Examples for Molecular crystals

[rulong Zhou]

Do you have examples about prediciton of Molecular crystals? I did a
test using the example C6H6 molecule, but it stopped with errors:

error: invalid row index = 0
error: evaluating binary operator `-' near line 486, column 60
error: evaluating binary operator `.^' near line 486, column 83
error: evaluating argument list element number 1
error: evaluating binary operator `^' near line 486, column 87
error: evaluating assignment expression near line 486, column 21
error: evaluating for command near line 485, column 1
error: evaluating for command near line 469, column 1
error: evaluating if command near line 465, column 1
error: evaluating if command near line 458, column 1
error: called from `createORGStruc' in file `/home/MSI/rlzhou/USPEX/
test/FunctionFolder/createORGStruc.m'
error: evaluating if command near line 24, column 1
error: evaluating while command near line 1, column 1

I only changed the calculationtype to 11 in the INPUT-EA.txt and
provide MOL_1 in the same directory as INPUT-EA.txt.

[USPEX]

Hello,

in just a few days there will be a new USPEX release with the new manual. There will be three tests (examples) for molecular crystals there! You should just wait a little and modify the methane test :) which would be the closest one for your system.

Sincerely,

 Andriy

[rulong Zhou]

I wrote my MOL_1 file according the manual and the MOL_1 file in
examples T11. However, the program stopped with errors when generating
structures. The errors are:

error: invalid row index = 48
error: evaluating assignment expression near line 18, column 9
error: evaluating if command near line 12, column 1
error: evaluating for command near line 11, column 1
error: evaluating if command near line 10, column 1
error: evaluating if command near line 2, column 1
error: called from `NEW_ZMATRIXCOORD' in file
`/home/MSI/rlzhou/USPEX-9/test/FunctionFolder/USPEX/NEW_ZMATRIXCOORD.m'
error: evaluating assignment expression near line 49, column 8
error: evaluating if command near line 10, column 1
error: evaluating for command near line 9, column 1
error: called from `GetOrientation' in file
`/home/MSI/rlzhou/USPEX-9/test/FunctionFolder/USPEX/GetOrientation.m'
error: evaluating assignment expression near line 206, column 11
error: evaluating for command near line 203, column 1
error: evaluating if command near line 202, column 1
error: evaluating while command near line 161, column 1
error: evaluating if command near line 155, column 1
error: called from `initialize_POP_STRUC' in file
`/home/MSI/rlzhou/USPEX-9/test/FunctionFolder/USPEX/initialize_POP_STRUC.m'
error: evaluating if command near line 76, column 1
error: evaluating if command near line 40, column 1
error: evaluating while command near line 22, column 1
error: called from `ev_alg' in file `/home/MSI/rlzhou/USPEX-9/test/ev_alg.m'
error: evaluating if command near line 3, column 1

I checked the logfile. I seems no problem about the molecular
structure except that the goodcenter =0 while in the example
goodcenter =1.

Can you help me to fix these problems?

2012/5/24, USPEX <us...@googlegroups.com>:

[rulong Zhou]

My molecule contains carbon rings. It seem's that it hard to write the MOL_1 file when the molecule containing rings. Can you give me an example MOL_1 file of a molecule contaning a ring, e.g. benzene.

[rulong Zhou]

There is a bug in USPEX. When I do a test calculation using GULP, sometimes Gulp terminated with errors. I checked the output and input of GULP and found that the x, y, z coordinates of some atoms connected as one string. e.g. 0-4.4409e-16 0.5.  GULP can not read y coordinate from that line. I hope you can fix it?

[Andriy Lyakhov]

Hello,

please contact Qiang Zhu about creating the MOL_1 file, he should be able to help you.

As for gulp error - I have never seen it before. Does it appear in molecular or bulk crystal calculations?

Sincerely,

 Andriy

[Qiang Zhu]

Below is an example of MOL_1 for benzene. The last two columns are only used for DMACRYS code. If you don't use it, just keep it without any change.

