DEAP Project Director:
Mark Boulay (Queen's University)
phone: (613) 533-6197
fax: (613) 533-6813
Department of Physics, Stirling Hall,
Queen's University at Kingston,
Kingston, Ontario, Canada
K7L 3N6
For general information email:
mark.boulay@queensu.ca

DEAP

Liquid argon based Dark Matter detection

Dark matter makes up about 25% of our universe, yet it has never been detected. The goal of the DEAP experiment is to directly observe and identify this dark matter component of the universe. This will be achieved by observing the elastic scattering of dark matter particles, probably in the form of Weakly Interacting Massive Particles (WIMPs), from argon nuclei. Argon in its liquid form is a favorable detection medium for Dark Matter searches because it has a high stopping power against ionizing radiation and good light yield, it allows for any desired detector shape and, due to its low cost, for a large detector mass. A very low background can be reached due to ease of purification and scintillation characteristics which are suitable for achieving very powerful pulse shape discrimination. A prototype detector, DEAP-1, has been operating since 2007. Background suppression against electromagnetic events using pulse shape discrimination could be demonstrated at the level of 10-7. The DEAP-3600 detector currently under construction will be sensitive to Dark Matter interaction cross sections down to 10-46 cm2 per nucleon.

DEAP collaboration members at the collaboration meeting in Kingston March 2012

3-D rendering of the DEAP-3600 detector under construction at the moment.

A picture of the DEAP-1 detector without the PMTs installed.

People on top of the support structure.

The support structure for the DEAP-3600 and MiniClean experiments in the CubeHall at SNOLAB

The DEAP-1 Prototype

DEAP-1 is a prototype for DEAP-3600 and has been running at 3100 m.w.e. in SNOLAB since 2007. The main purpose of DEAP-1 was to demonstrate the power of pulse shape discrimination for the suppression of backgrounds due to β and γ interactions. The detector is shielded against neutrons by 8000 kg of water. It is now being run to study surface contaminations and to prototype components for DEAP-3600.

Liquid Argon scintillation

The passage of ionizing radiation through liquid argon does not produce any permanent chemical change. It does however ionize and excite argon atoms, which in that state form strong bonds with regular argon atoms, leading to ionized or excited dimers (excimers) [1]. Highly excited excimers quickly de-excite non-radiatively to lower energies until they are in either a singlet or a triplet lowest energy state. The decay of those excimers into the repulsive ground state is the origin of argon scintillation light, which at a wavelength of 128 nm[2] is not energetic enough to re-excite another argon atom and thus passes through the argon unhindered. The singlet state decay is an allowed transition with a lifetime of 6.7 ns, while the triplet state decay is “forbidden” and thus has a much longer lifetime of about 1600 ns[3]. This is the basis for the background suppression used in DEAP.

Pulse shape discrimination

WIMPs interact in the detector by nuclear recoils. Separation of β-γ interactions from nuclear recoils is therefore critical for background suppression. Such a separation is possible in liquid argon (LAr) by pulse-shape discrimination (PSD) based on scintillation timing. Electromagnetic interactions preferentially excite the argon excimer triplet state while nuclear recoils tend to excite the singlet state. Since the singlet state lifetime is so short compared to the triplet state lifetime, the signal intensity at the beginning of the scintillation pulse can be used to separate these two classes of events.

Our PSD parameter Fprompt is the ratio of measured light within 150 ns of the leading edge of the signal to the total amount of light. Waveforms are accumulated for a total of 9 microseconds (old DAQ) or ~14 microseconds (new DAQ). Data taken in DEAP-1 with an AmBe neutron source shows how, at energies above 20 keV electron equivalent, nuclear recoil events from the neutrons and γ events form bands around Fprompt values of about 0.8 and 0.3. At lower energies, the two bands are no longer well separated due to worsening statistics of photo electron distribution in the prompt time window and due to noise. The low energy threshold in DEAP is determined by the energy at which too many γ events leak into the nuclear recoil region. Optimization in light yield for better statistics thus translates directly into a lower energy threshold and is an important design consideration for DEAP-3600.

Backgrounds

In rare event search experiments like DEAP all possible backgrounds have to be understood and dealt with by either eliminating them or showing that they will not mimic a WIMP signal. The following sources of background need to be addressed in DEAP: By far the largest background will be β particles from the decay of 39Ar, which is produced cosmogenically and decays at a rate of about 1 Bq per kg of argon. This background, as well as any γ radiation, can be very effectively mitigated by pulse shape discrimination (PSD). Sources of argon depleted in 39Ar are also being explored. A more dangerous background are neutrons, which mimic the expected WIMP signal in the PSD parameter Fprompt and in energy. The detector will be submerged in a water tank to stop neutrons on their way to the detector. Neutrons emitted from the PMT glass and detector materials are absorbed in the acrylic light guides, filler material and the acrylic vessel. Events from alpha particles, most notably from radon and radon decay products, mimic the WIMP signal in Fprompt, but, provided they lose all of their energy in the LAr volume, not in energy. Surface alpha events, as shown in figure 11, lose part of their energy outside of the LAr volume, and can thus mimic a WIMP signal in energy as well. A fiducial volume cut which based on event position reconstruction discards events close to the surface reduced the rate of this background in DEAP-3600. Alpha events close to the surface can also lead to scintillation of the wavelength shifter TPB (cases b and c in figure 11). The fiducial volume cut is also effective against these events in DEAP-3600, but they might affect the WIMP sensitivity of DEAP-1. This background is currently under investigation.

The DEAP-3600 detector

The DEAP-3600 detector, shown to the right, will be the most sensitive experiment for the direct detection of dark matter particles, with sensitivity to spin-independent WIMP-nucleon scatters with cross-sections as low as 10-46 cm2 per nucleon. This represents an increase in sensitivity of a factor of 500 over current experimental limits (see figure 12). This high sensitivity will be achieved due to the very large target mass possible for liquid argon, and the very low background level achievable at the unique SNOLAB facility, the deepest site with the lowest rate of cosmic-ray muons and associated neutron backgrounds at which to perform such an experiment. Neutron backgrounds are further minimized by operating the detector within a water tank. The dangerous background due to surface alpha events is reduced by removing a 500 μm layer of material from the inside surface of the acrylic vessel before filling it with argon, which effectively removes alpha emitters present there.

DEAP-3600 will allow for a three year background-free run with a 1000 kg sensitive argon target.