Atomic-level polarization in electric fields of defects for electrocatalysis

The thriving field of atomic defect engineering towards advanced electrocatalysis relies on the critical role of electric field polarization at the atomic scale. While this is proposed theoretically, the spatial configuration, orientation, and correlation with specific catalytic properties of materials are yet to be understood. Here, by targeting monolayer MoS2 rich in atomic defects, we pioneer the direct visualization of electric field polarization of such atomic defects by combining advanced electron microscopy with differential phase contrast technology. It is revealed that the asymmetric charge distribution caused by the polarization facilitates the adsorption of H*, which originally activates the atomic defect sites for catalytic hydrogen evolution reaction (HER). Then, it has been experimentally proven that atomic-level polarization in electric fields can enhance catalytic HER activity. This work bridges the long-existing gap between the atomic defects and advanced electrocatalysis by directly revealing the angstrom-scale electric field polarization and correlating it with the as-tuned catalytic properties of materials; the methodology proposed here could also inspire future studies focusing on catalytic mechanism understanding and structure-property-performance relationship.

2. There is no any simulation to support the polarized electric field distribution of the antisite defect observed by DPC in the experiment.The determination on the polarized in atomic field distribution images also should be addressed more clearly in Fig. 4.
3. Compared to the atomic electric field, the charge density mapping in atomic-scale is a more direct indicator for the asymmetric charge distribution of antisite defect.This could also be obtained by the DPC technique but missed in this work.
Reviewer #2 (Remarks to the Author): I would like to complement the efforts put in by the authors in bringing out this very nice piece of challenging work.The importance of this study is showing the potential of differential phase contrast STEM imaging in revealing the atomic level surface charges which have an important consequence on several electrochemical phenomena.While the idea and the execution of the study supported by relevant theoretical modeling bring out the central idea proposed here, I have a few suggestions on the way the sentences are currently constructed as well as a few technical suggestions which in my opinion would make the claims in this study stronger.Technical inputs: 1.The authors start by claiming in the abstract "For the first time, it has been experimentally proven that atomic-level polarization in electric fields can enhance catalytic HER activity", further in the text, they cite literature with the statement "A large number of advanced electrocatalysts have atomic defect structures on their surfaces.These defects can alter the electric field/charge distribution of electrocatalysts with enhanced catalytic performances7 -13" This is very correct and acceptable.Defects certainly influence the surface-potential/EDL.A very recently published article Xu, Y., Ma, Y. B., Gu, F., Yang, S. S., & Tian, C. S. (2023).Structure evolution at the gate-tunable suspended graphene-water interface.Nature, 1-5 also establishes something in these lines.From the DFT calculations done by the authors the link between the formation of defects and variation in surface potentials is shown whereas to show how this influences the catalytic activity, is it possible to establish something in the lines of this study of Swift, M. W., Swift, J. W., & Qi, Y. (2021).Modeling the electrical double layer at solid-state electrochemical interfaces.Nature Computational Science, 1(3), 212-220.If not a complete study, a 1-dimensional variation of potential across the cross-section of electrolyte-MoS2 with the presence of a defect perhaps supports the claimed results substantially?From the previously published studies of the same group, it appears like they have ample experience in this methodology and computation to go one step further and show the variation of surface potential across a monolayer-electrolyte interface in the presence and absence of defects.2. On a monolayer like MoS2, while imaging at 300kV, it is highly likely that you are creating defects while imaging.Do the authors observe this when they obtain a series of images from the same region?3.If newer defects are formed potentially by the knock-on damage by the electron beam, you are constantly changing the surface potential by creating newer defects.Have any dose optimisation tests been done?4. The creation of electrodes to carry out 3-electrode electrochemical measurements is certainly a commendable job.The device fabrication involves spincoating and liftoff.Doesn't the fabrication process alter the surface potentials/state of defects in the MoS2.Wouldn't the creation of electrodes first on the substrate and transferring the MoS2 later by a transfer method ( Schneider, Grégory F., et al. "Wedging transfer of nanostructures."Nano letters 10.5 (2010): 1912-1916) give a more accurate estimation of surface potentials in the electrochemical measurements? 5. How reproducible are the electrochemical measurements?Are the Tafel slopes indicating a direct correlation with the defects based on the synthesis procedure, or is there an influence of the device fabrication method of spin-coating and lift-off and subsequent processes?6.Overall, the authors observe a variation in the surface potential due to the presence of defects clearly from electrochemical measurements.They also simulate the same by introducing defects.If they manage to make it a little more clearer in the written text or support with an additional simulation that the surface potential may alter in presence of electrolyte and in addition introducing defects alters further and if they manage to establish a link with the electrochemical experimental observation, this would make the outcome even more stronger.Nonetheless, not having this strictly doesn't demerit any of the observations in this study.Suggestions on rephrasing the sentences: If I understand what the authors are trying to communicate, my advice is to simply a few of these sentences.Apart from these modifications, there are several typographical errors and misplacing articles which may be checked at the end. 1. Please consider rewriting the first few sentences in the abstract as: "The thriving field of atomic defect engineering towards advanced electrocatalysis relies on the critical role of electric field polarization at the atomic scale.While this is proposed theoretically, the spatial configuration, orientation, and correlation with specific catalytic properties of materials are yet to be understood."Instead of the present text 2. Please rephrase the following sentence: "As such, analysing the non-periodic…."as: "Analyzing the non-periodic electric field on the electrocatalyst with high spatial accuracy is crucial for understanding the catalytic mechanism.Unfortunately, since the technical challenges of atomic imaging currently hinder the characterization of the non-periodic electric fields surrounding specific atomic defects, understanding such microscopic mechanisms largely relies on theoretical calculations14-16." 3. The sentence "Therefore, MoS2 with point defects…" may be simplified as: "Therefore, MoS2, with point defects as a catalyst, is an ideal material system to explore the effect of electric field polarization of atomic defect sites on the as-tuned catalytic property and performance."4. The sentence "In fact, the asymmetric charge density distribution…."May be rephrased as: "The asymmetric charge density distribution on the catalyst surface is directly related to its electric field, which is the decisive origin of catalytic performances6,11-16,26.Here, the DPC technology, a recent advancement in STEM imaging27-30, was selected to characterize the electric field distributions of the MoS2-based materials." 5.With the experimental results presented in this study, the authors may go as far as making a stronger conclusion by modifying the last sentence in conclusions as: "This is the first example to explore the relationship between the atomic-level polarization in electric fields of catalyst surfaces and their catalytic activity.This study not only correlates the influence of atomic-level defects on catalytic activity but also paves the way to potential real-time catalysis studies using MEMS-based devices in microscopes." Reviewer #4 (Remarks to the Author): In this study, Xu et al. have undertaken a comprehensive investigation of electric field polarization and its impact on the hydrogen evolution reaction (HER) efficiency in defective MoS2.Their work combines experimental techniques and theoretical calculations to shed light on the factors influencing catalytic activity.While the experimental section is commendable, there are some suggestions for refining the theoretical portion of the study before it can be considered for acceptance.
1. Figure 3d should include labels for the hydrogen adsorption free energies, even though this information is available in the supplementary material.This will enhance the clarity of the figure and make it easier for readers to interpret.
2. It is better to provide a quantitative explanation for the observed hydrogen adsorption energies in the context of electronic structure, possibly using theories such as band center theory.This would help readers better understand the underlying mechanisms driving the variations in adsorption energies for VMo-MoS2, S2Mo-MoS2, and MoS2.
3. To improve the clarity of the charge density diagrams, it is recommended to include a color bar that demonstrates charge accumulation and depletion.This will make it easier for readers to interpret the diagrams and understand the spatial distribution of charges.
4. In the Methods section, it is imperative to explicitly mention the use of the computational hydrogen electrode (CHE) method for calculating Gibbs free energy instead of simply using 1/2 H2 directly.Additionally, cite the relevant literature, such as J. Phys.Chem. B 2004, 108(46), 17886−17892 andJ. Electrochem. Soc. 2005, 152 (3), J23-J26.by J. K. Nørskov, to provide a reference for this approach, enhancing the transparency of the computational methodology.
5. Listing the detailed calculation parameters in the climbing image nudged elastic band (cNEB) method would be better.
6. Did the authors include the Van der Walls interactions and solvation correction in the DFT calculations?
Overall, these revisions will strengthen the theoretical aspect of the paper and improve its accessibility to the scientific community.

