Dual-Phase Xenon TPC: Principles and Advances

• Dual-phase xenon TPCs are radiation detectors that employ liquid and gaseous xenon phases to enable precise 3D position reconstruction and powerful background suppression.
• They achieve sub-keV energy thresholds and mm-scale spatial resolution by capturing prompt S1 scintillation and delayed S2 electroluminescence signals.
• Recent advances enhance charge gain and scalability for multi-ton detectors, vital for dark matter, neutrino, and rare decay experiments.

A dual-phase xenon time projection chamber (TPC) is a radiation detection instrument that utilizes both liquid and gaseous xenon phases to achieve precise measurement of rare low-energy events, enabling 3D position reconstruction and powerful background discrimination. Originally developed for direct detection of weakly interacting massive particles (WIMPs), dual-phase xenon TPCs now represent the most sensitive technology for dark matter, low-energy neutrino, and rare decay searches. These detectors operate by recording both prompt vacuum-ultraviolet scintillation (S1) in the liquid target and delayed proportional electroluminescence (S2) in the gas above, which is produced by ionization electrons extracted from the liquid by a strong electric field. The dual-phase architecture is critical for achieving high charge gain, sub-keV energy thresholds, mm-scale spatial resolution, and multi-ton scalability (Baudis, 2023, Schumann, 2014).

1. Detector Structure and Fundamental Operating Principles

A dual-phase xenon TPC consists of a cylindrical active volume of liquid xenon (LXe) overlaid by a thin gas gap of xenon (GXe), fully instrumented with photosensors (typically PMT arrays) above and below. The core elements are:

• Electrodes: Cathode at the bottom (often –10 kV to –100 kV), a grounded or defined gate near the LXe surface, and an anode grid in the GXe gap above. The precise liquid–gas interface must be stabilized with mm precision (e.g., using a weir or “diving bell” system) (Hogenbirk et al., 2016, Xie et al., 24 Nov 2025).
• Field-shaping rings: Copper or stainless-steel rings embedded in PTFE (Teflon) spacers, resistively chained to maintain a uniform drift field along the z-axis (Hogenbirk et al., 2016, Aprile et al., 2010).
• Photosensors: Arrays of VUV-sensitive PMTs (or more recently, SiPM tiles (Baudis et al., 2020)) at the top (GXe) and bottom (LXe) for S2 and S1 light collection, respectively, achieving photon detection efficiencies $g_1$ in the 0.05–0.15 $\rm PE/photon$ range (Xie et al., 24 Nov 2025).

A particle interaction in LXe yields S1 (prompt VUV, $\lambda \approx 175$–178 nm) and simultaneously excites ionization electrons. The electrons are drifted upward by an electric field ($E_{\mathrm{drift}} = 0.1$–1 kV/cm), extracted from the liquid by a strong field at the interface ($E_{\mathrm{ext}}\geq 5$–10 kV/cm), and accelerated in the gas gap where each electron generates proportional electroluminescence (S2).

S1 provides the event t₀; the delay to S2 gives the depth (z) via $z = v_{d} \Delta t$, where $v_{d}$ is the drift velocity. The S2 light pattern on the top array yields $(x, y)$ via reconstruction algorithms, enabling full 3D position determination and fiducialization (Schumann, 2014).

2. Electric Fields, Charge Transport, and Extraction Efficiency

The dual-phase architecture relies critically on careful control of electric fields across multiple regions:

• Drift Region: Uniform field ($E_{\mathrm{drift}} = 0.1$–1 kV/cm) between cathode and gate ensures collection of primary electrons with minimal recombination and diffusion. Typical drift velocities are $v_{d} \approx 1.5$–2.0 mm/μs at $\rm PE/photon$0 kV/cm (Baudis et al., 2017, Baur et al., 2022).
• Extraction Region (Liquid–Gas Interface): A strong field across the last mm of liquid ($\rm PE/photon$1) is required to extract electrons into the gas. Extraction efficiency $\rm PE/photon$2 transitions from near zero below a threshold ($\rm PE/photon$31.5 kV/cm) up to unity at $\rm PE/photon$4 kV/cm (Edwards et al., 2017). Empirically, $\rm PE/photon$5 is given by:

$\rm PE/photon$6

for $\rm PE/photon$7 kV/cm.

• Electroluminescence (S2) Region: In the gas gap, a field $\rm PE/photon$8–12 kV/cm is applied. The S2 photon yield per electron is

$\rm PE/photon$9

with $\lambda \approx 175$0 photons/(electron·kV·cm$\lambda \approx 175$1), $\lambda \approx 175$2 mm (Hogenbirk et al., 2016, Schumann, 2014).

