Development of decay energy spectroscopy using low temperature detectors

Applied Radiation and Isotopes 70 (2012) 2255–2259

Contents lists available at SciVerse ScienceDirect

Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso

Development of decay energy spectroscopy using low temperature detectors Y.S. Jang a,b, G.B. Kim a, K.J. Kim a, M.S. Kim a, H.J. Lee a,c, J.S. Lee a, K.B. Lee a, M.K. Lee a, S.J. Lee a, H.C. Ri b, W.S. Yoon a,c, Y.N. Yuryev a,1, Y.H. Kim a,c,n a

Korea Research Institute of Standards and Science (KRISS), Daejeon, South Korea Department of Physics, Kyungpook National University, Daegu, 702-701, South Korea c University of Science and Technology, Daejeon, 305-333, South Korea b

a r t i c l e i n f o

abstract

Available online 2 March 2012

We have developed a high-resolution detection technique for measuring the energy and activity of alpha decay events using low-temperature detectors. A small amount of source material containing alpha-emitting radionuclides was enclosed in a 4p metal absorber. The energy of the alpha particles as well as that of the recoiled nuclides, low-energy electrons, and low-energy x-rays and g-rays was converted into thermal energy of the gold absorber. A metallic magnetic calorimeter serving as a fast and sensitive thermometer was thermally attached to the metal absorber. In the present report, experimental demonstrations of Q spectroscopy were made with a new meander-type magnetic calorimeter. The thermal connection between the temperature sensor and the absorber was established with annealed gold wires. Each alpha decay event in the absorber resulted in a temperature increase of the absorber and the temperature sensor. Using the spectrum measured for a drop of 226Ra solution in a 4p gold absorber, all of the alpha emitters in the sample were identified with a demonstration of good detector linearity. The resolution of the 226Ra spectrum showed a 3.3 keV FWHM at its Q value together with an expected gamma escape peak at the energy shifted by its g-ray energy. & 2012 Elsevier Ltd. All rights reserved.

Keywords: Cryogenic detector Metallic magnetic calorimeter Q spectrometer Alpha spectroscopy High energy resolution

1. Introduction Research in cryogenic particle detectors has been motivated primarily by nuclear/particle physics and astrophysics, mainly because their energy sensitivities and thresholds are beyond the limit of conventional semiconductor-based detectors (Booth et al., 1996). Over the past two decades, remarkable improvements have been achieved for their energy resolution in many applications in radiation detection and measurement (Enss, 2005). In calorimetric detection, a typical cryogenic detector consists of an absorber and a temperature sensor. The energy transferred to the absorber by an incident particle or radiation causes a small temperature increase. With a sensitive thermometer attached to the absorber, the temperature increase can be accurately measured according to the initial energy input to the absorber. Temperature sensors such as Transition Edge Sensors (TESs) and Metallic Magnetic Calorimeter (MMCs) have been a major driving force in the successful development of cryogenic detectors. TESs utilize the sharp resistance change at the phase

n Corresponding author at: University of Science and Technology, Daejeon, 305-333, South Korea. E-mail address: [email protected] (Y.H. Kim). 1 Now at Korea Research Institute of Standards and Science (KRISS), Daejeon, South Korea

0969-8043/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.apradiso.2012.02.109

transition of a superconducting film (Irwin, 1995; Irwin and Hilton, 2005). At the transition temperature, a small change in temperature results in a resistance change that can be measured by a feedback mechanism. MMCs use a paramagnetic material in a small magnetic field as a temperature sensor (Enss et al., 2000; Fleischmann et al., 2005). The magnetization of the magnetic material, which is a function of temperature, is measured via a superconducting circuit. The energy resolution in alpha spectroscopy has been remarkably improved by the development of TES and MMC technologies (Croce et al., 2011; Horansky et al., 2010; Ranitzsch et al., 2011). They have achieved an energy resolution of 1–3 keV for thin layer alpha sources. This energy resolution range allows for the identification of radionuclides whose alpha lines are so close to each other that they cannot be separated by conventional silicon-based detectors. The resolution of conventional detectors is theoretically limited to 8–10 keV for alpha particle detection (Steinbauer et al., 1994). In applications to radionuclide metrology, a 4p measurement scheme with a metal absorber has been suggested using an MMC as the temperature sensor (Loidl et al., 2004). When 55Fe is completely enclosed by a metal absorber, the total energy transferred to the absorber from an electron capture decay event is converted into thermal energy. With the calorimetric detection method, the energy release and the decay counts can be measured

