Rank deficiency detected by HFTI, rank = 0 (12)

[Bhuvan]

Hello!

I was performing flux melting calculations using alphaMELTS 1.9 on Windows using Depleted MORB Mantle Composition of Workman & Hart 2005 and the hydrousfluid composition taken from the workshop files. It was working fine till yesterday.

Surprisingly, today I encountered the Rank deficiency detected by HFTI, rank = 0 (12) with a THE QUADRATIC MINIMIZATION ALGORITHM HAS FAILED TO CONVERGE error. The initial subsolidus calculation using Option 3 was fine. The output was as follows:

Initial alphaMELTS calculation at: P 10000.000000, T 1200.000000
olivine: 56.637167 g, composition (Ca0.00Mg0.89Fe''0.10Mn0.00Co0.00Ni0.00)2SiO4
orthopyroxene: 24.202071 g, composition opx Na0.01Ca0.05Fe''0.18Mg1.67Fe'''0.01Ti0.01Al0.16Si1.92O6
clinopyroxene: 15.707648 g, composition cpx Na0.10Ca0.66Fe''0.11Mg0.91Fe'''0.01Ti0.01Al0.31Si1.88O6
spinel: 3.592112 g, composition Fe''0.16Mg0.85Fe'''0.04Al1.62Cr0.33Ti0.01O4

Selecting option 4 and then trying to run the calculation gives the Rank Deficiency error. I had tried to run the calculation on the WSL version of alphaMELTS 1.9 with the same input files and got the same (which was working fine till last week!). For your reference, I am attaching the settings file and the input files here. 

Any leads on how to solve this issue would be much appreciated.

Thank you!

[Paula]

Were you using exactly the same files when it was working before? Same amount of flux added at each step etc? If so, then it must be a bug in the code. Unfortunately that will not be fixed, as alphaMELTS 1.9 is not being actively developed anymore.

If the files differed, then in what way e.g. P, T conditions, amount of flux? If you post what you typed at the menu that would help. Thanks.

[Bhuvan]

Hello Paula!

Thank you for your reply!

UPDATE: I copy pasted the files into a different directory and the calculation works again! I'm not sure what was wrong with the original directory that these files were in.

Yes, I was using the exact same .melts files with the same amount of flux added in each step. I am adding the input melts files here for your convenience.

Code Select Expand

Title: Depleted MORB Mantle of Workmann & Hart 2005
Initial Composition: SiO2 44.71
Initial Composition: Al2O3 3.98
Initial Composition: Fe2O3 0.191
Initial Composition: FeO 8.008
Initial Composition: MnO 0.13
Initial Composition: MgO 38.73
Initial Composition: CaO 3.17
Initial Composition: Na2O 0.28
Initial Composition: Cr2O3 0.57
Initial Composition: TiO2 0.13
Initial Composition: NiO 0.24
Initial Trace: K 60
Initial Trace: P 20
Initial Trace: Ni 1886
Initial Trace: Cr 3900
Initial Trace: Mn 822
Initial Trace: Rb 0.050
Initial Trace: Ba 0.563
Initial Trace: Th 0.0079
Initial Trace: U 0.0032
Initial Trace: Nb 0.1485
Initial Trace: Ta 0.0096
Initial Trace: La 0.192
Initial Trace: Ce 0.550
Initial Trace: Pb 0.018
Initial Trace: Pr 0.107
Initial Trace: Nd 0.581
Initial Trace: Sr 7.664
Initial Trace: Zr 5.082
Initial Trace: Hf 0.157
Initial Trace: Sm 0.239
Initial Trace: Eu 0.096
Initial Trace: Ti 716.3
Initial Trace: Gd 0.358
Initial Trace: Tb 0.070
Initial Trace: Dy 0.505
Initial Trace: Ho 0.115
Initial Trace: Y 3.328
Initial Trace: Er 0.348
Initial Trace: Yb 0.365
Initial Trace: Lu 0.058
Initial Temperature: 1350.0
Initial Pressure: 20000
Log fO2 Path: FMQ
Log fO2 Delta: 0.0

Assimilant:

Code Select Expand

Title: hydrousfluid   
Initial Composition: SiO2 20.0
Initial Composition: TiO2 0.001
Initial Composition: Al2O3 0.001
Initial Composition: Fe2O3 0.001
Initial Composition: FeO 0.001
Initial Composition: MgO 0.001
Initial Composition: CaO 0.001
Initial Composition: Na2O 5.0
Initial Composition: K2O 2.0
Initial Composition: H2O 73.0 
Initial Mass: 0.01 
Initial Trace: La 37 
Initial Trace: Ce 80 
Initial Trace: Pr 9.7 
Initial Trace: Nd 38.5 
Initial Trace: Sm 10
Initial Trace: Eu 3
Initial Trace: Gd 7.62
Initial Trace: Tb 1.05
Initial Trace: Dy 5.6
Initial Trace: Ho 1.06
Initial Trace: Er 2.62
Initial Trace: Yb 2.16
Initial Trace: Lu 0.3
Log FO2 Path: FMQ
Log fO2 Delta: 0.0

The batch file:

Code Select Expand

1
DMM.melts
3
0
olivine
x
4
250
1
-1
hyd.melts

Thanks again for your time!

[Paula]

Is there a typo in your batch file? The assimilant file that you linked earlier was called hyd1.melts. If you are running it with a batch file and it fails to open the enrichment file then the results could be rather unpredictable.

[Bhuvan]

I'm sorry I had posted a batch file.

hyd.melts and hyd1.melts are exactly the same with just the filenames being different. I have two different batch files referring to each of them, used separately.