Firefly and PC GAMESS-related discussion club

 
Learn how to ask questions correctly  
 
 

 

Alex Granovsky
gran@classic.chem.msu.su

I'm guessing the problem is that two graphene sheets are not
geometrically connected so you need to connect them between
each other by a "bond" or two. This "artificial" bond(s) will help
Firefly to construct a complete coordinate system. You need to specify
a pair of atoms, one of these atoms belonging to the first sheet and
other to the second. Ideally these atoms should be in the proximity.
Did you read page 56 of the current Firefly's manual?

I'd recommend to use atoms near a center of the sheets. For example,
you can connect atoms 6 and 50. In addition, I'd add a second "bond"
near the boundary of the sheets, e.g., by connecting atoms 25 and 66
or so. These "bonds", as well as related auto-generated angular and
torsional variables will be added to the list of automatically
generated variables forming a full set of variables.

Next, check your geometry once again! Atoms 36, 43, 80 and 87
seems to be displaced.

>Also I wish to know that DO we run the optimization with DIFFERENT
distances between layers and then find the lowest energy one OR the
optimization started with ANY distance between the layers will
automatically get it to the lowest energy configuration? I strongly
feel the latter one is correct.

Indeed, the latter one is correct.

Kind regards,
Alex Granovsky

On Wed Oct 15 '14 7:48am, Siddheshwar Chopra wrote
--------------------------------------------------
>Dear Alex,
>Thank you for the information that there is negligible difference in CPU times for DFT and TDDFT. Sir please have a look at the attached output file. I am unable to run this bilayer graphene file due to the error:

> ERROR: BAD DELOCALIZED COORDINATES GENERATED!!!
>        588 PRIMITIVE INTERNAL COORDINATES ARE BEING USED.
>        252 LINEARLY INDEPENDENT COMBINATIONS OF THESE ARE FOUND.
>        258 COORDINATES ARE NEEDED TO HAVE A TOTAL OF 3N-6 OR 5.

> IF THE NUMBER OF LINEARLY INDEPENDENT COOORDS IS TOO SMALL,
> ADD SOME COORDINATES BY -NONVDW- TO CONNECT MONOMER PIECES
> (ESPECIALLY IF YOU ARE SHORT BY A MULTIPLE OF 6), OR ELSE
> USE -IXZMAT- TO ADD EXTRA PRIMITIVE COORDINATES.

> IF THE NUMBER OF LINEARLY INDEPENDENT COOORDS IS TOO BIG,
> TRY USING -IRZMAT- TO REMOVE SOME PRIMITIVE COORDINATES.

>I also read the manual, but failed to understand how to use "NONVDW". Please help me in running this input file. Where am I doing wrong?
>Also I wish to know that DO we run the optimization with DIFFERENT distances between layers and then find the lowest energy one OR the optimization started with ANY distance between the layers will automatically get it to the lowest energy configuration? I strongly feel the latter one is correct.