Instruments and Experimental Techniques, Vol. 46, No. 2, 2003, pp. 153–160. Translated from Pribory i Tekhnika Eksperimenta, No. 2, 2003, pp. 16–23. Original Russian Text Copyright © 2003 by Ashitkov, Barabash, Belogurov, Carugno, Konovalov, Massera, Puglierin, Saakyan, Stekhanov, Yumatov.
NUCLEAR EXPERIMENTAL TECHNIQUE
A Liquid-Argon Ionization Detector for the Study of Double b Decay V. D. Ashitkov*, A. S. Barabash*, S. G. Belogurov*, G. Carugno**, S. I. Konovalov*, F. Massera***, G. Puglierin**, R. R. Saakyan*, ****, V. N. Stekhanov*, and V. I. Umatov* * Institute of Theoretical and Experimental Physics, ul. Bol’shaya Cheremushkinskaya 25, Moscow, 117259 Russia e-mail:
[email protected] ** Dipartimento di Fisica e INFN, Universita di Padova, Padova, Italy *** Istituto Nazionale di Fisica Nucleare (INFN), Sezione di Bologna, Bologna, Italy **** Laboratori Nazionali del Gran Sasso dell’INFN, L’Aquila, Italy Received September 11, 2002
Abstract—A multisection liquid-argon ionization detector has been developed by the DBA collaboration to study double β decay of 100Mo. The experiment has been carried out in the Gran Sasso underground laboratory in Italy. The detector design and main characteristics are described. The limits on the 42Ar and 222Rn concentrations in liquid Ar are ≤ 4.3 · 10–21 g/g and ≤1.2 · 10–3 Bq/kg, respectively.
INTRODUCTION The current interest in 2β(0ν) decay has been aroused by the fact that the existence of this process is closely related to the following fundamental aspects of elementary-particle physics [1–3]: the nonconservation of the lepton number, the existence of neutrino mass and its nature; the existence of right-handed currents in electroweak interaction; the existence of a Majoron; the Higgs-sector structure; and the existences of leptoquarks, a heavy sterile neutrino, and a composite neutrino. All of these questions lie beyond the scope of the standard model of electroweak interaction; that is why the detection of 2β(0ν) decay will give rise to “new physics.” The primary interest in this process is, of course, associated with the problem of neutrino mass: the detection of 2β(0ν) decay will indicate that the rest mass of even one neutrino is nonzero, and that this mass is a Majorana-type mass. Currently, only the lower bounds on the half-lives of different nuclei have been measured by the neutrinoless channel (T1/20ν). These bounds are used to find the limitations on the Majorana neutrino mass, parameters of the admixture of righthanded currents, coupling constant of a Majoron with a neutrino, etc. However, the uncertainties in computing the nuclear matrix elements make it impossible to achieve sufficiently reliable limitations on these fundamental quantities. Hence, the detection of two-neutrino double β decay (2β(2ν)) gains particular importance. A liquid-argon detector [4, 5] was designed to study double β decay. Nuclei of the 100åÓ isotope were selected for the study because of the high energy of its ββ transition (E2β = 3033 keV). In 1996, the experimen-
tal setup was installed in the Gran Sasso underground laboratory (Italy). In this work we describe the equipment used in the experiment. A special emphasis is laid on the structural features, calibration procedure, and operation of the setup components. EXPERIMENTAL SETUP The experiment was carried out in the Gran Sasso underground laboratory located at a depth of 3500 m w.e. (meters of the water equivalent). The experimental setup (Fig. 1) comprised a liquid-argon ionization detector enclosed in a passive lead shielding, a gas-supply system, a radon-protection system based on liquidnitrogen vapor, and a mechanical crane used for assembling and dismantling the detector. The equipment was mounted on a special concrete platform with dimensions of 4 × 6 × 0.6 m. For the suppression of mechanical vibrations, the platform was set up on a rubber sheet. The platform carried a special “house,” which reliably prevented the impact of the underground moisture on the equipment. The data-acquisition and readout system for the liquid-argon detector was placed in the other “house” located 2 m away from the platform. In order to suppress the effect of radon from the ambient air, the nitrogen vapor from the detector thermostat was heated to 20°C in an electrical heater and conveyed inside the passive shielding. LIQUID-ARGON IONIZATION DETECTOR Titanium and its alloys were the main structural materials of the detector. All the insulators of its electrode system were made of fluoroplastic. A diagram of
0020-4412/03/4602-0153$25.00 © 2003 åAIK “Nauka /Interperiodica”
154
ASHITKOV et al. Gas-supply system
Passive (neutron) shielding Liquidnitrogen supply system
Liquid ionization detector
Passive shielding (lead)
Radonprotection system based on nitrogen vapor
Data readout and acquisition system Fig. 1. Diagram of the experimental setup.
