Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter

Nuclear Inst. and Methods in Physics Research, A xxx (xxxx) xxx

Contents lists available at ScienceDirect

Nuclear Inst. and Methods in Physics Research, A journal homepage: www.elsevier.com/locate/nima

Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter Riham Mariam, Matias Rodrigues ∗, Martin Loidl CEA, LIST, Laboratoire National Henri Becquerel (LNE-LNHB), CEA-Saclay, 91191 Gif sur Yvette Cedex, France

ARTICLE

INFO

Keywords: Metallic Magnetic Calorimeter (MMC), Actinide, X-ray spectrometry

ABSTRACT Emissions of L-X-rays by actinides are quite intense due to the large internal conversion coefficients of the gamma transitions. These emissions can be beneficial for the quantitative analysis of actinides by photon spectrometry, however the L X-rays emitted by radionuclides are generally not well known. Several studies explored the L-X-ray regions of actinides and measured total intensities of L-X-ray groups with large uncertainties up to 10%. However, they cannot resolve the individual L-X-ray lines due to the limited energy resolution (of the order of 200 eV (FWHM) at 25 keV) of the semiconductor detectors. The present work shows and discusses the measurement of individual L-X-ray emission intensities emitted by Cm-244 using an ultra-high-resolution spectrometer. The used spectrometer is a Metallic Magnetic Calorimeter (MMC) conceived for photon spectrometry below 100 keV with an energy resolution of 37 eV at 20 keV and a nearly constant efficiency over the L X-ray energy range. A high-resolution spectrum with a counting statistics of 4.6 × 106 counts was obtained. The spectrum processing allowed to determine the individual emission intensities of 30 L-X-ray lines despite the narrow energy spacing.

Contents 1. 2. 3.

Introduction ....................................................................................................................................................................................................... Spectrum processing and discussion ..................................................................................................................................................................... Decay data for Cm-244 ....................................................................................................................................................................................... References..........................................................................................................................................................................................................

1. Introduction Most actinides decay to many nuclear excited states. Some of the related gamma transitions have large internal conversion coefficients (ICCs) which lead to intense emission of X-ray photons during the atomic rearrangement. The detection of these photons could be used to analyze samples with mixed actinides. However, the knowledge of these emission intensities of L-X-rays is limited in particular for Cm-244: there is only one measurement with large uncertainties [1]. According to the literature, the knowledge of these emissions concerns only the L-X-ray groups, and it suffers from large uncertainties due to the limited resolution of semiconductor detectors and to the Kedge discontinuity in the HPGe detector detection efficiency in the L-X-ray energy range [2,3]. Cm-244 decays by alpha emission with a probability of 76.7% to the first excited state of Pu-240 at 42 keV. The 𝛾 transition (𝐸2 electromagnetic transition) is highly converted, involving mainly the L-shell with an ICC 𝛼L = 658 (13). In addition, ∗

1 2 3 3

with a mean fluorescence yield 𝜔L = 0.521 (20) [2], this 𝛾 transition leads to intense X-ray emissions below 25 keV. The L-X-ray spectrum is very complex because of the many possible transitions to fill the hole in the atomic inner subshells (L1 , L2 , L3 ) of the daughter. According to the ICC of each sub-shell, there are 0.442 (18), 12.3 (5) and 10.62 (44) primary vacancies per 100 decays created respectively in the L1 , L2 and L3 subshells. In order to recover the electronic ground state, different atomic radiative and non-radiative processes take place depending on their probabilities. An electron from an outer subshell can fill the vacancy with an emission of X-ray photon: Li – Yj (i = 1, 2, 3; j = 1, … , 7; Y = M, N, O, P). The vacancy created by the IC in Li can be filled by an outer subshell which has the same quantum number L. This radiationless transition, called Coster–Kronig (CK) ejects an electron from an outer subshell Yk (Y = M, N, O, P), therefore the atom is doubly ionized. The CK Li −𝐿j Yk transition is defined by the probability 𝑓𝑖𝑗 (for Pu: 𝑓12 = 0.03, 𝑓23 = 0.214, 𝑓13 = 0.68: respectively L1 −L2 Y2 ,

Corresponding author. E-mail address: [email protected] (M. Rodrigues).

https://doi.org/10.1016/j.nima.2019.04.020 Received 31 July 2018; Accepted 3 April 2019 Available online xxxx 0168-9002/© 2019 Published by Elsevier B.V.

