The use of single-metal atom-based photocatalysts for the production of ammonia through photocatalytic nitrogen fixation

Abstract

The conventional synthetic ammonia industry is characterized by its high energy consumption, necessitating the exploration of a new environmentally sustainable method for NH₃ synthesis. The photocatalytic nitrogen reduction reaction (pNRR) allows NH₃ production under room conditions. The optimization of photocatalysts, particularly through the use of single-metal atom catalysts (SMACs), plays a significant role in enhancing the performance of pNRR. SMACs have garnered growing attention in photocatalysis for their exceptional catalytic activity, selectivity, stability, and complete atom utilization. These catalysts involve isolated atoms supported on substrates without aggregating into nanoparticles. In this review, we elucidate the mechanisms and pathways of pNRR, focusing on the latest advances in carbon-based and non-carbon-based SMACs, and conclude with an overview of the existing challenges and prospects of pNRR for sustainable ammonia production.

---

Ping Zhang

Ping Zhang is a professor at the School of Chemical Engineering, at Northwest Minzu University. In December 2014, he majored in organic chemistry at Northwest Normal University and obtained a doctor of science degree. Main research fields: photocatalysis and environmental treatment; photocatalytic CO2 reduction; study of photocatalytic nitrogen reduction. He has presided over more than 10 scientific research projects and published more than 50 papers. He has won the 9th Gansu Youth Science and Technology Award, the Young Teachers’ Talent Award in Gansu Province, and the Young Innovative and Entrepreneurial Talents in Longyuan.

Yongchong Yu

Yongchong Yu received a bachelor's degree from Guizhou Normal University in June 2022. He is currently a master's student of Professor Ping Zhang's at the School of Chemical Engineering of Northwest Minzu University. The main research directions are the design and synthesis of nanomaterials, photocatalytic N2 reduction, photocatalytic pollutant degradation, and photocatalytic CO2 reduction.

1. Introduction

NH₃ is a pivotal chemical compound globally, crucial in producing nitrogen fertilizers essential for ensuring an ample food supply for human societies worldwide.^1–3 Apart from its widespread application in agriculture, ammonia finds utility across various domains like military operations and chemical engineering, while also being recognized as an efficient renewable fuel source.^4–6 The significance of ammonia is underscored by its extensive large-scale production over the past century. Notably, the industrial manufacturing of ammonia predominantly hinges on the Haber–Bosch process, which was first introduced in the early 1900s. However, this process operates at elevated reaction temperatures (673.15–773.15 K) and pressures (100–200 atm), necessitating substantial reliance on fossil fuels.^7–12 Furthermore, the substantial hydrogen (H₂) demand in the Haber–Bosch process is met through steam-reforming natural gas (CH₄ + 2H₂ → 4H₂ + CO₂), contributing to an annual release of approximately 1.6% of CO₂ emissions into the atmosphere.^13,14 Such carbon dioxide discharges have the potential to heighten the prevailing greenhouse effect and pose environmental challenges, diverging from the pursuit of sustainable development goals such as carbon neutrality and peak carbon emissions. Hence, the quest for an environment-friendly, economical, and sustainable way to achieve green synthesis is of paramount importance.

A great number of methods have been explored for ammonia synthesis, such as photocatalysis, biocatalysis assisted, photoelectrocatalysis combined, photothermal catalysis combined, and electrocatalysis. Since 1977, Schrauzer and Guth¹⁵ have been at the forefront of converting N₂ to NH₃ on TiO₂ surfaces under ultraviolet light irradiation, contributing significantly to understanding TiO₂-based nitrogen-fixing catalysts. In the 21st century, extensive efforts have been made towards advancing photocatalytic nitrogen fixation. The process relies on the clean, cost-effective, readily available light, water, and air components, as demonstrated by the following equation.

In theory, photocatalytic nitrogen fixation (PNF) is regarded as an energetically favorable process involving water splitting and N₂ fixation, driven by solar energy. This approach presents a promising avenue for sustainable ammonia production without energy losses during power generation.^16,17 By avoiding greenhouse gas by-products, PNF serves as a green alternative to conventional ammonia synthesis, offering a solution to environmental concerns and energy scarcity. Consequently, the employment of PNF emerges as a clean, energy-efficient, and environmentally benign strategy for NH₃ production and environmental preservation.^18,19 In recent years, the burgeoning interest in diverse semiconductor designs has propelled photocatalysis to the forefront of research in the field of energy and chemical fuel production, particularly for harnessing the boundless and eco-friendly solar energy derived from sunlight.^20,21 This encompasses a wide array of applications such as the photocatalytic treatment of wastewater,²² which includes the breakdown of dyes, antibiotics like tetracycline,^23,24 organic and inorganic pollutants, and pesticides.^25–27 Additionally, photocatalysis has been instrumental in processes like hydrogen generation,^28,29 CO₂ reduction,^30,31 and so on. The current landscape boasts a plethora of semiconductor materials, ranging from transition metal compounds,^32,33 non-metallic polymers,^34,35 metal–organic frameworks,^36,37 layered double hydroxides,^38,39 and single-metal atom catalysts,^40–44 all of which have demonstrated considerable efficacy in facilitating photonitrogen fixation.

In 2011, Zhang et al. introduced a groundbreaking catalyst system by developing a metal platinum single-atom Pt/FeOx catalyst, thereby pioneering the concept of a single-metal atom catalyst (SMAC).⁴⁵ Following this seminal work, scholars have extensively explored the realm of metal single atoms. A distinguishing characteristic of SMAC is the homogeneous dispersion of the co-catalyst (comprising precious metal/non-precious metal) as individual atoms across the surface of the semiconductor carrier. These isolated single atoms serve as active centers, thereby maximizing atomic efficiency. Distinct from conventional metal–metal bonding, single atoms primarily form bonds with heteroatoms present on the carrier.⁴⁶ Recently, various SMACs have been successfully synthesized, prompting thorough investigations into the active center characteristics and photocatalytic reaction mechanisms.⁴⁷ SMACs exhibit distinctive unsaturated coordination surroundings and electronic configurations, setting them apart from nanoparticles, clusters, and bulk catalysts. These particular features afford SMACs notable benefits in reactant adsorption/activation, carrier separation, and catalytic performance.⁴⁸ For example, He et al.⁴⁹ used Au single-atom isomorphic porphyrin-based COFs (COFX Au, X = 1–5) for pNRR. The COFs anchored with Au atoms exhibited high NH₃ generation activity (427 μmol g⁻¹ h⁻¹), Similarly, Hu and colleagues⁴² fabricated single-atom Fe-porous g-C₃N₄ (FPx) samples through a one-step annealing method employing bubble template and direct metal atomization, eliminating the demand for sacrificial agents or cocatalysts. The FPx material significantly improved photocatalytic performance, achieving a rate of 62.42 μmol g⁻¹ h⁻¹, which was more than five times higher than the photocatalytic activity of bulk g-C₃N₄. In summary, SMACs present a range of compelling strengths. (1) The unsaturated coordination conditions and distinct electronic structure can influence reactants, intermediates, and products’ adsorption/desorption behavior and intensity, thus enhancing catalytic efficiency and selectivity. (2) Theoretically, isolated SMACs have the highest atomic utilization efficiency, rapidly expanding operational locations and mitigating metallic usage. (3) SMACs can be regarded as a versatile stage for exploring the structure–activity relationship in heterogeneous catalysis at the atomic and molecular levels. (4) SMACs can ensure an even distribution of active sites while maintaining the stereostructural characteristic of heterogeneous catalysts, positioning this as an optimal approach for the sustainable advancement of catalytic reaction technologies.^{46,48,50–53}

The utilization of SMACs as photocatalysts in the pNRR process has been extensively studied due to their numerous benefits. For example, Li and colleagues successfully developed a well-defined Pt–N₃ single-atom catalyst (Pt SACs/CTF) by incorporating Pt into the N₃ position within two-dimensional ultra-thin covalent triazine framework (CTF) nanosheets. This catalyst exhibited a notable photocatalytic efficiency of 170.4 μmol g⁻¹ h⁻¹ for ammonia production without sacrificial agents.⁵⁴ A critical aspect of the pNRR process lies in developing effective photocatalyst materials. SMACs, known for their exceptional activity and controlled charge transfer pathways, were shown to enhance the effectiveness of photocatalytic nitrogen fixation for ammonia synthesis. The introduction of single-metal atom attachment on metal–organic frameworks via defects was demonstrated to significantly improve the efficiency of the pNRR process. Additionally, the interaction between single-metal atoms and carriers was highlighted as equally crucial for pNRR. Ren et al.⁵⁵ illustrated this by creating defects within the UIO-66 (Zr) framework and introducing Ru single-atoms on the framework, establishing electron metal–support interactions (EMSI) through covalent bonds that enhanced charge transfer efficiency and substantially boosted photocatalytic yield. Researchers addressed the challenge of low reaction activity in the pNRR process through the selective design of catalytic materials, the construction of suitable active sites to lower the reaction activation energy, and the coupling of semiconductor band structures to enable efficient transfer of photogenerated charge carriers. This approach effectively improved the efficiency of pNRR for ammonia synthesis.

In this paper, we summarize the recent progress in research on applying single-metal atoms in pNRR. The basic principle of pNRR is introduced, and the mechanism of metal single atoms anchoring on carbon-based and non-carbon-based photocatalysts to improve nitrogen fixation performance is also discussed. Finally, the shortcomings and prospects of single-metal atom photocatalysts in pNRR are discussed.

2 The reaction process and mechanism of pNRR

2.1. Basic process of the pNRR

Before photocatalytic reactions occur, it is essential to transport N₂ molecules from the external conditions to the catalyst's surface, where N₂ interacts with the catalyst and adsorbs. By reducing the reaction energy barrier using a catalyst, ammonia synthesis can be catalyzed under the energy provided by light. The reaction process of pNRR to synthesize ammonia mainly involves the following steps: (1) when sunlight irradiates the photocatalyst, the energy of visible light (hν) is greater than the catalyst's band gap width (E_g), which excites electrons (e⁻) from the valence band to the conduction band (CB), creating positive holes (h⁺) in the catalyst, forming electron–hole pairs inside the catalyst;^56–59 (2) some of the electron–hole pairs recombine inside or on the surface of the catalyst, while the rest diffuse to the catalyst surface under chemical potential and react with substances adsorbed on the surface in redox reactions. In the oxidation reaction, due to the valence band (VB) holes being strong oxidants (1.0–3.0 V vs. NHE), water combines with the holes and gets oxidized, producing protons and oxygen; the CB electrons act as strong reducing agents (0.5 to −1.5 V vs. NHE), which combine with N₂ molecules and protons to be reduced by the photogenerated electrons, resulting in the generation of ammonia molecules.^2,60–63 The produced ammonia eventually desorbs from the surface of the catalyst, diffusing externally into liquid or gas phase systems.

The occurrence of photocatalytic redox reactions is contingent on the reduction potential of the adsorbate and the positioning of the adsorbent, specifically, the band position within the semiconductors, as illustrated in Fig. 1.^64–67 The conduction band of a semiconductor must possess a greater energy level than the reduction potential of N₂, whereas the valence band should be lower than the oxygen evolution potential. The critical energy transition states were identified in the initial electron transfer (N₂ + e⁻ → N₂ −4.16 V vs. NHE) and proton-coupled electron transfer (N₂ + H⁺ + e⁻ → N₂H −3.2 V vs. NHE) processes, which significantly influence the overall kinetic reaction.^16,68,69 These factors present primary obstacles that must be addressed to activate the N₂ molecules for NH₃ formation. It is imperative to maintain a small bandgap in semiconductors, ideally in the visible light range, as it can also meet the thermodynamic reduction potential of N₂ to NH₃.^70–72 Additionally, it is imperative to inhibit the recombination of charge carriers within semiconductor photocatalysts to optimize the reaction's solar conversion efficiency and apparent quantum yield (AQY) of the response.

Fig. 1 The schematic diagram of the photocatalyst for converting N₂ to NH₃ and its oxidation–reduction potential (V vs. NHE, pH = 0) is marked on the left side.¹⁶

2.2 The mechanism of pNRR

This process could have followed either a N₂ reduction reaction (NRR) or a N₂ oxidation reaction (NOR). The pNRR process involves several steps. When exposed to light, the conduction band (CB) energizes electrons, creating holes in the valence band (VB). Some of these electrons and protons are released into the system. Meanwhile, the additional photogenerated holes (h⁺) facilitate the oxidation of water to produce H⁺ and O₂ (eqn (1)). Simultaneously, the reduction of hot electrons leads to the production of NH₃ from N₂ (eqn (2)). Consequently, NH₃ is formed under controlled conditions by utilizing sunlight as the energy source (eqn (3)).

Water splitting: 2H₂O + 4h⁺ → O₂ + 4H⁺ (1)

Ammonia synthesis: N₂ + 6H⁺ + 6e⁻ → 2NH₃ (2)

Overall reaction: 2N₂ + 6H₂O → NH₃ + 3O₂ (3)

Fig. 2a illustrates the various mechanisms of N₂ photofixation observed under alkaline conditions (pH 14). The primary emphasis of nitrogen fixation lies in converting N₂ to NH₃, which involves dissociative and associative pathways, and interaction of reaction intermediates.^73–76 The dissociative way (Fig. 2a, path 1) involves the initial activation of the triple bond in the N₂ molecule. This activation leads to the generation of two separate atoms on the surface of the catalyst that undergo individual protonation to yield NH₃, mirroring the process used in conventional NH₃ production methodologies. Given that the breaking of the N≡N triple bond requires an energy of 9.79 eV (945 kJ mol⁻¹), this specific pathway under consideration manifests a significant energy barrier for reactions, posing a formidable challenge in maintaining its presence on the surface of a catalyst under moderate environmental conditions.^76,77

Fig. 2 (a) Different NH₃ evolution routes and their intermediates (b) their spherical and rod-shaped representations.^84,85

Implementing this procedure becomes particularly intricate when dealing with materials characterized by layered structures. In contrast to the dissociative system, the associative system operates akin to a chemical mechanism such as the enzymes involved in N₂ fixation. This process involves gradually preparing adsorbed N₂ molecules that link two atoms before forming NH₃.⁷⁸ The hydrogenation step in the associative approach can be broken down into associative distal pathways and other associative paths 2–8, as illustrated in Fig. 2, confirming the nitrogen-reduction reaction ways need an initial nitrogen adsorption configuration. Generally, there are two kinds of N₂ adsorption patterns: side-mounted and end-mounted. The associative distal (Fig. 2a, path 2) and alternative associative pathways (Fig. 2a, path 3) are accessed through the side-mounted configuration. In contrast, the associative enzymatic route is achieved via the end-mounted configuration (Fig. 2a, path 4). The formation of N₂ configurations proved to be intricate without the application of advanced in situ FTIR and first-principle computations that provided further insights into the adsorption configurations and mechanisms. The surface atoms of the catalytic adsorbent were found to predominantly participate in the hydrogenation process, leading to the formation of a distant NH₃ molecule. Subsequently, a second NH₃ molecule was generated as another atom that participated in the hydrogenation process (path 2). Alternatively, two atoms undergo hydrogenation through proton-coupled electron transfer resulting in the sequential liberation of two NH₃ molecules in the final step (path 3). Remarkably, the hydrogenation of the intermediate *NNH₂ (where * represents the N₂ adsorption site) can occur through either the distal way or an alternative pathway leading to the formation of *HN–NH₂ (path 6). Following this, in (path 3), the intermediate *HN–NH₂ may proceed along the distal pathway, resulting in the formation of *NH with the release of an NH₃ molecule (path 8). At potentials below 0.71 V, it is anticipated that the *NNH₂, *NHNH₂, and intermediate hydrazine (*NH₂NH₂) intermediates will dissociate, giving rise to several reaction paths (paths 5, 7, and 9). Back and Jung's calculations (Fig. 2a) have shown that the Watt and Chrisp technique was previously utilized for the identification of hydrazine synthesis or the comparison of Gibbs values of *HN–NH and *NH₂NH₂, potentially originating from the protonation of *N₂H.^78,79 Moreover, the enzymatic method (path 4) operates on a hydrogenation mechanism akin to the alternate pathway, where both atoms adhere to a catalyst surface in an end-mounted coordinating configuration, which is a rarity among heterogeneous catalysts. Li et al.⁷⁹ explored an additional route known as the Langmuir–Hinshelwood (LH) mechanism. In brief, H₂ was initially adsorbed onto the surface of a nitrogen fragment. For specific transition metal nitride (TMN) catalysts, the Mars–van Krevelen mechanism is advantageous for the electrolytic synthesis of ammonia. Furthermore, the atoms within the TMN lattice continually combine with proton–electron pairs (PEPs) to produce NH₃ molecules, consequently producing vacancies on the surface.^80–83 The vacancies were effectively absorbed chemically and stimulated N₂ molecules to facilitate the ensuing nitrogen reduction reaction (NRR). Generally, the prediction of multiple electrons and the process of hydrogenation are deemed unfavorable because of the complexities of different bonding mechanisms and reaction pathways, which continue to pose challenges. Consequently, as illustrated in Fig. 2, nine alternative reaction pathways were investigated based on innovative mechanistic investigations conducted through first-principles calculations.

The dissociative process depicted in Fig. 2b, involving N≡N triple-bond cleavage, requires a substantial amount of energy input as the initial step in the Haber–Bosch process. Two distinct associative mechanisms for N₂ reduction have been proposed: distal and alternating pathways, each of which suggests a unique intermediate formation. In the alternating pathway, two hydrogen atoms are added sequentially to N₂, resulting in the release of the first ammonia and the formation of a hydrazine-bound intermediate, as shown in Fig. 2b, after 4 hydrogenation steps. In contrast, within the distal route, a sole nitrogen atom experiences a series of three hydrogenation reactions to form the initial NH₃ molecule, while the remaining nitride –N atom undergoes three hydrogenation processes to generate a second NH₃ molecule. The examination of thermodynamics in pNRR can be classified into 2 distinct types: whole-reaction and half-reaction thermodynamics, both of which remain unaltered by the presence of a catalyst. However, photon-induction mechanisms rely on specific photocatalysts. Comer and colleagues⁸⁶ undertook a comprehensive thermodynamic study of prevalent nitrogenous substances through the utilization of explicit entropy and enthalpy to evaluate the viability of diverse N₂ fixation techniques. It is possible to estimate the free energy and anticipate the initial potential of the reaction by determining the adsorption energy and vibrational energy of the potential intermediates formed during N₂ reduction. In the photocatalytic process, various constraints are imposed by photocatalysts. Achieving a reduction potential of 4.2 V vs. NHE is essential to disrupt N₂ bond breakage on the photocatalyst surface, which is a task that proves challenging for most semiconductors. Conversely, many electrical strategies offer more practical alternatives.^87,88

Additionally, the conversion of N₂ into a singular nitrogen molecule, such as ammonia or nitrate, is relatively straightforward in terms of thermodynamics. Conversely, converting diazo compounds into single nitrogen compounds presents specific challenges. Nitrogen fixation, classified as a reduction or oxidation reaction, may produce nitrogen reduction and nitrate oxidation products when titanium-based photocatalysts are used under ambient conditions.⁸⁹ Consequently, discerning the predominant oxidation or reduction processes becomes complex. The transformation of redox products involves a temporal dimension, leading to complex pathways between oxidation and reduction reactions. Several studies have documented the oxidation of NH₃ under room conditions. Photocatalysis in such settings may aid in the oxidation of reduced products, as the shift from reduced to oxidized forms requires a driving force. Nonetheless, this process requires lower voltage and thermodynamic energy than traditional nitrogen fixation techniques.^90,91

In proton environments, the hydrogen evolution reaction (HER) constantly challenges the nitrogen reduction reaction (NRR). Certain intriguing materials exhibiting significant nitrogen catalytic capabilities are posited to possess hydrogen binding energies approaching the optimal threshold for facilitating hydrogen evolution, thus potentially aiding the facilitation of competitive HER processes. Consequently, in the pursuit of metal co-catalysts for nitrogen reduction Norskov⁹² group investigated the correlation between the catalytic efficacy of N₂ reduction and intermediate binding energies, employing density functional theory (DFT) to scrutinize FCC metals. Notably, the HER displays a higher positive limiting potential than the NRR.^93,94

The limitation potential discrepancy between the HER and NRR is less pronounced on the (111) facet than on the (211) facet. A significant potential difference exceeding 0.4 V was observed between these two reactions, indicating a level of selectivity in the HER process (Fig. 3). It is crucial to impede the HER while advancing N₂ reduction catalysts. The critical step is the weakening of the metal-to-nitrogen bond during the reductive adsorption of N₂ to *N₂H. The reaction rate is determined by either the hydrogenation of *NH to *NH₂ or the transformation of *NH₂ to NH₃, contingent on the strength of the metal-to-nitrogen bond. An effective heterogeneous nitrogen reduction catalyst should exhibit strong nitrogen affinity, facilitating rapid nitrogen reduction while maintaining a low affinity for intermediate species.

Fig. 3 (a) The restriction of the achievable NRR is examined about the binding energy of *N on (111) and (211) crystallographic planes of different transition metal substrates; (b) a scholarly comparison was conducted analyzing the potential volcano plots’ limiting potential of HER (depicted in blue) and NRR (depicted in black) on the (111) and (211) surfaces across diverse transition metals.⁹²

3 Preparation process and formation mechanism of SMACs

This study focuses on preparing SMACs based on semiconductive supports, given detailed coverage of the various synthetic approaches for SMACs in the existing literature.^95–97 The vulnerability of active sites in SMACs to relocating and forming clusters or nanoparticles due to high surface energy poses challenges in stabilizing them at high loading levels. To address this, innovative methods for creating SMACs have been briefly outlined.

3.1 Wet chemical method

Wet chemical methods employing uncomplicated processes hold potential for the large-scale synthesis of SMACs. Typically, these methods involve two processes to attach SMAs to the catalyst support. In the liquid phase, the hydrothermal method is employed to affix metal atoms to the support or reduce metal ions to individual atoms. The fabrication of SMACs involves utilizing freeze-drying techniques and gas-phase reduction methods. For example, Li et al.⁹⁸ devised an N-coordination approach for creating Fe–N₅ sites uniformly dispersed on N-doped porous nanofibers (Fe SA/PNCNFs-0.1) as a high-performance catalyst through a wet chemical synthesis method (Fig. 4a). In detail, porous N-doped carbon nanofibers (PNCNFs) were meticulously crafted through a series of electrospinning, oxidation, and pyrolysis procedures. Subsequently, integrating the Fe source onto PNCNFs facilitated thorough contact between the fiber matrix and Fe species. This was achieved by preabsorbing PNCNFs onto Fe species through a wet impregnation method. The decomposition of Fe acetylacetone at elevated temperatures resulted in the localization of Fe on nitrogen-doped carbon, with certain Fe species being reduced to a zero valence in the presence of carbon. To eliminate the Fe particles, HCl etching was performed meticulously. However, the nanoparticles that developed within Fe SA/PNCNFS-0.4 at an elevated impregnation level underwent increased etching, resulting in a reduced presence of single atoms.

Fig. 4 (a) Schematic of the structure of Fe SA/PNCNFs-0.1;⁹⁸ (b) schematic of Ga-doped C₃N₄ single-atom photocatalyst;⁹⁹ (c) schematic of the preparation process of Pt-MCT-3;¹⁰⁰ (d) schematic of preparation CW-Pd;¹⁰¹ (e) schematic of Ni atoms modified TiO₂.¹⁰²

3.2 Pyrolysis method

The pyrolysis method served as a crucial approach for producing single-atom photocatalysts by heating a combination of metal and organic semiconductor precursors at elevated temperatures. Establishing a robust coordination bond between the metal centers and the organic precursor is typically necessary to prevent the aggregation of isolated metal sites during the pyrolysis process. For instance, Jiang et al.⁹⁹ successfully incorporated Ga (0.014–0.045 wt%) into g-C₃N₄ through the direct pyrolysis of a mixture of GaCl₃ and urea at 550 °C (Fig. 4b).

3.3 Photodeposition method

The advantage of photodeposition lies in its ability to produce high-quality thin films with complex structures, while controlling various parameters such as thickness, composition, and microstructure. It demonstrates unique advantages in the synthesis of single-atom catalysts. Liu et al.¹⁰⁰ developed a hybrid system by combining nitrogen defects with single Pt atoms, achieved through a straightforward self-assembly and photodeposition approach, named Pt-MCT-3. The MCT precursors were produced through the in situ self-assembly process involving trithiocyanuric acid, melamine, and cyanuric acid. A specific volume of chloroplatinic acid solution was then added dropwise into the solution of MCT-3. After light deposition for 30 minutes, an efficient single-metal atom photocatalyst Pt-MCT-3 was obtained (Fig. 4c).

3.4 Other methods

In addition to the methods outlined above, alternative strategies were devised for the production of SMACs. Atomic layer deposition (ALD) presented a precise technique for synthesizing such photocatalysts by sequentially depositing single metal sites onto semiconductor surfaces in a controlled manner. Zhang et al.¹⁰¹ deposited Pd single atoms onto carbonized wood (CD) using ALD. By heating the Pd(hfac)₂ precursor to approximately 85 °C, the desired vapor pressure was achieved. Deposition was conducted at 200 °C, with the ALD delivery lines maintained at 100 °C. Nitrogen was used as a purging and carrier gas throughout the deposition process, resulting in the creation of an efficient SMAC photocatalyst (Fig. 4d). Xiao et al.¹⁰² successfully achieved the precise incorporation of single Ni atoms onto the TiO₂ photocatalyst using a controllable molten-salt method (MSM). This method provided ample ionized anions and cations with a strong polarizing force to disrupt covalent, ionic, or metallic bonds within the molten salt environment (Fig. 4e). The molten salts, comprising LiCl and KCl, created a liquid medium and spatial constraint that enabled the isolated dispersion of Ni species and facilitated the formation of robust Ni–O bonds on TiO₂. Moreover, chemical etching was employed as a method to fabricate single-atom photocatalysts with a substantial metal content.

4 SMACs for pNRR

The physical and chemical characteristics of a catalyst undergo a significant transformation when the dispersion of the doping metal is heightened to an individual atomic level, leading to the emergence of a distinct class of catalysts known as single-metal atom catalysts (SMACs).^103–106 Recently, SMACs have demonstrated remarkable stability, targeting activity, selectivity, and reactivity, making them prospective candidates for applications in the pNRR. Recent studies have illustrated the superior photocatalytic performance of SMACs in the pNRR, achieving a notable atomic efficiency.^107–117 Notably, SMACs containing a singular metal atom readily integrate into the matrix carrier, facilitating interactions with specific external species at the unimpregnated coordination sites. Strong covalent or ionic bonds between the matrix and metal atoms prevent the agglomeration or dissolution of SMACs. The remarkable catalytic efficiency of SMACs can be ascribed to their low coordination number, a characteristic that promotes the adsorption and activation of dinitrogen on the catalytic sites. Furthermore, the adsorption characteristics of the substrate can be regulated by changing the electronic states caused by charge transfer. Varied coordination structures of metal atoms yield distinct free energies during reactions, enhancing selectivity and reaction efficiency. Currently, SMACs exhibit promising prospects for advancement in the domain of pNRR. Here, we discuss the NRR activity of diverse SMACs dispersed on various matrices.

4.1 Single-metal atoms supported on a carbon-based substrate for pNRR

Carbon-based materials exhibit appealing band structures, robust chemical stability, high abundance on the Earth, and ease of production, rendering them well suited for photocatalytic nitrogen fixation. Diverse carbon-based materials, such as graphene oxide (GO) featuring oxidative functional groups, graphene and its derivatives such as graphene quantum dots (GQD), GO, reduced graphene oxide (rGO), graphitic nitride carbon, N-doped carbon, reduced graphene oxide, cellulose, graphene diyne, carbon nanotubes (CNTs), and ultra-thin carbon nanosheets,^34,118–123 among others, have demonstrated exceptional performance, establishing them as ideal nanocomposite materials. When dispersed in the matrix and carbon nanotubes, they exhibit outstanding electron transfer efficiency, cost-effectiveness, stability, and environmental friendliness, even in acidic and alkaline environments. Augmentation of SMAC active sites with a larger surface area can enhance ammonia production by boosting light absorption for effective fixation.^124–128 Moreover, carbon substrates facilitate electron guidance to metal atoms by improving the contact of surface-active sites, thereby improving N₂ adsorption sites and subsequently increasing NH₃ production.¹²⁹

4.1.1 Single atoms of noble metals on a carbon substrate for pNRR. Recently, metals like Ru,^{108,130–132} Pd,^133,134 and Ag ^135,136 have been employed in contemporary endeavors aimed at facilitating environmentally friendly ammonia synthesis processes. Li et al.¹³¹ successfully anchored Ru atoms onto nitrogen-doped carbon substrates containing cyanide defects. Through a one-pot calcination process, Ru single atoms were loaded onto a g-C₃N₄ catalyst, demonstrating remarkable performance in the NRR with a value of 125.5 μmol g_cat⁻¹ h⁻¹. The experimental results indicated that incorporating single-atom Ru reduces the band gap of carbon nitride, expands the photoresponse range of catalysts, and enhances its absorption capacity.

Similarly, Ding et al.¹³² synthesized Ru/MOF/C₃N₄ bifunctional catalysts in situ, which not only demonstrated success in the NRR but also in the CO₂RR. Initially, C₃N₄ sheets were introduced into a Zn(NO₃)₂·6H₂O solution containing ligands such as 2,5-dihydroxyterephthalic acid, DMF, and H₂O to effectively attach Zn ions onto the C₃N₄ surface, promoting the homogeneous separation of MOFs during the in situ synthesis procedure. Subsequently, RuCl₃ with unsaturated coordination sites was incorporated into the metal–organic framework (MOF) possessing unsaturated coordination sites. This allowed Ru ions to coordinate with the unsaturated sites on the MOF, immobilizing the dispersed Ru to produce a single Ru atom (Fig. 5a). Theoretical calculations using DFT demonstrated that a singular Ru atom generates energy levels within the bandgap of the MOFs, thus hindering electron–hole complex formation and augmenting the abundance of photocarriers. Furthermore, the researchers conducted Gibbs free energy calculations for each intermediate product in various models of distinct reaction pathways, showing a decrease in the Gibbs free energy of the NRR in both the distal and alternating mechanisms (Fig. 5b and c).

Fig. 5 (a) Diagrams of the synthesis of Ru/MOF/C₃N₄; (b and c) free energy diagrams were calculated for the photocatalytic NRR processes, specifically for pNRR (Distal) and pNRR (Alternating) pathways;¹³² (d) adsorption configuration and (e) the charge density contrast observed in an N₂ molecule adhered to an N₂C NV site on a PCN surface is depicted by yellow and light blue isosurfaces, denoting areas of charge accumulation and depletion in three-dimensional space. The specific charge density value measured is 0.0025 elementary charges per cubic angstrom.¹³⁷ (f and g) A photocatalyst comprising g-C₃N₄ adorned with single Pd sites was investigated for its efficacy in eliminating NOx pollutants.¹³⁸

Gold's distinctive electronic configuration, which facilitates effective charge transfer and exhibits low reactivity towards hydrogen, which inhibits the hydrogen evolution reaction, has established it as a promising catalyst contender in the pNRR. Studies have demonstrated that gold nanoparticles or clusters exhibit remarkable pNRR performance. In line with this, Guo et al.¹³⁷ used an etching method to uniformly embed Au nanoparticles into the mesopores of nitrogen-deficient hollow nitrogen-doped carbon spheres to form a plasma hybrid catalyst. Incorporating Au nanoparticles within the mesoporous shell of nitrogen-doped carbon spheres significantly improved the absorption of visible light, leading to the generation of hot electrons for N₂ reduction while effectively suppressing carrier recombination. This design promoted concurrent oxidation and reduction reactions within the system. 7.4 wt% of gold embedded in nitrogen-doped carbon spheres achieved a NH₃ yield rate of 783.4 mmol h⁻¹ g_cat⁻¹ under visible light. Theoretical investigations have indicated that N₂ molecules exhibit favorable capture and adsorption properties when interacting with NVs at the N₂C site of PCN in a dinuclear end-on coordination configuration, resembling the coordination of N₂ bound to metal complexes (Fig. 5d). The discrepancy in charge density was computed to assess the electron transfer dynamics between the nitrogen vacancies (NVs) and the N₂ molecules adhered to them. The electrons residing at the carbon atoms neighboring the NV locations are shifted to the adsorbed nitrogen molecule, thereby triggering the activation of the N≡N triple bond (Fig. 5e). Scholars have extended their investigations from Ru- and Au-based systems to Pd-based systems, thereby broadening the horizons of catalyst research. Liu et al.¹³⁸ utilized urea as a precursor to immobilize Pd atoms on g-C₃N₄ carriers embellished with adjustable carbon defects, resulting in the formation of Pd–N₃ active sites (Fig. 5f). These sites demonstrated superior and consistent performance in the photocatalytic removal of NO, surpassing the original g-C₃N₄ by 4.4 times (Fig. 5g).

Steady-state. Silver (Ag), a noble metal, is extensively used in pNRR. Wang et al.¹³⁶ conducted a study using first-principles research to elucidate the mechanisms underlying the improvement of pNRR performance. In their study, silver single atoms were evenly distributed on N-doped graphene carbon (Ag/g-C₃N₄) synthesized via a straightforward hydrothermal process employing block-like cyanamide (CN). Subsequently, using reduction deposition, a one-pot technique was employed to make uniformly dispersed Ag/g-C₃N₄ as a highly effective N₂ photofixation catalyst at extremely low temperatures (Fig. 6a). When the loading amount of Ag was 4%, the catalytic activity of ammonia was the highest (1.45 mmol L⁻¹ g_cat⁻¹ h⁻¹), which was nearly 15 times higher than the original CN value (Fig. 6b). Theoretical investigations have indicated that the increase in photocatalytic activity can be attributed as follows: (1) the small silver particles exhibit a reduced recombination rate of the photogenerated electron–hole pairs in comparison with their larger counterparts, owing to the abbreviated transfer distance within their structure to the Ag–CN interface. This proximity facilitates enhanced contact between the small Ag particles, improving charge-separation efficiency. (2) Small Ag particles offer the advantage of exposing supplementary active sites. Furthermore, incorporating Ag as a cocatalyst enhances the disjunction of photogenerated electron–hole pairs, thereby diminishing the potential for NH₂ production and bolstering its catalytic generation (Fig. 6c).

