Rare Earth Occurrences in Streams of Processing a Phosphate Ore

Round 1

Reviewer 1 Report

Please identify the objective of the characterization of the samples. Is it to look at processes to recover the REE?If so, iIn which products?

The authors should  provide some information concerning the method used  to produce the apatite concentrate or at least  identify the mine that produces it, so it would be possible to assess if the apatite concentration method influences the recovery of REE.

Author Response

A Word file has been attached. 

Author Response File: Author Response.docx

Reviewer 2 Report

I have reviewed the manuscript “Rare earth occurrences in phosphate rock processing” by

I have reviewed the manuscript “Rare earth occurrences in phosphate rock processing” by Xiao Sheng Yang et al. submitted to the journal Minerals. The manuscript presents the results of investigations of the waste products generated during processing of apatite ore. The studies have been carried out using the techniques of Inductively Coupled Mass Spectrometry (ICP-MS), Mineral Liberation Analysis (MLA) and electron microprobe (EPMA) in order to estimate the concentrations of rare earth elements (REE = Lanthanides + Y + Sc) and identify REE-bearing mineral phases in the ore, the flotation concentrate, flotation tailings and phosphogypsum. The obtained results are discussed mainly in the light of the relative changes in the REE contents and modes of occurrence in the ore and different products and wastes generated at each step of the ore processing treatment.

In my opinion the obtained results are interesting and the scientific problem dealt with will be of interest for a wider audience. However, I feel that the present version of the manuscript needs substantial changes before publishing. Below are my comments grouped into general and more detailed comments.

General comments:

1. I recommend the change of the title as “Rare Earth Occurrences in Phosphate Rock Processing” seems to be linguistically incorrect. Besides, the contribution is a case study (the research was conducted on a specific material) and not a broad review as suggested by a very general sense of the title.

 

2. I recommend additional references to be used in the introductory and discussion chapters:

Binnemans, K., Jones, P.T., Blanpain, B., Van Gerven, T., Pontikes, Y.: Towards zero-waste valorisation of rare-earth-containing industrial process residues: a critical review. J. Clean. Prod. 99, 17–38 (2015)

Grabas, K., Pawelczyk, A., Strek, W. et al.: Study on the Properties of Waste Apatite Phosphogypsum as a Raw Material of Prospective Applications .Waste Biomass Valor. (2018). https://doi.org/10.1007/s12649-018-0316-8

 

3. It is not clear what is understood by rare earth elements (REEs). The Table 1 suggests that REE are lanthanides with Y and Sc. If that is correct, please state it clearly in the abstract and the text the first time the acronym appears. Also the expressions “rare element oxides (REO) distribution” and similar (e.g. lines 8, 15-16, 102 and others) is unfortunate. REEs can be distributed but REOs are just a measure of the REEs concentrations. I also suggest using the plural form of “REEs” when all these elements are collectively used in the text.

 

4. There is no information what the investigated ore is, except that it mainly built of silicates (Table 3). Some details on the rock type, petrography and location are necessary. “A phosphate mine in Europe” (lines 59-60) is a very vague statement. If possible, some details on the ore processing technology that resulted in the formation of the investigated concentrate and phosphogypsum as well as the commercial use of the end product should be provided.

 

5. Limits of detection (LODs) should be given for ICP-MS and EPMA measurements in the “Materials and methods” section. When an element was below LOD, it should be marked so in the appropriate tables, e.g. by marking it as n.d. (not detected). It should also be mentioned if some elements were consistently below LODs for any of the method applied. In some cases the SEM technique is mentioned further in the text (e.g. line 155) but neither this method nor the purpose of using it is mentioned in this section. The name and version of software used for MLA measurements and calculations should be given.

 

6. The chapter “Materials and methods” gives no explanation of the overall analytical strategy. The aims of each analytical step should be specified maybe with a figure presenting a flowchart for the whole analytical operation. See also the next comment.

 

7. Identification of minerals in the studied samples is not supported by any evidence like complete analyses of chemical compositions, electron microscope images with EDS spectra etc. Was it done basing on the formula calculations from EPMA data? If so, what were the calculation strategies? While it can be straightforward in case of apatites, calcite, dolomite etc., the procedure for calculating pyrochlore- and aeschynite-group minerals is quite challenging and without good evidence must be considered dubious. Besides, mineralogical nomenclature used in the manuscript must be tidied up. The Authors mix the names for mineral species with those for mineral groups and/or supergroups, e.g. the names such as apatite, monazite, allanite, aeschynite, britholite are names of mineral groups and pyrochlore may refer to a group or even a supergroup. Although for the purpose of this study precise distinction between for instance different minerals of the aeschynite-group is neither important nor necessary, it should be clearly stated what the taxonomic names really stand for. Perhaps in some cases it will be enough to explain that for instance the name aeschynite stands for mineral phases with stoichiometry of the aeschynite-group minerals.

