Low-dimensional materials for ammonia synthesis

Abstract

Ammonia is an essential chemical due to its immense usage in agriculture, energy storage, and transportation. The synthesis of “green” ammonia via carbon-free routes and renewable energy sources is the need of the hour. In this context, photo- and/or electrocatalysis proves to be highly crucial. Low-dimensional materials (LDMs), owing to their unique properties, play a significant role in catalysis. This review presents a vast library of LDMs and broadly categorizes their catalytic performance according to their dimensionality, i.e., zero-, one-, and two-dimensional catalysts. The rational design of LDMs can significantly improve their catalytic performance, particularly in reducing small molecules like dinitrogen, nitrates, nitrites, and nitric oxides to synthesize ammonia via photo- and/or electrocatalysis. Additionally, converting nitrates and nitrites to ammonia can be beneficial in wastewater treatment and be coupled with CO₂ co-reduction or oxidative reactions to produce urea and other valuable chemicals, which are also discussed in this review. This review collates the works published in recent years in this field and offers some fresh perspectives on ammonia synthesis. Through this review, we aim to provide a comprehensive insight into the catalytic properties of the LDMs, which are expected to enhance the efficiency of ammonia production and promote the synthesis of value-added products.

---

Apabrita Mallick

Dr Apabrita Mallick earned her PhD in Chemical Science from the Indian Institute of Science Education and Research (IISER), Kolkata, in 2022. Thereafter, she joined as a post-doctoral researcher at the Advanced Nanorobots and Multiscale Robotics Lab, led by Prof. Martin Pumera at VSB–Technical University of Ostrava, Czech Republic. Her current research interests encompass micro/nanorobotics, photocatalysis, low-dimensional materials, environmental remediation, nitrate-to-ammonia conversion, small molecule activation, and the generation of value-added products.

Carmen C. Mayorga-Martinez

Dr Carmen Mayorga is currently a professor at the School of Biomedical Engineering, UPC-Peruvian University of Applied Sciences. Previously, she was a senior scientist at VSB–Technical University of Ostrava and led the Kralupy unit of the Centre for Advanced Functional Nanorobots, UCT-Prague. She was a research fellow in the nanobioelectronics and biosensors group/ICN2, Barcelona-Spain, and at Nanyang Technological University, Singapore. Currently, her main research fields include the development of bio/sensors based on 2D-materials platforms functionalized with bioreceptors (enzyme, DNA, and antibodies) as well as micro/nano robotics at different scales and different propulsion modes for biomedical applications and environmental monitoring. Moreover, she is also interested in 2D-materials catalysis for energy applications.

Martin Pumera

Professor Martin Pumera is Head of the Advanced Nanorobots and Multiscale Robotics Laboratory at VSB–Technical University Ostrava, Czech Republic. He founded the Center for the Advanced Functional Nanorobots at UCT Prague, where he served as a director (2017–2023). He was a tenured group leader at the National Institute for Materials Science, Japan, in 2006. In 2010, Martin joined Nanyang Technological University, Singapore, where he worked as a tenured associate professor for almost a decade. Prof. Pumera has diverse research interests in nanomaterials and microsystems, in the specific areas of micro and nanomachines, electrochemical energy storage and conversion, machine intelligence, and 3D printing.

---

1. Introduction

Ammonia is a valuable chemical because of its diverse applications, such as in fertilizer industries, as a viable precursor to produce fine chemicals, and as a source of clean hydrogen energy storage, making it a potential alternative to fossil fuels.^1–3 The current large-scale synthetic industrial technique for ammonia production is the Haber–Bosch process, however, this process demands high energy input. This process consumes about 1–2% of global energy resources per year and contributes to carbon emissions corresponding to ∼1% of global greenhouse gas emissions per year. This has prompted scientists to ponder over alternative, sustainable methods and, thus take a step toward attaining a carbon-neutral economy.^4,5 The inspiration for devising such reactions is acquired from the nitrogen cycle occurring in nature, where biological nitrogenase enzymes catalytically reduce dinitrogen to ammonia.⁶ Consequently, different biohybrid catalysts were engineered to photoreduce nitrogen and produce ammonia. Various materials other than biohybrids evolved with time and participated in ammonia synthesis.^7,8 Photo-, electro-, and photoelectrocatalytic processes have been mostly employed for synthesizing ammonia by reducing dinitrogen, nitrates, nitrites, and nitric oxides. The reaction conditions, mechanisms, benefits, and drawbacks of these catalytic processes are discussed in the review.

Over recent years, low-dimensional materials (LDMs) have emerged as effective catalysts in photo- and electrocatalysis. LDMs are typically defined as materials smaller than 100 nm in a minimum of one dimension.⁹ Standard examples of LDMs include 0D spherical nanoparticles, 1D nanorods, nanowires, and 2D nanosheets. In lower dimensional (0D, 1D, and 2D) materials, their quantum-confined structures modify the electronic properties such that the LDMs become more efficient catalysts than their bulk counterparts, offering increased activity and stability.^10,11 LDMs have long been studied for their structural and optoelectronic properties, but their utilization in catalysis is quite recent.¹² This review explores the current state-of-the-art utilization of LDMs for ammonia synthesis by photo- and/or electroreduction of nitrogen-containing small molecules like dinitrogen, nitrates, nitrites, nitric oxides, etc. For a better understanding of the roles of dimensionalities, the catalytic synthesis of ammonia by LDMs has been categorized according to the dimensionality of the LDM (0D, 1D, and 2D) catalysts used for the photo- or/and electrocatalytic production of ammonia (Fig. 1). These low-dimensional materials present a wide range of physical and chemical characteristics that can boost the catalytic reactions.

Fig. 1 Schematic overview of low-dimensional materials used as catalysts for ammonia synthesis.

In recent years, several reviews have been published, showcasing various techniques for ammonia synthesis. For instance, two reviews by Pang et al. and Ruan et al. focus on two-dimensional materials for nitrogen reduction by electrocatalytic pathways.^13,14 Qing and co-workers reviewed recent works on electrocatalytic N₂ reduction to produce ammonia,¹⁵ and Shi and colleagues provided an account of photocatalytic procedures to generate ammonia.¹⁶ Xiong et al. summarized the works on ammonia synthesis via electrochemical nitrate reduction.¹⁷ All these reviews on ammonia synthesis primarily focus on either 2D materials for electrocatalytic or photocatalytic synthesis of ammonia. Further elucidation of the novelty of this review compared with some of the existing reviews published in the last few years in the context of ammonia synthesis highlights the catalytic method, materials used, and N-sources used as reactants as presented in Table 1. Based on our current knowledge, a comprehensive review encompassing all low-dimensional materials, including 0D, 1D, and 2D, for ammonia synthesis is still pending. Some reviews also concentrate on LDMs for CO₂ reduction or HER.^9,18,19 Nevertheless, a review of LDMs for ammonia synthesis is lacking until now. This review presents all three classes of emergent low-dimensional catalysts for photochemical, electrochemical, and photoelectrochemical ammonia synthesis. Further, with the growing importance of ammonia synthesis in the global market and the increasing number of publications in this field, a thorough analysis and discussion on this subject is imperative and has been undertaken in this review.

Table 1 Selected examples of recently published reviews and perspectives on ammonia synthesis

Journal	Catalytic approach	Highlights	N-Source	Ref.
Nat. Energy	Electrocatalysis	Metal-based catalysts, reactor design	N2	20
Nat. Catal.	Electrocatalysis	Reaction mechanism, protocols	NOx	21
Nat. Catal.	Electrocatalysis	Challenges, protocols, and future perspectives	N2	22
Nat. Synth.	Photo- and electrocatalysis	Mechanisms and catalyst design	N2	23
Chem. Rev.	Electrocatalysis	Theoretical and experimental N2RR	N2	15
Chem. Soc. Rev.	Electrocatalysis	2D catalysts	N2	13
Chem. Soc. Rev.	Photocatalysis	Materials, structure, and reaction engineering of MOx, BiOX, g-C3N4, and organic frameworks	N2	16
Chem. Soc. Rev.	Photo-, photoelectro- and photothermocatalysis	Fundamentals and challenges of N2RR, reaction mechanisms, quantification methods, techno-economic applications	N2	24
Adv. Mater.	Photo-, electro-, and photoelectrocatalysis	Reaction mechanism, catalyst engineering	N2	25
Adv. Mater.	Photo- and electrocatalysis	Reaction and catalyst engineering	N2	26
Adv. Mater.	Electrocatalysis	Reaction mechanism, catalyst design	NO3^−	17
Adv. Energy Mater.	Electrocatalysis	Fe-based single-atoms	N2, NO3^−	27
Adv. Funct. Mater.	Electrocatalysis	Active hydrogen in N2RR and NOxRR	N2, NO3^−, NO	28
ACS Energy Lett.	Electrocatalysis	Li and alkaline earth metal catalysts	N2	29
ACS Nano	Electrocatalysis	Group VIII-based catalysts	NO3^−	30
ACS Nano	Electrocatalysis	Graphene derivatives	N2	31
ACS Catal.	Photocatalysis	Fe-based catalysts	N2	32
ACS Catal.	Electrocatalysis, catalysis under temperature and pressure	Ru-based catalysts	N2	33
Small	Electrocatalysis	Single-atom catalysts	N2	34
Small Methods	Electrocatalysis	Reaction mechanism, catalyst engineering	N2, NO3^−	35
Small Methods	Photo- and photoelectrocatalysis	Reaction mechanism, catalyst engineering	N2	36
Chem. Eng. J.	Electrocatalysis	Reaction mechanism, catalyst engineering	NO3^−	37
Chem. Eng. J.	Photo-, electro-, and thermocatalysis	Ru-based single-atoms	N2	38
J. Mater. Chem. A	Electrocatalysis	2D catalysts	N2	14
J. Mater. Chem. A	Photo- and photoelectrocatalysis	Three-phase interface heterojunction catalysts	N2	39
This Review	Photo-, electro-, and photoelectrocatalysis	0D, 1D, 2D, and heterostructured catalysts, reaction pathways, applications	N 2 , NO 3 ^− , NO 2 ^− , and NO

---

The review is divided into eight sections. Section 2 focuses on the different approaches undertaken for synthesizing ammonia over the past years. The most common synthetic strategy of ammonia is the industrial Haber–Bosch process; however, this process is both energy-intensive as well as harmful to environment. Hence, other green strategies have been devised over time, which include enzyme catalysis and photo- and/or electrocatalysis. Here, we will also discuss the reaction pathways for ammonia synthesis.

In Section 3, we explore how the different dimensionalities of the LDMs play significant roles in catalysis. Zero-, one-, and two-dimensional (0D, 1D, and 2D) materials have different optoelectronic properties, impacting the catalytic process for ammonia synthesis. Through this review, we will present the latest advancements in developing LDMs, their characteristics, and the advantages they offer for catalytic ammonia production. The LDMs will be categorically summarized in the review, along with descriptions of the catalytic processes. These LDMs are often integrated to produce heterostructures that form promising catalysts for ammonia formation. Heterostructures render the interfaces with unique properties, leading to improved charge transfer and slowing down the recombination of charge carriers, thereby promoting catalytic ammonia synthesis.⁴⁰ In the next three sections of the review, we describe up-to-date progress on LDMs in ammonia synthesis and categorize them according to the energy input, i.e.; photocatalysts, electrocatalysts, and photoelectrocatalysts.

Section 4 delves into a comprehensive array of 0D, 1D, and 2D low-dimensional photocatalytic materials used for ammonia synthesis. Structure engineering of the photocatalysts for enhancing catalytic performance via the introduction of defects, construction of heterojunctions,²¹ and increasing surface area by hybridization with other materials will be explored further.^41–43 The role of co-catalysts in boosting the photocatalytic efficiency of ammonia synthesis will also be discussed here.⁴⁴

Section 5 focuses on electrocatalysts that efficiently convert N-containing species to ammonia. Like photocatalysts, the low-dimensional electrocatalysts (0D, 1D, and 2D) have also been modified for better performance via defect engineering, formation of heterostructures and regulation of crystal planes, and the roles of such modifications will be illustrated with distinct examples.^45–47 LDMs have also been used to produce 3D-printed electrodes that can electrocatalytically reduce nitrates and nitrites to form ammonia.⁴⁸ 3D printing techniques can precisely tune the shape, structure, and geometry of LDMs; hence, we will explore these emerging 3D-printed LDM catalysts for enhancing catalytic performance.

Section 6 emphasizes low-dimensional photoelectrocatalysts that combine the advantages of photocatalysts and electrocatalysts. In photoelectrocatalysis the combination of light and electric bias helps improve the separation and migration of electron–hole pairs, reducing recombination losses.^49,50 Narrow band gap materials or heterostructures formed by LDMs are mostly used for designing photocathodes, and these LDMs will be explored in detail in this section.^51,52

In Section 7 we discuss the advantages and applications of ammonia synthesis reactions. For instance, nitrates are well-known contaminants in groundwater, so reducing nitrates apart from yielding valuable-added ammonia also aids in wastewater treatment.⁵³ The reduction process can also be coupled with valuable oxidation reactions, such as water oxidation, methanol oxidation, glycerol oxidation, and plastic degradation via PET oxidation.^54–56 These reactions will also be explored in the review. The reduction of dinitrogen or nitrates, coupled with the co-reduction of CO₂ can produce urea via C–N coupling. Urea is the most common fertilizer used in agriculture.⁵⁷ This process is extremely beneficial as it can help realize carbon/nitrogen neutrality in the environment. The design of catalysts for two co-reduction processes is difficult, but not impossible. Dual-site, bimetallic, single-atom catalysts have been employed to achieve the formation of urea.^58–60 The catalysts explored in all these aspects will be discussed in this section.

The final section presents future perspectives and the outlook for ammonia synthesis. All sections in this review are supported and illustrated with representative literature from the last decade (2015–2025). This review aims to present a vast library of emergent low-dimensional materials for ammonia generation. It also provides a comprehensive insight into the choice of materials for catalysts, the strategic designs employed to reach maximum efficiency, and implementations toward next-generation applications.

2. Approaches to make ammonia: methods and reactions of ammonia synthesis

The synthesis of ammonia is important as ammonia has versatile uses, such as zero-carbon fertilizers, energy storage, and the synthesis of value-added products.^2,3,61 Global production of ammonia amounts to ∼170 million metric tons per year.⁶² The industrial synthesis of ammonia mainly involves gaseous nitrogen, abundantly available in the atmosphere. But the process is not very simple. The dissociation of N≡N requires a very high energy of ∼941.8 kJ mol⁻¹, and the standard enthalpy of formation of ammonia at 25 °C is −46 kJ mol⁻¹.^63,64 Hence, the synthesis of ammonia becomes challenging as the reaction involves high temperature, high pressure, or reactive reagents, and also the process is not energy efficient. This section focuses on the most widely used strategies explored till now to synthesize ammonia (Fig. 2).

Fig. 2 Schematic representation of different routes to synthesize ammonia. Industrial Haber–Bosch process, nitrogenase enzyme-catalyzed reaction, reproduced with permission from ref. 65, copyright 2017, Wiley-VCH, photocatalysis, electrocatalysis, and photoelectrocatalysis from the conversion of N₂, NO_x⁻, and NO_x.

2.1. Haber–Bosch process

To meet the need for ammonia-based fertilizers, in the early twentieth century, two Nobel Prize-winning chemists, Fritz Haber and Carl Bosch, invented a high-pressure reaction to directly combine N₂ and H₂ to form ammonia (eqn (1)).⁶⁶ To date, most industries rely on this process for the commercial production of ammonia. The modified Haber–Bosch reaction is typically conducted over Fe- or Ru-based catalysts in the temperature zone of 623–773 K and the pressure region of 15–35 MPa.⁶⁶ This process is energy-intensive, accounting for about 1–2% of global energy consumption. The H₂ required for this process is mostly derived from the combustion of fossil fuels. It releases CO₂, contributing to ∼1% of the total greenhouse gas emissions of the world.⁴ The reaction for ammonia production following the Haber–Bosch process is as follows:

2.2. Enzyme-catalyzed reactions

In an attempt to reduce the carbon footprint generated by Haber–Bosch reactions, researchers have taken inspiration from biological nitrogen fixation reactions. In this reaction, nitrogenase, a bacterial enzyme, directly converts atmospheric N₂ to NH₃ (eqn (2)).⁶⁷ Instead of H₂, this reaction uses protons and electrons to form the N–H bonds in ammonia. Nitrogenase comprises two proteins: Fe-based protein (FeP) and MoFe-based protein (MoFeP).^65,67 FeP hydrolyzes ATP to ADP, and the electrons generated in this process are transferred to MoFeP. MoFeP reduces N₂ to NH₃. Stoichiometrically, six pairs of protons and electrons should be used for the synthesis of NH₃ from N₂. But the reaction takes up 8 pairs of H⁺/e⁻. This apparent wastage of 2e⁻ and 4 ATP molecules contributes to the energy required for thermodynamically driving the reaction. However, excess energy usage leads to a large chemical overpotential of this reaction.

The high demand for ammonia has led researchers to find a sustainable method for its synthesis that will be energy-efficient, cost-effective, and environmentally benign. In recent years, numerous methods, such as plasma, thermochemical, photochemical, electrochemical, and photoelectrochemical techniques have been explored to decarbonize ammonia synthesis. This review will focus on ammonia generation from different nitrogen-containing species (like N₂, NO, NO₃⁻, and NO₂⁻) via photocatalysis, electrocatalysis, and photoelectrocatalysis.

2.3. Photocatalysis

Solar energy is a clean and inexhaustible energy source, which has triggered the wide utilization of photocatalysis in the sector of chemical fuels. Literature evidence suggests that over the initial years, it was thought that the photoreduction of N-containing species occurs only in nature. Soil and sand samples were explored to study the effect of minerals on the photoreduction of nitrogen in nature.^68,69 Fascinated by the working mechanism of the Fe and FeMo proteins in nitrogenase enzyme reactions, chemists have designed photocatalysts that absorb light to generate electrons and protons. These electrons activate the N-containing species (N₂, NO_x, and NO_x⁻), which react with protons from water to form ammonia. Brown et al. developed such an analog CdS: FeMo protein biohybrid, which can reduce nitrogen to ammonia upon light activation.⁷ CdS nanocrystals act as photosensitizers, providing electrons to the FeMo protein for nitrogen reduction. In this work, light is used as the energy source, whereas in the biological nitrogenase system energy is acquired by hydrolysis of ATP. Researchers have extended this know-how to design other photocatalytic systems like metal oxides,^70,71 metal sulfides,⁷² bismuth oxyhalides,^73,74 layered double hydroxides,^75,76 and carbon-based materials^77,78 for the photocatalytic synthesis of ammonia. According to the literature, the photocatalytic N₂, NO_x, and NO_x⁻ reductions by semiconductors occur on photoactivation, when the excited electrons move to the conduction band, leaving holes in the valence band.⁷⁹ The electrons reduce N-species to ammonia while the holes participate in the water oxidation reaction. The recombination of electrons and holes is one of the significant problems in photocatalytic reactions.⁸⁰ Thus, hole scavengers are sometimes used as sacrificial agents to prevent recombination. The band gap of the materials and the position of conduction and valence bands play a significant role in photocatalysis. The reduction potentials of N₂RR, NO_xRR, and water oxidation must lie in between the conduction and valence band positions. NO_x⁻ (nitrates and nitrites) are common sources of pollutants in wastewater and the photoreduction of NO_x⁻ can help in wastewater treatment, which will be discussed in detail later in Section 7.1. The direct reduction of nitrate and nitrite wastes into ammonia is thus both economically and environmentally important. In some cases, however, the competitive hydrogen evolution reaction (HER) and reduction of NO₃⁻ to N₂ limits the selectivity of the ammonia produced. A detailed discussion on different classes of photocatalytic materials like metal oxides, metal nanoparticles, single-atoms, quantum dots, and their composites, which can photoreduce N-containing species to ammonia with high selectivity, will be presented in Section 4 of this review.

2.4. Electrocatalysis

Electrochemical reduction of N₂ or NO_x⁻ to ammonia occurs under ambient conditions, i.e., at room temperature and atmospheric pressure at a specific applied voltage. Sometimes, electrochemical reactions are considered to be more energy-efficient than photochemical reactions. This is because, in photocatalysis, some of the absorbed photons remain unutilized due to the variable wavelengths of incident light sources and the recombination of electrons and protons.⁸¹ The design of electrocatalysts is quite challenging because the activity, selectivity, efficiency, and stability depend on the choice of materials. The material structure (surface area, porosity, crystal facets) and morphology affect the turnover and yield, while the electronic properties (heteroatom doping, defect engineering) influence the active site of the catalyst.⁸² The surface properties of the catalyst can affect the adsorption–desorption of N₂ or NO_x⁻ species, which in turn affects the bond dissociation and the consecutive hydrogenation reactions for ammonia formation. The availability of electrons and protons can also be controlled by the design of the catalyst, which might be able to suppress the competing HER reactions and also help in attaining selectivity of the ammonia, reducing the chance of obtaining lesser reduced products like NO₂⁻ or NO from the electroreduction of NO₃⁻.

In the electrochemical experiments, initially N₂ or NO_x⁻ is adsorbed on the surface of the electrode in a simple electrochemical cell. Electrons supplied by the external circuit reduce the N-containing molecule, which undergoes subsequent hydrogenation by the protons to produce ammonia. In most cases, water present in the reaction medium acts as the source of hydrogen for ammonia production. A judicious choice of electrolyte is crucial for electrocatalytic reactions because the electrode reactions of the electrochemical cell are dependent on the electrolyte. In past years, aqueous and non-aqueous electrolytes have been used where protons (H⁺), hydroxides (OH⁻), oxides (O²⁻), and nitrides (N³⁻) act as mobile charge carriers. The main bottlenecks in electrocatalytic ammonia synthesis are the yield and the Faradaic efficiency (FE). Competing reactions on the surface of electrocatalysts, like water reduction to hydrogen evolution reaction (HER), dominate over N₂ or NO_x⁻ reduction reactions (N₂RR and NO_xRR) at higher overpotentials, thus decreasing the selectivity of the product.⁸³ To eliminate the drawbacks, over the past few years, researchers have been working to upgrade electrocatalysis cells, electrocatalysts, electrolytes, and working potentials, which will be discussed in detail in Section 5 of this review.

2.5. Photoelectrocatalysis (PEC)

The general redox reactions for photocatalysis and electrocatalysis are similar. In PEC, these reactions become more favourable because of the combined effects, i.e., electric and light energy.⁸⁴ In an electrochemical cell, N₂RR or NO_xRR occur at the cathode while water splitting occurs at the anode. The experimental setup of PEC is very similar to the electrocatalytic set-up; only in PEC, the cathode is replaced with a photocathode. The photocathode utilizes energy from a light source to provide the photovoltage. In this way, the electrical energy required for N₂RR and NO_xRR either decreases to a certain extent or is completely diminished depending on the Fermi level of the photocathode.⁴⁹ Generally, in PEC a three-electrode set-up is employed where the catalyst is loaded on the working electrode (photocathode).³⁶ The reduction reactions (N₂RR or NO_xRR) occur at the photocathode and water splitting at the anode (counter electrode). A proton exchange membrane is placed between the cathode and anode, which ensures only protons are supplied to the cathodic side for ammonia formation, and no other oxidative species can migrate to the cathodic part so that the ammonia produced does not undergo any oxidation. In photocatalysis and electrocatalysis, the competing HER reduces the Faradaic efficiency (FE) of NH₃. In PEC, the competing HER can be suppressed using photoelectrodes. For instance, an Au-decorated ordered silicon nanowire array photocathode can perform NO₂RR to NH₃ with FE as high as 95.6% at 0.2 V vs. RHE.⁸⁵ This is because the Au and Si surfaces of the electrode are not active for water reduction. The absence of the competing HER makes the FE for NH₃ production comparatively higher. The PEC reaction mechanisms of N₂RR and NO_xRR are similar to photocatalysis and electrocatalysis. The design of materials for PEC follows the same principles as electrocatalysts and photocatalysts. Improvement in the photoelectrocatalytic NH₃ yield and selectivity can be achieved by modification of the catalysts such as, doping with heteroatoms, making heterojunctions, defect engineering, and modification of morphology.^86–88 Thus, we can infer that the structures and properties of the materials used for designing the catalysts have profound roles in the catalytic reactions and products formed. Section 6 of this review will extensively discuss the emergent low-dimensional materials for photoelectrocatalytic ammonia synthesis.

Among the above-mentioned ammonia synthesizing techniques, the most widely used methods in recent literature include photocatalysis, electrocatalysis, and photoelectrocatalysis. These processes are more sustainable and easily achievable and only require catalyst materials and energy sources to produce ammonia. Though all these processes follow the general principles of catalysis, however, the reaction conditions, mechanisms, and catalyst design differ. To get a brief idea about the working principles, catalysts, reaction mechanisms, benefits and drawbacks, and environmental impact, a comparative study of these processes is presented in Table 2 and Fig. 3.

Table 2 Comparison among photocatalysis, electrocatalysis, and photoelectrocatalysis methods used for ammonia synthesis by LDM-based catalysts

Factors	Photocatalysis	Electrocatalysis	Photoelectrocatalysis
Energy source	Solar energy (light)	Electrical energy	Light and electrical energy
Catalyst materials	Semiconductors, metallic nanoparticles, plasmonic materials, quantum dots, metal oxides, single-atoms, carbon-based materials	Metals, alloys, metallic oxides, single-atoms, carbon-based materials, MXenes, transition metal dichalcogenides	Heterostructures composed of photocatalysts and electrocatalysts
Design of catalytic systems	i. Band gap engineering.	i. Good conductivity and corrosion resistance.	i. Heterostructures of photocatalysts and electrocatalysts for simultaneous light and electrical energy utilization.
	ii. Efficient charge separation.	ii. Electrode surface modification to adsorb reactants and intermediates.	ii. Interface engineering for charge transfer. Catalyst engineering with photo- and electrocatalysts.
	iii. Surface area modification for better light absorption and adsorption of N2/NOx.	iii. Electrode porosity for better mass transfer.	iii. Reactor design for efficient light and electrical energy distribution.
	iv. Catalyst engineering via the introduction of defects, doping, cocatalysts, heterojunction and heterostructure formation, particle size, and morphology modification.	iv. Catalyst engineering via the introduction of defects, doping, cocatalysts, facets, heterojunction, and heterostructure formation.
		v. Good electrolytes to enhance ionic conductivity.
Reaction mechanism	i. Light absorption leads to the separation of electron–hole pairs and photoexcitation of electrons from the valence band to the conduction band.	i. Application of electric potential.	i. Light absorption and electric potential generate required charge carriers for photoelectrocatalysis.
	ii. Electrons in the conduction band reduce N2/NO/NO2^−/NO3^− to form NH3.	ii. Electrons reduce N2/NO/NO2^−/NO3^− to NH3 at the cathode.	ii. Synergistic effects enhance electron transfer.
	iii. Holes oxidize water or other organic reactants.	iii. Efficient electron transfer at electrode surface facilitated by catalysts.	iii. Optimization of photo- and electrical parameters for efficient reduction of N2/NO/NO2^−/NO3^− at the photocathode.
		iv. The oxidation half-reactions occur at the anode.
Cost	Low	High	High
Efficiency	Low	High	Low
Advantages	i. Light energy can be harnessed from abundant solar energy.	i. Ammonia production with high efficiency and greater selectivity.	i. Combined advantages of photo- and electrocatalysts.
	ii. Low operational cost.	ii. Reaction parameters can be controlled and adjusted easily for electrocatalytic processes.	ii. Potential for synergistic reactions.
	iii. Sustainable synthesis of ammonia due to reduced greenhouse gas emissions.	iii. Lower dependency on weather conditions as electric energy can be produced on demand.	iii. Flexible reaction conditions.
Disadvantages	i. For solar-powered photocatalytic reactions, ammonia production is dependent on regions with high solar irradiance.	i. Requires electrical energy, which is sometimes harnessed from non-renewable sources.	i. Complexity of system design.
	ii. Efficient light absorption and conversion are required.	ii. Electrode degradation with time.	ii. Requires optimization of both light and electrical inputs.
	iii. Long-term stability and recyclability of photocatalysts have to be considered.	iii. Higher operational costs due to electricity, maintenance, and replacement of electrochemical setup.	ii. Higher costs of photoelectrochemical setup and maintenance.
Environmental impact	i. Low environmental footprint due to reduced use of fossil fuels for photocatalytic ammonia production.	i. Dependent on electricity, sometimes produced by non-renewable energy sources.	i. Can provide an optimized balance between light and electrical energy sources and reduce environmental impact.
	ii. Minimal waste generation and greenhouse gas emissions.	ii. Emissions associated with electricity production and potential waste generation from electrode degradation.
Future applications	i. Solar-powered green ammonia generation.	i. Potential for industrial-scale ammonia productions.	i. Hybrid systems for enhanced sustainable ammonia production under variable reaction conditions.
	ii. Potential for integration with renewable energy sources.	ii. On-grid applications.	ii. Off-grid and on-grid applications.
	iii. Off-grid applications.

---

Fig. 3 Comparison of (A) photocatalysis, (B) electrocatalysis, and (C) photoelectrocatalysis methods for ammonia synthesis.

2.6. Reaction mechanisms of N₂ reduction to NH₃

The reaction of catalytic reduction of dinitrogen (N₂) to ammonia (NH₃) using natural ingredients like air and water is proposed to be as follows (eqn (3)–(5)):^79,89

The half-cell N₂ reduction in the acidic and basic media can be depicted as follows:^36,90

Acidic condition: N₂ + 6H⁺ + 6e⁻ → 2NH₃ (4)

Basic condition: N₂ + 6H₂O + 6e⁻ → 2NH₃ + 6OH⁻ (5)

The reactions (eqn (6)–(9)) for the reduction of N₂ to NH₃ can change with the reaction medium and charge carriers as follows:¹⁵

a. In the presence of H⁺

N₂ + 6H⁺ + 6e⁻ → 2NH₃

3H₂ → 6H⁺ + 6e⁻ (6)

b. In the presence of OH⁻

N₂ + 6H₂O + 6e⁻ → 2NH₃ + 6OH⁻

3H₂ + 6OH⁻ → 6H₂O + 6e⁻ (7)

c. In the presence of O²⁻

N₂ + 3H₂O + 6e⁻ → 2NH₃ + 3O²⁻

3O²⁻ → 3/2O₂ + 6e⁻ (8)

d. In the presence of N³⁻

N₂ + 6e⁻ → 2N³⁻

3H₂ + 2N³⁻ → 2NH₃ + 6e⁻ (9)

In all reaction conditions, dinitrogen reduction to ammonia is a six-electron reduction reaction, which is a kinetically “uphill” process. The design of photo- and/or electrocatalysts plays a vital role as they provide an alternative lower energy pathway to the production of ammonia. Nitrogen is a highly stable molecule owing to its triple bond and breaking the first bond is considered the rate-determining step. The mechanism of N₂ reduction to ammonia is currently being investigated by different research groups and studies are still underway. It has been proposed that there are two possible pathways for N₂ reduction to NH₃: (i) associative and (ii) dissociative (Fig. 4A).^36,91 In the dissociative pathway, the N–N bond breaks before hydrogenation. However, this process is highly energy-demanding and the dissociative pathway is not considered favourable. In the associative mechanism, the first step involves the protonation of N₂. Protonation depends on the configuration of N₂ and can be either to the distal N atom, forming N₂H* species, or to the two N atoms which can alternatively form N₂H₂* species. In the associative distal pathway, the reaction proceeds via three consecutive hydrogenation steps, forming one NH₃ molecule. Then, hydrogenation occurs at the second N atom, forming NH₃ after three more steps. In the associative alternating pathway, hydrogenation occurs alternately at both N atoms, and two molecules of ammonia are produced almost simultaneously. Photocatalytic and electrocatalytic reduction of N₂ to ammonia can follow any of these pathways, though simple calculations are not enough to predict the exact reduction pathway.

Fig. 4 General reaction pathways of ammonia synthesis. (A) N₂ reduction to ammonia via dissociative, associative alternating, and associative distal pathways, (B) NO_x⁻ and NO_x reductions to ammonia via associative, dissociative, alternating, and distal pathways through N- and O-ends.

2.7. Reaction mechanisms of NO_x⁻ and NO_x reduction to NH₃

As discussed above, the bond dissociation energy of N₂ is very high (∼941.8 kJ mol⁻¹). In comparison, the bond dissociation energies for NO_x and NO_x⁻ species are lower. They can be disintegrated into deoxygenated species with a lower bond dissociation energy (for NO₃⁻ ∼204 kJ mol⁻¹); hence, nitrogen oxides can offer a sustainable route to the synthesis of ammonia at a lower energy cost. Also, the valence states of N in nitrogen oxides are higher (+5 in NO₃⁻, +3 in NO₂⁻) than in dinitrogen where the valence state of N is 0; hence, deep reduction reactions are more plausible in the case of nitrogen oxides. The catalytic reactions are similar to that of dinitrogen but require a different number of electrons and protons for each species (eqn (10)–(12)).⁹²

Nitrates and nitrites are present in the form of electrolytes for the electrocatalytic reduction, which involves 8e⁻ and 6e⁻ respectively for the conversion to ammonia. The reduction proceeds via the dissociation of NO_x⁻ species into deoxygenated species followed by hydrogenation, which is quite similar to the adsorption–desorption mechanism of nitrogen reduction. The reduction of nitrogen oxide species to ammonia is illustrated in Fig. 4B and it exhibits multi-step processes of electron and proton transfer.^93,94 Initially, NO₃⁻ is adsorbed on the catalyst surface as *NO₃⁻, which is followed by the loss of an electron to form *NO₃. This *NO₃ intermediate is then converted into *NO₂via the transfer of two electrons and two protons. In most cases, the reduction of nitrate to nitrite is the rate-determining step for NO₃⁻ reduction. The *NO₂ intermediate is further reduced to *NO. The subsequent steps of reduction to ammonia passing through *NO intermediate is the same for NO_x⁻ and NO. The next possible steps can follow either dissociative or associative pathways. In the dissociative pathway, the N–O bond undergoes dissociation, forming adsorbed nitrogen (*N) and oxygen (*O) species on the catalyst surface. The *N is then sequentially hydrogenated to yield ammonia. The associative pathway can proceed through any one of the three possible routes depending on the adsorption type of *NO intermediate. The adsorption of *NO on the catalyst surface can be through (i) N-end, (ii) O-end, or (iii) NO-side. The hydrogenation of the N-end and O-end routes can again proceed through either alternating or distal hydrogenation processes. In the case of the distal-O or N hydrogenation pathway, at first one of the O or N atoms is at first fully hydrogenated to form H₂O or NH₃, followed by the hydrogenation of the other atom. For the case of the alternating-O or N hydrogenation pathway, the O and N atoms are alternately hydrogenated stepwise to form H₂O and NH₃, respectively.⁹⁴

3. Low-dimensional materials (LDMs): dimensionality of nanostructures and heterostructures

Nanostructured materials (NSMs) can act as effective catalysts because of their morphologies, electronic properties, and surface characteristics.⁹⁵ Dimensionality is an important parameter because different dimensional materials of similar chemical composition exhibit separate physical and chemical properties. For instance, carbon has different allotropes: fullerene, carbon nanotube, graphene, and graphite. Fullerene is 0D, carbon nanotube is 1D, graphene is 2D, and graphite is 3D, and all four have different properties. NSMs have building units of size ranging from nanometer to sub-micron scale. The classification of NSMs has been proposed by Gleiter (1995) and Skorokhod (2000); later, Pokropivny and Skorokhod proposed a modified classification.^96,97 According to the classification by Pokropivny and Skorokhod, NSMs can be classified as 0D, 1D, or 2D. The structure and morphology of some common low-dimensional materials are illustrated in Fig. 5. In this review, we will explore the implementation of the NSMs and multicomponent heterostructures, which are built from the 0D, 1D, and 2D materials for photocatalytic, electrocatalytic, and photoelectrocatalytic N₂RR and NO_xRR to produce ammonia.

Fig. 5 Dimensional classification of low-dimensional materials (LDMs). 0D common structures. (A) Metallic nanoparticles, representative SEM image of bimetallic Au/Ni nanoparticle. Reproduced with permission from ref. 98. Copyright 2019, American Chemical Society. (B) Semiconductor nanoparticles, representative TEM image of CdS semiconductor nanoparticles. Reproduced with permission from ref. 99. Copyright 2020, American Chemical Society. (C) Core–shell structures, representative TEM image of Fe₃O₄@silica RF core–shell particles. Reproduced with permission from ref. 100. Copyright 2009, Royal Society of Chemistry. (D) Nanocrystals, representative TEM image of maghemite nanocrystals. Reproduced with permission from ref. 101. Copyright 2001, American Chemical Society. (E) Single atoms, representative HAADF-STEM image of Fe single atoms. Reproduced with permission from ref. 102. Copyright 2023, Wiley-VCH. (F) Quantum dots, representative TEM image of carbon quantum dots. Reproduced with permission from ref. 103. Copyright 2023, Royal Society of Chemistry. 1D common structures. (G) Nanowires, representative SEM image of Ag nanowires. Reproduced with permission from ref. 104. Copyright 2020, American Chemical Society. (H) Nanotubes, representative TEM image of carbon nanotube. Reproduced with permission from ref. 105. Copyright 2019, Elsevier. (I) Nanofibers, representative SEM image of Fe₂TiO₅ nanofibers. Reproduced with permission from ref. 106. Copyright 2022, Wiley-VCH. (J) Nanorods, representative TEM image of g-C₃N₄ nanorods. Reproduced with permission from ref. 107. Copyright, 2022, Elsevier. (K) Nanobars, representative SEM image of LiNi_0.4Co_0.2Mn_0.4O₂ nanobars. Reproduced with permission from ref. 108. Copyright 2012, Royal Society of Chemistry. (L) Nanoribbons, representative TEM image of graphene nanoribbons. Reproduced with permission from ref. 109. Copyright 2015, AIP Publishing. 2D common structures. (M) Nanosheets, representative TEM image of ZnCr-LDH nanosheets. Reproduced with permission from ref. 76. Copyright 2020, Wiley-VCH. (N) Nanoplates, representative TEM image of hexagonal Co(OH)₂ nanoplates. Reproduced with permission from ref. 110. Copyright 2013, Royal Society of Chemistry. (O) Nanofilms, representative TEM image of Cu-TCPP MOF ultra-thin nanofilm. Reproduced with permission from ref. 111. Copyright 2013, Royal Society of Chemistry. (P) Nanolayers, representative cross-TEM image of a-ZrTiO_x nanolayers. Reproduced with permission from ref. 112. Copyright 2013, Royal Society of Chemistry. (Q) Nanodiscs, representative TEM image of Bi₂Se₃ nanodisc. Reproduced with permission from ref. 113. Copyright 2012, American Chemical Society. (R) Nanoprisms, representative SEM image of Ag nanoprisms. Reproduced with permission from ref. 114. Copyright 2024, American Chemical Society.

3.1. Zero-dimensional (0D) nanostructured materials

0D materials are mostly nano-dimensional structures with an average size of less than 10 nm along all three dimensions and are isotropic.¹¹⁵ They comprise metal nanoparticles, metal alloy nanoparticles, nanospheres, core–shell structures, yoke–shell structures, quantum dots, single-atoms, nanocrystals, and metal oxides.^98–103,116 The fabrication of 0D catalysts has been well explored and various methods, such as thermal evaporation, sputtering, chemical vapour deposition, electrodeposition, solvothermal, sol–gel, annealing, galvanic replacement, gas phase deposition, and template-based methods are quite common fabrication techniques.^117,118 The optical and physicochemical properties of these 0D materials can be tuned depending on their size, making them potential catalysts for photo- and electrocatalysis reactions.

