1 Introduction
Ammonia (NH
3) is an important chemical raw material, which is not only used in the production of chemical products such as fertilizers and industrial fuels, but also an important clean energy carrier because of its high hydrogen content (17.6 wt%) and energy density (4.32 kWh·h
-1)
[1,2]. At present, the traditional Haber-Bosch process with nitrogen (N
2) and hydrogen (H
2) as raw materials is mainly used for industrial ammonia synthesis
[3]. Due to the high bond energy (941 KJ·mol
-1) of the triple bond (N N bond) in the nitrogen molecule, the Haber process for ammonia synthesis requires the use of heterogeneous iron-based catalysts at high temperatures (300 ~ 500 ℃) and high pressures (150 ~ 300 atm), and consumes 1% ~ 2% of the world's total energy each year. At the same time, carbon dioxide produced by hydrogen production accounts for 1.5% of the world's total annual emissions, exerting tremendous pressure on the environment
[4⇓⇓~7]. Therefore, it is necessary to find a low energy consumption and green sustainable alternative technology to realize ammonia synthesis under mild conditions.
To this end, researchers are committed to exploring new methods to catalyze the reduction of N
2 to NH
3, such as photocatalytic reduction, biological nitrogenase reduction and electrocatalytic reduction
[8⇓~10]. Electrocatalytic Nitrogen Reduction Reaction (NRR) has many advantages, such as low energy consumption, controllable Reaction, simple equipment and green environmental protection, so it is very popular. The reaction can directly use water as a hydrogen source and nitrogen as a nitrogen source under mild conditions to synthesize ammonia under the action of a catalyst. However, the N
2 has low electron affinity (-1.9 eV) and high ionization energy (15.8 eV), which increases the resistance to Lewis acid/base. There is a large energy gap (10.82 eV) between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the N
2, which is not conducive to electron transfer
[11,12]. In addition, most of the electrocatalytic nitrogen reduction Reaction is carried out in the aqueous solution containing electrolyte, so it is hindered by the competition of Hydrogen Evolution Reaction (HER)
[13,14]. Based on the above reasons, there are some problems in the current electrocatalytic nitrogen reduction reaction, such as low ammonia yield, low Faradic Efficiency (FE) and poor selectivity. How to achieve efficient electrocatalytic synthesis of ammonia is a challenging problem in this field. The key to solve this problem is to construct a reasonable nitrogen reduction reaction system, develop efficient NRR electrocatalysts, optimize the selectivity of electrocatalysts and improve the electron utilization, which is also a hot topic in NRR research. By reasonably designing and adjusting the active components, crystal structure and surface properties of the electrocatalyst, the catalyst can preferentially bind nitrogen atoms rather than hydrogen atoms, thus effectively inhibiting the generation of HER and further improving the efficiency of NRR
[15]. In addition, in the process of electrocatalysis, the catalyst is prone to deactivation and decomposition, which requires that the designed catalyst should have good stability and long service life. In order to obtain efficient and stable NRR catalysts, researchers have used many design strategies, such as alloying, heterojunction, size effect, interface effect and defect engineering, to effectively promote the nitrogen reduction activity of catalysts, and have made great progress and meaningful research results.
In this paper, the basic reaction mechanism of electrocatalytic nitrogen reduction is introduced, how to reasonably construct the electrocatalytic nitrogen reduction system is introduced, and the latest research progress of NRR electrocatalyst in recent years is summarized.The design ideas and strategies to improve the apparent or intrinsic activity of catalysts were reviewed, and the challenges and potential applications in this emerging field were prospected.
2 Reaction mechanism of NRR
The reaction mechanism of NRR is very complex and mainly depends on the catalytic surface and the applied potential. The reaction process of NRR involves three basic steps: :(1)N2 the adsorption of molecules on the active sites of the catalyst; (2) cracking and hydrogenation of N N N bond; (3) desorption of the generated ammonia molecules (or other intermediate products) from the surface of the catalyst.
At present, the proposed electrocatalytic nitrogen reduction reaction mainly follows two pathways, namely the dissociation pathway and the association pathway
[16]. In the dissociation pathway, the N N N bond is first cleaved when the 1a),N
2 molecule is adsorbed on the surface of the heterogeneous catalyst, resulting in two isolated nitrogen atoms on the active site, and then the two nitrogen atoms on the surface of the catalyst are successively protonated and finally converted into NH
3 molecules. Because the N N N bond is very stable, this reaction path requires a huge amount of energy. The association path is different from the dissociation path, and the N N N bond will be broken only when the NH
3 molecule needs to be released. In addition, the association pathway can be further divided into the distal association pathway (scheme 1b) and the alternative association pathway (scheme 1C) according to different hydrogenation sequences. In the distal association pathway, the nitrogen atom located at the distal end is first protonated successively to form a NH
3 molecule, and then the remaining N atom located at the surface of the catalyst is protonated again to release a second NH
3 molecule. In the alternative association pathway, the two nitrogen atoms first undergo 1H
+/1e
-stepwise electron-proton transfer simultaneously, and then the two nitrogen atoms undergo alternative hydrogenation. With the continuous addition of protons, the N N N bond is broken and a new N — H bond is formed, and then two NH
3 molecules are gradually formed. According to the above analysis, hydrazine (N
2H
4) may be produced as a by-product in the association mechanism of alternating hydrogenation. In addition, density functional theory (DFT) calculations by Nørskov et al. Show that the reaction step of protonation of N
2 molecules to
*N
2H is the rate-limiting step of NRR on the surface of catalysts with weak binding capacity to N
2 molecules
[17]. For the catalysts with good activity, such as noble metal catalysts Ru, Re and Rh, their surface can strongly adsorb N
2 molecules, and their rate-limiting step is the reaction step from
*NH to
*NH
2 or from
*NH
2 to NH
3.
Recently, Abghoui and Sk Skúlason proposed a new Mars-van Krevelen (MvK) pathway for metal nitride (TMNs) catalysts, which is different from the traditional dissociation and association pathways
[18]. According to this pathway, the lattice nitrogen atoms on the surface of metal nitride catalysts can be continuously hydrogenated and reduced to NH
3 molecules, and the nitrogen vacancies formed on the surface of catalysts can be further replenished and restored by adsorbing N
2 molecules. Unlike the two previously discussed pathways, the MvK pathway focuses on the replenishment of nitrogen vacancies and the regeneration of the catalyst, rather than on the adsorption of N
2. In addition, based on the MvK mechanism, Abghoui et al. Conducted NRR simulations on a series of transition metal nitrides and predicted their NRR reaction onset potentials. The results show that ZrN, NbN, CrN and VN have the greatest potential to be used in NRR reaction. Although many metal nitrides have been proved to be promising NRR catalysts by theoretical analysis, it is very important to determine the nitrogen source of metal nitrides for ammonia production and to verify the reaction mechanism of MvK. Zhang Haimin's team of Institute of Solid State Physics, Chinese Academy of Sciences, further verified the Mvk reaction mechanism of plasma etched titanium nitride (TiN-PE) NRR process from the experimental point of view by isotope labeling method
[19]. When
15N
2 is used as the feed gas,The resulting NH
3 consists of
14N (36% abundance) in the
14NH
4 and
15N (64% abundance) in the
15NH
4. This clearly demonstrates that the NRR process on TiN-PE is carried out through the MvK mechanism.
In addition, a new nitrogen reduction mechanism, surface hydrogenation mechanism, has recently been proposed, which demonstrates that noble metal catalysts (such as Au and Pd) can electrochemically reduce N
2 to NH
3 at very low potentials
[20]. Different from the existing reaction mechanism, the first step of surface hydrogenation is the reduction of H
+ and the formation of surface hydrogenation, rather than the adsorption and activation of N
2. Subsequently, under the synergistic effect of the catalyst and the surface
*H, the N
2 can react with the surface
*H,An intermediate of that
*N
2H
2 is generate, and the
*N
2H
2 can be further reduced to finally generate the NH
3. In this reaction mechanism, the reduction of H
+ determines the initial potential of the reaction, and the reaction of N
2 with
*H to form
*N
2H
2 determines the reaction rate. However, in aqueous solution, the overpotential of the electrocatalytic nitrogen reduction reaction on the catalyst surface is greater than that of the hydrogen evolution reaction, and either reaction path will compete strongly with HER, so it is necessary to develop efficient NRR electrocatalysts.
