1 Introduction

Hydrogen peroxide (H2O2) is a high-value and environmentally friendly oxidizing agent with a wide range of applications in chemical synthesis, paper and pulp, and wastewater [1, 4]. Currently, 95% of H2O2 production predominantly depends on the anthraquinone oxidation (AO) process [5−7]. However, this process involves multistep reactions, including a sequential hydrogenation/oxidation of anthraquinone molecules and separations (extraction of H2O2 from organic solvents), which require enormous energy input and sophisticated facilities [8,9,10]. The hydrogenation process is initially performed over a Pd catalyst followed by rapid O2 oxidation to produce concentrated H2O2. Then, the H2O2 is removed by solvent extraction. Despite being able to yield concentrated H2O2, the large energy input, multiple facilities, and the usage of Pd catalyst increase the cost, and the high concentration of H2O2 has a risk of storage and transportation, further increasing the cost. These drawbacks have triggered researchers to explore alternative green H2O2 synthesis technologies. A popular alternative to the traditional AO process is the direct synthesis using H2 and O2 as the starting reactants [11,12,13,14,15]. This process is performed over noble metal catalysts such as Pd, which can enable continuous and decentralized H2O2 production. However, the most substantial barrier to the development of this process is the safety issues originating from the explosive nature of H2 and O2. Thus, to enable large-scale application, other efficient and economic routes for H2O2 electrosynthesis are highly desirable.

Fortunately, the electrochemical strategy through a two-electron oxygen reduction reaction (2e− ORR) pathway provides an attractive route to produce H2O2, which is portable and safety [16,17,18,19]. Moreover, the electrocatalytic oxygen reduction process just needs water and O2 as the starting material, and it could be coupled with renewable energy sources [20,21,22]. Figure 1 briefly describes the different synthesis methods of H2O2. The electrochemical strategy is more cost-effective and environmentally friendly than the traditional AO and direct process. In the electrochemical process, the H2O2 product can be directly generated by the reduction of O2 at the cathode. The electrochemical H2O2 production via a 2e− ORR process was first reported in the 1930s [7, 23]. Since then, the on-site H2O2 generation from ORR has been widely used for the pulp and paper bleaching process [24,25,26]. Despite these advantages, the sluggish reaction kinetics and the competing reactions limit the overall energy efficiencies. Thus, the pre-requirements of the H2O2 electrochemical production process are the rational design of specialized catalysts with high activity, high selectivity, and good stability. The emerging electrocatalyst is divided into noble metal catalysts, transition metal-based catalysts, and carbon-based catalysts. A timeline illustrating the important finding is shown in Fig. 2. H2O2 is electrochemically produced on the electrode either using a classic H-cell or flow cell with a gas diffusion layer. Therefore, beyond the catalyst-level designs, the electrode and the reaction reactor with the capacity for rapid mass transfer and reactants/products circulating can further enable an efficient H2O2 production rate.

Fig. 1
figure 1

Schematic of H2O2 production via traditional anthraquinone process, the direct synthesis, and electrochemical synthesis with related demerits and merits. Reprinted with permission [2]. Copyright 2021, Elsevier. (Color figure online)

Full size image
Fig. 2
figure 2

Timeline of some significant findings of 2e– ORR electrocatalyst. PtHg4: Reproduced with permission [11]. Copyright 2013, Springer Nature. Pd2Hg5: Reproduced with permission [61]. Copyright 2014, American Chemical Society. HPC: Reproduced with permission [131]. Copyright 2015, Wiley VCH. Pt-S4: Reproduced with permission [79]. Copyright 2016, Springer Nature. Co Complex: Reproduced with permission [93]. Copyright 2017, American Chemical Society. g-N-CNHs and O-CNTs: Reproduced with permission [138, 146]. Copyright 2018, Elsevier Inc. and Copyright 2018, Springer Nature. Pt-CuSx and CoS2: Reproduced with permission [77, 98]. Copyright 2019, Elsevier Inc. and Copyright 2019, American Chemical Society. PtP2 and CoN4: Reproduced with permission [64, 120]. Copyright 2020, Springer Nature. CoNOC and GOMC: Reproduced with permission [121, 132]. Copyright 2021, Springer Nature and Copyright 2021, Elsevier Inc. Ni2-xP-VNi and OCG: Reproduced with permission [100, 134]. Copyright 2022, Wiley VCH. and Copyright 2022, Royal Society of Chemistry. Black: noble metal-based catalysts. Blue: carbon-based catalysts. Red: transition metal-based catalysts. (Color figure online)

Full size image

The research on electrochemical production of H2O2 via a selective 2e− ORR process is an emerging field. Recent intensive studies have led to the development of various promising catalysts, electrodes, and reaction reactors [20, 27]. Although extensive review articles regarding tailoring the 4e− ORR pathway to the 2e− ORR pathway for H2O2 production have been reported, the attention paid to the complementary of catalysts with other cell components into an efficient device for H2O2 production is rare. In this review, the recent advances in the disclosed catalyst design and the improvements made to the electrode and cell design that enables unprecedented H2O2 electrochemical production are discussed. In the last part, a perspective on some major challenges and opportunities for the rational design of high-efficient catalysts, electrode engineering, and the reaction reactor is presented to accelerate the development of H2O2 electrochemical production in future studies.

2 Fundamentals of H2O2 Production from Selective Oxygen Reduction Reaction

2.1 Reaction Mechanism of Two-Electron Oxygen Reduction Reaction

In general, the ORR is a multistep process that can proceed with the dissociative mechanism and the associative mechanism [28]. The dissociative mechanism refers to the breakage of O–O bond to form O upon oxygen adsorption, which is reduced successively to OHads and H2Oads. The associative mechanism means that the O–O bond is maintained and the final product can be H2O or H2O2 depending on the ability of the catalyst to dissociate the O–O bond in the *OOH intermediate [28,29,30]. For the ORR to produce H2O2, it is generally believed that the O–O bond cleavage is unfavorable, and the associative mechanism is dominant. As displayed in Fig. 3a, the O2 molecule was firstly adsorbed onto the active sites, followed by a proton-coupled electron transfer process to form *OOH [29, 31]. Subsequently, *OOH would be reduced to the final product of H2O2, which is desirable for the electrosynthesis of H2O2 [32]. However, provided that the cleavage of O–O bond occurs, the following reduction of *O would be proceeded, resulting in the formation of undesirable H2O [33, 34]. Therefore, protecting the O–O bond and modulating the adsorption energy of *OOH is crucial for achieving high H2O2 production.

Fig. 3
figure 3

a Schematic for the proposed oxygen reduction reaction process. Reproduced with permission [28]. Copyright 2019, American Chemical Society. b Schematic illustration of the double-layer structure during ORR in acidic (left) and alkaline (right) conditions. Insets illustrate the inner- and outer-sphere electron transfer processes. Reprinted with permission [29]. Copyright 2011, American Chemical Society. (Color figure online)

Full size image

In addition, previous studies have shown that different electrolytes can lead to distinct H2O2 electrosynthesis performance because solvation effects, surface-adsorbed species (such as OHads and electrolyte anions), and other factors can affect their ORR pathway. Researchers have studied the ORR performance of Pt/C catalyst in varied pH electrolyte. It was found that in the alkaline medium, the hydroxyl species was strongly adsorbed on the catalyst surface, which is a source of protons and transfers electrons to the water-solvated molecular O2 (O2·(H2O)n). The mechanism of the outer-sphere electron transfer for the ORR is dominant in the alkaline medium. Solvated O2 and anions filled the outer Helmholtz plane (OHP) (shown in Fig. 3b). O2 chemisorption is not a prerequisite, and this process is not specific and can proceed on all catalysts. Therefore, all catalysts can be used for H2O2 production in alkaline electrolytes. Especially for metal compounds and other catalysts that is semiconductor, this mechanism gives the reasons that can be used for H2O2 electrosynthesis. The first step of this mechanism is the O2 molecules are solvated (O2·(H2O)n), followed by proton and electron transfer to form *OOH intermediates. The selectivity to H2O2 or H2O depends on the cleavage capability of O–O bonds in the *OOH intermediate. For the metal compounds, the binding energy of *OOH on the catalyst surface is weak, thus promoting the desorption of *OOH to form H2O2. It can be illustrated as:

$${\text{M}}-{\text{OH }} + {\text{ }}\left[ {{\text{O}}_{{\text{2}}} \cdot\left( {{\text{H}}_{{\text{2}}} {\text{O}}} \right)n} \right]_{{{\text{aq}}}} + {\text{ e}}^{ - } \to {\text{M}}-{\text{OH }} + {\text{ }}(^{*} {\text{OOH}})_{{{\text{ads}}}} + {\text{ OH}}^{ - } + {\text{ }}\left( {{\text{H}}_{{\text{2}}} {\text{O}}} \right)_{{n - {\text{1}}}}$$
(1)
$$\left( {^{*} {\text{OOH}}} \right)_{{{\text{ads}}}} + {\text{ e}}^{ - } \to ({\text{HO}}_{{\text{2}}} ^{ - } )_{{{\text{ads}}}}$$
(2)

