Screenshot instructions:
Windows
Mac
Red Hat Linux
Ubuntu
Click URL instructions:
Rightclick on ad, choose "Copy Link", then paste here →
(This may not be possible with some types of ads)
From: Fabian Dortu <fabian.dortu@wa...>  20020906 09:11:39

I would like to answer to the following remark: >Personally, I don't understand the information presented in the crystal >properties. The information is different than that generally used with >protein crystallography, which uses cell dimensions (a, b, c) and >angles alpha, beta, gamma), and the space group. The notation used in crystal properties(CP) is the one used in the code 'abinit' (www.abinit.org). Actually, we use *adimensional* base vectors and multiplication factors. It is only because it is more easy to figure (0.5 0.5 0.0)*4.35667 (0.0 0.5 0.5)*4.23445 (0.5 0.0 0.5)*4.34459 than (2.178335 2.178335 0.0) (0.0 2.117225 2.117225) (2.172295 0.0 2.172295) . Ok now, suppose we have a *cartesian* referential x,y,z. Actually it is the referential used in Jmol. **The base vectors given in CP are the base vectors in this referential. These base vectors are *absolute*. I mean that the orientation of the unit cell is fixed. The atomic positions can be loaded either in *cartesian* coordinate or in *crystallographic* coordinate (=reduced coordinate of the base vectors). **Now suppose that the base vectors are given in the nation a,b,c,alpha,beta,gamma. The unit cell is no more fixed in space. If you want to fix it you have to make 2 hypothesis. For instance: a is parallel to x. b is in plane (xy). If you don't fix the unit cell, you can only give the atoms in *crystallographic* coordinate. If you fix the unit cell in the referential then you can also give the *cartesian* coordinate. ** What about the space group? I still have to learn more about the notation and how to link it to the symmetry operations. For now, the space group is asummed to be "1". It means that we have to specify *all* the atoms in the unit cell. Now, you have the picture. I have to think how I can design a nice CP dialog box that combines the 2 possible notations. I you have ideas, you are welcome!! Fab  ********** Fabian Dortu Collegeberg 16 3000 Leuven Belgium Phone : 32475599268 emails : Fabian.Dortu@... Fabian.Dortu@... *********************************************** 
From: E.L. Willighagen <egonw@sc...>  20020906 09:20:17

On Friday 06 September 2002 13:11, Fabian Dortu wrote: > I would like to answer to the following remark: > >Personally, I don't understand the information presented in the crystal > >properties. The information is different than that generally used with > >protein crystallography, which uses cell dimensions (a, b, c) and > >angles alpha, beta, gamma), and the space group. > > The notation used in crystal properties(CP) is the one used in the code > 'abinit' (www.abinit.org). > Actually, we use *adimensional* base vectors and multiplication factors. > It is only because it is more easy to figure > > (0.5 0.5 0.0)*4.35667 > (0.0 0.5 0.5)*4.23445 > (0.5 0.0 0.5)*4.34459 > > than > > (2.178335 2.178335 0.0) > (0.0 2.117225 2.117225) > (2.172295 0.0 2.172295) That sounds ok. Merely a matter of convenience... > Ok now, suppose we have a *cartesian* referential x,y,z. Actually it is > the referential used in Jmol. > > **The base vectors given in CP are the base vectors in this referential. > These base vectors are *absolute*. I mean that the orientation of the > unit cell is fixed. > The atomic positions can be loaded either in *cartesian* coordinate or > in *crystallographic* coordinate (=reduced coordinate of the base > vectors). Aka. partial atomic coordinates. > **Now suppose that the base vectors are given in the nation > a,b,c,alpha,beta,gamma. The unit cell is no more fixed in space. If you > want to fix it you have to make 2 hypothesis. For instance: > > a is parallel to x. > b is in plane (xy). Yes, that is normally used... > If you don't fix the unit cell, you can only give the atoms in > *crystallographic* coordinate. If you fix the unit cell in the > referential then you can also give the *cartesian* coordinate. > > > ** What about the space group? I still have to learn more about the > notation and how to link it to the symmetry operations. For now, the > space group is asummed to be "1". It means that we have to specify *all* > the atoms in the unit cell. What I use is to add the assymetric atoms... based on the symmetry group the symmetry related atoms are added... This is done in two steps: 1. Add symmetry related atoms based on space group 2. Move atoms inside unit cell by translation along one or more cell axes In case of molecular crystals, I used the molecular centre of mass in step 2. to move the molecule inside the unit cell. > Now, you have the picture. I have to think how I can design a nice CP > dialog box that combines the 2 possible notations. You could also have a CP that allows to edit both... Egon 
From: E.L. Willighagen <egonw@sc...>  20020912 11:34:32

On Friday 06 September 2002 13:11, Fabian Dortu wrote: > Ok now, suppose we have a *cartesian* referential x,y,z. Actually it is > the referential used in Jmol. > > **The base vectors given in CP are the base vectors in this referential. > These base vectors are *absolute*. I mean that the orientation of the > unit cell is fixed. > The atomic positions can be loaded either in *cartesian* coordinate or > in *crystallographic* coordinate (=reduced coordinate of the base > vectors). > > **Now suppose that the base vectors are given in the nation > a,b,c,alpha,beta,gamma. The unit cell is no more fixed in space. If you > want to fix it you have to make 2 hypothesis. For instance: > > a is parallel to x. > b is in plane (xy). > > If you don't fix the unit cell, you can only give the atoms in > *crystallographic* coordinate. If you fix the unit cell in the > referential then you can also give the *cartesian* coordinate. Have you seen that the program I wrote based on Jmol 1 contains code to do this? Egon 
From: Fabian Dortu <fabian.dortu@wa...>  20020912 12:12:37

On Thu, 20020912 at 13:34, E.L. Willighagen wrote: > On Friday 06 September 2002 13:11, Fabian Dortu wrote: > > Ok now, suppose we have a *cartesian* referential x,y,z. Actually it is > > the referential used in Jmol. > > > > **The base vectors given in CP are the base vectors in this referential. > > These base vectors are *absolute*. I mean that the orientation of the > > unit cell is fixed. > > The atomic positions can be loaded either in *cartesian* coordinate or > > in *crystallographic* coordinate (=reduced coordinate of the base > > vectors). > > > > **Now suppose that the base vectors are given in the nation > > a,b,c,alpha,beta,gamma. The unit cell is no more fixed in space. If you > > want to fix it you have to make 2 hypothesis. For instance: > > > > a is parallel to x. > > b is in plane (xy). > > > > If you don't fix the unit cell, you can only give the atoms in > > *crystallographic* coordinate. If you fix the unit cell in the > > referential then you can also give the *cartesian* coordinate. > > Have you seen that the program I wrote based on Jmol 1 contains > code to do this? > > Egon > OOups, no, I didn't check actually. I will have a look. Fabian  Contrary to popular belief, Unix is user friendly. It just happens to be selective about who it makes friends with. Dave Parnas ********** Fabian Dortu Collegeberg 16 .~. 3000 Leuven /V\ Belgium /( )\ ^^^^ Phone : 32475599268 emails : Fabian.Dortu@... Fabian.Dortu@... *********************************************** 
Sign up for the SourceForge newsletter:
No, thanks