| 12:31:09 AM PST - Mon, Feb 6th 2012 |
|
Hi,
I have been observing the evolution of NWChem for quite sometime
and really interested in the use NWChem for high level calculations in distributed cluster. And recently I'd started to try it.
As my first test, I want to compare the CCSD(t) energy calculation in molpro and NWChem.
The calculation here is to determine the molecular interaction between CO2 and N2O.
I have used the input file below for molpro calculation:
memory,200,m
geometry={
nosymm;
angstrom;
He
X 1 1.0
C 1 r0 2 90.0
X 3 1.0 1 90.0 2 b1
X 3 1.0 1 90.0 4 90.0
O1 3 rc 5 90.0 1 a1
O2 3 rc 5 90.0 6 180.0
X 1 r0 2 90.0 3 180.0
X 8 1.0 1 90.0 2 0.0
X 8 1.0 9 90.0 1 90.0
O3 8 r1 10 90.0 1 a2
N3 8 r2 10 90.0 11 180.0
N4 8 r3 10 90.0 11 180.0}
rc=1.162047
r1=1.113058
r2=0.072042
r3=1.199342
r0= 3.5000
a1= 60.0000
a2=120.0000
b1= 90.0000
basis={
default=avtz;
s,he,0.9,0.3,0.1;
p,he,0.9,0.3,0.1;
d,he,0.6,0.2;
f,he,0.3}
text,total energy for vdw complex
dummy he
hf
ccsd(t)
etot=energy
text,energy of co2
dummy he,o3,n3,n4
hf
ccsd(t)
en2o1=energy
text,energy of n2o
dummy he,c,o1,o2
hf
ccsd(t)
en2o2=energy
de=(etot-en2o1-en2o2)*tocm
show etot,en2o1,en2o2,de
---
Just clueless about how to make a workable nwchem input for it especially in BSSE and bond functions with He.
Appreciate your help very much.
best,
Lu Yunpeng
|