- Email: [email protected]
Rejection of surface background in thermal detectors: The ABSuRD project
Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Rejection of surface background in thermal detectors: The ABSuRD project L. Canonica a,n, M. Biassoni b, C. Brofferio b, C. Bucci a, S. Calvano a, M.L. Di Vacri a, J. Goett c, P. Gorla a, M. Pavan b, M. Yeh d a
INFN, Laboratori Nazionali del Gran Sasso, Assergi, AQ, Italy Università di Milano Bicocca e INFN Sezione di Milano Bicocca, Milano, Italy c Los Alamos National Laboratory, Los Alamos, NM, USA d Brookhaven National Laboratory, Upton, NY, USA b
art ic l e i nf o
a b s t r a c t
Available online 25 May 2013 Keywords: Thermal detectors Rare events searches Scintillation Surface background
Thermal detectors have recently achieved a leading role in the fields of Neutrinoless Double Beta Decay and Dark Matter searches thanks to their excellent energy resolution and to the wide choice of absorber materials. In these fields the background coming from surface contaminations is frequently dominant. ABSuRD (A Background Surface Rejection Detector) is a scintillation-based approach for tagging this type of background. We discuss the innovative application of this technique in non-scintillating bolometric detectors which will allow for a more favorable signal to background ratio. & 2013 Elsevier B.V. All rights reserved.
1. Introduction
2. Rejection of surface background
A bolometer is a particle detector that measures the energy deposited by incident radiation based on the temperature change of the detector. It generically consists of three parts: an absorber, a temperature sensor and a weak thermal link connecting the absorber to a thermal reservoir. Due to the excellent energy resolution that bolometers can achieve (of the order of a few per mil) and due to a vast choice of absorber materials possible, limited only by the heat capacity at the operating temperature, thermal detectors are presently successfully employed in many dark matter (DM) [1–4] and neutrinoless double beta decay ð0νDBDÞ searches [5]. A search for extremely rare events requires the removal of any spurious signal that can mimic the rare event (e.g. radioactive source or cosmic rays). The residual backgrounds counting rate measures the sensitivity of the experiment. As a consequence, one of the main issues of present and future 0νDBD and DM experiments is the identification, control and reduction of the radioactive background. In this work we propose a scintillation-based approach for rejecting surface background and discuss the innovative application of this technique in non-scintillating bolometric detectors.
It has been demonstrated that the main limitation to the sensitivity of bolometric experiments to rare events physics arises from surface alpha contaminations (i.e. in the copper structures holding the detectors), namely degraded α interactions in the absorber [5]. In a bolometric 0νDBD decay experiment, radioactive contaminations close to the absorber surfaces can cause counts in the energy region where the signal is expected. An α particle generated near the surface of a detector component will release only a part of its energy before escaping from the material. The α will then eventually hit another component of the detector. If one of the two components is non-active (e.g. the holding structure) the total α energy cannot be reconstructed, generating a continuum background at all energies. A purely bolometric detector, measuring only the heat signal, cannot discriminate radiation type or surface/bulk events. The discrimination between surface and bulk events and the identification of the particle type interactions (α vs. β=γ) is fundamental for reducing as much as possible surface background sources, and thus increasing the experimental sensitivity. In this work we propose an innovative approach for rejecting surface background. The idea of the ABSuRD project is to encapsulate a purely thermal detector with a scintillating foil and to add a bolometric light detector to detect the light produced by the interaction of degraded α particles with the scintillating foil (see Fig. 1). A surface α particle releasing part of its energy on the crystal and part on the scintillating foil can be tagged by analyzing the coincidence signal of heat (in the absorber) and light (in the light
n
Corresponding author. Tel.: +39 086 243 7431. E-mail address: [email protected] (L. Canonica).
0168-9002/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nima.2013.05.114
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
287
Light detector
Fig. 1. Diagram of the proposed method.
detector) and rejected as background. Extremely low energy thresholds ð o1 keVÞ are needed for the bolometric light detector, in order to be able to detect small light signals produced by degraded α's. As an example, an α particle of 5.3 MeV (generated by the decay of 210Po) produces about 10 keV of photons in commercial scintillating foil at room temperature. Unfortunately, plastic scintillators have an extremely non-linear response to α particles [8,9] and the amount of light produced by a degraded α particle of 1.5 MeV is about an order of magnitude smaller than the one produced by a 5.3 MeV α. Detecting such a small amount of light represents a serious challenge for most of the currently used light detectors at mK temperature. In order to detect low energy α particles it is therefore desirable to increase the amount of scintillation light. This can be achieved by simply superposing a thin plastic scintillator foil over a reflecting foil.
