The XENON Project
The XENON Project is a series of experiments first proposed by Elena Aprile in 2001 which uses liquid noble time projection chambers as dark matter detector.
The first part of the project XENON10 contained about 10kg of xenon. It was soon followed by a 100kg scale detector, XENON100. The next step, a ton scale detector, XENON1T, is currently under construction, while the project has already been extended to include a fourth step, a multi-ton scale detector called XENONnT.
All detectors were or are located in the Laboratori Nazionali del Gran Sasso (LNGS), an particle physics laboratory located under the Gran Sasso mountain in Italy at a depth of 1.4km or 3600 mwe (meter water equivalent).
The XENON10 detector, shown in Figure 2, was the first and smallest phase of the XENON Project and the first experiment to use liquid xenon in attempts to detect dark matter. It developped extremely fast and released its first result in 2007, reaching the world best limit on dark matter searches at the time (WIMP-nucleon cross section of ) and establishing liquid noble detectors as a viable technology for dark matter searches.
XENON100 was built between 2007 and 2009, at which point it started taking data. Several world best limits were released in 2010, 2011 and 2012 (best limit on the WIMP-nucleon cross section at ), unfortunately without significant signal above the expected background, indicating no sign of dark matter yet. However, these results probed a significant chunk of the parameter space where dark matter is thought to be.
Although XENON100 has now reached its scientific goal and is primarily used as an R&D tool for the upcoming XENON1T experiment, there is still a considerable amount of data that is being or still remains to be analyzed.
Here is a recap of XENON100 most important Scientific Results (from the most to least recent):
- The first axion results produced by XENON100: 2014-09-09: First Axion Results from the XENON100 Experiment, arxiv:1404.1455, cite as: E. Aprile et al. (XENON100), Phys. Rev. D 90, 062009
- Limits on spin-dependent WIMP-nucleon cross sections from 225 live days of dark matter data: 2013-07-09: Limits on spin-dependent WIMP-nucleon cross sections from 225 live days of XENON100 data, arXiv:1301.6620, cite as: E. Aprile et al. (XENON100), Phys. Rev. Lett. 111, 021301 (2013)
- Main dark matter results after 225 blind live days of dark matter data: 2012-07-25: Dark Matter Results from 225 Live Days of XENON100 Data, arXiv:1207.5988, cite as: E. Aprile et al. (XENON100), Phys. Rev. Lett. 109, 181301 (2012), arXiv:1207.5988
- XENON100 data analysis of inelastic dark matter: 2011-04-15: Implications on Inelastic Dark Matter from 100 Live Days of XENON100 Data, arXiv:1104.3121 cite as: E. Aprile et al. (XENON100), Phys. Rev. D 84, 061101 (2011)
- Results from 100 live days of XENON100: 2011-04-13: Dark Matter Results from 100 Live Days of XENON100 Data, arXiv:1104.2549 cite as: E. Aprile et al. (XENON100), Phys. Rev. Lett. 107, 131302 (2011)
- First results from XENON100: 2010-05-11: First Dark Matter Results from the XENON100 Experiment, arXiv:1005.0380 cite as: E. Aprile et al. (XENON100), Phys. Rev. Lett. 105, 131302 (2010).
XENON1T is the ongoing phase of the XENON project. Its construction started in 2014 and is now nearly complete, with only the TPC missing. XENON1T is on schedule to take science data at the end of 2015. With its 1 ton fiducial volume of liquid xenon and a much lower background, this detector has a very strong discovery potential with a projected sensitivity improved by a factor 100 (projected WIMP-nucleon cross section limit of ).
The Detector Principle
As previously mentionned, in all phases of the project, the detector used is a special case of Time Projection Chamber (TPC) which consists of a cylinder made of highly reflective material (teflon) and filled with liquid xenon, with a gaseous xenon layer on top. Incoming particles interact directly with xenon atoms and by doing so induce nuclear or electron recoils in the xenon, depending if the entire xenon atom recoils from the interaction or if only an electron from the xenon atom is ejected. These recoils in turn produce scintillation (light) and ionization (electron) signals. The scintillation signal (S1) is immediately detected by photomultiplier tubes at the bottom and top of the TPC while the electrons (ionization signal) are drifted along the length of the TPC where they reach the gas phase and by colliding with xenon atoms, emit a second scintillation signal (S2).
The position of the initial interaction between an incoming particle and a xenon atom (called an event) can be reconstructed in 3D.
Because electrons drift vertically the x-y location of the S2 signal is the same as the one of the S1 signal. In addition, the S2 signal happens just beneath the top PMTs which makes the x and y coordinates of an event easily determined by the very localized top PMT hit pattern (figure 7).
The z position of an event can simply be inferred from the S1-S2 time separation, which is the time difference it took from an electron to drift upwards into the gas phase. Because the drift speed is constant and accurately known and that the light travel time is instantaneous on this scale, the drifted distance and so the z position of the event can be accurately determined.
Identifying the initial interacting particle is done based on using the ratio between the S1 and S2 signals. Indeed S2 signals are always bigger than S1 signals. In addition, the ionization yield (how much charge is going to be emitted from an initial interaction) is different for electron and nuclear recoils, and an S2 coming from an initial nuclear recoil is going to be smaller than an S2 coming from an initial electronic recoil. Consequently the ratio S2/S1 is always going to be much smaller for nuclear recoils than electronic recoils, making the S2/S1 ratio a very efficient discrimination method.
To visualize this discrimination effect, the detector is exposed to calibration sources, gamma sources to study electronic recoils and neutron sources to study nuclear recoils. By purposedly irradiating the detector with a big amount gammas and neutrons, the detector’s response to these particles can be observed. When plotting the ratio S2/S1 as a function of energy, two very distinct regions appear (figure 9), one for nuclear recoil and one for electronic recoil.
Because WIMPs are expected to interact via nuclear recoil, the region of interest for the search for WIMPs is within the nuclear recoil band and within the expected energy range of a WIMP induced recoil.
XENON100 Core Results
XENON100 disclosed its main results after 225 live days of data taking. In order to remove as much bias as possible, the analysis was blinded, meaning that it was developed offline and only calibration was looked at during that time. Only once the full 225 days were on file and the analysis package completely developed, was the analysis applied to the dark matter data.
Figure 13 shows the result of such analysis. Two events remained in the region of interest after the analysis was performed even though only one was expected from background.
A profile likelihood statistical analysis of these events rejected them as background allowing XENON100 to reach a never before achieved limit on the WIMP nucleon cross section of .
The next experiment, XENON1T, currently under construction at LNGS, will have a high discovery potential with an increase in sensitivity by 2 orders of magnitude, consequently probing very deep into the preferred parameter space from SUSY.