Cryogenic characterization of a LiAlO 2 crystal and new results on spin-dependent dark matter interactions with ordinary matter
A.H. Abdelhameed, G. Angloher, P. Bauer, A. Bento, E. Bertoldo, R. Breier, C. Bucci, L. Canonica, A. D'Addabbo, S. Di Lorenzo, A. Erb, F.v. Feilitzsch, N. Ferreiro Iachellini, S. Fichtinger, D. Fuchs, A. Fuss, V.M. Ghete, A. Garai, P. Gorla, D. Hauff, M. Ješkovský, J. Jochum, J. Kaizer, M. Kaznacheeva, A. Kinast, H. Kluck, H. Kraus, A. Langenkämper, M. Mancuso, V. Mokina, E. Mondragon, M. Olmi, T. Ortmann, C. Pagliarone, V. Palušová, L. Pattavina, F. Petricca, W. Potzel, P. Povinec, F. Pröbst, F. Reindl, J. Rothe, K. Schäffner, J. Schieck, V. Schipperges, D. Schmiedmayer, S. Schönert, C. Schwertner, M. Stahlberg, L. Stodolsky, C. Strandhagen, R. Strauss, I. Usherov, F. Wagner, M. Willers, V. Zema, J. Zeman, M. Brützam, S. Ganschow
EEur. Phys. J. C manuscript No. (will be inserted by the editor)
Cryogenic characterization of a LiAlO crystal and new results onspin-dependent dark matter interactions with ordinary matter A. H. Abdelhameed , G. Angloher , P. Bauer , A. Bento , E. Bertoldo (cid:63) ,a,1 , R. Breier ,C. Bucci , L. Canonica , A. D’Addabbo , S. Di Lorenzo , A. Erb , F. v. Feilitzsch ,N. Ferreiro Iachellini , S. Fichtinger , D. Fuchs , A. Fuss , V.M. Ghete , A. Garai ,P. Gorla , D. Hauff , M. Ješkovský , J. Jochum , J. Kaizer , M. Kaznacheeva ,A. Kinast , H. Kluck , H. Kraus , A. Langenkämper , M. Mancuso (cid:63) ,b,1 , V. Mokina ,E. Mondragon , M. Olmi , T. Ortmann , C. Pagliarone , V. Palušová ,L. Pattavina , F. Petricca , W. Potzel , P. Povinec , F. Pröbst , F. Reindl , J. Rothe ,K. Schäffner , J. Schieck , V. Schipperges , D. Schmiedmayer , S. Schönert ,C. Schwertner , M. Stahlberg , L. Stodolsky , C. Strandhagen , R. Strauss ,I. Usherov , F. Wagner , M. Willers , V. Zema , J. Zeman (The CRESSTCollaboration)andM. Brützam , S. Ganschow Max-Planck-Institut für Physik, D-80805 München, Germany Comenius University, Faculty of Mathematics, Physics and Informatics, SK-84248 Bratislava, Slovakia INFN, Laboratori Nazionali del Gran Sasso, I-67100 Assergi, Italy Physik-Department and Excellence Cluster Universe, Technische Universität München, D-85748 Garching, Germany Institut für Hochenergiephysik der Österreichischen Akademie der Wissenschaften, A-1050 Wien, Austria Atominstitut, Vienna University of Technology, A-1020 Wien, Austria Eberhard-Karls-Universität Tübingen, D-72076 Tübingen, Germany Department of Physics, University of Oxford, Oxford OX1 3RH, United Kingdom also at: Departamento de Fisica, Universidade de Coimbra, P3004 516 Coimbra, Portugal also at: GSSI-Gran Sasso Science Institute, 67100, L’Aquila, Italy also at: Walther-Meißner-Institut für Tieftemperaturforschung, D-85748 Garching, Germany also at: Dipartimento di Ingegneria Civile e Meccanica, Università degli Studi di Cassino e del Lazio Meridionale, I-03043 Cassino, Italy also at: Chalmers University of Technology, Department of Physics, SE-412 96 Göteborg, Sweden Leibniz-Institut für Kristallzüchtung, D-12489 Berlin, GermanyReceived: date / Accepted: date
Abstract
In this work, a first cryogenic characterization ofa scintillating LiAlO single crystal is presented. The re-sults achieved show that this material holds great potentialas a target for direct dark matter search experiments. Threedifferent detector modules obtained from one crystal grownat the Leibniz-Institut für Kristallzüchtung (IKZ) have beentested to study different properties at cryogenic tempera-tures. Firstly, two 2.8 g twin crystals were used to build dif-ferent detector modules which were operated in an above-ground laboratory at the Max Planck Institute for Physics(MPP) in Munich, Germany. The first detector module wasused to study the scintillation properties of LiAlO at cryo-genic temperatures. The second achieved an energy thre-shold of (213.02 ± (cid:63) Corresponding authors a [email protected] b [email protected] scattering cross section for dark matter particle masses be-tween 350 MeV/c and 1.50 GeV/c . Secondly, a detectormodule with a 373 g LiAlO crystal as the main absorberwas tested in an underground facility at the Laboratori Nazio-nali del Gran Sasso (LNGS): from this measurement it waspossible to determine the radiopurity of the crystal and studythe feasibility of using this material as a neutron flux moni-tor for low-background experiments. Keywords
Dark matter · Cryogenics · Spin-Dependent · Lithium · NeutronsCompiled on May 7, 2020
