phase angle in viscoelasticity simulation by NEMD

[Vigoss1]

Dear all,

I’m running the MD simulation about the viscoelasticity by NEMD. But result shows that the phase angle is negative. Attached are the simulation strain/stress result (which is wrong) and the theoretical right result. Could anyone tell me how to modify the code? Your help will be greatly appreciated!

Part of the input code is as below:

# turn on NEMD shear and equilibrate some more
reset_timestep 0

fix 4 all nvt/sllod temp {T} {T} ${tempdamp}
fix 5 all deform 1 xy wiggle 3 500 remap v

variable strain equal xy/lx

thermo 25
thermo_style custom step temp etotal press v_strain pxy density
thermo_modify norm no
run 50000

[athomps]

What you call the phase angle is simply a timelag in the material response. The magnitude of the timelag is probably affected to some extent by the amplitude and period of the strain forcing, as well the the system size. The reason you think it is “negative/wrong” is simply because in your simulation, the forcing variable is strain, whereas for the theoretical result, the forcing variable is stress. There is probably another version of the theory in which the forcing variable is strain.