Fluid Components in Cordierite from the Rocks of Epidote-Amphibole Facies of the Muzkol Metamorphic Complex, Tajikistan: Pyrolysis-Free GC-MS Data

Round 1

Reviewer 1 Report

The authors studied the composition of volatile components of cordierite from the Muzkol metamorphic complex in detail by using pyrolysis-free gas chromatography-mass spectrometry (GC-MS) with simultaneously IR and Raman spectroscopy, determined that it is composed of aqueous-carbonic acid fluid and homologous series of various organic compounds, and analyzed the changes of volatile components from the crystal center to the edge. The test and analysis procedures are detailed and reliable.

1. In addition to the cordierite thermometer, whether to use other thermometers to calculate the temperature, and whether it is consistent with the regional temperature and pressure.

2. What are the implications of the different concentrations of organic compounds in the volatile components of different zones of cordierite for the mineral formation and background changes (temperature, pressure, etc)？

3. Check spelling and punctuation. Line 261, It should be added ". " at the end.

Author Response

Dear colleague,
Thank you for your work on our article and your the valuable remarks on the work. 
The answers to the questions mentioned in the review are given in the file attached.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript entitled “Fluid components in cordierite from the rocks of epidote-amphibole facies of the Muzkol metamorphic complex, Tajikistan: pyrolysis-free GC-MS data” is devoted to comprehensive analysis of fluid preserved in the cavities and channels of jewelry-quality cordierite crystal.

The Authors provide high quality characterization of rock composition and volatile components trapped in the rock. The Authors use perfectly designed methodology of the research and the scope of analytical methods, which is close to those applied for the previous studies of other minerals (Sobolev at al., 2019a, Zatolokina et al, 2021). This is certainly the strongest part of the present research.

In the current manuscript, the Authors provide lack of conclusions. The presented key result is the presence of 170 organic compounds in the sample with only general words about their origin and estimations of mineral-forming conditions. For example, the estimation of the trends of changes in P and T parameters at different stages of metamorphism based on the distribution of fluid components could be added, as it is mentioned in the introduction. In general, I recommend to improve conclusions and give some estimations on mineral-forming conditions based on the results of CO_2, water, as well as organic compounds distribution within the sample.

I want to thank Authors for the precise and excellent analytical work. The conclusions should be improved. I recommend the manuscript for publication after major revision of the text. The details are given in the file attached.

Comments for author File: Comments.pdf

Author Response

Dear colleague,
Thank you for your work on our article and your the valuable remarks on the work. 
The answers to the questions mentioned in the review are given in the file attached.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors present a very interesting analysis of cordierite from the rocks of epidote-amphibole facies of the Muzkol metamorphic complex. They have combined the techniques of Raman and infrared spectroscopy with GC-MS analysis. Although the authors claim this to be one of the first of these types of studies, there have been many studies using these techniques for over 40 years as documented in many of their references.

The presence of H₂O and CO₂ in the channels of cordierite is well known and well investigated. However, the interesting part of this study is the huge variety of organic compounds also detected in the GC-MS analyses. I have also noted presence of complex organic compounds in my analysis of fluid inclusions from ore deposits and have attached a paper that the authors may not be aware of. The observation of these complex molecules prompted me to ask the following questions that I feel the authors of this paper also need to address:

1. Firstly, I find it difficult to understand how these complex chemicals have survived the high temperatures and pressures of amphibolite grade metamorphism. It is well known in petroleum systems that at temperatures above the "oil window" (i.e. > 350 C) all the complex organic molecules are converted to gases. So the presence of these organic molecules may suggest that they were trapped after the peak metamorphic episode and hence may not be relevant to the metamorphic conditions.

2. The authors state that no fluid inclusions were observed but it is still possible that sub-microscopic inclusions are present and these could contribute some of the water and CO₂.

3. The authors report the presence of large-chain aliphatic hydrocarbons and heterocyclic organic compounds. Given that the channels in cordierite are only about 0.6 nm diameter, it is difficult to imagine how these sometimes larger organic molecules could enter and exit these smaller channels. This requires further explanation.

