Nondestructive Elemental Depth-Profiling Analysis by Muonic X-ray

Letter pubs.acs.org/ac

Nondestructive Elemental Depth-Profiling Analysis by Muonic X‑ray Measurement Kazuhiko Ninomiya,*,† Michael K. Kubo,‡ Takashi Nagatomo,§ Wataru Higemoto,∥ Takashi U. Ito,∥ Naritoshi Kawamura,§ Patrick Strasser,§ Koichiro Shimomura,§ Yasuhiro Miyake,§ Takao Suzuki,⊥ Yoshio Kobayashi,# Shinichi Sakamoto,○ Atsushi Shinohara,† and Tsutomu Saito⊗ †

Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan College of Liberal Arts, International Christian University, Mitaka, Tokyo 181-0015, Japan § Institute of Materials Structure Science, High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305-0801, Japan ∥ Advanced Science Research Center, Japan Atomic Energy Agency, Tokai, Ibaraki 319-1195, Japan ⊥ College of Engineering, Shibaura Institute of Technology, Saitama-shi, Saitama 337-8570, Japan # Graduate School of Engineering Science, University of Electro-Communications, Chofu, Tokyo 182-8585, Japan ○ J-PARC Center, Japan Atomic Energy Agency, Tokai, Ibaraki 319-1195, Japan ⊗ National Museum of Japanese History, Sakura, Chiba 285-8502, Japan ‡

ABSTRACT: Elemental analysis of materials is fundamentally important to science and technology. Many elemental analysis methods have been developed, but three-dimensional nondestructive elemental analysis of bulk materials has remained elusive. Recently, our project team, dreamX (damageless and regioselective elemental analysis with muonic X-rays), developed a nondestructive depth-profiling elemental analysis method after a decade of research. This new method utilizes a new type of probe; a negative muon particle and high-energy muonic X-rays emitted after the muon stops in a material. We performed elemental depth profiling on an old Japanese gold coin (TempoKoban) using a low-momentum negative muon beam and successfully determined that the Au concentration in the coin gradually decreased with depth over a micrometer length scale. We believe that this method will be a promising tool for the elemental analysis of valuable samples, such as archeological artifacts.

E

level that are generated in midair by primary cosmic rays. Large accelerator facilities are needed to produce intense muon beams for scientific research. Positively charged muons, which can be regarded as light proton, are widely used in material science research for muon spin rotation, relaxation, and resonance (μSR) methods. By contrast, negatively charged muons can generate muonic atoms that contain one muon in an atomic orbital instead of an electron. Muonic atoms have unique properties due to the large mass of the muon; the atomic muon orbit is much smaller than that of an electron, the muon is in close proximity to an atomic nucleus, and the binding energy of each atomic muon level is very high. As a result, the energies of the characteristic muonic X-rays emitted after muonic atom formation are very high, and such high-energy photons can penetrate several centimeters of material. For example, the energy of muonic hydrogen KαX-ray is 1.9 keV, 75 keV for carbon and 1.5 MeV for copper. This is an important and

lemental analysis is one of the most fundamental and essential techniques for all research fields related to the natural sciences. Many elemental analysis methods have been developed, for example, chemical separation, photofluorescence and absorption spectrometry, mass spectrometry, and applications of some nuclear processes such as nuclear activation. These elemental analysis methods are highly developed and extremely sensitive and require small amounts of samples, but some of them are destructive. Nondestructive analysis methods are of great benefit to the practicing scientist, and some useful methods that leave the sample unharmed have been developed. For example, 3D micro X-ray fluorescence (XRF) analysis1 and neutron activation analysis have been established. However, an analysis method that is nondestructive, three-dimensionally position selective and can detect multiple elements including light ones for a bulk sample has remained elusive. Here, we propose a new probe for elemental analysis: negatively charged muons. Negatively charged muons are elementary particles that have the same charge as an electron but are about 200 times heavier (105.658 MeV/c2). Muons are among the main constituents of cosmic rays found at ground © XXXX American Chemical Society

