During MD tensile simulation, the stress values rise upto some GPa (high stress value). I think it is because of the following reasons:
1. System is almost defect free or only with few defects compared to real world materials
2. Peierls stress from the system
3. Applied high strain rate
4. Low temperature (not always)
5. Time and length scale issues from MD technique
If there is any deviations, please let us discuss them.
Mr. Arumugakani,
Can you specify the MD method and the program package?
I think the reason is similar to this: "Why we use Gigabytes"
Just convention.
Madam Bojidarka B. Ivanova
I'm using 'LAMMPS' classical MD simulator in which I used fix deform to do tensile deformation.
Mr. Kostadin Georgiev Gaminchev ,
The strength of the real world material is usually in MPa. But, MD simulation shows the strength reaches GPa. Its not convention.
Mr. Kedharnath Arumugakani,
depends on materials used. For examples in my calculations (atomic units) and mostly scientific articles for molecular dynamics (and not only!) the appropriate strength is in GPa, and I think that from LAMMPS and such big software projects know that apply as well.
Actually you can change the code to show you results in MPa :))))
Cheers
Have a nice day!
your question is why so such a large stress values required for deformation of nano materials.I think this is due to large surface to volume ratio. U can also find failure of hall petch relation at sub nono- scale. Think in this way, you can understand.
Mr. J. Veerababu
Yes, I could see the inverse Hall Petch relation when grain size was varied. But, I used polycrystalline model which had lots of defects and grain boundaries still I could see higher stress values (few GPas).
I think there is no thumb rule such that all nano scale materials must have stress values in GPa. it depends on size of your system (within nano level itself), shape, temperature, and material.
https://www.researchgate.net/post/Can_anyone_explain_the_main_reasons_factors_for_the_different_behaviours_of_nanomaterials
Kedharnath Arumugakani
The 5 points mentioned with the question cover the main reasons for the observed high stress values (in GPa) in MD simulations in LAMMPS.
Another factor could be that the defects present in the MD simulations are still a simplified version of the defects actually present in the material.
Also, the interatomic potential used for the MD simulations can effect the obtained results. A given material can have different interatomic potential based on the fitting parameters, data used and different functional forms used for fitting. For example, if the data used to develop the potential is taken from the experimental values then the stress values from MD calculations are more likely to match the experimental values. However, if the data is based on DFT (stress values are usually higher than experimental values) values then the results from MD simulations will tend to be closer to the DFT values.
Dear Pulkit Garg
Thank for your answer. Can you please arrange the above 5 points in an order such that which point would impact the strength to a larger extend in MD simulation. I like to know which impacts more on MD systems.
Yes, I agree that we model ideal defect structures or simple defects in MD simulations. But even with a single edge dislocation in 10^3 nm^3 box volume, the dislocation density is very high compared to the experimental results.
I understand that potential functions affects the results from MD simulations but I haven't tried playing with potentials. I thought EAM potentials are fitted using both experimental and DFT results. In most of the cases, lattice constants, elastic constants etc are nearly same in experimental and DFT results. If you have experience or noticed any data with difference in experimental and DFT results that lead to different potential functions, Please help me with the article.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress