Fuel Cell Reactors for the Clean Cogeneration of Electrical Energy and Value-Added Chemicals

Abstract

Fuel cell reactors can be tailored to simultaneously cogenerate value-added chemicals and electrical energy while releasing negligible CO₂ emissions or other pollution; moreover, some of these reactors can even “breathe in” poisonous gas as feedstock. Such clean cogeneration favorably offsets the fast depletion of fossil fuel resources and eases growing environmental concerns. These unique reactors inherit advantages from fuel cells: a high energy conversion efficiency and high selectivity. Compared with similar energy conversion devices with sandwich structures, fuel cell reactors have successfully “hit three birds with one stone” by generating power, producing chemicals, and maintaining eco-friendliness. In this review, we provide a systematic summary on the state of the art regarding fuel cell reactors and key components, as well as the typical cogeneration reactions accomplished in these reactors. Most strategies fall short in reaching a win–win situation that meets production demand while concurrently addressing environmental issues. The use of fuel cells (FCs) as reactors to simultaneously produce value-added chemicals and electrical power without environmental pollution has emerged as a promising direction. The FC reactor has been well recognized due to its “one stone hitting three birds” merit, namely, efficient chemical production, electrical power generation, and environmental friendliness. Fuel cell reactors for cogeneration provide multidisciplinary perspectives on clean chemical production, effective energy utilization, and even pollutant treatment, with far-reaching implications for the wider scientific community and society. The scope of this review focuses on unique reactors that can convert low-value reactants and/or industrial wastes to value-added chemicals while simultaneously cogenerating electrical power in an environmentally friendly manner.

Graphical Abstract

A schematic diagram for the concept of fuel cell reactors for cogeneration of electrical energy and value-added chemicals

1 Introduction

Electricity and value-added chemicals are among the indispensable necessities of modern human society [1,2,3,4,5,6]. However, energy generation and chemical production have led to the fast depletion of fossil fuel resources [7,8,9] and related environmental issues [10, 11]. Therefore, the global search for a more efficient and environmentally friendly approach to clean power generation and the production of downstream value-added chemicals has drawn extensive attention in the past few decades because of increasing demand. To date, most strategies have fallen short in addressing these issues to reach a win–win solution that simultaneously meets production needs and environmental concerns [12,13,14,15,16,17,18,19].

Inspired by this, the ideal method should be tailored to simultaneously cogenerate value-added chemicals with high conversion efficiency and high-density power in the same reactor without any external electrical energy or emission of pollutants. Considering their combined function as both chemical reactors and electric generators, fuel cells, which emerged in 1838 [20, 21], have been adopted as a base to realize the concept of “cogeneration” [22,23,24]. In fuel cell reactors, value-added chemicals and electrical energy can be simultaneously harvested with negligible emissions of CO₂ or other pollution; furthermore, this kind of reactor can even “breathe in” poisonous gas as a feedstock. Therefore, these unique reactors can result in great contributions to a low-carbon economy [25, 26], which coincides with the concept of “carbon neutrality” [27,28,29]. Multinational groups of scientists have collaborated on clean energy that surpasses carbon neutrality by absorbing CO₂ from the atmosphere. These unique reactors would provide the potential to release less CO₂ and/or to capture CO₂, which could be one strategy to achieve carbon neutrality commitments.

As a kind of electrical power-generating device, fuel cells can be understood as electrochemical reactors that convert chemical energy from fuels and oxygen (or oxygen in the air) into electrical energy through electrochemical reactions with a maximum theoretical energy efficiency of 83%; this is in contrast to the only 58% of internal combustion engines due to the omission of the Carnot cycle in fuel cells [30, 31]. As chemical synthesizers, fuel cells can be classified as “electrosynthesis” devices [32,33,34], i.e., electrochemical cells for the production of chemical compounds. By controlling the cell potential, catalyst components, surface structure, and reagent concentration, electrosynthesis has considerable advantages over ordinary redox reactions, such as a regulatable reaction rate and selectivity, reduced waste heat, and higher selectivity and yield [35]. Electrosynthesis has been a zealously researched science topic [36,37,38] and is also economically beneficial in industrial applications, e.g., wastewater treatment [39, 40]. In fuel cell reactors, both oxidation reactions (at the anode) and reduction reactions (at the cathode) can be utilized, often invoking radical intermediates. The intermediates can be produced in initial reactions at the electrode surface, which then diffuse into the solution and participate in subsequent reactions [34, 35], thereby providing the possibility to broaden the range beyond only direct redox reactions at the electrode.

All these superior attributes of fuel cell reactors guarantee energy efficiency when used as electricity generation systems, and the flexible selection of oxidation and reduction reactions at the separated anode and cathode also makes it possible to produce a multitude of value-added chemicals. Furthermore, the use of membranes in a sandwich structure to separate the anodic and cathodic parts ensures the purity of the product [23, 41], as shown in Fig. 1.

Fig. 1

Schematic diagram showing the concept of fuel cell reactors that cogenerate electrical energy and value-added chemicals

2 Scope of the Review

This review paper starts with an overall view of all fuel cell reactors for cogeneration, covering the in-depth and distinct aspects of the reactors from their fundamentals to their evaluation criteria. Then, the main categories of fuel cell reactors, including polymer electrolyte membrane fuel cell (PEMFC) reactors and solid oxide fuel cell (SOFC) reactors, will be introduced. Furthermore, other electrochemical reactors based on batteries, which are not actual fuel cells but have borrowed the concept of cogeneration in their discharge processes, are also included in this review. Next, the structural design and electrode material selection of various fuel cell reactors will be comprehensively described and discussed in terms of their universality and individuality. Finally, state-of-the-art reactor technologies will be systematically summarized, and the review paper will end with our vision of future developmental perspectives in this field.

Certain aspects of this review have been covered in other reviews in the past decades: fuel cells for chemicals and energy cogeneration [42], solid electrolyte membrane reactors–current experience and future outlook [23, 24], methane conversion to C₂ hydrocarbons in solid electrolyte membrane reactors [43, 44], research progress in ethane dehydrogenation to cogenerate power and value-added chemicals in solid oxide fuel cells [45], fuel cells for carbon capture applications [46], alkaline anion exchange membrane fuel cells for cogeneration of electricity and valuable chemicals [22], progress in solid oxide fuel cells with nickel-based anodes operating on methane and related fuels [47], and insights into the interfacial effects in heterogeneous metal nanocatalysts toward selective hydrogenation [48]. However, the above review papers either focused on one unique type of fuel cell reactor/reaction with a limited literature summary or reactors without cogeneration, which coproduce value-added chemicals and electrical energy, sorted out separately. Therefore, it is imperative to publish a review paper covering all the types of fuel cell reactors that simultaneously cogenerate value-added chemicals and electrical energy in an environmentally friendly manner. A comprehensive summary combined with a thorough analysis of the latest research breakthroughs and advances, aiming for a solid introduction to the universality and individuality of these unique reactors, is urgently needed.

3 Fundamentals of Cogeneration

The “electrogenerative process”, i.e., “cogeneration”, was defined in 1963 by Langer and Landi [49] as an electrochemical reaction with negative free reaction energy $\left( {\Delta_{{\text{R}}} G^{\Theta } } \right)$ for the overall process that simultaneously produces desired chemicals and electrical energy.

Under the hypothesis that all the pure reactants are converted into pure products, with each species in its standard state, the change in free reaction energy $\left( {\Delta_{{\text{R}}} G^{\Theta } } \right)$ can be easily calculated according to the stoichiometric factor ν_i of each reactant i according to the following equation when the standard free energy of formation $\left( {\Delta_{{\text{F}}} G^{\Theta } } \right)$ [50, 51] of reactant i is available:

$$\Delta_{{\text{R}}} G^{\Theta } = { }\sum v_{i} { }\left( {\Delta_{{\text{F}}} G^{\Theta } } \right).$$

(1)

The standard equilibrium voltage U₀ (V) can also be obtained as follows:

$$\Delta_{{\text{R}}} G^{\Theta } = - nFU_{0} ,$$

(2)

where n is the number of electrons and F is the Faraday constant (96 485 C mol^–1).

According to the thermodynamic understanding of electrochemical cogeneration, reactions should be chosen according to the following criteria.

• 1. The overall reactions must follow ${\Delta_{{\text{R}}} G^{\Theta}}$ < 0, i.e., must be thermodynamically favorable.

• 2. The chemicals reacting at the anode must be oxidizable, e.g., must have olefinic double or triple bonds; accordingly, the chemicals at the cathode must be reducible, e.g., must have oxidizing groups or elements with high valence.

• 3. The reactants should be easily soluble in the electrolyte or easily adsorbable at the catalytic sites on the corresponding electrode for the electrode reaction to take place.

Several fuel cell reactors have been developed to shorten complicated chemical formation processes to single-step processes [52, 53]. Compared to conventional catalytic routes, those in fuel cell reactors possess the following advantages.

• 1. Because the reactants are fed separately into the system, i.e., the oxidation and reduction reactions are never competing at the same sites and the risk of explosion can be reduced.

• 2. The size of the reactors and the corrosion that occurs in them can be reduced; moreover, by taking advantage of the electrochemical reactions, the required operating temperature can also be much lower.

• 3. By altering either the electrode potential or external load, the type of product and the corresponding chemical selectivity in the process can be varied; in addition, the variety of particle sizes and the surface status of the catalysts in the electrodes play important roles in selectivity and efficiency.

• 4. Since the reactants are recyclable, their use is economical, thereby realizing fuel cell reactors that can meet all the requirements while providing an additional benefit in the form of power generation.

4 Evaluation Criteria of Cogeneration Systems

According to the concept of cogeneration, the two main functions of such systems are the harvesting of value-added chemicals and the production of electrical energy, so the corresponding evaluation standards for chemical synthesis reactors and energy-producing systems (fuel cells) [54, 55] can be borrowed, i.e., conversion and selectivity, as well as power density and energy density.

4.1 Conversion and Selectivity

In chemical reaction engineering, conversion and its related terms, i.e., yield and selectivity, are very important for effective holistic reactors. These terms are expressed as ratios of how much of a reactant has reacted (X—conversion, 0 < X < 1), how much of the desired product has been formed (Y—yield, 0 < Y < 1), and how much of the desired product has been formed in proportion to the total amount of reactant(s) consumed (S—selectivity) [56,57,58].

According to the chemical reaction taking place, the stoichiometric coefficients satisfy the following relation:

$$\mathop \sum \limits_{i = 1}^{n} v_{i} A_{i} = \mathop \sum \limits_{j = 1}^{m} u_{i} B_{i} ,$$

(3)

with the following assumptions.

This definition can also be applied for multiple parallel reactions, either per reaction or simply for the limiting reaction. A batch reaction assumes that all reactants are added at the beginning; a semibatch reaction assumes that some reactants are added at the beginning, while the rest are fed during the batch process; and a continuous reaction assumes that reactants are fed and products leave the reactor continuously in a steady-state process.

Conversion can be defined for both (semi)batch and continuous reactors as the overall conversion and instantaneous conversion. When fuel cells are used as reactors, the internal reactions are continuous [23, 43]. For continuous processes, both the overall and instantaneous conversions are the same. Furthermore, for multiple reactants, the conversion can be defined as overall or per reactant:

$$X_{i} = \frac{{n_{{i,{\text{ in}}}} - { }n_{{i,{\text{ out}}}} }}{{n_{{i,{\text{ in}}}} }} = 1 - \frac{{n_{{i,{\text{ out}}}} }}{{n_{{i,{\text{ in}}}} }}.$$

(4)

In general, the yield is defined as the amount of product(s) produced per amount of product(s) that could be produced:

$$Y_{{\text{p}}} = \frac{{n_{{{\text{p}},{\text{ out}}}} - { }n_{{{\text{p}},{\text{ in}}}} }}{{n_{{{\text{k}},{\text{ in}}}} }} \cdot \left| {\frac{{u_{{\text{k}}} }}{{v_{{\text{p}}} }}} \right| .$$

(5)

The overall selectivity is defined as the amount of the reactant required to form a product per total amount of that reactant consumed:

$$S_{{\text{p}}} = \frac{{n_{{{\text{p}},{\text{ out}}}} - { }n_{{{\text{p}},{\text{ in}}}} }}{{n_{{{\text{k}},{\text{ in}}}} - { }n_{{{\text{k}},{\text{ out}}}} }} \cdot \left| {\frac{{u_{{\text{k}}} }}{{v_{{\text{p}}} }}} \right|.$$

(6)

For continuous reactors, the three concepts can be combined:

$$Y_{{\text{p}}} = X_{i} \cdot S_{{\text{p}}} .$$

(7)

For example, according to the reactions in Sect. 6.2.2.3, ethylene can be produced from ethane in a fuel cell reactor [59, 60] with the simultaneous cogeneration of electrical energy; the conversion, yield, and selectivity can be calculated as follows:

$${\text{C}}_{2} {\text{H}}_{6} { }\,{\text{conversion}} = \left[ {\frac{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{6} \,{\text{converted}}}}{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{6} \,{\text{introduced}}}}} \right] \times 100\% ,$$

(8)

$${\text{C}}_{2} {\text{H}}_{4} \,{\text{selectivity}} = \left[ {\frac{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{4} \,{\text{produced}}}}{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{6} \,{\text{converted}}}}} \right] \times 100\% ,$$

(9)

$${\text{C}}_{2} {\text{H}}_{4} \,{\text{yield}} = \left[ {\frac{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{4} \,{\text{produced}}}}{{{\text{moles}}\,{\text{of}}\,{\text{C}}_{2} {\text{H}}_{6} \,{\text{introduced}}}}} \right] \times 100\% {.}$$

(10)

In addition, some more complicated reactions, such as methane reforming inside SOFC reactors [61, 62], which is described in Sect. 6.2.2.1 can be calculated as follows:

$${\text{CO}}_{2} \,{\text{conversion}} = \frac{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}\left[ {{\text{CO}}} \right]}}{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 2}}\right.\kern-\nulldelimiterspace} \!\lower0.7ex\hbox{$2$}}\left[ {{\text{CO}}} \right] + \left[ {{\text{CO}}_{2} } \right]}} \times 100\% ,$$

(11)

$${\text{CO}}\,{\text{selectivity}} = \frac{{\left[ {{\text{CO}}} \right]}}{{\left[ {{\text{CO}}} \right] + \left[ {{\text{CO}}_{2} } \right]}} \times 100\% ,$$

(12)

$${\text{H}}_{2} \,{\text{selectivity}} = \frac{{\left[ {{\text{H}}_{2} \,{\text{converted}}} \right]}}{{\left[ {{\text{H}}_{2} \,{\text{converted}}} \right] + \left[ {{\text{CO}}\,{\text{converted}}} \right]}} \times 100\% {.}$$

(13)

To calculate the conversion, selectivity, and yield, the species and production concentrations need to be defined both qualitatively and quantitatively. Most of these parameters can be determined by directly detecting the outcomes with nuclear magnetic resonance (NMR) spectroscopy [63, 64] and/or chromatographic techniques [65,66,67], such as high-performance liquid chromatography (HPLC), gas chromatography (GC), or even gas chromatography-mass spectrometry (GC–MS).

