MoS₂‐Based Catalysts for N₂ Electroreduction to NH₃ – An Overview of MoS₂ Optimization Strategies

Abstract

The nitrogen reduction reaction (NRR) has become an ideal alternative to the Haber‐Bosch process, as NRR possesses, among others, the advantage of operating under ambient conditions and saving energy consumption. The key to efficient NRR is to find a suitable electrocatalyst, which helps to break the strong N≡N bond and improves the reaction selectivity. Molybdenum disulfide (MoS₂) as an emerging layered two‐dimensional material has attracted a mass of attention in various fields. In this minireview, we summarize the optimization strategies of MoS₂‐based catalysts which have been developed to improve the weak NRR activity of primitive MoS₂. Some theoretical predictions have also been summarized, which can provide direction for optimizing NRR activity of future MoS₂‐based materials. Finally, an outlook about the optimization of MoS₂‐based catalysts used in electrochemical N₂ fixation are given.

Optimization strategies for the application of MoS2‐ based electrocatalysts in the Nitrogen Reduction Reaction (NRR) are summarized. They include enhancing electrical conductivity, enlarging the specific surface area, taking advantage of the interaction effect between other elements and MoS2 and inhibiting competitive reactions. Some enhancement strategies based on theoretical calculations and predictions have also been reviewed, which can provide researchers with ideas for enhancing the NRR activity of MoS2‐based catalyst materials.

1. Introduction

Ammonia (NH₃) not only plays a significant role in industry, agriculture and other industries related to daily work,[ ¹ , ² , ³ , ⁴ ] but also is a potential hydrogen storage material. ^[5] In industrial production, the Haber‐Bosch process is used forNH₃ synthesis. But the Haber‐Bosch process has extremely tough requirements, including high temperature (400–500 °C) and pressure (200–250 bar). The process is also accompanied by massive CO₂ emission.[ ⁶ , ⁷ , ⁸ , ⁹ ] Therefore, it is urgent to explore a NH₃ synthesis method that has less impact on the environment. Recently, the nitrogen reduction reaction (NRR) has attracted significant attention for artificial N₂ fixation because the NRR process can be performed at ambient conditions and save energy. Hence, it is regarded as an ideal alternative to the Haber‐Bosch process.[ ¹⁰ , ¹¹ , ¹² ]

However, the N≡N bond is hard to break (bond energy about 941 kJ mol⁻¹) and the competing hydrogen evolution reaction (HER) has a huge impact. ^[13] It is thus necessary to identify suitable catalysts to improve the efficiency of ambient NH₃ synthesis. Noble metal catalysts show excellent activity for NRR due to their favorable conductivity and strong chemical binding with reactants.[ ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ ] But the use of this type of catalyst has been vastly limited by its scarcity. Consequently, numbers of non‐noble catalysts have been researched.[ ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ , ⁴⁶ ] In non‐noble catalysts, two‐dimensional materials have been widely explored in the NRR field due to the large surface area and novel electronic properties. MoS₂, as an emerging layered two‐dimensional material, can conduct electricity like graphene. Compared to other two‐dimensional materials, MoS₂ not only has the advantages of adjustable electronic structure, optimal N₂ adsorption energy and good stability in liquid media, but also has the potential of large‐scale production as the preparation process is simple. Therefore, it has been largely used in the electrocatalysis field.[ ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ , ⁵¹ ]

Recently, Sun and co‐workers reported bulk MoS₂ nanosheets for electrochemical N₂ fixation. ^[52] The nanosheet structure of MoS₂ is markedly displayed in transmission electron microscopy (TEM) images (Figure 1A). In this work, they firstly performed density function theory (DFT) calculations to verify the feasibility of MoS₂ for the NRR process. The calculation results showed that the energy barrier of the potential‐determining step (PDS) was 0.68 eV (Figure 1B), which demonstrated MoS₂ was a potential NRR catalyst. The edge of MoS₂ was attested to be the N₂ adsorption site on the basis of Bard charge analysis (a large number of positive charges gather at the edge of MoS₂, as shown in Figure 1C). According to electrochemical experiment, the MoS₂ nanosheet could achieve an NH₃ yield of 8.08×10⁻¹¹ mol s⁻¹ cm⁻¹ in 0.1 m Na₂SO₄, but the Faraday efficiency (FE) was only 1.17 % (Figure 1D), mainly limited by the HER process. Therefore, MoS₂ is indeed a promising and potential NRR electrocatalyst. How to optimize MoS₂ to increase its NRR activity has become a hot topic among researchers.

Figure 1.

(A) The transmission electron microscope image of MoS₂ nanosheets. (B) Free energy profile proposed by DFT calculations. The asterisk (*) denotes the N₂ adsorption site. (C) The view from atop the isosurface of deformation charge density. Red and green represent charge accumulation and drop at the edge of Mo atoms, respectively. Isosurface is 0.0025 a.u. (D) The NH₃ yields and FE of MoS₂/CC (carbon cloth) at various potentials. Reproduced with permission from Ref. [52]. Copyright 2018 Wiley‐VCH.

Herein, we summarize the optimization strategies of MoS₂ since it was first used as a NRR catalyst, including enhancing electrical conductivity/enlarging specific surface area, taking advantage of the interaction effect between other elements and MoS₂ and inhibiting competitive reaction. We also summarize recent theoretical calculations on and predictions of MoS₂‐based catalysts. In addition, some advice on how to improve the NRR progress activity of MoS₂‐based catalysts is finally presented.

2. The Mechanism of NH₃ Synthesis

The NRR process can be simply divided into the following process: (1) N₂ is adsorbed on the active site of the catalyst surface; (2) to the N atom, hydrogen is constantly added; (3) a NH₃ molecule is released. The mechanism of the NH₃ synthesis process can involve dissociative (Figure 2A) or associative mechanisms (Figures 2B and C). During the pathway of the dissociative mechanism, the N≡N is fractured in advance, then each individual N atom gets hydrogenated. The Haber‐Bosch process is thought to follow the dissociative pathway. The associative mechanism can be differentiated in the alternating (Figure 2B) and the distal pathway (Figure 2C) differing in how the hydrogenation proceeds. For the alternating pathway, a N atom connects with the surface of the catalyst; followingly, two N atoms are hydrogenated separately until two NH₃ molecules are released. For the distal pathway, the N atom not bound to the catalyst is firstly hydrogenated to release an NH₃ molecule, then the other N atom is hydrogenated and the second NH₃ molecule is released. The adsorption modes of the alternating pathway and the distal pathway are called end‐on adsorption.

Figure 2.

The NRR mechanisms on the surface of catalyst: (A) dissociative pathway; (B) alternating pathway; (C) distal pathway; (D) side‐on adsorption. Blue, white represent N, H atoms, respectively.

Another mode of adsorption is called side‐on adsorption (Figure 2D). In the side‐on adsorption pathway, two N atoms are both bound to the catalyst surface and alternatingly hydrogenated to produce two NH₃ molecules.

3. Strategies for Enhancing the NRR Activity of MoS₂

3.1. Enhancing Electrical Conductivity/Enlarging Specific Surface Area

It has been demonstrated that the edges of MoS₂, that is, free Mo atoms, are the N₂ adsorption sites of MoS₂; thus, creating more active sites to increase N₂ adsorption is beneficial to the improvement of the NRR activity. After MoS₂ had first been used in the electrocatalytic NH₃ synthesis process, ^[52] Sun and co‐workers further used defect‐rich MoS₂ nanoflowers as a catalyst for the NRR process. ^[53] Thanks to the defect structure, the electronic structure of the MoS₂ nanoflowers enhanced the activity of the reaction sites, which was more advantageous to the absorption of N₂. The MoS₂ nanoflowers could achieve a high FE of 20.48 % with an NH₃ yield of 29.28 μg h⁻¹ mg_cat. ⁻¹ in 0.1 m Na₂SO₄. Liao et al. reported ultra‐thin MoS₂ nanosheets with high specific surface area as an NRR catalyst. ^[54] The large specific surface area leads to an increase in active sites. Therefore, compared with bulk MoS₂, the NH₃ yield of the ultrathin MoS₂ nanosheets achieved a significant growth (41.66 μg h⁻¹ mg_cat. ⁻¹), but the FE (1.10 %) showed only little improvement due to the strong competing influence of the HER process. Chen et al. employed high temperature annealing to get porous atomic layered MoS₂. ^[55] Benefiting from the formation of pores, more N₂ adsorption sites on the inert basal plane were exposed. The reasonable use of the inert basal plane of MoS₂ provided a new concept for optimizing MoS₂‐base catalysts for NRR. At the same time, the multilayer structure allowed most of the N₂ adsorption sites to take part in the reaction. Thus, an NH₃ yield of 3405.55 μg h⁻¹ mg_cat. ⁻¹ and a FE of 44.36 % were achieved.

In addition, it is well known that MoS₂ is a semiconductor material, ^[56] so taking measures to increase MoS₂ electrical conductivity is also an option to optimize the NRR activity of MoS₂‐based catalysts. Loading MoS₂ on a suitable substrate is one of the measures to tackle the lack of N₂ adsorption sites and the poor electrical conductivity. As reduced graphene oxide (rGO) has the advantages of high specific surface area, favorable electroconductibility and high stability, rGO has been widely used in electrocatalysis in recent years. ^[57] In our recent work, we loaded MoS₂ nanosheets onto rGO to improve the specific surface area and electrical conductivity of MoS₂. ^[58] Having the reaction taking place in 0.1 m LiClO₄ furthermore had the effect of inhibiting the HER (the mechanism will be described specifically in strategy 3.3). Under the synergistic effect of the above strategies, we obtained an NH₃ yield of 24.82 μg h⁻¹ mg_cat. ⁻¹ and a FE of 4.56 %. As MoS₂ nanodots has the advantages of more N₂ adsorption sites and larger specific surface area compared to nanosheet structures, Zhang and co‐workers loaded MoS₂ nanodots uniformly on rGO. ^[59] By means of X‐ray photoelectron spectroscopy (XPS), the MoS₂ nanodots/rGO material was demonstrated to contain strong C−S−C bonds which effectively promoted electron transfer. Thus, MoS₂ nanodots/rGO exhibited a FE of 27.93 % at −0.35 V versus the reversible hydrogen electrode (RHE) potential with an NH₃ yield of 16.41 μg h⁻¹ mg_cat. ⁻¹ at −0.75 V (vs. RHE).

Through a pyrolysis strategy, Zhao et al. generated MoS₂ grown, in situ, on a C₃N₄ layer, which led to an NH₃ yield of 19.86 μg h⁻¹ mg_cat. ⁻¹ with a FE of 6.87 %. ^[60] The reason for the high catalytic activity is the strong interaction between C₃N₄ and MoS₂, formed by interfacial Mo−N coordination to promote electron transmission. In addition, the source of N₂ was also determined in electrochemical experiments. They demonstrated that C₃N₄ had no catalytic capacity for NRR, so the N source for electrocatalytic NH₃ synthesis only came from the N₂ in the air. Shao and co‐workers immobilized 1T‐MoS₂ on g‐C₃N₄ (1T‐MoS₂/g‐C₃N₄) to boost NRR activity. ^[61] The 1T phase (comprised of edge‐sharing [MoS₆] octahedra) had the advantage of more N₂ adsorption sites along the plane and higher electron conductivity than other phases of MoS₂. ^[62] Therefore, an appropriate load of 1T‐MoS₂ was conducive to the adsorption and activation of N₂. 1T‐MoS₂/g‐C₃N₄ achieved a high catalytic efficiency of NRR (29.97 μg h⁻¹ mg_cat. ⁻¹, 20.48 %).

Mao et al. first prepared n‐butyl triethyl ammonium bromide functionalized polypyrrole/graphene oxide (BTAB/PPy/GO) as a precursor, ^[63] in which the substrate polypyrrole/graphene oxide was used to enhance electroconductibility and to expand the specific surface area of the catalyst. Interestingly, BTAB/PPy/GO itself showed little NRR activity (Figure 3A), but it did affect the morphology and phase of 1T‐MoS₂. 1T‐MoS₂/BTAB/PPy/GO (schematic synthesis is shown in Figure 3C) thus could attain a high NRR activity (13.60 μg h⁻¹ mg_cat.⁻¹, 1.96 %). However, after long‐term use, the metastable 1T‐MoS₂ was transformed into Mo₂N due to the electrochemical reaction with N₂. Having formed new Mo−N bonds, this prevented the catalyst from binding N₂. Figure 3B shows the concentration of NH₄ ⁺ versus operation time; the concentration of NH₄ ⁺ stayed steady after 18 h of reaction, which indicated that the active sites of 1T‐MoS₂/BTAB/PPy/GO had been deactivated.

Figure 3.

(A) The NH₃ yield and FE of different catalysts at −0.49 V after electrolysis (2 h): (a) carbon cloth (CC), (b) graphene oxide (GO), (c) PPy/GO (PPy=polypyrrole), (d) BTAB/PPy/GO (BTAB=n‐butyl triethyl ammonium bromide), (e) MoS₂, (f) MoS₂/GO, (g) MoS₂/PPy/GO and (h) 1T‐MoS₂/BTAB/PPy/GO. (B) Time‐dependent concentration of NH₄ ⁺ over 30 h in 0.1 m KOH. (C) Schematic synthesis of 1T‐MoS₂/BTAB/PPy/GO. Reproduced with permission from Ref. [63]. Copyright 2020 American Chemical Society.

Xu et al. anchored 1T‐MoS₂ on Ti₃C₂ MXene through a hydrothermal reaction. ^[64] Because the 1T‐MoS₂ could be fully loaded on the Ti₃C₂ MXene, in turn exposing more N₂ adsorption sites and the composite had high electrical conductivity, the 1T‐MoS₂/Ti₃C₂ MXene achieved an NH₃ yield of 30.33 μg h⁻¹ mg_cat. ⁻¹ and a FE of 10.94 % in 0.1 m HCl.

3.2. Taking Advantage of the Interaction Effect Between Other Elements and MoS₂

As the NRR performance of MoS₂ is not satisfactory, taking advantage of the interaction effect between other elements and MoS₂ to enhance the NRR activity of MoS₂ has become a widely studied strategy. The interaction effect, including synergy effects and the addition of other elements, can change the electronic structure of MoS₂; thus, heteroatom doping has become the priority approach in this strategy. Zeng et al. achieved a high NH₃ yield of 128.17 μg h⁻¹ mg_cat. ⁻¹ and a FE of 11.34 % by doping bimetallic Ni‐Fe in MoS₂. ^[65] The reason for the high activity was that the bimetallic dopant reacted with the S atoms to generate a nanohollow structure (Figure 4A), affording more N₂ adsorption sites because of the synergistic effect of the S‐Fe‐Ni system. As shown in Figure 4B, Zhao et al. doped a carefully balanced amount of Fe nanodots into MoS₂, using carbon cloth as a substrate, so that the Fe atoms would not aggregate and degrade during the NRR process. ^[66] The synergistic effect (Fe nanodots can regulate the chemical state of MoS₂) made Fe‐MoS₂/CC afford an NH₃ yield of 12.5 μg h⁻¹ cm⁻² with a FE of 10.8 %.

Figure 4.

(A) Scanning Electron Microscope (SEM) image of Ni‐Fe@MoS₂ nanocubes. Reproduced with permission from Ref. [65]. Copyright 2020 Royal Society of Chemistry. (B) Illustration of the synthesis of Fe‐MoS₂/CC. Reproduced with permission from Ref. [66]. Copyright 2019 Royal Society of Chemistry. (C) Structure diagrams of 2H‐MoS₂ and Fe‐Mo‐S materials. Reproduced with permission from Ref. [67]. Copyright 2020 Elsevier. (D) Schematic synthesis of MoS₂ and AuNPs@MoS₂ nanosheets. Modified with permission from Ref. [68]. Copyright 2019 Elsevier.

Most of the current MoS₂ synthesis processes involve ammonium ions, which leads to unclear N sources in the NRR process. Partially replacing S atoms in 2H‐MoS₂ with Fe atoms (Figure 4C), Guo et al. used an ammonium‐free hydrothermal process to grow Fe@2H‐MoS₂ on carbon cloth, ^[67] thus successfully avoiding the problem of unclear N sources. The substituted Fe atoms affected the electronic state of Mo and S atoms, which facilitated the adsorption of N₂. Therefore, compared with bulk MoS₂, the NH₃ production rate of Fe@2H‐MoS₂ was increased by 10 times and the FE was increased by 5 times.