In the past few decades, great effort has been devoted tothe investigation of dark matter [1]. One path which could a r X i v : . [ phy s i c s . i n s - d e t ] M a y lead to the identification of this elusive particle(s) is thatof direct detection experiments. The goal of most experi-ments in this class is to detect interactions of a dark matterparticle with nuclei of a target material [2]. The CRESST(Cryogenic Rare Event Search with Superconducting Ther-mometers) experiment, like most other direct searches, hasprimarily focused on probing spin-independent dark matter-nucleus interactions [3]. CRESST [4] is currently operat-ing CaWO and Al O crystals at cryogenic temperatures inthe LNGS underground laboratory located below the GranSasso massif in Italy. One advantage of this experiment isthat the technology is not necessarily tied to the target em-ployed; it is relatively easy to change the target crystal andthereby take advantage of the properties of different targetnuclei.In 2019, the CRESST Collaboration published the first re-sults obtained with a lithium-based crystal operated in anabove-ground laboratory [5], showing great potential for darkmatter searches using lithium-containing crystals. Lithiumis an attractive material because it is the lightest elementthat can be tested with the CRESST technology, which con-sists of a scintillating crystal equipped with a tungsten basedTransition Edge Sensor (TES) operated at cryogenic temper-atures. Since CRESST is heavily oriented towards the searchfor dark matter particles with sub-GeV mass, the adoptionof crystals containing light elements can boost this explo-ration due to the favorable kinematics. Furthermore, lithiumis one of the best elements to investigate spin-dependent in-teractions, being mainly constituted of Li (92.41 % naturalabundance [6]), which has J N = / (cid:104) S p (cid:105) = .
497 [7].We do not investigate spin-dependent interactions with Libecause of the current lack of (cid:104) S p / n (cid:105) values in the availableliterature.Another appealing property of these crystals is the possibil-ity to detect the Li(n, α ) H reaction: Li + n → α + H + .
78 MeV . (1)In fact, one of the most challenging sources of backgroundfor a direct dark matter search experiment are neutrons which,like dark matter particles, induce nuclear recoils. Throughthe detection of the above reaction, which shows a distinc-tive signature in a scintillating bolometer [8, 9], it is viableto precisely measure the neutron flux inside the experimen-tal setup and, with the support of Monte Carlo simulations,it might be possible to reconstruct the energy spectrum ofthe neutrons.There are many crystals containing lithium that can be em-ployed, such as Li MoO [8, 10], Li Mg (MoO ) [11],Li WO [12], and LiF [13, 14]. Amongst these, a crystalwith promising properties is LiAlO . First, the CRESST tech-nology for the direct deposition of a TES on the crystalsurface can be applied. Second, LiAlO is a scintillator atroom temperature and shows a light emission with a 340 nm peak [15] at which the CRESST light detectors have a highabsorption [16]. Finally, LiAlO also contains Al (100.0%natural abundance [6]), another interesting element to studyspin-dependent interactions, with J N = / (cid:104) S p (cid:105) = . All the detector modules used in this work are based onLiAlO targets obtained from one single crystal grown atIKZ. The original crystal had a 5 cm diameter and was pro-duced using the Czochralski technique [18]. The primarychallenge for the growth of this kind of material stems fromits high melting temperature of 1780 ◦ C, which entails a stro-ng Li O evaporation. Li O evaporates not only from themelt, but also from the growing crystal: in unfavorable ther-mal conditions, this evaporation is so strong that an Li-freeshell of α -Al O , a few millimeters thick, can form aroundthe LiAlO crystal. To avoid crystal decomposition whichwould arise from this effect, the axial thermal gradient in thesetup must be kept as steep as possible. However, a steeptemperature gradient implies an increased superheating ofthe melt associated with an intensified Li O evaporation fromthe melt itself: this shifts the melt composition from thedesired one towards an Al O -rich melt. Because of non-identical melt and crystal compositions, the crystallizationwith Al O -rich melt involves solute segregation. To a cer-tain extent, this results in the degradation of the grown crys-tals, in the form of a non-uniform macro distribution of theconstituting elements and/or micro-inhomogeneities like se-cond-phase inclusions, mainly LiAl O , due to reduced in-terface stability. There is no perfect set of growth conditionsand parameters which can avoid all the effects of Li O evap-oration: a practical solution will necessitate a compromiseamong crystal perfection, crystal size, and cost of the pro-cess.The crystal used in this work was grown inside a cylindricaliridium crucible of 100 mm diameter in an argon protec-tive atmosphere. The raw materials used for the crystal pro-duction are Li CO and Al O compounds with a 4N/5Npurity. Special attention was paid to the preparation of theraw material in order to prevent Li O losses before the crys-tal growth: these materials were weighed and mixed in astoichiometric ratio and calcinated at temperatures between700 ◦ C and 750 ◦ C in platinum crucibles. The temperatureand duration for this preparation was deduced from thermo-gravimetric measurements of the starting materials [19].
