[lammps-users] melting through LJ potentail

Hi

I want to make Cu-Zr metallic glass using LJ(12-6)potential.

To do this, I tried to melt in two ways

  1. increase temperature with NPT

But, I couldn’t find the melting temeperature ( their was only linear relationship between temperature vs potential and volume )

here is my inputfile script

units metal
boundary p p p
atom_style atomic

read_data initial.txt

pair_style lj/cut 7.037
pair_coeff 1 1 0.82873 2.81524
pair_coeff 1 2 0.61044 2.54405
pair_coeff 2 2 0.44966 2.27286

velocity all create 10 654321

fix 1 all npt 10.0 2500 0.001 aniso 0.0 0.0 0.0 0.0 0.0 0.0 1
fix 3 all rdf 1000 rdf.out 200 1 1 1 2 2 1 2 2

compute 1 all pe/atom
compute 2 all ke/atom
compute 3 all coord/atom 3.3

thermo 1000
thermo_modify norm yes
thermo_style custom step temp press pe ke etotal enthalpy vol lx ly lz

dump 1 all custom 10000 CuZr_.dat type x y z c_1 c_2
dump 2 all custom 10000 CuZr-coor_
.dat type c_3

timestep 0.001
run 1000000

unfix 1
fix 1 all npt 2500 2500 0.001 aniso 0.0 0.0 0.0 0.0 0.0 0.0 1

run 500000

unfix 1
fix 1 all npt 2500 10.0 0.001 aniso 0.0 0.0 0.0 0.0 0.0 0.0 1

run 225000

unfix 1
fix 1 all npt 10 10.0 0.001 aniso 0.0 0.0 0.0 0.0 0.0 0.0 1

run 500000

write_restart amorphous_making.lmp

Do i have a incrrect command to melt Cu-Zr??

  1. The other way is to make liquid directly

But, temp. temperature was decrease to 1000L and increas to 4500K again

And when i see the final atom configurations I still maintain the B2 structure. (normally liquid is randomly distributed …)

Here is the inputfile

units metal
boundary p p p
atom_style atomic

read_data initial.txt

pair_style lj/cut 7.037
pair_coeff 1 1 0.82873 2.81524 (1 : Zr 2 : Cu)
pair_coeff 1 2 0.61044 2.54405
pair_coeff 2 2 0.44966 2.27286

velocity all create 2500.0 49284598 rot yes dist gaussian

fix 1 all nvt 2500 2500 10

compute 1 all pe/atom
compute 2 all ke/atom
compute 3 all coord/atom 3.3

thermo 100
thermo_modify norm yes
thermo_style custom step temp press pe ke etotal enthalpy vol lx ly lz

timestep 0.001
run 100000

write_restart luquid.lmp

I don’t know why it doesn’t melt…

Can you give me a advice??

Thanks !!

Sincerely.

Mirim, Lee

Determining melting temperatures accurately is non-trivial.
But to simply melt a solid, raise the temperature while
keeping the volume fixed. Small or tiny systems can
have pathologies due to periodic boundaries.

Steve

2009/5/21 이미림 <[email protected]...>: