I was trying to solve one problem of metal nanowire of Cu. the procedure what I followed is as firstly relaxed to energy minimization using conjugate gradient method. then equilibrate at 10 K using NOSE HOOVER thermostat for 100 ps before loading in tension. the
loading applied by using one end fixed and the n applying
velocities to atoms along the loading direction using ramp
velocity profile. and used no periodic boundary conditions
to simulate to capture the relavent surface effects. I am
using EAM potentials given in the library of LAMMPS for
this work. one input file I am attaching if I am doing some
thing wrong somebody can tell me in write direction.
problems which I am having related to this problems is
1.to equilibrate at 10 K using nose hoover thermostate. I
am not sure I have to run for how long is 100 ps is enough.
becouse the temperature was still fluctuating because I
tried for 200 500 ps etc also. temperature was not
stablized.
as I am applying the velocity as loading if I am not
wrong for applying the velocity as a function of load I
have to define various fix command or how can I give the
velocity as a function of time as a input. or what is the
correct way to give input as a load as afucntion of time,
which command should I have to use?
3.For post processing as i am using VMD, I am not able to
see the perticular atom no in the GUI. If I want to plot
the stress vs strain for a particular atom of a set of atom
which post processor is good for such kind of post
processing.
4.I want to generate my own input using Hartree-fock method
to generate potentials, is there any free code is available
online?
as there are possibility of changing structure from one
phase to another i.e from fcc to bcc or some other. how to
visualize such kind of things? or if I have to write some
code for that; what is the correct way to procede further
in that direction. of If some post processor can do such
kind of thing?
If some one can give valuable suggestions It will be great.
I was trying to solve one problem of metal nanowire of Cu. the procedure what I followed is as firstly relaxed to energy minimization using conjugate gradient method. then equilibrate at 10 K using NOSE HOOVER thermostat for 100 ps before loading in tension. the
loading applied by using one end fixed and the n applying
velocities to atoms along the loading direction using ramp
velocity profile. and used no periodic boundary conditions
to simulate to capture the relavent surface effects. I am
using EAM potentials given in the library of LAMMPS for
this work. one input file I am attaching if I am doing some
thing wrong somebody can tell me in write direction.
problems which I am having related to this problems is
1.to equilibrate at 10 K using nose hoover thermostate. I
am not sure I have to run for how long is 100 ps is enough.
becouse the temperature was still fluctuating because I
tried for 200 500 ps etc also. temperature was not
stablized.
You will always see fluctuations, especially if your system is small.
Try running NVE and see if it is more stable.
as I am applying the velocity as loading if I am not
wrong for applying the velocity as a function of load I
have to define various fix command or how can I give the
velocity as a function of time as a input. or what is the
correct way to give input as a load as afucntion of time,
which command should I have to use?
If you want the velocity of boundary atoms to be a function
of time, you’ll have to write your own fix to do this.
3.For post processing as i am using VMD, I am not able to
see the perticular atom no in the GUI. If I want to plot
the stress vs strain for a particular atom of a set of atom
which post processor is good for such kind of post
processing.
AtomEye is good for nanowires and materials in general.
You can color atoms by various values.
4.I want to generate my own input using Hartree-fock method
to generate potentials, is there any free code is available
online?
as there are possibility of changing structure from one
phase to another i.e from fcc to bcc or some other. how to
visualize such kind of things? or if I have to write some
code for that; what is the correct way to procede further
in that direction. of If some post processor can do such
kind of thing?
Try looking at the coordination number of atoms. It will
change for fcc vs bcc. See the compute coord/atom command.