Distinguishing Features and Identification Criteria for K-Dioctahedral 1M Micas (Illite-Aluminoceladonite and Illite-Glauconite-Celadonite Series) from Middle-Infrared Spectroscopy Data
Round 1
Reviewer 1 Report
The manuscript is interesting, well written, well organized,
extremely rich of information.
I good paper.
Not easy to read, however, due to the huge amount of data.
It is nearly a chapter of a book, rather than a paper.
I have a general, very general objection concerning the overall discussion.
Nowadays this kind of identification through spectroscopic features of various compounds, with, very often, minor chemical or structural differences from one to the other, might, or, better, should be supported by simulation. Quantum mechanical simulation.
Simulation is very complementary to experiment.
Producing the spectrum (IR, Raman, EPR) of a system whose composition is well defined, as it is selected in input to the computer code, permits to establish a very clear one to one correspndence from local features (say a functional group) and a specific peak.
I am not asking, obviously, at this stage, to perform quantum mechanical calculations for the various sets of compounds.
But that would enormously improve the quality of the manuscript.
That remains, I confirm, interesting and publishable..... but a bit partial.
Author Response
We greatly appreciate the high appraisal of our manuscript and agree that it may appear to be nearly a chapter in a book. In fact, our decision to write this paper was in part inspired by requests from a number of colleagues to present our recent results on the FTIR identification criteria for K-dioctahedral micas 1M in an extended and generalized form.
We agree that quantum-mechanical simulations are very helpful and supplementary to the empirical or semi-empirical approach in interpreting the IR experimental data, and we are grateful for the understanding that such simulations would be far beyond the scope of the present paper. In future, however, it would be indeed interesting and beneficial to compare our results with those of the quantum-mechanical simulations.
Reviewer 2 Report
This manuscript (minerals-680450) describes the application of mid-infrared spectroscopy to distinguish micas. This manuscript seems to be an extension of the authors’ previous works published in Clay and Clay Minerals 2017, 65, 234-251. The experimental results are likely carefully done, and the research subject is interesting.. However, my concerns are mainly on the interpretation of their measurements. Several questions have to be answered and I am mentioning them as follows.
1) This manuscript is in lack of information from recent published works (within 2 years or later than 2017) in the literature. The authors should update and be familiar with recent advances in the scientific community. Actually there exist a number of recent studies focusing on the IR application of minerals. For example, Ventura et al. (see Physics and Chemistry of Minerals 2019, 46, 759-770) use FTIR to study the principle O-H-stretching region of minerals, and the two-mode behavior was proposed (p.764).
2) I notice that the authors used the KBr pellet method for their IR measurements. However, KBr tends to absorb moisture. Although the authors tried to remove the content of water, the minor peak at ca. 3400 cm-1 is still observable in most of their samples (For example, in Figure 1: samples 60, BHS11, 551, 103, 69…). Such minor peak at ca. 3400 cm-1 may imply a large absorption band at ca. 3600 cm-1. Please check the work of Lopez-Pastor et al (J. Phys. Chem. B 2006, 110, 10896-10902) for the detail. This manuscript should provide an alternative way to measure the water content to remove my concerns. The authors should address this question.
3) In pages 18-20, the authors describe that the O-H bending bands locate in the region 950-600 cm-1. Based on my understanding, the OH banding of water locates at ca. 1600 cm-1, and O-H banding mode is red-shifted to 1165 cm-1 for topaz (see Physics and Chemistry of Minerals 1997, 24, 551-554). Could the authors provide the evidence (or experimental clues) for their O-H bending assignment?
4) The spectrum of Z1 (in the bottom of Figure 1) is empty in the region of 600-400 cm-1.
Based on the above reasons, I suggest that this manuscript should be modified and checked after suitable revision.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 3 Report
The manuscript minerals-680450 entitled “Distinguishing features and identification criteria for K-dioctahedral 1M micas (illite-aluminoceladonite and illite-glauconite-celadonite series) from middle-infrared spectroscopy data” has been reviewed. The manuscript is focused on the determination of specific features and identification criterion based on Mid-infrared spectroscopy for distinguishing K-dioctahedral 1M micas. The study was based on a large set of natural samples which are previously characterized in the literature by Infrared and other technics. The present paper provides numerous characteristics IR features and band positions in the Si-O bending and stretching region, as well as the OH-stretching region to discriminate both Illite - Al-celadonite and illite-glauconite-celadonite series. The authors present some interesting relationships between position shifts of bands and the crystal-chemistry of the samples. The paper also demonstrates that the decomposition of the OH-stretching region allow to determine the octahedral cation content of K-dioctahedral 1M micas.
The paper is well written and provides some lacking knowledge in the literature that could be interesting for many mineralogists, however, there are lots of lacking data. Notably, as the paper is only focused on IR spectroscopy, the authors have to be provided all spectra of each sample (at least in the supplementary file) and parameters used to decompose the IR spectra. So, I can recommend the present manuscript only if the authors provide all data concerning the sample spectra and decomposition parameters. Half of the presented spectra are previously published in the previous paper Zviagina et al. (2017) Clays and Clay Minerals, 65, 4, 234–251.
Here below are suggestions and comments to support my statement.
Major Comments:
The infrared spectrum of each sample must be presented in the paper at least in supplementary files. Indeed, the paper is focused on the IR analysis, but the authors did not present the spectrum of each natural sample. The authors only provided the band positions in some unreadable big tables. Concerning the decomposition of IR spectrum; to be sure, Did the authors calculate the cation content using the relative area of the components? Because, the cation content has to be calculated from the area of the bands and not from the intensity of bands.In any case, the authors have to be provide (in supplementary files) all the parameters of components (height, width, and %of Gaussian) used to decompose each spectrum. Indeed, the calculation of the cation content is very dependent to the parameter used, notably, the widths of components. For each sample, as the OH groups are assumed to be in the same mineral phase, all widths of each component have to be the same (or very close to each other).
As notice in many papers in the literature, the decomposition of the infrared spectrum is very user-dependent. In order to constrain the decomposition of the spectrum a signal treatment, as the second derivative, is a powerful tool to constrain the number and the approximate position of the components that composed a complex massif of bands. The authors used a second derivative to evidence some very weak shoulder as mentioned in the text, but did they used their second derivative to constrain the number and the position of the IR bands?
The authors validated their approach for calculating octahedral composition by comparing their results from FTIR with the formula units for 10 samples. Why these samples? And does it works too with the other samples, for example Fe-illites?
Minor Comments:
I recommend to place some tables in supplementary files for increasing the clarity of the paper to not lose the reader in the numerous large tables. Figure 7: The figure is unreadable due to the label of numerous positions of the features. As all the positions are reported in the table 4, I recommend to remove some position labels for more clarity. Figure 8 e) for the attribution of the band at 3531 cm-1 replace Fe2+OHFe3+ by Fe3+OHFe3+ as in the table 5.Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors have made suitable revision based on my previous comments. I recommend to publish this manuscript as it is.