3 Factors affecting the electrocatalytic reaction system of NRR
In the current research, many researchers ignore the design of the electrocatalytic reaction system, and only focus on the NRR electrocatalyst. However, the key to the electrocatalytic nitrogen reduction reaction is the synergy between the electrocatalyst and the reaction system. In addition, the kinetics and thermodynamics of nitrogen reduction depend on many factors, such as electrode potential, solvent effects, and pH of the electrolyte. Therefore, to improve the performance of NRR, not only the catalyst should be designed reasonably, but also the whole reaction system should be investigated, the structure of the electrolytic cell should be designed reasonably and the appropriate electrolyte should be selected.
3.1 Design of electrolytic cell
In the electrocatalytic nitrogen reduction system of aqueous solution, there are mainly three types of electrolytic cells at present, namely, the polymer electrolyte membrane (PEM) type electrolytic cell (Fig. 1A), the single-chamber electrolytic cell (Fig. 1b) and the double-chamber H-type electrolytic cell (Fig.1c)
[7]. Inside the single-chamber cell, the cathode and anode of the cell are in the electrolyte solution, which causes the ammonia produced to be re-oxidized on the anode. This problem can be avoided by the double-chamber H-type electrolytic cell. In the H-type electrolytic cell, the working electrode and the reference electrode are both in the cathode chamber, so that the applied voltage can be more accurate, and the cathode and the anode are separated by an ion exchange membrane, so that the occurrence of side reactions is effectively inhibited. Therefore, in recent years, most of the electrocatalytic ammonia synthesis reactions have been carried out in a two-chamber H-cell system. In addition to the two types of electrolytic cells reported above, there is another type of reaction cell which is a PEM-type electrolytic cell, but there are few reports on this type of electrolytic cell
[21]. Xiang et al. Used KHCO
3 solution as anode electrolyte to provide protons in a PEM-type electrolytic cell, and electrocatalytically driven NRR on a Fe
2O
3/CNTs catalyst
[22]. The reaction system is beneficial to the separation of ammonia, but the yield of ammonia is low.
3.2 Effect of applied potential
The electrocatalytic nitrogen reduction reaction occurs only when the applied potential is lower than the equilibrium electrode potential, so the overpotential can be seen as the driving force for the NRR. In a previous study, the researchers found that the minimum potential required for the electrocatalytic NRR was 1.17 V at ambient conditions and without any electrocatalyst, which was at least 0.5 V greater than the overpotential of the hydrogen evolution reaction
[23]. Therefore, the voltage applied in the reaction system is very important for NRR. In general, the optimal potentials are different for different reactions. When the applied voltage exceeds the optimal potential, both the rate of NH
3 formation and FE decrease because the hydrogen evolution reaction is dominant at this time, but NRR can also be dominant when the appropriate working potential is selected. For example, Chen et al. Used Fe-CNT electrocatalyst to carry out nitrogen reduction reaction. When the applied voltage was between -1.5 and -2.0 V vs RHE, the hydrogen evolution reaction was dominant, and the Faraday efficiency of ammonia would be greatly reduced
[24]. However, when the applied voltage is less than -1. 0 V vs RHE, the nitrogen reduction reaction will dominate again, resulting in a sharp decrease in the Faraday efficiency of hydrogen evolution.
3.3 Effect of electrolytes
3.3.1 Effect of pH value of electrolyte
In the process of electrocatalytic nitrogen reduction reaction, the selection of appropriate electrolyte solution can limit the proton supply in the solution, thus inhibiting the occurrence of hydrogen evolution reaction
[14,25]. Generally, after the desorption of H
2 or NH
3 generated on the surface of the catalyst, the active sites can adsorb protons H or N
2 molecules again. If the potential around the electrons and protons is lower than 0 V vs RHE, the hydrogen evolution reaction will take place preferentially because H adsorbed on the active sites is dominant both kinetically and thermodynamically
[12]. Conversely, if there are no electrons and protons around, the N
2 will be adsorbed, and when electrons and protons are subsequently supplied, they may preferentially bind to the N
2.
Like the catalysts described above, which have an optimal applied potential range, NRR also has an ideal pH range. As early as 1993, some researchers studied the effect of pH value on the reduction reaction performance of nitrogenase, and gave the optimal pH range of MoFe nitrogenase as 5 ~ 9.8
[26]. This is because in the acidic medium with low pH value, there are a lot of protons in the system, and the hydrogen evolution reaction will take place preferentially. In addition, Fengxiao Feng of the University of Central Florida in the United States studied the NRR performance of Pd/C catalysts in electrolytes with different pH values
[27]. Compared with the V vs RHE and 0. 1 mol/L NaOH (pH = 12. 9), the Pd/C catalyst showed better NRR performance in 0. 1 mol/L PBS.At 0. 1 V vs RHE potential, the ammonia synthesis rate was 4.5μg·h
-1·m
, and the Faraday efficiency was 8. 2%. This is because in the neutral medium, the competitive hydrogen evolution reaction is effectively suppressed due to the high resistance of charge transfer and proton diffusion. In addition, in the present study, the optimum pH range of the catalyst is different depending on the reaction system, but there is no report on the universal optimum pH range for each NRR reaction system.
3.3.2 Counterion effect
The low solubility of nitrogen in aqueous solution is one of the main problems in the electrocatalytic nitrogen reduction reaction. In order to solve this problem, many researchers use ions in electrolytes as counter ions to improve the solubility of N
2 in solution. For example, Song et al. Used nitrogen-doped carbon nanotubes as NRR catalysts to study the counterion effects of lithium perchlorate, sodium perchlorate and potassium perchlorate electrolytes on NRR systems
[28]. The size of counter ions increases in the order of Li
+<Na
+<K
+, and the NRR performance is negatively correlated with the size of counter ions. Ammonia yield and Faraday efficiency were higher in the system with LiClO
4, followed by NaClO
4 and KClO
4. The researchers speculate that this is because counterions with smaller atomic radii can enhance the electric field in the Stern layer and increase their association with N
2 molecules, so the catalyst has higher NRR activity in the LiClO
4 electrolyte.
3.3.3 Synergistic effect of catalyst and electrolyte
Zhao Huijun and Zhang Haimin, Institute of Solid State Physics, Chinese Academy of Sciences, used uniformly distributed molybdenum disulfide nanosheets (MoS
2/BCCF) as catalysts to investigate the effects of different pH and different concentrations of Li
2SO
4 and Na
2SO
4 as electrolytes on the catalytic activity of NRR
[29]. It was found that the ammonia synthesis rate and Faraday efficiency in the electrolyte of 0.1 mol/L Li
2SO
4(pH=3) were 8 and 18 times higher than those in the electrolyte of 0.1 mol/L Na
2SO
4(pH=3), respectively. This is because the hydrogen-producing active sites of the MoS
2 are marginal sulfur sites, and when the Li
2SO
4 is used as the electrolyte, the S sites at the marginal positions of the catalyst and the Li
+ in the electrolyte form a strong "Li — S" bonding effect, which inhibits the hydrogen production of the marginal sulfur sites and improves the ability of activating nitrogen on the marginal molybdenum sites, thereby improving the catalytic performance of the catalyst for N
2 reduction to ammonia. The researchers skillfully used the synergistic effect of electrolyte and catalyst to significantly improve the efficiency of electrocatalytic ammonia production, which provided a new idea for the construction of efficient electrocatalytic ammonia synthesis system at room temperature and pressure in the future.
By choosing the appropriate electrolyte and constructing the optimal catalyst-electrolyte system, the hydrogen evolution reaction can also be inhibited to a certain extent through the synergistic effect of the electrolyte and the electrocatalyst, thereby promoting the electrocatalytic nitrogen reduction reaction.
4 Research Progress of Electrocatalysts for NRR
With the optimization of the system, the design of catalysts with rich active sites to improve the activity of NRR plays an important role in the field of electrochemical nitrogen reduction. In the past decade, there are many kinds of catalysts for electrocatalytic NRR, including noble metal catalysts, non-noble metal catalysts and metal-free catalysts and their corresponding monatomic catalysts.