Response:
Thank you very much for reviewing our manuscript.We greatly appreciate your inspiring, useful, helpful, and constructive comments.Your concern regarding the damage to the monolayer MoS2 structure caused by an operating voltage of 300 kV is quite reasonable.During the experimental process, we also carefully explored the impact of different voltages on HAADF imaging of monolayer pristine MoS2.First, we used the ThermoFisher Themis Z microscope to characterize the atomic structure of monolayer MoS2 under a voltage of 60 kV, as illustrated in Figure R1.Observing the S atom on the atomic resolution HAADF image in Figure R1 is very difficult, implying that the spatial resolution of images gathered under 60 kV conditions is inadequate, leading to certain atomic defect structures (such as antisite defects) going undiscovered.Therefore, we characterize the atomic electric field distribution of atomic defect structures using DPC technology to pursue an improved resolution.
Then, we found that at 300 kV, the atomic structure of monolayer MoS2 and different defects can be clearly characterized, as shown in Figures 1e-g of the original version of the manuscript.Furthermore, we continuously acquired two HAADF images of the same local region of monolayer MoS2 under 300 kV and 15 pA conditions, as shown in Figure R2.The results showed that the atomic structure of monolayer MoS2 is relatively stable, and the electron beam causes no obvious damage when HAADF images are collected.Besides, some literature has also reported the use aberrationcorrected scanning transmission electron microscopy (AC-STEM) for atomic resolution HAADF imaging of monolayer MoS2 under 300 kV or 200 kV (Nat. Commun. 2022, 13, 3063;J. Am. Chem. Soc. 2023, 145, 11348), and the characterization results also indicate that electron beam will not cause obvious structural damage monolayer MoS2.
These results indicate that the monolayer MoS2 prepared in this work possesses relatively stable structure under electron beam, and collecting their HAADF and DPC images under 300 kV conditions should not cause any observious structural damage and thus should not affect the reported results here.
The related results of Figure R2 have been added as Supplementary Fig. 4 in Page 5 of the revised Supplementary Information (SI) file.For the revision, the following discussion is added on Pages 6 and 7 of the main text: "Firstly, there is no significant damage from the electron beam when collecting high-angle annular dark field (HAADF) images (Supplementary Fig. 4), which proves that the atomic structure of the prepared monolayer MoS2 was relatively stable".For your convenience, all changes have been highlighted by yellow in the revised manuscript and SI file.

Response:
Thank you for this helpful suggestion.As you suggested, necessary simulations are important support for the polarized electric field distribution of the antisite defect results in experiments.Firstly, we performed DPC technology with four segments detector imaging simulation of electric field distribution using Dr. Probe V1.10 software package (Ultramicroscopy 2018, 193, 1).The model is a monolayer of MoS2 with antisite defects structure with a thickness of 0.55 nm that was equidistantly divided into three layers for simulations based on multislice method.The simulations used a frozen lattice configuration of 100 variants per slice and were performed at 300 kV experimental conditions.To account for partial spatial coherence of the probes, the images were convolved with a probe size of 0.06 nm.Then, we also draw the vector maps with Avizo software, and the simulated electric field distribution of antisite defects structure are obtained in Figure R3.We also found significant asymmetric polarization in the electric field distribution of antisite defects structure from the simulation image (red arrows represent the electric field polarization region), further supporting the reliability of the DPC experimental results.
In addition, the display of the electric field distribution in Figures 4f-h  "Besides, the simulation image results also show that the electric field distribution of the antisite defect structure has obvious asymmetric polarization (Supplementary Fig. 21)", "In addition, overlay images of HAADF-STEM and corresponding DPC map in Fig.

