In-Situ Crystallization and Continuous Modification of Chromian Spinel in the “Sulfide-Poor Platinum-Group Metal Ores” of the Norilsk-1 Intrusion (Northern Siberia, Russia)

Round 1

Reviewer 1 Report

Please see attached file

Comments for author File: Comments.docx

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Review report on the manuscript: “In-situ crystallization and continuous modification of Cr-spinel in the sulfide-poor PGE ores in the Upper zone of the Norilsk-1 intrusion” by Ivan F. Chayka et al. (Manuscript ID: 796686) submitted for publication to Minerals MDPI.

Dear Editor,

Thank you for consulting me on this article. The authors are weighing in on a hotly debated issue; that is, the genesis of the PGE-rich ores in the upper parts of the Norilsk-1 intrusion (Russia). In my opinion, they contribute valuable information to the debate. In particular, I am impressed with the detailed micro-textural descriptions and the length of the analytical work the authors carried out. Another strong point of this investigation is that it gives proper credit to all the previous works on the topic. I note that the manuscript is well written and the conclusions inferred by the authors are clearly supported by the data they presented. I have only some minor/technical comments to make, which I hope will be useful to authors during revisions. I believe this interesting work deserves to be published as soon as possible.   

• Please replace Cr-spinel with chromian spinel in the title. I would avoid the use of abbreviations in the title (i.e., PGE). There are many geoscientists who are not familiar with the terminology, but might be interested in your work.
• Abstract: the term Cr-spinel must be explained in its first appearance in the text (including the abstract).
• Line 57: I think that economic geologists are more interested in the formation of these deposits than petrologists.
• Line 490: please replace chromite with Cr-spinel.
• Line 509: I think it should be “… regarding the origin of …”
• Line 521: please replace chromite with Cr-spinel or indicate somehow that these names refer to the same mineral.
• Line 529: this paper (reference No. 64) deals with alpine chromitites. There are some minor differences between ophiolitic and alpine complexes. So, in line 527 you can write “… ophiolitic or alpine chromitites …”.
• Line 534: I think it should be pneumatolitic instead of pneumatolithic. Pneumatolitic derives from Pneumato-lysis (Lysis = to unbind, not Lithos), which is a kind of hydrothermal geological process.
• Lines 538-540: Please modify this sentence. It does not make any sense the way it is.
• Perhaps you can provide a shorter version of the last section (Conclusions).

Good luck!

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper thoroughly describes and discusses the textures, composition and petrographic relationships of Cr-spinels from the upper zone of the Norislk intrusion. The rigorous set of observations and analysis presented provide solid evidence for the discussion of the non-cumulative origin of the studied rocks and the Cr-spinels. The methodologies are adequately presented and easy to follow. The use of false coloring imaging related to chemical variations between crystals was properly implemented. It is of interest to note however that the use of an automated technique to achieve this analysis, such as QEMSCAN or MLA, may have provided additional information on petrographic relationships. The only downside of the numerous, rigorous and thorough descriptions is the sheer size of this paper. If possible, the paper could benefit from additional efforts to try to streamline the text. One of the main conclusions, stating that the Cr-spinel remains of magmatic origin, though in a highly viscous environment unaffected by gravitational accumulation processes, is mostly supported by the correlations of chemical variations between the Cr-spinels and various zones of the host rock.  I would be curious to find out if such features could instead be explained by involving more complexity in the fluid evolution, or multiple distinct metasomatic events, though the suggested mechanism is convincing based on the current knowledge on these rocks.  

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

Report on the manuscript “In-situ crystallization and continuous modification of Cr-spinel in the “sulphide-poor” PGE ores in the Upper zone of the Norilsk-1 intrusion” by IF Chayka et al.

General comment

The manuscript is aimed to the study of the Cr-spinel/low sulphide mineralization at the top of the Norilsk intrusion, well known as a world-class Ni and PGE repository. It is an interesting subject, and the authors provide a very accurate petrographic and mineralogical descriptions at the sample scales and below, supported by a wealth of analytical data. These rich data allow interesting interpretations, which are however rather poorly presented in the discussion sections, which demand a serious rewriting (numerous suggestions in the detailed comments below). Regrettably, the work lacks a petrologic interpretation of the studied rock package, what would have been a plus when coming to the interpretations. Should it be possible for the authors to address this topic, I would encourage them to add it into their text: but this not a binding request. Also, in all this work, the alteration affecting the pristine assemblages appears of tremendous importance. Yet, the silicate alteration assemblages and sequences are not well documented, and there is no information on the composition of the alteration minerals. This may be felt as a serious flaw of the study, because, in relation with the very nice observations on the spinels, the question arises of the conservation, or not, of the Mg# when silicates (Ol, Cpx) were, for instance, chloritized. Again, should this information be available, I would encourage the authors to introduce it; but, again, this is not a binding request.

