic acid, and adipic acid production from glucose) and iv) CELBICON (grant 679050), turning CO₂ into carbon monoxide and hydrogen. Avantium is a Netherlands company actively investing in these directions.

From an industrial perspective, the results are still far from the conditions for exploitability, or they are insufficiently reliable to invest in further development. There are a few start-up companies in the area (for example, i) CERT Systems to convert electrocatalytically CO₂ into carbon-based fuels and chemicals, such as ethanol and ethylene, ii) Jupiter Ionics for an electrochemical process to produce green ammonia, iii) CERT for CO₂ electrolysis to ethylene, iv) Oxylys Energy to convert CO₂ into green e-methanol, v) NitroVolt for electrochemical Li-mediated ammonia synthesis, vi) HPNow for the electrochemical production of hydrogen peroxide. However, the effective exploitability of these technologies, based on the claims, is not sufficiently proven. Major companies are instead reluctant to invest in novel electrocatalytic processes, which are still considered a long-term possibility (2050 or beyond).

The introduction of radically new technologies always requires significant time to market. In addition, there are other external constraints, such as the availability of cost of renewable electrical energy, which determine exploitability. However, the time to market is potentially shorter in the case of electrocatalysis because the implementation does not depend on scaling by size as in traditional chemical processes but mainly on numbers. Furthermore, some novel technologies, such as 3D printing, may accelerate the scalability of electrodes and reactors³⁴. Nevertheless, the transition is slower than usual, even when compared to the deep transformation that occurred about 70 years ago in chemical production³⁵ due to the change in feedstocks.

The reflection, which is the aim of this personal account, is whether the current approach in electrocatalysis is appropriate to make a fast transition or instead it is necessary to revise it. In this respect, the first point to clarify is the concept of electrocatalysis vs electrochemistry³⁶.

1.2 Electrocatalysis vs Electrochemistry

These two terms (electrocatalysis and electrochemistry) are often used as synonymous and considered equivalent, but they underpin different physicochemical phenomena and mechanisms. Rarely is there sufficient proof and evidence to discriminate among them, as shown for water photoelectrolysis.³⁷

Electrochemistry refers to the redox process occurring at the surface of the electrode upon application of a potential. In organic electrosynthesis, the processes are also often mediated by redox transfer agents present in solution^{9e, 38}. The electron transfer occurs at the interface, typically through an inner- or outer-sphere mechanism as described, for example, by Marcus theory³⁹, still remaining at the core of the description of the electrochemical mechanisms⁴⁰. Electron transfer occurs at the intersection of the two potential energy surfaces. It is thus determined from the reorganisation energy, which includes the reorganisation of the solvent (outer sphere) and ligands or strongly coordinated molecules (inner sphere).

The theory was developed for single electron transfer, but two relevant cases requiring adaptation of the theory are multiple electron transfer processes and proton-coupled electron transfer⁴¹. In these electrochemical approaches, the electrode surface is not explicitly involved in the mechanism. However, for simultaneous multiple electron transfer, the activation energy increases significantly. Thus, it may be more convenient energetically to form intermediate species for a cascade sequence of single electron transfer reactions. These intermediates may chemisorb or be stabilised at the catalyst surface. For the classical reaction of hydrogen evolution reaction (HER), the first electron transfer occurs on a proton, forming a chemisorbed hydrogen species (H_ads) that, through a concerted proton-electron transfer or alternatively reacting with a second H_ads species, generates H₂ (Volmer, Heyrovsky, and Tafel steps). Thus, some aspects of catalysis are included in electrochemical approaches and modelling. Still, there is no effective description of many catalytic aspects, particularly nanostructure, and its implications in terms of reaction mechanism.

In electrocatalysis, instead, a surface catalytic mechanism is used to model and understand the process, including the so-called Sabatier principle and its variations, e. g. that a volcano plot exists in correlating reaction rate and an activity descriptor⁴². This approach has been extensively used in electrocatalysis to identify optimal electrocatalysts following the pioneering studies by Nørskov and coworkers⁴³. Although widely used by many researchers, there are limitations in the approach, particularly for relevant multielectron electrocatalytic reactions⁴⁴. More relevant is that the computational approach does not effectively differentiate between catalysis and electrocatalysis. More aspects will be discussed later.

