Comprehensive review on single-atom catalysts in electrochemical hydrogen-evolution reaction: computational modelling and experimental investigation

ABSTRACT

The growing demand for renewable energy sources has driven research efforts towards highly efficient electrochemical processes, especially the Hydrogen Evolution Reaction (HER) which is an essential component in renewable hydrogen production. Single Atom Catalysts (SACs) have been recognised as promising catalysts for exceptional HER activities owing to their unique electronic and geometrical features. This comprehensive review highlights the effective correlation between computational modelling and experimental research in fabricating new SACs and better understanding their catalytic mechanisms in the HER. Furthermore, the review briefly describes the choice of metals and tailoring the electronic structure, as well as exploring the influence of the reaction parameters on the catalytic activity of SACs for HER. Concluding remarks alongside future perspectives on the improvement of SACs for HER are emphasised. It is hoped that this review will pave the way for rational design of highly effective, low-cost and stable SACs for sustainable hydrogen production.

KEYWORDS:

• SACs
• HER
• DFT modelling
• electronic structure
• catalytic mechanism

1. Introduction

Energy and environmental crises are global priorities that require immediate attention and proper solutions [1, 2]. Energy consumption, mostly from limited fossil fuels, has raised concerns about resource depletion, geopolitical conflicts, and greenhouse gas emissions. These emissions cause climate change, which threatens humankind through rising temperatures, harsh weather, and ecosystem disruption [3–6]. A rapid shift to sustainable, renewable energy sources is necessary to mitigate climate change, optimise the supply of energy, and promote accessibility to clean energy. Technology, regulatory reforms, and an ongoing dedication to a greener, more sustainable future are needed to balance energy needs with sustainability [7, 8].

In the pursuit of a sustainable energy future, the electrochemical hydrogen evolution reaction (HER) is crucial for harnessing the potential of hydrogen as a clean and efficient energy carrier [9–12]. Advanced catalysts have proven critical in energy conversion technologies to tackle global warming and fossil fuel depletion [13–17]. Recently, Single Atom Catalysts (SACs) have revolutionised catalysis, inspiring researchers with their unique features and efficiency. SACs are made up of individual metal atoms dispersed on a support material [18–20]. SACs can optimise the HER catalytic activity and minimise the use of less abundant precious metals. SACs have a distinct advantage over the bulk catalysts owing to their large surface area, controllable electronic structure, and improved catalytic performance. These features arise from the precisely controlled active sites at the atomic level, which allow for optimal interactions with reactants [21]. However, there still exist some limitations such as the stability issue of SACs. The single atoms on SACs always agglomerate and make clusters, which remarkably decreases the catalytic performance of the catalyst. Further, the reactivity and electronic structure of SACs are highly sensitive to the local environment, rendering them sensitive to deactivation or poisoning by the reactants and intermediates. Moreover, it is difficult to maintain their structural integrity under harsh reaction conditions [19].

In the field of sustainable energy, HER is a vital step for hydrogen production, and the design of highly efficient SACs offers a possible route for improving the efficiency and cost-effectiveness of this fundamental reaction, thus leading to the development of clean energy technologies [22–27]. For instance, Chen et al.[28] anchored single molybdenum (Mo) atoms on N-doped graphene, which revealed exceptional performance compared to bulk Mo₂C and MoN catalysts and even better stability than the commercial Pt/C. The estimated Gibbs free energy (ΔG_H*) of the Mo₁N₁C₂ site was close to zero, indicating that the as-fabricated SAC is highly efficient for H₂ production. The estimated density of states (DOS) of the Mo₁N₁C₂ site further revealed that anchoring Mo atoms onto the surface of graphene can remarkably improve the domination of d-electron near the Fermi level, thereby enhancing the catalytic performance. According to Qiu et al.[29] single nickel (Ni) atom incorporated into graphene by substituting one carbon (C) site within the graphene lattice revealed exceptional HER performance and cyclic stability. DFT simulations validate that the substitutional dopants, specifically those occupying the C sites in the graphene lattice (Ni_sub), reveal exceptional HER performance compared to the Ni atom attached to the defect sites of graphene (Ni_def). Beside these, MoS₂-based SACs with unique designs and electronic features exhibited improved catalytic activity for the HER compared to pristine MoS₂. This has drawn significant attention of many researchers. It was confirmed via the both theoretical and experimental investigations that the coordinatively unsaturated sulfur (S) atoms found at the margins of the pristine MoS₂ catalyst serve as the active site for the HER, whereas the basal surfaces of the catalyst were not catalytically active. Activating the in-plane S atoms is an effective method for enhancing the HER activity of MoS₂, on account of the limited number of edge sites and an excess of unutilised S atoms within the basal plane of MoS₂ [30, 31].

