I’m using the latest compiled version of PSI4 (1.2a1.dev781). So the problem is I cannot optimize transition state of the hydrogen abstraction reaction.
The method is df-mp2/6-311g(s). Two minima’s are great optimized in psi4, but TS is the problem. I just use the optimized TS geometry given by NWChem. If I use freq(‘mp2’) calculation with psi4 on this geometry it gives me one imaginary frequency corresponded to the hydrogen abstraction TS. But when I try to optimize(‘mp2’) on NWChem’s geometry, psi4 gives me the minima corresponds to reaction products. I have to calculate the whole reaction using psi4 cause the energies from it and NWChem may vary.
Is it any way to control OPTKING while optimization of TS? e.g. print the RFO vectors of the first step and choose one.
My input file looks like this:
molecule {
0 3
symmetry c1
N -0.04237 -1.35927 0.21347
O 0.96141 -0.44745 -0.12726
O -1.22835 -1.06655 -0.04633
C 0.45329 -2.71419 0.01093
H 1.42968 -2.80422 0.48029
H -0.27087 -3.38400 0.46584
H 0.52516 -2.89936 -1.06301
H 0.83096 3.60733 -0.72678
H -1.02978 2.13560 -1.03629
H -1.06617 1.90189 0.79180
H 0.08395 1.11808 -0.18673
N 0.27896 3.34419 0.07303
H 0.80063 3.39322 0.93279
C -0.51634 2.21792 -0.08605
}
memory 21 gb
set {
basis 6-311gs
reference uhf
mp2_type df
scf_type df
guess core
d_convergence 1e-1
maxiter 120
geom_maxiter 100
full_hess_every 0
opt_type ts
rfo_root 1
step_type rfo
rfo_follow_root 1
g_convergence gau_tight
flexible_g_convergence True
dynamic_level 4
intrafrag_step_limit 0.02
intrafrag_step_limit_min 0.02
intrafrag_step_limit_max 0.05
interfrag_step_limit 0.01
#opt_coordinates both
}
optimize('MP2')