There is thus a gap on both sides in developing a unified electrochemical and electrocatalysis theory that is able to include both words and can be effectively used to design novel electrocatalysts for the emerging field mentioned above⁴⁵. Many scientists probably do not agree with this statement. Still, the absence of proper recognition of this critical gap is one of the crucial factors slowing the progress in the field. There is also a lack of proper distinction between catalysis and electrocatalysis, which should be approached differently, including from a theoretical perspective.

While the concepts discussed are heterodox and far from having a consensus, it is surprising that there is an absence of discussion on them as well as regarding the limits in current electrochemistry vs electrocatalysis and electrocatalysis vs catalysis approaches. Thus, this personal account aims to stimulate a discussion and reflection that is considered crucial to accelerate the deployment of electrocatalytic technologies for energy and chemical production.^{3c, 5a, 21c}

2.1 From Electro to Photoelectro-Catalysis

We are in a deep transition⁴⁶, e. g., one that radically changes the energy and chemical production scenario towards a sustainable, circular and resilient future⁴⁷. There are discording ideas on whether and when this may occur, but overcoming current centralised production models for a decentralised⁴⁸ production is one of the disruptive directions. Decentralised production offers many advantages in terms of minimised impact, use of local resources, integration within communities, faster time to market, greater flexibility, avoided or strongly reduced dependence on externalities, etc. Realising net-zero communities requires the development of such a type of decentralised production.

Integration of the direct capability of harnessing solar light is necessary, e. g., to realise photoelectrocatalytic (PEC) devices. Figure 1 illustrates schematically the type of PEC devices where the photoactive unit is integrated into the anodic part (PECa) or is present externally as a photovoltaic unit (PV/EC), eventually integrated into the cell. Detailed comments on the differences between these (and other) typologies of PEC cells were reported elsewhere^21b. In brief, in PECa design, the photoactive functionalities are integrated into one of the electrodes, typically the anode, as in PECa design. There is the possibility of having both anode and cathode photoactive, but there are many difficulties to overcome in a practical realization, making this approach ineffective. The design scheme of the PECa compact cell differs from the first because the cathode and anode are directly located at the two sides of a membrane. This design allows for the reduction of transport limitations, which greatly determine the performance. In addition, as outlined in Figure 2, it is possible to eliminate the electrolyte and realize zero-gap cells using gas-diffusion electrodes. More details elsewhere.⁴⁹ Instead, the PV/EC configuration is based on a separate PV cell that drives the electrocatalytic (EC) unit.

However, compared to electrocatalytic devices, where the current density is dictated by the cell design and operations, in a PEC system, the current density is associated with that provided by the photoactive element (an external photovoltaic unit, or a photo-cathode or -anode), the resistance of the cells, and coupling between the photo and electro components. While typical current densities of industrial interest in electrocatalytic processes are above 500 mA/cm², they are over one order of magnitude lower in PEC devices. Furthermore, while electrocatalytic devices operate continuously (24 h), even if the carbon footprint of the available electrical energy may negatively impact the effective greenhouse (GHG) emissions, PEC devices need to operate when sunlight is available. This would require rethinking design in terms of low manufacturing cost for discontinuous operations rather than in terms of conventional aspects such as solar-to-chemical conversion efficiency^21b.

In terms of how photoelectrocatalytic approaches differ in effectiveness compared to traditional electrochemical methods, it may be indicated that essentially, in all the three schemes of PEC devices presented in Figure 1, e. g. including when the photoactive elements are integrated into the electrodes (PECa design), the mechanism of the generation of a photocurrent which drives the electrocatalytic processes. Thus, essentially, the effectiveness and mechanism are the same for EC and PEC approaches. Still, the latter has limitations in terms of potential and (internal) current densities, which are associated with the characteristics of the photoactive elements. In the EC approach, they can be externally modulated to the optimal value, while this is more difficult to achieve in PEC design. Thus, in contrast to EC devices, where current densities can be up to 1 A/cm², PEC devices typically have up to two orders of magnitude lower current densities.^21b

A distributed production also requires minimising downstream operations (conversion and separation) as much as possible, making them eventually compatible with the PEC production unit (as productivity, pressure, etc.). These aspects significantly influence the choice and design criteria even during the initial evaluation. It is thus surprising that there is scarce attention given to these aspects in the literature. Similarly, PEC devices are typically studied as independent elements from their integration in the value chain. They could also be used to design novel approaches, as outlined in the following sub-section.

2.2 From Air to Chemicals

The PEC devices presented in Figure 1 are also called artificial leaf or artificial photosynthesis devices^{21b, 24b-24f} because, as in natural photosynthesis, they can use CO₂, water and solar light to make chemicals. Often, the same definition is used for PEC-type devices for water splitting. However, except in some cyanobacteria, H₂ production is not a target product for natural photosynthesis. Note also that artificial photosynthesis is different from natural one in several important aspects: i) the solar-to-chemical efficiency is significantly higher, already over 10 % or above, ^{24d, 50} while it is typically below 1 % for natural photosynthesis, ii) the target products (formic acid, methanol, C2 chemicals, etc.) are not the products obtained in natural photosynthesis, and often they can be toxic for natural systems, iii) the photosynthesis machinery is less complex and sophisticated (for example, self-regulation systems are not present), but more robust and easier to realise. Thus, mimicking natural processes is a commonly used term, but effectively, the artificial leaf has to be conceptually different. Differently, in artificial photosynthesis, mimicking natural processes is an effective goal, but these systems have significantly lower efficiencies than the PEC devices. Also, photosynthesis particulate-type devices have significantly lower solar-to-chemical efficiencies in CO₂ conversion.^{21b, 51}

An important difference between natural and artificial leaves is that the former are able to capture CO₂ and water from the air directly. Thus, realising an artificial tree, e. g., the further extension of the concept of an artificial leaf, requires incorporating the possibility of capturing CO₂ and water from the air, and eventually also N₂, and using them in an integrated PEC device. A further extension regards the possibility of making not only simple chemicals but also the complex molecules we need, such as carbohydrates, lipids, proteins, etc. Enzymes are capable of such syntheses, but the slow step is to convert CO₂ by the photosynthesis machinery to the C-substrate used further in the enzymatic cycles. By coupling a PEC device to convert CO₂ to such a type of C-substrates (acetate, in particular) to be fed to enzymes, it is possible to intensify the overall rate by over three orders of magnitude. Synthetic foods produced in this way would thus require much less resources and land compared to agriculture, drastically reducing the environmental impact at the same impact. This is thus a potentially disruptive technology for a sustainable future.

Centi and Perathoner⁵² discussed this possibility in a perspective paper. The concept they presented is graphically shown in Figure 2. It represents a vision for the future, while the paper discusses the current status of the research on the specific elements of this technology and the gaps to overcome. There is progress in the specific sub-topics, but the attempt to integrate them into a single unit has not yet been investigated.

This device combines i) the capabilities of functionalised membranes (or equivalent systems) in capturing and concentrating H₂O, CO₂ and eventually N₂ to the surface of an electrocatalyst while preventing contact with molecules that may inhibit the activity such as O₂; ii) the possibility of a PEC device to use sunlight to convert CO₂ on the cathodic side (to acetate preferably), while fixing N₂ to nitrate on the anodic side; the cathodic sides is a feed for microbial further conversion, while the anodic side can be used as a fertiliser; iii) the abilities of microbial processes in synthesising complex molecules such as carbohydrates or proteins, eventually assisted by photo/electro components to bypass limitations by using co-enzymes (NADH and ATP).

While the cited paper⁵² reports the status of research on the single components of the hybrid artificial-tree-type integrated device presented in Figure 2, specific examples or case studies on the integrated device are still not present in the literature. Thus, it is necessary to foster research in this direction.

2.3 Exploit Both Anodic and Cathodic Reactions

(Photo)electrocatalytic devices have separate zones for oxidation and reduction reactions. This fact offers creative possibilities for process intensification and the design of innovative solutions⁵³. Instead, literature studies typically focus on a single electrode with minor attention on the whole device enough and exploiting both cell sides.^{3a, 54} The research focus of electrocatalysis for the target reactions is on the cathodic development, with OER being the typical anodic reaction used (with conventional catalysts, although often it is the limiting process).⁵⁴ OER is a 4e⁻ redox process with a slugging kinetic⁵⁵ that requires a relatively high overpotential, e. g. lower energy efficiency. It was estimated that OER may account for up to 90 % of the total electricity input in CO₂ electrolysers.⁵⁶ There is increasing attention, but not systematic, to using different anodic reactions than OER to improve the kinetics and efficiency. At the same time, they could produce an added value compared to O₂ production.