Received: December 9, 2014 Accepted: April 22, 2015

A

DOI: 10.1021/acs.analchem.5b01169 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry superior feature of muonic X-rays compared to electronic Xrays. In addition, the muon stopping depth can be controlled by adjusting the incident muon energy, which can be easily achieved by tuning the magnet system for muon beam transportation in accelerator facilities. In this way, when we select optimum muon momentum, all incident muons are stopped in the material and emit a series of cascading muonic X-ray.2 A new elemental analysis method that is nondestructive, position-selective and can analyze multiple elements for a bulk sample should be possible with negative muons. Some pioneering work on elemental analysis using negative muons has been pursued in the past. The possibility of elemental analysis with muonic X-rays was first discussed over 40 years ago.3,4 Later some fundamental studies were reported for archeological artifacts5−7 and applications in medical science were considered.8 However, because of several limitations of muon beams, such as low intensity and large muon momentum spread, a detailed and quantitative investigation of elemental analysis with muons has never been performed. Our project team, dreamX, has been developing an elemental analysis method with muonic X-rays over the past decade.7 Recently, intense and high-quality muon beams have become available at the muon science facility (MUSE) in the Japan Proton Accelerator Research Complex (J-PARC).9 At this facility, we performed elemental analysis studies of some archeological artifacts10,11 and developed an elemental analysis method for sealed materials.12 In this study, we applied the elemental analysis method to archeological samples and achieved quantitative nondestructive micrometer-scale depth profiling using standard materials.

Figure 1. Schematic view of the experimental setup. The vacuum chamber was directly connected to the beamline duct. The sample was oriented at 45° to the muon beam. Two high-purity germanium detectors were set in the outer side of the vacuum chamber near the 100 μm aluminum foils.

inner part was demonstrated by Auger electron analysis with ion spattering. However, an elemental composition bias on the order of micrometers has never been investigated without destroying the sample, until now. In this study, we performed nondestructive elemental depth profiling by tuning the incident muon energy to change the muon stopping depth. We selected 200, 270, 300, 490, and 1000 keV for the muon irradiation energy, which correspond to central stopping depths of 2.1, 3.1, 3.4, 6.6, and 20.6 μm for muons in the gold plate.

■

EXPERIMENTAL SECTION The muon irradiation experiments were performed in Japan at MUSE, which is located at J-PARC/MLF (Materials and Life Science Experimental Faculty), the world’s most intense pulsed muon facility. An intense low-energy muon beam is available at MUSE because of the high energy of the primary proton beam.13 The muon irradiated samples were set into the vacuum chamber that was directly connected to the beamline. Characteristic muonic X-rays emitted after muons stop in the sample were measured by two high-purity germanium detectors (GMX20P4-70, SEIKO EG&G ORTEC) to get higher statistics during limited beam time. Because the energies of muonic Xrays are very high and the absorption of muonic X-rays by foil vacuum window of the chamber and air is low, the detectors can be placed on the outer side of the vacuum chamber. The schematic view of the experimental setup is shown in Figure 1. The shape of muon beam was normal distribution with the size of 5 cm in vertical and 3 cm in horizontal direction in 1σ. There are no vacuum foils between the sample and the superconducting solenoid magnet system where the muons are generated; only a thermal shield of 12.5 μm aluminum foil is placed in the sample path. With this setup, we can use extremely low energy muons below 100 keV, which correspond to 5 MeV/c in momentum. In this study, we selected an old Japanese gold coin for elemental analysis. The inside of Tempo-koban (19th century) coin is an alloy of Au−Ag with 57 wt % Au, as determined by electron beam excitation X-ray analysis.14 Interestingly, the Tempo-koban has a gold-rich layer of several micrometers near the surface that is responsible for the gold color which is achieved by the color-dressing technique “Iroage”.14 The changes of elemental composition from the surface to the

■

RESULTS AND DISCUSSION The characteristic muonic X-ray spectra of the Tempo-koban with 200 and 1000 keV incident muons are shown in Figure 2. Muonic X-rays originating from muon capture in Au and Ag atoms were clearly identified. Some gamma-ray lines from radioactive nuclides produced by muon absorption of Au and Ag nuclides were also identified.15 The X-ray intensity corresponds to the amount of muons stopped in each element, that is, the elemental composition of the sample. We can easily determine that the Tempo-koban was mainly made of Au and Ag and that the presence of other elements was relatively low. In this study, we focused on the relatively strong muonic X-rays for elemental identification of Au and Ag: μAu(5-4) and μAg(43). The numbers in parentheses indicate the change of principal quantum numbers of the muon by muonic X-ray emission. The X-ray intensity ratios of μAu(5-4)/μAg(4-3) were clearly different to the energies of incident muons as shown in Figure 2. This shows that the elemental composition of the Tempokoban changed with the muon stopping depth. For quantitative analysis, we also performed muon irradiation on three standard Au−Ag alloys with 50, 60, and 80 wt % Au. The relationship between the X-ray intensity ratios of μAu(54)/μAg(4-3) and the elemental compositions is shown in Figure 3. The muonic X-ray intensity of μAu(5-4) increases linearly with the elemental composition of Au in the sample. A proportional relationship is established between the ratios of muonic X-ray intensity and elemental composition, at least in the region of 50−80 wt % Au in the Au−Ag alloy. Such a B

