Applications of a perturbation-aware local correlation method to coupled cluster linear response properties
Ruhee D'Cunha, T. Daniel Crawford

TL;DR
This paper evaluates the accuracy and efficiency of PNO++, combined PNO++, and PNO methods for coupled cluster response properties across various molecular systems, highlighting their relative strengths and limitations.
Contribution
The study extends and tests PNO++ and combined PNO++ methods on larger molecules, comparing their performance to PNO in response property calculations.
Findings
PNO++ and combined PNO++ reduce localization errors in linear response properties.
PNO performs better for correlation energies, but combined PNO++ offers similar accuracy.
All methods show comparable errors for complex 3D molecules, with slight variations depending on the system.
Abstract
We have investigated the efficacy of two recently proposed variations of the pair-natural-orbital approach to reducing the scaling of coupled cluster property calculations. In particular, we have extended our implementations of the PNO++ and combined PNO++ methods, which make use of field-aware pair-densities to define the virtual-orbital spaces used to describe electron correlation effects, in order to test their accuracy, efficiency, and robustness on larger molecular systems than previously investigated. For fluoroalkane chains up to 1-fluoroheptane we find that the PNO++ and combined PNO++ methods yield smaller localization errors in response properties than PNO for similarly compact virtual spaces, and, while the PNO method performs better than the PNO++ method for correlation energies, the combined PNO++ method recovers similar accuracy for correlation energies to the PNO method.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsEngineering Applied Research