Since MoS₂ is a semiconductor material, which hinders the process of electron transfer, doping with a conductive metal is a potential strategy of resolving this. In addition to iron‐doping, other elements have also been used to produce outstanding results with this strategy. Recently, Zhang et al. have grown ultra‐small Au nanoparticles (NPs) on MoS₂ nanosheets to enhance the electron transfer ability of MoS₂ (Figure 4D). ^[68] They found that the Au on the surface of MoS₂ caused unsatisfied valences, thus adjusting the electronic structure of MoS₂s so as to enhance the absorption of N₂. This, in turn led to an increased NRR activity with an NH₃ yield of 25 μg h⁻¹ mg_cat. ⁻¹ and a FE of 9.7 %. Suryanto et al. modified 2H‐MoS₂ with Ru clusters, producing an electrocatalytic material for the NRR process which achieved an NH₃ yield rate of 1.14×10⁻¹⁰ mol cm⁻² s⁻¹ with a FE of 17.6 % (50 °C). ^[69] The high NRR activity was ascribed to the synergistic interaction between the isolated sulfur vacancies on MoS₂ (the hydrogenation centers) and the Ru cluster (the sites of N₂ binding).

In addition, MoS₂ has also been doped with some non‐metals that have also been proven to promote electron transfer. Yang et al. synthesized C/MoS₂ porous nanospheres by a simple one‐pot hydrothermal method. ^[70] Benefitting from the synergistic effect between C and MoS₂, the electron transfer during the N₂ electroreduction to NH₃ was significantly improved. The C/MoS₂ material thus achieved a FE of 8.2 %. Zeng et al. reported an impressive NH₃ yield of 69.82 μg h⁻¹ mg_cat. ⁻¹ and a high FE (9.14 %) by doping N atoms (by using ammonium fluoride during the synthesis) into MoS₂ nanoflower structures. ^[71] The N atoms altered the electronic structure of MoS₂ and produced abundant S vacancies, which promoted electron transfer.

Due to the appropriate orbital energy and symmetry, the unoccupied and occupied d orbitals of the transition mental (TM) can separately receive electrons from and feed electrons back to the N₂ antibonding orbitals, respectively, activating the N−N bond. Based on this mechanism, Guo et al. proposed a concept of the strong‐weak electron polarization (SWEP) pair, ^[72] consisting of two catalytic centers with vastly different electron acceptance and back‐feeder abilities. As the d orbitals of transition metals and the p orbitals of boron show similar behavior, SWEP pairs were created by filling the S vacancy in MoS₂ with B atom (adjusting the electronic structure of MoS₂). The SWEP pair in B−MoS₂/CFC (CFC=carbon fiber cloth) polarized the non‐polar N−N bond (Figure 5A), thus promoting the breaking of the first bond of N≡N. B−MoS₂/CFC thus achieved an NH₃ yield of 44.09 μg h⁻¹ mg_cat. ⁻¹ with a FE of 21.72 % (Figure 5B).

Figure 5.

(A) The synergistic polarization of N≡N bond by boron/molybdenum hybrid catalysts in the step of dinitrogen adsorption: strong‐weak electron polarization (SWEP). (B) NH₃ yields and FE of B−MoS₂ under different potentials. Reproduced with permission from Ref. [72]. Copyright 2020 Elsevier. (C) Schematic depiction of Co‐doped defected MoS_2‐x with adsorbed N₂. The N₂ molecule (light blue spheres) binds to three Mo atoms (dark blue spheres) near the S (yellow spheres) vacancy position. With Co (dark red sphere) replacing Mo in the catalyst, N₂ binds only one of the remaining Mo atoms near the S vacancy. (D) Free energy profile of NH₃ synthesis at defected MoS_2‐x (red) and Co‐doped MoS_2‐x (blue) basal planes. Reproduced with permission from Ref. [75]. Copyright 2019 American Chemical Society.

Besides heteroatom doping, compound loading is also one of the strategies to optimize the NRR activity of MoS₂. Guo et al. reported FeS@MoS₂ under the action of independent conductive substrate carbon fiber cloth (CFC) as an NRR catalyst. ^[73] The high NRR activity (NH₃ yield of 8.45 μg h⁻¹ cm⁻²) of FeS@MoS₂/CFC was ascribed to the synergistic effect arising from FeS nanoparticles providing a large number of N₂ adsorption sites. Yang et al. loaded CoS₂ nanoparticles onto MoS₂ nanosheets as a heterostructured catalyst. ^[74] The strong interaction between CoS₂ and MoS₂ can adjust the interface charge distribution, which effectively promoted the adsorption of N₂. Thus CoS₂/MoS₂ attained an NH₃ yield of 54.7 μg h⁻¹ mg⁻¹ and a FE of 20.8 %.

Inspired by the role of dinitrogenase in biological NH₃ synthesis, Zhang et al. simulated the active site of nitrogenase by introducing Co atoms to the S vacancy of the MoS₂ basal plane(Figure 5C). ^[75] The S vacancy could make Mo atoms below the basal plane manifest to generate more N₂ adsorption sites. The energy barrier of the rate‐limiting step (Figure 5D), according to DFT calculations, changed from −1.62 eV (MoS₂) to −0.59 eV (Co−MoS_2‐x), which further proved that Co‐doping could enhance the activity of the NRR process. Thus the catalyst could achieve an NH₃ yield over 0.6 mmol h⁻¹ g⁻¹ and a FE over 10 %. Similar to the work of Zhang et al., ^[75] Zeng et al. also prepared Co−MoS₂ by making use of the positive interaction between Co and S. ^[76] The difference was that they loaded Co−MoS₂ onto a zeolitic imidazolate framework (ZIF) as a heterojunction catalyst, which could benefit from the inhibitory effect of ZIF on the HER process. They further designed a N₂ diffusion cathode because N₂ has a low solubility in the electrolyte. Through the bidirectional optimization of the electrocatalyst and the reaction environment, the Co−MoS₂ could obtain a satisfactory NRR activity in 0.1 m Na₂SO₄ (127.88 μg h⁻¹ mg_cat. ⁻¹, 11.29 %).

