Question about MSD value in cooling and heating simulation using read_dump

Hi ,

I ran a cooling simulation of a large LJ system composed of 19652 atoms for a specific cooling

rate, 1*10^8 K/s. for a heating simulation for the same rate I used the read dump command from

the last dump file from the cooling simulation.

Going over the log files, I saw that the MSD value is different in the few first values of the

heating compared to the value from the log file that corresponds to the dump file.

the difference is ~10 orders of magnitude.

I understand that the read_dump in my case is

limited to the atom coordinates but from what I read in the MSD compute

documentation the MSD is calculated from the displacement in each direction.

My question is, why is there such a large deviation in values from the last few log files in the cooling simulation and the first few valuew in the heating simulation?

in the end of the mail are the dumpfile header and input and logs for each of the runs and marked in red are the values of the MSD i’m referring to.

Thanks very much in advance

Stas Shor

Materials physics group website: https://sites.google.com/site/makovgroup/home
MSc Student ,Department of Materials Engineering
Ben-Gurion University of the Negev,
Beersheba, Israel

Dump file header and first few lines:

ITEM: TIMESTEP
500000000
ITEM: NUMBER OF ATOMS
19652
ITEM: BOX BOUNDS pp pp pp
-0.637518 90.0575
-0.637518 90.0575
-0.637518 90.0575
ITEM:ATOMS id type xs ys zs c_myBond[1] c_myBond[2]
6635 1 0.0202624 0.0123278 0.0363909 0.522957 0.630023
5446 1 0.0103892 0.0328791 0.00402261 0.299563 0.53606

Hi ,

I ran a cooling simulation of a large LJ system composed of 19652 atoms for
a specific cooling

rate, 1*10^8 K/s. for a heating simulation for the same rate I used the read
dump command from

the last dump file from the cooling simulation.

Going over the log files, I saw that the MSD value is different in the few
first values of the

heating compared to the value from the log file that corresponds to the dump
file.

the difference is ~10 orders of magnitude.

I understand that the read_dump in my case is

limited to the atom coordinates but from what I read in the MSD compute

documentation the MSD is calculated from the displacement in each
direction.

My question is, why is there such a large deviation in values from the last
few log files in the cooling simulation and the first few valuew in the
heating simulation?

because this is the beginning of the simulation, and thus your atoms
have not yet accumulated a significant displacement.

axel.