Another scheme is to “mark” the product with unique molecules as probes, which provides more possibilities for designing the monitoring part of fuel cell reactors. Sombatmankhong et al. reported a strategy to understand the formation, decomposition, and removal of hydrogen peroxide in the cathodic chamber by sensing a fluorescence signal in it using confocal microscopy [68]. Dichlorofluorescein (DCFH) can be used as a highly sensitive fluorescent probe that can be detected by confocal laser scanning microscopy. The quantitative hydrogen peroxide production, effect of operating potential on current efficiency, and electrochemical cogeneration performance have been characterized and investigated with this detection scheme.

4.2 Energy Density and Power Density

In a given region of space or in a system, the energy density is the amount of energy stored per unit volume. In most situations, only the extractable or useful energy is measured, i.e., the inaccessible energy (e.g., rest-mass energy) is ignored [69, 70].

The power density, i.e., the rate of energy flux per unit of area/mass, is a key determinant of the nature and dynamics of energy systems and is an important but largely overlooked parameter [71, 72], because any understanding of complex energy systems must rely on quantitative measurements of many fundamental variables. In energy transformers, such as batteries, fuel cells, motors, and other similar systems (e.g., power supply units), the power density refers to a volume. Therefore, this parameter is also called the volume power density and is expressed as W m^–3 [73, 74]. Sometimes the mass power density is an important consideration where weight is constrained, such as in electrical vehicles [71, 75].

Different from pure fuel cell systems, in which the energy density is mainly determined by the fuel type and operating conditions [76,77,78], in our cogeneration systems, “fuels” (or the oxidizing agents fed to the cathodes) have two roles: as feedstocks for realizing value-added chemicals and for producing electrical energy [43]. Therefore, controlling both the input type and operating conditions should be considered first to produce chemicals with higher value and purity, and the energy density or power density should be a secondary evaluation standard after conversion and selectivity [79,80,81].

5 System Approaches for Cogeneration

5.1 Solid Oxide Fuel Cell (SOFC) Reactors

As electrochemical energy conversion devices, SOFCs offer extraordinary promise for delivering significant electrical efficiency and important environmental benefits. In terms of fuel flexibility, as well as efficient (with fuel regeneration of higher than 70%) and clean electric power generation [82, 83], SOFCs produce valuable electricity via reactions between a fuel and pure oxygen (or oxygen in the air); this is accomplished by the diffusion of oxide ions through an ion-conducting solid electrolyte layer from the cathode to the anode or by protons produced from the fuel that diffuse in the opposite direction [84, 85]. SOFCs are composed of a dense electrolyte layer sandwiched between two porous electrodes, i.e., the cathode and anode, as shown in Fig. 2, using an oxide ion (Fig. 2a) and/or a proton (Fig. 2b) conductor as the electrolyte. The external circuit is completed when the electrons generated from fuel oxidation at the anode are accepted for oxygen reduction at the cathode [86, 87].

Fig. 2

Schematic diagram of a solid oxide fuel cell (SOFC) showing a an oxide ion-conducting electrolyte and b a proton-conducting electrolyte during operation. A key advantage of SOFCs is their fuel flexibility (i.e., they can also utilize hydrocarbons as a fuel source)

Figure 2 schematically shows the configuration of an SOFC. In this system, the solid electrolyte can be an oxygen ion (O^2–) conductor, proton (H⁺) conductor, or another type of conductor. The materials for oxygen ion conductors, proton conductors and another type of conductors are summarized in Tables 1, 2, and 3, respectively. A typical cell consists of a dense solid electrolyte membrane and two porous electrodes. To realize whole redox reactions in the cell, the cathode in the figure is exposed to an oxygen-containing gas or other kinds of oxidizing agent in a gaseous state; the other electrode, i.e., the anode, is exposed to the reacting mixture, which is mostly composed of a reducing agent in a gaseous state that acts as the fuel.

Table 1 O²⁻-conducting electrolyte membranes employed in fuel cell reactors, as reported in the open literature

Table 2 H⁺-conducting electrolyte membranes employed in fuel cell reactors, as reported in the open literature

Table 3 Mixed-ion/other ion-conducting electrolyte membranes employed in fuel cell reactors, as reported in the open literature

For simplicity, most of the overall chemical reactions can be written stoichiometrically as follows:

$${\text{A}} + n{\text{O}}^{2 - } \leftrightarrow m{\text{B,}}$$

(14)

or

$${\text{A}} - n{\text{H}}^{ + } \leftrightarrow m{\text{B}}{.}$$

(15)

As shown in Fig. 3, taking an SOFC with an oxygen ion (O^2–) conductor as an example, the two electrodes can be connected to a voltmeter (Case a), to an external resistive load (Case b), or to an external power source (Case c) [23]. Due to the different chemical potentials of oxygen at the two sides (the anode and cathode) in the cell, the driving force for oxygen transport across the electrolyte may operate the cell in one of the following modes [23], which are marked as a, b, and c in Fig. 3.

• (a) In the open-circuit configuration, which can also be applied in oxygen activity sensors, the spontaneous movement of oxygen to the low-oxygen side is counterbalanced by the large resistance of the voltmeter; hence, there is no net current through the electrolyte. The difference in chemical potential is converted into the open-circuit electromotive force (emf) of the cell.

• (b) In the closed-circuit configuration, oxygen ions travel from the cathode to the anode to react with the “fuel” (reducing agent), which is oxidized to H₂O and/or CO₂ or partially oxidized to value-added products, according to Reaction (14). Simultaneously, the electric circuit is closed, and if the goal is to produce electricity only, chemical energy can be directly converted into electrical energy.

• (c) If, on the other hand, the primary goal is not electricity production but the electrochemical production of (partially) oxidized products, the external power source can be used to impose a current, which is equivalent to an oxygen flux, through the cell in the desired direction. This operation mode is called electrochemical oxygen “pumping”. The term “pumping” can be used to describe an externally driven flux regardless of the flow direction, especially when oxygen is forced to flow opposite to the thermodynamically expected direction. This case represents an electrical energy-consuming mode.

Fig. 3

Copyright © 2013, the American Chemical Society

Three modes of operation of solid electrolyte cells: a as oxygen activity sensors, b as fuel cells, and c as electrolyzers or electropromotion reactors. Reprinted with permission from Ref. [88].

It has been stated for decades that one important goal for energy technologies and chemical reactor development is efficiency and environmental benefits. At the forefront of these kinds of technologies are SOFCs, which have been most intensively studied for the direct conversion from chemical energy in a fuel (hydrocarbon, hydrogen, ammonia, hydrothion, etc.) into electricity with high thermodynamic efficiency and minimal environmental pollution [89, 90].

To simultaneously harvest value-added chemicals and electrical energy based on the strength of the superiorities mentioned above, many efforts have been made to selectively oxidize hydrocarbons to valuable compounds with the simultaneous cogeneration of power in SOFC reactors [91, 92]. To estimate the thermodynamic tendency of the reaction, the open-circuit voltages (OCVs) for a single cell based on the selective oxidation reaction can be calculated as shown below.

Taking electrochemical oxidative ethane dehydrogenation as an example [59]:

$${\text{C}}_{2} {\text{H}}_{6} \left( {\text{g}} \right) + \frac{1}{2}{\text{O}}_{2} \left( {\text{g}} \right) = {\text{C}}_{2} {\text{H}}_{4} \left( {\text{g}} \right) + {\text{H}}_{2} {\text{O}}\left( {\text{g}} \right),$$

(16)

$$E = E^{0} \left( T \right) - \frac{RT}{{2F}}\ln \frac{{P_{{{\text{H}}_{2} {\text{O}}}} P_{{{\text{C}}_{2} {\text{H}}_{4} }} }}{{P_{{{\text{C}}_{2} {\text{H}}_{6} }} P_{{{\text{O}}_{2} }}^{\frac{1}{2}} }},$$

(17)

$$E^{0} = - \frac{{{\Delta }G^{0} \left( T \right)}}{2F},$$

(18)

where E⁰ is the thermodynamic reversible potential, as determined from the Gibbs free energy ${\Delta }G^{0} \left( T \right)$; this value can be obtained from heat capacity software, such as HSC Chemistry [93]. It has been found that under closed-circuit conditions, the conversion, i.e., selective oxidization of the fuel, is strongly dependent on the current [94], which is adjustable by changing the loading (resistance) in the external circuit so that the conversion increases with increasing current.

In reactors based on SOFCs, most cogeneration reactions employ the anodic side as a zone for value-added chemical production, i.e., selective oxidation of the “fuels”. Selective oxidation mainly uses the O^2– ions transferred from the cathodic side (Reaction (14)) [95, 96] or dehydrogenizes the fuel to eliminate some H atoms and generate unsaturated bonds (Reaction (15)) [97, 98]; notably, O^2–-conducting and H⁺-conducting electrolyte membranes are used, respectively, for such reactors. However, due to the poor controllability of the oxidation degree, the latter system has much higher selectivity, which will be discussed in detail in Sect. 6.1.

“Selective oxidation” can be understood not only as “selective” groups in the fuel but also as “selective” unique compositions of the mixed gaseous fluid oxidized in the cogeneration process. The latter definition is beneficial for adjusting the components according to a theoretical ideal, which is significant for application and achieving economic value. One possible strategy for a cogeneration system adopts a novel CH₄–CO₂ dry reforming [99] process before selectively oxidizing the in situ produced H₂ in high-performance proton-conducting SOFCs to coproduce CO-concentrated syngas and electricity; this can be achieved by optimally using the heat generated in the selective H₂ oxidation step.

An additional layered structure reported by Luo et al. [61] can be incorporated to form an anode support in a layered SOFC with a Ni_0.8Co_0.2-La_0.2Ce_0.8O_1.9 (NiCo-LDC) composite for a dry reforming process. Figure 4 is a schematic illustration of this novel layered SOFC design. The cross-sectional scanning electron microscopy (SEM) images of the layered proton-conducting SOFCs (an H-SOFC or PC-SOFC) show how the designed construction was achieved. A feed stream composed of CH₄ and CO₂ was fully reformed through the high-performance NiCo-LDC catalyst layer, yielding syngas (CO and H₂ with a ratio of 1:1) in anode reactions in which part of the H₂ in the syngas was consumed by selective oxidation on the anodic catalyst Ni-BaZr_0.1Ce_0.7Y_0.1Yb_0.1O_3−δ (BZCYYb). In this selective oxidation step, not only were electrical power and CO-enriched syngas generated but the total amount of heat required for internal dry reforming was provided. According to the thermodynamic data below, thermal independence could be obtained when H₂ utilization was higher than 52% (corresponding to a total fuel utilization of 26%). Thus, the energy input for conventional methane reforming [100, 101] can be significantly counterbalanced.

$${\text{CH}}_{4} \left( {\text{g}} \right) + {\text{CO}}_{2} \left( {\text{g}} \right) = 2{\text{CO}}\left( {\text{g}} \right) + 2{\text{H}}_{2} \left( {\text{g}} \right),\quad {\Delta }H = 260{ }\,{\text{kJ}}\,{\text{mol}}^{ - 1} { }\left( {700\,^{^\circ } {\text{C}}} \right);$$

(19)

$$2{\text{H}}_{2} \left( {\text{g}} \right) + {\text{O}}_{2} \left( {\text{g}} \right) = 2{\text{H}}_{2} {\text{O}}\left( {\text{g}} \right),\quad {\Delta }H = - 495\,{\text{kJ}}\,{\text{mol}}^{ - 1} { }\left( {700\,^{^\circ } {\text{C}}} \right).$$

(20)

Fig. 4

Reproduced with permission from Ref. [61]. Copyright © 2016, the Royal Society of Chemistry

a Schematic showing the configuration of the novel layered H-SOFC. Cross-sectional SEM images of the layered H-SOFC: b microstructure of the NiO-BZCYYb/BZCYYb/NBCaC-BZCYYb triple layer, c magnified image of the NBCaC-BZCYYb cathode, d microstructure of the NiCoO-LDC/NiO-BZCYYb interphase; and e EDX elemental mappings of (d).

