Liquid Xenon TPC: Principles & Advances

• Liquid xenon TPCs are position-sensitive detectors that use a liquid xenon medium to generate prompt scintillation (S1) and delayed ionization (S2) signals with sub-keV thresholds.
• They enable rare event searches including dark matter and neutrinoless double beta decay by combining high spatial resolution and excellent background rejection.
• Technical advancements in electrode design, cryogenic purification, and signal processing have allowed scalability to ton-scale detectors with near-Fano-limited energy resolution.

A liquid xenon time projection chamber (LXe TPC) is a position-sensitive, ionization-based detector that employs liquid xenon as its active medium to enable simultaneous measurement of both primary scintillation (S1) and charge (S2) signals from rare particle interactions. The dual-phase LXe TPC is the dominant architecture in modern searches for dark matter, neutrinoless double beta decay, and applications in low-background nuclear and medical physics. Key technical advances have established LXe TPCs as the leading technology for rare-event detection at the ton scale and beyond, capable of achieving high spatial granularity, sub-keV energy threshold, excellent background rejection, and robust scalability.

1. Principle of Operation and Signal Formation

An LXe TPC consists of a monolithic volume of ultra-pure liquid xenon, bounded by planar or cylindrical electrodes and instrumented with photosensors (typically PMTs or silicon photomultipliers). When a particle interacts in LXe, it creates excited xenon states (Xe*), leading to prompt VUV scintillation (S1), and ionized Xe atoms that release free electrons. The S1 signal is detected immediately by the photosensors. The free electrons are drifted upward by a uniform electric field (typically 0.2–1.3 kV/cm), with a velocity $v_d(E)$ that is a function of the drift field and temperature, for example $v_d = 1.53 \mathrm{\ mm/\mu s}$ at $E = 0.22\ \mathrm{kV/cm}$ and $1.88\ \mathrm{mm/\mu s}$ at $1.26\ \mathrm{kV/cm}$ in Xurich II (Baudis et al., 2017).

At the liquid-gas interface (dual-phase operation), a strong extraction field ($E_\mathrm{extr}$, typically >5 kV/cm) extracts electrons into the gas xenon (GXe) region. There, electrons accelerate and produce proportional scintillation (S2) via electroluminescence (typically 24–35 PE/electron in Xurich II, 34 PE/electron in RELICS (Xie et al., 24 Nov 2025), up to 60 PE/electron in large-scale systems).

The drift time, $\Delta t$ between S1 and S2, yields the $z$-coordinate; the S2 light pattern on the top sensor array is used for $(x,y)$ reconstruction with mm-scale precision (Stephenson et al., 2015). The combined energy scale,

$$E_{CES}=W\left(\frac{S1}{g_1}+\frac{S2}{g_2}\right)$$

(with $W=13.7$ eV) provides optimal energy resolution, exploiting the anti-correlation between light and charge due to recombination effects (Baudis et al., 2017, Stephenson et al., 2015, Schumann, 2014).

2. Detector Architecture and Subsystems

A standard dual-phase LXe TPC comprises:

• Active volume: Cylindrical or rectangular, typically defined by PTFE reflectors for high VUV reflectivity and copper or stainless-steel field-shaping rings for field uniformity. Volumes from $\sim$0.05 kg (prototype) to $>10$ tonnes (XENONnT, DARWIN) are realized (Adrover et al., 2024).
• Electrodes: Meshes or wires define cathode, gate, anode (plus screening grids in large TPCs). High-transparency stainless-steel meshes (100 μm wires, 2–5 mm pitch) are used, with optical transparency $\geq$90% (Linehan et al., 2021). Gate-anode gap and precise mechanical tolerances are critical for field uniformity, especially in electroluminescence regions (Edwards et al., 2017, Linehan et al., 2021).
• Photosensors: Arrays of VUV-sensitive PMTs (e.g., Hamamatsu R9869, R8520, R11410) or SiPMs arranged above the gas, below the liquid, or on barrel walls (Adrover et al., 2024, Stiegler et al., 2020).
• Cryogenics and purification: Precise temperature control (stability $\leq0.02$ K (Baur et al., 2022)), continuous xenon recirculation through high-temperature getters for sub-ppb O₂/H₂O levels, yielding electron lifetimes $\tau_e>200$ μs (MiX, Xurich II, RELICS, XeBRA) and up to ms-scale in tonne-scale TPCs (Stephenson et al., 2015, Baudis et al., 2017, Xie et al., 24 Nov 2025, Baur et al., 2022).