4. Given the above questions, then the possibility of contamination must also be discussed. It is interesting to note that only H₂O and CO₂ were observed by infrared and Raman spectroscopy. While GC-MS is a more sensitive technique, it is also more prone to contamination. I am aware of the previous studies of these authors where they have experimentally created similar complex organic molecules at high pressures and temperatures. However, I am not convinced that they have carefully checked for contamination. I would like to see the GC-MS analyses repeated using highly cleaned quartz or silica powder to check that no organic molecules are produced when analysing blanks.

The Discussion section of this paper is relatively thin and mainly reiterates the analytical results. It needs to properly address the above questions and come up with a viable model for how these very complex organic mixtures could form in  the Muzkol metamorphic complex.

The paper is well written. However, I recommend that a major revision of the paper is required to address the issues I have outlined above.

Comments for author File: Comments.pdf

Author Response

Dear colleague,
Thank you for your work on our article and your the valuable remarks on the work. 
The answers to the questions mentioned in the review are given in the file attached.

Unfortunately, we have not received the article with your study. But we would like to have a chance to read it.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear Authors,

I agree with all the corrections you’ve done. 

In the second version of the manuscript, I still see lack of conclusions. The presented key result is the presence of 170 organic compounds in the sample including N,S-organics with only general words about their origin. What is the main point of identification of such compounds in the crystal? Also the estimation of the trends of changes in P and T parameters at different stages of metamorphism based on the distribution of fluid components could be added, as it is mentioned in the introduction. In general, I recommend to improve conclusions and give some estimations on mineral-forming conditions based on the results of CO_2, water, as well as organic compounds distribution within the sample.
To be honest my expertise in the field of metamorphic petrology is poor, so it is also possible to publish the manuscript in its present form.  

Author Response

Dear colleague,

Thank you so much for your valuable advices on the manuscript! The conclusions in the work are based on the data obtained, with no related data on all mineral associations presented, thus, authors can not widen the Conclusion part. Authors plan on the further investigation on the matter. The identification of organic compounds is performed using the chromatography-mass spectrometry data analysis. Interpretation of the obtained GC-MS data with identification of peaks and selection of individual components from overlapping peaks was performed using both the AMDIS (Automated Mass Spectral Deconvolution and Identification System) software version 2.73, and manual mode with background correction for the NIST 2020 and Wiley 12 mass spectrum libraries with the help of the NIST MS Search program version 2.4, the search parameters were standard. The estimation on P and T parameters was not performed due to the application of the model, presented in Dufour and Kotov work (1972).

Dufour M.S., Kotov N.V. Thermodynamic conditions for the manifestation of metamorphism and metasomatism in the rocks of the eastern part of the Central Pamirs. Izv. of the USSR Academy of Sciences, Ser.geol., 1972, №10, P. 24-36

Author Response File: Author Response.docx

Reviewer 3 Report

The authors have made some minor additions to the manuscript but I feel they have not adequately addressed the questions I raised in my previous review. These may have been discussed in the cover letter but that was not passed onto me. So once again  I will comment on the points that concern me.

1. The authors concede that they may have analysed some sub-microscopic inclusions  but they do not discuss the effect of these on their results or conclusions. This is important as secondary inclusions may be trapped at any time after crystal formation.

2. There needs to be more comment on what steps the authors took to reduce contamination of the samples. The sections were obviously prepared by mechanical cutting but this is not included in the methods and no information is given on the use of coolant fluids or whether the surfaces were cleaned before analysis. This needs to be added.

3. I still have major problems accepting the implication that large organic molecules can be trapped in the cordierite channels which the authors state are only 2.2 angstoms in radius. This is equivalent to the diameter of a benzene molecule and thus I find it hard to believe that more complex heterocyclic and aromatic hydrocarbons could originate from these cordierite channels.

