Platinum single-atom catalyst coupled with transition metal/metal oxide heterostructure for accelerating alkaline hydrogen evolution reaction

Single-atom catalysts provide an effective approach to reduce the amount of precious metals meanwhile maintain their catalytic activity. However, the sluggish activity of the catalysts for alkaline water dissociation has hampered advances in highly efficient hydrogen production. Herein, we develop a single-atom platinum immobilized NiO/Ni heterostructure (PtSA-NiO/Ni) as an alkaline hydrogen evolution catalyst. It is found that Pt single atom coupled with NiO/Ni heterostructure enables the tunable binding abilities of hydroxyl ions (OH*) and hydrogen (H*), which efficiently tailors the water dissociation energy and promotes the H* conversion for accelerating alkaline hydrogen evolution reaction. A further enhancement is achieved by constructing PtSA-NiO/Ni nanosheets on Ag nanowires to form a hierarchical three-dimensional morphology. Consequently, the fabricated PtSA-NiO/Ni catalyst displays high alkaline hydrogen evolution performances with a quite high mass activity of 20.6 A mg−1 for Pt at the overpotential of 100 mV, significantly outperforming the reported catalysts.

In this reviewer's opinion, it's no longer acceptable to publish computational catalysis investigations without at least testing the effects of implicit and explicit solvation. Since the authors are using VASP, they at least should recalculate their most important predictions using VASPsol. Additionally, they should examine the convergence of these predictions with respect to the number of explicit solvating water molecules (at the active site should suffice).

Modeling strongly correlated NiO with GGA-PBE (a semi-local exchange-correlation functional)
NiO is a Mott insulator due to electron-electron interactions, which are described poorly by DFT with semi-local exchange-correlation functionals like GGA-PBE. While it is not clear from Figure 4, I suspect that the authors predict NiO to be metallic when, at 0 K (to which static DFT calculations correspond), it has a band gap of 3-4 eV. With that being said, since the authors are trying to predict roomtemperature properties, perhaps the use of GGA-PBE for NiO is less problematic. To benchmark the accuracy of their calculations for NiO, the authors should calculate a room-temperature property for which reference experimental data exists, e.g., the NiO formation free energy.
Minor comments/questions 1. "Despite the significant progress that has been presented in nonprecious catalysts, the HER performances are still second to platinum (Pt)-based materials due to its optimal binding ability with hydrogen.7-10" on page 2 While this may be the case in base, there are several excellent electrocatalysts for the HER in acid, e.g., the nickel phosphides.
3. "Compared with the original NiO/Ni ( Figure S2), the exposed PtSA-NiO/Ni nanosheets morphology on Ag NWs should be attributed to the H2-assisted delamination effect 21,34 during Pt electroreduction process in alkaline condition, which will provide more sites for Pt atoms immobilization and improve the HER performance." on page 6 I'm not sure I see the difference between Figure 2(b-c) and Figure S2 to which the authors are referring -can they please clarify? 4. "These results further confirm the formation of single-atom Pt anchored NiO/Ni composition, and the interfacial coupling of Pt single atom with NiO/Ni does not change the phase structure of NiO/Ni." on page 8 The authors should mention explicitly the fact that their theoretical predictions are limited by the fact that they use a crystalline model for amorphous NiO. Dear Editor: We are glad to receive the reviewers' comments from you about our manuscript entitled " Tailoring Water Dissociation Energy by Platinum Single-Atom Catalyst Coupled with Transition Metal/metal Oxide Heterostructure for Accelerating Alkaline Hydrogen Evolution Reaction" (NCOMMS-21-03305). We thank you for your concern about our work. Here, we have noticed that three reviewers have commented on our paper. They give positive comments on our work, and further structural characterization and performance tests are required. We give our sincere thanks to all of the reviewers for taking the time to read and comment on our work.
We understand that the strict comments suggest the high responsibility of reviewers for journal publishers and scientific research. According to the all of comments and suggestions proposed by reviewers, we carefully think about the deficiency of this work and make some significant improvements, so that our work can be competent for publishing on Nature Communications.
