Hi, siesta users,
I am using siesta to do some exercises. In the geometry optimization of
BaTiO3, Si bulk, the program preceed only one CG move and then stopped without
any warnning or error. I have set the MD.NumCGsteps to 50, 300 and other
non-zero numbers. This problem really confused me. Is there anybody ever
encontered this problem? Please give me a hand.
The Si.structure.fdf:
SystemName Bulk Silicon
SystemLabel Si
NumberOfSpecies 1
NumberOfAtoms 2
%block ChemicalSpeciesLabel
1 14 Si
%endblock ChemicalSpeciesLabel
LatticeConstant 3.43 Ang
%block LatticeVectors
0.00 0.50 0.50
0.50 0.00 0.50
0.50 0.50 0.00
%endblock LatticeVectors
AtomicCoordinatesFormat ScaledByLatticeVectors
%block AtomicCoordinatesAndAtomicSpecies
0.00 0.00 0.00 1
0.25 0.25 0.25 1
%endblock AtomicCoordinatesAndAtomicSpecies
%block kgrid_Monkhorst_Pack
4 0 0 0.5
0 4 0 0.5
0 0 4 0.5
%endblock kgrid_Monkhorst_Pack
MD.TypeOfRun CG
#MD.VariableCell .true.
MD.NumCGsteps 300
#MD.MaxStressTol 0.0005 eV/Ang**3
#MD.MaxForceTol 0.05 eV/Ang
MD.UseSaveCG = .true.
MD.UseSaveXV = .true.
The Si.structure.out :
Siesta Version: siesta-4.0b-494
Architecture : x86_64-unknown-linux-gnu--unknown
Compiler flags: mpif90 -g -O2
PP flags : -DMPI -DFC_HAVE_FLUSH -DFC_HAVE_ABORT
PARALLEL version
* Running on 24 nodes in parallel
>> Start of run: 26-AUG-2018 19:03:57
......
superc: Internal auxiliary supercell: 5 x 5 x 5 = 125
superc: Number of atoms, orbitals, and projectors: 250 3250 4000
====================================
Begin CG opt. move = 0
====================================
superc: Internal auxiliary supercell: 5 x 5 x 5 = 125
superc: Number of atoms, orbitals, and projectors: 250 3250 4000
outcell: Unit cell vectors (Ang):
0.000000 2.715000 2.715000
2.715000 0.000000 2.715000
2.715000 2.715000 0.000000
outcell: Cell vector modules (Ang) : 3.839590 3.839590 3.839590
outcell: Cell angles (23,13,12) (deg): 60.0000 60.0000 60.0000
outcell: Cell volume (Ang**3) : 40.0258
New_DM. Step: 1
Initializing Density Matrix...
New grid distribution: 1
1 1: 10 1: 3 1: 2
2 1: 10 1: 3 3: 4
3 1: 10 1: 3 5: 6
4 1: 10 1: 3 7: 8
5 1: 10 1: 3 9: 9
6 1: 10 1: 3 10: 10
7 1: 10 4: 6 1: 2
8 1: 10 4: 6 3: 4
9 1: 10 4: 6 5: 6
10 1: 10 4: 6 7: 8
11 1: 10 4: 6 9: 9
12 1: 10 4: 6 10: 10
13 1: 10 7: 8 1: 2
14 1: 10 7: 8 3: 4
15 1: 10 7: 8 5: 6
16 1: 10 7: 8 7: 8
17 1: 10 7: 8 9: 9
18 1: 10 7: 8 10: 10
19 1: 10 9: 10 1: 2
20 1: 10 9: 10 3: 4
21 1: 10 9: 10 5: 6
22 1: 10 9: 10 7: 8
23 1: 10 9: 10 9: 9
24 1: 10 9: 10 10: 10
InitMesh: MESH = 20 x 20 x 20 = 8000
InitMesh: (bp) = 10 x 10 x 10 = 1000
InitMesh: Mesh cutoff (required, used) = 100.000 112.482 Ry
ExtMesh (bp) on 0 = 54 x 47 x 46 = 116748
New grid distribution: 2
1 6: 10 1: 5 1: 2
2 6: 10 4: 5 3: 5
3 6: 10 1: 5 6: 7
4 6: 10 1: 3 8: 10
5 1: 5 1: 5 1: 2
6 6: 10 1: 3 3: 5
7 1: 5 6: 10 1: 2
8 1: 5 6: 7 3: 5
9 1: 5 1: 5 6: 7
10 1: 5 1: 3 3: 5
11 1: 6 6: 7 8: 10
12 1: 5 4: 5 3: 5
13 1: 5 1: 3 8: 10
14 6: 10 6: 7 3: 5
15 1: 6 6: 10 6: 7
16 7: 10 6: 10 6: 7
17 6: 10 4: 5 8: 10
18 1: 5 4: 5 8: 10
19 6: 10 6: 10 1: 2
20 6: 10 8: 10 3: 5
21 1: 5 8: 10 3: 5
22 1: 6 8: 10 8: 10
23 7: 10 8: 10 8: 10
24 7: 10 6: 7 8: 10
New grid distribution: 3
1 6: 10 1: 5 1: 2
2 1: 5 1: 3 3: 5
3 1: 5 1: 5 6: 7
4 6: 10 9: 10 3: 5
5 6: 10 1: 5 6: 7
6 6: 10 1: 3 3: 5
7 1: 5 1: 5 1: 2
8 6: 10 4: 5 3: 5
9 6: 10 6: 10 6: 7
10 1: 5 1: 3 8: 10
11 1: 5 9: 10 8: 10
12 1: 5 4: 5 3: 5
13 6: 10 6: 10 1: 2
14 6: 10 6: 8 3: 5
15 6: 10 4: 5 8: 10
16 1: 5 6: 8 8: 10
17 6: 10 1: 3 8: 10
18 6: 10 6: 8 8: 10
19 1: 5 6: 10 1: 2
20 1: 5 9: 10 3: 5
21 1: 5 6: 10 6: 7
22 1: 5 6: 8 3: 5
23 6: 10 9: 10 8: 10
24 1: 5 4: 5 8: 10
Setting up quadratic distribution...
ExtMesh (bp) on 0 = 49 x 49 x 46 = 110446
PhiOnMesh: Number of (b)points on node 0 = 50
PhiOnMesh: nlist on node 0 = 4120
stepf: Fermi-Dirac step function
siesta: Program's energy decomposition (eV):
siesta: Ebs = -76.445406
siesta: Eions = 380.802124
siesta: Ena = 115.502327
siesta: Ekin = 78.997046
siesta: Enl = 35.832836
siesta: DEna = 0.000000
siesta: DUscf = 0.000000
siesta: DUext = 0.000000
siesta: Exc = -64.548619
siesta: eta*DQ = 0.000000
siesta: Emadel = 0.000000
siesta: Emeta = 0.000000
siesta: Emolmec = 0.000000
siesta: Ekinion = 0.000000
siesta: Eharris = -215.515303
siesta: Etot = -215.018534
siesta: FreeEng = -215.018534
scf: iscf Eharris(eV) E_KS(eV) FreeEng(eV) dDmax Ef(eV)
scf: 1 -215.5153 -215.0185 -215.0185 1.82315 -4.1671
timer: Routine,Calls,Time,% = IterSCF 1 0.686 38.93
scf: 2 -215.4754 -215.4703 -215.4703 0.01765 -3.9736
scf: 3 -215.4748 -215.4715 -215.4715 0.01191 -3.9829
scf: 4 -215.4746 -215.4723 -215.4723 0.00803 -3.9892
scf: 5 -215.4745 -215.4728 -215.4728 0.00541 -3.9935
scf: 6 -215.4744 -215.4732 -215.4732 0.00364 -3.9964
scf: 7 -215.4744 -215.4735 -215.4735 0.00244 -3.9984
scf: 8 -215.4744 -215.4737 -215.4737 0.00164 -3.9998
scf: 9 -215.4744 -215.4739 -215.4739 0.00110 -4.0007
scf: 10 -215.4744 -215.4740 -215.4740 0.00073 -4.0013
scf: 11 -215.4743 -215.4741 -215.4741 0.00049 -4.0017
scf: 12 -215.4743 -215.4742 -215.4742 0.00033 -4.0020
scf: 13 -215.4743 -215.4742 -215.4742 0.00022 -4.0022
scf: 14 -215.4743 -215.4742 -215.4742 0.00014 -4.0023
scf: 15 -215.4743 -215.4743 -215.4743 0.00009 -4.0024
SCF Convergence by dMax criterion
max |DM_out - DM_in|: 0.00009461
SCF cycle converged after 15 iterations
Using DM_out to compute the final energy and forces # This is the last line
of the output file.
Any suggestion, comment or help will be highly appreciated. Thank you!
Tao Hu
-- Mob. : 17824225125
Institute of Materials Science & Devices
Suzhou University of Science and Technology
No. 1 Kerui Road, Suzhou 215009, China
