timestep in MD simulation
tao liu
liut... at gmail.com
Mon Jan 4 06:46:44 UTC 2016
Dear All,
I am running a NVT MD simulation on MgSiO3 (250 atoms) at high pressure and
temperature by cp2k2.7 on ARCHER. The problem is relating to TIMESTEP.
Because it is a high pressure and temperature simulation, so I try
different TIMESTEP = 0.5 fs, 0.75 fs, 1.0 fs, 1.5 fs, 2.0 fs to run MD
simulation.
0.5 fs, 1.5 fs, and 2.0 fs works well.
But 0.75 fs, and 1 fs don't work, crash at thousand steps because reaching
the max SCF cycles. Even though I restart it, it will crash in a few of
steps.
Could anyone tell me if I got anything wrong? I prefer to use 1 fs, but
doesn't work. Thanks.
I pasted input file and bottom of .ener .stress and output file here.
Regards,
Tao
*'input file'*
<https://en.wikipedia.org/wiki/Car%E2%80%93Parrinello_molecular_dynamics#cite_note-3>&GLOBAL
PROJECT
pv-lad
RUN_TYPE
md
PRINT_LEVEL
low
&END
GLOBAL
&MOTION
&PRINT
&STRESS
&END
STRESS
&velocities
&end
velocities
&END
PRINT
&MD
ENSEMBLE NVT
STEPS 5000
TIMESTEP 1
TEMPERATURE 4000.0
&THERMOSTAT
&NOSE
LENGTH 3
YOSHIDA 3
TIMECON 10.0
MTS 2
&END NOSE
&END
&END MD
&END MOTION
&FORCE_EVAL
METHOD
Quickstep
STRESS_TENSOR
analytical
&DFT
BASIS_SET_FILE_NAME
BASIS_MOLOPT
POTENTIAL_FILE_NAME
POTENTIAL
&MGRID
CUTOFF
500
REL_CUTOFF
40
&END
MGRID
&PRINT
&DFT_CONTROL_PARAMETERS
&END DFT_CONTROL_PARAMETERS
&END PRINT
&QS
ls_scf
EPS_DEFAULT 1.0E-10
EXTRAPOLATION use_prev_p
MAP_CONSISTENT
&END QS
&LS_SCF
PURIFICATION_METHOD TRS4
MU 0.15
MAX_SCF 200
S_PRECONDITIONER ATOMIC
&END
&SCF
&OT ON
MIXED_PRECISION .TRUE.
&END OT
EPS_SCF 1.0E-5
MAX_SCF 200
SCF_GUESS atomic
&OUTER_SCF
EPS_SCF 1.0E-7
MAX_SCF 500
&END
&END SCF
&XC
&XC_FUNCTIONAL LDA
&END XC_FUNCTIONAL
&END XC
&END DFT
&PRINT
&STRESS_TENSOR
NDIGITS 4
&END STRESS_TENSOR
&END PRINT
&SUBSYS
&CELL
ABC 26.0369999999999990 9.1050000000000004 6.2900000000000000
MULTIPLE_UNIT_CELL 1 1 1
SYMMETRY ORTHORHOMBIC
&END CELL
&COORD
Mg 0.0835357257057528 0.2080505102144339
0.7695373512802808
Mg 0.4168690590390862 0.2080505102144339
0.7695373512802808
...
...
...
...
...
