Input YAML Reference

ARC input files are YAML dictionaries passed directly to ARC(**input_dict). Top-level keys must match arguments accepted by arc.main.ARC. Entries under species are converted to ARCSpecies objects, and entries under reactions are converted to ARCReaction objects.

Use this page as a practical checklist of accepted keys. For deeper behavior, see the API reference pages for ARC, ARCSpecies, ARCReaction, and Level.

Minimal Shape

project: my_project

species:
  - label: ethanol
    smiles: CCO

level_of_theory: wb97xd/def2svp
ess_settings:
  gaussian: local

Top-Level Keys

Project and execution:

project - project name. Required for new projects.
project_directory - local directory for project output.
verbose - logging level.
running_jobs - restart state for submitted jobs.
output - restart/status state for species.
output_multi_spc - restart/status state for multi-species workflows.

Species and reactions:

species - list of species dictionaries.
reactions - list of reaction dictionaries.
dont_gen_confs - species labels for which conformer generation should be avoided when XYZ is supplied.
n_confs - number of low-energy force-field conformers to keep. Default: 10.
e_confs - conformer energy window in kJ/mol. Default: 5.0.
allow_nonisomorphic_2d - allow optimization even when generated 3D geometry is not isomorphic to the 2D graph.

Levels of theory:

level_of_theory - shortcut for composite or sp//opt style levels.
composite_method - composite method level.
conformer_level - alias for conformer_opt_level.
conformer_opt_level - conformer optimization level.
conformer_sp_level - conformer single-point level.
ts_guess_level - level for TS guess comparison/optimization.
opt_level - geometry optimization level.
freq_level - frequency level.
sp_level - single-point energy level.
scan_level - rotor scan level.
irc_level - IRC level.
orbitals_level - orbitals level.
arkane_level_of_theory - level used for Arkane atom and bond corrections.
adaptive_levels - level choices by heavy-atom count.

Job selection and resources:

job_types - dictionary of job type booleans.
specific_job_type - run one job family; takes precedence over job_types.
job_memory - memory per job in GB.
max_job_time - wall-time limit in hours.
keep_checks - keep ESS checkfiles.
trsh_ess_jobs - troubleshoot failed ESS jobs. Default: true.
trsh_rotors - troubleshoot failed rotor scans. Default: true.
skip_nmd - skip normal-mode-displacement checks.
report_e_elect - report electronic energy.

Software and statmech:

ess_settings - map ESS names to server names for this project.
thermo_adapter - thermochemistry adapter. Default: Arkane.
kinetics_adapter - kinetics adapter. Default: Arkane.
compare_to_rmg - compare calculated data to RMG database values. Default: true.
compute_thermo - compute thermodynamic properties. Default: true.
compute_rates - compute rate coefficients. Default: true.
compute_transport - compute transport properties. Default: false.
bac_type - bond additivity correction type: p, m, or null.
freq_scale_factor - harmonic frequency scaling factor.
calc_freq_factor - calculate a scaling factor if one cannot be found. Default: true.
bath_gas - bath gas for OneDMin/Lennard-Jones workflows.

Kinetics conditions and TS search:

T_min - minimum temperature tuple, e.g. [500, K].
T_max - maximum temperature tuple, e.g. [3000, K].
T_count - number of temperature points. Default: 50.
ts_adapters - TS search adapters to try, such as heuristics, AutoTST, GCN, xtb_gsm, and orca_neb.

Species Entries

Each entry under species may use the following keys:

Identity and structure:

label - species label.
smiles - SMILES string.
inchi - InChI string.
adjlist - RMG adjacency list.
xyz - XYZ string, XYZ dictionary, list of XYZ entries, or path to a supported coordinate/input/output file.
yml_path - path to a species YAML file.
mol - RMG Molecule object when using the Python API.
species_dict - dictionary representation used in restarts/API workflows.

