---
title: 'DarkSide–20k: Dual-Phase LAr TPC Detector'
url: https://www.emergentmind.com/topics/darkside-20k-ds-20k
type: topic
---

# DarkSide–20k: Dual-Phase LAr TPC Detector

DarkSide--20k (DS--20k) is a next-generation, dual-phase liquid argon time projection chamber (LAr TPC) designed for direct detection of Weakly Interacting Massive Particles (WIMPs), with a 20-tonne fiducial mass and a projected “instrumental background-free” exposure of 200 tonne-years. Located at Gran Sasso National Laboratory (LNGS) in Italy, DS--20k integrates stringent low-radioactivity requirements, advanced cryogenics, novel cryogenic silicon photomultiplier (SiPM) instrumentation, and aggressive background-rejection strategies to push the sensitivity frontiers of WIMP-nucleon cross sections down to $10^{-48}\,\mathrm{cm}^2$ at 0.1 TeV/$c^2$[2312.03597].

## 1. Scientific Objectives and Sensitivity Goals

DS--20k addresses the unresolved problem of the particle nature of dark matter, which constitutes approximately 85% of the total matter density in the universe. The primary aim is direct detection of nuclear recoils induced by WIMPs in the 1 GeV--10 TeV/$c^{2}$ mass range. With a ten-year exposure ($200\,\mathrm{t}\cdot\mathrm{yr}$), the experiment targets fewer than 0.1 background events in the WIMP search region (44--89 keVee), achieving a WIMP-nucleon cross-section sensitivity floor of:
\[
\sigma_N \approx 1\times10^{-48}~\mathrm{cm}^2 \quad \text{for}~m_\chi = 0.1~\mathrm{TeV}/c^2
\]
The principal formula for the zero-background cross-section limit at confidence level CL is 
\[
\sigma_{\text{limit}} = \frac{-\ln(1-\mathrm{CL})}{M T \epsilon}
\]
where $M$ is the fiducial mass, $T$ the exposure time, and $\epsilon$ the analysis efficiency. For $T=10$ years, $M=20$ t, and $\epsilon=0.6$, the 90% C.L. sensitivity is $\approx 1.9\times10^{-48}$ cm$^{2}$[2312.03597].

## 2. Detector Architecture and Cryogenic Implementation

The core instrument is a dual-phase LAr TPC featuring:

- **Active argon mass:** 51 t (extracted from deep CO$_2$ wells, depleted in $^{39}$Ar)
- **Fiducial mass (WIMP search):** 20 t
- **Drift length:** $\sim$120 cm; **Drift field:** 200 V/cm (requiring $-75$ kV cathode bias)
- **Inner neutron veto:** 32 t of UAr in a 15-cm thick Gd-loaded PMMA shell, with neutron captures producing $\gamma$ cascades up to 8 MeV
- **Outer cosmic veto:** $\sim$600 t of LAr for external neutron/muon rejection

A central innovation is the use of a mechanically integrated assembly of TPC and veto, supported in a membrane cryostat with redundant insulation and LN$_2$ cooling[2312.03597][2210.00322]. Continuous recirculation and purification of UAr (up to 1000 slpm) ensures electronegative impurity levels below 0.06 ppb O$_2$-equiv, with measured system efficiencies of >95% in full-scale prototypes[2408.14071].

## 3. Background Rejection Strategies

Instrumental backgrounds are suppressed by a hierarchy of active and passive measures:

| Source                | Mitigation Technique                                        | Residual Background       |
|-----------------------|------------------------------------------------------------|--------------------------|
| $^{39}$Ar ($\beta$)   | UAr depletion (factor $1{,}400\pm200$), Aria distillation | $\sim$0.7 mBq/kg in UAr  |
| External $\gamma$     | 20 t fiducialization/self-shield, cryostat design         | < 0.1 events in ROI      |
| Neutron/muon          | Gd-PMMA inner veto, 600 t LAr outer veto                   | <0.1 events in 200 t·yr  |

Pulse-shape discrimination (PSD) between electron recoils (ER) and nuclear recoils (NR) in the 44--89 keVee ROI achieves $>$2.4$\times$10$^{-8}$ ER rejection efficiency. The combined effect of material selection, cleanroom assembly, double veto, and PSD is a projected instrumental background expectation $<$0.1 events over the full exposure[2312.03597]. Cosmogenic activation during UAr production, purification, and transport is minimized with baseline protocols validated by direct batch assay (DArT/ArDM) and is far subdominant to the intrinsic residual activity[2312.02183].

## 4. SiPM-Based Optical Readout and Performance

DS--20k replaces conventional PMTs with SiPM arrays for both TPC and veto detection. Each Photo-Detection Module (PDM) consists of 24 SiPMs (8$\times$12 mm$^2$ per SiPM), four of which are summed to form a channel; 16 PDMs are combined on a 400-cm$^2$ Photo-Detection Unit (PDU). 
Critical optical metrics:

| Parameter                       | Value/Spec                |
|----------------------------------|---------------------------|
| Photon-detection efficiency (PDE)| $\approx$45% (including fill factor) |
| Single-photoelectron charge res. | $\mathcal{O}(1\%)$        |
| Signal-to-noise ratio (SNR)      | $>$8 at 7 VoV             |
| Dark count rate                  | $<10^{-3}$ Hz/cm$^2$      |
| Correlated noise                 | $<$40% total              |
| Time resolution                  | ns-scale (PSD & 3D reco)  |

Component assembly is distributed among low-radon, clean environments in the UK and Poland, with rigorous module-level QA/QC under ISO5--ISO7 air and Rn$<$5 Bq/m$^3$[2312.03597]. The SiPM system enables high light yields ($\sim$8--10 pe/keV for S1), sub-ns timing, and low thresholds, directly supporting PSD and 3D event reconstruction.

## 5. Purification, Calibration, and Thermal/Hydrodynamics

Efficient target LAr purification employs a dedicated recirculation loop with getter-based chemical (O$_2$, N$_2$, H$_2$O) and radon removal. Computational fluid dynamics (CFD) studies inform double-ring LAr inlet placement, outlet geometry, and turnover time ($\tau_{\rm turn}\sim40$ days), delivering uniform mixing critical for continuous purification and for rapid distributed calibration with short-lived $^{83\text{m}}$Kr (1.83 h), which homogenizes in $\sim$13 min post-injection[2503.08468]. Dual-phase operation stability is preserved via gas-pocket control, with heat transfer at the liquid--gas interface characterized at $62$ W (best estimate) to $144$ W (upper bound), and the minimum gas inlet temperature of $89$ K to prevent anode condensation. 

## 6. Data Acquisition and Triggerless Readout

The DS--20k DAQ utilizes a fully triggerless, continuous-acquisition architecture. Key elements:

- 2\,720 SiPM channels are digitized at 125 MSa/s, 16 bit, across 48 CAEN digitizers.
- Firmware-level zero suppression (500 ns windowing), followed by software matched-filtering and peak finding in front-end processors, reduces a raw $680$ GB/s stream to $\sim$3 GB/s, and ultimately to $<$60 MB/s physics data.
- Data are aggregated in 1-s time slices with $<0.5\%$ duplication overhead to ensure continuity over TPC drift time ($\sim$5 ms).

This design sustains single-photon sensitivity at high dynamic range, supporting real-time event-building and later offline analyses[2502.15651].

## 7. Timeline, Construction Status, and Outlook

Detector infrastructure at LNGS and the external cryostat have been under construction since 2023. Key milestones include:

| Milestone                          | Year   |
|-------------------------------------|--------|
| Cryostat and support installation   | 2023--2024 |
| UAr extraction/purification begins  | 2024   |
| TPC and optical-plane assembly      | 2025   |
| Commissioning (TPC, veto, DAQ)      | 2025–2026 |
| Physics data-taking initiation      | 2026 (planned) |

The program expects at least a decade of data collection, targeting a total exposure of $200\,\mathrm{t}\cdot\mathrm{yr}$[2312.03597]. With projected sensitivity covering 1 GeV/$c^2$ to 10 TeV/$c^2$ WIMP masses and a minimum cross-section reach dictated by a near“zero-background” regime, DS--20k is positioned to deliver world-leading argon-based direct dark matter constraints while validating underpinning technologies for future larger-mass LAr detectors.

Source: https://www.emergentmind.com/topics/darkside-20k-ds-20k