4.1 Noble metal catalyst
Noble metals have good electrical conductivity and active polycrystalline surfaces, which can combine with a variety of reactants, so they can be used as efficient electrocatalysts for many reactions. The intensity of nitrogen reduction reaction mainly depends on the adsorption intensity of N
2 on the catalyst surface and the ability of N N bond cleavage, while the adsorption intensity of N
2 depends on the electronic state of the active catalyst surface. The d orbitals of noble metals can accept lone pair electrons from N
2 and feed them back into the π orbitals of N
2 molecules, thus weakening the N N N bond
[30,31]. Therefore, noble metals or noble metal-based electrocatalysts, such as Pt, Au, Ru, Ag, Pd, Rh, etc., exhibit excellent NRR performance
[32,33][34⇓⇓⇓⇓~39][40⇓⇓~43][44⇓~46][47⇓⇓⇓⇓~52][53].
For example, Yan Junmin's team at Jilin University used spherical aberration correction transmission electron microscopy and UV-Vis spectroscopy to characterize the high-index section and surface atomic arrangement of the synthesized tetracosahedral gold nanorods (THH Au NRs)
[37]. The surface of the polyhedral THH Au NRs is composed of a mixture of high-index { 730 } planes consisting of (210) and (310) subplanes (Fig. 2a). The THH Au NRs catalyst exhibited better NRR electrocatalytic activity than the gold nanoparticles with ammonia synthesis rates of V vs RHE when a -0.2 V vs RHE voltage was applied (Fig. 2b, C). This is due to the fact that the exposed unsaturated { 730 } step surface of the catalyst has higher surface energy and lower coordination number, which provides abundant active sites and can efficiently capture N
2, so that nitrogen atoms are preferentially adsorbed on the crystal surface than protons, and chemisorbed Au — N bonds are formed, thus facilitating the cleavage of N N bonds. In addition, the results of DFT calculation show that the hydrogenation of THH Au NRs on the surface of THH Au NRs catalyst is carried out according to the alternating association mechanism (Fig. 2D), so the by-product hydrazine hydrate (N
2H
4·H
2O) is produced in the nitrogen reduction reaction, and its synthesis rate is 0.102μg·h
-1·cm
-2, resulting in poor product selectivity.
Although noble metal catalysts have good activity for nitrogen reduction, their high cost and scarcity of resources have seriously limited their large-scale application. The noble metal is compounded with a non-noble metal, so that the usage amount of the noble metal can be remarkably reduced, the cost of the catalyst is reduced, and the catalytic performance of the material is further optimized. For example, ZIF-8 supported AuCu alloy catalysts with high NRR activity and stability were prepared by wet chemical method using NaBH
4 as reducing agent and adjusting the ratio of Au/Cu
[39]. The sample with Au ∶ Cu = 3 ∶ 7 supported on ZIF-8 (S3/ZIF-8) has higher activity, which is attributed to the support effect of ZIF-8 with high specific surface area, good alloy dispersion and the modulation of surface electronic structure by alloying. In situ electrochemical Ftir spectroscopy was used to monitor the reaction process in real time, and the
*N
2H
y species (1 ≤ y ≤ 4) of the product intermediates formed on the AuCu alloy were captured, indicating that the NRR reaction catalyzed by the material follows the association pathway. In addition, the density functional theory (DFT) results show that the rate-determining step of the NRR reaction on the AuCu @ ZIF-8 surface is the first protonation of the N
2 molecule, and the nitrogen reduction process is theoretically verified to be an alternating hydrogenation process. In addition, when air is used as a nitrogen source, the ammonia synthesis rate of the catalyst can reach the
of the 23.3μg·h
-1·m in an acidic medium. This indicates that synthetic ammonia has taken an important step towards green industrialization.
In addition to gold catalysts, noble metals such as Ru and Pd have also been shown to have the potential to electrocatalyze nitrogen reduction. Inspired by the metal-sulfur bond in nitrogenase, Muhammad Ibrar Ahmed et al. Used surface modification strategy to adjust the electronic structure of the catalyst, thereby enhancing the adsorption of Muhammad Ibrar Ahmed and inhibiting hydrogen evolution reaction
[43]. They anchored different aliphatic thiol-modified ruthenium nanocrystals on reduced graphene oxide to achieve Ru-S bonds. Under acidic conditions, the faradaic efficiency of V vs RHE and the rate of synthesis of 50μg·h
-1·mg
-1 were obtained at-0. 1 V vs RHE using dodecanethiol modified ruthenium nanocrystal catalyst. The modified catalyst showed excellent stability and recoverability for the production of NH
3, as confirmed by rigorous control experiments including
15N isotope labeling experiments. DFT calculations show that the electronic structure modification and surface modification can inhibit the adsorption and desorption of N
2 on the intermediate products. The above-mentioned Fengxiao Feng team at the University of Central Florida in the United States loaded Pd nanoparticles on carbon black and studied its NRR performance. In neutral medium, at 0.1 V vs RHE potential, the ammonia synthesis rate was V vs RHE
, and the Faraday efficiency was 8.2%
[27]. DFT calculations show that Pd can interact with hydrogen atoms to form α-PdH, which can reduce the energy barrier of N
2 activation and accelerate the protonation process of N
2. In order to verify the promotion effect of hydride, Cheng Fangyi's team of Nankai University directly synthesized nanoporous palladium hydride (np-PdHx) and used it as NRR catalyst. At -0.15 V vs RHE potential, the ammonia synthesis rate of the catalyst was V vs RHE
, and the Faraday efficiency reached 43.6%
[52]. By means of in situ Raman analysis and DFT calculation, as well as further isotope hydrogen labeling experiments, it is confirmed that the lattice hydrogen in the hydride is the active hydrogen source, and the barrier of forming
*N
2H is low, so the catalyst has good NRR activity.
The noble metal catalysts have outstanding performance in NRR research, but some of them are more likely to promote hydrogen evolution reaction because they are more inclined to adsorb hydrogen atoms. In order to improve the NRR activity of noble metal catalysts, the catalysts can be optimized by alloying, defect engineering and other design strategies in the follow-up study.
4.2 Non-noble metal catalyst
Compared with precious metals, transition metal materials are relatively inexpensive, abundant in the crust, and also have d-orbital electrons that can feed back to N N N bonds. In addition, in nature, the molybdenum-iron protein containing transition metal elements in nitrogenase is the catalytic active site of biological nitrogen fixation
[54,55]. Inspired by this, researchers have applied various transition metal-based materials, such as oxides, nitrides, carbides, sulfides and so on, to electrocatalytic nitrogen reduction, showing great potential for electrocatalytic nitrogen reduction.
4.2.1 Transition metal oxides
Transition metal oxides, especially iron-based compounds, have special surface structures, which can better adsorb N
2. For example, by controlling the chemical state and composition of the catalyst, the researchers found that the Faraday efficiency of V vs RHE catalyst could reach 8.29% at -0.3 V vs RHE potential
[56]. This is mainly due to the fact that the interface between Fe and the Fe
3O
4 has different binding forces to the intermediate of the
*N
xH
y, which effectively avoids the poisoning effect caused by the binding of the intermediate.The interface between Fe and Fe
3O
4 nanocrystals can adsorb N
2 to form
*N
2H, and further desorb
*NH
2 to form NH
3, thereby enhancing the NRR activity.
In addition to iron-based catalysts, other transition metal oxides such as molybdenum oxide, niobium oxide, manganese oxide and tungsten oxide have also been reported to be effective in driving the nitrogen reduction reaction
[57,58][59,60][61][62⇓~64][65⇓⇓⇓⇓⇓⇓~72]. For example, Zheng Gengfeng's team at Fudan University found that V vs RHE can provide empty d orbitals for adsorbed N
2, and can feed back d electrons to the π orbitals of N bonds to activate N bonds. When -0.65 V vs RHE voltage is applied, the ammonia synthesis rate is 11.6μg·h
-1·m
, and the Faraday efficiency is as high as 32%
[60].
In a word, transition metal oxides have shown great application value in NRR research, but their NRR performance is not optimistic due to their poor conductivity and low intrinsic activity, so researchers may focus on improving these shortcomings to improve their catalytic activity.
