English

• 1. Introduction

  Ammonia (NH₃) has been used as fertilizer in the past century and contributed to population growth. Besides, NH₃ is recognized as a clean energy transporter for chemical energy storage because of its high gravimetric density of hydrogen (17.8 wt%) [1–4]. As the most widely formed industrial compound in the world, 89.9% of NH₃ is synthesized by the Haber-Bosch method [2,5,6]. The Haber-Bosch process has been revolutionary in agriculture, but it requires extreme temperature and high pressure to break the N≡N triple bond, including pressures of 20–40 MPa and temperatures of 400–600 ℃ even using Fe-based catalysts. Another drawback is that low temperature is required to capture nitrogen (N₂) for the Haber-Bosch process. Besides, hydrogen gas (H₂) produced from fossil fuels has extremely endothermic processes. Keeping in mind, the cost of H₂ owns about 75% of NH₃ production cost [7]. The remaining 25% of expenditure is for separating and synthesizing NH₃. Moreover, an estimated 300 million metric tons of carbon dioxide (CO₂) are produced, seriously aggravating the greenhouse effect [7–10]. Hence, it is urgent to discover a more efficient approach that can operate under more moderate conditions for synthesizing NH₃.

  Over the past decade, several attempts, such as non-thermal plasma catalysis, low-temperature thermal catalysis, photocatalysis, electrochemical catalysis, and enzymatic catalysis, have been made to fix N₂ toward NH₃ under favorable reaction conditions [11]. Among them, electrochemical N₂ reduction reaction (ENRR) using renewable electricity can provide a promising candidate for sustainable NH₃ synthesis [12]. ENRR has several potential advantages. First, NH₃ can be synthesized electrochemically under mild conditions. Second, NH₃ can be synthesized locally and on a small-scale using electrochemical synthesis and renewable energy derived from solar or wind sources. Finally, water is typically employed directly as the H source in the electrochemical synthesis of NH₃ without requiring fossil energy [2]. Despite experiments being reported in 1960, ENRR to NH₃ has recently received major attention, and therefore, the number of studies has dramatically expanded.

  The N₂ reduction process on electrode surface can be summed up as follows: N₂ molecules adsorb on the electrode surface, which is further reduced by the external circuit electrons while the protons are simultaneously supplied to produce NH₃. Significant advancements have been made in ENRR over the past several years under benign conditions. However, the poor NH₃ production rates and low Faradaic efficiency (FE) severely limit its practical applications [13,14]. The main limitations are shown as follows: (1) Due to small overpotential, hydrogen evolution reaction (HER) is the key competing reaction that impedes FE progression [15]; (2) Low electrochemical performance is caused by N₂ molecules being difficult to adsorb and activate due to their high dissociation energy (941 kJ/mol) and low proton affinity [16]; (3) The electrocatalytic performance is significantly affected by the slight solubility and poor mass transfer of N₂ in electrolyte solutions.

  Since electrode materials are essential for ENRR, it has undergone extensive research, including unsupported and supported noble metals, metal nitrides, metal oxides, and non-metals, expecting to improve the excellent electrochemical efficiency. In addition to electrocatalysts, electrolytes, as the significant component for electrolysis, significantly influence the performance and catalytic process. Liquid electrolytes such as aqueous alkaline and acidic electrolytes, ionic liquids (ILs), alkali metal hydroxides or chloride salts, and so on have been adopted in the electrochemical synthesis of NH₃. However, the rational selection of a new system for heterogeneous electrocatalytic reduction of N₂ needs to comprehend the mechanisms of electrode materials and electrolytes, which are still immature. Previously, many reviews have focused on the advances of electrocatalysts for ENRR, while the performance of electrocatalysts in various electrolytes has received comparatively little attention. Therefore, a timely review of the most recent developments in this field is urgently needed.

  Herein, the recent progress on the electrolytes for ENRR toward NH₃ is reviewed (Fig. 1). First, the fundamentals of ENRR are provided, with a brief discussion of reaction pathways, reaction kinetics, and thermodynamics, as well as methods for NH₃ detection. Then, the important recent contributions of electrolytes and mediators in converting N₂ to NH₃ with some representative electrocatalysts in various types of electrolytes are summarized. Finally, the perspective of this technology is provided which also offers thoughts on current issues and opportunities in this focus field.

  Figure 1

  

  Figure 1.  Schematic diagram for the recent progress on the electrolytes for ENRR toward NH₃.

  DownLoad: Full-Size Img PowerPoint

  2. Fundamentals of electrocatalytic nitrogen reduction

  2.1 Reaction pathways

  Typically, the ENRR is carried out in a three-electrode system with cathodic and anodic cells separated by a membrane for proton exchange. N₂ is used as the N source, while the H source is supplied by acid electrolytes or water electrolysis in basic electrolytes [17]. The oxygen evolution reaction (OER) is an anodic reaction, and ENRR happens in the cathodic cell, dependent on pH of electrolytes.

  In acidic electrolytes, electrode reactions are represented as following equations.

  Anodic reaction:

  (1)

  Cathodic reaction:

  (2)

  Under alkaline conditions, N₂ molecules directly combine with water and electrons in the cathode to produce NH₃, unlike in acidic electrolytes. The anodic reaction and cathodic reactions are shown as follows [18].

  Anodic reaction:

  (3)

  Cathodic reaction:

  (4)

  The overall reaction:

  (5)

  Due to its complexity, the reaction mechanisms for the ENRR to NH₃ are not entirely understood. Generally, the associative and dissociative mechanisms are two main categories used to classify the reaction mechanism [19,20]. The specific reaction pathways adopted by N₂ reduction reaction are determined by the configurations in which N₂ molecules are adsorbed. As shown in Fig. 2, N₂ adsorption using an end-on pattern occurs through distal and alternate pathways for the ENRR. Alternatively, the N₂ molecule is adsorbed with a side-on pattern through enzymatic and consecutive pathways for the reduction reaction [21]. Before hydrogenation, the dissociative pathway breaks the triple bond of the adsorbed N₂ molecule, leaving N atoms on the catalyst surface. The left N atom subsequently hydrogenates to form NH₃ [22]. Such dissociative pathway generally occurs in Haber-Bosch process and is difficult to realize in ambient conditions due to its high energy consumption. In an associative alternating pathway, the adsorbed N₂ with two N centers are protonated until the final N–N bond is dissociated. In particular, the first NH₃ is produced with the last N–N bond dissociation, while another NH₃ is formed by proton addition one by one. The first NH₃ is produced through consecutive hydrogenation, which occurs predominantly at the N atoms located far away from the catalyst for associative distal pathway. Then, the generation of the second NH₃ is accomplished by adding protons to the N atom. According to the thermodynamics prospectus, the associative mechanism is the preferred pathway for the ENRR to produce NH₃ [23].

  Figure 2

  

  Figure 2.  Schematic diagram of ENRR to NH3 over catalyst via associative and dissociative pathways. Reproduced with permission [21]. Copyright 2019, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  2.2 Reaction kinetics and thermodynamics

  The hydrogenation of N₂ to produce NH₃ is thermodynamically preferred (△G_f^Ɵ = −16.48 kJ/mol, △H_f^Ɵ = −92.22 kJ/mol), but kinetics hindered. The negative electron affinity (−1.9 eV), high ionization potential (15.83 eV), significant energy gap (10.82 eV), low proton affinity (493.8 kJ/mol), and high dissociation energy (941.0 kJ/mol) between the lowest unoccupied and highest occupied molecular orbitals of the N₂ molecule all work together to prevent electron transfer. Additionally, the partially hydrogenated diazene (N₂H₂) (△H_f^Ɵ = 212.90 kJ/mol) and hydrazine (N₂H₄) (△H_f^Ɵ = 95.34 kJ/mol) molecules exhibit extremely high formation enthalpies. As a result, even in a very acidic environment, N₂ is difficult to activate and cannot immediately protonate [17,24].

  The ENRR is a multi-step reduction involving the transfer of six electrons and six protons. The respective equilibrium potentials for HER and ENRR in acidic and alkaline conditions are shown as follows [25,26].

  (6)

  (7)

  (8)

  (9)

  Equilibrium potentials for Eqs. 6 and 8 that create H₂ are comparable to those for Eqs. 7 and 9 for N₂ reduction. In contrast, during a single half-reaction, the HER process only takes two electrons to make one hydrogen. On the other hand, the average of several stages is used to compute the standard electrode potential (E^Ɵ) for the 6e− reductions of N₂ to NH₃. The thermodynamic parameters for 2e− reductions from N₂ to NH₃ are provided in Eqs. 10–12 in terms of E^Ɵ values. E^Ɵ values are calculated using experimentally determined standard generating Gibbs free energy (△G_f^Ɵ, 298 K). Some E^Ɵ values in the primary source have questionable attribution and different values have been reported. References to △G_f^Ɵ (298.0 K) are used for aqueous conditions whenever available to prevent ambiguity. The △G_f^Ɵ (298.0 K) values used to determine E^Ɵ, unless otherwise stated, were obtained from the National Bureau of Standardization database [27].

  (10)

  (11)

  (12)

  Poor FEs for NH₃ production are consistent with the highly endoergic process in Eq. 10 [19]. The thermodynamic restrictions imposed for hypothetical one-electron reduction intermediates can be used to understand the N₂ reduction cycle. E^Ɵ establishes oneelectron reduction by the gas phase electron affinity for N₂ (Eq. 13) [28].

  (13)

  Hence, HER is more energetically favorable than the ENRR to NH₃, as demonstrated by Eqs. 6–9. Significant barriers exist to forming potential intermediate species linked with the first reduction processes, that must be overcome through electrolytes and catalyst design.

  The thermodynamic technique has also demonstrated an energy link, known as the scale relationship, between the intermediates in catalytic reactions involving several intermediates [29]. As the binding energies of N₂H and NH₂ are not independent, ENRR requires a limiting overvoltage of 0.4 V [11]. Therefore, it is possible to find alternative reaction routes that can link the intermediates in various ways [29,30].

  Besides, from a kinetic perspective, the ENRR may undergo either proton-coupled electron transfer (PCET) process or sequential proton-electron transfer (SPET) process. Yet, compared to SPET, the PCET process is thermodynamically preferable. Besides, due to the involving of proton and electron during the PCET processes, the HER process would be more competitive as it only requires two protons and two electrons [31,32]. Additionally, it has been theoretically established that H atoms adhere to electrocatalyst surfaces far more readily than N₂ molecules, resulting in H atoms occupying most of the active sites [32]. Hence, the HER displays substantially faster kinetics and is more favorable compared to the ENRR process, acting as the bottleneck problem for ENRR [14]. Additionally, the NH₃ yield rate is constrained by the difficulty of N₂ activation due to the thermodynamic stability and chemical inertness of the N₂ molecules [26,33].

  Generally, the PCET process, the desorption of products, and the transport and adsorption of reactants including proton donors or N₂ molecules all occur concurrently at the electrode/electrolyte interface where both the HER and ENRR take place [25,34]. The components of the electrolytes and the local microenvironment close to the electrode/electrolyte interface may significantly influence the mass diffusion, the protons and electron availability for the two processes, which have a significant impact on the ENRR and HER performances [35]. Hence, electrolyte engineering and microenvironment regulation at the electrode/electrolyte interface with a focus on adjusting the diffusion, availability, and local concentration of proton donors are crucial to the trade-off between ENRR and HER and, in the end, facilitates the ENRR process [26,36].

