LobsterAutomation

This is a package to reproduce the publication on Lobster automation: J. George, G. Petretto, A. Naik, M. Esters, A. J. Jackson, R. Nelson, R. Dronskowski, G.-M. Rignanese, G. Hautier, ChemPlusChem, https://doi.org/10.1002/cplu.202200123

Start_Scripts includes two scripts to start all Lobster computations with pymatgen, fireworks, and atomate.

You need to download the data from our second zenodo.org repository first. You can get the data here: https://doi.org/10.5281/zenodo.6373369

wget https://zenodo.org/record/6373369/files/Results.zip?download=1

Please unzip the data and put it in the folder "LobsterAutomation".

You can then use the "Analysis_Scripts" to reproduce our publication.

• analyse_results_14_1_11_without_Yb5d.py will produce the automatic analysis for Yb14Mn1Sb11 with a basis set that does not include 5d (as in the manuscript, aka "small basis").
• analyse_results_14_1_11__without_Yb5d_anion_anion.py will produce the automatic analysis including all bonds for Yb14Mn1Sb11 with a basis set that does not include 5d (as in the manuscript, aka "small basis")
• analyse_results_14_1_11_with_Yb5d.py will produce the automatic analysis for Yb14Mn1Sb11 with a basis set that does include 5d (as shown in SI, aka "large basis").
• analyse_results_14_1_11__with_Yb5d_anion_anion.py will produce the automatic analysis including all bonds for Yb14Mn1Sb11 with a basis set that does include 5d (as shown in SI, aka "large basis")
• analyse_14_1_11_chemenv.py will produce the geometric analysis of coordination environments with ChemEnv
• analyze_reslts_14_1_11_DOS_Mn.py will compare the DOS for Mn with the two different basis sets
• analyze_reslts_14_1_11_DOS_Sb.py will compare the DOS for Sb with the two different basis sets
• analyze_reslts_14_1_11_DOS_Yb.py will compare the DOS for Yb with the two different basis sets
• analyse_results_GaN.py will produce all plots and outputs for the GaN phases
• analyse_results_NaCl.py will produce all plots and outputs for the NaCl phases
• analyse_oxynitrides.py will produce all plots and outputs for the oxynitride systems

In Analysis_Scripts/Output, you can find the outputs.

To produce the data, you can use the scripts in "Start_Scripts".

• start_workflows_bulk.py will start all computations except for the oxynitride computations
• start_defect_structures.py will start the oxynitride computations

The following version numbers are needed for the workflows:

• pymatgen 2022.2.1
• atomate 1.0.3
• enumlib 2.0.4 (Installation with conda)

To analyze the results, you would need Lobsterpy 0.2.1 (which is compatible only with a later version of pymatgen)

• LobsterPy

Computing times

The reported computing times are for Intel Xeon E7-4820 v4 CPU type (128 GB RAM per node). We are reporting here the computing times for all the compounds with ISPIN=2 presented in the paper (for the sake of brevity we have excluded oxynitrides and calculations with ISPIN=1). Objective is just to provide an general overview of time required for performing the calculations and not to account here all the computing times of our calculations.

NaCl

Material ID	Structure type	Calc Type	Computing time (secs)	NCORES	NODES
mp-22851	NaCl	VASP	704.975	80	2
		lobster	38	40	1
mp-22862	CsCl	VASP	798.869	80	2
		lobster	30	40	1

GaN

Material ID	Structure type	Calc Type	Computing time (secs)	NCORES	NODES
mp-804	Wurtzite	VASP	2568.488	80	2
		lobster	265	40	1
mp-830	Zinc-blende	VASP	2646.004	80	2
		lobster	211	40	1
mp-1007824	BN	VASP	2491.302	80	2
		lobster	200	40	1
mp-2853	NaCl	VASP	2494.205	80	2
		lobster	165	40	1

Yb14MnSb11

Material ID	Structure type	Calc Type	Computing time (secs)	NCORES	NODES
mp-568088	--	VASP	19729.590	80	2
		lobsterbasis_0	10,215	40	1
		lobsterbasis_1	92,898	40	1