HERs involve the transfer of two electrons. Although the reaction mechanism is essentially identical in acidic and alkaline conditions, protons in an acidic solution are derived from H₃O⁺, whereas protons in an alkaline solution are derived from H₂O. The basic reaction is as follows:

Volmer reaction:

Heyrovsky reaction:

Tafel reaction:

In this formula, * represents the reaction site on the catalyst surface. The Volmer reaction is the initial electrochemical reaction step in which hydrated protons combine with electrons to form adsorbed hydrogen atoms (H*) on the catalyst surface. The Heyrovsky reaction is the subsequent electrochemical desorption step in which the adsorbed hydrogen atoms combine with H₃O⁺ or H₂O and accept an electron from the electrode surface to generate hydrogen gas (H₂). This is followed by the Tafel reaction, which is a composite desorption step in which two H* molecules on the catalyst combine directly to form H₂. Accordingly, HER must include electron transfer and desorption steps. Therefore, there are two fundamental reaction mechanisms for HER: the Volmer-Heyrovsky mechanism and the Volmer-Tafel mechanism. Moreover, analyzing the Tafel slope reveals the electrochemical reaction kinetic process during the actual catalytic reaction.

In HER, adsorption and desorption of hydrogen atoms on the catalyst surface are two opposing processes. If the adsorption force is considerably weak, it is not conducive to hydrogen generation. Conversely, if it is considerably strong, it is conducive to hydrogen generation but not to its release [4]. Therefore, there must be a balance between adsorption and desorption on the catalyst surface. Catalytic activity is usually measured using the free energy of hydrogen adsorption (ΔG_H), and the ΔG_H of a good catalyst should be close to zero. Additionally, the common metal catalyst ΔG_H has a “volcanic” relation with the catalytic activity [5]. Pt-based noble metals exhibit a suitable ΔG_H and a high reaction rate and catalytic activity in practical catalytic reactions. However, the scarcity and high cost of Pt-based noble metals are the biggest obstacles to their large-scale commercial applications. Therefore, developing a cost-effective, abundant, and stable nonprecious metal catalyst for replacing these precious metals is crucial to the application of electrocatalytic hydrogen production technology [6].

Transition metal phosphides (TMPs) have undergone nearly two centuries of development since their discovery and are widely used in numerous fields owing to their excellent mechanical strength, electrical conductivity, and chemical stability. Studies on their crystal structure have shown that TMP crystals comprise prismatic structural units with a triangular prismatic geometry, forming an isotropic crystal structure. This crystal structure tends to have more coordinatively unsaturated atoms on its surface, yielding an increased HER catalytic activity [7].

Because phosphides have a higher electronegativity and can absorb electrons from metal atoms, the P atoms in TMPs are considered to be doped into the transition metals. Thus, in the HER reaction, negatively charged Р acts as an active site to capture positively charged protons. Accordingly, the catalytic activity of phosphide can be optimized by increasing the P concentration [8]. However, when the P content is too high, the free electrons in the compound and the conductivity of the catalyst material decrease, yielding a decrease in its catalytic activity. Conversely, the metal (M) is the hydride-accepting center, and the higher the metal content, the greater the stability and electrical conductivity of the compound. However, an excess of metal will affect the ratio of P atoms in TMPs, disrupting the equilibrium between the adsorption and desorption of H atoms on the surface of P atoms. Therefore, controlling the proportion of M−P is an important topic in the study of TMPs [9].

Recently, various methods have been developed to prepare TMP-based electrocatalysts, yielding diverse morphologies, orientations, and compositions. In conjunction with the development of preparation methods, effective material engineering strategies to promote catalytic performance have been summarized herein. Additionally, based on the previous overview of TMPs, we present the preparation methods of commonly used TMP materials, the means to improve the performance of TMPs, and the future development direction of TMPs. At the same time, we propose the concept of “stimulatory factors.” These “stimulatory factors,” as described in the third section, are the core of this paper and represent a precise and concise summary of the various factors affecting the performance of TMPs. We hope that readers will be able to easily identify the most important “stimulatory factors” so that the research on TMPs will have a clear context.

