⚛️ phenomenology

DM induced neutron disappearance as the origin of the LZ nuclear recoil event

This paper proposes that dark matter-induced neutron disappearance, where an incoming dark matter particle annihilates a bound neutron into an invisible scalar, offers a viable explanation for the high-energy nuclear recoil event observed by LUX-ZEPLIN without contradicting constraints from collider experiments or Borexino data.

Original authors: P. Uttayarat, J. Julio, R. Primulando

Published 2026-09-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: P. Uttayarat, J. Julio, R. Primulando

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

The universe is filled with a mysterious substance called dark matter, an invisible material that makes up most of the mass in the cosmos but refuses to interact with light or ordinary matter in any way we can easily detect. For decades, scientists have built massive, ultra-sensitive detectors deep underground, hoping to catch a rare glimpse of a dark matter particle bumping into an atom. The most recent and sophisticated of these experiments, located in a mine in South Dakota, uses a tank of liquid xenon to watch for these tiny collisions. When a dark matter particle hits an atom, it should knock the atom backward, creating a small flash of light and a tiny electrical signal. This recoil is the smoking gun that would finally reveal the nature of the invisible universe.

Recently, the LUX-ZEPLIN collaboration reported a single, startling event in their data: a xenon atom recoiled with an energy of about 248 keV, a value far higher than the gentle taps most theories predict. While this single hit is not enough to claim a discovery on its own, its high energy and isolation from the usual background noise have sparked intense curiosity. Standard theories of dark matter struggle to explain why a particle would hit an atom so hard without also producing a flood of weaker, lower-energy hits that the detector should have seen. To solve this puzzle, a team of physicists has proposed a radical new idea: the dark matter particle didn't just bounce off an atom; it destroyed a piece of the atom itself.

In this new scenario, the incoming dark matter particle does not scatter off a neutron, which is one of the building blocks inside the xenon nucleus. Instead, it annihilates the neutron, turning it into an invisible particle that vanishes from the detector. Imagine a billiard ball striking another ball, but instead of bouncing, the first ball causes the second to disappear entirely, leaving the remaining cluster of balls to recoil from the sudden loss of mass. When the dark matter particle consumes a neutron inside a xenon nucleus, the remaining nucleus is left with a specific amount of energy and momentum, causing it to recoil with a precise speed. Because the laws of physics dictate exactly how much energy is released when a neutron is removed, this process should produce a very specific, sharp signal rather than a broad smear of energies.

The researchers calculated what this signal would look like for different types of dark matter particles. They found that if the dark matter has a mass of 5 or 50 times that of a proton, the annihilation of a neutron could produce a recoil energy that matches the mysterious 248 keV event observed by LUX-ZEPLIN. Unlike other theories that predict a continuous stream of hits at various energies, this model predicts that the recoils would appear as distinct lines, like rungs on a ladder. In the real world, the motion of the dark matter particles as they zip through the galaxy would blur these sharp lines slightly, turning them into a narrow, localized peak. This explains why the detector saw a single high-energy event without seeing a corresponding flood of lower-energy events, a problem that has plagued other explanations.

To test if this idea holds water, the team looked at what else this process would do. If dark matter can eat neutrons in xenon, it should be able to do the same in other elements found in nature. Specifically, they checked data from the Borexino experiment, which uses a liquid scintillator containing carbon atoms. If dark matter were annihilating neutrons in carbon, it would leave behind a carbon nucleus in an excited state that immediately releases a photon of light with a very specific energy of 4.44 MeV. The Borexino team has been watching for exactly this kind of signal for years. The new analysis shows that if the dark matter parameters required to explain the LUX-ZEPLIN event were correct, Borexino should have seen a clear signal by now. The fact that Borexino has not seen this signal places a tight constraint on the theory, ruling out some of the most likely versions of the model, particularly for the heavier dark matter candidate.

Despite these constraints, the idea remains a compelling possibility for explaining the strange high-energy event. The researchers showed that there are still specific combinations of particle masses and interaction strengths that could fit the LUX-ZEPLIN data while avoiding the limits set by Borexino. This suggests that the mystery of the high-energy recoil is not yet solved, but the solution may lie in a process where dark matter destroys matter rather than just bouncing off it. If future experiments with even larger detectors confirm this pattern, or if they find similar high-energy hits in other materials, it would open a completely new window into the physics of the dark sector. For now, the single event at 248 keV stands as a quiet challenge, inviting scientists to look beyond simple collisions and consider the more dramatic possibility of matter vanishing into the dark.