Several review articles have been reported on the design and fabrication of SACs for HER [32–35]. Different from those reviews, this minireview aims to provide insights into the effective correlation between computational modelling and experimental research in elaborating the catalytic mechanisms of SACs in the HER. In addition, the review briefly describes about the tuning of electronic structure and metal atom choice, as well as the influence of the reaction parameters on the HER catalytic activity of SACs. Finally, the conclusion and future perspectives on the improvement of SACs for HER are highlighted. The review aims to pave the way for rational design of highly effective, low-cost and stable SACs for sustainable hydrogen production.

2. Theoretical insights of SACs in HER

At the forefront of electrochemical water splitting, the HER is a critical component in renewable energy research. SACs have gained significant attention for their exceptional efficiency and specificity in catalysing HER [36]. The introduction of DFT into this field has been transformative, providing profound insights into the atomic-level interactions and electronic properties of SACs. DFT, a quantum mechanical modelling methodology, is pivotal in investigating the electronic structures of materials, especially in the context of SACs for HER [36, 37]. It plays an active role in predicting and elucidating the behaviours of these catalysts. This theory is crucial for understanding how single atoms interact within their environments and serves as a foundational tool for the efficient design and optimisation of catalysts. The HER process involves the adsorption of hydrogen ions from an electrolyte and their reduction to form hydrogen gas [38]. In SACs, this process is facilitated by single catalytic atoms, typically transition metals, which are arrayed on suitable substrates as illustrated in Figure 1 [36]. DFT is key in clarifying the electronic dynamics at these active sites, thereby explaining the heightened catalytic activity of SACs compared to their bulk counterparts.

Figure 1. Descriptors for hydrogen evolution on single-atom catalysts in nitrogen-doped graphene. Reproduced with permission from ref.[36] Copyright 2020, The American Chemical Society.

A significant contribution of DFT in SACs research is the analysis of their electronic structure [12]. DFT calculations reveal insights into the atomic electronic configurations, energy levels, and how these factors influence reactivity towards hydrogen adsorption and desorption [12, 39]. Understanding the electronic structure allows researchers to predict and tailor the catalytic actions of diverse SACs, enhancing their effectiveness for HER [12]. The theoretical insights provided by DFT are invaluable in the design and optimisation of SACs. The DFT simulations enable the modelling of various single-atom and substrate combinations, predicting their interactions and resultant catalytic activity [40]. This foresight is crucial in identifying promising SAC candidates, significantly streamlining the research process.

Despite its strengths, DFT faces challenges, particularly in accurately modelling dispersion interactions and electron correlation effects. Dispersion interactions refer to the attractive forces between molecules or atoms caused by fluctuations in electron distributions. These interactions are essential in understanding the behaviour of molecules and materials, particularly in condensed phases and at interfaces. Whereas, electron correlation effects refer to the mutual influence of electrons on each other's motion owing to their repulsive nature, which cannot be fully captured by mean-field theories such as DFT. These effects become significant in systems with strong electron–electron interactions, such as transition-metal complexes and correlated electronic materials. These aspects are crucial for a true representation of SACs, necessitating the development of new functionals and computational methods to enhance the accuracy of DFT simulations. To address the limitations of traditional DFT, there are continuous advancements in computational techniques. Innovations like hybrid functionals and machine learning algorithms, along with more sophisticated modelling approaches, are enhancing the accuracy of catalytic behaviour predictions. These advancements are deepening our understanding of existing SACs and opening avenues for the discovery of novel catalytic materials.

The full potential of DFT is realised when it is integrated with experimental studies. Computational predictions guide the experimental synthesis and characterisation of SACs, enabling the validation and refinement of theoretical models. This synergy between theoretical and empirical approaches is accelerating the development of effective and robust SACs for HER [37]. In addition, the integration of DFT with other computational and experimental methods holds promising prospects for catalysis research. The fusion of machine learning and artificial intelligence with DFT is particularly exciting. This combination is anticipated to expedite the screening and performance prediction of a broad spectrum of SACs, potentially revolutionising the discovery of more effective catalysts for HER and other critical chemical reactions. The HER proceeds via one-electron transfer mechanism and includes one proton–electron coupled elementary steps (Equations 1(1) $H^{+}+e^{-}+\ast \to H\ast$(1) and 2(2) $H\ast \to {\displaystyle \frac{1}{2}{H}_2+\ast }$(2) ): (1) $H^{+}+e^{-}+\ast \to H\ast$(1) This equation represents the initial step in the HER, where a proton (H⁺) and an electron (e⁻) react with a catalytic site (∗) to form an adsorbed hydrogen atom (H*). This step is crucial as it initiates the overall reaction process on the catalyst surface. (2) $H\ast \to {\displaystyle \frac{1}{2}{H}_2+\ast }$(2) Here, the adsorbed hydrogen atom (H*) undergoes further reaction to produce hydrogen gas (H₂) and regenerate the catalytic site (∗). This step illustrates the desorption of hydrogen from the catalyst surface – a key aspect in HER kinetics. The zero-point energy, enthalpy, and entropy corrections were added to H* to convert electronic energy of H* to free energy. The Gibbs free energy (G) at U = 0 V is modified by Equation (3(3) $G=E_{\mathrm{DFT}}+E_{\mathrm{ZPE}}\text{–}\mathrm{TS}+\int C_p\mathrm{dT}$(3) ). (3) $G=E_{\mathrm{DFT}}+E_{\mathrm{ZPE}}\text{–}\mathrm{TS}+\int C_p\mathrm{dT}$(3) These symbols represent various thermodynamic and kinetic parameters such as the electronic energy (E_DFT) calculated by DFT, the zero-point energy (E_ZPE), the temperature (TS), and the integral of the heat capacity with respect to temperature (∫CpdT). These parameters are essential for determining the energetics and kinetics of the HER process, providing insights into the feasibility and rate of hydrogen evolution. The G of the adsorbed hydrogen Δ(H*) is a key descriptor for the HER catalytic activity close-to-zero value of ΔG(H)* implies that reaction barriers in both adsorption and desorption steps are comprised which favours HER, serving as an indicator for good photocatalyst for HER. Moreover, the Δ(H*) at a given potential U_RHE can be written as (Equation 4(4) $\mathrm{\Delta G}\left(\mathrm{H}\ast \right)=G\left(\mathrm{H}\ast \right)-G(\ast )-{\displaystyle \frac{1}{2}{\mathrm{H}}_2-e{U}_{\mathrm{RHE}}}$(4) ): (4) $\mathrm{\Delta G}\left(\mathrm{H}\ast \right)=G\left(\mathrm{H}\ast \right)-G(\ast )-{\displaystyle \frac{1}{2}{\mathrm{H}}_2-e{U}_{\mathrm{RHE}}}$(4) This equation calculates the Gibbs free energy change (ΔG) associated with the adsorbed hydrogen species ($\mathrm{H}$ *). It considers the difference in the Gibbs free energy between the adsorbed hydrogen ($\mathrm{H}$ *) and the bare catalytic site (∗), as well as the energy change for the formation of hydrogen gas (H₂) from two hydrogen atoms. Additionally, it incorporates the reference potential (eU_RHE) to account for the electrode potential. This equation is fundamental in assessing the thermodynamic driving force for the HER on single-atom catalysts.

Upon formation, the H₂ molecule exhibits weak interactions with the surface via dispersion forces and is subsequently released to the gas phase at finite temperatures. Given the nearly identical energy of the physisorbed intermediate to the separated systems (Metal (M) + H₂), the minimum on the potential energy surface can be neglected. The ΔG (H*) of the reaction is then determined using a singular descriptor – the strength of the M─H bond. This methodology yields a volcano curve (Figure 2) illustrating the relationship between the exchange current (i₀) of the HER and the M─H bond strength, in line with Trasatti's proposition from half a century ago [41].

Figure 2. (a) 2D volcano plot, derived from DFT results on 55 Single Atom Catalysts (SACs) for the HER, assumes an H* intermediate. (b) 3D volcano plot considers both H* and HH (dihydrogen) intermediates. The colour scheme indicates activity levels, with red for high activity, blue for low activity, and instances of extremely low activity (log(i₀) < −10) depicted in black. Reproduced with permission from ref. [42] Copyright 2021, The American Chemical Society.

In short, DFT has established itself as an indispensable component in the study and advancement of SACs for the HER. Its unparalleled ability to provide detailed insights into the electronic structure and reactivity of these catalysts at an atomic scale is remarkable. As computational methodologies continue to evolve, their amalgamation with empirical research is set to drive significant progress in catalysis, leading to more efficient, sustainable, and cost-effective hydrogen production technologies.

3. Computational modelling of SACs

Computational modelling of SACs for HER entails using mathematical and computational tools for simulating and predicting the atomic-level behaviour of these catalysts in facilitating the HER process [43]. Computational modelling of SACs represents a fascinating convergence of theoretical chemistry, material science, and cutting-edge computational technology [44–48]. SACs, where isolated single atoms serve as active catalytic sites on diverse support materials, offer remarkable efficiency and precision in chemical reactions, rendering them highly desirable in fields ranging from renewable energy to pharmaceuticals and environmental remediation [49–52].

A profound grasp of SACs and their distinctive attributes necessitates a deep dive into the microscopic intricacies of these materials. Each individual atom in a SACs possesses its own unique electronic and geometric configurations, profoundly influencing its catalytic behaviour. In stark contrast to traditional catalysts, where only surface atoms engage in reactions, every atom in a SAC has the potential to act as an active site, resulting in unparalleled efficiency and selectivity. At the heart of computational modelling for SACs DFT is employed to explore the electronic structure of atoms and molecules. DFT calculations empower scientists to anticipate how these solitary atoms interact with substrates and reactants, providing invaluable insights into their catalytic potential. This method furnishes a detailed depiction of the electronic distribution around the atom and the associated energy levels critical to catalytic processes [53]. Additionally, SACs are dynamic, with the ability to undergo structural changes during the reaction. It is essential to consider these dynamics when comparing theoretical and experimental results [54, 55]. The dynamic dimension is introduced through Molecular Dynamics (MD) simulations, elucidating the behaviour of atoms and molecules over time. MD simulations complement DFT by offering a real-time view of atomic processes. Theory is instrumental in offering phase diagrams that reveal the stability of distinct structural motifs under diverse conditions. For instance, a study on Rh₁/TiO₂, employing first-principles atomistic thermodynamics, illustrates how redox conditions impact the stability of Rh single atoms [56]. Experimentally observed changes in local coordination and reactivity include the formation of Rh(OH)_ads complexes in reducing H₂ atmospheres and the favoured incorporation of Rh atoms into the oxide lattice under strong oxidising conditions (Figure 3).

Figure 3. The surface stability diagram for a Rh atom on TiO₂(110) displays variations based on Δµ(H) and Δµ(O), representing H and O chemical potentials. Colours denote different configurations: blue, light blue, and green show regions where Rh preferentially substitutes a six-coordinated surface Ti with zero, one, or two O vacancies (Rh₁@TiO₂, Rh₁@TiO_2−x, Rh₁@TiO_2−2x). Orange and pink zones indicate preferences for the supported Rh structure (Rh₁/TiO_2−x and Rh1/TiO_2−2x), respectively. Reproduced with permission from ref. [56]. Copyright 2021, Springer Nature.

