⚛️ phenomenology

An axion constraint from the diffuse supernova neutrino background indicated by Super-Kamiokande

Motivated by Super-Kamiokande's recent $2.6\sigma$ indication of the diffuse supernova neutrino background, this study uses hydrodynamic simulations to derive a competitive $1\sigma$ upper limit on the axion-proton coupling constant ( $|g_{ap}|<1.3\times10^{-9}$ ) that offers a robust constraint independent of single supernova event properties.

Original authors: Kanji Mori, Tomoya Takiwaki, Kazunori Kohri, Masamitsu Mori

Published 2026-08-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kanji Mori, Tomoya Takiwaki, Kazunori Kohri, Masamitsu Mori

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Deep within the heart of a dying star, a final, violent transformation occurs. When a massive star runs out of fuel, its core collapses under its own weight, crushing atoms together until they form a city-sized ball of neutrons, a protoneutron star. This object is so hot and dense that it glows with a brilliance far exceeding that of the sun, but not in light; it shines in neutrinos. These are ghostly particles that rarely interact with anything, allowing them to stream out of the star's core and into the universe, carrying away the star's immense heat. For decades, scientists have known that if new, invisible particles exist, they could be born in this furnace and escape even faster than neutrinos, acting like a cosmic drain that cools the star down too quickly. If such a drain existed, the burst of neutrinos we see would be shorter and weaker than expected.

The question of whether these invisible particles exist has long been a mystery, but a new study offers a fresh way to look for them. Instead of waiting for a single, nearby star to explode, researchers have turned their attention to the faint, steady hum of neutrinos coming from all the supernovae that have ever exploded in the history of the universe. This background glow, known as the diffuse supernova neutrino background, acts as a cosmic average of every stellar death. Recently, the Super-Kamiokande experiment in Japan, a massive tank of water buried deep underground, detected a hint of this background glow. Motivated by this potential discovery, a team of physicists has used this faint signal to place a new, independent limit on the existence of axions, a hypothetical particle that could solve some of the biggest puzzles in physics.

The researchers focused on axions, particles originally proposed to explain why the strong nuclear force behaves differently than expected in certain conditions. If axions exist, they would be produced in the hot cores of collapsing stars and would carry energy away, speeding up the cooling process. This would change the amount of neutrinos released over time. To test this, the team ran detailed computer simulations of a collapsing star, modeling how the core behaves when axions are present. They simulated a star with a mass roughly ten times that of our sun, tracking how the neutrino light changes over the first fifteen seconds after the core bounces back from its collapse. They found that when axions are included, the star cools faster, and the stream of neutrinos it emits drops off more quickly and with less total energy than in a standard star without axions.

The team then took these simulation results and calculated what the total signal from all the supernovae in the universe would look like if axions were present. They compared this theoretical prediction against the actual data recently reported by the Super-Kamiokande collaboration. The real-world data showed a specific amount of neutrino flux, or flow, arriving at Earth. The researchers found that if the axion particle interacted with protons too strongly, the predicted signal would be significantly lower than what the detectors actually observed. By finding the point where their model stopped matching the observation, they established a new boundary for how strongly axions can interact with ordinary matter.

The study concludes that the interaction between axions and protons must be extremely weak, with a coupling constant smaller than 1.3 times 10 to the power of negative 9. This limit is comparable in strength to the most famous constraint on axions, which was derived from the neutrino burst observed from Supernova 1987A, a single exploding star seen in our galaxy. However, the new method has a distinct advantage. The old limit relied entirely on the specific details of one star that exploded thirty years ago, meaning the result could be influenced by the unique properties of that particular star, such as how fast it was spinning or how its core was structured. The new limit, based on the diffuse background, averages out the signals from countless stars across cosmic history. This makes the result more robust, as it does not depend on the quirks of a single event.

While the findings are promising, the researchers are careful to note the uncertainties involved. Their simulations relied on specific models of how axions are produced inside the star, and other production methods or different types of stars could slightly alter the result. Furthermore, the study did not fully account for the complex ways neutrinos change their identity as they travel through space, a process that could slightly reduce the signal. Despite these factors, the team suggests that their limit is conservative, meaning the true limit could be even stricter. This approach opens a new door for testing not just axions, but any other invisible particle that might be stealing energy from dying stars. By listening to the collective whisper of the universe's dead stars, scientists can now probe the fundamental laws of physics with a clarity that a single explosion could never provide.

Technical Summary: Axion Constraints from the Diffuse Supernova Neutrino Background

Problem Statement
Core-collapse supernovae serve as extreme laboratories for probing physics beyond the Standard Model, particularly through anomalous energy loss from newly formed protoneutron stars (PNS). While the neutrino burst from SN 1987A provided the first stringent constraints on axion-nucleon couplings, this limit relies on a single event, making it potentially sensitive to the specific progenitor properties and explosion dynamics of that particular supernova. Recent reports by the Super-Kamiokande (SK) Collaboration indicate a detection of the Diffuse Supernova Neutrino Background (DSNB) with a statistical significance of $2.6\sigma$ . This paper investigates whether this cumulative signal, representing the integrated neutrino emission from the cosmic population of core-collapse events, can provide a robust, independent constraint on axion cooling that avoids the biases inherent in single-event analyses.

Methodology
The authors employ a multi-step approach combining long-term neutrino-radiation hydrodynamic simulations with cosmological integration:

Supernova Simulations: The study utilizes one-dimensional general-relativistic neutrino-radiation hydrodynamic simulations of a $9.6 M_\odot$ zero-metallicity progenitor star, developed by Mori and Mori. These simulations incorporate axion cooling effects lasting $\sim 15$ seconds post-bounce.
- Axion Model: The interaction is described by the Lagrangian $\mathcal{L} = \frac{g_{ap}}{2m_p} \bar{p}\gamma^\mu\gamma_5 p \partial_\mu a$ , focusing on nucleon bremsstrahlung ( $N N \to N N a$ ) via the one-pion-exchange (OPE) approximation, including nucleon spin fluctuation effects.
- Parameter Space: Seven models were calculated with axion masses $m_a \leq 11$ meV, corresponding to axion-proton coupling constants $|g_{ap}| \leq 8.5 \times 10^{-10}$ under the Kim-Shifman-Vainshtein-Zakharov (KSVZ) model assumption.
DSNB Flux Calculation: The DSNB spectrum $\Phi_{\bar{\nu}_e}(E)$ is reconstructed by integrating the neutrino emission over cosmic history.
- The neutrino spectrum for individual events is modeled using a pinched quasithermal distribution.
- The core-collapse event rate $\dot{\rho}(z)$ is estimated using an empirical formula up to a redshift $z_{max} = 4.5$ .
- A reduction factor $p_{\bar{\nu}_e}(g_{ap}, E) = \Phi_{\bar{\nu}_e}(g_{ap}, E) / \Phi_{\bar{\nu}_e}(g_{ap}=0, E)$ is calculated by comparing axion-inclusive simulations against the standard model ( $|g_{ap}|=0$ ). This factor is applied to a fiducial DSNB spectrum derived from the "KIHNN" binary population synthesis code.
Comparison with Observation: The calculated DSNB flux $F_{\bar{\nu}_e}$ is integrated over the energy range $13.3 \text{ MeV} \leq E \leq 60 \text{ MeV}$ (matching the SK analysis threshold and model availability). This theoretical flux is compared against the SK-reported flux of $3.6^{+1.6}_{-1.5} \text{ cm}^{-2}\text{s}^{-1}$ . For coupling constants exceeding the simulation range, the model prediction is extrapolated using a linear fit.

Key Results

Cooling Effect: Axion emission accelerates PNS cooling, causing the $\bar{\nu}_e$ luminosity to deviate from the standard case earlier (around $t_{pb} \sim 1$ s) and reducing the total neutrino emission timescale. Consequently, the DSNB flux is suppressed as $|g_{ap}|$ increases.
Constraint: By comparing the SK observation with the linearly extrapolated model, the authors derive a $1\sigma$ upper limit on the axion-proton coupling constant:
$|g_{ap}| < 1.3 \times 10^{-9}$
Derived Limits: In the context of the KSVZ model, this corresponds to an axion decay constant $f_a > 0.35 \times 10^9 \text{ GeV}$ and an axion mass limit $m_a < 16 \text{ meV}$ .
Comparison: This limit is comparable in magnitude to the conventional SN 1987A bound ( $|g_{ap}| < 6 \times 10^{-10}$ ) but is derived from a statistical ensemble rather than a single event.

Significance and Claims
The paper claims that the DSNB offers a distinct and robust probe for new physics. Unlike the SN 1987A constraint, which is sensitive to the temporal evolution and duration of a single burst, the DSNB constraint probes the time-integrated neutrino emission from a large population of core-collapse events. This approach naturally mitigates uncertainties related to the specific progenitor mass, rotation, or explosion mechanism of a single supernova.

The authors note that their derived limit is likely conservative due to several factors:

Progenitor Mass: The simulations used a low-mass ( $9.6 M_\odot$ ) progenitor with a small compactness parameter, which likely yields lower PNS temperatures and axion luminosities compared to heavier stars that contribute to the DSNB.
Neutrino Oscillations: The study assumed no neutrino oscillations ( $P=1$ ). Including oscillations could reduce the observed flux, potentially tightening the limit, though collective oscillation effects remain uncertain.
Theoretical Uncertainties: The treatment of axion production channels (e.g., beyond OPE, pion-induced reactions, strange matter) introduces theoretical uncertainties that future self-consistent simulations could refine.

Ultimately, the work demonstrates that the DSNB can serve as a competitive, single-event-independent constraint on feebly interacting particles, with the methodology applicable to axion-like particles, sterile neutrinos, and dark photons.

Technical Summary: Axion Constraints from the Diffuse Supernova Neutrino Background

More like this