Similar to the above approach for selectively oxidizing the sweep gas immediately after the reactions in the reforming layer, another alternative structure can be designed in the opposite order as the route of successive reactions, i.e., to reform the off-gas from a fuel cell reactor, in which partial oxidation occurs simultaneously with the harvesting of electrical energy from the difference in Gibbs free energy of the feed gas and the selectively oxidized products. Shao et al. [79] integrated a downstream catalyst in a single-chamber SOFC to maximize its cogeneration function. A bilayer electrolyte fuel cell reactor operated in an atmosphere of CH₄:O₂ = 2:1 achieved an OCV of 1.07 V and a maximum peak power density of approximately 1 500 mW cm^–2 at 700 °C while also achieving the partial oxidation of methane to CO, CO₂, H₂, and H₂O. Then, the effluent gas passed through a GdNi/Al₂O₃ catalyst at 850 °C, and synthesis gas (syngas) was obtained. In the whole reaction from methane to syngas, the conversion was higher than 95%, with a greater than 98% CO and H₂ selectivity and a H₂/CO ratio of approximately two. The authors also found that neither the syngas formation rate nor the H₂/CO molar ratio was affected by the polarization current density. This design successfully split Reaction (21) into two steps as follows (Fig. 5):

$$2{\text{CH}}_{4} + {\text{O}}_{2} \to 2{\text{CO}} + 4{\text{H}}_{2} .$$

(21)

Fig. 5

Copyright © 2011, Wiley

Fuel cell reactor system used for the cogeneration of synthesis gas and electric power from methane. The fuel cell and the GdNi-Al₂O₃ catalyst were located in a single quartz tube reactor (Φ = 14 mm and L = 300 mm). The feed gas was composed of methane and oxygen at a molar ratio of 2:1, which was introduced from one side of the reactor, first passed through the fuel cell, and then the catalyst layer. Both the fuel cell and catalyst were heated in electric furnaces. Reprinted with permission from Ref. [79].

Step one:

$${\text{cathode,}}\,\,{\text{O}}_{2} + 4{\text{e}}^{ - } \to 2{\text{O}}^{2 - } ;$$

(22)

$${\text{anode,}}\,\,{\text{CH}}_{4} + \frac{1}{2}{\text{O}}_{2} \to {\text{CO}} + 2{\text{H}}_{2} ,$$

(23)

$${\text{CO}} + {\text{O}}^{2 - } \to {\text{CO}}_{2} + 2{\text{e}}^{ - } ,$$

(24)

$${\text{H}}_{2} + {\text{O}}^{2 - } \to {\text{H}}_{2} {\text{O}} + 2{\text{e}}^{ - } .$$

(25)

Step two:

$${\text{CH}}_{4} + \frac{1}{2}{\text{O}}_{2} \to {\text{CO}} + 2{\text{H}}_{2} ,$$

(26)

$${\text{CH}}_{4} + {\text{CO}}_{2} \to 2{\text{CO}} + 2{\text{H}}_{2} ,$$

(27)

$${\text{CH}}_{4} + {\text{H}}_{2} {\text{O}} \to {\text{CO}} + 3{\text{H}}_{2} .$$

(28)

Selectivity is guaranteed by reduction of the “overoxidized” CO₂ and H₂O in Step one back to CO and H₂ in Step two, and part of the combustion energy in Reaction (21) is converted to electrical energy in Step one.

To explain and guide the design of reactors based on SOFCs, some modeling results have been published [102,103,104]. Barnett et al. [102] evaluated strategies for scaling electrochemical partial oxidation (EPOX) processes from the laboratory scale to the practical application scale using computational models. As shown in Fig. 6, the reaction in their research comprised the partial oxidation of hydrocarbon fuel streams to simultaneously produce syngas (H₂ and CO) and electrical energy. Because a practical limitation in this reaction was that carbon tended to deposit on Ni-based anode catalysts, the authors explored the use of barrier layers to overcome carbon deposition. Syngas and electricity could be delivered through a designed tubular cell from methane as the primary fuel. The effects of the inlet velocity, barrier layer length, and cell potential were also investigated. The anodic catalysts are summarized in Table 4.

Fig. 6

Copyright © 2008, Elsevier

a Illustration of an anode-supported tubular solid oxide fuel cell reactor with a porous barrier layer in the entrance region. b Solution profiles for the nominal tube geometry and operating conditions. The middle panel shows the gas-phase composition of the fuel stream along the channel length. The lower panel shows the local current density, temperatures of the fuel and air streams, and the MEA structure as functions of axial position. The upper panels show the gas-phase composition in the pore spaces and the adsorbed-species coverages on Ni surfaces within the Ni-YSZ anode at three axial positions along the tube. The bottom of these graphs is at the channel interface, and the top is at the dense electrolyte interface. Reprinted with permission from Ref. [102].

Table 4 Anode catalysts for oxygenation in fuel cell reactors, as reported in the open literature

Similar research has been published by Ni et al. [103] (Fig. 7). The authors employed a 2D model developed for a tubular direct carbon solid oxide fuel cell (DC-SOFC) reactor for the cogeneration of CO and electricity. Factors including the distance between the carbon chamber and anode electrode (D_ce), the operating temperature, and the cell length were all considered. A comparison between electrolyte-supported and anode-supported DC-SOFCs was also made to understand the effects of the support type on the CO generation characteristics and electrical power output.

Fig. 7

Copyright © 2016, Elsevier

Schematic of an a electrolyte-supported DC-SOFC and b anode-supported DC-SOFC. Reprinted with permission from Ref. [103].

5.2 Polymer Electrolyte Membrane Fuel Cell (PEMFC) Reactors

PEMFCs have been developed as electrical power sources for portable and stationary power applications, even for vehicles [105]. Most PEMFCs work at temperatures between 20 and 100 °C and pressures between 1 and 5 bar (1 bar = 100 kPa) [106]; thus, not only is a highly proton-conductive polymer membrane, e.g., Nafion™, needed but also a supporting electrolyte or additional product-separating unit is required. As shown in Fig. 8, in the PEMFC anode, hydrogen, methanol, or other fuels can be oxidized, and protons can be produced (Reaction (29) or (30)) and transferred through the Nafion membrane to the cathode, at which pure oxygen or oxygen in the air is employed as the oxidizing agent of the overall fuel cell reaction to generate H₂O with protons (Reaction (31)).

$${\text{Anode:}}\,\,{\text{H}}_{2} \leftrightarrow 2{\text{H}}^{ + } + 2{\text{e}}^{ - } ,$$

(29)

or

$${\text{C}}_{x} {\text{H}}_{y} {\text{O}}_{z} + \left( {2x - z} \right){\text{H}}_{2} {\text{O}} \leftrightarrow \left( {4x - 2z + y} \right){\text{H}}^{ + } + \left( {4x - 2z + y} \right){\text{e}}^{ - } + x{\text{CO}}_{2} .$$

(30)

$${\text{Cathode:}}\,\,{\text{O}}_{2} + 4{\text{H}}^{ + } + 4{\text{e}}^{ - } \leftrightarrow 2{\text{H}}_{2} {\text{O}}{.}$$

(31)

Fig. 8

Schematic of a typical fuel cell reactor based on a polymer electrolyte membrane fuel cell (PEMFC) reactor

The proton-producing reaction at the anode can be understood as a “dehydrogenation reaction”, which can also be used for (partial) oxidation. Borrowing the concept of PEMFCs, the types of reagents for the cathode can be extended from oxygen to other substances with oxidizing ability. As mentioned above, most reactions with an inorganic or unsaturated organic material as an oxidizing agent can occur at the cathode in a PEMFC [68, 106,107,108,109,110,111] (Fig. 9), making use of the “hydrogenation reaction” at the cathode. Therefore, in some reactors, the zones for value-added chemical generation are the cathodic sides, which use H⁺ from the anode or produce OH^– for the anode [22, 112]:

$${\text{C}} + {\text{hH}}^{ + } \leftrightarrow {\text{lD,}}$$

(32)

$${\text{C}}{-}{\text{hOH}}^{ - } \leftrightarrow {\text{lD}}{.}$$

(33)

Fig. 9

Copyright © 2012, Elsevier

Schematic representation of the gas diffusion electrodes under investigation. Reprinted with permission from Ref. [111].

The standard free enthalpies $\left( {\Delta_{{\text{R}}} G^{\Theta } } \right)$ [113] of the overall fuel cell reactions can be calculated to be negative, from which the equilibrium voltages U₀ can also be obtained. In this way, the cogeneration of desired products and electrical energy in the same fuel cell reactor can be successfully approached.

Similar to PEMFC reactors, another option for cell design is a free electrolyte liquid phase with an additional proton-permeable membrane placed in the electrolyte between the anode and cathode to prevent the crossover of reagents or products [113, 114].

In principle, as shown in Fig. 10, alkaline anion exchange membrane fuel cell (AAEMFC) reactors have similar structures to PEMFC reactors; the main difference is the use of an alkaline anion exchange membrane (AAEM) rather than a polymer electrolyte membrane (PEM) [22]. The corresponding catalysts are summarized in Table 4.

Fig. 10

Schematic of a typical fuel cell reactor based on AAEMFCs

Similar to Mode (b) in Fig. 3 of Sect. 5.1 (closed-circuit operation mode in a fuel cell), selective oxidation on the anodic side in combined reactors based on PEMFC or AAEMFC reactors employs dehydrogenation using protons transferred through the membrane to the cathode or OH^– produced on the cathodic side [106, 108, 109, 115, 116]:

$${\text{A}} + n{\text{OH}}^{ - } \leftrightarrow m{\text{B,}}$$

(34)

or

$${\text{A}} - n{\text{H}}^{ + } \leftrightarrow m{\text{B}}{.}$$

(35)

One special case in this type is a biofuel cell reactor, in which microbes (including enzymes) are employed as the catalyst [117,118,119,120]. In these reactors, waste can be easily used as fuel [121], realizing simultaneous waste disposal, as shown in Fig. 11.

Fig. 11

Copyright © 2012, Science

Conversion of wastes into bioelectricity and chemicals by using microbial electrochemical technologies. Reprinted with permission from Ref. [121].

5.3 Other Types of Fuel Cell reactors

In addition to the predominant candidates, namely, PEMFCs and SOFCs, alkaline fuel cells (AFCs), phosphoric acid fuel cells (PAFCs), molten salt fuel cells (MSFCs), flow fuel cells (FFCs), and photocatalytic fuel cells (PFCs) have been employed as a basis for cogeneration reactors. Similarly, most reactors based on AFCs, PAFCs, MSFCs, and PFCs use H⁺ transferred through the electrolyte for reactions in the anode because this approach is more flexible than that of FFCs.

5.3.1 Alkaline Fuel Cell (AFC) Reactors

Initially, AFCs were designed with two electrodes separated by a porous matrix saturated in an aqueous alkaline solution, e.g., potassium hydroxide (KOH) [122, 123]. Because the aqueous alkaline solution can become “poisoned” by the conversion from KOH to K₂CO₃ due to CO₂ generated from fuel oxidation at the anode, more flexible choices have been provided by adopting different combinations of the electrolyte, catholyte, and anolyte. This kind of fuel cell has also been used as a basic cogeneration reactor model.

Zhou and Wu et al. [112] (Fig. 12) reported a reactor based on an alkaline ethanol fuel cell (AEFC) reactor for harvesting electricity and simultaneously reducing Cr₂O₇^2– in wastewater, i.e., the contaminant (Cr(VI)) also acts as the oxidant:

$${\text{Cr}}_{2} {\text{O}}_{7}^{2 - } + 14{\text{H}}^{ + } + 6{\text{e}}^{ - } \to 2{\text{Cr}}^{3 + } + 7{\text{H}}_{2} {\text{O}},\quad E_{0} = 1.33\,{\text{V}}{.}$$

(36)

Fig. 12

Copyright © 2013, the American Chemical Society

Schematic diagram of the fuel cell reactor based on the AEFC. Reprinted with permission from Ref. [112].

An agar-gel electrolyte containing saturated KNO₃ was chosen as the catholyte (1 M H₂SO₄) and anolyte (0.5 M KOH) (1 M = 1 mol L⁻¹). The electrolytes and the cathodic catalysts are summarized in Tables 3 and 5, which are in Sects. 6.1.3 and 6.3.2, respectively.

Table 5 Cathode catalysts for other reactions in fuel cell reactors, as reported in the open literature

Similarly, Wen et al. [124] assembled an alkaline-acid Zn-H₂O fuel cell for the simultaneous generation of electricity. The OCV was approximately 1.25 V, and H₂ was produced with almost 100% Faradic efficiency (FE). The energy was harvested from both electrochemical Zn oxidation and electrochemical neutralization, as demonstrated in Fig. 13 and the equations below.

$${\text{Cathode:}}\,\,2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{H}}_{2} { }\left( {2.0{\text{M H}}_{2} {\text{SO}}_{4} } \right),\quad E_{{\text{c}}} = 0.035{ }\,{\text{V}}.$$

(37)

$${\text{Anode:}}\,\,{\text{Zn}} + 2{\text{OH}}^{ - } + 2{\text{e}}^{ - } \to {\text{ZnO}} + {\text{H}}_{2} {\text{O }}\left( {4.0\,{\text{M}}\,{\text{NaOH}}} \right),\quad E_{{\text{a}}} = - 1.285{ }\,{\text{V}}.$$

(38)

$${\text{Overall}}\,{\text{reaction:}}\,\,{\text{Zn}} + 2{\text{H}}^{ + } + 2{\text{OH}}^{ - } \to {\text{ZnO}} + {\text{H}}_{2} + {\text{H}}_{2} {\text{O,}}$$

(39)

$$V_{{{\text{cell}}}} = E_{{\text{c}}} - E_{{\text{a}}} = 1.32\,{\text{V,}}$$

(40)

$$E_{{{\text{theo}}}} = C_{{{\text{theo}}}} \cdot V_{{{\text{cell}}}} = 820\,{\text{mAh}}\,{\text{g}}^{ - 1} \times 1.32{ }\,{\text{V}} = 1 \, 082.4\,{\text{Wh }}\,{\text{kg}}^{ - 1}.$$

(41)

Fig. 13

Copyright © 2018, Wiley

Schematic illustration of the as‐proposed Zn-H₂O fuel cell, with a Zn plate in 4.0 M NaOH and Pt/CNTs loaded on a carbon‐cloth cathode in 2.0 M H₂SO₄, separated by a bipolar membrane. Reprinted with permission from Ref. [124].

Denayer et al. [125] (Fig. 14) employed a tank in series reactor (TSR) model for an AFC that cogenerated H₂O₂ and electricity in potentiostatic mode. The authors revealed the influence of mass transfer on the distribution of the current density and concentration.

Fig. 14

Copyright © 2012, Elsevier

TSR model of an AFC. Reprinted with permission from Ref. [125].

5.3.2 Phosphoric Acid Fuel Cell (PAFC) Reactors

PAFCs employ liquid phosphoric acid as the electrolyte and were the first kind of fuel cell to be successfully commercialized [122, 123]. PAFCs are also the generation preceding PEMFCs; liquid phosphoric acid was replaced with a solid PEM to perform the same function of conducting protons.

Otsuka et al. [126] (Fig. 15) adopted a cell with the configuration [alkene, H₂O|(Pd + vapor-grown carbon fiber (VGCF)) anode|H₃PO₄/silica wool|VGCF-cathode|O₂, NO] to realize NO₂ reduction and electricity generation. The alkene oxidation rate was accelerated by more than ten times with the addition of NO into the O₂ stream to the cathode. The cathode overpotential in this reactor was very small due to the very fast rate of reduction of NO₂ to NO and H₂O.

Fig. 15

Reproduced with permission from Ref. [126]. Copyright © 2003, The Electrochemical Society

Reaction mechanism for the electrochemical oxidation of propylene in the a [C₃H₆-O₂] and b [C₃H₆-(O₂ + NO)] fuel cell reactor systems.

Similarly, Pescarmona et al. [127] constructed a H₂-NO PAFC reactor for the cogeneration of hydroxylamine (NH₂OH) and electricity.

5.3.3 Molten Salt Fuel Cell (MSFC) Reactors

MSFCs use an electrolyte composed of a molten salt mixture suspended in a porous matrix [128, 129], mostly with operating temperatures at 650 °C or above [130, 131], especially for molten-carbonate fuel cells (MCFCs).

Datta et al. [132] (Fig. 16) used MSFC reactors in a supported molten salt electrocatalysis technique for the partial oxidation of ethylene with homogeneous (liquid phase) catalysts. In this system, to allow operating temperatures of up to 165 °C, the authors chose Nafion membranes impregnated with different electrolytic materials (molten salt containing dispersed or dissolved transition metal complex catalysts occupying a fraction of the pore space) as electrolytes. The reactions and mechanism of the reactor are provided in Sect. 6.2.2.4.

Fig. 16

Reproduced with permission from Ref. [132]. Copyright © 1996, The Electrochemical Society

Schematic diagram of a cogeneration reactor based on a supported molten salt electrocatalytic (SMSEC) fuel cell.

Freni et al. [133] (Fig. 17) presented an MCFC reactor system operated with and without a cogeneration bottoming cycle. The thermodynamic yield for direct internal methane catalytic partial oxidation (CPOX) could be accomplished with a theoretical energy efficiency ranging from 100% to 54.9%. This system appeared to be very flexible for application in plants to produce syngas for CH₂OH or NH₂ synthesis with electricity as a byproduct.

Fig. 17

Copyright © 1999, Elsevier

Schematic description of a CPOX-MCFC system. Reprinted with permission from Ref. [133].

5.3.4 Flow Fuel Cell (FFC) Reactors

FFCs use features of laminar flow to form stable interfaces between the anolyte and catholyte, removing the need for expensive semipermeable membranes [134,135,136]. FFCs, especially laminar flow fuel cells (LFFCs), can also be used as cogeneration reactors due to the following advantages.

• 1. Stable interfaces between the catholyte and anolyte [136].

• 2. Reduced cost.

• 3. Flexible assembly [135, 137].

• 4. Easy water management [133, 138].

• 5. Electrolyte pH flexibility [135, 139, 140].

• 6. Independently controllable pH for both the anolyte and catholyte [141].

Wouters et al. [142] (Fig. 18) used a microfluidic fuel cell-type reactor for nitrobenzene reduction and electricity production.

Fig. 18

Copyright © 2016, Elsevier

Exploded view of the cogeneration colaminar flow cell (CLFC). (1) POM piece with milled flow channels. (2) Electrocatalyst-coated graphite electrodes were placed in a Delrin® polyoxymethylene (POM) piece and glued to electrical wires with conductive epoxy. (3) Topas® cyclic olefin copolymer (COC) plate and rubber seal covering the channel. (4) Polymethylmethacrylate (PMMA) piece with holes that are used as nanoports for tubing. (5) Aluminum plates that hold the construction together. Reprinted with permission from Ref. [142].

Anode: methanol oxidation

$${\text{CH}}_{3} {\text{OH}} + 6{\text{OH}}^{ - } \to {\text{CO}}_{2} + 5{\text{H}}_{2} {\text{O}} + 6{\text{e}}^{ - } .$$

(42)

The overall cell reaction is as follows:

$${\text{C}}_{6} {\text{H}}_{5} {\text{NO}}_{2} + {\text{CH}}_{3} {\text{OH}} + 6{\text{OH}}^{ - } + 6{\text{H}}^{ + } \to {\text{C}}_{6} {\text{H}}_{5} {\text{NH}}_{2} + 7{\text{H}}_{2} {\text{O}} + {\text{CO}}_{2} .$$

(43)

Godoi et al. [143] employed a flow battery, which was designed with a hollow fiber of carbon nanotubes (cathode), Zn wire (anode), and 1‐ethyl‐3‐methylimidazolium tetrafluoroborate (electrolyte), to produce CH₄ from CO₂ under ambient conditions while also promptly outputting the gaseous product through the hollow fiber; this system exhibits a FE of up to 94%. The mechanistic scheme is shown in Fig. 19.

Fig. 19

Copyright © 2021, Springer

Proposed mechanistic scheme for the reactions on the a Zn anode and b–d CHF cathode. The inset in (a) shows the numbered [EMIM] + . Reprinted with permission from Ref. [143].

5.3.5 Multiple-Membrane Fuel Cell (MMFC) Reactors

Xia et al. [144] (Fig. 20) reported a different kind of fuel cell reactor with a multiple-membrane structure (three layers): the cationic and anionic products were formed and then combined after diffusion.

Fig. 20

Copyright © 2019, Science

Schematic illustration of the two different H₂O₂ synthesis methods using H₂ and O₂. a Synthesis of H₂O₂ using diluted H₂ and O₂ at high pressure. Methanol, which is used to improve the solubility of the reacting gases in the medium, must then be removed downstream. Other studies that avoid alcohols have been performed in acidic solutions of either HCl or H₂SO₄, with NaBr or NaCl as promotors. b Electrosynthesis of H₂O₂ using pure H₂ and O₂ streams separately introduced to the anode and cathode, respectively. The solid electrolyte (SE) consisted of either functionalized styrene–divinylbenzene copolymer microspheres or inorganic Cs_xH_3–xPW₁₂O₄₀. Electrochemically generated cations (H⁺) and anions (HO₂^–), driven by the electric field, cross in the porous SE layer and recombine to form H₂O₂. DI water flows through the porous SE layer then dissolves H₂O₂ without any impurities. Reprinted with permission from Ref. [144].

2e^– O₂ reduction reaction (2e^–-ORR):

$${\text{O}}_{{2}} + {\text{H}}_{{2}} {\text{O}} + {\text{2e}}^{-} \to {\text{HO}}_{{2}}^{-} + {\text{OH}}^{-} .$$

(44)

H₂ oxidation reaction (HOR):

$${\text{H}}_{{2}} \to {\text{2H}}^{ + } + {\text{2e}}^{-} .$$

(45)

H₂O₂ formation:

$${\text{HO}}_{{2}}^{-} + {\text{H}}^{ + } \to {\text{H}}_{{2}} {\text{O}}_{{2}} .$$

(46)

5.4 Photoelectrochemical Cell (PEC) Reactors

Photoelectrochemical cells (PECs) electrooxidize water to hydrogen and oxygen by irradiating the anode with electromagnetic radiation. This approach has been referred to as artificial photosynthesis and has been suggested as a way of storing solar energy in the form of hydrogen for use as a fuel [145, 146].

$${\text{Cathode:}}\,\,4{\text{H}}^{ + } + 4{\text{e}}^{ - } \to 2{\text{H}}_{2} .$$

(47)

$${\text{Anode:}}\,\,4{\text{h}} + 2{\text{O}}^{2 - } \to {\text{O}}_{2} .$$

(48)

In other words, the cogeneration of chemical and electrical energy, even nonspontaneous reactions, can proceed in this kind of cell by assimilating optical energy.

Li and Wang et al. [147] reported a reactor (Fig. 21) using one photochemical-chemical loop linked with redox couples (Fe²⁺/Fe³⁺ and I^–/I^3–) for H₂S chemical absorption redox reactions and photoelectrochemical H₂ production as a proof of concept. H₂S was directionally oxidized into α-S rather than polysulfide, which was guaranteed by the “oxidizing agent”, namely Fe³⁺ or I^3–, and could be restored to its reduced state to close the loop. Therefore, in this reactor, the hazardous H₂S waste could be converted to value-added α-S and electrical energy.

Fig. 21

Copyright © 2014, Wiley

a Conventional photoelectrochemical approach and b proposed photoelectrochemical-chemical loop for H₂S splitting on an n-type photoelectrode linked with redox couples. Reprinted with permission from Ref. [147].

As a flexible template, Li and Wang et al. [148] further improved the reactors in Fig. 21: the authors used I^–/I^3– instead as the redox couple in the photoanode loop for the oxidation of H₂S to S; in the cathode, the reduction of O₂ was accomplished with anthrahydroquinone (H₂AQ) to produce hydrogen peroxide (H₂O₂) and anthraquinone (AQ), which was then electroreduced to H₂AQ. The photooxidation of H₂S to S and the indirect photoreduction of O₂ to H₂O₂ were integrated into one photochemical-chemical cycle, as shown in Fig. 22.

Fig. 22

Reproduced with permission from Ref. [148]. Copyright © 2014, The Royal Society of Chemistry

Schematic illustration of the selective production of H₂O₂ and S from O₂ and H₂S on the n-type electrode.

Zheng et al. [149] (Fig. 23) designed and implemented a new type of unassisted PEC system, essentially a light-driven fuel cell reactor, that could produce chemicals (with H₂O₂ as the main product) at both the BiVO₄ photoanode and C cathode. Their work was the first to demonstrate two-sided H₂O₂ generation with a maximum power density of 0.194 mW cm^–2. Both the water oxidation reaction (WOR, Reaction (49)) at the anode and the oxygen reduction reaction (ORR, Reaction (50)) are two-electron pathways.

$${\text{Anode:}}\,\,2{\text{H}}_{2} {\text{O}} \leftrightarrow {\text{H}}_{2} {\text{O}}_{2} + 2\left( {{\text{H}}^{ + } + {\text{e}}^{ - } } \right),\quad E_{{{\text{OX}}}}^{{\text{O}}} = - 1.763\,{\text{V.}}$$

(49)

$${\text{Cathode:}}\,\,{\text{O}}_{2} + 2\left( {{\text{H}}^{ + } + {\text{e}}^{ - } } \right) \leftrightarrow {\text{H}}_{2} {\text{O}}_{2} ,\quad E_{{{\text{red}}}}^{{\text{O}}} = 0.695\,{\text{V.}}$$

(50)

$${\text{Overall}}\,{\text{reaction:}}\,\,2{\text{H}}_{2} {\text{O}}\left( {\text{l}} \right) + {\text{O}}_{2} = 2{\text{H}}_{2} {\text{O}}_{2} ,\quad \Delta E_{{{\text{cell}}}} = - 1.068\,{\text{V}}{.}$$

(51)

Fig. 23

Copyright 2018, Wiley

a Schematic illustration of the design of a light-driven fuel cell reactor with spontaneous H₂O₂ generation. The light bulb illustrates the simultaneously produced electricity that flows through the external circuit. b Band diagram of the system. The conduction band (CB) and valence band (VB) edge positions of BiVO₄ straddle the redox potentials of O₂/H₂O₂ and H₂O/H₂O₂, suggesting the possibility of unassisted H₂O₂ production. The theoretical V_oc of the light-driven fuel cell reactor is 0.693 V, as estimated from the CB of BiVO₄ and O₂/H₂O₂ redox potential. Reprinted with permission from Ref. [149].