  2.3 Methods for ammonia detection

  The key parament to evaluate the electrocatalyst performance is based on the quantity of NH₃ produced. Ion-selective electrode detection, isotope detection, ion chromatography, and spectrophotometry are primary techniques for NH₃ detection [26]. The most common spectrophotometric techniques are known as the indophenol blue method and Nessler reagent method. These two methods are generally adopted for NH₃ detection because of their high sensitivity levels and low cost. Indophenol blue method has higher color rendering, low detection limits (10× 10⁻⁶ mol/L), and excellent reproducibility. In contrast, the Nessler reagent method is better in reducing reagent consumption and saving time, but it is a mercury-containing color reagent that failed to protect the environment. However, specific metal ions that have been dissolved in the solution can lead to inaccurate measurements. To get rid of interference, it has been proposed that a controlled experiment should be carried out [37]. The ion-selective electrode is straightforward, rapid, and has a broad dynamic range for detection (0.01– 17, 000 ppm). During the experiment, you must maintain control of both the temperature and the stirring speed. Ion chromatography is notable for its high repeatability, precision, and low detection limitation of 3 × 10⁻⁷ mol/L. Because of the short retention time, electrolytes containing Li⁺ and Na⁺ are inappropriate, and organic electrolytes are also irrelevant [26]. Furthermore, the NH₃ source for the catalyst responsible for the ENRR can only be identified by experiments involving the ¹⁵N₂ isotope. To determine the number of products, employing at least two distinct detection techniques is recommended since this can yield more convincing results [38]. Moreover, Nafion membrane is frequently used for ENRR to separate NH₃ from anode products. It has been found Nafion membrane can allow NH₃ to travel through, be adsorbed, and even interact with the membrane [39]. This results in a measurement error and speeds up the membrane deterioration process, both of which are detrimental to NH₃ quantification and long-term stability. In this case, using a salt bridge (saturated KCl) to replace Nafion membrane is demonstrated to be a feasible method. Besides, further research on membrane engineering is needed to solve this problem, which is essential to NH₃ quantification and electrocatalyst evaluation.

  3. Characteristics of electrolytes

  Recent advances in ENRR toward NH₃ have depended significantly on the theoretical and empirical development of novel types of electrolytes along with compatible electrocatalysts. In the recent past, several efforts have been made to increase the NH₃ production rate and FE using different electrolytes under ambient conditions. Choosing an appropriate electrolyte is crucial in boosting the electrocatalysts performance for N₂ reduction to NH₃. The supporting electrolytes break down into charged ions during ionization, which scatters evenly throughout the solvent. When an electrochemical system is active, charged ions flow predictably, forming a conductive route with the external electron transfer circuit. The conductivity of the electrolytes is crucial for achieving charge equilibrium rapidly and reducing the resistance to charge transfer. N₂ has very poor solubility in aqueous electrolytes (0.66 mmol/L at room temperature), resulting in insufficient reactants [37]. The solubility of N₂ has a close relationship with the proton concentration in the electrolytes. Using a fixed saturated N₂ concentration and different HCl concentrations, higher proton concentrations may reduce the N₂ concentration near the catalyst surface [40]. Apart from the proton concentration, the concentrations of salt in the electrolyte have a notable impact on N₂ solubility. At concentrations higher than 0.6 mol/L LiClO₄, a decrease in N₂ solubility and diffusivity is observed, as well as an increase in SEI conductivity as measured [41]. Therefore, there is animmediate need for an electrolyte with a high gas solubility to circumvent this limitation.

  Moreover, the electrolyte serves as the source of the protons involved in the electrocatalytic reduction reaction. According to density functional theory (DFT) calculations and the findings, H atoms predominate the active site both thermodynamically and kinetically [42,43]. In this scenario, the competitive HER is mostly produced by the overstock of protons (Fig. 3a). Consequently, it is of considerable importance to manage the concentration of the protons or raise the proton transfer barrier to the surface of catalyst by adjusting the electrolyte (Fig. 3b). Overall, when improving the electrolysis performance based on the crucial roles of electrolytes, several electrolyte characteristics, such as gas solubility, ion conductivity, pH, and proton transfer rate, affect electrocatalyst activity and selectivity, must be taken into consideration.

  Figure 3

  

  Figure 3.  (a) Schematic illustration of the competitive adsorption and surface coverage on the catalyst during the ENRR and HER process (blue, N₂; purple, proton; gray, catalyst). (b) Illustration of the limiting proton transfer rate by decreasing bulk solution proton concentration and increasing the surface proton transfer barrier. Reproduced with permission [36]. Copyright 2017, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  4. Aqueous electrolytes for electrochemical nitrogen reduction

  Aqueous electrolytes feature low cost, environmental friendliness, excellent conductivity, and a large electrochemical working window, making it possible to carry out electrochemical reduction processes within a wide potential range [44]. Besides, it is simple to collect the necessary protons to create important intermediates such as NNH* in aqueous electrolytes (* denotes an active site on the surface of the electrocatalyst) [45]. Benefiting from these advantages, aqueous electrolytes have been extensively investigated in ENRR. Aqueous electrolytes are generally divided into alkaline, neutral, acidic, and water-in-salt (WiS) electrolytes. Moreover, Table 1 compares the electrochemical performance of catalysts in various types of aqueous and non-aqueous electrolytes [46–70]. The following will briefly introduce some representative works linked to these.

  Table 1

  

  Table 1.  Comparison of ENRR performances in various electrolytes.

  DownLoad: CSV

  4.1 Alkaline electrolytes

  4.1.1   Electrocatalytic performances in alkaline electrolytes

  Alkaline electrolytes are widely used in ENRR due to their capability to suppress HER and promote selectivity toward NH₃ formation. For instance, the evaluation of ENRR in 0.1 mol/L KOH and 0.1 mol/L HCl over CF-ZIF-1100–1 h catalyst shows that the NH₃ yield is higher in a basic electrolyte than that of an acidic medium [71]. Besides, the NH₃ production rate in an acidic medium drops with temperature, but it increases with the rise of temperature in 0.1 mol/L KOH. These findings indicate that higher proton concentration in acidic electrolytes significantly enhanced the HER, thus suppressing the ENRR. Chen et al. [72] investigated the effect of pH of electrolytes on the ENRR to produce NH₃ over a 30% Fe₂O₃—CNT electrocatalyst at a constant potential. The rate of NH₃ production is observed to be 1.06 × 10⁻¹¹ mol cm⁻² s⁻¹ in 0.5 mol/L KOH, which is higher than the rate of NH₃ production achieved in 0.25 mol/L KHSO₄ electrolyte (Fig. 4). It is believed that the transfer of H₂O is the determining factor of the production rate of NH₃ in alkaline ENRR. Herein, the ENRR requires a careful pH adjustment of the electrolyte environment to achieve desired results.

  Figure 4

  

  Figure 4.  (a) Current density profiles in different electrolytes. (b) Average NH₃ formation rates and FEs in different electrolytes. Reproduced with permission [47]. Copyright 2017, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  4.1.2   Electrocatalysts for alkaline electrochemical nitrogen reduction

  To enhance the alkaline ENRR performance, Guo et al. developed a ZIF-67-derived CoP hollow nanocage (CoP HNC) as an electrocatalyst with 1.0 mol/L KOH solution (pH 14, Fig. 5) as the electrolyte [73]. The CoP HNC catalyst with a structure of 3D hierarchical of nanoparticles, nanosheets, and nanocage provides a surface with rich active sites. It exhibited a high FE of 7.36% at 0 V vs. RHE, and the rate of NH₃ yield improved exponentially without hydrazine. It reached 10.78 µg mg_cat.⁻¹ h⁻¹ at −0.4 V vs. RHE. Ma et al. [74] attained a high NH₃ yield of 45.7 g mg_cat.⁻¹ h⁻¹ in 0.1 mol/L KOH (pH 13) with 3D unalloyed nanoporous Au as the extremely efficient and durable electrocatalyst. It was believed that the numerous surface steps and bends on the Au ligament surface of NPG were the N₂ reduction's active sites, making it an efficient catalyst for ENRR with a FE as high as 3.41%.

  Figure 5

  

  Figure 5.  (a) Schematic procedure, (b) HIM image, (c) TEM image, and (d) HRTEM image of CoP HNC. Reproduced with permission [73]. Copyright 2018, Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  Bao et al. used the seeded growth method to produce tetrahexahedral Au nanorods with a stepped (730) face and a multi-step face composed of (210) and (310) sub-faces for ENRR in 0.1 mol/L KOH electrolyte (Figs. 6a and b) [70]. The maximal rate of NH₃ production and FE were 1.648 g h⁻¹ cm⁻² and 3.879%, respectively. Meanwhile, hydrazine hydrate was produced to be 0.102 g h⁻¹ cm⁻². NH₃ production rate and FE reduce considerably as HER dominates the electrode reaction when the potential is lower than −0.2 V vs. RHE (Fig. 6c). The significant number of high-index (310) and (210) facets exposed on the catalysts, which provide considerable active sites for N₂ reduction, was respond for the extraordinarily high activity of Au nanorods. According to DFT calculations, N₂ electroreduction on Au nanorods followed an associative alternating pathway rather than an associative distal pathway, as evidenced by the concurrent synthesis of hydrazine of NH₃. Mukherjee et al. [71] performed ENRR using ZIF-8-derived N-doped and defect-rich nanoporous carbon as an electrocatalyst prepared by one-step thermal activation. At benign temperature conditions, the as-prepared electrocatalyst reveals a remarkable rate of NH₃ synthesis and FE of 3.4 × 10⁻⁶ mol cm⁻² h⁻¹ and 10.2%, respectively, using alkaline electrolyte. The N-doped nanoporous carbon electrocatalyst also demonstrated excellent stability for a continuous test of 18 h with a constant rate of NH₃ production. Further, the catalytic activity of N-doped nanoporous carbon with Fe doping was investigated. The rate of NH₃ production was significantly decreased, as the Fe doping blocked the active sites of the electrocatalyst and facilitated HER.

  Figure 6

  

  Figure 6.  (a) Geometric models and (b) TEM image of Au nanorods. (c) Average yield of NH₃ and hydrazine hydrate and FEs of Au nanorods in 0.1 mol/L KOH. Reproduced with permission [70]. Copyright 2016, Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  4.2 Neutral electrolytes

  4.2.1   Electrocatalytic performances in neutral electrolytes

  The method of ENRR is greatly desired in neutral electrolytes in light of averting corrosion issues in both acidic and alkaline electrolytes. Moreover, neutral electrolytes with lower proton concentrations can be applied quickly and easily to reduce the competition from HER by lowering the coverage of *H, thus improving electrocatalytic selectivity. The ENRR activity in acidic and neutral electrolytes were investigated using Au catalyst [65,75]. The catalytic activity decreases at potentials greater than −0.5 V vs. RHE in both acidic and neutral electrolytes. Theoretically, the presence of plentiful protons in acidic electrolytes would cause a large amount of *H to cover the metals that prevent *N₂ adsorption, whereas the presence of protons in neutral electrolytes might increase the activity of ENRR on those metals. Consequently, the current densities, NH₃ production, and FEs vary with the different pH values of electrolytes over Pd/C electrocatalyst and operate best in phosphate buffer solution (PBS) of neutral pH (Fig. 7a) [76]. An NH₃ production rate of ~4.5 µg mg⁻¹ h⁻¹ and FE of 8.2% were achieved at −0.1 V vs. RHE. Mukherjee et al. [77] used three electrolytes, including 0.01 mol/L HCl with pH 1, 0.5 mol/L LiClO₄ with pH 8, and 0.1 mol/L KOH with pH 13, to compare their catalytic activity for ENRR. The optimal NH₃ production is 115 mg cm⁻² h⁻¹ in a neutral medium, and the maximum FE is 18.5% for the atomic Ni site on N-doped carbon. Wu et al. analyzed the most recent findings and developments in Fe-based catalysts for N₂ reduction processes and found that these are well-suited for NH₃ production, especially in the presence of neutral electrolytes at lower potentials [78,79]. Liu et al. [80] reported a FE of 18.6% ± 0.8% and an NH₃ yield rate of 62.9 ± 2.7 µg h⁻¹ mg⁻¹ were produced using a Fe single-atom catalyst for ENRR in 0.1 mol/L PBS electrolyte. Hu et al. [81] employed Fe/Fe₃O₄ catalyst for ENRR and attained a maximum 0.19 µg cm⁻² h⁻¹ of NH₃ production rate with an 8.29% of FE at −0.3 V vs. RHE in 0.1 mol/L PBS electrolyte (pH 7.2). Chen et al. successfully created NiFe-MoS₂ nanocubes (NiFe@MoS₂ NCs) as a catalyst and evaluated their ENRR performance by adopting an electrolyte of 0.1 mol/L Na₂SO₄ solution [82]. Excellent electrocatalytic performances due to trimetallic synergistic interactions and the hollow NC structure with a high NH₃ yield rate\break (128.17 µg h⁻¹ mg⁻¹) and FE (11.34%) are obtained at a temperature of 40 ℃, at −0.3 V vs. RHE (Fig. 7b). To investigate the impact of H⁺ concentration on ENRR performance, Li et al. examined the current density, NH₃ yield rate, and FE in 0.05 mol/L H₂SO₄ and 0.1 mol/L Na₂SO₄ electrolytes [83]. Particularly the cathodic current density at negative potentials, is significantly higher in Ar-saturated 0.05 mol/L H₂SO₄ aqueous solution than in Ar-saturated 0.1 mol/L Na₂SO₄ aqueous solution, indicating the active HER in acidic electrolytes. As shown in Fig. 7c, similar patterns may be seen in the chronoamperometry test at −0.45 V. Lastly, the NH₃ yield rate as well as FE are much larger in 0.1 mol/L Na₂SO₄ than those in 0.05 mol/L H₂SO₄ electrolyte (Fig. 7d). Yu et al. [84] reported a FE of 42.68% and an NH₃ yield rate of 13.09 µg h⁻¹ mg⁻¹ using a Sn-based electrocatalysts for ENRR in 0.1 mol/L Na₂SO₄ electrolyte at −0.2 V vs. RHE. Zheng et al. designed a series of dual metal single atom catalysts containing adjacent M-N-C dual active centers to uncover the structure-activity relationship through combined theoretical and experimental study [85]. The catalyst revealed promising ENRR activity with the limiting potential of −0.64 ~0.62 V in aqueous electrolyte. As an effective electrocatalyst for ENRR to NH₃, a hybrid of TiO₂ and juncus effusus-derived carbon microtubes with a three-dimensional cross-linked hollow tubular structure is presented, which provides a significant NH₃ yield of 20.03 µg h⁻¹ mg_cat.⁻¹ and a high FE of 10.76% at −0.50 V vs. RHE, as well as greater electrochemical and structural durability in 0.1 mol/L Na₂SO₄ [86].

  Figure 7

  

  Figure 7.  (a) NH₃ yield rates and FEs in different electrolytes. Reproduced with permission [76]. Copyright 2018, Springer Nature. (b) NH₃ yields of Ni-Fe-MoS₂ and MoS₂. Reproduced with permission [82]. Copyright 2020, Royal Society of Chemistry. (c) Chronoamperometric curves of NH₂-MIL-88B-Fe in 0.1 mol/L Na₂SO₄ and 0.05 mol/L H₂SO₄ measured at −0.45 V. (d) NH₃ yield rates and FEs in 0.1 mol/L Na₂SO₄ and 0.05 mol/L H₂SO₄ electrolytes. Reproduced with permission [83]. Copyright 2020, Springer Nature.

  DownLoad: Full-Size Img PowerPoint

  4.2.2   Electrocatalysts for neutral electrochemical nitrogen reduction

  By engineering efficient electrocatalysts, the ENRR performance in natural electrolytes can be significantly enhanced. Fe₂O₃ nanorods with porous nature on carbon cloth (p-Fe₂O₃/CC) were created by Wang et al. [87] using the hydrothermal technique and high-temperature calcination (Fig. 8a). The binder-free porous structure of p-Fe₂O₃/CC exhibits active sites in large numbers and available open channels for N₂ adsorption, contributing to the reduction of NH₃. As a result, the p-Fe₂O₃/CC catalyst presents excellent stability, with a maximal NH₃ production rate of 13.56 µg mg_cat.⁻¹ h⁻¹ and a FE of 7.69% in 0.1 mol/L Na₂SO₄ (Fig. 8b). Moreover, the highest NH₃ production rate and FE were achieved over p-Fe₂O₃/CC catalyst with a hydrothermal time of 6 h. The high efficiency of p-Fe₂O₃/CC was ascribed to its higher value of double layer capacitance, which offers a larger electrochemical active surface area and exposes active sites for N₂ reaction (Fig. 8c).

  Figure 8

  

  Figure 8.  (a) schematic illustration for preparing p-Fe₂O₃/CC catalyst. (b) NH₃ production rates and FEs of p-Fe₂O₃/CC catalyst. (c) NH₃ production rates and FEs of p-Fe₂O₃/CC, Fe₂O₃–4/CC, Fe₂O₃–8/CC, and CC catalysts. Reproduced with permission [87]. Copyright 2019, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  Zhang et al. [46] synthesized TiO₂/Ti catalyst using a hydrothermal method, and achieved a maximum NH₃ production rate of 9.16 × 10⁻¹¹ mol s⁻¹ cm⁻² in 0.1 mol/L Na₂SO₄ electrolyte, which is superior to most reported electrocatalysts for ENRR (Table 1). The highest FE of TiO₂/Ti electrocatalyst for NH₃ production was obtained to be 3.34% at −0.6 V and reduced to 2.50% at −0.7 V due to the competing HER. Zhang et al. [50] loaded TiO₂-reduced graphene oxide (rGO) on carbon paper (CP) to prepare TiO₂-rGO/CP electrocatalyst for the ENRR to NH₃. The TiO₂-rGO/CP electrocatalyst revealed a maximum rate of NH₃ production (15.13 µg mg_cat.⁻¹ h⁻¹) and high FE (3.3%) in 0.1 mol/L Na₂SO₄ at −0.90 V vs. RHE. The NH₃ yield rate was also compared with CP, rGO/CP, TiO₂/CP, and TiO₂-rGO/CP electrocatalyst (Figs. 9a and b). A negligible performance of the CP as an electrocatalyst is observed, and the performance of the TiO₂-rGO/CP was superior to those of the TiO₂/CP and rGO/CP electrocatalysts. The superior performance is attributed to the well-dispersed TiO₂ nanoparticles over rGO, which provides maximum exposure of active sites for N₂ adsorption and reduction to NH₃. Furthermore, the NH₃ production rate and FE maintain for six successive recycling tests at −0.9 V, indicating the good electrochemical stability of the TiO₂-rGO/CP electrocatalyst (Fig. 9c).

  Figure 9

  

  Figure 9.  (a) NH₃ production rates and FEs for TiO₂-rGO/CP in 0.1 mol/L Na₂SO₄. (b) NH₃ production rates using various electrodes at −0.90 V. (c) NH₃ production rates and FEs at −0.90 V during recycling tests. Reproduced with permission [50]. Copyright 2018, Royal Society of Chemistry.

  DownLoad: Full-Size Img PowerPoint

  Zhang et al. developed MnO nanoparticles on a Ti mesh (MnO/Ti) for natural ENRR and achieved a very high NH₃ production rate of 1.11 × 10⁻¹⁰ mol s⁻¹ cm⁻² and FE of 8.02% in 0.1 mol/L Na₂SO₄ at −0.39 V vs. RHE [47]. After ten repeated cycles, no loss of NH₃ production rate, and FE are observed at −0.39 V, suggesting outstanding electrochemical stability. DFT calculations demonstrate low adsorption energy of N on the Mn atom than that of H on the O atom of the MnO (200) surface, indicating the high selectivity of MnO/Ti catalyst for N₂. Multi-shelled hollow Cr₂O₃ microspheres (MHCMs) served as useful electrocatalysts for ENRR to NH₃ [49]. The highest value of NH₃ production rate by MHCMs electrocatalysts is 25.3 g mg_cat.⁻¹ h⁻¹ with a FE of 6.78% in 0.1 mol/L Na₂SO₄. DFT calculations were executed to understand the mechanism on the MHCMs electrocatalyst surface for ENRR. The calculations and results suggest that the first N₂ adsorbed on the Cr³⁺ species of MHCMs, followed by the hydrogenation, took place by (H + e⁻) pair addition. Thus, both partially alternative routes and distal associative are involved during the ENRR over the MHCMs electrocatalysts. To improve the ENRR performance in natural aqueous electrolytes, Li et al. [88] hydrothermally synthesized Y₂O₃ nanosheets. The NH₃ production rate and FE are attained to be 1.06 × 10⁻¹⁰ mol s⁻¹ cm⁻² and 2.53%, respectively, in 0.1 mol/L Na₂SO₄ aqueous solution utilizing Y₂O₃ nanosheets at −0.9 V vs. RHE. DFT simulations show that the *N₂ hydrogenation to generate *N₂H is the potential deciding step of the reaction. Song et al. [89] used the chemical vapor deposition method to synthesize N-doped carbon nano-spikes (CNS) as an effective electrocatalyst for ENRR to NH₃. Using CNS electrocatalyst, the maximum NH₃ production rate (97.18 µg h⁻¹ cm⁻²) and FE (11.56%) are obtained at −1.19 V vs. RHE in 0.25 mol/L LiClO₄. The remarkable electrocatalytic activity of CNS catalyst for ENRR is ascribed to its sharp tip texture that amplifies the strong electric field and significantly affects molecular orbital energy levels in N₂.

  4.3 Acidic electrolytes

  4.3.1   Electrocatalytic performances in acidic electrolytes

  Typically, acidic electrolytes exhibit a higher solubility of N₂ compared to alkaline and natural media [90]. Nevertheless, HER often dominates the electrochemical process due to the adequate supply of protons. Therefore, the acidic ENRR performance can be efficiently adjusted by regulating the mass transfer of protons. For instance, Cheng et al. [91] used molten salt synthesis and fabricated Mo₂C/C catalyst by embedding Mo₂C nanodots on ultra-thin carbon nanosheets with a particle size of 2–3 nm (Figs. 10a). Figs. 10b and c show the activity of N₂ reduction over Mo₂C/C catalyst supported on hydrophilic CC as well as hydrophobic CC in aqueous solutions of 0.5 mol/L Li₂SO₄ with pH of 2 and 3. It is anticipated that hydrophobic CC and an electrolyte with a higher pH value are thought to reduce HER by limiting the number of protons available at the electrode surface, hence boosting FE. The results of the experiments corroborated with this theory show that when the accessibility of protons decreased, the highest FE increases from 1.6% to 7.8%. Yet, higher FEs are met with lower NH₃ formation rates. This suggests that while severe HER suppression can increase faradic efficiency, more is needed to accomplish a high NH₃ generation rate. These results illustrate the challenges to establish a high NH₃ production rate and FE at an electrode surface. Han et al. obtain the highest NH₃ production rate of 4.80 × 10⁻¹⁰ mol s⁻¹ cm⁻² and FE of 0.8% at −0.50 V vs. RHE in 0.1 mol/L HCl with MoO₃ nanosheets as electrocatalysts [63,92]. Besides, Excellent performance in the ENRR to NH₃ on Nb₂O₅ nanofibers was reported, and a maximum NH₃ production rate of 43.6 µg mg_cat.⁻¹ h⁻¹ and FE of 9.26% at −0.55 V vs. RHE in 0.1 mol/L HCl is attained. DFT simulations show that the surface Nb-edge atoms of Nb₂O₅ (181) are responsible for polarizing and activating N₂ molecules, which is necessary for their reduction to NH₃.

  Figure 10

  

  Figure 10.  (a) An illustration of a schematic for synthesizing Mo₂C/C using the molten salt approach. (b) FEs and (c) NH₃ yield rates over Mo₂C/C on hydrophilic CC and hydrophobic CC with pH of 2 and 3 in 0.5 mol/L Li₂SO₄. Reproduced with permission [91]. Copyright 2018, Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  4.3.2   Electrocatalysts for acidic electrochemical nitrogen reduction

  To improve the performance of acidic ENRR, Du et al. [64] reported the Cr₂O₃ nanofiber as the catalyst for N₂ reduction. The Cr₂O₃ nanofiber reached a high production rate of 28.13 µg mg_cat.⁻¹ h⁻¹ and FE of 8.56% in 0.1 mol/L HCl at −0.7 V vs. RHE with strong electrochemical durability after six cycles during electrolysis. Lv et al. [60] used electrospinning followed by calcination in the air to prepare a Bi₄V₂O₁₁/CeO₂ hybrid catalyst for ENRR to NH₃ under benign conditions. The molar ratio of Ce and Bi can affect the crystallinity of Bi₄V₂O₁₁ in a hybrid catalyst profoundly. The obtained Bi₄V₂O₁₁/CeO₂ hybrid with an amorphous phase (BVC-A) hybrid catalyst exhibited crystalline CeO₂ and amorphous Bi₄V₂O₁₁ when the molar ratio of Bi and Ce was 2:1 (Fig. 11a). The BVC-A hybrid catalyst outperformed in terms of NH₃ production rate than Bi₄V₂O₁₁/CeO₂ hybrid with a crystal phase (BVC-C) with the Bi and Ce molar ratio of 4:1, Bi₄V₂O_11, and CeO₂ (Fig. 11b). In 0.1 mol/L HCl, a maximum NH₃ generation rate of 23.32 µg mg_cat.⁻¹ h⁻¹ and a FE of 10.16% are achieved with the BVC-A hybrid catalyst (Fig. 11c). The inclusion of amorphous Bi₄V₂O₁₁ species, which can offer abundant faulty sites and a low energy barrier for N₂ reduction to NH₃, is credited with the outstanding performance of the BVC-A hybrid catalyst.

  Figure 11

  

  Figure 11.  (a) Schematic illustration for prrparing BVC-A and BVC—C catalysts. (b) Electrochemical performances of Bi₄V₂O₁₁, CeO₂, BVC-A, and BVC—C at −0.2 V vs. RHE. (c) NH₃ formation rates and FEs of BVC-A in 0.1 mol/L HCl. Reproduced with permission [60]. Copyright 2018, Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  By thermal nitriding in an NH₃ flow environment, Zhang et al. [93] created VN nanosheet arrays on a Ti grid (VN/TM) and VN nanowire arrays on CC (VN/CC). In N₂-saturated 0.1 mol/L HCl, an NH₃ production rate of 8.40 × 10⁻¹¹ mol s⁻¹ cm⁻² with a corresponding FE of 2.25% is realized at −0.50 V vs. RHE. Additionally, the catalyst maintains a steady voltage of −0.50 V vs. RHE during recycling. Zhang et al. sythesized an array of MoN nanosheets on carbon fabric (MoN/CC) and similarly used a method similar to that of VN nanostructures [67]. Ren et al. used a N₂-saturated 0.1 mol/L HCl electrolyte and investigated the Mo₂N nanorods as an electrocatalyst to obtain an NH₃ production rate of 78.4 µg mg_cat.⁻¹ h⁻¹ and FE of 4.5% for N₂ reduction at −0.3 V vs. RHE [58]. The Mars Van Krevelen mechanism is responsible for the N₂ reduction over the Mo₂N catalyst. Li et al. [66] used the facile co-reduction method to deposit Au and CeO_x nanoparticles on RGO (Au/CeO_x-RGO) for ERNN. Au nanoparticles amorphous formation is observed, having a particle size of ~5 nm. Compared with Au/CeO_x and Au/RGO, Au/CeO_x-RGO electrocatalyst is a more efficient catalyst in acidic ENRR (Fig. 12a). Its superior performance is attributed to amorphous Au nanoparticles on the substrate having abundant unsaturated coordination sites. The maximum NH₃ synthesis rate of 8.3 µg mg_cat.⁻¹ h⁻¹ and FE of 10.10% are obtained using Au/CeO_x-RGO catalyst having 1.31 wt% loading of Au at −0.2 V vs. RHE in 0.1 mol/L HCl. A noteworthy decline in the rate of NH₃ production, as well as the FE, are observed due to the HER at higher negative potentials (Fig. 12b).

  Figure 12

  

  Figure 12.  (a) NH₃ synthesis rates achieved using several catalysts at −0.2 V vs. RHE. (b) NH₃ synthesis rate and FEs of the Au/CeO_x-RGO composite in 0.1 mol/L HCl. Reproduced with permission [66]. Copyright 2017, Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  Liu et al. [61] used the same approach to formulate an electrocatalyst composed of ZIF-8 derived N-doped nanoporous carbon and 0.05 mol/L H₂SO₄ electrolyte to achieve a maximum NH₃ production rate (2.6 × 10⁻¹⁰ mol s⁻¹ cm⁻²) as well as FE (0.9%) at −1.1 V vs. RHE. Under optimal conditions, Cu₃P nanoribbon is developed as a highly effective electrocatalyst for the conversion of N₂ to NH₃. Such Cu₃P nanoribbon exhibits remarkable FE up to 37.8% and a significant yield of 18.9 µg h⁻¹mg_cat.⁻¹ at −0.2 V when measured in N₂-saturated 0.1 mol/L HCl [94]. Chen et al. fabricated plasma-etched Ti₂O₃ with oxygen vacancies which presents outstanding N₂ fixation performances with the maximum NH₃ yield and FE of up to 37.24 µg h⁻¹ mg_cat.⁻¹ and 19.29%, respectively in 0.1 mol/L HCl [95]. Ti₂O₃ nanoparticles function effectively as a pure Ti₃⁺ system for NH₃ electrosynthesis. According to the investigation, the Ti₃⁺ ions serve as the primary catalytically active sites and considerably boost ENRR activity. In an acidic electrolyte, Ti₂O₃ performs exceptionally well with high NH₃ yield and FE, outperforming the majority of titanium-based catalysts previously reported [96]. Lv et al. [97] also used the thermal treatment method and fabricated polymeric carbon nitride (PCN) for ENRR. The PCN catalyst reveals excellent electrocatalytic efficiency with an NH₃ synthesis rate of 8.09 µg mg_cat.⁻¹ h⁻¹ and FE of 11.59% in 0.1 mol/L HCl electrolyte (Figs. 13a and b). DFT calculations suggest that the N-vacancy of the PCN catalyst is the active site for N₂ molecule, which dramatically increases the N₂ bond length from 1.10 Å (for free N₂ molecule) to 1.26 Å due to the electrons transfer between adsorbed N₂ and adjacent carbon atoms (Fig. 13c). This N₂ bond length falls between diazene (1.20 Å) and hydrazine (1.47 Å) and facilitates the reduction of N₂ into NH₃.

  Figure 13

  

  Figure 13.  (a) NH₃ formation rates and (b) FEs of PCN in 0.1 mol/L HCl which is N₂-saturated. (c) The charge density difference of the N₂-adsorbed PCN with N-vacancies and the geometry of N₂ adsorption on PCN with N-vacancies. Reproduced with permission [97]. Copyright 2018 Wiley-VCH.

  DownLoad: Full-Size Img PowerPoint

  Yu et al. [98] performed a thermal reduction to yield a boron-doped graphene electrocatalyst. The graphene framework retains its conjugated planar structure and sp² hybridization but induces electron deficiency by doping boron atoms (Figs. 14a and b). The boron being positively charged atoms not only offers active sites for N₂ adsorption but also prohibits the binding of protons, resulting in an outstanding electrocatalytic performance and high FE for N₂ reduction to NH₃. According to DFT calculation, the BC₃ structure of boron-doped graphene reveals the lowest energy barrier between the various boron doping positions such as B₄C, BC₃, BC₂O, and BCO₂ and is beneficial for N₂ adsorption and its reduction to NH₃. The boron-doped graphene having 6.2% boron concentration shows a maximum NH₃ production rate of 9.8 µg h⁻¹ cm⁻² and FE of 10.8% in 0.05 mol/L H₂SO₄ electrolyte at −0.5 V vs. RHE under benign conditions (Fig. 14c).

  Figure 14

  

  Figure 14.  (a) Schematic illustration for preparing boron-doped graphene. (b) TEM image, (c) NH₃ yields and FEs of boron-doped graphene. Reproduced with permission [98]. Copyright 2018, Elsevier.

  DownLoad: Full-Size Img PowerPoint

  Through interfacial engineering approach, Rh atomic layer-coated SnO₂ hetero-structured catalyst was rationally designed with the maximum NH₃ yield rate (149 µg mg_cat.⁻¹ h⁻¹) and FE (11.69%) in 0.05 mol/L H₂SO₄ electrolyte (Fig. 15) [99]. According to DFT and X-ray absorption spectra calculations, the SnO₂ heterostructure and the Rh atomic layer exhibit strong electron interactions, efficiently controlling the interfacial electron transfer and d-band center. The adsorption energy of H is significantly lowered, and the reaction kinetics for N₂ reduction is dramatically expedited as a result of the downshift of the d-band center. By reducing H adsorption and further enhancing the electrocatalytic N₂ reduction, this work offers unique insight into the regulation of interfacial electrons.

  Figure 15

  

  Figure 15.  (a) UV–vis spectra of NH₃ production rate after electrolysis for 2 h at −0.35 V vs. RHE. (b) Comparison of NH₃ yield rates and FEs on pure Rh, SnO₂, Rh_0.3@SnO₂, Rh_0.5@SnO₂, Rh_1.3@SnO₂, and mechanically mixed Rh_0.5/SnO₂ catalysts at −0.35 V vs. RHE. (c) NH₃ yield rates and FEs on Rh_0.5@SnO₂ at −0.35 V vs. RHE after consecutive tests. (d) Chronoamperometry curve for 20 h on Rh_0.5@SnO₂. Reproduced with permission [99]. Copyright 2022, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  4.4 Water-in-salt electrolytes

  WiS, as a novel type of electrolyte, has extremely high salt concentrations and have been explored for boosting the efficiency of ENRR. Two benefits would be anticipated from such a system. First, the high quantity of salt has a substantial solvation effect, locking down water molecules, reducing the amount of free water available, and effectively suppressing HER [100,101]. Also, the WiS electrolyte provides a chance to fine-tune the H₂O concentration for electrokinetic studies of ENRR, a parameter that is previously challenging to change but crucial to understanding the intricate mechanisms. It also has safety benefits and low toxicity.

  It should be noted that WiS electrolytes can increase the electrochemical stability window to over 3.0 V, which could overcome the restricted electrochemical windows of HER/OER for conventional aqueous electrolytes [102,103]. As a result, WiS electrolyte is a desirable choice for ENRR research in the future. Wu et al. recently described an NH₃ production method with 20 mol/L lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) as a super-concentrated WiS system. As the catalyst for ENRR, oxygen vacancy-rich TiO₂ loaded with Au nanoparticles was used. A high FE of 37.2% and NH₃ yield of 20.3 µg h⁻¹ cm⁻² were achieved at −0.4 V vs. RHE. According to theoretical calculations, the H₂O molecules are effectively stopped from accessing the catalytic surface by the extremely concentrated LiTFSI salt solution, which deactivates them by intense ionic contact. Hence, in this WiS electrolyte, the energy barrier for the HER is raised, and the H₂O diffusion is severely constrained. Since the current density and the onset potential decrease drastically due to the unique action of the WiS electrolyte, the HER is severely slowed down [104].

