Coherent elastic neutrino–nucleus scattering
Below about 50 MeV a neutrino can scatter off an entire nucleus at once, with a probability that grows with the square of the neutron number. The only trace is a tiny nuclear recoil, which took 43 years to detect.
Beginner
A gentle push on a whole nucleus
Neutrinos usually interact with single particles: one neutron, one proton or one electron. At low energies, however, a neutrino can do something else. It can bounce off an entire atomic nucleus at once, as if the nucleus were a single solid ball. Physicists call this coherent elastic neutrino–nucleus scattering, shortened to CEvNS and pronounced 'sevens'. 'Elastic' means that nothing breaks apart: the nucleus simply recoils, a little like a heavy boat nudged by a passing wave.
Why the whole nucleus joins in
A neutrino with an energy of a few million electronvolts behaves like a wave that is larger than a nucleus. Such a wave cannot pick out individual neutrons. Instead, the tiny contributions from all the neutrons add up in step, like many people pushing a car together rather than one at a time. Because the contributions add before they are squared, the chance of a collision grows with the square of the number of neutrons. A nucleus with 80 neutrons is therefore about 6,400 times more likely to be hit than a single neutron. This makes CEvNS the most likely way for a low-energy neutrino to interact with matter.
The catch: almost no visible effect
If CEvNS is so common, why was it found only in 2017? The answer is that the recoil is extremely small. Imagine a ping-pong ball hitting a bowling ball: the bowling ball barely moves. A heavy nucleus struck by a neutrino moves so slightly that it deposits only a few thousand electronvolts, or even just a few hundred, into the detector. That is far less energy than the light of a single camera flash spreads over one pixel. Detecting such whispers requires extremely sensitive instruments, shielding against natural radioactivity and an intense source of neutrinos.
A prediction that waited 43 years
The physicist Daniel Freedman predicted the process in 1974, soon after a new kind of weak interaction, the neutral current, had been discovered1. He also warned that measuring it might be 'an act of hubris' because the signal is so faint. In 2017 the COHERENT collaboration finally saw it, using a crystal of caesium iodide weighing just 14.6 kilograms, about the size of a large bottle of water2. The neutrinos came from an accelerator in Oak Ridge, Tennessee.
Later milestones
- 2020: COHERENT saw the same process in liquid argon, a much lighter nucleus, confirming the predicted dependence on neutron number3.
- 2024: two large detectors built to hunt for dark matter, XENONnT and PandaX-4T, saw first hints of CEvNS from neutrinos made in the Sun56.
- 2025: the CONUS+ experiment at a Swiss nuclear power plant observed CEvNS from reactor antineutrinos, which produce even smaller recoils4.
Why it matters
Because CEvNS is so much more likely than other neutrino interactions, detectors can be small enough to carry, instead of filling caverns with thousands of tonnes of liquid. That makes it attractive for studying nuclear reactors and for testing the Standard Model with high precision. CEvNS also matters for the search for dark matter. Detectors that look for dark-matter particles bumping into nuclei will increasingly see neutrinos doing exactly the same thing. This growing background is called the neutrino fog: a natural limit that future experiments must learn to see through.
Quiz
Test your knowledge
4 questions on this level · Print as worksheet
Intermediate
The process
In coherent elastic neutrino–nucleus scattering, a neutrino exchanges a Z boson with a nucleus as a whole: ν + A → ν + A. The interaction is a neutral-current process, so it works for all three flavours equally and has no energy threshold. The only observable is the recoil energy of the nucleus.
Coherence requires the momentum transfer to be small enough that the neutrino cannot resolve the individual nucleons, roughly when the de Broglie wavelength exchanged exceeds the nuclear radius of a few femtometres. In practice this holds for neutrino energies below about 50 MeV. In that regime the amplitudes from all nucleons add in phase. The Z boson couples to neutrons with a strength proportional to −1/2 and to protons with a strength proportional to 1/2 − 2 sin²θW, which is nearly zero because the weak mixing angle has sin²θW ≈ 0.24. The total is therefore dominated by the neutrons.
Freedman's prediction
Daniel Freedman published the idea in 1974, a year after Gargamelle discovered weak neutral currents1. He realised that the cross-section would be the largest of any neutrino process at low energies and that the process mattered for the stellar collapse that leads to supernovae. He also foresaw the difficulty: a neutrino of 30 MeV striking a caesium nucleus transfers at most about 15 keV, and a reactor antineutrino of 3 MeV striking germanium at most a few hundred eV. Moreover, only part of the recoil energy becomes visible as light or charge, an effect called quenching.
Observation
- 2017, COHERENT, caesium iodide: a 14.6 kg CsI[Na] crystal at the Spallation Neutron Source in Oak Ridge recorded 134 ± 22 events at 6.7σ2. The source produces pions that stop and decay at rest, giving a pulsed beam of muon and electron neutrinos up to 52.8 MeV, ideal for separating signal from background.
