Thermodynamic Constraints on REE Mineral Paragenesis in the Bayan Obo REE-Nb-Fe Deposit, China
Round 1
Reviewer 1 Report
This is a nice study of rare earths (RE) minerals formation by hydrothermal processes simulating the paragenensis in a specific location in Bayan Obo, China, which is based on thermodynamic modelling combined with SEM-EDS analysis of the minerals under study. The authors made a considerable effort to collect the relevant data and extent the respective databases to properly simulate the precipitation processes under variable conditions (T,p, compositions) of the source hydrothermal fluids. As such, the study is worth to be published in Minerals. I would just appreciate a brief description of how the aqueous phase has been modelled. My second concern is the modelling of solid phases containing RE which form solid solutions. However, in the present study bastnaesite and parisite are described just as specific fixed compositions of the respective solution phases, while monazite and fluorocerite are considered as different stoichiometric phases of the corresponding RE. Why the individual LnPO4 and LnF3 were not leastwise considered as end members of ideal solution phases? This should be at least mentioned in the Methods section or in the Discussion.
Although I am not a native speaker, the English language should be improved in my view. For instance, I would prefer a form "Replacement of st1 with st2" rather than "Replacement of st2 to st1". This is at least the result of Google translator from my language.
Author Response
Response to Reviewer 1 Comments
This is a nice study of rare earths (RE) minerals formation by hydrothermal processes simulating the paragenensis in a specific location in Bayan Obo, China, which is based on thermodynamic modelling combined with SEM-EDS analysis of the minerals under study. The authors made a considerable effort to collect the relevant data and extent the respective databases to properly simulate the precipitation processes under variable conditions (T,p, compositions) of the source hydrothermal fluids. As such, the study is worth to be published in Minerals.
Point 1: I would just appreciate a brief description of how the aqueous phase has been modelled. My second concern is the modelling of solid phases containing RE which form solid solutions. However, in the present study bastnaesite and parisite are described just as specific fixed compositions of the respective solution phases, while monazite and fluorocerite are considered as different stoichiometric phases of the corresponding RE. Why the individual LnPO4 and LnF3 were not leastwise considered as end members of ideal solution phases? This should be at least mentioned in the Methods section or in the Discussion.
Response 1: Take calculation in Figure 8 as an example, there are 104 kinds of aqueous species in sum calculated in the modelling by GEMS (Figure 1). Once independent composition of the system has been set (IComp. in Figure 8 includes C, Ca, Ce, Cl, F, H, La, Mg, Nd, P, O, Pr), all possible aqueous species containing these elements would be taken into calculation. The composition of aqueous species at equilibrium state was obtained when the Gibbs Energy of all species reach minimum value. If we conduct the species calculation manually, over 10 species in sum would be rather complex, especially considering calculation under different P and T. Since there are over 100 species in the hydrothermal solution, and their thermodynamic properties have been embedded in the MINEs database, mostly cited from the well-known and public SUPCRT 92 database, we didn’t provide property list to aqueous species, and there is no supplementary table of varing composition of these species in each step of addition. The data amount was enormous.
Figure 1 All calculated aqueous species in Figure 8 of the manuscript
（Figure 1 in the attached file）
As for your second concern, this is limited by the measurement of standard thermodynamic properties: the properties of REE-phosphates and REE-fluocerites were measured on artificially synthetic phosphate or fluocerite with end composition of each light lanthanides; those of bastnaesite and parisite were measured on corresponding natural mineral with complex but fixed composition. And there is only one paper yet that was able to provide properties of these fluorocarbonate, which means we have no alternative data source. However, the fluorocarbonate measured by Gysi (2015) has rather similar composition to bastnaesite and parisite at Bayan Obo, which is extremely Ce-enriched. Therefore, it is reasonable that the calculations of bastnaesite and parisite in this study would simulate the real REE precipitation process at Bayan Obo. For the same reason, there is no properties of natural monazite (solid solution) provided in this study. But as your kind suggestion, in the modified manuscript, the LnPO4 and LnF3 as end member of ideal solution phase would be stressed.
Actually, in the manuscript, we mentioned if the CePO4 was used up in the hydrothermal solution, the formation of bastnaesite and parisite would be ended. Because all these REE minerals are most enriched in Ce. And even though the LaPO4, PrPO4 and NdPO4 exist when all CePO4 was consumed, we describe “the replacement of monazite by other minerals” as “the complete replacement of CePO4 by other minerals”. It is because the initial amount of other LREE was set as equal values to Ce, which are much more than they really are in natural hydrothermal fluids measured by Buhn (1999).
Generally, if we provide thermodynamic behaviour of end members, or solid solution with composition close to the real condition, the precipitation process would be predictable. In the condition that calculated solid solution has very different composition from the real natural samples, the modelling would be problematic and suspicious.
Point 2: Although I am not a native speaker, the English language should be improved in my view. For instance, I would prefer a form "Replacement of st1 with st2" rather than "Replacement of st2 to st1". This is at least the result of Google translator from my language.
Response 2: Thanks for your suggestion. We are sorry for the inappropriate expression in the last version of manuscript. The usage of “replacement of mineral1 to mineral2” has been completely changed into “replacement of mineral1 by mineral2”.
Author Response File: 
Author Response.docx
Reviewer 2 Report
I am not a native speaker; however I believe that some common sentences are wrong:
You always write it in an indirect form:
Replacement of bastnaesite to monazite but it should mean replacement of monazite by bastnaesite. You use this style very often-please rephrase.
I would add in Figure 1 in the legend the color of the open pits (grey). And please Kuangu Anticline: do not write the A in bold
Author Response
Response to Reviewer 2 Comments
Point 1: I am not a native speaker; however I believe that some common sentences are wrong: You always write it in an indirect form:
Replacement of bastnaesite to monazite but it should mean replacement of monazite by bastnaesite. You use this style very often-please rephrase.
Response: Sorry for the inappropriate phrasing. All of the “replacement of mineral1 to mineral2” has been modified into the “replacement of mineral1 by mineral2”, according to your kind suggestion.
Point 2: I would add in Figure 1 in the legend the color of the open pits (grey). And please Kuangu Anticline: do not write the A in bold
Response: Thanks. The legend of open pits has been added in Fig.1. And the “Kuangou Anticline” in the former version of manuscript has been modified into consistent format with other location names, which is written in completely bold form “Kuangou Anticline”.
Author Response File: Author Response.docx
