Quick tip: Constraining molecular coordinates

Tuesday, July 12, 2011

Have you ever run a relaxation or a path optimization technique (e.g. nudged elastic band or string metadynamics) on an isolated system only to find that you were getting spurious answers arising from rotation or translation of your system? Have you ever wanted to fix one part of your molecule and only relax another portion of it? Have you ever tried to optimize donor-acceptor distances between two molecules or molecular fragments?  Well, there are actually two distinct ways to optimize structures with some measure of constraint on positions in the PWscf portion of Quantum-ESPRESSO.

 

The simple way: fixing cartesian coordinates

In the ATOMIC_POSITIONS namelist of your input file, you can add additional flags in the fifth through seventh column alongside each atom to freeze coordinates.  The column ordering of these flags follows the same as the coordinates in the 2nd-4th column and a ‘0’ freezes the particular x-, y-, or z-position of each atom, while a ‘1’ corresponds to allowing the atom to move freely in that direction.  You don’t have to set this flag for each atom, but once you set a single coordinate flag on a given atom, you should set the flag for each x-, y-, and z-coordinate.   This method for freezing coordinates works alongside all other approaches including both nudged elastic band and structural relaxations.

 

The (slightly) harder way: defining internal coordinate constraints

Although PWscf does not currently support a z-matrix or internal coordinate form for the atomic positions, it is possible to impose constraints that are defined in terms of internal coordinate parameters.  To carry out a constrained relaxation, we set up a relaxation with ion_dynamics=‘damp’and add an additional CONSTRAINTS namelist to the end of the input file after K_POINTS:

    0 CONSTRAINTS

The first line of this namelist should include the number of constraints you wish to apply and the tolerance on achieving those constraints:

    1 <num_constraints> <tolerance>

The remaining lines are where you define the constraints in terms of the atom number of each atom in the constraint as defined by its order in yourATOMIC_POSITIONS input:

    2+ <constr_type>, <at #s comma delimited>, <constraint value>

 

More specifically, for each constraint, line 2 and onwards can be defined with any combination of the following:

 

distance:    ‘distance’, <#at1>, <#at2>, <tgt distance> 

                                                                (in bohrs!)

angle:       ‘planar_angle’, <#at1>,<#at2>,<#at3>, <tgt angle> 

                                       ijk angle with 2=j vertex      (in degrees!)

dihedral:    ‘torsional_angle’,<at#1>,<at#2>,<at#3>,<at#4>,<tgt dihed> 

                                       ijkl/1234 order with 2&3 central   (in degrees!)

 

Some additional constraints on other quantities such as coordination number are also available, and I encourage you to read the INPUT_PW file that is in the Doc folder of your Quantum-ESPRESSO installation root for more details.

 

I hope that this quick tip has helped you to better understand how to make the most of your structural relaxations in PWscf.  Please email me if you have any additional questions not answered here!

About Us

The Kulik group focuses on the development and application of new electronic structure methods and atomistic simulations tools in the broad area of catalysis.

Our Interests

We are interested in transition metal chemistry, with applications from biological systems (i.e. enzymes) to nonbiological applications in surface science and molecular catalysis.

Our Focus

A key focus of our group is to understand mechanistic features of complex catalysts and to facilitate and develop tools for computationally driven design.

Contact Us

Questions or comments? Let us know! Contact Dr. Kulik: