Firefly and PC GAMESS-related discussion club



Learn how to ask questions correctly


Re^2: BAD DELOCALIZED COORDINATES

Masoud Nahali
masoudnahali@gmail.com


Dear Pedro

Thanks for your attention and help. What does this line "NONVDW(1)=1,9,2,9,3,9,4,9,1,15,2,15,3,15,4,15" mean ? in "ifzamt" keyword 1, 2, and 3 are related to bond, angle, and dihedral respectively but it seems that in the line above you had another mean ! and they are only the bonds 1-9 , 2-9 , 3-9 , 4-9 , 1-15 , 2-15 ... ! are they ? Also, is it necessary that we connect the all adsorbate atoms to the surface ? which atoms should be connected to the surface by NONVDW ? Does FF optimize the bonds that were not prepared by NONVDW ?        


On Tue Feb 8 '11 3:25pm, Pedro Silva wrote
------------------------------------------
>Adding extra coordinates with NONVDW has no effect on the pysics of the process, but it helps FF build appropriate delocalized coordinates.
>It works like this: assuming you have a surface with 8 atoms (numbered 1 through 8) and that the atom you want to adsorb a benzene (12 atoms, numbered 9 therough 20), you may build new non-physical bonds with:

>NONVDW(1)=1,9,2,9,3,9,4,9,1,15,2,15,3,15,4,15

>FF then takes the original bonds it could find (e.g. between 1 and, 1 and 4. etc...), adds the new bonds, computes all possible angles and dihedrals defined by ALL bonds, and builds DLCs with these coordinates.


[ Previous ] [ Next ] [ Index ]           Wed Feb 9 '11 2:00am
[ Reply ] [ Edit ] [ Delete ]           This message read 912 times