3. Demonstrator run In order to study the feasibility of this method, a prototype detector has been operated at low temperature. The main goal of the run is to demonstrate, with a non-optimized scintillating foil and light detector, the potential for tagging and rejecting α particles generated on the surface of the absorber crystal. The experimental setup was hosted in a dilution refrigerator placed in the Hall C of National Laboratory of INFN at Gran Sasso, operating at 10 mK. The studied bolometer is a 3 3 6 cm3 TeO2 crystal, held by means of PTFE clamps to a structure of Oxygen Free High Conductivity (OFHC) copper frames and columns, which serve both as mechanical support and heat sink. The crystal is surrounded by a commercial scintillating foil (Saint-Gobain BC400 series) and by a reflecting foil (3 M VM2002), to increase the light collection efficiency. The crystal is provided with a Neutron Transmutation Doped Ge thermistor (NTD), to convert the phonon signal into a detectable voltage pulse. To detect the scintillation light, the top of the crystal faces a germanium bolometric light detector also equipped with an NTD thermistor [7]. It consists of a 50 mm diameter, 300 μm thick pure Ge crystal absorber facing the surface of the crystal. For a picture of the complete set-up, see Fig. 2. A liquid solution containing an α source of 147Sm is deposited on one of the surfaces of the crystal, facing the scintillating and
Fig. 2. The 3 3 6 cm3 TeO2 is enclosed by four copper walls. It is surrounded by the scintillating foil and by the reflecting foil. The top face of the crystal is visible to the Germanium Light Detector.
reflective foils. For an α decay (Q-value ¼2.3 MeV) occurring on the crystal surface, the α particle can escape from the crystal, releasing in the absorber only the energy related to the recoiling nucleus. The remaining part of the energy is released into the scintillating foil, giving rise to a light signal detected by the light detector. Both heat and light pulses produced by particle interactions are amplified and digitized using an 18-bit NI-6284 PXI ADC unit. The trigger is software generated on each bolometer. A complete description of the experimental setup can be found in Ref. [5] and references therein for what concerns the electronics and the DAQ and in Ref. [6] for what concerns the cryogenic setup and the shields. The energy calibration of the TeO2 crystal is accomplished by the insertion of a removable 232Th gamma source between the cryostat OVC and the external lead shields. The light detector is calibrated by means of a 55Fe X-ray source permanently facing the light detector. 3.1. Results The amplitude and the shape parameters of the voltage pulses are computed during off-line analysis by using the optimum filter technique [10]. The signal amplitudes are evaluated as the maximum of the filtered pulse. The pulse shape parameters that are considered are the rise time ðτR Þ and the decay time ðτD Þ. The rise time of the pulse is defined as the time interval during which the signal rises from 10% up to 90% of the pulse height. Similarly, τD is defined as the time interval during which the signal decreases from 90% down to 30% of the pulse height. All relevant pulses from the detector have approximately the same rise time and decay time. This allows us to apply a pulse shape cut to select only physical pulses and reject noise events. τD as a function of the heat energy is shown for the TeO2 crystal in Fig. 3: the small red dots represent physical pulses, while the bigger blue dots are noise events. The energy spectrum of the TeO2 crystal obtained in 215 h of measurement is reported in Fig. 4. Since the 147Sm α source was
288
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
Table 1 FWHM energy resolutions of the TeO2 crystal and of the light detector evaluated at the 40K γ line and the 55Fe line, respectively.
100
Decay Time
80
Detector
Energy (keV)
FWHM (keV)
TeO2 Light detector
1461 5.9
6.9 0.4
60 40 20 0 0
500
1000
1500
2000
2500
102
Fig. 3. Decay time vs. Energy scatter plot for the TeO2 crystal. The central band (small red dots) is populated by physical events and it is possible to apply a pulse shape cut to reject non-physical events (big blue dots). (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
counts / 3(keV)
Energy (keV)
10
102
counts / 3(keV)
1 0
500
1000
1500
2000
2500
Energy (keV)
10
Fig. 6. Energy spectrum of the bolometric TeO2 detector. The events occurring in coincidence with an event in the light detector are shown in yellow. They are plotted together with the total count rate (blue) of the TeO2 crystal. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
1 0
500
1000
1500
2000
2500
Energy (keV)
103 counts / 0.08 (keV)
Fig. 4. Energy spectrum of the TeO2 crystal. The contribution of the 147Sm α source is clearly evident below 2.3 MeV. The 40K γ line at 1.46 MeV is visible as well.
