LUX-ZEPLIN (LZ) Experiment

Updated 28 November 2025

LUX-ZEPLIN (LZ) is a multi-tonne, dual-phase liquid xenon time projection chamber designed to detect WIMP dark matter and explore rare-event physics with high sensitivity.
Its innovative architecture integrates a central LXe TPC, an instrumented LXe skin, and an outer gadolinium-loaded scintillator tank, all enclosed in a water shield to rigorously mitigate background signals.
Advanced calibration systems and sophisticated statistical techniques enable precise energy reconstruction and event discrimination, setting world-leading limits on dark matter interaction cross sections.

The LUX-ZEPLIN (LZ) experiment is a multi-tonne, dual-phase liquid xenon time projection chamber (TPC) located at the Sanford Underground Research Facility (SURF) in Lead, South Dakota. Designed primarily as a direct search for weakly interacting massive particle (WIMP) dark matter, LZ also supports a broad rare-event physics program, encompassing searches for neutrinoless double beta decay, non-standard neutrino interactions, and new physics in low-energy electronic recoils. LZ achieves world-leading sensitivity through a combination of large target mass, ultra-low radioactivity construction, meticulous background control, and integrated calibration and veto systems (Collaboration et al., 2019, Akerib et al., 2018, Aalbers et al., 2 May 2024).

1. Detector Architecture and Subsystems

LZ is centered on a cylindrical two-phase xenon TPC containing 7 t of active liquid xenon (LXe) within a total 10 t xenon inventory. The TPC has an active region of 1.46 m diameter and 1.46 m height. Prompt scintillation (S1) and ionization electrons from particle interactions are detected by two arrays of PMTs (253 on top, 241 on bottom), with PTFE field cage panels for high UV reflectivity. The vertical electric field (∼200–310 V/cm) drifts electrons to the liquid-gas interface, where they are extracted and produce electroluminescence in the gas, generating the S2 signal (Collaboration et al., 2019, Akerib et al., 2018, Khaitan, 2015).

Surrounding the TPC:

The xenon "skin": 2 t LXe, instrumented with 131 PMTs for γ-ray and β tagging.
Outer Detector (OD): 17 t gadolinium-loaded liquid scintillator (GdLS), in 10 acrylic tanks, viewed by 120 inward-facing 8-inch PMTs. The OD achieves neutron tagging via Gd(n,γ) cascades and also vetoes γ-ray backgrounds (Turner et al., 2021).
The full assembly is enclosed by a 228 t ultrapure water tank, instrumented for muon veto and passive γ/neutron attenuation.

Cryogenics and purification systems enable continuous LXe recirculation (>3 t/day through a Zr getter), achieving electron lifetimes >1 ms and maintaining impurity concentrations compatible with low backgrounds (Collaboration et al., 2019, Akerib et al., 2020).

2. Calibration, Signal Reconstruction, and Energy Scale

LZ employs a comprehensive calibration suite (Aalbers et al., 2 May 2024):

Dispersed internal sources: ^83mKr, ^131mXe, ^220Rn for S1/S2 gain and spatial uniformity; tritiated methane and ^14C for low-energy β response; controlled injection/removal via a custom gas handling panel.
Calibration Source Deployment (CSD) rods: Deploy γ (e.g., ^57Co, ^54Mn, ^22Na, ^228Th) and neutron (^241AmLi, ^241AmBe) sources at multiple z and φ locations.
DD neutron generator and ^88Y/Be photoneutron source for in situ nuclear recoil response (1–200 keV_nr).
Dedicated LED-based optical calibration for PMT gain, timing, and afterpulsing, including both TPC and OD photodetectors (Turner et al., 2021).

Signals are reconstructed from digitized waveforms (100 MS/s, 14-bit precision), employing dual-gain amplifiers (G=0.5, G=20) to span single-photon to 10^5-phd signals with <0.5 mV RMS electronic noise (Khaitan, 2015). Event position is reconstructed in (x,y) from the S2 PMT pattern (σ_x,y<0.5 cm) and in z from S1–S2 drift time (σ_z<1 mm); spatial corrections for light collection are mapped via ^83mKr and ^131mXe calibrations (Aalbers et al., 2 May 2024).