Fig. 6 (a) Illustration of the synthesis of Ag/g-C₃N₄. (b) NH₃ production rate graph with different Ag/CN; (c) the yellow denotes charge accumulation and cyan denotes charge depletion;¹³⁶ (d) illustration of the synthesis of Ag/g-C₃N₄ at different temperatures; (e) NH₃ production rate at various temperatures; (f) the impact of particle size on both the catalytic SBH value and the photocatalytic performance is under examination; (g) the size of Ag nanoparticles can be controlled, leading to the establishment of an inherent electric field between silver and C₃N₄ material.¹³⁹

Similarly, in Li et al.¹³⁹ the deposition of silver particles with an average diameter of 9.3–33.2 nm on N₂ photofixed catalyst g-C₃N₄ was studied using the ultra-low temperature reduction deposition method (Fig. 6d). The Ag/CN (−40) exhibited the best NH₃ precipitation rate (1.95 mmol L⁻¹ g_cat⁻¹ h⁻¹) (Fig. 6e). The researchers delved into the mechanism of this process and observed that as the preparation temperature increased, the size of the silver particles on g-C₃N₄ increased gradually, while the work function and Schottky barrier height (SBH) value decreased. Larger silver nanoparticles led to a smaller SBH, facilitating interactions between silver and g-C₃N₄ at the interface. As the dimensions of the silver particles enlarged, the electrons and (or) holes found themselves confronted with a lengthier journey to interact with the photocatalyst's surface. This discrepancy in the propulsive force for extended-range movement heightened the likelihood of charge recombination, thus leading to a decline in the effectiveness of the photocatalytic process (Fig. 6f and g).

4.1.2 Single atoms of non-noble metals on a carbon substrate for pNRR. Apart from noble metal SMACs, non-noble metals like Fe, Mo, Co, and Cu have recently gained attention as photocatalysts for pNRR under environmental conditions. The shift from precious to non-precious metal catalysts presents a compelling opportunity to broaden the realm of catalyst exploration. Notably, Fe stands out for its positioning at the apex of the volcano plot based on DFT, highlighting its potential as a promising transition metal species. Building on this premise and drawing insights from nitrogenase, scientists⁴² have successfully fabricated Fe porous g-C₃N₄ (FPx) by dispersing a solitary iron atom onto nitrogen-doped carbon with a porous structure that exposes more active sites. This unique design not only alters the adsorption mechanism of N₂ from physical to chemical but also switches the pNRR pathway from the associative distal pathway to the associative alternating pathway (Fig. 7a). The synthesis of the single-atom iron porous g-C₃N₄ (FPx) specimen involved a one-step annealing process using NH₄Cl to generate gaseous NH₃ and HCl as bubble templates, facilitating the direct atomization of Fe particles to form isolated Fe sites (Fig. 7b).

Fig. 7 (a) Demonstration of the amalgamation process of FeSA/NC; (b) comparison of accelerated nitrogen reduction with decelerated nitrogen reduction;⁴² (c) optimized N₂ adsorption model on pure PCN and Mo-PCN; (d) N₂ reduction catalyzed by Mo-PCN through distal and alternating mechanisms.¹⁴⁰

Scholars have recently explored deeply the possibilities of molybdenum-based systems in catalysis following the advancements in iron-based SMAC. In light of the coexistence of Fe and Mo metal centers within nitrogenase, numerous studies have strived to integrate molybdenum atoms into porous carbon frameworks as a means to enhance the development of NRR catalysts. Guo and colleagues¹⁴⁰ synthesized a collection of Mo single-atom catalysts through the calcination of economically feasible urea as the primary precursor, in conjunction with different quantities of Na₂MoO₄·2H₂O. The individual Mo center exhibited coordination with two nitrogen donors, resulting in the formation of a two-coordinated MoN₂ species, which was immobilized on the poly-nitride carbon structure formed in situ. The low coordination Mo centers can be used as active sites for N₂ chemical adsorption and activation, realizing high photocatalytic activity for NH₃ release in pure water. Experimental studies and calculations using density functional theory have shown that the coordinatively unsaturated metal site within single-atom catalysts exhibits a high affinity for N₂ molecules, facilitating strong adsorption through an end-on configuration. This interaction leads to the elongation of the N≡N bond from 1.11 to 1.15 Å (Fig. 7c). Consequently, photoexcited electrons can transfer to the weakened N≡N bond, enhancing the efficacy of nitrogen fixation in natural settings. The analysis presented in Fig. 7d indicates that in both distal and alternating mechanisms, the crucial step governing the rate of the process is the conversion of N₂ to N₂H*, characterized by an endothermic nature with an energy barrier of 0.81 eV. These findings suggest the coexistence of two distinct reaction pathways during the photocatalytic N₂ fixation process.

Liu and colleagues conducted an extensive DFT investigation, employing the N≡N dipole within the adsorbed N₂ as a metric to assess the efficiency of catalytic sites in the catalyst for NRR.¹⁴¹ The study revealed that employing 2D phthalocyanine (2D Mo–Pc) to support Mo single atoms led to superior performance compared with other transition metals. The enhanced performance observed can be ascribed to the dominant dipole moment produced within the N≡N triple bond due to the interaction between Mo and N atoms. This dipole moment effectively promotes the migration of d electrons from individual Mo atoms to the π* anti-bonding orbitals of the N₂ molecule, thereby enhancing the reaction. Additionally, the reduced affinity of Mo with hydrogen minimizes the competing HER, consequently improving the selectivity of the 2D Mo–Pc catalyst towards the NRR.

Theoretical studies have underscored the importance of employing single atoms of transition metals belonging to the 3d, 4d, and 5d series on two-dimensional rectangular monolayers of tetracyanoquinodimethane (TM-rTCNQ) for NRR. Among these transition metals, Mo, Tc, and W have been identified as promising SMACs due to their remarkable structural stability and high selectivity, rendering them favorable for NRR applications.¹⁴² Significant adsorption and activation of N₂ were discovered for all three SMACs. The observed phenomenon can be ascribed to the augmented orbital hybridization and charge transfer mechanisms, which are fostered by the interplay between the catalyst and nitrogen species. Through a wide evaluation process, the Mo-rTCNQ monolayer was determined to be the most effective SMAC for catalyzing NRR, showcasing a distal mechanism. Tetracyanoquinodimethane has the potential to be fluorinated to produce tetrafluorotetracyanoquinodimethane (F₄TCNQ), which serves as a promising organic charge transfer material and robust π electron acceptor. Researchers conducted a detailed investigation on the viability of a single transition metal atom (TM = 3d–5d series transition metal atoms) incorporated into a rectangular F₄TCNQ (M-rF₄TCNQ) monolayer for the pNRR using advanced first-principles high-throughput screening calculations.¹⁴³ Among the identified four SMACs capable of catalyzing the NRR (M-rF₄TCNQ, where M = Ti, Mo, Nb, or Tc), their efficacy was considered to be the robust adsorption and activation of N₂, primarily arising from the “acceptance–donation” interaction between the transition metal and the adsorbed nitrogen molecule. The research demonstrated that Ti-rF₄TCNQ could efficiently promote the NRR process through an enzyme-like mechanism. Moreover, Mo-rF₄TCNQ was also recognized as a feasible SMAC for NRR, facilitating the reaction through a distal pathway.¹⁴⁴

Loading cobalt single atom has occurred as a viable method to enhance the catalytic performance of semiconductors, thereby facilitating the efficient separation of electron–hole pairs generated during photoinduced reactions and providing essential sites for nitrogen reduction. This approach is crucial for achieving optimal energy synergy in catalytic processes.^145,146 Recently, Wang et al.¹⁴⁷ employed H₃PW₁₂O₄₀ (PW₁₂) incorporated with ZIF-67 as a starting material to synthesize nitrogen-doped graphitic carbon (NGC) nanocages hybridized with tungsten carbide and cobalt nanoparticles (WC–Co/NGC) through a calcination process (Fig. 8a). The NH₃ production rate (142 μmol g⁻¹ h⁻¹) of WC/Co–NGC-2 under visible light irradiation was approximately six times higher than that of Co–NGC with WC unloaded (Fig. 8b). The advancement observed in the pNRR process can be attributed to the inclusion of WC particles and the cooperative influence arising from the interaction between WC and Co. The 3D nanocages exhibited a high specific surface area and an appropriate band structure location, contributing to efficient N₂ fixation (Fig. 8c).

Fig. 8 (a) The synthetic procedure of WC–Co/NGC; (b) the rate of ammonia formation through photocatalysis, and (c) a schematic illustration depicting the process of photocatalytic nitrogen fixation utilizing WC–Co/NGC is presented;¹⁴⁷ (d) ammonia yield across various samples; (e) cycling measurements were conducted to assess the pNRR performance of Co@g-C₃N₄-1 nanosheets;(f–g) The adsorption geometry of nitrogen molecules on pristine g-C3N4 and cobalt-decorated g-C3N4, denoted as Co@g-C3N4-1 was investigated.¹⁴⁸

In another work by Fu et al.,¹⁴⁸ a molecular scaffolding approach was utilized to immobilize and support single-site Co on porous crimped g-C₃N₄ (Co@g-C₃N₄), proposing these as highly efficient photocatalysts for solar-photon-driven N₂ fixation under ambient circumstances. As revealed in Fig. 7d, it was observed that the bulk g-C₃N₄ showed a significantly low N₂ fixation activity of 8.2 μmol h⁻¹, while the supramolecular synthesized g-C₃N₄ showed a higher activity of 19.5 μmol h⁻¹. Upon integrating single cobalt sites, the optimal activity of Co@g-C₃N₄-1 reached 50.2 μmol h⁻¹. Notably, the catalyst displayed good stability over 3 cycles under severe nitrogen fixation circumstances with visible light irradiation (Fig. 8e). DFT calculations were performed to identify the theoretically catalytically active sites. Intriguingly, the coordinated cobalt site emerged as the active center with a robust adsorption energy of 0.99 eV compared with bare g-C₃N₄ (Fig. 8f and g).

The utilization of copper (Cu) catalyst in the photocatalytic reduction of nitrogen gas for producing green ammonia under mild conditions and atmospheric pressure has been extensively investigated. However, the application of single-atom copper in the NRR process remains relatively limited in current research. Huang et al.¹⁴⁹ successfully synthesized a copper single-atom catalyst by modifying copper on g-C₃N₄. The resulting Cu–CN material demonstrated a notable concentration of active Cu–N₂ sites (Fig. 9a). X-ray absorption fine structure (srXAFS) provided additional evidence supporting the presence of a solitary Cu atom configuration within Cu–CN (Fig. 9b). Notably, the ammonia production rate of copper-altered carbon nitride (Cu–CN) when subjected to visible light in pure water was recorded at 186 μmol g⁻¹ h⁻¹ (Fig. 9c); DFT simulations indicated that an individual copper atom can effectively extract paired electrons from the extended π electron system of p-CN, with the majority of these limited copper atoms likely to form associations with the T-defect vacancies within the polymerized carbon nitride structure (Fig. 9d).

Fig. 9 (a) HAADF-STEM image of Cu–CN; (b) Cu foil and Cu–CN FT-EXAFS spectra were analyzed; (c) quantitative determination of NH₃ generated under visible light; (d) the suggested arrangement of copper cyanide within a T-shaped defect (illustrated in color online) is being proposed.¹⁴⁹

Alkaline-earth metal elements located in the s-block of the periodic table have received limited attention as potential active sites for nitrogen (N₂) photofixation. A recent study by Pan et al.¹⁵⁰ focused on the synthesis of alkaline earth metal single atom-modified mesoporous g-C₃N₄ materials (including Ca, Mg, Sr, Ba) via calcination and continuous stirring techniques (Fig. 10a). Notably, the research found that the 0.5 Ca/m-g-C₃N₄ catalyst exhibited superior performance, generating 42.23 μg g_cat⁻¹ h⁻¹ of NH₃ (Fig. 10b). Furthermore, nitrogen-programmed temperature desorption experiments (N₂-TPD) revealed that atomically dispersed Ca acted as the active site on Ca/m-g-C₃N₄, demonstrating efficient N₂ molecule adsorption. The isolated Ca atoms not only functioned as the primary center for N₂ activation but also optimized the band structure of m-g-C₃N₄, thereby enhancing the photocatalytic production of NH₃ (Fig. 10c). The reduction mechanism of N₂ molecules predominantly involves binding via remote pathways, where N₂ adsorption during the photoinduced nitrogen reduction reaction occurs with concurrent chemical transformations. Specifically, N₂ molecules along the distant pathway are adsorbed in a terminal configuration on Ca single-atom sites, leading to a weakening of the N≡N bond (Fig. 10d).

Fig. 10 (a) Schematic illustration for the preparation of Ca/m-g-C₃N₄; (b) different alkaline earth metal NH₃ production; (c) N₂-TPD spectra of the as-prepared different photocatalysts; (d) the band structure and pNRR pathway of continuous hydrogenation process on Ca active sites of 0.5 Ca/m-g-C₃N₄.¹⁵⁰

4.2 Single atoms of non-noble metals on a non-carbon substrate for pNRR

The exploration of single-atom catalysts on non-carbon substrates for NRR has been a prominent area of research. While doped carbon-based substrates are typically utilized for dispersing single-metal atoms, non-carbon substrates are essential for controlling the dispersion of metal atoms and influencing the NRR process. This section delves into the theoretical and practical aspects of NRR photocatalysts that employ single-metal atom catalysts supported on non-carbon materials, elucidating their significance and mechanisms in detail.

4.2.1 Metal oxides as supports. Transition metals such as Fe, Co, Ti, and Ni and their respective oxides in the Earth's crust represent cost-effective substrates suitable for implementation as catalysts across various catalytic applications. These include, but are not limited to, the facilitation of the oxygen reduction reaction, nitrogen reduction reaction, hydrogen evolution reaction, oxygen evolution reaction, and hydrogenation processes.¹⁴⁴ Notably, nanostructured mesoporous TiO₂ exhibits favorable characteristics such as a substantial surface area and porous structure, which can significantly enhance various catalytic processes as well as improve the efficacy of adsorption and separation mechanisms.

By employing the molten salt technique for the incorporation of Ru single atoms into TiO₂, the electronic metal–support interactions (EMSI) established between Ru single atoms and TiO₂ play a crucial role in stabilizing oxygen vacancies. This stabilization facilitates the activation and adsorption of N₂ molecules for NRR¹⁵¹ (Fig. 11a). Remarkably, a notable NH₃ production of 18.9 μmol g⁻¹ h⁻¹ is attainable under gentle conditions without the need for sacrificial agents. The collective findings from empirical observations and theoretical computations indicate that the doping of Ru exhibits the optimal synergistic effect between OVs, thereby facilitating the concurrent adsorption, N₂ activation, and electron transfer to N₂. In the associative distal pathway context, the initial hydrogenation of proton and electron pairs occurs on the N atom that is distant from the active sites, yielding ammonia. Conversely, in the associative alternating mechanism, these pairs interact with N atoms on opposing sides in an alternating fashion. Within the Ti–V_O–Ti site, the transition from *N₂ to *N-NH presents the most substantial energy barrier at 0.69 eV among all involved steps, establishing the transformation of *N–N to *N-NH as the pivotal rate-limiting stage. The hydrogenation of *N-NH encompasses two distinct procedures, namely the formation of *NNH₂via the distal pathway and *NH–NH through the alternating pathway. The energy changes (ΔG) associated with NH–NH* and NH₂* generation are recorded at −0.37 and −0.08 eV, respectively. Hence, the preference for N₂ molecule hydrogenation via the associative alternate pathway at the Ti–V_O–Ti site is evident. Following the introduction of Ru, the transition from *N–N to *N–NH emerges as the controlling step with an energy barrier of approximately 0.42 eV (Fig. 11b and c). Furthermore, EMSI plays a role in modulating the local electronic structure by facilitating electron transfer from the Ru atom to TiO₂, consequently diminishing the energy barriers of rate-limiting steps and promoting the activation of N₂ and subsequent hydrogenation reactions.¹⁵¹

Fig. 11 (a) Schematic illustration for the preparation of Ru/TiO₂-V_O; (b) the free energy diagram depicting the NRR on a TiO₂-V_O catalyst, along with the corresponding atom configurations; (c) the free energy diagram depicting the NRR occurring on a Ru₁/TiO₂-V_O catalyst, accompanied by the corresponding atomic configurations denoted by distinct colors: Ti in gray, Ru in green, O in red, N in blue, and H in white;¹⁵¹ (d) diagram illustration for the preparation of Fe-SA/WO_2.72−x; (e) photocatalytic ammonia production rates over different catalysts; (f) Gibbs free energy diagrams of different paths on WO_2.72 and Fe-SA/WO_2.72−x.⁴¹

In addition, Hu et al.⁴¹ fabricated Fe-SA/WO_2.72−x photocatalysts through a series of methods including solvent heat treatment, impregnation, and annealing under an H₂/Ar atmosphere at a controlled temperature (Fig. 11d), Notably, the Fe-SA-4/WO_2.72−x catalyst demonstrated the highest NH₃ production rate at 186.5 μmol g⁻¹ h⁻¹ among all samples (Fig. 10e). The study employed DFT calculations to investigate the hydrogenation process of N₂* to NNH* on pristine WO_2.72−x, revealing a rate-limiting energy of 1.67 eV. However, the presence of Fe–O–W sites on Fe-SA/WO_2.72−x resulted in a notable weakening of the N≡N bond of adsorbed N₂ (*N₂), resulting in a substantial reduction in the energy barrier for hydrogenation and transformation to *N₂H. Consequently, the incorporation of Fe single atoms on WO_2.72−x and the presence of Ov were identified as key factors contributing to the improved nitrogen fixation (Fig. 11f).