DOI: 10.1021/acs.analchem.5b01169 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

Figure 2. Muonic X-ray spectra for the Tempo-koban with 200 and 1000 keV incident muon energy. The irradiation time of each irradiation was 32 and 7 h. Muonic X-rays emitted from muons that changed principle quantum number from 5 to 4 in Au atoms were written as μAu(5-4). Other Xrays are described in the same way. Each peak was identified from the reported muonic X-ray energy.15 The muonic X-rays emitted after muon capture in Au and Ag atoms were clearly identified. The small figures on the right side are the expanded spectra of the μAg(4-3) and μAu(5-4) X-ray peak regions that were corrected by intensities of μAg(4-3).

solid line. In this model, the Tempo-koban comprises four layers; 2 μm of Au 90 wt %, 2 μm of Au 80 wt %, 3 μm of Au 60 wt %, and finally Au 57 wt % in the deeper region. This depth profile was taken from the results of previous destructive analysis.14 The density of each layer was obtained from the reported literature.17 The stopping depth of 200, 270, 300, 490, and 1000 keV incident muons in the sample was calculated to be 2.2, 3.1, 3.4, 6.7, and 21.2 μm, respectively, using the Bethe− Bloch equation.18 Note that the sample was set at 45° against the beam axis, which makes the sample thickness effectively 1.4 times larger than if it were set perpendicular to the beam. The Au concentration with depth of the Tempo-koban determined from the calibration curve was summarized in Figure 4. The uncertainties of the analyzed values were derived from statistical errors of the muonic X-ray counts and systematic errors arising from the detection efficiency calculation (5%). In the analysis, we used the added X-ray counts of the two detectors. A small contribution from X-ray absorption by the sample itself was also taken into account from photon transportation simulations using a Monte Carlo calculation method as implemented in the EGS5 code.19 As shown in Figure 4, a high Au concentration was identified near the surface. Such a high concentration of Au was found only at the surface of the sample and the Au concentration of the sample rapidly decreased with depth. The concentration of Au reached about 57 wt % within 5 μm of the surface which continued into the deeper region. The same result was also obtained from gamma-ray intensities. The Au concentration of the deeper region agreed with the previous analysis performed by muon irradiation using higher energy incident muons10 and other destructive analysis methods.14 The depth profile of the Au concentration is consistent with that determined by electron impact analysis,14 although a small deviation in the absolute values was present. Such a deviation came from the difference in analysis area. Because the spatial spread of the muon beam in this work was larger than the size of the Tempo-koban itself,14 the results obtained by muon irradiation are average values for the whole sample surface, whereas the electron impact analysis represents only a tiny area investigated by the well-focused electron beam. The uniformity of elemental composition in archeological artifacts was conceivable because of a small chip on the surface of the coin and/or insufficient manufacturing technique in “Iroage”.

Figure 3. Relationship between elemental mass ratios and the ratio of the muonic X-ray intensities for Au−Ag alloys. A proportional relationship exists between the region of 50−80 wt % Au in the Au− Ag alloy.

relationship has been obtained in other alloy systems.16 From this relationship, we can determine the elemental composition ratios of alloys of Au−Ag quantitatively using only the muonic X-ray intensities. To estimate the muon stopping depth in the Tempo-koban, we postulate the sample structure as shown in Figure 4 by the

Figure 4. Depth-profiling elemental compositions of the Tempokoban. The solid line corresponds to the Au composition from a range of calculations taken from the literature.14 The open circles show the analyzed values as determined by muonic X-ray measurements. The Au-rich layer near the surface was clearly identified. The concentration of Au at each depth was quantitatively determined without sample destruction. Note that the muon stopping depths were calculated for pure Au metal and the Tempo-koban was placed at 45° to the beam direction.