Heteroatom doping improves the catalytic activity of TMPs by introducing different types of atoms. Doping can be divided into three categories based on the types of heteroatoms: metal atom doping, nonmetal atom doping, and metal-nonmetal atom codoping. Ge et al. [22] synthesized Mn-doped CoP catalyst materials and investigated the effect of doping atoms on the performance of the catalyst for hydrogen evolution in both acidic and alkaline electrolytes. CoP was doped with Mn atoms, causing the d-band center of Co atoms to shift in a negative direction, indicating that the interaction between Co sites and hydrogen may be weakened. In addition, the electrochemical performance of the as-prepared Mn–CoP showed that the overpotentials required to achieve a current density of 10 mA/cm² in 0.5 mol/L H₂SO₄ and 1 mol/L KOH are 65 and 100 mV, respectively (Figs. 1a and b). Hua et al. [23] synthesized V-doped CoP rice flake arrays (V–CoP/CC) supported on a carbon cloth using a facile water bath/phosphating method, as shown in Figs. 1c and d. V-doped CoP can optimize the electronic structure and enhance the electron transport efficiency in the catalyst. Accordingly, the V–CoP/CC electrode exhibited excellent HER electrocatalytic activity in 1 mol/L KOH solution. Fig. 1e shows the impedance test result for V–CoP/CC versus CoP. An overpotential of 98 mV is required to achieve a current density of 10 mA/cm². Ma et al. [24] synthesized Fe-doped CoP nanoarrays (Fe–CoP/CC) using a carbon cloth as a substrate (Fig. 1f). The LSV test in 0.1 mol/L potassium borate (pH 9.2) required an overpotential of 175 mV to achieve a current density of 10 mA/cm², demonstrating excellent HER catalytic performance. In addition, the two-electrode water electrolyzer comprising Fe–Co–Pi–Bi/CC as the anode and FeCoP/CC as the cathode demonstrated a current density of 10 mA/cm² at a voltage of 1.95 V. Wang et al. [25] prepared W-doped CoP nanosheet arrays (W–CoPNAs/CC) supported on a carbon cloth. Fig. 1g shows that W–CoPNAs/CC exhibits a hierarchical porous structure. The LSV testing of W–CoPNAs/CC in acidic, alkaline, and neutral electrolytes showed that overpotentials of 89, 94, and 102 mV, respectively, were required to achieve a current density of 10 mA/cm², confirming an excellent catalytic performance. Because of the porous nature of the arrays of W-doped CoP nanosheets, the number of active sites increased and the catalytic performance improved. Wang et al. [26] synthesized N–CoP₂ nanowires supported on a carbon cloth using simple hydrothermal, thermal phosphating, and ammonia atmosphere calcination methods. Further, they tested the HER performance in 0.5 mol/L H₂SO₄; the overpotential required to reach a current density of 10 mA/cm² was 38 mV, which is considerably lower than that of undoped CoP (125 mV). This performance improvement is attributed to the CoP₂ lattice shrinkage and ΔG_H* reduction caused by N doping. Sun et al. [27] prepared an amorphous B-doped CoP ternary alloy Co_2.90B_0.73P_0.27 supported on nickel foam via an ultrasonic method, which has an excellent catalytic performance in alkaline electrolytes. Accordingly, the overpotential required to reach a current density of 10 mA/cm² was 12 mV (Fig. 1h), the Tafel slope was as low as 42.1 mV/dec, and the stability was maintained for 20 h at a high current density of 1 A/cm², even surpassing that of commercial Pt/C. This excellent performance is attributed to the synergistic effect of B and P, which weakens the hydrogen adsorption on the surface and inhibits the oxidation of Co.

In Section 3.1 on heteroatom doping, we can conclude that the introduction of heteroatoms destroys the periodicity of the TMP lattice, modifies the local electronic structure, shifts the d-electron center away from the Fermi level, and lowers the reaction energy barrier, accelerating the reaction kinetics [28,29]. By doping atoms with high electrical conductivity with TMPs, the electrical conductivity and electrocatalytic performance of the material can be improved. Additionally, this results in a synergistic effect between the substrate and active components, which enhances the stability of TMP particles and reduces the agglomeration phenomenon to a certain extent [30-32].