Fig. 1
Close-up of module A . It is possible to see the 2.8 g LiAlO crystal instrumented with the NTD sensor through the CRESST-IIIlight detector. During the crystal growth, the axial temperature gradi-ent was increased step-wise by changing the thermal insula-tion, until the opaque Al O shell disappeared entirely anda shiny transparent crystal was obtained. This was achievedby applying a pulling rate of 1.5 mm/h when growing alongthe (100) direction, together with a crystal rotation between10 and 25 rpm to improve the melt mixing. A more detaileddescription of the growth procedure, crystal defects, and tun-ing of the parameters can be found in [20]. Two (20x10x5) mm crystals with a mass of 2.8 g each werecut from the LiAlO single crystal produced at IKZ, andwere used to assemble two different detector modules.The first crystal was used to assemble module A , a detec-tor module (see Figure 1) designed to characterize LiAlO atcryogenic temperatures. In this case, the crystal was instru-mented with a Neutron Transmutation Doped (NTD) ger-manium thermistor [21] glued to one surface. The crystalwas held in position inside a copper frame by two strings ofPTFE tape. Electrical and thermal connections to the NTDwere provided via 25 µ m diameter gold bond wires. Thetemperature variation of the NTD was obtained by measur-ing the resistance of the thermistor. To do so, a constantbias current was sent through the NTD and the voltage dropof the sensor was measured with a commercial differential GP 12 Allzweck-Epoxidkleber, Gößl + Pfaff
Fig. 2
Close-up of module B , constituted by a 2.8 g LiAlO crystalinstrumented with a TES directly deposited on the surface. voltage amplifier . A CRESST-III light detector (LD) [16]was facing the crystal, held in position by two CuNi clamps;this LD was made of a sapphire plate with a 1 µ m siliconlayer epitaxially grown on one face (Silicon-on-Sapphire)with a TES as thermal sensor deposited on the silicon side.The readout of the light detector is obtained with a commer-cial SQUID system , combined with a CRESST-like detec-tor control system [22]. An Fe X-ray source with an activ-ity of ∼ ∼ T CLD =22 mK.The second crystal constituted the main absorber of mo-dule B (Figure 2), a detector designed to reach a low energythreshold (< 1 keV). The crystal was held in position inside acopper frame by two CuNi clamps. On one face of the crys-tal, a TES with a design similar to that of the light detectorwas deposited. The TES is constituted by a thin strip of tung-sten with two large aluminum pads partially overlapping thetungsten layer. The aluminum pads serve two different pur-poses, as phonon collectors and as bond pads. The bond padsare connected via a pair of 25 µ m aluminum bond wiresthrough which the bias current is injected. The tungsten filmis also connected by a long and thin strip of gold to a thickergold bond pad on which a 25 µ m gold wire is bonded. Thegold strip serves as a weak thermal link between the sensorand the heat bath at ∼
10 mK. On the same surface, but sep-arated from the TES, there is an evaporated heater made of athin strip of gold with two aluminum pads deposited on top. Stanford Research System - SR560 Low-noise voltage preamplifier Applied Physics System model 581 DC SQUID
Fig. 3
Measurement of resistance versus temperature with 5 differentbias currents applied to the TES on module B . At ∼ These pads are bonded with a pair of 25 µ m aluminum bondwires through which a tunable current is injected to maintainthe TES at the desired temperature. The heater is also usedto inject heater pulses to monitor the detector response overtime and for calibration purposes.This TES had a critical temperature T CB (cid:39) T CB is rather high compared to usual transition tempera-tures of CRESST TESs ( ∼
15 mK); this can negatively affectthe performance of the calorimeter, resulting in a higher en-ergy threshold.The two modules have been operated together inside aLeiden Cryogenics dilution refrigerator at the Max PlanckInstitute for Physics in Munich, Germany. The dilution re-frigerator is located in an above-ground laboratory withoutshielding against environmental and cosmic radiation. Themodules have been mechanically and thermally connectedto the coldest point of the dilution refrigerator ( ∼
10 mK). Since there is no literature available on the cryogenic per-formance of LiAlO , the starting point was to study its ba-sic properties. This was done using module A , which al-lowed an initial overview on scintillation, light yield (LY),and Quenching Factors (QFs) [23–25] for different parti-cle interactions inside the crystal.The energy calibration of the light detector was performedusing the peaks originating from the Fe source (Figure 4).After calibration, the baseline resolution of the light detec-tor is σ LDbaseline = ( . ± . ) eV, while the resolution at5.895 keV is σ Fe =(123.9 ± module A , an AmBe neutron sourcewas installed at a distance of ∼
50 cm from the center of the The Quenching Factor for the interaction of an arbitrary particle x isdefined as: QF x = LY x / LY γ dilution refrigerator. For the energy calibration of the NTDthe neutron capture peak appearing at 4780 keV (Equation 1)was used, where the energy resolution is σ capture =(19.96 ± form one band starting from zero energy andwith a light yield of (1.180 ± = Energy LD Energy
NTD (2)Neutrons scattering within the crystal exhibit a band star-ting from zero energy as well, but with a much reducedlight yield (0.284 ± ∼ ± β / γ band and the neutron band,the neutron capture by Li appears. Assuming a linear lightemission up to this energy, the QF for the neutron capture is0.599.The separation between the β / γ band and the neutron bandstarts to become evident at ∼
170 keV; thus, in the energy re-gion of interest for dark matter search ( ∼ Li(n, α ) H reaction. Thisfamily of events has not been observed in the LNGS mea-surement (see Section 7), but the neutron source employedin that case had an extremely reduced activity with respectto the above-ground measurement.
As explained in Section 4, module B was designed to studyspin-dependent interactions of low-mass dark matter parti-cles with nuclei of a LiAlO crystal in a cryogenic measure-ment. A low threshold is a key parameter to reach this goal,due to the steep increase of expected dark matter recoils atlower energies. For this reason the TES was directly evap-orated onto the LiAlO surface, applying, for the first time,the CRESST technology on a crystal containing lithium . Fig. 4
Energy spectrum of events registered by the CRESST-III lightdetector in the energy region of X-ray emission by the Fe source.Two peaks are visible: one at 5.895 keV, resulting from the sum of K α and K α lines, and one at 6.490 keV, resulting from the sum of K β andK β lines. The fit function (red solid line) consists of the sum of twoGaussian functions ( µ and µ are the expected values, σ and σ thestandard deviations) plus a constant factor c to account for the flat back-ground. In principle, σ should not be lower than σ , but we attributethis anomaly to the presence of an energy loss in the left shoulder ofthe 5.895 keV peak. The 5.895 keV peak is used to obtain the energycalibration of the light detector. Fig. 5
Energy measured by the CRESST-III light detector versus en-ergy measured by the NTD for each event registered by module A in thepresence of an AmBe neutron source during 9.44 hours of data collec-tion. Two bands starting from zero appear: the one with the higher lightemission is constituted by β / γ events interacting inside the LiAlO crystal, while the one with lower light emission is caused by the scat-tering of neutrons within the crystal. At 4780 keV a different family ofevents appears, due to the neutron capture of Li. In the vicinity of theneutron capture there is an additional family of events, with a slightlyhigher energy. This family currently is of unknown origin and the mod-eling of the anomalously high light yield is particularly challenging.
A total of 22.2 hours of data without any source ("back-ground data") were collected for module B using a contin-uous DAQ with a sampling rate of 25 kHz. The events weretriggered with a dedicated software based on the optimumfilter [26].The energy calibration is implemented using the5.895 keV peak from the Fe source, similar to the one usedfor the LD of module A . During the run, heater pulses offour different known amplitudes were injected to interpolatethe energy calibration in the whole energy region of inter-est. The peaks corresponding to the heater pulses are identi-fied in the dataset: each peak is fit with a Gaussian functionwhich returns the mean and the error of the mean. After-wards, the amplitude of heater pulses (A injected ) versus the
Fig. 6
Injected amplitude of 4 different heater pulses versus amplituderegistered by the TES (black crosses) expressed in arbitrary units. Eachamplitude registered by the TES with relative error are obtained fromthe peaks appearing in the raw spectrum via a Gaussian fit. The fourpoints are fit with Equation 3, which is used for the energy calibrationof the detector. amplitude measured by the TES (A phonon ) is plotted and thefollowing function is fit to the data points:A phonon = p0 · p1 · A injected · IR L + p1 · A injected (3)where p0 is the gain of the SQUID, I is the bias current ofthe TES, R L is the load resistor, and p1 is a coefficient whichtranslates the temperature change of the TES, induced bythe heat pulse, into a variation of the TES resistance. Equa-tion 3 is derived from the circuit scheme used to read out theTES [22]. For this measurement I=9 µ A, while R L =40 m Ω ;p0 and p1 are the free parameters of the fit. Finally, the meanvalue registered by the TES A phonon =(2379.2 ± phonon to energy via Equation 3. This description assumes that theTES resistivity changes linearly with the temperature in theenergy interval considered (0-6 keV). With this method anaccurate energy calibration (Figure 6) was obtained takinginto account the intrinsic non-linearity of the read-out sche-me. The baseline resolution is σ Bbaseline =(39.75 ± E BT =(213.02 ± counts/(keV · kg · day), two orders ofmagnitude lower than the observed event rate in the 1-5 keVrange.Figure 7 shows the calibrated energy spectrum of the22.2 hours background measurement. The X-ray peaks from Fe decay clearly emerge. A moderate rise of events below300 eV is also evident. The energy resolution at 5.895 keVis equal to σ K α =(191.5 ± σ HP =(41.6 ± σ HP =(57.0 ± is of the order of 2 · counts/(keV · kg · day), similar to theone observed in [28]. This high value is expected, since thedetector is operated in an above-ground laboratory withoutany shielding or veto systems.From the measured spectrum, dark matter exclusion limitsfor spin-dependent interactions are calculated. The energyregion of interest ranges from E BT to 4000 eV without ap-plying any cut to the particle events registered by the detec-tor. All the events with energies above 4000 eV contributeto the dead time, reducing the exposure from 22.2 hours to17.2 hours, corresponding to a total exposure of 2.01 · − kg · day, with an exposure for Li of 1.95 · − kg · day and anexposure for Al of 8.22 · − kg · day. The exclusion lim-its are calculated using Yellin’s optimal interval method [29,30] and are shown in Figure 8. The baseline resolution ofthe detector σ Bbaseline and the energy threshold E BT are takeninto account to evaluate the minimum value of dark mat-ter mass for which it is possible to draw exclusion limits.These limits are valid for both proton-only interactions andfor neutron-only interactions, as discussed in the theoreti-cal framework presented in [5]. The calculation of the ex-clusion limits adopts the standard dark matter halo model,which assumes a Maxwellian velocity distribution and a lo-cal dark matter density of ρ DM = . ( GeV / c ) / cm [31].Furthermore, v esc =
544 km / s is assumed for the galacticescape velocity [32] and v (cid:12) =
220 km / s for the solar orbitvelocity [33]. For the proton-only exclusion limits (cid:104) S p (cid:105) = . Li and (cid:104) S p (cid:105) = . Al are used, whilefor the neutron-only exclusion limits (cid:104) S n (cid:105) = . Liand (cid:104) S n (cid:105) = . Al [7, 17] are used. The data anal-ysis efficiency is computed generating a known flat energyspectrum of events. These events are created by superim-posing the ideal detector response on recorded data and thenprocessed with the same analysis chain used for the realdata. The fraction of surviving events over the total simu-lated events at each energy bin represents the data analysisefficiency. Since the determination of the amplitude and thetriggering are done in one step by the optimum filter and nofurther data selection criteria applied, in this case the dataanalysis efficiency is equivalent to the trigger efficiency.
After the successful tests at MPP, the bulk of the originalLiAlO crystal sample was mechanically polished obtaininga 373 g crystal. Such crystal size is ideal to study the crys-tal radiopurity and to assess the feasibility of using LiAlO crystal as a monitor for the neutron flux in a shielded ex-perimental setup. For this reason, this crystal was used ina new detector module, module C , which was installed inthe MPP Test-Cryostat facility located in the undergroundlaboratory of Laboratori Nazionali del Gran Sasso (LNGS),under 3600 m water equivalent overburden to shield against Fig. 7
Energy spectrum collected during 22.2 hours of backgroundmeasurement with module B without any cut applied to the data set.In black: events induced by injected heater pulses. In light blue: par-ticle events only. At 5.895 keV the peak caused by the X-ray emis-sion of Fe decay appears; the energy resolution at 5.895 keV isequal to σ K α =(191.5 ± σ HP =(41.6 ± σ HP =(57.0 ± Fig. 8 Top:
Exclusion limits set by various direct detection experi-ments for spin-dependent interactions of dark matter particles withneutrons. The result obtained from module B data with Li+ Alis shown in solid red. The first result obtained by CRESST using Li is plotted in dotted red [5], while the result obtained with Oin CRESST-III is shown in dashed red [4]. For comparison, limitsfrom other experiments are also shown: EDELWEISS [34] and CDM-Slite [35] using Ge, LUX [36] and XENON1T (Migdal effect) [37]using
Xe+
Xe.
Bottom:
The same, but for spin-dependent in-teractions of dark matter particles with protons. The result obtainedfrom module B data with Li+ Al is shown in solid red. The firstresult obtained by CRESST using Li is plotted in dotted red [5].Additionally, limits from other experiments are also shown: CDM-Slite with Ge [35]; PICO with F [38]; XENON1T (Migdal effect)with
Xe+
Xe [37]; Collar with H [39]. Finally, a constraint fromBorexino data derived in [40] is shown in dotted black. cosmic radiation [41].As visible in Figure 9, an NTD [21], a (5 × ×
1) mm Sicarrier with a thin gold stripe heater deposited on it, anda CaWO carrier crystal on which a CRESST-II TES hadbeen evaporated [42] are both glued to the top surface ofthe LiAlO crystal. The NTD and the CRESST-II TES areboth being used as phonon sensors. This choice is motivatedby the fact that the NTD has a higher dynamic range thanthe TES, while the TES can generally achieve a lower en-ergy threshold than the NTD. Therefore, with this detectormodule it is possible to study both the low energy part of thespectrum ( ∼ ∼
10 MeV).This allows the potential setting of competitive limits forspin-dependent dark matter search and the detailed study ofthe neutron capture by Li during the same measurement.The crystal was held in position inside a copper holder us-ing three PTFE clamps on the bottom and three on the top.Reflective foil was used to surround the crystal, in order tomaximize light collection. A CRESST-II light detector [43]was facing the top surface of the LiAlO crystal, completingthe detector module.The MPP Test-Cryostat facility is located in the corridorconnecting Hall A and Hall B of LNGS. The model of di-lution refrigerator installed in this facility is the same as theone used for the above-ground measurement at MPP. Thedetector module operated in this dilution refrigerator em-ploys the same kind of wiring, NTD readout, and TES read-out as in the previous above-ground measurement. The detector operation of module C at LNGS was dividedinto two parts: one focused on the efficacy of measuringneutrons, the other on measuring the radioactive impuritiesin the crystal. For these type of measurements, the data col-lected with the NTD (that does not saturate in the energy re-gion of interest) and the CRESST-II light detector have beenanalyzed. The CRESST-II TES was also simultaneously op-erated as a phonon sensor to study the low-energy part of thespectrum (<1 MeV).At the beginning of the run an AmBe neutron source emit-ting ∼
10 neutrons/s was installed at a distance of ∼
60 cmfrom the center of the dilution refrigerator and 13.1 hoursof data were collected. To ensure the stability of the NTDsensor, heater pulses with seven different amplitudes wereinjected, two of which were close to the energy region ofinterest for the neutron capture by Li (Equation 1). The de-tector response was calibrated using these heater pulses andthe 4.78 MeV peak corresponding to the neutron capture.After calibration, the energy resolution at 4780 keV is GP 12 Allzweck-Epoxidkleber TM Enhanced Specular Reflector
Fig. 9
Detector module C was operated at LNGS. A 373 g crystal is in-strumented with two phonon sensors glued on the top surface: an NTDand a CRESST-II TES. On the same surface there is a glued heaterwhich ensures the stability of the detector operation. The crystal is sur-rounded by reflective foil and a CRESST-II light detector is facing thetop surface of LiAlO . Fig. 10
QF versus the energy registered by the NTD sensor for 13.1hours of effective live time in the presence of a weak AmBe source. Forenergies (cid:46) β / γ band which was usedto normalize the QF. At energies (cid:38) α decays and one prominent line at4.78 MeV corresponding to the neutron capture of Li can be seen. σ capture =(18.3 ± α decays. These two classes ofevents are used to build two histograms (Figure 11): neu-tron capture events are selected from an energy interval of Fig. 11
Distribution of QF values for events originated by α decayswith a mean value of 0.38 ± ± ± σ capture centered around 4780 keV, while all other eventsabove 4 MeV are considered alpha events. It is possible tosee that the two distributions are partially overlapping. How-ever, even with a simple cut on the QF value one can ex-clude the vast majority of unwanted α decay events: if onlyevents with a QF>0.44 (the mean value of the neutron cap-ture distribution) are accepted, 93.3% of α events are cutwhile halving the detection efficiency for the neutron cap-ture. The efficiency in discarding α events can then also beconsiderably increased defining a cut on the energy detectedby the NTD phonon sensor: clearly, this cut is more effectivethe higher the energy resolution of the NTD.In a low-background environment only a few neutron eventsare expected. suggests that there may not be enough eventsto build a distribution based on the QF values. However, it ispossible to perform a neutron calibration and then, based onthe data, define a region where neutron events are expectedduring the background data campaign. From the total num-ber of events inside this region, it is then feasible to quotea neutron flux value (or upper limit) with the respective un-certainty. This result demonstrates the high potential for adetector of this kind in monitoring small neutron fluxes di-rectly inside an underground experimental setup.The long term goal for CRESST is to directly detect neu-trons inside the experimental setup using a specifically de-signed detector based on a lithium-containing crystal, therebyproviding a relevant input to the background model of theexperiment. From this data, using dedicated Monte Carlosimulations, the total neutron flux (or an upper limit) can beassessed while also possibly reconstructing the energy spec-trum of the incoming neutrons. The measurement presentedin this work is a first step in this direction.After the neutron measurement, the AmBe source wasremoved to measure the radiopurity of the crystal. In thiscase, a 58.4 hours background measurement was carried out.After stability and data quality cuts, the effective measuringtime is 35.6 hours. In this measurement it was not possi- Fig. 12
Energy spectrum registered by the NTD during a backgroundmeasurement of 35.6 hours effective time. From this spectrum at least4 different families of α decays in the 4-7 MeV region can be easilydistinguished and above 7 MeV additional events, likely due to Bi-
Po decays, appear. ble to use the neutron capture peak to calibrate the NTD re-sponse, but the heater pulses that were previously calibratedare used instead. In Figure 12, the energy spectrum mea-sured by the NTD is shown. From this spectrum, at least 4different families of α decays in the 4-7 MeV region can beeasily distinguished. At even higher energies some eventsare recorded, likely due to Bi-
Po decays. The totalnumber of events above 3 MeV is 483: this means an upperbound on the total alpha activity of (10.1 ± crystal. The radiopurity of this crystalis ∼ crystalproduced within the CRESST Collaboration (TUM40) [44],but in line with standard commercial CaWO crystals. Thegoal for the future is to drastically improve the radiopurityof LiAlO , starting from a careful selection of the raw mate-rials used for the crystal growth. Additionally, a 20.8 hourscalibration using a Am gamma source installed close tothe outer shield of the dilution refrigerator was carried outto test the performance of theCRESST-II TES. During the calibration and the backgroundmeasurement, heater pulses with nine different amplitudeswere injected. The 59.54 keV gamma peak from the
Amsource used for the energy calibration has a resolution of σ Am =(3.044 ± ± crystal: this is expected dueto the large increase in mass as showed by the scaling lawdescribed in [45]. This work details the results of three different detectors, allof which employ a LiAlO target crystal, a material thathas never been employed in cryogenic experiments thus far.The cryogenic properties of the material were tested in anabove-ground laboratory with a 2.8 g crystal and new lim-its on spin-dependent dark matter interactions are set witha crystal instrumented with a TES deposited on LiAlO . Alarge-size detector with a mass of 373 g was operated in anunderground cryogenic facility at LNGS in the presence ofa weak neutron source, in order to assess the feasibility tomonitor the neutron flux directly inside cryogenic setups.The results presented in this work demonstrate the high po-tential of LiAlO crystals as cryogenic detectors in the fieldof low-background applications and contribute to the ongo-ing search for dark matter. Acknowledgements
This work is supported through the DFG bySFB1258 and the Origins Cluster, and by the BMBF05A17WO4.