3.2. One-dimensional (1D) nanostructured materials

1D materials are nanostructured along any two dimensions while the third dimension can be in the microscale range.¹¹⁵ 1D catalysts are advantageous over 0D catalysts as they offer a unique anisotropic structure, more density of active sites, and more surface area than 0D materials.¹¹⁹ 0D materials are more susceptible to Ostwald ripening due to the high surface energy. In 1D nanostructures, the lower-energy facets are preferentially exposed on the surface, which reduces the surface energy of the whole material, making them more stable.¹²⁰ The structure of 1D materials enhances mass transport and diffusion, and provides them with superior catalytic properties.¹²¹ 1D materials comprise nanowires, nanotubes, nanorods, nanoribbons, nanobelts, and nanofibers.^104–109 The most common synthetic techniques of 1D materials include hydrothermal, sol–gel, self-assembly, electrospinning, and template-assisted methods.^122–124 In this review, we will explore the recently developed 1D catalysts and how their structural and electronic properties aid in the catalytic synthesis of ammonia.

3.3. Two-dimensional (2D) nanostructured materials

In 2D materials, only one dimension is confined to the nanoscale regime while the other two dimensions can range from micrometer to centimeter scale.¹¹⁵ Nanosheets, nanoplates, nanodisks, and nanoprisms are the most commonly designed 2D materials for catalysis.^{76,110–114,125} These 2D nanostructures have exposed surface area, increasing their contact with reactants.^126,127 Most of the active sites are present on the surface and are easily accessible for catalysis. Defects and unsaturated sites can be introduced into the 2D structures, which enhances adsorption–desorption and facilitates mass transport.¹²⁶ In 2D materials, electrons can be trapped within the atomic layers, which brings good conductivity and electron mobility to the structures, thereby facilitating electron transport and charge carrier separation.¹²⁸

According to Sabatier's Principle, an optimized catalytic reaction occurs when the interaction between the reactants and the catalyst surface is optimal, i.e., the adsorption energy is intermediate.¹²⁹ Thus, it can be said that the catalytic properties of a material are dependent on its surface and electronic properties.^130,131 Hence, engineering the band structure of a material proves to be an important tool in catalyst design.¹³² The electronic structures of low-dimensional nanostructures differ from their bulk counterparts due to the quantum confinement of the electronic density states. As a result, the photon absorption properties of these materials are enhanced, leading to better catalytic activities in these systems.^133,134 The electronic properties are also dependent on the morphology of the materials, including kinks, edges, facets, and phases.⁹ Additionally, the surface area to volume ratio is higher in the low-dimensional catalysts, which ensures the availability of more active sites on the surface of the catalyst.⁹ The successful utilization of 0D, 1D, and 2D materials for catalysis has also prompted researchers to prepare hybrid structures using these nanomaterials and exploit them in catalysis. Moreover, the integration of two components provides synergistic effects from each counterpart toward catalysis. These individual components of heterostructures are held together mainly via van der Waals forces.⁹ Heterostructures have emerged as state-of-the-art catalysts as they offer superior-quality interfaces with unique properties compared to normal nanostructures. The newly formed interfaces between the components can lead to improved charge transfer, slowing down the recombination of charge carriers.¹²⁶ These heterostructures can be 0D–1D, 0D–2D, 1D–2D, 2D–2D, and so on. In the next sections, we will discuss the 0D, 1D, and 2D materials, their heterostructures, and the interplay between the electronic properties and their catalytic activities.

4. Low-dimensional photocatalysts for ammonia synthesis

The design of photocatalysts for the reduction of N-containing species to ammonia requires a few criteria to be met: (i) adsorption of the N-containing species on the surface of the catalyst; (ii) absorption of light by the catalyst; (iii) tuning the band gap corresponding to the wavelength of light absorbed, which can promote the excitation of electrons from the valence band to the conduction band; (iv) position of the valence and conduction bands such that the potentials of the redox reactions are aligned in the gap; and (v) recombination of charge carriers is minimum. These requirements can be fulfilled by choosing the proper catalyst from a wide range of photocatalytic materials, such as semiconductors, metal nanoparticles, metal oxides, metal sulfides, and carbon-based materials. Structural engineering of the photocatalysts via the introduction of defects, construction of heterojunctions, increasing surface area by hybridization with other materials, and the use of cocatalysts can enhance photocatalytic performance.

Photosynthesis of ammonia, in principle, requires only water, light, and N₂ or NO_x⁻ hence, it is quite attractive. However, the photocatalysts developed earlier exhibit low conversion efficiency for ammonia generation. This might be due to the availability of less active sites, limited light absorption properties, and quick recombination of photogenerated charges. Active sites adsorb and activate the N-containing substrate. The activation of N₂ is challenging due to the high dissociation energy of N≡N. Nitrogen has four bonding (2σ and 2π) and four anti-bonding (2σ* and 2π*) orbitals. Activation of N₂ requires a donation of electrons to the bonding orbitals and an acceptance of electrons from anti-bonding π* orbitals. The thermodynamic reduction potential of N₂ to NH₃ is 0.148 V vs. RHE, which is quite close to the hydrogen evolution reaction (0 V vs. RHE), so the competing HER sometimes hinders the N₂RR, and this remains a major concern in most cases.¹³⁵ The drawbacks of the low solubility of N₂ in an aqueous medium and the high dissociation energy required for the cleavage of N≡N can be overcome by replacing the N-containing precursor from N₂ with nitrates or nitrites.²¹ Moreover, high concentrations of nitrates and nitrites in wastewater can be a major health concern; thus, reducing them to ammonia can diminish health risks. However, nitrate reduction to ammonia requires 8e⁻; hence, the risk of obtaining lesser-reduced products like N₂ persists. For achieving higher NH₃ selectivity, side reactions of N₂ production and HER must be suppressed. This may be achieved by surface modification of catalysts or modification of materials based on the mechanistic pathway of ammonia synthesis.

The principle of photocatalysis relies on harvesting solar energy (Fig. 6). Hence, the band gap of the catalysts should also be tuned for direct utilization of solar light, preferably visible light. However, the band gap tuning sometimes reduces the energy of photogenerated electrons and, thereby, their reduction capability.¹³⁶ Hence, developing an optimized photocatalyst that effectively absorbs light while maintaining its reduction capability is essential. For activation of N-containing molecules, the active sites must be electron-rich and able to promptly transfer electrons to adsorbed N-containing moieties. Inducing defects and unsaturated sites with abundant localized electrons in the catalysts can address this issue as they effectively transport electrons, and consequently, activate and reduce the N-containing moieties. The synthetic and modification strategies of LDMs are mostly simple and single-step, but structure engineering of the catalysts and the formation of heterostructures sometimes require multi-step fabrication strategies. 0D photocatalysts like metal and metal oxides mostly rely on different reduction methods of metallic salts, like photoreduction, H₂ reduction, or chemical reduction, whereas single-atom-based photocatalysts use different deposition techniques for integration with support materials.^137–141 Fabrication of 1D materials based on metal oxide, metal oxyhalides, and carbon-based materials mostly uses different thermal treatment procedures like hydrothermal, solvothermal, and annealing.^142–144 2D materials based on semiconductor nanosheets, graphitic carbon nitrides, and graphdiynes rely on different thermal synthetic procedures.^145–148 The catalytic properties of these materials can be enhanced by defect engineering like the introduction of vacancies,⁴³ doping of heteroatoms,⁷¹ formation of heterojunctions,¹⁴⁹ heterostructures,¹⁵⁰ and use of cocatalyst materials.¹⁴⁶ The fabrication strategies of the most widely used 0D, 1D, 2D, and heterostructured photocatalysts are presented in Table 3. This section will explore the emergent low-dimensional photocatalysts (Fig. 6) used recently to convert N₂, NO_x, or NO_x⁻ to NH₃.

Fig. 6 Low-dimensional photocatalysts for ammonia synthesis. Classification of photocatalysts into 0D, 1D, and 2D, and schematic representations of few selected photocatalytic LDMs.

Table 3 Synthetic and modification techniques for selected ammonia generating low-dimensional photocatalysts

Classification	Materials	Photocatalysts	Synthesis/Modification	Ref.
0D LDMs and heterostructures	Metallic nanoparticles (NPs)	AuNPs–TiO2	In operando photodeposition	137
		Pd/Cu/Ag–TiO2	H2 reduction	138
		CuPd–TiO2	Chemical reduction	139
		Ru–g-C3N4	Green tea reduction	151
		Ru–ZrO2−x	Impregnation method	152
		Ru–K2Ta2O6−x	Thermal decomposition	153
	Metal oxide nanocrystals (NCs)	MONCs–TiO2 (M = Mg, Ca, Sr, Ba)	In operando photodeposition	154
		Cu2ONCs–TiO2	In operando photodeposition via pseudo-Fehling's reaction	155
	Quantum dots (QDs)	TiO2QDs–Fe3S4	Heterojunction formation	149
		Bi2Ti2O7 QDs	Defect engineering via hydrothermal method	156
		InP QDs	Organometallic reactions, ligand exchange	157
	Single-atoms (SAs)	Ru SA–HxMoO3−y	Defect engineering, H-spillover method	43
		Ru SA–CeO2	Impregnation-calcination	158
		Ru1 SA–d-UiO-66	Defect engineering, photochemical deposition	140
		Pt1 SA–BiOBr	Defect engineering, photochemical deposition	141
1D LDMs and heterostructures	Carbonaceous materials	g-C3N4	Defect engineering via molten-salt method	107
		Ti^3+–TiO2/HAp-C	Thermal annealing and reduction	142
	Metal oxides	MoO3−x NWs	Defect engineering via hydrothermal method	143
		Cu2O NPs-W18O49 NWs	Defect engineering, in situ reduction, heterojunction formation	150
		CdS NPs-WO3 NRs	Heterojunction formation	159
	Metal oxyhalide	Bi5O7Br NTs	Defect engineering, thermal treatment	144
	Pyrochlore	Bi NPs–Bi2Sn2O7 NWs	Defect engineering, heterojunction formation, hydrothermal treatment	160
2D LDMs and heterostructures	Metal oxides	B-TiO2 NS	Hydrothermal treatment, B-doping	145
		C-TiOx NS	Bottom-up substitutional C-doping	71
		Ru NPs–W18O49	Defect engineering via hydrothermal method, chemical reduction	161
		Sb–MoO3−x NS	Defect engineering, Sb-doping, heat treatment	162
	Metal oxyhalides	M–BiOBr NS (M = Fe, Mo, Ni)	Defect engineering, metal doping, hydrothermal treatment	73
		Bi NPs–BiOBr nanoplates	Schottky-junction formation, solvothermal reaction	146
		Co–BiOCl nanoplatelets	Co-doping, hydrothermal treatment	44
	Graphitic carbon nitrides	B-C3N4 NS	B-doping, thermal treatment	77
		Ni3B NPs-VN-C3N4 NS	Defect engineering, Schottky junction, chemical reduction, electrostatic self-assembly	163
		B-C3N5 NS	Defect engineering, B-doping, thermal polymerization	147
	Graphdiynes (GDYs)	GDY-Fe3O4	Modified Glaser–Hay coupling, microwave-hydrothermal approach	148
		GDY-CoOx QD	In situ growth-deposition	164
Organic framework-based heterostructures	Porous-organic frameworks (POFs)	MPc-POF (M = Fe, Co, Zn)	Schiff base reaction, solvothermal reaction	165
	Covalent-organic frameworks (COFs)	Porphyrin COF-Au SA	Thermal reactions	166
		DPPCOOH-COF	Schiff base condensation reaction	167
		Bi/COF-TaTp	Solvothermal reactions	168
	Metal-organic frameworks (MOFs)	MIL-100 (Fe) MOF	Defect engineering of ligands, nonthermal plasma-assisted synthesis	169
		Ca^2+-d-UiO-66 MOF	Ca^2+-doping, ligand incorporation, solvothermal reaction	170
		MXene-MIL-125(Ti) MOF	In situ etching, delamination heterojunction formation, ligand pre-coupling	171
		NH2-MIL-125 MOF-Co(OH)2-ZIF-8	Heterojunction formation, nanoarchitectonics, solvothermal reactions	172

---

4.1. 0D photocatalysts and heterostructures

0D photocatalysts for ammonia synthesis mainly consist of three classes of materials: semiconductor nanoparticles, quantum dots, single-atoms, and heterostructures formed with these 0D particles. Due to their plasmonic properties, 0D structures prove to be effective catalysts because the photons can be harvested at lower energy by nanoparticles, and their adsorption properties can be easily tuned by altering their shape and architecture.⁹ The separation of charge carriers is also possible because of the strong electric fields created at local “hotspots” induced by surface plasmons.¹⁷³ These properties allow the 0D materials to be incorporated with 1D or 2D materials to create heterostructures that can function as efficient photocatalysts for ammonia synthesis.

Metallic nanoparticles have plasmonic properties and their size and shape can be easily tuned, making them important materials for catalysis. Quantum dots also prove to be promising photocatalytic materials owing to their unique characteristics like high absorption coefficient, efficient charge transfer properties, large surface-to-volume ratio, and stability. Single-atoms are another class of materials that have recently emerged as prospective photocatalysts due to their highly efficient atom economy, low coordination environment, unique electronic structural properties, and atomic-level understanding of reaction mechanisms. Metallic nanoparticles, quantum dots, and single-atoms, upon heterogenization with different 1D and 2D support materials, can produce excellent photocatalysts for ammonia synthesis. For instance, the well-explored TiO₂ nanosheets can be used as substrates for incorporating metallic nanoparticles, metal oxide nanoparticles, and single-atoms, which can simultaneously modify the interface as well as act as active sites for catalysis. The heterostructures formed are mostly of dimension 0D–2D. Heterostructures of semiconductors like TiO₂ and CdS decorated with metallic nanoparticles act as efficient photocatalysts for ammonia generation from N-containing species like N₂ and NO_x.^174–176 Ideally for NO reduction reactions, a high concentration (>10 000 ppm) of NO is required for photoproduction of ammonia. However, limited NO conversion and ultra-low solubility of NO in water (1.94 ± 0.03 mmol L⁻¹ at 25 °C) are the major bottlenecks for NO conversion to NH₃.¹⁷⁷ To address these challenges and for direct photocatalytic conversion of NO, Fe(II)EDTA is employed as the NO chemical absorbent, forming Fe(II)EDTA–NO chelates, and formaldehyde (HCHO) acts as the antioxidant to prevent the formation of Fe(III) from Fe(II) oxidation.¹³⁷ TiO₂ decorated with Au nanoparticles are adorned with active sites to facilitate photogenerated charge separation, thereby promoting ammonia generation. The simultaneous chemical adsorption and photocatalytic reduction system enable continuous NO adsorption, NO reduction, and Fe(II)EDTA regeneration on-site. Notably, the efficiency of NO conversion and the selectivity of ammonia produced remain unaffected even in the presence of H₂O, SO₂, and metal ions (K⁺, Ca²⁺, Cd²⁺, and Pb²⁺). In flue gas, the simultaneous presence of NO and SO₂ presents certain challenges of removal and recovery. SO₂ is soluble and easily oxidized, hence, SO₂ acts as a potential reductant in the NO photoreduction reaction. The formation of the SO₂–NO redox pair promotes easy conversion of both NO and SO₂ in continuous flow.¹⁷⁸ Thus, high selectivity is achieved simultaneously for both NO-to-NH₃ upcycling (97%) and SO₂-to-SO₄²⁻ purification (92%). Metallic nanoparticles or bimetallic nanoalloys-loaded TiO₂, such as Cu/TiO₂ and Cu–Pd/TiO₂, in the presence of hole scavengers like oxalic acid and methanol, can perform photocatalytic 8e⁻ conversion of NO₃⁻ to NH₃ without forming N₂ while simultaneously producing H₂, which lowers the selectivity of NH₃.^138,139 To address this problem, subnanometric metal oxide nanoparticles like BaO, CaO, and MgO are synthesized in operando on the TiO₂ nanosheets.¹⁵⁴In operando growth of 0D subnanometric metal oxide nanocrystals (MO_NC) at the oxygen defect sites of TiO₂ nanosheets promotes NO₃⁻ reduction to synthesize NH₃ selectively. The construction of MO_NC at defect sites is preferable because, due to the limited number of electron lone pairs, the nanocrystals cannot undergo further agglomeration. These MO_NC@OVs act as active sites for the 8e⁻ photoreduction of NO₃⁻ to NH₃. Water splitting, the primary side reaction, produces trace amounts of H₂. However, theoretical calculations indicate that the activation energy of NO₃⁻ reduction is 1.42 eV less than water splitting, favouring NO₃⁻ reduction over the side reaction of water splitting. Compared with the HER, the selectivity of NH₄⁺ has been reported to be 97.67%. The formation of a unique interface between metal oxides and TiO₂ promotes charge separation as well as charge transfer properties of the hybrid catalyst. The light absorption property of the catalyst is improved and the conduction band is also elevated, enhancing the photocatalytic performance of NO₃⁻ reduction. A similar work along these lines features the in situ formation of dynamic Cu₂O sub-nanoclusters on TiO₂ nanosheets (TNS) as the ammonia-producing photocatalyst.¹⁵⁵ This work introduces a different approach to ammonia formation, utilizing dynamic Cu₂O nanoclusters following pseudo-Fehling's reaction. The structure of Cu₂O is unstable under photocatalytic working conditions. Hence, dynamic reconstruction of active Cu₂O sites is required to perform the photocatalytic reactions using actual Cu₂O. The novel active Cu₂O sub-nanoclusters are constructed on-site utilizing photoinduced pseudo-Fehling's reaction in a photocatalytic system containing redox pairs of Cu²⁺ and reducing sugars (formaldehyde and formic acid). The photogenerated electrons (e⁻) and holes (h⁺) participate in the Cu²⁺ to Cu₂O reduction and HCHO to HCOOH oxidation for the formation of Cu₂O NCs and construction of Cu₂O–TNS interface, respectively. The value-added oxidation of HCHO to HCOOH is accompanied by synchronous reduction of different N-sources (NO₃⁻, NO₂⁻, and NO) to value-added NH₃ by the active dynamic Cu₂O sites. This strategy ensures the formation of realistic dynamic active sites and provides a better understanding of the reaction mechanisms underlying the photocatalytic reactions. The photoactivity of TiO₂ semiconductor nanoparticles can also be enhanced by doping with heteroatoms. Co-doping of TiO₂ with synergistic transition metals Ce^3+/4+-cation and S²⁻-anion adjusts the band gap of TiO₂ such that maximum photoactivity is obtained upon visible light irradiation.¹⁷⁹ The ratio of Ce³⁺/Ce⁴⁺ is adjusted with hydrazine to form oxygen vacancies (V_O), which helps maintain the charge and lattice electroneutrality of the TiCeOS catalyst. The Ce³⁺ centers act as the active sites for the adsorption and activation of N₂ while the electron-hopping between heterovalent Ce³⁺/Ce⁴⁺ promotes electron transfer for the photoreduction of N₂. Simultaneously, the V_Os act as the active sites for trapping water molecules and subsequent proton generation from water to protonate N₂ to NH₃. The introduction of metallic nanoparticles like Ru can tune the band gap of the photocatalytic materials, boost electron transfer, and facilitate charge separation. Ru-modified g-C₃N₄ in the presence of visible light can selectively photoreduce NO₃⁻ to NH₃.¹⁵¹ The theoretically calculated activation energy for the rate-determining step of NH₃ synthesis is 0.75 eV, which is much less when compared to the activation energies of competing HER (0.98 eV) and N₂ synthesis (1.36 eV), making the generation of NH₃ more selective. The Ru sites have more Bader charge; hence, the density of electrons is higher on the Ru atoms, making them the active sites for photocatalysis. Ru nanoparticles can also act as cocatalysts for ammonia production when loaded on defective semiconductor-based ZrO_2−x nanoparticles with oxygen vacancies (V_O).¹⁵² The defective ZrO_2−x nanoparticles have a narrow band gap, and excellent reducing and electron donation properties, which makes it an outstanding photocatalyst for ammonia generation. The V_Os stabilize the dispersed Ru nanoparticles, which act as cocatalysts and pose an upward band bending of ZrO_2−x, inducing an interfacial Schottky barrier that promotes the separation of photogenerated charge carriers. The Schottky barrier at the Ru and ZrO_2−x interface also provides a unidirectional pathway for photogenerated electron transport. The Ru nanoparticles trap these electrons, thereby ensuring the supply of requisite electrons for N₂ reduction to NH₃. Ru nanoparticles loaded on defective perovskite and pyrochlore structures can also promote N₂ photoreduction to NH₃.¹⁵³ In this work, different perovskite- and pyrochlore-structured tantalates with low-valent Ta and abundant oxygen vacancies (V_O) were fabricated by high-temperature solid-state reduction. These visible light active dark tantalates have narrow band gaps and upon the introduction of Ru nanoparticles, band bending occurs to construct an interfacial Schottky barrier. The Schottky barrier promotes adsorption of N₂ molecules and electron transfer to reduce N₂ to NH₃. Among all the synthesized tantalates, Ru-loaded defective pyrochlore K₂Ta₂O_6−x has the highest electron-donation properties and chemical stability, and therefore is most effective for ammonia photosynthesis in the gas–solid phase at low pressure.

0D quantum dots (QDs) also prove to be efficient photocatalysts for ammonia production. TiO₂ nanoparticles are one the most widely used semiconductors for photocatalytic applications. Controlling the size of the nanoparticles and reducing them to below 10 nm creates TiO₂ QDs with outstanding photocatalytic properties owing to enhanced charge separation, modified textural properties, and altered redox potentials.¹⁴⁹ Upon the formation of an S-scheme heterojunction of TiO₂ with photoactive Fe₃S₄ crystalline spinels, the photofixation of N₂ to produce NH₃ is boosted in the presence of simulated sunlight. This enhanced photocatalytic activity is attributed to the heterojunction formation, which leads to more visible light absorption, accelerated photoinduced charge separation, and enhanced catalyst surface area. Formation of an S-scheme heterojunction between 0D g-C₃N₄ QDs and 3D macro- and mesoporous TiO_2−x with oxygen vacancies (V_O) and enhanced charge transfer properties promotes the photo/electrocatalytic reduction of NO to NH₃.¹⁸⁰ The photogenerated charge carriers are migrated by the S-scheme heterojunction formed in the g-C₃N₄ QDs/3D-TiO_2−x and the accumulated electrons in the conduction band of g-C₃N₄ quantum dots reduce NO to NH₃. Pyrochlore Bi₂Ti₂O₇ QDs with oxygen vacancies (V_Os), produced hydrothermally from bismuth nitrate and titanium sulfate also produce NH₃ from N₂ upon photoirradiation.¹⁵⁶ When ammonia production is compared with Bi₂Ti₂O₇ nanosheets, the QDs are found to be more photoactive. Despite possessing the same amount of V_Os, the QDs are found to be more photoactive as the synergistic roles of shallow levels arising from V_O of Bi₂Ti₂O₇ QDs and the quantum confinement effect promote adsorption and activation of N₂ molecules to produce NH₃. Ammonia production is also possible from nitrate (NO₃⁻) and nitrite (NO₂⁻) anions using visible-light-active indium phosphide (InP) quantum dots.¹⁵⁷ InP QDs have tuneable absorption properties, high charge separation abilities, efficient mobility of charge carriers, and flexibility in modifying the surface chemistry. The photoexcited charge carriers generated by the III–V InP QDs directly induce NO₃⁻ reduction to produce NH₃. The kinetic experiments from this study confirm that the reduction of NO₃⁻ to NO₂⁻ is the most energy-demanding rate-determining step in the conversion of NO₃⁻ to NH₃. The conversion of NO₂⁻ to NH₃ is faster and almost 100% conversion is achieved in this step. Additionally, in this study, water is used as the source of protons for ammonia production.

Single-atom (SA)-based catalysts are one of the emergent classes of materials currently undergoing widespread investigation. Compared to nanoparticles, single-atoms provide access to a higher density of active sites and contribute to better atomic usage, which in turn escalates the catalytic process.^27,43 The high catalytic activity, selectivity, and stability of SA catalysts can also be attributed to homogeneity and the low coordination capability of the single-atoms.¹⁸¹ The Ru SAs can tune the electronic structure of the oxygen vacancies present in TiO₂ and improve the adsorption of N₂ for photocatalytic N₂RR. The synergistic effect of the two active sites, Ru SA and oxygen vacancies (V_Os) powers the N₂RR, demonstrating that both components of the heterostructures participate in the catalytic reaction. For Ru SAs on CeO₂ support, CeO₂ generates electron–hole pairs upon photoirradiation and the Ru sites pull the electrons toward them, accumulating photogenerated electrons around the Ru sites (Fig. 7A–E).¹⁵⁸ This further modulates the local electron density of the adsorbed N₂ molecules on Ru sites, lowers the energy barrier of the rate-limiting step, and, thereby, promotes the hydrogenation of adsorbed N₂via the associative distal pathway. Ru SAs have also been implanted on other supports like molybdenum oxides and metal–organic frameworks to photocatalytically synthesize ammonia by N₂RR. Here, the active single Ru site anchored on the UiO-66 nodes participates in producing ammonia (Fig. 7F–I).¹⁴⁰ Ru SAs can be embedded in TiO₂ by electronic metal–support interactions, and this catalytic system eradicates the use of external sacrificial agents for producing ammonia (Fig. 7J–M).¹⁸² Pt SAs with tuneable oxidation states photodeposited on a BiOBr support with oxygen vacancies (V_Os) can also efficiently produce ammonia upon photoirradiation.¹⁴¹ The electron–metal support interactions (EMSI) between the Pt and BiOBr and the tuneable oxidation state of Pt SA promote charge transfer between the modifiers (V_O and Pt SA) and the BiOBr support. The accelerated electron transfer and variation of the local electronic structure of V_O by Pt SA leads to selective adsorption and activation of N₂ as well as the reduction of the energy barrier of the rate-limiting step, and promote hydrogenation of *N₂ intermediate to produce NH₃via a multielectron alternating reduction pathway. It is to be noted that though most of the reported SA photocatalysts use noble metals, the metal utilization efficiency of single-atoms in catalysis lowers the cost of noble atoms. However, efforts to prepare a cost-effective, metal-free SA, like a boron-based SA photocatalyst are underway.¹⁸³ Theoretical studies on single-atom B on graphitic carbon nitride support demonstrate that it can efficiently reduce dinitrogen to selectively produce ammonia-suppressing competitive HER via the “σ donation–π* back-donation” properties of the designed catalyst. The activation barrier and overpotential of dinitrogen reduction are notably less for this metal-free SA catalyst when compared with most existing metallic catalysts.

Fig. 7 Zero-dimensional (0D) photocatalysts for ammonia synthesis. (A) Schematic representation of the isolated 0D Ru sites anchored on CeO₂. (B) HAADF-STEM image of Ru–CeO₂. The inset features the corresponding simulated STEM image. (C) UV-Vis DRS of CeO₂ and Ru–CeO₂. (D) PL spectra of Ru–CeO₂ with different Ru loadings. With an increase in Ru loading the PL intensity decreases. (E) NH₃ yield comparison for CeO₂ and Ru–CeO₂ catalysts with different Ru loadings. The maximum yield obtained is 18 mmol g_cat⁻¹ h⁻¹ for Ru/CeO₂-3 where the Ru loading is 1.87%. Reproduced with permission from ref. 158. Copyright 2024, Wiley-VCH. (F) Schematic representation of the 0D Ru single-atom in the Ru₁/d-UiO-66 catalyst. (G) Aberration-corrected HAADF-STEM image of the catalyst. (H) UV-Vis DRS of UiO-66, d-UiO-66, and Ru₁/d-UiO-66. (I) NH₃ yield comparison for UiO-66, d-UiO-66, and Ru₁/d-UiO-66 catalysts, and control experiments conducted in the absence of light, N₂, and water. Reproduced with permission from ref. 140. Copyright 2023, Wiley-VCH. (J) Schematic representation of the Ru single-atom-bonded TiO₂, Ru₁/TiO₂-V_O, prepared by the molten salt method. (K) HADDF-STEM image of Ru₁/TiO₂-V_O, where the Ru single atoms are marked by white circles. (L) UV-Vis DRS of TiO₂, TiO₂-V_O, and Ru₁/TiO₂-V_O. (M) NH₃ yield comparison for TiO₂, TiO₂-V_O, and Ru₁/TiO₂-V_O, and control experiments conducted in the absence of light, N₂, catalysts, and water. Reproduced with permission from ref. 182. Copyright 2023, Elsevier.

4.2. 1D photocatalysts and heterostructures

1D materials, being microscale along one dimension, provide more surface area and easily accessible catalytic active sites than 0D materials, which enhances their catalytic performance. Various 1D carbonaceous materials have been investigated over the past few years because of their stability, good conductivity, and tuneable electronic structure, which exhibit remarkable catalytic activity.¹⁸⁴ The porous structure of the carbonaceous materials also facilitates the adsorption of N-containing species, which enhances the catalytic reaction rate. The onset potential for HER is comparatively more on carbon-based catalysts, which is beneficial for N₂RR and NO_xRR. The introduction of heteroatoms like B, N, and S, which have different electronegativity than C, results in charge distribution in the structure and induces more active sites in the material for the adsorption of reactants. Graphitic carbon nitrides, g-C₃N₄, generally form layered two-dimensional structures that can transfer electrons in two dimensions. However, orderly 1D nanostructures of g-C₃N₄ can constrain charge transfer in one dimension, thus reducing the charge recombination. The introduction of defects, i.e., metal dopant K and –C≡N, into the g-C₃N₄ nanorods further enhances their catalytic activities (Fig. 8A–D).¹⁰⁷ From theoretical studies, it has been observed that the active site is –C≡N, which donates electrons while K centres trap electrons. The synergistic effect of dual defects also promotes light absorption, charge separation, and proton adsorption, and enhances the photocatalytic N₂RR. Carbon-coated hydroxyapatite (Hap-C) nanorod, another carbon-based material derived from bones, exhibits considerable photoluminescence under ultraviolet light.¹⁴² Hydrothermal deposition of Ti³⁺–TiO₂ on the surface of Hap-C promotes the absorption of visible light, increases electron transfer, and reduces agglomeration. Due to these characteristics, the Ti³⁺-TiO₂/HAp-C nanorods can accelerate photocatalytic N₂/H₂O ammonia synthesis compared to pristine Ti³⁺-TiO₂ and HAp-C nanorods separately.

Fig. 8 One-dimensional (1D) photocatalysts for ammonia synthesis. (A) Schematic representation of the structure of graphitic carbon nitride (CN) and graphitic carbon nitride nanorods (CNNR). (B) TEM image of CNRR. (C) UV-Vis DRS of CN and CNRR. (D) NH₃ yield comparison for different carbon nitride photocatalysts. Reproduced with permission from ref. 107. Copyright 2022, Elsevier. (E) Schematic representation of NH₃ synthesis by Bi₅O₇Br nanotubes. (F) TEM image of Bi₅O₇Br-40 nanotubes prepared at 40 °C. (G) UV-Vis DRS of different Bi₅O₇Br prepared at various temperatures. (H) Comparison of NH₃ evolved by using different Bi₅O₇Br catalysts prepared at various temperatures. Reproduced with permission from ref. 144. Copyright 2020, American Chemical Society. (I) Schematic representation of the CdS-decorated WO₃ nanorods, forming a heterojunction. (J) FE-SEM image of WO₃/CdS nanorods. (K) Schematic representation of the electron transfer process at the WO₃/CdS heterojunction for photocatalytic “overall N₂ fixation”. (L) NH₃ yield comparison between CdS, WO₃, and WO₃/CdS catalysts. Reproduced with permission from ref. 159. Copyright 2022, Wiley-VCH.

Oxygen vacancies, the most frequently formed defect, can act as active sites for the adsorption and activation of precursor molecules, and can also alter the optical and electronic properties of photocatalysts. In the case of photocatalysts like Bi₅O₇Br nanotubes with oxygen vacancies (V_O), the adsorption of N₂ can elongate the N≡N bond and form V_O–N=N, which consequently helps in the activation of N₂ (Fig. 8E–H).¹⁴⁴ The concentration of V_O has to be optimum because a too high amount of V_O can entrap the charge carriers to such an extent that it may reduce the rate of photocatalysis. Asymmetric defects induced by V_Os can also enhance the segregation of photogenerated electron–hole pairs through charge redistribution and thereby enhance photocatalytic N₂ reduction reactions. One such example of asymmetric defects is demonstrated by the MoO_3−x nanometric wires fabricated by the hydrothermal process using glycine for inducing defects.¹⁴³ Mo-based catalysts are quite popular N₂ fixation catalysts as the unoccupied d-orbitals of Mo present a strong affinity towards N₂. The defects infused by the V_Os in the structure of MoO_3−x result in charge redistribution and promote N₂ adsorption and activation.

Heterojunction engineering and heteroatom doping are two more approaches besides defects engineering to enhance photocatalytic ammonia synthesis. One such example of a catalyst that has been fabricated using all three approaches is Cu ion-doped W₁₈O₄₉ ultrathin nanowires (Cu-W₁₈O₄₉-x UTNW).¹⁵⁰In situ reduction of the Cu-W₁₈O₄₉-x UTNW with ascorbic acid as an antioxidant and NaBH₄ as a reducing agent form Cu₂O-W₁₈O₄₉-x UTNM, the actual photocatalyst for N₂ reduction. This catalyst is adorned with oxygen vacancies (V_O) which promote N₂ adsorption and activation, and facilitate N≡N bond dissociation. The in situ-generated Cu₂O forms an S-scheme heterojunction with W₁₈O₄₉-x UTNW, which modifies the internal electric field and enhances the separation and transfer of the photogenerated carriers. All these strategies developed for fabricating the nanowires result in efficient photofixation of N₂ to form NH₃. The formation of a heterojunction between two materials induces a strong electronic coupling that promotes interfacial charge transfer and, at the same time, inhibits the recombination of electron–hole pairs. One such recent work reported the construction of a heterojunction between two redox semiconductors, CdS nanoparticles and WO₃ nanorods, forming a 0D–1D-type heterostructure (Fig. 8I–L).¹⁵⁹ The reductive-type CdS has a negative conduction band (CB) position and the electrons in CB_CdS have a strong reduction ability. The oxidative component WO₃ has a higher positive valence band (VB) position and the holes in VB_WO3 have a strong oxidation ability. The heterojunction builds an interface and an electric field between CdS and WO₃ components. Upon photoexcitation, this built-in electric field promotes the separation of electrons and holes, and migrates them via interfacial charge transfer to CdS and WO₃, respectively. These electrons with strong photoreducing ability can efficiently convert N₂ to NH₃, and the holes can simultaneously oxidize N₂ to NO₃⁻ under mild conditions, thereby completing the “overall nitrogen fixation” reactions. Another example of such heterojunction formation is Bi/Bi₂Sn₂O₇ nanocomposites where etching of oxygen vacancies (V_O) in the Bi–O bonds present in Bi₂Sn₂O₇ pyrochlore is utilized for the in situ preparation of metallic Bi.¹⁶⁰ The Schottky junctions formed at the interface of metallic Bi and semiconducting Bi₂Sn₂O₇ lead to the separation of photogenerated charge carriers. Moreover, the contact between the semiconducting Bi₂Sn₂O₇ and metallic Bi facilitates directional electron transfer from Bi₂Sn₂O₇ to Bi, enriching the concentration of photogenerated electrons at the active sites of metallic Bi. These electrons can effectively reduce N₂ to form NH₃. These results highlight the significance of engineering one-dimensional heterostructured catalysts for photocatalytic production of ammonia.