Auxiliary systems include slow-control (automated alarms, interlocks), high-voltage distribution and safety, and integrated calibration sources (e.g., $^{83\mathrm{m}}$Kr, $^{37}$Ar) for spatial and energy calibration (Xie et al., 24 Nov 2025, Baur et al., 2022).

3. Signal Processing, Calibration, and Energy Resolution

Pulse identification utilizes width-based filters and template-matching algorithms:

• $A_1(i)=\sum_{j=i-w_1/2}^{i+w_1/2}S_j$ ($w_1\approx80$ ns for S1)
• $A_2(i)=\sum_{j=i-w_2/2}^{i+w_2/2}S_j-\max_k A_1(k)$ ($w_2\approx1.1\,\mu\mathrm{s}$ for S2)

Waveform baseline subtraction, $\chi^2$-template filtering against S1 shapes, and customized peak-finding loops are deployed (Baudis et al., 2017). For 3D event reconstruction, neural-network-based S2 pattern algorithms reach sub-mm $(x,y)$ precision (Stephenson et al., 2015, Baur et al., 2022).

Calibration with internal $^{83\mathrm{m}}$Kr (continuous, spatially uniform 32.1 keV + 9.4 keV transitions) enables precise mapping of S1/S2 yield vs. position and drift time (non-uniformities typically $<2\%$ post-correction) (Baudis et al., 2017). Gains $g_1$ (PE/photon) and $g_2$ (PE/electron) are determined by charge-light anti-correlation (Doke plots); for example, $g_1=0.191\pm0.006$ PE/photon, $g_2=24.4\pm0.4$ PE/electron (Xurich II at $E_\mathrm{extr}\approx10.3$ kV/cm) (Baudis et al., 2017).

Energy reconstruction based on combined S1+S2 yields near-Fano-limited resolutions: $\sigma/E=1\%$ at 1.33 MeV (MiX), $5.8\%$ at 32.1 keV in Xurich II (Stephenson et al., 2015, Baudis et al., 2017). Resolutions remain stable at $\sim$5–6% across 0.2–1.3 kV/cm drift field when using the combined estimator, while single S1 or S2 channels show field-dependent degradation (Baudis et al., 2017).

4. Microphysics and Performance

• Light Yields: S1 yield is anti-correlated with the drift field: Xurich II measured $Y_{S1}=15.0$ (9.4 keV), $14.0$ (32.1 keV) PE/keV at zero field, falling to $10.8$ and $7.9$ PE/keV at 1 kV/cm, consistent with LUX and XENON100 (Baudis et al., 2017).
• Charge Yields: S2 charge yield increases with field due to suppression of recombination: $28$ e$^-$/keV (9.4 keV), $31$ e$^-$/keV (32.1 keV) at 1 kV/cm (Baudis et al., 2017).
• Single-electron S2 gain: Ranges from $g_{2}\approx16$ PE/e$^-$ (quartz TPC (Sato et al., 2019)) up to $34$ PE/e$^-$ (RELICS prototype (Xie et al., 24 Nov 2025)), depending on field and electroluminescence geometry.
• Electron Drift Velocity: Consistent with $v_d=\Delta z/\Delta t$, showing canonical field dependence (e.g., $v_d\approx1.5$–$2.0$ mm/μs for $E=$ 200–500 V/cm) (Baudis et al., 2017, Stephenson et al., 2015, Baur et al., 2022).
• Electron Lifetime: Purity-limited, with $\tau_{e}$ exceeding 200 μs in small systems and ms-scale in low-background large TPCs; $\tau_{e}$ is inversely proportional to O₂.
• Energy Threshold: For NR, S1 threshold of 2 PE ($\sim10.5$ photons) sets a threshold of $(2.3$–$2.7)$ keV$_\mathrm{nr}$ (Baudis et al., 2017). S2-only analyses with high $g_2$ can reach even lower thresholds (Xie et al., 24 Nov 2025).
• Spatial Resolution: $z$ from $t_{drift}$ yields $<1$ mm precision; $(x,y)$ from S2 light maps achieves few-mm scale (Stephenson et al., 2015, Aprile et al., 2010).

