The Activity of Ultrafine Cu Clusters Encapsulated in Nano-Zeolite for Selective Hydrogenation of CO2 to Methanol
and Valentin Valtchev 1,2,*
Round 1
Reviewer 1 Report
This paper discusses the selective hydrogenation of CO2 to methanol over Cu nanoparticles encapsulated inside nano-zeolites. The topic will be of interest to a large audience is worthy of investigation. The introduction of the manuscript clearly exposes the objectives of the paper, based on the relevant literature reports. The synthesis and characterization of the silicalite-1 zeolite and Cu-modified versions are clearly detailed and appropriately discussed. Catalytic data are collected at 3 MPa pressure and above 200 °C. In situ DRIFTS data collected at atmospheric pressure and room temperature are then presented. I have several concerns with the DRIFTS data, collected under conditions irrelevant to the catalytic data.
1- A major band at around 3000/cm is observed at room temperature under CO2 (Figure 6A), correctly assigned to OH and water stretching modes. However, another band at 1614/cm is then assigned only to a Cu-born formate. The present 1614/cm is likely mostly due to the presence of molecular water (bending mode) present in the zeolite and associated with that at around 3000 /cm. (see Figure 5 and 6 in Meunier et al., Inorg. Chem. 2019, 58, 8021−8029, showing the quantitative relation between these two vibrations modes of adsorbed water). The authors should mention this point, or remove adsorbed water to reveal the true contribution of formates.
2- The peak supposedly at 2073/cm cannot be seen. Baseline distortion actually question the other assignment proposed at 1746/cm.
3- The spectral changes observed upon H2 introduction are quite complex and the interpretation proposed is not convincing. A major negative band near 3000/cm is not discussed. Baseline distortion effects make again impossible to assess the presence of adsorbate bands. The 1638/cm band is again likely mostly water, not formate.
Overall, the material synthesis and catalytic sections are rigorous and worth publishing, while the DRIFTS section is not. I would suggest to totally remove the DRIFTS section, or collect DRIFTS data under condition that would not create such strong baseline distortion (above 150°C to readily remove water, e.g. See Meunier et al. Catalysts 2022, 12, 793).
Author Response
This paper discusses the selective hydrogenation of CO2 to methanol over Cu nanoparticles encapsulated inside nano-zeolites. The topic will be of interest to a large audience is worthy of investigation. The introduction of the manuscript clearly exposes the objectives of the paper, based on the relevant literature reports. The synthesis and characterization of the silicalite-1 zeolite and Cu-modified versions are clearly detailed and appropriately discussed. Catalytic data are collected at 3 MPa pressure and above 200°C. In situ DRIFTS data collected at atmospheric pressure and room temperature are then presented. I have several concerns with the DRIFTS data, collected under conditions irrelevant to the catalytic data.
Response: Yes, we admit that the conditions for the in situ FTIR characterization is away from the real reaction conditions, and we share the concerns of the Reviewer. The lower pressure was a compromise to the experimental set-up, for we could manage only a low-pressure cell with IR-transparent windows. Consequently, we were forced to record the spectra of the adsorbed species only at a lower temperature, since they have weak signals due to the low amounts at the low pressures. The discussions on the concerns are as below.
1-A major band at around 3000 cm-1 is observed at room temperature under CO2 (Figure 6A), correctly assigned to OH and water stretching modes. However, another band at 1614 cm-1 is then assigned only to a Cu-born formate. The present 1614 cm-1 is likely mostly due to the presence of molecular water (bending mode) present in the zeolite and associated with that at around 3000 cm-1. (see Figure 5 and 6 in Meunier et al., Inorg. Chem. 2019, 58, 8021−8029, showing the quantitative relation between these two vibrations modes of adsorbed water). The authors should mention this point, or remove adsorbed water to reveal the true contribution of formates.
Response: Yes, we thank the reviewer to point out that H2O has simultaneously the stretching at 3000 cm-1 and the bending at 1614 cm-1. Now we added this point in the text, with the literature references that the Reviewer has shared. Furthermore, we inspect the 1614 cm-1 peak (and the 3000/cm peak) closely, and suspect based on the asymmetrical shapes that they may contain multiple components, including formate, perhaps also carbonate, in addition to the bending vibration of H2O. The discussion of this possibility is updated in the text.
2-The peak supposedly at 2073 cm-1 cannot be seen. Baseline distortion actually question the other assignment proposed at 1746 cm-1.
Response: Yes, these peaks are so weak that the 1746 cm-1 peak is just above the noise and the 2073/cm peak is barely seen. We have changed the text to provide a discussion of the possibilities rather than to claim their existences.
3-The spectral changes observed upon H2 introduction are quite complex and the interpretation proposed is not convincing. A major negative band near 3000 cm-1 is not discussed. Baseline distortion effects make again impossible to assess the presence of adsorbate bands. The 1638 cm-1 band is again likely mostly water, not formate.
Response: Yes, there are two negative bands, at 3740 cm-1 and 3000 cm-1, which are related to water. H2O are formed during the course of reactions, and dynamically adsorbs/desorbs from the catalyst, according to the spectra. Meanwhile, the 1638/cm peak (and the 2856/cm peak) remains positive, indicating it probably contains some formate.
Overall, the material synthesis and catalytic sections are rigorous and worth publishing, while the DRIFTS section is not. I would suggest to totally remove the DRIFTS section, or collect DRIFTS data under condition that would not create such strong baseline distortion (above 150°C to readily remove water, e.g. See Meunier et al. Catalysts 2022, 12, 793).
