[lammps-users] energy minimization

LAMMPS LAMMPS Mailing List Mirror
1



Dear All,

I am doing energy minimization on gold nanowire. I minimized the wire twice. I tabulated (below) the length and bounds of the simulation box. The dimensions of the box at the beginning of the 2nd minimization step should be the same as those at the end of the 1st minimization step. But in my case its not coming out to be the same. Input script is attached below.
Can anyone help me
Xlo Xhi Lx
End of 1st minimization 2.39 160.80 158.41
Beginning of 2nd minimization 3.24 160.15 156.91

units metal
boundary s s s

atom_style atomic
lattice fcc 4.08

region box block 0 40 0 6 0 6
create_box 2 box
create_atoms 2 box

pair_style eam
pair_coeff * * Au_Foiles.eam

neighbor 2 bin
neigh_modify delay 1

region lower block INF 2 INF INF INF INF
region upper block 38 INF INF INF INF INF
group upper region upper
group lower region lower

# initial minimization
dump 1 all xyz 100 dump.0K.minimize
thermo 100
thermo_style custom step xlo xhi lx
minimize 1.0e-14 1.0e-16 100000 100000


# second minimization step

thermo 50
thermo_style custom step xlo xhi lx
minimize 1e-14 1.0e-16 100000 100000

Thanks in advance
Sivashankar



|

2

Why shouldn’t the box change - you are using shrink-wrap boundaries?

Steve

3

Hi steve,
The wire is completely minimized in the first minimization step. If it is so, then what drives the wire to contract (even if i use shrink-wrapped boundaries) in the second minimization step which has the same minimum energy as that of at the end of the first minimization step. The wire should not have two different configurations at the same minimum energy?.

Thanks
Sivashankar

4

I don’t know. Have you dumped out the atom coords and verified
that they have moved and that the box has shrunk around them
accordingly?

Steve

5

Hi steve,
the atom coordinates remain the same, but the length of the wire is changing.
it seems i should rely on the atom coordinates than the out put of thermo_style command. how can i calculate the length of the wire using atom coordinates.

thanks
sivashankar

6

You can take min/max values of atom vectors with the compute reduce
command.

Steve