[molpro-user] JKFIT basis set for Na+ ? (density fitting basis set)

[Grant Hill]

Try using def2-qzvpp/jkfit for Na.

E.g.,

***,Test of JK fitting for Na
memory,40,m

geom={Na}

basis={
default,wcvdz

set,dfjk
na=def2-qzvpp/jkfit
}

{df-hf,df_basis=dfjk;wf,10,1,0}

---

HTH,

Grant

On 7 Nov 2014, at 16:35, Hanusha Bhakhoa <hbhakhoa at gmail.com> wrote:

> Dear All,
> 
> I am trying a local correlated density fitting computation at CCSD(T) level [DF-LUCCSD(T)]. 
> 
> My molecule (neutral molecule) is rather big (36 atoms) which comprises of Na+ cation. Therefore, I want to consider the core electrons (2s and 2p) of Na+ in the local correlated computation. 
> 
> I opted for a correlation consistent basis set which considers core correlation, namely the wcvdz for sodium. The following is the basis set section from my input file:
> 
> "basis={
> default,avdz
> Na=wcvdz
> set,mp2fit
> default,avdz/mp2fit
> Na=wcvdz/mp2fit
> set,jkfit
> default,avdz/jkfit
> Na=def2-SVP-JKFIT 
> }"
> 
> However, my output file suggests that the JKfit basis set for sodium is not available in the library (despite the fact that molpro library provide some jkfit basis set for sodium). Running the above molpro computation results in 
> 
> "Basis set:  MP2FIT                  
>  Basis size:  1657
>  Attributes:
>  Sphericals:   T
>  LIBRARY EXHAUSTED
>   Searching for Na I DEF2-SVP-JKFIT      
>   Library contains the following bases:
>  ? Error
>  ? Basis library exhausted
>  ? The problem occurs in Binput"
> 
> Please also note that I tried other JKfit basis set namely the univ-JKFIT , univ-JFIT , etc. Yet, it seem that the software does not recognize these basis set for sodium.
> 
> Is it a problem related to the software or I am going the wrong way?
> 
> Please advice.
> 
> Thank you.
> 
> Kind regards,
> Hanusha
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user