pyscf.md package#

Submodules#

pyscf.md.distributions module#

pyscf.md.distributions.MaxwellBoltzmannVelocity(mol, T=298.15, rng=Generator(PCG64) at 0x7F8458FA0AC0)[source]#

Computes velocities for a molecular structure using a Maxwell-Boltzmann distribution. Args:

mol : gto.mol object

Tfloat
Temperature, in Kelvin, that the distribution represents.

rngnp.random.Generator
Random number generator to sample from. Must contain a method normal. Default is to use the md.rng which is a np.random.Generator

Returns:: Velocities as a ndarray of dimension (natm, 3) in atomic units.

pyscf.md.integrators module#

class pyscf.md.integrators.Frame(ekin=None, epot=None, coord=None, veloc=None, time=None)[source]#

Bases: object

Basic class to hold information at each time step of a MD simulation

Attributes:

ekinfloat: Kinetic energy
epotfloat: Potential energy (electronic energy)
etotfloat: Total energy, sum of the potential and kinetic energy
coord2D array with shape (natm, 3): Geometry of the system at the current time step
veloc2D array with shape (natm, 3): Velocities of the system at the current time step
timefloat: Time for which this frame represents

class pyscf.md.integrators.NVTBerendson(method, T, taut, **kwargs)[source]#

Bases: _Integrator

Berendsen (constant N, V, T) molecular dynamics

Args:

method : lib.GradScanner or rhf.GradientsMixin instance, or has nuc_grad_method method.

Method by which to compute the energy gradients and energies in order to propagate the equations of motion. Realistically, it can be any callable object such that it returns the energy and potential energy gradient when given a mol.

Tfloat: Target temperature for the NVT Ensemble. Given in K.
tautfloat: Time constant for Berendsen temperature coupling. Given in atomic units.

Attributes:

accelndarray: Current acceleration for the simulation. Values are given in atomic units (Bohr/a.u.^2). Dimensions is (natm, 3) such as

[[x1, y1, z1], [x2, y2, z2], [x3, y3, z3]]

class pyscf.md.integrators.VelocityVerlet(method, **kwargs)[source]#

Bases: _Integrator

Velocity Verlet algorithm

Args:

method : lib.GradScanner or rhf.GradientsBase instance, or has nuc_grad_method method.

Method by which to compute the energy gradients and energies in order to propagate the equations of motion. Realistically, it can be any callable object such that it returns the energy and potential energy gradient when given a mol.

Attributes:

accelndarray: Current acceleration for the simulation. Values are given in atomic units (Bohr/a.u.^2). Dimensions is (natm, 3) such as

[[x1, y1, z1], [x2, y2, z2], [x3, y3, z3]]

pyscf.md.integrators.kernel(integrator, verbose=3)[source]#

Module contents#

Molecular Dynamics#

Simple usage:

>>> from pyscf import gto, dft
>>> import pyscf.md as md
>>> mol = gto.M(atom='N 0 0 0; N 0 0 1', basis='def2-tzvp')
>>> mf = dft.RKS(mol)
>>> mf.xc = 'pbe,pbe'
>>> integrator = md.NVE(mf, dt=5, time=10).run()

pyscf.md.set_seed(seed)[source]#: Sets the seed for the random number generator used by the md module