ILIAS-N3: Direct Dark Matter Detection

Background Simulations, Neutron-Shield and Muon-Vetoes



1. Introduction

The sensitivity of current and future dark matter detectors to WIMP-nucleus interactions depends crucially on their ability to suppress and reject all types of backgrounds. Some of these backgrounds, i.e. gammas, betas and partly alphas, can be rejected using energy deposition, event localization and intrinsic properties of the detector materials in discriminating electron-like and nuclear-like events. Neutron-induced events, in particular low-energy single nuclear recoils, being indistinguishable from those expected from WIMP-nucleus interactions, are the most important background, which remains after all other types of events are rejected by off-line analysis.
The aim of this working group is to assess various backgrounds in detectors based on different technologies, and formulate requirements for passive shielding and active veto systems to reduce these backgrounds, thus helping to estimate sensitivity and to design future dark matter detectors.


2. Sources of background

We can consider three main sources of background:
  • external neutron and gamma background arising from U/Th/K traces in the rock surrounding dark matter detectors;

  • internal background of neutrons, gammas, betas and alphas from radioactive contamination of detector components and shielding;
  • external and internal background from neutrons produced by cosmic-ray muons and their secondaries in the rock, shielding and detector components.
Some special cases of external and internal backgrounds should also be mentioned which are also common to practically all experiments:
  • gamma, alpha and neutron backgrounds from radon, accumulated in air and detector components;
  • background arising from surface contamination of the target materials and detector vessels;
  • activation of detector components by cosmic-rays at the surface.
The backgrounds have to be simulated with Monte Carlo codes and the rates of events in various detectors due to these backgrounds have to be calculated.

3. Plan of work

Extensive studies of various types of background for dark matter experiments have been started as parts of the national programmes. The objectives of this working group are:

  • To coordinate the efforts of several groups from different institutions in studying the background for dark matter experiments.
  • To share information about Monte Carlo codes, helping each other in learning codes, modifying them for specific purposes and using codes for simulations.
  • To report the results of simulations within the working group, the N3 network and the ILIAS project.
  • To make information about Monte Carlo codes and results of background simulations available to large scientific community.

To achieve these objectives, the working group plans:

  1. To identify Monte Carlo codes suitable for simulations of various types of the background.

  2. To compare the results of simulations using different codes with each other and with available experimental data.
  3. To critically assess the code performance and select the codes best suited for each particular task.
  4. To create a web-page which will have an information concerning the working group progress including code comparison and results.
  5. To create a code repository which will contain modified programs (subroutines, include files, input files, interfaces, geometry descriptions, physics lists, cross-section libraries) to be used by all interested parties (note that the working group has no rights and no intention to distribute original codes which can be purchased from the code developers; code developers do not carry any responsibility for modifications done as part of our work).
  6. To carry out simulations (as part of national programmes) of various backgrounds in dark matter detectors and to compare the results.
  7. To study efficiency of different shields and active veto systems to suppress and reject the background; to formulate the requirements for shielding and vetoes to be used in dark matter detectors with different technologies.
  8. To calculate the event rates due to various backgrounds in dark matter detectors taking into account different configurations for passive shielding and active veto system.
  9. To write yearly reports on the status of work.
  10. To disseminate the results of the study through publications in the refereed journals, conference proceedings, preprints and the group wep-page.

The work will be done in cooperation with other working groups within N3 network (for instance, WG6 - HPMP), and the research projects within ILIAS - JRA1 and JRA2.

4. Participants

  • DAPNIA/SPP, Saclay (France) - EDELWEISS I/II, cryogenic Ge detectors.
  • IN2P3, Lyon (France) - EDELWEISS I/II, cryogenic Ge detectors.
  • Orsay (France) - Cross-section measurements.
  • LSM, Modane (France) - EDELWEISS I/II, cryogenic Ge detectors.
  • University of Karlsruhe (Germany) - CRESST I/II, cryogenic CaWO4 detectors.
  • Technical University, Munich (Germany) - CRESST I/II, cryogenic CaWO4 detectors.
  • University of Tuebingen (Germany) - CRESST I/II, cryogenic CaWO4 detectors.
  • Imperial College, London (UK) - ZEPLIN I/II/III/MAX - xenon detectors.
  • RAL, Chilton (UK) - ZEPLIN I/II/III/MAX - xenon detectors.
  • University of Sheffield (UK) - ZEPLIN I/II/III/MAX - xenon detectors; DRIFT I/II - low-pressure gas TPC.
  • University of Zaragoza - IGEX, Ge detectors; ANAIS, NaI detectors; ROSEBUD.
  • LNGS-INFN, Assergi (Italy) - background measurements and simulations for the Gran Sasso Laboratory.
  • University of Milan (Bicocca) and INFN-Milan (Italy) - TeO2 crystals, also for double beta decay experiment.
  • University of Novi Sad (Serbia and Montenegro) - Cross-section measurements.
5. Publications

There was a number of publications by the participants of the working group on background issues for dark matter detection:

  1. Proceedings of the 4th International Workshop on the Identification of Dark Matter - IDM2002 (York, UK, 2-6 September 2002)
    H. Wulandari et al.
    p. 464
  2. Proceedings of the 4th International Workshop on the Identification of Dark Matter - IDM2002 (York, UK, 2-6 September 2002)
    G. Chardin and G. Gerbier
    p. 470
  3. Proceedings of the 4th International Workshop on the Identification of Dark Matter - IDM2002 (York, UK, 2-6 September 2002)
    V. Kudryavtsev et al.
    p. 476
  4. Nucl. Instrum. and Meth.
    V. A. Kudryavtsev et al.
    in Phys. Res. A, 505 (2003) 688
  5. Talk at TAUP-2003
    G. Gerbier
    see TAUP-2003 web-site
  6. Talk at TAUP-2003
    S. Cebrian
    see TAUP-2003 web-site
  7. PhD thesis (2003)
    H. Wulandari
    Technical University of Munich
  8. Astroparticle Physics (2004)
    J. M. Carmona et al.
    in press
  9. Astroparticle Physics (2004)
    M. J. Carson et al.
    in press; hep-ph/0404042

These publications reflect the work done so far and provide a good start for a working group activity.

6. Code assessment

  • Codes for spontaneous fission and (Α, n) reactions:

    • - SOURCES (LANL); correct technique but: limited cross-section library (quite old), 6.5 MeV limit (modified by the Sheffield group), soft spectrum does not agree with most measurements.
      - MCNP; good for neutron propagation, but can it be used for spontaneous fission and ( ,n) reactions (TUM, Lyon, Saclay and others)?
      - Is there any other code that can be used for this?

  • Muon-induced neutrons (need for a single code to produce, transport and detect all particles):
    • - FLUKA; widely used for neutron transport, radiative damage etc. but simplified generation of nuclear recoils (Sheffield, TUM, Milan, Zaragoza).
      - GEANT4; potential for excellence but needs more tests (ICL).
      - GEANT3; probably good enough after modifications, nuclear recoil production and photonuclear interactions are under question (Karlsruhe, Lyon and Saclay).

  • Codes for neutron transport and detection:
      - MCNP; nuclear recoil production is slightly complicated but possible (TUM, Lyon, Saclay and others).
      - GEANT4; some bugs are still present but the code is good in general (Sheffield, RAL, ICL, Zaragoza, Milan).

  • Codes for cosmogenic activation:
      - COSMO (Milan).

  • In-house codes for neutron production, transport and detection (FAUST - RAL, other codes from EDELWEISS, TUM): how good are they?

7.Comparison with experimental data

  • Published neutron flux measurements at Gran Sasso, Modane, Canfranc, CPL (Korea) and others.

  • Planned neutron flux measurements at Gran Sasso, Modane, Canfranc, Boulby as proposed in JRA1 (ILIAS).
  • Measurements of muon-induced neutrons at various depths underground: neutron yields, energy spectra, lateral distributions etc.: LVD, KARMEN, LSD etc.
  • Measurements of muon-induced neutrons using active vetoes for dark matter detectors (existing and planned): Canfranc, Modane, Boulby, Gran Sasso.

8.Some graphs showing results of on-going simulations

Figure 1. Neutron energy spectra from rock activity behind shielding: thick red curve - spectrum at the rock/cavern boundary; thin red curve - spectrum after 30 cm of lead; light blue, dark blue and green curves - spectra after 20, 30 and 40 g/cm2 of hydrocarbon shielding, respectively, behind lead (from Carson et al., Astropart. Phys., in press - simulations for salt rock at Boulby).

Figure 2. Energy spectra of muon-induced neutrons at the rock/cavern boundary (red filled circles) and after the lead (30 cm) and hydrocarbon (40 g/cm2) shielding (blue open circles) (from Carson et al., Astropart. Phys., in press - simulations for salt rock at Boulby).



Figure 3. Neutron energy spectra at the EDELWEISS II detectors due to various sources of background (from Gerbier, talk at TAUP-2003).





Figure 4. Effect of shielding materials used in CRESST shown as the neutron flux at the boundary between the shield and the detector area inside the shield (open circles). As a comparison the flux of neutrons entering the hall including back scattering is shown (filled circles) (from Wulandari, PhD Thesis, TUM, 2003).



Figure 5. Simulation of neutrons from radioactivity in rock and comparison with measurements at Can-franc Underground Labo-ratory (from Cebrian, talk at TAUP-2003; also Carmona et al., Astroparticle Phys., in press).



Figure 6. Simulated gamma spectra for bulk contaminations of the TeO2 crystals, the copper structure, the Roman lead shield and the outer lead shield. Each spectrum is obtained by summing the simulated anticoincidence spectra of all CUORE detectors (from CUORE proposal, http://crio.mib.infn.it/wig/).

documents