Molecular Dynamics#

Performing a JEDI analysis on molecular dynamics simulations is a little more involved than the “classic” analysis we introduced over the previous pages. To illustrate, we will — like in the very first tutorial — use a nitrogen molecule and simulate it at 400 Kelvin. For running the calculations, we will also use EMT for the same reasons: it is readily available and reasonably fast to run.

Full example#

Here, you find the whole code example to test. Below, every step is discussed in detail. Please note, that most of this script is just boilerplate to get the example working in a self-contained manner. The actual JEDI analysis is at the very end of this file.

from ase import Atoms, units
from ase.calculators.emt import EMT
from ase.io.trajectory import Trajectory
from ase.md.langevin import Langevin
from ase.optimize import BFGS
from ase.vibrations import Vibrations

from strainjedi import Jedi

# create an N2 molecule and optimise its geometry.
n2 = Atoms("N2", [(0, 0, 0), (0, 0, 1.1)])
n2.calc = EMT()
BFGS(n2).run(fmax=0.01)

dist = n2.copy()
dist.calc = EMT()

# timestep of 5 fs, T = 400 K, friction coeff = 0.002 a.u.
# save every 4th step in moldyn3.traj
dyn = Langevin(
    dist,
    timestep=5 * units.fs,
    temperature_K=400,
    friction=0.002,
    trajectory="moldyn3.traj",
    loginterval=4,
    fixcm=False,
)

# run the simulation for 200 timesteps.
dyn.run(200)
traj = Trajectory("moldyn3.traj")

# obtain the Hessian
vib = Vibrations(n2)
vib.run()
hessian = vib.get_vibrations()

for idx, atoms in enumerate(traj):
    print(f"performing JEDI analysis for step {idx}...", end=" ")
    j = Jedi(n2, atoms, hessian)
    j.run(printout=False)
    j.visualize(output_dir=f"moldyn_vis/{idx}", single_mode="bl")
    print("done!")

Create The Structure#

First, we will start by creating our nitrogen atom and perform a basic geometry optimization. Do note that this is to provide a self-contained example; as established previously, it is just as possible to use an existing calculator.

from ase import Atoms
from ase.calculators.emt import EMT
from ase.optimise import BFGS

n2 = Atoms("N2", [(0, 0, 0), (0, 0, 1.1)])
n2.calc = EMT()
BFGS(n2).run(fmax=0.01)

Run The Simulation#

The ASE framework provides a very simple interface to run MD simulations and then store the entirety of that simulation. Feel free to contribute examples for the calculator of your choice — this makes JEDI more accessible to everyone.

from ase import units
from ase.md.langevin import Langevin

dist = n2.copy()
dist.calc = EMT()

temp = 400  # Kelvin
# timestep of 5 fs, T = 400 K, friction coeff = 0.002 a.u.
# save every 4th step in moldyn3.traj
dyn = Langevin(
    dist, timestep=5 * units.fs, temperature=temp * units.kB, friction=0.002, trajectory="moldyn3.traj", loginterval=4
)

# run the simulation for 200 timesteps.
dyn.run(200)

This will store the results of the molecular dynamics simulation in moldyn3.traj. Now, we can read this ASE-specific format into JEDI.

Perform a JEDI Analysis#

First, we will have to read in the trajectory file we created earlier.

from ase.io.trajectory import Trajectory

traj = Trajectory("moldyn3.traj")

Then, also obtain the necessary inputs (the Hessian).

from ase.vibrations import Vibrations

vib = Vibrations(n2)
vib.run()
hessian = vib.get_vibrations()

Finally, we can start a JEDI analysis for all time steps. To avoid creating a lot (and we really mean a lot!) of visualization windows, we will instead store them inside a moldyn_vis directory. Also, let’s tell the visualizer to only create figures pertaining to the bond length.

from strainjedi import Jedi

for idx, atoms in enumerate(traj):
    print(f"performing JEDI analysis for step {idx}...", end=" ")
    j = Jedi(n2, atoms, hessian)
    j.run(printout=False)
    j.visualize(output_dir=f"moldyn_vis/{idx}", single_mode="bl")
    print("done!")