CLUPAN v0.7 documentation

9. Structure selection for DFT calculations

«  8. Optimization of clusters   ::   Contents   ::   10. Ground state search  »

9. Structure selection for DFT calculations

9.1. Theoretical backgrounds

See [CE1] to find how to select DFT structures.

9.2. search_poscar_variance

  • Searching for a DFT structure for validating and improving a trial CE using the simulated annealing.
  • DFT structures used for constructing the trial CE is specified by CORRELATION file.
  • Available only for binary system.

Input files

Output files

  • POSCAR_new

9.3. MKPOS.in

Searching for a structure with the lattice constructed by the 2x2x1 expansion of a 4-atoms unit cell (16-atoms cell). The structure has 12 A-atoms (spin=+1) and 4 B-atoms (spin=-1). The structure is optimized for clusters specified by CLUSTERINDEX.

ISUB = 1
NUCELL = 2 2 1
NCHANGE = 4
SPIN = 1 -1
NALLCLUSTER = 53
CLUSTERINDEX = 0 1 2 3 4 5 8 16 35 49 48
TEMPINIT = 1000
TEMPFINAL = 1
TEMPMUL = 0.99
NSTEP = 20

9.3.1. ISUB tag

Sublattice indexes for considering atomic configurations. Atomic configurations are considered on ISUBth lattice sites of UPOSCAR (line 6).

  • Default : 1
  • Example : ISUB = 1 3

9.3.2. NCHANGE tag

Number of atoms. In N-nary systems, set N-1 kinds of atoms. The number of the other kind of atoms is automatically set to (number of lattice sites - NCHANGE).

  • Default : none
  • Example : NCHANGE = 16

9.3.3. SPIN tag

Spin values of atoms. In binary systems, SPIN tag has two kinds of values. The same correspondence between spins and atom types used in correlation should be used here.

  • Default : none
  • Example : SPIN = 1 -1

In the above example, spin values of NCHANGE atoms are set to -1.

9.3.4. NUCELL tag

A structure is searched for using the NUCELL expansion of the unit cell.

  • Default : none
  • Example : NUCELL = 2 2 2

9.3.5. NALLCLUSTER tag

Number of all clusters included in CORRELATION file.

  • Default : none
  • Example : NALLCLUSTER = 53

9.3.6. CLUSTERINDEX tag

Indexes of clusters used for the trial CE.

  • Default : none
  • Example : CLUSTERINDEX = 0 1 2 3 4 5 8 16 35 48 49

9.3.7. TEMPINIT tag, TEMPFINAL tag

Initial and final temperatures in the simulated annealing.

  • Default : none
  • Example : TEMPINIT = 1
  • Example : TEMPFINAL = 0.001

9.3.8. TEMPMUL tag

Exponential base for an automatic setting of temperatures in the simulated annealing. Set values ranged from 0 to 1.

  • Default : none
  • Example : TEMPMUL = 0.9

9.3.9. NSTEP tag

Number of steps for simulated annealing at one temperature.

  • Default : none
  • Example : NSTEP = 20

9.4. variance_energy

  • Calculating the mean variance of the predicted energy.

Input files

Output files

  • Mean variance of the predicted energy

9.5. variance_stpop

  • Under development.
  • Generating VARIANCE and MEAN files required in search_poscar_variance.
  • VARIANCE and MEAN files can be generated by calculating correlation functions using correlation for random structures generated by mkposcar and calculating their means and variances.

Input files

Output files

  • VARIANCE
  • MEAN

«  8. Optimization of clusters   ::   Contents   ::   10. Ground state search  »