4.2.2 Transition metal dichalcogenide as supports. Transition metal dichalcogenides (MoS₂ WS₂, etc.) exhibit great potential as catalysts in photocatalysis, presenting a viable alternative option to noble metals and showing the capability to surpass their performance. The exceptional catalytic performance of transition metal dichalcogenides in their metallic phase can be attributed to their remarkable characteristics. The materials exhibit notable conductivity (Fig. 12a), facilitating the creation of trapping sites for photoinduced charges and facilitating the processes of charge separation, migration, and collection, thereby enhancing quantum efficiency (Fig. 12b). With abundant active sites present on both the edge and basal planes (Fig. 12c), these materials are capable of serving as crucial locations for adsorption, activation, and reaction of reactant molecules, effectively reducing activation energies and accelerating reaction kinetics (Fig. 12d). Importantly, their atomically thin structure allows for the formation of a dense interface structure of light-harvesting semiconductor, ensuring quick transfer of photoinduced charges. In addition, these materials also possess other ideal interface surface properties and behaviors, such as active site self-optimization, abundant sulfur vacancies as induced photothermal effects, and more. These distinct properties and behaviors enhance the potential of these materials as efficient cocatalysts in photocatalysis.^91,152,153

Fig. 12 (a) Good conductivity; (b) facilitating the creation of trapping locations for charges generated by light exposure, and enhancing the processes of charge separation, movement, and gathering; (c) abundant active sites are present on both the edge and basal plane surfaces; (d) engaging actively in the adsorption, activation, and reaction sites of reactant molecules is crucial for reducing activation energies and enhancing the kinetics of chemical reactions.¹⁵⁵

Based on this, Hu et al.¹⁵⁴ conducted a study where Mn single-atom doped MoS_2−x nanoflowers were synthesized utilizing manganese chloride as the precursor for manganese atoms and thiourea as the sulfur source through a hydrothermal method (Fig. 13a). The research employed DFT to investigate the pNRR process on Mn-MoS_2−x. The outcomes indicated a remarkable decrease in the adsorption of N₂ and intermediate post-single-atom Mn doping. The energy barrier of Mn-MoS_2−x-Sv was found to be higher than Mo edge sites yet lower than pure MoS₂-based Mo sites. Notably, the energy barrier of pNRR on Mn-MoS_2−x-Sv was significantly reduced to 0.87 eV, representing the lowest potential among MoS₂ structures. This reduction was attributed to the fixation of N₂ on exposed Mo sites facilitated by the presence of double S vacancies (Fig. 13b). Furthermore, under light irradiation without sacrificial agents in a pure water medium, Mn-MoS_2−x-4 exhibited a visible light NH₃ generation rate of 148.3 μmol g⁻¹ h⁻¹, a marked improvement compared with the original MoS₂ (30.5 μmol g⁻¹ h⁻¹), demonstrating a 4.9 times increase in efficiency (Fig. 13c).

Fig. 13 (a) Schematic illustration for the preparation of Mn-MoS_2−x; (b) Gibbs free energy diagram of N₂ reduction of Mo-MoS_2−x; (c) photocatalytic NH₃ generation rate of Mn-MoS_2−x-x in N₂ saturated water under visible light irradiation;¹⁵⁴ (d) the top view of the Ni dopant configuration represented by the black solid line and the S vacancy represented by the red dashed line W as blue and element S as yellow.¹⁵⁶ (e) Photocatalytic performance for visible-light-induced ammonia production reaction depending on AuNP-nano-Ti₃C₂ MXene heterozygotes weight.¹⁵⁷ (f) The NH₄⁺ yield rate of the control experiments under Ar, Air, N₂, and dark ambient conditions.¹⁵⁸

The monolayer of WS₂ stands out as a highly promising and economical catalyst across various electrocatalysis disciplines. Despite its inherent inertness, the WS₂ basal plane boasts a notable surface-to-volume ratio, rendering it a prime candidate for enhancement through heteroatom doping methods. These strategies encompass the directed placement of dopant atoms into interstitial sites or the substitution of primitive lattice atoms within the WS₂ framework. Ma et al.¹⁵⁶ conducted a comprehensive examination of the pNRR of transition metal (TM) atoms, including Nb through Cd, Sc through Zn, W, Au, and Pt, attached to WS₂ nanosheets featuring vacancies in their sulfur atoms. Through spin-polarized DFT calculation, it was revealed that a singular Ni atom positioned within a sulfur vacancy (Ni-WS₂) exhibited heightened pNRR activity (Fig. 13d). Additionally, the NRR energy diagram of the Ni–WS₂ surface indicated its inclination to proceed along the alternating associative pathway for NH₃ preparation autonomously, devoid of external energy input.

4.2.3 MXenes as supports. Recently, MXene-derived materials have exhibited notable effectiveness in a diverse range of catalytic uses. The distinctive two-dimensional structure of these materials enables proficient anchoring of individual metal atoms due to the abundant cation vacancies present in MXenes, which possess highly reactive and reductive attributes. Researchers have investigated the enhanced performance of gold-based single-atom catalysts in conjunction with two-dimensional materials known as MXenes, owing to their expansive specific surface area, abundant functional groups (–OH, –O, and –F), high conductivity, and capability to adsorb N₂ chemically. Consequently, MXenes are recognized as promising candidates for the conversion of N₂ to NH₃.^159–162 The relatively low light absorption coefficient of MXenes within the visible light spectrum can be enhanced through the incorporation of plasma nanomaterials such as Au.^163,164 Shin et al.¹⁵⁷ demonstrated the synthesis of ammonia under visible-light conditions by utilizing nanoscale Ti₃C₂ MXene in tandem with plasma Au. At a wavelength of 523 nm, the optical density (OD 0.1) reached 5334 μmol g_cat⁻¹ h⁻¹. The enhanced photocatalytic efficiency was attributed to the increased light absorption of Au and efficient N₂ fixation facilitated by chemically reduced nanoscale Ti₃C₂ MXene (Fig. 13e). Gao et al.¹⁵⁸ effectively produced MXene-derived in situ TiO₂ incorporated with cobalt for the pNRR through a dual-stage calcination technique. Notably, the production of NH₃ reached 110 μmol g⁻¹ h⁻¹ in pure water under UV-vis light irradiation without the need for a sacrificial agent (Fig. 13f). The remarkable achievement observed can be ascribed to the integration of Cobalt, which serves to modulate the adsorption energy levels of nitrogen and ammonia, thereby facilitating the attainment of a state of chemical adsorption equilibrium.

4.2.4. Covalent/metal–organic framework (COF/MOF) as supports. COFs and MOFs are structured porous materials formed by the establishment of covalent linkages between organic monomers and coordination interactions involving metal clusters and organic ligands. The amalgamation of organic frameworks and single atoms in the context of pNRR presents an opportunity to synergistically leverage their unique strengths while circumventing inherent drawbacks. Specifically, the unique catalytic sites of single atoms compensate for the inherent limited activity of framework supports in N₂ reduction. Simultaneously, the ability to redesign and adjust the structures of frameworks ensures an optimal setting for anchoring single atoms and preventing their aggregation. Moreover, by controlling the chemical compositions of frameworks, atoms’ physical and chemical characteristics can be precisely manipulated, resulting in improved catalytic efficiency. Additionally, the customizable pore structures within frameworks facilitate mass transfer pathways during photocatalysis and modulate the interfaces for reactions between reactants and single atomic catalytic sites. Notably, the properties such as band gap, band structure, highest occupied molecular orbital, and lowest unoccupied molecular orbital levels of single-atom catalysts upheld by frameworks can be finely tuned at the molecular scale, thereby enhancing light absorption, photoinduced charge separation and migration and, ultimately, the photocatalytic performance in N₂ reduction.^165–168

He et al.⁴⁹ synthesized Au single-atom modified covalent organic frameworks (COFs) following a specific procedure. Initially, COFs were synthesized using acetic acid, benzaldehyde, and diamine tetraacetic acid. Subsequently, KAuCl₄ was employed as the source of Au atoms, which were then stirred at 85 °C for 24 hours. The resulting precipitate was isolated through centrifugation at 8000 rpm for 5 minutes and washed thrice with tetrahydrofuran (THF). The final product, denoted as COFX Au, was subjected to vacuum drying at ambient conditions overnight to ensure the complete removal of solvent residues (Fig. 14a). After being functionalized with Au atoms, COFX-Au exhibited preserved crystal structures and porosity. Analysis using aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) displayed an even distribution of single Au atoms as luminous points on the COF frameworks (Fig. 14b). The characteristics of Au atoms were further scrutinized through synchrotron-radiation-based X-ray absorption fine-structure spectroscopy (Fig. 14c). The coordination environment of Au atoms was determined to be 3.0 Au–N (±0.9) and 1.9 Au–Cl (±0.5) based on extended X-ray absorption fine-structure (EXAFS) fitting analyses. These results revealed that the N atoms from porphyrin subunits functioned as anchoring sites for single Au atoms, facilitating the formation of square-planar coordination structures. The most noteworthy compound was COF5-Au featuring a robust electron-deficient pentafluorophenyl group attached to the terminal position of the porphyrin Au center. This configuration exhibited superior NH₃ release capabilities, demonstrating a heightened NH₃ generation rate of 427.9 μmol g⁻¹ h⁻¹ (Fig. 14d). These findings reinforce the notion that electron-withdrawing entities play a pivotal role in improving ammonia production.

Fig. 14 (a) Schematic illustration of COFX-Au; (b) HAADF-STEM of COFX-Au; (c) XAFS of COFX-Au; (d) mass production ammonia rate with different Au loadings.⁴⁹ (e) Schematic illustration of PtSACs/CTF; (f) the band structure and (g) the N₂ fixation mechanism of Pt-SACs/CTF; (h) the diagram illustrating the distal and alternating mechanisms, along with the corresponding free energy profiles, during the nitrogen reduction process on the Pt-SACs/CTF catalyst is presented.⁵⁴

Similarly, Li et al.⁵⁴ successfully prepared single-atom Pt anchored at the –N₃ sites of stable and ultrathin CTF nanosheets, denoted as Pt-SACs/CTF, using photodeposition techniques (Fig. 14e). The synthesis method led to the production of NH₃ from an N₂-saturated aqueous solution at a rate of 171.40 μmol g⁻¹ h⁻¹, without the need for a sacrificial agent, with an AQE of 1.4% at 420 nm. The CB potential of Pt-SACs/CTF (Fig. 14f and g) met the requirements for the ammonia synthesis reaction (N₂/NH₃ = −0.28 V vs. NHE). When exposed to light, the photogenerated electrons on the CB of the stable ultrathin CTF-PDDA-TPDH nanosheets were transferred to Pt-SACs, enabling the conversion of N₂ molecules adsorbed on Pt-SACs to ammonia, which then dissolved in water to form NH₄⁺. DFT analysis revealed that the fixation of N₂ in the Pt-SAC/CTF catalyst occurred through an alternating mechanism. In this mechanism, the initial protonation step of forming NNH* from adsorbed N₂ is electron withdrawing, whereas all subsequent fundamental steps are exergonic. The alternating mechanism's onset potential was 1.26 eV (Fig. 14h).

Shang et al.¹⁶⁹ also explored the pNRR performance of Fe–N–C catalysts derived from a porphyrin-based metal–organic framework (PMOF). A notable feature of PMOF is the utilization of Al as a stable metal node. Fe is uniformly distributed at the atomic level within the porphyrin rings, facilitating N₂ adsorption and activation. The Fe–N site emerges as the pivotal center for photocatalytic activity. The integrated findings from experiments and theoretical analyses indicate that the Fe–N site within Al-PMOF(Fe) serves as the primary photocatalytic site, alleviating the challenges associated with the rate-controlling way in photocatalytic nitrogen reduction reactions. The optimal configuration of N₂ within Al PMOF (Fe) involves the end-to-end adsorption of the N–N bond onto the Fe single-atom positioned between two PMOF layers, with the reaction proceeding through a series of alternating pathways (Fig. 15a and b).

Fig. 15 (a) The illustration presents the charge disparity map, depicting regions of positive density difference in yellow and negative density difference in cyan when nitrogen gas is adsorbed onto the aluminum-based porous metal–organic framework containing iron. Herein, C in gray, N in blue, H in white, O in red, Al in pink, and Fe in purple; (b) the free energy diagram of the associative alternating pathways in Al-PMOF(Fe);¹⁶⁹ (c) quantitative determination of NH₃ on Pd EG BiOBr; (d) photocatalytic mechanism of Pd EG BiOBr; (e and f) the illustration presents the charge disparity map, depicting regions of positive density difference in yellow and negative density difference in cyan when nitrogen gas is adsorbed onto the aluminum-based porous metal–organic framework containing iron.⁶¹

4.2.5 Bismuth-based material as support. Bismuth (Bi)-based layered materials have emerged as prospective and innovative semiconductors for different applications, particularly in photocatalysis.^170–172 The preferred hybridization between Bi 6s and O 2p orbitals imparts the majority of Bi-based materials with the ability to harness visible light. Furthermore, the exceptional stability of Bi³⁺ and cost-effectiveness of Bi metal position Bi-based layered materials as a favorable catalyst.¹⁷³ Advancements in pNRR have been noted in recent years; however, the outcomes achieved thus far have yet to meet the desired levels of success. Notably, Bi, a main group metal with semiconductor characteristics, is being increasingly recognized as a potentially effective catalyst for N₂ fixation. Metals exhibiting diminished reactivity, exemplified by bismuth, possess the capacity to preferentially augment the reductive adsorption of nitrogen to produce N₂H*, without succumbing to the impact of the binding energy of ensuing intermediate species. This property also limits surface electron accessibility,⁹² suppressing the Hydrogen Evolution Reaction (HER). Currently, Bi-based materials demonstrate remarkable activity and selectivity in synthetic NRR processes. Since 2015, the group led by Zhang¹⁷⁴ has successfully fabricated BiOBr nanosheets featuring exposed (001) surfaces and O_Vs (BOB-001-O_V) for N₂ reduction under visible light irradiation, sans the need for sacrificial agents. Notably, the catalyst exhibited a commendable NH₃ production of 104.2 μmol g_cat⁻¹ h⁻¹. Subsequently, materials based on bismuth have garnered significant interest among academic researchers.

Recently, Liu et al.⁶¹ developed a catalytic system consisting of atomically dispersed palladium (Pd) anchored on bismuth oxybromide (BiOBr) through ethylene glycol (EG) as a molecular connector. The incorporation of 0.20 wt% of Pd EG BiOBr led to a notable enhancement in the pNRR efficiency, resulting in a NH₃ production rate of 124.63 μmol h⁻¹ (Fig. 15c). The adsorption of N₂ through chemisorption and the activation of the chemically inert N≡N triple bond on the surface of BiOBr, facilitated by atomically dispersed Pd, were investigated through DFT. The extensibility of the N≡N triple bond length to 1.139 Å signifies an intensified activation of the N≡N bond over the Pd atom when contrasted with the N₂ molecule in isolation (1.093 Å). Analysis of the charge density difference (Fig. 15e and f) illustrates the electron distribution and interactions between the Pd atom and adsorbed nitrogen. The observed reduction in electron density on the Pd atom and corresponding increase in electron density on the adsorbed N₂ molecule indicates a possible initiation of N≡N triple bond cleavage (Fig. 14f). Moreover, the electron redistribution on the C and O atoms of EG (Fig. 15e and f) implies that EG may modulate the electron density of the Pd atom to improve the adsorption capacity of N₂. These findings suggest that atomically dispersed Pd is more effective at adsorbingand activating the inert N≡N triple bond. In the illustrated mechanism denoted as Fig. 14d, the EG molecule acts as a bridge, promoting the generation of atomically dispersed Pd species. Additionally, the EG molecule facilitates electron transfer from BiOBr to both Pd and activated N₂ molecules. Subsequently, the electrons combine with H⁺ to yield the formation of NH₃. Table 1 presents an overview of the photocatalytic efficiency of SMACs in the context of pNRR as documented in the existing literature.

Table 1 Summary of the photocatalytic efficiency of SMACs for pNRR

Photocatalysts	Amount [mg]	Ammonia production rate	Light source	Efficiency of pNRR	Sacrificial agent	Ref.
Ru/g-C3N4	50	125.5 μmol gcat^−1 h^−1	300 W Xe lamp (λ = 420)	—	—	131
Ru/MOF/C3N4	10	8 mmol g^−1	300 W Xe lamp (λ = 420)	AQY = 7.7%	Methanol	132
Au/HCNS-NV	50	783.4 mmol gcat^−1 h^−1	300 W Xe lamp (λ = 425)	AQE = 0.64	Methanol	137
Pd/g-C3N4	20	2200 μmol gcat^−1 h^−1	300 W Xe lamp (λ > 425)	NO conversion 56.3%	—	138
Ag/g-C3N4	100	1.45 mmol gcat^−1 h^−1	300 W Xe lamp (λ > 400)	—	Alcohol	136
Ag/CN	100	1.95 mmol L^−1 gcat^−1 h^−1	300 W Xe lamp (λ > 400)	—	Alcohol	139
Fe-g-C3N4	30	62.42 μmol gcat^−1 h^−1	300 W Xe lamp (λ = 420)	AQE = 0.1%	—	42
Mo-PCN	3	50.9 mmol gcat^−1 h^−1	300 W Xe lamp (λ = 400)	AQE = 0.7%	—	140
WC–Co/NGC	50	142 μmol g^−1 h^−1	300 W Xe lamp (λ > 420)	—	1 mmol L^−1 Na2SO3	147
Co@g-C3N4	20	50.2 μmol h^−1	500 W Xe lamp (λ > 420)	AQE = 5.4%	Methanol	148
Cu–CN	50	186 μmol g^−1 h^−1	300 W Xe lamp (λ > 420)	AQE = 1.01%	—	149
Ca/m-g-C3N4	20	42.23 μg gcat^−1 h^−1	300 W Xe lamp (full-spectrum)	—	—	150
Ru/TiO2-VO	30	18.9 μmol g^−1 h^−1	300 W Xe lamp (full-spectrum)	—	—	151
Fe-SA/WO2.72−x	5	186.5 μmol g^−1 h^−1	300 W Xe lamp (full-spectrum)	AQE = 0.41%	—	41
Mn-MoS2−x-4	20	148.3 μmol g^−1 h^−1	300 W Xe lamp (λ = 400)	AQE = 0.26%	—	154
Co-Mxene-TiO2	30	110 μmol g^−1 h^−1	300 W Xe lamp (λ = 420)	—	—	158
Au-nano-Ti3C2	10	5334 μmol gcat^−1 h^−1	300 W Xe lamp (400 < λ < 780)	—	—	157
COFX-Au	100	427.9μmol gcat^−1 h^−1	300 W Xe lamp (full-spectrum)	AQE = 0.29%	K2SO3	49
Pt-SACs/CTF	50	171.40 μmol g^−1 h^−1	300 W Xe lamp (420 < λ < 780)	AQE = 1.4%	—	54
Al-PMOF(Fe)	10	127 μg gcat^−1 h^−1	100 mW cm^−2 Xe lamp (λ = 420)	AQE = 0.05%	Methanol	169
Pd EG BiOBr	50	124.63 μmol h^−1	300 W Xe lamp (λ = 420)	—	Methanol	61

---

5 Theoretical perspective

The design of efficient photocatalysts for pNRR has largely relied on the interplay between N₂ molecules and the elemental components of the catalyst. This interplay plays a crucial role in influencing the adsorption and activation mechanisms of N₂. The distinctive characteristics of N₂ generally result in several interactions, including σ-donation, π-back donation, and their combination. A molecule of N₂ is characterized by a strong triple bond between two N atoms, each possessing electrons in the s and p orbitals. One pair of electrons resides in the 2s orbital, while three lone-pair electrons are in the p orbitals. The hybridization of orbitals leads to the formation of four bonding and four antibonding orbitals, consisting of two σ and two π orbitals. This hybridization results in a robust N≡N bond (Fig. 16a). The short bond length of 109.8 pm contributes significantly to the stability of the compound. This heightened stability poses a challenge in activating N₂ because of the notable energy gap between the Highest Occupied Molecular Orbital (HOMO) and the Lowest Unoccupied Molecular Orbital (LUMO) of N₂. The influence of N₂ activation on the catalytic surface is crucial for pNRR, emphasizing the need to develop tailor-made catalysts. A thorough comprehensive understanding of N₂ interactions with different elements is essential for designing such catalysts. Extensive theoretical and experimental investigations have explored various transition metals (TMs), demonstrating that TMs with several free d-electrons exhibit a more advantageous π-back donation effect (Fig. 16b and c). The vacant d-orbitals of TMs are capable of accepting lone-pair electrons from N₂, resulting in the formation of a σ bond. Simultaneously, the occupied d-electrons are back-donated to the empty π* orbital of N₂. Consequently, Ru, Fe, Mn, and Mo have been recognized as high-performing catalysts for pNRR within this specific group.^175,176

Fig. 16 (a) Diagram of nitrogen molecular orbital;¹⁷⁵ (b and c) interaction between TM and N₂ molecules occurs through side-on and end-on adsorption in orbital interactions;¹⁷⁶ (d) volcano plot for TMs (flat and stepped) surfaces versus absorption energy of *N.¹⁸⁰

5.1 Density functional theory (DFT)

First-principles calculations have been instrumental in narrowing the vast chemical space for effective photocatalyst discovery and providing a detailed atomic-level understanding of chemical reaction tasks often challenging for experimental approaches. Several modeling studies using DFT have demonstrated the promise of theoretical frameworks in elucidating and designing novel catalysts for pNRR. DFT has been extensively studied and has proved capable of accurately predicting the adsorption energies of reaction intermediates across numerous benchmark investigations.^177–179

Skúlason and colleagues¹⁸⁰ conducted a comprehensive study on various tightly packed and stepped TM surfaces, laying a foundational basis for the development of non-noble metal catalysts in NRR. The Ru single-atom and other transition metals positioned at the peak of a volcano plot (Fig. 16d) demonstrated elevated rates of NRR activity. However, HER posed a notable challenge to NRR by reducing selectivity and covering the surface with *H adsorbates during primary reactions. The surfaces of early transition metals, such as Y, Sc, Ti, and Zr, were shown to be predominantly coated with *N, effectively suppressing HER and enhancing selectivity. Alternative forms of transition metals, including oxides, carbides, and nitrides, have also been proposed to enhance NRR efficiency and deter competitive HER pathways. Similarly, Li et al.¹⁸¹ systematically studied the NRR catalytic activity of transition metals (TM = Sc–Cu, Y–Ag, and Hf–Au) anchored on monolayer graphyne (GY) using DFT calculations. The study highlighted TM@GY (TM = Sc, V, Mn, Y, Tc, and Os) as exhibiting remarkable NRR capabilities, with Mn, Tc, and Os@GY demonstrating particularly high selectivity toward NRR. This study provided a method for identifying highly effective catalysts for NRR under standard environmental conditions. Abghoui et al.¹⁸² proposed several TM nitrides (including VN, ZrN, NbN, and CrN) for NRR through the Mars–van Krevelen mechanism, demonstrating notably low limiting potentials (η = 0.5 V for VN). SMACs have been employed to enhance the catalytic activity of transition metals by overcoming surface limitations posed by NRR. This approach aimed to optimize the electronic characteristics of TMs and improve atom efficiency.^183,184 SMACs doped on graphene exhibited great potential for the Nitrogen Reduction Reaction (NRR), demonstrating elevated catalytic efficiency while inhibiting the competing HER. Choi et al.¹⁸⁴ identified Ti@N₄ and V@N₄ catalysts as having the most favorable energy barriers of 0.69 and 0.87 eV, respectively, through an evaluation of various TM-doped graphene variants with different coordination numbers (Fig. 17a). Ling et al.¹⁸⁵ identified W-doped graphene with C₃ coordinated atoms as the most effective catalyst, displaying the lowest potential-determining step (PDS) of 0.25 eV compared with other catalysts (Fig. 14b). NRR involves numerous intermediate stages within each mechanistic pathway, and SMACs may struggle to activate highly stable dinitrogen compounds. Biatom catalysts (BACs) leverage two TMs as active sites capable of effectively adsorbing N₂ and enhancing NRR. Gu et al.¹⁸⁶ studied various homonuclear and heteronuclear BACs on a phthalocyanine (Pc) material. Among these, V–V doped on Pc exhibited the most significant activity (Fig. 17c).

Fig. 17 (a) PDS of SACs for N-doped graphene loaded with different TMs;¹⁸⁴ (b) different materials of TMs supported N-doped graphene;¹⁸⁵ (c) structures of BACs supported phthalocyanine;¹⁸⁶ (d) schematic of describing the SGCNN model.¹⁹²

5.2 Machine learning

Recently, there has been a significant rise in the utilization of Machine Learning (ML) within pNRR research, leveraging advancements in artificial intelligence, computational capacities, big data, and ML methodologies.^187,188 ML methodologies have greatly enhanced the development of efficient catalysts by playing a crucial role in evaluating potential catalysts and optimizing catalyst structures.¹⁸⁹ This innovative approach involves the application of various ML models specifically tailored for the advancement of non-precious metal catalysts. Generally, the ML process is influenced by three critical factors: the database utilized, the choice of algorithms, and the creation of distinctive descriptors.¹⁹⁰ Multiple elements affect ML learning and prediction processes. The algorithm selected is vital for managing data extraction, organization, and distribution. Additionally, descriptors tailored to different materials and properties are instrumental in capturing the fundamental physical and chemical attributes essential for catalytic activity.¹⁹⁰ ML applications have focused extensively on nanomaterials, revealing complex correlations between catalyst structure and key performance indicators, such as stability, selectivity, and activity. Nonetheless, high-throughput DFT techniques have proved beneficial for evaluating potential catalysts, particularly when managing extensive datasets with numerous systems, thereby underscoring the crucial role of ML in such scenarios.¹⁹¹ Advances in ML methodologies involved the creation of ML interatomic potentials based on first-principles data, utilizing ab initio datasets. These potentials have been employed to perform dynamic simulations of intricate substance systems. Given the high computational cost associated with DFT and ab initio molecular dynamics (AIMD) simulations, ML-driven AIMD simulations can be utilized to assess the long-term stability of catalysts during operation, offering reduced simulation costs. Recently, Yeo et al.¹⁹² introduced a novel method known as the slab-graph convolutional neural network (SGCNN), designed for surface–adsorbate systems involved in catalytic reactions. This innovative model enhanced the ability of the graph-based convolutional neural network (GCNN), which was effective at predicting bulk characteristics but less effective in surface reaction catalysis (Fig. 17d).

5.3 Big data collection

Machine Learning (ML) utilizes big data generated from experiments or simulations, with particular emphasis on DFT-based calculations. These calculations provide a precise dataset that can be used to effectively understand and enhance catalyst performance. As part of a standard procedure, a dataset is first created using DFT calculations, including significant information on catalyst surface electronics (e.g., electronegativity, neighboring atoms, bond length, electron affinity, bond angle, atomic number, etc.) and computational metrics (e.g., adsorption energies of intermediates, theoretical overpotential, energy barriers, etc.). Diverse open databases, such as the Materials Project, Novel Materials Discovery (NOMAD),¹⁹³ and the International Crystal Structure Database (ICSD),^194,195 are available to access the necessary data pool. Subsequently, a machine learning algorithm is used to establish a mapping relationship between input and output matrices, enhancing fitting results during the training phase. Although machine learning provides valuable insights, challenges such as computational limitations and the potential incompleteness of databases persist due to differences in characterization technologies and inadequate data for various catalysts.

5.4 Artificial intelligence (AI)

The incorporation of artificial intelligence (AI) holds the potential to accelerate the identification of substances with specific desired attributes. However, the limitations of current ML algorithms highlight the necessity for more sophisticated models capable of containing a wide range of properties without overfitting, even when tested under realistic, non-ideal circumstances.¹⁹⁶ Foundational insights are further enhanced through AI, particularly ML, offering an opportunity to expedite the discovery of new materials. Customized descriptors for NRR catalysts play a crucial role, and enhancing ML models by integrating core principles of condensed matter physics has been shown to improve forecasts and broaden datasets for more precise results.^197,198 Essentially, the journey toward developing highly effective photocatalysts for NRR requires a comprehensive approach. This effort encompasses leveraging various materials, addressing catalyst development challenges, enhancing the integration of AI and ML, emphasizing synthesis uniformity, deepening the understanding of mechanisms, and exploring innovative methods to improve catalysis. This collaborative endeavor aims to bring the field closer to achieving efficient and environmentally friendly pNRR processes.¹⁹¹

6 Summary and perspectives

This article briefly reviews the SMACs used in the production of green ammonia through the pNRR process. The review encompasses a comprehensive analysis of different SMACs supported on various substrates, categorized as carbon and non-carbon materials, detailing their synthesis techniques, physicochemical characteristics, and performances in pNRR reactions. The superior pNRR activity exhibited by SMACs on carbon-based substrates is substantiated by theoretical investigations and elucidated with precision. Moreover, the review delves into theoretical projections regarding diverse SMACs supported on non-carbon substrates. While SMACs have made notable advancements in the pNRR sector in recent decades, there persist challenges within this promising domain that necessitate resolution to enhance efficacy:

(I) Recent research has demonstrated effective mitigation of nanoparticle and cluster formation through meticulous regulation of precursor ion concentrations at the substrate surface. While achieving optimal atomic dispersion, this approach often sacrifices loading capacity. Hence, to develop a single-atom catalyst boasting exceptional atomic dispersion, high loading efficiency, and scalability for industrial applications, a comprehensive interdisciplinary investigation is imperative.

(II) The selection and preparation of a suitable support system play pivotal roles in the efficacy of single-metal atom catalysts. Before application, careful consideration and treatment of the support material are imperative. Furthermore, enhancing the stability of singular metal atoms can be achieved by facilitating metal–support interactions (MSIs) or forming strong bonds with adjacent coordination atoms. The objective is to advance the quantity of active sites and improve the reactivity of said sites concurrently. Currently, a limited number of semiconductors serve as supports; therefore, it is imperative to expand the development of other semiconductor systems for this purpose.

(III) HAADF-STEM and XAS techniques are employed in the analysis of metal single-atom photocatalysts to ascertain the precise atomic distribution of individual atoms while simultaneously negating the existence of metal nanoclusters or nanoparticles. These characterization methods, however, are limited to assessing the catalyst's state before and after a reaction. The direct observation of processes such as migration, aggregation, and dissociation of active sites on single-atom photocatalysts during reactions remains unexplored. The dynamic observation of metal single-atom sites during preparation and reaction stages is crucial for comprehending their characteristics and operational mechanisms, which is pivotal for developing robust and highly active single-atom photocatalysts. Although various studies have employed in situ or operando techniques to monitor the evolution of metal single-atom photocatalysts, the outcomes are susceptible to environmental conditions, and definitive mechanisms are yet to be elucidated. Enhancing the efficacy of characterization techniques is instrumental in advancing catalyst development. Despite progress, there exist hurdles in effectively characterizing single-metal atom photocatalytic systems for N₂ fixation, necessitating the utilization of straightforward, effective, and dependable methodologies.

(IV) Currently, significant advancements are being made in the field of pNRR. Nonetheless, the challenge persists in transitioning this technology from laboratory settings to commercial applications due to the limited absorption and conversion efficiency of actual solar energy. To address this critical issue, it is imperative to enhance catalyst efficiency and devise photocatalytic reduction nitrogen reaction engineering strategies. By optimizing the effective irradiation surface area of the reactor, and adjusting sunlight incidence angles and reaction device configurations, solar energy conversion efficiency can be elevated at a fundamental level, consequently enhancing photocatalytic reaction rates. We encourage researchers to delve into this realm which may unveil a novel era in photocatalytic nitrogen fixation.

Author contributions

Ping Zhang: funding acquisition, investigation, supervision. Yongchong Yu: writing – original draft, writing – review & editing. Reyila Tuerhong: software. Xinyu Du: validation. Keyi Chai: investigation. Su Xiaoping: supervision, validation. Shujuan Meng: resources, supervision. Qing Su: resources, supervision. Han Lijuan: funding acquisition, resources. The manuscript was written through the contributions of all authors. All authors have approved the final version of the manuscript.

Data availability

Data availability does not apply to this article as no new data were created or analyzed in this study.

Conflicts of interest

There are no conflicts to declare

Acknowledgements

The authors are grateful for financial aid from the National Natural Science Foundation of China (22378326), Supported by the Fundamental Research Funds for the Central Universities (331920240059), the Science and Technology Project of Lanzhou (2024-3-42), the Scientific Research Project of Introducing Talents of Northwest Minzu University (xbmuyjrc202215).

