For more than 300 electron: fixing electrons in MO


Clicked A Few Times
Hello

Say I have a 19 atom Pt cluster, which has 190 electrons in total (10 electrons per Pt). Correspondingly, there are 19*5=95 d-orbital of Pt. I want to fill these 95 d-orbitals with 95 up-spin electrons and 81 down spin electrons. Remaining 14 electrons go into seven of the 6s levels. Besides, the LANL2DZ_ecp basis set that I am using offers 18 electrons as valence electrons for Pt. However, I want to only fix the earlier mentioned 5d electronic occupations. How can I deal with this problem using OCCUP/or any other keyword? I tried mult keyword however, upon energy minimization the occupations do not stay fixed to the set values (refer below).

I also read in this forum that OCCUP keyword can only be used with single point energy calculations. Which restricts my future calculations of adsorbates on this Pt cluster as I require geometry optimization for the adsorbates. Any hints on how to go about my problem of fixing occupation?

Thank you.

Test calculation with 6 atom Pt system and occupations corresponding to energy minimum:
MULT=3
                 Occupations of the irreducible representations
----------------------------------------------

                    irrep           alpha         beta
-------- -------- --------
ag 27.0 27.0
au 27.0 27.0

MULT=5
                 Occupations of the irreducible representations
----------------------------------------------

                    irrep           alpha         beta
-------- -------- --------
ag 27.6 26.2
au 27.2 27.0

MULT=7
                 Occupations of the irreducible representations
----------------------------------------------

                    irrep           alpha         beta
-------- -------- --------
ag 28.0 24.0
au 29.1 26.9

MULT=9
                 Occupations of the irreducible representations
----------------------------------------------

                    irrep           alpha         beta
-------- -------- --------
ag 28.0 24.0
au 29.1 26.9

Input file:
echo
geometry autosym
Pt -0.831261 -0.804522 -1.403607
Pt -0.831261 -0.804522 1.403607
Pt -0.822937 -3.235626 0.000000
Pt 1.458032 0.013694 0.000000
Pt 1.466355 -2.417410 -1.403607
Pt 1.466355 -2.417410 1.403607
end

basis "large" cartesian
Pt library lanl2dz_ecp file /usr/local/NWChem/data/libraries/
end

ecp
Pt library lanl2dz_ecp file /usr/local/NWChem/data/libraries/

end
set "ao basis" "large"

dft
iterations 1000
direct
mult 3
XC xpbe96 cpbe96
convergence ncyds 1000 damp 70 ncydp 500 diis 16
smear 0.001
end

DRIVER
XYZ a.xyz
MAXITER 500
END

task dft energy

Clicked A Few Times

Gets Around
Quote:Nwchemy Jun 16th 9:25 am
Hello

Input file:
echo
geometry autosym
Pt -0.831261 -0.804522 -1.403607
Pt -0.831261 -0.804522 1.403607
Pt -0.822937 -3.235626 0.000000
Pt 1.458032 0.013694 0.000000
Pt 1.466355 -2.417410 -1.403607
Pt 1.466355 -2.417410 1.403607
end

basis "large" cartesian
Pt library lanl2dz_ecp file /usr/local/NWChem/data/libraries/
end

ecp
Pt library lanl2dz_ecp file /usr/local/NWChem/data/libraries/

end
set "ao basis" "large"

dft
iterations 1000
direct
mult 3
XC xpbe96 cpbe96
convergence ncyds 1000 damp 70 ncydp 500 diis 16
smear 0.001
end

DRIVER
XYZ a.xyz
MAXITER 500
END

task dft energy


Hi.

Just set NWCHEM_BASIS_LIBRARY environment variable point to the NWChem data libraries directory instead of using
Pt library lanl2dz_ecp file <whatever_dir>
detail explanation - http://nwchemgit.github.io/index.php/Compiling_NWChem#General_site_installation

Vladimir.


Forum >> NWChem's corner >> General Topics