  Similarly, Wang et al. carried out ENRR in highly concentrated LiCl electrolytes. This proof-of-concept system accomplishes a high FE of 71.9% even when utilizing a metal-free electrocatalyst. They believed the effectiveness of H₂O as both a proton source and a solvent is severely hampered by establishing a hydration shell caused by the solute ions of highly concentrated electrolytes' significant attraction for the nearby H₂O molecules. Hence, a significant N₂ flux and an effective inhibition of the HER occur at the reaction interface, enhancing activity and selectivity during the following reduction phase [105]. Chu et al. evaluated the ENRR performances in various concentrations of LiClO₄ WiS electrolytes and obtained a good FE of 62.5% and NH₃ production of 181.3 µg h⁻¹ mg⁻¹ in 12 mol/L LiClO₄ WiS electrolytes. They discovered that the water molecules are slowed down by being encased in solvation sheaths as the electrolyte concentration was raised, effectively blocking the HER. The decline in N₂ affinity on the catalyst surface is caused by the extremely low diffusion rate at extremely high concentrations, which limits the movement of N₂ to the catalyst surface. Meanwhile, WiS can support transfer electrons to antibonding orbitals of N₂ molecules favorably and promote N₂ activation, as shown in XPS studies. As a result, the WiS system effectively promotes ENRR activity and selectivity [106].

  4.5 Summary

  Aqueous electrolytes are generally the most popular choice for N₂ reduction reactions due to their ease of production, low cost, and minimal toxicity. ENRR has demonstrated an astonishing advance by effectively selecting suitable aqueous electrolytes. The previous literatures have reviewed that acidic electrolytes are generally more favorable for ENRR in terms of selectivity and activity compared to alkaline and neutral solutions. This may be ascribed to two reasons. First, in acidic electrolytes, the solubility of N₂ is generally higher than those in alkaline and natural solutions. Besides, as the rate-determining step for ENRR is the formation of *N₂H through a PCET process, acidic electrolyte may be beneficial for ENRR if the electrocatalysts are designed to has weak adsorption of H species. Therefore, more efforts can be devoted to acidic ENRR. However, the comparatively poor N₂ solubility in the aqueous phase, even in acidic electrolytes, continues to impede the advancement of ENRR. In addition, the lower FE for ENRR resulting from the unavoidable HER side reactions are still far from satisfactory. Hence, there is still a need of further advancement in each category of aqueous electrolyte in terms of solubility of N₂. Besides, it is also crucial to develop alternative techniques or optimize the catalysts to offset the disadvantage of aqueous electrolytes by combining the HER suppression techniques [32,78,107,108].

  5. Non-aqueous electrolytes

  Since non-aqueous electrolytes have much lower water content than aqueous electrolytes and are good options considering that many protons would be unfavorable for the required reactions. Moreover, N₂ is far more soluble in non-aqueous solutions than in aqueous solutions [90]. Interestingly, alternative reaction pathways to promote ENRR are possible in non-aqueous electrolytes [109]. As a result, non-aqueous electrolyte research has great potential and receives great interest. Three categories of non-aqueous electrolytes have been established by their varied chemical compositions: organic electrolytes, ILs, and mixed water-organic electrolytes. Table 2 compares the electrochemical performance of catalysts in various types of non-aqueous electrolytes [110–116]. These will be introduced as follows.

  Table 2

  

  Table 2.  Compositions of non-aqueous electrolytes and their ENRR performances.

  DownLoad: CSV

  5.1 Organic electrolytes

  Organic electrolytes have received extensive attention in electrocatalysis. As mentioned above, to slow down HER, we can either raise the proton transfer barrier to the catalyst surface or lower the concentration of proton in the bulk solution [36].

  For example, the HER is suppressed by employing DMF to reduce proton donors [117], and a protective layer to prevent transport of proton in the context of electrocatalysis for the production and consumption of fuels [118]. In addition, the chemical structure of proton donor substantially affects the HER activity in acetonitrile. With the donor, tetraethylammonium hydroxide (TEAH⁺), which is less bulky, exhibits higher HER activity at all observed potentials than the bulkier donor, N,N-diisopropylethylamine (protonated) (DIPEAH⁺) [119]. Hence, organic electrolytes offer essential approaches to regulating the transfer of both electrons and protons. Gorodisky et al. [120]. reported the Ti(OH)₃−Mo(Ⅲ) system for catalytic N₂ reduction. It has been suggested that clusters of Ti(Ⅲ) and Mo(Ⅲ) hydroxides must have adhered to the cathode, which forms a complex with Mo(Ⅲ) for electrons to move from the anode to the N₂ molecule at a suitable negative potential. However, a more precise mechanism is still unknown. The highest NH₃ and N₂H₄ yields are observed when the electrolyte contains up to 3% water, indicating that water is also important for NH₃ synthesis. Lee et al. [116] employed a solvent mixture of ethanol and 0.2 mol/L lithium trifluoromethanesulfonate (LiCF₃SO₄) in tetrahydrofuran as the electrolyte to increase the selectivity of NH₃ under ambient conditions, in which ethanol is used as the proton source. Ag NCs were first positioned on an Au electrode and then covered in a superhydrophobic ZIF layer to produce the ZIF/Ag-Au electrocatalyst. The super-hydrophobic ZIF is employed to stop water molecules from getting to catalyst surface. The ZIF with strong gas adsorptive ability is used to concentrate the catalyst surface with reactant molecules and N₂. The ZIF/Ag-Au electrocatalyst enhances N₂ adsorption, promotes the ENRR, and blocks the competing HER. The NH₃ production rate and FE of the ZIF/Ag-Au electrode superior to those of the Ag-Au electrode. Remarkably, the Ag-Au@ZIF electrocatalyst achieves a FE of 18% ± 4% when converting N₂ to NH₃ electrochemically.

  Organic electrolytes are the most widely studied for the Li-mediated ENRR process. Li metal and N₂ combine to form Li₃N at room temperature, as is well known. After that, Li₃N produces NH₃ as a result of a reaction with water and ethanol as proton sources. Li may be used in the ENRR to NH₃ as a catalyst, as the initial potential for HER is substantially larger than that for reducing Li⁺ to Li. Li metal very quickly reacts with water to form H₂, and therefore this process cannot be used in aqueous solutions. As a result, aprotic solvents are frequently employed in this reaction. One of the earliest demonstrations was provided by Tsuneto et al. [121] using Zn, Ag, Cu, Ni, Co, Fe, Mo, Ti, Pb, and Al as cathode materials and THF solution (0.18 mol/L ethanol and 0.2 mol/L LiClO₄) as electrolyte for ENRR in a electrochemical cell with single-compartment having a metal cathode and a Pt anode. With respective values of 8.2% and 8.4% for the production of NH₃ when N₂ was present as atmospheric pressure, the highest FE was offered by Ti and Ag among them. When the electrolysis was carried out without adding ethanol to the electrolyte or when Ar was used as the working medium, controlled investigations revealed that only a trace amount of NH₃ was produced. Additionally, the amount of NH₃ production was insignificant when water replaced ethanol as the electrolyte or NaClO_4, or Bu₄NClO₄ replaced LiClO₄. Ma et al. [122] described a rechargeable Li-N battery based on the reaction of Li and N₂ that utilized a Li-foil anode, a CC cathode, an ether-based electrolyte comprising 1 mol/L LiCF₃SO₃, and a glass fiber separator. When N₂ molecules mix with Li⁺ at the cathode during discharge, Li-containing discharge products (mainly Li₃N) are formed; these discharge products break down during charging. NH₃ is formed when the discharge products are hydrolyzed, and the best FE for NH₃ production is 59%.

  5.2 Ionic liquid electrolytes

  ILs are suitable nonaqueous electrolytes for the ENRR due to their high solubility of N₂ and low water content. Zhou et al. used two different ILs ([C₄mpyr][eFAP] and [P_6,6,6,14][eFAP]) to study the performance of ENRR over Fe deposited F-doped SnO₂ glass, Fe deposited nickel foam, and Fe deposited stainless steel cloth [111,123]. The lack of N atoms in [P_6,6,6,14][eFAP] IL makes it possible to rule out IL as a N source. The ILs contain only 100 ppm of water, which serves as the sole source of protons for the ENRR. All systems obtain a minimum NH₃ generation rate of 2.9 mg h⁻¹ m⁻² and a 60% FE (Figs. 16a–c). Suryanto et al. employed carbon fiber paper with α-Fe nanorods as electrodes, and they produced NH₃ by reacting an aprotic fluorinated solvent with an IL [70]. The electrode material was produced hydrothermally by β-FeOOH growth on carbon fiber paper, which was then transformed into α-Fe by annealing in the presence of a steady flow of H₂. A thin Fe₂O₃ layer of ~5 nm thickness is formed on the surface of α-Fe because of exposure to air oxygen. Applying a negative voltage will electrochemically decrease this layer to the active α-Fe surface. Because it has a high N₂ solubility, the authors of this work utilized the aprotic fluorinated solvent 1H,1H,5H-octauoropentyl-1,1,2,2-tetrafluoroethyl ether (FPEE) to control the delivery of protons to the electrode. These solvents, however, typically have poor solubility when it comes to salts that are frequently utilized in aprotic electrolyte media. As a supporting electrolyte, [C4mpyr][eFAP], a strongly fluorinated IL, was used to solve this issue. In these tests, the electrolyte's water content was 106 ppm. It was discovered that the ideal applied potential is −0.65 V vs. NHE and IL mole fraction is 0.23. Under these conditions, the NH₃ formation rate and FE reach 1.58 × 10⁻¹¹ mol s⁻¹ cm⁻² and 23.8%, respectively. These values increase to 2.35 × 10⁻¹¹ mol s⁻¹ cm⁻² and 32% when the water content in the electrolyte reached 114 ppm; however, both values decrease as the water concentration increased (Figs. 16d and e) [110].

  Figure 16

  

  Figure 16.  (a) Schematic of Fe-SS electrode based N₂ reduction process. (b) FEs for electro-reduction of N₂-saturated ILs on different electrodes at −0.8 V vs. NHE. (c) NH₃ production rates on several electrodes and ILs at −0.8 V vs. NHE. Reproduced with permission [111]. Copyright 2017, Royal Society of Chemistry. (d) FEs and the NH₃ yields at X_IL = 0.16. (e) NH₃ yields and FEs at −0.65 V vs. NHE. Reproduced with permission [110]. Copyright 2018, American Chemical Society. (f) Corresponding FEs and NH₃ yield rates. After chronoamperometry, the electrode was either coated with visible degradation products or was not, as indicated by the pink and light-green shading, respectively. Reproduced with permission [124]. Copyright 2022, Nature.

  DownLoad: Full-Size Img PowerPoint

  Hoang-Long et al. presented a high-efficiency, reliable technique made possible by compacting ionic stacking at the electrode-electrolyte interface [124]. A high concentration of imide-based Li salt is used as the electrolyte to create the interface, which prevents electrolyte breakdown and promotes steady N₂ reduction (Fig. 16f). Consequently, stabilized NH₃ yield rates of 150 ± 20 nmol s⁻¹ cm⁻² and NH₃ conversion efficiencies near 100% are acquired. This research emphasized the connection between the physicochemical characteristics of the electrode-electrolyte interface and the ENRR performance, which could serve as a roadmap for creating a reliable and highly efficient method for NH₃ production.