5. Advanced Detector Concepts and R&D

To address scaling and background challenges, a variety of innovative designs have been realized:

• Hermetic TPCs: Fully-sealed volumes using PTFE/quartz to inhibit $^{222}$Rn and exogenous impurity infiltration, achieving $<0.1\,\mu\mathrm{Bq/kg}$ Rn concentrations, enabling operation below the neutrino floor (Dierle et al., 2022, Sato et al., 2019).
• Quartz Chambers and Single-layer Graphene Electrodes: Minimize radiogenic backgrounds and photoelectron emission, achieving order-of-magnitude reduction in single-electron backgrounds relative to traditional stainless-steel electrodes (Wei et al., 2020).
• Single-phase and Radial TPCs: S2 electroluminescence directly in liquid (single-phase) is established with strategic micro-wire geometry, eliminating liquid-gas interface engineering and potentially improving scaling and background systematics. Measured S2 gains are lower than dual-phase (e.g., $g_2=1.9\pm0.3$ PE/e$^-$ (Tönnies et al., 2024)), but discrimination between ER and NR is preserved (Tönnies et al., 2024, Wei et al., 2021, Qi et al., 2024).
• Electron Extraction Optimization: Systematic study of extraction efficiency as function of liquid and gas phase fields reveals no saturation up to $E_{liquid}=7$ kV/cm; high field engineering of the gate-anode region is essential for maximizing S2 response (Edwards et al., 2017, Linehan et al., 2021). For example, efficiency increases from $0.21$ at $2.4$ kV/cm to $1.0$ at $7.1$ kV/cm (Edwards et al., 2017).
• 3D Event Reconstruction and ML Algorithms: Neural networks trained on S2 light-shape simulations deliver mm-scale fiducial volume definition and enable high-fidelity background rejection (Stephenson et al., 2015, Baur et al., 2022).

6. Applications in Rare-Event Searches and Medical Imaging

LXe TPCs are the reference technology for:

• Dark Matter Direct Detection: Dual-phase TPCs with full 3D event reconstruction, S2/S1 ER/NR discrimination ($>$99% ER rejection at 50% NR acceptance (Schumann, 2014, Aprile et al., 2010)), and powerful background suppression underpin world-leading WIMP searches (XENON1T, LZ, PandaX, DARWIN) (Schumann, 2014, Adrover et al., 2024, Baudis et al., 2017).
• Neutrinoless Double Beta Decay: Single-phase charge-collection or dual-phase architectures (EXO-200, nEXO), with 1% FWHM energy resolution at $Q_{\beta\beta}$, three-dimensional fiducialization, and advanced background rejection (Sorel, 2019, Stiegler et al., 2020). Open field cages and “skin” LXe tagging further improve background index by $\sim5\%$, critical for ton-scale 0νββ sensitivity (Stiegler et al., 2020).
• Coherent Elastic Neutrino-Nucleus Scattering (CEνNS): LXe TPCs with sub-keV thresholds (e.g., RELICS prototype demonstrates detection of $0.27$ keV events (Xie et al., 24 Nov 2025)), ultra-low backgrounds, and fine position resolution enable precision reactor neutrino CEνNS studies.
• Medical Imaging (PET, Compton telescopes): Sub-mm position resolution, 6–10% FWHM energy resolution at 511 keV, and fast timing (LXe singlet lifetime $\sim$2.2 ns) make LXe TPCs suitable for advanced PET and 3γ imaging modalities (Miceli et al., 2011, Oger et al., 2011).

7. Scalability, Performance, and Future Directions

With successful drift lengths up to 2.6 m and active masses up to 40 t in design (DARWIN), LXe TPCs have proven scalability (Adrover et al., 2024). Critical to scaling is:

• Electron lifetime management: Continual recirculation and purification enable ms-scale lifetimes, dominating charge survival and uniformity over extended drifts.
• High-voltage engineering: Robust gate-anode meshes and precision field cage manufacturing maintain field uniformity, minimizing S2 fluctuations and maximizing extraction—the latter unsaturated up to highest fields tested (Linehan et al., 2021, Edwards et al., 2017).
• Material radiopurity and Rn suppression: Sealed or mechanically hermetic chambers, purification cycling architectures, and material screening are driving ppb-level or sub-ppb $^{222}$Rn and U/Th contamination (Dierle et al., 2022, Sato et al., 2019, Wei et al., 2020).
• Expanded photodetector arrays and ML event reconstruction: Large-area SiPM and PMT coverage provide high photon detection efficiency (PDE), integral for competitive g₁/g₂ and low threshold. ML event reconstruction and simulation-driven corrections further enhance resolution and background rejection.

Emergent single-phase TPC variants trade strict field uniformity and S2 gain for radical simplification of liquid handling, interface control, and potential noise backgrounds—under active investigation for next-generation ton-scale detectors (Qi et al., 2024, Wei et al., 2021, Lin, 2021).

In conclusion, the LXe TPC remains the foundation technology for low-background rare-event searches, driven by continual refinement in field management, material control, scalable cryogenics, and advanced data analysis protocols (Baudis et al., 2017, Stephenson et al., 2015, Baur et al., 2022, Adrover et al., 2024). The performance achieved in current and prototype systems underpins the design and projection of multi-tonne future observatories targeting fundamental questions in physics.