counts / 0.08(keV)
103
102
102
10
1 0
1
2
3
4
5
6
7
8
9
10
Energy (keV)
10
1 0
1
2
3
4
5
6
7
8
9
10
Energy (keV) Fig. 5. Energy spectrum of the bolometric light detector. The peak at 6 keV is due to the 55Fe source.
dissolved in acid liquid solution, the α energies are smeared out: the acid solution can partially diffuse inside the crystal, and the emitted α can thus release part of its energy in the crystal. The relative fraction of energy deposited in the crystal depends on the depth at which the α is emitted and this generates the continuum background below the 147Sm Q-value in Fig. 4. Fig. 5 shows the
Fig. 7. Energy spectrum of the bolometric light detector. The events occurring in coincidence with an event in the TeO2 detector are shown in yellow. They are plotted together with the total count rate (blue) of the light detector. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
energy spectrum of the light detector, where the peak of the 55Fe source is clearly visible at ∼6 keV. The detector performance is shown in Table 1. The energy resolution of the TeO2 bolometer is evaluated at the 40K γ line at 1.46 MeV, while the energy resolution of the light detector is evaluated at the low energy X-rays (5.9 and 6.4 keV lines). In order to study the α particles emitted on the surface of the absorber crystal and interacting in the scintillating foil, also a coincidence cut has been applied. The coincidence of a heat signal (pulse on the TeO2 crystal due, as an example, to the recoiling
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
nucleus in the decay) and a light signal (pulse on the light detector, due to the α particle escaping from the crystal and interacting in the scintillating foil) is defined in a time window of 50 ms. Fig. 6 shows the energy spectrum of the events in the TeO2 crystal occurring in coincidence with an event in the light detector. Those events (yellow in Fig. 6) are shown together with the total count rate (blue) in the TeO2 crystal. It is clear that the coincidence events populate only the lower energy region of the TeO2 spectrum: this is the energy region in which we expect the energy released by a nuclear recoil of a 147Sm decay. Fig. 7 shows the energy spectrum of the events in the light detector. The coincidence events (yellow) are shown together with the total events detected by the light detector (blue). The coincidence events are concentrated at an energy of about 400 eV and there are very few coincidence events at higher energies. Using the coincidence analysis we thus have a clear indication that the events in the TeO2 crystal are due to nuclear recoils (energy ∼80 keV) of the 147Sm decay while the corresponding coincidence events detected in the light detector are produced by the interaction of the α exiting from the crystal into the scintillating foil. Even if further improvements in the experimental setup are needed, we can state that the proposed technique has the capability of tagging this kind of surface events. 4. Conclusions
detector is encapsulated in a scintillating foil and after adding a bolometric light detector it has been operated at low temperature. Using a coincidence analysis between the events of the crystal absorber and the light detector, we are able to identify and to tag the events caused by α decays occurring on the surface of the crystal. By means of this technique it will be possible to improve the reduction of background in the next generation of experiments searching for rare events with bolometric detectors.
Acknowledgments We would like to thank M. Pedretti and W. Seidel for useful discussions and S. Pirro and L. Cardani for constructive help in various stages of this experiment.
References [1] [2] [3] [4] [5] [6] [7] [8]
In this work we analyze the performance of a prototype bolometric detector designed for tagging and rejecting α contamination generated on the surface of the detector. The bolometric
289
[9] [10]
G. Angloher, et al., Astroparticle Physics 23 (2005) 325. Z. Ahmed, et al., Physical Review Letters 102 (2009) 011301. A. Benoit, et al., Physics Letters B 616 (2005) 25. S. Cebrian, et al., Physics Letters B 563 (2003) 48. C. Arnaboldi, et al., Physical Review C 78 (2008) 035502. S. Pirro, Nuclear Instruments and Methods in Physics Research Section A 559 (2006) 672. S. Pirro, et al., Nuclear Instruments and Methods in Physics Research Section A 559 (2006) 361. F. D Becchetti, et al., Nuclear Instruments and Methods in Physics Research Section A 138 (1976) 93. S.K. Saraf, et al., Nuclear Instruments and Methods in Physics Research Section A 288 (1990) 451. E. Gatti, et al., Rivista del Nuovo Cimento 9 (1986) 1.
