Randomized trial establishes ground-state energy in helium atom, suggesting implications for DFT methodologies.
This paper documents a self-contained density-functional theory (DFT) code, built from scratch, that computes the ground-state energy of the helium atom taking nothing as input but the nuclear charge. The program is constructed in stages, each adding one piece of physics: a hydrogen-like radial Schrödinger solver (Numerov integration with a shooting/bisection eigenvalue search), the Hartree potential from the radial Poisson equation, the local-density-approximation (LDA) exchange of Slater, and the Perdew–Zunger parametrisation of the Ceperley–Alder correlation. Each stage is validated against a known analytic or reference limit and locked in by an automated test. The converged results are E = −2.8616 Ha for the self-interaction-corrected Hartree level (which, for a two-electron singlet, coincides with Hartree–Fock), E = −2.7236 Ha after adding Slater exchange, and E = −2.8342 Ha for the full LDA, compared with the Hartree–Fock limit −2.8617 Ha and the exact value −2.9037 Ha. The discussion shows that self-interaction-corrected Hartree is identical to restricted Hartree–Fock for a two-electron singlet, that the LDA total energy is less accurate than Hartree–Fock for so small a system because of its self-interaction error, and that the LDA effective potential has the wrong long-range asymptotics. Source code, test suite, and figure scripts are released for replication.
No takes yet. Share an insight, caveat, or question.
Sajana Yasas Dehigaspitiyage Don (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: