Photoelectrochemical Performance of a CuBi₂O₄ Photocathode with H₂O₂ as a Scavenger

Round 1

Reviewer 1 Report

The following issues must be addressed:

1.       Introduction part should contain examples about other materials used as photocathode such as: GaN (DOI: 10.1016/j.ijleo.2017.06.076), WO3 (DOI: 10.1016/j.jeurceramsoc.2005.07.048), NiO (DOI: 10.1016/j.electacta.2012.01.081);

2.       Provide the purity of each substance;

3.       The authors discuss about film thickness but it seems that the layer is not homogenous; in which point/area have you measured the film thickness?

4.       Please discuss in more details the photocurrent stability.

5.       Provide and discuss the equivalent circuit.

Author Response

Response to reviewer’s comments

Thank you for your valuable comments on our manuscripts. All questions have been answered in the relevant section of the text and the changed parts are shown in yellow, and green color for the first, and second revisions in the manuscript.

Reviewer #1:

The following issues must be addressed:

 Response to Reviewer #1:

1. Introduction part should contain examples about other materials used as photocathode such as: GaN (DOI: 10.1016/j.ijleo.2017.06.076), WO3 (DOI: 10.1016/j.jeurceramsoc.2005.07.048), NiO (DOI: 10.1016/j.electacta.2012.01.081);

Thank you for your valuable comment to improve our manuscript quality. The correction has been done accordingly. In the introduction part, we added contain photocathodes such as Cu₂O, GaN, NiO, CaFe₂O_4, CuNb₃O₈, CuFeO₂, LaFeO₃, and CuBi₂O₄ photocathode materials. Furthermore, we have cited the relevant works in our revised manuscript.

2. Provide the purity of each substance;

The correction has been done accordingly. In the revised manuscript we added the materials part in the supporting information to show the purity of each substance.

3. The authors discuss about film thickness but it seems that the layer is not homogenous; in which point/area have you measured the film thickness?

We measured SEM cross section (film thickness) of CuBi2O4 photocathode at below point/area: SEM HV:10.0 KV, WD:16.59 mm, SEM MAG:50.0 kx

The quality of our SEM equipment for high resolution have a problem, that’s why maybe the quality of SEM cross section not good enough. Furthermore, our film thickness which measured with SEM cross section, are with good agreement with some other articles.

Please see Figure S1 in (https://pubs.rsc.org/en/content/articlelanding/2019/ta/c9ta07892d)

Please see Figure 4 in (https://opg.optica.org/oe/fulltext.cfm?uri=oe-27-4-A171&id=404773)

Below figure shows SEM cross-section of the #1-CuBi₂O₄ photocathode at point/area SEM MAG:100.0 kx.

4. Please discuss in more details the photocurrent stability.

Thank you for your valuable comment. The correction has been done accordingly. In the revised manuscript we discussed more detail the photocurrent stability (please see the revised manuscript).

5. Provide and discuss the equivalent circuit.

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly. In the revised version of the manuscript, we added and discussed the equivalent circuit model (please see the inset Figure 5a).

Author Response File: Author Response.docx

Reviewer 2 Report

The paper reports a photoelectrochemical and spectroscopic characterization of CuBi2O4 perovskytes produced by a backing thermal oxidation of precursor solutions using a dropping method. Films of different thickness are produced by succesively dropping-heating steps to find the most optimal one. The paper is well structured, but it needs an intensive revision concerning the style and the grammatic of the text. But more essentially, there are some interpretations made by the authors, which should be revised as marked in the attached copy. 

Comments for author File: Comments.pdf

Author Response

Reviewer #2:  

The paper reports a photoelectrochemical and spectroscopic characterization of CuBi2O4 perovskytes produced by a backing thermal oxidation of precursor solutions using a dropping method. Films of different thickness are produced by succesively dropping-heating steps to find the most optimal one. The paper is well structured, but it needs an intensive revision concerning the style and the grammatic of the text. But more essentially, there are some interpretations made by the authors, which should be revised as marked in the attached copy. 

Response to Reviewer #2:

1. The reader cannot understand why the larger exposition area to the electrolyte should reduce the efficiency of the cell. Perhaps by the formation of surface states promoting recombination? If this, please explain. In this case why do you expect to have more efficient films with a larger thickness?

We rewrote that part in the revised manuscript. Furthermore, the effect of the thickness on the PEC performance for the #1-CuBi₂O₄, #2-CuBi₂O₄, and #4-CuBi₂O₄ photocathodes was measured and discussed in the revised manuscript (Figure S5).