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
Rejection of surface background in thermal detectors: The ABSuRD project L. Canonica a,n, M. Biassoni b, C. Brofferio b, C. Bucci a, S. Calvano a, M.L. Di Vacri a, J. Goett c, P. Gorla a, M. Pavan b, M. Yeh d a
INFN, Laboratori Nazionali del Gran Sasso, Assergi, AQ, Italy Università di Milano Bicocca e INFN Sezione di Milano Bicocca, Milano, Italy c Los Alamos National Laboratory, Los Alamos, NM, USA d Brookhaven National Laboratory, Upton, NY, USA b
art ic l e i nf o
a b s t r a c t
Available online 25 May 2013 Keywords: Thermal detectors Rare events searches Scintillation Surface background
Thermal detectors have recently achieved a leading role in the fields of Neutrinoless Double Beta Decay and Dark Matter searches thanks to their excellent energy resolution and to the wide choice of absorber materials. In these fields the background coming from surface contaminations is frequently dominant. ABSuRD (A Background Surface Rejection Detector) is a scintillation-based approach for tagging this type of background. We discuss the innovative application of this technique in non-scintillating bolometric detectors which will allow for a more favorable signal to background ratio. & 2013 Elsevier B.V. All rights reserved.
1. Introduction
2. Rejection of surface background
A bolometer is a particle detector that measures the energy deposited by incident radiation based on the temperature change of the detector. It generically consists of three parts: an absorber, a temperature sensor and a weak thermal link connecting the absorber to a thermal reservoir. Due to the excellent energy resolution that bolometers can achieve (of the order of a few per mil) and due to a vast choice of absorber materials possible, limited only by the heat capacity at the operating temperature, thermal detectors are presently successfully employed in many dark matter (DM) [1–4] and neutrinoless double beta decay ð0νDBDÞ searches [5]. A search for extremely rare events requires the removal of any spurious signal that can mimic the rare event (e.g. radioactive source or cosmic rays). The residual backgrounds counting rate measures the sensitivity of the experiment. As a consequence, one of the main issues of present and future 0νDBD and DM experiments is the identification, control and reduction of the radioactive background. In this work we propose a scintillation-based approach for rejecting surface background and discuss the innovative application of this technique in non-scintillating bolometric detectors.
It has been demonstrated that the main limitation to the sensitivity of bolometric experiments to rare events physics arises from surface alpha contaminations (i.e. in the copper structures holding the detectors), namely degraded α interactions in the absorber [5]. In a bolometric 0νDBD decay experiment, radioactive contaminations close to the absorber surfaces can cause counts in the energy region where the signal is expected. An α particle generated near the surface of a detector component will release only a part of its energy before escaping from the material. The α will then eventually hit another component of the detector. If one of the two components is non-active (e.g. the holding structure) the total α energy cannot be reconstructed, generating a continuum background at all energies. A purely bolometric detector, measuring only the heat signal, cannot discriminate radiation type or surface/bulk events. The discrimination between surface and bulk events and the identification of the particle type interactions (α vs. β=γ) is fundamental for reducing as much as possible surface background sources, and thus increasing the experimental sensitivity. In this work we propose an innovative approach for rejecting surface background. The idea of the ABSuRD project is to encapsulate a purely thermal detector with a scintillating foil and to add a bolometric light detector to detect the light produced by the interaction of degraded α particles with the scintillating foil (see Fig. 1). A surface α particle releasing part of its energy on the crystal and part on the scintillating foil can be tagged by analyzing the coincidence signal of heat (in the absorber) and light (in the light
n
Corresponding author. Tel.: +39 086 243 7431. E-mail address: [email protected] (L. Canonica).
0168-9002/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.nima.2013.05.114
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
287
Light detector
Fig. 1. Diagram of the proposed method.
detector) and rejected as background. Extremely low energy thresholds ð o1 keVÞ are needed for the bolometric light detector, in order to be able to detect small light signals produced by degraded α's. As an example, an α particle of 5.3 MeV (generated by the decay of 210Po) produces about 10 keV of photons in commercial scintillating foil at room temperature. Unfortunately, plastic scintillators have an extremely non-linear response to α particles [8,9] and the amount of light produced by a degraded α particle of 1.5 MeV is about an order of magnitude smaller than the one produced by a 5.3 MeV α. Detecting such a small amount of light represents a serious challenge for most of the currently used light detectors at mK temperature. In order to detect low energy α particles it is therefore desirable to increase the amount of scintillation light. This can be achieved by simply superposing a thin plastic scintillator foil over a reflecting foil.
