Sub-GeV dark-matter–electron scattering to the solar-neutrino floor with skipper-CCD and qubit sensors
Limits sit orders of magnitude above the floor and an unexplained low-energy excess in silicon is not understood, but the route is engineering and background control, so difficulty 44 is well below dark-matter identity (73). Application value 14 equals the old dark-matter entry's, which was justified by exactly this sensor spin-off (skipper CCDs, qubit and TES sensors). Leverage 46 because the problem decides the thermal sub-GeV dark-matter benchmarks that a large share of the light-dark-matter programme targets; it is below dark-matter identity (70) because it covers one mass window.
The problem
Build detectors sensitive enough to see light dark matter bumping electrons, down to where solar neutrinos become the limit.
- 1Light dark matter can kick an electron
- 2Detect one electron: skipper CCDs
- 3Background mimics the same tiny signal
- 4Remove noise and unexplained excesses
- 5Reach solar-neutrino floor
The target. Dark matter with mass between about 1 MeV and 1 GeV is too light to produce a detectable nuclear recoil in standard experiments. But it can scatter off bound electrons, which are lighter and more easily excited, ionizing atoms and freeing a few electrons. The quantity to constrain is the scattering cross-section σ_e as a function of mass. Two cases are considered: a heavy mediator (the interaction acts like contact, form factor F_DM = 1) and a light mediator (the interaction is long-range, F_DM ∝ 1/q²).
Path to understanding
foundations → problemThe research entries this problem rests on, in the order to read them. Each links to its explainer.
- Start with
- Step 2
- Step 3
- ΛCDM, the standard model of cosmologyTheory · host
- Then
- Sub-GeV dark-matter–electron scattering to the solar-neutrino floor with skipper-CCD and qubit sensorsThe problem
State of the art
SENSEI set world-leading limits in 2020 with a single ~2 g skipper CCD and 24 days of data in the MINOS cavern (arXiv:2004.11378), and SENSEI at SNOLAB followed with 100.72 g-days after masking and blinding (arXiv:2312.13342; the 534.9 g-day figure in its first version is the pre-masking total), adding a daily-modulation search in 2025 (arXiv:2510.20889). DAMIC at SNOLAB, upgraded to skipper readout, confirmed an unexplained bulk excess of about 7 events per kg-day near 80 eVee in a 3.25 kg-day exposure (arXiv:2306.01717), so low-energy backgrounds in silicon are not understood. The DAMIC-M prototype at Modane reached about 1.3 kg-day with a 50 times lower single-electron rate and excluded the freeze-in and freeze-out hidden-sector benchmarks over parts of the MeV mass range (arXiv:2503.14617, 2025). Essig et al. (arXiv:1801.10159) showed that coherent elastic scattering of solar neutrinos off nuclei, which yields one- to few-electron ionisation signals, is an irreducible background for sub-GeV electron-recoil searches above roughly kilogram-year exposures. At the nuclear-recoil end, LZ reached the neutrino fog for 3–9 GeV masses, seeing ⁸B CEνNS at 4.5σ in 5.7 tonne-years (arXiv:2512.08065; PRL 137, 091806, 2026). No candidate signal is accepted. Multi-kilogram skipper-CCD (Oscura) and qubit- or TES-based (TESSERACT) programmes had published no dark-matter–electron results by October 2026, so current limits remain several orders of magnitude above the floor over most of the 1 MeV–1 GeV range.
Realistic applications
A signal would be the first non-gravitational evidence for dark matter and fix its mass scale, while a null result to the floor would exclude thermal-relic freeze-in and freeze-out benchmarks. The skipper-CCD and quantum-sensor technology feeds low-light imaging, X-ray spectroscopy and qubit-based sensing.
What counts as solved
Resolved when experiments report 90% CL limits (or a ≥5σ signal with confirmation in a second target) reaching the solar-neutrino floor (CEνNS-induced few-electron events) for 1 MeV–1 GeV masses, for both heavy and light mediators, with background models validated in sideband data.
How it gets verified
ObservationSettled by new measurements or observations.
Needs new detector data: experiments must report 90% CL limits reaching the solar-neutrino floor for masses from 1 MeV to 1 GeV (heavy and light mediators), or a signal at 5 sigma or more confirmed in a second target, with background models validated in sideband data.
Claims under verification
+ add a claimWhere the work is
Engineering and background characterisation dominate: bigger skipper arrays, lower dark counts, deeper sites. Theory is in-medium response and empirical share is the measurement itself.
Bottlenecks
ranked by votes- #1EngineeringDark counts and unexplained low-energy excess
The single- and few-electron spectrum is dominated by dark current, surface and instrumental effects; the SNOLAB excess shows these are not yet characterised.
Community grades
Sign in to add your own grade.
Connected problems
Progress
- Dec 8, 2025LZ light-DM searchsource
No excess at 3–9 GeV/c² in 5.7 tonne-years; ⁸B solar-neutrino CEνNS seen at 4.5σ.
- Mar 18, 2025DAMIC-M hidden-sector limitssource
About 1.3 kg-day with 50× lower single-electron rate excludes benchmark freeze-in/freeze-out models over part of the MeV range.
- Jun 2, 2023Bulk low-energy excess confirmed with skipper CCDssource
DAMIC at SNOLAB confirms an unexplained excess of about 7 events per kg-day near 80 eVee.