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

Trace elements during melt-rock reaction

Started by bertopadron, April 03, 2019, 02:47:39 AM

Previous topic - Next topic

bertopadron

Hi,
I am trying to model trace element enrichment during hydrous metasomatism. All is fine using option 12 but when I try to do same using cycles of continuous melting + "assimilation", major elements are properly mixed but trace elements are directly those of the enrichment file trondh_CV201h.melts in my case (i.e. no mixing). This is actually what is written in the alphamelts guide ("Trace element concentrations should be the bulk assimilant values" page 47), although shortly after it is said that "Major and, if appropriate, trace elements will be mixed (to mix trace elements only use ALPHAMELTS_FLUX_MELTING)". Is there any option to gradually mix trace elements from a single enrichment melts file in each step of interaction?

I am using macosx_alphamelts_1-9.

!Default values of environment variables (version 1.6)
!Variables preceeded by '!' are 'unset' (i.e. 'false')

ALPHAMELTS_VERSION pMELTS
ALPHAMELTS_MODE isothermal
ALPHAMELTS_DELTAP 0
ALPHAMELTS_DELTAT 0
ALPHAMELTS_CONTINUOUS_MELTING true
ALPHAMELTS_MINPHI 0.01
ALPHAMELTS_ASSIMILATE true
ALPHAMELTS_DO_TRACE true
ALPHAMELTS_DO_TRACE_H2O true
ALPHAMELTS_TRACE_INPUT_FILE trace_data.txt
ALPHAMELTS_CELSIUS_OUTPUT true
ALPHAMELTS_SAVE_ALL true
ALPHAMELTS_QUICK_OUTPUT true

batch file
1
McKenzie_ONions_DM.melts
4
50
0
olivine
x
1
0.3
trondh_CV201h.melts
0

Thanks in advance.
José Alberto

Paula

#1
Sorry I didn't see this before. Not quite sure I understand the issue. Are the trace elements exactly the same as the enrichment file (which could be a bug)? Or is it that the trace elements are incompatible and (mostly) being removed with the liquid during continuous melting, so the next step is dominated by the assimilant trace element budget?

The "Trace element concentrations should be the bulk assimilant values" is referring to when you have multiple assimilant files for different minerals (somewhat mimicking the assimilation input in the graphical user interface). In this case the major elements will be different in each file and individual phase properties will be used e.g. calculate the enthalpy of the assemblage for heat-balanced assimilation. As the trace elements do not have any effect on the thermodynamics of melting, they are treated in bulk.

If you want to change the trace elements being assimilated at each step you will need to use multiple (bulk) assimilant files.

Paula

bertopadron

Dear Paula,
Thanks for your reply and apologise for my slow feedback. I was checking the forum after posting the question but then I (incorrectly) thought I was going to get an email when somebody replied. Now I realise that there is an option in the forum for getting an automatic alert when someone replies  ;)  Anyways.

My guess is there must be some kind of bug because:

1) the problem exists even before getting any melt in the system.
2) When I run the same thing manually with only one step (option 12) I get the right mixing between the two melts files.

I can not find the way to attach files in the forum, but I can send them to you if is useful to find the issue.
Thanks again in advance.
Alberto

Paula

I was going to say you need to turn on notifications  ;)

I disabled attachments, as they are a bit of a pain on the server end. You can either copy and paste, or email me (though I'm a bit snowed under this week). Also let me know what you are typing at the menu.

The source mixer routine (option 12) and the 'remixer' (assimilation / flux melting) aren't the same. Anyway, when I get a chance I'll have a look.

Cheers,
Paula

bertopadron

Excellent the alerts makes communication much easier!  ;D

Here is what I typed in macosx_alphamelts_1-9
If you check the "Bulk Trace" in Trace_main_tbl.txt you will see that the traces initially correspond to the trondh_CV201h.melts with no mixing with Balmuccia.melts. It starts to change when the melt starts to be extracted at mass 113.272952 (melting starts shortly before at 112.127001).
I hope my explanation makes sense.
Thank you very much in advance.
Alberto

----------------------------------------
! alphamelts run by josealbertopadron-navarta
! Time: Jun 3 2019 at 09:26
! Command line: -p /Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/output_files -f /Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/alphamelts_default_env.txt -b /Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/auto_batch_continuous_melting_Balmuccia.txt
! Settings file: /Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/alphamelts_default_env.txt
----------------------------------------

----auto_batch_continuous_melting_Balmuccia.txt------
1
/Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/Balmuccia.melts
4
80
0
olivine
x
1
0.3
/Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/trondh_CV201h.melts
0
----------------------------------------

----- alphamelts_default_env.txt ----

ALPHAMELTS_VERSION pMELTS
ALPHAMELTS_MODE isothermal
ALPHAMELTS_DELTAP 0
ALPHAMELTS_DELTAT 0
ALPHAMELTS_CONTINUOUS_MELTING true
ALPHAMELTS_MINPHI 0.01
ALPHAMELTS_ASSIMILATE true
ALPHAMELTS_DO_TRACE true
ALPHAMELTS_DO_TRACE_H2O true
ALPHAMELTS_TRACE_INPUT_FILE /Users/josealbertopadron-navarta/Documents/alphaMELTS/macosx_alphamelts_1-9/finero/input_files/trace_data.txt
ALPHAMELTS_CELSIUS_OUTPUT true
ALPHAMELTS_SAVE_ALL true
ALPHAMELTS_QUICK_OUTPUT true
----------------------------------------

----- Balmuccia.melts ---------------
Title: Average spinel lherzolite from Balmuccia Hartmann and Wedepohl 1993
Initial Composition: SiO2 44.8
Initial Composition: TiO2 0.104
Initial Composition: Al2O3 3.19
Initial Composition: Cr2O3 0.37
Initial Composition: FeO 7.05
Initial Composition: Fe2O3 1.39
Initial Composition: MnO 0.14
Initial Composition: MgO 38.9
Initial Composition: CaO 3.12
Initial Composition: K2O 0.003
Initial Composition: Na2O 0.16
Initial Composition: H2O 0.019
Initial Trace: La 0.12
Initial Trace: Ce 0.40
Initial Trace: Pr 0.131
Initial Trace: Nd 0.45
Initial Trace: Sm 0.19
Initial Trace: Eu 0.092
Initial Trace: Gd 0.33
Initial Trace: Tb 0.08
Initial Trace: Dy 0.45
Initial Trace: Ho 0.122
Initial Trace: Er 0.33
Initial Trace: Tm 0.060
Initial Trace: Yb 0.31
Initial Trace: Lu 0.052
Initial Trace: Cs 0.00132
Initial Trace: Rb 0.017
Initial Trace: K 23.
Initial Trace: Ba 1.10
Initial Trace: Th 0.009
Initial Trace: Pb 0.0232
Initial Trace: U 0.002
Initial Trace: Ta 0.013
Initial Trace: Nb 0.05
Initial Trace: Sr 3.5
Initial Trace: Zr 4.8
Initial Trace: Hf 0.13
Initial Trace: Ti 623.
Initial Trace: Na 1187.
Initial Trace: Y 2.9
Initial Trace: Zn 51.
Initial Trace: Cu 35.
Initial Trace: Li 2.4
Initial Trace: Ga 3.0
Initial Trace: Sc 15.9
Initial Trace: V 83.
Initial Trace: Mn 1053.
Initial Trace: Co 106.
Initial Trace: Cr 2568.
Initial Trace: Ni 2040.
Initial Temperature: 950
Initial Pressure: 12000.00
Log fO2 Path: None

-----------------------------------

---------- trondh_CV201h.melts -------------
Title: Trondhjemite Lazaro et al. 2011 Lithos CV201h
Initial Composition: SiO2 72.7
Initial Composition: TiO2 0.07
Initial Composition: Al2O3 16.4
Initial Composition: FeO 0.53
Initial Composition: Fe2O3 0.01
Initial Composition: MnO 0.01
Initial Composition: MgO 0.31
Initial Composition: CaO 1.73
Initial Composition: Na2O 5.77
Initial Composition: K2O 1.08
Initial Composition: H2O 5.0
Initial Trace: La 1.80
Initial Trace: Ce 2.60
Initial Trace: Pr 0.34
Initial Trace: Nd 1.37
Initial Trace: Sm 0.50
Initial Trace: Eu 1.13
Initial Trace: Gd 0.56
Initial Trace: Tb 0.094
Initial Trace: Dy 0.58
Initial Trace: Ho 0.105
Initial Trace: Er 0.23
Initial Trace: Tm 0.030
Initial Trace: Yb 0.159
Initial Trace: Lu 0.024
Initial Trace: Cs 0.16
Initial Trace: Rb 31
Initial Trace: K 14445
Initial Trace: Ba 4220
Initial Trace: Th 2.6
Initial Trace: Pb 6.7
Initial Trace: U 3.1
Initial Trace: Ta 0.26
Initial Trace: Nb 1.95
Initial Trace: Sr 97
Initial Trace: Zr 71
Initial Trace: Hf 2.6
Initial Trace: Ti 360.
Initial Trace: Na 66840.
Initial Trace: Y 3.3
Initial Trace: Zn 12.4
Initial Trace: Cu 3.3
Initial Trace: Li 4.6
Initial Trace: Ga 16.8
Initial Trace: Sc 0.37
Initial Trace: V 15.4
Initial Trace: Mn 77
Initial Trace: Co 0.99.
Initial Trace: Cr 119.
Initial Trace: Ni 12.9
Initial Temperature: 950.00
Initial Pressure: 12000.00
Log fO2 Path: None


Paula

Hi Alberto,

That is happening because you need to do at least one (batch) calculation with Balmuccia.melts on its own before starting to assimilate, just to populate the trace elements in the system. I suppose it's kind of a bug, but not really something that can be fixed. I can add a warning in future.

Generally it's safer to do option 3 before option 4 for most calculations. The trouble is that when you do that in this case, it fails. You can fix this by treating K2O as a trace element only. It's better to do this for pMELTS calculations anyway because felspar is not always stable and there are relatively few other phases that can take K2O.

Hope that helps,
Paula


bertopadron

Hi Paula,
Thanks for your precise reply. Effectively if I run option 3 first I have to discard K, otherwise it fails. This is quite unfortunate as it is the enrichment in phlogopite due to mixing with the enrichment file what I am interested in. My roundabout for the moment is to recompute trace element concentration based on computed amount of phases using a python script and the right mixing of trace elements between the two melt files.
Please let me know if there is any other alternative.
Thanks again for all your help.
Alberto

Paula

Hi Alberto,

I haven't tried it but you could probably work around it something like this:

  • Call source mixer (option 12) and mix in enough that the bulk composition is what you would expect after one assimilation step.
  • Call Execute (option 4); when you are asked for an enrichment file the first time, give a made up file name that does not exist.
  • When you are asked for an enrichment file the second time, give the correct file name...
The easiest way to make sure Source Mixer gives the same results as one cycle of Remixer would, to add an 'Initial mass:' lines in the melts and enrichment files.

Cheers,
Paula

bertopadron

Hi Paula,
I tried your work around, but it failed. I tried with another input file (more depleted) and the funny thing is that for this case the problem when running option 3 first it was the convergence for OH (comprehensible as the input file was nearly free of H2O). So I deleted the line of H2O in the input file and then run 3 (worked fine!) and then 4 with the enrichment file. The good news is that by doing this way I get the right mixing of the trace elements with the enrichment file and the partition between the phases (very similar to the ones I computed using the python script). So it was a happy end.
Thanks again for all your help.
Cheers,
Alberto