Figure 3 presents a 2D surface stability diagram that depicts the relationship between the stability of substitutional and supported sites on the TiO₂(110) surface concerning the chemical potentials of oxygen (Δμ(O)) and hydrogen (Δμ(H)). A prominent feature is the thick black line, marking the boundary between the stability regions of these two sites. The analysis highlights that higher Δμ(O) values promote the stability of substitutional sites, while lower Δμ(O) values favour the stability of supported sites. Notably, as Δμ(H) increases (reflecting higher hydrogen coverage), the boundary between these stability regions shifts toward lower Δμ(O) values, particularly within the temperature range associated with reduction conditions (400–600 K). This shift is explained by the significant adsorption energy of hydrogen on Rh in the presence of two oxygen vacancies, effectively extending the stability region of supported sites. Consequently, an elevation in hydrogen pressure stabilises both substitutional and supported sites. However, under extremely oxygen-poor conditions, the stability of the supported site is notably favoured, making it more stable.

For intricate systems where DFT and MD may lack precision, Quantum Monte Carlo (QMC) methods come into play. Although computationally more demanding, QMC offers heightened accuracy and is frequently employed to cross-verify results obtained from DFT calculations. The realm of modelling SACs confronts a significant challenge: the delicate equilibrium between accuracy and computational cost. Striving for higher levels of precision in simulations often demands substantial computational resources, imposing limitations on the scope of the study. Moreover, accurately capturing interactions in SACs, particularly under real-world reaction conditions, remains an ongoing challenge.

In response to these challenges, a surge in the integration of machine learning (ML) and artificial intelligence (AI) with conventional computational methods has been witnessed [57]. ML algorithms, trained on extensive datasets of previously computed material properties, can swiftly predict the behaviour of new SACs, eclipsing traditional methods in terms of speed. This approach not only expedites the discovery of novel SACs but also aids in optimising existing ones for specific applications [57]. The implications of computational modelling in the context of SACs are profound and multifaceted. One of its most immediate impacts lies in catalyst design and optimisation. Predictive modelling empowers researchers to craft new SACs with bespoke properties tailored for specific industrial reactions. This not only enhances efficiency but also curtails costs and environmental impact. Material discovery constitutes another pivotal domain where computational modelling of SACs plays a pivotal role. It assists in the identification of novel materials capable of supporting single atoms, thereby amplifying overall catalytic performance [44, 58]. This facet assumes particular significance in the realm of catalysts for sustainable energy applications, including hydrogen production and CO₂ reduction.

While considering the future, computational modelling’s trajectory in the area of SACs appears poised for a paradigm shift characterised by high-throughput computational screening. This approach entails the automated assessment of an extensive array of potential SAC configurations, expeditiously pinpointing the most promising candidates. Such an approach, when harmonised with the integration of computational predictions and empirical data, holds the promise of drastically accelerating the discovery and development of novel catalysts. Moreover, as computational prowess continues its ascent and algorithms evolve toward greater sophistication, one can anticipate more accurate and intricate models. These advancements will empower scientists to tackle ever-more intricate systems and reactions, pushing the boundaries of what is achievable in the realms of catalysis and material science. In summary, computational modelling in the context of SACs stands as a discipline at the forefront of contemporary scientific inquiry. Its reverberations extend across diverse industries, propelling the development of catalytic systems that are not only efficient and sustainable but also economically viable. As our understanding of these captivating materials continues to deepen, new vistas of technological advancement and environmental sustainability beckon. The ongoing evolution of computational methodologies, when coupled with empirical inquiry, promises to revolutionise our comprehension and application of SACs in the forthcoming years [43, 59, 60].

