Cobalt single-atom-decorated nickel thiophosphate nanosheets for overall water splitting

Abstract

Formidable challenges in the development of efficient single-atom catalysts including accessible fabrication processes, unambiguous interactions with matrixes, and accelerated kinetics of catalytic mechanisms, pose an obstacle in maximizing the catalytic performance. Herein, it is proposed, for the first time, to bond single-atom Co on the surface of single-layered nickel thiophosphate (SA Co NiPS₃) forming a uniform unsaturated coordination mode of isolated Co. Theoretical studies indicate that the SA Co NiPS₃ electrode possesses a higher intrinsic activity than that of traditional Co alternative-doped NiPS₃via the enhancement of conductivity and the reduction of energy barriers for bifunctional hydrogen/oxygen evolution reactions (HERs/OERs). Notably, a water splitting electrolyzer assembled using SA Co NiPS₃ nanosheets as both the cathode and the anode achieves 50 mA cm⁻² at 1.60 V, which outperforms the Pt/C//RuO₂ benchmark (1.74 V@50 mA cm⁻²). Besides, this electrocatalyst exhibits an outstanding durability up to 6 days with negligible decay. This work shines fresh light on igniting 2D metal thiophosphate electrocatalytic activity that even surpasses noble-metal-based electrocatalysts via the single-atom engineering.

---

1. Introduction

Hydrogen, as a good substitute for traditional carbon-based fuels in future energy consumption, has gained increasing research interest, mainly due to its high energy density, zero-emission, and renewability.^1,2 Water splitting for hydrogen evolution via electrocatalysis is one of the most effective methods because of high energy conversion efficiency, cost-effectiveness, and environmental green nature.^3–9 Noble-metal-based materials such as Pt and RuO₂ have proved to be the most active electrocatalysts for hydrogen evolution reactions (HERs) and oxygen evolution reactions (OERs), respectively; however, scarcity and high cost of noble metals hamper their large-scale implementation.^10,11 Exploration of catalysts with competitive properties even surpassing noble-metal-based materials on catalytic activity and stability is a critical challenge for wide use of water splitting.

Recently, various noble-metal-free materials have been explored as alternative overall water splitting electrocatalysts including alloys,¹² sulfides,^13–15 carbides,¹⁶ phosphides,¹⁷ hydroxides,¹⁸ and (oxy)hydroxides,¹⁹ among many others. Noteworthy, lamellar metal thiophosphates (MPS₃),²⁰ which have an analogous sandwich configuration to the typical metal disulfides such as MoS₂ and WS₂, can also be employed as highly attractive electrocatalysts due to their tunable electronic structure and compositional diversity. Inspired by the modification of MoS₂, large-sized and mono-layered NiPS₃ were prepared via an electrochemically exfoliating process, delivering a robust overall water splitting performance.²¹ Besides, atom-scale modulation of electronic configuration via in situ substitution doping of elements in the MPS₃ lattice has been demonstrated as an effective approach to motivate the intrinsic catalytic activity. For example, when half of Fe atoms in FePS₃ have been substituted by Ni atoms forming NiFePS₃, the electrocatalytic performance was significantly enhanced on HER, OER and overall water splitting for the unique lattice distortion, which decreased the kinetic energy barriers of water splitting.²² Unfortunately, a large gap in their electrocatalytic water splitting activity still exists compared to the noble-metal-based benchmark due to the difficulty in maximizing the intrinsic activity of their active sites.²³

In parallel, single atoms anchored on the surface of supports can realize a theoretical efficiency of 100% of atomic utilization for catalysis. This single-atom (SA) catalyst strategy has been implemented on various 2D materials to activate the inert basal planes drastically in recent years.²⁴ According to the results of computational and experimental studies, the electron density distribution of the 2D metal disulfide matrix can be flexibly adjusted via electronic interaction between isolated metal atoms and the matrix, thereby optimizing the intermediate adsorption and desorption possibilities of single-atom decorative metal disulfides.^25,26 Qi and coworkers presented the utilization of Co SAs as a fuse to trigger the electrocatalytic activity of ultrathin MoS₂; Co SAs covalently bound to the distorted 1T MoS₂ catalyst exhibited extraordinary HER activities.²⁷ In another study, Liu's group obtained similar results: Co SAs promoting mono-layered MoS₂ displayed an even much higher HER activity than that of the state-of-the-art Pt catalyst.²⁸ Additionally, Co SAs can act as promoters for other 2D materials such as TaS₂,²⁹ Mo₂C³⁰ and N-doped graphene.³¹ Although Co partial substitution of interlayered metals in MPS₃ has been proved to be an efficient method to enhance the electrocatalytic performance, the noble-metal alternative target is still out of reach because of the restricted electronic modulation causing by the coverage of the outer P–S layer.³² Nevertheless, to the best of our knowledge, fully inspiring the intrinsic activity of ultrathin NiPS₃via SA Co anchoring has not been reported yet, and more importantly, this origin underlying the electrocatalytic activity influenced by the SA Co merging is blank and it can be explored reasonably by the intricate interactions of electronic states between SA Co and NiPS₃, following the metal disulfides transcendentally.

