Material radioassay and selection for XENON1T

To attain the high sensitivity needed to detect a dark matter particle with a xenon time-projection chamber, all other sources of particle interactions need to be eliminated or minimized. These interactions are classified as background events. Radiogenic backgrounds, in particular, come from radioactive isotopes within the detector materials that decay and lead to alpha, beta, or gamma emissions. Neutrons from spontaneous fission of heavy isotopes or from secondary reactions within the detector materials also contribute to the radiogenic background and can mimic a dark matter signal.

To minimize the radiogenic background, the goal of the XENON1T radioassay program is to measure the radioactivity of all materials that are needed to build the detector and to select only the most radiopure materials for the final construction. To do this, we use mass spectrometry techniques and high-purity germanium spectrometers that are capable of measuring radioactivity at the level of 10-6 decays per second in a kilogram of material (Bq/kg). As comparison, a typical banana has an activity of ~102 Bq/kg!

Because natural radioactivity is present in the soil, the water, and in the air, it is also present in the XENON detector materials. The Figure shows a measurement obtained with a germanium spectrometer of the gamma rays emitted from a sample of photomultiplier tubes. The background (purple spectrum) is subtracted from the sample (pink spectrum) in order to quantify the expected activity from a XENON1T component or material sample.

A high-purity germanium spectrometer measurement of gamma rays emitted from a sample of XENON1T photosensors. Some prominent isotopes from different sources are labeled: primordial uranium and thorium decay chains (green), potassium (red), man-made (orange) and cosmogenic (orange) isotopes.

The most common radioactive isotopes present in the Earth are primordial uranium and thorium, each of which decays into a series of other radioactive isotopes (marked in green in the Figure). Potassium (red) is also a common, primordial isotope that is found in soil, and subsequently in food and in your body. Other isotopes that are found in detector materials come from interactions with cosmic rays (yellow) or from man-made activities (blue), i.e. industrial or medical use, nuclear power plant emissions, nuclear accidents, and military testing.

The measured activities of each material selected for detector construction are used in simulations of XENON1T to determine the expected background. This allows for a prediction of the attainable sensitivity of the detector to dark matter interactions. The radioassay measurement results from over 100 material samples are presented in our new paper “Material radioassay and selection for the XENON1T dark matter experiment”.

Results from a Calibration of XENON100 Using a Source of Dissolved Radon-220

With 2 tonnes of target material, XENON1T is currently the largest liquid xenon detector in the search for dark matter. The detector’s immense volume greatly increases our chances of successfully observing that rare scatter of a dark matter particle off a xenon atom that we have sought for more than a decade, like casting a larger net to catch more fish. However, the size also makes it considerably more difficult for us to know how our detector would respond to the scatters of dark matter or to the scatters of electrons or gamma rays that would obscure our view of dark matter. These electrons and gammas come from ambient radon, impurities in the xenon, and even the detector components themselves. While their effects can be reduced through an appropriate selection of location and the materials, they cannot be completely eliminated. Hence, it is necessary to understand how our detector responds to electrons and gammas in contrast to dark matter. A thorough understanding of these backgrounds is arguably the single most crucial part of our endeavor.

Historically, in XENON10 and XENON100, we used external radioactive sources of cobalt and thorium, which emit gammas. Unfortunately, a drawback of XENON1T’s size is that it hobbles the rate of interactions in the central region of the detector, which is the most important for us in understanding our detector. For this reason, we needed to identify an internal, or dissolved, radioactive source that can be injected directly into the detector’s liquid xenon target volume.

One possibility is a source of 220Rn, which we characterize using the XENON100 detector. The source is well suited to calibrate our detector because it can imitate the effects of ambient radon (222Rn). The radioactive decay of 212Pb (a daughter of 220Rn) generates an electron just like 214Pb (a daughter of 222Rn), and thus their responses are very similar. The only thing we have to verify is that we can spread the 212Pb atoms throughout the detector volume. Due to the particular design of XENON100, many gamma events from other 220Rn daughters prevent us from identifying which events arise from 212Pb. So, we use the distribution of 212Bi, the daughter of 212Pb, to clearly show that 212Pb reaches even the centermost part of the detector. This distribution is shown in the figure below. Once that calibration is done, we do not have to proactively clean our detector, as would be the case with alternatives. Since the half-life of 212Pb is about half a day, we just wait a few days for the detector to return to normal.

The source also provides alpha particles that prove useful in understanding our background. The alpha particles of 216Po and 220Rn enable us to easily pinpoint their locations at the time of decay. And because they happen in quick succession (about 0.1 seconds), the daughter 216Po can be paired with its parent 220Rn. These pairs show us the trajectory of 216Po in the liquid xenon and how long it takes to decay. With both the time and the displacement, we can calculate the average velocity of the 216Po atom. If we then determine the velocities of all identified RnPo pairs, we can create a map of atomic motion in the full xenon volume of a given detector, as shown here:

In the case of XENON100, we find that atoms move at speeds up to 8 mm/s. They move in a single convection cell with a small contribution that results from an electric field applied to the xenon volume. This aspect tells us that an appreciable fraction of 216Po atoms is left ionized after the decay of 220Rn.

In general, a known map of atomic motion motivates two additional techniques to help us understand the backgrounds. Firstly, such a map enables us to identify the presence of “dead regions” through which liquid xenon does not flow very well. This feature is important for keeping the xenon free of impurities. Secondly, with sufficiently slow atomic motion or sufficiently large detector volumes, we could match 214Pb to its parent 222Rn or 218Po following a method similar to the one described previously for 220Rn. This tagging process would clearly distinguish 214Pb background events from the coveted dark matter event.

All of the finer details about this dissolved radon source can be found in the dedicated publication in Phys. Rev. D: E. Aprile et al. (XENON), Results from a calibration of XENON100 using a source of dissolved radon-220, Phys. Rev. D 95, 072008 (2017). This publication is of course also available as a pre-print as arXiv:1611.03585.

Search for magnetic inelastic dark matter with XENON100

Search for magnetic inelastic dark matter with XENON100

There is the long-standing claim of the DAMA/LIBRA collaboration about a detection of dark matter via the highly significant observation of an annually modulating signal in radiopure sodium iodide crystals. This signal, however, is in conflict with exclusion limits from various other dark matter experiments, including XENON100. Several alternatives to the classical WIMP scenario have thus been proposed in order to reconcile these null results with DAMA/LIBRA. One of these models is magnetic inelastic dark matter (MiDM). The MiDM model is motivated by comparing certain properties of the different detector targets and how they possibly influence the expected event rates. Iodine, used in DAMA/LIBRA, is distinguished by its high atomic mass and high nuclear magnetic moment. This enhances the signal of MiDM compared to other targets, such as xenon, and opens up new parameter space for the DAMA/LIBRA signal that is not in conflict with other null results.

The exclusion limit (at 90% confidence level, CL) on MiDM interactions from of XENON100 for a dark matter mass of m=123GeV/c2. Also shown are the 68% and 95% (green) CL regions of the MiDM best fit to the DAMA/LIBRA modulation signal. The limit from COUPP data is shown as well.

In the framework of MiDM the dark matter particle is expected to scatter inelastically off the nucleus, thereby gets excited, and de-excites after a lifetime of the order of order μs with the emission of a photon with an energy of δ~100 keV. Given the mean velocity of the dark matter particle, it travels a distance of O(m) before it de-excites. Furthermore it is assumed that the dark matter particle has a non-zero magnetic moment, μχ. The combination of a low-energy nuclear recoil followed by  an electronic recoil from the de-excitation creates a unique delayed coincidence signature which has been searched for the first time using the XENON100 science run II dataset with a total exposure of 10.8 ton × days. No MiDM candidate event has been found, thus we calculate an upper limit on the interaction strength. The Figure shows the resulting limit for a dark matter mass of 123 GeV/c2 which completely excludes the DAMA/LIBRA modulation signal as being due to MiDM.The sensitivity of this type of analysis will be greatly improved for current ton-scale (e.g., XENON1T) and future multi-ton dual-phase LXe TPCs (e.g., XENONnT, LZ and DARWIN). This is not only due to the increased target mass, but also thanks to the higher probability of detecting the de-excitation inside the larger active volume.

XeSAT2017: Online krypton and radon removal for the XENON1T experiment

This talk by Michael Murra (slides) was presented at the XeSAT2017 conference in Khon Kaen, Thailand, from 3. – 7. April 2017.

The  main background for the XENON1T experiment are the intrinsic contaminants krypton and radon in the xenon gas. Instead of purifying the xenon once before starting the science run we were able to operate our distillation column in a closed loop with the XENON1T detector system running during its commissioning phase. This resulted into reducing the krypton concentration quickly below 1 ppt (parts per trillion, 1 ppt = 10^(-12) mol/mol) without emptying and refilling of the detector.

In addition, the column was operated in the same closed loop in inverse mode in order to reduce Rn-222 by about 20% during the first science run.

This so-called online removal for both noble gases along with the working principle of the distillation system are presented within this talk.

Search for Two-Neutrino Double Electron Capture of Xenon-124 with XENON100

Besides the hunt for dark matter particles, the XENON detectors can be used to search for many other rare processes. One interesting case arises from one of the xenon isotopes itself, namely 124Xe, which is slightly abundant in natural xenon (0.1%). While it is considered stable since its direct decay into 124I is energetically forbidden, there is a rare process in nature, so far only indirectly observed, which would lead to a decay of 124Xe into the isotope 124Te. This requires, in the most probable case, the simultaneous capture of two electrons from the closest atomic shell turning two protons into two neutrons. Since this happens rarely, the corresponding half-life is predicted to be as large as 1022 years, which overshoots the lifetime of the universe by some 12 orders of magnitude. Nevertheless, as the XENON detectors are built for the rare event detection of dark matter particles, they are also very well suited for a search of such a rare process. What would one expect to be the trace of such a decay within the detector? Although the nuclear reaction