  5.3 Mixed organic-water electrolytes

  A mixture of water and organic solvents was often used as an electrolyte for ENRR, which can also reduce the competition from HER and boost the solubility of N₂ in the electrolyte. The NH₃ production rate increases initially as the proton donor concentration is increased from 0.015 mol/L to 0.03 mol/L but decreases when the concentration is increased to 0.06 mol/L in a methanol/LiClO₄/H⁺ solution on a Pt-plate coated with a polyaniline film. After electrolysis, a maximum concentration of 57 µmol/L is produced with a maximum current efficiency of 16% [112]. Recently, Ren et al. [115] proposed a high-efficiency electrode-electrolyte interface engineering strategy to promote ENRR by adopting an innovative alcohol-water electrolyte system. This new-type mixed alcohol-water electrolyte was found possessing the ability to control the proton-donating ability and adjust the microenvironment at the electrode-electrolyte interface, thus significantly suppressing the competitive HER and enhancing the selectivity and reaction rate for NH₃ generation. As a result, using the methanol-water electrolyte system, a distinguished FE of 75.9% ± 4.1% and a record high NH₃ yield rate of 262.5 ± 7.3 µg h⁻¹ mg_cat.⁻¹ were achieved on the FeOOH nanospindles anchored on CNTs (FeOOH/CNTs) catalyst, further demonstrating its superiority for ENRR.

  A workable strategy to improve ENRR selectivity is to limit the proton donor concentration. Kim et al. utilized a mixture of 2-propanol and water (9:1, v/v) as an electrolyte and observed that 2-propanol better suppressed HER and dissolved more N₂ than pure water for ENRR [113]. The mixed electrolyte (2-propanol and water) reveals a high FE of 0.98% compared to 0.07% in pure water (Fig. 17a). Kim et al. [114] employed 0.05 mol/L H₂SO₄ as the anodic electrolyte and 0.1 mol/L LiCl/ethylenediamine (EDA) as the cathodic electrolyte. The effect of EDA on ENRR performance towards NH₃ is studied. The presence of EDA has a considerable influence on the FE, resulting in a significantly higher FE. It is conceivable because of the large electrochemical window of EDA to attain a much higher FE of 17.2% (Fig. 17b). In addition, alkali metal ions could interact strongly with molecular N₂ in an electrolyte and hinder the HER. It is also discovered that the selection of counter ions in the aqueous electrolyte has a significant effect, and the order of Li⁺ > Na⁺ > K⁺ in terms of NH₃ production rate shows that the smallest counter ions are the most advantageous (Figs. 17c and d) [89]. This might be the case because the ENRR is affected by the steric effect and a significant interaction between counter ions and N₂ [54].

  Figure 17

  

  Figure 17.  (a) Effect of the electrolyte medium on NH₃ synthesis rate and FE. Reproduced with permission [113]. Copyright 2016, Electrochemical Society. (b) A schematic representation of ethylenediamine (EDA)-based NH₃ synthesis. Reproduced with permission [114]. Copyright 2016, Electrochemical Society. (c, d) Comparison of electrolyte counterion effect of Li⁺, Na⁺, and K⁺ on NH₃ synthesis rate and FE. Reproduced with permission [89]. Copyright 2018, Science.

  DownLoad: Full-Size Img PowerPoint

  5.4 Summary

  Since non-aqueous electrolytes improve reactant gas capture and partially suppress HER, the yield and FE of the ENRR are often superior to those in aqueous electrolytes. Nonaqueous electrolytes, however, still have a lot of challenges. First, their volatility, flammability, and probable toxicity under ambient conditions vary from the objective of green synthesis. In addition, the cost of nonaqueous electrolytes is quite high, particularly for ILs, and the current density is limited. Furthermore, the complicated structure of non-aqueous solvents such as ILs and mixed electrolytes makes it challenging to carry out thorough and clear investigations in the electrocatalytic process. A unique reaction mechanism for the different electrolytes has not been proposed. Therefore, a thorough investigation of the potential of non-aqueous electrolytes for ENRR in large-scale applications is necessary.

  6. Conclusion and perspectives

  The environmentally friendly N₂ electroreduction procedure, which has the potential to be used to store renewable energy in chemical bonds, is anticipated to replace the outdated Haber-Bosch process. The scientific community has been working hard to speed up and improve the N₂ electroreduction process since 2015. The majority of research has concentrated on developing electrocatalysts, which has given researchers insights into how to boost the activity and selectivity of the ENRR while stifling the rival HER. If the catalyst design is the exclusive focus of the research, the significance and requirements of the overall reaction system may be missed. As discussed in this review, electrolytes serve as a proton source, a medium to trap N₂, a method of transferring protons and intermediates during the reaction process, and a way of proton transfer during the development of effective systems for ENRR. Several exemplary works pertinent to the modulation of electrolytes are listed according to the types of solvents, including aqueous electrolytes and non-aqueous electrolytes. Even though FE and the yield of ENRR have greatly improved, the industrialization of ENRR is still a long way off. Several important aspects remain to be considered (Fig. 18).

  Figure 18

  

  Figure 18.  Perspectives for enhancing ENRR.

  DownLoad: Full-Size Img PowerPoint

  There is a further need to explore these tough but essential areas to know how electrolytes affect electrocatalytic performance properly in the future. More complex in situ methods, such as in situ spectroscopy and in situ mass spectrometry, should be developed in order to identify the real active sites and keep track of significant intermediates. These kinds of advanced in-situ techniques greatly improve catalytic performance and provide essential resources to support deepgoing analysis of the electrolyte in electrochemical reduction processes. Besides, it is urgently necessary to make significant advancements in theoretical calculations and computational analysis to provide a deeper insight into the electrocatalytic mechanism, electrolyte effects (e.g., pH, buffer ability), and the change that occurred on the surface of the catalyst under different electrolytes. Computational analysis can identify the active N₂ activation sites in different electrolytes and reveal the least-energetic chemical pathway. The method of DFT-guided synthesis can also be used to tailor the electrolytes to improve the quantity of N₂ activation sites, encourage a reaction with lower free energy, and require less overpotential. According to the findings of the published experiments, the operating parameters, electrolytes, and catalysts have a considerable impact on the generation and distribution of intermediates, as well as the mass transfer of reactants and products, etc. But most of these components have been overlooked. Further research is required to develop multiscale techniques that can take into account the impacts of mass transit, electrolytes, and catalysts to better understand the ENRR in this setting. For the ENRR method to be used in practice, an N₂-electrolyzer that incorporates the anode reaction into the overall NH₃ production process is required. Because the OER at the anode can absorb a portion of the excess potential in the aqueous phase, it is recommended to use flowing organic electrolytes (ethanol, methanol, and urea) to reduce anodic overpotential. Hence, in this interdisciplinary subject, collaboration between chemists, electrochemists, materials scientists, and chemical engineers may lead to significant advancement. Only if ENRR is made economically viable will they be a useful and practical option to store renewable energy and drastically lower energy use and CO₂ emissions.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (No. 22272150), the Major Program of Zhejiang Provincial Natural Science Foundation of China (Nos. LD22B030002 and LZ23B030002), the Zhejiang Provincial Ten Thousand Talent Program (No. 2021R51009), and the Key Science and Technology Project of Jinhua City (No. 2020-1-044).

  
  
• 1. [1]

     T. Xu, B. Ma, J. Liang, et al., Acta Phys. Chim. Sin. 37 (2021) 2009043.

  2. [2]

     G. Qing, R. Ghazfar, S.T. Jackowski, et al., Chem. Rev. 120 (2020) 5437–5516. doi: 10.1021/acs.chemrev.9b00659

  3. [3]

     X. Li, G. Zhang, P. Shen, X. Zhao, K. Chu, Inorg. Chem. Front. 10 (2023) 280–287. doi: 10.1039/d2qi02118h

  4. [4]

     W. Song, J. Wang, L. Fu, et al., Chin. Chem. Lett. 32 (2021) 3137–3142. doi: 10.1016/j.cclet.2021.02.043

  5. [5]

     Q. Liu, T. Xu, Y. Luo, et al., Curr. Opin. Electrochem. 29 (2021) 100766. doi: 10.1016/j.coelec.2021.100766

  6. [6]

     C. Smith, A.K. Hill, L. Torrente-Murciano, Energy Environ. Sci. 13 (2020) 331–344. doi: 10.1039/c9ee02873k

  7. [7]

     A. Klerke, C.H. Christensen, J.K. Nørskov, T. Vegge, J. Mater. Chem. 18 (2008) 2304–2310. doi: 10.1039/b720020j

  8. [8]

     J. Liang, Q. Liu, A.A. Alshehri, X. Sun, Nano Res. Energy 1 (2022) e9120010. doi: 10.26599/nre.2022.9120010

  9. [9]

     Y. Tan, L. Yan, C. Huang, et al., Small 17 (2021) 2100372. doi: 10.1002/smll.202100372

  10. [10]

      S. Licht, B. Cui, B. Wang, et al., Science 345 (2014) 637–640. doi: 10.1126/science.1254234

  11. [11]

      R. Shi, X. Zhang, G.I. Waterhouse, Y. Zhao, T. Zhang, Adv. Energy Mater. 10 (2020) 2000659. doi: 10.1002/aenm.202000659

  12. [12]

      Y. Luo, P. Shen, X. Li, Y. Guo, K. Chu, Nano Res. 15 (2022) 3991–3999. doi: 10.1007/s12274-022-4097-9

  13. [13]

      J. Kibsgaard, J.K. Nørskov, I. Chorkendorff, ACS Energy Lett. 4 (2019) 2986–2988. doi: 10.1021/acsenergylett.9b02286

  14. [14]

      B.H. Suryanto, H.L. Du, D. Wang, et al., Nat. Catal. 2 (2019) 290–296. doi: 10.1038/s41929-019-0252-4

  15. [15]

      J.H. Montoya, C. Tsai, A. Vojvodic, J.K. Nørskov, ChemSusChem 8 (2015) 2180–2186. doi: 10.1002/cssc.201500322

  16. [16]

      N. Cao, G. Zheng, Nano Res. 11 (2018) 2992–3008. doi: 10.1007/s12274-018-1987-y

  17. [17]

      C.J. van der Ham, M.T. Koper, D.G. Hetterscheid, Chem. Soc. Rev. 43 (2014) 5183–5191. doi: 10.1039/C4CS00085D

  18. [18]

      J. Hou, M. Yang, J. Zhang, Nanoscale 12 (2020) 6900–6920. doi: 10.1039/d0nr00412j

  19. [19]

      H. Xu, K. Ithisuphalap, Y. Li, et al., Nano Energy 69 (2020) 104469. doi: 10.1016/j.nanoen.2020.104469

  20. [20]

      D. Ma, Z. Zeng, L. Liu, Y. Jia, J. Energy Chem. 54 (2021) 501–509. doi: 10.1016/j.jechem.2020.06.032

  21. [21]

      D. Ma, Z. Zeng, L. Liu, X. Huang, Y. Jia, J. Phys. Chem. C 123 (2019) 19066–19076. doi: 10.1021/acs.jpcc.9b05250

  22. [22]

      Q. Wang, J. Guo, P. Chen, J. Energy Chem. 36 (2019) 25–36. doi: 10.30564/jams.v2i2.514

  23. [23]

      J. Yang, W. Weng, W. Xiao, J. Energy Chem. 43 (2020) 195–207. doi: 10.1016/j.jechem.2019.09.006

  24. [24]

      H.P. Jia, E.A. Quadrelli, Chem. Soc. Rev. 43 (2014) 547–564. doi: 10.1039/C3CS60206K

  25. [25]

      J. Deng, J.A. Iñiguez, C. Liu, Joule 2 (2018) 846–856. doi: 10.1016/j.joule.2018.04.014

  26. [26]

      X. Cui, C. Tang, Q. Zhang, Adv. Energy Mater. 8 (2018) 1800369. doi: 10.1002/aenm.201800369

  27. [27]

      D.D. Wagman, W.H. Evans, V.B. Parker, et al., J. Phys. Chem. Ref. Data 18 (1989) 1807–1812. doi: 10.1063/1.555845

  28. [28]

      N. Bauer, J. Phys. Chem. 64 (1960) 833–837. doi: 10.1021/j100836a001

  29. [29]

      M.T. Koper, Chem. Sci. 4 (2013) 2710–2723. doi: 10.1039/c3sc50205h

  30. [30]

