mat and ucf files for defects and doping

[الأستاذ لطرش‎]

Dear developers

I'm studying this compound CoFe1.05Mn0.95Al, how to modify mat and ucf file to understand the precentages i'm using

material:num-materials = 5
#---------------------------------------------------
# Material 1
#---------------------------------------------------
material[1]:material-name=Co
material[1]:damping-constant=1.0
material[1]:atomic-spin-moment=0.8006 !muB
material[1]:uniaxial-anisotropy-constant=0.00000000e+00
material[1]:material-element=Co
material[1]:initial-spin-direction = 0.0,0.0,1.0
material[1]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[1]:unit-cell-category=1
#---------------------------------------------------
#---------------------------------------------------
# Material 2
#---------------------------------------------------
material[2]:material-name=Fe_1
material[2]:damping-constant=1.0
material[2]:atomic-spin-moment=0.1618 !muB
material[2]:uniaxial-anisotropy-constant=0.00000000e+00
material[2]:material-element=Fe
material[2]:initial-spin-direction = 0.0,0.0,-1.0
material[2]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[2]:unit-cell-category=2

#---------------------------------------------------
#---------------------------------------------------
# Material 3
#---------------------------------------------------
material[3]:material-name=Mn
material[3]:damping-constant=1.0
material[3]:atomic-spin-moment=2.5156 !muB
material[3]:uniaxial-anisotropy-constant=0.00000000e+00
material[3]:material-element=Mn
material[3]:initial-spin-direction = 0.0,0.0,1.0
material[3]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[3]:unit-cell-category=3
#---------------------------------------------------
#---------------------------------------------------
# Material 4
#---------------------------------------------------
material[4]:material-name=Fe_2
material[4]:damping-constant=1.0
material[4]:atomic-spin-moment=2.7926 !muB
material[4]:uniaxial-anisotropy-constant=0.00000000e+00
material[4]:material-element=Fe
material[4]:initial-spin-direction = 0.0,0.0,1.0
material[4]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[4]:unit-cell-category=3
# Material 5
#---------------------------------------------------
material[5]:material-name=Al
material[5]:damping-constant=1.0
material[5]:uniaxial-anisotropy-constant=0.00000000e+00
material[5]:material-element=Al
material[5]:initial-spin-direction = 0.0,0.0,-1.0
material[5]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[5]:non-magnetic = remove
material[5]:unit-cell-category=4
#---------------------------------------------------

UCF file

# Unit cell size:
5.71 5.71 5.71
# Unit cell vectors:
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
# Atoms num, id cx cy cz mat lc hc
4 4
0 0.00 0.00 0.00 0 0 0
1 0.25 0.25 0.25 1 0 0
2 0.50 0.50 0.50 2 0 0
3       0.75    0.75    0.75    3       0       0

[gabo...@gmail.com]

For CoFeMnAl, it looks like your material file (before you added Fe_2 as material[4] and set Al as material[5]) had 4 atoms (1 atom Co, 1 atom Fe, 1 atom Mn, and 1 atom Al) given by:

material[1], material[2], material[3], and material[4]

And your .ucf has a cell with those atoms given by:

0,  1 , 2, and 3.

The LCD of 1.05 and 0.95 in CoFe1.05Mn0.95Al should be 20 [1].

So, it looks like a supercell of at least 80 atoms (20 atoms Co, 21 atoms Fe, 19 atoms Mn, and 20 atoms Al) would be needed for CoFe1.05Mn0.95Al or Co20Fe21Mn19Al20 (from Fe1.05 x 20 = Fe21, Mn0.95 x 20 = Mn21).

Therefore, your material file likely needs adjusted to have:

material[1], material[2], ..., material[20] for Co

material[21], material[22], ..., material[60] with 21 of these being for Fe and 20 for Mn

material[61], material[62], ...,  material[80] for Al

The .ucf would need modified correspondingly with atoms:

0, 1, ..., 79

Hope that helps.

[1]

Kind Regards,

Gavin

VAMPIRE user

On Friday, May 1, 2026 at 12:21:00 PM UTC-6 goldf...@gmail.com wrote:

Dear developers

I'm studying this compound CoFe1.05Mn0.95Al, how to modify mat and ucf file to understand the precentages i'm using

material:num-materials = 5
#---------------------------------------------------
# Material 1
#---------------------------------------------------
material[1]:material-name=Co
material[1]:damping-constant=1.0
material[1]:atomic-spin-moment=0.8006 !muB
material[1]:uniaxial-anisotropy-constant=0.00000000e+00
material[1]:material-element=Co

material[1]:material[1]:initial-spin-direction = 0.0,0.0,1.0

[الأستاذ لطرش‎]

Thank you for you response Kevin

I was reading the vampire guide and found the "material[A]:host-alloy and material[A]:alloy-fraction[B] = 0.05".

In my case i'm using CPA (Coherent Potential Approximation) witch is a different way to treat defects and doping than a supercell approach. The problem now is the calculated Curie temperature is nearly double (1150) the value found by Mean Field Approximation (523), calculated by SPRKKR.  