The main corrections have been added into the revised manuscript with yellow background, and the responses to the reviewer's comments are as following:

Reviewer #1:
This work describes single Pt-atom coupled with NiO/Ni junction for HER in alkaline electrolytes. The major finding is that Pt single atom located at NiO/Ni junctions accelerate water dissociation and hydrogen desorption, thereby improve HER kinetics. The authors present a large set of experimental and theoretical results to support their claims. Although the results are interesting and could be good contributions to the field, I feel there are some critical problems which I will explain below, such that the experiments are not well designed and the conclusions are not well supported.
Q1: This work is on catalyst material development and if this catalyst is truly exceptional, compelling results in the more challenging neutral electrolytes should be provided.
A1: Thank you for your comments on our work. Your doubt and valuable advice help us improve the quantity of this manuscript. The electrocatalytic HER performances of the catalysts in neutral electrolytes containing 1.0 M phosphate buffer solutions (PBS, pH = 7.0) have been added in Figure S40 and Table S4 in the revised manuscript and also shown as Figure R1 in this letter.
Based on the structural characterizations and theoretical investigations, the Pt single-atom catalyst coupled with NiO/Ni heterostructure possesses extremely highly intrinsic HER activity. As shown in Figure R1a-b, the PtSA-NiO/Ni shows the highest HER performance among all catalysts in 1.0 M PBS neutral electrolytes, and only needs a quite low overpotential of 27 and 159 mV to achieve the current density of 10 and 100 mA cm -2 , respectively, significantly, suggesting the superior neutral HER performance over the PtSA-NiO, PtSA-Ni, NiO/Ni and the Pt/C catalyst. Moreover, Figure R1c presents a small Tafel slope (31.94 mV dec -1 ) for PtSA-NiO/Ni, lower than that of PtSA-NiO (47.26 mV dec -1 ), PtSA-Ni (40.68 mV dec -1 ), and Pt/C catalyst (42.40 mV dec -1 ), revealing fast HER kinetics for NiO/Ni heterostructure coupled Pt single atoms in neutral electrolytes. The above merits of the PtSA-NiO/Ni, including low overpotential and Tafel slope, are superior to most previously reported catalysts in the neutral solution ( Figure R1d and Table S4), further confirming the advance by the constructing single-Pt sites in NiO/Ni hybrid system.  Table S4. Q2: The use of flexible cloth fabric is a big concern, no information about the surface properties and chemical structure is provided and it is unclear how this cloth would affect performance.
A2: Thanks for your valuable advice and the corresponding data have been added in the Figure S1 and Figure S15 section in the revised manuscript and shown as Figure   R2-4 in this letter. Figure R2a  polyester fibers which have good elasticity, wrinkle resistance, shape retention, excellent wash-and-wear performance, and durability 1 . In addition, polyester fibers have good resistance to strong alkalies and acids at room temperature and belong to an excellent insulator. The majority composition of polyester fibers is composed of terephthalic acid and ethylene glycol (PET) as shown in the insert in Figure R2f. 2 The favorable hydrogen release deriving from the interconnected pore structure and high stability originating from the good resistance to strong alkalies will contribute cloth fabric to be used as an excellent substrate for electrocatalytic hydrogen evolution in alkaline media.

As shown in
Further, the wettability of the cloth fabric substrate supported catalyst was investigated by measuring the contact angle of the electrode. As shown in Figure R3, it was difficult to measure the contact angle of the electrode as the water was absorbed by the felt instantaneously, indicating the super hydrophilic nature. The excellent hydrophilicity will boost the electrolyte accessibility, accelerate the mass transfer, reduce the charge transfer resistance of the electrode and increase the durability of the electrode. 3  Besides, the influence of the cloth fabric substrate on the HER performance of the electrocatalyst was investigated by loading Ag NWs on the different substrates, including steel sheet, PET, cloth fabric, and glassy carbon electrodes. As shown in Figure R4, the various substrates supported Ag NWs based electrodes show the negligible difference in HER performances before onset overpotential (potential required to reach the current density of -5 mA cm -2 ), suggesting the negligible effect of the substrates on the intrinsic HER activity of the electrocatalyst. When the current density greater than -5 mA cm -2 , the cloth fabric substrate supported Ag NWs demonstrate slightly high HER performances comparing with other Ag NWs based electrodes, which should be attributed to the favorable mass transfer and hydrogen release derived from interconnected pore structure and the super hydrophilic nature of the cloth fabric substrate supported catalyst electrode as above discussion. Herein, several catalytic systems with different current collectors were introduced as comparable groups. The PtSA-NiO/Ni@Ni foam, PtSA-NiO/Ni@Carbon cloth, and PtSA-NiO/Ni@Cu foam were prepared using the same procedure as the PtSA-NiO/Ni@Ag NWs except for choosing Ni foam, carbon cloth, and Cu foam as the current collector instead of Ag NWs, respectively. As shown in Figure R5a, the response current density of PtSA-NiO/Ni@Ag NWs was higher than other control groups at the same overpotential in the polarization curve, suggesting a superior electrocatalytic HER activity. The PtSA-NiO/Ni@Ag NWs catalytic system could deliver a current density of 10 mA cm -2 at an overpotential of 27 mV, which is lower than that of the PtSA-NiO/Ni@Ni foam (46 mV), PtSA-NiO/Ni@Carbon cloth (40 mV), and PtSA-NiO/Ni@Cu foam (36 mV). Moreover, the mass activity of PtSA-NiO/Ni@Ag NWs normalized to the loaded Pt mass at an overpotential of 100 mV is 20.6 A mg -1 (Figure R5b), which is about 2.1, 1.2, and 1.5 times greater than that of PtSA-NiO/Ni@Ni foam (9.9 A mg -1 ), PtSA-NiO/Ni@Carbon cloth (16.7 A mg -1 ) and PtSA-NiO/Ni@Cu foam (14.2 A mg -1 ), respectively, suggesting that introducing Ag NWs into PtSA-NiO/Ni can extremely maximize the alkaline HER activity of Pt-based catalysts.
To get insight into the origin of the extraordinary HER performance of Ag NWs supported PtSA-NiO/Ni, the HER reaction kinetics of the fabricated Pt-based catalysts were measured by electrochemical impedance spectroscopy (EIS). As depicted in  NiO/Ni junction has been widely studied due to the good HER activity in alkaline electrolytes. Typically, Gong et al. 7 reported the Ni/NiO junction attached to oxidized carbon nanotube (NiO/Ni-CNT) as shown in Figure R6c, which requires a high overpotential of about 120 mV to deliver a current density of 20 mA cm -2 as shown in  Figure R9. A core-shell structure was clearly shown as discussed above.
The "core" layer was composed of Ag elements (Figure R9e), and Ni, O, and Pt elements share the same area located at the "shell" layer ( Figure R9b-d). The powerful ultrasonic treatment results in some damage and exfoliation of NiO/Ni shell layer as director of the red arrow in Figure R9. Interestingly, the evolution of Pt element distribution in the NiO/Ni shell area before and after ultrasonic treatment shows the same feature as that of Ni and O elements, further proving that the Pt atoms mainly deposit on NiO/Ni.     Glass carbon electrodes have been widely used in HER catalysts fabrication as a current collector. 10,11 However, polymer binder (e.g., polytetrafluoroethylene and Nafion) and conductive agent are usually required to hold the active material (HER catalysts) on glass carbon electrode due to the weak binding ability between catalysts and planar glass carbon electrode. As a similar process, PtSA-NiO/Ni supported by Ag NWs loaded glass carbon electrode (PtSA-NiO/Ni@AgNWs@GCE) was fabricated by coating catalyst ink on the current collector using Nafion as an adhesive followed by an electrochemical procedure. In detail, 5 mg Ag NWs, 50 μl Nafion (5 wt% The alkaline HER performances of PtSA-NiO/Ni@AgNWs@GCE were investigated under a standard three-electrode system. As shown in Figure R12a-b, the PtSA-NiO/Ni@AgNWs@GCE shows a negligible difference in HER performances with PtSA-NiO/Ni@AgNWs@CFS when the response current density is lower than -50 mA cm -2 . However, when the current density is greater than -50 mA cm -2 , the HER performances of PtSA-NiO/Ni@AgNWs@GCE demonstrate inferior to that of PtSA-NiO/Ni@AgNWs@CFS. To get insight into the origin of the extraordinary HER performance of PtSA-NiO/Ni@AgNWs@CFS, the reaction kinetics of the fabricated electrodes were measured by EIS. As depicted in Figure R12c, the PtSA-NiO/Ni@AgNWs@CFS exhibits a much low Rct value (0.61 Ω cm -2 ) than PtSA-NiO/Ni@AgNWs@GCE (0.72 Ω cm -2 ), indicating the smaller interfacial charge-transfer resistance for PtSA-NiO/Ni@AgNWs@CFS. Furthermore, the ECSA of the different electrodes was measured using the cyclic voltammetry technique to obtain Cdl. As shown in Figure R12d, the PtSA-NiO/Ni@AgNWs@CFS possessed the higher Cdl (9.0 mF cm -2 ) than PtSA-NiO/Ni@AgNWs@GCE (7.2 mF cm -2 ), suggesting more accessible active sites in PtSA-NiO/Ni@AgNWs@CFS.