Zheng et al. embedded Fe atoms in single‐molecular layered MoS₂ (sMoS₂) which then contained a [Fe−S₂−Mo] motif similar to the core structure of the Fe−Mo−S cluster in nitrogenase. Benefitting from the interactions in the Fe−S₂−Mo motif, Fe−sMoS₂ could achieve an NH₃ yield of 24 μg cm⁻² h⁻¹ with a FE of 27 % in 0.5 m K₂SO₄. ^[77]

Due to the structure of MoS₂, some research has shown that heteroatom doping can also inhibit competing reactions of the NRR process; more detail will be given in the next section.

3.3. Inhibiting Competitive Reaction

As the reaction potentials of the hydrogen evolution reaction (HER) and NRR are similar to each other,[ ⁷⁸ , ⁷⁹ ] HER is the biggest competitive reaction in electrocatalytic N₂ reduction. In the original experiment by Sun et al., ^[52] it was the influence of the HER that caused the low FE of the pure MoS₂ nanosheets. Therefore, inhibition of the HER process became one of the most effective strategies to enhance the NRR activity. Liu et al. used the interaction between Li and S to inhibit the HER process. ^[80] Specifically, when S‐rich MoS₂ nanosheets were used as NRR catalyst in a 0.1 m Li₂SO₄ electrolyte, the adsorption free energy of H was reduced (0.03 eV→0.47 eV) due to the interaction between Li⁺ and the S edge site in MoS₂ (Figure 6A). Thus, the adsorption process of H^. was inhibited. At the same time, the strong action of Li positively charged the edge site of MoS₂, which enhanced adsorption of N₂ (the N₂ adsorption free energy was increased from −0.32 eV to −0.70 eV). Compared to conducting the experiment in a 0.1 m Na₂SO₄ electrolyte, the Li‐based electrolyte led to a high NRR activity of the S‐rich MoS₂ nanosheets (43.4 μg h⁻¹ mg_cat. ⁻¹, 9.81 %) (Figure 6B). Through electrochemical experiments, Liu et al. verified the stability of the catalyst (Figure 6C and D). In our recent work (mentioned in section 2.1), ^[58] the Li−S interaction was also used to suppress the HER process. Similarly, in a 0.25 m Li₂SO₄ electrolyte, Patil et al. grew 1T‐MoS₂ on nickel foil (1T‐MoS₂/NF) for use in the electrochemical N₂ reaction, ^[81] also using the strong Li−S interaction to inhibit the HER process. In addition, 1T‐MoS₂/NF featured pseudo‐six‐membered rings, in which the interaction between N and Li could enhance the adsorption capacity of N₂. This work provided researchers with a new idea to design N₂ fixation catalysts. 1T‐MoS₂/NF reached a FE of 27.66 % and an NH₃ yield of 1.05 μg min⁻¹ cm⁻².

Figure 6.

(A) The Li−S interactions as signified by the deformation charge density of *N₂ at a MoS₂ edge. Yellow and blue represent charge accumulation and loss, respectively. (B) NH₃ yield and FE of MoS₂/BCCF acquired for N₂‐saturated 0.1 m Li₂SO₄ and 0.1 m Na₂SO₄ electrolytes at −0.2 V (vs. RHE) after electrolysis for 2 h. (C) NH₃ yield and FE for a continuous reaction over 12 h. (D) NH₃ yield and FE during a reusability test comprising five cycles of 2 h each. Reproduced with permission from Ref. [80]. Copyright 2019 Wiley‐VCH. (E) A sketch map of strain engineering by replacing S on the edge site of the MoS₂ with F. Blue, yellow, green represent Mo, S, F ions, respectively. Reproduced with permission from Ref. [83]. Copyright 2020 Royal Society of Chemistry.

In this strategy, the above‐detailed reports used the action of electrolyte to inhibit the HER process. In addition, optimizing the structure of MoS₂ to inhibit the HER process has also been studied. Su et al. dispersed Fe atom on MoS₂ nanosheets to simulate nitrogenase‐like NH₃ synthesis with superior NRR activity (8.63 μg h⁻¹ mg_cat. ⁻¹, 18.8 %). ^[82] They modified the S edge position in MoS₂ with Fe atoms, which effectively inhibited the HER process by increasing the energy barrier of HER from 0.03 eV to 0.15 eV.

It is well known that MoS₂ possesses a layered two‐dimensional (2D) graphene‐like structure and the decrease of MoS₂ layer spacing is beneficial to the inhibition of the HER process. By substituting S with F atoms in defect‐rich MoS₂, Liang et al. reported F‐MoS₂ as electrocatalyst for the NRR process (Figure 6E). ^[83] Compared with sulfur, fluorine has a smaller size and higher electronegativity, leading to a decreased layer spacing of MoS₂ and thus reduced HER activity. In addition, this work centered on defect‐rich MoS₂ with marginal defects (larger specific surface area than bulk MoS₂), which further increased the number of active sites for N₂ fixation. Therefore, F‐MoS₂ could achieve an NH₃ yield of 35.7 μg h⁻¹ mg_cat. ⁻¹ and a FE of 20.6 %.

Duan et al. prepared an NRR catalyst by completely encapsulating ball‐like MoS₂ nanoflowers with a ZIF‐71 coating. ^[84] In this MoS₂@MOF interface, the ZIF‐71 coating did not only effectively concentrate N₂ through the inherent micropores, but also acted as a hydrophobic barrier to inhibit the HER process. At the same time, the ball‐like MoS₂ nanoflowers provided abundant active edge sites due to the unique ultrathin subunits. Thus, MoS₂@ZIF‐71 achieved an NH₃ yield of 56.69 μg h⁻¹ mg_MoS2 ⁻¹ and a FE of 30.91 %.

