"""Langevin dynamics class."""
from typing import IO, Optional, Union
import numpy as np
from ase import Atoms, units
from ase.md.md import MolecularDynamics
from ase.parallel import DummyMPI, world
[docs]
class Langevin(MolecularDynamics):
"""Langevin (constant N, V, T) molecular dynamics."""
# Helps Asap doing the right thing. Increment when changing stuff:
_lgv_version = 5
def __init__(
self,
atoms: Atoms,
timestep: float,
temperature: Optional[float] = None,
friction: Optional[float] = None,
fixcm: bool = True,
*,
temperature_K: Optional[float] = None,
trajectory: Optional[str] = None,
logfile: Optional[Union[IO, str]] = None,
loginterval: int = 1,
communicator=world,
rng=None,
append_trajectory: bool = False,
):
"""
Parameters:
atoms: Atoms object
The list of atoms.
timestep: float
The time step in ASE time units.
temperature: float (deprecated)
The desired temperature, in electron volt.
temperature_K: float
The desired temperature, in Kelvin.
friction: float
A friction coefficient in inverse ASE time units.
For example, set ``0.01 / ase.units.fs`` to provide
0.01 fs\\ :sup:`−1` (10 ps\\ :sup:`−1`).
fixcm: bool (optional)
If True, the position and momentum of the center of mass is
kept unperturbed. Default: True.
rng: RNG object (optional)
Random number generator, by default numpy.random. Must have a
standard_normal method matching the signature of
numpy.random.standard_normal.
logfile: file object or str (optional)
If *logfile* is a string, a file with that name will be opened.
Use '-' for stdout.
trajectory: Trajectory object or str (optional)
Attach trajectory object. If *trajectory* is a string a
Trajectory will be constructed. Use *None* (the default) for no
trajectory.
communicator: MPI communicator (optional)
Communicator used to distribute random numbers to all tasks.
Default: ase.parallel.world. Set to None to disable communication.
append_trajectory: bool (optional)
Defaults to False, which causes the trajectory file to be
overwritten each time the dynamics is restarted from scratch.
If True, the new structures are appended to the trajectory
file instead.
The temperature and friction are normally scalars, but in principle one
quantity per atom could be specified by giving an array.
RATTLE constraints can be used with these propagators, see:
E. V.-Eijnden, and G. Ciccotti, Chem. Phys. Lett. 429, 310 (2006)
The propagator is Equation 23 (Eq. 39 if RATTLE constraints are used)
of the above reference. That reference also contains another
propagator in Eq. 21/34; but that propagator is not quasi-symplectic
and gives a systematic offset in the temperature at large time steps.
"""
if friction is None:
raise TypeError("Missing 'friction' argument.")
self.fr = friction
self.temp = units.kB * self._process_temperature(temperature,
temperature_K, 'eV')
self.fix_com = fixcm
if communicator is None:
communicator = DummyMPI()
self.communicator = communicator
if rng is None:
self.rng = np.random
else:
self.rng = rng
MolecularDynamics.__init__(self, atoms, timestep, trajectory,
logfile, loginterval,
append_trajectory=append_trajectory)
self.updatevars()
def todict(self):
d = MolecularDynamics.todict(self)
d.update({'temperature_K': self.temp / units.kB,
'friction': self.fr,
'fixcm': self.fix_com})
return d
def set_temperature(self, temperature=None, temperature_K=None):
self.temp = units.kB * self._process_temperature(temperature,
temperature_K, 'eV')
self.updatevars()
def set_friction(self, friction):
self.fr = friction
self.updatevars()
def set_timestep(self, timestep):
self.dt = timestep
self.updatevars()
def updatevars(self):
dt = self.dt
T = self.temp
fr = self.fr
masses = self.masses
sigma = np.sqrt(2 * T * fr / masses)
self.c1 = dt / 2. - dt * dt * fr / 8.
self.c2 = dt * fr / 2 - dt * dt * fr * fr / 8.
self.c3 = np.sqrt(dt) * sigma / 2. - dt**1.5 * fr * sigma / 8.
self.c5 = dt**1.5 * sigma / (2 * np.sqrt(3))
self.c4 = fr / 2. * self.c5
def step(self, forces=None):
atoms = self.atoms
natoms = len(atoms)
if forces is None:
forces = atoms.get_forces(md=True)
# This velocity as well as rnd_pos, rnd_mom and a few other
# variables are stored as attributes, so Asap can do its magic
# when atoms migrate between processors.
self.v = atoms.get_velocities()
xi = self.rng.standard_normal(size=(natoms, 3))
eta = self.rng.standard_normal(size=(natoms, 3))
# When holonomic constraints for rigid linear triatomic molecules are
# present, ask the constraints to redistribute xi and eta within each
# triple defined in the constraints. This is needed to achieve the
# correct target temperature.
for constraint in self.atoms.constraints:
if hasattr(constraint, 'redistribute_forces_md'):
constraint.redistribute_forces_md(atoms, xi, rand=True)
constraint.redistribute_forces_md(atoms, eta, rand=True)
self.communicator.broadcast(xi, 0)
self.communicator.broadcast(eta, 0)
# To keep the center of mass stationary, we have to calculate
# the random perturbations to the positions and the momenta,
# and make sure that they sum to zero.
self.rnd_pos = self.c5 * eta
self.rnd_vel = self.c3 * xi - self.c4 * eta
if self.fix_com:
self.rnd_pos -= self.rnd_pos.sum(axis=0) / natoms
self.rnd_vel -= (self.rnd_vel *
self.masses).sum(axis=0) / (self.masses * natoms)
# First halfstep in the velocity.
self.v += (self.c1 * forces / self.masses - self.c2 * self.v +
self.rnd_vel)
# Full step in positions
x = atoms.get_positions()
# Step: x^n -> x^(n+1) - this applies constraints if any.
atoms.set_positions(x + self.dt * self.v + self.rnd_pos)
# recalc velocities after RATTLE constraints are applied
self.v = (self.atoms.get_positions() - x - self.rnd_pos) / self.dt
forces = atoms.get_forces(md=True)
# Update the velocities
self.v += (self.c1 * forces / self.masses - self.c2 * self.v +
self.rnd_vel)
# Second part of RATTLE taken care of here
atoms.set_momenta(self.v * self.masses)
return forces