Hello,
I am using the LAMMPS version “64-bit 27 May 2021” for my simulation.
I am defining the pair by command “Pair_style hybrid” and met with the error “Pair coeff for hybrid has invalid styles”.
Below are my scripts:
pair_style hybrid lj/cut 5 lj/cut 10 tersoff lj/cut 10 lj/cut/coul/cut 10 lj/cut/coul/cut 5
pair_coeff 1 1 lj/cut 1 4.477 0
pair_coeff 1 2* lj/cut 0 0
pair_coeff * * C.tersoff NULL C C NULL NULL
pair_coeff 23 45 lj/cut 0 0
pair_coeff 4 4*5 lj/cut/coul/cut 0 0
pair_coeff 5 5 lj/cut/coul/cut 0.0067 3.166
info coeffs
Could someone help me to debug my coding?
Great thanks!
The line after that error message should tell you which line is incorrect.
please note that there is no need to define lj/cut or lj/cut/coul/cut multiple times in a hybrid pair style if you need different cutoffs, you can just specify them with the pair_coeff command, but if you do, that changes the syntax slightly.
you should be able to resolve all these issues by just carefully reading the documentation for pair style hybrid.
Thanks for your reply, I think I just found the error.
Hi everyone,
I am calculating mean square deviation to estimate diffusion coefficient in glasses. Which ensamble to be used during the glass formation??
I am preparing glass via conventional
- velocity mixing (nvt) at 4500K for 10 ps
- Relaxation at 4500K (npt) 100 ps
- Cooling from 4500K to 300K at zero pressure (npt) 1K/ps
- Relaxing at 300 K at zero pressure (npt)
- Statistical averaging over 100 ps (nvt) at 300 K.
So, I am putting msd at step no 3 i.e. cooling step by varying the temperature over which cooling needs to be performed i.e. 4000K, 3500K, … 1500K.
Is the procedure right to estimate msd at cooling stage via npt ensamble??
Thanks in anticipation
Mukesh
In the future, please post a new message from scratch if you have a new topic and do not reply to an existing message, even if you change the subject line. the fact that you quoted a completely unrelated message confuses programs that sort emails into threads. also, it makes it look as if you are too lazy to type the email address of the mailing list. probably not the impression that you want to convey. 
Hi everyone,
I am calculating mean square deviation to estimate diffusion coefficient in glasses. Which ensamble to be used during the glass formation??
during that process you are not in any of the well-known statistical mechanical ensembles.
I am preparing glass via conventional
- velocity mixing (nvt) at 4500K for 10 ps
- Relaxation at 4500K (npt) 100 ps
- Cooling from 4500K to 300K at zero pressure (npt) 1K/ps
- Relaxing at 300 K at zero pressure (npt)
- Statistical averaging over 100 ps (nvt) at 300 K.
So, I am putting msd at step no 3 i.e. cooling step by varying the temperature over which cooling needs to be performed i.e. 4000K, 3500K, … 1500K.
Is the procedure right to estimate msd at cooling stage via npt ensamble??
there is a significant difference between using fix npt and having an NPT ensemble. when you are cooling down and while you are cooling down, you do NOT have an NPT ensemble. the integrator in fix npt is only one requirement to have an NPT ensemble, but you must not have any other system manipulations and they system must have reached equilibrium at the given temperature and pressure until you can claim that you are sampling an NPT ensemble.
as for scientific correctness, you need to discuss this with somebody that knows and understands the science of your simulation like your adviser/supervisor/colleagues. this is a mailing list about using the software correctly, not about how to realize your science correctly.
all I can say is that it seems odd to me that you want to measure the mobility of atoms while you are still trying to establish your system. the whole process of quenching a melt to a glass is a bit artificial and you are supposed to cool rapidly which contradicts the need to run for an extended period in equilibrium to get well converged MSD values.