4.2.2 Transition metal nitrides
Transition metal nitrides, such as molybdenum nitride, vanadium nitride and tungsten nitride, have good thermodynamic stability, high conductivity and excellent electronic structure, and show excellent NRR activity in the process of electrocatalytic nitrogen reduction
[73,74][75][76]. For example, Sun Xuping's team at the University of Electronic Science and Technology of China prepared Mo
2N nanorods by calcining V vs RHE precursors in an ammonia atmosphere. In 0.1 mol/L HCl solution, the ammonia synthesis rate was 78.4μg·h
-1·m
at -0.3 V vs RHE potential, and the Faraday efficiency was 4.5%
[73]. Through DFT calculations, the team showed that the Mo
2N catalyst achieved catalytic reduction of N
2 through the Mars-van Krevelen path. In addition, since the nitrogen atoms in the crystal lattice participate in the reaction, it is essential to improve the stability of the nitride catalyst. For example, when Yang et al studied the nitrogen reduction performance of VN nanoparticles, they found that the ammonia synthesis rate of the catalyst was 3.3×10
-10mol·s
-1·cm
-2 in the first hour, but after 5 hours of reaction, the ammonia synthesis rate decreased by 62%
[77]. They believed that the newly formed VN
0.7 on the surface of VN in the process of nitrogen reduction was the active component, and with the extension of catalytic reaction time, the active component was reduced to inactive VN again, resulting in the decrease of ammonia synthesis rate. Therefore, how to improve the stability of nitride catalyst is an urgent problem to be solved. In addition, when nitrogen is contained in the catalyst, attention should be paid to the nitrogen source of the labeled product. When T. Leo Liu's team at Utah State University used tetragonal T. Leo Liu to electrocatalytically reduce N
2, they found that the ammonia produced was produced by the decomposition of T. Leo Liu
[78]. When they examined the source of nitrogen in ammonia products by
15N labeling experiments, only a large amount of
14NH
4+ was observed, suggesting that ammonia formation may not come from electrocatalytic nitrogen reduction, but from chemical decomposition of Mo
2N. This work sounds an alarm for the use of metal nitrides and even nitrogen-containing materials in electrocatalytic nitrogen reduction reactions. Therefore, when nitrogen-containing materials are used as NRR catalysts,
15N labeling experiments should be carried out to confirm the source of nitrogen in ammonia products.
4.2.3 Transition Metal Carbide
Transition metal carbides have been widely used in hydrogen evolution reactions due to their good hydrogen evolution performance, but researchers have found that they can also catalyze nitrogen reduction well by designing the catalyst structure. For example, the team led by Wang Haihui from South China University of Technology embedded Mo
2C nanodots on ultra-thin carbon nanosheets (Mo
2C/C) with low density, which first proved the feasibility of its catalytic nitrogen reduction in a strong hydrogen evolution environment. The optimal ammonia synthesis rate of the Mo
2C/C is 11.3μg·h
-1·m
, and the Faraday efficiency is 7.8% (Fig. 3A, B)
[79]. This is because Mo
2C nanodots have abundant active centers and unique electronic structures, which are conducive to the adsorption, activation and hydrogenation of N
2. Meanwhile, the Mo
2C nano-dots are embedded on the carbon nano-sheet in low density, which can effectively prevent the active catalyst from falling off and deactivating due to the overflow of hydrogen, reduce the covering of hydrogen on the surface of the catalyst, and increase the probability of nitrogen diffusing to the surface of the catalyst. In addition, the researcher confirmed the important role of protons in the nitrogen reduction reaction (Fig. 3C). When the hydrogen evolution reaction is enhanced, the ammonia synthesis rate will also increase, while the inhibition of hydrogen evolution can improve the Faraday efficiency of NRR, but it will greatly reduce the ammonia synthesis rate. At present, there are few studies on the application of transition metal carbides in nitrogen reduction reaction.Therefore, researchers can develop more carbide catalysts with efficient NRR performance by designing and regulating the structure of catalysts and adjusting the balance between HER and NRR.
4.2.4 Transition metal sulfides
Sulfur plays a very important role in biological nitrogen fixation, so chalcogenides have also attracted wide attention. As a new type of ultrathin two-dimensional (2D) transition metal sulfide nanomaterials, MoS
2 nanosheets have attracted much attention due to their unique structure and properties. Sun Xuping's team at the University of Electronic Science and Technology of China, Chen Liang, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, and Tang Bo's team at Shandong Normal University found through DFT calculations that in the process of NRR catalyzed by MoS
2.The edge Mo atom with positive charge is the active site for activating the N
2 molecule, and the electron can be transferred from the adsorbed N
2 to the edge Mo site and form a N — Mo bond, thereby weakening the N N N bond
[80]. Thus, MoS
2 catalysts with more marginal sites exhibit superior NRR performance under ambient conditions. Based on the theoretical prediction, they grew MoS
2 nanosheet arrays in situ on carbon cloth, and because the vertical structure exposed many edge positions, the catalyst had good NRR activity. In the V vs RHE, the ammonia synthesis rate of the MoS
2 is 8.08×10
-11mol·s
-1·cm
-2 at the potential of-0. 5 V vs RHE, but the Faraday efficiency is only 1. 17% due to the competitive reaction. Therefore, in future studies, researchers can also try to improve the NRR activity of sulfides by regulating the crystal phase and creating sulfur vacancies.
Transition metals and their derivatives have been widely used in the field of electrocatalytic synthesis of ammonia because of their low price and easy synthesis.However, the ability of transition metals to activate N2 is weak, and the electrons in their d orbitals are more likely to combine with hydrogen atoms to form M — H bonds, resulting in competitive reactions. In addition, the catalytic mechanism of transition metal compound electrocatalysts in NRR research is still unclear, and further research is needed.
4.3 Metal-free catalyst
Metal-free materials are favored by researchers because of their low cost, adjustable composition, excellent stability and environmental friendliness. Carbon-based materials have become efficient electrocatalysts with excellent activity and stability due to their excellent conductivity and large specific surface area
[81]. Among them, heteroatom-doped carbon materials can adjust the electronic structure and produce more defects, induce charge transfer in carbon materials, and thus promote the adsorption of reactants on electron-deficient sites
[82]. Sun Xuping's team at the University of Electronic Science and Technology of China synthesized defect-rich fluorine-doped graphene nanosheets by fluorination of graphene. When the material was used as an NRR electrocatalyst, its ammonia synthesis rate was 9.31μg·h
-1·m
, and its Faraday efficiency was 4.2%
[83]. Density functional theory calculations show that the defects caused by the fluorination of graphene not only provide a large number of reaction sites for NRR, but also effectively activate the N
2 molecules. In addition, Zheng Gengfeng's team at Fudan University used boron-doped graphene BC
3 (boron atoms replacing carbon atoms in graphene skeleton) as catalysts for electrocatalytic nitrogen reduction, with an ammonia synthesis rate of 9.8μg·h
-1·cm
-2 and a Faraday efficiency of 10.8%
[84]. Compared with the undoped graphene, the boron-doped graphene samples show better catalytic performance, because the electronegativity of B is less than that of C, so the B doping sites are positively charged, which inhibits the combination of Lewis acid (H
+) and makes it easier to bind Lewis base (N
2). DFT calculations showed that the difference of electron density caused by boron doping had a great influence on the catalytic activity of NRR, and the BC
3 had a suitable nitrogen binding energy (0. 01 eV) and the lowest reaction energy barrier (0. 43 eV). Recently, Ling Tao's team at Tianjin University has systematically studied the electrocatalytic nitrogen reduction potential of carbon materials doped with sulfoxy elements (O, S, Se, Te) through the combination of experiments and theories
[85]. They found that the charge accumulation effect induced by doping atoms can promote the adsorption of N
2 on carbon atoms, and the NRR performance of Te and Se doped carbon catalysts is better than that of other S and O doped catalysts.
In addition, in addition to the single element doping modification, the synergistic effect of double heteroatom doped carbon materials can also significantly improve the electrochemical performance. On the basis of nitrogen-doped porous carbon materials, our group introduced B, F, P and S heteroatoms into nitrogen-doped porous carbon materials, respectively
[86]. It was found that the porous carbon catalyst co-doped with N and B had the best NNR performance, because the presence of B element promoted the formation of B — N bonds, thus improving the NNR activity. In a word, the doping of non-metallic materials can change the local charge and spin density of carbon atoms, and further adjust the adsorption and desorption strength of reactants, reaction intermediates and products on the surface of the catalyst, thus further improving its NRR performance.
