[molpro-user] Part 1: Density-Fitting + ECP basis set for K

Hanusha Bhakhoa hbhakhoa at gmail.com
Fri Feb 13 04:31:05 GMT 2015


Dear all,


Greetings.


I want to perform a local correlated ccsd(t) computation using the
density-fitting approximation for a system containing C, N, H and K atoms.


So, basically either an all-electron or an ECP basis set can be used for
the K atom.


*Using the ECP10MWB basis set for K and avdz basis set for C, N, H atoms:*

(1)    Trial 1 was okay with the following jkfit basis set:



basis

default,avdz

K=ECP10MWB



set,dfjk,

default,avdz/jkfit

spdfg,K,def2-qzvpp/jkfit,c

end



{df-hf,df_basis=dfjk;wf,126,1,0}



(2)    Trial 2 was okay with the following mp2fit basis set:



basis

default,avdz

K=ECP10MWB



set,mp2fit,

default,avdz/mp2fit

spdf,K,def2-atzvpp/mp2fit;c

end



{df-hf,df_basis=mp2fit;wf,126,1,0}



(3)    Trial 3 (combination of the jkfit and mp2fit in a single block) was
okay.



basis

default,avdz

K=ECP10MWB



set,mp2fit,

default,avdz/mp2fit

spdf,K,def2-qzvpp/mp2fit;c



set,dfjk,

default,avdz/jkfit

spdfg,K,def2-qzvpp/jkfit;c

end



{df-hf,df_basis=dfjk;wf,126,1,0}



(4)    However, when I add the command ‘DF-LCCSD(T)’ to the input file, I
end up with an error termination:



basis

default,avdz

K=ECP10MWB



set,mp2fit,

default,avdz/mp2fit

spdf,K,def2-qzvpp/mp2fit;c



set,dfjk,

default,avdz/jkfit

spdfg,K,def2-qzvpp/jkfit;c

end



{df-hf,df_basis=dfjk;wf,126,1,0}

DF-LCCSD(T)



Error:



Cannot find default basis ECP10MWB for atom K

 Type=FIT

 Context=MP2FIT



 Please specify a default basis or define basis sets for all atoms!



Am I missing something in the input file? -- Can you please advise how to
solve this problem?



Thank you.



Kind regards,

Hanusha


---------------------------------

Hanusha Bhakhoa

MPhil/PhD scholar

Computational Chemistry Group

University of Mauritius
*http://sites.uom.ac.mu/ccuom <http://sites.uom.ac.mu/ccuom>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20150213/deae73e1/attachment.html>


More information about the Molpro-user mailing list