[solved] Difference of spacegroup description in TOPAS vs. BGMN

iangie

Dear TOPAS users,

I am currently translating an structure file in BGMN syntax to TOPAS syntax.
The describtion of spaceg group and atome positions in BGMN seams a little bit different from that in TOPAS:

For example, in BGMN:

SpacegroupNo=5 HermannMauguin=C121
PARAM=A=0.53_0.525^0.535 PARAM=B=0.915_0.91^0.92 PARAM=C=1.5_1.4^1.58 BETA=100.2
......
// ------atom positions------
E=AL+3 Wyckoff=a y=0.0 TDS=tdsoct
E=MG+2 Wyckoff=a y=0.6540 TDS=tdsoct
E=FE+3 Wyckoff=a y=0.3210 TDS=tdsoct
E=SI+4 Wyckoff=c x=0.4320 y=0.3330 z=0.1245 TDS=tdstet
E=SI+4 Wyckoff=c x=0.4320 y=0.6620 z=0.1245 TDS=tdstet
......

BGMN seams allow manuplate the symmetry of each atom site seperately by adding Wyckoff position in atom sites. In BGMN spacegrp.dat file, symmetry opereations are defined based on SpacegroupNo, HermannMauguin, and Wyckoff:

SpacegroupNo=5 HermannMauguin=C121 Setting=1 Lattice=Monoclinic UniqueAxis=b CellChoice=1
Wyckoff=c N=4
x y z
-x y -z
Wyckoff=b N=2
0 y 1/2
Wyckoff=a N=2
0 y 0

SpacegroupNo=5 HermannMauguin=C121 Setting=8 Lattice=Monoclinic UniqueAxis=b CellChoice=2
Wyckoff=c N=4
x y z
-x y -z
Wyckoff=b N=2
0 y 1/2
Wyckoff=a N=2
0 y 0

While in TOPAS:

str
...
space_group C2/m
....
site Al1 num_posns 2 x =0;:0.00000 y =0;:0.00000 z =0;:0.00000 occ 1 beq 0.5

It seams TOPAS do not deal with atom symmetry seperately, the symmetry operation for atoms in unit cell is defined in space_group only.

Is my understanding correct? Is TOPAS also able to treat atom symmetry sepsrately?
Thank you very much.

johnsoevans

Iangie,

Not sure I quite understand your question. The normal thing to do in topas is to just quote the atoms in the asymmetric unit and let the space group generate the rest.

e.g. site Al x 0.21 y 0 z 0 occ Al 1.0 beq 0.5

If you want to you can express coordinates using parameters/equations which may or may not be symmetry operators. e.g. you could choose to have two atoms related by the expression below:

site Al1 x al_x 0.21 y 0 z 0 occ Al 1.0 beq 0.5
site Al2 x =1-al_x; y 0 z 0 occ Al 1.0 beq 0.5

Here the second site has an x coordinate related by an equation to that of the first site. If you put a ":0" after the semicolon in the equation then topas will write the result of the equation at the end of refinement. There's lots of flexibility to set things up how you like.

Does that help?

John

iangie

Thanks John,

So the symmetric operation is same for all sites in unit cell, right?
Do you know why BGMN require Wyckoff position specified for each site?

Thank you. :)

johnsoevans

Iangie,

All symmetry operations get applied to all the atoms in the list. What's different is how many equivalent sites get generated from one site. In topas this is what the "num_posns 2" tells you - one atom in the list (asymmetric unit) generates 2 in the unit cell. It's the same number as you see in the Wyckoff label (e.g. 2c). Can't comment on BGMN.

John

iangie

I see. I really appreciate. Thanks John.

alan-coelho

Iangie,

I would also add that the number of positions generated is dependent upon whether the site coordinates is refined or not. For example, in all space groups (except for P1) then the following will generate a special position:

site NAME x 0 y 0 z 0 ....

In all space groups then the following will generate a general position:

site NAME x @ 0 y @ 0 z @ 0 ....

The program interprets the site coordinates and determines that the coordinates are free to move as they are defined that way. Depending on the definition of the coordinates the program may look at many equations including the fact that the site may be defined within a rigid body or a number of rigid bodies.

Thus the number of positions generated is implied by the definition of the site coordinates.

iangie

Thanks for your inputs, Alan! :)