SCALED T
&END COORD
&TOPOLOGY
MULTIPLE_UNIT_CELL 1 1 1
&END TOPOLOGY
&KIND Mg
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PADE-q10
&END KIND
&KIND Si
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PADE-q4
&END KIND
&KIND O
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PADE-q6
&END KIND
&PRINT
&CELL
&END CELL
&KINDS
BASIS_SET
POTENTIAL
&END KINDS
&END PRINT
&END SUBSYS
&END FORCE_EVAL
*'*.ener'*
52241 99241.000000 4.411213874 3885.493746208
-5533.754419320 -5511.857390810 25.345344531
52242 99242.000000 4.342485222 3824.956044365
-5533.605897687 -5511.856215770 18.092279346
52243 99243.000000 4.276184769 3766.557154129
-5533.468202487 -5511.855599935 13.521783815
52244 99244.000000 4.219259478 3716.416111920
-5533.347713891 -5511.855260868 20.452569381
52245 99245.000000 4.174711953 3677.177676762
-5533.244993727 -5511.854024077 13.410673904
52246 99246.000000 4.157625727 3662.127754916
-5533.174592854 -5511.853076546 18.038370990
52247 99247.000000 4.175426934 3677.807447113
-5533.137836015 -5511.847932992 13.264672622
52248 99248.000000 NaN
NaN 156032.896882306 NaN
575.632388886
*'*.stress'*
52241 99241.000 1393467.1901913243 -4049.0319591435
16542.2251158790 -4049.0319591435 1342881.5952529698
6198.4426952424 16542.2251158790 6198.4426952424 1366629.9398137301
52242 99242.000 1397547.5508231323 -4579.9413527279
13810.5798040272 -4579.9413527279 1347383.1869124325
4835.2659481228 13810.5798040272 4835.2659481228 1367257.9328355824
52243 99243.000 1402562.0095211340 -2469.8827275863
11326.0268584547 -2469.8827275863 1355053.6419066861
2967.7081481965 11326.0268584547 2967.7081481965 1368330.4512780211
52244 99244.000 1408171.8438126745 888.5049041032
7965.5485159208 888.5049041032 1364397.2652421803
822.2322445308 7965.5485159208 822.2322445308 1372140.1351925349
52245 99245.000 1409730.5634058635 5552.5388752075
4525.6394441786 5552.5388752075 1370727.1093890155
-1862.2415766141 4525.6394441786 -1862.2415766141 1376614.3797829675
52246 99246.000 1412460.3088731237 9057.5282769796
628.5805615638 9057.5282769796 1378499.6190023583
-4538.0776370565 628.5805615638 -4538.0776370565 1383519.0328408128
52247 99247.000 1411091.8367309219 13731.7124595861
-2346.7333744248 13731.7124595861 1387502.2123396317
-7483.0678576032 -2346.7333744248 -7483.0678576032 1382544.9170446743
52248 99248.000 NaN
NaN NaN NaN
NaN NaN NaN
NaN NaN
*'.out'*
SCF 198 158962.937507772 1965.560497129 2.771007
*********************************************************
Using MIXING_FRACTION=0.450 to mix KS matrix: iscf=199
KS_nw=0.450*KS + 0.550*KS_old
*********************************************************
Lanczos converged: F
change in mixed matrix: 141.05387
HOMO upper bound: 104.64941
LUMO lower bound: -177.33021
Predicts a gap ? F
Est. extremal eigenvalues -69.03105 87.29441 converged: F
TRS4 it 1 1.00000000 0.897E+02 0.047 4763.272 0.100E-05
TRS4 it 2 1.00000000 0.857E+02 0.052 4321.115 0.100E-05
TRS4 it 3 1.00000000 0.635E+02 0.094 4126.422 0.100E-05
TRS4 it 4 1.00000000 0.439E+02 0.087 4013.760 0.100E-05
