Dear LAMMPS users ,
I am doing a MD simulation about condensation of vapor and formation of liquid droplets .
I need to calculate the size of these liquid droplets at different times .
Can I do this with LAMMPS ?
Thanks for your attention.
Regards,
Saeed.
Dear LAMMPS users ,
I am doing a MD simulation about condensation of vapor and formation of
liquid droplets .
I need to calculate the size of these liquid droplets at different times .
Can I do this with LAMMPS ?
yes and no. that depends on how you define what constitutes a droplet
and how you want to measure "size".
i can imagine that a somewhat crude approach could be done through
creative use of computes/fixes via the "chunk" features, but anything
sophisticated that would factor in orientation of molecules, number of
contacts and shape of clusters would only be possible in post
processing.
axel.
To reinforce what Axel has stated, you will need to post process the data
using a clustering algorithm
https://en.wikipedia.org/wiki/Cluster_analysis
in which you will have to provide a definition of "droplet". From that you
should be able to then determine the formation rate and size distribution of
"droplets" in your simulation.
Remember, "droplets" can be rather large (in simulation terms). Here's a
link to a tangentially related paper that discusses water droplet sizes
found in crude oil:
http://www.uni-leipzig.de/~diff/pdf/volume9/diff_fund_9(2009)7.pdf
Jim
James Kress Ph.D., President
The KressWorksTM Foundation C
An IRS Approved 501 (c)(3) Charitable, Nonprofit Organization
"ENGINEERING THE CURE" TM
(248) 605-8770
Learn More and Donate At:
Website: http://www.kressworks.org
Facebook: KressWorks Foundation
Twitter: @KressWorksFnd
Confidentiality Notice | This e-mail message, including any attachments, is
for the sole use of the intended recipient(s) and may contain confidential
or proprietary information. Any unauthorized review, use, disclosure or
distribution is prohibited. If you are not the intended recipient,
immediately contact the sender by reply e-mail and destroy all copies of the
original message.
The compute cluster/atom will flag every atom with
the “ID” of a cluster it is currently in, using a cutoff
distance criterion you provide. If you use the
cluster ID as the chunk ID in the compute chunk/atom
command, then you can count atoms in clusters
at any timestep, and output them via the fix ave/chunk
command. You can also sum per-atom energy per
cluster, etc and output that.
Steve
Thanks a lot for your help .
Can you please let me know how I can use cluster ID as the chunk ID in the compute chunk/atom command ?
My apologies for wasting your time .
Regards,
Saeed.
Thanks a lot for your help .
Can you please let me know how I can use cluster ID as the chunk ID in the
compute chunk/atom command ?
Hi .
Could you please let me know How I can change the number of neighbors when I want to calculate cluster sizes with compute cluster/atom command ?
I mean I want to have cluster with for example r = 4 and N= 3 , 5 , 8 and I want to calculate this cluster size but I don’t know how I can change the number of neighbors .
Thanks for your attention.
Best ,
Saeed.
The cluster/atom command works based on a
cutoff distance you provide. The size of the clusters
then depends on your geometry and your cutoff.
I think you are asking for it to be the other way,
you provide the size of the cluster and the code
determines the cutoff. That’s not the way it works.
Steve
Dear Steve ,
Thank you so much for your reply .
Actually I want to define a cluster with for example r(cutoff) = 4.0 and number of neighbors = 3 or 5 or 8 and then I want to calculate the size of this cluster .
Is It possible ?
Best ,
Saeed.
If by “number of neghbors” you mean
the # of atoms in the cluster, then no
it isn’t possible. If you define the cutoff,
that defines the size of the clusters that
will be found. If there are 10 atoms within
4 Angs of each other, there are 10
atoms in the cluster. You can’t choose
there to be 3.
Steve