Direct Synthesis of Hydrogen Peroxide under Semi-Batch Conditions over Un-Promoted Palladium Catalysts Supported by Ion-Exchange Sulfonated Resins: Effects of the Support Morphology

Round 1

Reviewer 1 Report

The manuscript by Frison et al describes interesting results about the direct synthesis of H2O2. However, two questions require to be addressed before publication.

- The minor one is the quality of presentation. The manuscript is not easy to read due to problems such as the changes in font size, the messages “Error! Reference source not found”, the structure and size of parameters such as apparent rates (lines 207-210 and 238) and units (lines 266, 272), quality of some figures (numbering of insert in Fig 5 and 6), and finally, some of the references are incomplete (publisher of 22 and 27; type of publication in 6; letter A in journal names of 14, 24, 30, 31; article number of 41, as it has no page) or wrong (publisher of 1).

- The main issue is related with the lack of a true explanation about the behavior of the different catalysts. The study of particle size has been limited to only one of the catalysts, although this property might be of crucial importance to explain the changes in activity and selectivity with ageing. An alternative might be the measurement of CO physisorption to see the differences in Pd surface. If not, the manuscript is only a collection of interesting results with no insights into their origin.

Author Response

The minor remarks of the Referee have been integrated into the manuscript. In particular, he incompatibility with links producing errors to the references to figures should be solved and  structure/size of units and parameters, as well the references have been amended.

As to the main issue pointed out by the Referee, we modified the text to better emphasize the role of the support in controlling the performance of the catalysts, which in our opinion is enough to explain the differences in the behaviour of the catalysts. As to the CO chemisorption measurements  they cannot be applied to this kind of catalysts, because it is a dry state technique. In this state the contact of CO with the metal nanoparticles cannot be ensured in resin-supported catalysts and the result can be no chemisorption at all (according to previous findings on macroporous resin based catalysts). A sentence was added in the text, with proper literature citations, to illustrate these circumstances.

Reviewer 2 Report

The paper by Frison et al. reports an original an valuable investigation of a palladium based catalyst for the production of H2O2. The paper is sound and well written, but I recommend some minor changes before publication:

Title: why semi-batch conditions?

All references to figures contain errors (Error! Reference source not found).

Scheme 1: What happens to Na?

Figure 4: the apparent consumption of H2 increases almost linearly with time for all compounds, but in the case of Pd/µS-pDVB35 one observes a saturation of the produced H2O2. Please, specify that the reaction is shifted towards the production of H2O.

Lines 217-219 seem to be contradictory.

There are so many figures in Table 3 and poor explanation of their meaning. Explain what are R, C and S. The same is true also for Table 4.

Figure 6: the symbols of Pd/mS-pSDVB10 are hardly visible.

In all figures replace the comma in the numbers with a dot.

Please, check that the captions of Figures 7 and 8 are correct.

At least in the provided pdf there are many characters larger than others or vertically displaced by the line center.

Author Response

The remarks suggested by the Referee have been carefully integrated into the text. In particular, the incompatibility with links producing errors to the references to figures should be solved and the text has been amended to avoid vertical displacement and different font size.

Scheme 1 has been amended to include Na.

“semi-batch conditions” has been removed from the title. “semi-batch conditions” has been originally mentioned in the title to point out the kind of reactor used in the experiments, conceptually different from batch reactors, often used for this kind of investigation.

The behaviour of the catalyst observed in Figure 4 is commented in the final part of the paragraph, before the discussion of TEM characterization. The total selectivity towards water has been made explicit with a suitable addition to the text.

The inserts of the figures (9 and 10 in the amended version) have been amended.

“Lines 217-219 seem to be contradictory”: Pd/mS‑pSDVB10 are the least active catalysts according to their specific H2 consumption rates are the lowest. On the contrary, Pd/C catalyst shows much higher activity, but with low H2O2 productivity, therefore indicating that H2 is mainly converted into water. The sentence has been reformulated in order to make this point clearer.

The captions of Tables 3 and 4 have been updated, according to Referee’s suggestions, to clarify the meaning of the exploited symbols.

Round 2

Reviewer 1 Report

The authors have improved the quality of the presentation, although the mistakes in the references are still present, probably due to problems in their references manager. For example, Ullmann's Encyclopedia is not published by the American Cancer Society (ref. 1), it is not clear what is reference 6 (article? patent?), and several journal titles are incomplete.

Regarding the question of CO adsorption, I agree that swelling is a problem for dry determinations, but I wonder which is the swelling coefficient of the macroreticular resins described in this manuscript, mainly in the case of poly(divinylbenzene), where the crosslinking degree is very high. Moreover, the access to the nanoparticles, once they are formed, might be less swelling dependent than the sulfonic sites in the preparation procedure.

I understand that this subject is not easy at all, although the full characterization of all the catalysts, either by TEM or by CO adsorpotion, at least those supported on sulfonic resins, would be advisable. Is XPS suitable for characterization of the Pd oxidation state? This would be useful in the case of aged catalysts.

Author Response

The swelling degree of pDVB is quite high in spite of its high cross-linking degree and this is the consequence of its peculiar morphology, which can be described as an open cell gas-in-solid "meso-foam" (ref.s 39-42).
Whereas in conventional resins swelling implies the penetration of the liquid agent inside the polymer framework so that it solvates and separates the polymer chains forming a gel-phase more or less extended, for pDVB the phenomenon is believed to be of a very different nature.

In fact in dry pDVB the meso and nano-bubbles (the pores) dispersed in the continuous solid phase are "squeezed" by the elastic forces acting on the polymer chain so that the gas-solid interface (the pore walls) is featured by the presence of creases and wrinkles. When pDVB is swollen, the meso- and nano-bubbles expand so that the gas-solid interface gets flatter. Moreover, the SEC characterization of swollen pDVB shows that anyway some gel phase is formed upon swelling even in this material (ref.s 39-42). As the metal nanoparticles are formed in swollen pDVB, in the dry state they can be well entrapped within the creases and wrinkles formed on the pore walls and excluded from the contact with gases even though they are not actually located inside the polymer framework. We have already attempted CO adsorption measurements of other samples (not included in this work), but we did not detect any adsorbed CO (hence these results were not published), which is in line with the literature cited in the paper.

As to the possibility of XPS analysis of course it could be useful, but the request from the referee seems a quest for replacing a characterization technique with another one. We are under the impression that he/she believes that characterization of the materials is not sufficient for publication. On the one hand we are perfectly aware that a much  more thorough characterization could be possible, but on the other we also believe that the experimental results, including the pieces of characterization reported in the paper, are fully compatible with our interpretation of the catalytic behavior as the consequence of the differences in the morphologies of the support. This was never
contended by the referee and for us is more than enough to publish this at least as a working hypothesis. By the way we are not aware of any previous report on aging of catalysts for the DS, which in our opinion makes the publication of these results even more useful even though causes have not been fully disclosed yet. In this sense this can be considered as a preliminary work and confirmation (or refutation) of our conclusions will be found with further research.