3. Demonstrator run In order to study the feasibility of this method, a prototype detector has been operated at low temperature. The main goal of the run is to demonstrate, with a non-optimized scintillating foil and light detector, the potential for tagging and rejecting α particles generated on the surface of the absorber crystal. The experimental setup was hosted in a dilution refrigerator placed in the Hall C of National Laboratory of INFN at Gran Sasso, operating at 10 mK. The studied bolometer is a 3 3 6 cm3 TeO2 crystal, held by means of PTFE clamps to a structure of Oxygen Free High Conductivity (OFHC) copper frames and columns, which serve both as mechanical support and heat sink. The crystal is surrounded by a commercial scintillating foil (Saint-Gobain BC400 series) and by a reflecting foil (3 M VM2002), to increase the light collection efficiency. The crystal is provided with a Neutron Transmutation Doped Ge thermistor (NTD), to convert the phonon signal into a detectable voltage pulse. To detect the scintillation light, the top of the crystal faces a germanium bolometric light detector also equipped with an NTD thermistor [7]. It consists of a 50 mm diameter, 300 μm thick pure Ge crystal absorber facing the surface of the crystal. For a picture of the complete set-up, see Fig. 2. A liquid solution containing an α source of 147Sm is deposited on one of the surfaces of the crystal, facing the scintillating and
Fig. 2. The 3 3 6 cm3 TeO2 is enclosed by four copper walls. It is surrounded by the scintillating foil and by the reflecting foil. The top face of the crystal is visible to the Germanium Light Detector.
reflective foils. For an α decay (Q-value ¼2.3 MeV) occurring on the crystal surface, the α particle can escape from the crystal, releasing in the absorber only the energy related to the recoiling nucleus. The remaining part of the energy is released into the scintillating foil, giving rise to a light signal detected by the light detector. Both heat and light pulses produced by particle interactions are amplified and digitized using an 18-bit NI-6284 PXI ADC unit. The trigger is software generated on each bolometer. A complete description of the experimental setup can be found in Ref. [5] and references therein for what concerns the electronics and the DAQ and in Ref. [6] for what concerns the cryogenic setup and the shields. The energy calibration of the TeO2 crystal is accomplished by the insertion of a removable 232Th gamma source between the cryostat OVC and the external lead shields. The light detector is calibrated by means of a 55Fe X-ray source permanently facing the light detector. 3.1. Results The amplitude and the shape parameters of the voltage pulses are computed during off-line analysis by using the optimum filter technique [10]. The signal amplitudes are evaluated as the maximum of the filtered pulse. The pulse shape parameters that are considered are the rise time ðτR Þ and the decay time ðτD Þ. The rise time of the pulse is defined as the time interval during which the signal rises from 10% up to 90% of the pulse height. Similarly, τD is defined as the time interval during which the signal decreases from 90% down to 30% of the pulse height. All relevant pulses from the detector have approximately the same rise time and decay time. This allows us to apply a pulse shape cut to select only physical pulses and reject noise events. τD as a function of the heat energy is shown for the TeO2 crystal in Fig. 3: the small red dots represent physical pulses, while the bigger blue dots are noise events. The energy spectrum of the TeO2 crystal obtained in 215 h of measurement is reported in Fig. 4. Since the 147Sm α source was
288
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
Table 1 FWHM energy resolutions of the TeO2 crystal and of the light detector evaluated at the 40K γ line and the 55Fe line, respectively.
100
Decay Time
80
Detector
Energy (keV)
FWHM (keV)
TeO2 Light detector
1461 5.9
6.9 0.4
60 40 20 0 0
500
1000
1500
2000
2500
102
Fig. 3. Decay time vs. Energy scatter plot for the TeO2 crystal. The central band (small red dots) is populated by physical events and it is possible to apply a pulse shape cut to reject non-physical events (big blue dots). (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
counts / 3(keV)
Energy (keV)
10
102
counts / 3(keV)
1 0
500
1000
1500
2000
2500
Energy (keV)
10
Fig. 6. Energy spectrum of the bolometric TeO2 detector. The events occurring in coincidence with an event in the light detector are shown in yellow. They are plotted together with the total count rate (blue) of the TeO2 crystal. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
1 0
500
1000
1500
2000
2500
Energy (keV)
103 counts / 0.08 (keV)
Fig. 4. Energy spectrum of the TeO2 crystal. The contribution of the 147Sm α source is clearly evident below 2.3 MeV. The 40K γ line at 1.46 MeV is visible as well.