4. Choice of metal and tailoring the electronic structure

SACs which combine the advantages of homogeneous and heterogeneous catalysts, possess the highest level of atomic utilisation efficiency, a unique electronic state, selectivity, durability, and high catalytic performance [61]. SACs have received a lot of attention in recent years for potential applications in electrochemical HER. This reaction is critical in the development of a hydrogen economy, in which molecular hydrogen is used as a clean energy carrier for a wide range of applications such as energy storage and transportation [62]. Atomically dispersed metals are susceptible to migration and aggregation during the electrochemical process owing to the large surface energies of single atomic metal sites. Therefore, more effort should be devoted to understanding the interaction between the metal support and the physical/chemical environment of single atoms to find new and more conductive support materials to enhance the dispersion and stability of single atoms. The choice of metal in SACs plays a crucial role for HER, as it affects the efficiency of water electrolysis, which is very much dependent on the electrocatalysts activity. Electrocatalysts significantly enhance the HER by reducing the activation energy required for the reaction.

In electrocatalytic HER, various metal-based SACs (especially Pt, Pd, Ru, Fe, Co, Ni, Mo, W, V) have been systematically reviewed. Noble metals such as platinum (Pt), iridium (Ir), palladium (Pd), and rhodium (Rh) are considered among the most efficient materials for catalysis, due to their high activity [63]. However, these metals are expensive and scarce, which has led to extensive research on developing alternative SACs based on earth-abundant metals like nickel (Ni), iron (Fe), cobalt (Co), and molybdenum (Mo). These metals exhibit varying degrees of HER activity. The rational design and controlled synthesis of catalysts based on a thorough understanding of the structure – activity relationship and reaction mechanism are essential for a cost-effective HER catalytic process that maximises the usage efficiency of noble metals [64]. Notably, research in this field revealed that precise control of the oxidation states of single-atom Pt catalysts through electronic metal–support interaction significantly modulates the catalytic activities in either acidic or alkaline HER. In-depth studies on single platinum atom catalysts have revealed a remarkable enhancement in catalytic activity, surpassing conventional platinum/carbon catalysts by up to 37 times [62]. Platinum single sites trapped in the defects of carbon substrates, referred to as trapped Pt-SSCs, have emerged as a promising electrocatalytic solution for the HER [65]. In a meticulous process, the wet-chemical approach was used to synthesise Pt-GDY1 and Pt-GDY2 samples from GDY, using K₂PtCl₄ precursors and annealing at 273 and 473 K respectively (Figure 4a). HAADF-STEM images revealed the atomic dispersion of platinum (Pt) in both Pt-GDY1 and Pt-GDY2, exemplified in (Figure 4b, c). Notably, Pt-GDY2 exhibited a remarkable enhancement in the catalytic performance for the hydrogen evolution reaction (HER), outperforming both Pt-GDY1 and conventional Pt/C catalysts, as demonstrated in (Figure 4d) [66]. The Pt-SA/ML-WO₃ catalyst was also synthesised by subjecting PtCl₂-loaded nanosheets of ML-WO₃·H₂O to heat treatment, employing a space-constrained method (Figure 4e). The HAADF-STEM analysis distinctly revealed numerous individual and brighter dots, identified as Pt atoms on the Pt-SA/ML-WO₃ catalyst (Figure 4f). Importantly, the electrocatalytic HER activity of Pt-SA/ML-WO₃ surpassed that of ML-WO₃·H₂O, ML-WO₃, and even a commercial catalyst (20 wt% Pt/C), as evidenced by the polarisation curves with iR-correction presented in (Figure 4g) [67]. Tailoring the d-band centre has emerged as a promising strategy to enhance catalytic activity, by facilitating intermediate adsorption, formation of surface-active species and creating defects. Specifically, partially filled 5d orbitals of Pt on nitrogen-doped graphene enable efficient charge transfer, which contributes to the high conductivity of the material [62]. The Pt coating plays a pivotal role in electrochemical actuation by providing a protective layer that protects less noble transition metal atoms from oxidation and dissolution.

Figure 4. (a) Schematic illustration for the synthesis of Pt-GDY1and Pt-GDY2. Atomic-resolution HAADF-STEM images for Pt-GDY1 (b) and Pt-GDY2 (c). (d) HER polarisation curves for Pt-GDY1, Pt-GDY2. Reproduced with permission from ref.[66] Copyright 2018, Wiley. (e) Schematic illustration for the synthesis of Pt-SA/ML-WO₃. (f) HAADF-STEM image of Pt-SA/ML-WO₃. (g) HER polarisation curves for Pt-SA/ML-WO₃. Reproduced with permission from ref.[67] Copyright 2021, Wiley.

Electronic structure plays a pivotal role in determining the catalytic properties of single-atomic catalysts, especially in the context of HER. This is because, the electronic structure of a material influences its conductivity, band structure, and charge distribution, which are all directly related to the efficiency of charge transfer during electrocatalytic processes. It can also create effective active sites with low reaction energy during catalytic reactions [68]. Various strategies to modify the electronic configuration, via doping, vacancy creation, heterostructures, strain, and phase transitions, have been proposed on account of their significant contribution to enhancing the performance of various catalytic processes of HER (Figure 5). Doping induces sharp changes in the electronic structure of the catalyst material, leading to changes in its catalytic activity, while vacancy defects regulate electronic structures. Heterointerfaces facilitate partial electron transfer, optimising catalyst electronic structures during reactions. Strain engineering manipulates electronic configuration by adjusting atom distances, and phase transitions significantly impact electronic structures and catalytic properties. However, achieving high electrical conductivity, stability, and catalytic activity in a single phase remains challenging [69].

Figure 5. Schematic diagram representing reaction parameters affecting single atom catalysts in electrochemical Hydrogen Evolution Reaction.

5. Influence of reaction parameters

5.1. The effect of support material

The mobility and tendency of metal single atoms to aggregate, especially in the absence of strong anchoring to a support surface, present substantial challenges in catalyst fabrication and catalytic reaction conditions. During synthesis, the mobility of metal single atoms might cause them to aggregate, forming clusters or nanoparticles, altering the catalyst's features and potentially reducing its catalytic activity or selectivity. Similarly, under catalytic reaction conditions, the high mobility of single atoms can cause migration and aggregation, affecting the catalyst's performance [70, 71]. Furthermore, supports also play a role in influencing the electronic characteristics of metal nanoparticles through the interaction between the metal and support, as well as in facilitating charge transfer processes between them [32]. This not only reduces their surface free energy, achieving notable stability, but also strongly influences their catalytic properties, thereby significantly tailoring their catalytic performance. The highly desirable support properties of SACs for HER include high thermal stability, extremely high electrical conductivity, highly porous structure, excellent chemical and electrochemical stability and large surface area to anchor more single atoms [32, 72]. Because of their various crystal phases, surface pH, and anionic or cationic vacancies that act as anchor sites for metal nanoparticles, metal oxides are the most often employed supports for single-atom catalysts [73]. The selection of support materials for SACs has expanded to include metal nanomaterials, carbon-based materials, and surfaces of bulk metals [74].

5.2. The effect of coordination environment

The coordination environment significantly influences the behaviour of SACs in HER. The coordination environment refers to the type and arrangement of bonds surrounding the SAC, which has comprehensive effects on stability, selectivity, material loading, and overall activity in HER. Changes in the coordination environment, such as defects, heteroatoms, additional ligands, or support variations, can impact the electronic structure of catalytically active sites, potentially leading to different reaction pathways that hinder elucidation process [75]. Researchers often utilise computational modelling tools to investigate the effect of various coordination environments on SACs performance in HER, thereby directing the development of more efficient catalysts. Understanding and tailoring the coordination environment enables the optimisation of SACs towards better catalytic performance and stability in HER, enabling the advancement of hydrogen-based renewable energy sources [19]. Notably, the coordination environment's profound effect on the activation energy barrier for proton transfer during HER directly shapes the reaction pathway and overall catalytic performance. Therefore, strategic optimisation of the coordination environment in a well-designed setting can enhance HER activity by improving the binding energy of reactants. In particular, the active central atoms in SACs, with high unsaturated coordination, showed much higher catalytic activity compared to the surface atoms of conventional supported nano catalysts [76].