New .mat file

material:num-materials = 4

material[3]:host-alloy

material[3]:damping-constant=1.0
material[3]:atomic-spin-moment=2.5156 !muB
material[3]:uniaxial-anisotropy-constant=0.00000000e+00
material[3]:material-element=Mn
material[3]:initial-spin-direction = 0.0,0.0,1.0
material[3]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0

material[3]:alloy-fraction[2] = 0.05

material[3]:unit-cell-category=3
#---------------------------------------------------
#---------------------------------------------------

# Material 4
#---------------------------------------------------

material[4]:material-name=Al
material[4]:damping-constant=1.0
material[4]:uniaxial-anisotropy-constant=0.00000000e+00
material[4]:material-element=Al
material[4]:initial-spin-direction = 0.0,0.0,-1.0

material[4]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0

material[4]:non-magnetic = remove
material[4]:unit-cell-category=4
#---------------------------------------------------

[gabo...@gmail.com]

I have read before that CPA can handle partial occupancies (i.e., at [1]).

However, I'm unfamiliar with using a CPA program and using its output with VAMPIRE.  If two atoms (in this case Fe and Mn) were occupying the same atomic position, how is the output provided by the CPA program?  Does the CPA provide you say an effective atomic-spin-moment and exchange values to use?  Or does it provide output for partial occupancy contributions separately?  If separately, I don't recall if the .ucf will allow you to specify an atomic position twice with two different atom id.  I suppose you could try it to see if the VAMPIRE program gives an error or not when you try that.  If it does error to show its not allowed, is there a way you have to combine the results from CPA (such as perhaps for example by averaging them)?

Yeah, I believe you should be able to use alloy-fraction with the cell as long a you're fine with the atom arrangement VAMPIRE provides you with for that.  If you need to control more specifically the atom arrangement, you may have to do it as a supercell.  Your new material file (with material[1] through material [4]) below looks fine as long as Mn (material[3]) can replace any of the Fe atoms (material[2]).  However, if some Fe sites shouldn't be substituted with Mn.  Then, I think your previous supercell sublattice approach would be the way to go, where you had the Fe_1 (material[2]) that won't be substituted by Mn while 5% of Fe_2 (material[4]) could be substituted with Mn (material[3]) using material[4]:alloy-fraction[3] (based on [2]).

In any case, you would probably want to visualize VAMPIRE's output of the structure to check that it looks as desired.  For examples of visualizing if it helps, there should be a post with rasmol at [3] and jmol at [4].

[1] https://www.mail-archive.com/wi...@zeus.theochem.tuwien.ac.at/msg05007.html

[2] https://vampire.york.ac.uk/tutorials/system-generation/random-alloy/

[3] https://groups.google.com/g/vampire-users/c/XGacgJv5r-A/m/HVS20iduDQAJ

[4] https://groups.google.com/g/vampire-users/c/XGacgJv5r-A/m/wFB150dbCgAJ

Kind Regards,

Gavin

VAMPIRE user

[الأستاذ لطرش‎]

CPA is intergrated with SPRKKR code, the result of exchange is given as an effective J value but the  moment are provided by scf are separated 

i tried this 

(base) ecp3m@Poste-3:~/SPRKKR/CoFeMnAl/cpa/vampire$ grep "makes up" log
02-05-2026 [22:48:36] Material 1 Co makes up 33.3333 % of all atoms ( 1000 atoms )
02-05-2026 [22:48:36] Material 2 Fe_1 makes up 33.3333 % of all atoms ( 1000 atoms )
02-05-2026 [22:48:36] Material 3 Mn makes up 31.6333 % of all atoms ( 949 atoms )
02-05-2026 [22:48:36] Material 4 Al makes up 0 % of all atoms ( 0 atoms )
02-05-2026 [22:48:36] Material 5 Fe_2 makes up 1.7 % of all atoms ( 51 atoms )

material:num-materials = 5
#---------------------------------------------------
# Material 1
#---------------------------------------------------
material[1]:material-name=Co
material[1]:damping-constant=1.0
material[1]:atomic-spin-moment=0.8006 !muB
material[1]:uniaxial-anisotropy-constant=0.00000000e+00
material[1]:material-element=Co

material[3]:alloy-fraction[5] = 0.05

material[3]:unit-cell-category=3
#---------------------------------------------------
#---------------------------------------------------

# Material 4
#---------------------------------------------------
material[4]:material-name=Al
material[4]:damping-constant=1.0
material[4]:uniaxial-anisotropy-constant=0.00000000e+00
material[4]:material-element=Al
material[4]:initial-spin-direction = 0.0,0.0,-1.0
material[4]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0
material[4]:non-magnetic = remove

material[3]:unit-cell-category=4
#---------------------------------------------------
# Material 5
#---------------------------------------------------
material[5]:material-name=Fe_2
material[5]:damping-constant=1.0
material[5]:atomic-spin-moment=2.7 !muB
material[5]:uniaxial-anisotropy-constant=0.00000000e+00
material[5]:material-element=Fe
material[5]:initial-spin-direction = 0.0,0.0,1.0

material[5]:uniaxial-anisotropy-direction = 0.0 , 0.0, 1.0

 

[Gavin Abo]

Thanks for the CPA information.

If you need the exact composition of CoFe1.05Mn0.95Al, you have probably already noticed from your log output that the number of atoms look like it will give a composition that is slightly off:

Co1000/1000Fe(1000+51)/1000Mn949/1000Al1000/1000 = CoFe1.051Mn0.949Al

So, there are likely structure parameters (such as alloy-fraction in the .mat file, atomic positions in the .ucf, and/or dimensions in the input file) that have to be adjusted if the exact composition is desired.

Kind Regards,

Gavin
VAMPIRE user