■

CONCLUSIONS Nondestructive quantitative elemental analysis of an archeological artifact was performed. The micrometer-order depthC

DOI: 10.1021/acs.analchem.5b01169 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry profiling of the Tempo-koban revealed different elemental compositions between the surface and at depths of several micrometers. We are now possible using a new elemental analysis method based on muonic X-rays that is nondestructive and regioselective for bulk material by using a high-intensity negatively charged muon beam.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: +81-6-68506999. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by the Grant-in-Aid for Young Scientists B (JSPS KAKENHI, Japan, Grant Number 26800213).

■

REFERENCES

(1) Mantouvalou, I.; Malzer, W.; Schaumann, I.; Lühl, L.; Dargel, R.; Vogt, C.; Kanngiesser, B. Anal. Chem. 2008, 80, 819−826. (2) Vogel, P. Phys. Rev. A 1980, 22, 1600−1609. (3) Rosen, L. Science 1971, 173, 490−497. (4) Daniel, H. Nuclear-Medizin 1969, 4, 311−319. (5) Köhler, E.; Bergmann, R.; Daniel, H.; Ehrhart, P.; Hartmann, F. J. Nucl. Instrum. Methods Phys. Res. 1981, 187, 563−568. (6) Daniel, H.; Hartmann, F. J.; Köhler, E. Fresenius Z. Anal. Chem. 1985, 321, 65−67. (7) Kubo, M. K.; Moriyama, H.; Tsuruoka, Y.; Sakamoto, S.; Koseto, E.; Saito, T.; Nishiyama, K. J. Radioanal. Nucl. Chem. 2008, 278, 777− 781. (8) Hosoi, Y.; Watanabe, Y.; Sugita, R.; Tanaka, Y.; Nagamine, K.; Ono, T.; Sakamoto, K. Br. J. Radiol. 1995, 68, 1325−1331. (9) Kadono, R.; Miyake, Y. Rep. Prog. Phys. 2012, 75, 026302. (10) Ninomiya, K.; Nagatomo, T.; Kubo, M. K.; Strasser, P.; Kawamura, N.; Shimomura, K.; Miyake, Y.; Saito, T.; Higemoto, W. J. Phys.: Conf. Ser. 2010, 225, 012040. (11) Ninomiya, K.; Nagatomo, T.; Kubo, K.; Ito, T. U.; Higemoto, W.; Kita, M.; Shinohara, A.; Strasser, P.; Kawamura, N.; Shimomura, K.; Miyake, Y.; Saito, T. Bull. Chem. Soc. Jpn. 2012, 85, 228−230. (12) Terada, K.; Ninomiya, K.; Osawa, T.; Tachibana, S.; Miyake, Y.; Kubo, M. K.; Kawamura, N.; Higemoto, W.; Tsuchiyama, A.; Ebihara, M.; Uesugi, M. Sci. Rep. 2014, 4, 5072. (13) Shimomura, K.; Koda, A.; Strasser, P.; Kawamura, N.; Fujimori, H.; Makimura, S.; Higemoto, W.; Nakahara, K.; Ishida, K.; Nishiyama, K.; Nagamine, K.; Miyake, Y. Nucl. Instrum. Methods Phys. Res., Sect. A 2009, 600, 192−194. (14) Ueda, M.; Taguchi, I.; Saito, T. Institute for Monetary and Economic Studies, 1996, Discussion Paper 1996-E-26. (15) Evans, H. J. Nucl. Phys. A 1973, 207, 379−400. (16) Daniel, H.; Hartmann, F. J.; Naumann, R. A. Phys. Rev. A 1999, 59, 3343−3348. (17) Kraut, J. C.; Stern, W. B. Gold Bull. 2000, 33, 52−55. (18) Hatano, Y.; Katsumura, T.; Mozumder, A. Charged Particle and Photon Interactions with Matter: Recent Advances, Applications, and Interfaces; CRC Press: Boca Raton, FL, 2011. (19) Hirayama, H.; Namito, Y.; Bielajew, A. F.; Wilderman, S. J.; Nelson, W. R. SLAC-Report-730, Stanford Linear Accelerator Center: Stanford, CA, 2005.

D

DOI: 10.1021/acs.analchem.5b01169 Anal. Chem. XXXX, XXX, XXX−XXX