From the above discussion, the preparation of the traditional glass carbon electrode-loaded catalysts usually involves coating catalyst fines on the current collectors using adhesives such as Nafion. However, the incorporation of these insulating adhesives will inevitably bury active sites and increases the dead volume and the contact resistance between the catalyst and the current collector, 12 Table S3.

Reviewer #2
The manuscript submitted by Zhou et al. re-ports the synthesis of Pt single atoms immobilized NiO/Ni heterostructure as an alkaline HER catalyst with high stability and small overpotential. They found that the metallic Ni sites and O vacancies modified NiO sites prefer the adsorption for both OH* and H*, which should be facilitated for water dissociation. Moreover, the Pt atoms fixed at the NiO/Ni interfaces could promote the H* conversion and H2 desorption, thus accelerating overall alkaline HER. The reported investigation is interesting and also expands the research area of alkaline HER to achieve better electrochemical performance. Both the experimental and calculated data are well explained. Therefore, the manuscript should be considered to be published after considering the comments below.   This is reflected by the fact that even for Pt foil, the fitting results using symmetric peaks do not fit well with the experimental results. In addition, the peak area between 4f7/2 and 4f5/2 of the same species should be 4:3, which was neglected by the authors.  Ni foam, carbon cloth, and Cu foam as the conductive supporter, respectively. As shown in Figure R5a, the response current density of PtSA-NiO/Ni@Ag NWs was higher than other control groups at the same overpotential in the polarization curve.
The PtSA-NiO/Ni@Ag NWs catalytic system could deliver a current density of 10 mA cm -2 at an overpotential of 27 mV, which is lower than that of the PtSA-NiO/Ni@Ni foam (46 mV), PtSA-NiO/Ni@carbon cloth (40 mV), and PtSA-NiO/Ni@Cu foam (36 mV). Moreover, the mass activity of PtSA-NiO/Ni@Ag NWs normalized to the loaded Pt mass at an overpotential of 100 mV is 20.6 A mg -1 as shown in Figure R5b, which is about 2.1, 1.2, and 1.5 times greater than that of PtSA-NiO/Ni@Ni foam (9.9 A mg -1 ), PtSA-NiO/Ni@carbon cloth (16.7 A mg -1 ) and PtSA-NiO/Ni@Cu foam (14.2 A mg -1 ), respectively, suggesting that introducing Ag NWs into PtSA-NiO/Ni can extremely maximize the alkaline HER activity of Pt-based catalysts. By EIS and ECSA measurement shown in Figure R5c-d, the extraordinary HER performance of Ag NWs supported PtSA-NiO/Ni derive from the higher HER reaction kinetics (0.61 Ω cm -2 for Rct) and the larger electrochemical specific area (9.0 mF cm -2 for Cdl) than others, confirming the unique nanostructure feature and high electrons conductivity of Ag NWs. Based on the above analysis, carbon cloth and Cu foam substrate serve as the potential substitute of Ag NWs under the consideration of the low cost, large-scale manufacturing, and the commercial application for PtSA-NiO/Ni-based catalyst, due to the superior HER performances for carbon cloth and Cu foam supported PtSA-NiO/Ni over most previously reported catalysts deriving from the highly intrinsic HER activity of single-Pt anchored NiO/Ni hybrid system as shown in Figure R15. Figure R15. Comparison of the HER activity for carbon cloth and Cu foam supported Pt SA -NiO/Ni with reported catalysts, originating from Table S3.

Reviewer #3:
The major claims of this paper are (1) the authors synthesized Pt SACs on a NiO/Ni heterostructure, (2) this HER catalyst is exceptionally efficient in alkaline media, and (3) the Pt/NiO/O interface provides "dual" active sites, which facilitate dissociative water adsorption. This work is novel and of interest to the community and wider field because it reports a new catalyst for alkaline HER that is competitive with the state-of-the-art. The experimental evidence is sufficient to justify claims 1 and 2, however, further theoretical evidence is required to justify claim 3 (see major and minor comments/questions below). With that being said, I think this paper will inspire new strategies for optimizing electrolyzers using SACs, heterostructures, and morphology. For these reasons, I recommend publication after the major and minor comments/questions below are addressed: Q1: Regarding Figure 4 A1: Thank you for your encouragement in our work. Your professional advice helps us improve the quantity of this manuscript. The corresponding data have been added in Figure S27 section and Table S2 in the revised manuscript and shown as Figure   R16 and Table R1 in this letter.
According to your professional advice, the H and OH adsorption energies of Ni and NiO with the different coverages of 1/1, 3/4, 1/2, 1/4, and 1/8 are calculated, respectively, as shown in Figure R16 and Table R1.   Additionally, they should examine the convergence of these predictions concerning the number of explicit solvating water molecules (at the active site should be sufficed).
A2: Thanks for your professional advice and the corresponding data have been added in Figure 4h-i and Figure S33 section in the revised manuscript and shown as Figure   R17 in this letter.
The effects of implicit solvation were considered by using VASPsol software, 16 and the revised energy barrier of water desorption and the adsorption free energies of H* for PtSA-NiO/Ni, PtSA-NiO, and PtSA-Ni systems were shown in Figure R17a-  shown in Figure 18.
In fact, in our DFT, the key factor affecting the catalytical speed is the energy barrier of water desorption (as shown in Figure 4h) and the adsorption free energies of H * (as shown in Figure 4i). So, the accuracy of energy is the key point in our DFT calculation. According to the reviewer's opinion, we calculate the room-temperature where ω is the phonon frequencies and g(ω) is the phonon density of states; while Fel(T, V) is the thermal electronic contribution to the free energy, which can be From the calculation result of Helmholtz free energy (Figure R19), we found the energy changes from -11.38 to -11.46 eV per formula unit with the temperature from 0 to 300 K. It proves the energy does not change too much (0.08 eV per formula unit). It should not affect our calculation accuracy apparently, because the energy barrier of water desorption and the adsorption free energies of H * only have a relationship with the interaction between the NiO slab and H2O/H * , but they do not include the energy of the NiO slab. Figure 18. The electronic density of states of pure NiO with bulk structure using static GGA + PBE at 0 K. Figure 19. Helmholtz free energy versus temperature from 0 K to 300 K for NiO. The enthalpy unit is eV per formula unit.
Q4: "Despite the significant progress that has been presented in nonprecious catalysts, the HER performances are still second to platinum (Pt)-based materials due to its optimal binding ability with hydrogen. Despite the significant progress that has been presented in nonprecious catalysts, 20,21 the platinum (Pt)-based materials are still regarded as the most active catalysts for HER due to their optimal binding ability with hydrogen. [22][23][24][25] Q5: "Transmission electron microscopy (TEM, Figure S3a-b) images, high-resolution TEM (HRTEM, Figure S3c) image, fast Fourier transform (FFT, Figure S3d), and elemental mapping ( Figure S4) Figure R20, suggesting the high stability of NiO/Ni heterostructure. Q6: "Compared with the original NiO/Ni ( Figure S2), the exposed PtSA-NiO/Ni nanosheets morphology on Ag NWs should be attributed to the H2-assisted delamination effect 21,34 during Pt electro-reduction process in alkaline condition, which will provide more sites for Pt atoms immobilization and improve the HER performance." on page 6. I'm not sure I see the difference between Figure 2(b-c) and Figure S2 to which the authors are referring -can they please clarify?
A6: Thanks for pointing this out. The corresponding descriptions have been modified and clarified in the revised manuscript.
During the single-atom Pt electro-reduction process, some quantities of hydrogen bubbles are generated and released due to the high cathodic potentials between 0 V and -0.50 V versus reversible hydrogen electrode (RHE) in alkaline conditions. 26 In this case, the unchanged PtSA-NiO/Ni nanosheets morphology on Ag NWs (Figure   2b-c) compared with the original NiO/Ni (Figure S3) indicates the high structural stability of the catalyst for HER application, and the exposed NiO/Ni nanosheet could provide more sites for Pt atoms immobilization and improve the HER performance.