The authors provide a rich and significant set of illustrations. Unfortunately, they are poorly exploited, and, in particular, the figure captions are as a rule too short and lack explanations and specifications. A significant effort has to be done to improve the manuscript from this point of view.

My conclusion is therefore that the manuscript provides a very interesting material which is rather well documented, and contain data that could have led to significant interpretations, but unfortunately remained confusedly exploited. This work is worth to be published in Minerals, but this cannot be done without a very serious work of rewriting, including a better formatting of the concluding interpretations.

Detailed comments

• L56: I think that the genesis of chromitites in ophiolites (so-called podiform deposits) is rather well known! In general, may I add, the main problem is less to understand the genesis of chromite in itself (admitting that it may be a bit more problematic in the case of layered bodies) than to find the genetic link with PGM formation. In the Norilsk case, there is the additional problem of concentrating Pd relative to Pt.

Geological background

• Figure 1: as far as I know, most of the intrusions are not outcropping, and this is not (in my feeling) sufficiently underlined. In addition, text refers to entities (Kharaelakh Syncline, NorilsK Syncline) which are not clearly visible in Fig. 1A. Finally, I think that cross-sections are really necessary in order to assist the reader in the understanding of the nice description of the deposit setting.

Samples and methods

• Figure 2: the images presented in Fig. 2 are precious, but it is highly desirable that they would be accompanied by interpretive sketches! In the same way, a section of the drill hole, with location of the samples, and indication of the relationships between the rock types would be appreciated.
• L177: Fe²⁺to Fe³⁺balance: I would have think that you would have done a control on a charge basis? On what basis was the stoichiometric calculation done (oxygen? Or ∑cations?)
• L180-sq: it is an interesting procedure! And it evidently provided very interesting data. Yet, why not have tried to perform X-ray mapping, with EPMA, at the crystal scale (in the case that the variable hues be determined by a combination of other cations: Al, for example, has the same effect as Mg)? I wish also to draw your attention on the great potential of Micro-XRF mapping for this kind of work at the sample scale (in my lab, we use the Bruker-Nano M4 Tornado instrument).

Petrography

• Table 1: what is the meaning of the “*”? I see that it is done in the text, but a reminder should be done here. And to what is addressed the exponent ⁴? In a general way, why do you not specify the Cr-spinel host (or hosts) when describing each sample? Are all the minerals the same size (the samples the same granulometry)? If not, could it be possible to give the information (here, or in the text body)? Why not mention the minor minerals (phlogopite, Opx, apatite)?

MR14: “distinctly bordered from Cr-spinel-free sites”=??. Do you mean that these cloud-like disseminations (why not “clusters”?) exhibit distinct boundaries?

MR-30 (1): “relatively dense, suspended and homogeneous” =what do you mean by “suspended”?

MS24-321.9: what sulphides? Same question for MS24-323.4-(1) and MS24-325.4

Medvezhnij Rutchej: here is needed an account of the relationships between the rock types!

• L210; “undergone”—> underwent
• L214: “the most distinct” = the most representative?
• Figure 4: the images are nice; but they are not so useful for the reader, because of a dramatic lack of labelling/specifying the displayed minerals! This is particularly disturbing in 4-C. In addition, some magnification would also be useful, for example in 4-B the domain 3, where plagioclase seems to be more or less dendritic??
• L249: as suggested above, a figure with a section of the drill hole would be usefully presented here. In the Table 1, gabbro-dolerite is observed above the leucogabbro: why not mention it in the present part of the text??
• L253: “which finally change with the gabbro- and gabbro-diorite-like rocks”: do you mean a transition?
• L254-256: unclear; do you mean that these rock types are the “prevalent” hosts of Cr-rich layers?
• L270-sq: sample 323-4

Figure 4-C: it is difficult to connect fig. 4-C with the description in the text. What is seen, seems to be a boundary between a very fine-grained leucogabbro and a coarser-grained one, with at the boundary a feature very reminiscent of the stockscheider textures in the granite systems, but here with sulphides (pentlandite?)? What makes the black matrix of the protruding sulphides(?)? From this comparison, it could be suggested that the coarser grained facies was in magmatic contact with an earlier fine-grained unit? Is there a specialization of these different facies in terms of spinels and PGM?

