yambo Documentation

PWscf to yambo interface (p2y)

The p2y interface can read and process bandstructure output generated from the PWscf (Quantum-Espresso) package (http://www.pwscf.org). Linking libraries and routines from different codes is tricky, so read the following instructions thoroughly and carefully.

There are in fact two separate interface procedures you can follow:

The direct approach, using qexml (valid for PWscf 3.1 or later):
1. Perform standard nscf or bands calculation with pw.x to generate [prefix].save/ directory and contents.
2. Convert content of [prefix].save/ directory into SAVE/(n)s.db1, SAVE/(n)s.wf files using p2y.

The indirect approach, using pw_export (valid for PWscf 2.1.5 or later):
1. Perform standard nscf or bands calculation with pw.x to generate [filename].save/ directory OR *.save, *.pot, *.rho files.
2. Convert output of step (1) into XML-like format using the pw_export.x program: generate [filename].export/ directory containing index.xml, *.grid, *.wfc files.
3. Convert output of step (2) into SAVE/(n)s.db1, SAVE/(n)s.wf files using p2y.

We recommend that you use the direct approach, which is better supported and updated.

Please Note: The direct interface with PWscf version 4.0/4.0.1 is not available in the tarfile. It will be available in the SVN version by mid September. In the meantime, you might either try the indirect approach, or use the previous version of pw.x (e.g. 3.2.3)

Kleinman-Bylander form factors are not created with p2y: if these are important to you, you might be better off using the Abinit convertor, a2y. Otherwise, you might think about writing the code yourself...!

Follow the links for detailed instructions for setting up and using p2y:

Direct approach.
Indirect approach.
Tips and troubleshooting advice.

Running p2y

Although the default bin/p2y executable should work as is, various command line options are supported to help you during the conversion process. These are:

 -h             :Short Help
 -H             :Long Help
 -N             :Skip MPI initialization
 -F opt         :XML index file name
 -O opt         :Output directory
 -S             :DataBases fragmentation
 -a real        :Lattice constants rescaling factor
 -t             :Force no TR symmetry
 -y             :Force no symmetries
 -w             :Force no wavefunctions
 -v             :Verbose wfc I/O reporting

Most of these are self explanatory, so just a few comments:

 -F <opt>       :XML index file name

Specify the name of the XML-like index file name. Usually you should not need to set this flag, since the default values are determined when you compile p2y. The default is index.xml for the indirect approach, and data-file.xml for the direct approach.

 -S             :DataBases fragmentation

This option will generate subdirectories, by k-point and spin component, inside the SAVE directory. This is recommended for large systems with few k-points.

 -v             :Verbose wfc I/O reporting

This will force some extra reporting of the progress for reading and writing the wavefunctions, which can be useful to see if the conversion is working immediately.