The energy scale is established as: $E = W \left(\frac{\mathrm{S1}_c}{g_1} + \frac{\mathrm{S2}_c}{g_2}\right)$ where $W=13.7$ eV, $g_1\approx0.114$ phd/photon, $g_2\approx34$ phd/electron in recent operations. Energy resolution is typically 4–6% at 122 keV and 1–2% at 2.5 MeV (Aalbers et al., 2 May 2024).

3. Background Mitigation, Assay, and In Situ Controls

LZ’s sensitivity is predicated on maintaining background rates of $(6.3 \pm 0.5)\times10^{-5}$ events/(keV $_{ee}$ ·kg·day) in the WIMP ROI (Aalbers et al., 2022). This is achieved via:

Ultra-clean materials: extensive multi-technique radioassay campaign (HPGe γ-spectrometry, ICP-MS, NAA, GDMS) with batch certification for all PMTs, cryostat, structural, and electronic components (Akerib et al., 2020).
Rigorous cleanliness protocols: class 1000 cleanroom assembly, dust-removal (max depositions <500 ng/cm²), radon-reduced air (Rn <0.5 Bq/m³), and plate-out controls (PTFE <500 µBq/m² ^210Pb).
Integrated radon and dust monitoring: continuous radon-emanation and decay chain α-spectroscopy. In situ fits of α, β, and γ spectra and time-resolved tracking of cosmogenically activated isotopes (^127Xe, ^37Ar) provide real-time modeling.
Active vetoes: OD provides >95% neutron-veto efficiency at >200 keV and skin >90% γ-tagging (Aalbers et al., 22 Oct 2024, Turner et al., 2021).

Background sources include distributed β emitters (^214Pb, ^85Kr, ^212Pb), surface activity, solar/atmospheric neutrinos, accidental S1/S2 coincidences, and radiogenic/cosmogenic neutrons. All are constrained by auxiliary measurements and profiled as nuisance parameters in statistical analyses. Neutron backgrounds are further mitigated by GdLS and skin vetos (combined inefficiency <10^–2), with cosmogenic neutrons rendered negligible by 4300 m.w.e. overburden (Aalbers et al., 2022, Collaboration et al., 2019).

4. WIMP Dark Matter Search Program and Statistical Techniques

LZ's primary scientific mission is the search for spin-independent and spin-dependent WIMP-nucleus scattering. After 280+ live days, the combined exposure surpasses 4.2 t·yr (Aalbers et al., 22 Oct 2024, Aalbers et al., 2022):

Nuclear recoils (NR) are identified via S1/S2 pulse shape and charge-to-light ratio, using NEST-calibrated profiles, with 50% NR acceptance at ∼5.4 keV_nr and >99.5% ER discrimination.
Fiducial volumes are optimized for backgrounds; typical cuts yield 5.5 t with <1 mm wall definition error.
Unbinned, extended profile-likelihood ratio (PLR) analyses in the (S1_c, log_10 S2_c) plane incorporate floating nuisance parameters for background rates, tagging efficiencies, and response. Power-constraint techniques prevent accidental overcoverage (Collaboration et al., 2023, Aalbers et al., 22 Oct 2024, Aalbers et al., 2022).
The latest SI WIMP-nucleon upper limit is $\sigma_{SI} < 2.2\times10^{-48}$ cm $^2$ at $m_\chi = 40$ GeV/c $^2$ ; SD-neutron and SD-proton limits are $3.7\times10^{-43}$ cm $^2$ and $9.8\times10^{-42}$ cm $^2$ , respectively (Aalbers et al., 22 Oct 2024). No WIMP excess has been observed.

LZ has also set leading limits on non-relativistic effective field theory WIMP-nucleon operators, including extended energy-region analyses to 270 keV_nr for momentum-dependent interactions (Collaboration et al., 2023).