/ 33 4c-h more clearly show the electric field distribution of individual atoms in monolayer
MoS2 (Supplementary Fig. 22)", and "DPC-STEM image simulation was performed using the Dr. Probe V1.10 software package 37 ".This could also be obtained by the DPC technique but missed in this work.

Response:
Thank you very much for your constructive comment.DPC mapping with atomic resolution reflects the direction and intensity of electric fields distribution around individual atoms.In fact, we can further process DPC mapping to obtain its charge density distribution, which is called differentiated differential phase contrast (dDPC) mapping.Notably, dDPC mapping requires mathematical differentiation of the DPC signal, which can lead to a significant weakening of the dDPC signal strength, especially on charge density distribution of individual atoms.First of all, we further obtained dDPC mapping images in Figure R5 of the corresponding regions of Figures 4b,d in the manuscript, and the results showed that the charge distribution of antisite defects structure was clearly asymmetric, which is different from the elliptical charge distribution of the surrounding Mo atom.This result is consistent with the theoretical calculation.However, we also found some abnormal distortions in the charge distribution intensity of other atoms, which is related to the surface structure in our synthesized materials.However, it is more likely that the current dDPC technology with four-segment detector does not have enough precision in characterizing the charge distribution of individual atoms.shows that the charge density distribution of antisite defects structure was asymmetric, which is also consistent with the DFT calculations".Uq (V/SHE) = -4.6V -Φq / e where -4.6 V is the absolute electrode potential of the SHE benchmarked in VASPsol, and -Φq is the work function of the charged system.Subsequently, under these conditions, we performed Electrostatic Potential (ESP) analyses on the models with two distinct defects, respectively.Multiwfn software was employed during the process of generating the plots.It is evident that, in comparison to the defect-free MoS2 at a significant separation, there is a significant variation in the potential at defect sites.This observation aligns well with the theoretical proposition that regions with lower electrostatic potential are more prone to electron donation.This correlation aligns effectively with the adsorption energies previously calculated in our study.
Red represents a low electrostatic potential, indicating that this region more easily gives electrons and is more nucleophilic than other regions, while blue represents a high electrostatic potential, indicating that this region is easier to obtain electrons and is more electrophilic than other regions.Calculations use the 0.001 electrons/bohr 3 density isosurface.as Refs.44-47.For the revision, the following discussion is added on Pages 15 and 24 of the main text: "Besides, to elucidate the impact of the antisite defects on the surface potential, this distribution of surface electrostatic potential was computed to more intuitively investigate the influence of antisite defects on the surface electric field potential under experimental conditions (Supplementary Fig. 18).In comparison to the defect-free MoS2 at a significant separation, there is a significant variation in the potential at defect sites.This observation aligns well with the theoretical proposition that regions with lower electrostatic potential are more prone to electron donation", and "An implicit solvent model was employed.The linearized Poisson−Boltzmann model with a Debye length of 3.0 Å mimics the compensating charge.The electrode potential of the two models to 0.2 V vs SHE was adjusted in accordance with experimental conditions, following the procedure outlined in Equation.The solvent environment was modeled by the VASPsol code 44-47 .Uq (V/SHE) = -4.6V -Φq / e where -4.6 V is the absolute electrode potential of the SHE benchmarked in VASPsol, -Φq is the work function of the charged system."