4 The complex organic mixtures reported by the authors appear to closely resemble liquid petroleum mixtures and it is hard to believe that such mixtures would survive high temperature metamorphic processes as claimed by the authors. For example the Burton thermal cracking process converts petroleum into hydrocarbon gases at temperatures of 370 - 400 C and a pressure of 620 kPa.

As well as a discussion of these issues I have also made some additional comments in the attached manuscript that I think need to be addressed before the paper can be published.

Comments for author File: Comments.pdf

Author Response

Dear colleague,

Thank you for your second review on the manuscript! Here are the answers both on the questions in review letter and in the manuscript file.

Comment 1: The authors concede that they may have analysed some sub-microscopic inclusions  but they do not discuss the effect of these on their results or conclusions. This is important as secondary inclusions may be trapped at any time after crystal formation.

Response 1: For authors did not detect the fluid inclusions in the studies cordierite, it is not possible to estimated its effect on the data obtained. In cordierite, as well as in many other minerals, the presence of submicroscopic inclusions is possible. The complication is the possibility to detect them. E.g., in natural diamonds, alongside fluid macro-inclusions detected by optical microscopy, we have established nanoscale fluid inclusions of a similar composition (Sobolev et al., 2019). However, the detection of nanoscale fluid inclusions in diamonds was possible only using the transmission electron microscopy (TEM) method and because the nitrogen in the inclusions was solid form. The application of TEM on cordierite in order to detect inclusions from nanoscale to submicron is of no use due to the non-solid aggregate state of the inclusions contents. However, in the Kurepin work (2010) it was stated, that the temperature evaluations using Grt-Crd and other cation exchange cordierite-bearing geothermometers do not depend on the presence of volatile components in cordierite.

• Sobolev N.V., Logvinova A.M., Tomilenko A.A., Wirth R., Bul'bak T.A., Luk'yanova L.I., Fedorova E.N., Reutsky V.N., Efimova E.S. Mineral and fluid inclusions in diamonds from the Urals placers, Russia: Evidence for solid molecular N-2 and hydrocarbons in fluid inclusions // Geochimica et Cosmochimica Acta. - 2019. - Vol.266. - P.197-219. https://doi.org/10.1016/j.gca.2019.08.028
• Kurepin, V.A. Cordierite as an indicator of thermodynamic conditions of petrogenesis. Contrib Mineral Petrol 160, 391–406 (2010). https://doi.org/10.1007/s00410-009-0484-4

Comment 2: There needs to be more comment on what steps the authors took to reduce contamination of the samples. The sections were obviously prepared by mechanical cutting but this is not included in the methods and no information is given on the use of coolant fluids or whether the surfaces were cleaned before analysis. This needs to be added.

Response 2: Procedures features and our GC-MS analysis methodology allow us to exclude artificial contamination of the sample. The area of the cordierite plate selected using an optical microscope suitable for analysis (without substitution products, without signs of structure twinning and blockiness, with uniform extinction in crossed nichols, etc.) was drilled out using diamond bit and immediately loaded with tweezers into the cell of sample destruction, i.e. there was no contact and storage in any container. Special measures were taken to exclude any interaction of the sample with organic pollutants. After sealing the cell, the sample was heated in a helium flow and a blank analysis was performed without destroying the sample. The previous analysis allowed to control the release of gases sorbed by the sample surface, including atmospheric components, and at the end of this process to record the system form (data is presented in the attached letter). According to the results of the further analysis, the degree and completeness of the heavy hydrocarbons and polycyclic aromatic hydrocarbons (PAHs) elution from the analytical column was determined while programming the temperature in the chromatograph thermostat. If necessary, the analytical column was thermoconditioned until reaching the required blank.

Comment 3: I still have major problems accepting the implication that large organic molecules can be trapped in the cordierite channels which the authors state are only 2.2 angstoms in radius. This is equivalent to the diameter of a benzene molecule and thus I find it hard to believe that more complex heterocyclic and aromatic hydrocarbons could originate from these cordierite channels.

