# the region temperature calculated by <compute temp/region> command shows invalid result

Dear Sir,

[Problem]
Although the individual atoms have kinetic energy, the region temperature calculated by <compute temp/region> command shows 0 Kelvin.
Such results are hard to think actually and I would like to teach improvement if it is a software bug or setting problem.

The outline of the simulation and the result are as follows;
· Investigate the moving behavior of flake when flake is placed on a substrate with a temperature gradient
· Dividing the flake n to calculate the average temperature for each region (n1 to n5)
· Although the distribution of kinetic energy is almost unchanged at each time step,

the average temperature for each is different

the average temperature of 33000 steps: (n1,n2,n3,n4,n5)=(295, 265, 261, 237, 245)

the average temperature of 300000 steps:(n1,n2,n3,n4,n5)=(0, 0, 0, 0, 274)

attached file are raw data of the atoms of flake and k line shows the kinetic energy.

regards,

relating script is below;

units metal
boundary f p f
atom_style atomic
lattice sc 1
mass

pair_style airebo 3.0
pair_coeff * *

neighbor 0.3 nsq
neigh_modify delay 1

region n block INF INF INF INF 29 INF #cubic
region n1 block INF -159 INF INF 29 INF
region n2 block -159 -153 INF INF 29 INF
region n3 block -153 -147 INF INF 29 INF
region n4 block -147 -141 INF INF 29 INF
region n5 block -141 INF INF INF 29 INF

group n region n
group n1 region n1
group n2 region n2
group n3 region n3
group n4 region n4
group n5 region n5

compute comn n com

compute pall all pe/atom
compute kall all ke/atom

compute pe all pe
compute pflake all reduce/region n sum c_pall
compute ke all ke

compute tn all temp/region n
compute t_n1 all temp/region n1
compute t_n2 all temp/region n2
compute t_n3 all temp/region n3
compute t_n4 all temp/region n4
compute t_n5 all temp/region n5

compute stress all stress/atom

variable stress_xx atom c_stress[1]
variable stress_yy atom c_stress[2]
variable stress_zz atom c_stress[3]

thermo 100
thermo_style custom step etotal pe ke c_tall c_tb c_tc c_td c_te c_tf c_tg c_th c_ti c_tj c_tk
thermo_modify temp tall

#minimize 0.0 1.0e-6 100 500000
#reset_timestep 0

timestep 0.001

fix nvtb b nvt temp 300 300 0.1
fix nvtk k nvt temp 200 200 0.1
fix nve free nve

velocity all create 1 581277 temp tall

velocity a set 0 0 0
velocity l set 0 0 0
velocity n set 0 0 0

fix fa a setforce 0 0 0
fix fl l setforce 0 0 0

fix fn n setforce 0 0 0

dump 2 n cfg 1000 n..cfg id type xs ys zs vx vy vz x y z c_pall c_kall v_stress_xx v_stress_yy v_stress_zz id
dump 3 all cfg 1000 graphene.
.cfg id type xs ys zs vx vy vz x y z c_pall c_kall v_stress_xx v_stress_yy v_stress_zz id

fix data all ave/time 1 1000 1000 c_pe c_ke c_tall c_pflake file data.dat

fix temp all ave/time 1 100 100 c_tb c_tc c_td c_te c_tf c_tg c_th c_ti c_tj c_tk c_tn file temp.dat
fix temp_cubic all ave/time 1 100 100 c_t_n1 c_t_n2 c_t_n3 c_t_n4 c_t_n5 file temp_cubic.dat

unfix fn

run 5000000

n.33000.xlsx (58.3 KB)

n.300000.xlsx (64.5 KB)

Dear Sir,

[Problem]
Although the individual atoms have kinetic energy, the region temperature
calculated by <compute temp/region> command shows 0 Kelvin.
Such results are hard to think actually and I would like to teach
improvement if it is a software bug or setting problem.

people they don't make no sense without suitable explanations.

The outline of the simulation and the result are as follows;
· Investigate the moving behavior of flake when flake is placed on a
· Dividing the flake n to calculate the average temperature for each region
(n1 to n5)
· Although the distribution of kinetic energy is almost unchanged at each
time step,
the average temperature for each is different
the average temperature of 33000 steps: (n1,n2,n3,n4,n5)=(295, 265, 261,
237, 245)
the average temperature of 300000 steps:(n1,n2,n3,n4,n5)=(0, 0, 0, 0, 274)

attached file are raw data of the atoms of flake and k line shows the
kinetic energy.

have you checked, whether are are actually atoms inside those regions?

axel.