Figure 4-D: please specify 1 to 3 in the caption, as it is not done in the text. Apparently, 1 is representative of the leucogabbro type: but one is stricken by the absence of correlation with Fig. 4-C?

Clearly, a more accurate description of the sample is needed!

Textures

• Figure 5: what is intended by these lines of red spots?

5-A: are the fine-grained parts spinel accumulations? Here again, you should have better specified the figure

5-C: upper left, a feature which would have been worth a magnification: dendritic spinel?

• L311-312: I am unable to see “clusters and “suspended” pattern formed by thousands of fine grains” in Fig. 4-C!!
• L312-313: do you mean that there is independency between the silicate phase textures/boundaries, and the shape of the envelope of spinels clusters, which overprint them?

L313-314 are a repetition of 312-313 in other terms (better chosen). All in all, a rewriting of these lines is demanded.

• L317-321: again, obscure sentences: what do you exactly mean? Maybe it would be sufficient to state that you occasionally observe spinel chains passing through the silicate boundaries, as nicely shown in 5-E?
• L324-sq: this feature is clearly at the opposite of your clusters embayed in the silicate framework; the latter should be indicative of spinel crystals floating in the melt prior the incipient silicate crystallization, while the former clearly shows that the crystals around which cluster the spinels were crystallized earlier.
• L331-sq: The term “circle” is a little bit exaggerated… “subcircular aggregates” better, in my opinion. From your Figure 6, it may be to the contrary deduced that small euhedral crystals of the main phases (plagioclase, Cpx, Ol) floating in the melt preceded the growth of the spinels, that were accumulated on their boundaries (surface effects). Only 5-F is problematic in this respect.

Morphology

• Table 2: “simplectite” symplectite

It is a striking feature that the spinel is not zoned, and this could be emphasized.

• Figure 7: the image numbers are not all coincident with the text??

7-A: the magnification is not pertinent in order to illustrate the crystal shapes (it is in fact a repetition of the textural relationships in preceding figures). However, it shows intriguing sulphide features (filamentous?), which would have been worth a magnification.

7-D apparently corresponds to such a magnification of 7-A better showing the spinel habit?

• L339-sq: I do not see in Fig. 7-B neither the “tightly packed intergrowths and clusters of angular, straight-edged grains”, nor the “intergrown with ilmenite, forming intricately-shaped chromite- ilmenite aggregates” announced in the text (in fact, it is a good counterpart to 7-A), whereas in Fig. 7-C the definition (grey hues not distinctive) and the magnification are not sufficient to appreciate the reality of theses intergrowths. By contrast, Fig. 7-B shows an interesting spinel-Ap-Ccp-Pn assemblage, which would have been worth a commentary (primary magmatic assemblage?).

It seems that (apart from the ilmenite-intergrowth case) you are dealing with euhedral smaller crystals (give the size range!) on one hand, and larger subhedral crystals on the other hand, as very well shown in Fig. 7-E, F. You could start by saying that, on the basis of these two figures, and then present the others: it should be simpler and easier to understand.

I note in Fig. E-F a very remarkable zoning/gradient of composition (whereas the morphological “transition” is not so evident: for my part, I rather see an abrupt change): why not a first comment?

• L343: amaoeboid spinel is shown in Fig. 7-G, not D!
• L350: from Fig. 7-H, it would appear that pyrite replaced magnetite? A better magnification of the magnetite would be appreciated: skeletal growth?
• L353: “simplectite” —> symplectite

Chemistry

I think that the first section (chemical ranges) duplicate the second one (heterogeneity), and finally brings more confusion than anything else. Although I have made some comments of this first section (below), I think that it could be deleted without inconvenient (provided that the crystal chemistry relationships = substitution vectors be addressed in this second section).

Compositional ranges etc

At the beginning of this section, you should specify what will be the reference data base as presented in your Fig. 8; worldwide layered intrusions make sense, as do spinels from other levels from the Norilsk and Talnakh ore bodies (albeit short comments on their significance would be welcome); but why especially the Gudchikhinsky and Tuklonsky picrites??