Free-electron lifetime, $\lambda \approx 175$3, quantifies bulk purity (attachment to O$\lambda \approx 175$4-like impurities). For optimal operation, $\lambda \approx 175$5 μs is needed (attenuation length $\lambda \approx 175$6 m); typical TPCs achieve $\lambda \approx 175$7–1000 μs, with $\lambda \approx 175$8 S2 signal loss at full drift (Hogenbirk et al., 2016, Baudis et al., 2017, Xie et al., 24 Nov 2025).

3. Signal Formation, Event Reconstruction, and Energy Calibration

Each event generates two signals:

• S1: Prompt VUV scintillation from xenon excimer decays (singlet and triplet, τ ≈ 2 ns and 27 ns). Light yield at zero field $\lambda \approx 175$9 photons/keV, reduced to 2–5 photons/keV at typical drift fields due to suppressed recombination (Schumann, 2014, Huang et al., 2021).
• S2: Proportional (electroluminescent) scintillation in the gas gap—$E_{\mathrm{drift}} = 0.1$0–$E_{\mathrm{drift}} = 0.1$1 detected photoelectrons per extracted electron is standard (SEG of 25–34 PE/e$E_{\mathrm{drift}} = 0.1$2 reported in modern prototypes) (Xie et al., 24 Nov 2025).

Energy is reconstructed using the combined energy scale (CES) exploiting anti-correlation:

$E_{\mathrm{drift}} = 0.1$3

with $E_{\mathrm{drift}} = 0.1$4 eV (average work required per quantum). Combined energy resolutions $E_{\mathrm{drift}} = 0.1$5 below 6% at tens of keV and approaching 2% at 511–662 keV have been achieved, with sub-2% resolution at MeV scales for $E_{\mathrm{drift}} = 0.1$6 applications (Hogenbirk et al., 2016, Lin et al., 2013, Baudis, 2023).

Spatial reconstruction is achieved by:

• $E_{\mathrm{drift}} = 0.1$7 from time separation S2–S1: $E_{\mathrm{drift}} = 0.1$8 mm.
• $E_{\mathrm{drift}} = 0.1$9 from S2 hit-pattern: central resolutions $E_{\mathrm{ext}}\geq 5$0 mm for SiPM-based arrays (Baudis et al., 2020), several mm for PMT-based top arrays (Lin et al., 2013).

4. Discrimination, Background Suppression, and Performance Metrics

Background rejection in dual-phase TPCs is realized via:

• ER/NR Discrimination: The logarithm of the S2/S1 ratio (in PE) provides powerful separation, exploiting higher recombination for NR than ER. Discrimination power $E_{\mathrm{ext}}\geq 5$1 ER rejection at 50% NR acceptance is standard at energies relevant to WIMP scattering (Schumann, 2014, 0908.0790, Baudis, 2023).
• 3D Fiducialization: Sub-mm to mm-volume exclusion (“skin” and top/bottom cuts) excludes external backgrounds; multiple-scatter events (more than one S2) are vetoed.
• Intrinsic Radioactivity Control: Fiducial regions, radiopure materials (PTFE, low-background PMT/SiPMs), purification (SAES hot gas getters, distillation for Kr$E_{\mathrm{ext}}\geq 5$2 and Radon reduction), and hermetic chamber designs are employed for background $E_{\mathrm{ext}}\geq 5$3 events/(t·y·keV) (Dierle et al., 2022, Sato et al., 2019).

Key performance metrics from leading TPCs:

• Light yield (122 keV, 0 V/cm): $E_{\mathrm{ext}}\geq 5$4 PE/keV (Hogenbirk et al., 2016), $E_{\mathrm{ext}}\geq 5$5 PE/keV at sub-10 keV (Baudis et al., 2017).
• Charge yield: 28–31 electrons/keV at 9–32 keV, $E_{\mathrm{ext}}\geq 5$6 transmission with optimal fields (Baudis et al., 2017).
• Energy threshold: $E_{\mathrm{ext}}\geq 5$7 keV (S1), $E_{\mathrm{ext}}\geq 5$8 keV S2-only (RELICS prototype) (Xie et al., 24 Nov 2025).
• Electron lifetime: 200–1000 μs in modern LXe systems (Hogenbirk et al., 2016, Xie et al., 24 Nov 2025).
• Position resolution: $E_{\mathrm{ext}}\geq 5$9 mm (SiPM), $z = v_{d} \Delta t$0 mm (PMT) (Lin et al., 2013, Baudis et al., 2020).

5. Engineering Optimization and Scaling to Multi-ton Detectors

For large TPCs ($z = v_{d} \Delta t$1 up to 50 t), critical engineering drivers are:

• Electrode Design: Extraction region fields $z = v_{d} \Delta t$2–7 kV/cm, mesh spacings $z = v_{d} \Delta t$35 mm, tensioned SS meshes (pitch $z = v_{d} \Delta t$4–$z = v_{d} \Delta t$5 mm, $z = v_{d} \Delta t$6–$z = v_{d} \Delta t$7m) to minimize grid sag and ensure uniform fields (deflection $z = v_{d} \Delta t$8 mm over $z = v_{d} \Delta t$9–1.5 m) (Linehan et al., 2021).
• HV Delivery: Cathode voltages up to 100 kV, robust HV feedthroughs, and stress simulations to avoid breakdown (Adrover et al., 2024, Linehan et al., 2021).
• Cryogenics and Xenon Handling: Gas-phase purification, recirculation rates $v_{d}$0 kg/h, and precise level-metering systems to maintain surface alignment within $v_{d}$1 mm (Adrover et al., 2024).
• Radon Emanation Controls: Hermetic concepts (PTFE/quartz shells, cryo-fitted seals) isolate the sensitive LXe from Rn-emanating volumes, achieving $v_{d}$2–$v_{d}$3 reduction in $v_{d}$4Rn backgrounds (Dierle et al., 2022, Sato et al., 2019).

Scaling challenges include uniform field and level control over meter-scale lengths, mechanical tolerances on grid installation, and maintaining ultra-high LXe purity.

6. Recent Advances, Alternative Modes, and Future Directions

Recent advances:

• SiPM Readout: SiPM arrays now match PMTs in VUV performance and offer channel multiplicity for finer spatial resolution (Baudis et al., 2020).
• Alternative Target Phases: Demonstrated equivalence of scintillation and S2 gain in crystalline-vapor dual-phase designs, opening possibilities for in-situ radon-chain tagging via parent–daughter event topology (Kravitz et al., 2022).
• Low-Energy Sensitivity: Improved single-electron gain (SEG) up to 34 PE/e$v_{d}$5, direct sub-keV S2-only detection, and robust background modeling at the 0.27 keV level (Xie et al., 24 Nov 2025).
• Single-Phase LXe Proportional Scintillation: Demonstrated S1/S2 anti-correlation, with ER/NR discrimination maintained at $v_{d}$6 leakage, but S2 gain $v_{d}$7 PE/e$v_{d}$8 limits low-energy reach compared to dual-phase (Qi et al., 2024).

Next-generation TPCs (DARWIN/XLZD, PandaX-xT) target LXe masses up to 50 t, sub-1 keV thresholds, and $v_{d}$9Rn concentrations below 0.1 μBq/kg—enabling sensitivity to the solar neutrino floor and multi-purpose rare-event physics (Baudis, 2023, Dierle et al., 2022).

---

References:

• (Hogenbirk et al., 2016) Commissioning of a dual-phase xenon TPC at Nikhef
• (Schumann, 2014) Dual-Phase Liquid Xenon Detectors for Dark Matter Searches
• (Edwards et al., 2017) Extraction efficiency of drifting electrons in a two-phase xenon time projection chamber
• (Baudis et al., 2017) A Dual-phase Xenon TPC for Scintillation and Ionisation Yield Measurements in Liquid Xenon
• (Aprile et al., 2010) Design and Performance of the XENON10 Dark Matter Experiment
• (Xie et al., 24 Nov 2025) Development of a dual-phase xenon time projection chamber prototype for the RELICS experiment
• (Baudis et al., 2020) The first dual-phase xenon TPC equipped with silicon photomultipliers and characterisation with $(x, y)$0Ar
• (Lin et al., 2013) High Resolution Gamma Ray Detection in a Dual Phase Xenon Time Projection Chamber
• (Adrover et al., 2024) Commissioning of the 2.6 m tall two-phase xenon time projection chamber of Xenoscope
• (Linehan et al., 2021) Design and production of the high voltage electrode grids and electron extraction region for the LZ dual-phase xenon time projection chamber
• (Dierle et al., 2022) Reduction of $(x, y)$1Rn-induced Backgrounds in a Hermetic Dual-Phase Xenon Time Projection Chamber
• (Kravitz et al., 2022) Operation and performance of a dual-phase crystalline/vapor xenon time projection chamber
• (Qi et al., 2024) Feasibility of Liquid-phase Xenon Proportional Scintillation for Low-energy Physics
• (Baudis, 2023) Dual-phase xenon time projection chambers for rare-event searches