7 IOA Deposits - REE fractionation and apatite vein replacement reaction

Iron Oxide-Apatite (IOA) systems (e.g. Kiruna-type) contain fluids rich in rare earth elements (REE) reacting with preexisting magmatic apatite and magnetite, triggering a destructive replacement texturally visible in ore deposits where fluorapatite is replaced by monazite/xenotime overgrowths and magnetite oxidizes/re-equilibrates to hematite. Figure xyz shows how this typical reaction would look like in a geochemical model. We are essentially looking at an irreversible reaction of the type:

\[\text{Apatite} + \text{Magnetite} \rightarrow \text{Monazite} + \text{Xenotime} + \text{Hematite}\]

Figure 7.1: (left) Conceptual model of a single flow-through reactor representing a multipass leaching model of an IOA deposit. (right) The simulated reaction of apatite and magnetite to monazite and hematite using GEMS and the MINES database. From: Gysi, A.P., Hofstra, A.H., Harlov, D.E., Miron, G.D. (2019) Rare earth element (REE) metasomatism in iron-oxide-apatite mineral deposits: stability of hydrothermal monazite and xenotime. AGU Fall Meeting Abstracts, 2019, V33C-0244).

Monazite hosts the light REE and xenotime the heavy REE, and the fractionation of these two minerals is a key to understand the geochemical evolution of the ore-forming fluids. In this module we will learn how to set up a multipass leaching model using a Process Simulation representing an open vein system akin a reactive transport model but focusing on one box and without considering hydrologic parameters. The latter is a more advanced topic and will be covered in a future tutorial.

First we create the parent system in SysEq and use the ‘R mode’ in Process simulations to compute a multipass single flow-through reactor model. The case study involves calculation of the alteration assemblages and monitor the fluid REE concentrations to learn how to plot a sum of elements, prepare a REE-rich fluid as PCO in Thermodynamic database mode. We will also learn how to further analyze the model and create additional fluid compositions with varying initial pH.

7.1 Create a parent record in the Ca-Fe-P-O-H-Cl-F-REE system

The first step is to setup a parent system record in System Equilibrium calculation mode (SysEq) using the project Module 7 and calculate equlibrium to check the mineralogy, pH and aqueous speciation.

• Select the SysEq option in the left panel and create and select ‘Create a new record from scratch’. We will create a new record using a temperature of 100 \(^\circ\)C and a pressure of 1000 bar. Let’s call the record Fl-Rock_IOA

• For the input recipe we need to add our IOA deposit rock (60g Magnetite, 40g Fluorapatite), and our REE-bearing saline acidic fluid, see Fig. 7.2. The fluid recipe is as follows:

  • 1 kg H₂O
  • 1.0 M NaCl
  • 0.01 m HCl (pH ~2.0)
  • 100 ppm of each REE.

• Now calculate equilibrium and check the results. Fig. 7.3 shows the calculated pH and monazite/xenotime ideal solid solution compositions.

Figure 7.2: SysEq input for a new record and the fluid and rock compositions. This dialog can be reached by clicking on the yellow vial, make sure to select the correct property; for example use DComp to select magnetite and apatite then add the quantity and units; select IComp for individual REE and choose y for mg (ppm = mg/kg); choose Compos for custom records like Acid_HCl, H2O, and NaCl.

Figure 7.3: SysEq equilibrium calculation ouputs showing the record created, the yellow vial (Input recipe), and the results window with the compositions of ideal monazite and xenotime solid solutions, stable phases, and at the bottom pH, P-T, pe, and ionic strength.

7.2 Create our rock and fluid PCO in Thermodynamic database mode

• Switch to Thermodynamic database mode, select the Compos option and create a new record from scratch (Fig. 7.4). We will call it Rock_IOA.
• Fig. 7.5 shows the PCO wizard, make sure to select DComp, then the elements (IComp), then the minerals (magnetite and apatite) that are in the rock.

Figure 7.4: Compos input for adding a new rock or flud record in Thermodynamic database mode. This dialog can be reached by clicking on Compos module on the left panel.

Figure 7.5: Composrecord wizard. Choose DComp if you want to add a dependent component (compound like a mineral) as input recipt; choose IComp if you want to add individual elements IComp as inputs (this can always be ticked just make sure to enter zeros if you don’t want it to be taken into account in your input recipt); choose the number of user defined formula if you want to enter your own formula (e.g. SiO2, HCl, etc.). The next window is self explanatory, just make sure to select DComp or IComp as needed.

• Fig. 7.6 shows the resulting PCO for the rock and all the steps. In the Settings tab you will see either the DComp chosen or if selected, the user-defined formula can be entered on the blank lines. Once you have entered units and quantities, move to Page 1 tab. Here you will make sure that the total rock is normalized to 100g (add 0.1 kg), enter the name and information about the record, switch off/zero the IComp values to avoid adding additional elements to the input recipe, then click re-calculate to check the total composition. Finally click Save to save the record.

• The same procedure is used to create a PCO for the fluid, which we will call Fluid_IOA. Select the Compos option and create a new PCO record from scratch (Fig. 7.7). Make sure to select all of the REE elements as IComp and the compounds HCl, NaCl, and H2O from the DComp selection window.

• Fig. 7.8 shows the resulting PCO for the fluid and all the steps. In the Settings tab you will see the DComp chosen together with units and quantities. On the Page 1 tab make sure to enter the REE as IComp (100 ppm each), and put a zero in the normalization mass to let GEMS calculate the total mass of fluid, which should appear once you click on re-calculate. Finally click Save to save the record.

  

  Figure 7.6: Composrecord. Start with the Settings tab to input your DComp or user-defined formula for your minerals or fluid compounds and add the units (1-2). In the Page 1 tab follow the sequence: (3) enter record information; (4) enter the normalization mass (i.e., 0.1 kg); (5) make sure the IComp values are zero unless additional elements need to be added to the recipe; (6) click the re-calculate button to check the total composition shown in window (7).

Figure 7.7: Composrecord wizard used to create the IOA fluid. Make sure to select DComp in the first window to see the selected compounds for the fluid (i.e. HCl@, NaCl@, H2O@, O2@).

Figure 7.8: Composrecord showing the selected compounds for the fluid (i.e. HCl@, NaCl@, H2O@, O2@) in the Settings tab (1-2), and the IComp values in the Page 1 tab (3-4). Make sure to add 100 ppm for each REE (4), input a zero in step (5), re-calculate in step (6) and inspect the results.

7.3 Setting up a multipass leaching model in Process simulations

Here we will create a multipass leaching model using a single flow-through reactor in Process simulations. But before we start we need to create a parent record in SysEq and add the two PCOs we created above to our system. We will make two parent records with 100 g IOA rock and 1000 g of fluid, one record at 100 °C and the other at 350 °C and 1000 bar.

• Switch back to System Equilibrium calculation mode (SysEq) to add the two PCOs to our system and create the parent record. You already learned how to do this in the core tutorials, but Fig. 7.9 shows what to select. Let’s create a record called FR_multipass at 100 °C and 1000 bar, and select the two PCOs we created above (i.e. Rock_IOA and Fluid_IOA) as shown in Fig. 7.9. Click ok and calculate equilibrium to check the results. You can clone the record to create another parent system at 350 °C and 1000 bar.

Figure 7.9: Equilibrium Calculation mode (SysEq) showing how to create a parent record in preparation forthe multipass leaching model, and where to enter the rock and fluid PCOs created in the previous step. The resulting record is called FR_multipass and is calculated at 100 °C and 1000 bar. Note the initial fluid/rock ratio and amount of fluid given here will be used in the Process simulation; for advanced users it can also be changed in the Process simulation code later on. Make sure to always add a comment and title for each SysEq record created, it will help you to keep track of your records.

• Switch to Process simulation mode and create a new record from scratch. We can call it FR_multipass and select the Process simulation code (R) for the type of process to simulate as shown in Fig. 7.10. In the next wizard step, make sure to select the sequence shown in Fig. 7.11. First create the record at 100 °C and 1000 bar. Then repeat the process above to create another Process simulation record at 350 °C and 1000 bar.

• In the plotting window in Fig. 7.11 you can select the property to plot on the x- and y-axis. Here we select phM for phase mass in grams, and under Scalars select J which is used for fluid aliquots (right click and select J so that it appears as xp[J], i.e. x-axis value).

• Accept all the following dialogues. Then click on Save this record to database, which creates your new process simulation record.

• Under Process simulation when you select your record, under the Controls tab make sure your setup is correct, including temperature, pressure, ‘iTm’ (i.e. each record will be assigned a new number, from 1000-1200 in 10 steps means a maximum of 20 records will be generated starting with 1000, 1010, and so forth). In the Sampling tab make sure you have xp[J] =: J which represents the x-axis, and a series of variables such as yp[J][0] =: phM[{Apatite}] which represents what to plot on the y-axis.