Technical Summary: DM-Induced Neutron Disappearance as the Origin of the LZ Nuclear Recoil Event

Problem Statement
The LUX-ZEPLIN (LZ) collaboration recently reported a single, highly isolated nuclear recoil (NR) event at an anomalously high energy of approximately $248 \pm 23(\text{stat}) \pm 23(\text{sys})$ keV. This event exhibits a global statistical significance of roughly $2.6\sigma$ and stands out against expected low-energy background models. Standard elastic dark matter (DM) models struggle to explain this observation because they generically predict rising NR rates toward the detector threshold, which would imply a corresponding signal at lower energies that is absent. While inelastic DM and momentum-suppressed scattering models have been proposed to address the lack of low-energy events, they typically produce continuous spectra that extend into side-band regions where LZ reports zero events. This paper investigates an alternative mechanism: DM-induced neutron disappearance, which naturally produces a localized, line-like spectrum rather than a continuous one.

Methodology
The authors propose a scenario where an incoming DM particle ( $\chi$ ) annihilates a bound neutron ( $n$ ) within a target xenon nucleus, producing an invisible scalar particle ( $\phi$ ) via the process $\chi + n \to \phi$ . The residual nucleus recoils with energy and momentum fixed by the kinematics of this $2$-to-$1$ process and the specific removal energy of the daughter nuclear state.

Kinematics: The recoil energy ( $E_R$ ) is determined by the energy release ( $\Delta = m_\chi + m_n - m_\phi$ ) and the neutron removal energy ( $E_m$ ) of the specific nuclear transition. In the zero-velocity limit, this results in a discrete line spectrum. The finite velocity of the DM halo broadens these lines into symmetric distributions, with the width scaling linearly with the DM mass.
Rate Calculation: The event rate is calculated by folding the effective cross-section with the DM velocity distribution and the nuclear hole spectral function ( $S_h$ ). The spectral function accounts for the probability of removing a neutron from specific orbitals to reach various daughter states, incorporating spectroscopic factors and radial wavefunctions derived from a Woods-Saxon potential.
Model Realization: The effective operator describing the interaction is embedded in an ultraviolet (UV) complete model involving a color-triplet scalar diquark and a singlet Dirac fermion. This allows the authors to derive constraints on the coupling strength ( $y$ ) from collider searches (CMS dijet) and astrophysical observations (neutron stars).
Constraints: The authors recast data from the Borexino experiment to constrain accompanying nuclear de-excitation signals. Specifically, they analyze the potential for the same interaction to occur in $^{13}\text{C}$ , which would produce a prompt 4.44 MeV photon.

Key Contributions and Results
The paper identifies specific parameter regions for benchmark DM masses of 5 GeV and 50 GeV that can account for the LZ event near 248 keV without generating excess events at lower energies or in the high-energy sidebands.

Spectral Features: Unlike continuous spectra from standard scattering, this mechanism produces a set of discrete lines corresponding to accessible daughter states. For the 5 GeV benchmark ( $\Delta = 14$ MeV), multiple isotopes contribute, with peaks potentially appearing in sidebands. For the 50 GeV benchmark ( $\Delta = 7.8$ MeV), only isotopes with low removal energies (specifically $^{131}\text{Xe}$ and $^{129}\text{Xe}$ ) contribute, naturally suppressing events in both the low-energy signal region and the high-energy sideband.
Preferred Regions: Using an extended unbinned likelihood analysis, the authors map the preferred regions in the $(y, \Delta)$ plane. They find that the LZ excess can be explained within these regions, though the 50 GeV benchmark's best-fit point is partially excluded by Borexino constraints.
Constraints:
- Collider: Constraints from CMS dijet searches and neutron star observations limit the coupling $y$ , excluding portions of the parameter space but leaving the LZ preferred regions partially allowed.
- Borexino: For $\Delta \gtrsim 9.38$ MeV, the interaction can excite $^{12}\text{C}^*$ in the Borexino detector, leading to a 4.44 MeV photon signal. The analysis of 1494 live days of Borexino data places competitive bounds on $y$ , excluding the best-fit point for the 50 GeV benchmark.

Significance
The paper claims that DM-induced neutron disappearance provides a viable framework for interpreting localized nuclear recoil excesses in direct-detection experiments. Its primary significance lies in the distinct spectral signature: a set of discrete lines broadened by halo velocities, which can explain a single high-energy event without the continuous low-energy tail characteristic of standard elastic or inelastic scattering models. The authors note that this mechanism is indifferent to the mother nucleus, suggesting that complementary probes in other nuclei (such as $^{13}\text{C}$ in liquid scintillators or $^{2}\text{H}$ ) could provide further tests. The work concludes that if the LZ excess persists, this scenario offers a distinguishable alternative to inelastic dark matter explanations, particularly through the absence of events in specific side-band regions.

Technical Summary: DM-Induced Neutron Disappearance as the Origin of the LZ Nuclear Recoil Event

More like this