JUNO’s First Results Kill a Candidate Neutrino Mass Pattern and Sharpen DUNE’s Test

Neutrinos are the only known elementary particles whose masses cannot be explained by the Standard Model of particle physics, and the pattern of those masses remains one of the deepest puzzles in fundamental physics. Now, the first oscillation results from the Jiangmen Underground Neutrino Observatory (JUNO) in China have delivered a precision measurement that is already reshaping the landscape of theoretical neutrino models. A new theoretical study published in The European Physical Journal C by Debasish Borah of the Indian Institute of Technology Guwahati, Pritam Das of Salbari College, and Debajyoti Dutta of Bhattadev University shows that JUNO’s improved numbers do more than refine known parameters: they eliminate one of the few remaining candidate structures for the neutrino mass matrix and dramatically sharpen the ability of the upcoming Deep Underground Neutrino Experiment (DUNE) to test the survivors.

The theoretical framework at stake is known as texture-zeros. If neutrinos are Majorana particles, meaning they are their own antiparticles, their three-by-three mass matrix is complex and symmetric, containing six independent complex parameters. A texture-zero hypothesis posits that one or more of the six elements of this matrix vanish exactly. Such zeros are not arbitrary; they typically arise from underlying flavor symmetries that constrain how leptons transform. The appeal is predictivity: imposing a zero forces correlations among the neutrino masses, mixing angles, and CP-violating phases, which can then be confronted with experimental data. In the diagonal charged lepton basis, where the charged lepton mass matrix is already diagonalized, a symmetric Majorana mass matrix with three or more texture-zeros is incompatible with observed neutrino oscillations, leaving only one-zero and two-zero candidates as viable structures.

JUNO’s first results, released in late 2025, measured the solar mixing angle theta-12 as sin-squared theta-12 equal to 0.3092 with an uncertainty of 0.0087, and the solar mass-squared difference Delta-m-squared-21 as 7.50 times ten to the minus five electronvolts squared with an uncertainty of 0.12 in the same units, both at one-sigma confidence. This represents a factor of 1.6 improvement in precision over the combination of all previous measurements. The experiment achieves this by detecting reactor antineutrinos from nuclear power plants at a baseline of roughly 53 kilometers, where the oscillation pattern is governed almost entirely by the solar parameters, free from the degeneracies that plague accelerator-based measurements.

Borah and collaborators revisited all previously allowed one-zero and two-zero textures using both the NuFit 6.0 global fit of neutrino oscillation data from 2024 and the new JUNO numbers. They assumed neutrinos to be Majorana particles and the charged lepton mass matrix to be diagonal, and they imposed additional constraints from neutrinoless double beta decay searches and from the cosmological upper bound on the sum of absolute neutrino masses. The result is striking: while the validity of all one-zero textures is essentially unaffected by JUNO’s data, one of the previously allowed two-zero textures, labeled B1 in the standard classification, is ruled out in the normal mass hierarchy scenario. Before JUNO, the larger uncertainties in theta-12 and Delta-m-squared-21 allowed the B1 texture to fit within the broader allowed parameter space. JUNO’s precision shrinks that space so drastically that no parameter points remain consistent with all global oscillation data.

The physical mechanism behind this exclusion is elegant. When a Majorana mass matrix element is constrained to be zero, the condition can be written as a homogeneous linear equation involving the PMNS mixing matrix elements and the complex neutrino masses, with Majorana phases absorbed into the mass eigenvalues. For the B1 texture, two such equations can be combined to eliminate the heavy atmospheric mass scale and yield an analytical expression for the ratio of the two lighter neutrino masses, independent of Majorana phases. JUNO’s best-fit solar mass splitting dictates that for any reasonable absolute mass scale below roughly 0.03 electronvolts, this physical ratio must be extremely close to unity, approximately 1.0007. However, when the allowed mixing angle and CP phase ranges from the analysis are used, the texture-predicted ratio falls to approximately 0.87, a clear deviation from JUNO’s prediction. Under the older NuFit 6.0 uncertainties, the same expression could shift dynamically and intercept the physical ratio curve at a real positive mass value around 0.01 electronvolts. JUNO’s halving of the theta-12 uncertainty eliminates that flexibility entirely.