A general remark, however, is that the anodic reaction, from an industrial perspective, should be carefully chosen to combine with the value chain of the cathodic reaction in terms of market, volumes of production, field of interest, etc. A typical example is the use of wastewater purification as the anodic reaction,⁵⁷ which typically does not result in practical applications (as coupling to CO₂RR, for example), while it is one of the advanced oxidation processes (AOP) in use as such (e. g. without valorisation of the cathodic reaction). Electrocatalysts typically used are boron-doped diamond (BDD), PbO₂, SnO₂ and Ti-based anode (e. g., Ti₄O₇, blue titanium oxide).

However, possible useful integration can be identified, even if not still studied. In biogas production, the CO₂ produced (typical biogas composition is two-thirds CH₄ and one-third CO₂) could be photoelectrocatalytically converted to methane at the cathode (thus lowering carbon footprint and increasing productivity). Instead, the wastewater produced in the digestor (where the biogas is produced) could be fed to the anodic part of the electrocatalytic cell to reduce the chemicals that inhibit the wastewater treatment by active sludges. The added value derives from the improved wastewater treatment operations.

The OER alternatives proposed to reduce costs and improve sustainability range from raw chemical generation (e. g. H₂O₂) to waste oxidation and molecule upgrade.⁵⁸ Apart from hydrogen peroxide, which could be preferably produced at the cathode by oxygen reduction (ORR) than at the anode by water oxidation (WOR)^{28, 59} and requires high overpotentials, the other anodic reactions proposed in the literature (apart from waste total oxidation) include unsafe (even if established) inorganic reactions (such as NaCl to Cl₂), or organic chemicals conversion which have a difficult coupling (as value chain) to CO₂RR cathodic reaction (for example ethanol to acetate, glycerol or 1,2-propandiol to lactic acid).⁵⁸ However, they can be used in some biorefineries. There is still a need to explore new possibilities, but having a precise vision of the industrial context in which the co-valorisation (cathode and anode) technology would be located.

However, an aspect often underestimated is that the reaction rates (and mass/electron balance) at anodic and cathodic sides should match when pairing two electrocatalytic reactions. An electrocatalytic cell is a closed circuit where both reactions are mutually influenced. While separate reactions are typically investigated even when co-valorisation of anodic/cathodic reactions is proposed, only results with the full cell, e. g. when both reactions are simultaneously investigated, provide the right results even in terms of optimal electrocatalysts. There is a lack of studies in this direction. When a photoactive element is present, e. g. PEC devices, a further critical aspect and variable is added. Experience has proven that the electrocatalyst performances (and selection) in a PEC cell are different from those in an electrocatalytic cell with an externally applied bias.⁶⁰

Typically, electrocatalytic reactions are studied at room temperature without analysing the effect of the reaction temperature. In addition, the same temperature is used at both the cathodic and anodic sides. However, it is technically feasible to have cells with a temperature difference up to 50–100 °C that allows a better match between the reactions at the cathode and anode sides. This possibility was explored in the frame of an EU project (TERRA, grant 677471) and offers new clues to design advanced electrocatalytic processes.

Pairing CO₂RR with the electro-oxidation of organic (biobased) chemicals (OOR) is an emerging direction. The oxidation of 5-hydroxymethylfurfural (HMF, a biobased platform molecule⁶¹) to 2,5-furan-dicarboxylic acid (FDCA, a monomer for alternative biobased polymers substituting terephthalic acid⁶²) is an example of coupling CO₂RR with OOR. The substitution of OER with this anodic oxidation decreases the overpotential, as shown in Figure 3, besides producing a valuable added-value chemical.

A techno-economic assessment (TEA) of various possibilities (about 300) for coupling CO₂RR with the selective electro-oxidation of organics (OOR) has been reported by Na et al⁶³. They developed an automatic framework approach for process simulations, although with limits in accounting for some process systems, such as solidification, acid treatments, and GDL-type electrolyser devices. The approach was used to estimate the levelized costs of chemicals, concluding that for coupling with CO₂RR, formic acid, n-propanol, acetaldehyde, allyl alcohol, glycolaldehyde, and ethylene glycol are preferable. Other chemicals such as FDCA, 2-furoic acid, ethyl acetate, lactic acid, formic acid, glycolic acid, and oxalic acid were also indicated as suitable. The analysis, besides some limitations in the TEA approach, does not account for the relevance of the value chain, both in terms of costs of raw materials (including the critical issue of renewable electrical energy cost) and the use of the products, including the difference in the market size. Nevertheless, it shows pairing CO₂RR with OOR is an important direction deserving more effort in finding optimised devices and reaction combinations and testing them in pilot plants.