In acidic electrolytes, the high mobility of protons leads to low concentration of adsorbed hydroxyl species on the catalyst surface. O2 molecules are chemisorbed on the catalyst surface, then getting electrons from the electrode. This mechanism is called the inner-Helmholtz plane (IHP) process (shown in Fig. 3b). Therefore, the catalysts used for acidic H2O2 production are conductors, which can facilitate O2 chemisorption and electrons transfer. This mechanism elaborates that noble metal-based catalysts and the emerging M–N–C catalysts with high ORR activity can be used for H2O2 production in acidic electrolyte. The elemental steps for acidic H2O2 production are:

$${\text{O}}_{{2}} \to {\text{O}}_{{2}} ,{\text{ ads}}$$
(3)
$${\text{O}}_{{{\text{2}},{\text{ ads}}}} + {\text{ e}}^{ - } + {\text{ H}}^{ + } \to \;{}^{*}{\text{OOH}}$$
(4)
$$^{*} {\text{OOH }} + {\text{ e}}^{ - } + ~{\text{H}}^{ + } \to \;^{*} {\text{OOH}}$$
(5)

These ORR pathways give an insight into the pH effect on the H2O2 electrosynthesis performance and guide us to design catalysts. However, there is rare study about the H2O2 production mechanism in neutral electrolyte. In the neutral electrolyte, the hydroxyl species is barren as well as the protons. Thus, we should further investigate the ORR pathway to reveal the underlying pH effect on the H2O2 production.

2.2 Performance Evaluation: Rotating-Ring Disk Electrode Versus Practical Devices

The current H2O2 electrosynthesis performance evaluation mainly relies on the rotating ring-disk electrode (RRDE) technique in a three-electrode system. This technique is an effective yet facile electrochemical method for quantifying ORR activity, electron transfer number (n), and H2O2 selectivity (%) in a laboratory setting. Typically, a glassy carbon disk with a Pt ring electrode is used as the working electrode. The ink dripped on the disk was used for H2O2 production, while the ring is set with 1.2 V (vs. RHE) to quantify the H2O2 amount [35,36,37]. The negative current at the disk is used to evaluate the electrocatalyst activity. The ORR current on the ring is negligible as the ORR occurs on the disk and the H2O2 is oxidized on the ring electrode. Therefore, the positive current on the ring is in an index of the H2O2 selectivity. The electron transfer number (n) and H2O2 selectivity (%) can be determined by measuring the current at the disk and the ring as the following equations:

$$n = \frac{{4\left| {I_{{{\text{disk}}}} } \right|}}{{\left| {I_{{{\text{disk}}}} } \right| + I_{{{\text{ring}}}} /N}}$$
(6)
$${\text{H}}_{2} {\text{O}}_{2} \% = \frac{{200 \times I_{{{\text{ring}}}} /N}}{{\left| {I_{{{\text{disk}}}} } \right| + I_{{{\text{ring}}}} /N}}$$
(7)

where Iring and Idisk refer to ring current and disk current, respectively. N represents the collection efficiency of the ring electrode, which is generally obtained via the oxidation–reduction reaction of [Fe(CN)6]4−/[Fe(CN)6]3−. In the ORR, the value of n closer to 2 indicates the pathway is toward the 2e− ORR pathway to produce H2O2, while the value of n closer to 4 indicates the 4e− ORR pathway to produce H2O. The higher H2O2 selectivity means that the 2e− ORR pathway is dominant.