 

8. Scandium was detected in the bulk samples (Table 1). Why it has not been detected by EPMA? Was it not measured at all?

 

9. Lines 110-112. While the total concentrations of REEs in the ore and flotation concentrate measured for bulk samples (ICP-MS method) and from combined MLA and EPMA analyses are in a good agreement, the same is not entirely true when individual REEs are considered. For instance the results obtained by two methods for Nd, Sm, Gd and Dy show differences that should be discussed.

 

10. Lines 142-182. I do not see the purpose of the two separate subchapters entitled “Modal Mineralogy” and “Determination of all the REE-bearing phases (...)”. How can modal mineralogy be estimated without determining mineral phases first? The Authors should consider merging the two subchapters. Also the terms “mineral-REEmineral-mix”, “miner-REE-mix”, “Ca-P-F-phase”, “Ca-P-F-compound” and “Fe-Al-Si-P-compound” are confusing and used inconsistently. Perhaps acronymic names (e.g. CPF mineral for an unidentified mineral that contains Ca, P and F) could be used after explaining clearly what they mean? The distinction should also be made between a mineral that has not been identified but was found to have certain elements (e.g. Ca, P and F) and polymineral aggregate composed of identified minerals.

 

11. Figure 1. It would be more informative with the EDS spectrum of the REE-bearing phase added.

 

12. Aeschynite-like mineral was detected in phosphogypsum but not in the ore. That should be explained.

 

13. In my opinion the discussion does not utilize the full potential of the obtained results and is dominated by information provided by the literature. For instance I could not find the conclusion if the studied material offers any perspective for commercial recovery of REEs. Although various techniques of REE recovery are cited, the Authors do not propose a strategy that might be advantageous in the studied case. The discussion does not address the problem of accurate estimation of REE concentrations in different products and wastes generated during the processing of apatite ores. The Authors observe differences in the results obtained using only MLA (often used by industry) and by combination MLA with EPMA (rather restricted to academic users) but do not propose the best analytical strategy for commercial purposes.

 

13. According to the Minerals’ Instructions for authors, abbreviations should be defined in parentheses the first time they appear in the abstract, main text, and in figure or table captions and in many cases they are not, e.g. lines 8-9, 28, 43, 55 and others. References must be numbered in order of appearance in the manuscript not in the alphabetical order. Some journal names of the cited references are not abbreviated.

 

14. Graphic presentation of the obtained data is poor what negatively influences readability. The manuscript contains 14 Tables and only two figures (one of them gives the SEM image)! I would suggest to reconsider whether all the tables are necessary (perhaps some can be merged or moved to supplementary sections or even omitted at all) and to introduce more diagrams and graphs in such a way to make the paper more readable.

 

15. Perhaps, after modifications of the contribution, the abstract could need some changes.

 

Specific comments

 

Lines 12-13 and 16-17 contain similar information on the contents

Lines 63-64. Where were the ICP-MS analyses performed?

Lines 65-67. What software was used for MLA calculations?

Line 75 onwards. Some other expression should be used instead of “REE compositions”. Maybe “individual REEs concentrations” or something alike would be better?

Line 77. “All the REE except Pm exist” should be changed to “All the REEs except Pm were detected”

Line 83. The method of measurement should be added into the table caption.

Lines 93-94 and Table 3. It would be better to state what minerals were taken into account in the MLA analyses and which of them were found to contain measurable amounts of REEs and separate the two groups in the table.

Line 108. Do the Authors mean wt%?

Tables 4-8, 11-13. The elements are listed in a strange order. They should be rearranged as in Table 1. It should also be clearly stated which elements have been below detection limits. For tables with EPMA results it should be clearly stated if they contain average contents and from how many analyses the average data were calculated.

Line 126 and Tables 7-8. Units should be given. The caption should be more informative. As they are now, the tables may give the misleading impression that they present REE contents in mineral phases. The columns with totals are redundant since the data have been normalized.

Line 132. Why decimals (as zero) are given in the text for numbers which are clearly rounded to whole units?

Line 137 and Table 9. The caption is the same as for Table 8. The last column should be renamed as it is not clear what “total” means in this context.

Line 138. The wet processing is not a subject discussed in this subchapter so it should be removed from the title.

Line 140. “Reported” is obviously not the right word in this context.

Line 144. “Dominated” should be changed to “dominating”. Bracketed numbers can be removed because the average number is given.

Table 10. Mineral phases should be grouped into REE-bearing and REE-free ones. Celestine is given two times with different abundances. Does the “Average” column refer to the concentrations in the bulk samples, i.e. not divided into grain-size fractions? Otherwise what is the purpose of calculating average values from different grain-size fractions? Some data, concerning REE-free mineral phases that are presented in the table are not given in the text. What is the last meaningful decimal place in these calculations? There are numbers given in the table with four, three and two decimal places.

