ERROR: Fix ehex kinetic energy rescaling too large

LAMMPS LAMMPS Beginners
,
1

Dear all

As a beginner in LAMMPS, I’m using the NEMD method to calculate the thermal conductivity of single - layer graphene. I’ve used the “fix ehex” command, but I encountered the following error: “Fix ehex kinetic energy rescaling too large”. I haven’t found any solutions to this error on this forum or other websites. I hope someone can give me some guidance.

At the same time, when calculating the thermal conductivity with LAMMPS, I obtained the temperature gradient within a relatively small number of running steps and calculated the thermal conductivity through Fourier’s law. However, I clearly noticed that the value of the thermal conductivity is two orders of magnitude lower. Here is my in - file. I would be extremely grateful if you could give me some advice and suggestions on my in - file.

Best regards,
Kun

units       metal       
boundary    p p p       
atom_style  atomic      
neighbor    0.3 bin  
neigh_modify every 1 delay 10 check yes  

variable  T equal 300 
variable  A equal 1.42 
variable  DT equal 0.0002 
variable  POWER equal 0.005 

lattice custom  ${A} a1  3  0  0   a2   0 1.732  0 a3 0 0 2.357   &
basis   0     0   0  & 
basis   0.333 0   0  & 
basis   0.5   0.5 0  &
basis   0.833 0.5 0

region graphenesheet block 0 255.78 0 24.12 -0.5 0.5 units box   
create_box      1 graphenesheet                         
create_atoms    1 region graphenesheet      
mass            1 12.0107      
change_box      all z delta -150 150 units box            


pair_style  tersoff
pair_coeff  * * SiC.tersoff C

write_data      1Model.data

timestep    ${DT} 
velocity    all create ${T} 89684 dist gaussian 
min_style   cg
minimize    1e-10 1e-10 1000 1000  

write_data      2min.data


dump        1 all custom 10000 4graphene.lammpsstrj id type x y z fx fy fz 
fix         NVT all nvt temp ${T} ${T} 0.02
thermo_style custom step temp press pe ke lx ly lz density 
thermoss      100000 


run 1000000  
undump 1
write_data  3chiyu.data

unfix NVT   
fix  NVE all nve  
run 100000

reset_timestep 0


region SOURCE block 0 1 INF INF INF INF units box 
region SINK block 25 26 INF INF INF INF units box 



fix HEAT_IN  all ehex 1 ${POWER} region SOURCE 
fix HEAT_OUT all ehex 1 -${POWER} region SINK 



compute  KE all ke/atom    
variable KB equal 8.617e-5 
variable TEMP atom c_KE/1.5/${KB} 


compute BLOCKS all chunk/atom bin/1d x lower 0.02 units reduced
fix     T_PROFILE all ave/chunk 10 10000 100000 BLOCKS v_TEMP file temp${T}.txt
thermo      100000        
thermo_style custom step temp press lx ly vol

dump        1 all custom 100000 5conductivity.lammpsstrj id type x y z fx fy fz 

run 5000000

NEMD.in (4.0 KB)

2

Check out the documentation. It has a note

If heat is subtracted from the system too aggressively so that the group’s kinetic energy would go to zero, then LAMMPS will halt with an error message. Increasing the value of nevery means that heat is added/subtracted less frequently but in larger portions. The resulting temperature profile will therefore be the same.

This is a question about your research and how to apply the various methods correctly. That is about the science of your research, not about LAMMPS. And discussing with you how to do your research correctly, is the responsibility of your adviser or tutor and off-topic for this forum.

1 Like
3

I’m extremely grateful for your prompt reply. Please forgive me as a beginner for my inexperience in using the official documentation. After referring to the documentation, can I understand that the above - mentioned error occurs due to the excessive energy of the system? Reducing the heat - flow power or decreasing the frequency of heat - flow increase and decrease will be helpful in resolving this error. Thank you again for your reply.