      D.E. Brown, T. Edmonds, R.W. Joyner, J.J. McCarroll, S.R. Tennison, Catal. Lett. 144 (2014) 545–552. doi: 10.1007/s10562-014-1226-4

  31. [31]

      G.F. Chen, S. Ren, L. Zhang, et al., Small Methods 3 (2019) 1800337. doi: 10.1002/smtd.201800337

  32. [32]

      Y. Ren, C. Yu, X. Tan, et al., Energy Environ. Sci. 14 (2021) 1176–1193. doi: 10.1039/d0ee03596c

  33. [33]

      D.R. MacFarlane, P.V. Cherepanov, J. Choi, et al., Joule 4 (2020) 1186–1205. doi: 10.1016/j.joule.2020.04.004

  34. [34]

      L. Suo, O. Borodin, T. Gao, et al., Science 350 (2015) 938–943. doi: 10.1126/science.aab1595

  35. [35]

      L. Hu, Z. Xing, X. Feng, ACS Energy Lett. 5 (2020) 430–436. doi: 10.1021/acsenergylett.9b02679

  36. [36]

      A.R. Singh, B.A. Rohr, J.A. Schwalbe, et al., ACS Catal. 7 (2017) 706–709. doi: 10.1021/acscatal.6b03035

  37. [37]

      C. Tang, S.Z. Qiao, Chem. Soc. Rev. 48 (2019) 3166–3180. doi: 10.1039/c9cs00280d

  38. [38]

      Y.C. Hao, Y. Guo, L.W. Chen, et al., Nat. Catal. 2 (2019) 448–456. doi: 10.1038/s41929-019-0241-7

  39. [39]

      Y. Ren, C. Yu, X. Tan, et al., Small Methods 3 (2019) 1900474. doi: 10.1002/smtd.201900474

  40. [40]

      H. Tao, C. Lian, H. Jiang, et al., AIChE J. 68 (2022) e17549. doi: 10.1002/aic.17549

  41. [41]

      O. Westhead, M. Spry, A. Bagger, et al., J. Mater. Chem. A 11 (2023) 12746—12758. doi: 10.1039/D2TA07686A

  42. [42]

      E. Skúlason, T. Bligaard, S. Gudmundsdóttir, et al., Phys. Chem. Chem. Phys. 14 (2012) 1235–1245. doi: 10.1039/C1CP22271F

  43. [43]

      J.G. Howalt, T. Bligaard, J. Rossmeisl, T. Vegge, Phys. Chem. Chem. Phys. 15 (2013) 7785–7795. doi: 10.1039/c3cp44641g

  44. [44]

      M.C. Leech, K. Lam, Nat. Rev. Chem. 6 (2022) 275–286. doi: 10.1038/s41570-022-00372-y

  45. [45]

      L.V. Haynes, D.T. Sawyer, Anal. Chem. 39 (1967) 332–338. doi: 10.1021/ac60247a013

  46. [46]

      R. Zhang, X. Ren, X. Shi, et al., ACS Appl. Mater. Interfaces 10 (2018) 28251–28255. doi: 10.1021/acsami.8b06647

  47. [47]

      Z. Wang, F. Gong, L. Zhang, et al., Adv. Sci. 6 (2019) 1801182. doi: 10.1002/advs.201801182

  48. [48]

      L. Zhang, X. Ji, X. Ren, et al., Adv. Mater. 30 (2018) 1800191. doi: 10.1002/adma.201800191

  49. [49]

      Y. Zhang, W. Qiu, Y. Ma, et al., ACS Catal. 8 (2018) 8540–8544. doi: 10.1021/acscatal.8b02311

  50. [50]

      X. Zhang, Q. Liu, X. Shi, et al., J. Mater. Chem. A 6 (2018) 17303–17306. doi: 10.1039/c8ta05627g

  51. [51]

      X. Xiang, Z. Wang, X. Shi, M. Fan, X. Sun, ChemCatChem 10 (2018) 4530–4535. doi: 10.1002/cctc.201801208

  52. [52]

      X. Li, T. Li, Y. Ma, et al., Adv. Energy Mater. 8 (2018) 1801357. doi: 10.1002/aenm.201801357

  53. [53]

      Q. Liu, X. Zhang, B. Zhang, et al., Nanoscale 10 (2018) 14386–14389. doi: 10.1039/C8NR04524K

  54. [54]

      M. Nazemi, S.R. Panikkanvalappil, M.A. El-Sayed, Nano Energy 49 (2018) 316–323. doi: 10.1016/j.nanoen.2018.04.039

  55. [55]

      X. Zhu, Z. Liu, Q. Liu, et al., Chem. Commun. 54 (2018) 11332–11335. doi: 10.1039/c8cc06366d

  56. [56]

      V. Kordali, G. Kyriacou, C. Lambrou, Chem. Commun. (2000) 1673–1674.

  57. [57]

      D. Yang, T. Chen, Z. Wang, J. Mater. Chem. A 5 (2017) 18967–18971. doi: 10.1039/C7TA06139K

  58. [58]

      X. Ren, G. Cui, L. Chen, et al., Chem. Commun. 54 (2018) 8474–8477. doi: 10.1039/c8cc03627f

  59. [59]

      B.H. Suryanto, D. Wang, L.M. Azofra, et al., ACS Energy Lett. 4 (2018) 430–435.

  60. [60]

      C. Lv, C. Yan, G. Chen, et al., Angew. Chem. 130 (2018) 6181–6184. doi: 10.1002/ange.201801538

  61. [61]

      Y. Liu, Y. Su, X. Quan, et al., ACS Catal. 8 (2018) 1186–1191. doi: 10.1021/acscatal.7b02165

  62. [62]

      W. Qiu, X.Y. Xie, J. Qiu, et al., Nat. Commun. 9 (2018) 1–8. doi: 10.1038/s41467-017-02088-w

  63. [63]

      J. Han, Z. Liu, Y. Ma, et al., Nano Energy 52 (2018) 264–270. doi: 10.1016/j.nanoen.2018.07.045

  64. [64]

      H. Du, X. Guo, R.M. Kong, F. Qu, Chem. Commun. 54 (2018) 12848–12851. doi: 10.1039/c8cc07186a

  65. [65]

      M.M. Shi, D. Bao, B.R. Wulan, et al., Adv. Mater. 29 (2017) 1606550. doi: 10.1002/adma.201606550

  66. [66]

      S.J. Li, D. Bao, M.M. Shi, et al., Adv. Mater. 29 (2017) 1700001. doi: 10.1002/adma.201700001

  67. [67]

      L. Zhang, X. Ji, X. Ren, et al., ACS Sustain. Chem. Eng. 6 (2018) 9550–9554. doi: 10.1021/acssuschemeng.8b01438

  68. [68]

      K. Kugler, M. Luhn, J.A. Schramm, K. Rahimi, M. Wessling, Phys. Chem. Chem. Phys. 17 (2015) 3768–3782. doi: 10.1039/C4CP05501B

  69. [69]

      Z. Wang, Y. Li, H. Yu, et al., ChemSusChem 11 (2018) 3480–3485. doi: 10.1002/cssc.201801444

  70. [70]

      D. Bao, Q. Zhang, F.L. Meng, et al., Adv. Mater. 29 (2017) 1604799. doi: 10.1002/adma.201604799

  71. [71]

      S. Mukherjee, D.A. Cullen, S. Karakalos, et al., Nano Energy 48 (2018) 217–226. doi: 10.1016/j.nanoen.2018.03.059

  72. [72]

      S. Chen, S. Perathoner, C. Ampelli, et al., ACS Sustain. Chem. Eng. 5 (2017) 7393–7400. doi: 10.1021/acssuschemeng.7b01742

  73. [73]

      W. Guo, Z. Liang, J. Zhao, et al., Small Methods 2 (2018) 1800204. doi: 10.1002/smtd.201800204

  74. [74]

      H. Ma, Z. Chen, Z. Wang, Nanoscale 13 (2021) 1717–1722. doi: 10.1039/d0nr07362h

  75. [75]

      M. Nazemi, M.A. El-Sayed, J. Phys. Chem. Lett. 9 (2018) 5160–5166. doi: 10.1021/acs.jpclett.8b02188

  76. [76]

      J. Wang, L. Yu, L. Hu, et al., Nat. Commun. 9 (2018) 1795. doi: 10.1038/s41467-018-04213-9

  77. [77]

      S. Mukherjee, X. Yang, W. Shan, et al., Small Methods 4 (2020) 1900821. doi: 10.1002/smtd.201900821

  78. [78]

      H. Zhu, X. Ren, X. Yang, et al., SusMat 2 (2022) 214–242. doi: 10.1002/sus2.70

  79. [79]

      M. Wang, S. Liu, T. Qian, et al., Nat. Commun. 10 (2019) 341. doi: 10.1038/s41467-018-08120-x

  80. [80]

      F. Lü, S. Zhao, R. Guo, et al., Nano Energy 61 (2019) 420–427. doi: 10.1016/j.nanoen.2019.04.092

  81. [81]

      L. Hu, A. Khaniya, J. Wang, et al., ACS Catal. 8 (2018) 9312–9319. doi: 10.1021/acscatal.8b02585

  82. [82]

      L. Zeng, X. Li, S. Chen, et al., J. Mater. Chem. A 8 (2020) 7339–7349. doi: 10.1039/c9ta13336d

  83. [83]

      X. Yi, X. He, F. Yin, et al., J. Mater. Sci. 55 (2020) 12041–12052. doi: 10.1007/s10853-020-04777-2

  84. [84]

      Y. Yan, H. Qu, X. Zheng, et al., Chin. Chem. Lett. 33 (2022) 4655–4658. doi: 10.1016/j.cclet.2021.12.054

  85. [85]

      X. Zheng, Y. Liu, Y. Yan, X. Li, Y. Yao, Chin. Chem. Lett. 33 (2022) 1455–1458. doi: 10.1016/j.cclet.2021.08.102

  86. [86]

      H. Chen, J. Liang, K. Dong, et al., Inorg. Chem. Front. 9 (2022) 1514–1519. doi: 10.1039/d2qi00140c

  87. [87]

      Z. Wang, K. Zheng, S. Liu, et al., ACS Sustain. Chem. Eng. 7 (2019) 11754–11759. doi: 10.1021/acssuschemeng.9b01991

  88. [88]

      X. Li, L. Li, X. Ren, et al., Ind. Eng. Chem. Res. 57 (2018) 16622–16627. doi: 10.1021/acs.iecr.8b04045

  89. [89]

      Y. Song, D. Johnson, R. Peng, et al., Sci. Adv. 4 (2018) e1700336. doi: 10.1126/sciadv.1700336

  90. [90]

      Q. Zhang, B. Liu, L. Yu, Y. Bei, B. Tang, ChemCatChem 12 (2020) 334–341. doi: 10.1002/cctc.201901519

  91. [91]

      H. Cheng, L.X. Ding, G.F. Chen, et al., Adv. Mater. 30 (2018) 1803694. doi: 10.1002/adma.201803694

  92. [92]

      J. Han, X. Ji, X. Ren, et al., J. Mater. Chem. A 6 (2018) 12974–12977. doi: 10.1039/C8TA03974G

  93. [93]

      X. Zhang, R.M. Kong, H. Du, L. Xia, F. Qu, Chem. Commun. 54 (2018) 5323–5325. doi: 10.1039/c8cc00459e

  94. [94]

      Q. Liu, Y. Lin, S. Gu, et al., Nano Res. 15 (2022) 7134–7138. doi: 10.1007/s12274-022-4568-z

  95. [95]

      H.J. Chen, Z.Q. Xu, S. Sun, et al., Inorg. Chem. Front. 9 (2022) 4608–4613. doi: 10.1039/d2qi01173e

  96. [96]