The competing reactions (four­electron pathway and one­electron pathway) can be repressed by the selection of electrode materials.

For some systems, the reactor design above cannot meet the requirements of mass transfer and light absorption; thus, effort has been put into improving the photoelectrode structure. Jia et al. [150] (Fig. 24) reported a rotating disk photocatalytic fuel cell (RDPFC) reactor using a highly efficient aqueous-film rotating disk; the thin aqueous film on the upper part was continuously refreshed via mass transfer for the simultaneous production of electricity and hydrogen with the concomitant degradation of a high concentration of dye in wastewater.

Fig. 24

Copyright © 2014, Elsevier

a Schematic diagram of the aqueous-film RDPFC reactor. b Direct photoexcitation pathway of TiO₂ under UV light irradiation. Configuration of the anode compartments in c the aqueous-film RDPFC and d the conventional immersion reactor. The cell is filled with the sample solution. The figure is not to scale. Reprinted with permission from Ref. [150].

5.5 New Types of Cogeneration Reactors

The concept of “cogeneration”, i.e., the simultaneous generation of electrical energy and production of value-added chemicals, can be expanded into reactors based on energy-converting devices other than fuel cells. For example, considering only batteries, the discharge reaction of the whole battery can be understood as a fuel cell reaction that converts chemical energy to electrical energy through electrochemical reactions at both the anode and cathode. Here, we take a nitrogen fixation reaction in a battery as an example.

Zhang et al. [27] (Fig. 25) experimentally proved a proposed reversible Reaction (54) and reported the successful implementation of this reaction based on a reversible N₂ cycle in a rechargeable lithium-nitrogen (Li-N₂) battery. A Li anode, ether-based electrolyte, and carbon cloth cathode composed the assembled N₂-fixation battery system, which is not only a strategy for next-generation electrochemical energy storage systems (Reactions (52) to (53)) but also provides promising candidates for N₂ fixation.

$${\text{Anode:}}\,\,6{\text{Li}} \leftrightarrow 6{\text{Li}}^{ + } + 6{\text{e}}^{ - } .$$

(52)

$${\text{Cathode:}}\,\,6{\text{Li}}^{ + } + {\text{N}}_{2} + 6{\text{e}}^{ - } \leftrightarrow 2{\text{Li}}_{3} {\text{N.}}$$

(53)

$${\text{Overall:}}\,\,6{\text{Li}}^{ + } + {\text{N}}_{2} \leftrightarrow 2{\text{Li}}_{3} {\text{N}}{.}$$

(54)

Fig. 25

Copyright © 2017, Elsevier

Based on a rechargeable lithium-nitrogen battery, an advanced strategy for reversible nitrogen fixation and energy conversion has been successfully implemented at room temperature and atmospheric pressure. It shows a promising nitrogen fixation FE and superior cyclability. Reprinted with permission from Ref. [27].

Furthermore, to the best of our knowledge, NH₃ can be produced from Li₃N by reacting with water:

$${\text{Li}}_{3} {\text{N}} + 3{\text{H}}_{2} {\text{O}} \to {\text{NH}}_{3} + 3{\text{LiOH}}{.}$$

(55)

The production of NH₃ can be verified by the hydrolysis reaction upon combination with Nessler’s reagent to form NH₂Hg₂OI, which produces a yellow complex:

$$2\left[ {{\text{HgI}}_{4} } \right]^{2 - } + {\text{NH}}_{3} + 3{\text{OH}}^{ - } \to {\text{NH}}_{2} {\text{Hg}}_{2} {\text{OI}} + 7{\text{I}}^{ - } + 2{\text{H}}_{2} {\text{O}}.$$

(56)

Considering only discharge reactions, Li can be oxidized as a fuel with the simultaneous reduction of N₂ and production of NH₃ in the same electrochemical reactor. Because Li is recyclable, ammonia synthesis can be carried out on the cathodic side in this reactor. Nitrogen fixation is very important for industrial engineering and agriculture, and this reactor design may provide a solution.

Similarly, Zhi et al. [151] (Fig. 26) and Ma et al. [152] reported an Al-N₂ battery.

$${\text{Anode:}}\,\,{\text{2Al}} + {\text{14AlCl}}_{{4}}^{ - } \leftrightarrow {\text{8Al}}_{{2}} {\text{Cl}}_{{7}}^{ - } + {\text{6e}}^{ - } .$$

(57)

$${\text{Cathode:}}\,\,{\text{8Al}}_{{2}} {\text{Cl}}_{{7}}^{ - } + {\text{N}}_{{2}} + {\text{6e}}^{ - } \leftrightarrow {\text{2AlN}} + {\text{14AlCl}}_{{4}}^{ - } .$$

(58)

$${\text{Overall:}}\,\,{\text{2Al}} + {\text{N}}_{{2}} \leftrightarrow {\text{2AlN}}.$$

(59)

Fig. 26

Reproduced with permission from Ref. [151]. Copyright © 2020, the Royal Society of Chemistry

Schematic diagram showing the achievement of both energy storage and N₂ fixation using the Al-N₂ battery system.

More similar cases have been revealed for other batteries involving low-priced reagents at the cathode and the reversible and/or irreversible generation of value-added intermediate products during discharge.

Wang et al. [153] reported a novel fuel-gas CO-generating Li-CO₂ battery with a porous fractal Zn cathode capable of tuning CO formation within a wide discharge current range and obtaining a maximum FE of 67% (Fig. 27). In this battery, the proposed CO production mechanism is as follows:

$$2{\text{Li}}^{ + } + 2{\text{CO}}_{2} + 2{\text{e}}^{ - } \to {\text{CO}} + {\text{Li}}_{2} {\text{CO}}_{3} .$$

(60)

Fig. 27

Reproduced with permission from Ref. [153]. Copyright © 2018, the Royal Society of Chemistry

Li-CO₂ battery with a porous fractal Zn (PF-Zn) cathode producing CO and electrical energy. a Schematic diagram showing Li-CO₂ batteries generating fuel-gas CO. b Galvanostatic discharge potentials at several discharge currents from 0.010 to 0.225 mA. c Galvanostatic discharge potential during long-term discharge at 0.1 mA. Open-circuit potential before 0 min. d Max Faradaic efficiency (FE) of CO at several currents during 10 h of discharge. e FE of CO during long-term discharge at 0.1 mA.

In most Li-CO₂, Na-CO₂, and metal-CO₂/O₂ batteries, solid discharge products, such as Li₂C₂O₄ [154] and peroxodicarbonate [154, 155], or solid-state carbon [156] can be obtained, as suggested by the following reaction:

$$4{\text{Li}}^{ + } + 3{\text{CO}}_{2} + 4{\text{e}}^{ - } \to {\text{C}} + 2{\text{Li}}_{2} {\text{CO}}_{3} .$$

(61)

The obtained carbon can exhibit different morphologies, e.g., a network of fibers [157]; if the shape and ingredients are controllable and the product has a high value, the discharge process can successfully realize cogeneration.

With additional reports on Metal-CO₂ battery cogeneration reactions [158,159,160,161,162,163], Xie et al. [162] presented metal-CO₂ batteries as the crossroad for realizing CO₂ cycling, in which simultaneous energy transformations and chemical changes exist in the same system (Fig. 28). This could be understood as an extension of cogeneration reactions.

Fig. 28

Reproduced with permission from Ref. [162]. Copyright © 2020, Wiley

Metal-CO₂ batteries as the crossroad between the Li/Na‐CO₂ system and Zn/Al‐CO₂ system to better support sustainable human life.

6 Material Design Strategies for Cogeneration Systems

6.1 Electrolyte Membranes for Fuel Cell Reactors

6.1.1 O²⁻-conducting Electrolyte Membranes

As mentioned above, in some cogeneration reactors, especially those based on SOFCs, O²⁻-conducting electrolyte membranes are employed as separators to split the anodic and cathodic reactions and to transfer the O²⁻ generated in the cathode to the anode. These ions act as the oxidizing agent for selective oxidation for achieving the simultaneous cogeneration of electrical energy and value-added chemicals.

Yang et al. [94] (Fig. 29) reported the use of an SOFC as a reactor with Bi₄Cu_0.2V_1.8O_11−δ as the O²⁻-conducting electrolyte membrane for the selective electrochemical oxidation of propane with the cogeneration of electric power and acrylic acid; notably, the selectivity reached 73%.

Fig. 29

Reproduced with permission from Ref. [94]. Copyright © 2009, the Royal Society of Chemistry

Mechanism of the selective oxidation of propane in the Bi₄Cu_0.2V_1.8O_11−δ (BICUVOX.10) SOFC reactor.

However, due to the poor controllability of selective oxidation from deep oxidation, the selectivity of reactors with O²⁻-conducting electrolyte membranes is not satisfactory.

6.1.2 H⁺-Conducting Electrolyte Membranes

Due to their benefits for the simultaneous cogeneration of chemicals and electricity with low or even no CO₂ emissions [60, 97, 194], PC-SOFCs have been developed to convert fuel to desired (value-added) chemicals and generate cleaner energy [113]. Additionally, compared with SOFC reactors with O²⁻-conducting electrolyte membranes, PC-SOFC reactors can better control the degree of oxidation.

As shown in Fig. 30a, due to the presence of oxygen in the reaction, it is not easy to control the degree of deep oxidation from ethane to CO₂ and thereby achieve high ethylene selectivity [195, 196], i.e., a large number of byproducts, such as CO and CO₂, are produced, which decreases the selectivity of ethylene and the total value of the products [197]. This problem can be overcome in PC-SOFCs because the dehydrogenation reactions are all realized by protons conducted through the electrolyte from the anode to the cathode.

Fig. 30

Reproduced with permission from The Royal Society of Chemistry. Reproduced with permission from Ref. [97]. Copyright © 2011, the PCCP Owner Societies

Schematic working principles of hydrocarbon SOFCs with a an oxide ion electrolyte and anode for hydrocarbon deep oxidation and b a proton-conducting electrolyte and dehydrogenation anode.

Proton-conducting electrolytes normally have higher ionic conductivity than oxide ion-conducting electrolytes; thus, PC-SOFCs, which can be operated at intermediate or low temperatures, have attracted increasing attention in recent years [198,199,200,201]. As shown in Fig. 30b, PC-SOFCs are employed as membrane reactors when the anode catalyzes the selective dehydrogenation of a dry hydrocarbon fuel to provide protons. More cogeneration reactors based on SOFCs with proton-conducting electrolyte membranes are summarized in Table 2.

Similarly, in most PEMFCs, Nafion membranes [202, 203] are used as H⁺-conducting electrolyte membranes (shown in Fig. 31). Nafion, a sulfonated tetrafluoroethylene-based fluoropolymer-copolymer, was discovered by Walther Grot of DuPont in the late 1960s [204, 205]. The unique ionic properties of Nafion are a result of the incorporation of perfluorovinyl ether groups terminated with sulfonate groups onto a tetrafluoroethylene (Teflon) backbone. Nafion has gained a considerable amount of attention as a proton-conducting membrane for PEMFCs [206, 207] because of its excellent thermal and mechanical stability [208, 209]. When PEMFCs are adopted for cogeneration reactors, the selection of proton-conducting electrolyte membranes is optimized; thus, these membranes have been applied in many practical devices, as listed in Table 2.

Fig. 31

Copyright © 2010, Elsevier

An idealized Nafion pore. Reprinted with permission from Ref. [210].

6.1.3 Mixed-Ion/Other Ion-Conducting Electrolyte Membranes

In addition to the above cogeneration reactors that use O²⁻- or H⁺-conducting electrolyte membranes with O²⁻ or H⁺ as the reactants or products of electrode reactions, another option in reactor design is to assume that electrolyte membranes are only used as separators between the anodic and cathodic sides to make a closed circle. In this case, mixed-ion or other ion-conducting electrolyte membranes could be an effective design choice.

As mentioned in Sect. 5.3.1, the AEFC reactor assembled by Zhou and Wu et al. [112] used an agar-gel electrolyte containing saturated KNO₃ as the “electrolyte membrane”, with Cr(VI) in H₂SO₄ (aq.) as the catholyte and C₂H₅OH in KOH (aq.) as the anolyte. The agar-gel electrolyte served only to split the two reactions and to constitute a closed circle, rather than perform mass transport for the reactions, as clearly demonstrated in Fig. 32.