Number of atoms: 12

   0.00000   0.00000   0.00000   6 0 0 0  1 1 1 1  1

   0.70488   1.22088   0.00000   6 1 0 0  0 1 1 2  1

   2.11462   1.21615   0.00000   6 2 1 0  0 0 1 3  1

   2.81129  -0.00943   0.00000   6 3 2 1  0 0 0 4  1

   2.09416  -1.22315   0.00000   6 4 3 2  0 0 0 5  1

   0.68454  -1.20424   0.00000   6 5 4 3  0 0 0 6  1

  -1.11074   0.00033   0.00000   1 1 2 3  0 0 0 7  1

   0.14922   2.18264   0.00000   1 2 1 7  0 0 0 8  1

   2.67293   2.17637   0.00000   1 3 2 8  0 0 0 9  1

   3.92200  -0.01655   0.00000   1 4 3 9  0 0 0 10 1

   2.64011  -2.19046   0.00000   1 5 4 10 0 0 0 11 1

   0.11660  -2.15880   0.00000   1 6 5 11 0 0 0 12 1

A short description on the format:

Number of atoms: 12  (Here you just put the number of atoms in the molecule), and then

      X          Y            Z      Species    i     j   k    ifr  ifa  ift      sort   Symbol

 0.00000  0.00000  0.00000      6         0    0    0    1   1   1         1         1
 0.63512  0.63512 -0.63512      1         1    0    0    0   1   1         2         1
-0.63512 -0.63512 -0.63512      1         1    2    0    0   0   1         3         1
 0.63512 -0.63512  0.63512      1         1    2    3    0   0   0         4         1
-0.63512  0.63512  0.63512      1         1    2    3    0   0   0         5         1

X, Y, Z

It records the Cartesian coordinate of the piece of molecule. One can just pick it from *.xyz format. 

Species

The index of element.

i, j, k

To specify the topological relation in Zmatrix style.

ifr, ifa, ift

Optimization flag (This is only used for SIESTA). In Zmatrix style, the molecules are constructed by internal coordinates (bond length, bond angle, and dihedral angle). If you put them 0, it means the molecule would be treated as the rigid molecule. 

By default, you can construct the matrix like
1 1 1
0 1 1
0 0 1
……
…..

The last two columns (sort and symbol are only used in DMACRYS, if you don't use DMACRYS as the local optimizer, please SKIP IT !!!)
 

Sort

It is to specify the order of atoms when reading the atomic coordinate in DMACRYS. For instance, the order of glycine in MOL_1 is:
C1 O1 O2 C2 H1 H2 H3 N1 H4 H5
Output file will give the atomic information in the following order:

C1 C2 H1 H2 H3 H4 H5 O1 O2 N1If the specify the order in the INPUT is
C H O N 

Symbol

This is the symbol to specify the element’s chemical environment in DMACRYS
For instance, H in CH4 is H_F1, H_F2 in NH3
By default, put 1; but sometimes it would be 2 or something else if you adopt DMACRYS as the local optimizer.

Cheers,

Qiang Zhu

[Andriy Lyakhov]

On Wednesday, May 30, 2012 12:27:13 PM UTC-4, rulong Zhou wrote:

There is a bug in USPEX. When I do a test calculation using GULP, sometimes Gulp terminated with errors. I checked the output and input of GULP and found that the x, y, z coordinates of some atoms connected as one string. e.g. 0-4.4409e-16 0.5.  GULP can not read y coordinate from that line. I hope you can fix it?

Hello,

I believe I found the problem. It's the way octave handles num2str operator for numbers like   -4.4409e-16. It is not a problem for matlab, but apparently is a problem to octave. I changed the way we write the input files and this problem should be fixed now. Please download the USPEX from our web-site again.  

Sincerely,

 Andriy 

[Qiang Zhu]

Dear Rulong,

If you are trying to use GULP for molecular crystals, you might meet problems. We indeed interfaced GULP for molecular crystal. Unfortunately, GULP does not support fixed-molecule optimization well at the moment. Therefore, it would be problematic in the output. Sometimes, GULP might lead to some meaningful results if one plays some tricks in the force field (which is definitely NONTRIVIAL). For the practical use, one would better try abinitio codes like SIESTA or VASP, which is very expensive, but robust and meaningful.

Cheers,

Qiang