TRS4 it 5 0.96654122 0.289E+02 0.090 3488.278 0.100E-05
TRS4 it 6 0.95568580 0.191E+02 0.078 3775.442 0.100E-05
TRS4 it 7 0.95672906 0.125E+02 0.082 3520.711 0.100E-05
TRS4 it 8 0.96273637 0.816E+01 0.081 3564.536 0.100E-05
TRS4 it 9 0.98135529 0.537E+01 0.078 3735.249 0.100E-05
TRS4 it 10 0.98941451 0.355E+01 0.080 3692.521 0.100E-05
TRS4 it 11 0.99104224 0.234E+01 0.076 3897.213 0.100E-05
TRS4 it 12 0.99253780 0.154E+01 0.077 3886.772 0.100E-05
TRS4 it 13 0.99272503 0.102E+01 0.078 3853.237 0.100E-05
TRS4 it 14 0.99881619 0.707E+00 0.078 3893.548 0.100E-05
TRS4 it 15 0.99760219 0.494E+00 0.078 3867.742 0.100E-05
TRS4 it 16 0.99128755 0.319E+00 0.079 3805.687 0.100E-05
TRS4 it 17 0.98412362 0.194E+00 0.082 3627.809 0.100E-05
TRS4 it 18 0.95759578 0.119E+00 0.078 3733.491 0.100E-05
TRS4 it 19 0.96372829 0.735E-01 0.077 3804.604 0.100E-05
TRS4 it 20 0.98971046 0.351E-01 0.080 3743.113 0.100E-05
TRS4 it 21 0.99122111 0.764E-02 0.078 3818.051 0.100E-05
TRS4 it 22 0.99122111 0.275E-03 0.079 3795.518 0.100E-05
Final TRS4 iteration 22 0.99122111 0.275E-03
Chemical potential (mu): -37.10018
Lanczos converged: F
Lanczos converged: F
HOMO/LUMO/gap -37.15175 -36.90884 0.24291
SCF 199 157261.886998689 -1701.050509083 2.799975
*********************************************************
Using MIXING_FRACTION=0.450 to mix KS matrix: iscf=200
KS_nw=0.450*KS + 0.550*KS_old
*********************************************************
Lanczos converged: F
change in mixed matrix: 139.91979
HOMO upper bound: 102.76805
LUMO lower bound: -176.82863
Predicts a gap ? F
Est. extremal eigenvalues -68.87729 85.63565 converged: F
TRS4 it 1 1.00000000 0.905E+02 0.047 4789.695 0.100E-05
TRS4 it 2 1.00000000 0.860E+02 0.048 4640.079 0.100E-05
TRS4 it 3 1.00000000 0.629E+02 0.093 4197.692 0.100E-05
TRS4 it 4 0.99871352 0.431E+02 0.084 4141.198 0.100E-05
TRS4 it 5 0.96728110 0.284E+02 0.084 3755.189 0.100E-05
TRS4 it 6 0.95162355 0.186E+02 0.079 3730.035 0.100E-05
TRS4 it 7 0.95464254 0.121E+02 0.077 3690.826 0.100E-05
TRS4 it 8 0.95691490 0.785E+01 0.077 3676.631 0.100E-05
TRS4 it 9 0.96656933 0.508E+01 0.077 3733.906 0.100E-05
TRS4 it 10 0.98102775 0.329E+01 0.080 3647.389 0.100E-05
TRS4 it 11 0.98165356 0.215E+01 0.084 3493.869 0.100E-05
TRS4 it 12 0.98033688 0.142E+01 0.078 3737.030 0.100E-05
TRS4 it 13 0.98043352 0.922E+00 0.082 3590.334 0.100E-05
TRS4 it 14 0.98047952 0.557E+00 0.078 3774.055 0.100E-05
TRS4 it 15 0.98190189 0.302E+00 0.079 3753.784 0.100E-05
TRS4 it 16 0.99003522 0.149E+00 0.080 3713.618 0.100E-05
TRS4 it 17 0.99753877 0.652E-01 0.078 3861.570 0.100E-05
TRS4 it 18 0.99875278 0.190E-01 0.083 3665.070 0.100E-05
TRS4 it 19 0.99753877 0.147E-02 0.080 3805.444 0.100E-05
TRS4 it 20 0.98989026 0.193E-04 0.080 3729.016 0.100E-05
Final TRS4 iteration 20 0.98989026 0.193E-04
Chemical potential (mu): -36.52358
Lanczos converged: F
Lanczos converged: F
HOMO/LUMO/gap -36.71009 -36.28310 0.42699
SCF 200 156032.896882306 -1228.990116383 2.659866
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160103/62917c29/attachment.htm>
More information about the CP2K-user
mailing list