Fig. 32

Copyright © 2013, American Chemical Society

Schematic diagram of AEFC. Reprinted with permission from Ref. [112].

In the example above, the authors classified their device as alkaline based on the anolyte rather than the electrolyte membrane. The most important difference between this structure and those in Sects. 6.1.1 and 6.1.2 is that the electrode reactions happen independently, with only electron shifts in the external circuit and no ion transfer through the electrolyte membranes, similar to Zn-H₂O fuel cell reactors [125]. The advantage of this kind of reactor is the flexibility of reaction selection on the two sides, and the disadvantage is the consumption of catholyte and anolyte in these reactions, which makes a circulation mechanism necessary for continuous operation.

6.2 Catalyst Materials for Value-Added Chemical Production at the Anode in Fuel Cell Reactors

Anode catalysts in fuel cell reactors, which produce target chemicals on the anodic side, should provide sites for selective oxidation to take place. In reactors for targeted product synthesis, according to the definition of “selective oxidation”, i.e., oxygenation and dehydrogenation, the corresponding catalysts can be classified into two categories, which are described below. For each category, typical examples and catalyst designs are presented.

6.2.1 Anode Catalysts for Oxygenation in Fuel Cell Reactors

Oxygenation reactions in organic chemistry can be defined as reactions increasing the oxidation number of C atoms or other atoms. This kind of chemical change should be accomplished on the anodic side of fuel cell reactors. Properly speaking, the dehydrogenation reaction should be a subset of the oxygenation reaction, but in considering holistic fuel cell reactors, we classify the catalysts for dehydrogenation separately in Sect. 6.2.2; such catalysts accomplish the target reactions by exporting dehydrogenated H atoms through proton-conducting electrolyte membranes.

In this part, several typical oxygenation reactions that take place at the three-phase boundary (shown in Fig. 33) will be introduced. These reactions employ O²⁻ ions imported from the cathode as the oxidizing agent, similar to anodic reactions in SOFCs with O²⁻-conducting electrolyte membranes; this is shown in the following figure. As mentioned in Sect. 6.1.1 The selectivity is insufficient due to the poor controllability of selective oxidation from deep oxidation.

Fig. 33

Copyright © 2002, Elsevier

Schematic illustration of the three-phase boundary at the anode. Reprinted with permission from Ref. [190].

6.2.1.1 Steam-Carbon Fuel Cell for the Cogeneration of H₂

The steam-carbon fuel cell concept has been proposed and investigated [95, 227, 228] by adopting and modifying the fluidized bed direct carbon fuel cell (FB-DCFC) concept [229, 230]. A bed of solid carbon particles in the anode compartment contacts the electrode surface, and the cathode compartment is exposed to steam [231].

The overall cell configuration (Fig. 34) is as follows:

$${\text{H}}_{2} ,{\text{ H}}_{2} {\text{O}}\left( {\text{g}} \right)/{\text{Pt}}/{\text{YSZ}}/{\text{Pt}}/{\text{C}}\left( {\text{s}} \right),{\text{ CO}},{\text{CO}}_{2} .$$

(62)

$${\text{Cathodic}}\,{\text{reaction:}}\,\,{\text{H}}_{2} {\text{O}} + 2{\text{e}}^{ - } \to {\text{H}}_{2} + {\text{O}}^{2 - } .$$

(63)

$${\text{Anodic}}\,{\text{reactions:}}\,\,{\text{CO}} + {\text{O}}^{2 - } \to {\text{CO}}_{2} + 2{\text{e}}^{ - } ,$$

(64)

$${\text{C}}\left( {\text{s}} \right) + {\text{CO}}_{2} \to 2{\text{CO}}.$$

(65)

$${\text{Net}}\,{\text{cell}}\,{\text{reaction:}}\,\,{\text{H}}_{2} {\text{O}} + {\text{C}}\left( {\text{s}} \right) \to {\text{CO}} + {\text{H}}_{2} .$$

(66)

Fig. 34

Reproduced with permission from Ref. [95]. Copyright © 2011, The Electrochemical Society

Schematic describing the operating principle of steam-carbon fuel cell reactors.

The OCV can be calculated with the Nernst equation:

$$E_{{{\text{OCV}}}} = E_{0} \left( T \right) - \frac{RT}{{nF}}\ln \frac{{P_{{{\text{H}}_{{2,{\text{ Cathode}}}} }} P_{{{\text{CO}}_{{{\text{Anode}}}} }} }}{{P_{{{\text{H}}_{2} {\text{O}}_{{{\text{Cathode}}}} }} }}.$$

(67)

The OCV can be eliminated by Reaction (66), and the cell potential can guarantee that the reactor is operated as a fuel cell, which enables the cogeneration of electrical power and hydrogen without the need for externally applied energy.

6.2.1.2 Oxidation of NH₃ to NO

A few papers have reported types of fuel cells that use NH₃ as a kind of fuel to simultaneously produce electrical energy and nitric oxide. For example, Vayenas et al. obtained nitric oxide with a yield higher than 60% along with simultaneous electrical energy production in an SOFC reactor: NH₃, NO, N₂, Pt/ZrO₂(8% Y₂O₃)/Pt, air [92, 185, 232]; notably, selectivity for NO was obtained between 900 and 1 200 K. Above 1 200 K, direct NH₃ decomposition to N₂ and H₂ became very fast. The selectivity was limited by the flux through the electrolyte, and thinner electrolyte walls must be used to maintain high selectivity at higher NH₃ molar flow rates.

6.2.1.3 Producing HCN from CH₃OH and NH₃

Hydrogen cyanide, which is important for the production of a variety of chemicals, is produced on a large scale worldwide. There are three main commercial hydrogen cyanide production routes, namely, the Andrussow ammoxidation process [233, 234] (Reaction (68)), the Blausäure-Methan-Ammoniak ammonolysis process [235] (Reaction (69)), and the Nitto ammoxidation process [236] (Reaction (70)):

$${\text{CH}}_{4} + {\text{NH}}_{3} + 1.5{\text{O}}_{2} \to {\text{HCN}} + 3{\text{H}}_{2} {\text{O,}}$$

(68)

$${\text{CH}}_{4} + {\text{NH}}_{3} \to {\text{HCN}} + 3{\text{H}}_{2} ,$$

(69)

$${\text{CH}}_{3} {\text{OH}} + {\text{NH}}_{3} + {\text{O}}_{2} \to {\text{HCN}} + 3{\text{H}}_{2} {\text{O}}{.}$$

(70)

However, all three classical reactions require high-temperature and expensive noble metal catalysts, such as Pt and Ru.

Atkinson et al. accomplished this process by using an SOFC reactor and simultaneously obtained electrochemical energy [166]. The authors used a cermet anode composed of Ni and Ce_0.9Gd_0.1O_1.95 (CGO10), a CGO10 electrolyte, and a La_0.6Sr_0.4Co_0.2Fe_0.8O_3−δ (LSCF)/CGO10 composite cathode at an intermediate temperature (500–650 °C). The electrode reactions are as follows.

$${\text{Cathode:}}\,\,4{\text{e}}^{ - } + {\text{O}}_{2} \to 2{\text{O}}^{2 - } .$$

(71)

$${\text{Anode:}}\,\,{\text{CH}}_{3} {\text{OH}} + {\text{NH}}_{3} + 2{\text{O}}_{2} \to {\text{HCN}} + 3{\text{H}}_{2} {\text{O}} + 4{\text{e}}^{ - } .$$

(72)

HCN in the anode off-gas reached a maximum yield of 40%, and the selectivity of HCN reached 47.5%. CO₂, N₂, H₂, and CH₄ were obtained as byproducts.

6.2.2 Anode Catalysts for Dehydrogenation in Fuel Cell Reactors

For dehydrogenation, anode catalysts in fuel cell reactors possess the vital function of pulling off unique H atoms from the reactants; these H atoms are transferred in the form of protons through the proton-conducting electrolyte membrane to participate in cathodic reactions. Most of these kinds of reactions occur at the anodes of PC-SOFC reactors. Representative reactions are classified and summarized below, and details for each type of reaction can be found in Table 6.

Table 6 Anode catalysts for dehydrogenation in fuel cell reactors, as reported in the open literature

6.2.2.1 Oxidation of CH₄ to CO₂ and H₂ or to CO and H₂ (Synthesis Gas (Syngas))

Since no oxygen ions can be moved to the anode to directly oxidize carbon species, the direct utilization of hydrocarbon fuels in PC-SOFCs as the only source of electrochemical energy remains challenging. However, this challenge can be overcome through an internal reforming process by reaction with either steam or CO₂ [61, 257,258,259]:

$${\text{CH}}_{4} + 2{\text{H}}_{2} {\text{O}} \to {\text{CO}}_{2} + 4{\text{H}}_{2} ,$$

(73)

$${\text{CH}}_{4} + {\text{CO}}_{2} \to 2{\text{CO}} + 2{\text{H}}_{2} .$$

(74)

In particular, the latter reaction, which is the production of syngas from CH₄ and CO₂, is defined as “dry reforming” [260,261,262]. An efficient approach based on the SOFC system for this energy conversion process is to include a microreformer [62] in the anode chamber [263, 264] for the internal reforming of biogas (IRB), which can effectively decrease the thermal gradients in the anode.

The high-temperature operation of SOFCs is beneficial to the endothermic reaction step and kinetics of all reactions. IRB can reduce both the system cost and complexity and promote intrinsic thermal coupling between endothermic (IRB, Reaction (75)) and exothermic (electrooxidation, Reactions (76) and (77)) reactions.

$${\text{CH}}_{4} \left( {\text{g}} \right) + {\text{CO}}_{2} \left( {\text{g}} \right) = 2{\text{CO}}\left( {\text{g}} \right) + 2{\text{H}}_{2} \left( {\text{g}} \right),\quad {\Delta }H = + 259\,{\text{kJ}}\,{\text{mol}}^{ - 1} { }\,\left( {800\,^{^\circ } {\text{C}}} \right);$$

(75)

$$2{\text{H}}_{2} \left( {\text{g}} \right) + {\text{O}}_{2} \left( {\text{g}} \right) = 2{\text{H}}_{2} {\text{O}}\left( {\text{g}} \right),\quad {\Delta }H = - 496\,{\text{kJ}}\,{\text{mol}}^{ - 1} \,\left( {800{ }\,^{^\circ } {\text{C}}} \right);$$

(76)

$$2{\text{CO}}\left( {\text{g}} \right) + {\text{O}}_{2} \left( {\text{g}} \right) = 2{\text{CO}}_{2} \left( {\text{g}} \right),\quad {\Delta }H = - 556\,{\text{kJ}}\,{\text{mol}}^{ - 1} \,\left( {800{ }\,^{^\circ } {\text{C}}} \right).$$

(77)

From the above reactions, it is clear that heat balance can be easily achieved with the utilization of more than 52% H₂.

Luo et al. [81] employed the characteristics of a PC-SOFC and a double perovskite oxide material (PrBaMn₂O_5+δ (PBM) [265], with excellent coke/sulfur resistance and crystal-structure stability in a reducing atmosphere) to accomplish in situ methane reforming, effective CO₂ utilization, and the cogeneration of electrical power through a NiCo/PBM bifunctional nanoarchitecture on BZCYYb in a porous nickel cermet anode using readily available H₂S-containing hydrocarbon fuels [252, 266].

Alternatively, the strategy of adding a reforming layer can be used in other fuel cell reactor structure designs [62]. Shao et al. [216] designed a La₂NiO₄ catalyst layer on the conventional anode of a PC-SOFC for CO₂ dry reforming, in which CO₂ is reformed with CH₄ to yield CO and H₂ before entering the anode of the PC-SOFC. The authors employed a Ruddlesden-Popper structure as a catalyst layer on a conventional NiO + BaZr_0.4Ce_0.4Y_0.2O_3−δ (NiO + BZCY4) anode to accomplish the in situ CO₂ dry reforming of methane with a corresponding CO selectivity of up to 93.3% (Table 6).

Another option is to selectively oxidize H₂ in the syngas to coproduce electricity and CO-enriched syngas with few CO₂ byproducts. Luo et al. [61] achieved the NiCo-LDC internal dry reforming of methane using a PC-SOFC with a layered structure. It has proven that multiple-twinned bimetallic nanoparticles (Fig. 35) have superior activity toward in situ dry reforming. From the results revealed in Fig. 35, compared to conventional designs, this layer-structured SOFC demonstrated high internal reforming performance (CO₂ conversion reached 91.5% at 700 °C) and drastically improved CO₂ resistance. With a CH₄–CO₂ feed stream of 1 A cm⁻², the structure achieved up to 100 h of galvanostatic stability. Moreover, by effectively and exclusively converting H₂ by electrochemical oxidation, CO-concentrated syngas with no CO₂ could be harvested in the anode effluent with a maximum power density of more than 910 mW cm⁻² at 700 °C and a polarization resistance as low as 0.121 Ω cm².