In this work, for the first time, we successfully synthesized SA Co anchored on mono-layered NiPS₃ (SA Co NiPS₃) by a gentle cyclic voltammetry leaching process over Co nanodisks/NiPS₃ precursors (Co@NiPS₃). This isolated Co atoms have a uniform unsaturated coordination mode through three Co–S covalent bonds, which was confirmed by aberration-corrected scanning transmission electron microscopy (STEM), X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure (XAFS) spectroscopy. As expected, electrochemical evaluates over SA Co NiPS₃ electrodes in 1.0 M KOH exhibit superior HER and OER activities with very low overpotentials of 1.19 and 1.46 V, respectively at a current density of 50 mA cm⁻². More significantly, when tested as overall water splitting electrocatalysts, a SA Co NiPS₃-based two-electrode setup delivers a current density of 50 mA cm⁻² at only 1.60 V and good stability with only ∼3.4% decay over 6 days of uninterrupted measurement, which obviously surpass commercial Pt/C//RuO₂ electrodes. These interesting results open up an avenue in designing SA-based MPS₃ catalysts with remarkable water splitting performance.

2. Results and discussion

First, density functional theory (DFT) calculations were implemented to predict the effect of the two different configurations of Co atoms on the electronic properties and electrocatalytic hydrogen evolution activity of NiPS₃. The original NiPS₃ monolayer exhibits a sandwich structure comprising octahedral NiS₆ clusters surrounded by three inverted trigonal bipyramidal P₂S₆ clusters, as shown in Fig. 1a. Based on the theoretical optimized results among different structures regarding single Co atom-decorated NiPS₃, there are Co atoms substituting Ni atoms (CoNiPS₃) and Co atoms anchored on the surface of NiPS₃ (SA Co NiPS₃). According to the calculated results, CoNiPS₃ is the more preferential configuration compared to SA Co NiPS₃ because of the milder formation energy (−2.7833 and −6.9561 eV for CoNiPS₃ and SA Co NiPS₃, respectively), which can explain the fact that it is blank for the SA metal decorating on NiPS₃ until now. As shown in Fig. 1b, all the total density of states (TDOS) calculated for NiPS₃, CoNiPS₃, and SA Co NiPS₃ are discontinuous near the Fermi levels (E_f), suggesting the semiconductor features of NiPS₃ kept well after Co merging with different coordination modes. In particular, after decorating Co atoms on/into NiPS₃ cells, CoNiPS₃ and SA Co CoPS₃ show smaller energy bands of 1.45 eV and 1.38 than that of NiPS₃ (1.53 eV), which implies that more charge carriers are involved in the catalytic reaction, thus significantly boosting the HER performance.³³ Then, DFT calculations were further carried out to evaluate the effects of saturated and unsaturated Co atoms on the catalytic activity of mono-layered NiPS₃via monitoring the hydrogen adsorption free energy (ΔG_H*) of three different catalysts. It is generally accepted that an almost zero ΔG_H* means a good HER electrocatalyst, which possesses the approachable H adsorption and satisfactory H₂ desorption.³⁴ As displayed in Fig. 1c, the ΔG_H* value obviously decreases from 0.397 eV (NiPS₃) to 0.168 eV (CoNiPS₃) when one Ni atom was in situ substituted by one Co atom, indicating the lower energy barriers for H₂ formation over CoNiPS₃, in line with previous reports on MPS₃.^22,35 Surprisingly, ΔG_H* of SA Co NiPS₃ displays a much closer value to the thermoneutral value of only 0.078 eV, which benefits the compromise of reaction barriers between adsorption and desorption processes, resulting in a further enhanced HER performance. A similar tendency was also discovered on the theoretical value of H₂O molecule adsorption energy, as displayed in Fig. 1d. The SA Co NiPS₃ catalyst displays the strongest binding strength with the initial H₂O (0.234 eV) transcending that of NiPS₃ (0.132 eV) and CoNiPS₃ (0.192 eV), guaranteeing a smooth activation of H₂O for the efficient HER, and the theoretical conclusion matching well with the experimental results. The above-mentioned theoretical calculation results motivate us to implement the assembly of the Co SA-anchored single-layered NiPS₃ with uniform unsaturated coordination mode as electrocatalysts for water splitting applications.

Fig. 1 (a) Optimized structure of NiPS₃, NiCoPS₃, and SA Co NiPS₃. (b) Calculated TDOS for NiPS₃, NiCoPS₃, and SA Co NiPS₃. (c) DFT calculated free-energy diagram for the HER of NiPS₃, NiCoPS₃, and SA Co NiPS₃. (d) H₂O molecular absorption energy diagram for NiPS₃, NiCoPS₃, and SA Co NiPS₃.

