a si

##tersoff

###Recipe:

Begin with crystal 6x6x6 with 5.43 Ang lattice constant:

1. NVT melt 3500 K: 500000 steps
2. Instantly change KE from 3500K -> 1400K
3. NVT anneal 1400K: 500000 steps
4. Instantly change KE from 1400K -> 300K
5. NPT anneal 300 K: 20000000 steps (10 ns). Time-avg box size during this step
6. Deform box to time-avg 0-pressure box size from step 5)
7. NVT buffer 300K: 500000 steps
8. NVE 300 K: time-avg the atomic pos every 100000 steps.
9. Energy minimize atomic positions from end of step 8).

###Mean Square Displacements at steps 1), 3), 5), 8)

The MSD in steps 1) and 3) show liquid behavior. The MSD in step 5) shows no sign of liquid, and is quite higher than MSD of step 8). I believe this is because step 5) is in NPT ensemble and 8) is in NVE.

###step 8) positions snapshots

it is best to open this in a separate browser window, and step thru like

https://raw.github.com/jasonlarkin/disorder/master/si/amor/m_si_amor_x0_timeavg_annealHe_tersoff_6x_1.png https://raw.github.com/jasonlarkin/disorder/master/si/amor/m_si_amor_x0_timeavg_annealHe_tersoff_6x_2.png https://raw.github.com/jasonlarkin/disorder/master/si/amor/m_si_amor_x0_timeavg_annealHe_tersoff_6x_3.png https://raw.github.com/jasonlarkin/disorder/master/si/amor/m_si_amor_x0_timeavg_annealHe_tersoff_6x_4.png https://raw.github.com/jasonlarkin/disorder/master/si/amor/m_si_amor_x0_timeavg_annealHe_tersoff_6x_5.png

Based on the MSD, the structure is not a liquid, but based on these snapshots, there is meta-stability.

##published Di