2256

Y.S. Jang et al. / Applied Radiation and Isotopes 70 (2012) 2255–2259

with a high accuracy (Loidl et al., 2008). This method has been used to study the shape and the end point of the 241Pu beta spectrum, which is a forbidden and non-unique transition (Loidl et al., 2010). The 4p detection method has also been applied to alphaemitting nuclides (Lee et al., 2010). In this case, the total disintegration energy (Q) of 241Am decay is converted to a temperature change of the 4p absorber. The calorimetric measurement of the decay energy was confirmed through a subsequent experiment with an additional external 241Am source introduced to the absorber. This 4p measurement of alpha emitters suggests an analytic tool for the identification and integral counting of radionuclides. When this measurement method was employed for a commercially available plutonium sample, the Q spectrum showed two clearly separated peaks for 239Pu and 240Pu with an energy resolution of 4–6 keV at the FWHM (Jang et al., in press). The present experimental work aims to extend the 4p measurement method to various samples over a wide energy range and for a combination of beta-decaying nuclides. Moreover, a new MMC setup fabricated on a meander pickup coil (Fleischmann et al., 2005) is employed for a 4p gold absorber for the first time. The meander MMC was designed for high-resolution measurements with a large absorber suitable for the 4p measurement of alpha emitters.

2. Detector setup 2.1. Metallic magnetic calorimeters MMCs employ Au:Er, which consists of a small concentration of erbium in a gold host, as the sensor material. The diluted magnetic ions in the metallic host maintain paramagnetic properties at temperatures on the order of tens of mK (Fleischmann et al., 2000; Enss et al., 2000). The gold host provides a short thermalization time to reach a thermal equilibrium in the sensor material between the sub-thermal systems of electrons and magnetic spins. The temperature dependence of the magnetization is well understood, accounting for the exchange interactions between the magnetic spins. The development of MMCs in particle detection was done using a setup consisting of a Au:Er sensor placed inside the loop of a Superconducting QUantum Interference Device (SQUID). In this early MMC design, the SQUID loop itself was used as a pickup coil (Fleischmann et al., 2005). The state of the art development of SQUID technologies guarantees an accurate and fast measurement of magnetization changes with low noise upon any temperature changes caused by energy absorption. The SQUID converts the change in magnetic flux into a measurable voltage signal on the basis of a quantum interference measurement operating at low temperatures. The detection principle of an MMC can be characterized as DE-DT-DM-DF-DV where E, T, M, F and V indicate the physical quantities of energy, temperature, magnetization, magnetic flux and voltage, respectively. An MMC setup with a twostage SQUID system obtained an energy resolution of 2.7 eV FWHM for 55Fe x-rays (Fleischmann et al., 2009). In a new design of MMCs, a Au:Er film is sputtered onto a meander-shaped pickup coil (Burck et al., 2008). The microfabricated meander coil forms a superconducting loop to provide a persistent current in order to apply a constant magnetic field to the Au:Er film. The current signal in the superconducting loop due to the magnetization change of the Au:Er film is read by a current sensing SQUID inductively connected to the coil. A simplified measurement circuit of the present experiment is illustrated in Fig. 1. This new MMC setup is particularly useful for detectors

sensor SQUID 2.8 × 2.8 × 0.05 mm3

Nb pickup coil

1 mm2 gold stems Au:Er gold wires 25 μm ×1.8 mm gold wires

radionuclides

gold stems Nb coil Au:Er sapphire substrate

Bakelite substrate

sample holder Fig. 1. A schematic diagram of the meander-type MMC used in the 4p decay energy spectrometer. (A) A simplified detector design. The simplified circuit shows a two-sided meander coil and an input coil of the sensor SQUID. The Au:Er film is sputtered on both sides of the meander pickup coil connected to the sensor SQUID in a gradiometric configuration. Sixteen stems are fabricated for an easy thermal connection to an absorber. A 4p gold absorber containing radionuclides is located next to the meander coil and is thermally connected onto the top surface of the stems on one side of the meander via six 25 mm gold wires. (B) A cross-sectional view of the spectrometer (not to scale).