124Xe + 2e124Te* + 2νe

would suggest that neutrinos are the signal to search for, as they are a direct product of the decay with a total energy of 2.8MeV, their weak interaction cross-section makes them not detectable. But there are two electrons now missing from the atom’s shell, which is usually from the closest one (K-shell). So there are two “holes” left at an energetically favored position. In a cascade-like process, electrons from upper shells are now dropping down, filling these holes. This releases their former higher binding energy of a characteristic value in the form of secondary particles such as X-rays or Auger electrons. These particles cascade is releasing a summed energy of 64 keV, which is the signature we expect to see with our detector.

Looking for this signal in our well-known XENON100 data from 2011 to 2012 with 225 live days of exposure, we found no signal excess above our background. This way, a lower limit on the half-life of the decay with a value of 6.5×1020 years could be determined using a Bayesian analysis approach. This is close to the optimistic theoretical predictions, but a bit less sensitive than the XMASS detector, which estimated the half-life to be larger than 4.7 x 1021 years.
However, the results from XENON100 can be seen as the preparation for the next step, XENON1T. As XENON1T has about 2kg of 124Xe in its two-ton active xenon target (a factor of 70 more compared to the 29g used in XENON100) it will be more sensitive to this rare decay. Moreover, the background in XENON1T is a factor of 30 smaller in the region of interest. After only five live days of measurement it is thus expected to explore regions no experiment has explored before, and after 2 live years of measurement, we can probe half lives up to 6 x 1022 years (see Fig.1). It has to be emphasized that this data comes for free while searching for dark matter particles, since both searches require the same settings.

hp

Figure 1: Expected sensitivity of XENON1T as a function of live time in days. The aimed duration for the dark matter search is marked at 2 ton years, which would translate into
two years of measurement using 1 ton of the detector mass as a fiducial volume. After 5
days new parameter space is explored.

The XENON1T detector is also prepared to search for competing decay modes of the double electron capture, as it has an improved response to high energy signals. The so far unobserved emission of two positrons and two neutrinos as well as a mixture with one positron emitted and one electron captured simultaneously. While any detection of these decay modes would certainly lead to a deeper understanding of standard nuclear physics another possible decay branch could open the door to physics beyond the Standard Model: The neutrinoless double beta decay. If this hypothetical mode, where no neutrinos would be emitted, would be detected, it would reveal that they are their own anti-particles and annihilate in this process of double beta decay. This would prove the violation of lepton number conservation and, additionally, it could tell something about the mass of neutrinos, which is known to be very small (<eV) but is not determined today. Unfortunately, the expected life time of these decays given by theoretical calculations is even larger than for the process with the emission of two neutrinos.

First Signals in the XENON1T Time Projection Chamber

While the functionality of each of the 248 PMTs had been tested during the different commissioning stages of the XENON1T dark matter detector, the signal detection with both PMT arrays and the full data acquisition system remained to be tested. For this, and for the LED_event1_cutsubsequent calibration of the time projection chamber (TPC), an LED illumination system has been set up with 3 individual channels, each branching out into six optical fibers distributed in a circumference around the TPC. Light shining through the fibers is collected by the PMTs, whose output signals are then magnified by a factor 10 with operational amplifiers and digitized with fast analog-to-digital converters.

The figure on the right shows the first detection of blue LED light by the XENON1T PMT arrays. A time delay between the LEDs has been set, resulting in the three peaks seen in the top panel, which correspond to the combined waveforms of all PMTs. The bottom panel shows the signals detected by each individual channel.

On March 17th, the TPC was filled with warm xenon gas for the first time, allowing to acquire the first scintillation signals with the detector. For these measurements, only the PMTs have been biased and no electric drift field was applied. The figure below shows the detection of an event occurring between the so-called screening mesh in front of the top PMT array and the photosensors (see the January 19 post for details on the TPC structure) and constitutes the first detection of an S2-like signal in XENON1T. The left panels show the hit pattern on the top and bottom arrays, while the right top and bottom panels display the summed waveform and the individual PMT hits, respectively.

first_s2

Water Tank Filling

We started to fill the water tank:

In this view from the top, the cryostat with the actual detector is visible on the left. Photomultiplier tubes of the water Cherenkov muon veto are seen at the bottom and side of the water tank, to the right of the image.

In this view from the top, the cryostat with the actual detector is visible on the left. Photomultiplier tubes of the water Cherenkov muon veto are seen at the bottom and side of the water tank, to the right of the image.

The water acts as a passive shielding against external radioactivity. In addition, using the photomultipliers that can be seen towards the right of the picture, the water acts as an active muon detector. Muons may induce events in the xenon detector that may mimic dark matter signals. We therefore turn a blind eye (“veto”) for a short time whenever a muon travels through the water tank.