Brownian Dynamics
-----------------

In the limit of high friction, stochastic dynamics reduces to Brownian
dynamics, also called position Langevin dynamics. This applies to
over-damped systems, *i.e.* systems in which the inertia effects are
negligible. The equation is

.. math:: {{\mbox{d}}\mathbf{r}_i \over {\mbox{d}}t} = \frac{1}{\gamma_i} \mathbf{F}_i(\mathbf{r}) + {\stackrel{\circ}{\mathbf{r}}}_i
          :label: eqnbrowniandyn

where :math:`\gamma_i` is the friction coefficient
:math:`[\mbox{amu/ps}]` and
:math:`{\stackrel{\circ}{\mathbf{r}}}_i(t)` is a noise
process with
:math:`\langle {\stackrel{\circ}{r}}_i(t) {\stackrel{\circ}{r}}_j(t+s) \rangle = 2 \delta(s) \delta_{ij} k_B T / \gamma_i`.
In |Gromacs| the equations are integrated with a simple, explicit scheme

.. math:: \mathbf{r}_i(t+\Delta t) = \mathbf{r}_i(t) +
          {\Delta t \over \gamma_i} \mathbf{F}_i(\mathbf{r}(t)) 
          + \sqrt{2 k_B T {\Delta t \over \gamma_i}}\, {\mathbf{r}^G}_i,
          :label: eqnbrowniandynint

where :math:`{\mathbf{r}^G}_i` is Gaussian distributed
noise with :math:`\mu = 0`, :math:`\sigma = 1`. The friction
coefficients :math:`\gamma_i` can be chosen the same for all particles
or as :math:`\gamma_i = m_i\,\gamma_i`, where the friction constants
:math:`\gamma_i` can be different for different groups of atoms. Because
the system is assumed to be over-damped, large timesteps can be used.
LINCS should be used for the constraints since SHAKE will not converge
for large atomic displacements. BD can be activated by using
:mdp-value:`integrator=bd` and the simulations are run using the
:ref:`mdrun <gmx mdrun>` program.

In BD there are no velocities, so there is also no kinetic energy. Still
:ref:`gmx mdrun` will report a kinetic energy and temperature based on
atom displacements per step :math:`\Delta x`. This can be used to judge
the quality of the integration. A too high temperature is an indication
that the time step chosen is too large. The formula for the kinetic
energy term reported is:

.. math:: \frac{1}{2} \sum_i \frac{\gamma_i \Delta x_i^2}{2 \, \Delta t}
      :label: eqnbrowniandynekin
