A small number with a larger meaning
The paper’s central question is how four valence neutrons in helium-8 may arrange themselves. In the authors’ interpretation, they form an interaction-driven bound structure made of two neutron pairs — a finite-system analogue of a quartet — rather than behaving like two wholly independent pairs. The proposed picture combines spin-singlet s-wave pairing with additional 3P2 attraction, and is presented as consistent with the quartet component of multimodal superfluidity.
The key result is a three-point energy difference, XH, calculated along the helium-4, helium-6 and helium-8 chain. At ℏω = 15 MeV, this quantity moved toward approximately −1 MeV as the calculation was enlarged: it was −1.019 MeV at Nmax = 8 and −0.962 MeV at Nmax = 10, compared with an experimental value of −1.1496 MeV. The negative curvature is the numerical pattern the study uses to discuss nonadditive binding.
The paper also reports two-neutron separation energies of 0.975 MeV for helium-6 and 2.125 MeV for helium-8 as experimental reference values.
How the result is built
This is a modeling study of the ground states of helium-4, helium-6 and helium-8. Its main calculations use the Daejeon16 nuclear interaction in the no-core shell model, including the Coulomb interaction. NuHamil generated the nuclear interaction matrix elements, while BIGSTICK performed shell-model calculations. The analysis compared basis sizes and broke the energy difference into kinetic, potential, partial-wave and subshell pieces.
At Nmax = 10, the roughly 1 MeV effect was the leftover from a much larger cancellation. The intrinsic kinetic term contributed +10.817 MeV, while the total potential contributed −11.780 MeV, leaving XH at −0.962 MeV. Within the potential, neutron-neutron interactions contributed −5.795 MeV, proton-neutron interactions −5.187 MeV and proton-proton interactions −0.797 MeV.
Why the extra attraction matters
In a restricted 0p3/2 valence model, V0 = −2.758 MeV and V2 = −1.478 MeV gave a neutron-neutron potential curvature, Xnn,V, of −4.634 MeV. But the idealized helium-6 and helium-8 configurations in that model still lay above their breakup thresholds. The calculation therefore produced nonadditive curvature without establishing absolute binding in the stripped-down model.
The researchers next separated the full neutron-neutron potential difference into partial-wave channels, the study’s labels for distinct interaction modes. In the order 1S0, 1D2, 3P0, 3P1 and 3P2, the contributions were −2.884, −0.367, +0.022, +2.632 and −5.171 MeV. The 3P2 term was the largest attractive contribution, while 3P1 was repulsive. The five listed channels accounted for −5.769 MeV of the full −5.795 MeV, and the irreducible three-subshell 3P2 piece was −1.034 MeV.
A controlled comparison made the role of that sector more concrete. At a representative attraction threshold of 1 MeV, the restricted full interaction had an attractive rank of 6, including magnetic multiplicity. Removing 3P2 reduced V2 to −0.652 MeV, dropped the rank to 1 and changed Xnn,V from −4.634 to −0.500 MeV. The result is consistent with a picture in which nonadditive binding involves more than one sufficiently attractive pair mode, rather than repeated use of a single one.
Wave-function clues
The paper also examines selected connected four-neutron density cumulants, another diagnostic drawn from the calculated wave functions. At Nmax = 10, the largest selected helium-8 cross-subshell cumulant was 3.32 × 10−2, compared with −8.70 × 10−4 for the isolated 0p3/2 subshell. The largest listed positive helium-6 comparator was 3.91 × 10−3. The authors include these contrasting values as part of the broader evidence for correlations extending across more than one subshell.
An analytic proof supplies a separate constraint on the proposed structure. It concludes that one attractive fermion-pair mode cannot provide more than additive interaction binding when two identical pairs occupy it. In practical terms, the two-pair picture cannot gain extra binding simply by duplicating the same pair mode; the analysis instead points toward cooperation among distinct attractive modes, including 1S0 and 3P2.
An intentionally limited conclusion
The authors limit their conclusion to the Daejeon16 NCSM calculation. The harmonic-oscillator basis used there does not include explicit scattering asymptotics, so the result is not a substitute for a continuum formulation or an interaction-independent result. It is therefore a case for the proposed interpretation within the stated calculation, not an interaction-independent conclusion.
The restricted 0p3/2 calculation also matters to how the result should be read: it produced nonadditive curvature but left the idealized isotope configurations above breakup thresholds. Taken together, the energy balance, interaction decompositions, analytic proof and selected cumulants are what the authors interpret as a two-pair quartet in helium-8.
The work is an arXiv preprint, version 1, dated 26 August 2026. It reports associated data as available through a Google Drive directory.
Paper data and sources
Original title: Evidence for Quartet Binding of Valence Neutrons in $^8$He
Authors: Young-Ho Song, Yuan-Zhuo Ma, Dean Lee
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text