The different drop-casting processes for 1, 2, and 4 times (layer) led to the formation of a thin layer of CuBi₂O₄ on the substrate with thicknesses of ~ approximately 450, 650, and 1000 nm, respectively. Increasing the thickness from 450 nm to 650 nm led to an increase in the photocurrent density. However, the photocurrent was decreased when a 4 times drop-casting process, producing a thickness of 1000 nm. Therefore, the best photocurrent density was obtained through 2 times drop-casting process (~650 nm) because the electron–hole pairs generated in the bulk of the films during the PEC water splitting recombine before reaching the surface.

2. please, explain here this notation, otherwise the reader is confused and it seems you are writing now about n-semiconductors

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly.

3. "used" is more appropriate

The correction has been done accordingly.

4. what side? both, from the front, from the back? please clear.

From the front side of the photocathode, please see section 2.2.

The PEC performance was assessed in an electrolyte solution of 0.5 M Na₂SO₄ under 100 mW/cm² irradiation (AM 1.5) from the front of the photocathode using a 300 W Xe lamp.

5. This part of the text is superflous because it repeats what you explained before

The correction has been done accordingly. With respect to the reviewer’s comment, we removed that part from the revised manuscript.

6. The low frequency limit is not enough to recognize the type of low frequency response. Thus, the calculation of the charge transfer from lesser than 1/4 of the total spectrum is high speculative. In addition, what you observe at the end of the capacitive loop, if any, is the polarisation resistance which is not necessary related to charge transfer rate in the complex intefacial behavior of an illuminated semiconductor-liquid interface.

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly. In the revised version of the manuscript, we added and discussed the equivalent circuit model (please see the insert Figure 5a).

7. what is the frequency for performing these experiments?

In the revised version of the manuscript, we added the Mott−Schottky measurement condition (please see section 2.2. Photoelectrochemical (PEC) processes measurement). The Mott−Schottky measurement was performed in the potential range of 0.7−1.6 V vs RHE, with frequencies 1000 Hz

8. This statement is problematic, because it is unprobable that the addition of a scavenger influences the photo-induced cathodic reaction, the water reduction. In my opinion, the function of the scavenger is to consume the surface holes, which otherwise recombine with the photoelectrons at the surface states. This is an very important issue that is not mentioned in the interpretation of the results and which cannot be ignored regarding the very defective surface which is expeted from the surface morphology. This also, can be the reason for the change of the slope in the Mott-Schottky plots after adding H2O2, because you have lower charging of the surface states and thus a change of the doble layer potential, which in turn, can be the origin of the change of ocp.

Concerning the reviewer’s comment, we rewrote that part in the revised manuscript, furthermore added more discussion about adding H₂O₂ as an electron scavenger. H₂O₂ was added to the electrolyte as an electron scavenger to test the CuBi₂O₄ photocathodes without limitations in the reaction kinetics, which would be the case for proton reduction. An electron scavenger will considerably improve the reaction kinetics by increasing the charge transportation rate. As an effective electron scavenger, H₂O₂ is expected to eliminate surface recombination and prevent limitations in reaction kinetics at the semiconductor–liquid interface. The electron scavenger (H₂O₂) leads to rapid electron transfer from the CuBi₂O₄ photocathodes to H₂O₂, enhancing the PEC performance.

DOI: 10.1021/jacs.7b07847

doi.org/10.1039/C9TA01489F

doi.org/10.1021/acsenergylett.1c02130

9. The scavenger cannot increase the number of photogenerated electrons! because they are generated by the photon absorption at the semiconductor surface!

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly. Concerning the reviewer’s comment, we rewrote that part in the revised manuscript furthermore added more discussion about adding H₂O₂ as an electron scavenger.

10. here, you come to consider recombination! that is right but you should reformulate your interpretation in the frame of it, in my opinion.

The correction has been done accordingly.

11. The figure suggests that you are illuminating from the back, it is right? please make it clear, because it is important to discuss the efficiency. From the front or from the back: conditions are absolutely different!

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly. We are sorry for the misunderstanding; we illuminated the photocathode from the front sides. Based on your comment, in the revised version of the manuscript, we drew again Figure 7 (the mechanism and band alignment of CuBi₂O₄ photocathodes).

