[lammps-users] Not conserved energy in NVE of alloy melt

Dear all,

I faced some problems; in NVE simulation of alloy melt (Al-Cu), the total energy is increased, the temperature and pressure are also increased.

The procedures are as follow;
1) NPT simulation of Al-Cu to get the equilibrium volume at p=0 pa and T = 2500 K: dt = 5 fs, time step = 50000
2) NVE simulation of Al-Cu melt at obtained volume V, and T = 2500 K: dt = 1fs, time step = 50000.

Input script is as follow;
# Getting the volume of liquid by NPT ensemble simulation
units metal
boundary p p p
atom_style atomic

lattice custom 3.7 a1 1.0 0.0 0.0 a2 0.0 1.0 0.0 a3 0.0 0.0 1.0&
    basis 0.0 0.0 0.0 basis 0.5 0.5 0.0 basis 0.5 0.0 0.5 basis 0.0 0.5 0.5
region box block 0 8.0 0 8.0 0 8.0
create_box 2 box
create_atoms 2 box basis 1 1 basis 2 2 basis 3 2 basis 4 2

pair_style eam/alloy
pair_coeff * * AlCu.eam.alloy Al Cu

neighbor 0.3 bin
neigh_modify delay 5

displace_atoms all random 0.35 0.35 0.35 4598345 units box
velocity all create 2500.0 4928459 rot yes dist gaussian
fix 1 all npt 2500.0 2500.0 10.0 xyz 0.0 0.0 10.0 drag 0.7
thermo 100

thermo_style custom step temp press vol etotal cpu
timestep 0.005

restart 50000 liq_2500_npt
dump 1 all atom 100 dump2500.npt
log log.liq_2500_npt
run 50000

# Collect the data by NVE ensemble simulation
clear
read_restart liq_2500_npt.50000

pair_style eam/alloy
pair_coeff * * AlCu.eam.alloy Al Cu

neighbor 0.3 bin
neigh_modify delay 5

fix 1 all nve
fix 2 all com 100 com.out
fix 3 all coord/original
fix 4 all msd 100 diff-2500.out
fix 5 all rdf 100 rdf-2500.out 500 1 1 1 2 2 1 2 2
fix Ther all thermal/conductivity 100 z 20
fix Visc all viscosity 100 x y 20

thermo_style custom step temp press vol etotal pe ke pxy
timestep 0.001
thermo 100

log log.liq_2500_nve
dump 1 all custom 100 dump-2500.coord tag x y z
dump 2 all custom 100 dump-2500.velocity tag vx vy vz
dump 4 all custom 100 dump-2500.coordorig f_3[1] f_3[2] f_3[3]
reset_timestep 0
run 50000

The output in NVE is as follow;
Step Temp Press Volume TotEng PotEng KinEng
        0 2496.5782 973.1178 34556.742 -5633.3981 -6293.9815 660.58339
      100 2503.7644 1559.8336 34556.742 -5633.4149 -6295.8997 662.48485
      200 2510.6452 -835.64242 34556.742 -5632.9604 -6297.2659 664.30547
      300 2483.0344 -760.0705 34556.742 -5632.3707 -6289.3704 656.99976
      400 2490.0842 1328.7958 34556.742 -5633.1299 -6291.995 658.86512
      500 2508.2062 199.57743 34556.742 -5633.479 -6297.1391 663.66012
      600 2488.7531 -902.05094 34556.742 -5634.125 -6292.6379 658.51292
      700 2495.9396 9.3507726 34556.742 -5633.8768 -6294.2912 660.41443
      800 2543.83 -2765.9139 34556.742 -5633.2548 -6306.3408 673.086
      900 2523.5406 1709.5687 34556.742 -5632.9736 -6300.6911 667.71753
     1000 2509.1854 2218.8799 34556.742 -5631.4962 -6295.4154 663.91921
     1100 2535.0913 768.50511 34556.742 -5631.3654 -6302.1392 670.7738
     1200 2547.4224 1150.3943 34556.742 -5629.5374 -6303.574 674.03656
     1300 2546.6773 996.52206 34556.742 -5630.3371 -6304.1765 673.8394
     1400 2498.0244 2679.999 34556.742 -5628.8876 -6289.8537 660.96605
...
    11700 2709.3762 7759.857 34556.742 -5549.9369 -6266.8257 716.8888
    11800 2743.6488 7037.5455 34556.742 -5550.0622 -6276.0194 725.95718
    11900 2747.909 5142.9637 34556.742 -5549.7642 -6276.8487 727.08441
    12000 2724.4031 6263.2292 34556.742 -5549.1656 -6270.0304 720.86485
    12100 2684.3066 8502.3985 34556.742 -5547.9889 -6258.2444 710.2555
    12200 2694.4988 6859.4477 34556.742 -5548.0323 -6260.9846 712.95231
    12300 2695.6327 8318.6089 34556.742 -5547.9506 -6261.2029 713.25233
    12400 2732.6903 8394.9672 34556.742 -5544.7971 -6267.8547 723.0576
    12500 2706.6753 8666.7851 34556.742 -5544.7074 -6260.8816 716.17415
    12600 2743.9752 7300.6021 34556.742 -5543.3171 -6269.3606 726.04353
    12700 2761.2971 5132.8108 34556.742 -5543.6048 -6274.2316 730.62683
    12800 2744.0508 6923.9369 34556.742 -5544.0718 -6270.1354 726.06353
    12900 2750.1352 7005.1982 34556.742 -5544.7651 -6272.4386 727.67343
    13000 2728.4839 8524.9718 34556.742 -5544.3387 -6266.2833 721.94461
    13100 2772.5724 5791.7983 34556.742 -5543.3643 -6276.9745 733.61023
    13200 2749.984 8711.5328 34556.742 -5543.2085 -6270.8419 727.63343
    13300 2730.2394 8202.013 34556.742 -5543.1243 -6265.5334 722.40911
    13400 2700.5879 10074.34 34556.742 -5544.0538 -6258.6173 714.56345
    13500 2772.806 6199.1337 34556.742 -5543.3711 -6277.0431 733.67203
...

How can I solve this problem?

Best,
Chol-Jun

Try equilibrating longer, with a smaller timestep? Maybe some
NVT before NVE?

Steve