ORCA Frequency Calculation Skill

When to Use

Use this skill when the user wants to:

• Compute vibrational frequencies of a molecule
• Get an IR spectrum
• Calculate zero-point energy (ZPE)
• Obtain thermochemical quantities (enthalpy, entropy, Gibbs free energy)
• Verify a transition state (exactly one imaginary frequency)
• Confirm a minimum (no imaginary frequencies)

Prerequisites

The input structure MUST be optimized at the same level of theory used for the frequency calculation. Running frequencies on an unoptimized structure will produce meaningless imaginary frequencies.

MCP Tool Examples — proven Expanse submission flow

Use catgo_workflow (graph-based), NOT catgo_workflow_engine (task-based). The graph-based tool auto-captures the viewer structure on create and supports connecting opt→freq via explicit node edges. Task-based add_task doesn't attach the viewer structure → "No input structure provided". Param keys also differ: graph-based uses method/basis, task-based uses orca_method/orca_basis.

1. Confirm structure is loaded and find Expanse session_id

catgo_view(action: "get_state")

curl -s http://localhost:8000/api/hpc/connections

Copy the session_id for host: login.expanse.sdsc.edu.

2. Create the workflow (auto-captures viewer structure)

catgo_workflow(action: "create", name: "Water frequencies B3LYP")

This creates a structure_input node with the current viewer structure. Note its node ID.

3. Add the freq node (or opt → freq chain)

Standalone freq (when structure is already optimized at the same level):

Inject extra_blocks: "%output jsongbwfile True jsonpropfile True end" so ORCA emits the JSON files OPI parses.

catgo_workflow(action: "batch", workflow_id: "<wf_id>", operations: [
  {"op": "add_node", "node_type": "freq", "label": "freq1",
   "params": {
     "software": "orca",
     "method": "B3LYP",
     "basis": "def2-SVP",
     "charge": 0,
     "multiplicity": 1,
     "extra_blocks": "%output jsongbwfile True jsonpropfile True end"
   }},
  {"op": "connect", "from_id": "<structure_input_id>", "to_id": "freq1",
   "from_handle": "structure", "to_handle": "structure"}
])

Opt → Freq chain (recommended — consistent PES guaranteed):

catgo_workflow(action: "batch", workflow_id: "<wf_id>", operations: [
  {"op": "add_node", "node_type": "geo_opt", "label": "opt1",
   "params": {
     "software": "orca",
     "method": "B3LYP",
     "basis": "def2-TZVP",
     "opt_convergence": "TightOpt",
     "dispersion": "D3BJ",
     "charge": 0,
     "multiplicity": 1
   }},
  {"op": "add_node", "node_type": "freq", "label": "freq1",
   "params": {
     "software": "orca",
     "method": "B3LYP",
     "basis": "def2-TZVP",
     "dispersion": "D3BJ",
     "charge": 0,
     "multiplicity": 1
   }},
  {"op": "connect", "from_id": "<structure_input_id>", "to_id": "opt1",
   "from_handle": "structure", "to_handle": "structure"},
  {"op": "connect", "from_id": "opt1", "to_id": "freq1",
   "from_handle": "structure", "to_handle": "structure"}
])

The freq node consumes opt1's optimized structure — no separate depends_on needed; the edge defines the dependency.

4. Optional: append a Gibbs energy node

ORCA reports thermochemistry at 298.15 K / 1 atm by default. For other conditions, chain a gibbs_energy analysis node:

{"op": "add_node", "node_type": "gibbs_energy", "label": "gibbs1",
 "params": {"temperature": 373.15, "phase": "gas"}},
{"op": "connect", "from_id": "freq1", "to_id": "gibbs1",
 "from_handle": "frequencies", "to_handle": "frequencies"}

5. Run with the full HPC run_config

run_config MUST include module_loads, orca_dir, account, partition, walltime, and the local-scratch SLURM template. Read server/templates/orca_generic.sh and pass its contents as default_template.

catgo_workflow(action: "run", workflow_id: "<wf_id>", run_config: {
  "execution_mode": "hpc",
  "default_session_id": "<expanse_session_id>",
  "base_work_dir": "/expanse/lustre/projects/sdp126/jyang25/ORCA/catgo",
  "default_job_params": {
    "nodes": 1, "ntasks": 4, "cpus_per_task": 1,
    "walltime": "00:30:00", "partition": "debug"
  },
  "cluster_configs": {
    "<expanse_session_id>": {
      "account": "sdp126",
      "partition": "debug",
      "module_loads": "module load cpu/0.17.3b\nmodule load gcc/10.2.0/npcyll4\nexport PATH=$HOME/openmpi-4.1.8/bin:$PATH\nexport LD_LIBRARY_PATH=$HOME/openmpi-4.1.8/lib:$LD_LIBRARY_PATH",
      "orca_dir": "/home/jyang25/orca_6_1_1_RRP8",
      "default_template": "<contents of server/templates/orca_generic.sh>",
      "default_job_params": {
        "nodes": 1, "ntasks": 4, "cpus_per_task": 1,
        "walltime": "00:30:00", "partition": "debug"
      }
    }
  }
})

For freq jobs longer than 30 min, bump walltime and switch partition to shared or compute. The local-scratch template stages I/O to $TMPDIR/orca_$SLURM_JOB_ID — necessary on Expanse (Lustre kills ORCA's many-small-file I/O during numerical Hessians).