In addition to carbon-based catalysts, other metal-free catalysts, such as black phosphorus and boron carbide, have also shown excellent NRR activity
[87][88]. Wang Haihui's team at South China University of Technology used phosphorus, a non-metallic element in the same main group as nitrogen, as the active center of the catalyst to achieve efficient selective reduction of nitrogen. In order to release the intrinsic catalytic activity of elemental phosphorus as much as possible and promote the adsorption and activation of nitrogen, they controllably prepared black phosphorus nanosheets (FL-BP NSs) with phosphene-like structure
[87]. Electrochemical tests showed that the Faraday efficiency of FL-BP NSs for ammonia synthesis from nitrogen reduction was 5. 07% at-0. 6 V vs RHE.The ammonia synthesis rate was V vs RHE
at-0. 7 V vs RHE.The performance of FL-BP NSs was better than that of most metal-based catalysts. In addition, the researchers fully confirmed that all the synthesized ammonia came from the supplied nitrogen by using the nuclear magnetic quantitative method of
15N
2 isotope tracing. DFT calculations show that the active sites of nitrogen reduction are concentrated on the zigzag (zz) and diff-zigzag (diff-zz) edge active sites of the few-layer black phosphorus nanosheets, and the active orbitals and electrons on the zz and diff-zz active sites promote the nitrogen reduction to preferentially hydrogenate via the association pathway.
4.4 Monatomic catalyst
When the size of the active metal is reduced to a single atom, the monatomic catalyst exhibits significantly different properties from those of conventional nanocatalysts. This is because the low coordination unsaturation of a single atom can improve the intrinsic activity of the active site, while the fully exposed atoms can improve the utilization of metal atoms, thus reducing the cost of the catalyst
[89,90]. Nitrogen-doped carbon materials are a good substrate for stabilizing metal monatomic catalysts, which is conducive to improving the stability of catalysts. Therefore, Martin Oschatz team of Max Planck Institute of Colloid and Interface prepared atom-dispersed Au monatomic (Au
1/C
3N
4) catalysts on carbon nitride.The optimum ammonia synthesis rate was 1305μg·h
-1·
$mg^{-1}_{cat}$, and the Faraday efficiency was 11. 1%, which was about 22. 5 times of that of the Au nanoparticle catalyst
[91]. DFT calculations show that there is efficient charge transfer between the single Au atom and the C
3N
4 substrate, and there is effective interaction between the single Au atom and the reaction intermediate.Its protonation reaction has a low energy barrier.It follows the alternative association pathway for nitrogen reductive hydrogenation. Zeng Jie's team at the University of Science and Technology of China successfully loaded Ru single atoms on nitrogen-doped carbon (Ru SAs/N-C) by using spatial confinement strategy
[92]. According to the method, the ruthenium atoms are uniformly dispersed by encapsulating the ruthenium atoms through the molecular pore canals of zeolitic imidazolate frameworks (ZIF-8), and the migration and agglomeration of the ruthium atoms are avoided by anchoring the ruthium atoms through a carrier framework. Compared with the Ru nanoparticle catalyst, the Ru monatomic catalyst has better NRR activity, with the ammonia synthesis rate of V vs RHE and the Faraday efficiency of 29. 6% at-0. 2 V vs RHE potential. This may be due to the large size of Ru nanoparticles and the serious hydrogen evolution on the surface, while the HER process is significantly inhibited when the size of Ru element is reduced to the atomic level. In addition, Sun Zhenyu's team at Beijing University of Chemical Technology found that the V vs RHE in the V vs RHE catalyst could effectively inhibit HER. At -0.21 V vs RHE potential, the synthesis rate of ammonia was as high as 3.665 mg·h
-1·m, and at -0.17 V vs RHE voltage, the Faraday efficiency was 21%
[93]. The results of DFT calculation show that the Ru atoms anchored by the oxygen vacancies on the surface of the ZrO
2 are the active sites, and the free energy of the monodisperse Ru sites adsorbing the intermediate
*NNH is lower than that of the
*H, which is beneficial to the further activation of the N
2 and makes the catalyst exhibit excellent NRR performance.
In addition to the above Au and Ru catalysts, the Pt catalyst with strong hydrogen evolution was brought back to the field of NRR research through the design of single atomization, and monodisperse Pt atoms or nanoparticles were anchored on tungsten trioxide nanosheets by photodeposition (Fig.The Pt monatomic catalyst showed excellent NRR performance (Figure 4 C), with an ammonia synthesis rate of V vs RHE
and a Faraday efficiency of 31.1% at -0.2 V vs RHE potential, which is much better than the activity of Pt nanoparticles, mainly due to the effective inhibition of HER by the Pt monatomic modified tungsten trioxide support
[94]. Combined with electrochemical In-situ infrared spectroscopy (In situ FT -IR) and DFT calculation (Figure 4D), it was found that the Pt site with Pt-3O coordination mode was the active center for nitrogen adsorption and activation, and ammonia was synthesized by nitrogen reduction through the alternative hydrogenation reaction path.
In addition to noble metal monatomic catalysts, Mo, Fe and other monatomic catalysts are also used in NRR research
[95][96]. Zhang Lizhi's team at Huazhong University of Science and Technology optimized the polarization of nitrogen at the interface through monatomic catalysts, promoted the breakage of N N bonds, and improved the efficiency of electrocatalytic synthesis of ammonia
[96]. The researchers used hydrothermal grafting method to load Fe single atoms on the surface of single-layer MoS
2, and the protruding iron atoms on the (Fe SACs-MoS
2),MoS
2 nanosheets can trigger an electric field, which can polarize the adsorbed N
2. Due to the interfacial polarization effect, the Fe SACs-MoS
2 can rapidly inject more electrons into the antibonding orbital of the N
2, which makes the catalyst show excellent NRR activity. In 0.1 mol/L KCl solution, the optimum catalytic activity and selectivity of the V vs RHE catalyst were obtained at-0.2 V vs RHE.The Faraday efficiency was about 31.6%, and the ammonia synthesis rate was (97.5±6)μg·h
-1·cm
-2. In addition, the researchers used similar synthesis methods to prepare Co (Cu, Rh, Ru)-SACs-MoS
2 catalysts, which can enhance the interfacial electric field strength and improve NRR performance, so the researchers believe that the polarization phenomenon triggered by monatomic catalysts is universal.
In summary, the simple and uniform structure of monatomic catalysts facilitates the identification of active sites, providing a tremendous opportunity to explain electrochemical processes and mechanisms at the atomic scale. However, it is difficult to prepare and characterize monatomic catalysts, and the in-situ characterization of components and active sites needs to be further solved.
5 Design of Electrocatalyst for NRR
At present, that main key problem of NRR electrocatalysts are poor activity and low Faraday efficiency. In order to improve the apparent or intrinsic activity of catalysts, researchers have designed and developed many efficient and durable electrocatalysts from the aspects of mass diffusion, electron transfer and surface/interface reactions. Therefore, we summarized the existing design strategies of efficient NRR catalysts, including size control, crystal face adjustment, morphology control and defect engineering, for your reference.
5.1 Size control of catalyst
In general, the size of metal nanocatalysts affects the physical and chemical properties of the materials, thus affecting the catalytic activity and selectivity of the materials. From the point of view of geometric structure, reducing the size of the catalyst exposes more unsaturated metal atoms, thus changing the structure and proportion of the active center of the material. From the point of view of electronic structure, the electronic energy level of metal atoms will change with the quantum size effect, which will affect the orbital hybridization and charge transfer between catalysts and reactants
[97,98]. In addition, reducing the particle size of the catalyst can also reduce the amount of active metal used, improve the utilization rate of the metal and reduce the use cost of the catalyst. For example, Yan Junmin's team at Jilin University synthesized Au nanoparticles (Au/TiO
2) of different sizes supported on TiO
2 by different reduction methods
[99]. Different from the V vs RHE reduction and photoreduction, the gold subnanoclusters synthesized by the reduction of tannic acid (TA) have good electrocatalytic performance for the reduction of nitrogen.The ammonia synthesis rate is 21.4μg·h
-1·m
and the Faraday efficiency is 8. 11% at-0. 2 V vs RHE potential, which is much better than the large-size gold nanoparticle catalyst. At the same time, this study proves that this subnanocluster has good stability, and the NRR reaction mechanism of the Au/TiO
2 cluster is analyzed. Because the Au active center in the Au — O — Ti bonding structure of the catalyst is positively charged, in acidic electrolyte, when the N
2 and H
+ are close to the surface of the catalyst, the active center will preferentially adsorb the electrically neutral N
2, thus inhibiting the HER and making the NRR have higher reaction efficiency. By adjusting the size of the electrocatalyst, low coordination sites can be formed, which can affect the binding strength of the reaction intermediates, thus optimizing the activity and selectivity of the electrocatalyst.