Please cite this article as: R. Mariam, M. Rodrigues and M. Loidl, Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter, Nuclear Inst. and Methods in Physics Research, A (2019), https://doi.org/10.1016/j.nima.2019.04.020.

R. Mariam, M. Rodrigues and M. Loidl

Nuclear Inst. and Methods in Physics Research, A xxx (xxxx) xxx

Fig. 1. Energy spectrum of the photons emitted in the decay of Cm-244.

Table 1 Comparison of the relative intensities of the groups 𝐼L1 obtained with the present measurement and with calculation. Intensity relative to the total L X-rays

SMX3

Calculation

𝐼L1 𝐼L2 𝐼L3

0.006622 (6) 0.4776 (9) 0.5151 (8)

0.0076 (26) 0.48 (9) 0.51 (10)

L2 −L3 Yk and L1 −L3 Yk [4]). This work presents a measurement of the LX-ray line intensities using a high resolution spectrometer developed by the LNHB and by the KIP of Heidelberg University. This spectrometer called SMX3 consists in an MMC detector dedicated to photon spectrometry below 100 keV. The basic idea of a cryogenic detector is the measurement of an energy deposit as a temperature rise in the detector. An MMC consists of a metallic absorber strongly thermally coupled to a paramagnetic sensor, which, located in a weak magnetic field of few mT, has a temperature dependent magnetization. The photon with an energy E interacting in the metallic absorber, characterized by a heat capacity C, results a temperature change ΔT = E∕C. The temperature variation in the absorber is converted to a magnetization variation ΔM(T) by the sensor. The magnetization change generates a magnetic flux change in a flux transformer magnetically coupled to a Superconducting Quantum Interference Device (SQUID) [5]. The metallic absorber used in SMX3 consists in a double layer of Au and Ag offering a constant and smooth intrinsic efficiency (> 98%) below 25 keV, unattainable with an absorber with a single layer of Au or Ag [6]. An electroplated source of Cm-244 with an activity of 71.54 (15) kBq prepared at LNHB, thin enough to avoid self-absorption, was integrated with SMX3. A tungsten collimator is used to define the solid angle and also to avoid photon interaction in the read-out chip. The alpha particles and the electrons from the Cm-244 decay are blocked using a beryllium window which allows transmitting 61.6% to 96.3% of the M-X-rays ranging from 2.2 keV to 5.92 keV and > 99.5% of the L X-rays ranging from 12 to 22 keV.

Fig. 2. Different regions of L-X-rays deconvoluted using Colegram; diagram line transitions (green solid lines): X-ray transitions when the atomic configuration is in a single vacancy state; satellite structures s (yellow dashed lines): X-ray transitions when the atomic configuration is in a multiple vacancy state; background (blue dashed line); X-ray fluorescences from Au or W (purple dash-dotted line). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

2. Spectrum processing and discussion SMX3 was cooled down to 10 mK using a dilution refrigerator. The output voltage of the SQUID electronics was digitized continuously during 10 days with an 16-bit resolution acquisition card at a sampling frequency of 250 kHz and stored on a hard disk. This signal was analyzed offline following the different steps described in [7] to obtain the spectrum of Cm-244 presented in the Fig. 1. This spectrum shows the different regions of the L-X-rays which were deconvoluted using the Colegram software [8]; the X-ray lines were fitted using a Voigt function (convolution of a Gaussian and a Lorentzian functions) with a fixed Lorentzian width taken from [9]. The FWHM of the Gaussian was determined on the X-ray fluorescence lines emitted by the source and detector holders (Fe, Ni, Cu, Mn) and at the 42 keV 𝛾 line. A second degree polynomial function was used to evaluate the variation of the

Fig. 3. Region of M-X-rays: diagram line transitions (green lines) and the transition 𝐿1 −𝐿3 (green solid line); background (blue dashed line).