4.3. 2D photocatalysts and heterostructures

Due to their light-harvesting properties, semiconductors play a major role in the photocatalytic synthesis of ammonia. When a semiconductor absorbs a photon with energy equal to or greater than its band gap, electrons are excited from the valence band to the conduction band, creating holes in the valence band. These electrons and holes can migrate to the surface of the semiconductor and initiate the corresponding reduction and oxidation reactions. But sometimes, the photogenerated electrons and holes can recombine either on the surface or in bulk, which in turn hampers the photocatalytic redox reactions. The recombination of charge carriers can be slowed down by introducing defects in the lattice or by forming heterostructures with other low-dimensional materials. TiO₂, ZnO, CdS, and CdSe are some common semiconductors, but the most widely used semiconductor in photocatalysis is TiO₂. TiO₂ is cheap, stable, non-toxic, and has band positions aligned to allow water oxidation and reduction reactions to occur simultaneously.¹⁸⁵ The semiconductors can be 0D, 1D, or 2D. Among these, 2D semiconductor nanosheets have more surface area and active sites, offering better chances of doping and making them potentially more active than 0D or 1D structures. In TiO₂, the most widely explored semiconductor photocatalyst, the oxygen vacancies (V_O) have a potential role in regulating its photocatalytic properties. Rutile TiO₂(110) structure is composed of alternating rows of Ti⁴⁺ and bridging O (O_b) atoms.⁷⁰ These O_b atoms form the vacant sites and the two excess electrons are transferred to each of the 3d orbitals of the two neighbouring Ti⁴⁺, which reduces Ti⁴⁺ to Ti³⁺. The position of the donor level of Ti³⁺ lies below the conduction band of TiO₂. This donor level traps electrons from the conduction band and reduces N₂ to NH₃, confirming Ti³⁺ to be the active site for photocatalysis. Hirakawa and co-workers have identified the active sites of TiO₂ for the reduction of NO₃⁻ to N₂ and NH₃, respectively.¹⁸⁶ They deduced that the Lewis acid sites (Ti⁴⁺) of TiO₂ can perform non-selective reduction of NO₃⁻ to N₂ and NH₃. The surface defects formed by oxygen vacancies (V_Os) can donate electrons to Ti⁴⁺ and reduce them to Ti³⁺, and these Ti³⁺ sites can selectively reduce NO₃⁻ to NH₃. So, by maintaining the ratio of surface defects and Lewis acid sites, a higher selectivity of NH₃ (97%) can be achieved. The high energy level of LUMO π* makes the donation of electrons to π* orbitals more energy-demanding. The rate-limiting step is the reduction of NO₃⁻ to NO₂⁻. Liberation of this NO₂⁻ intermediate from Lewis acid sites of TiO₂ is mainly responsible for N₂ generation. The surface defects of TiO₂ can entrap intermediates like NO₂⁻ and NO and suppress the formation of N₂. Heteroatom doping is another effective strategy for increasing active sites for adsorption and activation of N₂. In a typical case, anatase TiO₂ doped with B proves to be an excellent visible-light-driven catalyst for ammonia synthesis with a band gap of 1.92 eV and a thermodynamic barrier of 0.44 eV.¹⁴⁵ The B non-metal doping narrows the band gap of TiO₂ and makes it photoactive in the visible region. The advantages of using B as a dopant are manifold: (i) B forms interband states near the conduction band of TiO₂ and reduces the recombination of charge carriers; (ii) the hybrid orbitals of B can form π-back bonding with N to activate N₂; (iii) B atoms prohibit binding with H⁺ and thereby suppress the competing HER; and (iv) B atoms, along with the transition metal, act as active sites for N₂ reduction. In another work, C-doping by bottom-up approach from Ti₃SiC₂ MAX precursor forms two-dimensional C-doped TiO_x nanosheets (Fig. 9A–D).⁷¹ These C-TiO_x nanosheets show enhanced photocatalytic activity under visible light irradiation. C-doping increases the number of Ti³⁺ centres in the catalyst, induces photoactivity in visible light, and breaks N≡N, which is the rate-limiting step. Ti³⁺ centres act as the active sites and the two adjacent Ti³⁺ centres chemisorb and activate N₂ molecules. The optimal ratio of Ti³⁺/Ti⁴⁺ in this TiO_x nanosheet is maintained at 72.1% and, with the addition of Ru/RuO₂ co-catalyst, charge recombination is reduced. Apart from TiO₂, tungsten oxide semiconductors are also well-known photocatalysts. Oxygen vacancy (V_O)-rich W₁₈O₄₉ semiconductors with sea-urchin morphology, after low-content Ru modification by NaBH₄ reduction get transformed into rough 2D sheets.¹⁶¹ This sheet-like morphology exposes more active sites and facilitates the adsorption of N₂ molecules. The V_Os enhance the interfacial hydrogen spillover process, thus H* generated from water as the proton source is utilized in the hydrogenation of N₂ to NH₃. Ru captures electrons from W₁₈O₄₉ and acts as the active site for H₂O dissociation. The V_Os play multiple roles, such as facilitating H* migration from Ru for efficient hydrogen spillover, promoting the adsorption of N₂, acting as active sites for hydrogenation reaction, and lowering the overall energy barrier for NH₃ photosynthesis reaction. Plasmonic semiconductors also act as effective photocatalysts for ammonia production. The localized surface plasmon resonance (LSPR) promotes photoconversion of N₂ and the surface oxygen vacancies (V_O) present in these catalysts adsorb and activate the N₂ molecules. However, the V_Os can sometimes get oxidized by the photogenerated holes which might diminish the LSPR properties of the catalyst upon heat treatment. To solve this problem, in one of the recent works, plasmonic MoO_3−x nanosheets with Sb doping have been developed.¹⁶² Though Sb doping does not effectively increase the concentration of V_O, the low oxidation state of Sb can help stabilize the LSPR effect of the plasmonic MoO_3−x. The uncoupled electrons present in the d orbitals of low-valent Sb help maintain the LSPR in the NIR region, thus producing NH₃ from photocatalytic N₂ fixation. Bismuth-based semiconductors like bismuth oxyhalides and bismuth-based binary metal oxides are effective photocatalysts in the visible light range and have high chemical stability. However, their weak interactions with N₂ limit their use as catalysts. This drawback can be overcome with heteroatom doping, vacancy formation, formation of heterostructures, or modification of exposed crystal facets. Chen and co-workers modified BiOBr nanosheets with transition metals like Ni, Fe, and Mo by hydrothermal method and introduced oxygen vacancies (V_Os) via the solvothermal method (Fig. 9E–H).⁷³ The transition metals and V_Os tune the band gap of BiOBr and enhance electron transfer to the anti-bonding orbital of N₂, thereby triggering adsorption and activation of N₂. The simultaneous presence of V_Os and transition metals alters the band positions, diminishing the recombination of charge carriers. The modified BiOBr enhances the photoreduction of N₂ by six times compared to normal BiOBr, forming 46.1 μmol g⁻¹ h⁻¹ of NH₃. Using cocatalysts is another intriguing technique to increase the efficiency of photocatalytic ammonia synthesis. Cocatalysts are generally photocatalytic metallic particles or semiconductors and are used to promote electron transfer processes, inhibit the recombination of charge carriers, and switch to a preferable range of incident light for photocatalysis. In a typical BiOBr semiconductor photocatalyst, the efficacy of solar light-powered N₂RR is enhanced 65 times by using Bi nanoparticles as a cocatalyst.¹⁴⁶ Bi NPs lower the rate of competing HER and construct a Schottky junction at the Bi/BiOBr interface, which facilitates the interfacial transfer of electrons. The unidirectional transfer of electrons toward the Bi active sites accelerates the solar light N₂ conversion efficiency to ammonia. Instead of converting highly pure, air-separated, and expensive N₂, aerobic N₂ reduction of air (N₂/O₂) can be used to produce NH₃ cost-effectively. The addition of water oxidation cocatalysts like CoOOH with BiOCl nanosheets can also enhance photocatalytic NH₃ formation.⁴⁴ The cocatalyst facilitates the OER half-reaction and, consequently, boosts the N₂RR to NH₃. Solvothermally prepared trace Bi⁰-loaded Bi₂MO₆ (M = W, Mo) can prepare ammonia via aerobic photocatalytic reduction of N₂.¹⁸⁷ The role of a trace amount of Bi⁰ is crucial in this reaction. The formation of excess Bi⁰ is inhibited by the presence of O₂ in the reaction medium. The polarization dipole field produced by Bi⁰ favours an effective separation of photogenerated charge carriers and promotes ammonia synthesis from the photocatalytic reduction of N₂. Though the process is an inexpensive route for ammonia production, it has limitations because the O₂ present in the reaction medium might react with photogenerated e⁻s and h⁺s to create reactive oxygen species like O₂˙⁻, ˙OH, which can oxidize the synthesized ammonia or react with the catalyst itself. Therefore, the aerobic reduction of N₂ needs further optimization before it can be effectively used for ammonia synthesis. The problems of aerobic reduction and the expensive nature of pure N₂ can be circumvented by using nitrates as the N-source, often present in wastewater systems. In one of the recent works by our group, a BiOI-based “AmmoGen” microrobot has been prepared for the photocatalytic synthesis of NH₃ from NO₃⁻ upon visible light irradiation.¹⁸⁸ The efficiency of the conventional photo/chemical synthetic techniques for ammonia production is limited by intensive mass-transfer processes. To address this challenge, our group has envisioned a novel technology for ammonia generation, where the photocatalytic BiOI particles are hybridized with magnetic Fe₃O₄ nanoparticles to fabricate an “AmmoGen” microrobot, that can photoreduce nitrate to ammonia using renewable light energy sources. Experiments with “static” particles and “dynamic” microrobots demonstrate that the magnetic propulsion of the “AmmoGen” microrobots significantly improves the mass transfer process, and enhances the photocatalytic ammonia production. This work on microrobots, in principle, can help improve photocatalytic reactions and be utilized in the future for value-added small molecule synthesis. Piezoelectric materials like Bi₃TiNbO₉¹⁸⁹ and BaTiO₃@C¹⁹⁰ have also emerged recently as efficient photocatalysts for CO₂ reduction and organic pollutants degradation. In piezoelectric materials, a polarization-induced electric field is formed by the displacement of the positive and negative charge centres onto opposite sides and the resultant internal electric field (IEF) reduces the recombination of photogenerated electron–hole pairs, enhancing the photocatalytic activity of these materials. Layered bismuth-based piezoelectric SrBi₄Ti₄O₁₅ nanosheets with oxygen vacancies (V_O), formed by low-temperature hydrothermal treatment with glyoxal can effectively photoreduce N₂ to produce NH₃.¹⁹¹ These nanosheets manifest self-polarization, originating from the [TiO₆] octahedral distortion, with the polarization direction parallel to the [Bi₂O₂]²⁺ layer. This polarization field and V_O synergistically promote N₂ adsorption, activation, and reduction to NH₃via three electron-transfer pathways. TiO₂-decorated layered silicate magadiite piezo-photocatalytic nanosheets also exhibit significant enhancement in NH₃ production compared to TiO₂ or silicate magadiite separately.¹⁹² The TiO₂ nanoparticles deposited on silicate magadiite enhance the piezoelectric potential and the polarization-induced internal electric field increases the lifetime of the photogenerated charge carriers in these nanosheets, and the synergistic piezo and photo activities of the catalysts enhance the photocatalytic production of NH₃ from N₂ reduction.

Fig. 9 Two-dimensional (2D) photocatalysts for ammonia synthesis. (A) Schematic representation of the preparation of 2D C–TiO_x nanosheets by thermal oxidation etching at 200 °C. (B) SEM image of C₄–TiO_x obtained after four hours of thermal oxidation etching. (C) UV-visible absorption spectra of Ti₃SiC₂, TiO₂–air, and C₄–TiO_x. (D) NH₃ yield comparison between Ti₃SiC₂, TiO₂–air, and C–TiO_x catalysts obtained after varying durations of thermal etching. Reproduced with permission from ref. 71. Copyright 2021, Wiley-VCH. (E) Schematic representation of the transition metal (TM)-doped BiOBr nanosheets where TM is Fe, Mo, or Ni, obtained by solvothermal reactions. The structure in the dotted box represents the crystal lattice of the TM-doped BiOBr. The best performance is observed for TM = Fe. (F) SEM image of BiOBr-Fe-S-1 catalyst obtained by Fe doping in BiOBr in 1 molar ratio and after solvothermal reaction. (G) The light absorption spectrum of BiOBr-Fe-S-1. (H) NH₃ yield comparison between pristine BiOBr and doped BiOBr before and after solvothermal reactions for different molar ratios of Fe doping. Reproduced with permission from ref. 73. Copyright 2021, Elsevier. (I) Schematic representation of photocatalytic N₂ reduction to NH₃ by B-doped C₃N₅ nanosheets. (J) TEM image of B-C₃N₅. (K) UV-Vis DRS of and (L) NH₃ yields by g-C₃N₄, g-C₃N₅, and B-C₃N₅x photocatalysts (x is the mass percentage of ammoniumborate to 3-amino-1,2,4-triazole). Reproduced with permission from ref. 147. Copyright 2022, Elsevier.

Another commonly used photocatalytic 2D material is graphitic carbon nitride (g-C₃N₄). It consists of tri-s-triazine rings interconnected via tertiary amines, which makes it thermally and chemically stable.¹²⁶ The electronic structure of g-C₃N₄ makes it a prospective photocatalyst as it can absorb solar light due to its small band gap, unlike graphene. The structure of g-C₃N₄ bears a close resemblance to graphite due to π conjugation and the two-dimensional structures. Heteroatom doping can modify the electronic properties of g-C₃N₄ and the vacant active sites can increase the efficiency of the catalytic reactions. For instance, Wang and co-workers prepared B-doped porous g-C₃N₄ nanosheets by thermal treatment.⁷⁷ The B-dopant induces a small-scale p–n type heterojunction, which promotes efficient charge separation in the catalyst. B doping changes the electronic structure by inducing localized electron states in the band gap and promotes strong tail absorption in the visible light region. Also, doping passivates the active N atoms of g-C₃N₄, ensuring that the NH₃ produced does not originate from the exposed N atoms. The B atoms act as active sites for the adsorption and activation of N₂ and exhibit good photocatalytic activity with 313.9 μmol g⁻¹ h⁻¹ yield rate of NH₃. Incorporating N-vacancies (V_N) and adding cocatalysts can also improve the photocatalytic performance of g-C₃N₄ as exhibited by the Ni₃B/V_N-g-C₃N₄ catalyst.¹⁶³ The introduction of V_N enhances visible light absorption and electrical conductivity properties of g-C₃N₄. Noble metal-free cocatalyst Ni₃B nanoparticles act as active sites for the adsorption and activation of N₂. The Schottky junctions formed between Ni₃B and V_N-CN also facilitate the separation of photogenerated charge carriers and migration of electrons for photoreduction of N₂. Introducing one more N atom to g-C₃N₄ forms g-C₃N₅, a two-dimensional material with more π conjugation, which promotes superior charge separation, narrower band gap, and better solar absorption when compared to g-C₃N₄. Li et al. reported a B-doped g-C₃N₅ synthesized by a one-step thermal polymerization technique (Fig. 9I–L).¹⁴⁷ The five azole rings of g-C₃N₅ facilitate electron transfer and decrease the adsorption energy of N₂. B-doping increases the photocatalytic efficiency of g-C₃N₅ like in the previous case (B-doped g-C₃N₄) with B as the active site for photocatalysis and NH₃ yield of 421.18 μmol g⁻¹ h⁻¹, which is 1.72 times that of g-C₃N₅.

Graphdiyne (GDY) is another emergent 2D carbon material, comprising sp and sp² hybrid states endowed with unique properties like highly conjugated and super-large π structures, infinite number of pores, natural band gap, a hole-transport layer, high charge carrier mobility, electronic conductivity, and stability, which make them excellent photocatalysts.¹⁹³ Recent studies indicate that GDY can form heterojunctions with different low-dimensional materials to design excellent catalysts for ammonia photosynthesis. Fe site-specific magnetite when incorporated with GDY, the GDY modulates the coordination environment of magnetite to form two selective valence states, namely, tetrahedrally coordinated Fe and octahedrally coordinated Fe.¹⁴⁸ The coordination environment and valence charge transition regulate the photocatalytic properties of Fe₃O₄ and significantly enhance the photocatalytic nitrogen reduction to ammonia. Graphdiyne is also capable of modulating the surface plasmon resonance of quantum dots (QDs), which is evident from the heterojunction-based GDY-CoO_x QDs catalysts.¹⁶⁴ The natural porous structure, the acetylenic bonds, and the high reduction ability of the GDY incorporate the CoO_x QDs and the combined effects of enhanced surface plasmon resonance and modification of valence states of the metal atom enhance the photocatalytic nitrogen fixation reaction to produce ammonia.

4.4. Organic framework-based photocatalysts and heterostructures

Organic framework-based photocatalysts, viz., porous–organic frameworks (POFs), comprising covalent–organic frameworks (COFs), porous–aromatic frameworks (PAFs), covalent–triazine frameworks (CTFs), and metal–organic frameworks (MOFs), are emergent catalysts for photocatalysis owing to their unique physical and optoelectronic properties such as tuneable light absorption, large specific surface area, periodic arrangement of building blocks, precisely controlled pore size, and easy functional modifications. These properties make these materials unique for applications like catalysis, gas absorption and storage, and sensing. Different morphologies varying from 2D to 3D can be designed by tuning the structure and assembly of covalent bonds. This section will describe a few examples of such organic frameworks and their advantages in photocatalytic ammonia generation. Metal phthalocyanine-derived POFs (MPc-POFs (M = Fe, Co, Ni, Cu, Zn)) consist of conjugated planar π-electron organic macrocycles and their optical and electronic properties can be controlled for optimizing photocatalytic N₂RR.¹⁶⁵ Most photocatalytic N₂RR systems proceed via gas-in-solvent (GIS) systems where gaseous N₂ is purged in water. However, these systems are limited by certain disadvantages like low solubility of N₂ in water, sluggish N₂ transfer at the interface, and interaction of water with the catalyst. In an attempt to promote N₂ transfer and utilization at the active sites of the catalyst, these POFs use a solvent-in-gas (SIG) system as the reaction medium, where MPc-POFs perform the catalysis in direct N₂ gas, and the hydrogenation step is performed in suspended proton source provided as the dispersed phase. Herein, the active sites of the MPc-POFs have access to abundant N₂ molecules and can produce ammonia at an ultrafast rate of 1820.7 μmol g⁻¹ h⁻¹, which is approximately eight times higher than that obtained from conventional GIS (226.2 μmol g⁻¹ h⁻¹).

2D COFs are porous polymers where any photoactive species can be incorporated into the ordered pores of their frameworks such that they can function as catalysts for different photocatalytic reactions.^194,195 The organic units in the 2D π lattices of COF semiconductors can facilitate the separation and transport of photogenerated charge carriers.¹⁹⁴ The one-dimensional polygon channels in the porous COFs also help in mass transport. Single-atoms like Au can be easily anchored into porphyrin-based COFs (COFX-Au, X = 1–5) for photocatalytic NH₃ synthesis (Fig. 10A–D).¹⁶⁶ The performance of the photocatalytic NH₃ synthesis can be controlled by tuning the microenvironment of the single-atom Au catalytic centre resulting from the position of different functional groups at the proximal and distal positions of porphyrin units. In the first case, a strong electron-withdrawing group is attached in COF1-Au, which yields NH₃ at a rate of 37.0 ± 2.5 mmol g_Au⁻¹ h⁻¹. This is 171 times more than that of the porphyrin–Au molecular catalyst. The yield is also 2.8 times higher than COF4-Au, where an electron-donating group is used as the functional group. Thus, it can be deduced that the electron-withdrawing groups can facilitate the separation and transportation of photogenerated electrons within the entire COFX-Au framework. Following this deduction, NH₃ production is increased to 61.1 ± 2.7 mmol g_Au⁻¹ h⁻¹ by attaching two strong electron-withdrawing groups to the COF. The performance of photocatalytic N₂RR can also be enhanced by spatial confinement of N₂ near the surface of the COF for better adsorption and activation. This can be achieved by incorporating hydrophilic carboxyl groups (–COOH) into diketopyrrolopyrrole-based COF (DPPCOOH-COF) and forming a localized hydrogen bond network.¹⁶⁷ The –COOH groups promote water to form a layered or porous structure that interacts with the N₂ molecules and localizes them near the active pyrrole units. This leads to enhanced adsorption and activation of N₂ by the DPPCOOH-COF and reduces the energy for photocatalytic NH₃ generation. The photocatalytic activity of COFs is sometimes limited by the fast recombination of photogenerated charge carriers and the lack of sufficient active sites. Metal active sites can be easily incorporated into the COFs due to their ordered cavities and large specific surface area. Imine-linked 2D COFs have numerous N atoms, which endow the COFs with strong Lewis acidity, and this property of the imine COF-TaTp facilitates the adsorption of N₂, which is a weak Lewis base (Fig. 10E–H).¹⁶⁸ Furthermore, the strong interaction of the Bi metal with the imine nitrogen atom and hydroxyl functionalities of the imine COF facilitates the formation of Bi/COF-TaTp composites through N–Bi–O coordination. The incorporated Bi acts as the active site, which promotes the transfer of charge carriers and activation of N₂ molecules through the donation and back-donation mode. Additionally, the Bi sites inhibit the competitive HER and thus facilitate the photocatalytic NH₃ production from N₂RR.

Fig. 10 Organic framework-based photocatalysts for ammonia synthesis. (A) Schematic representation of nitrogen reduction to ammonia by the porphyrin-based covalent organic framework anchoring Au single-atoms (COF5-Au). The structure in the dotted box represents the microenvironment of the Au catalytic centre. (B) Aberration-corrected HAADF-STEM image of COF1-Au, where Au single-atoms are marked by red circles. (C) UV-Vis diffuse reflectance spectra of COFX-Au, where X = 1–5. (D) NH₃ yield comparison of COFX-Au catalysts. Reproduced with permission from ref. 166. Copyright 2023, American Chemical Society. (E) Schematic representation of the fabrication of Bi NP-decorated imine linked COF-TaTp (Bi/COF-TaTp). (F) TEM image of 5% Bi/COF-TaTp, (G) UV-Vis diffuse reflectance spectra of COF-TaTp, and 5% Bi/COF-TaTp. (H) NH₃ yield comparison between pristine COF and different percentages of Bi NPs. Reproduced with permission from ref. 168. Copyright 2022, Wiley-VCH. (I) Schematic representation of photocatalytic N₂ reduction to NH₃ by Ca²⁺-doped UiO-66. (J) SEM image of UN(Zr-0.30Ca)-4. (K) Light absorption spectrum of UN(Zr-0.30Ca)-4. (L) NH₃ yield comparisons between UN(Zr), UN(Zr-0.10Ca), UN(Zr-0.30Ca), UN(Zr-0.50Ca), and UN(Zr-0.75Ca) were obtained by feeding ZrCl₄ and CaCl₂ at Ca : (Zr + Ca) ratios of 0.1, 0.3, 0.5, and 0.75. Reproduced with permission from ref. 170. Copyright 2023, Elsevier.

Metal–organic frameworks (MOFs), another class of organic frameworks composed of metal ions or clusters and organic ligands are also important for photocatalytic applications. Sometimes, the catalytically active metal centres in MOFs are fully coordinated by bridging organic ligands and terminal ligands, and become inaccessible to the reactants. Hence, to address this issue, a defective Fe-based MIL (100) MOF has been designed by partial removal of these ligands, such as bridging organic ligands and terminal inorganic ligands (OH⁻ and H₂O), leading to the formation of defective structures.¹⁶⁹ The dual defects expose the coordinatively unsaturated Fe sites, which act as the catalytic active centres for adsorbing and activating N₂ molecules. The defects also modify the electronic structures, which favours the transfer of d-orbital electrons from Fe sites into the N₂ π* antibonding orbital to form the key intermediate *NNH in the photocatalytic N₂RR for NH₃ formation. As discussed previously, hydrogen evolution reaction (HER) is a major competitive reaction for photocatalytic N₂RR processes, which limits the production of NH₃. In an attempt to harmonize the competition between HER and N₂RR reactions, a core–shell MOF-based catalyst has been prepared by depositing Ni nanoparticles on the polydopamine shell and NH₂-MIL125 MOF core.¹⁹⁶ Photocatalytic HER occurs at the NH₂-MIL125 MOF core and the evolved H₂ is captured by the Ni nanoparticles and converted into adsorbed *H species. The produced *H is then transferred back to the NH₂-MIL125 core via NiO, which reacts with N₂ and produces NH₃. This “hydrogen state switching” strategy ensures the supply of requisite protons from HER for hydrogenation of N₂ to NH₃ formation, thus “transforming-competition-into-cooperation”. Like other photocatalysts, metal atom doping in MOFs can lead to enhanced photoproduction of NH₃. This is manifested in Ca²⁺-doped UiO-66, a Zr-based MOF, where low electronegative Ca²⁺ increases the electron donation capability of Zr active sites (Fig. 10I–L).¹⁷⁰ Furthermore, the incorporated diamino ligand modifies the band gap of Ca²⁺-UiO-66 to enhance the light absorption range and separation of charge carriers. All these factors contribute to amplified photoreduction of N₂ to produce NH₃ under the full spectrum of light. The formation of heterojunctions can also enhance the photoactivity of MOFs via inhibition of recombination of charge carriers. An example of such MOF is the heterojunction formation between MXene and MIL-125(Ti) MOF via ligand bridging.¹⁷¹ A ligand pre-coupling strategy is employed in this work to create the ligand-bridged MXene/MIL-125(Ti), where a coordination bond between the terminal oxygen group of MXene is coupled to the carboxyl group of the ligand in MIL-125(Ti). This ligand bridge forms a one-directional transport channel for electron transfer from MIL-125(Ti) to MXene. This strategy diminishes the interfacial charge transfer resistance and reduces the recombination of photogenerated electron–hole pairs, thus boosting the photocatalytic N₂ fixation to produce NH₃. The design of nanoarchitectonics using S-scheme heterojunction-based MOFs can also facilitate photocatalytic nitrate reduction to ammonia. A nanohouse-like catalyst structure is constructed employing the nanoarchitectonics technology, where NH₂-MIL-125 MOF with a nanoplate-like morphology serves as the floor.¹⁷² Hollow ZIF-8 cages form the surrounding walls and roof of the nanohouse, and Co(OH)₂ nanosheets are locked inside this nanohouse and connected to the ground by forming a heterojunction with NH₂-MIL-125 MOF. Each component of this unique sandwich-structured superstructure forming the nanohouse array plays a crucial role in photocatalytic NO₃⁻ reduction. The positively charged, hydrophobic, and porous ZIF-8 structure modifies the microenvironment of the NH₂-MIL-125/Co(OH)₂ heterojunction by increasing the NO₃⁻ enrichment, suppressing competitive HER, and promoting NH₄⁺ release. The S-scheme heterojunction between NH₂-MIL-125/Co(OH)₂ enhances the separation of photogenerated electron–hole pairs and boosts the photo-redox capability of the nanohouse catalyst, promoting NH₃ production from NO₃⁻.

5. Low-dimensional electrocatalysts for ammonia synthesis

Electrocatalysts are generally more energy-efficient than photocatalysts because all photons are not effectively utilized in photocatalysis. This inefficiency arises from mismatches between the wavelengths of incident light and the photoactive material's absorption spectrum, as well as from the recombination of photogenerated charge carriers, which reduces the overall efficiency. Electrocatalytic reactions occur under ambient conditions with the external input of electrical energy, which can be easily harnessed from renewable sources like solar, hydro, tidal, or wind energy. The prerequisites for electrocatalysts are: (i) good conductivity; (ii) high surface area and presence of abundant active sites for adsorption of N-containing species; (iii) resistance to corrosion at the working potential range and in the presence of electrolytes; and (iv) long-term stability. The low-dimensional electrocatalytic materials used for ammonia synthesis include transition metal nanoclusters, noble metal-based materials, metal oxides, phosphides, carbides and nitrides, single-atoms, and carbon-based materials. Electrocatalytic materials can be modified by defect engineering, forming heterostructures, or regulating the crystal planes and microenvironment of the catalyst to ensure better electrocatalytic performance. This section will discuss these strategies for designing low-dimensional electrocatalytic materials to synthesize ammonia from N₂ or NO_x sources (Fig. 11 and Table 4).

Fig. 11 Low-dimensional electrocatalysts for ammonia synthesis. Classification of electrocatalysts into 0D, 1D, and 2D, and schematic representations of few selected electrocatalytic LDMs.

Table 4 Synthetic and modification techniques for selected ammonia-generating low-dimensional electrocatalysts

Classification	Materials	Electrocatalysts	Synthesis/modification	Ref.
0D LDMs and heterostructures	Metallic nanoparticles (NPs) and nanoclusters (NCs)	Pd NCs	Seed-mediated method, facet engineering	197
		Au NPs	Seed-mediated method	198
		Ag NCs	Replacement reaction on Cu foam	199
		Aux–Ni NPs	Donor–acceptor construction, galvanic replacement method	98
		Hollow PdCu NPs	In situ reduction and nucleation	200
		Fragmented Bi^0 NPs	In situ reduction and fragmentation of Bi-MOF nanorods under applied potentials	201
	Metallic oxides	TiO2@N–C NPs	N-Impregnated carbon coating strategy	202
		Cu NPs-VO-TiO2	Defect engineering, chemical reduction	203
	Single-atoms (SAs)	Fe1 SA-NC	Thermal modulation	204
		Fe SA-N/P-C	In situ phosphatizing–adsorption–thermolysis process	102
		Cu1 SA-NC/CNT-FEM	Pyrolysis, probe sonication	205
		Cu(I) SA-N3C1	N-doping, wet chemical synthesis, pyrolysis	206
1D LDMs and heterostructures	Metallic nanowires (NWs)	Cu NWs	Chemical reduction	207
		Ru-CuNW	Cation exchange method, thermal annealing, electrochemical prereduction	208
		Rh NCs/SAs-Cu NWs	Electrochemical reduction, galvanic replacement	209
		TA-Au NWs	Surface modification with tannic acid (TA)	210
	Multimetal alloys	Cu6Sn5 alloy	Co-electrodeposition method	61
		IrNi, IrRhNi, and IrFeNi alloys	Chemical Co-reduction, solvothermal method	211
	Metallic oxides	P-TiO2/TP	P-Doping, hydrothermal growth, ion-exchange, phosphating process	212
		a-B2.6C@TiO2/Ti	Hydrothermal synthesis, sputter deposition	213
		Fe2TiO5	Defect engineering, electrospinning method	106
	Spinel oxides (AB2O4)	NiCo2O4 NWs/CC	Hydrothermal synthesis, thermal annealing	214
		Mn2–Co3O4 NTs	Hydrothermal synthesis, thermal annealing	215
		C-Co3O4 NTs	C-Doping, ‘‘surface locking’’ mechanism, in situ topotactic conversion	216
	Carbon nanotubes (CNTs)/nanofibers	MWCNTs/SWCNTs	Calcination and acid treatment	217
		Ni NPs-N-CNRs	N-Doping, electrospinning, carbonization, selective etching with acid	218
		Fe SA-N-carbon nanofibers	N-Doping, C-defects, electrospinning-pyrolysis method	45
2D LDMs and heterostructures	Graphene	Amorphous graphene	Laser irradiation	219
		Amorphous graphene	Transient laser heating	220
		Cu-cis-N2O2/graphene	Breaking of coordination symmetry, thermal annealing	221
	Graphdiyne (GDY)	Mo^0-GDY	Solvothermal method	222
		Ru SAs-GDY-graphene	Eglinton coupling reaction, microwave synthesis, impregnation method	223
		Cu SAs-GDY	Impregnation method	224
		Fe3C-GDY	Pyrolysis, interface engineering, heterojunction formation	225
	Carbon nitrides (CN)	B-BCN	Tuning B/N Lewis acid pairs	226
		BCN	Defect engineering, annealing, tuning frustrated Lewis pairs	227
	Metal and metal oxide-based nanosheets (NS)	Ru NS	Hydrothermal method, plasma treatment	228
		hcp-Co NS	Hydrothermal method	229
		CuCo NS	Co-Electrodeposition method	230
		Amorphous RuO2 NS	Defect engineering, molten salt synthesis	231
	MXenes	Ti3C2O2-VO	Defect engineering, functionalization of terminal group	232
		OH-terminated Ti3C2	Etching, exfoliation, functionalization of terminal group	233
		B-Ti3C2Tx	Etching, annealing, B-doping, pyrolysis	234
	MBenes	FeB2	Reflux method	235
	Transition metal dichalcogenides (TMDs)	F-MoS2	Strain engineering, F-doping, hydrothermal method	236
		V-MoS2	Defect engineering, V-doping, hydrothermal method	237
		Metastable 1T′′′ MoS2	Solid state reaction, thermal annealing, deintercalation of K ions	238
Organic framework-based heterostructures	Metal–organic frameworks (MOFs)	Cu-OUC MOF	Hydrothermal method	239
		Fe2M MOF (M = Fe, Co, Ni, Zn)	Solvothermal method	240
		NiCoBDC MOF-HsGDY	Interface engineering, dual-template, Glaser-coupling, solvothermal methods	241
		PCN-250-Fe3 MOF	Crystallization, thermal activation	242
3D-Printed electrodes based on LDMs	Metallic NPs	Cu Metallic NPs	Sintering, nanostructuring using electrochemical methods, chemical treatment	243
	Carbonaceous material-based heterostructures	MnOx coated 0D carbon black, 1D carbon nanotubes	Thermal activation, atomic layer deposition (ALD)	244
		TiO2-coated carbon	Thermal activation, atomic layer deposition (ALD)	245

---

For the electrocatalytic synthesis of ammonia, the optimal design of an electrocatalytic reactor is pertinent. The electrocatalytic reactor, in principle, consists of the electrocatalytic cell, electrolyte, electrodes, and ion exchange membrane. Electrocatalysts are loaded on the cathode and upon application of a voltage at the electrodes, the electrocatalytic reaction produces ammonia at the cathode by reduction of N₂ or NO_x. Simultaneously, water is oxidized at the anode to generate oxygen. The electrocatalytic cells can be classified as: (i) back-to-back cells; (ii) proton exchange membrane cells; (iii) single-chamber cells; or (iv) H-type cells.²⁴⁶ A back-to-back cell has two gas diffusion electrodes containing N₂ (cathode) and H₂O (anode), partitioned by cation/anion exchange membranes. The proton exchange membrane cell has a similar configuration except that the anodic component is in aqueous form. Because the electrolyte and the cathode have no direct contact, HER is limited in these kinds of cells. However, the reaction medium consists of gas and solid, which lowers the efficiency of the catalytic processes. The single-chamber cell, as evident from its name, has cathodic and anodic reactions occurring in the same electrolyte in one chamber. The electrolyte used here is liquid, but the simultaneous presence of the cathode and anode in one chamber can further oxidize the ammonia produced at the cathode. This problem can be overcome by using an H-type cell, the most frequently used electrochemical cell, where the cathodic and anodic compartments are separated by an ion-exchange membrane. The electrolyte is mostly aqueous and, depending on the reaction conditions may be acidic (H₂SO₄ or HCl), neutral (phosphate buffer or Na₂SO₄), or alkaline (KOH).²⁴⁷ The three-electrode setup is mostly used with a working electrode (e.g., glassy carbon plate, glassy carbon rotating disk, carbon paper, self-supported), reference electrode (e.g., Ag/AgCl), and counter electrode (e.g., Pt, graphite rod).²⁴⁷ Electrocatalysts are deposited on the working electrodes and proper optimization of the electrocatalysts is pertinent. The most common strategies to improve the efficiency of electrocatalysts for ammonia production include crystal facet engineering, heteroatom doping, and the introduction of vacancies and defects. Taking into account the earlier discussion about the disadvantages of photocatalytic N₂ reduction and the advantages of photocatalytic NO₃⁻/NO₂⁻ reduction, it can be said that they also hold for electrocatalytic reductions.

While devising catalysts for electrocatalytic N₂RR or NO_xRR to NH₃, the main objectives are to ensure good conductivity of the catalysts, availability of adequate active sites for electroreduction, and restraining the side reactions like HER or formation of by-products like N₂ and NO by partial reduction of NO_x⁻. In the following subsections, we will discuss the design of low-dimensional electrocatalysts and a strategy to overcome the bottlenecks, aiming to enhance both the faradaic efficiency and selectivity of the resulting NH₃. The most commonly used 0D, 1D, and 2D electrocatalysts use simple one-pot synthetic strategies, whereas their heterostructures rely on multi-step synthesis. Synthesis of 0D metallic and metal oxide nanoparticles or nanocrystals uses strategies like seed-mediated growth, galvanic replacement, or reduction methods.^{98,197,201,203} Single-atom engineering relies on different thermal modulation techniques like pyrolysis or thermolysis.^102,205 1D electrocatalysts like monometallic, bimetallic, multimetal alloys, metal oxides, carbon-based nanowires, nanorods, and nanofibers are prepared by chemical/electrochemical reduction, different deposition techniques, electrospinning methods, and thermal reactions.^{45,61,106,207,209,213,215} The most commonly used 2D electrocatalysts like graphdiynes, metal, metal oxides, and transition metal dichalcogenides use thermal synthetic strategies like solvothermal, hydrothermal, annealing, or pyrolysis.^{222,225,227,228,236} The synthesis of MXenes and MBenes relies on etching, thermal treatments, and the functionalization of terminal groups.^233–235 The electrocatalytic properties of these materials can be enhanced by defect engineering, heteroatom doping, facet engineering, using cocatalysts, formation of heterojunctions, and heterostructures.^{197,203,206,225} A more detailed and precise discussion of the different synthetic and modification strategies of ammonia-synthesizing electrocatalysts is presented in Table 4.

5.1. 0D electrocatalysts and heterostructures

Among different 0D materials, metallic nanoparticles have been extensively studied for electrocatalysis over the last few decades. The plasmonic character and possibilities of tuning the size, shape, and structure of nanoparticles have made them significant for catalysis. The packing of metal atoms in a particle can be different: tuning the size of nanoparticles exposes different crystallographic facets, which can provide different types of catalytic sites.²⁴⁸ Several metal nanoparticle-based catalysts have been reported recently; here, we will discuss a few of these and their roles in catalysis. Electrocatalytic denitrification of NO_x for simultaneous removal of N-containing oxides and ammonia production proves quite efficient on metal surfaces, particularly transition metals. Among different transition metals, Cu is the most efficient for electrocatalytic NO_xRR.^249,250 Cu produces NH₃ selectively as Cu can bind *NO intermediate and by not binding *H, it provides selectivity against competitive HER.⁹² Bimetallic nanoparticles are quite common as they offer two different sites for catalysis. Ni–Au nanoparticles can electrochemically reduce N₂ to NH₃, a reaction facilitated by donor–acceptor couples of Ni and Au (Fig. 12A–D).⁹⁸ Donor–acceptor pairs can adsorb and activate N₂ and also help the desorption of NH₃. Bimetallic PdCu nanocrystals can electrosynthesize ammonia selectively from nitrates (Fig. 12E–H).²⁰⁰ For synthesis, C₁₂N–COOH is employed as the template and dimethyl borane as the reducing agent to form the PdCu hollow nanospheres from Pd- and Cu-containing precursors. The hollow structure of PdCu can trap the intermediates and thus promote the nitrate reduction reaction (NO₃RR) with an effective FE of 87.3%. Crystal-phase engineering can effectively improve the performance of N₂RR and NO_xRR. Nanostructures with controlled facets can exhibit high efficiency for the production of NH₃. Several researchers have performed extensive studies to uncover the correlation between ammonia production and crystal facets. For instance, body-centered cubic (BCC) PdCu nanoparticles surpass face-centered cubic (FCC) PdCu nanoparticles in terms of N₂RR electrocatalysis due to strong d–d orbital interactions between Pd(4d) and Cu(3d) sites.⁴⁷ A comparative study between exposed (100), (111), and (110) facets of Pd reveals that the (100) facet is most active electrocatalytically as it is extremely stable under electrocatalytic reaction conditions, and the energy barrier for desorption of *NH₃ to NH₃ is lower than for other facets.¹⁹⁷ Pd nanocatalysts show face-dependent reduction capability for NO₃⁻ as well as NO₂⁻ in the order of Pd(111) > Pd(100) > Pd(hk0) for NO₃RR and Pd(100) > Pd(hk0) > Pd(111) for NO₂RR in alkaline conditions.²⁵¹ Cuboctahedral Pd with exposed (111) and (100) facets produce the highest quantity of ammonia. Exposure of the two facets makes the cuboctahedra bifunctional, where the (111) facets catalyse the conversion of NO₃⁻ to NO₂⁻ and the (100) facets promote the conversion of NO₂⁻ to NH₃. Theoretical calculations on Cu-based surfaces reveal that among Cu(100), Cu(111), and Cu(110), the best electrocatalytic NO₃RR performance is exhibited by Cu(111) in neutral and alkaline conditions, and by Cu(100) in strongly acidic conditions.²⁵² The local coordination environment and the electronic states are different for every exposed facet, and these results indicate that the pH and the exposed facets on the surface of the electrocatalyst affect NO₃RR accordingly. Fabrication of Au NPs with multiple high-index facets via a modified seed-mediated method can enhance the adsorption of *NNH intermediate, thus promoting the N₂RR reactions.¹⁹⁸ The adsorption of H*, an HER intermediate, is less on the high-indexed facets, which demotes the competing HER reaction, resulting in a moderately high FE of 73.32% for N₂RR at −0.3 V vs. RHE. Doping with heteroatoms can also impact the electrocatalytic performance. Cu-doped Fe₃O₄ particles can effectively reduce NO₃⁻ to NH₃ in a more thermodynamically facile manner with an effective FE ∼100% at −0.6 V vs. RHE.²⁵³ Doping the catalyst surface with Cu lowers the energy barrier and strengthens the adsorption of different reaction intermediates. Doping with 0D materials like Ru atoms can also enhance the catalytic properties of emergent materials like MXenes.²⁵⁴ Ru atoms can back-donate electrons to N₂, which increases its adsorption and activation. The thermodynamic energy required for the first hydrogenation step is also lowered and these synergistic roles of Ru active sites promote N₂RR. Apart from crystal facet engineering and doping, the surface structure of electrocatalysts can also be modulated by defect engineering. Surface defects like oxygen vacancy (V_O) can tune the electronic structure of the metal with unsaturated coordination, allowing them to act as active sites to enhance the adsorption of O-containing species like NO₃⁻ on the surface of the electrocatalyst. Incorporation of the ultra-low level of Pd in Cu₂O induces hierarchical cavity defects and surface defects on the surface of octahedral Cu₂O.²⁵⁵ The cavities and V_Os impart voids, which results in enhanced mass transport and adsorption of NO₃⁻ and H₂O reactants on the surface of Pd–Cu₂O. Moreover, the V_Os weaken the N–O bond of nitrates, and the Pd can act as active sites to generate active H_ads from H₂O, which helps in protonation to form NH₃. It is to be noted that different noble metals like nanocrystalline Pd,²⁵⁶ nanocrystalline Ag,¹⁹⁹ and their composites like Ru/g-C₃N₄²⁵⁷ and bimetallic Cu–Pt²⁵⁸ have been used for catalytic NO₃RR. However, considering the scarcity of noble metals and their expensive nature, these are not very cost-effective. For this reason, researchers have shifted their focus to lower-cost transition metal-based electrocatalysts. Several Bi-, Cu-, and Fe-based nanoparticle and single-atom catalysts with efficient and selective ammonia synthesis have been reported till now. One such work demonstrates that Bi-based-MOF, which acts as a precatalyst and, upon electroreduction, produces fragmented Bi⁰ nanoparticles in situ.²⁰¹ The stable Bi NPs produced can act as effective N₂RR catalysts in both acidic and neutral reaction conditions, though the FEs reported are quite low due to competing HER. Zhou et al. have reported selective electroreduction of NO₃⁻ to NH₃ by in situ-generated 0D-Cu⁰ cubes from its oxide (Cu₂O precursor) while the Cu/Cu₂O favours nitrite production.²⁵⁹ Nitrate reduction at the 0D-Cu cubes is preferred because it has a lower activation energy barrier for nitrates when compared to Cu/Cu₂O and Cu microspheres. The intrinsic activity of NO₃RR is higher on 0D-Cu cubes than on 0D-Cu microspheres because the 0D-Cu cube surface has more Cu(100) facets, whereas the surface of 0D-Cu microspheres is dominated by Cu(111) facets. Atomically dispersed 0D materials like metallic clusters and single-atoms are emergent catalysts for ammonia synthesis owing to their novel features like unique coordination environment and high atom economy. Atomic dispersed Au nanoclusters (Au-NCs) prove to be effective electrocatalysts for ammonia synthesis when embedded on TiO₂ nanosheet support.²⁶⁰ Pristine TiO₂ is not a very effective electrocatalyst for ammonia production due to lower charger transport and the absence of abundant catalytically active sites. However, the Au atoms in Au-NC/TiO₂ can significantly enhance the adsorption and activation of NO₃⁻ and promote the NO₃RR-to-NH₃ by lowering the energy activation barrier.

