Kinetic energy constructed from exact gradient expansion of second order in uniform gas limit
Abhishek Bhattacharjee, Hemanadhan Myneni, Manoj K. Harbola, and Prasanjit Samal

TL;DR
This paper introduces a new semilocal kinetic energy density functional for orbital-free DFT that is accurate, parameter-free, and computationally efficient, suitable for large-scale simulations of diverse systems.
Contribution
A novel GGA-level KEDF (KGE2) that preserves the second-order gradient expansion and achieves high accuracy without system-specific parameters.
Findings
Performs well for metals, semiconductors, and clusters
Matches the accuracy of nonlocal and meta-GGA functionals
Parameter-free and computationally efficient
Abstract
Orbital-Free Density Functional Theory (OFDFT) has re-emerged as a viable alternative to Kohn-Sham DFT, driven by recent advances in kinetic energy density functionals (KEDFs). Nonlocal (NL) KEDFs have significantly extended OFDFT's applicability, particularly for bulk solids, but their high computational cost and dependence of system-specific parameters limit their universality. In this work, we propose a semilocal KEDF at the Generalized Gradient Approximation (GGA) level that achieves accuracy comparable to state-of-the-art NL and meta-GGA functionals, while remaining entirely parameter-free. Our construction revives the Thomas-Fermi-von Weizsacker (TFvW) framework by modulating the relative contributions of TF and vW terms through physically motivated constraints and preserving the exact second-order gradient expansion. Despite its simple form, the proposed functional (KGE2)…
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
TopicsAdvanced Chemical Physics Studies · Boron and Carbon Nanomaterials Research · Machine Learning in Materials Science
