Related Experiment Video
Updated: Jan 26, 2026
21:47
Localizing Protein in 3D Neural Stem Cell Culture: a Hybrid Visualization Methodology
Published on: December 19, 2010
13.2K
Local-Hybrid Functional With a Composite Local Mixing Function Built From a Neural Network and a Strong-Correlation
Artur Wodyński1, Martin Kaupp1
1Institute of Chemistry, Theoretical Chemistry/Quantum Chemistry, Sekr. C7, Technische Universität Berlin, Berlin, Germany.
Journal of Computational Chemistry
|January 24, 2026
Summary
Local hybrid functionals (LHs) with neural-network local mixing functions (n-LMFs) now balance self-interaction and static-correlation errors effectively. The new LH25nP functional achieves record accuracy for main-group energetics and improves spin-related chemical problems.
Area of Science:
- Computational Chemistry
- Quantum Chemistry
- Materials Science
Background:
- Local hybrid functionals (LHs) offer a flexible approach to balancing self-interaction and static-correlation errors in density functional approximations.
- Recent advancements incorporate strong-correlation factors into local mixing functions (LMFs) for improved performance.
- Machine learning has been employed to develop neural-network LMFs (n-LMFs), showing promise for main-group energetics.
Purpose of the Study:
- To develop a novel local hybrid functional incorporating a neural-network local mixing function optimized for strong-correlation effects.
- To evaluate the performance of the new functional, LH25nP, for main-group energetics and reaction energies.
- To assess the functional's ability to address spin-related issues in various chemical systems.
Main Methods:
- Development of the LH25nP functional featuring a neural-network local mixing function (n-LMF) optimized with a fixed strong-correlation factor.
- Rigorous testing using the GMTKN55 benchmark set for main-group energetics and the W4-11RE set for reaction energies.
- Evaluation of fractional-spin errors and performance in spin-restricted bond dissociation and spin-contamination problems.
Main Results:
- LH25nP-D4 achieved a record low self-consistent WTMAD-2 value of 2.47 kcal/mol on the GMTKN55 set, the best for a rung 4 functional.
- The functional demonstrated the lowest mean absolute deviations (2.4 kcal/mol) on the W4-11RE reaction energy set for rung 4 functionals.
- Significant improvements were observed in fractional-spin errors, spin-restricted dissociation of covalent bonds, and spin-contamination issues in transition-metal complexes.
Conclusions:
- The LH25nP functional represents a significant advancement in local hybrid functionals, effectively mitigating self-interaction and static-correlation errors.
- Its performance on benchmark datasets and for spin-related problems suggests a departure from the typical trade-offs in functional development.
- Further optimization including transition-metal data is recommended to enhance transferability to organometallic systems.
Keywords:
DFTlocal hybrid functionalsneural‐network local mixing functionstrong‐correlation factorzero‐sum gameMore Related Videos
Related Concept Videos
01:25Network Function of a Circuit
681
Frequency response analysis in electrical circuits provides vital insights into a circuit's behavior as the frequency of the input signal changes. The transfer function, a mathematical tool, is instrumental in understanding this behavior. It defines the relationship between phasor output and input and comes in four types: voltage gain, current gain, transfer impedance, and transfer admittance. The critical components of the transfer function are the poles and zeros.
681
01:19Composition of Polyprotic Acid Solutions as a Function of pH
841
Polyprotic acids of the type H2M constitute two ionizable protons. As a result, on titration with a base, they exhibit two equivalence points in the titration curve. During titration, the species H2M, HM−, and M2− will be present in the solution at different points. The fractions of H2M, HM−, and M2− present at the various instances of the titration are denoted by α0, α1, and α2, respectively.
A graph with the alpha values is plotted against the volume of...
A graph with the alpha values is plotted against the volume of...
841
01:22Local Attraction
370
Local attraction refers to disturbances in compass readings caused by magnetic influences from nearby objects such as metal fences, buried pipes, vehicles, buildings, power lines, or natural iron ore deposits. Small items like wristwatches, steel tools, or belt buckles can also interfere with the compass by creating local magnetic fields that distort the Earth's natural magnetic field. These distortions lead to inaccurate readings, posing navigation and land surveying challenges.Local...
370
02:45Functional Groups
88.0K
Functional groups are a group of atoms with characteristic properties, which when linked to the carbon skeleton of a molecule, alter the properties of that molecule. For example, the presence of certain functional groups on a molecule will make them hydrophilic, whereas others will make them hydrophobic. These functional groups are an indispensable part of organic chemistry and important components of biological molecules, such as carbohydrates, proteins, lipids, and nucleic acids. Each...
88.0K
01:13Local Anesthetics: Pharmacokinetics
1.2K
The potency and duration of action of local anesthetics (LAs) are determined by their pharmacokinetics. Pharmacokinetics describes how LAs are absorbed, distributed, metabolized, and eliminated from the body. When administered to the vascular tissues, LAs are quickly absorbed and enter the systemic circulation, reducing their localized effects. Adding vasoconstrictors such as epinephrine to LAs reduces their absorption into the systemic circulation, making them clinically effective. The...
1.2K
02:18Network Covalent Solids
16.1K
Network covalent solids contain a three-dimensional network of covalently bonded atoms as found in the crystal structures of nonmetals like diamond, graphite, silicon, and some covalent compounds, such as silicon dioxide (sand) and silicon carbide (carborundum, the abrasive on sandpaper). Many minerals have networks of covalent bonds.
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
16.1K