Fig. 12 Zero-dimensional (0D) electrocatalysts for ammonia synthesis. (A) Schematic representation of the donor–acceptor couples of Ni and Au nanoparticles. (B) SEM image of Au and Ni nanoparticles. (C) Schematic representation of the electron transfer between donor–acceptor couple Au_x/Ni with different Au loadings. (D) NH₃ yields and Faradaic efficiencies using only Ni and Aux/Ni nanoparticles donor–acceptor couple catalysts at −0.14 V vs. RHE. Reproduced with permission from ref. 98. Copyright 2019, American Chemical Society. (E) Schematic of the simulated structure of hollow PdCu (PdCu-H) nanoparticle. (F) TEM image of PdCu-H particle at different magnifications and different locations. (G) NH₃ yields using hollow PdCu-H, PdCu-P nanoparticles, and commercial Pd–P. (H) Schematic illustration of NO₃⁻ reduction over PdCu-H catalyst. Reproduced with permission from ref. 200. Copyright 2023, Wiley-VCH. (I) Schematic representation of the 0D Fe single atom N, P co-modified carbon catalyst (Fe–N/P–C) obtained on pyrolysis from ZIF-8. (J) Aberration-corrected HAADF-STEM of Fe–N/P–C where the Fe single-atoms are marked by red circles. (K) NH₃ yields using N–C (N-modified carbon), N–P–C, Fe–N–C, and Fe–N/P–C catalysts. Reproduced with permission from ref. 102. Copyright 2023, Wiley-VCH. (L) Schematic representation of N₂ electroreduction to NH₃ using oxygen-doped MoC nanoparticles embedded in graphitized carbon shells (O-MoC@NC). (M) TEM image of O-MoC@NC-800 synthesized at 800 °C annealing temperature. (N) NH₃ yields and Faradaic efficiencies after 2 hours of reaction at different potentials. Reproduced with permission from ref. 261. Copyright 2019, American Chemical Society.

Single-atom (SA) confinement also proves to be an effective strategy for devising electrocatalysts that enhance the metal utilization efficiency in catalysis. Nowadays, the use of single-atom catalysts is quite frequent in HER, CO₂RR, NO₃RR, and metal-based SAs are prolific in electrocatalysis. Fe SAs and Cu SAs are the most widely used single-atom electrocatalysts for ammonia generation reported so far. The coordination environment of the single-atoms is vital because charge localization around the single-atom depends on the coordination. The coordination environment also impacts the electronic structure and geometry of the central single-atom and modulates the adsorption of reactants, which in turn has a direct effect on catalytic efficiency. The most common coordination is N₄. Fe–N₄ SAs have higher atomic site activity when compared to bulk or nanostructural catalysts due to a lower thermodynamic barrier.²⁶² Fe–N₄ coordination is also beneficial because due to the lack of neighbouring metal centres, N–N coupling cannot occur. This prevents the formation of N₂ and simultaneously promotes the selectivity of NH₃. To elucidate the structure–performance relationship, Liu et al. have prepared Fe active sites with three different coordination environments: square pyramidal Fe–N₄–OH, slightly broken square planar Fe–N₄, and trigonal pyramidal Fe–N₃.²⁰⁴ The interaction between Fe atoms and O atoms of NO₃⁻ is determined using DFT, which reveals a strong overlap between the 3d orbitals of Fe and 2p orbitals of O (NO₃⁻) in the case of Fe–N₃ as the d orbitals are more localized in N₃ coordination when compared with Fe–N₄ and Fe–N₄–OH. This ensures that the adsorption of NO₃⁻ is more favourable on Fe–N₃, making Fe–N₃ a more active catalyst for NO₃RR. Also, the localization of electrons is near the N atoms for Fe–N₄ and Fe–N₄–OH, whereas for Fe–N₃ the electrons are at both the Fe and N sites, enhancing the charge transfer required for NO₃RR. Nitrate reduction can also be enhanced by changing the coordination environment from N₄ to N₂O₂.²⁶³ The O atoms can regulate the d orbitals of Fe such that the adsorption energy of nitrates decreases on Fe–N₂O₂ when compared with Fe–N₄. The conversion of *NOH to *N is easier on Fe–N₂O₂. The conductivity and selectivity are also greater, which makes Fe–N₂O₂ a potentially better electrocatalyst than Fe–N₄. Besides N and O atoms, P atoms are also used for the coordination of single-atoms. The P atoms can break the local charge symmetry of the Fe atoms and facilitate the adsorption of nitrate as well as other key reduction reaction intermediates (Fig. 12I–K).¹⁰² The atomic interface of the Fe atom is asymmetric, which optimizes the electron density in such a way that it lowers the energy of formation of the NO₃RR intermediates and effectively improves the catalytic performance. Cu SAs also play a predominant role in electrocatalytic NO₃RR as Cu is electroactive, has tuneable electronic structures, and is also cost-effective. The coordination environment for Cu SAs is also crucial for catalysis like Fe SAs. The most frequently used Cu SAs have Cu–N₄ coordination where the symmetry is C_4v.^205,264,265 Symmetric coordination induces weak polarity, which reduces the attractive power of active sites for NO₃⁻ and lowers ammonia production.²⁶⁶ Asymmetric coordination like Cu–N₂O₂ in cis configuration has more polar active sites, which promotes the adsorption of NO₃⁻.²²¹ Moreover, the cis configuration splits the 3d orbitals of Cu, which reduces the energy of the formation barrier of *ONH intermediate. Modification of the coordination environment to Cu–N₃C₁ has manifold benefits.²⁰⁶ It increases the energy barrier of competing HER, reduces the desorption of intermediates produced, increases the adsorption of H*, and enhances the electrocatalytic hydrogenation process. However, the structural sensitivity of these metal SA catalysts can sometimes decrease the selectivity of ammonia. For instance, Co-based SAs can more selectively produce hydroxylamine (NH₂OH) from the electroreduction of NO, whereas metallic Co in a hexagonal close-packed (hcp) lattice produces NH₃ selectively.²⁶⁷ The hcp-Co has excellent electron and proton transfer properties, resulting in superior activity for ammonia production. Co-SA has a positively charged active centre and this modified electronic structure accounts for the exceptional hydroxylamine selectivity. hcp-Co can lead to vertical and strong NO adsorption, whereas moderate adsorption occurs on the Co-SA. Hence, the formation of NOH* intermediate is more favourable over hcp-Co, while HNO* formation is preferable on Co-SA. This local structural difference between hcp-Co and Co-SA leads to the selective formation of NH₃ and NH₂OH, respectively.

Besides photocatalysis, 0D semiconductor nanoparticles can also perform efficient electrocatalysis owing to their conductive properties. For instance, nitrogen-impregnated carbon increases the conductivity and causes partial oxygen defects on the surface of TiO₂.²⁰² This leads to the coexistence of Ti³⁺ and Ti⁴⁺, which increases charge transfer and, thus, facilitates the catalytic reactions. Here, in addition to ammonia, hydrazine is also obtained as a by-product for the electrocatalytic N₂RR, which somewhat limits the selectivity of ammonia. The formation of heterostructures can also improve the photocatalytic properties of TiO₂. In Cu nanoparticle-loaded oxygen-deficient TiO₂ (Cu NP–V_O–TiO₂), a strong metal support interaction is induced between Cu and TiO₂, improving electron density and electron transfer processes.²⁰³ This interaction also modifies the local charge distribution properties of the catalyst such that it is asymmetrical, resulting in polarization of the adsorbed N₂ and, consequently, improved N₂ activation. The experimental results reveal that the oxygen vacancies and Cu NPs act as active sites, and the metal support interaction helps in electron transfer from Cu to the oxygen-deficient TiO₂. In N₂RR, no hydrazine has been obtained as a by-product, thus confirming the good selectivity of NH₃. Further, the absence of hydrazine indicates that the catalytic mechanism proceeds via the associative distal pathway.

Core–shell nanostructures are a class of 0D nanomaterials composed of an inner core and outer shell with a distinct boundary between them. The interface connectivity between the core and shell materials sometimes offers new properties like a tuneable electronic surface and abundant active sites, which are beneficial for catalytic reactions. The core–shell structures have another advantage in catalytic reactions. The thickness of the shells and the composition of the core materials can affect the adsorption–desorption properties of intermediates and products so that the catalytic activity can be tuned easily by modification of the core–shell structures and compositions. Qu et al. synthesized a core–shell structure comprising oxygen-containing molybdenum carbides (O-MoC) and nitrogen-doped carbon layers (N-doped C) by pyrolysis of ammonium heptamolybdate and dopamine, which can electrocatalytically reduce N₂ to ammonia (Fig. 12L–N).²⁶¹ The interaction between O-MoC and N-doped C results in an electronic structure that makes the competing HER negligible. This increases the FE of the catalyst, and no by-products like hydrazine are produced here, which also increases the selectivity of NH₃ produced.

5.2. 1D electrocatalysts and heterostructures

The structure of 1D materials has diverse advantages, which makes them prospective in the field of catalysis. The aspect ratio of length: height for 1D structures is always greater than 1. The diameter of the 1D nanowires can be tuned to match the wavelength of photons and the diffusion length of charge carriers, tailored to specific catalytic applications.²⁶⁸ The high surface-to-volume ratio of these materials ensures the availability of enough active sites on the surface for catalysis. These characteristics alter the properties of 1D structures compared to their bulk counterparts, making them advantageous for catalysis. For instance, 1D Au nanowires coated with oxygen-rich tannic acid can electrochemically reduce N₂ to NH₃.²¹⁰ Tannic acid is loosely packed on the surface of Au nanowires and has high porosity, which improves access to most of the metal sites on the surface of the catalyst for adsorption and activation of N₂. The oxygen richness of the coating at the gold and tannic acid interface enhances N₂RR. Hydrazine has not been detected at any potential, thus confirming the selectivity of NH₃. Examples of other catalytic nanowires include single-atom-dispersed Cu nanowires where single-atoms (SAs) like Ru and Rh dopants are atomically dispersed in a Cu nanowire matrix to form a 0D–1D type heterostructure.^208,209 The Ru/Rh SAs offer diverse atomic and electronic properties that make them suitable for catalytic conversion of nitrates to ammonia. The NO₃RR reaction occurs at the Ru/Rh sites while the Cu sites suppress the competitive HER. The selectivity of NO₃⁻ to NH₃ reaches as high as 99.8%. The authors have also coupled NO₃RR to the air-stripping process and successfully obtained NH₃ and NH₄Cl, thus proving that the system can convert the nitrates present in wastewater to value-added products. Cu materials are widely used for NO₃RR reactions because the energies of the d orbital of Cu and the lowest occupied π* orbital of NO₃⁻ are similar, which is favourable for NO₃RR.²⁶⁹ But Cu materials perform NO₃RR at comparatively negative potentials because of the low nucleophilicity and affinity of nitrate on the Cu surface, which makes the competing HER more favourable.²⁷⁰ To address this issue, researchers have modified the electronic structure of Cu-based catalysts using metal heteroatoms or by doping with non-metals. For instance, B-doping in Cu nanowires can transfer electrons from the 3d orbitals of Cu to the empty 2p orbitals of B, thereby inducing electron localization and improving catalytic activity at the Cu sites.²⁷¹ DFT calculations indicate that boron incorporation also suppresses competing HER, increases the adsorption of intermediates like *NO₃, and enhances the conversion of *NO to *HNO. According to a recent study, CO₂ molecules enhance the reduction of nitrites to ammonia on Cu nanowire catalysts, achieving a FE of nearly 100% in a broad potential range.²⁰⁷ Reaction mechanism investigations show that CO₂ is first reduced to form *CO, which then facilitates the reduction of *NO intermediates to *N. This leads to a decrease in the energy barrier of deoxygenation of *NO intermediates, which is the rate-determining step of the nitrite electroreduction process. Additionally, the adsorbed *CO species can accelerate the hydrogenation of *NH₂ intermediates to form NH₃ at an enhanced rate. Besides the electroreduction of nitrite, CO₂ molecules can also accelerate the reduction of other nitro-containing compounds such as the conversion of nitrate to ammonia and nitrobenzene to aniline.

Monometallic nanomaterials with unconventional crystal phases, such as face-centered cubic (fcc) and hexagonal close-packed (hcp) phases exhibit enhanced catalytic activities over their common phase counterparts. However, monometallic materials sometimes lack sufficient active sites for the adsorption and stabilization of multiple intermediates, particularly in multi-step reduction reactions like NO_xRR and N₂RR. In this attempt, multimetal alloy nanomaterials with unconventional phases have been fabricated which can enhance the electrocatalytic reactions toward ammonia synthesis. One example of such alloy is Cu–Sn-based pine-needle structures that convert NO to ammonia in a flow cell.⁶¹ Theoretical investigation implicates that the energy barriers of protonation are low over Cu₆Sn₅-derived surface structures, which results in enhanced ammonia production. IrNi-based alloy nanobranches (NBs) with unconventional hcp phase demonstrate superior electrocatalytic NO₂RR performance toward ammonia synthesis.²¹¹ Solvothermally fabricated IrNi, IrRhNi, and IrFeNi alloy NBs consist of a Ni-rich core and an Ir-rich shell. Theoretical studies indicate that the Ir–Ni interactions within the hcp IrNi alloy can accelerate the electron transfer processes for NO₂RR. The hcp IrNi alloy surface also produces more active hydrogen, which reduces the energy barriers for the hydrogenation steps for ammonia formation.

One-dimensional carbon-based nanomaterials like carbon nanotubes (CNTs) and carbon nanorods (CNRs) exhibit excellent catalytic properties toward ammonia production, owing to their electron conductivity, large surface area, and stability. CNTs have a graphite sheet-like structure with sp² hybridized-C atoms, and the sheets can be rolled to form a cylinder-like structure.²⁷² The inner and outer walls of CNTs provide numerous active sites for catalysis. Single-walled and multi-walled carbon nanotubes (SWCNTs and MWCNTs) are promising 1D materials for efficient electrochemical NO₃RR. However, the work of Harmon et al. suggests that when heteroatoms like O and N are introduced into MWCNTs, the efficiency of catalytic NO₃RR decreases (Fig. 13A–C).²¹⁷ This observation proves that the catalytic reaction occurs at the C atoms present on the surface of the carbon nanotubes. The electron-rich O and N dopants reduce the number of active C sites on the surface, thus diminishing the catalytic reaction. Some Fe contaminant from the synthetic counterparts is present in SWCNTs, but not on the surface; hence, it does not affect the NO₃RR significantly. The authors also found that the surfaces of SWCNTs are more catalytically active and selective for NO₃RR than for MWCNTs. This might be because the SWCNTs have larger curvature than MWCNTs, which modifies the bonding and, consequently, the electronic structure of the active C atoms. For SWCNTs, the FE for NH₃ is 90% at −0.85 V vs. RHE. Other by-products like gaseous H₂ (FE 6%), liquid NO₂⁻ (FE 4%), and NH₂OH are also detected. Carbon nanorods (CNRs) are one-dimensional rod-shaped carbon materials with a moderate aspect ratio, high surface area, and good conductivity. The electrocatalytic properties of CNRs can be modulated by doping with heteroatoms like N, and the incorporation of transition metals like Ni.²¹⁸ Ni-embedded N-doped carbon nanorods (Ni-NCNR) can selectively electroreduce nitric oxide to form ammonia. The Ni atoms act as the active sites for adsorption and activation of NO. N-doping alters the electronic structure and enhances the interaction between the Ni active sites and CNR support; hence Ni incorporation and N-doping synergistically enhance NORR for NH₃ production. N-doped 1D carbon nanofibers with carbon defects can also accelerate electrocatalytic N₂RR for NH₃ synthesis.⁴⁵ The carbon defects enhance the water-splitting process, generating abundant protons for the protonation of N₂ to NH₃. Further, the C defects assist the coordination of the Fe atom with four N atoms. The Fe–N₄ sites, together with the adjacent C defects, promote the protonation reactions by reducing the energy barrier of the process, benefiting the overall N₂RR process.

Fig. 13 One-dimensional (1D) electrocatalysts for ammonia synthesis. (A) Schematic representation of electrochemical NO₃⁻ reduction to NH₃ by 1D carbon nanotube (CNT). (B) SEM images of single-walled carbon nanotubes (SWCNTs), multi-walled carbon nanotubes (MWCNTs), mildly oxidized MWCNTs (OCNTs), and reduced OCNTs (ROCNTs). (C) Faradaic efficiencies of possible products H₂, NO₂⁻, NH₂OH, and NH₃ obtained from NO₃⁻ reduction using SWCNT, MWCNT, OCNT, and ROCNT catalysts. Reproduced with permission from ref. 217. Copyright 2022, American Chemical Society. (D) SEM image of NiCO precursor. (E) SEM image of NiCO₂O₄ obtained from the precursor upon annealing. (F) TEM image of NiCo₂O₄ nanowire. (G) NH₃ yields and Faradaic efficiencies obtained at different potentials using NiCo₂O₄/CC nanowire catalyst. (H) Schematic illustration of Zn–NO₃⁻ battery using NiCo₂O₄/CC as the cathode. Reproduced with permission from ref. 214. Copyright 2022, Wiley-VCH. (I) Schematic representation of the fabrication of Mn-incorporated Co₃O₄ (Co₃O₄–Mn_x) nanotubes. (J) TEM image of Co₃O₄–Mn₂ nanotube, where Mn : Co ratio is 2 : 1. (K) NH₃ yields and Faradaic efficiencies obtained using Co₃O₄–Mn₂ and Co₃O₄ catalysts at various potentials. Reproduced with permission from ref. 215. Copyright 2023, Elsevier. (L) Schematic representation of the synthetic procedure of Fe₂TiO₅ nanofibers. (M) SEM image of Fe₂TiO₅ nanofibers. (N) Faradaic efficiencies of possible products H₂, NO₂⁻, and NH₃ obtained from NO₃⁻ reduction using Fe₂TiO₅ nanofibers at different potentials. Reproduced with permission from ref. 106. Copyright 2022, Wiley-VCH.

Spinel oxides have a general formula of AB₂O₄, where A and B are transition metal cations. A is a divalent cation and B is a trivalent cation, occupying tetrahedral and octahedral sites, respectively.²⁷³ Due to the distinct electronic structures and reactivities of A and B, their interactions with the reactants also differ significantly. Hence, spinel-type oxides have more catalytic activity compared to single-metal oxides. Bimetallic spinel oxides have a tuneable band gap, better electrical conductivity, and adsorption ability of reactants than single-metal spinel oxides.²⁷⁴ Such a redox-active bimetallic spinel oxide, NiCo₂O₄, can perform electrocatalytic NO₃RR to NH₃ under ambient conditions (Fig. 13D–H).²¹⁴ These spinel oxides have multivalent metals and high electronic conductivity, which are favourable for catalytic reactions. Also, the effect of bimetallic centres Ni and Co result in efficient NO₃RR, and the FE reaches a maximum value of ∼99% at −0.3 V vs. RHE, although small amounts of NO₂⁻ and N₂H₄ were obtained as by-products with NH₃. Ni²⁺ acts like a p-type dopant and replaces a Co³⁺ in Co₃O₄ to form NiCo₂O₄. This converts the semiconductor minority spin channel to conducting, and NiCo₂O₄ possesses half-metal characteristics. The DFT calculations show that the half-metal characteristic of NiCo₂O₄ facilitates electron transfer. Nitrates are easily adsorbed on the surface of Co₃O₄ compared to NiCo₂O₄ but, surprisingly, the NO₃RR is more favourable for NiCo₂O₄. This proves that the Sabatier Principle holds correct, i.e., the adsorption of reactants on the surface should be intermediate, and too weak or too strong interaction disfavours the catalytic process. The catalyst is also used as a cathode in a Zn–NO₃⁻ battery with a high yield of 48.5 μmol h⁻¹ cm⁻² NH₃ and an FE of 96.1%. Another such spinel oxide electrocatalyst, Mn-incorporated Co₃O₄ (Co₃O₄–Mn₂) nanotubes is synthesized via hydrothermal method and annealing (Fig. 13I–K).²¹⁵ Mn partially replaces the Co cation in the CoO₆ octahedron of Co₃O₄ to form Co₃O₄–Mn₂. Mn incorporation not only improves conductivity but also suppresses the HER. The surface sites of Co₃O₄–Mn₂ are less active than Co₃O₄. A Co : Mn ratio 1 : 2 produces the highest FE of 99.5% at −1.2 V vs. RHE. Doped spinel oxides like carbon-doped cobalt oxide (C/Co₃O₄) hollow nanotubes also exhibit highly efficient NH₃ synthesis from NO₂⁻ reduction with FE of almost 100% in the potential range of −0.1 to −0.6 V vs. RHE.²¹⁶ The C doping facilitates charge transfer by inducing a local electric field and reduces the energy barrier for the *N + e⁻ + H₂O → *NH + OH⁻ step, the rate-determining step of the NO₂RR process. This electrocatalytic system is further utilized to construct a Zn–NO₂⁻ battery that can concurrently degrade NO₂⁻, generate value-added NH₃, and create electricity.

Semiconductors are often used extensively as electrocatalysts. However, typical semiconductor materials have less electronic conductivity, which impedes their electrocatalytic activity. The electronic structure of semiconductors can be modified by introducing defects like oxygen vacancies (V_O) or by heteroatom doping. Besides modulating the electronic structure, doping TiO₂ nanobelts with the P atom induces charge redistribution and generates V_Os around the doping sites.²¹² As a result, P-TiO₂ exhibits better NO₂RR performance to produce ammonia compared with pristine TiO₂. Amorphous boron carbide sputtered on TiO₂ (a-B_2.6C@TiO₂) nanobelts also exhibits selective NH₃ synthesis via electrocatalytic NORR.²¹³ Theoretical studies indicate that the B–C bonding in the boron carbide layer effectively injects electrons to the NO π_2p* orbital, thus activating NO and ensuring the complete reduction of NO to NH₃ by lowering the energy barriers. Employing this electrocatalytic a-B_2.6C@TiO₂ as the cathode, a Zn–NO battery is assembled to produce ammonia and electricity concurrently. Pseudo-Brookite Fe₂TiO₅ nanofibers are narrow-band gap semiconductors with TiO₂-like atomic and electronic properties (Fig. 13L–N).¹⁰⁶ The highly reducing Fe atoms can easily replace the Ti⁴⁺ and induce abundant oxygen vacancies (V_Os) in the structure. The V_Os can reduce the adsorption energy of NO₃⁻ and boost the catalytic activity for NH₃ production. Iron phosphide (FeP), a transition metal phosphide-based narrow-band gap semiconductor, is also an effective electrocatalyst for NO₂RR from wastewater to generate ammonia.²⁷⁵ The NO₂⁻ ions bind to the (211) and (011) facets of the two adjacent Fe atoms present in FeP, which are the main active facets for NO₂RR. Further, FeP has moderate atomic hydrogen (H*) adsorption capability, which facilitates NH₃ formation, whereas excessive H* adsorption leads to competitive HER, reducing the selectivity of NH₃.

5.3. 2D electrocatalysts and heterostructures

Two-dimensional carbon-based electrocatalysts are emergent materials for the electrosynthesis of ammonia owing to their diverse benefits such as low cost, easy modification of atomic or molecular structures, and tolerance to acidic and alkaline electrolytes. State-of-the-art 2D carbon-based electrocatalysts for ammonia synthesis include graphene, graphdiyne, and carbon nitrides. These materials also provide superior support for anchoring 0D or 1D nanomaterials to promote their catalytic activity. Engineering their structures by introducing defects or heteroatom doping are various strategies to enhance their catalytic activity. Graphene, one of the most commonly used 2D catalysts, has a nanosheet-like structure with sp² hybridized carbon atoms, and excellent conductivity and chemical stability.²⁷⁶ The 2D π-conjugated structure of graphene also contributes to the adsorption of N-containing reactants on its surface and facilitates the catalytic reaction. Recently, amorphous graphene has been used extensively for catalytic purposes. The amorphous atomic structural features, as well as the graphene structural features, promote catalytic activity. Such an amorphous graphene electrocatalyst is synthesized by laser irradiation and is composed of disordered four- to eight-membered polygons (Fig. 14A–C).²¹⁹ The catalytic activity of amorphous graphene has been compared with crystalline reduced graphene oxide (rGO). It is observed that the adsorption properties of amorphous graphene are more due to structural deformations while rGO has lower adsorption properties due to aromaticity. Another graphene-based catalyst with amorphous/crystalline heterophase is synthesized by infrared laser induction.²²⁰ This graphene has intermediate crystallinity between amorphous carbon and crystalline graphene, and contains aromatic rings and distorted polygons. The presence of heterophase can modulate the electronic properties of the catalyst as the density of defects and state near the Fermi level increases, resulting in high capacitance. The heterophased graphene has abundant electrons and long-range disorders that favour 8e⁻ NO₃RR to NH₃. The formation of heterostructures with graphene has certain advantages: (i) promotion of the transfer of electrons to graphene is readily achievable because the Fermi level of graphene is at 0 V vs. NHE, whereas the position of the conduction band of the other component is generally higher than the Fermi level of graphene; (ii) separation of charge carriers is easier due to the rapid electron transfer from the second catalytic component to graphene; (iii) abundant anchoring sites are present in functionalized graphene for the dispersion of nanoparticles or single-atoms. Single-atoms (SAs) are one of the most emerging catalysts in recent times as discussed above. The SAs mostly coordinate to four N atoms, but this induces weak polarity to the SAs, which is unsatisfactory for catalytic processes. If this coordination with N can be partially substituted with heteroatoms like O or S, then the polarity on SA sites will increase, resulting in the adsorption of more NO₃⁻. Functionalized graphene with two N and two O in cis configuration can encapsulate Cu SAs.²²¹ As discussed before, the polarity of Cu sites causes more NO₃⁻ adsorption on the electrocatalytic surface and the cis configuration modifies the 3d orbitals of Cu so that it can form a π-complex with *ONH and reduce the energy barrier of the catalytic process. This reveals that functionalized graphene can also moderate the coordination of the SA catalyst and simulate its catalytic activity.

Fig. 14 Two-dimensional (2D) electrocatalysts for ammonia synthesis. (A) HRTEM images of amorphous graphene obtained by laser induction in air (ox-LIG) at different magnifications. (B) HRTEM image of reduced graphene oxide (rGO) at different magnifications. (C) NH₃ yields and Faradaic efficiencies obtained at different time intervals using amorphous graphene obtained by laser induction in air (ox-LIG), in an inert atmosphere (LIG), and reduced graphene oxide (rGO) at a potential of −0.73 V vs. RHE. Reproduced with permission from ref. 219. Copyright 2023, Wiley-VCH. (D) Schematic representation of defective boron carbon nitride (BCN) nanosheets with unsaturated B and N atoms as a frustrated Lewis pair (FLP). The pulling effect of FLPs captures and activates N₂, for N₂ reduction to NH₃. (E) STEM image of BCN. (F) NH₃ yields and Faradaic efficiencies obtained using BCN catalyst at different potentials. Reproduced with permission from ref. 227. Copyright 2022, Wiley-VCH. (G) Schematic representation of OH-terminated Ti₃C₂ MXene, where the hydrogen bonding between –OH groups and NO₃⁻ facilitates NO₃⁻ reduction to NH₃. (H) SEM image of Ti₃C₂ MXene. (I) Faradaic efficiencies of NH₄⁺ and NO₂⁻ obtained at different potentials using Ti₃C₂ MXene. Reproduced with permission from ref. 233. Copyright 2023, Wiley-VCH. (J) Lattice structures of the 2H, 1T′, and 1T′′ phases of MoS₂. (K) Comparison of NH₃ yields and Faradaic efficiencies for 1T′′′ and 1T′ MoS₂ at a potential of −0.3 V vs. RHE and for 2H MoS₂ at a potential of −0.4 V vs. RHE. Reproduced with permission from ref. 238. Copyright 2021, Wiley-VCH.

Graphdiyne (GDY), a carbon allotrope with one-atom-thickness is an emerging 2D carbon material. It comprises sp and sp² hybridized carbon atoms and has unique properties like high surface area, large cavity structure, large network plane, excellent hole transport properties, uneven surface charge, and multiple active sites, which make it an excellent two-dimensional catalyst.^193,277 Recent studies indicate that GDY-based low-dimensional materials exhibit excellent activity, selectivity, and stability toward electrocatalytic ammonia synthesis. Due to the presence of the reductive alkaline bond in the structure of GDY, it can self-reduce Pd²⁺ to form Pd-GDY.²⁷⁸ The coupling interaction between the Pd, C1, and C2 sites of GDY enhances the electron transfer process, thereby increasing the activity and selectivity of electrocatalytic N₂RR to produce ammonia with 100% selectivity in a neutral medium. Individual zero-valent atoms have unique catalytic properties owing to their electronic structure, high activity, and selectivity. One such zero-valent Mo⁰-GDY catalyst has been prepared based on the incomplete electron transfer properties between the GDY and Mo atom.²²² The atomically dispersed Mo⁰ atoms endow the catalyst with excellent activity toward electrocatalytic nitrogen reduction and hydrogen evolution reactions with highly efficient ammonia and hydrogen production, respectively. The confinement of single-atoms (SAs) like Ru, Rh, and Co in GDY structures also confers excellent properties toward electrocatalytic nitrogen reduction reactions to produce ammonia.^271,279 Single-atom (SA) and double-atom (DA)-adorned GDY like Cu₂-GDY also exhibits enhanced NORR for ammonia generation.²²⁴ This work shows that the d-band centre plays a pivotal role in the adsorption and hydrogenation of NO. The NO molecule activation on SAs and DAs is driven by electron “donation/back-donation” interactions between the metal atom and NO. On the SAs, the thermodynamic process (NH₃ and H₂O molecule desorption) dominates the entire NORR process, and Cu-GDY exhibits high NORR with selective NH₃ formation over H₂. In contrast, on the DAs, the electrochemical hydrogenation processes control the NORR, and Cu₂-GDY exhibits the highest selectivity toward NH₃ among Fe₂, Co₂, Ni₂, and Cu₂. These works explore the utilization of hybridized 0D atomically dispersed catalysts on 2D GDY-based structures toward electrocatalytic ammonia generation. Over the past few years, GDY-based heterojunction catalysts have also been developed to enhance electrocatalytic ammonia synthesis. For instance, the Fe₃C@GDY heterojunction catalyst has electron-donating triple bonds in GDY and electron-accepting Fe₃C, wherein, the sp-carbon–metal–carbon structures at the interface promote charge transfer and electrical conductivity in the catalyst.²²⁵ GDY can also regulate the coordination environment of the Fe atoms and thereby improve the adsorption and desorption of the reactants and intermediates, promoting electrocatalytic nitrate reduction to ammonia. The incomplete charge transfer between the donor–acceptor Fe₃C@GDY interface also enhances the activity and selectivity of the catalytic ammonia production. Compared to the pristine catalysts, the heterogenized GDY structures have faster electron transfer properties and the integration with GDY also provides enhanced stability to the heterogenized catalysts.

Boron (B)-based carbon 2D materials have also emerged as efficient electrocatalysts for ammonia synthesis. Doping carbon-based materials with boron or nitrogen results in charge redistribution, boosting the chemisorption of reactants.²⁸⁰ The B and N doping also tune the band gap, spin, and charge density of the boron carbon nitride (BCN), promoting N₂RR and NO_xRR over HER. The empty sp² orbital of B can interact with the lone pair of N electrons to adsorb and activate the N-species and hinder the binding of H⁺ to suppress the competitive HER. Surface anion vacancies can enhance the availability of electrons and provide suitable active sites for binding N₂ molecules. Defective BCN nanosheets with unsaturated B (e⁻-deficient) and N (e⁻-rich) atoms form the frustrated Lewis acid (LA) and base (LB) pairs (Fig. 14D–F).²²⁷ These adjacent LA and LB pairs can efficiently adsorb N₂ to form a six-membered ring as an intermediate and allow dissociation of N≡N at much lower energy due to the pull–pull effect by heterolysis. The presence of frustrated Lewis pair B and N as dual active sites increases the electrocatalytic activity of BCN and promotes N₂RR. The tunability of the Lewis pairs of B and N has substantial impacts on N₂RR.²²⁶ The B-enriched BCN exhibits better N₂RR activity than the N-enriched BCN. Theoretical studies indicate that the energy of each step of N₂RR by B-enriched BCN is relatively lower than that of N-enriched BCN, which results in enhanced N₂RR by B-BCN.

2D metal and metal oxide-based nanostructures are extensively used in electrocatalysis due to their high surface-to-volume ratio, unique electronic structures, and abundant exposed active sites. For instance, Ru nanosheets with low coordination numbers can exhibit electrocatalytic properties toward NORR to NH₃.²²⁸ The low coordination number of the Ru sites promotes the adsorption of NO molecules and lowers the energy barrier of the rate-determining hydrogenation step. The use of expensive noble metals can be circumvented by replacing them with redox-active transition metals, such as Co. Co nanosheets with hexagonal-close-packing (hcp) act as efficient catalysts for NORR compared to the face-centered cubic (fcc) phase of Co nanosheets.²²⁹ Enhanced electron donation from the d-π* orbitals of hcp-Co to the adsorbed *NO facilitates NORR. Additionally, the proton diffusion process in the hcp-Co is energetically favourable, ensuring the availability of more protons for the protonation of NO to NH₃. Bimetallic catalysts provide more active sites than monometallic catalysts for multistep N₂RR and NO_xRR and are often used to synthesize ammonia as they mimic bimetallic nitrogen reductase (MoFeP). A bio-inspired CuCo bimetallic nanosheet mimicking Cu–nitrogen reductase is formed by electrodeposition of the corresponding metals and this bimetallic catalyst can electroreduce NO_x⁻ to NH₃.²³⁰ The two active metal centres in the catalyst have separate roles—the Cu centre facilitates the adsorption of NO_x⁻ while the Co centres participate in the donation of electrons and protons. DFT calculations show that the adsorption of *NO (the rate-determining step) consumes less energy for CuCo, when compared to Cu and Co metals, separately. The FE for the Cu₅₀Co₅₀ catalyst reaches ∼100% at −0.2 V vs. RHE with a current density of 1035 mA cm⁻². Metal oxides with defects and vacancies can also enhance the catalytic activity of N₂RR and NO_xRR by altering the local coordination environment and electronegativity. Amorphous RuO₂ nanosheets exhibit superior NO₃RR when compared to their crystalline counterparts.²³¹ This superiority is attributed to atomic disorder in the structure of RuO₂, which endows it with numerous oxygen vacancies (V_Os). The rate-determining step is calculated to be the conversion of *NH₂ to *NH₃, which is facilitated on the surface of amorphous RuO₂. The formation of by-products like NH₂OH and H₂ are also suppressed on amorphous RuO₂, which boosts the selectivity of NH₃.

MXenes formed of carbides and nitrides are emerging 2D materials used in catalysis as they have unique properties like high electrical conductance and capacitance.^281,282 The introduction of defects on the surface reduces repulsive electrostatic forces and thus decreases the energy required for the adsorption of N₂ or NO_x⁻.²⁸³ Functionalization of MXenes suppresses competitive HER and improves the catalytic synthesis of ammonia.²³² Functionalized Ti₃C₂ MXenes with terminal oxygen groups formed by mild calcination reveal the active sites for electrocatalytic NO₃RR (Fig. 14G–I).²³³ Small amounts of N₂ and hydrazine by-products are obtained, resulting in moderate selectivity of NH₃ with an FE of 90.4% at −1.7 V vs. RHE. Under electrochemical conditions, the oxygen groups are converted to hydroxyl groups in situ to form interfacial hydrogen bonding, thus accelerating the electrocatalytic NO₃RR process. These surface hydroxyl groups help the adsorption of NO₃⁻ and contribute to hydrogenation, forming NH₃; hence, the –OH group functions as both an active site and a reactant. The adsorption of H on the O atoms also diminishes the HER process, increasing the selectivity of NH₃. Like other 2D materials, heteroatom doping of MXenes can also effectively increase the rate of NO₃RR and enhance the formation of NH₃. Doping of B atom into the lattice of B-Ti₃C₂T_x (T_x depicts the surface terminating group) MXenes can alter their electronic structure and accelerate the NO₃RR kinetics compared to the pristine Ti₃C₂T_x MXenes to produce ammonia with a high current density and at a low working potential.²³⁴ The B dopants facilitate the adsorption and activation of NO₃RR intermediates, reduce the energy barrier and thereby enhance ammonia production.

MBenes, a newly emerging class of 2D layered materials similar to MXenes, are gaining attention as promising electrocatalysts. Unlike MXenes, MBenes lack passivating surface functional groups, which allows the constituent metal and boron atoms to be fully exposed, enhancing their catalytic activity. The combined effects of metal and boron make MBenes attractive catalytic materials. The metal atoms participate in water splitting and the boron atom activates the N-containing species. A FeB₂ MBene synthesized via the reflux method exhibits efficient NO₃RR with an FE of 96.8% at −0.6 V vs. RHE.²³⁵ Fe acts as the *H donor and B atoms act as the *H acceptor, and this tandem action promotes the NO₃RR process. Theoretical studies reveal that the adsorption and activation of NO₃⁻ take place at the B sites rather than Fe, confirming B as the active site for catalysis. Water catalysis occurs at Fe and the *H that is generated is transmitted to B through the hydrogen spillover process for subsequent hydrogenation reactions. Another study screens a series of M₂B₂-type (M = IVB to V transition metals from the periodic table) MBenes for electrocatalytic NORR.²⁸⁴ Among the screened MBenes it has been observed that the Fe₂B₂, Mn₂B₂, and Rh₂B₂ can efficiently convert NO to NH₃ with smaller limiting potentials, whereas Nb₂B₂ and Hf₂B₂ have low limiting potentials for the NO conversion to NH₃. Mechanistic investigations indicate that hydrogenation of *NO to *NOH has lower energy than *HNO; hence, these MBenes have high selectivity for promoting the NORR to NH₃ over competitive HER.