It is effectively interesting to observe that the Norilsk spinels are not equivalent to those in layered intrusions.

• L361-362: obscure sentence, I do not see the interest, it could be deleted.
• Figure 8: the internal caption (in J) is too small and indeed very difficult to read.

Major elements: I find that your data could have been more thoroughly exploited. For example, the differences (to be explained) between the Medvezhy Ruchey (MR) and MS-24 data when Mg# is considered (in fact, it will be explained in the following section, which, as already said, is sufficient by itself). In both groups of samples, the Mg# distribution is clearly pluri-modal, with two or three modes each, which are not exactly the same, and mostly differ by their frequency, as seen as well in 8-A,B, and in 8-G, for example. In MR, the main mode is at Mg# < 20, with a secondary mode at ~ 30, whereas in MS-24, there is three modes, at ~ 10, 20-30, and ~ 50. This is clearly reflected in the TiO2 vs Mg#, where is seen that the low Mg# and high Mg# groups differ by the slope of the correlation trends. It would be interesting to interpret in terms of substitution vectors in the crystals. In the same way, Fig. 8-D and F show a tri-partition of the data, with a split of MR data and MS-24 data in between, and in Fig.8-I, it is fascinating to observe an array of linear trends for MR data, while MS-24 exhibits only one. Again, an interpretation in substitution terms could be of interest. This is the more visible in Fig. 9, clearly indicating that variably coupled sets of substitution vectors are involved, and not in the same way in the two settings (MR, MS-24): to be related to different rock type supports?

Heterogeneity etc

As said above, I think that the presentation of the data in this section is quite exhaustive and self-sufficient: it explains the confusing data of the preceding section, where the multi-modal histograms clearly demanded the disentangling which is performed in the present section. In view of the discussion to come, the only regret here is that no attempt has been made to check (mineral by mineral) the possible effect of the nature of the hosting rock (leucogabbro vs. picrite, for example). The present section however is also of confuse presentation and demands some rewriting: just start by saying that spinel compositions are highly variable and depend upon (i) the degree of alteration of the hosting mineral, and (ii) the nature of the unaltered mineral. Then, “suffice” to combine the statistical data in Fig. 12 and the nice illustrations in Fig. 11 (the two figures should be switched).

The results are remarkable. They show clearly that the Mg# was controlled by competition with the hosting mineral, with the spinel Mg enrichment inversely proportional to the Mg content of the host, and the same for the Al-Fe relationships. By contrast, alteration had evidently a smoothing effect as far as Mg is concerned, and was characterized by Fe enrichment. The opposite effects of alteration and pristine composition are at the origin of the variety and the complexity of the observed patterns, as nicely displayed by your false-coloured images.

• Figure 12: 12-F, same remark as for 8-J

PGM

• Figure 13: to what figure does refer the white frame in Fig. 13-A? This is a new example of the lack of pertinent information in your figure captions (of course, it is 13-B: but you not make easier the reading work …).

13-A is another example of the spatial gradient in Mg# already presented (and until now, not commented) in Fig. 7-E, F; but here, we have a fascinating pattern, showing clearly that these gradients (and shape modifications) are related to alteration processes (note that in the figure, the symbol Cpx* is barely visible, being partly masked by the white frame), as indeed already stated. But what is new, here, is to show that the alteration effects on spinels are by far more pervasive that those on silicates: Fig. 13-A features a “fluid corridor” and its metasomatic effects. Another important result is that the sulphides and PGM are correlated with this corridor. In this respect, your observations would be consistent with the classical observation that in many mafic-ultramafic systems, a “late” hydrothermal (S, As, …) reworking affected the pristine PGM assemblages.

13-G: do these “spicules” standing on Po are also Po?? This “sulphide band” is very suggestive of a liquid sulphide phase squeezed between the pristine silicates (filter press process?): here, could the apparent sequence Pn-Po-Ccp-PGM correspond to the classical fractionation sequence of a MSS?

• L493-494: your interpretation of the observed textures that spinels and PGM “were not separate processes, but to some extent overlapped with each other” is in contradiction with the data (my interpretation, above; your observation that the associated spinels compositions are related to alteration, below).
• L500-sq: the data presented in Fig. 14 show (in my opinion) that, whereas the Mg content was strongly affected by the alteration process, the trivalent cations distribution was conserved from the pristine spinel compositions, demonstrating the metasomatic/recrystallization nature of the process, as far as the spinels are concerned.

Discussion

The discussion is too long and may be shortened. Moreover, it must be split in order to correctly account the magmatic (sensu lato, including possible HT subsolidus processes) and the alteration processes. In your text, you alternate the two aspects; for example, L652-sq, following an interesting discussion of the LT alteration, are dealing with the control by the melt composition, with pertinent considerations, but evidently relative to before alteration times. This mixing up is rending the reading difficult, and weakens the message(s). A complete rewriting is necessary. The more, if it is considered that you neglected to address the source of the fluids responsible for the alteration, and that you also failed to address the question of the PGM mineralization: yet, you have data to discuss the possible transition from magmatic sulphide-PGM assemblages to the present state of sulpho-arsenide PGM in relation with the LT transformation of the Cr-spinels and the silicate alteration.

Textural evidence for Cr-spinels

The discussion here is very confusing. You seem prone everything and its opposite: magmatic and non-magmatic, melt and mush … Your nice observations deserve better!

I think that a pivotal point for the interpretation is the presence of textures which are necessarily magmatic, I mean the doleritic texture, and point to the existence of melts at some time in the ore history. Another pivotal point is the evidence for strong interaction with the wall rocks, and for multiple melt injections (e.g., Fig. 4-B, D).

On this basis, it may be proposed that the observed textures are consistent with a magmatic origin of the spinel, provided that a mechanism exists that can produce sub-instantaneous spinel oversaturation in the melt(s), leading to the sub-instantaneous formation of a great number of pristine euhedral spinels floating in the (possibly turbulent) melt. This may be likely related to interaction with the surrounding rocks, as proposed by some of the models you referred to in the preceding §. It is in my opinion difficult (if not impossible) to explain the “circle” features otherwise as them recording the aggregation of spinels around small silicate crystals floating in the melt. The main feature of the spinel distributions (clouds and lines without relationships with the silicate framework) are effectively well interpretable in terms of crystals floating in a rapidly crystallizing melt. You have indeed many observations pointing to such quick crystallization (the doleritic textures, in particular). The dense clusters with neat boundaries could represent the reworking of early spinel cumulates, likely due more to filter-pressing processes than to gravitational processes: the plagioclase cumulates and or the “picritic” compositions could also be representative of such processes, which are likely in the context of proximity to the wall rocks and multiple injections that seem to characterize your rock assemblages. Exchange with the local melt composition and/or HT subsolidus re-equilibrating would explain the primary variability in composition of the spinels. Thus, final evolution in a crystal mush context is a distinct possibility.

I evidently strongly suggest that this § should be rewritten in this perspective, and that possible disagreement of yours (or even opposite conclusions) be argued in relation with the above proposal.

• L564: again, I am unable to see the “suspended character of even the densest Cr-spinel disseminations” in Fig. 4-C?
• L608-610: “environment that the rocks formed in was initially chemically and rheologically heterogeneous. Under these conditions, different domains, now forming patchy textures, behaved as sub-systems with limited interaction between each other”. These sentences are basically equivalent to the proposal above, being however too much imprecise and allusive …

Re-equilibration

The discussion here is too long, and I am not sure that it demands a full §: as correctly stated in L611, the models much be supported by the mineralogical and chemical evidence, which consequently must be delivered while discussing the model, or be presented before: not after!

In addition, the § deals with both the pristine compositions (cogently arguing that re-equilibration at a late stage or HT subsolidus was responsible for the observed compositions in less altered settings) and the subsequent alteration. Yet, the § is supposed to support the model of formation of the spinel and their silicate environment.

I agree with your interpretation of the alteration features.

• L647: the “gains” in Ti could be viewed as a mere result of the Al loss, without input of Ti.
• L669: “the hypothesis of an initially heterogeneous environment”; this imprecise formulation evidently refers to the fact that the group of rocks you have studied did not represent one single melt batch: from your own descriptions, it is not a “hypothesis” but an obvious fact. Indeed, it would have been better (and closer to the reality) in the previous sections to take the problem from this angle, starting from the variety of studied lithotypes. Not having addressed the petrology of your marginal rock system is obviously an error in this perspective.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

A great paper and congratulations!

Please note that in line 99 "orthomagmatic" has a typo.