The practical H2O2 production performance of the catalyst can also be evaluated by utilizing electrolyzer configurations. The H2O2 electrosynthesis performance under such devices can represent the “real world” relative to the RRDE technique and can be used to test catalyst stability and bulk production over an extended time. The H2O2 concentration can be determined by a traditional Ce(SO4)2 titration method. The yellow Ce4+ ion has a strong absorption peak around 316 nm and can be reduced by H2O2 to colorless Ce3+. Thus, the concentration of H2O2 will be obtained by the concentration of Ce4+ before and after the reaction. Moreover, the faradaic efficiency (FE) in these real devices is calculated to evaluate the catalyst performance (H2O2 production rate, accumulated H2O2 concentration, and energy efficiency).

$${\text{2Ce}}^{{{4} + }} + {\text{ H}}_{{2}} {\text{O}}_{{2}} \to {\text{2Ce}}^{{{3} + }} + {\text{ 2H}}^{ + } + {\text{ O}}_{{2}}$$
(8)
$$C_{{{\text{H}}_{2} {\text{O}}_{2} }} { }\left( {{\text{mM}}} \right) = \frac{{V_{{{\text{Ce}}^{4 + } }} \times C_{{{\text{before}}}} {\text{Ce}}^{4 + } - (V_{{{\text{Ce}}^{4 + } }} + V_{{1,{\text{electrolyte}}}} ) \times C_{{{\text{after}}}} {\text{Ce}}^{4 + } }}{{2 \times { }V_{{2,{\text{electrolyte}}}} }}$$
(9)
$${\text{FE}}\% \, = \frac{2 \times C \times V \times F}{Q}$$
(10)

where CH2O2 is the actual produced H2O2 concentration, VCe4+ refers to the volume of Ce(SO4)2, V1, electrolyte is the volume of removed electrolyte from the electrolyzer. CbeforeCe4+ is the initial concentration of Ce(SO4)2, CafterCe4+ is the final concentration of Ce(SO4)2 after H2O2 is added, V2, electrolyte refers to the total electrolyte volume. In Eq. 10, C is the H2O2 concentration, V is the volume of the electrolyte, F is the faraday constant (96,485 C mol−1), and Q is the consumed charge. The value of FE is more meaningful to evaluate the H2O2 electrosynthesis performance compared to the H2O2% obtained by the RRDE technique.

There could be a distinct gap in the performance evaluation between the RRDE and electrolyzer set-ups because the RRDE technique tends to overperform while the catalyst performance in practical devices shows a more truthful H2O2 selectivity (faradaic efficiency). A recent work by Yang [7] gave a comparison of the FE of kinds of catalysts on RRDE (Fig. 4a), gas diffusion electrode (submerged, air-breathing) (Fig. 4b), and membrane electrode assembly (MEA) (Fig. 4c). An obvious difference in the selectivity between the RRDE and the practical devices can be clearly seen in Fig. 4d. The main reason behind this phenomenon may be attributed to mass transfer (including oxygen bubbles to the catalyst surface, ions transport, and the H2O2 transport away from the catalyst surface) and electron transfer (electron transfer from the substrate to the catalyst). For the RRDE technique, the produced H2O2 at the disk is instantly oxidized at the ring due to the electrode rotation, which can accelerate the H2O2 transfer and reduce the residence time of H2O2 on the catalyst surface. Moreover, the rotation can promote ion diffusion and reduce concentration polarization. Nevertheless, the solubility of O2 in the electrolyte is very low (70 mg O2 L−1). The insufficient O2 mass transfer restricts the ORR performance. In addition, in the case of the device, the accumulated H2O2 on the catalyst surface can accelerate the H2O2 corrosion process and the catalyst can easily deviate from the substrate, resulting in a large deviation in H2O2 performance evaluation. In this perspective, research efforts under conditions that are more representative of the “real world” are necessary to minimize the gap between fundamental research and practical implementation.

Fig. 4
figure 4

a Schematic of a RRDE setup. b GDE submerged in 0.1 M HClO4. c MEA using a three-electrode system. d Faradaic efficiency in various laboratory-scale electrochemical cells for H2O2 electrosynthesis. Reprinted with permission [7]. Copyright 2018, American Chemical Society. (Color figure online)

Full size image

Currently, catalyst development is mainly focused on tuning the 4e− ORR pathway toward the 2e− ORR pathway to produce H2O2. With the increasing demand for on-site H2O2 production, the facile electrode preparation has encouraged more search for the integrated electrode with high activity and high selectivity. This review will describe the current insights for H2O2 electrosynthesis, in which the rational catalyst design, their application, and reactor design strategies are highlighted.

3 Catalysts for H2O2 from the Oxygen Reduction Reaction

The common catalysts used for H2O2 electrochemical production include noble-metal-based catalysts, transition metal-based catalysts, and carbon-based catalysts. Extensive efforts have been made in the ORR electrocatalysts investigation due to the increasing demand for H2O2. In this section, the developed catalysts will be concluded and discussed in detail.