References
1. G. Arcadi et al. , The European Physical Journal C , 203 (2018),1703.07364.2. T. Marrodán Undagoitia and L. Rauch, Journal of Physics G ,013001 (2016), 1509.08767.3. CRESST, G. Angloher et al. , The European Physical Journal C , 25 (2016), 1509.01515.4. CRESST, A. H. Abdelhameed et al. , Physical Review D ,102002 (2019), 1904.00498.5. CRESST, A. H. Abdelhameed et al. , The European Physical Jour-nal C , 630 (2019), 1902.07587.6. J. Meija et al. , Pure and Applied Chemistry , 293 (2016).7. A. Pacheco and D. Strottman, Physical Review D , 2131 (1989).8. O. Barinova et al. , Nuclear Instruments and Methods in PhysicsResearch Section A , 54 (2010).9. L. Cardani et al. , Journal of Instrumentation , P10002 (2013),1307.0134.10. N. Casali et al. , Journal of Physics G , 075101 (2014),1311.2834.11. M. Martinez et al. , Journal of Physics: Conference Series ,012025 (2012).12. A. A. et al., Nuclear Instruments and Methods in Physics ResearchSection A , 162784 (2020).13. K. Miuchi et al. , Astroparticle Physics , 135 (2003).14. F. A. Danevich et al. , Nuclear Instruments and Methods , 89(2018), 1802.01888.15. T. Yanagida et al. , Journal of the Physical Society of Japan ,094201 (2017).16. J. Rothe et al. , Journal of Low Temperature Physics (2018).17. J. Engel, M. T. Ressell, I. S. Towner, and W. E. Ormand, PhysicalReview C , 2216 (1995), 9504322.18. B. Cockayne and B. Lent, Journal of Crystal Growth , 546(1981).19. R. Bertram and D. Klimm, Thermochimica Acta , 189 (2004).20. B. Veliˇckov et al. , Journal of Crystal Growth , 214 (2008).21. N. Wang, F. C. Wellstood, B. Sadoulet, E. E. Haller, and J. Bee-man, Physical Review B , 3761 (1990).22. G. Angloher et al. , Astroparticle Physics , 270 (2009). 23. V. I. Tretyak, Astroparticle Physics , 40 (2010), 0911.3041.24. A. G. B. A. e. a. Strauss, R., The European Physical Journal C (2014).25. C. C. D. G. e. a. Strauss, R., Journal of Low Temperature Physics , 905-910 (2014).26. E. Gatti and P. F. Manfredi, La Rivista del Nuovo Cimento , 1(1986).27. M. Mancuso et al. , Nuclear Instruments and Methods in PhysicsResearch Section A , 492 (2019), 1711.11459.28. CRESST, G. Angloher et al. , The European Physical Journal C , 637 (2017), 1707.06749.29. S. Yellin, Physical Review D , 032005 (2002), 0203002.30. S. Yellin, (2007), 0709.2701.31. P. Salucci, F. Nesti, G. Gentile, and C. F. Martins, Astronomy andAstrophysics , A83 (2010), 1003.3101.32. M. C. Smith et al. , Monthly Notices of the Royal AstronomicalSociety , 755 (2007), 0611671.33. D. Kerr, F. J.; Lynden-Bell, Monthly Notices of the Royal Astro-nomical Society , 1023 (1986).34. EDELWEISS, E. Armengaud et al. , Physical Review D ,082003 (2019), 1901.03588.35. SuperCDMS, R. Agnese et al. , Physical Review D , 022002(2018), 1707.01632.36. LUX, D. S. Akerib et al. , Physical Review Letters , 251302(2017), 1705.03380.37. XENON, E. Aprile et al. , Physical Review Letters , 241803(2019), 1907.12771.38. PICO, C. Amole et al. , Physical Review D , 022001 (2019),1902.04031.39. J. I. Collar, Physical Review D , 023005 (2018), 1805.02646.40. T. Bringmann and M. Pospelov, Physical Review Letters ,171801 (2019), 1810.10543.41. MACRO, M. Ambrosio et al. , Physical Review D , 3793 (1995).42. CRESST, R. Strauss et al. , The European Physical Journal C ,352 (2015), 1410.1753.43. G. Angloher et al. , Journal of Low Temperature Physics , 323(2016).44. CRESST, R. Strauss et al. , Journal of Cosmology and Astroparti-cle Physics , 030 (2015), 1410.4188.45. R. Strauss et al. , The European Physical Journal C77