Layered transition metal chalcogenides like MoS₂ have also been actively used as catalysts for N₂ and NO_x⁻ reductions. The positive charge on the Mo atom can polarize and activate the adsorbed reactants via Mo–N interactions. Heteroatom doping and defect engineering techniques have been employed to enhance the interaction between Mo and N species. Layered structured MoS₂ is a well-known catalyst for HER. Substituting S atoms with F introduces strain in the layered structure and compresses the interlayer spacing in the MoS₂ nanosheets.²³⁶ Additionally, F is more electronegative than S and these factors can suppress HER and promote N₂RR. The introduction of defects in the form of dopants like V can remarkably enhance the efficiency of electrocatalytic NO₃RR.²³⁷ V is highly conductive and can alter the electronic structure of the MoS₂ metalloenzyme. The V-MoS₂ electrocatalyst can lower the energy barrier of conversion of NO* to NOH* and also enhance the selectivity of NH₄⁺. Recent works reveal that the metastable phase of MoS₂ is more active for performing catalytic reactions and can efficiently reduce N₂ to NH₃ (Fig. 14J–K).²³⁸ Metastable MoS₂ provides access to partially filled t_2g orbitals, which can simultaneously form σ bonds with N and transfer electrons to N₂. The work compares the 2H (stable), 1T′, and 1T′′′ (metastable) phases of MoS₂ for the catalytic reactions, and the 1T′′′ phase with maximum electron density was found to be the most suitable for N₂RR. Mo–Mo clustering in the metastable phases is responsible for the enhanced electron density in these phases, facilitating activation of N₂ and accelerating N₂RR, which is almost nine times greater than the stable 2H phase. MoS₂ nanosheets deposited on graphite can also produce ammonia from the electroreduction of NO.²⁸⁵ The positively charged Mo-edge sites of MoS₂ promote the adsorption and activation of NO via an “acceptance–donation” mechanism, which promotes NORR to NH₃ and disfavours the binding of protons and coupling of the N−N bond, ruling out the formation of H₂ and N₂ and enhancing the selectivity of NH₃ produced.

5.4. Metal–organic framework (MOF)-based electrocatalysts and heterostructures

Metal–organic frameworks (MOFs), composed of metal ions or clusters linked by organic ligands in a three-dimensional arrangement, are one of the most explored emergent materials in catalysis. Their porous structure, high specific surface area, excellent stability, and presence of active sites offer unique features required for catalysis. For instance, a 2D Cu-based MOF, {[Cu(HL)]·H₂O}_n, (Cu-OUC, H₃L = 5-(2′-carboxylphenoxy)isophthalic acid) exhibits efficient electrocatalytic NO-to-NH₃ conversion.²³⁹ The Cu-OUC MOF activates NO via a two-way charge transfer mechanism of “electron acceptance and donation”, where the *NO formation step is the rate-determining step. Additionally, the Cu-OUC MOF also acts as the cathode for constructing a Zn–NO battery, which besides converting harmful NO into NH₃ also generates electricity. The reaction rates of ammonia formation depend on the type of metal centres present in the MOFs. In Fe₂M trinuclear cluster-based MOFs, where M = Co, Ni, Fe, Zn, the rates of adsorption and activation of nitrate and protons are in the order of Co > Fe > Ni > Zn (Fig. 15A–C).²⁴⁰ The unsaturated metal sites of the trinuclear complex help in the adsorption and reduction of nitrates while the dinitrogen ligand aids in the interfacial transfer of electrons due to the delocalized conjugation between the lone electron pairs of nitrogen and the π electrons of the ligand's aromatic ring. Ammonium sulfate is obtained by the electroreduction of nitrates and can be directly used as fertilizer for plants without further processing. MOFs have also been used as precursors for deriving metal- and carbon-based catalysts, which inherently have higher reactivity and stability when compared to conventional catalysts. Co–Fe/Fe₂O₃ has been derived from Co-doped Fe-MOF-74 by pyrolysis, where Co dopant increases conductivity and tunes the electronic properties such that Co–Fe/Fe₂O₃ shows efficient electrocatalysis.²⁸⁶ The temperature at which pyrolysis is carried out is crucial as experimental evidence suggests that the graphitic carbon produced at 900 °C shows the highest electrocatalytic performance while that produced at 1000 °C shows reduced catalysis due to the collapse of structural properties. The Co center activates NO₃⁻ and tunes the energy of the d orbitals of adjacent Fe atoms, facilitating charge transport; thus, both Co and Fe atoms participate in NO₃RR. Co also restrains HER, thus increasing the selectivity and FE to 99% and 85.2%, respectively. The conversion of NO₃⁻ to NH₃ involves deoxygenation and hydrogenation of nitrates with water via proton-coupled electron transfer. This coupling reaction between nitrates and water can be boosted if a multifunctional interface can be produced by incorporating a second component with the MOF. For instance, when the bimetallic Ni/Co-MOF interface is surface-coated with H-substituted graphdiyne, deoxygenation, and hydration reactions are promoted due to the formation of the multifunctional interface (Fig. 15D–F).²⁴¹ The intersection between the MOF and graphdiyne acts like a bifunctional membrane by facilitating the transport of electrons and reactants to the Ni and Co sites, where the concerted reaction occurs. An integrated-electrocatalysis cell using MOF-based gas diffusion electrodes (GDEs) can electroreduce NO (NORR) and electrooxidize NO (NOOR) simultaneously to produce NH₄NO₃ under dilute concentrations of NO.²⁸⁷ For preparing the MOFs-GDEs, UIO-66, ZIF-8, and ZIF-67 are modified with Cu nanowires for NORR and Ni/NiO for NOOR to enhance the production of NH₄⁺ and NO₃⁻, respectively. The enhanced NO reduction and oxidation is due to the transfer of adsorbed NO from the adsorption layer to the catalyst layer, which is experimentally verified by the enhanced NO mass transfer from gas to electrolyte across the modified electrode. Modifying MOFs without doping or other functionalization can improve fabrication time efficiency and enhance material properties for ammonia production. One such example is Fe metal center-based MOF, which in its thermally activated form can generate ammonia with FE of ∼90% at −1.0 V vs. RHE, which is much higher when compared with the non-activated pristine form of the MOF (Fig. 15G–I).²⁴² The activation of MOFs is expected to provide access to more active Fe sites for the adsorption of nitrates. The thermal activation process frees the pores of the MOF from trapped components and exposes the high-valent Fe(III) sites for the catalytic reaction. Such activation processes can further open avenues for tailoring the surface structures of desired electrocatalysts like MOFs for energy conversion reactions.

Fig. 15 Metal–organic framework (MOF)-based electrocatalysts for ammonia synthesis. (A) Schematic representation of NO₃⁻ reduction to NH₃ by Fe-based trinuclear cluster metal–organic framework (MOF) Fe₂M-MOF, where M = Fe, Co, Ni, or Zn. (B) SEM image and element mapping of C, Fe, and Co present in Fe₂Co-MOF. (C) NH₃ and NO₂⁻ yields and the corresponding Faradaic efficiencies using Fe₂M-MOFs at a potential of −1.1 V vs. RHE. Reproduced with permission from ref. 240. Copyright 2023, Wiley-VCH. (D) Schematic representation of Ni/Co-MOFs (NiCoBDC) with hydrogen-substituted graphdiyne (HsGDY) nanowire array for electrochemical NO₃⁻ reduction to NH₃. (E) SEM images of NiCoBDC@HsGDY nanoarray at different magnifications. (F) NH₃ yields at different potentials using NiCoBDC, NiCoBDC@HsGDY, CoBDC@HsGDY, and NiBDC@HsGDY. Reproduced with permission from ref. 241. Copyright 2023, American Chemical Society. (G) Schematic representation of the thermal activation of the Fe-MOF from pristine to activated form. (H) SEM images of activated Fe-MOF at different magnifications. (I) Comparison of Faradaic efficiencies for NO₃⁻ reduction between pristine and activated Fe-MOFs at different potentials. Reproduced with permission from ref. 242. Copyright 2023, American Chemical Society.

5.5. 3D-printed electrodes based on low-dimensional materials

To design active and stable electrocatalysts for ammonia production, researchers have focused on designing 3D-printed electrodes. Previously, 3D-printed electrodes have been employed in catalytic reactions like carbon dioxide reductions,¹⁸ oxygen evolution,²⁸⁸ and hydrogen evolution²⁸⁹ reactions; however, the use of 3D electrodes in N₂RR and NO_xRR is quite recent and has some advantages over conventionally prepared catalysts. The 3D printing technique enhances the designing of the catalysts and the reaction vessels by automating the prototype and simulations to prepare an optimized catalytic system.⁴⁸ This eliminates the manual time required for optimization for conventional catalysts, boosting the performance of the process. The 3D printing process allows for precise control over the shape, structure, and geometry of substrates, leading to enhanced catalytic performance. The 3D-printed substrates can be post-modified with desired active materials by sintering, sputter coating, electrodeposition, and atomic layer deposition.^290–293 For instance, in a recent study by our group, a Cu electrode is fabricated by fused fabrication filament (FFF) 3D printing (Fig. 16A–I).²⁴³ The 3D-printed electrode is subjected to thermal treatment and sintering to induce conductivity and the metallic phase to the electrode. An electrochemical modification is then induced to the electrode followed by NaHCO₃ treatment for forming nanostructures on the surface of the electrode with exposed (100) facets. The main composition of the electrode responsible for electrocatalytic NO₃RR has been observed to be Cu/Cu₂O. These 3D-printed Cu electrodes can perform efficient NO₃⁻-to-NH₃ conversion with FE 96.5% at −0.92 V vs. RHE with 95.6% selectivity for NH₃. Another work from our group compares the electrocatalytic properties of two 3D-printed electrodes: (i) 0-dimensional porous carbon black and (ii) 1-dimensional porous carbon nanotubes (Fig. 16J–M).²⁴⁴ The 0D carbon black 3D-printed electrode is electrocatalytically inert while the 1D carbon nanotubes actively participate toward NO₃RR. The defects and metallic impurities (TiO₂, Fe₃O₄) in 1D CNTs can impact the electrocatalytic reactions. However, the exact content and distribution of the impurities and defects are not quantifiable in industrial catalysts; hence, proper control over electrocatalysis is not always possible over industrial catalysts. To achieve better control, an ultra-thin layer of manganese oxide is deposited onto the 1D carbon nanotubes using the atomic layer deposition technique. This fine-tunes the surface properties of the catalysts and enhances their electrocatalytic performance. The carbon nanotubes are conductive and manganese oxide has electrocatalytic properties, and the synergistic effect from these two components makes the 3D-printed 1D CNT@MnO_x efficient toward NO₃RR. Surface-modified 3D-printed carbon substrates with atomic layer deposition (ALD) of TiO₂ can function as bifunctional platforms for electrocatalytic nitrite oxidation reaction (NO₂OR) and nitrite reduction reaction (NO₂RR).²⁴⁵ This 3D-printed heterostructure possesses the intrinsic surface properties of carbon nanotubes as well as Ti-dominated metallic impurities, and the formation of interfaces between the conductive carbon and ALD-coated TiO₂ enhances its electrocatalytic properties toward both NO₂OR and NO₂RR. The electrocatalytic properties of the TiO₂-coated 1D carbon electrode are associated with the ALD-coated TiO₂ layer, which can be tuned by modulating the number of ALD cycles. The 100-TiO₂ electrode, formed by non-continuous TiO₂ deposition on the 1D carbon framework is endowed with abundant carbon/TiO₂ interfaces. These interfaces play a vital function in the electrocatalytic reactions and enhance the formation of NH₄⁺ and NO₃⁻ from electro-reduction and oxidation, respectively.

Fig. 16 3D-printed electrodes designed with low-dimensional materials for ammonia synthesis. (A) Schematic representation of the electrocatalytic cell for NO₃⁻ reduction to NH₃ using a 3D-printed Cu electrode. (B) SEM image and (C) Cu elemental mapping of the electrode. (D) Schematic representation of the 3D-printed electrode. (E) Pictorial representation of the dimensions of the electrodes, where the length and width are 10 mm each, and the height is 1 mm. (F) Sintering of the electrodes. (G) 3D-printed Cu electrode before (left) and after (right) acid treatment. (H) Faradaic efficiencies of NH₃ and NO₂⁻ at different potentials using 3D-printed Cu electrodes. (I) NH₃ yields at different potentials using 3D-printed Cu electrodes. Reproduced with permission from ref. 243. Copyright 2023, American Chemical Society. (J) Schematic representation of NO₃⁻ reduction to NH₃ using 1-D@MnO_x fabricated by atomic layer deposition of MnO_x on a 3D-printed 1D carbon framework. (K) SEM image of 1D carbon framework. (L) HAADF-STEM image of 1D@500-MnO_x. (M) NH₃ yields at different potentials using 3D-printed 1D carbon and 1D@500-MnO_x electrodes. Reproduced with permission from ref. 244. Copyright 2023, Elsevier.

6. Low-dimensional photoelectrocatalysts for ammonia synthesis

Photoelectrocatalysts combine the dual benefits of both photocatalysts and electrocatalysts. Photocatalysts provide light energy for catalysis, and electrocatalysts offer the advantage of the electric bias. In conventional photocatalysts, the efficiency is sometimes limited by the recombination of the charge carriers. In photoelectrocatalysis, the recombination is paused by applying an external potential greater than the onset potential, which draws out charge carriers from the photoelectrode. N₂ and NO_x⁻ act as effective feedstock for the production of ammonia. Over the last few years, a diverse class of electrocatalysts has been studied for NH₃ generation. However, many of these electrocatalysts face a significant drawback: they require high overpotential to achieve considerable Faradaic efficiency (FE) for ammonia production. Photoelectrocatalysis can reduce this potential; the photoelectrodes can harness the light energy to generate a photovoltage and compensate for the reducing potential. The reduction potential for NO₃RR is near that of water splitting. Thus, from a thermodynamic viewpoint, the catalysts capable of photoelectrochemical water splitting are also prospective catalysts for NO₃RR. But kinetically, NO₃RR is more challenging and energy demanding as it requires 8e⁻s for complete conversion of NO₃⁻ to NH₃; otherwise, lower-reduced products like NO₂⁻, NO, N₂, and NH₂OH might be formed. Modification of the photocathode with proper catalysts and cocatalysts leads to active NO₃RR and selective production of NH₃. Herein lies the advantages of photoelectrocatalysts. The following subsections will discuss the design of emerging low-dimensional photoelectrocatalysts developed over recent years for ammonia generation (Fig. 17). The design of photoelectrocatalysts generally involves multi-step reactions for heterogenization with multiple-component photocatalysts and electrocatalysts. As the photoelectrocatalysts are composed of photocatalysts and electrocatalysts, the synthetic strategies resemble the photo- and electrocatalysts mentioned in Sections 4 and 5 of this review. The synthesis and modification techniques of some selected photoelectrocatalysts are listed in Table 5.

Fig. 17 Low-dimensional photoelectrocatalysts (PECs) for ammonia synthesis. Classification of photoelectrocatalysts into 0D, 1D, and 2D, and schematic representations of few selected photoelectrocatalytic LDMs.

Table 5 Synthetic and modification techniques for selected ammonia-generating low-dimensional photoelectrocatalysts

Classification	Materials	Photoelectrocatalysts	Synthesis/Modification	Ref.
0D LDMs and heterostructures	Metallic nanoparticles (NPs)	Au NPs-WS2@RGO	Hydrothermal, microwave methods, chemical reduction	294
		Ag NPs-black Si	Etching, deposition methods	51
		NiO–Au NPs-TiO2	Chemical reduction, thin film preparation, deposition methods	295
	Quantum dots (QDs)	Bi2S3 QDs-MoS2	In situ growth of QDs, hydrothermal and solvothermal methods	296
	Single-atoms (SAs)	Ru SAs–Cu2O	SA embedding by simple mixing	181
1D LDMs and heterostructures	Metal and metal oxide-based heterostructures	MoS2 nanoflakes-La2Zr2O7 nanofibers	Electrospinning, hydrothermal methods	87
		Ni–MoS2/Si nanowires	Ni-Doping, metal-assisted chemical etching, hydrothermal, cast-coating methods	88
		Si-α-Fe2O3 nanorods	Hydrothermal, thermal annealing methods	297
		B–Bi nanorolls	B-Doping, chemical reduction	298
2D LDMs and heterostructures	Metal oxide-based heterostructures	TiOx–CdS–Cu2ZnSnS4	Defect engineering of cocatalysts, chemical bath deposition, spray coating	299
	Carbonaceous materials	CoTiO3 nanorods/N-rGO nanosheets	Interface engineering, N-doping, p–n junction, reflux method	300
		CuPc–CeO2	Defect engineering, thin film fabrication, chemical deposition	301

---

6.1. 0D photoelectrocatalysts and heterostructures

Photoelectrocatalysts for N₂RR and NO_xRR toward ammonia synthesis encounter similar challenges like photocatalysts and electrocatalysts. The design of stable and selective photoelectrocatalysts for ammonia production using various strategies with improved efficiency is discussed here. Narrow-band gap materials like CuO, Cu₂O, and black P are mostly used for photocathodes.^49,302 Sometimes, photoelectrocatalysts comprise heterostructures of different low-dimensional materials. Plasmonic nanoparticles like Au loaded on 2D WO₃@rGO enhances photoelectrocatalytic performance (Fig. 18A–C).²⁹⁴ The excellent catalytic properties of AuWO₃@rGO for N₂RR are attributed to its electronic conductivity, porous structure, and high surface area. When low-band gap black silicon is integrated with plasmonic silver (Ag) nanoparticles, the hot electron generation and plasmonic resonant energy transfer properties of Ag, combined with the light absorption capabilities of black silicon, significantly enhance photoelectrocatalytic N₂RR.⁵¹ The Ag nanoparticles also protect the black silicon from oxidation. The plasmon hot electrons generated from plasmonic Au nanoparticles also activate the NO₃RR processes. In a typical NiO/Au plasmon/TiO₂ photosystem, NO₃⁻ are reduced to ammonia by the plasmon hot electrons at room temperature and neutral pH.²⁹⁵ These hot electrons also suppress H₂ formation and the hot holes are converted to atmospheric O₂. The plasmon electrons enhance the electrochemical process by achieving optimal efficiency at lower potentials. The lower working potentials confirm the suppression of competing HER, overcoming one of the major drawbacks of electrocatalytic NO₃RR. Quantum dots (QDs) also exhibit excellent photoelectrocatalytic properties, e.g., Bi₂S₃ QDs grown in situ over MoS₂ nanoflowers (Fig. 18D–G).²⁹⁶ Upon photoexcitation with visible light, interfacial charge transfer occurs from the QDs to MoS₂. This process results in the separation of photogenerated charge carriers, leading to a higher concentration of photogenerated electrons on the MoS₂ side. The electron-rich MoS₂ then actively participates in the catalytic reduction of N₂ to NH₃. Single-atoms (SAs) like Ru can also be effective for designing photoelectrocatalysts. Using Ru SAs embedded Cu₂O as photocathode and TiO₂ as photoanode in two-chamber cells can function as a dual system for ammonia synthesis and degradation of bisphenol A (Fig. 18H–K).¹⁸¹ The electrons supplied by the TiO₂ photoanode can reduce N₂ to NH₃ at the photocathode and, simultaneously, the bisphenol A can be degraded at the photoanode by the ˙OH radicals. This technique is highly prospective in water treatment and energy generation, and further bifunctional systems like this will be discussed in Section 7.1.

Fig. 18 Zero-dimensional (0D) photoelectrocatalysts for ammonia synthesis. (A) Schematic representation of the 0D plasmonic Au nanoparticles anchored on 2D WS₂@rGO. (B) HRTEM image of Au–WS₂@rGO; zoomed-in images show d-spacing corresponding to (111) and (002) planes of metallic Au and hexagonal phase of WS₂, respectively. (C) NH₃ yields and corresponding Faradaic efficiencies using WS₂, Au–WS₂, and Au–WS₂@rGO at a potential of −0.4 V vs. RHE. Reproduced with permission from ref. 294. Copyright 2023, Elsevier. (D) Schematic representation of Bi₂S₃ quantum dots (QDs) grown over MoS₂ nanoflowers. (E) TEM image of MoS₂ nanoflower. (F) TEM image of Bi₂S₃ QDs marked by red circles, grown over MoS₂. (G) NH₃ yields and the corresponding Faradaic efficiencies using different loadings of Bi₂S₃:MoS₂. Reproduced with permission from ref. 296. Copyright 2022, Elsevier. (H) Schematic representation of the photoelectrochemically driven N₂ reduction to NH₃, where TiO₂ is the photoanode and Ru single-atoms decorated over Cu₂O is the photocathode. (I) SEM image of TiO₂ photoanode. (J) HRTEM image of Cu₂O/Ru photocathode, where the Ru single-atoms are denoted by red circles. (K) NH₃ yields and the corresponding Faradaic efficiencies using different photoelectrocatalysts. Reproduced with permission from ref. 181. Copyright 2022, Elsevier.

6.2. 1D photoelectrocatalysts and heterostructures

Creating hybrid heterostructures and leveraging the interactions of the components stands out as one of the most effective strategies in photoelectrocatalysis. Incorporating MoS₂ nanoflakes on 1D La₂Zr₂O₇ nanofibers creates a heterostructure that combines their individual properties for effective photoelectrochemical ammonia synthesis.⁸⁷ MoS₂ offers the advantages of semiconductors while La₂Zr₂O₇ introduces interfacial oxygen vacancies, and the combined effects result in an efficient photocathode. The photogenerated electrons from MoS₂ are transferred to the vacant oxygen sites of La₂Zr₂O₇, which participates in N₂ reduction. The use of doped materials is another effective strategy for designing photoelectrocatalysts. Ni-doped MoS₂ coated on silicon nanowires acts as an efficient photocathode for ammonia production in porous water (Fig. 19A–E).⁸⁸ Combining two materials with limited catalytic properties into a heterostructure can enhance their catalytic performance by creating a unique interface between the components. Si nanowires and MoS₂ individually exhibit lower conductance. But when MoS₂ is doped with Ni and coupled with Si nanowires to form a heterostructure 1D–2D type electrode, the electron transfer rate is accelerated and so is the efficiency of photoelectrochemical N₂RR. Porous water has a high solubility of N₂, which removes the solubility constraint. MoS₂ and Si nanowires are low-band gap materials that form a type I heterojunction, and doping with Ni provides additional active sites and increases the motility of charge carriers; and all these factors together enhance N₂RR. Doping can also benefit the design of photoanodes. For instance, Mo-doped BiVO₄ photoanodes can transfer electrons from water to the Si photovoltaic-wired hematite-based photocathode (Fig. 19F–I).²⁹⁷ Upon irradiation, the photocathodes can reduce N₂ to NH₃. The process is coupled with H₂O₂-dependent oxyfunctionalization, which produces valuable materials simultaneously at the photocathode and photoanode. The doping of B in 1D Bi nanorolls photocathode also facilitates N₂RR by decreasing the energy barrier of the step N₂ → *NNH (Fig. 19J–M).²⁹⁸ The surface curvature of the nanorolls improves the adsorption of N₂. Upon irradiation, the TiO₂ photoanodes supply the necessary photogenerated electrons to the B-doped Bi photocathode N₂RR. Ordered Si nanowires, decorated with Au particles as the designed photocathode material, can operate at positive potential (0.2 V vs. RHE) using solar energy.⁸⁵ The hexagonal, ordered nanowire array of p-type semiconductor, Si allows efficient mass transport and reduces the recombination of charges. The Au particles act as the cocatalyst and aid in the selective formation of NH₃ with an FE of 95.6%.

Fig. 19 One-dimensional (1D) photoelectrocatalysts for ammonia synthesis. (A) Schematic representation of the fabrication of Ni-doped MoS₂/Si nanowires (Ni–MoS₂/Si NWs) photocathode. (B) SEM image of 1D Si nanowires. (C) TEM image of MoS₂. (D) TEM image of Ni–MoS₂. (E) NH₃ yields obtained using (a) Si nanowires, (b) MoS₂/Si NW, (c) Ni–MoS₂/Si NW, and (d) Ni–MoS₂/Si NW in porous coordinated polymer (PCP) at a potential of 0.25 V vs. RHE. Reproduced with permission from ref. 88. Copyright 2023, American Chemical Society. (F) Schematic illustration of Mo:BiVO₄ photoanode oxidizing H₂O to produce H₂O₂in situ, which activates peroxygenase for enantioselective oxyfunctionalization reactions. (G) Schematic illustration of Si-wired α-Fe₂O₃ photocathode for paired N₂ reduction to NH₃. (H) SEM image of α-Fe₂O₃. (I) NH₃ yields obtained using the above illustrated photoelectrochemical setup at different potentials. Reproduced with permission from ref. 297. Copyright 2023, Elsevier. (J) Schematic representation of the fabrication of 1D Bi-doped Bi nanorolls from 2D BiOBr nanosheet precursors. (K) TEM image and (L) HRTEM image of B-doped Bi nanorolls. (M) NH₃ yields and corresponding Faradaic efficiencies using B-doped Bi nanorolls. Reproduced with permission from ref. 298. Copyright 2021, Elsevier.

6.3. 2D photoelectrocatalysts and heterostructures

2D black phosphorus (P) nanosheets offer significant properties that are beneficial for N₂RR. Black P is also a direct band gap material with a wide range of light absorption properties. It has abundant surface and edge sites and weak hydrogen adsorption properties. 2D black P photocathodes can produce N₂ effectively and have excellent stability for up to 12 hours (Fig. 20A–D).³⁰² The formation of a strong interfacial junction between p-type perovskite oxide CoTiO₃ and N-doped reduced graphene oxide (N-rGO) can also optimize N₂RR.³⁰⁰ The interfacial heterojunction can promote the electron transfer encapsulated between the atomic layers on N-rGO by reducing the diffusion path length and the recombination of charge carriers. Additionally, the heterojunction broadens the wavelength window, enabling more efficient light harvesting. Defects and vacancies also boost the photoelectrocatalytic reactions. In an organic–inorganic hybrid like Cu phthalocyanine (CuPc)/CeO₂ heterostructure, the Ce³⁺/Ce⁴⁺ pairs modify the oxygen vacancies, which play a predominant role in the adsorption and activation of nitrates.³⁰¹ The 2D macromolecular CuPc has similar properties like p-type semiconductors, which are active under UV-visible light and can efficiently convert solar charge carriers. Also, the d orbitals of Cu match the energy level of LUMO π* of NO₃⁻, allowing charge transfer between Cu and NO₃⁻. However, CuPc lacks an adequate number of active sites on its surface, necessitating its integration with other catalysts such as CeO₂³⁰¹ or BiVO₄³⁰³ to perform NO₃RR. CeO₂ has Ce³⁺ and Ce⁴⁺ oxidation states and abundant oxygen vacancies, which can entrap photogenerated charge carriers and facilitate the adsorption of nitrates. BiVO₄ is a p-type semiconductor with a large band gap, which limits its photocatalytic performance, and coupling with CuPc makes BiVO₄/CuPc an effective catalyst. Defect-engineered materials like TiO_x with oxygen vacancies (V_Os) are effective cocatalysts for NO₃RR. For the construction of the photocathode, a p–n junction is fabricated by coating Cu₂ZnSn₄ with the CdS layer (Fig. 20E–H).²⁹⁹ Cu₂ZnSn₄ has excellent light harvesting properties, and CdS has efficient charge separation properties. TiO_x cocatalyst with Ti³⁺/V_O sites allows better adsorption of NO₃⁻ and *NO₂ intermediate on the photocathode. The efficient electron transfer reduces the work function and extends the carrier lifetime, which results in an FE of 89.1% for ammonia at 0.1 V vs. RHE. The benefit of the photoelectrochemical strategies mentioned above is that they couple photocatalysis and electrocatalysis, and the synergistic effect can enhance the charge separation, reduction capability of electrons, equilibrium of electron distribution, and also reduce the probability of oxidation of ammonia produced.³⁰⁴

Fig. 20 Two-dimensional (2D) photoelectrocatalysts for ammonia synthesis. (A) Schematic representation of the black phosphorus (BP) electrode fabrication using layer-by-layer assembly of exfoliated ultra-thin BP nanosheets. (B) Cross-section SEM image of the fabricated BP electrode. (C) NH₃ yields and the corresponding Faradaic efficiencies using BP electrodes at different potentials. (D) Schematic representation of the photoelectrochemical N₂ reduction to NH₃ by the BP electrodes fabricated on ITO. Reproduced with permission from ref. 302. Copyright 2020, Wiley-VCH. (E) Schematic representation of the step-by-step synthesis of TiO_x/CdS/CZTS electrodes. (F) Cross-section STEM image of TiO_x-250/CdS/CZTS electrodes, synthesized at 250 °C. (G) NH₃ yields and the corresponding Faradaic efficiencies using TiO_x-250/CdS/CZTS electrodes at different potentials. (H) Schematic representation of the mechanism of photoelectrochemical NO₃⁻ reduction to NH₃ by the TiO_x/CdS/CZTS electrodes. Reproduced with permission from ref. 299. Copyright 2022, Wiley-VCH.

7. Applications

Besides ammonia production by NO_xRR and N₂RR, these reactions are also important in many aspects. For instance, nitrates and nitrites are common pollutants found in wastewater and their reduction can help in environmental remediation. NO_xRR and N₂RR, when coupled with CO₂RR, C–N coupling can favour the formation of urea. Various relevant and important oxidation reactions can be coupled with these reduction reactions. All these aspects are highlighted in the following sections.

7.1. Water purification

Nitrates and nitrites, the precursors of ammonia synthesis, are common sources of pollutants in surface and groundwater. Hence, wastewater containing nitrates and nitrites can act as N-feedstock for ammonia production, helping in wastewater treatment and producing sustainable energy resources. The major sources of nitrate contamination in water originate from fertilizers used in agriculture, industrial waste (including ammonia-producing industries using the Haber–Bosch process), stormwater runoff from metropolitan areas, and sewage water. The concentration of nitrates in wastewater depends on the source of pollution: 1.95 M in low-level nuclear wastewater, 41.6 mM in industrial wastewater, and 7.4 mM in textile wastewater.³⁰⁵ Nitrates can lead to eutrophication and nitrites produced from the reduction of nitrates pose serious health risks to humans and the animal kingdom. Nitrates and their metabolites can also cause blue baby syndrome, hypertension, and cancers. Hence, the U.S. Environmental Protection Agency has set a limit of 10 mg L⁻¹ of N–NO₃⁻ and 1 mg L⁻¹ of N–NO₂⁻ in drinking water.³⁰⁶ Conventional denitrification approaches reduce nitrates present in water to gaseous nitrogen, avoiding side products like nitrous oxide or ammonia.^307–310 However, recent progress in this aspect highlights the benefits of the reduction of nitrates to ammonia as an important precursor to different value-added products, e.g., fertilizer and a prospective hydrogen storage fuel.³¹¹ Here, we will discuss a few state-of-the-art catalytic materials to demonstrate the feasibility of the process.

7.1.1. 0D nanostructured and heterostructured catalysts. Self-activated 0D Ni(OH)₂ nanoparticles deposited on Ni substrate can selectively reduce NO₃⁻ to NH₄⁺ in the laboratory as well as can be scaled up to treat wastewater from a chemical factory, Hunan Sinopec Catalyst Co., Ltd (China) (Fig. 21A–D).³¹² The pilot-scale wastewater treatment system reported in this work consists of a sedimentation tank, electrocatalytic reactor, and storage tank. A sedimentation tank is used to separate the solids suspended in wastewater. The electrocatalytic reactor is set up with 26 electrodes connected to a DC power supply and can purify 500 liters of wastewater. The reactor can work in fed-batch and continuous flow modes for NO₃⁻–N concentration 241–2527 mg L⁻¹, and the selectivity of NH₄⁺–N obtained is as high as 92.5%. The reduced ammonia can be separated from water using air-stripping methods. Photoelectrocatalytic 0D Co can also produce ammonia in simulated wastewater containing nitrates, nitrites, phosphates, sulfates, carbonates, and bicarbonates.³¹³ However, the FE for ammonia in simulated wastewater is only 12%, whereas the maximum FE reported is 92% in water containing only NO₃⁻. Ni single-atoms decorated on oxygen-vacant WO₃ can efficiently produce ammonia in simulated wastewater containing NO₃⁻ and other organic pollutants like phenol, benzyl alcohol, and formaldehyde (Fig. 21E–H).⁴¹ These organic pollutants act as hole-sacrificial agents and promote NO₃⁻ reduction. Subnanometric metal oxides on TiO₂ can also function as efficient photocatalysts for ammonia production in simulated wastewater.¹⁵⁴ CuO nanoparticles on Cu foam electrodes can directly produce ammonium-based fertilizer when simulated wastewater containing NO₃⁻ and Cl⁻ is fed into the constructed flow cell containing cathodic and anodic reaction chambers, an electrified PTFE membrane, and a trap chamber.⁵³ The performance of the electrochemical flow cell has also been monitored using industrial wastewater from Shandong Keyuan Pharmaceutical Co. Ltd, containing NO₃⁻, Cl⁻, Na⁺, K⁺, SO₄²⁻, and chemical oxygen demand (COD). Production of (NH₄)₂SO₄ from industrial wastewater is a proof-of-concept for electrochemical upcycling of NO₃⁻ to form valuable fertilizers.

Fig. 21 Low-dimensional catalysts for ammonia synthesis coupled with water purification. (A) Schematic representation of electrochemical reactor for wastewater purification containing NO₃⁻ using self-activated 0D Ni(OH)₂ particles formed on Ni substrate, Ni(OH)₂@Ni cathode. (B) TEM image of Ni(OH)₂@Ni cathode. (C) HRTEM image of the Ni(OH)₂@Ni cathode showing the lattice fringes with d-spacing of 0.233 nm and FFT (inset) corresponding to the (101) plane of Ni(OH₂). (D) Effects of NO₃⁻-N concentration on the conversion efficiency, Faradaic efficiency of NO₃⁻ reduction, and NH₄⁺–N selectivity under the fed-batch conditions. Reproduced with permission from ref. 312. Copyright 2021, American Chemical Society. (E) Schematic representation of the photocatalytic synthesis of value-added ammonia from nitrate-containing wastewater using Ni single-atoms-decorated defective WO₃. (F) SEM image of Ni/H_xWO_3−y catalyst. (G) HRTEM image showing the Ni single-atoms marked by white circles. (H) NH₃ yields using Ni/H_xWO_3−y catalyst and upon addition of different organic, cationic, and anionic pollutants. Reproduced with permission from ref. 41. Copyright 2024, Elsevier. (I) Schematic representation of the electrochemical NO₃⁻ to NH₄⁺ by natural hematite electrode. (J) SEM image of α-Fe₂O₃ deposited on Ni foam at different magnifications. (K) Comparative study of NO₃⁻ removal, NO₂⁻ selectivity, NH₄⁺ selectivity, and recovery of α-Fe₂O₃ and γ-Fe₂O₃. Reproduced with permission from ref. 314. Copyright 2024, Elsevier.

7.1.2. 1D nanostructured and heterostructured catalysts. Ru–Cu nanowire catalysts can reduce the nitrate concentration in water from an industrial level of 2000 ppm to a drinkable level of <50 ppm and efficiently produce ammonia.²⁰⁸ For separating ammonia, effluent with a high vapour pressure of ammonia is subjected to air stripping, and 99.7% of ammonia is successfully stripped out. For air stripping, in this work, the effluent is sealed in a flask with Argon for 24 hours. This stripped-out ammonia vapour is either trapped in HCl to form NH₄Cl or condensed to form ammonia water. The performance of catalytic materials for NO₃RR to NH₃ has also been demonstrated in simulated wastewater to extend the practical applicability. In another work published by the same group, similar Ru–Cu nanowires with slight modifications have been utilized for electrocatalytic NO₃RR from industrial wastewater containing 2000 ppm NO₃⁻.³¹⁵ In this work, they have developed a membrane electrode assembly (MEA) and a reactor with a porous solid electrolyte (PSE) and compared their electrocatalytic NO₃RR performance. The PSE reactor, which has a cation shielding effect, is more effective due to enhanced concentration of cations, and can produce purified water and ammonia simultaneously from electrocatalytic NO₃RR. Another work highlights the fabrication of a paired electrolysis platform, where defect-engineered 1D TiO₂ nanotubes with Co doping and oxygen vacancies act as the cathode.³¹⁶ The doped and vacant sites can synergistically promote electron transfer for NO₃RR to NH₃ at the cathode, and the anode simultaneously traps the NH₃ and converts it to (NH₄)₂SO₄, an ammonia-based fertilizer used in agriculture.

7.1.3. 2D nanostructured and heterostructured catalysts. The recovery of produced ammonia from wastewater is always challenging. A few works focused on this challenge and developed electrodes based on 2D materials to address this problem. In this attempt, a membrane-less electrochemical technique has been developed that electrochemically converts nitrate synchronized with ammonia recovery.³¹⁷ The system typically consists of a 3D-printed porous Cu–Ni metallic glass that can effectively electroreduce the nitrates present in wastewater to ammonia. This work successfully converts 70% of nitrate and recovers the ammonia from the reaction mixture as ammonium chloride. Hematite (α-Fe₂O₃) deposited on a Ni foam electrode and coupled with a water-resistant membrane can reportedly remove 97.6% of nitrates from wastewater, and consequent upcycling of these nitrates can yield ammonia (Fig. 21I–K).³¹⁴ 81.6% of this ammonia can be recovered in situ in the form of ammonium sulfates. To check practical applicability, the system is tested on wastewater from a photovoltaics industry containing nitrates and nitrites. After electrocatalysis, 74.8% of total inorganic nitrogen is removed and 77.1% of ammonium is recovered. The entire process leads to an energy consumption rate of 62.2 kW h per kg of NH₃.

7.2. Urea synthesis

Urea is the most important nitrogen-based fertilizer used in crop production. The current industrial process for urea synthesis uses harsh reaction conditions and leaves a huge carbon footprint. This includes (i) the synthesis of ammonia by the Haber–Bosch process and (ii) the combination of ammonia and CO₂ at a high temperature (150–200 °C) and pressure (150–200 bar).³¹⁸ Hence, a “greener” and more sustainable way to produce ammonia under ambient conditions is required, which would also aid in realizing carbon/nitrogen neutrality in the environment. Co-reduction of CO₂ and nitrogenous species like NO₃⁻, NO₂⁻, and N₂via electrochemical or photoelectrochemical pathways can lead to the sustainable production of urea by C–N coupling, thus maintaining carbon neutrality and the nitrogen cycle.^318,319 However, slow adsorption of substrates, multiple reaction steps, and competitive side reactions impede the catalytic reactions toward urea synthesis.³²⁰ Hence, to improve catalytic processes, the design of catalysts is crucial for the co-reduction process. A diverse range of catalysts, such as single-atoms, bimetallic, dual site, oxygen-vacant, and non-metallic materials, have been used in recent years for the co-reduction of nitrogenous species and CO₂ to form urea.