3.1 Noble Metal-Based Catalysts

Various noble metal-based catalysts for H2O2 production have been studied recently including noble metals and their alloys, single-atom catalysts, and so on. Noble metal-based catalysts have been a topic in fuel cells for their high activity and stability [41] reported the distinct H2O2 production performance of Pd–Se–B in various electrolytes. Pd–Se–B showed a surprising activity in neutral electrolyte. Moreover, it exhibited the highest H2O2 selectivity of 90% in neutral electrolyte than that in 0.5 M H2SO4 (55%) and 0.1 M KOH (55%) (Table 1). The markable changes of cyclic voltammetry (CV) curves in neutral, acidic, and alkaline electrolytes were indicative of the variable H2O2 selectivity. Under repeated hydrogen underpotential deposition, Pd2+ was reduced to metallic Pd, leading to the increasing ORR activity and decreasing H2O2 selectivity. While the Pd–Se–B exhibited negligible changes under the potential cycling in neutral electrolyte. This result is against the traditional mechanisms, indicating that the pH effect is of great significance to the H2O2 electrosynthesis.

Table 1 Electrode materials for H2O2 electrosynthesis at different pH values via different types of electrolytic cells
Full size table

3.2 Transition Metal-Based Catalysts

Compared with the noble metal-based catalysts, transition metal-based catalysts (such as Fe, Co, Ni, and Cu) have aroused interest due to their earth-abundant and tunable electronic structure of the central transition metal atoms [81,3.3.1 Porous Carbon-Based Materials

The most widely utilized structure regulation strategy is well-developed pore structure, pores volume, large surface area, and so on [126,127,128,129]. The structural properties play a substantial role during the electrocatalytic reaction process because they are closely linked to mass transport and the utilization of active sites [130]. For instance, a hierarchically porous carbon (HPC) derived from the metal–organic frameworks (MOFs) was presented for H2O2 electrosynthesis (Fig. 12a) [131]. The obtained HPC exhibited high catalytic activity and selectivity for electrochemical reduction of O2 to H2O2 over a wide pH range (1−7). The HPC catalyst with abundant micro-, meso-, or even macropores endowed it with plentiful exposed catalytic sites and shortened diffusion paths. Micropores could expose more active sites and provide additional active sites for ORR. The meso- and macropores allowed for fast transport of H2O2 from the catalyst surface to the bulk solution and reduced residence time, thus avoiding H2O2 further reduction to H2O. Especially for the carbon materials with a layered structure, most of the catalytic sites are buried in carbon layers in which interlayer spacing is too small to expose the catalytic sites. Using a template (silica, MgCl2, mesoporous aluminosilicate, and son) can create micropores or mesopores after template removal and thus expose more active sites [132]. Joo et al. [133] demonstrated that graphitic ordered mesoporous carbon (GOMC) nanocatalyst could expose abundant edge sites via mesoporous silica template introduction, rendering it with 28 times higher mass activity than that of a basal plane-rich CNT. Lee et al. [134] also demonstrated that 3D crumpled graphene showed an increased active surface area and could expose most of the buried active sites within 2D carbon layers through the MgCl2 introduction.

Fig. 12
figure 12

a Schematic illustration of HPC materials and their H2O2 production rates. Reproduced with permission [131]. Copyright 2015, Wiley. b ORR performance of N-FLG-8, N-FLG-12, and N-FLG-16. c Schematic diagram of two-electron and four-electron ORR pathways on N-FLG with different nitrogen configurations. d Relationship between H2O2 selectivity and atomic content of pyrrolic-N. Reproduced with permission [141]. Copyright 2020, Wiley. e TEM images of CNTs and O-CNTs. f ORR performance comparison of CNTs and O-CNTs. g ORR performance comparison of SP, O-SP, AB, and O-AB. Reproduced with permission [146]. Copyright, 2018, Springer Nature. h Illustration of active species in H-GOMC, GOMC, and O-GOMC. Reproduced with permission [132]. Copyright 2021, Elsevier. (Color figure online)