Fig. 2a displays the schematic of the preparation of the SA Co NiPS₃ catalysts via an ultrasonic assembly followed by a leaching method, which ensures the successful introduction of single Co atoms on the surface other than substitution doping of Ni. Uniform Co nanodisks (Fig. S1, ESI†), as a smart intermediation, were first assembled on the basal plane of mono-layered NiPS₃ (Fig. S2, ESI†) to achieve Co@NiPS₃ heterojunction via sonication-induced Co–S bonding, which was removed next by an electrochemical CV leaching process,³⁶ finally, leaving single Co atoms anchored on NiPS₃ (SA Co NiPS₃) plane. Phonons at ultrasonic power may provide sufficient energy to overcome the barrier for the formation of 2D/2D heterojunctions (Co@NiPS₃) via Co–S bond formation.³⁷ TEM image, Raman spectra (Fig. S3, ESI†), and XRD patterns (Fig. S4, ESI†) present that the Co nanodisks are well contacted with NiPS₃ after sonication. From Fig. 1b, the TEM image of SA Co NiPS₃ clearly shows a smooth lamellar morphology with a diameter of 50–100 nm, and the nearly transparent feature indicates the ultrathin thickness of the catalyst. The regular bright spots with a hexagonal symmetry from the SAED pattern (inset of Fig. 2b) indicate a single-crystalline structure exposed with the (001) facets along their primary surfaces. Meanwhile, the good structure stability of SA Co NiPS₃ can also be demonstrated via morphology comparison (Fig. S2b and 1b†), which even suffers from sonication as well as cyclic voltammetry leaching processes (Fig. S5, ESI†). To clarify the accurate distribution of the Co atoms, aberration-corrected HAADF-STEM characterization is executed as shown in Fig. 1c. First, one set of lattice fringes with distances of about 0.583 and 1.011 nm were marked, which matched well with that of the standard NiPS₃ crystal.³⁸ Second, the isolated bright spots on the hexagonal crystal base were highlighted by purple circles, which are assigned to Co atoms comfortably according to the atomic weight. Besides, the corresponding line profiles following the two orientations (marked by the blue rectangle) were extracted from the STEM image and monitored in Fig. 1d, further testifying the homogeneous distribution of Co single atoms on the basal plane of NiPS₃, and the location of the Co atom overlaps with that of the P atom.

Fig. 2 (a) Schematic of the synthesis process of the SA Co NiPS₃. (b) TEM images of SA Co NiPS₃; the inset shows the corresponding SAED pattern. (c) Aberration-corrected HAADF-STEM image of SA Co NiPS₃ together with the line profiles (d) extracted from the areas marked with blue rectangles of (c). (e) Co K-edge XANES of SA Co NiPS₃ and fitted curve. The inset shows the atomic structure of SA Co NiPS₃. (f) FT-EXAFS spectra of various Co-based catalysts at the Co K-edge. (g) Corresponding wavelet transform analysis for the k2-weighted Co K-edge EXAFS of Co foil (top) and SA Co NiPS₃ (bottom).

In previous reports about metal-based semiconductor catalysts, the metal atoms have been usually verified as the catalyst active center. For the SA metal catalyst system, the local coordination environment and the electronic structure of the SA metal also play a crucial role in the catalytic process.³⁹ To investigate the electronic properties and coordination features of Co species in SC Co NiPS₃, EXAFS and XAFS techniques were conducted on a series of Co-based samples. The XANES spectrum of the Co K-edge for SA Co NiPS₃ with Co foil, CoS_x and Co@NiPS₃ as contrasts (Fig. S6†) indicates that the Co SAs possess higher valence states than that of CoS_x and Co@NiPS₃ according to the profile of SA Co NiPS₃ (between those of the Co foil and CoS_x or Co@NiPS₃),⁴⁰ caused by the free electrons associated with the Co SAs, which were partially depleted to the bonding S atoms. It is generally accepted that the pre-edge characteristic is due to the 1s → 3d orbital forbidden transition, which would be excluded by dipole selection rules for a symmetry site.⁴¹ Taking the SA Co NiPS₃ as a model (the inset of Fig. 2e: Co atom was located upon the P atom through the coordination of three adjacent S atoms), we also simulated the XANES spectrum,⁴² as displayed in Fig. 2e. Clearly, the same pre-edge and white line energy values were marked at 7710.3 and 7717.9 eV, respectively, further demonstrating our expectation beginning: the single Co atom is right at the P atom atop site, consistent with the above-mentioned HAADF-STEM results (Fig. 2c). For EXAFS spectra, the best-fit results of Co@NiPS₃ and SA Co NiPS₃ are summarized in Fig. S7 and Table S1,† respectively. From Fig. 2f, undoubtedly, a single strong signal recorded at ∼1.78 Å in the R-space of the EXAFS spectrum suggests the exclusive existence of the Co–S bond in SA Co NiPS₃.⁴³ To more deeply uncover the dispersion of Co atoms on the basal planes of mono-layered NiPS₃, wavelet transform of the Co K-edge EXAFS oscillations on the Co foil and SA Co NiPS₃ are displayed in Fig. 2g. For the Co foil, only one wavelet transform focus at 7.2 Å⁻¹ was found, which corresponds with the previous reports on Co–Co active centers.⁴⁴ The single signal is concentrated at 6.5 Å⁻¹ over the SA Co NiPS₃ catalyst, in line with the Co–S bond features.²⁷ This visual analysis of wavelet transform EXAFS results again confirmed the fact that Co species were individually dispersed as single atoms, consistent with the HAADF-STEM images.