using an absorber with a large heat capacity. A meander MMC with a 1 mm2 area of Au:Er showed an energy resolution of 2.8 keV at the FWHM for 5.5 MeV alpha particles with a 2.5  2.5  0.07 mm3 gold absorber (Ranitzsch et al., 2011). 2.2. 4p Structure The 4p absorber used in this experiment was constructed via a simple method described in previous reports (Loidl et al., 2008; Lee et al., 2010). A small amount of a 226Ra solution in 1 M nitric acid was dropped and dried on a piece of 25 mm thick gold foil. After the foil was folded to cover the solution drop, it was pressed with a vise and put in an oven at 400 1C for 16 h in an argon atmosphere for diffusion welding. Then, it was cut in a square shape to contain all of the left-over from the original solution drop. The final dimensions of the foil became 2.8  2.8  0.05 mm3. This 4p absorber is larger than other 4p absorbers used in previous reports (Loidl et al., 2008; Lee et al., 2010; Loidl et al., 2010) and ensures that no leakage of radionuclides or alpha particles occurs at the sides. A thermal connection between the absorber and the temperature sensor was made with gold wires, as shown in Fig. 1. In a separate experiment (Yoon et al., in press), the thermal conductance between a gold absorber and a meander-type MMC was found to be the electronic thermal conductance along the gold wires, as expected by the Wiedemann–Franz law. This thermal conductance provides a much more efficient heat flow compared to the direct pressure contact used in an alpha spectrometer (Ranitzsch et al., 2011). Six annealed 25 mm gold wires with an average length of 1.8 mm were used for the thermal connection in this experiment. 3. 4p Spectrum The total available energy (Q) of an alpha decay event is determined by the mass difference between a parent nuclide and

Y.S. Jang et al. / Applied Radiation and Isotopes 70 (2012) 2255–2259

Fig. 3 shows the energy calibration and the linearity of the detector. The signal size relative to the 226Ra signals is compared with the corresponding Q values of their decay energy for five peaks including the g escape peak for 226Ra. As shown in the figure, the linear fit shows good agreement with the five points. The deviations between the calibration line and the five points are less than 1 keV. The extrapolation to zero signal size indicates some non-linearity. The calibration line does not pass through the origin but 61.7 keV at the zero signal size as shown in the inset. The thermal detectors in the simple model of an equilibrium calorimeter (McCammon, 2005) may not have an offset or may exhibit some nonlinearity in the measurement of an energy input. This nonzero extrapolation in the experiment may imply a limit of the thermal model in the linear theory of an equilibrium detector (McCammon, 2005). A nonlinearity in the energy measurement may appear over the full energy range. This nonlinearity may originate from the linear temperature dependence of heat capacity and the inverse of the magnetization in the range of the

the sum of the final products, an alpha particle and a ground-state daughter nuclide. When an alpha transition occurs to the ground level of the daughter’s nuclear state, the decay energy Q is shared by the alpha particle and the daughter nuclide. In the case that the excited levels of the nuclear states are involved in an alpha transition, the rest of the energy, aside from the kinetic energy of the two particles, is transferred to electrons and/or photons via deexcitation processes (i.e., gamma transitions) toward lower energy levels and eventually to the ground level. The gamma transitions may include internal conversion and g-ray emission processes. Atomic processes are also allowed in electron rearrangements, and may emit x-rays and Auger elections. Regardless of the path followed to the final state, the sum of the energy associated with alpha, gamma and atomic transitions is equal to the Q value. In the present 4p geometry, all of the energy associated with an alpha decay event is transferred into thermal energy of the gold absorber (Lee et al., 2010). This includes the kinetic energy of the alpha particle and the recoiled daughter as well as low-energy electrons and low-energy x-rays and g-rays. The de-excitation processes happen in a much shorter time than the rise time of the calorimetric signals, which are in the millisecond range in this experiment. High-energy electron and photon emissions may escape from the absorber without depositing all of their energy to the gold foil absorber with an effective thickness of 25 mm. However, the emission probabilities of those events are very low for the alpha emitters considered in this experiment except for 226 Ra. The 226Ra decay has a 186.211 (13) keV g-ray line with a 3.555 (19)  10 2 emission probability (Be´ et al., 2004). 226Ra must exhibit two peaks in the 4p spectrum at its Q value and at the Q value subtracted from the g-ray energy. Fig. 2 shows the full spectrum measured with the 4p absorber at a 40 mK base temperature and a 100 mA field current. Many sharp peaks appeared in the spectrum measured for a small drop of 226Ra solution. The peaks corresponding to 226Ra, 222Rn and 218 Po can be easily identified by comparing their relative size with their expected Q values. Most of the daughter nuclides, the decay products in the 226Ra decay chain, remained and also decayed in the 4p absorber, particularly at low temperatures where radon cannot diffuse out. Each of them contributed to a peak in the Q spectrum. Another peak is found at the Q value of 210Po, as expected from the data sheet of the 226Ra solution. Moreover, some daughter nuclides in the decay chain may undergo beta decay. These beta decay signals appear on the lowenergy side of the full spectrum. The broad peak near 8 MeV is attributed to consecutive decay events of 214Bi and 214Po. Because of the relatively short life time of 214Po, the two (beta and alpha) decay signals are not separated, but added in a pulse of a temperature increase.

3

1.25 1 Q(218Po)

1.2 Relative amplitude

0.5 1.15 0

0

2000

4000

6000

1.1

Q(222Rn)

Q(210Po)

1.05

226

Q(

Ra)

1 Q(226Ra)-E

0.95

Residual (10-3)

4600

4800

5000

5200 5400 5600 Energy (keV)

5800

6000

6200

4800

5000

5200

5800

6000

6200

1 0 -1 4600

5400

5600

Fig. 3. Relative signal size vs. alpha decay Q values. The line is a linear fit of the five points. The inset shows the extrapolation to zero.

226

10

2257

Ra 210

222

Rn

218

Po

Po

2

dN/dE

10

226

β signals

Ra γ escape 214

1

Bi + 214 Po

10

0

10

0

1000

2000

3000

4000 5000 6000 Decay energy (keV)

7000

8000

Fig. 2. Full spectrum of decay energy measurement with the 4p absorber.

9000

10000

2258

Y.S. Jang et al. / Applied Radiation and Isotopes 70 (2012) 2255–2259

temperature increase due to the energy input of several MeV under the measurement conditions.

4. Discussion and conclusion The energy resolution achieved in the study was characterized by fitting the measured spectrum with an exponential convolution of a Gaussian response. Any physical process that may cause an incomplete energy conversion other than a measurable temperature change accounts for the left-hand side low-energy tailing part of the resulting tailed Gaussian function (Philips and Marlow, 1976). The Gaussian width, s, indicates the FWHM of the detector response at 2.355s (Horansky et al., 2010). The measured spectrum near the 226Ra Q value is shown with a fit in Fig. 4. The small peak located at 186.22 keV less than the 226 Ra Q value is attributed to a g-ray escaping without depositing energy to the absorber, as indicated by two vertical lines. A 3.3 keV FWHM was obtained for the Q peak of 226Ra. A detailed comparison of the gray curve and the measured points shows some noticeable deviations on the left and right-hand side of the 226 Ra peak. This implies that other considerations should be added to the simple tailed Gaussian function in order to represent the measured spectrum. On the other hand, the random baseline noise of the measurement setup yielded a 1.3 keV FWHM equivalent to a decay energy. Furthermore, the Q peaks for higher energies showed 4.2 and 4.4 keV FWHM for the 222Rn and 218Po peaks, respectively. Similar results were found when the signals for 218Po were used to determine the relative size. The disagreement between the values indicates that the measured resolution was not limited by the electronic noise of the measurement circuit or the thermodynamic fluctuations of the detector, which is the intrinsic limitation of thermal detectors (McCammon, 2005). Although the reasons behind the broader resolution for larger Q values are not understood, one possible explanation is the lattice damage effect, depending upon the energies of the alpha particle and the daughter nuclide. A 1 keV FWHM for 5.3 MeV alpha particles has been found due to lattice damage in a superconducting tin absorber (Horansky et al., 2010). The introduction of a meander-type MMC and gold wires for thermal connections improved the energy resolution over those of previous 4p measurements, which is 6–10 keV at the FWHM. In addition, the significant low-energy tailing effect shown in the