While the industrial exploitability and integrability in biorefineries still need to be proven more convincingly⁶⁴, these indications show that it is possible to consider (in perspective) a distributed model of chemical production where (photo)electrocatalysis becomes the pillar for low-carbon chemical production beyond fossil fuels, i. e. an e-chemistry based on carbon circularity and biobased resources.^{3c, 4c, 5a, 5b} Improving the effort to exploit both anodic and cathodic reactions is a must to go in this direction.

A requirement, however, is the possibility of designing ad-doc electrocatalysts given a target reaction, as current mechanistic and theoretical approaches are still unable to provide a general toolbox of electrocatalysts suitable for novel OOR.

While, in principle, the same considerations valid for CO₂RR-ORR pairing could also be applied for NRR (N₂ fixation to ammonia),^{26, 65} the still insufficient performances have not yet stimulated the investigation of suitable pairing reaction. There is, however, an important difference to the previous cases. An attractive possibility is to pair NRR with N₂ oxidation to NOx (nitrate, in particular) (NOR) to develop in a single step the production of ammonium nitrate,⁵⁴ i. e. allow a distributed production of green fertilisers. The concept is shown in Figure 4.

This technology of direct electrocatalytic production of ammonium nitrate offers several potential advantages: i) it is well suited for distributed production (at farmer or small district level) using locally available renewable electricity sources (or can be coupled to a photoactive unit as presented in Figure 1), and ii) it produces an ammonium nitrate aqueous solution for direct local use as a fertiliser lowering the carbon footprint, avoiding long-distance transport of the fertiliser (and related externalities), decreasing cost (being avoided the energy-intensive step producing solid ammonium nitrate, with also associated safety issues). However, studies on NRR are still far from exploitability, and those on NOR are still very limited.⁶⁶ No results have yet been reported on an integrated device, as presented in Figure 3, notwithstanding the large applicative potential.

2.4 Tandem/Paired Electrocatalytic Reactions

Particularly in the field of biobased processes, there is a large potential for tandem/paired electrocatalytic reactions beyond the examples discussed above. Among the advantages are i) process intensification and develop low-carbon processes, ii) the use of renewable energy, and iii) in-situ generation of the redox reactants (hydrogen equivalents, active oxygen species), thus avoiding the production costs for reducing agents (typically H₂) or oxidants. Notwithstanding being known for years, there are still no commercial examples, being necessary to overcome many difficulties, from identifying suitable electrocatalysts combining high Faradaic selectivity at high current densities, high stability (fouling of the electrode or leaching are typical issues), pairing match the requirements of anodic and cathodic reactions (e. g., balance in terms of reaction rates, electron and H⁺/OH⁻ flow, etc.).

While research often focuses on a single electrode side, the cathodic one, exploiting both sides properly, is an unavoidable necessity.^{17b, 67} When the focus is on the cathodic part (in CO₂RR or HER, for example), OER is the typical anodic reaction. Still, it has slugging kinetics and requires relatively high overpotentials, being a 4e⁻ oxidation. Thus, alternative oxidation reactions have often been explored, including wastewater oxidation, which is mentioned in the introduction. There is recent interest in exploring also other energetically easy and kinetically fast reactions to speed up the process. For example, in the electrosynthesis of ammonia by nitrogen or nitrate reduction, hydrogen oxidation at the anode side has been proposed to enhance the performance.⁶⁸ While reducing the energy requirements of the cell (but not the overall energy valance, by considering the energy request to produce H₂), this is not the right approach to enhance the techno-economic viability and sustainability, which is the critical issue. Similarly, using nitrate as the feed for ammonia reduction rather than N₂,⁶⁹ improves the kinetics and performances but has a limited industrial relevance. Ammonia is mostly used to produce nitrate (as fertilizers). Thus, the interest is in converting N₂ to ammonia or, better yet, nitrate directly (as outlined in Figure 4) rather than in reducing nitrate to ammonia. Also, to eliminate nitrate from wastewater, there are better industrial alternatives. The message is thus to address relevant value chain and industrial cases rather than use academic shortcuts to prove higher performances.