References

1. V. Smil, Detonator of the population explosion, Nature, 1999, 400, 415–415.
2. M. Cheng, C. Xiao and Y. Xie, Photocatalytic nitrogen fixation: the role of defects in photocatalysts, J. Mater. Chem. A, 2019, 7, 19616–19633.
3. R. Shi, Y. Zhao, G. I. N. Waterhouse, S. Zhang and T. Zhang, Defect engineering in photocatalytic nitrogen fixation, ACS Catal., 2019, 9, 9739–9750.
4. L. Li, P. Zhang, N. Li, T. Reyila, Y. Yu, X. Su, C. Peng and L. Han, Application of g-C3N4-based photocatalysts for N2 photofixation, J. Environ. Chem. Eng., 2024, 12, 112142.
5. W. Liu, K. Yin, K. Yuan, S. Zuo, S. Yang, C. Yao and M. Chen, In situ synthesis of Bi2MoO6@C@attapulgite photocatalyst for enhanced photocatalytic nitrogen fixation ability under simulated solar irradiation, Colloids Surf., A, 2020, 591, 124488.
6. C. MacLaughlin, L. Liu, L. Greenlee and D. MacFarlane, Role for Standardization in electrocatalytic ammonia synthesis, ACS Energy Lett., 2019, 4, 1432–1436.
7. J. Li, X. Guo, L. Gan, Z.-F. Huang, L. Pan, C. Shi, X. Zhang, G. Yang and J.-J. Zou, Fundamentals and advances in emerging crystalline porous materials for photocatalytic and electrocatalytic nitrogen fixation, ACS Appl. Energy Mater., 2022, 5, 9241–9265.
8. R. Shi, X. Zhang, G. I. N. Waterhouse, Y. Zhao and T. Zhang, The Journey toward low Temperature, low pressure catalytic nitrogen fixation, Adv. Energy Mater., 2020, 10, 2000659.
9. F. Jiao and B. Xu, Electrochemical ammonia synthesis and ammonia fuel cells, Adv. Mater., 2019, 31, 1805173.
10. T. Kandemir, M. E. Schuster, A. Senyshyn, M. Behrens and R. Schlögl, The haber–bosch process revisited: On the real structure and stability of “ammonia iron” under working conditions, Angew. Chem., Int. Ed., 2013, 52, 12723–12726.
11. H. Li, C. Mao, H. Shang, Z. Yang, Z. Ai and L. Zhang, New opportunities for efficient N2 fixation by nanosheet photocatalysts, Nanoscale, 2018, 10, 15429–15435.
12. S. Licht, B. Cui, B. Wang, F.-F. Li, J. Lau and S. Liu, : Ammonia synthesis by N2 and steam electrolysis in molten hydroxide suspensions of nanoscale Fe2O3, Science, 2014, 345, 637–640.
13. Y. Liu, X. Ye, R. Li, Y. Tao, C. Zhang, Z. Lian, D. Zhang and G. Li, Boosting the photocatalytic nitrogen reduction to ammonia through adsorption-plasmonic synergistic effects, Chin. Chem. Lett., 2022, 33, 5162–5168.
14. H. Shen, M. Yang, L. Hao, J. Wang, J. Strunk and Z. Sun, Photocatalytic nitrogen reduction to ammonia: Insights into the role of defect engineering in photocatalysts, Nano Res., 2022, 15, 2773–2809.
15. G. N. Schrauzer and T. D. Guth, Photolysis of water and photoreduction of nitrogen on titanium dioxide, J. Am. Chem. Soc., 1977, 99, 7189–7193.
16. X. Chen, N. Li, Z. Kong, W.-J. Ong and X. Zhao, Photocatalytic fixation of nitrogen to ammonia: state-of-the-art advancements and future prospects, Mater. Horiz., 2018, 5, 9–27.
17. Y. Sun, Y. Ahmadi, K.-H. Kim and J. Lee, The use of bismuth-based photocatalysts for the production of ammonia through photocatalytic nitrogen fixation, Renewable Sustainable Energy Rev., 2022, 170, 112967.
18. R. Guan, X. Cheng, Y. Chen, Z. Wu, Z. Zhao, Q. Shang, Y. Sun and Z. Sun, Wettability control of defective TiO2 with alkyl acid for highly efficient photocatalytic ammonia synthesis, Nano Res., 2023, 16, 10770–10778.
19. T. Hou, H. Peng, Y. Xin, S. Wang, W. Zhu, L. Chen, Y. Yao, W. Zhang, S. Liang and L. Wang, Fe single-atom catalyst for visible-light-driven photofixation of nitrogen Sensitized by triphenylphosphine and sodium iodide, ACS Catal., 2020, 10, 5502–5510.
20. J. Yu, T. Zhang and N. Wu, Solar photocatalysis, Sol. RRL, 2021, 5, 2100037.
21. C. Mao, H. Li, H. Gu, J. Wang, Y. Zou, G. Qi, J. Xu, F. Deng, W. Shen, J. Li, S. Liu, J. Zhao and L. Zhang, Beyond the thermal equilibrium limit of ammonia synthesis with dual temperature zone catalyst powered by solar light, Chem, 2019, 5, 2702–2717.
22. W. Zhang, R. Liu, Z. Fan, H. Wen, Y. Chen, R. Lin, Y. Zhu, X. Yang and Z. Chen, Synergistic copper nanoparticles and adjacent single atoms on biomass-derived N-doped carbon toward overall water splitting, Inorg. Chem. Front., 2023, 10, 443–453.
23. W. Shao, M. Yu, X. Xu, X. Han, Y. Chen, J. Han, G. Wu and W. Xing, Design of a single-atom In–N3−S site to modulate exciton behavior in carbon nitride for enhanced photocatalytic performance, Small, 2024, 20, 2306567.
24. X. Xu, W. Shao, G. Tai, M. Yu, X. Han, J. Han, G. Wu and W. Xing, Single-atomic Co-N site modulated exciton dissociation and charge transfer on covalent organic frameworks for efficient antibiotics degradation via peroxymonosulfate activation, Sep. Purif. Technol., 2024, 333, 125890.
25. P. Zhang, Z. Wang, L. Zhao, L. Li, N. Li, X. Su and Q. Su, Construction of a novel 0D-3D boron nitride quantum dots/NH2-MIL-125(Ti) composite for photodegradation of Rhodamine B, Mater. Sci. Semicond. Process., 2023, 167, 107793.
26. X. Wang, P. Zhang, L. Li, N. Li, X. Su, X. Wei and L. Han, Preparation of a high-performance N-defect ZnO@g-C3N4 nanocomposite and its photocatalytic degradation of tetracycline, Mater. Today Commun., 2023, 36, 106732.
27. P. Zhang, L. Zhao, Z. Wang, L. Li, N. Li, X. Su, Q. Su and L. Han, Efficient visible light performance of MoS2QDs/TiO2 hollow sphere photocatalysts for the degradation of dyes and antibiotics, Vacuum, 2024, 219, 112695.
28. W. Zhan, Y. Yuan, L. Sun, Y. Yuan, X. Han and Y. Zhao, Hierarchical NiO@N-doped carbon microspheres with ultrathin nanosheet subunits as excellent photocatalysts for hydrogen evolution, Small, 2019, 15, 1901024.
29. H. Wang, X. Gao, Z. Lv, T. Abdelilah and A. Lei, Recent advances in oxidative R1-H/R2-H cross-coupling with hydrogen evolution via photo-/electrochemistry, Chem. Rev., 2019, 119, 6769–6787.
30. L. F. Garay-Rodríguez, L. M. Torres-Martínez and E. Moctezuma, Photocatalytic performance of K2Ti6O13 whiskers to H2 evolution and CO2 photo-reduction, J. Energy Chem., 2019, 37, 18–28.
31. X. Li, N. Li, Y. Gao and L. Ge, Design and applications of hollow-structured nanomaterials for photocatalytic H2 evolution and CO2 reduction, Chin. J. Catal., 2022, 43, 679–707.
32. H. Bai, S. H. Lam, J. Yang, X. Cheng, S. Li, R. Jiang, L. Shao and J. Wang, A schottky-barrier-free plasmonic semiconductor photocatalyst for nitrogen fixation in a “one-stone-two-birds” Manner, Adv. Mater., 2022, 34, 2104226.
33. Z. Liu, M. Luo, S. Yuan, L. Meng, W. Ding, S. Su, Y. Cao, Y. Wang and X. Li, Boron-doped graphene quantum dot/bismuth molybdate composite photocatalysts for efficient photocatalytic nitrogen fixation reactions, J. Colloid Interface Sci., 2023, 650, 1301–1311.
34. D. Zhang, W. He, J. Ye, X. Gao, D. Wang and J. Song, Polymeric carbon nitride-derived photocatalysts for water splitting and nitrogen fixation, Small, 2021, 17, 2005149.
35. F. Dong, Z. Zhao, T. Xiong, Z. Ni, W. Zhang, Y. Sun and W.-K. Ho, In situ construction of g-C3N4/g-C3N4 metal-free heterojunction for enhanced visible-light photocatalysis, ACS Appl. Mater. Interfaces, 2013, 5, 11392–11401.
36. K. Hu, Z. Huang, L. Zeng, Z. Zhang, L. Mei, Z. Chai and W. Shi, Recent advances in MOF-based materials for photocatalytic nitrogen fixation, Eur. J. Inorg. Chem., 2022, 2022, e202100748.
37. R. Bariki, S. K. Pradhan, S. Panda, S. K. Nayak, A. R. Pati and B. G. Mishra, Hierarchical UiO-66(−NH2)/CuInS2 S-scheme photocatalyst with controlled topology for enhanced photocatalytic N2 fixation and H₂O₂ production, Langmuir, 2023, 39, 7707–7722.
38. Y. Zhao, Y. Zhao, G. I. N. Waterhouse, L. Zheng, X. Cao, F. Teng, L.-Z. Wu, C.-H. Tung, D. O'Hare and T. Zhang, Layered-Double-Hydroxide nanosheets as efficient visible-light-driven photocatalysts for dinitrogen fixation, Adv. Mater., 2017, 29, 1703828.
39. G. Zhang, Y. Li, C. He, X. Ren, P. Zhang and H. Mi, Recent progress in 2D catalysts for photocatalytic and electrocatalytic artificial nitrogen reduction to ammonia, Adv. Energy Mater., 2021, 11, 2003294.
40. P. Liu, Z. Huang, S. Yang, J. Du, Y. Zhang, R. Cao, C. Chen, L. Li, T. Chen, G. Wang, D. Rao, X. Zheng and X. Hong, Support amorphization engineering regulates single-atom Ru as an electron pump for nitrogen photofixation, ACS Catal., 2022, 12, 8139–8146.
41. B. Hu, B.-H. Wang, L. Chen, Z.-J. Bai, W. Zhou, J.-K. Guo, S. Shen, T.-L. Xie, C.-T. Au, L.-L. Jiang and S.-F. Yin, Electronic modulation of the interaction between Fe single atoms and WO2.72−x for photocatalytic N2 reduction, ACS Catal., 2022, 12, 11860–11869.
42. T. Hu, G. Jiang, Y. Yan, S. Lan, J. Xie, Q. Zhang and Y. Li, Facile synthesis of Fe single-atom porous photocatalysts via direct metal atomization achieving efficient photocatalytic nitrogen fixation, J. Mater. Sci. Technol., 2023, 167, 248–257.
43. S. Liu, Y. Wang, S. Wang, M. You, S. Hong, T.-S. Wu, Y.-L. Soo, Z. Zhao, G. Jiang, Q. Jieshan, B. Wang and Z. Sun, Photocatalytic fixation of nitrogen to ammonia by single Ru atom decorated TiO2 nanosheets, ACS Sustainable Chem. Eng., 2019, 7, 6813–6820.
44. J. Zhang, W. Yang and C. Sun, Synergetic effect between non-metals and dual metal catalysts for nitrogen reduction reaction, Inorg. Chem. Front., 2023, 10, 4746–4753.
45. B. Qiao, A. Wang, X. Yang, L. F. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li and T. Zhang, Single-atom catalysis of CO oxidation using Pt1/FeOx, Nat. Chem., 2011, 3, 634–641.
46. G. Huang, G. Lin, Q. Niu, J. Bi and L. Wu, Covalent triazine-based frameworks confining cobalt single atoms for photocatalytic CO2 reduction and hydrogen production, J. Mater. Sci. Technol., 2022, 116, 41–49.
47. A. Chipojola Mtukula, X.-D. Zhang, S.-Z. Hou, J.-M. Huang, M. Xu and Z.-Y. Gu, Metal-organic frameworks for efficient electrochemical reduction of carbon dioxide, Eur. J. Inorg. Chem., 2023, 26, e202300170.
48. S. Si, H. Shou, Y. Mao, X. Bao, G. Zhai, K. Song, Z. Wang, P. Wang, Y. Liu, Z. Zheng, Y. Dai, L. Song, B. Huang and H. Cheng, Low-coordination single Au atoms on ultrathin ZnIn2S4 nanosheets for selective photocatalytic CO2 reduction towards CH4, Angew. Chem., Int. Ed., 2022, 61, e202209446.
49. T. He, Z. Zhao, R. Liu, X. Liu, B. Ni, Y. Wei, Y. Wu, W. Yuan, H. Peng, Z. Jiang and Y. Zhao, Porphyrin-based covalent organic frameworks anchoring Au single atoms for photocatalytic nitrogen fixation, J. Am. Chem. Soc., 2023, 145, 6057–6066.
50. C.-F. Li, W.-G. Pan, Z.-R. Zhang, T. Wu and R.-T. Guo, Recent progress of single-atom photocatalysts applied in energy conversion and environmental protection, Small, 2023, 19, 2300460.
51. H. Zhang, J. Wei, J. Dong, G. Liu, L. Shi, P. An, G. Zhao, J. Kong, X. Wang, X. Meng, J. Zhang and J. Ye, Efficient visible-light-driven carbon dioxide reduction by a single-atom implanted metal–organic framework, Angew. Chem., Int. Ed., 2016, 55, 14310–14314.
52. C. Gao, S. Chen, Y. Wang, J. Wang, X. Zheng, J. Zhu, L. Song, W. Zhang and Y. Xiong, Heterogeneous single-atom catalyst for visible-light-driven high-turnover CO2 reduction: the role of electron transfer, Adv. Mater., 2018, 30, 1704624.
53. L. Cheng, X. Yue, L. Wang, D. Zhang, P. Zhang, J. Fan and Q. Xiang, Dual-single-atom tailoring with bifunctional integration for high-performance CO2 photoreduction, Adv. Mater., 2021, 33, 2105135.
54. J. Li, P. Liu, Y. Tang, H. Huang, H. Cui, D. Mei and C. Zhong, Single-atom Pt–N3 sites on the stable covalent triazine framework nanosheets for photocatalytic N2 fixation, ACS Catal., 2020, 10, 2431–2442.
55. G. Ren, J. Zhao, Z. Zhao, Z. Li, L. Wang, Z. Zhang, C. Li and X. Meng, Defects-induced single-atom anchoring on metal–organic frameworks for high-efficiency photocatalytic nitrogen reduction, Angew. Chem., Int. Ed., 2024, 63, e202314408.
56. B. G. Seo, J. Park, B. J. Kim, G. D. Han, K. H. Park, H. Park and J. H. Shim, Hetero-structured palladium-coated zinc oxide photocatalysts for sustainable water treatment, ECS Meeting Abstracts, 2023, vol. MA2023-01, p. 2407.
57. L. Wang, Y. Xia and J. Yu, Hydrogen-bond activation of N2 molecules and photocatalytic nitrogen fixation, Chem, 2021, 7, 1983–1985.
58. M. He, Z. Fang, P. Wang, Y. You and Z. Li, Recent progress in photocatalytic chemical fixation of carbon dioxide, ACS Sustainable Chem. Eng., 2023, 11, 12194–12217.
59. G. Ren, S. Liu, M. Shi, Z. Zhang, Z. Li and X. Meng, Ultraviolet light-modulated defects on BiOBr to improve the photocatalytic fixation of nitrogen to ammonia, Sol. RRL, 2022, 6, 2200653.
60. K. Li, W. Cai, Z. Zhang, H. Xie, Q. Zhong and H. Qu, Boron doped C3N5 for photocatalytic nitrogen fixation to ammonia: The key role of boron in nitrogen activation and mechanism, Chem. Eng. J., 2022, 435, 135017.
61. J. Liu, F. Li, J. Lu, R. Li, Y. Wang, Y. Wang, X. Zhang, C. Fan and R. Zhang, Atomically dispersed palladium-ethylene glycol- bismuth oxybromide for photocatalytic nitrogen fixation: Insight of molecular bridge mechanism, J. Colloid Interface Sci., 2021, 603, 17–24.
62. H. Li, J. Wang, Z. Ruan, P. Nan, B. Ge, M. Cheng, L. Yang, X. Li, Q. Liu, B. Pan, Q. Zhang, C. Xiao and Y. Xie, Electron transfer bridge inducing polarization of nitrogen molecules for enhanced photocatalytic nitrogen fixation, Mater. Horiz., 2023, 10, 5053–5059.
63. C.-L. Chen, H.-Y. Wang, J.-P. Li, L.-S. Long, X.-J. Kong and L.-S. Zheng, Assembling lanthanide–transition metal clusters on TiO2 for photocatalytic nitrogen fixation, Inorg. Chem. Front., 2022, 9, 2862–2868.
64. M. A. Khan, M. A. Nadeem and H. Idriss, Ferroelectric polarization effect on surface chemistry and photo-catalytic activity: A review, Surf. Sci. Rep., 2016, 71, 1–31.
65. M. Shen, T. Ding, W. H. Rackers, C. Tan, K. Mahmood, M. D. Lew and B. Sadtler, Single-molecule colocalization of redox reactions on semiconductor photocatalysts connects surface heterogeneity and charge-carrier separation in bismuth oxybromide, J. Am. Chem. Soc., 2021, 143, 11393–11403.
66. X. Zhang, Z. Li, T. Liu, M. Li, C. Zeng, H. Matsumoto and H. Han, Water oxidation sites located at the interface of Pt/SrTiO3 for photocatalytic overall water splitting, Chin. J. Catal., 2022, 43, 2223–2230.
67. L. Chang, S. T. Yong, S. P. Chai, L. K. Putri, L. L. Tan and A. R. Mohamed, A review of methanol photoreforming: elucidating the mechanisms, photocatalysts and recent advancement strategies, Mater. Today Chem., 2023, 27, 101334.
68. M. Li, Y. Cui, X. Zhang, Y. Luo, Y. Dai and Y. Huang, Screening a suitable Mo form supported on graphdiyne for effectively electrocatalytic N2 reduction reaction: From atomic catalyst to cluster catalyst, J. Phys. Chem. Lett., 2020, 11, 8128–8137.
69. T. Nakao, K. Ogasawara, M. Kitano, S. Matsuishi, P. V. Sushko and H. Hosono, Ship-in-a-bottle synthesis of high concentration of N2 molecules in a cage-structured electride, J. Phys. Chem. Lett., 2021, 12, 1295–1299.
70. X. Niu, A. Shi, D. Sun, S. Xiao, T. Zhang, Z. Zhou, X. A. Li and J. Wang, Photocatalytic ammonia synthesis: mechanistic insights into N2 activation at oxygen vacancies under visible light excitation, ACS Catal., 2021, 11, 14058–14066.
71. A. R. Singh, B. A. Rohr, J. A. Schwalbe, M. Cargnello, K. Chan, T. F. Jaramillo, I. Chorkendorff and J. K. Nørskov, Electrochemical ammonia synthesis—The selectivity challenge, ACS Catal., 2017, 7, 706–709.
72. J. Wong, S. T. Omelchenko and H. A. Atwater, Impact of semiconductor band tails and band filling on photovoltaic efficiency limits, ACS Energy Lett., 2021, 6, 52–57.
73. K. Sridharan, S. Shenoy, S. G. Kumar, C. Terashima, A. Fujishima and S. Pitchaimuthu, Advanced two-dimensional heterojunction photocatalysts of stoichiometric and non-stoichiometric bismuth oxyhalides with graphitic carbon nitride for sustainable energy and environmental applications, Catalysts, 2021, 11(4), 426.
74. X. Li, X. Sun, L. Zhang, S. Sun and W. Wang, Efficient photocatalytic fixation of N2 by KOH-treated g-C3N4, J. Mater. Chem. A, 2018, 6, 3005–3011.
75. X. Chen, X. Zhang, Y.-H. Li, M.-Y. Qi, J.-Y. Li, Z.-R. Tang, Z. Zhou and Y.-J. Xu, Transition metal doping BiOBr nanosheets with oxygen vacancy and exposed {102} facets for visible light nitrogen fixation, Appl. Catal., B, 2021, 281, 119516.
76. S. Sun, Q. An, W. Wang, L. Zhang, J. Liu and W. A. Goddard III, Efficient photocatalytic reduction of dinitrogen to ammonia on bismuth monoxide quantum dots, J. Mater. Chem. A, 2017, 5, 201–209.
77. H. Ali, M. Masar, A. C. Guler, M. Urbanek, M. Machovsky and I. Kuritka, Heterojunction-based photocatalytic nitrogen fixation: principles and current progress, Nanoscale Adv., 2021, 3, 6358–6372.
78. P. Li, Z. Zhou, Q. Wang, M. Guo, S. Chen, J. Low, R. Long, W. Liu, P. Ding, Y. Wu and Y. Xiong, Visible-light-driven nitrogen fixation catalyzed by Bi5O7Br nanostructures: enhanced performance by oxygen vacancies, J. Am. Chem. Soc., 2020, 142, 12430–12439.
79. H. Li, J. Shang, Z. Ai and L. Zhang, Correction to “Efficient visible light nitrogen fixation with BiOBr nanosheets of oxygen vacancies on the exposed {001} Facets”, J. Am. Chem. Soc., 2018, 140, 526–526.
80. X. Kong, K. Chang, X. Tao and L. Huang, Oxygen vacancy and Z-scheme heterojunction cooperate to promote visible light nitrogen photofixation, Chem. Eng. J., 2023, 475, 146159.
81. L. Zhang, F. Xie, J. Liu, Z. Sun, X. Zhang, Y. Wang, Y. Wang, R. Li and C. Fan, Light-switchable oxygen vacancies enhanced nitrogen fixation performance on BiOBr: mechanism of formation, reconversion and function, Chem. Eng. J., 2022, 450, 138066.
82. J. Li, M. Zheng, F. Wei, C. Dong, Z. Xiu, W. Mu, X. Zhou, Y. Ding and X. Han, Fe doped InVO4 nanosheets with rich surface oxygen vacancies for enhanced electrochemical nitrogen fixation, Chem. Eng. J., 2022, 431, 133383.
83. W. Xie, J. Huang, L. Huang, S. Geng, S. Song, P. Tsiakaras and Y. Wang, Novel fluorine-doped cobalt molybdate nanosheets with enriched oxygen-vacancies for improved oxygen evolution reaction activity, Appl. Catal., B, 2022, 303, 120871.
84. M. H. Urgesa, D. F. A. Putra, A. Qadir, U. A. Khan, T.-C. Huang, Y. X. Chiu, J. H. Lin and R. T. Ginting, in Photocatalytic activities for environmental remediation and energy conversion, ed. H. Abdullah, Springer Nature Singapore, Singapore, 2023.
85. Y. Fu, Y. Liao, P. Li, H. Li, S. Jiang, H. Huang, W. Sun, T. Li, H. Yu, K. Li, H. Li, B. Jia and T. Ma, Layer structured materials for ambient nitrogen fixation, Coord. Chem. Rev., 2022, 460, 214468.
86. B. M. Comer and A. J. Medford, Analysis of photocatalytic nitrogen fixation on rutile TiO2(110), ACS Sustainable Chem. Eng., 2018, 6, 4648–4660.
87. Q. Liu, J. Yuan, Z. Gan, C. Liu, J. Li, Y. Liang and R. Chen, Photocatalytic N2 reduction: uncertainties in the determination of ammonia production, ACS Sustainable Chem. Eng., 2021, 9, 560–568.
88. Q. Hao, C. Liu, G. Jia, Y. Wang, H. Arandiyan, W. Wei and B.-J. Ni, Catalytic reduction of nitrogen to produce ammonia by bismuth-based catalysts: state of the art and future prospects, Mater. Horiz., 2020, 7, 1014–1029.
89. M.-H. Vu, C.-C. Nguyen and T.-O. Do, Synergistic effect of Fe doping and plasmonic Au nanoparticles on W18O49 nanorods for enhancing photoelectrochemical nitrogen reduction, ACS Sustainable Chem. Eng., 2020, 8, 12321–12330.
90. X. Yao, W. Zhang, J. Huang, Z. Du, X. Hong, X. Chen, X. Hu and X. Wang, Enhanced photocatalytic nitrogen fixation of Ag/B-doped g-C3N4 nanosheets by one-step in-situ decomposition-thermal polymerization method, Appl. Catal., A, 2020, 601, 117647.
91. B. Sun, Z. Liang, Y. Qian, X. Xu, Y. Han and J. Tian, Sulfur vacancy-rich O-doped 1T-MoS2 nanosheets for exceptional photocatalytic nitrogen fixation over CdS, ACS Appl. Mater. Interfaces, 2020, 12, 7257–7269.
92. J. H. Montoya, C. Tsai, A. Vojvodic and J. K. Norskov, The Challenge of electrochemical ammonia synthesis: A new perspective on the role of nitrogen scaling relations, ChemSusChem, 2015, 8, 2180–2186.
93. M. T. Tang, X. Liu, Y. Ji, J. K. Norskov and K. Chan, Modeling hydrogen evolution reaction kinetics through explicit water–metal interfaces, J. Phys. Chem. C, 2020, 124, 28083–28092.
94. X. Yu, H. Qiu, Z. Wang, B. Wang, Q. Meng, S. Sun, Y. Tang and K. Zhao, Constructing the Z-scheme TiO2/Au/BiOI nanocomposite for enhanced photocatalytic nitrogen fixation, Appl. Surf. Sci., 2021, 556, 149785.
95. Y. Huang, J. Xiong, Z. Zou and Z. Chen, Emerging strategies for the synthesis of correlated single atom catalysts, Adv. Mater., 2024, 2312182.
96. J. Xi, H. S. Jung, Y. Xu, F. Xiao, J. W. Bae and S. Wang, Synthesis strategies, catalytic applications, and performance regulation of single-atom catalysts, Adv. Funct. Mater., 2021, 31, 2008318.
97. Z.-H. Xue, D. Luan, H. Zhang and X. W. Lou, Single-atom catalysts for photocatalytic energy conversion, Joule, 2022, 6, 92–133.
98. M. Li, J. Yu, Q. Liu, J. Liu, R. Chen, J. Zhu, R. Li and J. Wang, Atomically dispersed Fe–N5 sites anchored in porous N-doped carbon nanofibers for effective hydrogen Evolution Reaction, ACS Sustainable Chem. Eng., 2022, 10, 13505–13513.
99. W. Jiang, Y. Zhao, X. Zong, H. Nie, L. Niu, L. An, D. Qu, X. Wang, Z. Kang and Z. Sun, Photocatalyst for high-performance H2 production: Ga-doped polymeric carbon nitride, Angew. Chem., Int. Ed., 2021, 60, 6124–6129.
100. D. Liu, C. Zhang, J. Shi, Y. Shi, T. T. T. Nga, M. Liu, S. Shen and C.-L. Dong, Defect engineering simultaneously regulating exciton dissociation in carbon nitride and local electron density in Pt single atoms toward highly efficient photocatalytic hydrogen production, Small, 2024, 20, 2310289.
101. B. Zhang, X. Song, Z. Wang, B. Xu, W. Ye, L. Zhang, H. Lin, T. Ji, X. Lu and J. Zhu, Atomic layer deposition of Pd nanoparticles and single atoms on self-supported carbon monolithic catalysts synergistically boosts the hydrogen evolution reaction, J. Mater. Chem. A, 2024, 12, 28503–28511.
102. M. Xiao, L. Zhang, B. Luo, M. Lyu, Z. Wang, H. Huang, S. Wang, A. Du and L. Wang, Molten-salt-mediated synthesis of an atomic nickel co-catalyst on TiO2 for improved photocatalytic H2 evolution, Angew. Chem., Int. Ed., 2020, 59, 7230–7234.
103. A. Wang, J. Li and T. Zhang, Heterogeneous single-atom catalysis, Nat. Rev. Chem., 2018, 2, 65–81.
104. M. B. Gawande, P. Fornasiero and R. Zboril, Carbon-based single-atom catalysts for advanced applications, ACS Catal., 2020, 10, 2231–2259.
105. C. Gao, J. Low, R. Long, T. Kong, J. Zhu and Y. Xiong, Heterogeneous single-atom photocatalysts: fundamentals and applications, Chem. Rev., 2020, 120, 12175–12216.
106. F. Kraushofer and G. S. Parkinson, Single-atom catalysis: insights from model systems, Chem. Rev., 2022, 122, 14911–14939.
107. Z. Geng, Y. Liu, X. Kong, P. Li, K. Li, Z. Liu, J. Du, M. Shu, R. Si and J. Zeng, Achieving a record-high yield rate of 120.9 for N2 electrochemical reduction over Ru single-atom catalysts, Adv. Mater., 2018, 30, 1803498.
108. B. Yu, H. Li, J. White, S. Donne, J. Yi, S. Xi, Y. Fu, G. Henkelman, H. Yu, Z. Chen and T. Ma, Tuning the catalytic preference of ruthenium catalysts for nitrogen reduction by atomic dispersion, Adv. Funct. Mater., 2020, 30, 1905665.
109. X. Wang, W. Wang, M. Qiao, G. Wu, W. Chen, T. Yuan, Q. Xu, M. Chen, Y. Zhang, X. Wang, J. Wang, J. Ge, X. Hong, Y. Li, Y. Wu and Y. Li, Atomically dispersed Au1 catalyst towards efficient electrochemical synthesis of ammonia, Sci. Bull., 2018, 63, 1246–1253.
110. H. Zou, W. Rong, S. Wei, Y. Ji and L. Duan, Regulating kinetics and thermodynamics of electrochemical nitrogen reduction with metal single-atom catalysts in a pressurized electrolyser, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 29462–29468.
111. Y. Chen, R. Guo, X. Peng, X. Wang, X. Liu, J. Ren, J. He, L. Zhuo, J. Sun, Y. Liu, Y. Wu and J. Luo, Highly productive electrosynthesis of ammonia by molecule-targeting single Ag sites, ACS Nano, 2020, 14, 6938–6946.
112. H. Yang, Y. Liu, Y. Luo, S. Lu, B. Su and J. Ma, Achieving high activity and selectivity of nitrogen reduction via Fe–N3 coordination on iron single-atom electrocatalysts at ambient conditions, ACS Sustainable Chem. Eng., 2020, 8, 12809–12816.
113. S. Zhang, M. Jin, T. Shi, M. Han, Q. Sun, Y. Lin, Z. Ding, L. R. Zheng, G. Wang, Y. Zhang, H. Zhang and H. Zhao, Electrocatalytically active Fe-(O-C2)4 single-atom sites for efficient reduction of nitrogen to ammonia, Angew. Chem., Int. Ed., 2020, 59, 13423–13429.
114. P. Kumar, T. A. Al-Attas, J. Hu and M. G. Kibria, Single atom catalysts for selective methane oxidation to oxygenates, ACS Nano, 2022, 16, 8557–8618.
115. L. Wang, J. Wang, X. Gao, C. Chen, Y. Da, S. Wang, J. Yang, Z. Wang, J. Song, T. Yao, W. Zhou, H. Zhou and Y. Wu, Periodic one-dimensional single-atom arrays, J. Am. Chem. Soc., 2022, 144, 15999–16005.
116. B.-B. Xu, X.-B. Fu, X.-M. You, E. Zhao, F.-F. Li, Z. Chen, Y.-X. Li, X. L. Wang and Y.-F. Yao, Synergistic promotion of single-atom Co surrounding a PtCo alloy based on a g-C3N4 nanosheet for overall water splitting, ACS Catal., 2022, 12, 6958–6967.
117. J. Yu, C. Chen, Q. Zhang, J. Lin, X. Yang, L. Gu, H. Zhang, Z. Liu, Y. Wang, S. Zhang, X. Wang and L. Guo, Au Atoms anchored on amorphous C3N4 for single-site raman enhancement, J. Am. Chem. Soc., 2022, 144, 21908–21915.
118. O. Pinto-Burgos, J. Castro-Gutiérrez, P. S. Poon, M. T. Izquierdo, A. Celzard, V. Fierro and J. Matos, Enhancing electrochemical capacitor performance of N-doped tannin-derived carbons by hydrothermal treatment in ammonia, J. Power Sources, 2024, 602, 234332.
119. J. Wang, H. Huang, P. Wang, S. Wang and J. Li, N, S synergistic effect in hierarchical porous carbon for enhanced NRR performance, Carbon, 2021, 179, 358–364.
120. S. Zhao, X. Lu, L. Wang, J. Gale and R. Amal, Carbon-based metal-free catalysts for electrocatalytic reduction of nitrogen for synthesis of ammonia at ambient conditions, Adv. Mater., 2019, 31, 1805367.
121. S. S. Sekhon and J.-S. Park, Biomass-derived N-doped porous carbon nanosheets for energy technologies, Chem. Eng. J., 2021, 425, 129017.
122. W. Zhang, S. Zhan, Q. Qin, T. Heil, X. Liu, J. Hwang, T. H. Ferber, J. P. Hofmann and M. Oschatz, Electrochemical generation of catalytically active edge sites in C2N-type carbon materials for artificial nitrogen fixation, Small, 2022, 18, 2204116.
123. P. Sun, J. Huang, F. Xu, J. Xu, T. Lin, W. Zhao, W. Dong and F. Huang, Boron-induced nitrogen fixation in 3D carbon materials for supercapacitors, ACS Appl. Mater. Interfaces, 2020, 12, 28075–28082.
124. X. Han, C. S. Gerke, S. Banerjee, M. Zubair, J. Jiang, N. M. Bedford, E. M. Miller and V. S. Thoi, Strategic design of MoO2 nanoparticles supported by carbon nanowires for enhanced electrocatalytic nitrogen reduction, ACS Energy Lett., 2020, 5, 3237–3243.
125. L. Prati, C. E. Chan-Thaw, S. Campisi and A. Villa, N-modified carbon-based materials: nanoscience for catalysis, Chem. Rec., 2016, 16, 2187–2197.
126. S. Cao and J. Yu, Carbon-based H2-production photocatalytic materials, J. Photochem. Photobiol., C, 2016, 27, 72–99.
127. Z. Li, K. Li, P. Du, M. Mehmandoust, F. Karimi and N. Erk, Carbon-based photocatalysts for hydrogen production: A review, Chemosphere, 2022, 308, 135998.
128. C. Feng, Z.-P. Wu, K.-W. Huang, J. Ye and H. Zhang, Surface Modification of 2D photocatalysts for solar energy conversion, Adv. Mater., 2022, 34, 2200180.
129. M. A. Shipman and M. D. Symes, Recent progress towards the electrosynthesis of ammonia from sustainable resources, Catal. Today, 2017, 286, 57–68.
130. S. Wang, D. Guo, M. Zong, C. Fan, X. Jun and D.-H. Wang, Unravelling the strong metal-support interaction between Ru quantum dots and g-C3N4 for visible-light photocatalytic nitrogen fixation, Appl. Catal., A, 2021, 617, 118112.
131. L. Li, Y. Yu, S. Lin, W. Chu, D. Sun, Q. Su, S. Ma, G. Du and B. Xu, Single ruthenium atom supported on g-C3N4 as an efficient photocatalyst for nitrogen fixation in ultra-pure water, Catal. Commun., 2021, 153, 106294.
132. Z. Ding, X. Li, C. Kang, S. Yan, D. Zhao, H. Cai, S.-Y. Zhang and Y.-J. Zeng, Single Ru atoms confined into MOF/C3N4 for dual improved photocatalytic carbon dioxide reduction and nitrogen fixation, Chem. Eng. J., 2023, 473, 145256.
133. M.-M. Shi, D. Bao, S.-J. Li, B.-R. Wulan, J.-M. Yan and Q. Jiang, Anchoring PdCu amorphous nanocluster on graphene for electrochemical reduction of N2 to NH3 under ambient conditions in aqueous solution, Adv. Energy Mater., 2018, 8, 1800124.
134. L. Han, Z. Ren, P. Ou, H. Cheng, N. Rui, L. Lin, X. Liu, L. Zhuo, J. Song, J. Sun, J. Luo and H. L. Xin, Modulating Single-atom palladium sites with copper for enhanced ambient ammonia electrosynthesis, Angew. Chem., Int. Ed., 2021, 60, 345–350.
135. J. Li, Q. Li, Y. Chen, S. Lv, X. Liao and Y. Yao, Size effects of Ag nanoparticle for N2 photofixation over Ag/g-C3N4: Built-in electric fields determine photocatalytic performance, Colloids Surf., A, 2021, 626, 127053.
136. Y. Wang, R. Zhao, F. Wang, Y. Liu, X. Yu, L. Chen, Y. Yao, S. Lu and X. Liao, Ultralow-temperature synthesis of small Ag-doped carbon nitride for nitrogen photofixation, Catal. Sci. Technol., 2020, 10, 7652–7660.
137. Y. Guo, J. Yang, D. Wu, H. Bai, Z. Yang, J. Wang and B. Yang, Au nanoparticle-embedded, nitrogen-deficient hollow mesoporous carbon nitride spheres for nitrogen photofixation, J. Mater. Chem. A, 2020, 8, 16218–16231.
138. G. Liu, Y. Huang, H. Lv, H. Wang, Y. Zeng, M. Yuan, Q. Meng and C. Wang, Confining single-atom Pd on g-C3N4 with carbon vacancies towards enhanced photocatalytic NO conversion, Appl. Catal., B, 2021, 284, 119683.
139. J. Li, Q. Li, Y. Chen, S. Lu, X. Liao and Y. Yao, Size effects of Ag nanoparticle for N2 photofixation over Ag/g-C3N4: Built-in electric fields determine photocatalytic performance, Colloids Surf., A, 2021, 626, 127053.
140. X.-W. Guo, S.-M. Chen, H.-J. Wang, Z.-M. Zhang, H. Lin, L. Song and T.-B. Lu, Single-atom molybdenum immobilized on photoactive carbon nitride as efficient photocatalysts for ambient nitrogen fixation in pure water, J. Mater. Chem. A, 2019, 7, 19831–19837.
141. Y. Kong, T. He, A. R. Puente Santiago, D. Liu, A. Du, S. Wang and H. Pan, Unravelling the reaction mechanisms of N2 fixation on molybdenum nitride: A full DFT study from the pristine surface to heteroatom anchoring, ChemSusChem, 2021, 14, 3257–3266.
142. S.-Y. Lv, C.-X. Huang, G. Li and L.-M. Yang, Electrocatalytic mechanism of N2 reduction Reaction by single-atom catalyst rectangular TM-TCNQ monolayers, ACS Appl. Mater. Interfaces, 2021, 13, 29641–29653.
143. Y. Ying, K. Fan, X. Luo, J. Qiao and H. Huang, Transition metal-tetracyanoquinodimethane monolayers as single-atom catalysts for the electrocatalytic nitrogen reduction reaction, Mater. Adv., 2020, 1, 1285–1292.
144. A. P. Hamsa, M. Arulprakasam and S. M. Unni, Electrochemical nitrogen fixation on single metal atom catalysts, Chem. Commun., 2023, 59, 10689–10710.
145. Q. Liu, C. Zeng, Z. Xie, L. Ai, Y. Liu, Q. Zhou, J. Jiang, H. Sun and S. Wang, Cobalt@nitrogen-doped bamboo-structured carbon nanotube to boost photocatalytic hydrogen evolution on carbon nitride, Appl. Catal., B, 2019, 254, 443–451.
146. S. Chen, Y. Hau Ng, J. Liao, Q. Gao, S. Yang, F. Peng, X. Zhong, Y. Fang and S. Zhang, FeCo alloy@N-doped graphitized carbon as an efficient cocatalyst for enhanced photocatalytic H2 evolution by inducing accelerated charge transfer, J. Energy Chem., 2021, 52, 92–101.
147. L. Wang, Q. Zhang, T. Wei, F. Li, Z. Sun and L. Xu, WC and cobalt nanoparticles embedded in nitrogen-doped carbon 3D nanocage derived from H3PW12O40@ZIF-67 for photocatalytic nitrogen fixation, J. Mater. Chem. A, 2021, 9, 2912–2918.
148. W. Zhang, Y. Fu, Q. Peng, Q. Yao, X. Wang, A. Yu and Z. Chen, Supramolecular preorganization effect to access single cobalt sites for enhanced photocatalytic hydrogen evolution and nitrogen fixation, Chem. Eng. J., 2020, 394, 124822.
149. P. Huang, W. Liu, Z. He, C. Xiao, T. Yao, Y. Zou, C. Wang, Z. Qi, W. Tong, B. Pan, S. Wei and Y. Xie, Single atom accelerates ammonia photosynthesis, Sci. China: Chem., 2018, 61, 1187–1196.
150. G. Pan, W. Zhang, T. Liu, Q. Tan, B. Wei, K. Ye, Y. Yang, D. Han, Z. Liu and L. Niu, Atomically dispersed s-block metal calcium site modified mesoporous g-C3N4 for boosting photocatalytic N2 reduction, Catal. Sci. Technol., 2023, 13, 111–118.
151. G. Ren, M. Shi, S. Liu, Z. Li, Z. Zhang and X. Meng, Molecular-level insight into photocatalytic reduction of N2 over Ruthenium single atom modified TiO2 by electronic Metal-support interaction, Chem. Eng. J., 2023, 454, 140158.
152. Y. Tang, W. Zhou, Q. Shang, Y. Guo, H. Hu, Z. Li, Y. Zhang, L. Liu, H. Wang, X. Tan, T. Yu and J. Ye, Discerning the mechanism of expedited interfacial electron transformation boosting photocatalytic hydrogen evolution by metallic 1T-WS2-induced photothermal effect, Appl. Catal., B, 2022, 310, 121295.
153. J. Qin, W. Zhao, X. Hu, J. Li, P. Ndokoye and B. Liu, Exploring the N2 adsorption and activation mechanisms over the 2H/1T mixed-phase ultrathin Mo1−xWxS2 nanosheets for boosting N2 photosynthesis, ACS Appl. Mater. Interfaces, 2021, 13, 7127–7134.
154. B. Hu, B.-H. Wang, Z.-J. Bai, L. Chen, J.-K. Guo, S. Shen, T.-L. Xie, C.-T. Au, L.-L. Jiang and S.-F. Yin, Regulating MoS2 edge site for photocatalytic nitrogen fixation: A theoretical and experimental study, Chem. Eng. J., 2022, 442, 136211.
155. R. Yang, Y. Fan, Y. Zhang, L. Mei, R. Zhu, J. Qin, J. Hu, Z. Chen, Y. Hau Ng, D. Voiry, S. Li, Q. Lu, Q. Wang, J. C. Yu and Z. Zeng, 2D transition metal dichalcogenides for photocatalysis, Angew. Chem., Int. Ed., 2023, 62, e202218016.
156. X. Ma, J. Hu, M. Zheng, D. Li, H. Lv, H. He and C. Huang, N2 reduction using single transition-metal atom supported on defective WS2 monolayer as promising catalysts: A DFT study, Appl. Surf. Sci., 2019, 489, 684–692.
157. H. H. Shin, W. Yang and D.-K. Lim, Gold and nanosized titanium carbide MXene heterozygotes as highly efficient visible-light responsive photocatalysts for ammonia synthesis, Carbon, 2023, 214, 118359.
158. W. Gao, X. Li, S. Luo, Z. Luo, X. Zhang, R. Huang and M. Luo, In situ modification of cobalt on MXene/TiO2 as composite photocatalyst for efficient nitrogen fixation, J. Colloid Interface Sci., 2021, 585, 20–29.
159. K. Huang, C. Li, H. Li, G. Ren, L. Wang, W. Wang and X. Meng, Photocatalytic applications of two-dimensional Ti3C2 MXenes: A Review, ACS Appl. Nano Mater., 2020, 3, 9581–9603.
160. Q. Zhong, Y. Li and G. Zhang, Two-dimensional MXene-based and MXene-derived photocatalysts: Recent developments and perspectives, Chem. Eng. J., 2021, 409, 128099.
161. S. H. W. Kok, J. Lee, L.-L. Tan, W.-J. Ong and S.-P. Chai, MXene—A new paradigm toward artificial nitrogen fixation for sustainable ammonia generation: synthesis, properties, and future outlook, ACS Mater. Lett., 2022, 4, 212–245.
162. J. Sun, W. Kong, Z. Jin, Y. Han, L. Ma, X. Ding, Y. Niu and Y. Xu, Recent advances of MXene as promising catalysts for electrochemical nitrogen reduction reaction, Chin. Chem. Lett., 2020, 31, 953–960.
163. R. Li, L. Zhang, L. Shi and P. Wang, MXene Ti3C2: An effective 2D light-to-heat conversion material, ACS Nano, 2017, 11, 3752–3759.
164. B. Chang, Y. Guo, D. Wu, L. Li, B. Yang and J. Wang, Plasmon-enabled N2 photofixation on partially reduced Ti3C2 MXene, Chem. Sci., 2021, 12, 11213–11224.
165. Y.-S. Wei, M. Zhang, R. Zou and Q. Xu, Metal–organic framework-based catalysts with single metal sites, Chem. Rev., 2020, 120, 12089–12174.
166. V. Hasija, S. Patial, P. Raizada, A. Aslam Parwaz Khan, A. M. Asiri, Q. Van Le, V.-H. Nguyen and P. Singh, Covalent organic frameworks promoted single metal atom catalysis: Strategies and applications, Coord. Chem. Rev., 2022, 452, 214298.
167. D. Wang and Y. Zhao, Single-atom engineering of metal-organic frameworks toward healthcare, Chemistry, 2021, 7, 2635–2671.
168. T. He and Y. Zhao, Engineering covalent/metal-organic frameworks with single atoms toward artificial photocatalytic nitrogen fixation, Chem Catal., 2023, 3, 100795.
169. S. Shang, W. Xiong, C. Yang, B. Johannessen, R. Liu, H.-Y. Hsu, Q. Gu, M. K. H. Leung and J. Shang, Atomically dispersed iron metal site in a porphyrin-based metal–organic framework for photocatalytic nitrogen fixation, ACS Nano, 2021, 15, 9670–9678.
170. Z. Chen, K. Mou, X. Wang and L. Liu, Nitrogen-doped graphene quantum dots enhance the activity of Bi2O3 nanosheets for electrochemical reduction of CO2 in a wide negative potential region, Angew. Chem., Int. Ed., 2018, 57, 12790–12794.
171. C. Dong, S. Lu, S. Yao, R. Ge, Z. Wang, Z. Wang, P. An, Y. Liu, B. Yang and H. Zhang, Colloidal synthesis of ultrathin monoclinic BiVO4 nanosheets for Z-scheme overall water splitting under visible light, ACS Catal., 2018, 8, 8649–8658.
172. D. Jiang, W. Ma, P. Xiao, L. Shao, D. Li and M. Chen, Enhanced photocatalytic activity of graphitic carbon nitride/carbon nanotube/Bi2WO6 ternary Z-scheme heterojunction with carbon nanotube as efficient electron mediator, J. Colloid Interface Sci., 2018, 512, 693–700.
173. S. Yang, X. Zhu and S. Peng, Prospect prediction of terminal clean power consumption in china via LSSVM algorithm based on improved evolutionary game theory, Energies, 2020, 13(8), 1–17.
174. H. Li, J. Shang, Z. Ai and L. Zhang, Efficient visible light nitrogen fixation with BiOBr nanosheets of oxygen vacancies on the exposed {001} Facets, J. Am. Chem. Soc., 2015, 137, 6393–6399.
175. S. Wang, F. Ichihara, H. Pang, H. Chen and J. Ye, Nitrogen fixation reaction derived from nanostructured catalytic materials, Adv. Funct. Mater., 2018, 28, 1803309.
176. S. C. Jesudass, S. Surendran, J. Y. Kim, T.-Y. An, G. Janani, T.-H. Kim, J. K. Kim and U. Sim, Pathways of the electrochemical nitrogen reduction reaction: From ammonia synthesis to metal-N2 batteries, Electrochem. Energy Rev., 2023, 6, 27.
177. S. Mallikarjun Sharada, R. K. B. Karlsson, Y. Maimaiti, J. Voss and T. Bligaard, Adsorption on transition metal surfaces: Transferability and accuracy of DFT using the ADS41 dataset, Phys. Rev. B, 2019, 100, 035439.
178. A. J. R. Hensley, K. Ghale, C. Rieg, T. Dang, E. Anderst, F. Studt, C. T. Campbell, J.-S. McEwen and Y. Xu, DFT-based method for more accurate adsorption energies: An adaptive sum of energies from RPBE and vdW density functionals, J. Phys. Chem. C, 2017, 121, 4937–4945.
179. R. B. Araujo, G. L. S. Rodrigues, E. C. dos Santos and L. G. M. Pettersson, Adsorption energies on transition metal surfaces: towards an accurate and balanced description, Nat. Commun., 2022, 13, 6853.
180. E. Skúlason, T. Bligaard, S. Gudmundsdóttir, F. Studt, J. Rossmeisl, F. Abild-Pedersen, T. Vegge, H. Jónsson and J. K. Nørskov, A theoretical evaluation of possible transition metal electro-catalysts for N2 reduction, Phys. Chem. Chem. Phys., 2012, 14, 1235–1245.
181. M. Li, Q. Fang, X. Zhao, C. Xia, A. Wang, Y. Xie, F. Ma, J. She and Z. Deng, Theoretical screening of a graphyne-supported transition metal single-atom catalyst for the N2 reduction reaction, Phys. Chem. Chem. Phys., 2023, 25, 18224–18232.
182. Y. Abghoui, A. L. Garden, J. G. Howalt, T. Vegge and E. Skúlason, Electroreduction of N2 to ammonia at ambient conditions on mononitrides of Zr, Nb, Cr, and V: A DFT guide for experiments, ACS Catal., 2016, 6, 635–646.
183. V. Giulimondi, S. Mitchell and J. Pérez-Ramírez, Challenges and opportunities in engineering the electronic structure of single-atom catalysts, ACS Catal., 2023, 13, 2981–2997.
184. C. Choi, S. Back, N.-Y. Kim, J. Lim, Y.-H. Kim and Y. Jung, Suppression of hydrogen evolution reaction in electrochemical N2 reduction using single-atom catalysts: A Computational guideline, ACS Catal., 2018, 8, 7517–7525.
185. C. Ling, Y. Ouyang, Q. Li, X. Bai, X. Mao, A. Du and J. Wang, A General two-step strategy–based high-throughput screening of single atom catalysts for nitrogen fixation, Small Methods, 2019, 3, 1800376.
186. X. Guo, J. Gu, S. Lin, S. Zhang, Z. Chen and S. Huang, Tackling the activity and selectivity challenges of electrocatalysts toward the nitrogen reduction reaction via atomically dispersed biatom catalysts, J. Am. Chem. Soc., 2020, 142, 5709–5721.
187. Z. Allal, H. N. Noura, O. Salman and K. Chahine, Machine learning solutions for renewable energy systems: Applications, challenges, limitations, and future directions, J. Environ. Manage., 2024, 354, 120392.
188. P. Thakkar, S. Khatri, D. Dobariya, D. Patel, B. Dey and A. K. Singh, Advances in materials and machine learning techniques for energy storage devices: A comprehensive review, J. Energy Storage, 2024, 81, 110452.
189. H. Yang, Z. He, M. Zhang, X. Tan, K. Sun, H. Liu, N. Wang, L. Guan, C. Wang, Y. Wan, W. Wang, H. Hu and M. Wu, Reshaping the material research paradigm of electrochemical energy storage and conversion by machine learning, EcoMat, 2023, 5, e12330.
190. J. Liu, W. Luo, L. Wang, J. Zhang, X.-Z. Fu and J.-L. Luo, Toward excellence of electrocatalyst design by emerging descriptor-oriented machine learning, Adv. Funct. Mater., 2022, 32, 2110748.
191. Y. Gu, Q. Zhu, Z. Liu, C. Fu, J. Wu, Q. Zhu, Q. Jia and J. Ma, Nitrogen reduction reaction energy and pathways in metal-zeolites: deep learning and explainable machine learning with local acidity and hydrogen bonding features, J. Mater. Chem. A, 2022, 10, 14976–14988.
192. M. Kim, B. C. Yeo, Y. Park, H. M. Lee, S. S. Han and D. Kim, Artificial intelligence to accelerate the discovery of N2 electroreduction catalysts, Chem. Mater., 2020, 32, 709–720.
193. C. Sutton, L. M. Ghiringhelli, T. Yamamoto, Y. Lysogorskiy, L. Blumenthal, T. Hammerschmidt, J. R. Golebiowski, X. Liu, A. Ziletti and M. Scheffler, Crowd-sourcing materials-science challenges with the NOMAD 2018 Kaggle competition, npj Comput. Mater., 2019, 5, 111.
194. A. Belsky, M. Hellenbrandt, V. L. Karen and P. Luksch, New developments in the Inorganic Crystal Structure Database (ICSD): accessibility in support of materials research and design, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 364–369.
195. R. Allmann and R. Hinek, The introduction of structure types into the inorganic crystal structure database ICSD, Acta Crystallogr., Sect. A: Found. Crystallogr., 2007, 63, 412–417.
196. A. N. Abdalla, M. S. Nazir, H. Tao, S. Cao, R. Ji, M. Jiang and L. Yao, Integration of energy storage system and renewable energy sources based on artificial intelligence: An overview, J. Energy Storage, 2021, 40, 102811.
197. A. N. Singh, R. Anand, M. Zafari, M. Ha and K. S. Kim, Progress in single/multi atoms and 2D-nanomaterials for electro/photocatalytic nitrogen reduction: experimental, computational and machine learning developments, Adv. Energy Mater., 2024, 14, 2304106.
198. B. B. Hoar, S. Lu and C. Liu, Machine-learning-enabled exploration of morphology influence on wire-array electrodes for electrochemical nitrogen fixation, J. Phys. Chem. Lett., 2020, 11, 4625–4630.

---