900 800 700

dN / dE

600

186.22 keV measured fit

500 400 300 200 100 0 4600

4650

4700

4750 4800 4850 Decay energy (keV)

4900

4950

Fig. 4. The decay energy spectrum of 226Ra. The small peak located at the energy 186 keV below the 226Ra Q value originates from g-ray escapes. The gray curve is a fit to the two peaks of the spectrum, obtained using a tailed Gaussian function with a Gaussian width of 3.3 keV at the FWHM.

previous experiment with a plutonium sample (Jang et al., in press) was not observed with this method. The MMC setup of the previous experiment was an old geometry in which an absorber is placed onto a SQUID chip under the influence of an external magnetic field. This feasibility study of a 4p measurement for Q spectroscopy demonstrated a decay energy spectrum in which each radionuclide represents one peak at its own Q value. High-energy resolution was obtained, that was much better than that of conventional silicon based detectors. With this resolution, the detector linearity and the capability of integral counting (Pomme´, 2007), all of the alpha emitters in a 4p absorber can be identified and numerated. Even though extended future efforts are required for further usage of this method for real samples, it is clear that the 4p Q spectrometric measurements may have some advantages over conventional alpha spectrometers, which encounter difficulties associated with unfavorable straggling properties or complex source preparation of alpha-emitting radionuclides.