Gaussian FWHM with the energy. The FWHM ranges from 32.43 eV at 6 keV to 37 eV at 20 keV. The background was fitted using an exponential function. The high statistics of 4.6 × 106 counts and the high energy resolution give access to the complex L-X-ray structure. Fig. 2 shows the different L X-ray groups (X L𝜄 ,X L𝛼, X L𝛽, X L𝜂, X L𝛾); 2

Please cite this article as: R. Mariam, M. Rodrigues and M. Loidl, Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter, Nuclear Inst. and Methods in Physics Research, A (2019), https://doi.org/10.1016/j.nima.2019.04.020.

R. Mariam, M. Rodrigues and M. Loidl

Nuclear Inst. and Methods in Physics Research, A xxx (xxxx) xxx

Table 2 Relative emission intensities of L X-rays measured by SMX3. The relative intensities are compared with the results from [1] and [13]. Transition IUPAC XL𝜄 XL𝛼 XL𝜂

XL𝛽

XL𝛾

𝐿1 − 𝐿3 𝐿3 −𝑀 1 𝐿3 −𝑀 2 𝐿3 −𝑀 3 𝐿3 −𝑀 4 𝐿3 −𝑀 5 𝐿2 −𝑀 1 𝐿3 −𝑁 1 𝐿3 −𝑁 4,5 𝐿1 −𝑀 2 𝐿3 −𝑁 7 𝐿3 −𝑂1 𝐿3 −𝑂5 𝐿3 −𝑃 1 𝐿2 −𝑀 4 𝐿1 −𝑀 3 𝐿1 −𝑀 4 𝐿1 −𝑀 5 𝐿2 −𝑁 1 𝐿2 −𝑁 3 𝐿2 −𝑁 4 𝐿1 −𝑁 2 𝐿2 −𝑁 6 𝐿2 −𝑂1 𝐿1 −𝑁 3 𝐿2 −𝑂4 𝐿1 −𝑁 4 𝐿1 −𝑂2 𝐿1 −𝑂3 𝐿1 − 𝑃1

Energy (eV)

Width (eV) [9]

Siegbahn L𝜄 Lt Ls L𝛼2 L𝛼1 L𝜂 L𝛽6 L𝛽2 L𝛽4 L𝛽7 ’ L𝛽7 L𝛽5 L𝛽1 L𝛽3 L𝛽10 L𝛽9 L𝛾5 L𝛾1 L𝛾2 L𝜈 L𝛾8 L𝛾3 L𝛾6 L𝛾4 ’ L𝛾4

Intensity SMX3

Intensity [1]

relative to the total X-rays (%) 5.047 12124 12510 13497 14085.4 14280.9 16333 16498 17259 17557 17635.9 17707 17955 18067.8 18296.2 18541 19134 19329 20707 21143 21420.4 21723.2 21832.5 21917.4 21983.3 22152.4 22259 22821 22891 23075

21.32 26.82 22.82 15.82 11.22 11.02 29.5 19.52 11.82 28.5 7.82 7.82 7.82 7.82 13.9 21.5 16.9 16.7 22.2 17.5 14.5 22.5 10.5 10.5 20.5 19.52 10.5 13.5 13.5 13.5

Group intensity SMX3

Group intensity [13]

relative to the Lray groups 𝜂𝛽

0.0268 (12) 2.569 (12) 0.0306 (9) 0.0313 (9) 3.655 (17) 34.56 (10) 1.139 (8) 0.6351 (50) 8.265 (30) 0.2233 (27) 0.0537 (12) 0.1457 (21) 1.434 (8) 0.1303 (20) 35.54 (12) 0.1939 (25) 0.00763 (42) 0.0146 (6) 0.3142 (34) 0.0208 (7) 8.704 (37) 0.0751 (16) 0.0621 (13) 0.0639 (13) 0.0615 (13) 2.006 (13)

Not detected not deconvoluted

5441 (30)

5,3 (8)

not deconvoluted

80,05 (36)

72 (7)