6. Monitor

catgo_workflow(action: "status", workflow_id: "<wf_id>")

7. Pull results when COMPLETED

Pull the outputs into a local directory, including the OPI JSON files:

mkdir -p ./local_run
for f in ORCA.out ORCA.hess ORCA.property.json ORCA.json; do
  curl -s -X POST http://localhost:8000/api/hpc/files/read-content \
    -H 'Content-Type: application/json' \
    -d "{\"session_id\":\"<expanse_session_id>\",\"file_path\":\"<work_dir>/$f\"}" \
    > ./local_run/$f
done

Parsed result fields (when fetched via catgo_workflow get_result or the results-enriched endpoint):

• frequencies: list of vibrational frequencies in cm⁻¹
• intensities: IR intensities in km/mol
• is_imaginary: boolean flags for each frequency
• zpe: zero-point energy in eV
• thermochemistry: dict with H, S, G at standard conditions

8. Parse with OPI

Replaces the hand-grep'd thermochemistry block. Requires pip install orca-pi.

import sys
sys.path.insert(0, ".claude/skills")  # for the _shared helper
from _shared.orca_opi import parse_local

out = parse_local("./local_run")

# IR table — replaces frequencies + intensities + is_imaginary trio
ir = out.get_ir()  # dict[int, IrMode]
for mode_idx, mode in ir.items():
    print(mode_idx, mode.wavenumber, mode.intensity, mode.dipole)

# Thermochemistry (units: hartree, hartree/K)
thermo = {
    "zpe_eh":           out.get_zpe(),
    "inner_energy_eh":  out.get_inner_energy(),
    "enthalpy_eh":      out.get_enthalpy(),
    "entropy_eh_per_K": out.get_entropy(),
    "free_energy_eh":   out.get_free_energy(),
    "G_minus_Eel_eh":   out.get_free_energy_delta(),
}

# Imaginary check from the raw frequency list (negatives = imaginary)
freqs = out.results_properties.geometries[0].thermochemistry_energies[0].freq
n_imag = sum(1 for f in freqs if f < 0)
print(f"Imaginary modes: {n_imag}")

Viewing the IR spectrum in the IDE

Use the shared helper to plot a stick spectrum and surface the PNG inline.

from _shared.orca_opi import quick_plot_ir, show_png
png = quick_plot_ir(out)             # writes ./local_run/ir_spectrum.png
show_png(png, "IR spectrum")         # prints `![IR spectrum](local_run/ir_spectrum.png)`

After running this, reply to the user with the markdown link the script printed so Claude Code renders the figure inline in chat.

Submission gotchas (real failures we hit)

• catgo_workflow_engine.add_task doesn't auto-attach the viewer structure → "No input structure provided".
• partition=workq (Shaheen default) is invalid on Expanse → use debug or shared.
• Missing account=sdp126 → "Invalid account or account/partition combination".
• Missing module_loads + orca_dir → orca not on PATH; numerical Hessians silently produce nothing.
• After re-connecting to Expanse, the session_id changes — re-discover via /api/hpc/connections and update default_session_id + cluster_configs key.
• Engine doesn't regenerate submit.sh on retry alone — call run with the new run_config to get a fresh script.

Interpreting Results

Minima verification

• All frequencies should be real (positive)
• Small negative frequencies (<50 cm-1) are numerical noise, usually harmless
• Large imaginary frequencies indicate the structure is NOT a minimum

Transition state verification

• Exactly ONE imaginary frequency (negative value)
• The imaginary mode should correspond to the expected reaction coordinate
• Use catgo_view to visualize the mode

Thermochemistry output

ORCA prints a thermochemistry block with:

Quantity	Symbol	Units
Zero-point energy	ZPE	eV (or kcal/mol)
Thermal energy	U	eV
Enthalpy	H = U + pV	eV
Entropy	S	eV/K
Gibbs free energy	G = H - TS	eV

For catalysis, feed the DFT energy and frequencies into gibbs_energy:

• phase: "adsorbed" -- harmonic approximation (no translational/rotational)
• phase: "gas" -- ideal gas (includes translation, rotation, vibration)

Frequency Scaling Factors

DFT frequencies are systematically overestimated. Common scaling factors:

Method	Scaling factor
B3LYP/def2-SVP	0.9813
B3LYP/def2-TZVP	0.9654
PBE/def2-SVP	0.9948
HF-3c	0.86

These are applied automatically by the gibbs_energy task when available.

Common Mistakes

• Running freq on unoptimized geometry (will show spurious imaginary modes)
• Using different method/basis for opt and freq (inconsistent PES)
• Ignoring imaginary frequencies and proceeding with thermochemistry
• Not using TightOpt for the preceding optimization (loose opt can leave residual forces that appear as small imaginary frequencies)

Canonical params (what the engine actually reads)

Parameter	Default	Description
method	B3LYP	DFT functional
basis	def2-SVP	Basis set
charge / multiplicity	0 / 1	Charge and 2S+1
dispersion	(none)	D4 | D3BJ | D3 | none. Use this field, NOT extra_keywords.
grid	DefGrid2	DefGrid1/2/3
wavefunction, uno, uco	—	Open-shell tweaks
num_cores / max_core_mb	4 / 4000	%pal nprocs / %maxcore

⚠️ extra_keywords and extra_blocks are NOT read by the engine. Forcing NumFreq, adding CPCM(Water), etc. via those keys silently does nothing. These are current node-def gaps for freq.

ORCA-Specific Notes

• ORCA uses analytical frequencies when available, numerical otherwise
• For large molecules (>100 atoms), frequencies become very expensive
• ORCA output lists frequencies as negative values for imaginary modes (not "i" notation)
• Forcing NumFreq is currently a gap — analytical Hessians are used by default for whatever functional supports them