- 2020, COHERENT, argon: the 24 kg liquid-argon detector CENNS-10 observed CEvNS at more than 3σ3. With 22 neutrons instead of 78 (caesium) or 74 (iodine), argon tests the N² scaling.
- 2024, dark-matter detectors: XENONnT and PandaX-4T reported first measurements of CEvNS from solar boron-8 neutrinos at 2.73σ and 2.64σ56.
- 2025, CONUS+: germanium detectors with very low thresholds, 20.7 m from the core of the Leibstadt reactor in Switzerland, observed 395 ± 106 events in 119 days of reactor operation, at 3.7σ4.
Why it took 43 years
Three things had to come together: intense, well-understood neutrino sources; detectors with thresholds of a few keV or less; and backgrounds from natural radioactivity and cosmic-ray neutrons reduced far enough that tiny recoils stand out. Spallation sources such as the SNS deliver neutrinos in short pulses, so most backgrounds can be rejected simply by looking only during the brief beam window. Reactors provide a far larger flux but continuously, which makes background control harder and explains why reactor CEvNS came eight years after the first observation.
Uses and consequences
CEvNS tests the Standard Model at low momentum transfer, for instance by measuring the weak mixing angle and the distribution of neutrons inside nuclei. It searches for non-standard interactions, a neutrino magnetic moment and sterile neutrinos through flavour-blind disappearance. Because the interaction rate per kilogram is high, compact detectors near reactors are considered for reactor monitoring. Large xenon and argon detectors would see CEvNS from all flavours of a galactic supernova. Finally, CEvNS from solar, atmospheric and diffuse supernova neutrinos produces recoils indistinguishable from those expected from many dark-matter candidates. This irreducible background, the neutrino fog, will limit future dark-matter searches7.
Quiz
Test your knowledge
4 questions on this level · Print as worksheet
Expert
Cross-section
At tree level in the Standard Model, the differential cross-section for a spin-zero-equivalent nucleus of mass M, with Z protons and N neutrons, is
Integrating gives σ ≈ (GF²/4π) QW² Eν² for F → 1, about 2 × 10⁻³⁸ cm² for 30 MeV neutrinos on caesium before form-factor suppression, more than two orders of magnitude above inverse beta decay on a free proton at the same energy. The form factor suppresses the rate once qR ≳ 1, with R ≈ 1.2 A^(1/3) fm; for Cs and I it reduces the SNS rate by tens of percent and makes CEvNS a probe of the neutron distribution (neutron skin) through FN(q²).
The visible energy is further reduced by quenching: in germanium only roughly 15–25% of the nuclear recoil energy appears as ionisation at sub-keV energies, and the quenching factor in this regime is itself a leading systematic uncertainty for reactor CEvNS.
Experimental status
- COHERENT at the SNS, Oak Ridge, uses π⁺ decay at rest: a prompt 29.8 MeV νμ followed by delayed νe and ν̄μ up to 52.8 MeV, synchronised with 60 Hz proton pulses of under 1 μs. First observation in 14.6 kg CsI[Na]: 134 ± 22 events, 6.7σ2. The full CsI data set gave 306 ± 20 events and 11.6σ, and a flux-averaged cross-section of (165 +30/−25) × 10⁻⁴⁰ cm², consistent with the Standard Model8. Liquid argon (CENNS-10, 24 kg): >3σ, 3.5σ in the primary analysis3.
- CONUS+ at Leibstadt (3.6 GW thermal), 20.7 m from the core: 395 ± 106 events against a predicted 347 ± 59, 3.7σ, the first observation with reactor antineutrinos and fully in the coherent regime4. An earlier claim from a germanium detector at the Dresden-II reactor was not confirmed.
- Solar ⁸B neutrinos on xenon: XENONnT with 3.51 t·yr exposure rejected the background-only hypothesis at 2.73σ5; PandaX-4T reported 2.64σ6. These are the first CEvNS measurements with astrophysical neutrinos and the first steps into the neutrino fog.
Physics reach
Fits to COHERENT data constrain sin²θW at q ≈ 50 MeV, the neutron radius of caesium and iodine, flavour-dependent non-standard interactions (breaking degeneracies in oscillation fits that affect mass ordering determination), and a neutrino magnetic moment or charge radius through spectral distortions at low T. Reactor CEvNS reaches the lowest q and is especially sensitive to magnetic moments and light mediators. Because CEvNS is flavour-blind, a measured deficit would indicate active-to-sterile oscillation directly. For a galactic supernova, tonne-scale xenon and argon detectors would record CEvNS from all six neutrino and antineutrino species, measuring the total emitted energy independently of flavour conversion in the star.