7.2.1. 0D nanostructured and heterostructured catalysts. Cu single-atoms with different coordination environments have been widely used for urea production. A comparative study with Cu–N–C coordination structures revealed that Cu–N₄ and Cu–N_4−x–C_x are more active sites for NO₃RR and CO₂RR when compared to Cu–N₃–C₁ and Cu–N₂–C₂.³²¹ The formation of *COOH and *NH intermediates are energetically more favourable on Cu–N₄ sites while the energies of other intermediates like *CO and *NH₂ are similar on all Cu–N–C sites. Cu single-atoms (SAs) can be electrochemically reconstituted to form Cu₄ clusters that act as active sites for C–N coupling and urea synthesis.⁵⁸ The Cu₄ clusters can be reversibly transformed to Cu SAs again by switching to open circuit potential and this makes the Cu₁–CeO₂ catalysts structurally and electrochemically stable. Cu SAs can also be used to form bimetallic sites Cu/Ti by coupling with TiO₂.³²² The benefit of using bimetallic sites is that the Cu site can effectively adsorb CO₂ and reduce *CO₂ to *CO; while the Ti sites can adsorb NO₃⁻ and reduce *NO₃⁻ to *NH₂ by photoelectrocatalysis. Due to the adjacent positioning of Cu and Ti, the *CO and *NH₂ intermediates can easily undergo C–N coupling to form urea. Bimetallic Cu/Zn sites can operate by relay catalysis mechanism and stabilize the intermediates *CO₂NO₂ (key intermediate for C–N coupling) on Zn sites and *COOHNH₂ (formed on protonation) on Cu sites preferentially.³²³ Urea formation with an FE of 75% is reported for wastewater-level nitrate concentrations. 0.28 kg of CO₂ emissions per kg of urea is calculated for the Cu/Zn catalyst, which is far less compared to 1.8 kg of CO₂ for conventional methods of urea synthesis. The study also compares the urea formation for bimetallic hybrid catalysts and their corresponding single-component catalysts. The reaction energy for C–N bond formation and the consecutive protonation step is higher for single-component catalysts, which leads to lower FE for urea. Electrocatalysts like CuWO₄ are also endowed with bimetallic sites that combine the advantages of WO₃ and CuO_x, and are capable of simultaneous reduction of NO₃⁻ and CO₂.⁵⁹ The high-valence W centers in WO₃ can stabilize *NO₂ intermediates, however, they are unable to reduce CO₂ to *CO at low overpotentials. Contrarily, Cu centers are well known for reducing CO₂ to *CO, but a continuous arrangement of Cu assists in C–C coupling reactions, forming a series of by-products. For the formation of urea, C–C coupling has to be suppressed, and C–N coupling reactions have to be promoted, which is achieved by CuWO₄. The alternating bimetallic reduction sites, Cu and W, can reduce the raw precursors, form reduced and hydrogenated intermediates, initiate C–N coupling, and increase the selectivity of urea produced. Besides the Cu and bimetallic catalysts, other single metallic catalysts based on Fe and Pt can also produce urea. In one such work, the utilization of 0D Fe₂O₃ nanoparticles has been highlighted for urea synthesis from the co-reduction of NO₃⁻ and CO₂ simultaneously.³²⁴ Here, minuscule γ-Fe₂O₃ nanoparticles (<2 nm), confined within the pores of a conductive metal–organic framework (MOF), Ni-HITP MOF can electrocatalytically produce urea, and the catalyst shows stability for 150 hours. Instead of having excellent catalytic properties, metal oxide nanoparticles suffer from a few drawbacks like instability, aggregation, and uneven size distribution. Encapsulation within the pores of Ni-HITP MOF offers better stability, uniform size, confinement, and conductivity to the γ-Fe₂O₃ nanoparticles for electrocatalysis. The Fe(III) centers can simultaneously reduce NO₃⁻ and CO₂ to *NO₂ and *COOH, the key intermediates. Further, these catalytically active adjacent Fe(III) centers can induce C–N coupling between *NH₂ and *COOH to produce *CONH₂, the key intermediate for urea formation. Pt nanostructures can also produce urea electrocatalytically from CO and NH₃.³²⁵ In this work, CO has been used instead of CO₂. In the first step, Pt triggers the oxidative coupling between CO and NH₃ to produce cyanate via proton-coupled electron transfer. In the next step, the produced cyanate reacts with ammonium via the Wöhler reaction to yield urea, the desired product. The reactions take place on the exposed Pt(100) and Pt(111) facets of the Pt/C catalyst. Oxidation of NH₃ and CO are energetically feasible on both Pt(111) and Pt(100) surfaces. However, the *NH₃ intermediate is more stable on Pt(111) while the *NH₂ intermediate is more stable on Pt(100) facet. Among the intermediates obtained from CO oxidation, *CO is stable on Pt(100); however, *CO₂ and *COOH are unstable and react with OH⁻ to form CO₂ and H₂CO₃, respectively. Hence, on the Pt(100) surface, *NH₂ and CO undergo C–N coupling to produce *OCNH, which is converted to *OCN (cyanate) under alkaline conditions. In the presence of NH₄⁺, this cyanate is converted to urea.

7.2.2. 1D nanostructured and heterostructured catalysts. Instead of single components, hybrid bimetallic catalysts can lower the energy barrier and selectively produce urea in higher amounts. Dual-active Fe sites, namely, amorphous Fe, and Fe₃O₄ on carbon nanotubes (CNTs), also act as efficient electrocatalysts for urea production (Fig. 22A–D).⁶⁰ The amorphous Fe@C and Fe₃O₄ active sites participate in the adsorption and activation of NO₃⁻ and CO₂, respectively, and lower the energy barrier for the formation of urea. Zn foil is an efficient electrode for the electrosynthesis of urea, but the fabrication of Zn nanostructures via typical methods is challenging. Coupling Cu with Zn to form a bimetallic core–shell nanowire structure is a viable alternative (Fig. 22E–H).³²⁶ Additionally, electron transfer from Zn to Cu produces electron-deficient Zn, which favours NO₃RR. The Cu@Zn bimetallic heterostructure thus favours the formation of intermediates and the C–N coupling required for urea synthesis. Non-metallic catalysts like F-doped carbon nanotubes (CNTs) also report excellent urea production by electroreduction of CO₂ and NO₃⁻.³²⁷ The availability of abundant C–F₂ moieties as active sites promotes the formation of *CO and *NH₂ intermediates, and the electronic structure of CNTs ensures excellent conductivity and charge transfer, beneficial for electrocatalysis. Another recent work reports 1D Cu phthalocyanine nanotubes, which have many active sites for electroreduction, namely, Cu center, pyridinic-N1, and pyrrolic-N2 and N3.³²⁸ Among them, urea synthesis is driven by the reduction of N₂ and CO₂ by dual active sites, the pyridinic-N1 site and Cu site to form *NN and *CO, respectively. Here also, the reaction proceeds through the *NCON intermediate, followed by subsequent protonation to form urea. The catalyst surface is modified via oxygen vacancy and facet engineering to boost the catalytic synthesis of urea. For the formation of urea, the C–N coupling reaction between *CO₂ and *NO₂ intermediates is crucial. The early-stage C–N coupling provides a better selectivity of urea when compared to the later-stage coupling between *CO₂ and *NH₂.³²⁹ For instance, defect-engineered 1D CeO₂ nanorods with abundant oxygen vacancies can entrap and insert *NO intermediate into the vacant sites and subject it to subsequent C–N coupling to form *NO–CO.³³⁰ The oxygen-vacant (V_O) CeO₂ nanorod catalysts were compared with the V_O-deficient CeO₂ catalysts to gain conducive insight into the *NO entrapping by the V_O sites. Further, this work indicates that the C–N coupled intermediate formation has a lower energy barrier when compared to the protonation intermediates in the case of pristine CeO₂. The oxygen vacancies can thus help in selective C–N coupling and promote the electrocatalytic synthesis of urea. The C–N coupling step promotes the selectivity of the formation of urea, however, the energy barrier of the protonation of *CO₂NH₂ is high, which is why it is the potential determining step (PDS). The energy barrier of PDS can be reduced by 1D indium oxyhydroxide, i.e., InOOH rods with oxygen vacancies.⁵⁷ The oxygen vacancies result in unsaturated In sites, which decrease the energy barrier of PDS, facilitate the protonation of *CO₂NH₂ to *COOHNH₂, and enhance the catalytic performance of InOOH for producing urea. The catalysts discussed above operate on the co-reduction of NO₃⁻ and CO₂ precursors to form urea. Though nitrates are readily available and have more advantages, some catalysts also report urea synthesis by N₂ and CO₂ co-reduction. Mott–Schottky heterostructures like Bi–BiVO₄ have a space–charge region at the interface due to the transfer of electrons from BiVO₄ to metallic Bi, forming local electrophilic and nucleophilic regions.³³¹ The space–charge region exposes more active sites and promotes adsorption and activation of N₂ and CO₂ at local electrophilic and nucleophilic sites, respectively. Thereafter, the C–N coupling between *N=N* and CO produces the *NCON* intermediate, which acts as a precursor for the formation of urea.

Fig. 22 Low-dimensional catalysts for urea synthesis from co-reduction of NO₃⁻ and CO₂. (A) Schematic representation of urea synthesis by dual active Fe(a)@C and Fe₃O₄ on carbon nanotubes (CNTs). (B) TEM image of Fe(a)@C-Fe₃O₄/CNT. (C) HRTEM image showing the lattice fringes with d-spacings of 0.25 nm and 0.294 nm corresponding to the (113) and (022) planes of Fe₃O₄ nanoparticles, respectively. (D) Urea yields and the corresponding Faradaic efficiencies using Fe(a)@C-Fe₃O₄/CNT at different potentials. Reproduced with permission from ref. 60. Copyright 2023, Wiley-VCH. (E) Schematic representation of the fabrication of core–shell Cu@Zn nanowires and catalytic formation of urea over these nanowires. (F) TEM image of Cu@Zn nanowire. (G) HRTEM image showing the lattice fringes with d-spacings of 0.23 nm and 0.21 nm corresponding to the Zn(100) and Cu(111) planes, respectively. (H) Urea yields at different applied potentials using the Cu@Zn nanowires. Reproduced with permission from ref. 326. Copyright 2022, American Chemical Society. (I) Schematic representation of grain boundary-rich Bi nanosheets reconstructed from Bi₂Se₃ nanosheets, utilized for the synthesis of urea from NO₃⁻ and CO₂. (J) TEM and (K) HRTEM images of grain boundary-rich Bi nanosheets showing lattice fringes with d-spacing of 0.328 nm corresponding to the (012) plane of Bi. (L) Comparison of urea yields by grain boundary-rich Bi, low grain boundary Bi, and bulk Bi at different applied potentials. Reproduced with permission from ref. 332. Copyright 2024, Wiley-VCH.

7.2.3. 2D nanostructured and heterostructured catalysts. 2D catalysts like MXenes, MBenes, and carbon nitrides are well-known catalysts discussed previously in Section 5.2 for electrocatalytic ammonia synthesis. Recent studies indicate that besides reducing N₂ or NO₃⁻, the heterostructures based on these 2D materials can co-reduce CO₂ and function as efficient electrocatalysts for urea synthesis. One of the recent works along this line reports a double transition metal-based Mo₂VC₂ MXene, which can co-reduce N₂ and CO₂ and, followed by C–N coupling, can produce urea.³³³ The selective formation of key intermediates for the C–N coupling is crucial for urea production. The Mo₂VC₂ MXene can selectively convert *CO₂ to *CO, which is one of the key intermediates. This *CO can further react with the intermediates formed from N₂, i.e., *N₂ and H-added intermediates like *NHN, *NHNH, *NH₂N, *NH₂NH, and *NH₂NH₂, via C–N coupling to produce urea. The work also highlights that double-transition metal-based MXenes can effectively function as better electrocatalysts for urea synthesis when compared with single-transition metal-based MXenes. Hence, single-transition metal-based MXenes like Mo₂C can also be functionalized with planar metal clusters like Cu via strong metal–support interactions for enhanced performance of urea electrosynthesis.³³⁴ The active sites on the Cu_n/Mo₂C catalyst enhance the electrocatalytic “one-step” N–C–N coupling process by inserting *CO intermediate into the NH₂*–NH₂* intermediate, promoted by the electrostatic interactions between the N atoms. This “one-step” coupling process offers an energy-efficient and favourable thermodynamic route to urea synthesis compared to the conventional two-step coupling processes. Bimetallic sites like Ru–Pd can also promote the photoelectrosynthesis of urea when encapsulated on 2D–2D heterostructures like WO₃/MXene.³³⁵ The photocathode composed of this 0D–2D–2D heterostructure can simultaneously reduce CO₂ and N₂ and is substantiated by water oxidation at the photoanode. However, the fabrication and engineering of this multi-component catalyst are complex and costly, and the design of simpler and cheaper photo- or electrocatalysts is more favourable for a better understanding of the catalytic mechanism over active sites. Another work reports the utilization of 2D MBenes to electrosynthesize urea from N₂ and CO₂ under ambient conditions.³³⁶ The MBenes with three different metal centers have been used here, namely, Ti₂B₂, Mo₂B₂, and Cr₂B₂. Electrocatalytic urea synthesis takes place at the basal planes of the MBenes. These MBenes can also suppress competitive N₂ reduction to NH₃. Among the three MBenes, the surfaces of Mo₂B₂ and Cr₂B₂ are resistant to oxidation and degradation and offer more reactive surface area, activity, and selectivity for electrocatalysis. Porous carbon nitrides are also another class of widely explored 2D electrocatalysts. These carbon nitrides can also electrosynthesize urea upon structure modification like doping and embedding metallic sites. In this approach, a new N₄B₄ porous carbon nitride with B doping has been synthesized.³³⁷ Upon further substitution doping, M₂B₂ structures have been formed. Fourteen metal (M) atoms have been used and, among them, M = Mn, Cr, and Os were found to be the most active for urea synthesis. The –C–N–C linkages in the carbon nitrides increase the surface area of the pores, which provides more area for electrocatalytic reactive sites. Metal doping increases the conductivity of the catalysts. The boron–boron, metal–boron, and metal–metal bidentate sites enhance the electrocatalytic performance. However, the B–Mn site provides the maximum activation effect on N₂, proving B₂Mn₂ to be the best-performing carbon nitride catalyst in this work. Doping of Fe atoms on InOOH nanosheets has also shown an effective increase in the rate of urea production as compared to bare InOOH nanosheets.³³⁸ The studies indicate that Fe doping increases the number of active sites and optimizes the electronic structure for the adsorption, activation, and hydrogenation of the *CONH₂ key intermediate, thereby enhancing the C–N coupling reaction for urea formation. As discussed previously, the catalyst surface plays a crucial role in the adsorption of substrates, the generation of active species, and, finally, releasing the catalytic products for any catalytic reaction. From this viewpoint, 2D catalysts with more active sites can be engineered from exfoliation of existing 2D materials, aiming toward better catalytic performance. Consequently, the reconstruction of Bi₂Se₃ nanosheets into 2D bismuth (Bi) nanosheets results in a high density of grain boundaries caused by the presence of defects (Fig. 22I–L).³³² The urea yield for the reconstructed Bi nanosheets is eight times that of the Bi₂Se₃ nanosheets. This enhancement in catalytic activity is because the grain boundaries lower the energy barriers for the formation of the key intermediates *CO and *NH₂, and also for the C–N coupling reaction. Hence, it can be said that tuning the structures of low-dimensional materials is always crucial from the aspect of catalysis for obtaining maximum catalytic activity.

7.3. Coupled oxidation reactions

An overall reaction consists of an oxidation half-reaction and a reduction half-reaction. Generally, the reduction reactions of N₂ and NO_x⁻ are accompanied by water-splitting, leading to oxygen evolution reaction (OER), and the formation of active hydrogen.^339–341 The active hydrogen produced from water splitting helps in the protonation process and promotes the formation of ammonia. However, the active hydrogen (H*) produced should be consumed immediately by the nitrogen intermediates, which is difficult as the interaction between two H* forms H₂, giving rise to competitive HER. The active H* can be successfully utilized by catalysts, such as Co(OH)₂ and CoP, where the water dissociation process is facilitated on the Co catalysts’ surface.^342,343 However, the OER reaction is sluggish and has a high thermodynamic potential of 1.23 V vs. RHE. As an outcome, this hampers the N₂RR and NO_xRR and slows down the entire redox process. Hence, efforts have been made to replace the OER with lower-potential oxidation reactions, like the oxidation of urea, hydrazine, methanol, ethanol, benzyl alcohol, glycerol, and PET, which can yield some value-added products. This section will explore different low-dimensional catalysts, focusing on this approach.

7.3.1. 0D nanostructured and heterostructured catalysts. Quantum dots (QDs) with semiconducting properties have tuneable band gaps and thus act as excellent catalysts for multielectron and multiproton processes.^344,345 Upon light irradiation, QDs produce sufficient electrons and protons to perform the 8e⁻ NO₃⁻ reduction to NH₃ as well as any coupled oxidation simultaneously. One such photocatalytic system, glutathione (GSH)-capped 0D CdSe QDs, can convert NO₃⁻ to NH₄⁺ in the presence of an LED light of wavelength 440 nm.³⁴⁶ Under this light irradiation, H˙ and GS˙ radicals are generated from glutathione. The H˙ radicals participate in the formation of the N–H bond of NH₄⁺. The GS˙ radicals provide the holes for value-added oxidative transformation of secondary aryl alcohols to the corresponding ketones. Bifunctional catalysts also play an immense role in coupling oxidation and reduction reactions in an electrochemical cell. One such example of 0D bifunctional catalysts is CuCo₂O₄ spinel oxide-based nanoparticles.³⁴⁷ CuCo₂O₄ nanoparticles, embedded on porous carbon fiber support, can electrocatalytically reduce nitrate and oxidize methanol at the cathode and anode of the cell, respectively. The carbon nanofiber acts as the support material, enhances charge transfer, converts NO₃⁻ to NO₂⁻, and suppresses the competitive HER. The CuCO₂O₄ nanoparticles reduce the adsorbed *NO₂ to NH₃. By leveraging the bifunctional nature of the spinel oxide catalyst and improving the thermodynamic favourability of NO₃RR, the water splitting at the anode has been substituted with the methanol oxidation reaction (MOR). Another advantage of coupling the MOR with NO₃RR is that it produces two value-added products instead of one: ammonia from NO₃RR and formate from MOR. Photoelectrochemical synthesis of ammonia is a green and sustainable technology, however, it suffers from a few bottlenecks like lower solar energy-to-ammonia conversion, and entails an extra bias for nitrate reduction coupled with water oxidation. To achieve bias-free photoelectrochemical ammonia production, researchers have recently designed 0D noble metal-based electrocatalysts that substitute water oxidation with the glycerol oxidation reaction (GOR). This cocatalysis technique has been used previously by CO₂ reduction catalysts to make the catalytic processes more energy efficient.^348–350 Along similar lines, a lead halide, triple cation-based perovskite photoelectrode with a smaller band gap and superior charge transfer properties has been fabricated, followed by the deposition of 0D electrocatalysts.⁵⁶ Nitrate is then reduced to ammonia at the cathode formed by the Ru nanoparticle-deposited titanate nanosheets. This NO₃RR is coupled with the GOR at the anode composed of Pt nanoparticle-deposited titanate nanosheets. These titanate nanosheets are more stable and corrosion-proof than conventional carbon material-based supports. This coupled photoelectrocatalytic system can lead to bias-free generation of ammonia utilizing solar energy sources, and this work considerably paves a lower-energy pathway for ammonia production, which is crucial in current catalytic ammonia synthesis research. Apart from green ammonia, this system also produces value-added products like glyceric acid and lactic acid from glycerol oxidation.

7.3.2. 1D nanostructured and heterostructured catalysts. 1D tungsten phosphide (WP) nanowires can operate as bifunctional catalysts by acting as reducing catalysts at the cathode and oxidizing catalysts at the anode. Exploiting this redox property of WP nanowires, NO₃RR has been coupled with oxygen evolution reaction (OER), hydrazine oxidation reaction (HzOR), and urea oxidation reaction (UOR) and a comparative study of the operating cell potentials of these reactions has been presented in this work (Fig. 23A–D).⁵⁵ The overall potential of NO₃RR-OER requires 1.54 V and NO₃RR-UOR requires 1.38 V, whereas NO₃RR-HzOR occurs at a much lower potential of 0.24 V vs. RHE. Thus, coupling NO₃RR with hydrazine oxidation leads to more energy-efficient ammonia production. The product corresponding to NO₃RR is NH₃, while the products for OER, UOR, and HzOR are O₂, (CO₂, N₂), and N₂, respectively. This system provides an alternative green route for ammonia synthesis at a much lower energy than other systems using sluggish water oxidation as the half-reaction. HzOR to N₂ requires 4 electrons while the NO₃RR to NH₃ is an 8-electron process. Hence, the kinetics of NO₃RR is more sluggish. To couple these two reactions to produce a galvanic cell, the design of appropriate catalysts is required that can operate at low overpotentials and generate a maximum output voltage. In an attempt to design such a cell, 1D bimetallic RuCo nanorods act as efficient precatalysts for both cathodic and anodic reactions.³⁵¹ Nitrate and hydrazine wastewater are fed into the cathodic and anodic chambers of the cell. At the anode, the RuCo nanorods oxidize hydrazine to N₂ at a very low overpotential of −0.031 V vs. RHE. Simultaneously, the same RuCo nanorods have been used to reduce NO₃⁻ to value-added NH₃ at 0.161 V vs. RHE at the cathode. However, post-reaction characterizations reveal the formation of heterostructures of Ru|Cu(OH)₂ 2D nanosheets over RuCo precatalysts, and it has been noted that these reconstructed nanosheets in situ are the active sites for HzOR and NO₃RR. Due to the overuse of plastic by mankind and its toxicity, plastic waste is one of the most common forms of pollutant noted in water, landfills, and the environment. From this viewpoint, in a recent work, NO₃RR has been coupled with the oxidation of ethylene glycol in polyethylene terephthalate (PET) for waste upcycling (Fig. 23E–H).⁵⁴ Low crystalline CoOOH catalysts reduce nitrates to produce ammonia at the cathode. Due to low crystallinity, the catalysts have both crystalline and amorphous regions. The amorphous area of the catalysts is endowed with a disordered atomic structure and dangled chemical bonds, which provide more active sites for the adsorption of NO₃⁻. The crystallinity enhances the conductivity of the catalyst and promotes electron transfer for NO₃RR. For PET conversion, the anode is formed by 1D Pd nanothorns. The nanothorns have a high curvature tip, which enhances the local electric field and, thus, ensures the availability of a concentration of OH⁻ near the active site. The OH⁻ converts the ethylene glycol in PET to carbonyl compounds. Pd nanothorns anode can not only result in the degradation of plastics but also form value-added products like glycolic acid and formate. These coupled redox reactions can lead to the formation of multiple value-added products at the same time in an energy-efficient fashion.

Fig. 23 Low-dimensional catalysts for ammonia synthesis via nitrate reduction coupled with oxidation reactions. (A) Schematic representation showing nitrate reduction (NO₃RR) coupled hydrazine oxidation (HzOR) by tungsten phosphide (WP) nanowires deposited on Ni foam (WP/NF) electrodes coupled with perovskite solar cell. (B) Low and (C) high magnification SEM images of WP nanowires. (D) Potential profiles comparing the cathodic and anodic nitrate reduction-coupled hydrazine oxidation reaction (NORR-HzOR) and nitrate reduction-coupled oxygen evolution reaction (NORR-OER). Reproduced with permission from ref. 55. Copyright 2023, Wiley-VCH. (E) Schematic representation of co-electrolysis of NO₃⁻ and PET plastic at the low-crystalline CoOOH-based cathode and Pd nanothorn-based anodes to produce NH₃ and glycolic acid simultaneously. (F) SEM image of the low-crystalline CoOOH grown on cobalt foam. (G) SEM image of Pd nanothorns grown on Ni foam. (H) Faradaic efficiencies of NH₃ and glycolic acid produced at different potentials. Reproduced with permission from ref. 54. Copyright 2023, American Chemical Society. (I) Schematic representation of paired electrochemical refining, where NiCu-based electrodes convert nitrate and glycerol to value-added ammonia and formate, respectively. (J) HRTEM image of reconstructed NiCu–OH cathode showcasing amorphous Ni(OH)₂ marked by red circles and lattice fringes with d-spacing of 0.20 nm corresponding to (111) planes of crystalline Cu, inset showing the corresponding FFT pattern. (K) STEM image of the reconstructed NiCuO anode showing lattice fringes with d-spacings of 0.209 nm (blue square) corresponding to the (200) of cubic NiO phase and 0.252 nm (red square), corresponding to the (111) plane of the monoclinic CuO phase. (L) NH₃ yields and the corresponding Faradaic efficiencies at the reconstructed NiCu–OH cathode at different voltages. Reproduced with permission from ref. 352. Copyright 2022, Royal Society of Chemistry.

7.3.3. 2D nanostructured and heterostructured catalysts. Arrays of 2D nanosheets of cobalt phosphide on carbon fiber cloth can electrochemically reduce NO₃⁻ to produce NH₃ under ambient conditions with a Faradaic efficiency of ∼100% in an alkaline medium.³⁵³ Mechanistic investigations indicate that the P centers of the catalyst optimize the active phase and the energy barriers for NO₃RR, and the Co centers adsorb NO₃⁻ and promote electron transfer for the catalytic process. Here, the oxidation reaction, OER at the anode, is replaced with benzyl alcohol oxidation. This approach lowers the potential of the oxidation reaction from E⁰ = 1.23 V vs. RHE for OER to E⁰ = 0.48 V vs. RHE, indicating the benzyl alcohol oxidation to be thermodynamically more favourable than OER. Apart from lowering the overpotential, coupling of the benzyl alcohol oxidation can also yield value-added products like benzaldehyde and benzoic acid, besides ammonia. The potential of the coupled oxidation reaction can also be lowered by replacing OER with sulfide oxidation reaction (SOR), requiring a lower potential of 0.65 V vs. RHE to attain 100 mA cm⁻².³⁵⁴ For obtaining nitrite reduction reaction (NO₂RR) coupled with SOR, 2D CoNiOOH nanosheets can work as bifunctional catalysts, which can simultaneously generate two value-added products, ammonia and elemental sulfur, from NO₂RR and SOR, respectively. The Ni dopant facilitates the adsorption of NO₂⁻ and S²⁻, lowers energy barriers, and enhances NO₂RR and SOR simultaneously. The synchronous production of two value-added products at the cathode and anode offers a greener energy-saving approach for chemical production. Another example of the NO₃RR-coupled glycerol oxidation reaction (GOR) catalyst includes non-noble bimetallic 2D NiCu–OH nanosheets fabricated by co-electrodeposition of amorphous Ni(OH)₂ and crystalline Cu₂(OH)₂(NO₃).³⁵² These 2D nanosheets act as the precatalysts for NO₃RR and GOR at the cathode and anode, respectively (Fig. 23I–L). NiCu–OH can be reconstructed to Cu nanoparticles and amorphous Ni(OH)₂ at the cathode under the operating reaction conditions for NO₃RR. The metallic Cu produced in situ at the cathode catalytically reduces NO₃⁻ to produce green NH₃. Concurrently, at the anode, under the reaction conditions of GOR, NiCu–OH is reconstructed to form Cu-vacant CuO/NiOOH composites. These bimetallic composites catalytically oxidize glycerol to value-added formate. These reconstructed 2D nanosheets have also been used to build a proof-of-concept paired electrochemical refinery that can perform NO₃RR and GOR at low cell voltage and simultaneously generate ammonium and formate. This work provides a proof-of-concept and paves an alternative way for constructing large-scale electrolyzers for ammonia synthesis.

8. Conclusions and perspectives

Over the past decade, the development of novel low-dimensional materials in electrocatalysis and photocatalysis has not only advanced the efficient synthesis of various value-added products but has also deepened our understanding of the complex mechanisms behind catalytic reactions, simplifying the design of more effective catalysts. The electronic properties of these materials can be easily tuned and their interfaces modulated for better adsorption and availability of catalytic sites, thereby making them suitable for photocatalysis and electrocatalysis. The combined effects of the materials and unique interfacial properties present in the heterostructures make them state-of-the-art catalysts in this field. Doping, defects, facet engineering, and formation of heterojunctions boost the catalytic properties of the low-dimensional materials. These catalytic materials can lead to selective production of green ammonia in a sustainable fashion by the reduction of N-containing species like N₂, NO_x, or NO_x⁻. The enzymes present in the nitrogen cycle act as the inspiration for fabricating active low-dimensional materials as catalysts. In this review, we have outlined the recent 0D, 1D, and 2D low-dimensional materials used for the photo- and/or electrocatalytic production of ammonia. These low-dimensional materials present an array of physical and chemical characteristics for boosting the catalytic reactions. For instance, the alignment of the band positions of the photocatalysts and the redox potentials of the desired reactions is crucial for performing the photocatalytic reactions. Low-dimensional materials offer the opportunity to design heterostructures such that the band positions can be engineered to align with the redox potentials of N₂RR and NO_xRR. The important features of low-dimensional materials like light absorption properties, charge carrier properties, and lower recombination of charges further make them ideal photocatalysts. Most state-of-the-art photocatalysts developed in the lab use simulated light sources. Direct harvest of solar light for the effective generation of ammonia still poses a challenge. Though few reports of solar light-harvesting photocatalysts have been reported, their efficiency is unsatisfactory. Hence, increasing the solar-to-fuel efficiency of the photocatalysts for the production of ammonia remains a major concern. Quantum materials with plasmonic structures have better light absorption properties and can be used to design efficient photocatalysts. The band structures of quantum materials are optimized to provide the electrons required for the catalytic reactions and, thus they act as electron reservoirs. The quantum states can provide better conductivity, transportation of charges, improved lifetime, and enhanced surface properties, which make them prospective materials for energy-converting catalytic reactions. Quantum materials can transfer charge and electron spin, which might boost the catalytic activity of reduction reactions of N-containing species to ammonia. Generally, the production of ammonia originates from the reduction of dinitrogen or nitrates, which requires 6 and 8 electrons, respectively, and, in principle, the reactions require high overpotentials to surpass the high energy barrier. In electrocatalysis, the low-dimensional materials are supposed to provide lower overpotential compared with ideal thermodynamic potentials. However, the overpotentials reported for most electrocatalysts developed for ammonia production are still much higher than the thermodynamic potential, which requires further initiatives for designing heterostructures from experimental and theoretical aspects. Determination of the active sites of the catalysts through theoretical calculations is also pertinent for assessing and comparing the performance of various catalytic materials. To effectively utilize electrocatalytically synthesized ammonia in fertilizers, it is crucial to select appropriate electrolytes. Electrolytes should be non-toxic for plants and also should not affect their growth. NaCl, KOH, and Na₂SO₄ are mostly used as electrolytes for this purpose. For large-scale practical applications, designing flow cells appears to be a promising strategy, and some progress has been made in this area. However, significant optimizations are still required to enhance efficiency, increase yield, and improve the stability of catalysts before they can be used for the mass production of ammonia. The stability of catalysts is often hindered due to side reactions and can be modulated by driving the flow of charge carriers to the active sites for the reaction. The input of two energy sources, photo and electrical energy for generating ammonia by photocathodes has also been explored. However, state-of-the-art photoelectrocatalysts developed at the lab-scale report a low efficiency of ammonia production. Coupling with an electric grid or battery to increase the current density might lead to a higher yield of ammonia.

Over the past years, the constant progress in ammonia synthesis has necessitated the establishment of a rigorous protocol for the measurement and quantification of NH₃. Though diverse and well-established quantification techniques exist in the literature, contamination from different nitrogen-containing molecules like NO_x or a non-negligible amount of ammonia from solvents or electrolytes, nitrogen or hydrogen leaching, and non-catalytic ammonia production can lead to errors in the quantification of ammonia. Additionally, ammonia can be in the form of ammonium ion (NH₄⁺) or un-ionized ammonia (NH₃) in liquid reaction media, depending on the pH and the temperature. At higher temperatures and pH (>11), ammonia predominantly remains in the gaseous state.³⁵⁵ Hence, in alkaline solutions, gaseous ammonia is present in the head space and the liquid phase as dissolved gas. This further complicates the quantitative analysis and preserving the samples for measurements. Reviews and protocols describing the correct method for ammonia measurement and quantification are present in the literature, following which correct protocols for ammonia quantification can be established in the lab.^{5,15,24,356,357} Among the existing techniques, the spectroscopic methods for ammonia quantification are the most common, less expensive and widely used. However, these methods depend on the pH of the reaction medium and the concentration of ammonia to be measured. For instance, Nessler's reagent method works in both alkaline and acidic solutions, over a concentration range of 0–8 mg L⁻¹.³⁵⁷ However, the high alkalinity of Nessler's reagent, the toxicity due to the presence of mercury, and the short lifetime somewhat limit the use of this method. The indophenol blue method, another spectroscopic technique, works in alkaline media and detects low-concentration ammonia ranging from 0–0.6 mg NH₃–N L⁻¹.¹⁵ Nevertheless, the sample preparation for this method is lengthy and the presence of organic nitrogen-containing molecules can perturb the quantification. The salicylate spectroscopic method despite being a stable method suffers from major drawbacks like lower sensitivity and higher cost. Ion chromatography is also widely used for the quantification of ammonia and it offers many advantages over the spectroscopic methods, including efficiency, reproducibility, and high sensitivity toward NH₄⁺ detection, and covers a wide detection range of 0.02–40 mg NH₃–N L⁻¹.¹³⁶ However, this process is comparatively expensive, requires complex instrumentation, proper selection of columns and eluents, and is sometimes incompatible with acidic or basic solutions, and organic solvents. Ammonia can also be detected by fluorescence upon reaction with o-phthaldialdehyde and sulfite, but the detection limit is only up to 1 nmol L⁻¹.³⁵⁸ Ammonium ion-selective electrode (ISE) is another efficient technique for the measurement of ammonia with concentrations ranging 0.03–1400 mg NH₃–N L⁻¹, however, the accuracy of this method is poor for concentrations less than 0.5 mg NH₃–N L⁻¹.¹³⁶ Extensive background control experiments are always necessary to check and confirm the presence of any non-synthesized ammonia contaminants. In this regard, nuclear magnetic resonance (NMR) techniques with isotope labeling experiments play a crucial role. Catalytic experiments with ¹⁵N-labeled N-sources as reactants analyzed by ¹H and ¹⁵N NMR methods can provide conclusive proof of the catalytic production of ammonia. The use of ¹⁵N-isotope-labeled reactant generates ¹⁵NH₄⁺, which can be confirmed from ¹H NMR spectra showing the chemical shifts of triplet coupling of ¹⁴N and doublet coupling for ¹⁵N.⁵ In the case of non-aqueous systems, the presence of organic solvents sometimes interferes with the detection of the NH₃ signal, which can be resolved using solvent signal suppression methods. The accurate concentration of ammonia can be measured by calibration curves with standard solutions containing ammonia. Isotope-labeling experiments can also be performed with liquid chromatography–mass spectrometry (LC–MS) techniques. An LC-MS coupled Berthelot assay can detect ¹⁴NH₃ and ¹⁵NH₃ by analyzing ¹⁴N-indophenol (m/z: 198) and ¹⁵N-indophenol (m/z: 199) formed by Berthelot reactions.³⁵⁹ Although many quantification techniques are available, based on the reaction conditions, reactants, solvents, and the concentration of ammonia produced, the appropriate method has to be chosen for the accurate quantification of ammonia.

