molecular dynamics simulation in ATK 11.8.

[fly]

If I execute an ab initio molecular dynamics simulation on 6-ZGNR
at room temperature (T=300 K) with a time step of 1 fs. Is it right for the following script? Do I need to make some revisions for some parameters in the following scrip? Who can provide the corresponding tutorial for molecular dynamics simulation in ATK 11.8. Thanks!


# -------------------------------------------------------------
# Bulk configuration
# -------------------------------------------------------------

# Set up lattice
vector_a = [10.0, 0.0, 0.0]*Angstrom
vector_b = [0.0, 23.56888, 0.0]*Angstrom
vector_c = [0.0, 0.0, 2.46100171044]*Angstrom
lattice = UnitCell(vector_a, vector_b, vector_c)

# Define elements
elements = [Carbon, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon,
            Carbon, Carbon, Carbon, Carbon, Hydrogen, Hydrogen]

# Define coordinates
cartesian_coordinates = [[  5.        ,  17.46788   ,   0.        ],
                         [  5.        ,   6.101     ,   0.        ],
                         [  5.        ,   6.81143   ,   1.23050086],
                         [  5.        ,   8.23229   ,   1.23050086],
                         [  5.        ,   8.94272   ,   0.        ],
                         [  5.        ,  10.36358   ,   0.        ],
                         [  5.        ,  11.07401   ,   1.23050086],
                         [  5.        ,  12.49487   ,   1.23050086],
                         [  5.        ,  13.2053    ,   0.        ],
                         [  5.        ,  14.62616   ,   0.        ],
                         [  5.        ,  15.33659   ,   1.23050086],
                         [  5.        ,  16.75745   ,   1.23050086],
                         [  5.        ,  18.56888   ,   0.        ],
                         [  5.        ,   5.        ,   0.        ]]*Angstrom

# Set up configuration
bulk_configuration = BulkConfiguration(
    bravais_lattice=lattice,
    elements=elements,
    cartesian_coordinates=cartesian_coordinates
    )

# -------------------------------------------------------------
# Calculator
# -------------------------------------------------------------
numerical_accuracy_parameters = NumericalAccuracyParameters(
    grid_mesh_cutoff=150.0*Hartree,
    k_point_sampling=(1, 1, 100),
    )

iteration_control_parameters = IterationControlParameters(
    tolerance=1e-05,
    )

calculator = LCAOCalculator(
    numerical_accuracy_parameters=numerical_accuracy_parameters,
    iteration_control_parameters=iteration_control_parameters,
    )

bulk_configuration.setCalculator(calculator)
nlprint(bulk_configuration)
bulk_configuration.update()
nlsave('analysis.nc', bulk_configuration)

initial_velocity = None
method = VelocityVerlet(
        time_step=1.0*femtoSecond,
        initial_velocity=initial_velocity
    )
molecular_dynamics = MolecularDynamics(
        bulk_configuration,
        constraints=[],
        log_filename='trajectory.nc',
        steps=1000,
        log_interval=1,
        method=method
    )
nlsave('analysis.nc', molecular_dynamics)
nlprint(molecular_dynamics)

[fly]

Who can offer me some advice? 

[zh]

The tips for performing MD are quite universal. The BOMD is implemented in ATK. You may refer to a book "Ab Initio Molecular Dynamics: Basic Theory and Advanced Methods" by Dominik Marx, Jürg Hutter.
The important parameter should be minded:
1) The choice of ensemble: NVE, NVT, or NPT.
2) The time step of MD;
3) The technique to control temperature;
4) The initial and final temperatures of the system under study.

The default setup in the MD simulation of ATK is to perform the simulation of a NVE.  From the manual, it seems that the functionality of MD part of ATK is limited. The setup for temperature is not explained there.

[fly]

Thank you for your reply. Can you present the full name for the abbreviations (BOMD, NVE, NVT, and NPT)? Thank you!

[Anders Blom]

These abbreviations are standard notation and important enough that rather than provide just a simple answer, I will encourage you to find out the proper way, since you need to learn to use and understand them anyway, in order to proceed with any meaningful MD studies.

It is worthwhile noticing that ATK since 11.8 also has support for MD simulation using DFTB and Slater-Koster models (and also a few EMT models, like the Brenner potential for hydrocarbons and Si). Also, in 12.2 we are improving the trajectory handling and allowing a better control of the initial velocities. Temperature control is missing yet, however, as well as a few other things needed to make ATK a fully-fledged MD code, but the basics are in place, and this is an area we continue to improve on, not least since we recently received an EU funding grant to do so.

[ruyam]

I have attached my python file. Its taking to much time for the results. Its more than 30hours still not getting the final result.
**I am a beginner with atk, can you please look into the file and suggest me the appropriate settings for such calculation? **

thanks in advance,
Mayur
Beginner, ATK

[ruyam]

PFA my python file

[Anders Blom]

Don't post new questions in old threads.