The active sites of the catalyst can be effectively increased by adjusting the crystal planes of the catalyst through favorable atomic arrangement and coordination. However, there is little research on the application of crystal face control in the electro-reduction of nitrogen, so the follow-up research can be carried out in this area.
5.2 Defective engineering of catalyst
In recent years, defect engineering has been considered as an effective strategy to control the performance of catalysts. When the catalyst is rich in defects, its electronic structure will change, resulting in changes in atomic arrangement and adsorption strength of intermediates, thus optimizing surface charge transfer
[100]. At present, the common ways of building defects mainly include vacancy defects and doping defects.
5.2.1 Vacancy defect
Oxygen vacancies (OVs) are anionic vacancies with low free energy of formation, which are usually easy to form in transition metal oxides. The formed oxygen vacancies can be used as active sites for adsorbing reactants and reaction intermediates, thereby reducing the activation energy barrier and promoting the electrochemical reaction
[101]. In addition, the introduction of vacancies can affect the electronic structure of the catalyst, enhance charge transfer, and expose more unsaturated coordination metal sites. In the NRR process, oxygen vacancies can capture metastable electrons and transfer them to the antibonding orbitals of adsorbed N
2 molecules, thus promoting the cleavage of N N N bonds.
For example, based on DFT theoretical calculations, Yan Junmin's team at Jilin University proposed a strategy to optimize the NRR performance of BiVO
4 catalysts by regulating the concentration of oxygen vacancies
[102]. They successfully prepared efficient BiVO
4 catalysts with different OVs concentrations by adjusting the pH value reasonably through a simple hydrothermal method. When the pH value is 7, the oxygen vacancy concentration of the prepared BiVO
4 catalyst is the highest, and the catalyst also has the best nitrogen reduction performance. At the potential of-0. 5 V vs RHE, the rate of ammonia synthesis is V vs RHE
, and the Faraday efficiency can reach 10. 04%, which confirms that the existence of oxygen vacancies plays an important role in nitrogen reduction. In addition, Xu yuanhong's team at Qingdao University used metastable Ti atoms on the surface edge of two-dimensional MXene material Ti
3C
2T
x as nucleation sites to synthesize TiO
2/Ti
3C
2T
x composite catalysts containing oxygen vacancies by one-step solvothermal method, and applied them to electrocatalytic synthesis of ammonia
[103]. It was found that the ammonia synthesis rate of the catalyst was V vs RHE V vs RHE at-0. 55 V vs RHE potential, and the Faraday efficiency was 16. 07% at-0. 45 V vs RHE potential, which was much higher than that of the single Ti
3C
2T
x or TiO
2 catalyst. In addition, they used DFT calculations to study the mechanism of nitrogen reduction of the catalyst and found that the oxygen vacancy in the catalyst is an active site, which can reduce the reaction barrier of nitrogen reduction and gradually generate NH
3 by remote hydrogenation.
The introduction of oxygen vacancies in the electrocatalyst can enhance the adsorption of N2 and optimize the NRR activity. Researchers can use the reported experience of oxygen vacancy construction to develop more efficient catalysts containing oxygen vacancies, thus promoting the development of NRR as a new field.
Nitrogen vacancies (NVs) can affect the electronic structure and energy band structure of nanomaterials, and are in a similar chemical environment with nitrogen atoms in the N
2, so NVs have the ability to enhance the activation of the N
2[104,105].
Yu Guihua's team at the University of Texas at Austin, in collaboration with Chen Gang's team at Harbin Institute of Technology, reported a defect engineering strategy to induce NVs in carbon-nitrogen polymers (PCN-NV
x,x is calcination time in Ar atmosphere), demonstrating the feasibility of their electrocatalytic NRR, with ammonia synthesis rate and Faraday efficiency as high as 8.09μg·h
-1·m
and 11.59%, respectively
[106]. Furthermore, the effect of nitrogen vacancy on the performance of electrocatalytic NRR was confirmed by theoretical calculation and comparative experiments. They propose that the N
2 molecule is chemisorbed on the nitrogen vacancy of PCN, and the lone pair electron of the PCN-NV
x feeds back to the anti-bonding orbital of the adsorbed N
2 molecule, which increases the N N N bond length and leads to the activation of the N
2. In addition, Qiao Shizhang's team at the University of Adelaide synthesized two-dimensional layered NV-W
2N
3 nanosheets rich in nitrogen vacancies by melting method and applied them to the electrocatalytic NRR process
[72]. At the potential of-0. 2 V vs RHE, the ammonia synthesis rate of the catalyst was V vs RHE
, and the Faraday efficiency was (11. 67 ± 0.93)%. Due to the high valence of W atoms and the constraint effect of the two-dimensional nanosheet morphology, the nitrogen vacancies on the W
2N
3 are still stable after a long time of electrochemical nitrogen reduction reaction. Further DFT calculations show that the nitrogen vacancy on the W
2N
3 provides an electron-deficient environment, which effectively promotes the combination of the lone pair electrons of the N
2 with the active site of the nitrogen vacancy, which is not only conducive to the adsorption of nitrogen, but also reduces the thermodynamic limiting potential of NRR, making the W
2N
3 have better NRR catalytic activity. In conclusion, the nitrogen vacancies in the catalyst can effectively induce the adsorption and activation of nitrogen, thus improving the electrocatalytic NRR performance.
Oxygen and sulfur are in the same main group, and they have very similar chemical properties. Therefore, researchers speculate that sulfur vacancies (SVs), similar to OVs, can affect the energy band structure and electronic structure of catalysts, and act as adsorption sites for N2 to induce the activation of N2, thus affecting their electrocatalytic NRR performance.
Sun Xuping's team at the University of Electronic Science and Technology of China synthesized MoS
2 nanoflowers rich in sulfur vacancies by hydrothermal method and applied them to the electrocatalytic nitrogen reduction process
[107]. The ammonia synthesis rate of the catalyst is V vs RHE
and the Faraday efficiency is 8. 34% at the potential of-0. 40 V vs RHE. The results show that the sulfur defects in the MoS
2 can reduce the apparent activation energy, produce more active unsaturated atoms, and have a strong impact on the electronic structure, adjust the activity of the reaction sites, and improve the electrocatalytic NRR performance. Through further DFT calculation, the researchers found that the energy barrier of NRR was reduced on the surface of the MoS
2 catalyst with rich defects, and the N
2 was first activated by the exposed Mo atoms on the edge of SVs, and then gradually protonated according to the mechanism of remote hydrogenation. In order to further improve the Faraday efficiency of the Douglas R. MacFarlane catalyst, Douglas R. MacFarlane team of Monash University constructed the Ru/2H-MoS
2 catalyst reasonably through polymorphic engineering. The Faraday efficiency of ammonia production reached 17.6%, and the ammonia synthesis rate was 1.14×10
-10mol·cm
-2·s
-1[108]. They believe that after the ruthenium cluster adsorbs the N
2, it combines with the H
+ adsorbed by the sulfur vacancy on the 2H-MoS
2 and protonates the N
2 to form the NH
3.
Vacancies (oxygen vacancies, nitrogen vacancies, sulfur vacancies) can effectively regulate the electronic structure and band gap of the catalyst, optimize the charge distribution of the catalyst, and thus enhance the NRR activity. In addition, vacancies can also be used as reactive sites to adsorb N2 and reduce the energy barrier, which is conducive to the rapid catalytic reaction.
5.2.2 Doping defect
In addition to vacancy engineering, the introduction of heteroatoms in the synthesis of electrocatalysts is also an effective way to enhance their electrocatalytic activity. In general, the doping of heteroatoms in catalysts can not only adjust the electronic structure of the catalyst surface, but also change the adsorption capacity of reactants or intermediates, which is conducive to the catalytic reaction
[109~111]. At present, heteroatom doping mainly includes metal atom doping and nonmetal atom doping
[112].
In nature, vanadium is an important active component of nitrogenase, which can bind separate protons and electrons to achieve N
2 immobilization under mild conditions. The adsorption of nitrogen atoms on Ti-based materials is stronger than that of hydrogen atoms, so Ti-based materials have better N
2 reduction ability. In view of this, Sun Xuping's team at the University of Electronic Science and Technology introduced vanadium into the TiO
2(V-TiO
2) and applied it to electrocatalytic nitrogen reduction
[113]. The researchers found that the optimum ammonia synthesis rate is V vs RHE
, and the Faraday efficiency can reach 15.3% at -0.5 V vs RHE potential. They believe that the doping of vanadium promotes charge transport, exposes more active sites, and the synergistic effect of V and Ti atoms on the (110) crystal face of the V-TiO
2 catalyst can enhance the adsorption of
*NH
2NH
2, thus improving the NRR performance of TiO
2. Wu et al. Doped Fe element into the TiO
2, and the synthesized catalyst contained a large number of oxygen vacancies, showing excellent electrocatalytic NRR activity
[114]. In the V vs RHE, the Faraday efficiency of the catalyst is 25. 6% at-0. 4 V vs RHE.The ammonia synthesis rate is 25.47μg·h
-1·m
, which is better than the original titanium-based and iron-based catalysts. With the help of DFT calculation, they believe that the introduction of Fe into the TiO
2 spontaneously increases the number of oxygen vacancies, thus further promoting the activation of N
2, and the synergistic effect of Ti
3+ and oxygen vacancies is the main reason for the high catalytic performance of the catalyst. In addition, our research group used a simple hydrothermal-calcination synthesis strategy to prepare urchin-like Al-doped Co
3O
4 microsphere free-standing catalysts with rich oxygen defects on the surface using nickel foam as the support (Figure 5A) (Figure 5B, C)
[115]. The introduction of aluminum element adjusts the electronic structure of the Co
3O
4 and promotes the generation of oxygen defects to generate more unsaturated active sites, thereby improving the NRR activity and efficiency of the catalyst; The functional urchin-like structure can effectively prevent the accumulation of bubbles and quickly remove small bubbles to avoid the covering of active sites (Fig. 5d), thus increasing the exposure of active sites to nitrogen and promoting the NRR performance of the catalyst. Further characterization by In situ FT IR revealed that the electrocatalytic NRR followed an alternating hydrogenation pathway (fig. 5e, f).
In the study of NRR, in addition to the above common doping of metal oxides, researchers are also actively studying the doping of other compounds, such as MoS
2. The edge of the MoS
2 showed a certain ability of nitrogen reduction, but the basal position had a weak catalytic ability because it was covered by sulfur element, so how to activate the catalytic activity of the basal surface was very important to improve the catalytic activity of the MoS
2. To this end, Lou Jun's team at Rice University, together with Mingjie Liu at Brookhaven National Laboratory, introduced sulfur vacancies into the basal plane of MoS
2, and used doped Co to replace Mo in the lattice to synthesize Co-doped MoS
2 polycrystalline nanosheets rich in sulfur vacancies. The catalyst showed excellent NRR activity, with the best Faraday efficiency of 10% and ammonia yield of 0.63 mmol·h
-1·g
-1[116]. This is due to the fact that Co doping accelerates the adsorption and dissociation of N
2 on the defective MoS
2, thus promoting the NRR process. The synergetic effect of the doping of the transition metal elements and the sulfur vacancy on the basal plane of the MoS
2 can effectively improve the electrocatalytic activity of NRR.
The use of metal doping not only provides an attractive nano-catalyst for electrocatalytic NRR, but also opens up a new way for the rational design of metal compounds containing defects.
(2) doping of non-metal atom
In general, the doping of non-metal atoms can not only adjust the band gap of the catalyst, enhance the charge density and conductivity, but also reduce the binding energy of the reactants and the starting potential required in the electrocatalytic reaction process
[117]. At present, the research on NRR usually chooses N, S, O, B and P atoms as heteroatoms to improve the electrocatalytic performance of the original materials.
For example, researchers have prepared nitrogen-doped porous carbon materials by pyrolyzing ZIF-8 precursors and used them for electrocatalytic reduction of ammonia
[118]. When nitrogen atoms are doped into carbon materials as heteroatoms, the electronic structure around the carbon atoms will change, so that the synthesized catalyst materials contain more defects and can adsorb N
2 molecules more strongly. The influence of pyrolysis temperature and time on the amount and location of nitrogen doping in the synthesized porous carbon materials was explored, and it was determined that the nitrogen-doped porous carbon prepared at 1100 ℃ for 1 H had the best NRR activity. In 0. 1 mol/L KOH solution, at -0. 3 V vs RHE potential, the V vs RHE synthesis rate reached 3.4×10
-6mol·cm
-2·s
-1, and the Faraday efficiency was 10. 2%. Further DFT calculations show that the N
3 sites containing three pyridine nitrogen atoms and the adjacent carbon vacancy defect sites in the carbon skeleton of the catalyst are the active sites for the nitrogen reduction reaction, which can effectively adsorb N
2 molecules and promote the cleavage of N N N, thus promoting the subsequent protonation process. In addition, the team also studied the effect of Fe and N co-doping on NRR activity in this system. Unlike the Oxygen reduction reaction (ORR) and the carbon dioxide reduction reaction (Oxygen reduction reaction), the Fe-N
4 sites in the porous carbon are not conducive to the nitrogen reduction process, which may be due to the Fe-N
4 blocking the formation of active sites for N
3 and promoting the competitive reaction, thus inhibiting the NRR reaction.
In addition to the common nitrogen doping, Martin Oschatz of the Max Planck Institute for Colloid and Interface, based on the multi-functional redox properties of titanium oxide containing oxygen vacancies, designed and synthesized a new carbon-doped titanium oxide (C-Ti
xO
y/C) electrocatalyst rich in oxygen defects by pyrolyzing the metal-organic framework precursor MIL-125 (Ti)
[119]. The catalyst has good electrocatalytic NRR performance, the optimal Faraday efficiency is 17.8%,NH
3, and the synthesis rate is 14.8μg·h
-1·mg
-1. It was found that when carbon atoms were doped into the catalyst, some oxygen vacancies were replaced and new Ti — C bonds were formed, which made the catalyst have a wider energy band structure and provided a large number of active sites for the activation of N
2, thus improving the overall electrochemical performance of the catalyst.
The electronic structure of the catalyst can be adjusted and the catalytic activity of the catalyst can be improved by doping and modifying the electrocatalyst with non-metallic elements. However, there are few studies on the modification of NRR catalysts by non-metal doping, and most of the studies focus on the doping of carbon materials. Therefore, researchers can develop other non-metal doped transition metal compounds to provide more active sites by using the synergistic effect of doping elements, thereby improving the activity of the catalyst.
5.3 Morphology control of catalyst
When the catalyst has a specific morphology and nanostructure, it can provide a larger specific surface area and increase the high active site density of the catalytic process. Studies have shown that some special morphologies can maximize the utilization of active centers and better control the mass transport and reaction pathways, thus enhancing the catalytic effect.
In general, the reactants react with the atoms on the outer surface of the nanocatalyst, while for the nanocatalyst with a hollow structure, the catalytic reaction can take place not only on the outer surface of the catalyst, but also on the inner surface. In view of this, Nazemi et al. Prepared hollow gold nanocages (AuHNCs) catalysts with large specific surface area
[120]. In the V vs RHE solution, the ammonia synthesis rate reached V vs RHE at-0. 5 V vs RHE potential, and the Faraday efficiency reached 30. 2% at-0. 4 V vs RHE potential. Compared with gold nanospheres, gold nanorods and gold nanocubes with similar size, the hollow gold nanocages showed better catalytic activity for NRR. This is mainly due to the unique hollow structure of AuHNCs, which can provide a larger active surface area for the reaction, resulting in more active sites exposed. At the same time, the nano-cage can limit the reactants in the cage, increase the residence time of N
2 molecules on the inner surface of the nano-cage, increase the steady-state concentration of N
2 in the cage, and promote the collision times of N
2 on the inner surface, which is more conducive to the conversion of N
2 into NH
3. In addition, the research team studied the effects of pore density and Ag content on the electrocatalytic NRR activity of hollow Au nanocages by adjusting the peak localized surface plasmon resonance from 635 to 795 nm. They found that the presence of Ag in the cavity reduced the electrocatalytic NRR activity of hollow Au nanocages due to the high activity of Ag for HER. Increasing the pore size of the nanocage structure will reduce the N
2 confined in the nanocage cavity, which will reduce the NH
3 yield and Faraday efficiency
[35].
