:orphan:

.. raw:: html

  <div class="container-fluid">
    <div class="row">
      <div class="col-md-10">
        <div style="text-align: justify; text-justify: inter-word;">

Diffusion Analysis in a Pore (MC Method)
========================================

The MC diffusion analysis needs the sampled object file using the mc
diffusion routine

.. code-block:: python

    import porems as pms
    import poreana as pa
    import matplotlib.pyplot as plt


    # Load molecule
    mol = pms.Molecule("benzene", "BEN", inp="data/benzene.gro")

    # Set step length
    len_steps = [1,2,5,10,20,30,40]

    # Set frame length
    len_frame = 2e-12

    # Sample transition matrix
    sample = pa.Sample("data/pore_system_cylinder.obj", "data/traj_cylinder.xtc", mol)
    sample.init_diffusion_mc("output/diff_mc_cyl_s.h5", len_steps, len_frame=len_frame)
    sample.sample(is_parallel=False)

``Finished frame 2001/2001...``


With the sampling obj-file the transition matrix can be plotted

.. code-block:: python

    # Set kwargs for the heatmap
    kwargs = {"vmin":0,"vmax":0.5, "xticklabels":30, "yticklabels":30, "cbar":True, "square":False}

    # Plot transition matrix for a step length of 10
    pa.diffusion.mc_trans_mat("output/diff_mc_cyl_s.h5", 10, kwargs)


.. figure::  /pics/diffusion_mc_01.png
      :align: center
      :width: 50%
      :name: fig1


After sampling, a model has to set and the MC Alogirthm started

.. code-block:: python

    # Set Cosine Model for diffusion and energy profile
    model = pa.CosineModel("output/diff_mc_cyl_s.h5", 6, 10)

    # Do the MC Algorithm
    pa.MC().run(model,"output/diff_mc.h5", nmc_eq=5000, nmc=5000, print_output=False, is_parallel=False)


``MC Calculation Start``

``...``

``MC Calculation Done.``


The results of the MC Alogrithm the diffusion can be calculated

.. code-block:: python

    # Print the results for the normal diffusion
    diff,diff_mean,diff_table = pa.diffusion.mc_fit("output/diff_mc.h5")


``Diffusion axial: 1.6913e-09 m^2/s``

``Mean Diffusion axial: 1.6777e-09 m^2/s``

``Standard deviation: 6.9341e-11 m^2/s``


.. figure::  /pics/diffusion_mc_02.svg
      :align: center
      :width: 50%
      :name: fig2


or the diffusion and free energy profile over the entire system can be
displayed

.. code-block:: python

    # Plot diffusion profile over the simulation box
    pa.diffusion.mc_profile("output/diff_mc.h5", infty_profile=True)

    # Plot free energy profile over the simulation box
    pa.freeenergy.mc_profile("output/diff_mc.h5", [10])


.. figure::  /pics/diffusion_mc_03.svg
      :align: center
      :width: 100%
      :name: fig3

Additionally, the pore area can be considered more closely

.. code-block:: python

    # Plot the lag time extrapolation for the pore ares
    pa.diffusion.mc_fit("output/diff_mc.h5", section="pore")

    # Plot diffusion profile in a pore
    pa.diffusion.mc_profile("output/diff_mc.h5", section="pore", infty_profile=True)



``Diffusion axial (Pore): 1.2534e-09 m^2/s``

``Mean Diffusion axial (Pore): 1.3417e-09 m^2/s``

``Standard deviation: 3.1949e-10 m^2/s``

.. figure::  /pics/diffusion_mc_04.svg
      :align: center
      :width: 100%
      :name: fig4


.. raw:: html

        </div>
      </div>
    </div>
  </div>