Full size image

3.3.2 Heteroatom Do**

Porosity has been demonstrated to affect the performance of pristine carbon catalysts. The incorporation of heteroatoms (such as N, B, and P) into the carbon framework also is shown to be an effective way to improve the H2O2 electrochemical production performance. Altering the electronic structure of carbon atoms could result in a pronounced enhancement of both activity and selectivity for H2O2 production [135,136,137,138,139,140]. For N-doped electron-rich carbon nanostructures, the carbon π electrons can be activated by conjugating with the lone-pair electrons from N dopants. Thus, O2 molecules get reduced on the positively charged C atoms neighboring N. Qiao’s group [141] developed nitrogen-rich few-layered graphene (N-FLG) with a tunable nitrogen configuration for electrochemical H2O2 generation (Fig. 12b). The experiment results showed that the high nitrogen do** content could effectively alter the electronic structure and facilitate the O2 adsorption. Combined spectroscopic results and electrochemical performance suggested that the OOH* intermediates could be substantially preserved with the presence of a high amount of pyrrolic-N, leading to a 2e− ORR pathway on the adjacent carbon atoms (Fig. 12c), while the 4e− ORR pathway was supposed to preferentially occur on the carbon atoms adjacent to the pyridinic-N. Moreover, a positive correlation between the content of pyrrolic-N and the H2O2 selectivity was experimentally observed (Fig. 12d). Recently, Wang et al. [143]. For example, Chen et al. [139] presented the synergic mechanism of the doped N and F atoms in the process of electrocatalytic H2O2 electroproduction. Nitrogen- and fluoride-co doped carbon nanocages (NF-Cs) showed excellent electrocatalytic performance for H2O2 electroproduction with high FE both in alkaline solution (pH 13) (89.6%) and in acid solution (pH 0.35) (88%). The strong synergistic effect between the doped N and F atoms facilitated the H2O2 electroproduction. The doped N atoms promoted O2 molecule adsorption on the catalyst surface, while the F atoms facilitated the desorption of the *OOH intermediate, thus enhancing the catalytic activity and selectivity for H2O2 production. Moreover, different electronegative elements co-do** strategy was found to have a distinct influence on the catalytic activity and selectivity toward H2O2 production. Hu et al. [123] investigated B and N co-doped sp2 carbon materials. The experimental and theoretical results jointly indicated that when B and N were bonded together, do** hardly influences the electronic structure of carbon atoms and the CNTs material remained inert. The B and N co-do** could turn CNTs into excellent ORR electrocatalysts when the B and N atoms were separated. This phenomenon could be explained by the distinguish electronegativity of B and N atoms. The separation of B from N prevented the neutralization because N is the electron donor while B is the acceptor, so they were still capable of conjugating with the sp2 carbon, as in the single-do**. These results imply that the heteroatomic do** strategy plays a pivotal role in regulating this activity–selectivity dilemma. The essence of heteroatom do** and defects construction is to break the electroneutrality of sp2 carbon to create charged sites favorable for O2 adsorption.

3.3.3 Oxygen Functionalization

In addition to heteroatomic do**, oxygen functionalization strategies prove to be a facile but powerful method to promote electrochemical H2O2 production. The oxygen species can modulate the electronic structure of carbon, thereby controlling the binding energy of OOH* intermediate species for optimal electrochemical H2O2 production. There are many methods to introduce oxygen functional groups to carbon-based materials, such as acid surface oxidation, H2O2 oxidation, alkaline treatment, and oxygen plasma treatment [7]. However, the harsh test conditions make it impractical for a large-scale test. For practical applications, the H-cell setup was designed to evaluate the H2O2 electrosynthesis performance of large-area electrodes (Fig. 13a). Once the H-cell setup occurred, it received considerable attention for it was easy to operate and could generate bulk H2O2 [27, 151,152,153,154]. The electrode was submerged in the liquid electrolyte. Yamanaka et al. [155] studied the H2O2 production using an H-cell in a one-pot batch reactor. In this system, the reported H2O2 production reached a maximum value of 0.289 mmol cm−2 h−1 at the beginning. However, the H2O2 electrosynthesis performance quickly deteriorates, leading to low reaction rates, H2O2 concentrations, and FE. The H2O2 electrosynthesis performance showed a dramatic decay in a one-pot batch reactor compared to that tested in the H-cell with a membrane because the H2O2 in bulk can be decomposed on the anode surface. For the dual-chamber reactor, the cathode chamber and the anode chamber were separated with a proton-exchange membrane (PEM). However, the long distance between the cathode and the anode leads to the ion diffusion path increase accompanied by the increased solution resistance. Moreover, the low solubility of O2 in liquid electrolytes further limits the achievable H2O2 production rate. Although H-cells have been widely used for preliminary catalysts screening, they still cannot accurately evaluate how electrocatalysts and electrodes behave in industrial reactors because H-cells are not continuous and the accumulated H2O2 on the electrode surface leads to the further reduction of H2O2.

