Hi everyone,
I’m trying to to simulate a tension test on Al sample. Attached you can see one of the stress-strain curves I got. Do you think there is plastic regime on this curve? İf so, can LAMMPS create dislocations by itself? Or is the plasticity just due to grain boundaries?
Looks pretty ugly. There is no way to tell without seeing your input file. Lammps doesn’t create anything on its own but only that you (the user) ask the code to do given an input model and parameters. Clueless user = meaningless results.
Carlos
Hi everyone,
I'm trying to to simulate a tension test on Al sample. Attached you can see
one of the stress-strain curves I got. Do you think there is plastic regime
on this curve? İf so, can LAMMPS create dislocations by itself? Or is the
LAMMPS will not "create" anything unless you ask it to do so. *but* it
also has no idea of the material that you are simulating. all it sees
are the coordinates, parameters and settings in your input script and
then it blindly computes what you ask it to do. whether this is
meaningful and what exactly happens, is the task of the user to to
figure out.
plasticity just due to grain boundaries?
no idea. have you visualized your results? have you tested your model?
does compare with published results for similar simulations using the
same parameters?
axel.
The thing I wanted to asked is if plastic flow can be realized without imposing it. I mean does LAMMPS simulate dislocation nucleation during deformation? This is the input file I’m using. Thanks for your answers.
Berkay
clear # to remove all atoms, restore all settings (does not remove variables)
units metal # check for desired units
dimension 3
boundary p p p # check for the boundary style of simulation box
atom_style atomic
lattice fcc 4.05
region whole prism 0 20 -50 50 0 20 0 0 0 units box
create_box 1 whole
#3 means that there will be 3 types of atoms for three grains (can be changed for more types to distinguish in OVITO)
region grain1 prism INF INF -50 -30 INF INF 0 0 0 units box
lattice fcc 4.05 orient x 1 -1 0 orient y 1 1 -2 orient z 1 1 1
create_atoms 1 region grain1
region grain2 prism INF INF -30 -10 INF INF 0 0 0 units box
lattice fcc 4.05 orient x 2 -1 0 orient y 3 6 -5 orient z 1 2 3
create_atoms 1 region grain2
region grain3 prism INF INF -10 10 INF INF 0 0 0 units box
lattice fcc 4.05 orient x 0 1 0 orient y 0 0 1 orient z 1 0 0
create_atoms 1 region grain3
region grain4 prism INF INF 10 30 INF INF 0 0 0 units box
lattice fcc 4.05 orient x 1 -1 1 orient y 1 -1 -2 orient z 1 1 0
create_atoms 1 region grain4
region grain5 prism INF INF 30 50 INF INF 0 0 0 units box
lattice fcc 4.05 orient x 6 -1 0 orient y 9 54 -37 orient z 1 6 9
create_atoms 1 region grain5
group grain1 type 1
group grain2 type 1
group grain3 type 1
group grain4 type 1
group grain5 type 1
pair_style eam/alloy
pair_coeff * * Al99.eam.alloy Al # * * for i,j and “Al Al Al” for atom types
neighbor 2.0 bin
neigh_modify delay 10 check yes
displace_atoms grain1 move 0 0 0 units lattice
displace_atoms grain2 move 0 0 0 units lattice
displace_atoms grain3 move 0 0 0 units lattice
displace_atoms grain4 move 0 0 0 units lattice
displace_atoms grain5 move 0 0 0 units lattice
delete_atoms overlap 0.35 all all
compute csym all centro/atom fcc
compute eng all pe/atom
compute eatoms all reduce sum c_eng
The thing I wanted to asked is if plastic flow can be realized without
imposing it. I mean does LAMMPS simulate dislocation nucleation during
deformation? This is the input file I'm using. Thanks for your answers.