The related results of
For your convenience, all changes have been highlighted by yellow in the revised manuscript and SI file.Thank you for your helpful and constructive comment.In fact, the higher the operating voltage of aberration-corrected scanning transmission electron microscopy (AC-STEM), the higher the spatial resolution of the collected image in theory.During the experimental process, we also carefully explored the impact of different voltages on HAADF imaging of monolayer MoS2.Firstly, we used the AC-STEM to characterize the atomic structure of the monolayer MoS2 under a voltage of 60 kV, as illustrated in Figure R2.Observing the S atom on the atomic resolution HAADF image in Figure R2 is very difficult, implying that the spatial resolution of images gathered under 60 kV conditions is inadequate, leading to certain atomic defect structures (such as antisite defects) going undiscovered.Then, we found that under the condition of 300 kV, the atomic structure of monolayer MoS2 and different defects can be clearly characterized by aberration-corrected scanning transmission electron microscopy (AC-STEM) as shown in Figures 1e-g of the original version manuscript.Furthermore, we continuously acquired two HAADF images of the same micro region of monolayer MoS2 under 300 kV conditions in Figure R3.The results showed that the atomic structure of monolayer MoS2 is relatively stable, and the electron beam causes no obvious damage when HAADF images are collected.Besides, some literature has also reported the use AC-STEM for atomic resolution HAADF imaging of monolayer MoS2 under 300 kV or 200 kV (Nat. Commun. 2022, 13, 3063;J. Am. Chem. Soc. 2023, 145, 11348), and the characterization results also indicate that electron beam will not cause monolayer MoS2 structural damage.These results indicate that the monolayer MoS2 atomic structure we prepared is relatively stable, and collecting their HAADF images under 300 kV conditions will not damage their atomic structure.
The related results of Figure R2 have been added as Supplementary Fig. 4 in Page 5 of the revised SI file.For the revision, the following discussion is added on Pages 6 and 7 of the main text: "Firstly, there is no significant damage from the electron beam when collecting high-angle annular dark field (HAADF) images (Supplementary Fig. 4), which proves that the atomic structure of the prepared monolayer MoS2 was relatively stable".

Response:
Thank you for your helpful and constructive comment.We obtained HAADF images of monolayer MoS2 through AC-STEM at a voltage of 300 kV, and the dose of the electron beam is a crucial parameter.In order to verify the effect of electron beam dose on the HAADF imaging damage of the atomic structure of monolayer MoS2, we explored the atomic-resolution HAADF imaging of monolayer MoS2 under different dose conditions.On the same monolayer MoS2 nanosheets, we collected atomicresolution HAADF images of under different current conditions of 15 pA and 70 pA respectively, as shown in Figure R4.The results show that the atomic structure of monolayer MoS2 has been the knock-on damaged by electron beam at high dose (at 70 pA).In contrast, the atomic structure of monolayer MoS2 remains intact in low dose mode (at 15 pA).Therefore, all HAADF and DPC data of our prepared monolayer MoS2 were collected at a dose of 15 pA, and the optimized electron beam dose did not cause any new defects.
For the revision, the following discussion is added on Page 20 of the main text: "The AC-STEM characterization was performed using a ThermoFisher Themis Z microscope equipped with two aberration correctors under 300 kV and 15 pA".Thank you very much for your correct and thoughtful question.Firstly, the surface potentials/states of defects in MoS2 can indeed be interfered with by various external factors, leading to inaccuracies in various optical, electrical, magnetic as well as electrocatalytic performance.These external influences come from the adsorption of active gases, small molecules and small impurities in the air, as well as residual polymers from photoresist et al.However, unlike three-electrode electrochemical measurements, the all 2D MoS2 samples for STEM and DPC tests in this paper are directly transferred to the copper microgrid, which will avoid the defect structure damage caused by 2D surface stress or electron beam in device preparation.
Therefore, such antisite defects can only come from the defects constructed by our annealing process.
Meanwhile, in this paper, the construction of micro-electrocatalytic devices will inevitably produce tiny residual polymer pollutants.Moreover, these polymer residues can change the surface potential of MoS2, thus affecting the accurate evaluation of catalytic properties.However, we are based on the same fine process and device preparation technology for testing the HER performance of all 2D MoS2 samples, which eliminates the interference of external factors caused by device preparation.It should also be noted here that the micro-electrochemical device process we prepared is carried out based on predecessors (Nat.Commun. 13, 3063 (2022); Nat.Catal. 5, 212-221 (2022); Nat.Mater. 18, 1098Mater. 18, -1104Mater. 18, (2019))).We explored a strict process to ensure that the 2D sample will not be damaged during the lithography process, as well as obvious surface pollution.However, our device manufacturing process or the creation of electrodes first on the substrate and transferring the MoS2 later (Nano letters 10.5 (2010): 1912-1916) both require electron beam to exposure an electrochemical window to be tested in order to contact the electrolyte to form a three-electrode test system.The defect damage caused by residual glue and stress from these two technologies cannot be effectively avoided, and causes some changes in the surface potential of monolayer MoS2.
Compared with pristine MoS2, the improvement of HER performance of the MoS2 with antisite defects can only come from the antisite defect active sites constructed by our annealing process.Moreover, all the 2D samples in this paper are under the same device preparation process.Through the horizontal comparison of HER performance, we can see that such HER performance changes caused by device preparation will occur in all the 2D samples, and will cancel when compared to each other.The difference in HER catalytic performance with or without defects can only come from the surface potential difference rooted in the antisite defect as well as the performance improvement derived from the electric field polarization described in this paper.
In order to further in-situ evaluate the HER performance changes of the antisite defect structure, and confirm that the improved HER performance only comes from artificially created defect structures, we will use the in-situ liquid phase electron microscopy technology in the future work to evaluate the difference in surface structure and electric field polarization of this defect in the process of hydrogen evolution under real conditions.Thus, the intrinsic reason of the HER performance difference caused by the antisite defect structure can be directly observed and verified.It should also be noted here that the micro-electrochemical device process we 18 / 33 prepared is carried out based on predecessors (Nat. Commun. 13, 3063 (2022); Nat.Catal. 5, 212 (2022);Nat. Mater. 18, 1098(2019)).In addition, the damage caused by residual glue and stress in the process of device fabrication cannot be effectively avoided, and causes some changes in the surface potential of MoS2.However, the photoresist contamination of device fabrication has little effect on the performance of HER tested by micro-electrochemical device.The obvious changes in HER performance (Tafel slopes) are mainly caused by antisite defects in during the synthesis of monolayer S2Mo-MoS2.
The related results of Figure R5 has been added as Supplementary Fig. 13 in Page 14 of the revised SI file.For the revision, the following discussion is added on Page 11 of the main text: "Furthermore, multiple S2Mo-MoS2-5 samples were conducted to electrocatalytic HER tests (Supplementary Fig. 13).The results show that the HER performance remained relatively stable, which illustrates the rough homogeneity of the number of antisite defects in S2Mo-MoS2-5 samples".We appreciate the suggestion from the reviewer.We have modified the sentences in Abstract of this article.For the revision, the following sentences is added on Page 2 in the Abstract: "The thriving field of atomic defect engineering towards advanced electrocatalysis relies on the critical role of electric field polarization at the atomic scale.While this is proposed theoretically, the spatial configuration, orientation, and correlation with specific catalytic properties of materials are yet to be understood."2. Please rephrase the following sentence: "As such, analysing the non-periodic…."as: "Analyzing the non-periodic electric field on the electrocatalyst with high spatial accuracy is crucial for understanding the catalytic mechanism.Unfortunately, since the technical challenges of atomic imaging currently hinder the characterization of the non-periodic electric fields surrounding specific atomic defects, understanding such microscopic mechanisms largely relies on theoretical calculations 14-16 ."

Response:
We appreciate the suggestion from the reviewer.We also have rephrased the sentences in this article.For the revision, the following sentences is added on Page 3 of the main text: "Analyzing the non-periodic electric field on the electrocatalyst with high spatial accuracy is crucial for understanding the catalytic mechanism.
Unfortunately, since the technical challenges of atomic imaging currently hinder the characterization of the non-periodic electric fields surrounding specific atomic defects, understanding such microscopic mechanisms largely relies on theoretical calculations 14-16 ." 3. The sentence "Therefore, MoS2 with point defects…" may be simplified as: "Therefore, MoS2, with point defects as a catalyst, is an ideal material system to explore the effect of electric field polarization of atomic defect sites on the as-tuned catalytic property and performance."

Response:
We appreciate the suggestion from the reviewer.We have modified the sentences in this article.For the revision, the following sentences is added on Page 4 of the main text: "Therefore, MoS2, with point defects as a catalyst, is an ideal material system to explore the effect of electric field polarization of atomic defect sites on the as-tuned catalytic property and performance."We appreciate the suggestion from the reviewer.We have modified the sentences in this article.For the revision, the following sentences is added on Page 15 of the main text: "The asymmetric charge density distribution on the catalyst surface is directly related to its electric field, which is the decisive origin of catalytic performances 6,11-16,26 .Here, the DPC technology, a recent advancement in STEM imaging 27-30 , was selected to characterize the electric field distributions of the MoS2based materials."We appreciate the suggestion from the reviewer.We have modified the sentences in Conclusions of this article.For the revision, the following sentences is added on Page 19 of the main text: "This is the first example to explore the relationship between the atomic-level polarization in electric fields of catalyst surfaces and their catalytic activity.This study not only correlates the influence of atomic-level defects on catalytic activity but also paves the way to potential real-time catalysis studies using micro-electro mechanical systems (MEMS)-based devices in microscopes."

Response:
Thank you for this constructive suggestion.According to the Reviewer's suggestions, we also tried to use band center theory to explain the variations in observed hydrogen adsorption energies for VMo-MoS2, S2Mo-MoS2, and MoS2.Unlike the d-band center theory for metal atoms, since the active site of HER is a non-metallic S atom with no d-band, we mainly calculate the p-band center of the active S atom here.According to the previous works (J. Mater. Chem. A, 2020, 8, 5688;Carbon 2018, 133, 260), the p-band center of active S sites can be calculated using:  Overall, these revisions will strengthen the theoretical aspect of the paper and improve its accessibility to the scientific community.
With AC-STEM imaging and DPC-STEM technique, they have firstly determined the polarized electric field of antisite defect for facilitating the catalytic activity in this reaction.I recommend to publish this work in Nature Communications, however, the observation of DPC-STEM here need be addressed and improved with following concerns.1.The voltage for the AC-STEM measurement in this work is 300 kV, which differs to the typical voltage of 60/80 kV used for the monolayer MoS2 in most of studies as the high voltage in STEM could easily cause the radiation (knock-on) damage for this material.Therefore, the authors should explain if the original state of investigate materials is reversed in the observation of STEM and DPC under such condition or not.

Figure R1 .
Figure R1.(a,b) HAADF image and corresponding Wiener filter image of monolayer pristine MoS2 under a voltage of 60 kV.

Figure R2 .
Figure R2.(a,b) Two HAADF images of monolayer pristine MoS2 in the same micro region were collected continuously under 300 kV and 15 pA conditions.The results showed that the atomic structure of monolayer MoS2 is not significantly changed under electron beam radiation when HAADF images are collected.
of the original manuscript was not clear enough, so we redrawn the electric field distribution in Figures 4f-h to more clearly determine the differences in electric field distribution among different atomic structures.Moreover, we further overlay the electric field distribution on Figures 4f-h and HAADF images on Figures 4c-e, as shown in Figure R4, which can better display the atomic polarization electric field distribution at the antisite defect structure.The related results of Figures R2 and R3 have been added as Supplementary Figs.21, and 22 in Pages 22 and 23 of the revised SI file.On page 30 of the revised manuscript file, (Ultramicroscopy 2018, 193, 1) has been cited as Ref. 37. For the revision, the following discussion is added on Pages 18, and 21 of the main text:

Figure R3 .
Figure R3.(a,b) Simulated HAADF image and corresponding DPC image of monolayer MoS2 with the antisite defect structure.The red arrows in (b) indicated the electric field polarization region.