Creating case examples for effective integration within a valuable value chain is important to push research and industrial interest in this area. One interesting example, which, however, is still not at the exploitability level for the many issues outlined above, is that at the core of the EU project, TERRA (cited before). Poly(ethylene 2,5-furanoate) (PEF) is a biobased polymer alternative to the fossil-based polyethene terephthalate.⁷⁰ PEF is produced from two monomers: FDCA and ethylene glycol. They can be produced in a paired electrocatalytic process at the anode and cathode, respectively, starting from sugars. Thus, a single electrocatalytic unit can generate simultaneously both the two monomers necessary for producing PEF, with a significant potential cost reduction in the process.

The electrocatalytic synthesis of adipic acid is another potentially large-scale example of paired electrosynthesis. It was at the core of the PERFORM EU project (cited before). Glucose is oxidised to glucaric acid at the anode, which is fed to the cathode for hydrodeoxygenation to adipic acid. There are some advances for the single reactions⁷¹, but the fully integrated process has not been demonstrated electrocatalytically, but only catalytically (Rennovia process⁷²).

Several examples of other possible electrochemical conversion of biobased molecules (polyols, carbohydrates, heterocyclic, and other chemicals) exist. Many were discussed by Lucas et al⁷³. They remarked, however, the gap between the lab-scale results and possible industrialisation. Several challenges were identified for industrialisation. The main issues indicated (among others) were i) the low faradaic efficiency and presence of competitive HER/OER reactions (with thus downstream costs for separation and loss of energy efficiency), ii) the low product concentrations and complex product streams, as well as the need to recycle the electrolyte, which determines costly downstream processing, iii) the insufficient current densities (thus high fixed costs, i. e. CAPEX), iv) electrode stability due to fouling, v) membrane crossover of chemical species. However, studies have been limited in these directions up to now, and also case examples to overcome them, at least in part, are scarce.

The cost of downstream processing is a key issue, and thus, a good case example to promote electrocatalytic technologies would be to identify possibilities to reduce these costs drastically. The EU project DECADE (grant 862030) addressed this question in two ways: i) by forming the same product at both the cathodic and anodic side and ii) by producing a stream which, with minimal downstream operations, can be directly used. The project concept is a PV/EC device, where at the cathodic side, the CO₂ is electrocatalytic reduced (on a gas diffusion cathodic electrode - GDE) to acetate, which is used to esterify the co-feed ethanol to ethyl acetate (EA). On the anodic side, the ethanol is instead oxidatively dimerised to EA. A stream of ethanol and EA with some ethyl formate (side product) is produced that can be directly used as a green solvent or gasoline fuel additive. The integrated photovoltaic module allows direct operations with solar energy. The device was developed up to the bench-scale level.

This is a scientific area with very limited available results. There are several studies on CO₂ electrocatalytic reduction to acetate and/or formate⁷⁴, but all these studies refer to the use of aqueous electrolytes. When ethanol is used as an electrolyte (plus some additives to increase conductivity), the electrocatalysts selective in aqueous electrolytes are completely inactive, even when some water is present together with ethanol. For the anodic part, there are few studies⁷⁵ of difficult reproducibility and/or applicability. There is, thus, novel electrocatalytic chemistry largely unexplored and, in general, still insufficient scientific bases to design tailored electrocatalysts for operations under less conventional conditions and electrolytes.

An analogous cell can be used to produce methyl formate (MF) from methanol by valorising the CO₂ emissions in methanol plants to a higher-added-value chemical. MF is commercially produced by methanol carbonylation, but a new green route could be the production of methanol from CO₂ and green H₂. Then, the electrocatalytic conversion of methanol to MF using the possibility to produce on both sides of the cell, by methanol oxidative dimerisation at the anode and by CO₂ reduction to formate, used to esterify methanol to MF.