counts / 0.08(keV)
103
102
102
10
1 0
1
2
3
4
5
6
7
8
9
10
Energy (keV)
10
1 0
1
2
3
4
5
6
7
8
9
10
Energy (keV) Fig. 5. Energy spectrum of the bolometric light detector. The peak at 6 keV is due to the 55Fe source.
dissolved in acid liquid solution, the α energies are smeared out: the acid solution can partially diffuse inside the crystal, and the emitted α can thus release part of its energy in the crystal. The relative fraction of energy deposited in the crystal depends on the depth at which the α is emitted and this generates the continuum background below the 147Sm Q-value in Fig. 4. Fig. 5 shows the
Fig. 7. Energy spectrum of the bolometric light detector. The events occurring in coincidence with an event in the TeO2 detector are shown in yellow. They are plotted together with the total count rate (blue) of the light detector. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
energy spectrum of the light detector, where the peak of the 55Fe source is clearly visible at ∼6 keV. The detector performance is shown in Table 1. The energy resolution of the TeO2 bolometer is evaluated at the 40K γ line at 1.46 MeV, while the energy resolution of the light detector is evaluated at the low energy X-rays (5.9 and 6.4 keV lines). In order to study the α particles emitted on the surface of the absorber crystal and interacting in the scintillating foil, also a coincidence cut has been applied. The coincidence of a heat signal (pulse on the TeO2 crystal due, as an example, to the recoiling
L. Canonica et al. / Nuclear Instruments and Methods in Physics Research A 732 (2013) 286–289
nucleus in the decay) and a light signal (pulse on the light detector, due to the α particle escaping from the crystal and interacting in the scintillating foil) is defined in a time window of 50 ms. Fig. 6 shows the energy spectrum of the events in the TeO2 crystal occurring in coincidence with an event in the light detector. Those events (yellow in Fig. 6) are shown together with the total count rate (blue) in the TeO2 crystal. It is clear that the coincidence events populate only the lower energy region of the TeO2 spectrum: this is the energy region in which we expect the energy released by a nuclear recoil of a 147Sm decay. Fig. 7 shows the energy spectrum of the events in the light detector. The coincidence events (yellow) are shown together with the total events detected by the light detector (blue). The coincidence events are concentrated at an energy of about 400 eV and there are very few coincidence events at higher energies. Using the coincidence analysis we thus have a clear indication that the events in the TeO2 crystal are due to nuclear recoils (energy ∼80 keV) of the 147Sm decay while the corresponding coincidence events detected in the light detector are produced by the interaction of the α exiting from the crystal into the scintillating foil. Even if further improvements in the experimental setup are needed, we can state that the proposed technique has the capability of tagging this kind of surface events. 4. Conclusions
detector is encapsulated in a scintillating foil and after adding a bolometric light detector it has been operated at low temperature. Using a coincidence analysis between the events of the crystal absorber and the light detector, we are able to identify and to tag the events caused by α decays occurring on the surface of the crystal. By means of this technique it will be possible to improve the reduction of background in the next generation of experiments searching for rare events with bolometric detectors.
Acknowledgments We would like to thank M. Pedretti and W. Seidel for useful discussions and S. Pirro and L. Cardani for constructive help in various stages of this experiment.
References [1] [2] [3] [4] [5] [6] [7] [8]
In this work we analyze the performance of a prototype bolometric detector designed for tagging and rejecting α contamination generated on the surface of the detector. The bolometric
289
[9] [10]
G. Angloher, et al., Astroparticle Physics 23 (2005) 325. Z. Ahmed, et al., Physical Review Letters 102 (2009) 011301. A. Benoit, et al., Physics Letters B 616 (2005) 25. S. Cebrian, et al., Physics Letters B 563 (2003) 48. C. Arnaboldi, et al., Physical Review C 78 (2008) 035502. S. Pirro, Nuclear Instruments and Methods in Physics Research Section A 559 (2006) 672. S. Pirro, et al., Nuclear Instruments and Methods in Physics Research Section A 559 (2006) 361. F. D Becchetti, et al., Nuclear Instruments and Methods in Physics Research Section A 138 (1976) 93. S.K. Saraf, et al., Nuclear Instruments and Methods in Physics Research Section A 288 (1990) 451. E. Gatti, et al., Rivista del Nuovo Cimento 9 (1986) 1.