Lines 155-156. The sentence “Meanwhile, other REE-bearing phases were found by SEM but they were not identified clear minerals, thus called mineral-REEmineral-mix” is not clear.

Lines 162-163. The discrepancy between the results obtained by MLA and EPMA methods should be discussed in the “Discussion” section.

Table 11 and 12. What is the last meaningful decimal place for the data presented? Do empty fields for certain elements indicate that they were not measured?

Lines 184-185. Were the experiments carried out by the Authors (not indicated in the Methods section)? If not, proper citation should be given. What does the value of 128% mean?

Lines 191-194. That part belongs rather to the Discussion section.

Lines 196-197. “From the analysis above it is seen that in the beneficiation process REO occur in the phases of apatite, allanite, monazite, zircon and pyrochlore.” They are REE-bearing minerals that occur in the ore (Table 3) so why a special analysis was needed to arrive at that conclusion?

Author Response

A Word file has been attached. 

Author Response File: Author Response.docx

Reviewer 3 Report

The reviewed manuscript is basically a technological report. I regret to tell the authors that I do not see too much scientific value in it. It is also obvious that the authors have problems with mineralogy and understanding of minerals. Hence, it is really surprising that the authors decided to submit their ms to a journal called Minerals. The authors have not convinced me that they are actually dealing with the phases they report. There is no single mineral analysis presented. Thus, it is even impossible to judge whether "Ca-P-F-compound", "Clay-like" phase or "REEmineral-mix" are really unidentified phases or not (just to mention one of many problems of this text). Therefore, I suggest the authors proper re-writing of this text, presenting of all data they posses and most importantly formulating a research hyopthesis. 

Author Response

A Word file has been attached. 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear Authors,

I am glad to confirm that nearly all of my doubts have been satisfactorily removed. The result looks much better and the text carries some new information and ideas. Well done.

However two issues still need to be addressed.

1. (old point 7). As concerns mineral identification I cannot fully agree with your proposition. It is not possible to identify precisely and unequivocally such chemically complicated and typically partly or completely metamict minerals as aeschynites, pyrochlores and britholites using just semi-quantitative EDS method. It is sometimes a very challenging task even with fully quantitative WDS approach. Therefore I propose to replace the text in lines 107-109 by something like (or some variation of this):
"It should be stressed that because the mineral identification was carried out by semi-quantitative EDS method, phases with complex chemical compositions such as allanite, aeschynite, britholite and pyrochlore should in  fact be referred to as phases with allanite-like, aeschynite-like, etc, compositions. However, although identification of such phases cannot be treated as entirely accurate, for the sake of clarity and readability of the text simplified and more general names will be used."

2. (old point 8), i.e. "the problem with scandium". My question was: "Scandium was detected in the bulk samples (Table 1). Why it has not been detected by EPMA? Was it not measured at all". Your answer is: "It has been detected by EPMA, such as for zircon, pyrochlore, but it is below the LOD."
I find your answer self-contradicting. If something is below LOD, it is regarded as not detected. To clarify my point - there are three possibilities when we deal with quantitative results:
- the result is below detection limit = we cannot be sure if an element is present in the sample,
- the result is above limit of quantification (LOQ) = the element was measured and the number can be used for further calculations; LOQ depends on the researcher's choice (if we want to be very strict with numbers or more lenient) but is typically LOQ=LOD+2*sigma or LOD+3*sigma, where sigma is a standard deviation,
- the result is between LOD and LOQ = we are sure the element is there but we cannot trust the exact numbers; That is usually noted as trace amounts without giving exact numbers in tables.

In your case: (i) from Table 4 it is clear that Sc is present in a mineral that goes into flotation concentrate; (ii) Sc is detected by ICP MS, the method with low LOD but is not by EPMA, the technique with higher LOD; (iii) zircon might be the Sc-bearing phase but it has not been proven by your results; (iv) any other phase, even those that have not been identified can be a carrier of Sc (you may want to take a look at a paper "Scandium of the Kovdor magnetite-apatite-baddeleyite deposit (Murmansk Region, Russia): Mineralogy, spatial distribution, and potential resource  with doi: 10.1016/j.oregeorev.2015.08.017. The paper concerns the presence of Sc in various minerals in the carbonatitic ore from Kovdor.
Anyway, I think you should add a sentence or two to comment that problem, because Sc is one of REEs as you define them and it is a very important and sought-after element.

I also put some minor, mostly editorial remarks on the manuscript (attached). But, although I am not an English native speaker, I need to stress that the language of your manuscript definitely requires a thorough review.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

A point-by-point response to your comments has been attached. Thank you very much for your time to review our manuscript. Your comments are really very helpful for improving the quality of our paper. 

Author Response File: Author Response.docx

Reviewer 3 Report

No other comments.

Author Response

Dear Reviewer,

Thank you very much for your time to review our manuscript. We have asked one of our colleagues, an English native speaker, to review our paper.