Figure R4 .
Figure R4.(a-c) The overlay images of HAADF-STEM and corresponding DPC map of pristine atomic structure (a), the antisite defect structure (b) and Mo vacancy structure (c) in monolayer MoS2.
Figure R6a, it can be clearly seen that a single Pt atom occupies the Mo site.In theory, the charge density distribution of a heavy metal Pt single atom should be significantly different from that of the surrounding Mo atoms.However, from the atomicresolution dDPC image obtained in the corresponding region of Figure R5a in Figure R6b, it is difficult to see an obvious difference in the charge density distribution of Pt single atoms compared to nearby Mo atoms.These results indicated that the accuracy of dDPC technology with four-segments detector may make it difficult to distinguish the difference i n the charge density distribution of individual atoms.The related results of Figure R5c,d have been added as Supplementary Fig. 23 in Page 24 of the revised SI file.For the revision, the following discussion is added on Page 18 of the main text: "Further, we obtain differentiated differential phase contrast (dDPC) mapping image of the antisite defect (Supplementary Fig. 23), and the result

Figure R5 .
Figure R5.(a,b) HAADF-STEM image and corresponding dDPC image of S2Mo-MoS2-5; (c,d) Enlarged regions of the positions indicated by the dashed lines in (a) and (b), the atomic structure of antisite defects and corresponding charge density distribution mapping.

Figure R6 .
Figure R6.(a,b) HAADF-STEM image and corresponding dDPC image of Pt-MoS2.The red dashed circles in (a) represent the single Pt atom, and black dashed circles in (b) represent the charge density distribution of the single Pt atom.

Figure R1 .
Figure R1.(a,b) Mapping Electrostatic Potential on electron density isosurface of two kinds of antisite defects.

Figure R2 .
Figure R2.(a,b) HAADF image and corresponding Wiener filter image of monolayer pristine MoS2 under a voltage of 60 kV.

Figure R3 .
Figure R3.(a,b) Two HAADF images of monolayer MoS2 in the same micro region were collected continuously under 300 kV conditions.The results showed that the atomic structure of monolayer MoS2 is relatively stable under electron beam radiation when HAADF images are collected.
reproducible are the electrochemical measurements?Are the Tafel slopes indicating a direct correlation with the defects based on the synthesis procedure, or is there an influence of the device fabrication method of spin-coating and lift-We appreciate the suggestion from the reviewer.Fisrt of all, we conducted several electrocatalytic HER tests on 2D S2Mo-MoS2-5 samples.As shown in Figure R5.The five S2Mo-MoS2-5 samples with antisite defects basically maintained relatively stable in the HER properties, especially through the comparison of polarization curves and Tafel slopes.The results show that the current density and the Tafel slope are basically the same, indicating that the homogeneity of all samples is good, and our electrochemical performance test is repeatable.Meanwhile, in order to compare the microscopic changes in the surface structure of 2D MoS2 before and after the micro-electrochemical device fabrication (such as spin-coating and lift-off), Raman and PL spectra were used to characterize five different monolayer 2D MoS2 samples with antisite defect.As shown in Figure R6, the Raman and PL spectra of the five samples basically remained unchanged, indicating no obvious additional pollution and no significant new structural defects during the device fabrication process.

Figure R5 .
Figure R5.(a) Polarization curves and (b) Tafel plots of five 2D S2Mo-MoS2-5 samples.The results show that the current density and Tafel slope are basically the same.

Figure R6 .
Figure R6.(a) Raman and (b) PL spectra of the five monolayer 2D S2Mo-MoS2 samples before and after device preparation.

4.
The sentence "In fact, the asymmetric charge density distribution…."May be rephrased as: "The asymmetric charge density distribution on the catalyst surface is directly related to its electric field, which is the decisive origin of catalytic performances 6,11-16,26 .Here, the DPC technology, a recent advancement in STEM imaging 27-30 , was selected to characterize the electric field distributions of the MoS2based materials."Response:

5.
With the experimental results presented in this study, the authors may go as far as making a stronger conclusion by modifying the last sentence in conclusions as: "This is the first example to explore the relationship between the atomic-level polarization in electric fields of catalyst surfaces and their catalytic activity.This study not only correlates the influence of atomic-level defects on catalytic activity but also paves the way to potential real-time catalysis studies using MEMS
Figure R8 (a) Atomic structures of reaction intermediates OOH* in oxygen evolution/reduction reactions.(b) Atomic structures of adsorbed H* on pristine MoS2.