Commands

General

This document presents details about the ADAQ commands. They are divide based on project, defect generation, workflow, and database.

Project

Make an ADAQ project with the following command:

$ adaq-set-up-project

If you are uncertin of your progress, run the following command:

$ adaq-next-step

If you want to calculate additional properties based on the screen results, you can use the following command to create a new ADAQ project for reruns:

$ adaq-set-up-rerun <path/to/old_data>

Defect Generation

To visualize any cell, use the following command:

$ adaq-show-cell <path>

where <path> is the path to the cell you would like to show, for example the relaxed unitcell (/unitcell/ht.task.*.finished/ht.run.*/CONTCAR.relaxfinal.bz2).

To generate defects, use the following command:

$ adaq-generate-defects

You can adjust the defect generation with the parameters in the parameter.py file. It contains the following parameters:

stacking_sequence - This is only relevant for layered material to keep track of with layer the different defect belong to. Leave as empty list for other materials.
minimum_defect_distance - Set the minimum distance between defect clusters and periodic copies. Can also be viewed as the supercell size. Given in Ångstrom with default: 20.0
vacancy - Generate vacancies, True or False.
substitutional - Generate substitutionals, True or False.
interstitial - Generate interstitials, True or False.
dopand_string - Which elements to dope with. default: “intrinsic”, one can also add “+” followed by the block in the peridoc table or write “all” for all dopants. It is also possible to specify a list of specific elements like “[Si,N]”.
only_single_dopands - Make only dopant cluster with single extrinsic element. If false, it is possible to make H and Li clusters in SiC. default: True
cluster_size - How large defect clusters should be, default: 2
min_distance - For defect clusters, provide the min pairwise defect distance in Ångstrom, default: 1.0
max_distance - For defect clusters, provide the max pairwise defect distance in Ångstrom, default: 3.5
int_distance - For interstitial generation, provide distance between interstitial positions in Ångstrom, default: 0.5. Read more about this parameter below.

It can be tricky to set the number of interstital positions. Hence, it can be tested before hand. Test difference value of the int_distance in the parameter.py file and visualize the number of interstital position that are generated with the command:

$ adaq-show-interstitials-in-unitcell

Workflow

Interface with the unitcell workflow:

$ adaq-workflow-relax-unitcell <arg1> <arg2> <arg3>

Arguments:

setup, send, run, receive
select which computer to run
number of taskmananger to run (optional)

Interface with the node scaling workflow:

$ adaq-workflow-node_scaling <arg1> <arg2> <arg3>

Arguments:

run, status, collect, result
select which computer to run
select number of nodes (default: 10, only needed for run argument)

Interface with the host workflow:

$ adaq-workflow-calculate-host <arg1> <arg2> <arg3>

Arguments:

setup, send, run, receive
select which computer to run
number of taskmananger to run (optional)

Interface with the screening workflow:

$ adaq-workflow-screen-defects <arg1> <arg2> <arg3>

Arguments:

setup, send, run, receive, check
select which computer to run
number of taskmananger to run (optional)

Note

This workflow omits interstitial-interstitial cluster. To turn this off, go to ADAQ/workflows/4_screen/run.py and comment out search.add(~search_defectinfo.defect_name.like('%Int%Int%'))

Interface with the full workflow:

$ adaq-workflow-full <arg1> <arg2> <arg3>

Arguments:

setup, send, run, receive
select which computer to run
select which defect to run for setup or number of taskmananger to run (optional)

Interface with the lookup tables:

$ adaq-lookup <arg1> <arg2>

Arguments:

select workflow, either screen or full
select action, either display or rename

Database

The following command goes through all data and makes the defects.sqlite database. It removes any older versions and make a new if the command is run again.

$ adaq-rebuild-database <arg1>

Arguments:

select what to include in the database, either light (skips the relaxed defect structures) or full (default). The longest part is storing relaxed structures. Using the struct argument, it is possible to store these separately in a database called structures.sqlite. When this database is present, the full database can be built faster.

It is also possible to extract data from the the database. To get a formation energy plot, run the following command:

$ adaq-database-plot-formation-energy <arg1>

Arguments:

is the defect key, which is unique for the defect

To get a eigenvalues energy plot, run the following command:

$ adaq-database-plot-eigenvalues <arg1> <arg2> <arg3> <arg4>

Arguments:

is the defect key, which is unique for the defect
select which workflow to extract the data from, ex: screen
select which charge, ex: -1
select which spin, ex: 1.0
select which state, ground (default) or excited

To show the defect cell, run the following command:

$ adaq-database-show-cell <arg1> <arg2> <arg3> <arg4>

Arguments:

is the defect key, which is unique for the defect
select which workflow to extract the data from, ex: screen
select which charge, ex: -1
select which spin, ex: 1.0
select which state, ground (default) or excited

Note

This command works even if the database is built with light since it access the relaxed files directly.

Get a copy of the defect cell, run the following command:

$ adaq-database-get-relaxed-structure <arg1> <arg2> <arg3> <arg4>

Arguments:

is the defect key, which is unique for the defect
select which workflow to extract the data from, ex: screen
select which charge, ex: -1
select which spin, ex: 1.0
select which state, ground (default) or excited

The file is saved as: CONTCAR_defect_charge_spin_state.vasp.