Figure 7.10: Process simulation mode showing how to create a new record from scratch and select the Process simulation code (R) for multipass leaching model. The resulting record is called FR_multipass_FR10.

Figure 7.11: Process simulation wizard to setup a mutlipass leaching model. (1) Select Leaching Compos source, (2) select the fluid, (3) select the rock, (4) enter the P-T conditions or steps if you want to change pressure-temperature. In the iTm iterator, one defines the fluid aliquots, for example for record values from 1000 to 1200 in 10 steps results in a total of 20 fluid aliquots (i.e. the first record will be assigned 1000, the second one 1010, etc.). In the plotting window under Scalars one finds J which is used for flud aliquots; other variables that can be selected there are for example temperature, pH etc. You can select the Property in the left panel, here we choose phM which is phase mass in grams.

7.4 Plotting and model modifications for more acidic fluid

• Click on the calculator icon Re-calculate and check record data to display the graph of the process simulation. Figs. 7.12-7.13 show the resulting calculated alteration mineralogy and fluid REE concentrations for the two parent records at 100 °C and 350 °C. Note that the fluid REE concentrations are plotted as a sum of all REE (which can be defined in the Sampling tab code), but you can also select individual REE to plot from the wizard.

• To make things interesting go back to the Thermodynamic database mode and create a new fluid PCO with more acidic pH (i.e. 1 M HCl instead of 0.1 M HCl) and repeate the steps above. Make a parent record with fluid and rock created in SysEq, then in Process simulation use these two parent records to do the multipass simulations at 100 °C and 350 °C. Fig. 7.12 shows the additional results.

• Fig. 7.13 shows the type of additional plots you can make. In this case we plot log molality of light (L) and heavy (H) REE in the fluid as a function of fluid aliquots. Note that the light REE are more mobile than the heavy REE at high temperature, and that pH plays a crucial role for REE mobilization. Here we use J for aliquots of fluid on the x-axis, and log(m_t[{element}]) for the y-axis, where element is the selected element. You can select individual REE to plot as well, or sum of all light and heavy REE as shown here. For individual elements remake the record and use the wizard. You can modify the plotting script in the Samplingtab according to Fig. 7.14.

• Here is the plotting code that can be copy pasted:

yp[J][0] =: lg(m_t[{La}]+m_t[{Ce}]+m_t[{Pr}]+m_t[{Nd}]+m_t[{Sm}]+m_t[{Eu}]+m_t[{Gd}]); 
yp[J][1] =: lg(m_t[{Tb}]+m_t[{Dy}]+m_t[{Ho}]+m_t[{Er}]+m_t[{Tm}]+m_t[{Yb}]+m_t[{Lu}]);

• Can you make such models? If this part became too advanced, I suggest to go back to the core tutorials and repeat the steps above for the Greisen model. Once you are familiar with the SysEq and Process simulation mode, the plotting scripts, and adding PCO records to the thermodynamic databse under Compos you can try to make your own models here again.

Figure 7.12: Simulated results for single flow-through reactor multipass leaching models at 100 °C and 350 °C, 1 kbar, showing the simulated alteration mineralogy for a fluid with initially 0.1 m HCl (left) and 1 m HCl (right). The models nicely show the replacement reactions between apatite and REE phosphates, and magnetite and hematite. pH plays a crucial role for REE mobilization.

Figure 7.13: Simulated results for single flow-through reactor multipass leaching models at 100 °C and 350 °C, 1 kbar, showing the simulated log molality light (L) and heavy (H) REE for a fluid with 1 m HCl at 100 °C (top) and 350 °C (bottom), 1 kbar. The models nicely show that pH plays a crucial role for REE mobilization.

Figure 7.14: Sampling tab in Process simulation mode showing the plotting script for the log molality of light (L) for yp[J][0] and heavy (H) for yp[J][1] rare earth elements in the fluid as a function of fluid aliquots xp[J].

7.5 Outcomes

Congratulations, you know now how to make a vein replacement model in GEMS using the MINES thermodynamic database! Couple of things to think about: - How would the results change if you change the initial fluid composition? - How would the results change if you change the initial fluid/rock ratio? - Why are the light and heavy REE fractionated in the modeled fluid, what is the role of pH and temperature?