AFM was employed to accurately monitor the thickness of the as-prepared ultrathin catalysts. The AFM picture of NiPS₃ (Fig. 3a) and the corresponding height profile (Fig. 3b) clearly displayed that the average thickness of exfoliated NiPS₃ is ∼0.95 nm, which agrees well with the thickness of mono-layered NiPS₃.⁴⁵ Suffering from the ultrasonic and leaching processes, mono-layered NiPS₃ well maintained the thickness, as shown in Fig. 3c and d, which exhibited superior structure stability simultaneously. HAADF-STEM (Fig. 3e) and corresponding STEM-EDX element mapping images (Fig. 3f–i) of the SA Co NiPS₃ catalyst present that Ni, Co, S and P were all uniformly distributed over the 2D basal surface. Additionally, the EDX mapping spectrum reveals that the atomic ratio of Ni : P : S is close to 1 : 1 : 2.89, which conforms to the theoretical value and the atomic ratio of Ni : Co is about 1 : 0.053, matching well with the result of cursory EDS (1 : 0.062) and accurate ICP-AES (1 : 0.057), as shown Fig. S8 and Table S2† respectively. For these mono-layer-based catalyst samples, XPS surface analysis was more accurate for uncovering the electronic properties and bonding configurations compared to other nanostructures. For blank single-layered NiPS₃ flakes (the top in Fig. 3j), six obvious peaks at 881.8, 875.6, 871.7, 864.1, 858.6, and 854.0 eV that deconvoluted from the high-resolution Ni 2p XPS curve could be assigned to the spin–orbit doublets of 2p_1/2 and 2p_3/2 and the corresponding satellites from Ni²⁺ species, respectively.⁴⁶ After Co nanodisk merging, the six primary XPS signals from the Co@NiPS₃ heterojunction kept well except the overall shifts to higher binding energies (∼0.3 eV) slightly, suggesting the decrease in the electron density around Ni atoms.⁴⁷ Further shifts toward higher binding energies (∼0.4 eV) can be found for the SA Co NiPS₃ sample in comparison with Co@NiPS₃, demonstrating the stronger electron delocalization from Ni atoms, which was influenced by the introduced Co SAs. This unidirectional electron migration behavior may guarantee enough electron requirements in a highly active HER process. Furthermore, Fig. 3k summarizes the comparison on Co 2p spectra features of Co nanodisks, Co@NiPS₃, and SA Co NiPS₃. After 30 h of continuous ultrasonic treatment, a distinct Co–S signal appeared at ∼776.9 eV, indicating that the strong interaction in Co@NiPS₃ was successfully established between Co nanodisks and NiPS₃ nanosheets. It is not difficult to find that the Co–Co bond disappears thoroughly and the Co–S bond heightens to a primary peak from the Co 2p curve of SA Co NiPS₃ after continuous 200th electrochemical leaching, which coincides with the results of XAFS and EDX.

Fig. 3 AFM images of NiPS₃ (a) and SA Co NiPS₃ (b). The corresponding height profiles of NiPS₃ (c) and SA Co NiPS₃ (d). (e) HAADF-STEM image and corresponding STEM-EDX element mappings on SA Co NiPS₃: Ni (f), P (g), S (h), and Co (i). High-resolution XPS spectra for Ni 2p (j) of NiPS₃, Co@NiPS₃, and SA Co NiPS₃. Co 2p (k) of Co nanodisks, Co@NiPS₃, and SA Co NiPS₃.

To understand the effect of SAs Co in NiPS₃ on their HER activity, the electrocatalytic activity towards the HER over the three NiPS₃-based catalysts and Pt/C benchmark were evaluated by a standard three-electrode system in an alkaline electrolyte (1.0 M KOH) (see the Experimental section for details, ESI†). Fig. 4a displays the polarization curves recorded by linear sweep voltammetry (LSV) tests. Clearly, the HER performance from the pristine NiPS₃ monolayer was comparable to that of ultrathin MPS₃ catalysts recently reported.^22,48–50 In comparison, the Co@NiPS₃ electrode only requires an overpotential of 153.5 mV (vs. RHE) to generate a current density of 10 mA cm⁻², much lower than that of the blank NiPS₃ monolayers. This boosting can be ascribed to the electronic modulation of the metal atoms, implied by the results of XPS. Impressively, after anchoring Co SAs, the HER activity of SA Co NiPS₃ significantly enhances closing to the commercial Pt/C benchmark (only 20 mV gap at 10 mA cm⁻²), which is among the best values in layered electrocatalysts (Table S3†). Based on the Butler–Volmer equation and its derivations, the Tafel plots of these four electrodes were fitted to obtain the Tafel slopes for confirming the rate-determining steps of the catalysts (Fig. 4b). As shown in Fig. 4b, the pure NiPS₃ and Co@NiPS₃ heterojunctions output the Tafel slopes of 77.2 and 102.1 mV dec⁻¹, respectively, which can be classified as the Volmer rate-determining step, a step of the primary discharge of protons.^51,52 In sharp contrast, the SA Co NiPS₃ electrode exhibits very low Tafel slopes of 38.8 mV dec⁻¹, close to that of Pt/C (36.0 mV dec⁻¹) at which the Heyrovsky reaction pathway is the rate-determining step. The lower Tafel slope from SA Co NiPS₃ than the original NiPS₃ monolayer as well as the Co@NiPS₃ heterojunction indicates much better charge transfer over the SA Co NiPS₃ flake, agreeing well with the electrochemical impedance spectroscopy tests that display the lowest reaction resistance of the SA Co NiPS₃ electrocatalyst (Fig. S9†). The potentiostatic test (Fig. S10†) presents no significant change in the current density (only 5.2% decline) under a potential of −0.10 V (vs. RHE) for 6 days, suggesting an outstanding catalytic stability from the SA Co NiPS₃ catalyst. Besides, the robust stability is further verified by the comparison of polarization curves before and after 10 000 cycles tests (Fig. S11†).