1.171 (31) 0.574 (11) Not detected 0.528 (34) Not detected Not detected Not detected Not detected 35.94 (34) 0.264 (11) 0.0367 (11) 0.0264 (23) 0.321 (11) Not detected 8.70 (7) Not detected Not detected Not detected Not detected 2.044 (23)

100

100

23,75 (11)

22,4 (23)

0.0145 (6) 0.0171 (6) 0.00689 (38)

0.0367 (23) Not detected Not detected

After the processing of each X-ray region, the number of absorbed photons for each emission was determined. The intrinsic efficiency of SMX3 presents a slight variation from ∼99.5% to 99.7% below 24 keV. The effect of this efficiency variation on the emission intensities is taken into account and its uncertainty is also accounted for in the estimation of the intensity uncertainty. Finally, the number of absorbed photons is normalized by the total number of absorbed photons in the L-X-ray region.

which have been analyzed separately using Colegram. The spectrum shows different physical processes that take place after the IC of the 𝛾 transition. The daughter atom Pu-240 recovers the ground state electronic configuration by atomic rearrangement of the primary vacancies created by the internal conversion. Therefore, the L-X-ray spectrum of the Cm-244 is very complex to deconvolute. A diagram line Li –Yj (i = 1, 2, 3; j = 1, … , 7; Y = M, N, O, P) occurs when an outer electron from Yj fills the single vacancy in Li . For example, X L𝜄 region in Fig. 2 shows the L𝜄 diagram line (L3 – M1 ) fitted using a Voigt in green. The spectrum shows also intense satellite structures, due to transitions in presence of multiple vacancy states of the atomic configuration. The multiple vacancies can be produced by three processes: Shakeoff, Shake-up or Coster–Kronig transitions [10]. Because of the intense CK transition probabilitie 𝑓23 and because the primary vacancies are mainly produced in L2 , the satellite structures of X-rays are observed for L3 –Y𝑗 transitions following a L2 −L3 Yk CK transitions. For example, the above mentioned L𝜄 : (L3 – M1 ) emission has a satellite structure due to the high CK probability 𝑓23 (in comparison the probability of Shakeoff and Shake-up is small [11]): the primary vacancy is created by IC in the 𝐿2 and is moved by CK transition to the L3 . Therofore, the largest CK probability concerns the ejection of an electron from M5 during the transition L1 –𝐿3 M5 (according to [12]). The region X L𝜂𝛽 shows an intense satellite structure at the L𝛽2 : 𝐿3 −𝑁 5 . It presents a different energy shift compared to the satellite structure of the L𝜄 : 𝐿3 − 𝑀 1 (Fig. 2).

3. Decay data for Cm-244 Table 2 shows the relative intensities of the measured transitions leading to an L-X-ray line in the decay of Cm-244, they are presented with two nomenclatures: IUPAC and Siegbahn. Using the measured L-Xray lines, we determined the L-X-ray group intensities relatively to the group of L𝜂𝛽 in order to compare to the relative intensities measured in [13]; a good agreement is observed. In addition, we compared the relative emission intensities measured with SMX3 to the intensities measured in [1]. Authors in [1] measured the L-X-ray lines intensities using the coincidence technique based on a semiconductor detector; the uncertainties shown in this work are dominated by the statistical uncertainties. We determine the intensities of all the emissions from the same 𝐿𝑖 subshell, relative to the total intensity of the L-X-ray region and compared them with the calculated values using the ICC, the CK probabilities and the fluorescence yields for each 𝐿𝑖 subshell of Cm244 [15] (Table 1). A good agreement between the calculation and the SMX3 measurement is observed. This work shows the performances of an MMC, in particular SMX3, to measure with a high resolution and accuracy the X-ray intensities for metrological applications and fundamental physics.

The L2 − Yj transitions can present a satellite structure due to multiple vacancy states created by CK transitions. Nevertheless, in the decay of Cm-244, this type of satellite is not observed due to the weak probability and to the small shift of energy of the satellites with respect to the diagram line [14]. Therefore, satellite structures for 𝐿2 − 𝑌𝑗 and 𝐿1 −𝑌𝑗 transitions are not observed. Most of the L-X-ray emissions range from 11 to 24 keV. But, an L-X-ray emission was detected at 5 keV in the region of M-X-rays, the transition 𝐿1 − 𝐿3 never observed before is shown in Fig. 3. It was also fitted using a Voigt function.