The neutrino fog
For WIMP-like dark matter, ⁸B CEvNS mimics a particle of about 6 GeV mass, while atmospheric and diffuse supernova neutrinos mimic heavier candidates. O'Hare's definition of the neutrino fog replaces the older 'floor' by the cross-section at which the discovery limit scales more steeply than 1/exposure because the signal must be separated statistically from a neutrino background with finite flux uncertainty7. Below that level, progress requires much larger exposures, better flux knowledge, or directional detection, since solar CEvNS points away from the Sun.
Compact detectors
Event rates per kilogram exceed those of inverse beta decay by one to two orders of magnitude, so kilogram-scale detectors suffice. The challenge shifts from target mass to threshold and backgrounds: low-noise germanium point-contact detectors, cryogenic bolometers with phonon readout, CCDs and scintillating crystals all aim at thresholds of order 10–100 eV. Reactor experiments must also suppress cosmogenic neutrons at shallow depth, as CONUS+ did with layered shielding and an active muon veto.
Quiz
Test your knowledge
4 questions on this level · Print as worksheet
Open questions
- How precisely can the quenching of sub-keV nuclear recoils in germanium and other materials be calibrated, and how much does it limit reactor CEvNS?
- Will CEvNS reveal physics beyond the Standard Model, such as new light mediators or a neutrino magnetic moment, at the low momentum transfers only it can reach?
- Can directional detectors or larger exposures let dark-matter searches see through the neutrino fog?
Terms to explore
- Coherent elastic Neutrino-nucleus scattering
- Coherent elastic Neutrino–Nucleus scattering (detection)
- CEvNS experiments
- COHERENT
- CONUS / CONUS+
- Solar ⁸B CEvNS in dark matter detectors
- Neutrino fog
- Nuclear recoil in CEvNS
- Neutral-current interaction
- Z boson
- Weak mixing angle
- Light quenching (Birks' law)
- Spallation Neutron Source (SNS)
- Decay-at-rest source
People
On the timeline
- 1973Gargamelle discovers weak neutral currents
- 1974Freedman predicts coherent neutrino-nucleus scattering
- 2017COHERENT observes coherent neutrino-nucleus scattering
- 2024Dark-matter detectors see solar neutrino CEvNS
- 2025CONUS+ observes CEvNS at a nuclear reactor
References
- D. Z. Freedman, 'Coherent effects of a weak neutral current', Phys. Rev. D 9, 1389 (1974), doi:10.1103/PhysRevD.9.1389
- COHERENT Collaboration, D. Akimov et al., 'Observation of coherent elastic neutrino-nucleus scattering', Science 357, 1123 (2017), doi:10.1126/science.aao0990
- COHERENT Collaboration, D. Akimov et al., 'First measurement of coherent elastic neutrino-nucleus scattering on argon', Phys. Rev. Lett. 126, 012002 (2021), doi:10.1103/PhysRevLett.126.012002
- N. Ackermann et al. (CONUS+), 'Direct observation of coherent elastic antineutrino–nucleus scattering', Nature 643, 1229 (2025), doi:10.1038/s41586-025-09322-2
- XENON Collaboration, E. Aprile et al., 'First indication of solar ⁸B neutrinos via coherent elastic neutrino-nucleus scattering with XENONnT', Phys. Rev. Lett. 133, 191002 (2024), doi:10.1103/PhysRevLett.133.191002
- PandaX Collaboration, Z. Bo et al., 'First indication of solar ⁸B neutrinos through coherent elastic neutrino-nucleus scattering in PandaX-4T', Phys. Rev. Lett. 133, 191001 (2024), arXiv:2407.10892
- C. A. J. O'Hare, 'New definition of the neutrino floor for direct dark matter searches', Phys. Rev. Lett. 127, 251802 (2021), doi:10.1103/PhysRevLett.127.251802
- COHERENT Collaboration, D. Akimov et al., 'Measurement of the coherent elastic neutrino-nucleus scattering cross section on CsI by COHERENT', Phys. Rev. Lett. 129, 081801 (2022), doi:10.1103/PhysRevLett.129.081801
Last updated: · Editorial standard
Cite this page
Text
Neutrino Atlas (2026). “Coherent elastic neutrino–nucleus scattering”. Retrieved 22 November 2026, from https://neutrino-atlas.com/learn/coherent-elastic-scattering/BibTeX
@misc{neutrinoatlas2026oherentelasticscattering,
title = {Coherent elastic neutrino–nucleus scattering},
author = {{Neutrino Atlas}},
year = {2026},
url = {https://neutrino-atlas.com/learn/coherent-elastic-scattering/},
note = {Accessed 2026-11-22},
language = {english}
}