Using DLPNO-CCSD(T) with canonical PCM in PSI4

1

Subject: Using DLPNO-CCSD(T) with canonical PCM in PSI4

Dear PSI4 developers and users,

I am an ORCA user interested in moving to a canonical PCM implementation in PSI4, while keeping a similar overall electronic‑structure workflow. I have not used PSI4 before and would appreciate guidance on what is currently possible and how to set it up.

Current ORCA workflow (CPCM, water) for a ~70‑atom system in water, I do:

  1. DFT geometry optimization:
    “B3LYP D3BJ aug-cc-pVDZ TightSCF OPT CPCM(Water)”

  2. DLPNO-CCSD(T) single point:
    “DLPNO-CCSD(T) aug-cc-pVTZ aug-cc-pVTZ/C RIJCOSX def2/J VeryTightSCF NormalPNO CPCM(Water)”

So both steps use ORCA’s conductor‑like CPCM(Water). However, I would like to use a standard PCM implementation (e.g., IEF‑PCM).

Questions for PSI4

  1. DFT+PCM geometry optimization
  • Can PSI4 perform a DFT (e.g., B3LYP‑D3(BJ)/aug‑cc‑pVDZ) geometry optimization with PCM in water, using your standard PCM (IEF‑PCM or recommended default)? Could you provide a minimal example showing how to:
    • enable PCM,
    • Choose the canonical PCM flavor?
  1. DLPNO-CCSD(T) single point + PCM
  • Is it possible to run a DLPNO-CCSD(T)/aug‑cc‑pVTZ single point with PCM(Water) in PSI4?

Any brief example inputs and recommended cavity/radii settings would be very helpful. Thank you for your time and for making PSI4 available.

Best regards,
AGhosh