BSSE and MP2 part error XYG3 functional


Clicked A Few Times
Hello again,

I am still trying to compute a CH3 +H2 reaction with XYG3 functional. I want to use two different orbital basis set : 6-31G** and aug-cc-pvQZ.

Here is my input:

title "nwpath3"
echo
start nwpath3

geometry geom1 units angstrom
LOAD format xyz xyz3.xyz
END

BASIS
* library 6-31G**
bqC library C 6-31G**
bqH library H 6-31G**
END

set geometry geom1
SCF
doublet
END

DFT
odft
tolerances tight
grid xfine
xc hfexch 0.8033 slater 0.1967 becke88 nonlocal 0.2107 lyp 0.6789 mp2 0.3211
dftmp2 semidirect
direct
mult 2
iterations 100
END

BSSE
mon first 1 3 4 5 mult 2
mon second 2 6
END

task dft energy


Where xyz3.xyz geometry is :

C 0.000000    0.000000    3.703909
H 0.000000    0.000000  -16.296091
H 0.000000    1.079000    3.703909
H -0.934441   -0.539500    3.703909
H 0.934441   -0.539500    3.703909
H 0.000000    0.000000  -17.039091


When I am trying to compute this input I have two diffferent error depending to the choice of basis set.

with aug-cc-pvQZ:
bsse_energy: no geometry


with 6-31G**:
mp2: MO vectors mismatch


It seems that is the MP2 part from XYG3 functional which cause the error...

What am I doing wrong?
Could you help me please?
Thank you very much!
Guillaume

Forum Vet
Guillaume
I wasn't able to reproduce your failures.
Could you upload your output files to a website?

Clicked A Few Times
Hello,

Here is the output file :

output.

I understand that it is due to a dft convergence failed for the CH3 fragment. For 6-31G** basis set, I achieved to obtain total energy when I compute separetely the fragments. It is not yet done for aug-cc-pvQZ but I am trying.

I am sorry if I wasted your time,
Guillaume

Forum Vet
I can see that a different xyz file was used for this output than the one reported in the August 27th posting ... what am I missing here?

Clicked A Few Times
Yes, I forgot. It is because I made several calculations. The only difference between the xyz file is the distance between CH3 and H2 which is 20A for the output dated 27th and 15A dated 1st.

I had the same error for both of cases..

Forum Vet
OK
First finding: if you changes the basis set to spherical, the convergence problems vanishes.


BASIS spherical
* library aug-cc-pvQZ
bqC library C aug-cc-pvQZ
bqH library H aug-cc-pvQZ
END

Clicked A Few Times
Yes, thank you very much.

I already computed my input and I have my result.

Best regards,
Guillaume

-


Forum >> NWChem's corner >> Running NWChem