Comparison of Catalysts with MIRA21 Model in Heterogeneous Catalytic Hydrogenation of Aromatic Nitro Compounds

Round 1

Reviewer 1 Report

The authors present the article comparison of catalyst with MIRA21 model in heterogeneous catalytic hydrogenation of aromatic nitro compounds, and this publication is of really high interest. However, I have some minor revision to submit to authors.

1- I agree that for a literature comparison a good mechanism understanding is mandatory. That is why in Figure 2, authors should correct their hydrazobenzene molecule (in which N=N should become N-N).

2- In page 6, line 185, authors should precise "second class" before "the parameters", as in this description it is not mentioned.

3- Still in page 6, before table 2, the authors weight parameters compared to others, but the reader does not understand well why, for example, some are 50% less in weight compared to the first class.

4- The authors chose to standardize their data speaking about TON (page 8) and to avoid TOF. However, this is not well discussed, as the two parameters do not lead to the same data. TON is about the way of catalysing before leaching, TOF is somehow directly linked to the kinetic. The author should insert the TOF as well in their discussion.

Finally, I have to highlight the merit of data availability statement where a website was created by the authors for the reader to go deeply into the MIRA21 performance.

 

For all these reasons, I suggest the publication of this paper after minor revisions.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The present paper by Viskolcz and coworkers deals with a problem, the analysis of an increasing amount of data, which is becoming more and more important. As such and in the aim of stimulating the discussion in the field, I think it should be accepted for publication in Catalysts. However, for the sake of objectivity, I should also say that the work fails in what should be its aims. I evidence in the following those that I think are the weak points of the work, in the hope that this will be a stimulus for the authors in future works.

1) The authors considered only 154 catalysts from 45 papers. This number is extremely low with respect to the actual literature. I made a very restrictive search on SciFinder limiting the reaction to the reduction of unsubstituted nitrobenzene to unsubstituted aniline by H2, excluding patents and limiting the period from 2000 to 2021. Still 961 reactions from 722 articles where identified. If the limits are loosen by extending the reaction to other nitroarenes (which actually constitute the problem), the number of reactions jumps to 65900. Loosing even the other constrains, the number is higher than 300000. Given these numbers, limiting the number of considered reaction to 154 does not allow getting any general conclusion.

2) The general criteria employed to select the publications were stated, but clearly not rigorously applied. For example, the introduction of cobalt catalysts on N-doped carbon is attributed to Zhang, but the first to report this strategy was Beller (not me) in several papers the earliest of which published on Nature and Science. None of the papers by this author is considered. Clearly, many very important papers were neglected.

3) The data had to be extracted "manually" from the original literature. I fully understand the problem of automatically extracting the data by some sort of artificial intelligence, but as far as this will not be possible, the analysis will be limited to very small data sets, that even a human intelligence can handle.

 

I understand that these comments cannot be considered at the actual stage because it would mean to restart the work from the beginning. However, a mention or discussion of these limits must be included.

 

Independent of what said above, there are some editorial points that must be amended.

1) Figure 2 is somewhat confusing. In its central section, azoxybenzene is derived from a reaction involving aniline, but no such reaction exists. The authors should rethink the figure.

2) It is not at all clear how the authors defined the "n catalyst" number in eq 4. In the cited review (ref 87), the problem of which number to attribute to the catalysts is discussed. Which value was considered here, the molar amount of catalytically active metal, the number of active metal atoms in the catalyst or what else? The definition the authors state having applied says "….reaction cycles occurring at the reactive centre..". From the latter it would appear that the number of actually active metal centers was employed, but this number is most of the times not known at all or estimated in very questionable manners. The authors should be more explicit.

3) In figure 8, the bars for the first and third quartile are adjacent. Is that correct? Should it be the second quartile?  

 

Minor points

page 2, line 74: "affect", not "affecting"

page 3, line 11-112: "activating", not "activate"

page 14, line 396: "operate", not "operation"

page 16, line 429: "included", not "include"

page 16, line 432: "activity", not "reactivation" (I guess. As it is now it makes no sense)

Page 18, line 487: "Given that palladium is much more expensive than platinum…" It is true that at the moment palladium is more expensive than platinum, but this has been so only in the last three years of so, when palladium had a sharp price increase.  Along the years, platinum was generally from more to much more expensive than palladium and the situation may revert to any moment. Suggesting to develop platinum catalysts instead of palladium ones is highly questionable. Historically, palladium was much investigated because it was cheaper than platinum.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors present a topic about using a homemade MIRA21 model for ranking the heterogeneous catalysts in aniline production. Authors compared 154 catalysts from 45 articles (96% Q1 journals and 14% Q2 journals) published in the last 20 years using data processing and made a list of ranked catalysts. This is an interesting method for catalyst development. However, it is suggested that the authors should improve the following aspects to form a publishable manuscript.

1. In the introduction part, the MIRA21 model was introduced without any description. Please make it clear.
2. The introduction part should be re-written because the novelty and originality of this study were not pointed out as compared to other reports.
3. The authors mentioned that this study aims to shorten the innovation path between industrial technologies and academic research. However, there is no “catalyst cost” parameter in the descriptor system of the MIRA21 model. Diversity catalysts with different supports and different active metals have a wide range of costs were investigated. Thus, the author should add “catalyst cost” as one parameter to compare the catalyst.
4. How can you consider the weighting for each class in the descriptor system?
5. Why “catalyst particle size” and “catalyst surface area” parameters were selected? The third class (catalyst physical properties) was considered as with 80% less weight with scores ranging from 4 to 6. Please explain how to score these parameters. The wide range of surface area and particle size of diversity catalysts was scored in the very narrow range. Authors should re-score the parameters in the third class.
6. In the fourth class, the “carrier effect” and “nature of effect” should be one parameter.
7. As you know, both active metals and supports have a strong influence on catalyst activity. How to score the effect of the carrier on the catalytic reaction? Why do authors not consider the “active metal effect” parameter?
8. Which tool was used for data mining? There is no mention in the manuscript.
9. There are two “catalyst” keywords. Do you think that “nitro” and “comparison” keywords are necessary? Why “MIRA21 ranking model” could not be a keyword?
10. The conclusion part should be re-written to be more concise and cohesive.
11. I would like to suggest that the author should use a professional English editing service (for example Editage, etc.) or the manuscript should be carefully checked by a native English speaker to correct the grammar, mistyping words, and enhance the soundness of scientific writing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Thank you very much for improving the manuscript. Although there are still some aspects that have required the further investigation, I would like to accept this version to publish on Catalysts.

In addition, I would like to give a suggestion to the authors. It is important to remember that when writing the scientific article, the focus point is the research and not the persons who conducted the research. Thus, it is better to avoid personal pronouns in the scientific writing even when statement is personal opinion from the authors of the study.