Geometry optimization in redundant internal coordinates is achieved using the
opt=Redundant or simply
opt keyword
Definition of a set of internally consistent, linearly maximally independent
internal coordinates is performed automatically by Gaussian based on
the structure supplied in the input file (both Z-Matrix or cartesian coordinates
can be used at that point). Taking as an example the input file for formaldehyde used
before
to illustrate the different input formats, the following redundant coordinates will be
generated by Gaussian:
----------------------------
! Initial Parameters !
! (Angstroms and Degrees) !
------------------------ -------------------------
! Name Definition Value Derivative Info. !
-----------------------------------------------------------------------------
! R1 R(1,2) 1.2 estimate D2E/DX2 !
! R2 R(1,3) 1. estimate D2E/DX2 !
! R3 R(1,4) 1. estimate D2E/DX2 !
! A1 A(2,1,3) 120. estimate D2E/DX2 !
! A2 A(2,1,4) 120. estimate D2E/DX2 !
! A3 A(3,1,4) 120. estimate D2E/DX2 !
! A4 L(3,1,4,2,-2) 180. estimate D2E/DX2 !
-----------------------------------------------------------------------------
The choice made by Gaussian can, to a large degree, be influenced by the
user through the definition of additional internal coordinates that should be part
of the overall set of coordinates. This can be achieved using the
opt=AddRed
keyword. The input file must then contain at least one additional line
of input following the regular geometry definition. If, in the above example, we want
the distance between hydrogen atoms H3 and H4 to be part of the coordinate set, we
could specify this in the following way:
#P HF/6-31G(d) opt=AddRed test1 HF/6-31G(d) opt formaldehyde 0 1 C1 O2 1 r2 H3 1 r3 2 a3 H4 1 r3 2 a3 3 180.0 r2=1.20 r3=1.0 a3=120. 3 4 |
#P HF/6-31G(d) opt=AddRed test1 HF/6-31G(d) opt formaldehyde 0 1 C1 O2 1 r2 H3 1 r3 2 a3 H4 1 r3 2 a3 3 180.0 r2=1.20 r3=1.0 a3=120. 3 4 1.60Given the additional constraint of a distance of 1.60 Angstrom between H3 and H4, Gaussian adjusts the values of other redundant coordinates such that the specified constraint is satisfied at the beginning of the optimization. This distance will, of course, be varied during the geometry optimization procedure.
last changes: 16.10.2004, HZ questions & comments to: zipse@cup.uni-muenchen.de