Re: [Jmol-users] Re: crystal cell terminology

[Miguel H. <mt...@mt...>]

> Miguel,
> 	About your unit cell terminology, I prefer "unit cell" and
> "multi cell" ("supercell" means something else to a crystallographer).

OK, good.

> 	But I also wanted to address a bigger question - how to deal
> with space group symmetry in order to "fill up" the unit cell. I've
> seen some discussions about the 230 different space groups and how to
> incorporate this information into Jmol.

To be clear ... we are not working on the space groups yet ... we are only
working on P1.

And in fact, I am still trying to get the unitcell atoms rendered inside
the cage.

But this will be very helpful ...

> If the input file is a SHELXL
> .res file, all the space group info is present in the "LATT" line and
> the "SYMM" lines. If LATT is 1, the cell is centrosymmetric (for
> every atom at x,y,z there is a symmetry related atom at -x,-y,-z); if
> LATT is -1, the cell is non-centrosymmetric and the space group does
> not have the -x,-y,-z symmetry operation. The SYMM lines list the  other
> symmetry operations.
> Thus, for the popular space group P2(1)/c, the LATT and SYMM lines are:
>
> 		LATT 1
> 		SYMM -X, .50+Y, .50-Z
>
> 	The combination of LATT and SYMM in this case, leads to four
> symmetry-related positions:
>
> 		x,y,z
> 		-x,-y,-z
> 		-x, 1/2+y, 1/2-z
> 		x, 1/2-y, 1/2+z
>
> which is all the info needed to "populate" the cell. Of course,
> depending on the particular co-ordinates, these expressions can lead  to
> molecules in the "next cell over" and it may be necessary to use,  e.g.,
> that third symmetry operation as 1-x, 1/2+y, 3/2-z to get the  molecule
> into the central unit cell (it is always legal to add or  subtract 1
> from the crystal coordinates).

OK, I think I got the general idea behind this.

And thanks to Richard Ball I now have a much better understanding of
fractional coordinates, so it makes some sense.

Egon will be able to make more lucid responses.

(Although be advised that for some unknown reason, he is currently unable
to post messages to sourceforge mailing lists ? )

> 	So there is no need to "teach" Jmol about all those space groups!

I *certainly* understand this ... that would be great.



Miguel