There are other possible examples of the possibility of producing the same product at both anodic and cathodic sides. The electrocatalytic production of glycolic acid (GC) is an interesting example. GC is produced by oxalic acid reduction at the cathode, while at the anode, glycerol is electro-oxidised to glycolic acid (GC). Glycerol is a byproduct of biodiesel production, while oxalic acid could be produced by biowaste fermentation. GC is a monomer for green poly-glycolic acid production, a high-value polymer. This is another case example of the value chain and reuse of waste to realise carbon circularity through advanced electrocatalytic processes. However, the results of the literature on developing such a type of device and identifying the electrocatalysts are very limited or made under non-relevant reaction conditions.

These examples demonstrate how electrocatalysis can be a crucial technology for future chemical production and energy. Still, a creative effort is required to investigate electrocatalysts, electrodes, and cell design outside of the traditional approaches and conditions. It should be guided by a more in-depth and open-minded analysis of value chains where these technologies could bring innovation to current production technologies.

2.5 Electrocatalytic Anodic Selective Oxidation

Previous sections have already evidenced the need to broaden the studies on electrocatalytic anodic selective oxidation. This is an area much less investigated than cathodic reaction, except for total oxidation. However, raising this as a general methodology to produce valuable large-volume chemicals is a crucial element in the vision for future sustainable chemical production.

An example is phenol synthesis, an important chemical and monomer currently produced industrially through a multi-step, energy-intensive process with an overall yield of around 7 %. Direct electrocatalytic oxidation of benzene to phenol using the activated oxygen species generated on the surface of an anodic catalyst is an interesting option to explore, although with limited studies. Lee et al.⁷⁶ reported a selectivity up to ~95 % (with a current efficiency of 76 %) in the direct hydroxylation of benzene to phenol over a VO_x anode at 50 °C. With alternating current and using a V_xO_y-Sn_0.9In_0.1P₂O₇ electrode, they also show the possibility of producing phenol both at the anodic and cathodic sides⁷⁷. The alternative approach is to generate insitu H₂O₂ and use it for the hydroxylation of benzene. For example, over nickel single-atom catalyst (on defective carbon nanofibres) acting as bifunctional electrocatalysts, able to make the two-electron ORR to H₂O₂ and H₂O₂-assisted benzene oxidation to phenol.⁷⁸ Productivity is generally low, but these approaches seem preferable over the alternative of direct photocatalytic oxidation of benzene to phenol,⁷⁹ which shares, however, several mechanistic analogies.

Direct epoxidation of olefins is another challenging and potentially large-volume reaction.⁸⁰ Recently, it was shown that ethylene and propylene can be epoxidised electrochemically to the corresponding epoxides (ethylene and propylene oxide, respectively) with selectivities around 70 % at industrially-relevant current densities.⁸¹ However, chlorine is used to mediate the process, with related problems of safety corrosion, etc. At the anode, the olefin and chlorine are converted to chlorohydrin [RCH(OH)CH₂Cl, where R is H for ethylene or CH₃ for the propylene] and HCl, while hydroxide and H₂ are produced at the cathode. Downstream mixing produces the epoxide, Cl⁻ and water. When integrated with a CO₂-to-ethylene step, an integrated CO₂-to–ethylene oxide process is possible.

Another example is the electrocatalytic production of ethylene and acetate from CO₂ and their (electro)catalytic coupling to produce vinyl acetate. These examples, together with the above case of phenol production (benzene could also be produced from CO₂), demonstrate the potential feasibility of producing large-volume monomers from CO₂, offering a full novel possibility for carbon circularity and reuse of CO₂. Electro-carbonylation of olefins and diolefins (ethylene and butadiene⁸²) is another route for making CO₂-based monomers and chemicals. Butadiene, and not only olefins, could also be produced electrocatalytically from CO₂.⁸³

The electrocatalytic oxidation (ECO) of organic compounds can proceed through either direct oxidation at the electrode surface or (ii) indirectly, using redox mediators, as discussed later. During electrolysis in an acidic or alkaline solution, H₂O (or OH^–) electrode transfer at the anode generates adsorbed hydroxyl radicals, which may either react with an organic molecule or oxidise a transition metal ion of the electrocatalyst, subsequently reacting with the organic molecule. While generally, these mechanisms lead to full oxidation of the organic molecule, the challenge is to control the process to realise selective oxidation. Few studies have explored ECO possibilities⁸⁴. Data for the predictive identification of suitable electrocatalysts are largely missing. Similarly, most of the studies on ECO are limited to small molecules, while those on more relevant chemicals are scarce.

Lignin and its derivative conversion by ECO is an interesting opportunity.⁵³ Lignin can be electrocatalytically depolymerised to oligomers.⁸⁵ These (and phenol derivatives) can also be electrocatalytically further upgraded to added-value chemicals. ^{64d, 86} Catechol can be selectively converted to 1,2-benzoquinone, resorcinol to 2,4-benzoquinone, p-cresol to 4-hydroxybenzyl alcohol and p-hydroxyl benzoic acid, phenol to hydroquinone and p-benzoquinone, as examples.⁵³ PbO₂ and Ni-based anodes are the most used, but deactivation is a common problem observed in many of these studies. There is a facile generation of radical species at the electrode surface, which starts the formation of polymeric species, such as humins, occurs, that deactivate the electrocatalyst by fouling.

An industrial-relevant example developed by BASF is the anodic conversion of p-methoxytoluene into p-anisaldehyde (Figure 5).⁸⁷ Toluene derivatives undergo anodic acetalisation in methanol to form dimethylacetal intermediates, which can be hydrolysed to the corresponding aldehyde. This electrochemistry can be adapted to a variety of other syntheses of valuable chemicals. p-Anisaldehyde is also produced in paired electrolysis, where the cathodic hydrogenation of dimethyl phthalate is associated with the anodic methoxylation of the p-toluene⁸⁸. BASF produces about 4000 t/y using this method, and other companies also use the method to convert other substituted toluene chemicals.^11a

The ECO of glucose and furanic compounds is another interesting area, as mentioned above. Various anodes have been explored, including, for example, Ni₂P/Ni/NF,⁸⁹ Ni₃S₂/NF,⁹⁰ and Ni₂P nanoparticles on nickel foam.⁹¹

Another relevant area is the Kolbe reaction, where one-electron oxidation at the anode results in the decarboxylation of carboxylates to give the corresponding alkyl radical, which may dimerise (due to the high concentration of radicals at the electrode surface) to produce R–R homo-dimer. On porous carbon graphite electrodes, two-electron oxidation is possible, resulting in the oxidation of the alkyl radical to the corresponding carbocation, R⁺ (Hofer–Moest reaction).⁹² When this reaction is combined with a cathodic production of carboxylates (from CO₂),⁹³ a very interesting synthetic chemistry is possible. Intramolecular biradical recombination of dicarboxylic acids to unsaturated compounds is alternatively a valuable option to produce alkenes, alkynes or cycloalkanes.⁹⁴ As several carboxylic acids are derived from biomass, the above electrocatalytic chemistry offers many valuable options for a biorefinery. The process can also be scaled up to continuous flow conditions, for example, to convert valeric acid to n-octane.⁹⁵

There is thus a variety of valuable electrocatalytic anodic reactions to convert platform biobased chemicals. Still, systematic studies are lacking, both in terms of identifying electrocatalyst characteristics (design) for the reactions of concern and clear identification of needs from an industrial perspective.

2.6 Electrocatalytic Mediated Synthesis

Electrocatalytic-mediated synthesis^{9a, 11, 96} (indirect electrolysis) is an approach well-used in organic electrosynthesis but less so in other electrocatalytic applications. The reason is that processing cost, including for separation when complex mixtures are present and the redox mediator should be recirculated, is a less critical aspect compared to the synthesis of commodities and large-volume chemicals. Nevertheless, it represents an opportunity that should be considered and evaluated.

In indirect oxidation, a redox mediator acts as the transfer agent to reduce the substrate (Figure 6), which may otherwise be more difficult to be converted directly. Figure 6 also shows some of the typical redox mediators used for anodic oxidation. Transition metal complexes can be used for cathodic reactions.^96c Among the advantages of indirect conversion are the elimination of the kinetic inhibition of the heterogeneous electron transfer between electrode and substrate, e. g. the overpotentials can be reduced, ii) electron transfer mediators can exhibit higher or totally different selectivity, iii) deactivation of the electrode can be reduced or eliminated, due to lower potentials, iv) some side reactions occurring at high potentials are avoided. On the other side, additional costs, particularly in downstream separation and redox mediator recovery, may be crucial for large-volume productions, different from organic syntheses of high-added-value chemicals.