      H. Chen, J. Liang, L. Li, et al., ACS Appl. Mater. Interfaces 13 (2021) 41715–41722. doi: 10.1021/acsami.1c11872

  97. [97]

      C. Lv, Y. Qian, C. Yan, et al., Angew. Chem. 130 (2018) 10403–10407. doi: 10.1002/ange.201806386

  98. [98]

      X. Yu, P. Han, Z. Wei, et al., Joule 2 (2018) 1610–1622. doi: 10.1016/j.joule.2018.06.007

  99. [99]

      Y.Q. Liu, L. Huang, Y. Fang, et al., ACS Appl. Mater. Interfaces 14 (2022) 12304–12313. doi: 10.1021/acsami.1c25240

  100. [100]

       W. Ye, H. Wang, J. Ning, Y. Zhong, Y. Hu, J. Energy Chem. 57 (2021) 219–232. doi: 10.1016/j.jechem.2020.09.016

  101. [101]

       J. Thompson, G. Morrison, Anal. Chem. 23 (1951) 1153–1157. doi: 10.1021/ac60056a029

  102. [102]

       T. Ali, H. Wang, W. Iqbal, et al., Adv. Sci. 10 (2023) 2205077. doi: 10.1002/advs.202205077

  103. [103]

       S.H. Yuen, A.G. Pollard, J. Sci. Food Agric. 3 (1952) 441–447. doi: 10.1002/jsfa.2740031002

  104. [104]

       L. Zhou, C.E. Boyd, Aquaculture 450 (2016) 187–193. doi: 10.1016/j.aquaculture.2015.07.022

  105. [105]

       M. Wang, S. Liu, H. Ji, et al., Nat. Commun. 12 (2021) 3198. doi: 10.1038/s41467-021-23360-0

  106. [106]

       P. Shen, X. Li, Y. Luo, et al., ACS Nano 16 (2022) 7915–7925. doi: 10.1021/acsnano.2c00596

  107. [107]

       Y. Ren, C. Yu, X. Tan, et al., Energy Environ. Sci. 15 (2022) 2776–2805. doi: 10.1039/d2ee00358a

  108. [108]

       Y. Ren, C. Yu, X. Song, et al., J. Mater. Chem. A 9 (2021) 13036–13043. doi: 10.1039/d1ta02681j

  109. [109]

       C.M. Gabardo, A. Seifitokaldani, J.P. Edwards, et al., Energy Environ. Sci. 11 (2018) 2531–2539. doi: 10.1039/c8ee01684d

  110. [110]

       B.H. Suryanto, C.S. Kang, D. Wang, et al., ACS Energy Lett. 3 (2018) 1219–1224. doi: 10.1021/acsenergylett.8b00487

  111. [111]

       F. Zhou, L.M. Azofra, M. Ali, et al., Energy Environ. Sci. 10 (2017) 2516–2520. doi: 10.1039/C7EE02716H

  112. [112]

       F. Köleli, T. Röpke, App. Catal. B: Environ. 62 (2006) 306–310. doi: 10.1016/j.apcatb.2005.08.006

  113. [113]

       K. Kim, N. Lee, C.Y. Yoo, et al., J. Electrochem. Soc. 163 (2016) F610. doi: 10.1149/2.0231607jes

  114. [114]

       K. Kim, C.Y. Yoo, J.N. Kim, H.C. Yoon, J.I. Han, J. Electrochem. Soc. 163 (2016) F1523. doi: 10.1149/2.0741614jes

  115. [115]

       Y. Ren, C. Yu, X. Han, et al., ACS Energy Lett. 6 (2021) 3844–3850. doi: 10.1021/acsenergylett.1c01893

  116. [116]

       H.K. Lee, C.S.L. Koh, Y.H. Lee, et al., Sci. Adv. 4 (2018) eaar3208. doi: 10.1126/sciadv.aar3208

  117. [117]

       A. Gennaro, A.A. Isse, M.G. Severin, et al., J. Chem. Soc. Faraday Trans. 92 (1996) 3963–3968. doi: 10.1039/FT9969203963

  118. [118]

       S. Malkhandi, B. Yang, A.K. Manohar, G.K.S. Prakash, S.R. Narayanan, J. Am. Chem. Soc. 135 (2013) 347–353. doi: 10.1021/ja3095119

  119. [119]

       M.N. Jackson, Y. Surendranath, J. Am. Chem. Soc. 138 (2016) 3228–3234. doi: 10.1021/jacs.6b00167

  120. [120]

       A. Gorodyskii, V. Danilin, O. Efimov, N. Nechaeva, V. Tsarev, React. Kinet. Catal. Lett. 11 (1979) 337–342. doi: 10.1007/BF02079722

  121. [121]

       A. Tsuneto, A. Kudo, T. Sakata, Chem. Lett. 22 (1993) 851–854. doi: 10.1246/cl.1993.851

  122. [122]

       J.L. Ma, D. Bao, M.M. Shi, J.M. Yan, X.B. Zhang, Chem 2 (2017) 525–532. doi: 10.1016/j.chempr.2017.03.016

  123. [123]

       S. Stevanovic, M.F. Costa Gomes, J. Chem. Thermodyn. 59 (2013) 65–71. doi: 10.1016/j.jct.2012.11.010

  124. [124]

       H.L. Du, M. Chatti, R.Y. Hodgetts, et al., Nature 609 (2022) 722–727. doi: 10.1038/s41586-022-05108-y

• 

  Figure 1  Schematic diagram for the recent progress on the electrolytes for ENRR toward NH₃.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Schematic diagram of ENRR to NH3 over catalyst via associative and dissociative pathways. Reproduced with permission [21]. Copyright 2019, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Schematic illustration of the competitive adsorption and surface coverage on the catalyst during the ENRR and HER process (blue, N₂; purple, proton; gray, catalyst). (b) Illustration of the limiting proton transfer rate by decreasing bulk solution proton concentration and increasing the surface proton transfer barrier. Reproduced with permission [36]. Copyright 2017, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Current density profiles in different electrolytes. (b) Average NH₃ formation rates and FEs in different electrolytes. Reproduced with permission [47]. Copyright 2017, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Schematic procedure, (b) HIM image, (c) TEM image, and (d) HRTEM image of CoP HNC. Reproduced with permission [73]. Copyright 2018, Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Geometric models and (b) TEM image of Au nanorods. (c) Average yield of NH₃ and hydrazine hydrate and FEs of Au nanorods in 0.1 mol/L KOH. Reproduced with permission [70]. Copyright 2016, Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  (a) NH₃ yield rates and FEs in different electrolytes. Reproduced with permission [76]. Copyright 2018, Springer Nature. (b) NH₃ yields of Ni-Fe-MoS₂ and MoS₂. Reproduced with permission [82]. Copyright 2020, Royal Society of Chemistry. (c) Chronoamperometric curves of NH₂-MIL-88B-Fe in 0.1 mol/L Na₂SO₄ and 0.05 mol/L H₂SO₄ measured at −0.45 V. (d) NH₃ yield rates and FEs in 0.1 mol/L Na₂SO₄ and 0.05 mol/L H₂SO₄ electrolytes. Reproduced with permission [83]. Copyright 2020, Springer Nature.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  (a) schematic illustration for preparing p-Fe₂O₃/CC catalyst. (b) NH₃ production rates and FEs of p-Fe₂O₃/CC catalyst. (c) NH₃ production rates and FEs of p-Fe₂O₃/CC, Fe₂O₃–4/CC, Fe₂O₃–8/CC, and CC catalysts. Reproduced with permission [87]. Copyright 2019, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 9  (a) NH₃ production rates and FEs for TiO₂-rGO/CP in 0.1 mol/L Na₂SO₄. (b) NH₃ production rates using various electrodes at −0.90 V. (c) NH₃ production rates and FEs at −0.90 V during recycling tests. Reproduced with permission [50]. Copyright 2018, Royal Society of Chemistry.

  下载: 全尺寸图片 幻灯片
  

  Figure 10  (a) An illustration of a schematic for synthesizing Mo₂C/C using the molten salt approach. (b) FEs and (c) NH₃ yield rates over Mo₂C/C on hydrophilic CC and hydrophobic CC with pH of 2 and 3 in 0.5 mol/L Li₂SO₄. Reproduced with permission [91]. Copyright 2018, Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 11  (a) Schematic illustration for prrparing BVC-A and BVC—C catalysts. (b) Electrochemical performances of Bi₄V₂O₁₁, CeO₂, BVC-A, and BVC—C at −0.2 V vs. RHE. (c) NH₃ formation rates and FEs of BVC-A in 0.1 mol/L HCl. Reproduced with permission [60]. Copyright 2018, Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 12  (a) NH₃ synthesis rates achieved using several catalysts at −0.2 V vs. RHE. (b) NH₃ synthesis rate and FEs of the Au/CeO_x-RGO composite in 0.1 mol/L HCl. Reproduced with permission [66]. Copyright 2017, Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 13  (a) NH₃ formation rates and (b) FEs of PCN in 0.1 mol/L HCl which is N₂-saturated. (c) The charge density difference of the N₂-adsorbed PCN with N-vacancies and the geometry of N₂ adsorption on PCN with N-vacancies. Reproduced with permission [97]. Copyright 2018 Wiley-VCH.

  下载: 全尺寸图片 幻灯片
  

  Figure 14  (a) Schematic illustration for preparing boron-doped graphene. (b) TEM image, (c) NH₃ yields and FEs of boron-doped graphene. Reproduced with permission [98]. Copyright 2018, Elsevier.

  下载: 全尺寸图片 幻灯片
  

  Figure 15  (a) UV–vis spectra of NH₃ production rate after electrolysis for 2 h at −0.35 V vs. RHE. (b) Comparison of NH₃ yield rates and FEs on pure Rh, SnO₂, Rh_0.3@SnO₂, Rh_0.5@SnO₂, Rh_1.3@SnO₂, and mechanically mixed Rh_0.5/SnO₂ catalysts at −0.35 V vs. RHE. (c) NH₃ yield rates and FEs on Rh_0.5@SnO₂ at −0.35 V vs. RHE after consecutive tests. (d) Chronoamperometry curve for 20 h on Rh_0.5@SnO₂. Reproduced with permission [99]. Copyright 2022, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 16  (a) Schematic of Fe-SS electrode based N₂ reduction process. (b) FEs for electro-reduction of N₂-saturated ILs on different electrodes at −0.8 V vs. NHE. (c) NH₃ production rates on several electrodes and ILs at −0.8 V vs. NHE. Reproduced with permission [111]. Copyright 2017, Royal Society of Chemistry. (d) FEs and the NH₃ yields at X_IL = 0.16. (e) NH₃ yields and FEs at −0.65 V vs. NHE. Reproduced with permission [110]. Copyright 2018, American Chemical Society. (f) Corresponding FEs and NH₃ yield rates. After chronoamperometry, the electrode was either coated with visible degradation products or was not, as indicated by the pink and light-green shading, respectively. Reproduced with permission [124]. Copyright 2022, Nature.

  下载: 全尺寸图片 幻灯片
  

  Figure 17  (a) Effect of the electrolyte medium on NH₃ synthesis rate and FE. Reproduced with permission [113]. Copyright 2016, Electrochemical Society. (b) A schematic representation of ethylenediamine (EDA)-based NH₃ synthesis. Reproduced with permission [114]. Copyright 2016, Electrochemical Society. (c, d) Comparison of electrolyte counterion effect of Li⁺, Na⁺, and K⁺ on NH₃ synthesis rate and FE. Reproduced with permission [89]. Copyright 2018, Science.

  下载: 全尺寸图片 幻灯片
  

  Figure 18  Perspectives for enhancing ENRR.

  下载: 全尺寸图片 幻灯片

  Table 1.  Comparison of ENRR performances in various electrolytes.

  下载: 导出CSV

  Table 2.  Compositions of non-aqueous electrolytes and their ENRR performances.

  下载: 导出CSV
•