You are here

ragged curves on pseudosection

6 posts / 0 new
Last post
Joined: 2015 Nov 21
ragged curves on pseudosection

Hi everyone,

I'm calculating pseudosection using td-tcds62-6axmn-03.txt dataset. Unfortunately, calculated pseudosection looks very strange (see below). Does anyone know what is the problem?

Jacek Sz.

Joined: 2018 Nov 07


Local minima related to the LiQ model? Plus labels :) 

It is a common proble for me with amphibole / cpx at HP/LT. I would try without the melt model. 



Joined: 2015 Nov 21

Hi Alex,

thanks for your advice. I'll try to do this. However, this problem doesn’t exist using different dataset.


Pierre Lanari
Pierre Lanari's picture
Joined: 2015 Oct 29

Hi Jacek, 

How did you define the number of oxygen in your bulk? I you are using O(?), I suggest you to add a little extra oxygen to avoid inimization issues with the ferric end-members. 

I would try to calculate the diagram between 550-600 °C and 5000-7500 bar using the following bulk: 

1    SI(66.84)AL(16.44)FE(3.14)MG(3.24)CA(0.55)NA(6.56)K(2.77)TI(0.43)MN(0.05)O(?)O(0.001)    * 

An other thing you could do is to run manually a few minimizations with Theriak in this P-T range and try to figure out what the problem actually is.



Joined: 2015 Nov 21

Hi Pierre,

thanks for your replay. I'm using the following bulk:

1   SI(66.8367)AL(16.4382)FE(3.1427)MG(3.2363)MN(0.0488)CA(0.5455)NA(6.5557)K(2.7697)TI(0.4264)H(100)O(?)  * 

So, if I understand you I could modify this bulk to:

1   SI(66.8367)AL(16.4382)FE(3.1427)MG(3.2363)MN(0.0488)CA(0.5455)NA(6.5557)K(2.7697)TI(0.4264)H(100)O(?)O(0.001)  * 

Am I right?



Joined: 2018 Nov 25

Hi Jacek,

Doug Tinkham discusses the issue with Ferric iron on his site ( under the section Ferric Iron Problems.

And yes, I believe that Pierre is correct in added a minor amount of  O to your bulk.

Tyler Ambrose

Postdoctoral Fellow

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer