Sample run of DIMER method and problems there in

[staar]

HI All,

This is the sample dimer run I did for the already found TS using NEB method. And the DIMCAR shows that it is not convergent and the calculation is going on. MODECAR is generated randomly by the run.

TRANSITION STATE POSCAR
Ag (110)
4.13999999999999968
1.4142135623730001 .0000000000000000 .0000000000000000
.0000000000000000 2.0000000000000000 .0000000000000000
.0000000000000000 .0000000000000000 4.2426406871192848
2 8
Selective dynamics
Direct
.0850010409260394 .2042979277222553 .4740434826856257 T T T
.2122979277222553 .0850010409260394 .4418827730647054 T T T
.0000000000000000 .0000000000000000 .5027363294221630 F F F
.5000000000000000 .0000000000000000 .5027363294221630 F F F
.0000000000000000 .5000000000000000 .5027363294221630 F F F
.5000000000000000 .5000000000000000 .5027363294221630 F F F
.2500000000000000 .2500000000000000 .5794737571239494 F F F
.7500000000000000 .2500000000000000 .5794737571239494 F F F
.2500000000000000 .7500000000000000 .5794737571239494 F F F
.7500000000000000 .7500000000000000 .5794737571239494 F F F

INCAR

SYSTEM = Typical Dimer Run
IBRION=3
POTIM=0.0
ISYM=0
NSW=600
EDIFF=1E-7
EDIFFG=-0.001
PREC=Acc
LREAL=.FALSE.
LWAVE=.FALSE.
LCHARG=.FALSE.
# DIMER PARAMETERS
ICHAIN=2
DdR=0.005
DRotMax=3
DFNMin=0.01
DFNMax=1.0
# OPTIMIZER PARAMETERS
IOPT=2

DIMCAR

Step Force Torque Energy Curvature Angle
1 0.36527 3.16209 -22.21719 1.32348 59.92008
1 0.36527 2.79483 -22.21719 -4.15650 39.46761
1 0.36527 1.22669 -22.21719 -5.26669 12.09358
2 0.13641 0.58634 -22.21355 -4.82130 4.36460
3 0.11262 0.63080 -22.22313 -4.43671 7.19743
4 0.06494 0.25315 -22.22641 -4.35304 1.90205
5 0.04837 0.16793 -22.22719 -4.60616 1.57309
6 0.16937 0.26048 -22.23236 -4.01375 3.03205
7 0.24201 0.38277 -22.24545 -3.45547 4.14278
8 0.20423 0.47952 -22.25513 -4.01147 5.57978
9 0.22811 0.78364 -22.27751 -3.37902 9.14871
10 0.28068 0.50867 -22.30596 -2.87559 8.51886
11 0.27556 0.62141 -22.33491 -3.06995 7.65180
12 0.21726 0.53998 -22.33775 -3.98292 3.58778

Could you please tell me where is the mistake I am doing. Its a established result and why it is not converging? Dimcar is not finished, still the calculation is running. Angle is not less than 1 as it is mentioned in website and forces are not dropping? Finally energy is decreasing and reaching to some other state.

Please guide me where I did the mistake?

[graeme]

I think this run is looking good. Clearly the initial mode was not along the lowest curvature direction, but it's a good sign that a strong negative mode was found in the first step. In subsequent steps, the dimer is following that mode uphill (while refining it), and the other modes downhill. Keep this run going. If it is converging to a nearby saddle, the force will eventually start dropping. If the curvature ever goes positive, then you worry that it is moving to a different saddle region, but for now, it all looks fine.