News:

alphaMELTS 1.9 is available at the download and information site.
alphaMELTS 2 standalone & for MATLAB/Python are available on request
(see the Version 2 tab on the download site for more details).
For news of all MELTS software see the MELTS Facebook page.

Main Menu

Modelling liquid immiscibility

Started by Bhuvan, January 10, 2020, 07:10:34 AM

Previous topic - Next topic

Bhuvan

Hi,
I am trying to understand the possibility of a silicate-silicate liquid immiscibility in the Skaergard intrusion. I'd like to begin with a melt of basaltic composition and fractionate olivine, clinopyroxene and plagioclase (Ã,±orthopyroxene) till the residual liquid has 10% of the initial mass. I would like to then simulate a liquid immiscibility leading to a SiO2 enriched melt and a ferrobasalt. I tried doing it by using both rhyoliteMELTS and pMELTS, at 2-5kbar pressures. I end up with a quadratic convergence failure error every time.

Any leads would be highly appreciated.

Paula

What are you doing with fO2 for these calculations? I suggest you distribute the Fe between FeO and Fe2O3 and then make sure the buffer is off for the calculations.

Paula

Bhuvan

I have not put any constraint on the fo2, but I did distribute the Fe between the two oxides. The buffer is off.

asimow

Hi Bhuvan,

Sorry for long delay. I have just looked at this problem. I was able to make it work.

I used the bulk composition recommended by Thy, which is chilled Dike margin Cm from Brooks and Nielsen (1990). Here is my input file (alphaMELTS 2 style):

Title: SkaergaardDikeCmfromBrooksandNielsen(1990)
Initial Composition: SiO2 48.25
Initial Composition: TiO2 2.92
Initial Composition: Al2O3 13.34
Initial Composition: Fe2O3 0.001
Initial Composition: FeO 13.69
Initial Composition: MnO 0.20
Initial Composition: MgO 6.31
Initial Composition: CaO 10.70
Initial Composition: Na2O 2.46
Initial Composition: K2O 0.45
Initial Composition: P2O5 0.29
Initial Temperature: 1200.00
Initial Pressure: 2000.0
log fo2 Path: FMQ
Mode: fractionate solids
Mode: multiple liquids

(For alphaMELTS 1.x, you would put the Mode statements in the .commands file instead of here in the .melts file)

I used rhyoliteMELTS, which found the liquidus at 1189.06 Ã,°C. Then I turned the fO2 buffer off and ran fractional crystallization in 1 Ã,°C steps. As soon as I released the fO2 buffer, the system exsolved a 2nd liquid, basically a fayalite-hematite binary liquid. This persists basically unchanged all the way down temperature to 1121.06 Ã,°C, where a 3rd liquid appears, basically a Ca-phosphate liquid. These three continue to coexist all the way down to 921.06 Ã,°C, at which point the main silicate liquid has evolved to a rhyolite with 74.27% SiO2, still coexisting with the Fe-silicate and Ca-phosphate liquids. Yes, at that point the model crashes because it runs out of MgO, but it has gone far enough for you to study the liquid immiscibility in some detail.

Now, it is somewhat puzzling why turning off the buffer triggers the initial resolution, but you can play with that.

Bhuvan

Dear Asimow,

Thank you for your elaborate reply! I will surely try it this way and let you know of any developments.