Porous metal materials are considered to be promising electrocatalysts because of their cross-linked nanoscale framework and pores, which can provide large accessible surface area, transport channels and sufficient active sites
[121,122]. Therefore, Wang Liang's team of Zhejiang University of Technology used the diblock copolymer (PVP-co-PS) micelle as a soft template to prepare a self-supporting gold/nickel foam (pAu/NF) with a cross-linked three-dimensional porous structure by micelle-assisted electrodeposition
[36]. The active component of pAu/NF is Au, and due to its special structure, it shows excellent activity and selectivity for NRR in V vs RHE solution under ambient conditions, with a NH
3 yield of 9.42μg·h
-1·cm
-2 and a Faraday efficiency of 13. 36% at-0. 2 V vs RHE potential. The porous metal film can be directly prepared on the metal substrate by the micelle-assisted electrodeposition method, which has very important guiding significance for the design of other NRR self-supporting electrocatalysts in the future.
5.4 Amorphous control of catalyst
Amorphous materials generally refer to disordered materials that lack crystalline characteristics, have many dangling bonds, and are in a metastable state. Amorphous materials contain many unsaturated coordination sites, which can be used as active sites for reaction. It has been found that many amorphous catalysts exhibit better catalytic performance than their crystalline forms in HER and ORR
[122~124]. Therefore, researchers believe that amorphous catalysts can enhance the adsorption of N
2, thereby improving NRR performance.
Chen Gang's team at Harbin Institute of Technology synthesized amorphous Bi
4V
2O
11/CeO
2 composite catalyst (BVC-a) with nanotubular structure by electrospinning
[126]. They believe that CeO
2 induces Bi
4V
2O
11 to form amorphous structure and produce a large number of defects, so amorphous Bi
4V
2O
11 contains abundant unsaturated bonds, which can reduce the energy barrier of NRR and provide sufficient catalytic sites for nitrogen fixation. In addition, the energy band alignment of the two semiconductors of the CeO
2 and the Bi
4V
2O
11 in the composite catalyst not only provides abundant active sites, but also promotes the transfer of interfacial charges from the CeO
2 to the Bi
4V
2O
11, thereby accelerating the reaction rate. Therefore, the BVC-a catalyst showed better NRR performance, the ammonia synthesis rate was 23.21μg·h
-1·
, and the Faraday efficiency was 10. 16%, which was significantly better than the crystalline phase composite catalyst and the single phase catalyst.
In addition, Yan Junmin's team at Jilin University also confirmed that the NRR activity of amorphous catalysts is better than that of crystalline catalysts
[127]. At room temperature, they synthesized amorphous Au nanoparticles anchored on reduced graphene oxide (a-Au/CeO
x-rGO). Among them, rGO acts as a substrate to anchor and disperse Au nanoparticles, and the CeO
x plays a key role in the process of inducing the amorphization of Au. Under the potential of-0. 2 V vs RHE, the ammonia synthesis rate of the catalyst was V vs RHE, and the Faraday efficiency was 10. 10%, which was significantly better than that of the corresponding crystalline material (c-Au/rGO). This is because the amorphous Au nanoparticles can induce the formation of more defect sites, which become the reactive sites of NRR, resulting in better catalytic activity. However, there is little research on amorphous materials in the current reports on NRR electrocatalysts, so researchers can pay more attention to the development of various amorphous materials and try to apply them to the electrocatalytic reduction of N
2.
6 Conclusion and prospect
Based on the latest progress of electrocatalytic synthesis of ammonia, the research progress of electrocatalytic nitrogen reduction catalysts in recent years is summarized in this paper.The design strategies of efficient NRR electrocatalysts under mild conditions are summarized in detail, which can provide guidance and reference for the selection of catalytic materials, structure design and electrolyte for ammonia synthesis by electrocatalytic nitrogen reduction in the future. Based on the above discussion, the NRR activities of different electrocatalysts under ambient conditions were compared, as shown in Table 1. Despite the tremendous progress in the development of NRR in the past few years, the design of effective NRR catalysts and the construction of excellent NRR electrocatalytic systems are still challenging, including high overpotential, low efficiency and catalyst instability. In order to develop the electro-catalytic nitrogen reduction technology for ammonia synthesis to a more practical level, the following aspects need to be focused on in the future:
(1) Study on the mechanism of NRR by theoretical calculation: The process of electrocatalytic nitrogen reduction was studied at the molecular or atomic level by combining experimental and theoretical calculations. In the current research, although some studies combine theory with experiment, most of the theoretical calculations focus on the free energy changes of active intermediates and the energy barriers of rate-limiting steps, while the adsorption energies of these N2 are assumed to be calculated in vacuum.The influence of the decisive factors of the electrochemical system (such as temperature, pH, mass transfer rate, proton supply, N2 solubility, etc.) Is neglected, which is quite different from the actual experimental operating conditions. Therefore, it is necessary to further improve the calculation method and model of NRR on the surface of heterogeneous catalysts, and to combine theoretical calculation with experiments to provide further guidance for the structural design of electrocatalysts.
(2) Reasonable design of catalyst and electrochemical system: At present, the catalytic activity and selectivity of NRR electrocatalyst for ammonia synthesis in aqueous solution are very low, one of the main factors is the competition of side reactions. Therefore, inhibition of HERs can significantly improve the catalytic activity and selectivity of NRR. Specifically, optimizing the size and morphology of the NRR electrocatalyst can create favorable coordination sites and affect the binding strength of reactants or key intermediates on the catalyst surface. Construct the defect engineering of the catalyst, capture metastable electrons through vacancies, and transfer them to the anti-bonding orbitals of the adsorbed N2 to promote the breakage of N N N bonds; It is also hopeful to construct stress engineering to regulate the surface atomic distance and bond length, and to change the electronic structure of the electrocatalyst, thus promoting the nitrogen reduction reaction. In addition, for future research, while exploring effective catalysts and building excellent electrocatalytic systems, we should also pay attention to efficiency and stability.
(3) Correct validity of product analysis: In current NRR studies, nitrogen isotope labeling experiments or strict blank control experiments are needed to prove the source of nitrogen in the NH
3 produced in the electrochemical system
[128]. ① First of all, it is necessary to remove other nitrogen pollution in the nitrogen, such as NO
x,
,
, etc. Therefore, it is necessary to purify the raw gas, which can be purified by washing with acid and alkali in turn before the reaction. (2) The blank control experiments should be done strictly, such as the comparison of catalyst electrolysis under nitrogen and argon, the comparison of catalyst under open circuit voltage, and the performance comparison of the carrier without catalyst. ③ When nitrogen atoms exist in the reaction system itself, isotope labeling experiments must be carried out to prove that the nitrogen in the produced ammonia really comes from the applied nitrogen molecules. By using isotopically labeled
15N
2, the origin of the nitrogen can be demonstrated and quantified using mass spectrometry, NMR techniques, and the like. In contrast, if the catalyst itself does not contain nitrogen atoms in principle, and nitrogen-containing reagents or surfactants are used in the synthesis, it is still necessary to prove the origin of the nitrogen atoms in the NH
3 produced, so that the use of nitrogen-containing chemicals in the synthesis is avoided as much as possible and contamination during washing and drying of the catalyst material is excluded.
(4) In-situ characterization of the NRR electrocatalyst in the electrocatalytic system: It is difficult to capture and characterize the catalytic reaction intermediates by experimental methods because of the complex structure of the intermediates in the NRR process and the short lifetime of the intermediates. In situ characterization of catalytic intermediates can effectively improve the reaction pathway and help to better understand the reaction mechanism and principle of NRR. In addition, the catalytic active sites on the surface of the catalyst can be established by in situ characterization, which provides guidance for the development of effective NRR electrocatalysts. However, in the current research, there are few studies to explore the real-time changes of the adsorbed intermediates on the catalyst surface in the NRR process under the operating conditions. Therefore, researchers should develop more in situ techniques to track and detect the generation and changes of reaction intermediates in real time, and further reveal and explain the catalytic mechanism scientifically and reasonably.
The current advances in the field of NRR are sufficient to demonstrate the feasibility of the electrochemical reduction of N2 to NH3 in aqueous solutions under ambient conditions. This new field of research has just started, and faces enormous challenges in both the design and preparation of catalysts and the precise and rigorous detection of product ammonia. However, we firmly believe that with the development and utilization of in-situ characterization technology and the continuous improvement of theoretical system, it will further promote the development of NRR system with high efficiency, high selectivity and high stability, and promote the early realization of the industrialization of artificial green ammonia synthesis.
National Natural Science Foundation of China(21421001)
National Natural Science Foundation of China(21573115)
National Natural Science Foundation of China(21875118)
PDF