4. Theoretical Calculations and Predictions

Since Sun and co‐workers used bulk MoS₂ as an electrocatalyst in the N₂ fixation process, ^[52] a large number of experimental studies on MoS₂‐based catalysts have been conducted. In order to facilitate a comparison, performance parameters such as NH₃ production rate and FE were listed in Table 1. Some calculations have also been made to predict the activity of MoS₂‐based catalysts, which provided researchers with ideas for enhancing their catalytic activity. So far, calculations and predictions have mainly focused on the heteroatom doping strategies.

Table 1.

Summary of performance parameters of MoS₂‐based catalysts in the NRR process.

Catalyst	Ref.	Electrolyte (Concentration [m])	NH3 yield rate	FE [%]
MoS2	[52]	Na2SO4 (0.1)	8.08×10−11 mol s−1 cm−1	1.17
MoS2 nanoflowers	[53]	Na2SO4 (0.1)	29.28 μg h−1 mgcat. −1	20.8
Ultra‐thin MoS2 nanosheets	[54]	Na2SO4 (0.1)	41.66 μg h−1 mgcat. −1	1.10
PLA‐MoS2	[55]	HCl (0.1)	3405.55 μg h−1 mgcat. −1	44.36
MoS2/rGO	[58]	LiClO4 (0.1)	24.82 μg h−1 mgcat. −1	4.56
MoS2 nanodots/rGO	[59]	Na2SO4 (0.1)	16.41 μg h−1 mgcat. −1	27.93
MoS2/g‐C3N4	[60]	Na2SO4 (0.1)	19.86 μg h−1 mgcat. −1	6.87
1T‐MoS2/g‐C3N4	[61]	HCl (0.1)	29.97 μg h−1 mgcat. −1	20.48
1T‐MoS2/BTAB/PPy/GO	[63]	KOH (0.1)	13.60 μg h−1 mgcat. −1	1.96
1T‐MoS2@Ti3C2	[64]	HCl (0.1)	30.33 μg h−1 mgcat. −1	10.94
Ni‐Fe@MoS2	[65]	Na2SO4 (0.1)	128.17 μg h−1 mgcat. −1	11.34
Fe‐MoS2/CC	[66]	KOH (0.1)	12.5 μg h−1 cm−2	10.8
Fe@2H‐MoS2	[67]	HCl (0.1)	n.a.	n.a.
Au NPs@MoS2	[68]	Na2SO4 (0.1)	25 μg h−1 mgcat. −1	9.7
Ru/2H‐MoS2	[69]	HCl (0.1)	1.14×10−10 mol cm−2 s−1	17.6
C/MoS2	[70]	Li2SO4 (0.1)	n.a.	8.2
N@MoS2	[71]	Na2SO4 (0.1)	69.82 μg h−1 mgcat. −1	9.14
B−MoS2/CFC	[72]	HCl (0.1)	44.09 μg h−1 mgcat. −1	21.72
FeS@MoS2/CFC	[73]	Na2SO4 (0.1)	8.45 μg h−1 cm−2	2.96
CoS2/MoS2	[74]	Li2SO4 (0.1)	54.7 μg h−1 mg−1	20.8
Co−MoS2‐x	[75]	H2SO4 (0.01)	0.6 mmol h−1 g−1	10
Co−MoS2	[76]	Na2SO4 (0.1)	127.88 μg h−1 mgcat. −1	11.29
Fe‐sMoS2	[77]	HCl (0.1)	24 μg cm−2 h−1	27
S‐rich MoS2	[80]	Li2SO4 (0.1)	43.4 μg h−1 mgcat. −1	9.81
1T‐MoS2/NF	[81]	Li2SO4 (0.25)	1.05 μg min−1 cm−2	27.66
Fe‐MoS2	[82]	K2SO4 (0.5)	8.63 μg h−1 mgcat. −1	18.8
F‐MoS2	[83]	Na2SO4 (0.1)	35.7 μg h−1 mgcat. −1	20.6
MoS2@ZIF‐71	[84]	Na2SO4 (0.1)	56.69 μg h−1 mgMoS2 −1	30.91

Azofra et al. deposited Fe on MoS₂ as an electrocatalyst according to the Fe−Mo−Co structure of nitrogenase. ^[85] DFT calculations were employed to predict the high selectivity of Fe‐MoS₂. Yang et al. used DFT calculations to investigate the NRR activity of some transition metal@MoS₂ systems. ^[86] By comparing the N₂ adsorption capacity and the potential‐determining step (PDS) energy barrier of the selected catalysts, V@MoS₂ was predicted to have the best activity. A DFT theoretical calculation on the performance of MoS₂ doped with transition metals (3d, 4d, 5d elements) was also performed by Zhai et al. ^[87] Through the metal binding energy (lower than −0.5 eV), they first excluded Zn, Cd and Hg as shown in Figures 7A, B and C. Low metal binging energies indicated weak interactions between metals and substrates. After that, Zr and Ta were also excluded due to the huge distortion. According to the law of free energy variation in the proton coupling, the two highest positive free energy steps ΔG (*N₂−*NNH) and ΔG (*NH₂−*NH₃) needed <0.49 eV, and only Re and Ti met the requirement as shown in Figure 7D. In the protonation pathway, the PDS energy barrier of Re@MoS₂ (−0.42 eV) was found smaller than that for Ti@MoS₂ (−0.73 eV) (Figure 7E and F). Therefore, Re@MoS₂ was identified as the most suitable catalyst. Moreover, the HER energy barrier of Re@MoS₂ was higher than that for NRR, which proved that Re@MoS₂ was more conducive to N₂ adsorption. The stability of Re@MoS₂ was also demonstrated by ab initio molecular dynamics simulations. Therefore, it seems reasonable to speculate that Re‐doped MoS₂ should demonstrate high NRR activities. Zhao et al. also used DFT calculations to predict the NRR activity of some transition metal atoms embedded in MoS₂ nanosheets. ^[88] The PDS energy barrier of Mo‐doped MoS₂ was determined at −0.53 eV (Figure 8A), which was better than other systems. Thus, they regarded Mo/MoS₂ as a potential catalyst for the NRR process.

Figure 7.

(A–C) The binding energies of 3d, 4d and 5d transition metals (TM) on Mo top sites and hollow sites of MoS₂, respectively. (D) Screening results of TM@MoS₂ systems for the NRR process by the free energy evolution of two thermodynamically detrimental protonation steps (ΔG(*N₂−*NNH) and ΔG(*NH₂−*NH₃)). The TM@MoS₂ systems which meet the requirement are marked with light cyan color in the lower left part of the graph. (E, F) Enzymatic pathway of side‐on adsorbed N₂ molecule on Re@MoS₂ and Ti@MoS₂ for the NRR process, respectively. Reproduced with permission from Ref. [87]. Copyright 2020 Royal Society of Chemistry.

Figure 8.

(A) Free energy chart for the NRR process at Mo‐doped MoS₂ nanosheets at zero and applied potential (limiting or onset potential). Reproduced with permission from Ref. [88]. Copyright 2018 Royal Society of Chemistry. (B) Berthelot spectrophotometric analysis for ammonium in the working electrolyte solutions after chronoamperometric tests. Reproduced with permission from Ref. [89]. Copyright 2020 Electrochemical Society. (C) Panel views of the Mo top site, the S top site and the hollow site of TM‐SAs in the 4×4×1 MoS₂ supercell. Reproduced with permission from Ref. [90]. Copyright 2020 Elsevier. (D) The limiting potentials of HER [UL(HER)] versus NRR [UL (NRR)] on Ti, Cu, Hf, Pt, and Zr‐decorated MoS₂. Ti, Cu, Hf, Pt, and Zr‐decorated MoS₂ in the region above the dotted line corresponds to HER being more favorable than NRR. Reproduced with permission from Ref. [91]. Copyright 2019 American Chemical Society.

Although theory predicted encouraging catalytic properties of Mo−MoS₂, Simonov and co‐workers recently demonstrated that Mo/MoS₂ was only partially successful in experiments, ^[89] as the desorption step of NH₃ was not considered. They first carried out an electrocatalytic experiment in water based media (0.1 m Li₂SO₄), but Mo/MoS₂ showed no NRR activity under these conditions (Figure 8B). After that, they experimented with an aprotic medium (1‐butyl‐1‐methylpyrrolidinium (trispentafluoroethyl) trifluorophosphate), in which the water concentration was tightly controlled. At the initial stage of the experiment, the accumulation of NH₃ was stable. However, after more than 2 h of continuous operation, Mo−MoS₂ lost its catalytic activity, because the resulting NH₃/NH₄ ⁺ adsorbed to the surface of Mo−MoS₂, clogging up the active sites. Therefore, it was concluded that the high NRR activity of Mo−MoS₂ predicted by DFT calculations was not feasible in practical experiments even though, through first‐principles high‐throughput calculations methods, Yang et al. had also predicted the NRR activity of Mo@MoS₂ to be the highest among the calculated systems. ^[90] In addition, they demonstrated that the N₂ adsorption activity site on the top of Mo atoms had the best NRR performance (Figure 8C). On this basis, other theoretical calculations were adopted to demonstrate high activity and selectivity of Mo@MoS₂−M. However, theoretical prediction lack in investigating the crucial NH₃ desorption step, the high activity of predicted Mo@MoS₂‐M still remains to be confirmed experimentally.

Guo et al. aimed at exploring the NRR potential of a part of transition metal‐doped (IIIB to IIB subgroups except for Tc and Hg) defective MoS₂. ^[91] DFT calculations were carried out to affirm the PDS energy barrier of each catalyst; Sc, Ti, Cu, Hf, Pt, and Zr‐decorated MoS₂ were considered as potential candidates for N₂ electroreduction to NH₃ due to the PDS energy barrier being smaller than −0.7 eV. To further screen, they tested the limiting potentials of HER [UL(HER)] versus NRR [UL(NRR)] for the six catalyst systems. As shown in Figure 8D, the HER activity of Pt‐ Cu‐modified MoS₂ was higher than N₂ fixation. Thus, Sc, Ti, Hf, and Zr‐doped MoS₂ materials were predicted to show a decent NRR activity. Tang and Li predicted that B‐doped MoS₂ should have a high NRR catalytic activity, ^[92] as B is difficult to combine with H under acidic conditions, which could effectively inhibit the HER process. They first contrasted the catalytic properties of B atom‐ and diatomic boron‐doped MoS₂. By calculating overpotentials (0.02 V vs. 0.30 V) and activation barriers (1.24 eV vs. 2.84 eV), they predicted that B₂@MoS₂ should have a better NRR activity. Therefore, defective MoS₂ with double S vacancies was used as substrate to avoid the aggregation of B₂. Meanwhile, the high stability, conductivity and selectivity of B₂@MoS₂ further proved the feasibility of B₂@MoS₂for the NRR process, predicting it to be a potentially useful electrocatalyst.