Fig. 4 (a) HER polarization curves, (b) corresponding Tafel plots, (c) OER polarization curves, (d) corresponding Tafel plots, and (e) polarization curves for water splitting of NiPS₃, Co@NiPS₃, and SA Co NiPS₃. Pt/C, RuO₂, and RuO₂//Pt/C were employed as the reference standard for the HER, OER, and overall water splitting, respectively. (f) The chronoamperometric curves of SA Co NiPS₃ for electrocatalytic water splitting at 1.6 V applied potential. The inset in (f) is the optical image of water electrolysis using SA Co NiPS₃ as both the cathode and the anode. All of the obtained potentials were directly calibrated into a reversible hydrogen electrode (RHE) and performed with iR correction (80%).

The electrocatalytic ability of the SA Co NiPS₃ NRs for OER application was also detected by LSV in 1.0 M KOH electrolyte. Fig. 4c records the OER activity of the three as-prepared electrodes above and RuO₂ as the reference. An obvious peak located at ∼1.38 V (vs. RHE) can be clearly observed in the LSV curves of the three NiPS₃-based anodes due to the Ni²⁺/Ni³⁺ redox reaction.⁵³ Focusing on the profile of polarization curves, beyond expectation, SA Co NiPS₃ and Co@NiPS₃ display an extraordinary OER activity with an onset potential of 146.7 mV (vs. RHE) and an overpotential of 151.8 mV (vs. RHE) at a current density of 50 mA cm⁻², surpassing the commercial RuO₂ benchmark. An exploration of the electrocatalytic OER kinetics over the three NiPS₃-based samples with RuO₂ was further probed based on the Tafel plot method, as shown in Fig. 4d. As expected, the Tafel slope of the SA Co NiPS₃ electrode (52.2 mV dec⁻¹) is far below that of Co@NiPS₃ (69.7 mV dec⁻¹), NiPS₃ (90.8 mV dec⁻¹), and RuO₂ benchmark (78.2 mV dec⁻¹). Overall, the SA Co NiPS₃ sample presents impressive OER catalytic activity, even superior to most of the reported 2D electrocatalysts in an alkaline environment (Table S4†). Meanwhile, the chronopotentiometry test suggests superior cyclic stability of the SA Co NiPS₃ catalyst with the 96.3% retention of voltage (Fig. S12†). Considering the results of the durability measurement before and after 10 000 cycles (Fig. S13†), together with the contrast of XPS characterization before and after HER/OER cycles³⁵ (Fig. S14–S17†), we have sufficient evidence to confirm the astounding activities and stabilities over the as-synthesised bi-functional SA-based electrocatalysts. As a common cognition in heterocatalysis, specific surface area and porosity features of the catalyst are crucial factors that significantly impact catalytic performance. N₂ adsorption–desorption isotherms were carried out to estimate the specific surface area and porosity of the three catalysts (Fig. S18 and S19†). Notably, the NiPS₃ (39.5 m² g⁻¹) and SA Co NiPS₃ (41.7 m² g⁻¹) nanoflakes have analogous surface areas and pore size distributions because of the good structural stability from the advanced preparation strategies. After cycle tests, the BET data maintained well, which also proved the stable performance of SA Co NiPS₃. From Fig. S20 in the ESI,† the electrochemically active surface area (ECSA) assessed by C_dl displays that SA Co NiPS₃ (31.8 mF cm⁻²) output much larger ECSA than that of Co@NiPS₃ (17.3 mF cm⁻²) and NiPS₃ (7.8 mF cm⁻²), implying that the enhanced electrocatalytic activity was triggered from the Co SAs anchoring. Motivated by the high catalytic activity from the obtained NiPS₃-based catalysts toward both the HER and the OER in a KOH (1 M) solution, we further employed these NiPS₃-based electrodes as both the anode and the cathode for overall water splitting in the same electrolysis media. As recorded in Fig. 4e, a current density of 50 mA cm⁻² can be obtained at a cell voltage as low as 1.603 V from the polarization curve of SA Co NiPS₃, beyond that of the other two NiPS₃-based samples (1.658 and 1.769 V for Co@NiPS₃ and NiPS₃, respectively). Impressively, the catalytic activity from the SA Co NiPS₃ electrode toward overall water splitting was much higher than that of Pt/C coupled with the IrO₂/C electrode system (1.731 V@50 mA cm⁻²). In addition, 6 days chronoamperometric (Fig. 4f) and 10 000 cycles testing (Fig. S21†) over the optimal electrocatalyst (SA Co NiPS₃) exhibit only a slight deactivation, proving the excellent stability of the SA Co NiPS₃ in an overall water splitting application. Besides, the outstanding stability of SA Co NiPS₃ was further confirmed by the comparison of the nondestructive characterization before and after water splitting tests, including XRD, aberration-corrected HAADF-STEM, EDS (Fig. S22†), and Raman spectroscopy (Fig. S23†). This performance from SA Co NiPS₃ toward overall water splitting was comparable to that of the other 2D-based electrocatalysts reported presently (Table S5†).

To gain insights into the effect of a single Co atom on the electrocatalysis activity of NiPS₃, DFT calculations were performed on SA Co NiPS₃ with Co in situ replaced NiPS₃ as the reference. For the HER, the advantage of hydrogen evolution over SA Co NiPS₃ was verified through the obtained lower ΔG_H* value from the SA Co NiPS₃ (Fig. 1c). In the four electron mechanism of the OER process as presented in Fig. 5a and b, a single bond between oxygen intermediates and the surface of catalyst is the only attended mode for simplification.⁵⁴ All the possible catalytic active sites influenced by the Co atom were fixed as the adjacent S, P, and Co atoms themselves. Bonding with the three different sites, the Gibbs free energies of the four steps over NiCoPS₃ and SA Co NiPS₃ are shown in Fig. 5c and d: low adsorbed energy difference between ΔG_O* and ΔG_OH* is desirable; the largest difference between two connected energy levels is identified as the rate determining step.⁵⁵ For the three different sites on NiCoPS₃, the rate determining step in each case is inconsistent: the second step for connecting with S–Co and P–S–Co (OH* → O*); the third step for connecting with Co (O* → OOH*), with the calculated determining potentials of 1.73, 1.01, and 0.49 V respectively. In view of the much lower determining potential, the substituted Co atom site in Co@NiPS₃ was identified for optimal catalytic activities, which should be due to the partial charge transfer effect of Ni atoms (Fig. 3j). Changing the Co coordination mode into an unsaturated state, the SA Co NiPS₃ sample demonstrated a similar rate determining step to Co@NiPS₃. However, the connection with Co path delivers an ultralow determining potential of 0.19 V, the lowest theoretical potential barrier value in all possible attended mode,^56,57 which agrees well with our experimental results.

Fig. 5 DFT calculations of the NiCoPS₃ and SA CoNiPS₃ for OER: (a) three different connect sites around the alternative Co atom and (b) the Gibbs free energy diagram for the four steps of OERs on NiCoPS₃. (c) Three different connect sites around the anchoring Co atom and (d) the Gibbs free energy diagram for the four steps of OER on SA Co NiPS₃ in alkaline media.

The promising catalytic performance of the designed SA-based NiPS₃ electrocatalyst could be ascribed to the following factors: (i) stable monolayered configuration provides abundant reaction sites and large ECSA, increasing electrochemical performance; (ii) the synergistic effect of bimetallic composition results in higher intrinsic activities, better conductivity, and lower energy barriers, which benefit for the electrocatalytic activity; (iii) the almost zero determining potentials from the SA Co site favor the OER activity, which acts as a crucial factor for the overall water splitting process; (iv) unsaturated SA Co design anchoring on the basal plane of NiPS₃ single layers causes monodirectional electron migration from nanoflakes to enthetic Co, triggering the activity of water splitting thoroughly. Overall, these combined factors synergistically contribute to the efficient and stable electrocatalytic water splitting performance of the SA Co NiPS₃ nanoflakes.

3. Conclusion

In conclusion, enlightened by the single-atom metals boosting the electrocatalytic property of layered MoS₂, we have designed and assembled single-atom Co decorating on the basal plane of NiPS₃ monolayers via a top-down assembly/leaching strategy. The as-prepared SA Co NiPS₃ catalyst exhibits impressive catalytic activities with ultralow overpotentials of 1.21 and 1.47 V at 50 mA cm⁻² for HERs and OERs, respectively. More importantly, this catalytic performance has great potential to achieve overall water splitting via delivering a very low cell voltage of 1.60 V at 50 mA cm⁻² with excellent durability, outperforming the noble metallic Pt/C//RuO₂ couple, which holds promise for applications in large-scale water splitting electrolyzers. Additionally, theoretical investigations demonstrate that Co SA anchored on the basal plane of NiPS₃ monolayers is a crucial active site for HERs with ultrafast hydrogen adsorption/desorption kinetics; meanwhile, for OERs with the lowest determining potentials, both originates from strong hybridization between the single Co atoms and water. Our work about the SA Co-decorated NiPS₃ highlights the area of modifying the monolayer metal thiophosphate catalysts on the atomic scale, which could significantly make a footprint in advanced technological development for large-scale industrial-grade applications.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

We acknowledge National Natural Science Foundation of China (52072182, U1732126, and 51872145), Natural Science Foundation of Jiangsu Province (BK20211278), China Postdoctoral Science Foundation (No. 2019M650120 and 2020M671554), and National Synergetic Innovation Center for Advanced Materials (SICAM).