Fig. 13
figure 13

a H-cell device. b Schematic of a continuous flow cell with a catalyst deposited on a GDE. Reproduced with permission [27]. Copyright 2020, American Chemical Society. c Electrosynthesis of H2O2 using a solid electrolyte. Reproduced with permission [157,158,159,160]. These GDEs can act as a membrane between the oxygen gas and the liquid electrolyte. The GDEs are fabricated by depositing the catalyst on the gas diffusion layer and the three-phase interfaces (TPIs) accelerate the oxygen diffusion and maintain a constant O2 flow on the catalyst layer (Fig. 13b). [161, 162] The O2 can be electrochemically reduced to H2O2 as soon as it approaches the catalyst layer, promoting a high concentration of H2O2. This can effectively circumvent the problem of the low solubility of oxygen. The integrated electrodes with the advantages of fast electron transfer were constructed to reinforce the gas-diffusion channels against destruction by coating them with a thin gas-diffusion layer [163, 164]. Moreover, the established gas-diffusion layers can mitigate the H2O2 corrosion process towards the catalyst and the substrate, thus enhancing the stability and durability of catalysts. More importantly, recent studies show that PTFE treatment can slow down the H2O2 decomposition due to the decreased dielectric constant originating from the PTFE [108]. The substrates were required to supply gas flow channels to ensure the efficient contact between gas/catalyst and substrate/the catalyst [165,166,167,168]. The substrate originating from carbon showed superior compression strength, gas permeability, and corrosion resistance compared to the metal substrate.

For practical applications, the accumulated high concentrations of H2O2 near the electrode can be easily decomposed and accelerate electrode corrosion and catalyst degradation. Based on the practical experiences in water electrolysis and fuel cells for several years, the essential of promoting electrolyte flow for a long cycle is proposed. Dual-chamber reactor developed into a flow type for low-cost H2O2 electrosynthesis was reported by Chen et al. [169]. The catalyst loaded on commercial carbon support typically consisting of a GDL to enhance the O2 diffusion is used as a cathode. Studies demonstrate that O2 can effectively reach the catalyst surface. The continuous flow of reactants and products can prevent the accumulation of H2O2 near the electrode surface and accelerate the H2O2 inflow to the bulk electrolyte, promoting a higher current density accompanied by a higher reaction rate. Pérez et al. [170] reported that combined with a turbulence promoter, the efficiency in the flow cell is close to 100% with low energy consumption. The study revealed that the increase of the electrolyte inflow velocity could bring a progressive enhancement of the current density and H2O2 production. The flow cells make the H2O2 practical-scale production reliable.

Since the direct electrosynthesis of pure H2O2 solutions was up to 20 wt%, there has seen a revival of interest in H2O2 electrosynthesis. In Wang and colleagues’ work [171], a solid-electrolyte fuel cell was used to produce pure H2O2 solutions with H2 and O2 separately delivered to the anode and cathode, respectively. Protons transferred from the cation exchange membrane (CEM) and HO2− ions transferred from the anion exchange membrane (AEM) were combined in the electrolyte chamber to generate H2O2 (Fig. 13c). In this case, the membrane between the chamber and the electrode avoids flooding in the case of direct contact of electrodes with water. The H2O2 concentrations can be varied by tuning the deionized water flow rate with no impurities introduced. Over 90% faradaic efficiency was achieved by using this device. Diverse “catholyte-free” flow reactors have been designed to obtain highly concentrated H2O2 via membrane electrode assembly, which combines the component of GDE, catalyst, and ion exchange polymer membrane into one unit. Based on the previously reported microfluidic cell, Kenis et al. [172, 173] pioneered a configuration without membranes that separate each side of the cell. Precise control over the cell assembly has been demonstrated to be effective in obtaining high current densities for H2O2 electrochemical production. Recently, **a et al. [174] also manifested a high efficiency of the membrane-free flow cell for H2O2 electrocatalytic generation (Fig. 13d). The maximum FE jumped to 200% because the hydrophilic carbon fiber paper could be used as cathode and anode to simultaneously produce H2O2. The hydrophilic layer can decrease dielectric constant of the aqueous solution, thus increasing the H2O2 decomposition overpotential and slow down the H2O2 decomposition process [108]. When the cell current arrived at 50 mA cm−2, a cell voltage was only approximately 1.7 V, whereas 1.98 V cell voltage was required for conventional cells, sharply reducing the energy consumption. Thus, biomass conversion reaction may be chosen for electrochemical oxidation as an attractive alternative to traditional water oxidation, which can not only produce a valuable product but also reduce the cell voltage to make energy consumption decreased.

