Carbon Monoxide Tolerant Pt-Based Electrocatalysts for H₂-PEMFC Applications: Current Progress and Challenges

Round 1

Reviewer 1 Report

Dear Authors,

The submitted manuscript "Carbon monoxide tolerant electrocatalysts for H2-PEMFCs application: Current progress and challenges" is well-written, well-structured, and easy to understand.

The paper can be accepted for publication in the journal of Catalysts without any further changes.

Regards

Author Response

We thank you for evaluating positively the quality of our manuscript.

Reviewer 2 Report

This work by Prof. Tsiakaras and Dr. Molochas proposes an overview on the current strategies for the design of Pt-based electrocatalysts with enhanced CO tolerance for application in hydrogen-fed proton exchange membrane fuel cells. The review addresses an interesting research topic in the current scenario, where a transition towards a more sustainable development is becoming imperative. The authors provide an original, keen and highly interesting overview, the structure of the review is well-organized, the language is clear, the cited references are pertinent and up-to-date. In my opinion, the review deserves to be published on Catalysts, after a careful revision according to the following comments.

 

• Title: The review focuses essentially on Pt-based electrocatalysts. Although it is well-known that Pt-based electrocatalysts are the most commonly used systems in PEMFCs, I suggest to add this information in the title. The following version might be a possible alternative: Carbon monoxide tolerant Pt-based electrocatalysts for H2-PEMFCs application: Current progress and challenges
• Par. 2.2 (p. 5, lines 178 et seq.) As cited by the authors, the nature of the interaction between CO and Pt surfaces has been well-described by Blyholder. In my opinion this is a fundamental aspect that needs to be emphasized and well-clarified in order to be fully understood by all the readers. I suggest to add and thoroughly comment a scheme which could depict the electronic interactions between Pt and CO at atomic level, as proposed by the Blyholder model and following integrations (e.g. Nilsson and Pettersson model).
• Par. 2.3 (p.6, lines 216 et seq.) Authors introduce chemisorption energies as energy descriptors. It is not clear if these values are to be considered as Gibbs free energies. In particular, it is important to clarify this point to better understand the approach of Back et al. described in the same paragraph (see comment below).
• Par. 2.3 (p.6, lines 236-239). The DFT-assisted approach by Back et al. is a valuable tool to predict CO tolerant electrocatalysts. In my opinion, this is a delicate point that might require a more detailed discussion. In particular the passage “they converted the adsorption energies to free energies and used pH-potential (Pourbaix) diagrams to predict the stability” should be discussed in more depth.
• Par. 3.1.4 (p.15, lines 602-605). Is it possible to find in the literature a valid explanation for the structure sensitivity of hydroxyl species formation?
• A general comment on the importance of CO concentration levels. It could be useful to indicate the typical CO concentration intervals in the reformate feed. In addition when reporting studies from the literature it could be important to indicate the CO concentration used in those investigations. This information in some cases is reported in the main text, but in many cases is just reported in the figure captions. I suggest to always report these data in the discussion.
• In general, each abbreviation should be shown at the first use of the expression. So, please, can the authors report the spelling for: HOR and ORR (p.2, line 59), RDE (p. 19, line 773), LSV (p. 19, line 786)

Minor points:

1. p.1, line 54. I am not familiar with the expression “in operando oxygen bleeding”. Can the authors provide more information about it?
2. p.5, line 234. It is not clear to me the kind of support DFT can provide to understand CO tolerance in mechanical terms. Did authors refer to mechanical stability of the electrocatalysts? If yes, can authors explain in which way DFT can be used to predict mechanical stability?
3. p.7, line 279. Please replace “valance” with more correct “valence”.
4. p.17, line 693. Did the expression “mesoporous size” refer to mesoporous volume or mesoporous radius? Can the authors be more precise about this?
5. p.19, line 783. The expression “TiMoO2-C composite” might be more appropriate than the used “Pt on Ti-O2 composite mixed with Mo”.
6. p.21, lines 844-848. I suggest to specify in the caption of Figure 11 that the numeric values associated with symbols refer to the corresponding entries of Table 1.
7. p.22, line 864. Please, use the more appropriate “adsorbed” rather than “absorbed”.

Author Response

2^nd Reviewer:

This work by Prof. Tsiakaras and Dr. Molochas proposes an overview on the current strategies for the design of Pt-based electrocatalysts with enhanced CO tolerance for application in hydrogen-fed proton exchange membrane fuel cells. The review addresses an interesting research topic in the current scenario, where a transition towards a more sustainable development is becoming imperative. The authors provide an original, keen and highly interesting overview, the structure of the review is well-organized, the language is clear, the cited references are pertinent and up-to-date. In my opinion, the review deserves to be published on Catalysts, after a careful revision according to the following comments.