As the synergistic interactions within metal clusters contribute to improving the catalytic activity of individual metal atoms, Zhang et al. attempted to predict the NRR performance of Fe₂/MoS₂ by theoretical calculations. ^[93] They used DFT calculations combined with the computational hydrogen electrode model, which concluded a high catalytic activity of Fe₂/MoS₂ (the overpotential was 0.21 V and the energy barrier of PDS was −0.37 eV). The catalytic principle of Fe₂/MoS₂ was also proposed (the electron density loss on the Fe cluster provided a stable Lewis active site for the adsorption of N₂, at the same time the electron feedback from the Fe clusters to N₂ promoted the activation of N−N), which further indicated that Fe₂/MoS₂ had potential as a high activity NRR catalyst. Matanovic et al. explored the HER and NRR activity of defect‐rich 2H‐MoS₂ through experiments and theoretical calculations. ^[94] By DFT calculations, they predicted that the two vacancies might have NRR selectivity, but FE might not be ideal due to the effect of the large overpotential. Therefore, it was concluded that the original 2H‐MoS₂ has a priority role in HER reaction. Further optimization is necessary for 2H‐MoS₂ application in the NRR process.

5. Conclusion and Outlook

MoS₂ is considered as a potential NRR catalyst due to the high specific surface area, adjustable electronic structure and elemental composition similar to the nitrogenase enzyme. However, due to the influence of the HER process and the defects of MoS₂, the NRR activity obtained by MoS₂ is not satisfactory. Therefore, optimization of MoS₂ with higher NRR catalytic activity has become a hot research direction. In this Minireview, we summarized the optimization strategies of MoS₂‐based catalysts. Optimization strategies include enhancing electrical conductivity/enlarging specific surface area, inhibition of competing reaction and taking advantage of the interaction effect between other elements and MoS₂. In addition, some enhancement strategies by theoretical calculations have also been reviewed, which provides researchers with ideas for enhancing the NRR activity of MoS₂‐based catalyst.

Despite some achievements that have been made with MoS₂‐based catalysts, it is still a long way from replacing the Haber‐Bosch process. Therefore, further optimization is necessary. As MoS₂ is a material with many unique characteristics, optimization research in the future can focus on the characteristics of MoS₂.

1. It is well known that MoS₂ is a 2D‐layered material. The decrease of MoS₂ layer spacing is conducive to inhibit the HER process. Therefore, the NRR selectivity can be improved by reducing the MoS₂ layer gap. Furthermore, ultra‐thin mono‐layer MoS₂ can observably increase the specific surface area. It is worth considering some other optimization measures on the basis of ultra‐thin mono‐layer MoS₂.

2. MoS₂ has a variety of crystal forms (1T, 2H, 3R), among which 1T‐MoS₂ and 2H‐MoS₂ were widely used in the NRR process. However, the base plane of 2H‐MoS₂ is inert, resulting in insufficient adsorption sites. The poor conductivity of 2H‐MoS₂ also limits the activity of NRR. Compared with 2H‐MoS₂, 1T‐MoS₂ shows as an octahedral structure, which not only overcomes the inert base surface to expose a denser active site, but also has high electrical conductivity. Nevertheless, as the inherent HER activity of 1T‐MoS₂ is a limitation, future optimization strategies can be developed based on inhibiting the HER activity. Overall, 1T‐MoS₂ has the most potential as an efficient catalyst for NRR.

3. Taking advantage of the characteristic of the adjustable electronic structure of MoS₂, the electrical conductivity and N₂ adsorption capacity can be improved by changing the electronic structure of MoS₂.

4. As N₂ is dissolved in water during the NRR process, it is converted to NH₃ in contact with the catalyst surface. However, the intrinsic surface of MoS₂ is hydrophobic, ^[95] which prevents N₂ from contacting the surface of the catalyst. Thus, taking measures to change the hydrophobicity of MoS₂‐based catalysts is also a potential research area.

Conflict of interest

The authors declare no conflict of interest.

Biographical Information

Liang Tian, graduate student at the Department of Materials and Chemical Engineering of the University of Jinan. His current research focuses on the application of non‐noble metal hierarchical heterostructure nanocatalysts for electrocatalytic N₂ reduction reactions.

Biographical Information

Jinxiu Zhao received her B.Sc. degree from the University of Jinan in 2017. She is now pursuing her PhD degree at the School of Chemical Engineering of the University of Jinan. Her current research mainly focuses on the electrochemical nitrogen reduction reaction and water splitting.

Biographical Information

Xiang Ren received his B.Sc. degree in Chemistry of Materials and English from the University of Jinan in 2012, his M.Sc. degree in Chemical Engineering and Technology from the University of Jinan in 2015, and a PhD degree from the University of Jinan/University of Electronic Science and Technology of China (USTC) in 2019. Now, he is an associate professor at the University of Jinan. His main research interests are energy catalysis, nanomaterial controlled‐synthesis, and electrochemical biosensors.

Biographical Information

Xu Sun received her PhD in Inorganic Chemistry at the University of Science and Technology of China (USTC). She is now working as an associate professor in the School of Chemistry and Chemical Engineering at the University of Jinan. Currently, her research focuses on the design and fabrication of novel nanomaterials for the construction of energy‐related devices.

Biographical Information

Qin Wei, a professor and DSc, has devoted herself to analytical teaching and scientific research. Her main research interests are the study of proteins and nucleic acids by photometry and electrochemical immunosensor preparation. She has published over one hundred articles on analysis, immunosensors in journals such as Biomaterials, Adv. Funct. Mater., Biosens. Bioelectron., Sens. Actuators B: Chem., Talanta, and applied successfully for many research projects.

Biographical Information

Dan Wu, a professor and DSc, received her PhD degree from Shandong University in 2007.. Her current research focuses on the application of nanomaterials to energy catalysis and sensors.

L. Tian, J. Zhao, X. Ren, X. Sun, Q. Wei, D. Wu, ChemistryOpen 2021, 10, 1041.