qutip/core/blochredfield.py
import os
import inspect
import numpy as np
from qutip.settings import settings as qset
from . import Qobj, QobjEvo, liouvillian, coefficient, sprepost
from ._brtools import SpectraCoefficient, _EigenBasisTransform
from .cy.coefficient import InterCoefficient, Coefficient
from ._brtensor import _BlochRedfieldElement
__all__ = ['bloch_redfield_tensor', 'brterm']
def bloch_redfield_tensor(H, a_ops, c_ops=[], sec_cutoff=0.1,
fock_basis=False, sparse_eigensolver=False,
br_dtype='sparse'):
"""
Calculates the Bloch-Redfield tensor for a system given
a set of operators and corresponding spectral functions that describes the
system's coupling to its environment.
Parameters
----------
H : :class:`qutip.Qobj`, :class:`qutip.QobjEvo`
System Hamiltonian.
a_ops : list of (a_op, spectra)
Nested list of system operators that couple to the environment,
and the corresponding bath spectra.
a_op : :class:`qutip.Qobj`, :class:`qutip.QobjEvo`
The operator coupling to the environment. Must be hermitian.
spectra : :obj:`.Coefficient`, func, str
The corresponding bath spectra.
Can be a :obj:`.Coefficient` using an 'w' args, a function of the
frequency or a string. The :class:`SpectraCoefficient` can be used
for array based coefficient.
The spectra can depend on ``t`` if the corresponding
``a_op`` is a :obj:`.QobjEvo`.
Example:
.. code-block::
a_ops = [
(a+a.dag(), ('w>0', args={"w": 0})),
(QobjEvo(a+a.dag()), 'w > exp(-t)'),
(QobjEvo([b+b.dag(), lambda t: ...]), lambda w: ...)),
(c+c.dag(), SpectraCoefficient(coefficient(array, tlist=ws))),
]
c_ops : list
List of system collapse operators.
sec_cutoff : float {0.1}
Cutoff for secular approximation. Use ``-1`` if secular approximation
is not used when evaluating bath-coupling terms.
fock_basis : bool {False}
Whether to return the tensor in the input basis or the diagonalized
basis.
sparse_eigensolver : bool {False}
Whether to use the sparse eigensolver
br_dtype : ['sparse', 'dense', 'data']
Which data type to use when computing the brtensor.
With a cutoff 'sparse' is usually the most efficient.
Returns
-------
R, [evecs]: :class:`qutip.Qobj`, tuple of :class:`qutip.Qobj`
If ``fock_basis``, return the Bloch Redfield tensor in the laboratory
basis. Otherwise return the Bloch Redfield tensor in the diagonalized
Hamiltonian basis and the eigenvectors of the Hamiltonian as hstacked
column.
"""
R = liouvillian(H, c_ops)
H_transform = _EigenBasisTransform(QobjEvo(H), sparse_eigensolver)
if fock_basis:
for (a_op, spectra) in a_ops:
R += brterm(H_transform, a_op, spectra, sec_cutoff, True,
br_dtype=br_dtype)
return R
else:
# When the Hamiltonian is time-dependent, the transformation of `L` to
# eigenbasis is not optimized.
if isinstance(R, QobjEvo):
# The `sprepost` will be computed 2 times for each parts of `R`.
# Compressing the QobjEvo will lower the number of parts.
R.compress()
evec = H_transform.as_Qobj()
R = sprepost(evec, evec.dag()) @ R @ sprepost(evec.dag(), evec)
for (a_op, spectra) in a_ops:
R += brterm(H_transform, a_op, spectra, sec_cutoff,
False, br_dtype=br_dtype)[0]
return R, H_transform.as_Qobj()
def brterm(H, a_op, spectra, sec_cutoff=0.1,
fock_basis=False, sparse_eigensolver=False, br_dtype='sparse'):
"""
Calculates the contribution of one coupling operator to the Bloch-Redfield
tensor.
Parameters
----------
H : :class:`qutip.Qobj`, :class:`qutip.QobjEvo`
System Hamiltonian.
a_op : :class:`qutip.Qobj`, :class:`qutip.QobjEvo`
The operator coupling to the environment. Must be hermitian.
spectra : :obj:`.Coefficient`, func, str
The corresponding bath spectra.
Can be a :obj:`.Coefficient` using an 'w' args, a function of the
frequency or a string. The :class:`SpectraCoefficient` can be used for
array based coefficient.
The spectra can depend on ``t`` if the corresponding
``a_op`` is a :obj:`.QobjEvo`.
Example:
coefficient('w>0', args={"w": 0})
SpectraCoefficient(coefficient(array, tlist=...))
sec_cutoff : float {0.1}
Cutoff for secular approximation. Use ``-1`` if secular approximation
is not used when evaluating bath-coupling terms.
fock_basis : bool {False}
Whether to return the tensor in the input basis or the diagonalized
basis.
sparse_eigensolver : bool {False}
Whether to use the sparse eigensolver on the Hamiltonian.
br_dtype : ['sparse', 'dense', 'data']
Which data type to use when computing the brtensor.
With a cutoff 'sparse' is usually the most efficient.
Returns
-------
R, [evecs]: :obj:`.Qobj`, :obj:`.QobjEvo` or tuple
If ``fock_basis``, return the Bloch Redfield tensor in the outside
basis. Otherwise return the Bloch Redfield tensor in the diagonalized
Hamiltonian basis and the eigenvectors of the Hamiltonian as hstacked
column. The tensors and, if given, evecs, will be :obj:`.QobjEvo` if
the ``H`` and ``a_op`` is time dependent, :obj:`.Qobj` otherwise.
"""
if isinstance(H, _EigenBasisTransform):
Hdiag = H
else:
Hdiag = _EigenBasisTransform(QobjEvo(H), sparse=sparse_eigensolver)
# convert spectra to Coefficient
if isinstance(spectra, str):
spectra = coefficient(spectra, args={'w': 0})
elif isinstance(spectra, InterCoefficient):
spectra = SpectraCoefficient(spectra)
elif isinstance(spectra, Coefficient):
pass
elif callable(spectra):
sig = inspect.signature(spectra)
if tuple(sig.parameters.keys()) == ("w",):
spectra = SpectraCoefficient(coefficient(spectra))
else:
spectra = coefficient(spectra, args={'w': 0})
else:
raise TypeError("a_ops's spectra not known")
sec_cutoff = sec_cutoff if sec_cutoff >= 0 else np.inf
R = QobjEvo(_BlochRedfieldElement(Hdiag, QobjEvo(a_op), spectra,
sec_cutoff, not fock_basis, dtype=br_dtype))
if (
((isinstance(H, _EigenBasisTransform) and H.isconstant)
or isinstance(H, Qobj))
and isinstance(a_op, Qobj)
):
R = R(0)
return R if fock_basis else (R, Hdiag.as_Qobj())