• Title: The review focuses essentially on Pt-based electrocatalysts. Although it is well-known that Pt-based electrocatalysts are the most commonly used systems in PEMFCs, I suggest to add this information in the title. The following version might be a possible alternative: Carbon monoxide tolerant Pt-based electrocatalysts for H2-PEMFCs application: Current progress and challenges

We agree with your suggestion. The title has been changed to ΄΄Carbon monoxide tolerant Pt-based electrocatalysts for H2-PEMFCs application: Current progress and challenges΄΄.

• 2.2 (p. 5, lines 178 et seq.) As cited by the authors, the nature of the interaction between CO and Pt surfaces has been well-described by Blyholder. In my opinion this is a fundamental aspect that needs to be emphasized and well-clarified in order to be fully understood by all the readers. I suggest to add and thoroughly comment a scheme which could depict the electronic interactions between Pt and CO at atomic level, as proposed by the Blyholder model and following integrations (e.g. Nilsson and Patterson model).

Indeed, we briefly discussed the electronic interactions between Pt and CO surfaces, as we tried to maintain the text range relatively small. However, we agree that this is a fundamental aspect requiring in-depth analysis. Therefore, we have presented the interactions between the density of states (DOS) of CO and Pt in detail, highlighting the interaction that determines the Pt-CO biding strength, i.e., the coupling between the Pt d-state and the 2π* CO antibonding state. Through this analysis, we pointed out the dependence of the metals’ d-band energy on the CO adsorption strength. Finally, based on this, we introduced the role of the electronic effect on the CO tolerance enhancement. Moreover, following your suggestion, we have added a scheme (Figure 2a in the revised manuscript) interpreting the aforementioned analysis for a better understanding of the readers. Generally, several variations of the Blyholder model have been published in the literature that could have been discussed. However, we desire to stick to the aim of chapter 2.2, which is the understanding of the CO tolerance mechanisms, so that readers can evaluate the discussion of the results of recent works presented in the overview chapter 3.

 

Figure: (a) The Pt and CO projected density of states (DOS) and the corresponding changes in the CO DOS at each coupling with metal s and d-states, and (i, ii): The electronic interactions between Pt and CO at each coupling step; (b) Qualitative interpretation of the electronic effect in terms of Pt d-state energy in a Pt-M CO tolerant electrocatalyst.

• 2.3 (p.6, lines 216 et seq.) Authors introduce chemisorption energies as energy descriptors. It is not clear if these values are to be considered as Gibbs free energies. In particular, it is important to clarify this point to better understand the approach of Back et al. described in the same paragraph (see comment below).

The reviewer is right; we did not clearly define the parameter. First of all, we decided to change the specific term ΄΄chemisorption energies΄΄ to the more general term ΄΄adsorption energies΄΄, which is used in the rest of the manuscript and more frequently in the literature. In this way, we could avoid creating possible confusion for readers. After all, in several parts of the text is pointed out that the type of adsorption we are dealing with is chemisorption. To clarify the commented issue, we highlighted that as an energetic descriptor can be used either the adsorption energy or the Gibbs free energy of adsorption. The validity of the description is maintained in every case, as the two parameters are related between them with a fundamental thermodynamic equation, presented and discussed in detail in the revised manuscript.  

• 2.3 (p.6, lines 236-239). The DFT-assisted approach by Back et al. is a valuable tool to predict CO tolerant electrocatalysts. In my opinion, this is a delicate point that might require a more detailed discussion. In particular the passage “they converted the adsorption energies to free energies and used pH-potential (Pourbaix) diagrams to predict the stability” should be discussed in more depth.

We have discussed in detail the experimental procedure and theoretical background that Back et al. based on to design their DFT calculations model. Specifically, we explained how and why the researchers converted the adsorption energies to Gibbs free energies of adsorption for their conclusions. Additionally, we explained how the pH-potential (Pourbaix) diagram is exploited for the extraction of the electrochemical stability of materials and how it is correlated to thermodynamics, resulting in simulations based on the calculation of ΔG_Pourbaix.

• 3.1.4 (p.15, lines 602-605). Is it possible to find in the literature a valid explanation for the structure sensitivity of hydroxyl species formation?

First of all, in the relevant sentence, we made a mistake; the CO oxidation activity is highest for Pt(110) facets and not for Pt(100), as we mentioned. The mistake has been corrected. Generally, many works have experimentally (either with electrochemical measurements or DFT calculations) proved the structure-sensitive nature of CO oxidation, and correspondingly the structure-sensitive hydroxyl formation. However, we could not find a specific explanation identified in at least two works concerning the insights of structure-sensitive formation of hydroxyl species. The simplest, and thus probably the most valid explanation is given by Koper in his review ΄΄Structure sensitivity and nanoscale effects in electrocatalysis΄΄ (DOI: 10.1039/c0nr00857e). Briefly, the explanation of Koper is based on the acknowledgment that facets present different surface energies. Therefore, their thermodynamic stability differs too, and correspondingly, their catalytic activity. By the way, based on the Langmuir-Hinshelwood mechanism of CO oxidation reaction, we thoroughly explained how the preferential OH formation is correlated with more active CO oxidation.      

• A general comment on the importance of CO concentration levels. It could be useful to indicate the typical CO concentration intervals in the reformate feed. In addition, when reporting studies from the literature it could be important to indicate the CO concentration used in those investigations. This information in some cases is reported in the main text, but in many cases is just reported in the figure captions. I suggest to always report these data in the discussion.

According to your suggestion, in the introduction section we have referred to the typical CO concentration intervals in the reformate feed. Additionally, to attach a practical aspect to this information, we have referred to the acknowledged maximum CO poisoning limits for the conventional Pt and PtRu anodes. Furthermore, according to your instruction, we have reported in every discussed work the CO concentrations used in each investigation. It is noted that in the discussed work of Brkovic (Figure 10a) this information was not available.   

• In general, each abbreviation should be shown at the first use of the expression. So, please, can the authors report the spelling for: HOR and ORR (p.2, line 59), RDE (p. 19, line 773), LSV (p. 19, line 786)

We have reported the spelling of the abbreviations in the text, according to your instruction.

Minor points:

54. 1, line 54. I am not familiar with the expression “in operando oxygen bleeding”. Can the authors provide more information about it?

First of all, we misused the word ΄΄in΄΄ in the phrase. The error has been corrected. An extended explanation has been added to the manuscript about the operando oxygen bleeding. Briefly, in this CO poisoning mitigation method, trace concentrations of oxygen are supplied in the anode during the operation of PEMFCs. The presence of oxygen in the anode promotes the oxidation of carbon monoxide carried in the hydrogen fuel (CO + O2 ⇌ CO2), mitigating the CO poisoning effect and thus, recovering the activity of PEMFCs. On the occasion, we have also included a brief definition of on-board preferential CO oxidation method.

234. 5, line 234. It is not clear to me the kind of support DFT can provide to understand CO tolerance in mechanical terms. Did authors refer to mechanical stability of the electrocatalysts? If yes, can authors explain in which way DFT can be used to predict mechanical stability?

We used incorrect wording to express the meaning we wanted to mention. We thank you for highlighting the relevant text. The miswording has been corrected. We are not referring to the mechanical stability of electrocatalysts. What we tried to refer to, is that DFT calculations can indicate which of the two CO tolerance mechanisms (electronic effect or bifunctional mechanism) is dominant in a CO tolerant electrocatalyst. For example, taking Pt as a reference, a positively shifted ΔE_CO of a modified catalyst indicates the introduction of the electronic effect, while a negatively shifted ΔE_OH indicates an active bifunctional mechanism. To clarify this issue, we have added a sentence in the relevant par. 2.3 summarizing the above explanation in the revised manuscript.

279. 7, line 279. Please replace “valance” with more correct “valence”.

The word has been replaced, according to the comment of the reviewer.

693. 17, line 693. Did the expression “mesoporous size” refer to mesoporous volume or mesoporous radius? Can the authors be more precise about this?

In the relevant text, the mesoporous size referred both to the mesoporous volume and the modal mesopore diameter, used to calculate the diffusion rate by the Knudsen equation. We have clarified this generality in detail in the revised manuscript.  

783. 19, line 783. The expression “TiMoO2-C composite” might be more appropriate than the used “Pt on Ti-O2 composite mixed with Mo”.

The expression has been revised to ΄΄Ti_xMo_1-xO₂-C composite΄΄ based on reviewer’s suggestion.

848. 21, lines 844-848. I suggest to specify in the caption of Figure 11 that the numeric values associated with symbols refer to the corresponding entries of Table 1.

A specification in the caption of Figure 11 has been added according to the comment of the reviewer.

864. 22, line 864. Please, use the more appropriate “adsorbed” rather than “absorbed”.

The word has been replaced according to reviewer’s suggestion.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I am very happy with the revisions that authors made. The review is an outstanding work and represents a highly informative contribution in the field. I strongly recommend it for publication on Catalysts.