References Be´, M.M., Chiste´, V., Dulieu, C., Browne, E., Chechev, V., Kuzmenko, N., Helmer, R., ¨ Nichols, A., Schonfeld, E., Dersch, R., 2004. Table of Radionuclides, Monographie BIPM-5. Tech. rep., BIPM. Booth, N.E., Cabrera, B., Fiorini, E., 1996. Low-temperature particle detectors. Ann. Rev. Nucl. Part. Sci 46, 471. ¨ Burck, A., Kempf, S., Schafer, S., Rotzinger, H., Rodrigues, M., Wolf, T., Gastaldo, L., Fleischmann, A., Enss, C., 2008. Microstructured magnetic calorimeter with meander-shaped pickup coil. J. Low Temp. Phys. 151, 337–344. Croce, M., Bacrania, M., Bond, E., Dry, D., Klingensmith, A., Moody, W., LaMont, S., Rabin, M., Rim, J., Beall, J., et al., 2011. Superconducting transition-edge sensor microcalorimeters for ultra-high resolution alpha-particle spectrometry. IEEE Trans. Appl. Supercond. 21 (3), 207–210. Enss, C. (Ed.), Springer-Verlag, Berlin Heidelberg. ¨ Enss, C., Fleischmann, A., Horst, K., Schonefeld, J., Sollner, J., Adams, J.S., Huang, Y.H., Kim, Y.H., Seidel, G.M., 2000. Metallic magnetic calorimeters for particle detection. J. Low Temp. Phys. 121 (3), 137–176. Fleischmann, A., Enss, C., Seidel, G.M., 2005. Metallic magnetic calorimeters. Top. Appl. Phys. 99, 151. Fleischmann, A., Schonefeld, J., Sollner, J., Enss, C., Adams, J.S., Bandler, S.R., Kim, Y.H., Seidel, G.M., 2000. Low temperature properties of erbium in gold. J. Low Temp. Phys. 118, 7. Fleischmann, L., Linck, M., Burck, A., Domesle, C., Kempf, S., Pabinger, A., Pies, C., ¨ Porst, J.-P., Rotzinger, H., Schafer, S., Weldle, R., Fleischmann, A., Enss, C., Seidel, G., 2009. Metallic magnetic calorimeters for x-ray spectroscopy. IEEE Trans. Appl. Supercond. 19, 63–68. Horansky, R.D., Stiehl, G.M., Beall, J.A., Irwin, K.D., Plionis, A.A., Rabin, M.W., Ullom, J.N., 2010. Measurement of ion cascade energies through resolution degradation of alpha particle microcalorimeters. J. Appl. Phys. 107, 044512. Irwin, K., 1995. An application of electrothermal feedback for high resolution cryogenic particle detection. Appl. Phys. Lett. 66, 1998. Irwin, K., Hilton, G., 2005. Transition-edge sensors. Top. Appl. Phys. 99, 63. Jang, Y.S., Lee, S.J., Kim, G.B., Kim, I.H., Kim, M.S., Lee, H.J., Lee, J.S., Lee, K.B., Lee, M.K., Ri, H.C., Yoon, W.S., Kim, Y.H., Development of decay energy spectroscopy for radionuclide analysis using cryogenic 4p measurements. J. Low Temp. Phys. In: Proceedings of the 14th International Workshop on Low Temperature Detectors, 1–5 August 2011, Heidelberg, Germany, doi:10.1007/ s10909-012-0477-y, in press. Lee, S.J., Lee, M.K., Jang, Y.S., Kim, I.H., Kim, S.K., Lee, J.S., Lee, K.B., Lee, Y.H., Kim, Y.H., 2010. Cryogenic measurement of alpha decay in a 4p absorber. J. Phys. G: Nucl. Part. Phys. 37, 055103. Loidl, M., Leblanc, E., Bouchard, J., Branger, T., Coron, N., Leblanc, J., de Marcillac, P., Rotzinger, H., Daniyarov, T., Linck, M., Fleischmann, A., Enss, C., 2004. Highenergy resolution x-ray, gamma and electron spectroscopy with cryogenic detectors. Appl. Radiat. Isot. 60, 363–368. Loidl, M., Leblanc, E., Rodrigues, M., Branger, T., Lacour, D., Bouchard, J., Censier, B., 2008. Validation study of a new technique for absolute activity measurement with 4p solid angle metallic magnetic calorimeters. Appl. Radiat. Isot. 66 (6– 7), 872–876. Loidl, M., Rodrigues, M., Censier, B., Kowalski, S., Mougeot, X., Cassette, P., Branger, T., Lacour, D., 2010. First measurement of the beta spectrum of 241Pu with a cryogenic detector. Appl. Radiat. Isot. 68 (7–8), 1454–1458. McCammon, D., 2005. Thermal equilibrium calorimeters- an introduction. Top. Appl. Phys. 99, 1. Philips, G.W., Marlow, K.W., 1976. Automatic analysis of g -ray spectra from germanium detectors. Nucl. Instrum. Methods 137, 525–536. Pomme´, S., 2007. Methods for primary standardization of activity. Metrologia 44 (4), S17–S26.

Y.S. Jang et al. / Applied Radiation and Isotopes 70 (2012) 2255–2259

Ranitzsch, P.C., Kempf, S., Pabinger, A., Pies, C., Porst, J.P., Schafer, S., Fleischmann, A., Gastaldo, L., Enss, C., Jang, Y.S., et al. Development of cryogenic alpha spectrometers using metallic magnetic calorimeters. Nucl. Instrum. Methods A. 652 (1), 299–301, in press. ¨ Steinbauer, E., Bauer, P., Geretschlager, M., Bortels, G., Biersack, J.P., Burger, P., 1994. Energy resolution of silicon detectors: approaching the physical limit. Nucl. Instrum. Methods B 85, 642–649.

2259

Yoon, W.S., Jang, Y.S., Kim, G.B., Kim, K.J., Kim, M.S., Lee, J.S., Lee, K.B., Lee, M.K., Lee, S.J., Lee, H.J., Yuryev, Y.N., Kim, Y.H. High energy resolution cryogenic alpha spectrometers using magnetic calorimeters. J. Low Temp. Phys. In: Proceedings of the 14th International Workshop on Low Temperature Detectors, 1–5 August 2011, Heidelberg, Germany, doi:10.1007/ s10909-012-0539-1, in press.