Interestingly, the B1 texture remains allowed in the inverted mass hierarchy scenario, where the physical mass ratio is governed by different relations among the mass eigenvalues. Because the two lighter masses are nearly degenerate in that scenario, the required ratio of at least 1.0523 can be satisfied for maximal atmospheric mixing. This hierarchy-dependent survival is a hallmark of texture-zero physics: the same algebraic structure can live or die depending on which mass ordering nature has chosen, making texture-zeros powerful discriminators of the neutrino mass hierarchy.

The second major contribution of the study is its analysis of how the surviving textures can be probed at DUNE, the next-generation long-baseline neutrino facility under construction in the United States. DUNE will employ a high-intensity broadband neutrino beam from the Long-Baseline Neutrino Facility at Fermilab, initially operating at 1.2 megawatts, with the beam peaking around 2.5 gigaelectronvolts to cover the first oscillation maximum. Its 1300-kilometer baseline enhances sensitivity through matter effects, and its massive far detector located deep underground will provide unprecedented precision on oscillation parameters. The researchers used the GLoBES simulation software with collaboration-provided detector specifications, fluxes, cross sections, and systematic uncertainties to generate simulated event spectra for a 13-year DUNE run.

The analysis reveals that DUNE, especially when combined with the proposed T2HK experiment in Japan, excludes a significantly larger fraction of the texture-predicted parameter space than current global-fit constraints. The researchers present their results in the plane of sin-squared theta-23 versus the Dirac CP phase delta-CP, the two parameters that dominate long-baseline sensitivity. For one-zero textures in the normal hierarchy, the textures labeled G4 and G6 show the most restricted allowed regions, indicating that DUNE plus T2HK excludes a larger fraction of their parameter space. In the inverted hierarchy, the texture G3 stands out as the most restrictive and testable scenario, with two small isolated allowed regions within the globally allowed space. Remarkably, when JUNO and reactor constraints on sin-squared theta-12 and sin-squared theta-13 are included as priors, the allowed parameter space for G3 is eliminated completely.

For two-zero textures, the picture is qualitatively similar but with even tighter restrictions. In the normal hierarchy, textures B2, B3, and B4 exhibit more localized allowed regions than A1, which permits comparatively broader parameter space. The inclusion of the JUNO prior on sin-squared theta-12 leads to moderate tightening of the contours, while an additional reactor constraint on sin-squared theta-13 produces further reduction. In the inverted hierarchy, the JUNO prior results in a noticeable reduction of allowed regions, particularly near their boundaries, with the B4 texture showing the most substantial shrinkage when both JUNO and reactor information are taken into account.

The broader significance of this work extends beyond the specific textures studied. Every texture-zero hypothesis that survives experimental scrutiny constrains the possible flavor symmetries that could underlie it in an ultraviolet-complete theory. The newly ruled-out B1 texture, for example, could have been explained by an Abelian discrete symmetry such as Z4 combined with a type-II seesaw mechanism involving three different SU(2)L triplet fields. Ruling out the texture with JUNO data therefore rules out all such flavor-symmetric interpretations simultaneously. As future JUNO data continue to accumulate and as DUNE and T2HK come online, the combined experimental program will further narrow the list of viable texture-zero scenarios, bringing particle physicists closer to understanding the origin of neutrino masses and the role of flavor symmetries in the fundamental laws of nature.

Subject of Research: Particle Physics

Article Title: Neutrino texture-zeros after JUNO’s first results: implications for long-baseline neutrino experiments

Article References: Borah, D., Das, P., & Dutta, D. (2026). Neutrino texture-zeros after JUNO’s first results: implications for long-baseline neutrino experiments. The European Physical Journal C, 86(10), Article 1175. https://doi.org/10.1140/epjc/s10052-026-16481-7

Image Credits: AI Generated

DOI: 10.1140/epjc/s10052-026-16481-7

Keywords: neutrino physics, JUNO, texture-zeros, DUNE, T2HK, Majorana neutrinos, long-baseline experiments, flavor symmetries, Neutrino, first, results, implications