Fig. 35

Reproduced with permission from Ref. [61]. Copyright © 2016, the Royal Society of Chemistry

TEM analyses of NiCo-LDC catalysts: a BF micrograph; b–d HRTEM images of NiCo bimetallic nanoparticles at the corresponding sites labeled in (a), with the dotted lines indicating the twin boundaries; and e schematic of a multiple-twinned nanoparticle. H₂ selective oxidation was evaluated at 700 °C in layered SOFCs fed with syngas containing various levels of H₂. f H₂ selectivity and exhaust gas composition as a function of H₂ percentage in the syngas and g corresponding voltage responses with different feeds at a constant current load of 2 A cm⁻².

From the examples above, we can categorize strategies for dry reforming into two types: one strategy is to accomplish the process in an anodic reaction, and the other is to add a reforming layer before the anodic reaction to optimize the composition for the subsequent electrochemical conversion. The corresponding catalyst designs for the PC-SOFC anodic reaction and the front reforming step are clear.

6.2.2.2 Olefine Acids Produced from Alkanes

As mentioned in Sect. 6.1.1 and demonstrated in Fig. 29, Yang et al. [94] reported a highly efficient catalyst, namely, MoV_0.3Te_0.17Nb_0.12O, as the anodic catalyst in an SOFC reactor for the production of acrylic acid from propane with a selectivity of up to approximately 73% along with the simultaneous production of electricity at 400–450 °C [94]. Selective oxidation from propane to acrylic acid can be realized by transferring O²⁻ through the electrolyte membrane from the cathode:

$${\text{C}}_{3} {\text{H}}_{6} + 4{\text{O}}^{2 - } \to {\text{CH}}_{2} = {\text{CHCOOH}} + 2{\text{H}}_{2} {\text{O}} + 8{\text{e}}^{ - } .$$

(78)

6.2.2.3 Olefins Produced from Alkanes

For industrial purposes, the predominant method for the production of ethylene is steam cracking [267,268,269]. Because the reaction is reversible, highly endothermic, and severely limited by thermodynamic equilibrium, high-temperature operating conditions are critical but consume energy.

As mentioned in Sect. 6.1.2, due to the high selectivity, i.e., low CO/CO₂ emission, of fuel cell reactors with proton-conducting electrolytes [62, 270,271,272], increasingly good results have been reported, especially in alkane elimination reactions. In this kind of cogeneration system, the anode serves as a selective electrocatalytic dehydrogenation reaction zone, which is a key challenge for direct hydrocarbon PC-SOFC reactors [60, 194].

Fu et al. [97] reported the use of a new, inexpensive, high-performance Cu-Cr₂O₃ nanocomposite catalyst derived from CuCrO₂ nanoparticles in a dehydrogenation anode for use in hydrocarbon PC-SOFC reactors (Fig. 36). The efficacy of the new catalyst was demonstrated through the efficient and selective conversion of ethane to ethylene, with an ethylene yield ranging from approximately 9% to 39% and a selectivity ranging from 99% to 90% at operating temperatures ranging from 650 to 750 °C, without any CO₂ emission. The simultaneously cogenerated electrical energy possessed a maximum power density of 170 mW cm⁻². The mechanism for ethane electrocatalytic dehydrogenation in the SOFC reactor was also proposed.

Fig. 36

Reproduced with permission from The Royal Society of Chemistry. Reproduced with permission from Ref. [97]. Copyright © 2011, the PCCP Owner Societies

Proposed mechanism for electrocatalytic ethane dehydrogenation to ethylene over a Cu-Cr₂O₃ composite anode in a PC-SOFC reactor.

A new kind of anode material composed of double-layered perovskite ((Pr_0.4Sr_0.6)₃(Fe_0.85Mo_0.15)₂O₇ (DLP-PSFM)) with a uniformly dispersed in situ exsolution of Co-Fe alloy nanoparticles was synthesized by Luo et al. [59] (Fig. 37) by annealing cubic perovskite (Pr_0.4Sr_0.6Co_0.2Fe_0.7Mo_0.1O_3−δ) in a reducing atmosphere. A maximal output power density of 496.2 mW cm⁻² in H₂ and 348.84 mW cm⁻² in C₂H₆ at 750 °C was achieved. Moreover, due to the considerably efficient catalysis on the in situ Co-Fe alloy nanoparticles that are homogeneously distributed on the DLP-PSFM backbone, a high ethylene yield was achieved, increasing from 13.2% at 650 °C to 41.5% at 750 °C with a remarkable ethylene selectivity of over 91% and no CO₂ emission. Lu et al. [253] also used proton-conducting fuel cells and (Pr_0.3Sr_0.7)(_0.9)Ni_0.1Ti_0.9O₃ as the anodic material to selectively convert propane to propylene and ethylene, respectively, as shown in Fig. 38.

Fig. 37

Copyright © 2016, the American Chemical Society

BaCe_0.7Zr_0.1Y_0.2O_3−δ electrolyte-supported PC-EFC single cell fabricated with the new DLP-PSFM anode material. Reprinted with permission from [59].

Fig. 38

Copyright 2010, Elsevier

Cogeneration of electricity and olefin via proton-conducting fuel cells using (Pr_0.3Sr_0.7)(_0.9)Ni_0.1Ti_0.9O₃ catalyst layers. Reprinted with permission from Ref. [253].

6.2.2.4 Aldehyde Production from Alcohol

Datta et al. used an MSFC with a homogeneous liquid phase to realize the partial oxidation of alcohol to aldehyde and simultaneously coproduce electric power.

$${\text{Anode:}}\,\,{\text{PdCl}}_{4}^{2 - } + {\text{C}}_{2} {\text{H}}_{5} {\text{OH}} \to {\text{CH}}_{3} {\text{CHO}} + {\text{Pd}}^{0} + 2{\text{HCl}} + 2{\text{Cl}}^{ - } ,$$

(79)

$$2{\text{Cl}}^{ - } + {\text{Pd}}^{0} + 2{\text{CuCl}}_{2} \to {\text{PdCl}}_{4}^{2 - } + 2{\text{CuCl,}}$$

(80)

$$2{\text{CuCl}} + 2{\text{HCl}} \to 2{\text{CuCl}}_{2} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } .$$

(81)

Overall for the anode:

$${\text{C}}_{2} {\text{H}}_{5} {\text{OH}} \to {\text{CH}}_{3} {\text{CHO}} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } .$$

(82)

Cathode:

$$\frac{1}{2}{\text{O}}_{2} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{H}}_{2} {\text{O}}{.}$$

(83)

Experiments demonstrated the efficiency of the Wacker catalyst (PdC1₂/CuCl₂) (Fig. 39) dispersed in molten salt for the production of acetaldehyde from ethanol at relatively mild temperatures, in which the overall reaction mechanism was assumed to be analogous to that of the Wacker process for ethylene [273].

Fig. 39

Reproduced with permission from Ref. [131]. Copyright © 1996, The Electrochemical Society

Proposed mechanism for acetaldehyde synthesis by oxidizing the dehydrogenation of ethanol on a Wacker catalyst.

6.2.2.5 Oxidation of CH₄ to CH₃OH

The conversion of methane and steam to methanol can be classified as an expanded form of dehydrogenation. Hibino et al. [219] reported this process in a fuel cell reactor where a mixture of methane and H₂O vapor was supplied to the anode and air was supplied to the cathode.

At the anode, there were two competing reactions:

$${\text{CH}}_{4} + 2{\text{H}}_{2} {\text{O}} \to {\text{CO}}_{2} + 8{\text{H}}^{ + } + 8{\text{e}}^{ - } ,$$

(84)

and

$${\text{CH}}_{4} + {\text{H}}_{2} {\text{O}} \to {\text{CH}}_{3} {\text{OH}} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } .$$

(85)

Therefore, when decreasing the oxidizing ability of active oxygen species for methane, the selective oxidation of methane to methanol (Reaction (85)) was achieved by maintaining a high current efficiency.

Furthermore, fundamental limitations [274] (Fig. 40), materials and device targets [275] have also been studied.

Fig. 40

Reproduced with permission from Ref. [274]. Copyright © 2018, the PCCP Owner Societies

Schematic figure of an electrochemical fuel cell converting CH₄ and O₂ into CH₃OH and H₂O.

6.3 Catalyst Materials for the Production of Value-Added Chemicals at the Cathode in Fuel Cell Reactors

Similar to fuel cells, i.e., the predecessors of fuel cell reactors, cathodes can be categorized as reaction zones for reduction. In most fuel cells, oxygen or air is fed to the cathodes, at which O₂ is converted to O²⁻ (O²⁻-conducting electrolyte membrane fuel cells) or H⁺ is transformed to H₂O (PC-FCs). When considering the utilization of this part as a value-added chemical production site, the types of reactions can theoretically be chosen from any reaction that has a reduction process, reaction conditions compatible with the whole system, and an overall negative Gibbs free energy. However, focusing on chemical-producing reactors, most reported studies of cathode applications have focused on imported protons (H⁺).

6.3.1 Cathode Catalysts for Consuming H⁺ in Fuel Cell Reactors

6.3.1.1 Hydrogenation Reactions for Unsaturated Organics

Schmidt et al. reported that several unsaturated organic alcohols and acids can be fed as oxidants to the cathode with hydrogen as fuel in a PEMFC [106], in which U₀ (mentioned in Sect. 3) is in the range of 0.4–0.6 V, i.e., the electrogenerative hydrogenation process takes place at a positive potential relative to the anodic hydrogen electrode [277]. Additionally, the electrogenerative hydrogenation process can be carried out at lower H₂ pressures and temperatures compared to conventional catalytic hydrogenation with molecular hydrogen.

In representative reactors based on PEMFCs that use hydrogenation reactions for the cogeneration of electricity, humidified hydrogen is fed to the anode, where protons can be produced. In the presence of H₂O, H₃O⁺ can be conducted from the anodic side to the cathodic side through the PEM. Then, the H₃O⁺ react with the unsaturated organics at the cathode. The electrode reactions are as follows.

$${\text{Anode:}}\,\,{\text{H}}_{2} + 2{\text{H}}_{2} {\text{O}} \to 2{\text{H}}_{3} {\text{O}}^{ + } + 2{\text{e}}^{ - } .$$

(86)

$${\text{Cathode:}}\,\,{\text{R}}_{1} - {\text{HC}} = {\text{CH}} - {\text{R}}_{2} + 2{\text{H}}_{3} {\text{O}}^{ + } + 2{\text{e}}^{ - } \to {\text{R}}_{1} - {\text{HC}} - {\text{CH}} - {\text{R}}_{2} + 2{\text{H}}_{2} {\text{O,}}$$

(87)

where R₁ and R₂ are equal to –H, –CH₃, –(CH₂)_n–CH₃, –CH₂–OH, –CHO, –COOH, etc.

Schmidt et al. [106] reported a fuel cell reactor employing Nafion® 117 as the PEM and 20% Pt/C on Vulcan XC-72 as the cathodic catalyst for hydrogenation reactions with several unsaturated organic alcohols (allyl alcohol, propargyl alcohol, 2-butyne-1,4-diol, and 2-butene-1,4-diol) and acids (maleic acid, acrylic acid, crotonic acid, and acetylenedicarboxylic acid); the results are listed in Table 7. The authors explained the adsorption of unsaturated alcohols/acids and hydrogenation of double or triple bonds as follows:

$$n{\text{Pt}} + {\text{R}}_{1} - {\text{HC}} = {\text{CH}} - {\text{R}}_{2} \to {\text{Pt}}_{n} - \left( {{\text{R}}_{1} + {\text{HC}} = {\text{CH}} - {\text{R}}_{2} } \right)_{{{\text{ad}}}}$$

(88)

$${\text{with}}\,\,{\text{R}}_{2} = - {\text{CH}}_{2} - {\text{OH}}\,{\text{or}}\,{ } - {\text{COOH,}}$$

(89)

$$2{\text{Pt}} - {\text{H}} + {\text{Pt}}_{n} - \left( {{\text{R}}_{1} + {\text{HC}} = {\text{CH}} - {\text{R}}_{2} } \right)_{{{\text{ad}}}} \to \left( {n + 2} \right){\text{Pt}} + {\text{R}}_{1} - {\text{HC}} - {\text{CH}} - {\text{R}}_{2} .$$

(90)

Table 7 Cathode catalysts for consuming H⁺ in fuel cell reactors, as reported in the open literature

Similar reactions can be found in the cathodes of PEMFC reactors for the cogeneration of cyclohexylamine (CHA) by the selective reduction of nitrobenzene (NB) [110],

$${\text{C}}_{6} {\text{H}}_{5} {\text{NO}}_{2} + 6{\text{H}}^{ + } + 6{\text{e}}^{ - } \to {\text{C}}_{6} {\text{H}}_{5} {\text{NH}}_{2} + 2{\text{H}}_{2} {\text{O,}}$$

(91)

$${\text{C}}_{6} {\text{H}}_{5} {\text{NH}}_{2} + 6{\text{H}}^{ + } + 6{\text{e}}^{ - } \to {\text{C}}_{6} {\text{H}}_{11} {\text{NH}}_{2} ,$$

(92)

$${\text{C}}_{6} {\text{H}}_{5} {\text{NO}}_{2} + 6{\text{H}}^{ + } + 6{\text{e}}^{ - } \to {\text{C}}_{6} {\text{H}}_{11} {\text{NO}}_{2} ;$$

(93)

and the hydrogenation of allyl alcohol to 1-propanol [115],

$${\text{CH}}_{2} = {\text{CHCH}}_{2} {\text{OH}} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{CH}}_{3} {\text{CH}}_{2} {\text{CH}}_{2} {\text{OH}}{.}$$

(94)

6.3.1.2 Reduction of Nitric Oxide

The main industrial process for the production of hydroxylamine, a commodity chemical mainly used in caprolactam synthesis, is nitric oxide (NO) hydrogenation in 20% sulfuric acid with a Pt catalyst in a catalytic-slurry reactor [278, 279]:

$$2{\text{NO}} + 3{\text{H}}_{2} \to 2{\text{NH}}_{2} {\text{OH,}}$$

(95)

$$2{\text{NH}}_{2} {\text{OH}} + {\text{H}}_{2} {\text{SO}}_{4} \to \left( {{\text{NH}}_{3} {\text{OH}}} \right)_{2} {\text{SO}}_{4} .$$

(96)

Analogous production can also be achieved in a PEMFC reactor with more moderate reaction conditions by employing the anode to realize the classical hydrogen oxidation reaction:

$${\text{H}}_{2} \to 2{\text{H}}^{ + } + 2{\text{e}}^{ - } ,{ }E^{0} = 0\,{\text{V }}\left( {{\text{vs}}.\,{\text{NHE}}} \right).$$

(97)

At the cathode, the electrochemical reduction products of NO in an acidic medium are as follows [114, 280]:

$$2{\text{NO}} + 4{\text{H}}^{ + } + 4{\text{e}}^{ - } \to {\text{N}}_{2} + 2{\text{H}}_{2} {\text{O}},\quad E^{0} = 1.68\,{\text{V}}\,{ }\left( {{\text{vs}}.\,{\text{NHE}}} \right);$$

(98)

$$2{\text{NO}} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{N}}_{2} {\text{O}} + {\text{H}}_{2} {\text{O}},\quad E^{0} = 1.59\,{\text{V}}\,{ }\left( {{\text{vs}}.\,{\text{NHE}}} \right);$$

(99)

$${\text{NO}} + 5{\text{H}}^{ + } + 5{\text{e}}^{ - } \to {\text{NH}}_{3} + 2{\text{H}}_{2} {\text{O}},\quad E^{0} = 0.73\,{\text{V }}\left( {{\text{vs}}.\,{\text{NHE}}} \right);$$

(100)

$${\text{NO}} + 3{\text{H}}^{ + } + 3{\text{e}}^{ - } \to {\text{NH}}_{2} {\text{OH}},\quad E^{0} = 0.38\,{\text{V}}\,{ }\left( {{\text{vs}}.\,{\text{NHE}}} \right);$$

(101)

and

$$2{\text{NH}}_{2} {\text{OH}} + {\text{H}}_{2} {\text{SO}}_{4} \to \left( {{\text{NH}}_{3} {\text{OH}}} \right)_{2} {\text{SO}}_{4} .$$

(102)

The selectivity for a particular product is sensitive to the electrode potential; for example, nitrous oxide (N₂O) is favored at high potentials (E > 0.5 V vs. NHE) while more hydrogenated products predominate at lower potentials [114, 280]. Thus, the selectivity/current efficiency for electrocatalytic reactions to produce hydroxylamine strongly depends on the electrode composition, i.e., the electrocatalyst.

These steps can also be accomplished in a H₂-NO PAFC [127] with the mechanism as shown in Fig. 41.

Fig. 41

Reproduced with permission from Ref. [127]. Copyright © 2016, the Royal Society of Chemistry

Proposed mechanism for the reduction of NO to NH₂OH over isolated Fe sites.

A modified TSR modeled after a H₂-NO fuel cell reactor operated in potentiostatic mode with a cocurrent flow direction was presented by Denayer et al. [107] (Fig. 42). The components, the gas channel energy balances in the liquid catholyte, the catalyst layers, and the charge balances at the electrolyte/electrode interfaces were all accounted for by the developed model. The simulation results revealed that the current density, temperature, and concentration distribution were influenced by the mass transfer and catalyst selectivity.

Fig. 42

Copyright © 2012, Elsevier

Cocurrent flow pattern for the TSR model of a planar H₂-NO fuel cell reactor. Reprinted with permission from Ref. [107].

As introduced in Sect. 5.3.2, Otsuka et al. [126] studied cells operated at 373 K with the configuration [alkene, H₂O(Pd + VGCF) anode|H₃PO₄/silica-woo|׀VGCF-cathode|O₂, NO]. The alkene oxidation rate, i.e., the current density, was increased by more than ten times by the addition of NO into the O₂ stream at the cathode. In this system, NO acted as an electrochemical reduction catalyst for the conversion of O₂ to H₂O, according to the following chemical reaction formulas:

$${\text{NO}} + \frac{1}{2}{\text{O}}_{2} \leftrightarrow {\text{NO}}_{2} ,$$

(103)

$${\text{NO}}_{2} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{NO}} + {\text{H}}_{2} {\text{O,}}$$

(104)

$${\text{O}}_{2} + {\text{NO}} \leftrightarrow \left( {{\text{NO}}_{3} } \right),$$

(105)

$$2{\text{NO}} \leftrightarrow {\text{N}}_{2} {\text{O}}_{2} ,$$

(106)

$${\text{NO}}_{2} + {\text{NO }} \leftrightarrow {\text{N}}_{2} {\text{O}}_{3} ,$$

(107)

$$2{\text{NO}}_{2} \leftrightarrow {\text{N}}_{2} {\text{O}}_{4} .$$

(108)

6.3.1.3 Hydrogen Peroxide Production

Hydrogen peroxide (H₂O₂) is widely used in the pulp and paper industry, medicine, water treatment, and detergents. The industrial chemical production method is based on the cyclic reduction of oxygen with hydrogen and uses anthraquinone as the catalyst [281, 282]; an alternative method has been designed based on the electrochemical reduction of oxygen (in the air) at carbonaceous cathodes through a two-electron reaction route [283, 284]. However, neither of the above strategies is safe, clean, and energy economical. Therefore, schemes to cogenerate H₂O₂ in the cathode of a PEMFC have been investigated.

Agladze et al. [108] accomplished this reaction in a PEMFC reactor with methanol as the fuel. At the cathode, oxygen was reduced to hydrogen peroxide according to the following formula when producing electrical energy:

$${\text{O}}_{2} + {\text{H}}^{ + } + 2{\text{e}}^{ - } \to {\text{H}}_{2} {\text{O}}_{2} .$$

(109)

One possible extension of this reaction is to utilize the in situ produced H₂O₂ as an oxidizing agent. Yang et al. [109] reported this kind of PEMFC reactor for the cogeneration of electrical energy and phenol, which is a valuable chemical for petrochemicals and is industrially produced via the so-called “cumene process” [285]. The authors used H₂ as the fuel and source of protons, which participated as the reactant for the generation of H₂O₂ and the subsequent phenol, as shown in Fig. 43.

Fig. 43

Copyright © 2005, Elsevier

Reaction scheme for phenol synthesis using in situ-generated H₂O₂ in a PEMFC reactor. Reprinted with permission from Ref. [109].

6.3.1.4 Oxidation with Byproducts at the Cathode

The partial oxidation of methane can be performed at the cathode in a fuel cell reactor. Hinino et al. [217, 218] (Fig. 44) reported a one-pass process for the production of methanol from methane at an intermediate temperature and low pressure in a micro- or nanosized electrocatalyst system. The oxidant for the methane oxidation step was O^*, which was a byproduct of the ORR at the cathode.

$${\text{Anode:}}\,\,{\text{H}}_{2} \to 2{\text{H}}^{ + } + 2{\text{e}}^{ - } ,$$

(110)

$${\text{or}}\,\,{\text{H}}_{2} {\text{O}} \to \frac{1}{2}{\text{O}}_{2} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } .$$

(111)

$${\text{Cathode:}}\,\,{\text{O}}_{2} + 2{\text{H}}^{ + } + 2{\text{e}}^{ - } \to \left( {{\text{H}}_{2} {\text{O}}_{2} } \right) \to {\text{O}}^{*} + {\text{H}}_{2} {\text{O,}}$$

(112)

$${\text{CH}}_{4} + {\text{O}}^{*} \to {\text{CH}}_{3} {\text{OH}}{.}$$

(113)

Fig. 44

Copyright © 2010, Elsevier

Schematic illustrations of a macrosized and b micro- or nanosized electrochemical cells. Reprinted with permission from Ref. [217].

The mechanisms for Reactions (112) and (113) on Pd-Au-Cu/C catalysts can be explained according to the Fenton mechanism [217, 218]:

$${\text{H}}_{2} {\text{O}}_{2} + {\text{H}}^{ + } + {\text{Cu}}^{ + } \to {\text{HO}}^{*} + {\text{H}}_{2} {\text{O}} + {\text{Cu}}^{2 + } ,$$

(114)

$${\text{R}} - {\text{H}} + {\text{HO}}^{*} + {\text{Cu}}^{2 + } \to {\text{R}} - {\text{OH}} + {\text{H}}^{ + } + {\text{Cu}}^{ + } .$$

(115)

Analogously, under these circumstances, Fenton chemistry predicts the hydroxylation of benzene to phenol [220]:

$${\text{H}}_{2} {\text{O}}_{2} + {\text{H}}^{ + } + {\text{Fe}}^{2 + } \to {\text{HO}}^{*} + {\text{H}}_{2} {\text{O}} + {\text{Fe}}^{3 + } ,$$

(116)

$${\text{PhH}} + {\text{HO}}^{*} + {\text{Fe}}^{3 + } \to {\text{PhOH}} + {\text{H}}^{ + } + {\text{Fe}}^{2 + } .$$

(117)

Furthermore, the oxidation of cyclohexane (CyH) to CyOH or CyO over SmCl₃/graphite cathodes can be expected [225, 226].

6.3.2 Cathode Catalysts for Other Reactions in Fuel Cell Reactors

As discussed in Sects. 5.3.1 and 6.1.3, cathodic reactions in cogeneration reactors can be applied to reduce ions from higher to lower valences [112, 124]. Because H⁺ imported from the anode is not consumed and O²⁻ is not produced at the anode, different analytes and catholytes are employed to participate in the reactions on the electrodes, and a mixed conducting electrolyte membrane closes the circle.

As classified above, in most fuel cell reactors that simultaneously produce electrical energy, value-added chemicals can be harvested at the anode or cathode with the import or export of H⁺ or O²⁻. The electrolyte membrane can play an important role in both mass transfer and closing the loop. However, there are also some exceptions, in which no mass transfer occurs through the electrolyte membrane. Some of the typical cogeneration reactions are summarized in Fig. 45.

Fig. 45

Summary of the typical cogeneration reactions in Sect. 6: a value-added chemicals produced at the anode and b value-added chemicals produced at the cathode

7 Conclusions and Perspectives

This review presents several typical works focusing on fuel cell reactors in which value-added chemicals and electrical energy are simultaneously produced, killing two birds with only one stone. In this kind of reactor, thermodynamically favorable reactions represent the “stone”: at least one of the electrode reactions is employed for chemical synthesis, and most of the free energy difference between the reactant and product is converted into the form of clean energy. Cogeneration reactions can be classified according to their baseline (i.e., the type of fuel cell used to approach cogeneration), the type of conducting electrolyte membrane, or the electrode reactions adopted for chemical production. However, from the perspective of value-added chemical production, most “synthesis reactions” involve O²⁻ consumption (or H⁺ production) at the anode or H⁺ use at the cathode. Due to the poor controllability of the oxidation degree, lower selectivity can be obtained during selective oxidation with O²⁻ due to deep oxidation also occurring. In addition to selectivity, conversion and yield need to be investigated in regard to various chemical synthesis reactors, while the energy density and power density are also integral for the other functions of a fuel cell reactor.

Based on the principles above, several typical cogeneration reactions and corresponding key assembly units (e.g., electrolyte membranes and catalysts) have been demonstrated and categorized. However, for fuel cell reactors with higher selectivity for the production of a unique chemical and better energy conversion efficiency, more effort should be put into the following directions.

• 1. The material optimization of key components is still imperative for improving product selectivity, energy conversion efficiency, and system stability. Fuel cells can be a fundamental reference, but unlike fuel cells, the target products of the catalysts in cogeneration reactors are value-added chemicals rather than complete oxidation products, e.g., CO₂ and H₂O; thus, they should be the focus of the electrocatalysis reactions. Additionally, the elucidation of mechanisms would be helpful for the performance optimization of next-generation materials.

• 2. Product detection remains difficult, especially when gaseous products and liquid products coexist with supporting electrolytic ions. In low-temperature (room temperature) systems, the gaseous products and/or liquid products can dissolve in the electrolyte and/or solvent, which then needs to be mostly neutralized and/or separated before being injected into an IC, HPLC, or other instrument or used in any practical application. In most systems, more than one product can be obtained, and target product separation could be important for value-added chemical production. As an essential part of the whole system design, universally applicable and/or in situ product detection strategies still have considerable room for improvement.

• 3. Due to the flexibility of reaction selection and the potential for power production, the application of cogeneration reactions is appealing. In addition to the harvesting of electrical energy and value-added chemical production, more environmental management strategies, such as the biosafe disposal of organic matter or toxic ions in industrial wastewater, could be explored.