My Community

Hi

I have a question of calculation in the O2 molecule and the Au25 cluster by using VASP.

According to literature, the most stable electronic state of O2 molecule should be in the triplet state (mag=2.00), but my optimization result shows that O2 molecule is in singlet state (mag=0.00).

I ensure that I have set the 'ISPIN=2' parameter in INCAR.



The similar problem was also encountered when I calculated the Au25 cluster.

Because electronic configuration of Au atom is 5d10 6s1, the Au25 cluster should be in the doublet state( mag=1.00).


However, I still got the singlet state result (mag=0.00).

This is an unphysical solution! !




Following is my INCAR file, which is used to calculate Au25 cluster :

SYSTEM = Au25

ISTART = 0 ! Job : 0-new 1-cont 2-samecut
ICHARG = 2 ! Charge Density: 0-ini wavefunction; 1-read CHGCAR; 2-superposition; +10-non self
LORBIT = 11 ! Files Written: 11-DOSCAR+PROCAR+Phase Factors
NWRITE = 2 ! Information written to OUTCAR: 2-Default

NPAR = 1 ! 1-one node; 2-two nodes; 4-four nodes;
NSIM = 4
LPLANE = .True. !
IALGO = 48 ! Ele Minim. Algorithm: 38-Normal; 48-VeryFast; Fast

NGX=190
NGY=190
NGZ=190


ISYM = 0 ! 0-broken symm.
ISMAEAR = 0 ! 0-Gaussian method
SIGMA = 0.1 ! smearing width (eV)


ENCUT = 500 ! Cutoff Energy (eV)
EDIFF = 1E-6 ! SCF criterion (eV)
EDIFFG = -0.01 ! Force criterion (eV)
NELMIN = 4 ! Numer of minimum SCF cycles


NSW = 35 ! OPT Cycle
IBRION = 2 ! 1-quasi, 2-CG, 3-Damp MD
POTIM = 0.5 ! Step size
ISIF = 0 ! caculate forces on atoms, relax atoms

ISMEAR = 0 ! Gaussian
SIGMA = 0.1 eV ! DOS related value


GGA = 91 ! 91-PW91; PE-PBE; RP-RPBE
VOSKOWN=1 ! the Algorithm for GGA-PW91
ISPIN = 2 ! 2-Spin polarized



PS. the Au25 cluster is put in a box with the size of 25.00*25.10*25.20.



Thanks in Advance and your suggestions.


x)
<span class='smallblacktext'>[ Edited ]</span>

Hi

I think I have found the reason.

Because I used the parameter IALGO = 48 in INCAR, this setting causes a wrong result.

If IALGO=48 is changed into IALGO=38, then the electronic state of O2 molecule becomes triplet state.

This result shows that the IALGO=48 cannot predict electronic configuration correctly.

However, I do not understand why IALGO=48 is unreliable in predicting the electronic configuration of gas molecules.

Does the result means that one should use IALGO=48 carefully in the surface systems?