Reaction Path following by QSD
Jorge Luis LLanio Trujillo
jolla at ci.uc.pt
Tue May 23 09:59:34 BST 2000
Dear people:
I decided to poster to the molpro mailing
list ONLY after many hours of efforts. I really need some help to solve
this problem(or at least some tips)
I want to generate the reaction path from TS to products and that from TS
to reactants(i.e. i want to reach reactant and product minima) and also i
want to obtain the geometries, energies and projected vibrational
frequencies at the points calculated along the path(if possible, well
suited for the prospect of VTST) .
Level of theory : CASSCF(8,8)/6-311G(2d,f)
I have 3 distances r1,r2,r3 ; 2 bonding angles c1,c2 & one dihedral angle
(fi).
My molpro input file looks like:
{Z-matrix input; basis set input; transition state(TS) starting geometry
; wave function definition(spin & spatial symmetry as well as
the orbital space partitioning) , NOT INCLUDED as it seems to
me not important in the context of my question}
....
...
.
optg
method,qsdpath,NUMHES,1,1
option,IDIR=1
print,history
I start the integration of the MEP at the transition state geometry(I am
sure that this is a real TS at CASSCF(8,8)/6-311G(2d,f) because in a
previous molpro run, starting at some initial guess, i reached it and
after being calculated the frequencies i got only one imaginary) and the
first step is taken along the "transition vector"(IDIR=1, see above)
To my surprise, i got only one point , which is indeed the starting
point(namely, the TS from which I began) and the integration terminate. It
seems to me that the integration should terminate at product(reactant)
minima structures. Any Tips?
What follows is a partial output of the optimization history:
OPTIMIZATION HISTORY:
ENERGIES -299.246708
Positions
1
R1 1.2529757
R2 1.3785954
C1 111.3507527
R3 1.9953596
C2 104.0520245
FI 86.2868481
Gradients
1
R1 -0.0000001
R2 0.0000000
C1 0.0000000
R3 0.0000001
C2 0.0000000
FI 0.0000000
IDIR = -1
Quadratic Steepest Descent - Reaction Path Following using updated
Hessian
In QSD: Reaction Path coordinates, scaled to the original units:
Optimization point 1
.....
THANKS IN ADVANCE !!
Jorge Luis LLanio(jolla at ci.uc.pt)
PhD. student
University of Coimbra
Portugal
More information about the Molpro-user
mailing list