In the HER reaction, the structure of the electrode has a considerable impact on the progression of the catalytic reaction. The electrode structure determines the number of available active sites. It also affects mass transport, concentration gradients, and stability. To maximize the catalytic activity of TMPs, researchers frequently start from the nanostructure and components of the catalyst simultaneously and endow the prepared products with particular shapes or large dimensions. A rational design and regulation of morphology and structure can effectively enhance the activity and stability of the catalyst. Sumboja et al. [33] used C₄H₆CoO₄ and C₁₅H₂₁CoO₆ as precursors to produce nanostructured cobalt phosphide (Co–P I) and nanoparticle cobalt phosphide (Co–P II) via the gas-solid phosphating method. Figs. 2a and b show that both Co–P I and Co–P II have alveolar sac-like structures. Additionally, the LSV test demonstrates the superior catalytic activity of Co–P I (Fig. 2c). Moreover, the special alveolar sac-like nanostructures increase the catalytically active surface area, affording increased HER activity. Yang et al. [34] developed CoP nanocrystals with a unique sea urchin-like structure by phosphating at low temperatures. The LSV test was conducted in an acidic medium. When the current density was 100 mA/cm², the overpotential was 180 mV. Chen et al. [35] used PPh₃ as the phosphorus source to synthesize rod-like and flower-like Co₂P by controlling the phosphating process (Figs. 2d and e). These Co₂P structures exhibit high porosity and specific surface area. Both rod- and flower-shaped Co₂P have low impedances in the EIS test (Fig. 2f). Liu and Li [36] described a new type of CoP hollow polyhedral catalyst in which MOF materials were used as a template and the hollow polyhedral CoP catalyst was successfully prepared in situ. This unique structure of CoP increases its specific surface area and porosity, affording a large number of active sites. Through the LSV test of the HER reaction in an alkaline medium, the required overpotential to achieve a current density of 10 mA/cm² is 159 mV, and the Tafel slope is 57 mV/dec. Su et al. [37] reconstructed the surface structure of CoP and attained stable Co(OH)_x@CoP hybridization using an in situ electrochemical activation method. Co(OH)_x formed in situ accelerates the dissociation of water, yielding a large number of hydrogen atoms, which are transferred to nearby exposed Co active sites to form Co–H bonds. The results of testing the electrocatalytic activity of CoP and Co(OH)_x@CoP for HER in 1 mol/L KOH electrolyte indicate that the overpotential required for CoP to reach a current density of 10 mA/cm² is 180 mV, whereas the overpotential required for Co(OH)_x@CoP is only 100 mV. This enhanced HER activity is attributed to the synergistic effect of the optimized surface/interface of the formed mixed Co(OH)_x@CoP species, which increases the concentration of surface hydroxyl groups and promotes water splitting.

In addition, modifying the structure of TMPs can alter the spatial dimensions of the materials, altering the charge distribution and inducing local electric fields [38-40]. Compared with traditional one-dimensional and two-dimensional materials, three-dimensional (3D) materials exhibit superior electron transport capacity and a larger specific surface area, which can increase the probability of the active material reacting with the electrolyte and thus improve the catalytic activity. Therefore, 3D materials have gradually become the mainstream of designing material structures. Chang et al. [41] used carbon paper as a substrate to grow S–Ni₅P₄NPA/CP nanoplate arrays with 3D structures (Figs. 2g and h). The 3D-structured S–Ni₅P₄NPA/CP nanoplate arrays exhibit a larger specific surface area and more active sites than nickel phosphide and nickel sulfide. After the LSV testing of HER performance, when the current density reaches 10 mA/cm² and 100 mA/cm², the overpotential is 56 mV and 104 mV, respectively, while the Tafel slope is only 43.6 mV/dec. S–Ni₅P₄NPA/CP has a lower impedance than Ni₅P₄NPA/CP after EIS testing (Fig. 2i). Wei et al. [42] reported a facile method of synthesis using algal bioprecursors to embed NiCoP nanoparticles on graphene aerogels (NiCoP–NPs@GA) using a phosphating step. In 0.5 mol/L H₂SO₄, the overpotential required to achieve a current density of 10 mA/cm² is 109 mV, and the Tafel slope is 63 mV/dec. In addition, the large number of pores in the 3D GA matrix effectively prevented the volume expansion and aggregation of NiCoP–NPs, affording its excellent catalytic performance. Yang et al. [43] reported that CoP nanoparticles were encapsulated in three-dimensional nitrogen-doped porous carbon (CoP@3D–NPC) catalysts. The unique 3D porous carbon could provide multidimensional electron/ion transport paths to reduce diffusion distances and the encapsulated CoP nanoparticles could prevent it from directly contacting the electrolyte to ensure interfacial stability. Further, they could exhibit outstanding hydrogen evolution catalytic performance in 0.5 mol/L H₂SO₄ and 1 mol/L KOH, and the overpotentials required to obtain a current density of 10 mA/cm² were 127 and 203 mV, respectively, while the corresponding Tafel slopes were 58 and 65 mV/dec, respectively.