Notes and references

1. H. Huang, M. Yan, C. Yang, H. He, Q. Jiang, L. Yang, Z. Lu, Z. Sun, X. Xu, Y. Bando and Y. Yamauchi, Adv. Mater., 2019, 31, 1903415.
2. W. Chen, G. Huang, H. Song and J. Zhang, J. Mater. Chem. A, 2020, 8, 20963–20969.
3. T. Ouyang, X. Wang, X. Mai, A. Chen, Z. Tang and Z. Liu, Angew. Chem., Int. Ed., 2020, 59, 11948–11957.
4. J. Wang, W. Liu, X. Li, T. Ouyang and Z. Liu, Chem. Commun., 2020, 56, 1489–1492.
5. J. Wei, M. Cao, X. Guo, S. Ye and Z. Liu, Small Struct., 2021, 2, 2100047.
6. J. Chen, Q. Long, K. Xiao, T. Ouyang, N. Li, S. Ye and Z. Liu, Sci. Bull., 2021, 66, 1063–1072.
7. K. Xiao, J. Wei, W. Han and Z. Liu, J. Power Sources, 2021, 487, 229408.
8. J.-H. Qin, P. Xu, Y.-D. Huang, L.-Y. Xiao, W. Lu, X.-G. Yang, L. Ma and S.-Q. Zang, Chem. Commun., 2021, 57, 8468–8471.
9. J. H. Qin, H. Zhang, P. Sun, Y. D. Huang, Q. Shen, X. G. Yang and L. F. Ma, Dalton Trans., 2020, 49, 17772–17778.
10. J. Li, H. Huang, X. Cao, H. Wu, K. Pan, Q. Zhang, N. Wu and X. Liu, Chem. Eng. J., 2021, 416, 127677.
11. J. Li, X. Liu and J. Zhang, ChemSusChem, 2020, 13, 2996–3004.
12. C. Zhang, X. Liang, R. Xu, C. Dai, B. Wu, G. Yu, B. Chen, X. Wang and N. Liu, Adv. Funct. Mater., 2021, 31, 2008298.
13. N. Yao, P. Li, Z. Zhou, R. Meng, G. Cheng and W. Luo, Small, 2019, 15, 1901993.
14. Y. Yang, H. Yao, Z. Yu, S. Islam, H. He, M. Yuan, Y. Yue, K. Xu, W. Hao, G. Sun, H. Li, S. Ma, P. Zapol and M. Kanatzidis, J. Am. Chem. Soc., 2019, 141, 10417–10430.
15. J. Zhang, L. Wang, X. Liu, X. Li and W. huang, J. Mater. Chem. A, 2015, 3, 535–541.
16. Z. Lv, W. Ma, M. Wang, J. Dang, K. Jian, D. Liu and D. Huang, Adv. Funct. Mater., 2021, 31, 2102576.
17. T. Zhao, X. Shen, Y. Wang, R. Hocking, Y. Li, C. Rong, K. Dastafkan, Z. Su and C. Zhao, Adv. Funct. Mater., 2021, 31, 2100614.
18. Z. Chen, M. Ju, M. Sun, L. Jin, R. Cai, Z. Wang, L. Dong, L. Peng, X. Long, B. Huang and S. Yang, Angew. Chem., Int. Ed., 2021, 60, 9699–9705.
19. J. Zhang, R. Cui, C. Gao, L. Bian, Y. Pu, X. Zhu, X. Li and W. Huang, Small, 2019, 17, 1904688.
20. J. Song, S. Qiu, F. Hu, Y. Ding, S. Han, L. Li, H. Chen, X. Han, C. Sun and S. Peng, Adv. Funct. Mater., 2021, 31, 2100618.
21. X. Li, Y. Fang, J. Wang, B. Wei, K. Qi, H. Hoh, Q. Hao, T. Sun, Z. Wang, Z. Yin, Y. Zhang, J. Lu, Q. Bao and C. Su, Small, 2019, 17, 1902427.
22. Q. Liang, L. Zhong, C. Du, Y. Zheng, Y. Luo, J. Xu, S. Li and Q. Yan, Adv. Funct. Mater., 2018, 28, 1805075.
23. H. Li, C. Tsai, L. Ai, L. Cai, A. Contryman, A. Fragapane, J. Zhao, H. Han, H. Manoharan and F. Abild-Pedersen, Nat. Mater., 2015, 15, 48–53.
24. Y. Wang, H. Su, Y. He, L. Li, S. Zhu, H. Shen, P. Xie, X. Fu, G. Zhou, C. Feng, D. Zhao, F. Xiao, X. Zhu, Y. Zeng, M. Shao, S. Chen, G. Wu, J. Zeng and C. Wang, Chem. Rev., 2020, 120, 12217–12314.
25. N. Xuan, J. Chen, J. Shi, Y. Yue, P. Zhuang, K. Ba, Y. Sun, J. Shen, Y. Liu, B. Ge and Z. Sun, Chem. Mater., 2019, 31(2), 429–435.
26. S. Ji, Y. Chen, X. Wang, Z. Zhang, D. Wang and Y. Li, Chem. Rev., 2020, 120, 11900–11955.
27. K. Qi, X. Cui, L. Gu, S. Yu, X. Fan, M. Luo, S. Xu, N. Li, L. Zheng, Q. Zhang, J. Ma, Y. Gong, F. Lv, K. Wang, H. Huang, W. Zhang, S. Guo, W. Zheng and P. Liu, Nat. Commun., 2019, 10, 5231.
28. Y. Luo, S. Zhang, H. Pan, S. Xiao, Z. Guo, L. Tang, U. Khan, B. Ding, M. Li, Z. Cai, Y. Zhao, W. Lv, Q. Feng, X. Zou, J. Lin, H. Cheng and B. Liu, ACS Nano, 2020, 14, 767–776.
29. Z. Li, Z. Wang, S. Xi, X. Zhao, T. Sun, J. Li, W. Yu, H. Xu, T. Herng, X. Hai, P. Lyu, M. Zhao, S. Pennycook, J. Ding, H. Xiao and J. Lu, ACS Nano, 2021, 15, 7105–7113.
30. D. Kuznetsov, Z. Chen, P. Kumar, A. Tsoukalou, A. Kierzkowska, P. Abdala, O. Safonova, A. Fedorov and C. Müller, J. Am. Chem. Soc., 2019, 141, 17809–17816.