References [1] P.N. Johnston, P.A. Burns, Absolute L X-ray intensities in the decays of 230 Th, 234 U, 238 Pu and 244 Cm, Nucl. Instrum. Methods A 361 (1995) 229–239. [2] M.-M. Bé, Bureau International des Poids et Mesures (Eds.), Table of Radionuclides, vol. 7: A=14 to 245, BIPM, Sèvres, 2013. 3

Please cite this article as: R. Mariam, M. Rodrigues and M. Loidl, Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter, Nuclear Inst. and Methods in Physics Research, A (2019), https://doi.org/10.1016/j.nima.2019.04.020.

R. Mariam, M. Rodrigues and M. Loidl

Nuclear Inst. and Methods in Physics Research, A xxx (xxxx) xxx

[3] M.C. Lépy, J. Plagnard, L. Ferreux, Measurement of 241 Am L X-ray emission probabilities, Appl. Radiat. Isot. 66 (6–7) (2008) 715–721. [4] J.L. Campbell, Fluorescence yields and Coster–Kronig probabilities for the atomic L subshells, At. Data Nucl. Data Tables 85 (2) (2003) 291–315. [5] A. Fleischmann, C. Enss, G.M. Seidel, Cryogenic particle detection, in: C. Enss (Ed.), Topics Appl. Phys., vol. 99, Springer, Berlin/ Heidelberg, 2005, pp. 151–216. [6] M. Rodrigues, R. Mariam, M. Loidl, A metallic magnetic calorimeter dedicated to the spectrometry of L X-rays emitted by actinides, EPJ Web of Conferences, vol. 146, 2017, p. 10012. [7] R. Mariam, M. Rodrigues, M. Loidl, Full-energy peak efficiency calibration of a metallic magnetic calorimeter detector for photon spectrometry below 100 keV, J. Low Temp. Phys. (2018). [8] H. Ruellan, M.C. Lépy, M. Etcheverry, J. Plagnard, J. Morel, A new spectra processing code applied to the analysis of 235 U and 238 U in the 60 to 200 keV energy range, Nucl. Instrum. Methods A 369 (2) (1996) 651–656. [9] J.L. Campbell, T. Papp, Widths of the atomic K–N7 levels, At. Data Nucl. Data Tables 77 (1) (2001) 1–56.

[10] M. Rodrigues, M. Loidl, L X-ray satellite effects on the determination of photon emission intensities of radionuclides, Appl. Radiat. Isot. 109 (2016) 570–575. [11] T. Mukoyama, Electron shake probabilities of heavy elements as a result of M-shell vacancy production: Shake probabilities as a result of M-shell vacancy production, X-Ray Spectrom. 44 (1) (2015) 7–12. [12] M.H. Chen, B. Crasemann, H. Mark, Relativistic radiationless transition probabilities for atomic K- and L-shells, At. Data Nucl. Data Tables 24 (1) (1979) 13–37. [13] Yu S. Popov, I.B. Makarov, D. Kh Srurov, E.A. Erin, M-and L-X-rays emission of actinides, Radiokhimiya 32 (2) (1990) 2–450. [14] F. Parente, M.H. Chen, B. Crasemann, H. Mark, L X-ray satellite energies, At. Data Nucl. Data Tables 26 (5) (1981) 383–466. [15] M.C. Lépy, B. Duchemin, J. Morel, Comparison of experimental and theoretical L X — ray emission probabilities of 241 Am, 239 Pu and 240 Pu, Nucl. Instrum. Methods A 353 (1994) 10–15.

4

Please cite this article as: R. Mariam, M. Rodrigues and M. Loidl, Determination of L-X-ray line emission intensities in the decay of Cm-244 with a metallic magnetic calorimeter, Nuclear Inst. and Methods in Physics Research, A (2019), https://doi.org/10.1016/j.nima.2019.04.020.