In Section 3.2 on the regulation of topography, we can conclude that the objective of designing a specific topography is to increase the specific surface area. An increase in specific surface area increases the contact area between the catalyst and the electrolyte, which yields more active sites and the intermediates produced by HER have a greater chance of adsorbing onto the active substances. The multidimensional structure can accelerate the flow of electrolytes and facilitate the transport of substances. These characteristics are important to achieve hydrogen evolution activity and stability.

A heterostructure is a composite material in which different catalyst materials come into contact to form a heterogeneous interface. Owing to the direct physical contact between each component or phase, the corresponding physicochemical properties of the interface are highly different from those of the bulk phase. Furthermore, these hybrid catalysts generally exhibit strong interfacial interactions. Generally, the rational design of multicomponent heterogeneous structures is crucial in the production of electrocatalysts with considerably enhanced electrocatalytic activity. In the experiments, TMPs can be combined with other substances possessing catalytic properties, and the structure-activity relationship between the heterostructure and catalytic activity enables the enhancement of the catalytic activity. Liu et al. [44] synthesized a novel self-supporting Ni₂P–NiSe₂/CC heterostructure catalyst with a unique sheet-like structure (Figs. 3a and b). Accordingly, the effect of the catalyst on HER catalytic activity was tested under an alkaline condition of 1 mol/L KOH. An overpotential of 66 mV was observed under a current density of 10 mA/cm² (Fig. 3c), and the stability and durability of Ni₂P–NiSe₂/CC were found to be excellent. According to calculations using the density functional theory, Ni₂P and NiSe₂ exhibit a strong electronic interaction, which optimizes the Gibbs free energy of H/H₂O adsorption and improves the HER activity. Boppella et al. [45] synthesized CoP/NiCoP composite nanosheets with a two-dimensional structure. Their NiCoP (111) and CoP (112) planes exhibit interplanar spacings of 0.22 and 0.19 nm, respectively. The well-defined lattice fringes demonstrate the formation of nanoheterojunctions between CoP/NiCoPs, and the coupling effect at the phase interface improves the HER reaction kinetics, optimizing the HER catalytic performance. Ge et al. [46] reported the use of two-dimensional zeolite imidazole framework arrays as templates to combine MoS₂ nanosheets with CoP for HER reaction; Fig. 3d shows its structure. In 0.5 mol/L H₂SO₄ and 1 mol/L KOH, a current density of 10 mA/cm² required overpotentials of 53 and 77 mV, respectively. Moreover, the Tafel slope reached 37 mV/dec in an electrolyte containing 0.5 mol/L of H₂SO₄, (Fig. 3e). Pan et al. [47] constructed a novel, highly active CoP/MoS₂–CNTs heterostructure catalyst via the in situ growth of CoP on the surface of MoS₂ and CNTs. The electrocatalytic performance of HER in 0.5 mol/L H₂SO₄ was excellent. Accordingly, only a 10 mV overcharge was required to obtain an overpotential of 10 mA/cm², and the Tafel slope was found to be 42 mV/dec. The results show that the electrocatalytic activity of the CoP/MoS₂ composite catalyst is superior to that of CoP and MoS₂ owing to a strong synergistic effect. Wang et al. [48] synthesized CoP₃/NiMoO₄ nanosheets (CoP₃/NiMoO₄–NF) in situ on nickel foam via hydrothermal treatment, annealing, and phosphating (Fig. 3f). The synergistic effect of CoP₃ and NiMoO₄ considerably promoted the activity