Charge, spin, and TS metadata:

charge - integer charge.
multiplicity - spin multiplicity.
number_of_radicals - radical count override.
is_ts - whether this species is a transition state.
ts_number - TS index.
rxn_label - associated reaction label.
rxn_index - associated reaction index.
irc_label - IRC label.

Thermo and output behavior:

compute_thermo - compute thermo for this species.
include_in_thermo_lib - include species in output thermo library.
e0_only - compute/store only E0 where applicable.
bond_corrections - custom bond correction data.
bdes - bond dissociation energy atom-pair list.
checkfile - ESS checkfile path.
run_time - restart/runtime metadata.

Conformers, rotors, and geometry controls:

force_field - force field for conformer generation. Default: MMFF94s.
directed_rotors - directed rotor scan requests.
preserve_param_in_scan - internal coordinates to preserve during scans.
consider_all_diastereomers - consider all diastereomers during conformer generation.
fragments - fragment atom indices.
active - active-space metadata.
external_symmetry - external symmetry number.
optical_isomers - number of optical isomers.
multi_species - multi-species label.
keep_mol - preserve the molecule object.
project_directory - species-specific project directory metadata.

Reaction Entries

Each entry under reactions may use the following keys:

label - reaction label, commonly A + B <=> C + D.
reactants - list of reactant labels.
products - list of product labels.
r_species - reactant ARCSpecies objects when using the Python API.
p_species - product ARCSpecies objects when using the Python API.
ts_label - associated TS species label.
ts_xyz_guess - TS guess or list of guesses.
xyz - alias for ts_xyz_guess.
family - RMG reaction family.
multiplicity - reaction PES multiplicity.
charge - reaction PES charge.
preserve_param_in_scan - scan constraints for the reaction/TS.
kinetics - user-supplied kinetics data.
reaction_dict - dictionary representation used in restarts/API workflows.
species_list - species object list used when reconstructing from dictionaries.

Level Dictionaries

Any level key can be a string:

opt_level: wb97xd/def2tzvp

or a dictionary:

sp_level:
  method: DLPNO-CCSD(T)-F12
  basis: cc-pVTZ-F12
  auxiliary_basis: aug-cc-pVTZ/C
  cabs: cc-pVTZ-F12-CABS
  software: orca

Accepted level dictionary keys are:

method
basis
auxiliary_basis
dispersion
cabs
method_type
software
software_version
compatible_ess
solvation_method
solvent
solvation_scheme_level
args
year

Use year only for Arkane correction matching, not inside QC method names.

Job Types

Current job type keys are:

conf_opt
conf_sp
opt
fine
freq
sp
rotors
irc
orbitals
onedmin
bde

Legacy aliases are normalized by ARC:

fine_grid -> fine
lennard_jones -> onedmin

Example:

job_types:
  conf_opt: true
  opt: true
  fine: true
  freq: true
  sp: true
  rotors: false

ESS Settings

ess_settings maps software names to configured server names:

ess_settings:
  gaussian:
    - local
    - cluster_a
  orca: local
  molpro: cluster_b

Supported ESS keys include cfour, gaussian, mockter, molpro, orca, qchem, terachem, onedmin, xtb, torchani, and openbabel. TS adapters such as heuristics, AutoTST, GCN, xtb_gsm, and orca_neb are configured through ts_adapters and related installation/settings.

Complete Example

project: ethanol_reference_demo
project_directory: /tmp/arc_ethanol_reference_demo

ess_settings:
  gaussian: local

job_types:
  conf_opt: true
  opt: true
  fine: true
  freq: true
  sp: true
  rotors: true
  irc: false

level_of_theory: wb97xd/def2svp
sp_level:
  method: wb97xd
  basis: def2tzvp
  software: gaussian

job_memory: 14
max_job_time: 24
compute_thermo: true
compute_rates: false
bac_type: null

species:
  - label: ethanol
    smiles: CCO
    multiplicity: 1
    charge: 0