31. H. Fei, J. Dong, M. Arellano-Jiménez, G. Ye, N. Kim, E. Samuel, Z. Peng, Z. Zhu, F. Qin, J. Bao, M. Yacaman, P. Ajayan, D. Chen and J. James, Nat. Commun., 2015, 6, 8668.
32. H. Huang, F. Li, Q. Xue, Y. Zhang, S. Yin and Y. Chen, Small, 2019, 15, 1903500.
33. B. Qiu, L. Cai, Y. Wang, Z. Lin, Y. Zuo, M. Wang and Y. Chai, Adv. Funct. Mater., 2018, 28, 1706008.
34. J. Greeley, T. Jaramillo, J. Bonde, I. Chorkendorff and J. Norskov, Nat. Mater., 2006, 5, 909–913.
35. B. Song, K. Li, Y. Yin, T. Wu, L. Dang, M. Cabán-Acevedo, J. Han, T. Gao, X. Wang, Z. Zhang, J. Schmidt, P. Xu and S. Jin, ACS Catal., 2017, 7, 8549–8557.
36. L. Fan, P. Liu, X. Yan, L. Gu, Z. Yang, S. Qiu and X. Yao, Nat. Commun., 2016, 7, 10667.
37. X. Shi, S. Posyasev, M. Huttula, V. Pankratov, J. Hoszowska, J. Dousse, F. Zeeshan, Y. Niu, A. Zakharov, T. Li, O. Miroshnichenko, M. Zhang, X. Wang, Z. Huang, S. Saukko, D. Gonzalez, S. Dijken, M. Altalo and W. Cao, Small, 2018, 14, 1704526.
38. D. Goossens, D. James, J. Dong, R. Whitfield, L. Noren and R. Withers, J. Phys.: Condens. Matter, 2011, 23, 065401.
39. J. Fu, J. Dong, R. Si, K. Sun, J. Zhang, M. Li, N. Yu, B. Zhang, M. Humphrey, Q. Fu and J. Huang, ACS Catal., 2021, 11, 1952–1961.
40. Z. Chen, C. Liu, J. Liu, J. Li, S. Xi, X. Chi, H. Xu, I. Park, X. Peng, X. Li, W. Yu, X. Liu, L. Zhong, K. Leng, W. Huang, M. Koh and K. Loh, Adv. Mater., 2019, 31, 1906437.
41. J. Zheng, S. Wu, L. Lu, C. Huang, P. Ho, A. Kirkland, T. Sudmeier, R. Arrigo, D. Gianoliod and S. Tsang, Chem. Sci., 2021, 12, 688–695.
42. W. Liu, L. Zhang, W. Yan, X. Liu, X. Yang, S. Miao, W. Wang, A. Wang and T. Zhang, Chem. Sci., 2016, 7, 5758–5764.
43. H. Kasai, K. Tolborg, M. Sist, J. Zhang, V. Hathwar, M. Filso, S. Cenedese, K. Sugimoto, J. Overgaard and B. Iversen, Nat. Mater., 2018, 17, 249–252.
44. W. Liu, W. Hu, L. Yang and J. Liu, Nano Energy, 2020, 73, 104750.
45. J. Wang, X. Li, B. Wei, R. Sun, W. Yu, H. Hoh, H. Xu, J. Li, X. Ge, Z. Chen, C. Su and Z. Wang, Adv. Funct. Mater., 2020, 30, 1908708.
46. Q. Liang, Y. Zheng, C. Du, Y. Luo, J. Zhang, B. Li, Y. Zong and Q. Yan, Small Methods, 2017, 1, 1700304.
47. S. Gage, C. Ngo, V. Molinari, M. Causa, R. Richards, F. Gentile, S. Pylypenko and D. Esposito, J. Phys. Chem. C, 2018, 122, 339–348.
48. J. Zhang, R. Cui, X. a. Li, X. Liu and W. Huang, J. Mater. Chem. A, 2017, 5, 23536–23542.
49. R. Gusmeao, Z. Sofer and M. Pumera, Adv. Funct. Mater., 2019, 29, 1805975.
50. Q. Liang, L. Zhong, C. Du, Y. Luo, J. Zhao, Y. Zheng, J. Xu, J. Ma, C. Liu, S. Li and Q. Yan, ACS Nano, 2019, 13, 7975–7984.
51. Y. Cui, Y. Xue, R. Zhang, J. Zhang, X. Li and X. Zhu, J. Mater. Chem. A, 2019, 7, 21911–21917.
52. Y. Luo, X. Li, X. Cai, X. Zou, F. Kang, H. Cheng and B. Liu, ACS Nano, 2018, 12, 4565–4573.
53. S. Hao, L. Chen, C. Yu, B. Yang, Z. Li, Y. Hou, L. Lei and X. Zhang, ACS Energy Lett., 2019, 4, 952–959.
54. B. Zhang, C. Zhu, Z. Wu, E. Stavitski, Y. Lui, T. Kim, H. Liu, L. Huang, X. Luan, L. Zhou, K. Jiang, W. Huang, S. Hu, H. Wang and J. Francisco, Nano Lett., 2020, 20, 136–144.
55. H. Xu, T. Liu, S. Bai, L. Li, Y. Zhu, J. Wang, S. Yang, Y. Li, Q. Shao and X. Huang, Nano Lett., 2020, 20, 5482–5489.
56. D. Hu, X. Wang, X. Chen, Y. Wang, A. Hong, J. Zhong, X. Bu, P. Feng and T. Wu, J. Mater. Chem. A, 2020, 8, 11255–11260.
57. F. Diongi, Z. Zeng, I. Sinev, T. Merzdorf, S. Deshpande, M. Lopez, S. Kunze, I. Zegkinoglou, H. Sarodnik, D. Fan, A. Bergmann, J. Drnec, J. de Araujo, M. Gliech, D. Teschner, J. Zhu, W. Li, J. Greeley, B. Roldan Cuenya and P. Strasser, Nat. Commun., 2020, 11, 2522.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ta08599a

---