5 Summary and Perspectives

In this review, significant advances in the development of electrocatalytic H2O2 production over various catalysts, electrodes, and other cell components are summarized and discussed in Table 1. SACs with lower cost are the most efficient catalysts among the available catalysts for H2O2 electrosynthesis. Much effort has been focused on the design and engineering of the electrode and electrochemical reactors to realize the industrial production of hydrogen peroxide via electrosynthesis. Despite significant advances, it is still in its infancy to realize the scale-up of H2O2 electrochemical production. Therefore, challenges and opportunities are present for the industrial production of H2O2 (Fig. 14).

  1. (1)

    Excavating ORR catalysts with new composition and structure is still at the heart of the H2O2 electrochemical production. To date, noble-based catalysts have been thought to be the most efficient ORR catalysts for H2O2 production. However, their high cost and scarcity restrict their industrial application. Thus, given the cost issue, precise synthesis of trace precious metal-based catalysts or precious metal-free-based high-efficient ORR catalysts is urgently needed. Recently, cost-effective amorphous materials and single-atom catalysts with maximally utilized active sites are emerging. Especially for the single-atom catalysts, they are regarded as the next-generation ORR catalysts with high ORR activity to enable the H2O2 electrochemical production large-scale [72, 84]. In the future, single atoms catalyst with amorphous structure can be expected to be explored. Catalysts must be deposited on the substrate by in situ growing on the substrate, electrodepositing process, or spraying the catalyst on the substrate. As one component of the device, the basic role of the substrate is to load the catalyst and collect the produced current. More importantly, they are required to provide gas flow channels to maximize the electrochemical utilization of gases. Metal substrates, such as SS316L, are easy to be corroded and the corrosion production increases the contact resistance. Therefore, more efforts should be devoted to develo** good catalytic and corrosion-resistant electrodes and studying their effect on the industrial production of H2O2.

  2. (2)

    The currently developed in situ techniques can provide new insight into identifying the active sites and structure evolution, which is of critical significance for designing efficient 2e− ORR catalysts. For example, to gain in-depth into the adsorbed oxygen intermediates on the catalysts during the ORR process, operando ATR-IR measurements could be used for characterizing OOHad and HOOHad to confirm the 2e− ORR pathway [175]. In situ X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) could be carried out to identify the change of coordination configuration and valence states near active sites [176]. In situ Raman spectroscopy could be used to probe the chemical bond stretching and bending vibrations.

  3. (3)

    With the development of computer technologies, thriving theoretical chemistry has been demonstrated as an innovative technology for identifying and develo** new efficient catalysts and predicting the potential reaction mechanism combined with in situ technologies [12]. For example, Hyeon et al. predicted the reaction energetics of H2O2 electrochemical production on M–N4/graphenes (M = Co, Ni, Fe, Pt, Ag, and Ru) and Co–N4/graphene with 4H*/2H* and O*/2O* adsorbed near the cobalt atom where Co–N4(O) possessed the optimal OOH* adsorption energies [120]. It is attractive to utilize theoretical chemistry to seek out new catalysts.

  4. (4)

    Electrochemical production of H2O2 via 2e− ORR using renewable energy is promising from the viewpoint of practical applications for water disinfection and wastewater treatment [177, 178]. The green electricity generated from solar and wind farms can realize a more efficient and cleaner onsite H2O2 production [20, 21]. Much of the literature about H2O2 production has focused on the cathode, while the anode is relatively neglected. Thus, we should combine the future study on H2O2 electrochemical production and seawater, oxidation of organic small molecules, which is more attractive than the conventional H2O2 electrosynthesis, integrating various applications [179,180,181].

Fig. 14
figure 14

Illustration of the challenges and opportunities in the development of the H2O2 electrochemical production. (Color figure online)

Full size image