5. Nuclear Recoil and Electronic Recoil Calibration

Absolute nuclear recoil response is anchored by a suite of dedicated neutron calibrations. In particular, the ^{88Y–^9Be} photoneutron source ("YBe") provides quasi-monoenergetic 152 keV neutrons for in situ calibration down to 1.8 keV_nr, verifying S1 and S2 yield models and the Lindhard-type quenching factor at low energies (Aalbers et al., 18 Sep 2025). The overall NR detection efficiency reaches 50% at ≃5 keV_nr.

Electronic recoil (ER) backgrounds are calibrated via internal tritium and ^14C sources. The ER detection efficiency rises from 0 at ∼0.5 keV_ee to 90% at ∼3 keV_ee. Time- and position-dependent corrections are maintained via continuous internal source injections and automated LED PMT calibrations (Aalbers et al., 2 May 2024, Collaboration et al., 2023).

Calibration constants (e.g., $g_1$ , $g_2$ , electron lifetime $\tau_e$ , drift velocity $v_d$ ) are tracked continuously, with spatial uniformity maintained to <3%. Key systematics—energy scale, neutron tagging efficiency, and ER/NR discrimination—are controlled to <2–5% (Aalbers et al., 2 May 2024).

6. Rare Event and Non-WIMP Science Reach

Beyond the dark matter program, LZ exploits its large mass and ultra-low backgrounds for a diverse rare-event portfolio:

Double-beta decay searches: LZ is uniquely positioned to probe both two-neutrino and neutrinoless double beta decay of ^134Xe (741 kg in active LXe, natural abundance 10.44%), achieving $T_{1/2}^{2\nu} > 1.7\times10^{24}$ yr and $T_{1/2}^{0\nu} > 7.3\times10^{24}$ yr at 90% CL for 1000 days exposure. These limits are orders of magnitude beyond current measurements and probe the full range of nuclear model predictions (LUX-ZEPLIN et al., 2021). The sensitivity for ^136Xe 0νββ decay ( $T_{1/2}^{0\nu} > 1.06\times10^{26}$ yr) nearly matches dedicated, enriched Xe detectors (Akerib et al., 2019).
Low-mass dark matter (S2-only, Migdal, Bremsstrahlung): Careful studies and mitigation of few-electron backgrounds and grid emission (with a coincident-photon tag, $\geq90\%$ acceptance) permit threshold reduction to 1–2 electrons, opening the 0.1–1 GeV/c² DM mass regime (Akerib et al., 7 Oct 2025, Tvrznikova, 2019).
Searches for new electron recoil physics: LZ sets leading or competitive limits on the solar axion-electron coupling, neutrino magnetic moment and millicharge, galactic ALPs, hidden photons, mirror DM, and leptophilic DM. Projected sensitivities reach $g_{ae}<1.6\times10^{-12}$ (solar axions), $\mu_\nu<6.2\times10^{-12}\mu_B$ , and $\kappa^2<7.4\times10^{-28}$ for hidden photons, covering unexplored parameter space below 100 keV (Collaboration et al., 2023, Collaboration et al., 2021).
Neutrino physics: LZ is sensitive to coherent elastic neutrino–nucleus scattering (CEνNS) of solar ^8B ν, with calibrated rates O(10–100) events per multi-year exposure, providing prospects for precision ν–nucleus cross-section measurements at low energies (Aalbers et al., 18 Sep 2025).

7. Impact and Future Directions

LZ represents the state-of-the-art in xenon-based rare-event detection:

The combination of mass, ultra-low backgrounds, and sophisticated calibration, background control, and simulation (BACCARAT, NEST, DER) underpins a discovery capability approaching the "neutrino floor" for the entire parameter space $m_\chi\gtrsim5$ GeV/c^2, as well as sensitivity to NR and ER rare-event signals previously considered unfeasible in a single instrument (Collaboration et al., 2020, Akerib et al., 2018).
The LZ design—modular, actively vetoed, and extensively assayed—provides a template for subsequent multi-tonne LXe TPCs (e.g., LUX-ZEPLIN upgrades, DARWIN).
Ongoing and planned operations, including a full 1000-day exposure, will improve statistical power, especially for multi-isotope double-beta decay searches and sub-GeV DM, and will extend the reach of axion and neutrino property searches with further reductions in low-energy backgrounds.