LAMMPS simulates whatever the physics of the model (coordinates,
potentials, parameters) determines. i repeat, it is up to *you* to
figure out whether this is genuine or an artifact. LAMMPS doesn't know
and doesn't care. you are confusing symptom and cause here.
axel.
p.s.: i don't see anything obviously wrong in the input. but i would
worry that the neighborlist setting is a bit aggressive for MD. do you
get any dangerous builds?
I don’t know anything about dangerous builds. I have modified a code for
different orientations. Do you have any suggestions for that case? Thanks a
lot.
Berkay
I don't know anything about dangerous builds.
look at your log files and read in the LAMMPS documentation what they stand for.
I have modified a code for different orientations. Do you have any suggestions for that case? Thanks a
lot.
i have no idea what you are trying to say with this.
axel.
Hi,
Usually elastic - plastic transition can be identified by a decay in the shear stress. Soo its better to make a plot of shear = 0.5(Pxx - 0.5(Pxx + Pyy)) vs strain.
Am, In my own words A Defect can be defined as an a rearrangement of the atoms from its original configuration and this rearrangement can induced by many means e.g pressure or temperature. When the rearrangement posses a symmetry then it can be categorized as a Phase Change, usually the rearrangement is not chaotic or random i.e there are certain planes (slip planes) where the atoms try to move and form dislocations. Usually the stacking fault energy and the gamma surface along different planes and crystallographic directions are a prediction of this motion.
Plasticity is the way the material relief the stress or free energy thus its naturally to notice a drastic change in the thermodynamic variables during the process…
Hope this helps
Oscar G.
Hi again,
Sorry for the late reply. I tried to plot shear stress over strain. The graph is the one below. So do you think the abrupt change around 0.035 strain can be the elastic-plastic transition? Thanks for the help.
Berkay
You could have send a better image of the same, at least a one which would have such noises removed from the data.
Anyways, the point of elastic-plastic transition is that plasticity is the result of the defects in an otherwise perfect solid , like dislocations. What you are trying to ask is if LAMMPS can create them “itself” without you creating/inserting them??
Couple of points:
1-) This is not a Lammps Q but one about your research
2-) What if someone was going to answer in the affirmative to your Q? Would you write in your manuscript:
“…and Fig 1 supported by the comments of some dude on the Lammps list ensure the system is undergoing an elastic-plastic transition…” This is more a religious approach than a scientific one. Somebody does the thinking and you just follow…
Read the lit and look for papers where analysis on detecting the elastic-plastic transition is done. Now that you have a simulation that is “working” you can apply those methods to your system.
Carlos
Thanks for the aggressive comments. I’m a beginner in LAMMPS area so that i only wanted know if i’m progressing right or not. I got my answers. Thanks for them.
Berkay
26 Haz 2013 14:47 tarihinde “Carlos Campana” <campanacue@…24…> yazdı:
Yes, old school comments. Those tend to be more difficult to forget. Nothing wrong with harsh criticism. Will do you good on your dissertation day and future life.
Carlos
Thanks for the aggressive comments. I'm a beginner in LAMMPS area so that
i only wanted know if i'm progressing right or not. I got my answers.
Thanks for them.
you seem to be confusing learning how to use LAMMPS with learning how to do
MD. this is - unfortunately - becoming a bit of a trend (i am wondering
whether this is a sign of the increased popularity of LAMMPS, or of the
activity and responsiveness of this mailing list)
we can help you with the first part, but the second is the task of your
teachers, your advisers and - especially - yourself. a mailing list is a
poor replacement for that, since most people care to solve LAMMPS problems
(which can be a lot of fun; chasing a bug can be like being indiana jones
entering the cave in the amazon to pick up the golden idol) and not tutor
people in doing MD (which is often exhausting and offers very little
gratification).
most of the time people don't mind a question from a beginner that is not
exactly on topic, but if people fail to take the hint that their approach
is heading in the right direction, they will have to put up with
increasingly less friendly responses and will have to survive the
occasional flame war as well.
so please take this under consideration before you post the next question
and perhaps be a pain in the neck of your adviser for a change. 
thanks,
axel.