Response 3: As was mentioned earlier, in accordance with modern concepts, СО₂ occupies the same cavities of structural channels as Н₂О (Armbruster and Bloss, 1982; Aines and Rossman, 1984; Armbruster, 1985; Le Breton, 1989). A free СО₂ molecule has a С–О bond length of 1.16 Å, and in cordierite structure it is reduced to 1.05 Å (Johannes and Schreyer, 1981). The molecule of N₂ also undergoes compression in cordierite structure from N–N = 1.0976 Å in the free state to N–N = 0.905 Å in cordierite.

Water and carbon dioxide were studied extensively, which was not the case with fluid components, including hydrocarbons. The number of works on hydrocarbons in cordierites is limited. In addition, mainly light hydrocarbons (methane, ethane, propane, and butane) are discussed in these works (Zimmermann, 1972, 1973, 1981; Beltrame et al., 1976).

The molecules of normal paraffin hydrocarbons have a zigzag shape. The carbon atoms that make up the molecule lie on the same plane. The chain of СН₂ groups of saturated hydrocarbons has a great mobility owing to the possibility of rotation near ordinary tetrahedrally arranged bonds (Zhdanova and Khalif, 1984). The С–С distance is 1.52 Ǻ, and the angle between the С–С–С atoms is 109°28′. The hydrogen atoms (СН₂ group) are arranged in pairs on the planes perpendicular to the direction of the chain elongation, with a С–Н distance of 1.17 Ǻ and an Н–С–Н angle of 105°. Two neighboring СН2 groups are localized on the planes arranged at a distance of 1.265 Ǻ. The section of the chain along the plane perpendicular to the direction of elongation provides a closed oval shape with an average radius of 2.6 Ǻ (Penkala, 1972). Thus, the size and geometry of chains of normal hydrocarbons allow them to penetrate the channels of the cordierite framework.

The benzene molecule is a regular hexagon. The distance between the carbon atoms in the molecule (1.39 Ǻ) corresponds to the arithmetic mean of ordinary and double bonds. The angle of the С–С–Н bonds is about 120°. The size of the benzene ring with respect to the maximum distance between the carbon atoms is ~3.8 Ǻ.

The formation of carbonyl groups (>С=О) in the benzene ring increases this effective size by 2.42 Ǻ (double the length of the С=О bond in case of two substituents). When one considers that oxygen from this group was borrowed from the framework, this estimate will decrease by at least two oxygen radii (~1.1 Ǻ). Therefore, the molecules derived from benzene can fit well the cavities between the six-membered rings, being, most likely, localized along the С-axis and, hence, along the larger diameter of this cavity. Probably, other radicals (–СН₂, –СН₃, and –ОН) can be also present in one benzene ring in addition to carbonyl radicals. The molecules of these benzene derivates, taking into account the bond lengths, are smaller than those of the rings with a carbonyl group (Bul’bak et al., 2002).

Thus, the ratio of sizes of cavities and channels in the cordierite structure to the critical sizes of the simplest representatives of the detected homologous series of volatiles suggests that they are localized in these cavities. Taking into account the critical diameters of more complex homologs, an assumption can be made about the deformation of the СН₂ groups of linear molecules of normal hydrocarbons and other chain molecules in the cavities of cordierite. The geometric discrepancy between the parameters of structural cavities of cordierite and the size of large hydrocarbon molecules, as well as the proven defectiveness of the material under study, indicates that the localization of particularly large gas molecules and all nonlinear molecules should be attributed to nonstructural.

• Aines, R.D., Rossman, G.R., 1984. The high temperature behavior of water and carbon dioxide in cordierite and beryl. Am. Mineral. 69 (3–4), P. 319–327.
• Armbruster, T., Bloss, F.D., 1982. Orientation and effects of channel H₂О and CО₂ in cordierites. Am. Mineral. 67, P. 284–291.
• Armbruster, T., 1985. Ar, N₂, and CО₂ in the structural cavities of cordierite, an optical and X-ray single-crystal study. Phys. Chem. Miner. 12 (4), P. 233–245,
• Le Breton, N., 1989. Infrared investigation of CО₂-bearing cordierites. Some implications for the study of metapelitic granulites. Contrib. Mineral. Petrol. 103 (3), P. 387–396,
• Johannes, W., Schreyer, W., 1981. Experimental introduction of CO₂ and H₂O into Mg-cordierite. Am. J. Sci. 281, 299–317
• Zimmermann, J.-L., 1972. Application pétrogénétique de l’étude de la libération de l’eau et du gaz carbonique des cordiérites. C. R. Acad. Sci. 275D, P. 519–522.
• Zimmermann, J.-L., 1973. Étude par spectrométrie de masse de la composition des fluides dans quelques cordiérites du sud de la Norvège, in: Soc. Géol. Fr.: Réunion An. Sci. Terre, Paris, P. 418.
• Zimmermann, J.-L., 1981. La libération de l’eau, du gaz carbonique et des hydrocarbures des cordiérites. Cinétique des mécanismes. Détermination des sites. Intérêt pétrogénétique. Bull. Minéral. 104, P. 325–338.
• Beltrame, R.J., Norman, D.I., Alexander, E.C., Jr., Sawkins, F.Y., 1976. Volatiles released by step-heating a cordierite to 1200 °C. EOS Trans. AGU 57 (4), P. 352.
• Zhdanova, N.V., Khalif, А.L., 1984. Drying of Hydrocarbon Gases [in Russian]. Khimiya, Moscow.
• Penkala, Т., 1972. Zapys Krystalochemii. Panstwowe wydamnictwo naukowe, Warsaw, P. 401–402.
• Bul’bak, T.A., Shvedenkov, G,Yu., Lepezin, G.G., 2002. On saturation of magnesian cordierite with alkanes at high temperatures and pressures. Phys. Chem. Miner. 29, P. 140–154.

Comment 4: The complex organic mixtures reported by the authors appear to closely resemble liquid petroleum mixtures and it is hard to believe that such mixtures would survive high temperature metamorphic processes as claimed by the authors. For example the Burton thermal cracking process converts petroleum into hydrocarbon gases at temperatures of 370 - 400 C and a pressure of 620 kPa.

Response 4: It was previously mentioned, that the results of experimental and thermodynamic modeling of C-H and C-O-H systems, and, also, the data on fluid inclusions in amphibolite formations and metamorphism granulite facies, in mantle xenoliths minerals and in natural and synthetic diamonds showed, that complex chemical compounds, including also high-molecular hydrocarbons, with increasing temperature and pressure become sustainable.

Reference list for the abovementioned data presented below:

• Chekaliuk - Naukova Dumka, Kiev, Ukraine 1967, 1971;
• Kenney et al. - PNAS, 2020, 99, 10976-10981;
• Zubkov - Irkutsk University, 2005;
• Zhang and Duan - Geochim. Cosmoch. Acta. 2009, 73, 2089–2102;
• Scott et al. - PNAS, 2004, 101(39), 14023-14026;
• Sharma et al., Energy Fuel., 2009, 23(11), 5571-5579;
• Dolgov et al. - Russian Geology and Geophysics, 1984. N8. P. 91-98;
• Sobolev et al. - Russian Geology and Geophysics, 1985. N4. P. 55-58;
• Tomilenko and Chupin – Nauka, Novosibirsk, 1983;
• Kolesnikov et al. - Chemistry Select, 2017, 2, 1336–1352;
• Sonin et al. - Dokl. Earth Sci., 2014, 454(1), 32-36;
• Sonin et al. – Sci. Rep., 2022 |12:1246|;
• Serovaiskii and Kutcherov - Sci. Rep., 2020, 10(1), 4559;
• Sobolev et al. - Russian Geology and Geophysics, 2018, 59, 1365–1379;
• Sobolev et al. – Geochim. et Cosmochim. Acta, 2019, 266, 197-219;
• Sobolev et al. - Engineering, 2019, V. 5(3), 471-478;
• Sokol et al. - Sci. Rep., 2017, 7, 1-19;
• Sokol et al. – Lithos, 2018, 318, 419–433;
• Sokol et al. - Phys. Earth Planet. Int., 2019, 291, 12-23;
• Tomilenko et al. - Eur. J. Mineral. 1998, 10, 1135–1141;
• Tomilenko et al. - Dokl. Earth Sci., 2018, 481, (2), 1004–1007;
• Tomilenko et al. - Dokl. Earth Sci., 2001, 379(5), 571–574;
• Tomilenko et al. - Dokl. Earth Sci., 2018, 481(1), 953–957;
• Tomilenko et al. - Dokl. Earth Sci., 2009, 426(4), 695-698;
• Tomilenko et al. - Dokl. Earth Sci., 2018, 481(1), 953–957;
• Tomilenko et al. - Dokl. Earth Sci., 2001, 379(5), 571–574;
• Tomilenko et al. - High Temper.-High Press., 2015, 44, 451–465.

Responses to additional comments:

1. What is meant by "pirate-free"?

Response 1: The misprint has been corrected.

2. How were these plates prepared? by diamond saw? Was a coolant used? Were the plates cleaned before analysis?

Response 2: Detailed assay on the sample preparation was mentioned above in the response for the second answer from the review letter. In addition, it must be mentioned, that coolant was not used and the plates were degassed during online pre-analysis, and this prodcedure was repeated until the necessary blank is reached.

3. S

Response 3: The corrections have been made.

4. other visible minerals

Response 4: The corrections were made.

5. How is the shock destruction carried out? More details are required. Is it a mechanical process or does it use ultrasound or another method?

Response 5: Shock destruction of the sample is a mechanical process. Cell for sample destruction (Fig. S1, node 4) is not a vacuum chamber, its characteristics are closer to the crusher with the possibility of heating. Cell for sample destruction helium is constantly flowing through it. Destruction of the sample is achieved by a one-shot downward impact on the shockproof stock (Fig. S1, node 12). Reciprocating motions of the flow are excluded. Cell thermostatic, can be heated to different temperatures at different speeds.

Figure S1 is presented in the Supplemetary materials.

6. This is not true if the channels remain open and fluids are free to enter the crystal at any time?
7. 't understand how this happens if the channels are free to exchange fluids at any time?
8. This is hard to reconcile with the typically stable and prolonged period of amphibolite grade metamorphism.

Response 6-8: The answers of the sixth to eighth questions may be combined. In Bulbak et al. work it was shown, that the fluid components exchange between aqueous cordierite and external fluid is ceased, i.e. the system become closed under the following conditions: Р^fl_com = 50-200 MPa and Т=200-350°С.

As was shown by Kurepin et al. (2010), the analysis of the phase equilibria in the Crd-H₂O-CO₂ system allows to conclude that cordierite could preserve its original volatile content at isobaric cooling in fluid-absent conditions. In such a case, cordierite became fluid-undersaturated even being fluid-saturated originally and remained thermodynamic stable up to low temperatures. Its volatile content could change at isobaric cooling only in the presence of free fluid.

• Bul'bak T.A, Shvedenkov G.Yu., Ripinen O.I. Kinetiks of exchange reaction water-CO₂ in the structural channels of (Mg, Fe2+)-cordierite. Geokhimiya, 2005, № 4, P. 429–437.
• Kurepin, V.A. Cordierite as an indicator of thermodynamic conditions of petrogenesis. Contrib Mineral Petrol 160, 391–406 (2010). https://doi.org/10.1007/s00410-009-0484-4

9. Howevnitrogenwas not detected by IR?

Response 9: Molecular nitrogen concentration is not high enough to be detected using IR.

Author Response File: Author Response.docx

Round 3

Reviewer 3 Report

I thank the authors for their detailed replies to my comments and questions. Could I suggest that some of their justifications be included in the Discussion to assist those who may be less familiar with this type of research.

Author Response

Dear colleague,

Thank you for the work on our manuscript! Your detailed analysis on the work was very valuable. According to your advice, we edited on the Discussion part.

Sincerely,

Ksenia Zatolokina