Ar (gas)
Vacuum
Liquid nitrogen
1 2 3 4
5
Fig. 2. Diagram of the liquid-argon ionization detector: (1) chamber filled with liquid argon; (2) thermostat volume filled with liquid nitrogen; (3) vacuum volume of the thermostat; (4) space filled with nitrogen vapor; and (5) system of heaters.
the detector is shown in Fig. 2. The detector was cooled with liquid nitrogen that was fed into a nitrogen space 2 from a 4000-l Dewar vessel. The liquid nitrogen was conveyed through a pipe enclosed in a vacuum heat insulation to reduce the liquid-nitrogen flow rate, which was particularly important in long-term experimental sessions. The nitrogen volume 2 was heat-insulated from the environment by a vacuum volume 3. The inner wall of the volume was coated with 12 layers of 10-µmthick aluminized Mylar. The flow rate of liquid nitrogen in the measurement mode was reduced to 6 l/h. The nitrogen volume 2 was connected with a chamber 4, in which a vessel 1 with liquid argon was housed, through a tube to equalize the nitrogen-vapor pressure in these spaces. Heaters 5 located on the outside of the vessel 1 regulated the liquid-argon temperature. The electrode system that formed the sensitive volume of the detector was placed in the vessel 1 with a 40-cm i.d. and a height of 70 cm. The active detection part of the detector was composed of similar measuring sections (Fig. 3). Each section consisted of two combined flat ionization chambers with shielding grids 2 and a common cathode 3. A foil made of the isotope under investigation (molybdenum) was inserted into annular frame cathodes. The sensitive volume was 30 cm in diameter and 56 cm high. The detector comprised 14 cathodes, 15 anodes, and 28 screening grids. The grid-to-anode and grid-to-cathode distances were 5.5 and 14.5 mm, respectively. Grids were used as anodes 4. The screening grids and anodes were made of 110-µm-diameter nichrome wires wound with a 1-mm spacing. The top and bottom
INSTRUMENTS AND EXPERIMENTAL TECHNIQUES
Vol. 46
No. 2
2003
A LIQUID-ARGON IONIZATION DETECTOR
4
1
1
155
3
4
2
2
2 4
3
Ar
Ionization detector
Ar
22 bottles
Ar
1
Fig. 3. Arrangement of the electrodes in the liquid-argon ionization detector: (1) titanium anode (continuous); (2) screening grid; (3) molybdenum-foil cathode; and (4) grid anode.
anodes 1 of the electrode system were solid titanium discs. The high voltage applied to the cathodes and the grids was –4.8 and –2 kV, respectively. The charge-sensitive preamplifiers converted the charge produced in each anode into an ionization pulse. ISOTOPES The detector cathodes were made of molybdenum foil ~50 mg/cm2 thick. The four cathodes of enriched molybdenum (98.4% 100Mo) were first investigated in three sessions of 202-, 238-, and 313-h measurements. For the radiation background to be suppressed, these cathodes were mounted at the center of the electrode system and alternated with the cathodes of natural molybdenum (containing 9.6% 100Mo). The concentration of radioactive impurities in the molybdenum-foil samples was under 0.015 Bq/kg for 214Bi, 0.0015 Bq/kg for 208Tl, and 0.04 Bq/kg for 234mPa. The total mass of 100Mo under investigation was 138.7 g. Later on, the 100Mo mass in the detector was increased to 306 g (eight cathodes of 100Mo). The alternation of natural- and enriched-molybdenum cathodes best suited the study of the 2νββ mode by the differential method. The total measurement time with eight 100Mo cathodes was 2706 h. GAS-SUPPLY SYSTEM The gas-supply system (Fig. 4) comprised 22 40-l stainless-steel bottles with gaseous argon under a presINSTRUMENTS AND EXPERIMENTAL TECHNIQUES
Fig. 4. Scheme of the gas-supply system: (1) gas-purity control system; (2) gas bottle (2 l in volume); (3) purification system; and (4) mechanical filters for rough gas purification.
sure of about 150 atm, a gas-purification system 3, and a gas-purity control system 1. Six of these bottles were housed in the stainless-steel Dewar vessels; after the measurements, they were used as a cryogenic pump to transfer the liquid argon from the chamber to the bottles. The system 3 for the elimination of electronegative impurities from the gas was based on a titanium sponge due to the low radiation background of this getter. For example, the purification system based on Oxisorb sorbent or a molecular sieve releases 100–1000 times more 222Rn, because the 238U content of their materials (approximately 10–6 to 10–7 g/g) exceeds that of titanium by the same factor. After passing the gaseous argon through the titanium sponge in the reactor once at a temperature of 850°ë, the concentration of electronegative impurities was under 1.9 · 10–9 equiv. é2 [6, 7]. The impurity content of argon was monitored with a two-phase detector 1 for electronegative impurities [7, 8]. The reactor vessel made of high-temperature steel was a 50-mm-diameter and 300-mm-high cylinder filled with a titanium sponge. The argon flow rate during the purification was ~0.6 m3/h. PASSIVE SHIELDING The passive shielding of 15-cm-thick lead was set upon a concrete platform. Note that the measured radon concentration in the ambient air was ~200 Bq/m3 (for comparison, the radon concentration in the air of other halls of the Gran Sasso underground laboratory was ~40 Bq/m3). For the radon effect to be suppressed, the Vol. 46
No. 2
2003
156
ASHITKOV et al. Amplifier ADCs
Uc Ug
Electronic units
Anode Grid Cathode
Preamplifier Preamplifier Shaping
Test
Computer
1
.. .
OR
electronics. The programs for initial data analysis and visualization were written in the same software environment. The data-selection trigger was generated if the signal in any measuring channel exceeded a 700-keV threshold. In this case, the digitized signals of all the channels were recorded on a magnetic tape. Such a recording circuit was convenient for the pulse-shape analysis, as well as for the identification and rejection of possible background signals.
Trigger
16
Fig. 5. Diagram of the electronics and the data-readout system.
nitrogen vapor from the detector-cooling system was directed into the internal space of the passive shielding. With the same aim in mind, hollow organic-glass blocks were placed near the detector inside the passive shielding (in order to reduce its internal space). As a result, the background due to 222Rn near the detector (inside the passive shielding) was suppressed down to a level of