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript investigated how drop casting time affect the photoelectrochemical performance of CuBi2O4 photocathode. Three CuBi2O4 thin film samples with different drop casting time were prepared, among which the twice drop casting one performs the best. After carefully reading this manuscript, I recommend a major revision since there are some flaws and issues that should be addressed by the authors.

i.               The title is too broad. If this is the pioneer work of CuBi2O4 photocathode for PEC water splitting, it may be appropriate. However, the fact is there are a bunch of papers studying the same material. So the title should give more information to readers.

ii.              Figure 3d: the tangent in Tauc plot for determining the bandgap value is not reliable for two reasons. First, there is actually higher linearity part in the curve. Second, it does not make any sense to extend the tangent to axis for the intercept because the baseline is above the axis. The right way should be extending the pre-edge baseline and drawing a reliable tangent, then locating the cross of those two lines.

iii.             Figure 5b: The problem also lies in the linear fitted line. Here isn’t the flatband value determined by the intercept of x-axis? So the actual flatband values should be larger than the value reported by the authors.

iv.             Figure 4: which sample did you use for the dark experiment? Please state explicitly in either main text or figure caption.

v.              The acceptor densities of all samples showed in SI are at least one order of magnitude higher than the reported values in literatures. How to explain this dramatic increase?

vi.             Why adding electron scavenger can lead to increase in acceptor density, as said in Section 3.4?

vii.           There are format flaws in most figures, such as multiple fonts in a same figure. For example:

Figure 1: “Drop casting” crosses the arrow;

Figure 2: every scale bar uses a unique font that is different with the others;

Figure 3: fonts of axis labels and axis ticks are different;

Figure 4: all y-axis labels are actually same, but different unit formats are used;

Figure S1: y-axis label: capitalize the initial letter.

Author Response

Reviewer #3:  

This manuscript investigated how drop casting time affect the photoelectrochemical performance of CuBi2O4 photocathode. Three CuBi2O4 thin film samples with different drop casting time were prepared, among which the twice drop casting one performs the best. After carefully reading this manuscript, I recommend a major revision since there are some flaws and issues that should be addressed by the authors.

Response to Reviewer #3:

1. The title is too broad. If this is the pioneer work of CuBi2O4 photocathode for PEC water splitting, it may be appropriate. However, the fact is there are a bunch of papers studying the same material. So the title should give more information to readers.

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly. In the revised version of the manuscript, we changed the title.

1. Figure 3d: the tangent in Tauc plot for determining the bandgap value is not reliable for two reasons. First, there is actually higher linearity part in the curve. Second, it does not make any sense to extend the tangent to axis for the intercept because the baseline is above the axis. The right way should be extending the pre-edge baseline and drawing a reliable tangent, then locating the cross of those two lines.

Thank you for your valuable comment. The correction has been done accordingly. In the revised version of the manuscript, we modified the tangent in the Tauc plot for determining the bandgap value to be more reliable.

iii.             Figure 5b: The problem also lies in the linear fitted line. Here isn’t the flatband value determined by the intercept of x-axis? So the actual flatband values should be larger than the value reported by the authors.

The correction has been done accordingly. We are sorry for the misunderstanding, which happened due to typo errors. In the revised version of the manuscript, we corrected the flat band values (please see Table S1).

1. Figure 4: which sample did you use for the dark experiment? Please state explicitly in either main text or figure caption.

The correction has been done accordingly. All the samples displayed negligible dark current density (Figures 4a and b). We state in the main text in the dark condition, the current density of all the samples is around zero.

1. The acceptor densities of all samples showed in SI are at least one order of magnitude higher than the reported values in literatures. How to explain this dramatic increase?

Thank you for your valuable comment. Based on your nice comment we remeasured the Mott-Schottky plot to calculate acceptor density (N_A) and flat band potentials. Please see Table S1, Figure 5b, and Figure 6c.

3. Why adding electron scavenger can lead to increase in acceptor density, as said in Section 3.4?

Concerning the reviewer’s comment, we rewrote that part in the revised manuscript furthermore added more discussion about adding H₂O₂ as an electron scavenger. H₂O₂ was added to the electrolyte as an electron scavenger to test the CuBi₂O₄ photocathodes without limitations in the reaction kinetics, which would be the case for proton reduction. An electron scavenger will considerably improve the reaction kinetics by increasing the charge transportation rate. As an effective electron scavenger, H₂O₂ is expected to eliminate surface recombination and prevent limitations in reaction kinetics at the semiconductor–liquid interface. The electron scavenger (H₂O₂) leads to rapid electron transfer from the CuBi₂O₄ photocathodes to H₂O₂, enhancing the PEC performance

DOI: 10.1021/jacs.7b07847

doi.org/10.1039/C9TA01489F

doi.org/10.1021/acsenergylett.1c02130

vii.           There are format flaws in most figures, such as multiple fonts in a same figure. For example:

Figure 1: “Drop casting” crosses the arrow;

Figure 2: every scale bar uses a unique font that is different with the others;

Figure 3: fonts of axis labels and axis ticks are different;

Figure 4: all y-axis labels are actually same, but different unit formats are used;

Figure S1: y-axis label: capitalize the initial letter.

Thank you for your constructive comment to improve our manuscript quality. The correction has been done accordingly.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The manuscript can be published in present form.

Reviewer 3 Report

Thanks for addressing my concerns.