Response: Yes, we admit that the current FTIR data are premature. And the assignments we provide are only preliminary and not rigorous. Currently we are not able to perform the experiment at higher temperature because the Cu loading amount is still low and cannot provide enough spectral intensity for the adsorbed species at higher temperature. Furthermore, in our systems higher temperature may also not help to remove water and provide a better baseline, because H2O are formed continuously in the reaction and get trapped by the zeolite. Despite these, we still wish the Reviewer to agree that we share the preliminary data and assignments to the public, since they offer possible evidences that the reaction may proceed through a particular pathway over the zeolite-encapsulated small particles, which is a new piece of information, and may help the other researchers with similar interests to carry out more experiments.
Reviewer 2 Report
The manuscript is well written and the study is sytematic and well organized, and it results interesting. However some minor corrections have to be addressed before its publication.
- Table 1 must be remade, it is not easy to understand and some figures have no units.
- If particles size is so important for catalytic results it would be better to be more precise reporting this, it is not convenient to write " around 180 nm", particle size distribution or other method for a better description of the sizes should be included (lines 156-162).
Author Response
The manuscript is well written and the study is systematic and well organized, and it results interesting. However, some minor corrections have to be addressed before its publication.
- Table 1 must be remade, it is not easy to understand and some figures have no units.
Response: Yes, it is specified in the table now that (1) the initial gel compositions are given in molar ratios; and (2) the Cu contents are measured for the products.
- If particles size is so important for catalytic results it would be better to be more precise reporting this, it is not convenient to write " around 180 nm", particle size distribution or other method for a better description of the sizes should be included (lines 156-162).
Response: Yes, the text has been rewritten with more precise descriptions of the particles’ shapes and sizes.
Reviewer 3 Report
This manuscript does not contain any information about stability/durability of the synthesized catalyst, although it is crucial for evaluating the catalyst. Without such data reported, further consideration would be difficult.
Author Response
This manuscript does not contain any information about stability/durability of the synthesized catalyst, although it is crucial for evaluating the catalyst. Without such data reported, further consideration would be difficult.
Response: Yes, we would agree that stability/durability of a catalyst is important, and is crucial in evaluating it. However, the present paper is about a model catalyst that is designed and prepared to test the hypothesis, that reducing the Cu particle sizes would positively change the activity in CO2 hydrogenation. In the paper we present that the steady-state (in a few hours on stream in a fixed-bed microreactor) STY of methanol over the small particle is comparable or higher than the benchmark catalyst in literature; and the in situ FTIR (in 30 min on stream) implies a preferential reaction pathway. These results are the positive supports to the hypothesis. The question on the durability or lifetime is another topic beyond the scope of the paper, therefore it is not tackled, yet.
Round 2
Reviewer 1 Report
I thank the authors for providing some answers to my earlier queries. Yet, I still think that the current state of the in situ IR data is not satisfactory to allow publication, unlike the other sections that are very good. I suggest to the authors to wholly remove it and redo experiments, for a future paper, above 150 °C that should dramatically reduce the concentration of adsorbed water. The current level of adsorbed water and baseline distortions render impossible any sound analyses of the IR spectrum. The origin of the sharp negative around 3000/cm should be investigated as it suggests a contamination or removal of a template. The 1121/cm band is unlikely to be due to a copper-bound methoxy, it is at the most a zeolite-bound methoxy and more likely a modification of a vibration of the zeolite lattice.
In conclusion, I propose to accept publication, after the IR section is removed.
Author Response
Response:
thanks to the Reviewer’s compliments to the other parts of our paper, as well as the critics to the IR study. We now have carried out a complementary study of the same material in the same atmospheres at 150°C, as the Reviewer suggested. Indeed, at 150°C in CO2, the H-bonded water is almost completely removed, while the 1614 cm-1 bands (carbonate, formate, and H2O) remain; so does the 1746 cm-1 (formyl), and the 2073 cm-1 (carbonyl). At 150oC in CO2 + 3H2, the formate- and carbonyl-related species are generally weaker but mostly still exist. Furthermore, a siloxyl -Si-OH band grows around 3450 cm-1, due to the continuous formation of H2O. We have also corrected the wrong claim for the 1121 cm-1 band. This does not belong to an adsorbate, but the zeolite. However, the big band around 3000/cm seem to be adsorbates on top of water, because they disappear at 150oC.
We agree with the Reviewer that these data are very pre-mature, and our attributions contains ambiguities. One reason is that our model catalyst contains only small amount of Cu clusters. But again, we wish to publish these preliminary data, in order to show the uniqueness of the zeolite-encapsulated small clusters, as well as to add IR data for the community to analyze and discuss on the attributions.
Reviewer 3 Report
I am not quite convinced by the authors' answer to my comment. However, I somwhat understand the authors' situation given the Catalysts journal's readership.
I still suggest for the authors to at least discuss deactivation that possibly occurs on the authors' catalyst during the hydrogenation reaction. They may refer and cite the following paper for the discussion about deactivation: Kim et al., Recently developed methods to enhance stability of heterogeneous catalysts for conversion of biomass-derived feedstocks, Korean J. Chem. Eng., 36(1), 1-11 (2019).
Author Response
Response:
Yes, we referred to the paper on the discussion that sintering is a common reason of deactivation of metal catalysts; and we added in the last sentence of the introduction that durability/lifetime of the model catalyst is not specifically studied in this paper.
