[molpro-user] convergence to wrong state

[ERSIN YURTSEVER]

I am calculating the potential energy curve for stacked pentacene with DF-MP2. The relevant input is below.If I start with the monomer-monomer separation from z=3.0 A and increase this distance by 0.1 A, I get very high energies. Around z=3.6 (which should be a near minimum) the energy is above two separate monomers.

On the other hand, starting at z=3.5, I get more reasonable answers. I guess I am running into a different state but in such closed-shell interactions should one define the state? 

Since it is a DF calculation, nosym option is in action.

The problem of the high energy starts at DF-HF level.

BASIS AND METHOD DEFINITION IS:

basis
default,avdz
default,vdz,h
end

df-hf,df_basis=vdz
df-mp2,df_basis_avdz


I appreciate suggestions on how to define the proper state.