Benchmark: compression energies of bulk fcc and rocksalt

Reference FHI-AIMS tight basis relativistic atomic_zora scalar equilibrium volumes compressed/expanded in the range 60% - 150% of the lattice constant.

Key-value pairs

key

description

name

Name of the system

relativistic

Non-relativistic and scalar-relativistic results included

project

Name of the project: “compression”

structure

fcc or rocksalt

Note that there are additional keys not explained above which are specific to the given calculator.

Results

First extract the data of the given code and insert it into a new database file. Then use the extract.py and energies.py scripts to write the csv formatted file using the data from the new database file.

# rocksalt aims tier2 basis relativistic none
ase-db compression.db project=compression,structure=rocksalt,calculator=aims,basis=tier2,relativistic=0 -i rocksalt.aims_tier2.nrel.db
python extract.py rocksalt.aims_tier2.nrel.db
python energies.py rocksalt.aims_tier2.nrel.db
# aims tier2 basis relativistic none
ase-db compression.db project=compression,structure=fcc,calculator=aims,basis=tier2,relativistic=0 -i fcc.aims_tier2.nrel.db
python extract.py fcc.aims_tier2.nrel.db
python energies.py fcc.aims_tier2.nrel.db
# aims tier2 basis relativistic atomic_zora scalar
ase-db compression.db project=compression,structure=fcc,calculator=aims,basis=tier2,relativistic=atomic_zora -i fcc.aims_tier2.srel.db
python extract.py fcc.aims_tier2.srel.db
python energies.py fcc.aims_tier2.srel.db
# elk nrel
ase-db compression.db project=compression,structure=fcc,calculator=elk,relativistic=0 -i fcc.elk.nrel.db
python extract.py fcc.elk.nrel.db
python energies.py fcc.elk.nrel.db
# elk srel
ase-db compression.db project=compression,structure=fcc,calculator=elk,relativistic=scalar -i fcc.elk.srel.db
python extract.py fcc.elk.srel.db
python energies.py fcc.elk.srel.db
# rocksalt exciting nrel
ase-db compression.db project=compression,structure=rocksalt,calculator=exciting,relativistic=0 -i rocksalt.exciting.nrel.db
python extract.py rocksalt.exciting.nrel.db
python energies.py rocksalt.exciting.nrel.db
# exciting nrel
ase-db compression.db project=compression,structure=fcc,calculator=exciting,relativistic=0 -i fcc.exciting.nrel.db
python extract.py fcc.exciting.nrel.db
python energies.py fcc.exciting.nrel.db
# exciting zora
ase-db compression.db project=compression,structure=fcc,calculator=exciting,relativistic=zora -i fcc.exciting.zora.db
python extract.py fcc.exciting.zora.db
python energies.py fcc.exciting.zora.db
# exciting iora*
ase-db compression.db project=compression,structure=fcc,calculator=exciting,relativistic='iora*' -i 'fcc.exciting.iora*.db'
python extract.py 'fcc.exciting.iora*.db'
python energies.py 'fcc.exciting.iora*.db'
# gpaw paw 09 nrel
ase-db compression.db project=compression,structure=fcc,calculator=gpaw,potentials_version=9.,relativistic=0 -i fcc.gpaw_paw09.nrel.db
python extract.py fcc.gpaw_paw09.nrel.db
python energies.py fcc.gpaw_paw09.nrel.db
# espresso gbrv 1.4
ase-db compression.db project=compression,structure=fcc,calculator=espresso,potentials=gbrv,potentials_version=1.4,relativistic=scalar -i fcc.espresso_gbrv_1.4.srel.db
python extract.py fcc.espresso_gbrv_1.4.srel.db
python energies.py fcc.espresso_gbrv_1.4.srel.db
# espresso sg15_oncv
ase-db compression.db project=compression,structure=fcc,calculator=espresso,potentials=sg15_oncv,potentials_version=20May2015,relativistic=scalar -i fcc.espresso_sg15_oncv.srel.db
python extract.py fcc.espresso_sg15_oncv.srel.db
python energies.py fcc.espresso_sg15_oncv.srel.db
# espresso sssp accurate 0.6
ase-db compression.db project=compression,structure=fcc,calculator=espresso,potentials=sssp_accurate,potentials_version=0.6,relativistic=scalar -i fcc.espresso_sssp_accurate.srel.db
python extract.py fcc.espresso_sssp_accurate.srel.db
python energies.py fcc.espresso_sssp_accurate.srel.db

The third step consists of eliminating all systems that show large scatter of results from the set. Use the https://svn.fysik.dtu.dk/projects/cmr2/trunk/compression/eliminate.sh script. This is due to the fact that it is difficult to obtain reliable results from all-electron codes.

The results can be plotted e.g. with https://svn.fysik.dtu.dk/projects/cmr2/trunk/compression/plot.py

Running the calculations again

Selected scripts used to obtain the results are available at https://svn.fysik.dtu.dk/projects/cmr2/trunk/compression