Orca Uv Vis

ORCA UV-Vis Calculation Pipeline

This skill covers two stages: generating an ORCA input file for a TD-DFT calculation, and post-processing the output to plot a Gaussian-broadened UV-Vis absorption spectrum.

Scope: Input generation, local post-processing, and (optionally) HPC submission via the CatGo workflow engine. The "Submitting to HPC" section below covers the proven Expanse flow. If the user is running on their own non-CatGo infrastructure, just generate the input file from the template in Stage 1 and skip the submission section.

Target version: ORCA 6.x. Output block layout in ORCA 5 is close but not identical; the parser below is written against the ORCA 6 format used by CatGo's own parser (server/catgo/utils/orca_output.py::OrcaUvVisOutput).

Stage 1: Input Generation

Before generating the input file, ask the user for:

• XYZ geometry file path (assume already optimized; if the user hasn't optimized, route them to orca-opt first)
• Solvent for CPCM (e.g. hexane, water, ethanol, dichloromethane, acetonitrile)
• Charge and multiplicity (default to 0 1 if not specified)

Generate an ORCA input file with these fixed settings:

• Functional: CAM-B3LYP
• Basis set: DEF2-TZVP
• Full TD-DFT (not TDA)
• 30 roots
• CPCM solvation inline on the keyword line
• No RI / auxiliary basis
• No special SCF convergence tricks

Template:

!CAM-B3LYP DEF2-TZVP CPCM(HEXANE)
%TDDFT
   NROOTS   30
END
%output jsongbwfile True jsonpropfile True end
*XYZFILE 0 1 geometry.xyz

Substitute the user's solvent, charge, multiplicity, and XYZ path. If the user provides inline coordinates instead of a file, use * XYZ <charge> <mult> followed by the coordinates and close with *.

The %output ... end line makes ORCA emit the JSON files OPI's Output.parse() consumes during post-processing (Stage 2). Keep it.

Building the %tddft block with OPI

OPI's BlockTddft exposes every TD-DFT knob (nroots, iroot, irootmult, maxdim, maxiter, etol, rtol, tda, lrcpcm, cpcmeq, donto, saveunrnatorb, spinflip, soc, socgrad, triplets, ...) as a typed Pydantic field. Bad keys raise at construction.

The catgo backend builds the route line, %pal, %maxcore, charge/multiplicity, and geometry from node params — it does not emit a %TDDFT block of its own. So OPI only contributes the %tddft and %output blocks here, which we paste into extra_blocks as text. Do not use Calculator.write_input() — that writes a full input file and would duplicate the route line / pal / geometry the backend already emits.

from opi.input.blocks import BlockTddft, BlockOutput

tddft_block = BlockTddft(nroots=30, tda=False, triplets=False, donto=True)
output_block = BlockOutput(jsongbwfile=True, jsonpropfile=True)

extra_blocks_text = tddft_block.format_orca() + "\n" + output_block.format_orca()
# Pass extra_blocks_text into the node's `extra_blocks` param.

format_orca() emits exactly one %...end block per call. The result for the snippet above is:

%tddft
    nroots 30
    tda False
    donto True
    triplets False
end
%output
    jsonpropfile True
    jsongbwfile True
end

TD-DFT with 30 roots at def2-TZVP is expensive — tell the user this is a heavy calculation and that they should expect to run it on a cluster or at least overnight on a workstation, not on a laptop.

Submitting to HPC (Expanse) — proven flow

Use this when the user wants the CatGo workflow engine to run the TD-DFT job on Expanse. Skip if they only want the input file.

Use catgo_workflow (graph-based), NOT catgo_workflow_engine (task-based). The graph-based tool auto-captures the viewer structure on create. Task-based add_task doesn't, so jobs fail with "No input structure provided". Param keys differ: graph-based uses method/basis, task-based uses orca_method/orca_basis.

1. Confirm structure is loaded and find 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

catgo_workflow(action: "create", name: "UV-Vis CAM-B3LYP TD-DFT")

3. Add the TD-DFT node

Use the orca_uvvis node type, which has dedicated TD-DFT params (nroots, triplets, tda, donto, solvation, solvent, calc_type, aux_basis) plus the standard dispersion field. Do NOT use extra_keywords or extra_blocks — they are NOT read by the engine and are silently dropped.

catgo_workflow(action: "batch", workflow_id: "<wf_id>", operations: [
  {"op": "add_node", "node_type": "orca_uvvis", "label": "tddft",
   "params": {
     "software": "orca",
     "method": "CAM-B3LYP",
     "basis": "def2-TZVP",
     "calc_type": "tddft",
     "nroots": 30,
     "tda": false,
     "triplets": false,
     "donto": true,
     "solvation": "CPCM",
     "solvent": "hexane",
     "dispersion": "D4",
     "charge": 0,
     "multiplicity": 1,
     "num_cores": 16,
     "max_core_mb": 4000
   }},
  {"op": "connect", "from_id": "<structure_input_id>", "to_id": "tddft",
   "from_handle": "structure", "to_handle": "structure"}
])

Canonical UV-Vis node params (verified against server/workflow/engines/orca.py:244-268)