I am trying to perform tensile testing of a bulk copper box. But when I plot the results of the test, I get a slight curve in the elastic region of the stress-strain plot. Is this normal? Also I used the Mishin potential to test copper. I created the required box and used velocity command to raise the temperature to 300K and "fix box/relax" and then "minimize" command to equilibriate the box. The other parameters used were 10^8/ps strain rate and testing was done at 300K.
The variations in the linear elastic zone based on the type of materials and applied load
I have observed similar behavior in my studies on dislocation nucleation in single crystals for copper and aluminum. One thing to always check is the strain rate effects caused by molecular dynamics; in some cases this can cause artifacts. In my studies, though, I found that this behavior was sensitive to the crystal orientation and stress state, i.e., tension versus compression. There is a really nice figure in Tschopp and McDowell, JMPS 2008 that shows stress-strain curves for various orientations just beyond dislocation nucleation.
In addition to Dr. Tschopp response, I noticed something that could also be source to your problems. After imparting a velocity equivalent to 300K, you are performing minimize and box/relax, this scheme would work nicely for a MS simulations. However for MD, you should consider performing NPT (assuming its a periodic domain, along which you apply strain). To see if its equilibrated, check the pressure it should be close to zero for a period defect free crystal. Hope this helps, just for reference you could check a nice collection of tutorials by Dr. Tschopp (https://icme.hpc.msstate.edu/mediawiki/index.php/LAMMPS_tutorials).
Regards
Ilaksh
Dear Dr. Tschopp, I have made the simulation box with 3 orthogonal axii as X-[112], Y-[11-1] and Z-[1-10] as my work involves investigating twin-dislocation interactions. Would this particular selection of axis orientations result in deviation in the elastic regime?
Dear Ilaksh Adlakha, Thank you for your very insightful answer. I shall try to incorporate your suggestions to see if it will resolve some of my issues.
There are two things to check here; the first is (as mentioned by others) that you have equilibrated properly. Beyond checking the pressure, make sure there are no resonant stress waves within your simulation box. The second is rate effects; at the rates associated with MD visco-elastic behaviour is quite prominent.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Material Characterization