“Gold-Standard” Δ-Machine Learned Transferable Potential for Linear Alkanes

The conformational properties of linear alkanes, C<sub><i>n</i></sub>H<sub>2<i>n</i>+2</sub>, have been of intense interest for many years. Experiments and corresponding electronic structure calculations were first reported in the mid-2000s and continue to the present time. These focus on the minimum chain length where the transition from the linear minimum to the hairpin minimum occurs. We recently reported a transferable, many-body permutationally invariant polynomial (MB-PIP) potential for linear alkanes trained on roughly 253 000 B3LYP electronic energies for C<sub>14</sub>H<sub>30</sub>. Here we report a Δ-ML approach to elevate this B3LYP-based and a new PBE0+MBD MB-PIP potential using roughly 4500 Pair Natural Orbital Local Coupled Cluster (PNO-LCCSD(T)-F12) energies. The new Δ-corrected potentials predict the difference in these minima accurately, compared to benchmark PNO-LCCSD(T)-F12 energies results, over the range C<sub>12</sub>H<sub>28</sub> to C<sub>28</sub>H<sub>58</sub>. Vibrational power spectra are also reported for C<sub>14</sub>H<sub>30</sub> using the original B3LYP-based and Δ-ML corrected potentials. These new MB-PIP potentials for linear alkanes are the most accurate ones currently available and can be used in studies of properties of linear alkanes.

Authors

Institutions

Publication Details

Journal
The Journal of Physical Chemistry Letters
Published
2025-11-20
DOI
https://doi.org/10.1021/acs.jpclett.5c02946
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

“Gold-Standard” Δ-Machine Learned Transferable Potential for Linear Alkanes

Qi Yu, Apurba Nandi, Paul L. Houston, Riccardo Conte et al.
The Journal of Physical Chemistry Letters
Machine Learning in Materials Science
article

“Gold-Standard” Δ-Machine Learned Transferable Potential for Linear Alkanes

Qi Yu, Apurba Nandi, Paul L. Houston, Riccardo Conte, Joel M. Bowman, Chen Qu
article en

Abstract

The conformational properties of linear alkanes, C<sub><i>n</i></sub>H<sub>2<i>n</i>+2</sub>, have been of intense interest for many years. Experiments and corresponding electronic structure calculations were first reported in the mid-2000s and continue to the present time. These focus on the minimum chain length where the transition from the linear minimum to the hairpin minimum occurs. We recently reported a transferable, many-body permutationally invariant polynomial (MB-PIP) potential for linear alkanes trained on roughly 253 000 B3LYP electronic energies for C<sub>14</sub>H<sub>30</sub>. Here we report a Δ-ML approach to elevate this B3LYP-based and a new PBE0+MBD MB-PIP potential using roughly 4500 Pair Natural Orbital Local Coupled Cluster (PNO-LCCSD(T)-F12) energies. The new Δ-corrected potentials predict the difference in these minima accurately, compared to benchmark PNO-LCCSD(T)-F12 energies results, over the range C<sub>12</sub>H<sub>28</sub> to C<sub>28</sub>H<sub>58</sub>. Vibrational power spectra are also reported for C<sub>14</sub>H<sub>30</sub> using the original B3LYP-based and Δ-ML corrected potentials. These new MB-PIP potentials for linear alkanes are the most accurate ones currently available and can be used in studies of properties of linear alkanes.

The Journal of Physical Chemistry LettersVol. 16(48)
Emory University (US), University of Luxembourg (LU), University of Milan (IT), Cornell University (US), Fudan University (CN), Emerson (United States) (US), Atlanta University Center (US)
Openalex Percentile: Top 71%
Machine Learning in Materials Science
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.