Despite the vast ongoing research on ammonia production, certain obstacles persist, necessitating additional progress. Currently, researchers are focusing on addressing these challenges and a few advancements have been achieved, and further upgradation in these aspects is required in the future. Despite being energy-intensive and leaving a significant carbon footprint, the Haber–Bosch (H–B) approach remains the industrial technology for ammonia synthesis. Recent efforts are now directed toward modification of the H–B plants with CO₂ capture facilities to reduce the carbon footprint. Different electrolyzers have been developed in recent years to increase the energy efficiency of the ammonia synthesis process. For instance, the Topsoe green ammonia demonstration uses a high-temperature solid oxide electrolyzer cell to power the ammonia synthesis process.³⁶⁰ This electrolyzer can split water to generate green hydrogen and separate nitrogen from the air, thus reducing energy input intake and increasing overall efficiency. The small-scale green ammonia demonstrator by the ThyssenKrupp industries applies the chlor-alkaline water electrolyzer cell for ammonia synthesis.³⁶¹ Polymer electrolyte membrane (PEM)-based electrolyzers are often used for industrial-scale ammonia synthesis. The green ammonia plant planned by Yara proposes to use a PEM electrolyzer produced by ITM Power.³⁶² The wind power-to-ammonia demonstrator at Harwell, Oxford, United Kingdom, in association with Siemens, uses a PEM water electrolyzer to produce H₂ and N₂ from air separation over a Johnson Matthey Fe-based catalyst.³⁶³ In an attempt to develop decentralized and energy-efficient technologies for ammonia generation, Europe has also proposed over 20 projects by 2030 for green ammonia generation.³⁶⁴ Green ammonia-generating plants are now being proposed to be constructed or currently are constructed in different parts of the world, such as the Australian Renewable Energy Agency (ARENA) in western Australia, Kapsom in north-eastern India, and Fertiberia in Spain.^365–367 The Yuri Renewable Hydrogen to Ammonia Project, led by Yara Pilbara Fertilisers Pty. Ltd (Yara), a fertilizer company, and ENGIE, a leader in low-carbon energy and services, with the support of ARENA, ENGINE, and Mitsui & Co. Ltd (Mitsui), was started in 2022, and is expected to be complete in 2028. This project proposes to industrially scale up the production of green ammonia using off-grid intermittent renewable H₂ obtained via electrolysis. This project also aims to build a 10 MW electrolyzer powered by 18 MW solar PV and supported by an 8 MW battery energy storage system.³⁶⁵ In the same year (2022), Fertiberia, a Spanish fertilizer producer with support from the Green H2F project also constructed a green ammonia plant in Spain with Iberdrola and Spain's National Hydrogen Center. This plant project has a 20 PEM MW electrolyzer powered by a 100 MW PV with a capacity of 3000 t_H2 year⁻¹. An additional 800 MW capacity is also under development and will be complete by 2027.³⁶⁸ Although the polymer electrolyte membrane (PEM) electrolyzers are quite promising H₂ production is somewhat limited by the low temperature. High-temperature electrolyzers including solid oxide or alkaline electrolyzers have also been considered but the higher cost limits their use for large-scale applications.²³ Different literature studies indicate that for the complete industrial production of decarbonized, sustainable production of green ammonia further developments are required to construct highly efficient electrolyzers (>80%).³⁶⁹ For entirely decarbonizing the Haber–Bosch approach to ammonia production, the upcoming goal is to generate ammonia from water, air, and renewable energy sources. In this regard, several photo- and/or electrocatalysts have been designed; still, they suffer from a few limitations like the cost and production of electrical energy and the low yield of ammonia produced in the range of mmol h⁻¹ g⁻¹. The yield of ammonia in the millimolar range is quite inferior compared to industrial standards. Hence, the photo- and electrocatalytic approaches developed so far are not sufficient for the lab to industrial-level scale-up of the technology. Further investigations in this regard are necessary and would require cumulative expertise from the fields of materials science, catalysis, reactor engineering, and the use of artificial intelligence and robotization.²⁴

Ammonia is highly water soluble due to the hydrogen bond between ammonia and water; hence, it is difficult to separate from aqueous medium. Most photocatalytic and electrocatalytic reactions are performed in an aqueous media, and the extraction of ammonia from aqueous phase remains an obstacle in most cases. Air and steam stripping are sometimes used to separate ammonia from aqueous medium, but these processes are energy-intensive.²⁰⁸ For air stripping of ammonia, maintenance of high airflow, temperature, and pressure of the packed bed are required to separate ammonia from water and convert it to the gaseous phase, eventually consuming a significant amount of energy.³⁷⁰ In the case of steam stripping, steam generation requires high temperatures consuming energy, while coupling heat recovery systems to improve the energy efficiency of this process can be complex and expensive.³⁷¹ Another effective way of ammonia separation is upscaling the produced ammonia to salts like ammonium sulfate or ammonium chloride. This process helps in easy ammonia retrieval and directly converts ammonia to fertilizers.^53,208,316 The development of membrane-based electrolyzers can also promote the separation of ammonia from the reaction medium. Recently, several lab-scale electrolyzers have been constructed for the electro/photoelectrocatalytic synthesis of ammonia. The electrolyzers have certain benefits over conventional catalytic setups. For instance, in conventional electrocatalysis electricity is sometimes harnessed from fossil fuels and produces greenhouse gases. Instead, the electrolyzers are powered by renewable energy sources like solar power or wind to produce electricity for the electrocatalytic reactions, thus making ammonia synthesis green and reducing the carbon footprint. Additionally, the electrolyzer concurrently couples electrosynthesis and gaseous product separation, which minimizes the undesired redox reaction between NH₃ and the oxidized products. Thus, the electrolyzers can produce highly pure ammonia with high yield rates and minimal loss. The electrolyzers can also participate in redox reactions and extract the oxidized and reduced products simultaneously. For instance, NH₃ and Cl₂ can be electrosynthesized simultaneously and separated by incorporating gas-extraction electrodes into a flow-type membrane-free electrolyzer, coupled with ammonia and chlorine trap channels and a waste stream channel.³⁷² Electrolyzers can be either membrane-based or membrane-free. The membrane-based electrolyzers can be proton exchange membranes (PEM), anion exchange membranes (AEM), or membrane–electrode assemblies (MEA). Membranes play a crucial role in determining the performance and scalability of ammonia-producing electrolyzers. In non-solid electrochemical/photoelectrochemical one-chamber cells, oxidation and reduction reactions occur in the single chamber and the ammonia produced upon reduction of N₂ or NO_x is susceptible to oxidation. To avoid this possibility, an electrolyzer consisting of a double-chamber cell or H-cell, separated by a proton exchange membrane (PEM) or an anion exchange membrane (AEM) is preferred. During the electrochemical reaction, the electrons generated at the anode-chamber can pass through the membrane to the cathode-chamber and participate in the N₂RR or NO_xRR to produce ammonia. However, the ammonia produced as NH₄⁺ in aqueous solutions cannot cross the membrane, thus avoiding oxidation and product loss. Hence, the choice of membranes that can limit NH₄⁺ crossover is extremely crucial. The most commonly used cation exchange membranes, Nafion 212 and Nafion 112 allow the transport of NH₄⁺; therefore they prove inefficient for ammonia generation electrolyzers.^367,373 Among anion exchange membranes, AEM PiperION-A80 exhibits negligible NH₄⁺ crossover in both acidic and neutral electrolytes and can be used for the electrosynthesis of ammonia in neutral or acidic reaction media. However, in basic electrolytes, AEM PiperION-A80 proves inefficient as it is permeable to NH₃ under basic conditions.³⁷³ Additionally, membranes allow the freedom of using two different electrolytes in the cathodic and anodic chambers.³⁷⁴ Anion exchange membranes are highly permeable to nitrates, hence, AEMs have to be avoided for NO₃RR systems. Instead, a proton-exchange membrane electrode assembly (PEMEA) constructed with Nafion 117 and coupled with electrocatalysts like oxide-derived Cu nanoneedle cathodes can efficiently produce ammonia from nitrates.³⁷⁵ The production of ammonia can be scaled up from the millimolar range obtained from conventional electrocatalytic reactions by copper–tin alloy catalyst to molar range by transferring to a membrane electrode assembly (MEA) electrolyzer using renewable electricity and the cathode and anode separated by anion-exchange quaternary ammonium poly(N-methylpiperidine-co-p-terphenyl) (QAPPT) membrane.³²² To meet the real application and industrial standards of ammonia production via the NO_xRR process, electrocatalysis experiments are now performed in electrolyzers using various electrolytes to obtain ampere-level current density.^230,376,377 Achieving long-term stability of ion-exchange membranes is still a challenge in many works, thus limiting long operating time for catalysis and high yield of ammonia. The construction of a 3D physically interlocked interface bipolar membrane can increase water dissociation sites, ionic transfer, and interfacial stability. By combining a Co nanoarray cathode with the bipolar membrane reactor, electrosynthesis of ammonia with increased yield can be achieved at 1000 mA cm⁻² for 100 hours of operation.³⁷⁸ MEA electrolyzers composed of cathodic CuZn ribbons, a proton exchange membrane (Nafion117), and an anodic IrO₂–Ti mesh, exhibit stable operation at 500 mA cm⁻² for 220 hours for producing ammonia.³⁷⁹ The electrolytes in the electrolyzers also have a significant impact on ammonia synthesis. The design of a flow electrolyzer with chain-ether-based electrolyte and gas diffusion electrodes can demonstrate 300 hours of continuous operation under ambient conditions.² The flow electrolyzers can also help separate synthesized ammonia by constructing a ‘two-in-one” flow cell electrolyzer that integrates the two chambers of NO₃RR electrolysis, and NH₃ capture through a commercial gas diffusion electrode.³⁸⁰ Apart from synchronized NH₃ production and capture, this electrolyzer rapidly transports NH₃ molecules away from the reaction interfaces, thereby promoting NO₃RR to NH₃. With such rapid development in catalyst and membrane design, electrolyzer device optimization; green, sustainable, carbon-free, low-cost, large-scale industrial methods for ammonia production are expected soon. A multidisciplinary approach is needed to upgrade lab-scale ammonia production into full-scale industrial ammonia fertilizer production. This includes expertise in photocatalysis, electrocatalysis, chemical separation, agricultural science, and technical engineering. These combined skills are essential to developing an efficient alternative to the Haber–Bosch process and addressing the current global demand for sustainable ammonia production. Exploring and identifying the most efficient methods for ammonia production is a critical challenge this field must address in the coming years. Advancing these technologies will be the key to driving growth in both the agriculture-based economy and the energy sectors that rely on ammonia as a fuel source.

Author contributions

M. P., A. M., and C. C. M. conceived and designed the review. A. M. wrote the initial draft. A. M., C. C. M., and M. P. reviewed, edited, and wrote the manuscript. M. P. supervised the entire review. All the authors contributed to the discussion and manuscript preparation.

Data availability

No primary research results, software, or code have been included, and no new data were generated or analyzed as part of this review.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This research was co-funded by the European Union under the REFRESH – Research Excellence For REgion Sustainability and High-tech Industries project number CZ.10.03.01/00/22_003/0000048 via the Operational Programme Just Transition. AM acknowledges the funding from the project MSCA CZ fellowship at VSB-TUO with project number CZ.02.01.01/00/22_010/0006136.