of HER. Furthermore, the nickel foam promotes the vertical growth of well-aligned nanosheet arrays, exposing an increased number of HER active centers. To achieve a current density of 10 mA/cm² in 1 mol/L KOH, a HER overpotential of 92 mV was required, with a Tafel slope of 50.7 mV/dec (Fig. 3g). In particular, CoP₃/NiMoO₄–NF demonstrated excellent durability, and the electrochemical activity could be maintained for more than 12 h. Wu et al. [49] designed a low-cost, high-performance bifunctional electrocatalyst. Heterogeneous Ni₂P–Fe₂P microflakes were synthesized by directly soaking nickel foam in hydrochloric acid and ferric nitrate solution, followed by phosphating, and the catalytic performance of HER under alkaline conditions was enhanced through Ni₂P–Fe₂P heterojunction. The experimental results show that in 1 mol/L KOH fresh water solution, only 168 mV and 186 mV overpotentials were required to achieve current densities of 100 mA/cm² and 500 mA/cm², respectively. In addition, the hydrophilic nature of the Ni₂P–Fe₂P heterojunction structure was able to provide sufficient space for the diffusion of the electrolyte, enhancing the gas mass transfer rate and affording its outstanding catalytic activity at a high current density. Accordingly, a versatile and cost-effective method for the preparation of heterogeneous metal phosphide catalysts was realized. Mai et al. [50] designed a two-dimensional (2D) heterostructured bimetallic phosphide and layered double hydroxide (LDH)-based electrode assembly (Co_xMo₁P/NiFe–LDH) by hydrothermal synthesis. The Co_5.0Mo₁P/NiFe–LDH hybrid catalyst was obtained through the chemical electrodeposition process (Fig. 3h). According to electrochemical tests, Co_5.0Mo₁P/NiFe–LDH exhibits excellent HER catalytic performance in 1 mol/L KOH, requiring only 98.9 mV of overpotential to reach a current density of 10 mA/cm². The Co_5.0Mo₁P/NiFe–LDH hybrid catalyst was used as the cathode to achieve a current density of 50 mA/cm² at a potential of 1.68 V, and the electrochemical activity could be maintained for 27 h, demonstrating excellent catalytic activity and stability. Additionally, the stability of Co_5.0Mo₁P/NiFe–LDH was exceptional (Fig. 3i). Yu et al. [51] successfully prepared N-doped carbon-coated CoP(MoP)–CoMoO₃ heterostructure catalysts [CoP(MoP)–CoMoO@CN] via thermal decomposition and phosphating strategies; these catalysts demonstrated excellent performance in 1 mol/L KOH alkaline medium. For the excellent catalytic performance of total water splitting, the HER of CoP(MoP)–CoMoO@CN nanocomposites required an overpotential of 198 mV to achieve a current density of 10 mA/cm², and the corresponding Tafel slopes were 95 mV/dec. Importantly, the use of the as-prepared CoP(MoP)–CoMoO@CN heterostructure catalyst as cathode and anode, respectively, yielded a current density of 10 mA/cm² at a potential of 1.55 V. The catalyst provides abundant active sites for the HER. This remarkable heterostructured CoP(MoP)–CoMoO@CN catalyst exhibits excellent catalytic activity.

In Section 3.3 on heterojunction, the mechanism of the heterojunction effect is summarized as follows. After the formation of the heterointerface, the periodicity of the metal phosphide lattice is destroyed, affording a change in the local electronic structure. Electronic interactions between species redistribute the electronic densities of their respective states. The modified electronic structure increases the conductivity of the system and accelerates charge transfer, accelerating the adsorption/desorption of intermediates and enhancing the reaction kinetics and catalytic activity [52-55].

The vacancy defect refers to the space left in the TMPs catalyst crystal when atoms or ions leave the lattice. Introducing vacancy defects is an effective method for regulating the catalytic performance of TMPs. Zhang et al. [56] synthesized Ni₂Р nanosheet arrays (V–Ni₂P/NF) with abundant cation vacancy defects through a multi-step reaction process (Fig. 4a). Density functional theory (DFT) calculations proved that the Ni vacancy defect changed the electronic structure near the atoms in the compound, bringing the adsorption energy of H on the surface of the V–Ni₂P/NF catalyst closer to zero and thereby decreasing the catalyst energy. The potential barriers and cation vacancies for the HER pushed the d-band center of the catalyst away from the Fermi level. The electrocatalytic performance of Ni₂P was successfully optimized viaby cation vacancy engineering, and the V–Ni₂P/NF catalyst was used in HER with good electrocatalytic performance under alkaline conditions, requiring an overpotential of 55 mV to achieve a current density of 10 mA/cm² (Fig. 4b). Kwong et al. [57] introduced Fe vacancy defects to tune the electronic structure of iron phosphide (FeP), where Mg was introduced as a sacrifice to form a Vc–FeP film. Vc–FeP had a higher phosphorus content than FeP and Mg–FeP, and the higher phosphorus content afforded more negative charges, which attracted more protons and enhanced the hydrogen evolution activity. In 1 mol/L KOH and 0.5 mol/L H₂SO₄, the overpotentials required to achieve a current density of 10 mA/cm² were 108 and 65 mV, respectively, and the Faradaic efficiency was close to 100%. Theoretical and experimental analysis showed that the HER activity of the enhanced catalyst resulted from the presence of iron vacancies, which optimized the electronic structure of the catalyst and brought the hydrogen adsorption-free energy close to zero, thereby enhancing the catalytic performance of the catalyst. Duan et al. [58] prepared (V–Ni₁₂P₅) with P vacancy defects using nickel foam as thea substrate. Fig. 4c depicts its structure. According to DFT calculations, a considerable number of electrons were accumulated in V–Ni₁₂P₅, which redistributed the electrons in the Ni₂P crystal and enhanced the catalytic activity of HER. The LSV test for the HER reaction of V–Ni₁₂P₅, in 1 mol/L KOH solution, yielded a current density of 10 mA/cm², requiring an overpotential of 27.7 mV, with a Tafel slope of only 30.88 mV/dec. Additionally, the stability of the catalyst was observed to be excellent after extensive testing (Fig. 4d). Zhang [59] and other research groups synthesized defect-rich structural DR-MoP catalysts through a novel and simple low-temperature calcination strategy (Fig. 4e). It is a three-pronged approach: (1) When the synthesis temperature is low, the grain growth is inhibited, the grain size becomes smaller, and the grain boundary widens; (2) Sodium hypophosphite is decomposed into PH₃, which will help aids the formation of a porous structure with manymore exposed areas; and (3) a lower synthesis temperature will lead tocauses an insufficient reaction, resulting inyielding residual Mo–O bonds of the precursor, allowing oxygen incorporation. By evaluating the performance of DR-MoP, it has been demonstrated that introducing defects into the crystal via the aforementioned method can provide manymore active sites, and that doping with O atoms can increase the intrinsic conductivity. When applied to the HER reaction, the DR-MoP catalyst demonstrated excellent catalytic activity and stability in both acidic and alkaline electrolytes, with required overpotentials of 156 and 104 mV at a current density of 10 mA/cm², corresponding to Tafel slopes of 49 and 50 mV/dec (Fig. 4f), respectively. DFT analysis revealed that the introduction of P vacancy defects reduced the degree of hybridization of M (transition metal) 3d and 2p orbitals and considerably increased the electron density of M and P atoms near the P vacancy [60-62].

In Section 3.4 on vacancy defects, the introduction of vacancy defects in TMPs alters its local atomic structure and coordination number and also increases the number of catalytically active sites. Additionally, vacancy defects modify the local atomic structure, yielding modifications to the electronic structure, which causes the d-band center to move away from the Fermi level and reduces the adsorption energy of hydrogen on the catalyst surface [63,64]. Simultaneously, the active site is increased and the HER reaction energy barrier is lowered.

Controlling the crystalline phase of TMPs (single-crystalline phase, amorphous phase, and polycrystalline metal phosphide with element doping) is also essential to optimize the activity of the catalyst. Li et al. [65] reported a new gas-solid synthesis method to synthesize single-crystal cobalt phosphide nanowires (CoP NWs) on porous Co foams. According to the TEM image shown in Fig. 5a, the interplanar spacing comprises the well-resolved lattice fringes at 0.254 nm, demonstrating the (200) plane of orthorhombic CoP. The corresponding fast Fourier transform (FFT) patterns of these CoP NWs reveal their single-crystal nature to further analyze their electrochemical performance. In 1 mol/L KOH, the HER requires an overpotential of 244 mV to achieve a current density of 100 mA/cm². Additionally, CoP NWs were used as the cathode and anode for water electrolysis, with the overpotentials required to achieve current densities of 10 and 100 mA/cm² being 156 and 178 mV, respectively (Fig. 5b), and the CoP NWs catalyst exhibiting excellent catalytic performance. Phosphides with different metal/phosphorus stoichiometric ratios have different intrinsic activities owing to their distinct crystal structures. Their disordered domains contain numerous defects and vacancies, which are considered electrocatalytically active centers. Zhang et al. [66] synthesized single-crystal (Fe_xNi_1-x)₂P nanosheets supported on nickel foam with exposed {011 1} planes. Compared to pristine amorphous Ni₂P, (Fe_xNi_1−x)₂P nanosheets exhibited superior HER activity. The required overpotentials to achieve a current density of 10 mA/cm² in acidic, neutral, and alkaline media, were 81, 90, and 103 mV, respectively. According to the DFT calculations, the {011 1} plane has great charge accumulation around the P active site, which reduces ΔG_H* and ΔG_H2O and enhances catalytic activity.

Different crystal plane indices in the catalyst have different effects on catalytic performance, and the atomic structure of the crystal plane partially determines the interaction strength between the crystal plane and the reactants or intermediates. The geometric and electronic structures of different crystal planes of different catalysts are distinct, yielding distinctive physicochemical properties. Zhang et al. [67] prepared ultrathin porous CoP nanosheets in the form of single crystals with exposed {200} crystal faces using a chemical transformation strategy (Fig. 5c). The ultrathin porous CoP nanosheet catalyst exhibits excellent electrochemical HER performance in acidic electrolysis, where overpotentials of 56 and 131 mV are required to obtain current densities of 10 and 100 mA/cm², respectively (Fig. 5d), the Tafel slope is 44 mV/dec, and the electrochemical activity can be maintained for 20 h, indicating that the catalyst has a stable performance when applied to HER. The excellent catalytic performance of the CoP nanosheet catalysts, as calculated by DFT, is primarily attributable to its ultrathin porous structure with abundant active sites.

In Section 3.5 on crystal phase/crystal plane, the intrinsic activity of TMPs can be modified by changing the phases of the metal phosphides. The preparation of single-crystal phase metal phosphide allows for the investigation of the effect of the preferred crystal surface on the catalytic performance; the amorphous phase metal phosphide is simple to use as the adsorption and desorption sites of the intermediate product owing to the large number of unsaturated bonds in the crystal. Thus, it demonstrates excellent catalytic performance. The atomic structure of the crystal face partially determines the interaction strength between the crystal face and the reactants or reaction intermediates, and the lower the crystal face index, the lower the catalytic activity. The high crystal face index has a high catalytic activity owing to a large number of kinked atoms and edge sites. Furthermore, crystals with a high facet index exist in an environment with short-range order, their atomic structure does not change, and their stability is excellent. Therefore, the catalytic performance can be enhanced by modifying the crystal structure of the catalyst. However, the instability of high-surface-energy crystal faces causes them to rapidly appear and disappear during catalysis. Therefore, the design of high-surface-energy crystal faces with high stability remains challenging.