5.3. The effect of electrolyte pH

According to research results, the pH of the electrolyte is a crucial factor in the HER evolution reaction. The pH of the electrolyte affects the availability of protons in the solution, impacting the reaction kinetics and efficiency of the HER. The influence of pH on the HER is clearly observable in highly active minerals. In general, low pH conditions promote proton adsorption, whereas very high or very low pH values can adversely affect catalyst stability. Furthermore, variation in pH can influence the surface chemistry and electrochemical behaviour of SACs, affecting their overall performance. Investigating this correlation is critical for tailoring SACs for a wide range of electrocatalytic applications, including hydrogen production to energy conversion, and highlights the importance of pH control in developing highly efficient and stable catalysts [32, 35]. For instance, as the pH level increases from 0 to 13, the HER efficiency of Pt, Ir, and Pd decreases significantly, by a factor of 210, 120, and 90, respectively. It is noteworthy that pH-neutral HER is influenced by reactant properties, electrolyte condition (buffered/unbuffered), and concentrations. The catalytic activity of SACs can be influenced by the electrolyte solution composition utilised in the HER. The presence of particular ions or additives in the electrolyte can exert an impact on both the catalytic stability and the reaction kinetics [77].

5.4. The effect of temperature

Temperature has a significant effect on the HER performance of SACs, which is a key factor in assessing the catalytic activity. Temperature affects the kinetics of the HER by influencing reactant adsorption, activation energies, and reaction rates. In general, increasing temperature improves HER kinetics, resulting in faster reaction rates and better catalytic activity of SACs. However, too high temperatures might cause catalyst deactivation or structural changes, reducing long-term performance [21]. For instance, precisely controlling the annealing temperature prevented Cu species from aggregating while simultaneously encouraging V_O migration and redistribution at Cu sites [78]. Elevated temperatures play a pivotal role in influencing reaction rates by providing molecules with greater kinetic energy, consequently facilitating faster diffusion of reactants. The increased thermal energy is especially important during the thermal activation/carbonization step. In this phase, the large heat conveys single atoms with increased energy, thereby stimulating their mobility [79].

6. Conclusion and future perspectives

In conclusion, this comprehensive review on SACs for HER provides valuable insights into the potential of these catalysts for improving the efficiency of hydrogen production. The experimental and computational modelling studies highlight the crucial role in identifying the HER mechanism and the role of SACs in catalysing the HER reaction. It is hoped that this review will pave the way to revolutionise the field of sustainable hydrogen production by enabling the rational design of highly efficient, stable, and economically viable SACs for HER.

Furthermore, the following future perspectives on the improvement of SACs for HER should be considered.

1. Designing and developing SACs with outstanding catalytic performance requires ongoing research and development. A more in-depth knowledge of the electronic structure and factors affecting the catalytic performance is required.

2. Enhancing the stability and durability of SACs in harsh operational environments is necessary for practical applications. The research should prioritise the development of SACs that would maintain their catalytic performance even when exposed to electrochemical environments for a long period of time.

3. The scalability of SACs fabrication techniques is another critical aspect that requires careful consideration. Further research is needed to explore scalable and cost-effective techniques for the large-scale fabrication of SACs to enable their use for industrial-scale hydrogen production.

4. Incorporating SACs into renewable energy systems, such as solar or wind power, can improve the future sustainability of hydrogen production. Further research should investigate the potential correlation between SACs and renewable energy technologies to enhance the production of hydrogen in efficient and environmentally friendly approaches.

5. Investigating the potential of SACs as versatile catalysts for various electrochemical processes beyond the HER is a promising approach. These applications might involve oxygen evolution reactions, carbon dioxide reduction, and other various energy conversion processes.

6. Conducting thorough techno-economic analysis is necessary for evaluating the economic feasibility of SACs in the context of large-scale hydrogen generation. This involves assessing the total costs, effectiveness, and ecological implications of integrating SACs into both conventional and prospective methods of hydrogen production.

In summary, the future of SACs in electrochemical HER research holds promising prospects for advancing sustainable and efficient hydrogen production. Addressing the identified challenges and exploring new opportunities will contribute to the development of SACs as key components in the transition to a clean and renewable energy future.

Acknowledgements

Author would like to acknowledge Prince Sultan University Riyadh Saudi Arabia.

Disclosure statement

No potential conflict of interest was reported by the author(s).