References

1. R. F. Service, Science, 2018, 361, 120–123.
2. S. Li, Y. Zhou, X. Fu, J. B. Pedersen, M. Saccoccio, S. Z. Andersen, K. Enemark-Rasmussen, P. J. Kempen, C. D. Damsgaard, A. Xu, R. Sažinas, J. B. V. Mygind, N. H. Deissler, J. Kibsgaard, P. C. K. Vesborg, J. K. Nørskov and I. Chorkendorff, Nature, 2024, 629, 92–97.
3. K. Zhang, A. Cao, L. H. Wandall, J. Vernieres, J. Kibsgaard, J. K. Nørskov and I. Chorkendorff, Science, 2024, 383, 1357–1363.
4. D. R. MacFarlane, P. V. Cherepanov, J. Choi, B. H. R. Suryanto, R. Y. Hodgetts, J. M. Bakker, F. M. F. Vallana and A. N. Simonov, Joule, 2020, 4, 1186–1205.
5. H. Iriawan, S. Z. Andersen, X. Zhang, B. M. Comer, J. Barrio, P. Chen, A. J. Medford, I. E. L. Stephens, I. Chorkendorff and Y. Shao-Horn, Nat. Rev. Methods Primers, 2021, 1, 56.
6. J. Kim and D. C. Rees, Biochemistry, 1994, 33, 389–397.
7. K. A. Brown, D. F. Harris, M. B. Wilker, A. Rasmussen, N. Khadka, H. Hamby, S. Keable, G. Dukovic, J. W. Peters, L. C. Seefeldt and P. W. King, Science, 2016, 352, 448–450.
8. Z. Chen, G. Quek, J.-Y. Zhu, S. J. W. Chan, S. J. Cox-Vázquez, F. Lopez-Garcia and G. C. Bazan, Angew. Chem., Int. Ed., 2023, 62, e202307101.
9. D. Voiry, H. S. Shin, K. P. Loh and M. Chhowalla, Nat. Rev. Chem., 2018, 2, 0105.
10. C. Katan, N. Mercier and J. Even, Chem. Rev., 2019, 119, 3140–3192.
11. T. Edvinsson, R. Soc. Open Sci., 2018, 5, 180387.
12. Q. Shao, P. Wang, T. Zhu and X. Huang, Acc. Chem. Res., 2019, 52, 3384–3396.
13. Y. Pang, C. Su, G. Jia, L. Xu and Z. Shao, Chem. Soc. Rev., 2021, 50, 12744–12787.
14. Y. Ruan, Z.-H. He, Z.-T. Liu, W. Wang, L. Hao, L. Xu, A. W. Robertson and Z. Sun, J. Mater. Chem. A, 2023, 11, 22590–22607.
15. G. Qing, R. Ghazfar, S. T. Jackowski, F. Habibzadeh, M. M. Ashtiani, C.-P. Chen, M. R. Smith III and T. W. Hamann, Chem. Rev., 2020, 120, 5437–5516.
16. Y. Shi, Z. Zhao, D. Yang, J. Tan, X. Xin, Y. Liu and Z. Jiang, Chem. Soc. Rev., 2023, 52, 6938–6956.
17. Y. Xiong, Y. Wang, J. Zhou, F. Liu, F. Hao and Z. Fan, Adv. Mater., 2024, 36, 2304021.
18. A. K. K. Padinjareveetil and M. Pumera, Adv. Mater. Interfaces, 2023, 10, 2201734.
19. B. Wang, S. Chen, Z. Zhang and D. Wang, SmartMat, 2022, 3, 84–110.
20. S. Li, X. Fu, J. K. Nørskov and I. Chorkendorff, Nat. Energy, 2024, 9, 1344–1349.
21. J. John, D. R. MacFarlane and A. N. Simonov, Nat. Catal., 2023, 6, 1125–1130.
22. B. H. R. Suryanto, H.-L. Du, D. Wang, J. Chen, A. N. Simonov and D. R. MacFarlane, Nat. Catal., 2019, 2, 290–296.
23. D. Ye and S. C. E. Tsang, Nat. Synth., 2023, 2, 612–623.
24. L. Collado, A. H. Pizarro, M. Barawi, M. García-Tecedor, M. Liras and V. A. D. L. P. O'Shea, Chem. Soc. Rev., 2024, 53, 11334–11389.
25. Y. Feng, L. Jiao, X. Zhuang, Y. Wang and J. Yao, Adv. Mater., 2025, 37, 2410909.
26. W.-Q. Li, M. Xu, J.-S. Chen and T.-N. Ye, Adv. Mater., 2024, 36, 2408434.
27. R. Nittoor-Veedu, X. Ju and M. Pumera, Adv. Energy Mater., 2024, 2402205.
28. G. Gan, G. Hong and W. Zhang, Adv. Funct. Mater., 2024, 2401472.
29. Y. Zhou, X. Fu, I. Chorkendorff and J. K. Nørskov, ACS Energy Lett., 2025, 10, 128–132.
30. S. Yin, Z. Guan, Y. Zhu, D. Guo, X. Chen and S. Wang, ACS Nano, 2024, 18, 27833–27852.
31. M. Majumder, H. Saini, I. Dědek, A. Schneemann, N. R. Chodankar, V. Ramarao, M. S. Santosh, A. K. Nanjundan, S. Kment, D. Dubal, M. Otyepka, R. Zbořil and K. Jayaramulu, ACS Nano, 2021, 15, 17275–17298.
32. T. Hu, X. Cheng, J. Luo, Y. Yan, Q. Zhang and Y. Li, ACS Catal., 2024, 14, 14539–14563.
33. H. Fang, D. Liu, Y. Luo, Y. Zhou, S. Liang, X. Wang, B. Lin and L. Jiang, ACS Catal., 2022, 12, 3938–3954.
34. X. Long, F. Huang, Z. Yao, P. Li, T. Zhong, H. Zhao, S. Tian, D. Shu and C. He, Small, 2024, 20, 2400551.
35. Y. Li, Q. Zhang, Z. Mei, S. Li, W. Luo, F. Pan, H. Liu and S. Dou, Small Methods, 2021, 5, 2100460.
36. K. Ithisuphalap, H. Zhang, L. Guo, Q. Yang, H. Yang and G. Wu, Small Methods, 2019, 3, 1800352.
37. N. H. Truong, J.-S. Kim, J. Lim and H. Shin, Chem. Eng. J., 2024, 495, 153108.
38. S. M. Ghoreishian, K. Shariati, Y. S. Huh and J. Lauterbach, Chem. Eng. J., 2023, 467, 143533.
39. A. H. Pizarro, J. Fermoso, M. García-Tecedor, M. Barawi, V. A. D. L. P. O'Shea and L. Collado, J. Mater. Chem. A, 2024, 12, 16987–17001.
40. C.-W. Tsao, M.-J. Fang and Y.-J. Hsu, Coord. Chem. Rev., 2021, 438, 213876.
41. Y. Wang, H. Yin, X. Zhao, Y. Qu, A. Zheng, H. Zhou, W. Fang and J. Li, Appl. Catal., B, 2024, 341, 123266.
42. J. Qin, X. Hu, X. Li, Z. Yin, B. Liu and K.-H. Lam, Nano Energy, 2019, 61, 27–35.
43. H. Yin, Z. Chen, Y. Peng, S. Xiong, Y. Li, H. Yamashita and J. Li, Angew. Chem., Int. Ed., 2022, 61, e202114242.
44. M. Mohebinia, X. Xing, G. Yang, D. Wang, C. Solares-Bockmon, Z. Ren and J. Bao, Mater. Today Chem., 2022, 24, 100993.
45. Y. Kong, L. Wu, X. Yang, Y. Li, S. Zheng, B. Yang, Z. Li, Q. Zhang, S. Zhou, L. Lei, G. Wu and Y. Hou, Adv. Funct. Mater., 2022, 32, 2205409.
46. D. Yin, D. Chen, Y. Zhang, W. Wang, Q. Quan, W. Wang, Y. Meng, Z. Lai, Z. Yang, S. Yip, C.-Y. Wong, X. Bu, X. Wang and J. C. Ho, Adv. Funct. Mater., 2023, 33, 2303803.
47. W. Tong, B. Huang, P. Wang, L. Li, Q. Shao and X. Huang, Angew. Chem., Int. Ed., 2020, 59, 2649–2653.
48. A. K. K. Padinjareveetil, J. V. Perales-Rondon and M. Pumera, Adv. Mater. Technol., 2023, 8, 2202080.
49. Y. J. Jang, A. E. Lindberg, M. A. Lumley and K.-S. Choi, ACS Energy Lett., 2020, 5, 1834–1839.
50. L. P. Camargo, P. R. C. da Silva, A. Batagin-Neto, V. Klobukoski, M. Vidotti and L. H. Dall'Antonia, Appl. Mater. Today, 2022, 28, 101540.
51. B. Wang, L. Yao, G. Xu, X. Zhang, D. Wang, X. Shu, J. Lv and Y.-C. Wu, ACS Appl. Mater. Interfaces, 2020, 12, 20376–20382.
52. M. A. Mushtaq, A. Kumar, G. Yasin, M. Arif, M. Tabish, S. Ibraheem, X. Cai, W. Ye, X. Fang, A. Saad, J. Zhao, S. Ji and D. Yan, Appl. Catal., B, 2022, 317, 121711.
53. J. Gao, N. Shi, Y. Li, B. Jiang, T. Marhaba and W. Zhang, Environ. Sci. Technol., 2022, 56, 11602–11613.
54. T. Ren, Z. Duan, H. Wang, H. Yu, K. Deng, Z. Wang, H. Wang, L. Wang and Y. Xu, ACS Catal., 2023, 13, 10394–10404.
55. C. Lim, H. Roh, E. H. Kim, H. Kim, T. Park, D. Lee and K. Yong, Small, 2023, 19, 2304274.
56. A. Tayyebi, R. Mehrotra, M. A. Mubarok, J. Kim, M. Zafari, M. Tayebi, D. Oh, S.-H. Lee, J. E. Matthews, S.-W. Lee, T. J. Shin, G. Lee, T. F. Jaramillo, S.-Y. Jang and J.-W. Jang, Nat. Catal., 2024, 7, 510–521.
57. C. Lv, C. Lee, L. Zhong, H. Liu, J. Liu, L. Yang, C. Yan, W. Yu, H. H. Hng, Z. Qi, L. Song, S. Li, K. P. Loh, Q. Yan and G. Yu, ACS Nano, 2022, 16, 8213–8222.
58. X. Wei, Y. Liu, X. Zhu, S. Bo, L. Xiao, C. Chen, T. T. T. Nga, Y. He, M. Qiu, C. Xie, D. Wang, Q. Liu, F. Dong, C.-L. Dong, X.-Z. Fu and S. Wang, Adv. Mater., 2023, 35, 2300020.
59. Y. Zhao, Y. Ding, W. Li, C. Liu, Y. Li, Z. Zhao, Y. Shan, F. Li, L. Sun and F. Li, Nat. Commun., 2023, 14, 4491.
60. J. Geng, S. Ji, M. Jin, C. Zhang, M. Xu, G. Wang, C. Liang and H. Zhang, Angew. Chem., Int. Ed., 2023, 62, e202210958.
61. J. Shao, H. Jing, P. Wei, X. Fu, L. Pang, Y. Song, K. Ye, M. Li, L. Jiang, J. Ma, R. Li, R. Si, Z. Peng, G. Wang and J. Xiao, Nat. Energy, 2023, 8, 1273–1283.
62. Q. Lai, T. Cai, S. C. E. Tsang, X. Chen, R. Ye, Z. Xu, M. D. Argyle, D. Ding, Y. Chen, J. Wang, A. G. Russell, Y. Wu, J. Liu and M. Fan, Sci. Bull., 2022, 67, 2124–2138.
63. H. Li, F. Pan, C. Qin, T. Wang and K.-J. Chen, Adv. Energy Mater., 2023, 13, 2301378.
64. M. Appl, Ammonia, Ullmann's Encyclopedia of Industrial Chemistry, 2006 DOI:10.1002/14356007.a02_143.pub2.
65. R. D. Milton, R. Cai, S. Abdellaoui, D. Leech, A. L. De Lacey, M. Pita and S. D. Minteer, Angew. Chem., Int. Ed., 2017, 56, 2680–2683.
66. J. Guo and P. Chen, Acc. Chem. Res., 2021, 54, 2434–2444.
67. O. Westhead, J. Barrio, A. Bagger, J. W. Murray, J. Rossmeisl, M.-M. Titirici, R. Jervis, A. Fantuzzi, A. Ashley and I. E. L. Stephens, Nat. Rev. Chem., 2023, 7, 184–201.
68. N. R. Dhar, E. V. Seshacharyulu and N. N. Biswas, Proc. Natl. Inst. Sci., 1941, 7, 115–131.
69. G. N. Schrauzer, N. Strampach, L. N. Hui, M. R. Palmer and J. Salehi, Proc. Natl. Acad. Sci. U. S. A., 1983, 80, 3873–3876.
70. H. Hirakawa, M. Hashimoto, Y. Shiraishi and T. Hirai, J. Am. Chem. Soc., 2017, 139, 10929–10936.
71. Q. Han, C. Wu, H. Jiao, R. Xu, Y. Wang, J. Xie, Q. Guo and J. Tang, Adv. Mater., 2021, 33, 2008180.
72. X. Gao, L. An, D. Qu, W. Jiang, Y. Chai, S. Sun, X. Liu and Z. Sun, Sci. Bull., 2019, 64, 918–925.
73. X. Chen, X. Zhang, Y.-H. Li, M.-Y. Qi, J.-Y. Li, Z.-R. Tang, Z. Zhou and Y.-J. Xu, Appl. Catal., B, 2021, 281, 119516.
74. P. Li, S. Gao, Q. Liu, P. Ding, Y. Wu, C. Wang, S. Yu, W. Liu, Q. Wang and S. Chen, Adv. Energy Sustainability Res., 2021, 2, 2000097.
75. S. Zhang, Y. Zhao, R. Shi, C. Zhou, G. I. N. Waterhouse, L.-Z. Wu, C.-H. Tung and T. Zhang, Adv. Energy Mater., 2020, 10, 1901973.
76. Y. Zhao, L. Zheng, R. Shi, S. Zhang, X. Bian, F. Wu, X. Cao, G. I. N. Waterhouse and T. Zhang, Adv. Energy Mater., 2020, 10, 2002199.
77. W. Wang, H. Zhou, Y. Liu, S. Zhang, Y. Zhang, G. Wang, H. Zhang and H. Zhao, Small, 2020, 16, 1906880.
78. Y. Lu, Y. Yang, T. Zhang, Z. Ge, H. Chang, P. Xiao, Y. Xie, L. Hua, Q. Li, H. Li, B. Ma, N. Guan, Y. Ma and Y. Chen, ACS Nano, 2016, 10, 10507–10515.
79. G. N. Schrauzer and T. D. Guth, J. Am. Chem. Soc., 1977, 99, 7189–7193.
80. Y. Guan, H. Wen, K. Cui, Q. Wang, W. Gao, Y. Cai, Z. Cheng, Q. Pei, Z. Li, H. Cao, T. He, J. Guo and P. Chen, Nat. Chem., 2024, 16, 373–379.
81. C. S. Diercks, Y. Liu, K. E. Cordova and O. M. Yaghi, Nat. Mater., 2018, 17, 301–307.
82. X. Cui, C. Tang and Q. Zhang, Adv. Energy Mater., 2018, 8, 1800369.
83. A. R. Singh, B. A. Rohr, J. A. Schwalbe, M. Cargnello, K. Chan, T. F. Jaramillo, I. Chorkendorff and J. K. Nørskov, ACS Catal., 2017, 7, 706–709.
84. M. Ali, F. Zhou, K. Chen, C. Kotzur, C. Xiao, L. Bourgeois, X. Zhang and D. R. MacFarlane, Nat. Commun., 2016, 7, 11335.
85. H. E. Kim, J. Kim, E. C. Ra, H. Zhang, Y. J. Jang and J. S. Lee, Angew. Chem., Int. Ed., 2022, 61, e202204117.
86. M. A. Mushtaq, M. Arif, X. Fang, G. Yasin, W. Ye, M. Basharat, B. Zhou, S. Yang, S. Ji and D. Yan, J. Mater. Chem. A, 2021, 9, 2742–2753.
87. M. S. Yu, S. C. Jesudass, S. Surendran, J. Y. Kim, U. Sim and M.-K. Han, ACS Appl. Mater. Interfaces, 2022, 14, 31889–31899.
88. C. Sun, Z. Shao, Y. Hu, Y. Peng and Q. Xie, ACS Appl. Mater. Interfaces, 2023, 15, 23085–23092.
89. X. Chen, N. Li, Z. Kong, W.-J. Ong and X. Zhao, Mater. Horiz., 2018, 5, 9–27.
90. B. Rytelewska, A. Chmielnicka, T. Chouki, M. Skunik-Nuckowska, S. Naghdi, D. Eder, A. Michalowska, T. Ratajczyk, E. Pavlica, S. Emin, Y. Fu, I. A. Rutkowska and P. J. Kulesza, Electrochim. Acta, 2023, 471, 143360.
91. M. A. Shipman and M. D. Symes, Catal. Today, 2017, 286, 57–68.
92. H. Wan, A. Bagger and J. Rossmeisl, Angew. Chem., Int. Ed., 2021, 60, 21966–21972.
93. X. Lu, H. Song, J. Cai and S. Lu, Electrochem. Commun., 2021, 129, 107094.
94. H. Liu, L. Bai, A. Bergmann, B. R. Cuenya and J. Luo, Chem, 2024, 10, 2963–2986.
95. T. Ouyang, S. Huang, X.-T. Wang and Z.-Q. Liu, Chem. – Eur. J., 2020, 26, 14024–14035.
96. H. Gleiter, Nanostruct. Mater., 1995, 6, 3–14.
97. V. V. Pokropivny and V. V. Skorokhod, Mater. Sci. Eng., C, 2007, 27, 990–993.
98. Z.-H. Xue, S.-N. Zhang, Y.-X. Lin, H. Su, G.-Y. Zhai, J.-T. Han, Q.-Y. Yu, X.-H. Li, M. Antonietti and J.-S. Chen, J. Am. Chem. Soc., 2019, 141, 14976–14980.
99. L. Meng, J. M. D. Lane, L. Baca, J. Tafoya, T. Ao, B. Stoltzfus, M. Knudson, D. Morgan, K. Austin, C. Park, P. Chow, Y. Xiao, R. Li, Y. Qin and H. Fan, J. Am. Chem. Soc., 2020, 142, 6505–6510.
100. B. Guan, X. Wang, Y. Xiao, Y. Liu and Q. Huo, Nanoscale, 2013, 5, 2469–2475.
101. T. Hyeon, S. S. Lee, J. Park, Y. Chung and H. B. Na, J. Am. Chem. Soc., 2001, 123, 12798–12801.
102. J. Xu, S. Zhang, H. Liu, S. Liu, Y. Yuan, Y. Meng, M. Wang, C. Shen, Q. Peng, J. Chen, X. Wang, L. Song, K. Li and W. Chen, Angew. Chem., Int. Ed., 2023, 62, e202308044.
103. G. Liu, M. Zavelani-Rossi, G. Han, H. Zhao and A. Vomiero, J. Mater. Chem. A, 2023, 11, 8950–8960.
104. M. Parente, M. van Helvert, R. F. Hamans, R. Verbroekken, R. Sinha, A. Bieberle-Hütter and A. Baldi, Nano Lett., 2020, 20, 5759–5764.
105. J. E. Omoriyekomwan, A. Tahmasebi, J. Zhang and J. Yu, Energy Convers. Manage., 2019, 192, 88–99.
106. H. Du, H. Guo, K. Wang, X. Du, B. A. Beshiwork, S. Sun, Y. Luo, Q. Liu, T. Li and X. Sun, Angew. Chem., Int. Ed., 2023, 62, e202215782.
107. G. Liu, Z. Tang, X. Gu, N. Li, H. Lv, Y. Huang, Y. Zeng, M. Yuan, Q. Meng, Y. Zhou and C. Wang, Appl. Catal., B, 2022, 317, 121752.
108. Z. Chen, D. Chao, J. Liu, M. Copley, J. Lin, Z. Shen, G.-T. Kim and S. Passerini, J. Mater. Chem. A, 2017, 5, 15669–15675.
109. R. Khare, D. B. Shinde, S. Bansode, M. A. More, M. Majumder, V. K. Pillai and D. J. Late, Appl. Phys. Lett., 2015, 106, 023111.
110. D. J. Lee, S.-H. Yu, H. S. Lee, A. Jin, J. Lee, J. E. Lee, Y.-E. Sung and T. Hyeon, J. Mater. Chem. A, 2017, 5, 8744–8751.
111. W. Bai, S. Li, J. Ma, W. Cao and J. Zheng, J. Mater. Chem. A, 2019, 7, 9086–9098.
112. G. Xia, S. Wang, X. Zhao and L. Zhou, J. Mater. Chem. C, 2013, 1, 3291–3296.
113. Y. Min, G. D. Moon, B. S. Kim, B. Lim, J.-S. Kim, C. Y. Kang and U. Jeong, J. Am. Chem. Soc., 2012, 134, 2872–2875.
114. S. Mahasivam, V. Bansal and M. Sastry, J. Phys. Chem. Lett., 2024, 15, 3923–3928.
115. Z. L. Shaw, S. Kuriakose, S. Cheeseman, M. D. Dickey, J. Genzer, A. J. Christofferson, R. J. Crawford, C. F. McConville, J. Chapman, V. K. Truong, A. Elbourne and S. Walia, Nat. Commun., 2021, 12, 3897.
116. J. N. Tiwari, R. N. Tiwari and K. S. Kim, Prog. Mater. Sci., 2012, 57, 724–803.
117. S. Sun, M. Lu, X. Gao, Z. Shi, X. Bai, W. W. Yu and Y. Zhang, Adv. Sci., 2021, 8, 2102689.
118. P. Innocenzi and L. Stagi, Acc. Mater. Res., 2024, 5, 413–425.
119. Y. Zhang, F. Gao, H. You, Z. Li, B. Zou and Y. Du, Coord. Chem. Rev., 2022, 450, 214244.
120. Y. Lu, S. Du and R. Steinberger-Wilckens, Appl. Catal., B, 2016, 199, 292–314.
121. H. Hou, G. Shao, W. Yang and W.-Y. Wong, Prog. Mater. Sci., 2020, 113, 100671.
122. X. Lu, C. Wang and Y. Wei, Small, 2009, 5, 2349–2370.
123. L. Liang, X. Kang, Y. Sang and H. Liu, Adv. Sci., 2016, 3, 1500358.
124. Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim and H. Yan, Adv. Mater., 2003, 15, 353–389.
125. M. Byakodi, N. S. Shrikrishna, R. Sharma, S. Bhansali, Y. Mishra, A. Kaushik and S. Gandhi, Biosens. Bioelectron.: X, 2022, 12, 100284.
126. T. A. Shifa, F. Wang, Y. Liu and J. He, Adv. Mater., 2019, 31, 1804828.
127. W. Huang, L. Hu, Y. Tang, Z. Xie and H. Zhang, Adv. Funct. Mater., 2020, 30, 2005223.
128. Z. He, J. Zhang, X. Li, S. Guan, M. Dai and S. Wang, Small, 2020, 16, 2005051.
129. P. Sabatier, La catalyse en chimie organique, Paris & Liége Ch. Béranger Editeur, 1920.
130. Y. Zhu, W. Peng, Y. Li, G. Zhang, F. Zhang and X. Fan, Small Methods, 2019, 3, 1800438.
131. J. K. Nørskov, T. Bligaard, J. Rossmeisl and C. H. Christensen, Nat. Chem., 2009, 1, 37–46.
132. Ž. Kovačič, B. Likozar and M. Huš, ACS Catal., 2020, 10, 14984–15007.
133. W. Hou and S. B. Cronin, Adv. Funct. Mater., 2013, 23, 1612–1619.
134. B. Bajorowicz, M. P. Kobylański, A. Gołąbiewska, J. Nadolna, A. Zaleska-Medynska and A. Malankowska, Adv. Colloid Interface Sci., 2018, 256, 352–372.
135. G. Zhang, Y. Li, C. He, X. Ren, P. Zhang and H. Mi, Adv. Energy Mater., 2021, 11, 2003294.
136. M. Li, H. Huang, J. Low, C. Gao, R. Long and Y. Xiong, Small Methods, 2019, 3, 1800388.
137. J. Li, J. Wang, S. Shen, R. Chen, M. Liu and F. Dong, Environ. Sci. Technol., 2023, 57, 5445–5452.
138. H. Kominami, A. Furusho, S.-Y. Murakami, H. Inoue, Y. Kera and B. Ohtani, Catal. Lett., 2001, 76, 31–34.
139. M. Yamauchi, R. Abe, T. Tsukuda, K. Kato and M. Takata, J. Am. Chem. Soc., 2011, 133, 1150–1152.
140. G. Ren, J. Zhao, Z. Zhao, Z. Li, L. Wang, Z. Zhang, C. Li and X. Meng, Angew. Chem., Int. Ed., 2024, 63, e202314408.
141. G. Ren, M. Shi, Z. Li, Z. Zhang and X. Meng, Appl. Catal., B, 2023, 327, 122462.
142. J. Bao, H. Maimaiti, J. Sun, L. Feng and X. Zhao, Int. J. Hydrogen Energy, 2024, 95, 1–11.
143. X. Yu, P. Qiu, Y. Wang, B. He, X. Xu, H. Zhu, J. Ding, X. Liu, Z. Li and Y. Wang, J. Colloid Interface Sci., 2023, 640, 775–782.
144. P. Li, Z. Zhou, Q. Wang, M. Guo, S. Chen, J. Low, R. Long, W. Liu, P. Ding, Y. Wu and Y. Xiong, J. Am. Chem. Soc., 2020, 142, 12430–12439.
145. C. Liu, D. Hao, J. Ye, S. Ye, F. Zhou, H. Xie, G. Qin, J. Xu, J. Liu, S. Li and C. Sun, Adv. Energy Mater., 2023, 13, 2204126.
146. Y. Huang, Y. Zhu, S. Chen, X. Xie, Z. Wu and N. Zhang, Adv. Sci., 2021, 8, 2003626.
147. K. Li, W. Cai, Z. Zhang, H. Xie, Q. Zhong and H. Qu, Chem. Eng. J., 2022, 435, 135017.
148. Y. Fang, Y. Xue, L. Hui, H. Yu and Y. Li, Angew. Chem., Int. Ed., 2021, 60, 3170–3174.
149. K. Pournemati, A. Habibi-Yangjeh and A. Khataee, ACS Appl. Nano Mater., 2024, 7, 2200–2213.
150. D. Cui, S. Wang, X. Yang, L. Xu and F. Li, Small, 2024, 20, 2306229.
151. D. Hao, J. Ren, Y. Wang, H. Arandiyan, M. Garbrecht, X. Bai, H. K. Shon, W. Wei and B.-J. Ni, Energy Mater. Adv., 2021, 2021, 9761263.
152. R. Fu, Y. Wang, G. Wang, Q. Zhan, L. Zhang and L. Liu, Green Chem., 2023, 25, 8531–8538.
153. Z. Zhao, R. Tan, Y. Kong, Z. Zhang, S. Qiu, X. Mu and L. Li, Angew. Chem., Int. Ed., 2023, 62, e202303629.
154. J. Li, R. Chen, J. Wang, Y. Zhou, G. Yang and F. Dong, Nat. Commun., 2022, 13, 1098.
155. J. Li, R. Chen, J. Wang, K. Wang, Y. Zhou, M. Xing and F. Dong, Angew. Chem., Int. Ed., 2024, 63, e202317575.
156. P. Li, R. Wu, P. Li, S. Gao, Z. Qin, X. Song, W. Sun, Z. Hua, Q. Wang and S. Chen, Adv. Sci., 2024, 11, 2408829.
157. V. Jain, S. Tyagi, P. Roy and P. P. Pillai, J. Am. Chem. Soc., 2024, 146, 32356–32365.
158. C. Feng, F. Raziq, M. Hu, H. Huang, Z.-P. Wu, S. Zuo, J. Luo, Y. Ren, B. Chang, D. Cha, S. Ayirala, A. Al-Yousef, T. D. Dao and H. Zhang, Adv. Energy Mater., 2024, 14, 2303792.
159. P. Xia, X. Pan, S. Jiang, J. Yu, B. He, P. M. Ismail, W. Bai, J. Yang, L. Yang, H. Zhang, M. Cheng, H. Li, Q. Zhang, C. Xiao and Y. Xie, Adv. Mater., 2022, 34, 2200563.
160. R. Wu, S. Gao, C. Jones, M. Sun, M. Guo, R. Tai, S. Chen and Q. Wang, Adv. Funct. Mater., 2024, 34, 2314051.
161. L. Zhang, R. Li, L. Cui, Z. Sun, L. Guo, X. Zhang, Y. Wang, Y. Wang, Z. Yu, T. Lei, X. Jian, X. Gao, C. Fan and J. Liu, Chem. Eng. J., 2023, 461, 141892.
162. K. Wu, Z. Wang, X. Zhang, C. Sun, Q. Li, H. Zhang, X. Bai, A. Khosla and Z. Zhao, Langmuir, 2024, 40, 13603–13612.
163. Q. Zhang, Q. Li, H. Li, X. Shi, Y. Zhou, Q. Ye, R. Yang, D. Li and D. Jiang, Inorg. Chem., 2023, 62, 12138–12147.
164. Y. Liu, Y. Xue, L. Hui, H. Yu, Y. Fang, F. He and Y. Li, Nano Energy, 2021, 89, 106333.
165. S. Liu, M. Wang, H. Ji, L. Zhang, J. Ni, N. Li, T. Qian, C. Yan and J. Lu, Adv. Mater., 2023, 35, 2211730.
166. T. He, Z. Zhao, R. Liu, X. Liu, B. Ni, Y. Wei, Y. Wu, W. Yuan, H. Peng, Z. Jiang and Y. Zhao, J. Am. Chem. Soc., 2023, 145, 6057–6066.
167. X. Zhang, Y. Liu, S. Feng, X. Gu, M. Zhou, H. Wang and J. Hua, Appl. Catal., B, 2024, 366, 125013.
168. M. Yu, Y. Chen, M. Gao, G. Huang, Q. Chen and J. Bi, Small, 2023, 19, 2206407.
169. X. Wang, G. Fan, S. Guo, R. Gao, Y. Guo, C. Han, Y. Gao, J. Zhang, X. Gu and L. Wu, Angew. Chem., Int. Ed., 2024, 63, e202404258.
170. J. Tan, H. Ren, Z. Zhao, X. Xin, Y. Shi, D. Yang and Z. Jiang, Chem. Eng. J., 2023, 466, 143259.
171. Z. Shen, F. Li, L. Guo, X. Zhang, Y. Wang, Y. Wang, X. Jian, X. Gao, Z. Wang, R. Li, C. Fan and J. Liu, Appl. Catal., B, 2024, 346, 123732.
172. Y. Xi, Y. Xiang, T. Bao, Z. Li, C. Zhang, L. Yuan, J. Li, Y. Bi, C. Yu and C. Liu, Angew. Chem., Int. Ed., 2024, 63, e202409163.
173. C. Clavero, Nat. Photonics, 2014, 8, 95–103.
174. K. T. Ranjit and B. Viswanathan, J. Photochem. Photobiol., A, 1997, 108, 73–78.
175. A. Pandikumar, S. Manonmani and R. Ramaraj, Catal. Sci. Technol., 2012, 2, 345–353.
176. K. T. Ranjit, R. Krishnamoorthy, T. K. Varadarajan and B. Viswanathan, J. Photochem. Photobiol., A, 1995, 86, 185–189.
177. I. G. Zacharia and W. M. Deen, Ann. Biomed. Eng., 2005, 33, 214–222.
178. R. Chen, J. Li, J. Wang, W. Yang, S. Shen and F. Dong, Environ. Sci. Technol., 2023, 57, 12127–12134.
179. P. Zhang, L. Chen, D.-H. Kuo, B. Wu, Z. Su, D. Lu, Q. Wu, J. Li, J. Lin and X. Chen, J. Mater. Chem. A, 2024, 12, 7163–7177.
180. L. Chen, D. Shen, B. Li, Z. Xiao, W. Sun, X. Liu, J. Ma, C. Li and W. Wang, Ceram. Int., 2023, 49, 23129–23139.
181. J. Zhang, G. Zhang, H. Lan, H. Liu and J. Qu, Chem. Eng. J., 2022, 428, 130373.
182. G. Ren, M. Shi, S. Liu, Z. Li, Z. Zhang and X. Meng, Chem. Eng. J., 2023, 454, 140158.
183. X. Lv, W. Wei, F. Li, B. Huang and Y. Dai, Nano Lett., 2019, 19, 6391–6399.
184. K. Zhao and X. Quan, ACS Catal., 2021, 11, 2076–2097.
185. S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angew. Chem., Int. Ed., 2013, 52, 7372–7408.
186. H. Hirakawa, M. Hashimoto, Y. Shiraishi and T. Hirai, ACS Catal., 2017, 7, 3713–3720.
187. X. Huang, R. Du, J. Ren, X. Li, M. Fu, S. Fu, T. Ma, L. Guo, R. A. Soomro, C. Yang and D. Wang, ACS Catal., 2024, 14, 13542–13549.
188. A. Mallick, J. Kim and M. Pumera, Small, 2025, 21, 2407050.
189. H. Yu, F. Chen, X. Li, H. Huang, Q. Zhang, S. Su, K. Wang, E. Mao, B. Mei, G. Mul, T. Ma and Y. Zhang, Nat. Commun., 2021, 12, 4594.
190. H. Zheng, X. Li, K. Zhu, P. Liang, M. Wu, Y. Rao, R. Jian, F. Shi, J. Wang, K. Yan and J. Liu, Nano Energy, 2022, 93, 106831.
191. S. Gao, H. Ji, P. Yang, M. Guo, J. Tressel, S. Chen and Q. Wang, Small, 2023, 19, 2206114.
192. K.-C. Ma, H.-Y. Lin, Y.-C. Chen, C.-H. Tsai, K.-H. Zheng and J. M. Wu, J. Mater. Chem. A, 2024, 12, 26866–26876.
193. Y. Fang, Y. Liu, L. Qi, Y. Xue and Y. Li, Chem. Soc. Rev., 2022, 51, 2681–2709.
194. N. Keller and T. Bein, Chem. Soc. Rev., 2021, 50, 1813–1845.
195. H. Wang, H. Wang, Z. Wang, L. Tang, G. Zeng, P. Xu, M. Chen, T. Xiong, C. Zhou, X. Li, D. Huang, Y. Zhu, Z. Wang and J. Tang, Chem. Soc. Rev., 2020, 49, 4135–4165.
196. Z. Zhao, Y. Nian, J. Shi, X. Xin, X. Huang, Y. Shi, J. Tan, Y. Zhang, Y. Han, D. Yang and Z. Jiang, ACS Catal., 2024, 14, 11626–11634.
197. H. Zhao, D. Zhang, H. Li, W. Qi, X. Wu, Y. Han, W. Cai, Z. Wang, J. Lai and L. Wang, Adv. Energy Mater., 2020, 10, 2002131.
198. L. Tan, N. Yang, X. Huang, L. Peng, C. Tong, M. Deng, X. Tang, L. Li, Q. Liao and Z. Wei, Chem. Commun., 2019, 55, 14482–14485.
199. Z. Liu, C. Wang, C. Chen, C. Li and C. Guo, Electrochem. Commun., 2021, 131, 107121.
200. X. Min and B. Liu, Small, 2023, 19, 2300794.
201. D. Yao, C. Tang, L. Li, B. Xia, A. Vasileff, H. Jin, Y. Zhang and S.-Z. Qiao, Adv. Energy Mater., 2020, 10, 2001289.
202. D. Kim, S. Surendran, G. Janani, Y. Lim, H. Choi, M.-K. Han, S. Yuvaraj, T.-H. Kim, J. K. Kim and U. Sim, Mater. Lett., 2022, 314, 131808.
203. W. P. Utomo, H. Wu and Y. H. Ng, Small, 2022, 18, 2200996.
204. L. Liu, T. Xiao, H. Fu, Z. Chen, X. Qu and S. Zheng, Appl. Catal., B, 2023, 323, 122181.
205. X. Wang, X. Wu, W. Ma, X. Zhou, S. Zhang, D. Huang, L. R. Winter, J.-H. Kim and M. Elimelech, Proc. Natl. Acad. Sci. U. S. A., 2023, 120, e2217703120.
206. Y. Xue, Q. Yu, Q. Ma, Y. Chen, C. Zhang, W. Teng, J. Fan and W.-X. Zhang, Environ. Sci. Technol., 2022, 56, 14797–14807.
207. Y. Zhang, Y. Wang, L. Han, S. Wang, T. Cui, Y. Yan, M. Xu, H. Duan, Y. Kuang and X. Sun, Angew. Chem., Int. Ed., 2023, 62, e202213711.
208. F.-Y. Chen, Z.-Y. Wu, S. Gupta, D. J. Rivera, S. V. Lambeets, S. Pecaut, J. Y. T. Kim, P. Zhu, Y. Z. Finfrock, D. M. Meira, G. King, G. Gao, W. Xu, D. A. Cullen, H. Zhou, Y. Han, D. E. Perea, C. L. Muhich and H. Wang, Nat. Nanotechnol., 2022, 17, 759–767.
209. H. Liu, X. Lang, C. Zhu, J. Timoshenko, M. Rüscher, L. Bai, N. Guijarro, H. Yin, Y. Peng, J. Li, Z. Liu, W. Wang, B. R. Cuenya and J. Luo, Angew. Chem., Int. Ed., 2022, 61, e202202556.
210. S. Liu, S. Yin, S. Jiao, H. Zhang, Z. Wang, Y. Xu, X. Li, L. Wang and H. Wang, Mater. Today Energy, 2021, 21, 100828.
211. Y. Wang, Y. Xiong, M. Sun, J. Zhou, F. Hao, Q. Zhang, C. Ye, X. Wang, Z. Xu, Q. Wa, F. Liu, X. Meng, J. Wang, P. Lu, Y. Ma, J. Yin, Y. Zhu, S. Chu, B. Huang, L. Gu and Z. Fan, Angew. Chem., Int. Ed., 2024, 63, e202402841.
212. L. Ouyang, X. He, S. Sun, Y. Luo, D. Zheng, J. Chen, Y. Li, Y. Lin, Q. Liu, A. M. Asiri and X. Sun, J. Mater. Chem. A, 2022, 10, 23494–23498.
213. J. Liang, P. Liu, Q. Li, T. Li, L. Yue, Y. Luo, Q. Liu, N. Li, B. Tang, A. A. Alshehri, I. Shakir, P. O. Agboola, C. Sun and X. Sun, Angew. Chem., Int. Ed., 2022, 61, e202202087.
214. Q. Liu, L. Xie, J. Liang, Y. Ren, Y. Wang, L. Zhang, L. Yue, T. Li, Y. Luo, N. Li, B. Tang, Y. Liu, S. Gao, A. A. Alshehri, I. Shakir, P. O. Agboola, Q. Kong, Q. Wang, D. Ma and X. Sun, Small, 2022, 18, 2106961.
215. D. Liu, L. Qiao, Y. Chen, P. Zhou, J. Feng, C. C. Leong, K. W. Ng, S. Peng, S. Wang, W. F. Ip and H. Pan, Appl. Catal., B, 2023, 324, 122293.
216. R. Zhang, S. Zhang, Y. Guo, C. Li, J. Liu, Z. Huang, Y. Zhao, Y. Li and C. Zhi, Energy Environ. Sci., 2022, 15, 3024–3032.
217. N. J. Harmon, C. L. Rooney, Z. Tao, B. Shang, N. Raychaudhuri, C. Choi, H. Li and H. Wang, ACS Catal., 2022, 12, 9135–9142.
218. D. Dhanabal, S. S. Markandaraj and S. Shanmugam, ACS Catal., 2023, 13, 9136–9149.
219. L. Huang, L. Cheng, T. Ma, J.-J. Zhang, H. Wu, J. Su, Y. Song, H. Zhu, Q. Liu, M. Zhu, Z. Zeng, Q. He, M.-K. Tse, D.-T. Yang, B. I. Yakobson, B. Z. Tang, Y. Ren and R. Ye, Adv. Mater., 2023, 35, 2211856.
220. L. Cheng, T. Ma, B. Zhang, L. Huang, W. Guo, F. Hu, H. Zhu, Z. Wang, T. Zheng, D.-T. Yang, C.-K. Siu, Q. Liu, Y. Ren, C. Xia, B. Z. Tang and R. Ye, ACS Catal., 2022, 12, 11639–11650.
221. X.-F. Cheng, J.-H. He, H.-Q. Ji, H.-Y. Zhang, Q. Cao, W.-J. Sun, C.-L. Yan and J.-M. Lu, Adv. Mater., 2022, 34, 2205767.
222. L. Hui, Y. Xue, H. Yu, Y. Liu, Y. Fang, C. Xing, B. Huang and Y. Li, J. Am. Chem. Soc., 2019, 141, 10677–10683.
223. X. Feng, J. Liu, L. Chen, Y. Kong, Z. Zhang, Z. Zhang, D. Wang, W. Liu, S. Li, L. Tong and J. Zhang, J. Am. Chem. Soc., 2023, 145, 10259–10267.
224. Y. Wu, J. Lv, F. Xie, R. An, J. Zhang, H. Huang, Z. Shen, L. Jiang, M. Xu, Q. Yao and Y. Cao, J. Colloid Interface Sci., 2024, 656, 155–167.
225. X. Zheng, Y. Xue, C. Zhang and Y. Li, CCS Chem., 2023, 5, 1653–1662.
226. B. Chang, L. Li, D. Shi, H. Jiang, Z. Ai, S. Wang, Y. Shao, J. Shen, Y. Wu, Y. Li and X. Hao, Appl. Catal., B, 2021, 283, 119622.
227. W. Lin, H. Chen, G. Lin, S. Yao, Z. Zhang, J. Qi, M. Jing, W. Song, J. Li, X. Liu, J. Fu and S. Dai, Angew. Chem., Int. Ed., 2022, 61, e202207807.
228. Y. Li, C. Cheng, S. Han, Y. Huang, X. Du, B. Zhang and Y. Yu, ACS Energy Lett., 2022, 7, 1187–1194.
229. D. Wang, Z.-W. Chen, K. Gu, C. Chen, Y. Liu, X. Wei, C. V. Singh and S. Wang, J. Am. Chem. Soc., 2023, 145, 6899–6904.
230. J.-Y. Fang, Q.-Z. Zheng, Y.-Y. Lou, K.-M. Zhao, S.-N. Hu, G. Li, O. Akdim, X.-Y. Huang and S.-G. Sun, Nat. Commun., 2022, 13, 7899.
231. Y. Wang, H. Li, W. Zhou, X. Zhang, B. Zhang and Y. Yu, Angew. Chem., Int. Ed., 2022, 61, e202202604.
232. T. Hu, M. Wang, C. Guo and C. M. Li, J. Mater. Chem. A, 2022, 10, 8923–8931.
233. J. Cai, J. Huang, A. Cao, Y. Wei, H. Wang, X. Li, Z. Jiang, G. I. N. Waterhouse, S. Lu and S.-Q. Zang, Appl. Catal., B, 2023, 328, 122473.
234. X. Luo, Y. Wu, H. Hu, T. Wei, B. Wu, J. Ding, Q. Liu, J. Luo and X. Liu, Small, 2024, 20, 2403399.
235. G. Zhang, X. Li, K. Chen, Y. Guo, D. Ma and K. Chu, Angew. Chem., Int. Ed., 2023, 62, e202300054.
236. J. Liang, S. Ma, J. Li, Y. Wang, J. Wu, Q. Zhang, Z. Liu, Z. Yang, K. Qu and W. Cai, J. Mater. Chem. A, 2020, 8, 10426–10432.
237. M. Yu, H. Huang, J. Hu, S. Wang, J. Li and D. Wang, J. Mater. Chem. A, 2022, 10, 23990–23997.
238. G. Lin, Q. Ju, X. Guo, W. Zhao, S. Adimi, J. Ye, Q. Bi, J. Wang, M. Yang and F. Huang, Adv. Mater., 2021, 33, 2007509.
239. R. Luo, B.-J. Li, Z.-P. Wang, M.-G. Chen, G.-L. Zhuang, Q. Li, J.-P. Tong, W.-T. Wang, Y.-H. Fan and F. Shao, JACS Au, 2024, 4, 3823–3832.
240. Y. Lv, S.-W. Ke, Y. Gu, B. Tian, L. Tang, P. Ran, Y. Zhao, J. Ma, J.-L. Zuo and M. Ding, Angew. Chem., Int. Ed., 2023, 62, e202305246.
241. J. Ma, Y. Zhang, B. Wang, Z. Jiang, Q. Zhang and S. Zhuo, ACS Nano, 2023, 17, 6687–6697.
242. A. K. K. Padinjareveetil, J. V. Perales-Rondon, D. Zaoralová, M. Otyepka, O. Alduhaish and M. Pumera, ACS Appl. Mater. Interfaces, 2023, 15, 47294–47306.
243. J. V. Perales-Rondon, D. Rojas, W. Gao and M. Pumera, ACS Sustainable Chem. Eng., 2023, 11, 6923–6931.
244. W. Gao, J. V. Perales-Rondon, J. Michalička and M. Pumera, Appl. Catal., B, 2023, 330, 122632.
245. W. Gao, J. Michalička and M. Pumera, J. Mater. Chem. A, 2024, 12, 32458–32470.
246. Z. Huang, M. Rafiq, A. R. Woldu, Q.-X. Tong, D. Astruc and L. Hu, Coord. Chem. Rev., 2023, 478, 214981.
247. C. Tang and S.-Z. Qiao, Chem. Soc. Rev., 2019, 48, 3166–3180.
248. F. F. Tao, L. Nguyen and S. Zhang, Metal nanoparticles for catalysis: advances and applications, The Royal Society of Chemistry, 2014.
249. B. H. Ko, B. Hasa, H. Shin, Y. Zhao and F. Jiao, J. Am. Chem. Soc., 2022, 144, 1258–1266.
250. J. Long, S. Chen, Y. Zhang, C. Guo, X. Fu, D. Deng and J. Xiao, Angew. Chem., Int. Ed., 2020, 59, 9711–9718.
251. J. Lim, C.-Y. Liu, J. Park, Y.-H. Liu, T. P. Senftle, S. W. Lee and M. C. Hatzell, ACS Catal., 2021, 11, 7568–7577.
252. T. Hu, C. Wang, M. Wang, C. M. Li and C. Guo, ACS Catal., 2021, 11, 14417–14427.
253. J. Wang, Y. Wang, C. Cai, Y. Liu, D. Wu, M. Wang, M. Li, X. Wei, M. Shao and M. Gu, Nano Lett., 2023, 23, 1897–1903.
254. W. Peng, M. Luo, X. Xu, K. Jiang, M. Peng, D. Chen, T.-S. Chan and Y. Tan, Adv. Energy Mater., 2020, 10, 2001364.
255. Y. Xu, K. Ren, T. Ren, M. Wang, Z. Wang, X. Li, L. Wang and H. Wang, Appl. Catal., B, 2022, 306, 121094.
256. Y. Han, X. Zhang, W. Cai, H. Zhao, Y. Zhang, Y. Sun, Z. Hu, S. Li, J. Lai and L. Wang, J. Colloid Interface Sci., 2021, 600, 620–628.
257. L. Lv, Y. Shen, J. Liu, X. Meng, X. Gao, M. Zhou, Y. Zhang, D. Gong, Y. Zheng and Z. Zhou, J. Phys. Chem. Lett., 2021, 12, 11143–11150.
258. G. A. Cerrón-Calle, A. S. Fajardo, C. M. Sánchez-Sánchez and S. Garcia-Segura, Appl. Catal., B, 2022, 302, 120844.
259. N. Zhou, Z. Wang, N. Zhang, D. Bao, H. Zhong and X. Zhang, ACS Catal., 2023, 13, 7529–7537.
260. M. Yang, T. Wei, J. He, Q. Liu, L. Feng, H. Li, J. Luo and X. Liu, Nano Res., 2024, 17, 1209–1216.
261. X. Qu, L. Shen, Y. Mao, J. Lin, Y. Li, G. Li, Y. Zhang, Y. Jiang and S. Sun, ACS Appl. Mater. Interfaces, 2019, 11, 31869–31877.
262. Z.-Y. Wu, M. Karamad, X. Yong, Q. Huang, D. A. Cullen, P. Zhu, C. Xia, Q. Xiao, M. Shakouri, F.-Y. Chen, J. Y. T. Kim, Y. Xia, K. Heck, Y. Hu, M. S. Wong, Q. Li, I. Gates, S. Siahrostami and H. Wang, Nat. Commun., 2021, 12, 2870.
263. W.-D. Zhang, H. Dong, L. Zhou, H. Xu, H.-R. Wang, X. Yan, Y. Jiang, J. Zhang and Z.-G. Gu, Appl. Catal., B, 2022, 317, 121750.
264. J. Yang, H. Qi, A. Li, X. Liu, X. Yang, S. Zhang, Q. Zhao, Q. Jiang, Y. Su, L. Zhang, J.-F. Li, Z.-Q. Tian, W. Liu, A. Wang and T. Zhang, J. Am. Chem. Soc., 2022, 144, 12062–12071.
265. J. Cheng, W. Sun, G. Dai, X. Yang, R. Xia, Y. Xu, X. Yang and W. Tu, Fuel, 2023, 332, 126106.
266. T. Xiang, Y. Liang, Y. Zeng, J. Deng, J. Yuan, W. Xiong, B. Song, C. Zhou and Y. Yang, Small, 2023, 19, 2303732.
267. P. Guo, D. Luan, H. Li, L. Li, S. Yang and J. Xiao, J. Am. Chem. Soc., 2024, 146, 13974–13982.
268. J. Deng, Y. Su, D. Liu, P. Yang, B. Liu and C. Liu, Chem. Rev., 2019, 119, 9221–9259.
269. S. Garcia-Segura, M. Lanzarini-Lopes, K. Hristovski and P. Westerhoff, Appl. Catal., B, 2018, 236, 546–568.
270. H. Jiang, G.-F. Chen, O. Savateev, J. Xue, L.-X. Ding, Z. Liang, M. Antonietti and H. Wang, Angew. Chem., Int. Ed., 2023, 62, e202218717.
271. F. Gou, H. Wang, M. Fu, Y. Jiang, W. Shen, R. He and M. Li, Appl. Surf. Sci., 2023, 612, 155872.
272. Q. He, S. Qiao, Y. Zhou, R. Vajtai, D. Li, P. M. Ajayan, L. Ci and L. Song, Adv. Funct. Mater., 2022, 32, 2106684.
273. J. O. Olowoyo and R. J. Kriek, Small, 2022, 18, 2203125.
274. Z. Deng, C. Ma, Z. Li, Y. Luo, L. Zhang, S. Sun, Q. Liu, J. Du, Q. Lu, B. Zheng and X. Sun, ACS Appl. Mater. Interfaces, 2022, 14, 46595–46602.
275. J. Yuan, H. Yin, X. Jin, D. Zhao, Y. Liu, A. Du, X. Liu and A. P. O’Mullane, Appl. Catal., B, 2023, 325, 122353.
276. A. Ambrosi, C. K. Chua, N. M. Latiff, A. H. Loo, C. H. A. Wong, A. Y. S. Eng, A. Bonanni and M. Pumera, Chem. Soc. Rev., 2016, 45, 2458–2493.
277. C. Huang, Y. Li, N. Wang, Y. Xue, Z. Zuo, H. Liu and Y. Li, Chem. Rev., 2018, 118, 7744–7803.
278. H. Yu, Y. Xue, L. Hui, C. Zhang, Y. Fang, Y. Liu, X. Chen, D. Zhang, B. Huang and Y. Li, Natl. Sci. Rev., 2021, 8, nwaa213.
279. H. Zou, W. Rong, S. Wei, Y. Ji and L. Duan, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 29462–29468.
280. X. Yu, P. Han, Z. Wei, L. Huang, Z. Gu, S. Peng, J. Ma and G. Zheng, Joule, 2018, 2, 1610–1622.
281. S. Wang, C. Song, Y. Cai, Y. Li, P. Jiang, H. Li, B. Yu and T. Ma, Adv. Energy Mater., 2023, 13, 2301136.
282. L. Verger, V. Natu, M. Carey and M. W. Barsoum, Trends Chem., 2019, 1, 656–669.
283. Y. Li, J. Ma, T. D. Waite, M. R. Hoffmann and Z. Wang, Environ. Sci. Technol., 2021, 55, 10695–10703.
284. Y. Xiao and C. Shen, Small, 2021, 17, 2100776.
285. L. Zhang, J. Liang, Y. Wang, T. Mou, Y. Lin, L. Yue, T. Li, Q. Liu, Y. Luo, N. Li, B. Tang, Y. Liu, S. Gao, A. A. Alshehri, X. Guo, D. M. Ma and X. Sun, Angew. Chem., Int. Ed., 2021, 60, 25263–25268.
286. S. Zhang, M. Li, J. Li, Q. Song and X. Liu, Proc. Natl. Acad. Sci. U. S. A., 2022, 119, e2115504119.
287. D. Pan, M. Austeria P, S. Lee, H.-S. Bae, F. He, G. H. Gu and W. Choi, Nat. Commun., 2024, 15, 7243.
288. Y. Ying, M. P. Browne and M. Pumera, Sustainable Energy Fuels, 2020, 4, 3732–3738.
289. K. P. A. Kumar, K. Ghosh, O. Alduhaish and M. Pumera, Electrochem. Commun., 2021, 122, 106890.
290. E. Redondo and M. Pumera, Appl. Mater. Today, 2021, 25, 101253.
291. C. Didier, A. Kundu and S. Rajaraman, Microsyst. Nanoeng., 2020, 6, 15.
292. C. Iffelsberger, S. Ng and M. Pumera, Appl. Mater. Today, 2020, 20, 100654.
293. V. Urbanová, J. Plutnar and M. Pumera, Appl. Mater. Today, 2021, 24, 101131.
294. G. Bharath, C. Liu, F. Banat, A. Kumar, A. Hai, A. K. Nadda, V. K. Gupta, M. A. Haija and J. Balamurugan, Chem. Eng. J., 2023, 465, 143040.
295. V. R. Silveira, R. Bericat-Vadell and J. Sá, J. Phys. Chem. C, 2023, 127, 5425–5431.
296. N. Gao, H. Yang, D. Dong, D. Dou, Y. Liu, W. Zhou, F. Gao, C. Nan, Z. Liang and D. Yang, J. Colloid Interface Sci., 2022, 611, 294–305.
297. C. H. Kim, J. Kim, F. Hollmann and C. B. Park, Appl. Catal., B, 2023, 336, 122925.
298. F. Xu, F. Wu, K. Zhu, Z. Fang, D. Jia, Y. Wang, G. Jia, J. Low, W. Ye, Z. Sun, P. Gao and Y. Xiong, Appl. Catal., B, 2021, 284, 119689.
299. S. Zhou, K. Sun, C. Y. Toe, J. Yin, J. Huang, Y. Zeng, D. Zhang, W. Chen, O. F. Mohammed, X. Hao and R. Amal, Adv. Mater., 2022, 34, 2201670.
300. L. Paramanik, S. Sultana and K. M. Parida, J. Colloid Interface Sci., 2022, 625, 83–99.
301. X. Li, W. Fan, Y. Bai, Y. Liu, F. Wang, H. Bai and W. Shi, Chem. Eng. J., 2022, 433, 133225.
302. D. Liu, J. Wang, S. Bian, Q. Liu, Y. Gao, X. Wang, P. K. Chu and X.-F. Yu, Adv. Funct. Mater., 2020, 30, 2002731.
303. Y. Bai, S. Gao, W. Xie, Z. Fang, H. Bai and W. Fan, Int. J. Hydrogen Energy, 2023, 48, 10882–10890.
304. P. Li, Y. Liu, M. A. Mushtaq and D. Yan, Inorg. Chem. Front., 2023, 10, 4650–4667.
305. P. H. van Langevelde, I. Katsounaros and M. T. M. Koper, Joule, 2021, 5, 290–294.
306. U. S. Environmental Protection Agency, Drinking Water Standards and Health Advisories, EPA 822-F-18-001, 2018, https://www.epa.gov/system/files/documents/2022-2001/dwtable2018.pdf.
307. L. Su, D. Han, G. Zhu, H. Xu, W. Luo, L. Wang, W. Jiang, A. Dong and J. Yang, Nano Lett., 2019, 19, 5423–5430.
308. M. Duca and M. T. M. Koper, Energy Environ. Sci., 2012, 5, 9726–9742.
309. S. Lee, S. Kim, C. Park, W. Kim, S. Ryu and W. Choi, Energy Environ. Sci., 2021, 14, 4437–4450.
310. C. Park, H. Kwak, G.-H. Moon and W. Kim, J. Mater. Chem. A, 2021, 9, 19179–19205.
311. J. M. McEnaney, S. J. Blair, A. C. Nielander, J. A. Schwalbe, D. M. Koshy, M. Cargnello and T. F. Jaramillo, ACS Sustainable Chem. Eng., 2020, 8, 2672–2681.
312. W. Zheng, L. Zhu, Z. Yan, Z. Lin, Z. Lei, Y. Zhang, H. Xu, Z. Dang, C. Wei and C. Feng, Environ. Sci. Technol., 2021, 55, 13231–13243.
313. N. C. Kani, J. A. Gauthier, A. Prajapati, J. Edgington, I. Bordawekar, W. Shields, M. Shields, L. C. Seitz, A. R. Singh and M. R. Singh, Energy Environ. Sci., 2021, 14, 6349–6359.
314. X. Wu, Z. Song, Z. Liu, X. Tang, F. Yao, F. Zhao, X. Min and C.-J. Tang, Appl. Catal., B, 2024, 359, 124467.
315. F.-Y. Chen, A. Elgazzar, S. Pecaut, C. Qiu, Y. Feng, S. Ashokkumar, Z. Yu, C. Sellers, S. Hao, P. Zhu and H. Wang, Nat. Catal., 2024, 7, 1032–1043.
316. Q. Zhang, Y. Li, M. Geng, J. Zhu, H. Sun and B. Jiang, Appl. Catal., B, 2023, 330, 122658.
317. G. Zhang, B. Li, Y. Shi, Q. Zhou, W.-J. Fu, G. Zhou, J. Ma, S. Yin, W. Yuan, S. Miao, Q. Ji, J. Qu and H. Liu, Nat. Sustainable, 2024, 7, 1251–1263.
318. M. Jiang, M. Zhu, M. Wang, Y. He, X. Luo, C. Wu, L. Zhang and Z. Jin, ACS Nano, 2023, 17, 3209–3224.
319. A. Khan, A. Abbas and R. Dickson, Green Chem., 2024, 26, 1551–1565.
320. Y. Wan, M. Zheng, W. Yan, J. Zhang and R. Lv, Adv. Energy Mater., 2024, 14, 2303588.
321. J. Leverett, T. Tran-Phu, J. A. Yuwono, P. Kumar, C. Kim, Q. Zhai, C. Han, J. Qu, J. Cairney, A. N. Simonov, R. K. Hocking, L. Dai, R. Daiyan and R. Amal, Adv. Energy Mater., 2022, 12, 2201500.
322. J. Zheng, S. Xu, J. Sun, J. Zhang, L. Sun, X. Pan, L. Li and G. Zhao, Appl. Catal., B, 2023, 338, 123056.
323. Y. Luo, K. Xie, P. Ou, C. Lavallais, T. Peng, Z. Chen, Z. Zhang, N. Wang, X.-Y. Li, I. Grigioni, B. Liu, D. Sinton, J. B. Dunn and E. H. Sargent, Nat. Catal., 2023, 6, 939–948.
324. D.-S. Huang, X.-F. Qiu, J.-R. Huang, M. Mao, L. Liu, Y. Han, Z.-H. Zhao, P.-Q. Liao and X.-M. Chen, Nat. Synth., 2024, 3, 1404–1413.
325. H. Xiong, P. Yu, K. Chen, S. Lu, Q. Hu, T. Cheng, B. Xu and Q. Lu, Nat. Catal., 2024, 7, 785–795.
326. N. Meng, X. Ma, C. Wang, Y. Wang, R. Yang, J. Shao, Y. Huang, Y. Xu, B. Zhang and Y. Yu, ACS Nano, 2022, 16, 9095–9104.
327. X. Liu, P. V. Kumar, Q. Chen, L. Zhao, F. Ye, X. Ma, D. Liu, X. Chen, L. Dai and C. Hu, Appl. Catal., B, 2022, 316, 121618.
328. J. Mukherjee, S. Paul, A. Adalder, S. Kapse, R. Thapa, S. Mandal, B. Ghorai, S. Sarkar and U. K. Ghorai, Adv. Funct. Mater., 2022, 32, 2200882.
329. C. Lv, L. Zhong, H. Liu, Z. Fang, C. Yan, M. Chen, Y. Kong, C. Lee, D. Liu, S. Li, J. Liu, L. Song, G. Chen, Q. Yan and G. Yu, Nat. Sustainable, 2021, 4, 868–876.
330. X. Wei, X. Wen, Y. Liu, C. Chen, C. Xie, D. Wang, M. Qiu, N. He, P. Zhou, W. Chen, J. Cheng, H. Lin, J. Jia, X.-Z. Fu and S. Wang, J. Am. Chem. Soc., 2022, 144, 11530–11535.
331. M. Yuan, J. Chen, Y. Bai, Z. Liu, J. Zhang, T. Zhao, Q. Wang, S. Li, H. He and G. Zhang, Angew. Chem., Int. Ed., 2021, 60, 10910–10918.
332. Y. Wang, S. Xia, R. Cai, J. Zhang, C. Yu, J. Cui, Y. Zhang, J. Wu and Y. Wu, Angew. Chem., Int. Ed., 2024, 63, e202318589.
333. Y. Yang, J. Peng, Z. Shi, P. Zhang, A. Arramel and N. Li, J. Mater. Chem. A, 2023, 11, 6428–6439.
334. Y. Zhang, L. Lu, T. Zhao, J. Zhao, Q. Cai and Z. Chen, J. Mater. Chem. A, 2024, 12, 16704–16715.
335. B. Govindan, K. Annamalai, A. Kumar, S. Palanisamy, M. Abu Haija and F. Banat, ACS Sustainable Chem. Eng., 2024, 12, 8174–8187.
336. X. Zhu, X. Zhou, Y. Jing and Y. Li, Nat. Commun., 2021, 12, 4080.
337. X. Cao, D. Zhang, Y. Gao, O. V. Prezhdo and L. Xu, J. Am. Chem. Soc., 2024, 146, 1042–1052.
338. H. Cai, Z. Wang, G. Meng, T. Wei, Y. Liu, J. Luo, Q. Liu, G. Hu and X. Liu, Inorg. Chem., 2024, 63, 20935–20939.
339. P. Li, Z. Jin, Z. Fang and G. Yu, Energy Environ. Sci., 2021, 14, 3522–3531.
340. Z. Fang, Z. Jin, S. Tang, P. Li, P. Wu and G. Yu, ACS Nano, 2022, 16, 1072–1081.
341. J.-Y. Zhu, Q. Xue, Y.-Y. Xue, Y. Ding, F.-M. Li, P. Jin, P. Chen and Y. Chen, ACS Appl. Mater. Interfaces, 2020, 12, 14064–14070.
342. X. Deng, Y. Yang, L. Wang, X.-Z. Fu and J.-L. Luo, Adv. Sci., 2021, 8, 2004523.
343. K. Fan, W. Xie, J. Li, Y. Sun, P. Xu, Y. Tang, Z. Li and M. Shao, Nat. Commun., 2022, 13, 7958.
344. X.-B. Li, C.-H. Tung and L.-Z. Wu, Angew. Chem., Int. Ed., 2019, 58, 10804–10811.
345. E. A. Weiss, ACS Energy Lett., 2017, 2, 1005–1013.
346. S.-L. Meng, J.-H. Li, C. Ye, Y.-L. Yin, X.-L. Zhang, C. Zhang, X.-B. Li, C.-H. Tung and L.-Z. Wu, Adv. Mater., 2024, 36, 2311982.
347. Z. Niu, S. Fan, X. Li, P. Wang, Z. Liu, J. Wang, C. Bai and D. Zhang, Chem. Eng. J., 2022, 450, 138343.
348. S. Verma, S. Lu and P. J. A. Kenis, Nat. Energy, 2019, 4, 466–474.
349. Á. Balog, E. Kecsenovity, G. F. Samu, J. He, D. Fekete and C. Janáky, Nat. Catal., 2024, 7, 522–535.
350. J. Kim, J.-A. Lin, J. Kim, I. Roh, S. Lee and P. Yang, Nat. Catal., 2024, 7, 977–986.
351. W. Zhu, X. Zhang, F. Yao, R. Huang, Y. Chen, C. Chen, J. Fei, Y. Chen, Z. Wang and H. Liang, Angew. Chem., Int. Ed., 2023, 62, e202300390.
352. S. Li, P. Ma, C. Gao, L. Liu, X. Wang, M. Shakouri, R. Chernikov, K. Wang, D. Liu, R. Ma and J. Wang, Energy Environ. Sci., 2022, 15, 3004–3014.
353. S. Ye, Z. Chen, G. Zhang, W. Chen, C. Peng, X. Yang, L. Zheng, Y. Li, X. Ren, H. Cao, D. Xue, J. Qiu, Q. Zhang and J. Liu, Energy Environ. Sci., 2022, 15, 760–770.
354. M. Yang, T. Wei, C. Zeng, J. Zhang, Y. Liu, J. Luo, G. Hu and X. Liu, Chem. Eng. J., 2024, 498, 155799.
355. R. Zaffaroni, D. Ripepi, J. Middelkoop and F. M. Mulder, ACS Energy Lett., 2020, 5, 3773–3777.
356. C. Zhang, J. Li, R. Chen, S. Shen, J. Wang, Y. Sun and F. Dong, ACS Catal., 2024, 14, 15721–15742.
357. Y. Zhao, R. Shi, X. Bian, C. Zhou, Y. Zhao, S. Zhang, F. Wu, G. I. N. Waterhouse, L.-Z. Wu, C.-H. Tung and T. Zhang, Adv. Sci., 2019, 6, 1802109.
358. N. Amornthammarong, J.-Z. Zhang and P. B. Ortner, Anal. Methods, 2011, 3, 1501–1506.
359. S. Z. Andersen, V. Čolić, S. Yang, J. A. Schwalbe, A. C. Nielander, J. M. McEnaney, K. Enemark-Rasmussen, J. G. Baker, A. R. Singh, B. A. Rohr, M. J. Statt, S. J. Blair, S. Mezzavilla, J. Kibsgaard, P. C. K. Vesborg, M. Cargnello, S. F. Bent, T. F. Jaramillo, I. E. L. Stephens, J. K. Nørskov and I. Chorkendorff, Nature, 2019, 570, 504–508.
360. J. B. Hansen and P. Han, Green Ammonia by Haldor Topsoe, Department of Energy, 2021, https://www.energy.gov/sites/default/files/2021-2008/2024-green-ammonia-haldor-topsoe.pdf.
361. Small-Scale Green Ammonia Plants Open up New Storage Possibilities for Wind and Solar Power, ThyssenKrupp Industrial Solutions, 2022, https://insights.thyssenkrupp-industrial-solutions.com/story/small-scale-green-ammonia-plants-open-up-new-storage-possibilities-for-wind-and-solar-power/.
362. Yara Selects Linde Engineering to Build Electrolysis Plant at Porsgrunn, Ammonia Energy Association, 2022, https://www.ammoniaenergy.org/articles/yara-selects-linde-engineering-to-build-electrolysis-plant-at-porsgrunn/.
363. S. Wu, N. Salmon, M. M.-J. Li, R. Bañares-Alcántara and S. C. E. Tsang, ACS Energy Lett., 2022, 7, 1021–1033.
364. B. Bonnet-Cantalloube, M. Espitalier-Noël, P. F. de Carvalho, J. Fonseca and G. Pawelec, Clean ammonia in the future energy system, Hydrogen Europe, 2023, https://hydrogeneurope.eu/wp-content/uploads/2023/2003/2023.<?pdb_no 2003?>2003<?pdb<?db_id PDB?> END?>_H2022Europe_Clean_Ammonia_Report_DIGITAL_FINAL.pdf.
365. ENGIE-YARA, Renewable Hydrogen and Ammonia Deployment in Pilbara, YURI Phase 0: Feasibility study public report, 2020, https://arena.gov.au/assets/2020/2011/engie-yara-renewable-hydrogen-and-ammonia-deployment-in-pilbara.pdf.
366. Kapsom, The World's First Green Ammonia Plant Was Made By Kapsom, Green ammonia produced from renewable,to a renewable energy drive, 2022, https://www.kapsom.com/the-worlds-first-green-ammonia-plant-was-made-by-kapsom/.
367. W. Bi, E. Gyenge and D. P. Wilkinson, Chem. Eng. J., 2023, 478, 147359.
368. H2F project, Green hydrogen, green ammonia and green fertiliser plant, https://www.fertiberia.com/en/greenammonia/h2f-project/.
369. P. Mayer, A. Ramirez, G. Pezzella, B. Winter, S. M. Sarathy, J. Gascon and A. Bardow, iScience, 2023, 26, 107389.
370. M.-H. Yuan, Y.-H. Chen, J.-Y. Tsai and C.-Y. Chang, Process Saf. Environ. Prot., 2016, 102, 777–785.
371. L. Zeng, C. Mangan and X. Li, Water Sci. Technol., 2006, 54, 137–145.
372. J. Gao, Q. Ma, Z. Wang, B. E. Rittmann and W. Zhang, Nat. Commun., 2024, 15, 8455.
373. L. M. Wilder, K. Wyatt, C. A. Skangos, W. E. Klein, M. R. Parimuha, J. L. Katsirubas, J. L. Young and E. M. Miller, ACS Appl. Energy Mater., 2024, 7, 536–545.
374. K. Kim, C.-Y. Yoo, J.-N. Kim, H. C. Yoon and J.-I. Han, J. Electrochem. Soc., 2016, 163, F1523.
375. S. Li, D. Han, G. Jiang, Z. Han, H. Lu, J. Gao, X. Wang, Y. Wang, C. Geng, Z. Weng and Q.-H. Yang, ACS Appl. Energy Mater., 2023, 6, 5067–5073.
376. W. Liao, J. Wang, G. Ni, K. Liu, C. Liu, S. Chen, Q. Wang, Y. Chen, T. Luo, X. Wang, Y. Wang, W. Li, T.-S. Chan, C. Ma, H. Li, Y. Liang, W. Liu, J. Fu, B. Xi and M. Liu, Nat. Commun., 2024, 15, 1264.
377. S.-N. Zhang, P. Gao, Q.-Y. Liu, Z. Zhang, B.-L. Leng, J.-S. Chen and X.-H. Li, Nat. Commun., 2024, 15, 10877.
378. Z. Xu, L. Wan, Y. Liao, M. Pang, Q. Xu, P. Wang and B. Wang, Nat. Commun., 2023, 14, 1619.
379. J. Lan, Z. Wang, C.-W. Kao, Y.-R. Lu, F. Xie and Y. Tan, Nat. Commun., 2024, 15, 10173.
380. L. Mi, Q. Huo, J. Cao, X. Chen, H. Yang, Q. Hu and C. He, Adv. Energy Mater., 2022, 12, 2202247.

---