problem of simulation of alloy melts

Hello all users,

I tried to model alloy melts by lammps, but found that the structure of the model remains crystalline.
So could anyone tell me where is the problem of my codes?

units metal
atom_style atomic
dimension 3
boundary p p p

lattice custom 2.8 a1 1.0 0.0 0.0 a2 0.0 1.0 0.0 a3 0.0 0.0 1.0 basis 0 0 0
region simulation block 0 10 0 10 0 10
create_box 2 simulation
create_atoms 1 region simulation basis 1 1
lattice custom 2.8 a1 1.0 0.0 0.0 a2 0.0 1.0 0.0 a3 0.0 0.0 1.0 basis 0.5 0.5 0.5
create_atoms 2 region simulation basis 1 2

group Mg type 1
group Al type 2

#EAM potential

pair_style eam/fs
pair_coeff * * Al-Mg.eam.fs Mg Al

timestep 0.001
neighbor 2.5 bin

thermo_style custom step temp enthalpy etotal pe ke press vol lx

thermo 1000

run_style verlet
velocity all create 1800 12345 dist gaussian rot yes loop geom

minimize 1e-6 1e-6 100000 100000

#equilibration

dump 1 all xyz 10000 al50mg50.xyz

fix 1 all npt temp 1800 1800 0.05 iso 0 0 0.5 drag 0.8
run 1000000
write_restart al50mg50.1800.restart
unfix 1

Thanks and regards,

Tang Chao

Melting is a complex thing to simulate. Many things can
effect how easy it is to melt. System size (you appear to
be running a small system), perioidic boundaries, nucleation,
etc. I suggest you read the literature and try to mimic what
is in some good paper as a first step.

Steve