[Paper Review] Status of the EDELWEISS-2 Dark Matter Search

A. Chantelauze|ArXiv.org|Oct 31, 2007

Dark Matter and Cosmic Phenomena3 citations

TL;DR

The EDELWEISS-2 experiment uses high-purity germanium bolometers in a deep underground lab to search for dark matter WIMPs via nuclear recoils, achieving background suppression through dual heat/ionization readout and active muon veto shielding. It reports a projected WIMP-nucleon cross-section sensitivity of 10⁻⁸ pb for 100 GeV/c² WIMPs, representing a 100× improvement over EDELWEISS-I, with the muon veto system successfully measuring an underground muon flux of 11.2 ± 0.3 μ/m²/d.

ABSTRACT

The Edelweiss programme is dedicated to the direct search for Dark Matter as massive weakly interacting particles (WIMPs) with Germanium cryogenic detectors operated in the Laboratoire Souterrain de Modane in the French Alps at a depth of 4800 mwe. After the initial phase Edelweiss I, which involved a total mass of 1 kg, the second step of the programme, Edelweiss II, currently operates 9 kg of detectors and an active shielding of 100 m^2 muon veto detectors and is now in its commissioning phase. The current status and performance of the Edelweiss II set-up in terms of backgrounds will be given, the underground muon flux measured with the muon veto system will be presented.

Motivation & Objective

• To directly detect weakly interacting massive particles (WIMPs) as a candidate for dark matter.
• To reduce backgrounds in germanium bolometer detectors through dual heat/ionization readout and active surface event rejection.
• To suppress electromagnetic and neutron-induced backgrounds using multi-layer shielding and a 100 m² muon veto system.
• To achieve a WIMP-nucleon cross-section sensitivity of ~10⁻⁸ pb for 100 GeV/c² WIMPs by 2009–2010.
• To validate new detector types (Ge/NbSi and Ge/NTD with interdigitized electrodes) for improved surface event rejection.

Proposed method

• Uses high-purity germanium cryogenic bolometers operating at ~20 mK to detect nuclear recoil energy from WIMP scattering.
• Measures both heat and ionization signals to distinguish electron recoils (background) from nuclear recoils (WIMP signal), using a Q = E_ionization / E_recoil discrimination parameter.
• Employs a 100 m² plastic scintillator muon veto system with 98% coverage to tag muons and reduce neutron backgrounds from muon spallation.
• Applies 20 cm lead shielding, 50 cm polyethylene shielding, and radiopure materials to suppress gamma and neutron backgrounds.
• Uses a reversed-design dilution cryostat with pulse tubes and a He reliquifier to minimize helium consumption and enable long-term operation.
• Performs Monte Carlo simulations to estimate residual nuclear recoil rates and background levels, predicting <10⁻³ evt/kg/d above 10 keV.

Experimental results

Research questions

• RQ1 What is the effective muon flux at 4800 mwe depth in the Modane underground laboratory, and how well does the muon veto system detect it?
• RQ2 Can the dual heat/ionization readout in germanium bolometers achieve sufficient electron recoil discrimination to suppress electromagnetic backgrounds?
• RQ3 What is the contribution of neutron backgrounds to the WIMP detection threshold, and can they be mitigated by active shielding and veto systems?
• RQ4 How effective are the new Ge/NbSi and Ge/NTD detectors with interdigitized electrodes in rejecting surface events?
• RQ5 What is the projected WIMP-nucleon cross-section sensitivity of EDELWEISS-2 with 28 and later 68 detectors?

Key findings

• The underground muon flux was measured at 11.2 ± 0.3 μ/m²/d using the muon veto system, in good agreement with GEANT-4 simulations (12.5 ± 0.4 μ/m²/d).
• The energy distribution of muon candidates fits a Landau distribution with a most probable energy of 10.8 MeV, confirming the system's response to muon interactions.
• The muon veto system achieves a raw data rate of 8 kHz, reduced to 3 Hz for event data through coincidence timing, with a 100 ns time window for event identification.
• The nuclear recoil rate above 10 keV is estimated at <10⁻³ evt/kg/d, corresponding to a WIMP-nucleon cross-section sensitivity of 10⁻⁸ pb for 100 GeV/c² WIMPs.
• The Ge/NbSi and Ge/NTD detectors with interdigitized electrodes demonstrated surface event rejection efficiency >95% in surface laboratory tests.
• Low-background physics runs with the initial 28-detector setup are expected to reach a sensitivity of ~10⁻⁷ pb by July 2008, with further improvement to 10⁻⁸ pb by 2009–2010 after adding 40 additional detectors.