Fiber-optic laser-induced breakdown spectroscopy of zirconium metal in air: Special features of the plasma produced by a long-pulse laser

Spectrochimica Acta Part B 142 (2018) 37–49

Contents lists available at ScienceDirect

Spectrochimica Acta Part B journal homepage: www.elsevier.com/locate/sab

Fiber-optic laser-induced breakdown spectroscopy of zirconium metal in air: Special features of the plasma produced by a long-pulse laser Ayumu Matsumoto a,⁎, Hironori Ohba a,b, Masaaki Toshimitsu a, Katsuaki Akaoka a, Alexandre Ruas a, Tetsuo Sakka c, Ikuo Wakaida a a b c

Collaborative Laboratories for Advanced Decommissioning Science, Japan Atomic Energy Agency, Tokai-mura, Naka-gun, Ibaraki 319-1195, Japan Tokai Quantum Beam Science Center, National Institutes for Quantum and Radiological Science and Technology, Tokai-mura, Naka-gun, Ibaraki 319-1106, Japan Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto University, Nishikyo-ku, Kyoto 615-8510, Japan

a r t i c l e

i n f o

Article history: Received 15 August 2017 Received in revised form 12 December 2017 Accepted 23 January 2018 Available online xxxx Keywords: Fukushima Daiichi decommissioning Long-pulse laser Fiber-optic LIBS Zirconium

a b s t r a c t The decommissioning of the Tokyo Electric Power Company (TEPCO) Fukushima Daiichi Nuclear Power Plant is an essential issue in nuclear R&D. Fiber-optic laser-induced breakdown spectroscopy (Fiber-optic LIBS) could be used for in-situ elemental analysis of the inside of the damaged reactors. To improve the performances under difficult conditions, using a long-pulse laser can be an efficient alternative. In this work, the emission spectra of zirconium metal in air obtained for a normal-pulse laser (6 ns) and a long-pulse laser (100 ns) (wavelength: 1064 nm, pulse energy: 12.5 mJ, spot diameter: 0.35 mm) are compared to investigate the fundamental aspects of fiber-optic LIBS with the long-pulse laser. The spectral features are considerably different: when the long-pulse laser is used, the atomic and molecular emission is remarkably enhanced. The enhancement of the atomic emission at the near infrared (NIR) region would lead to the observation of emission lines with minimum overlapping. To understand the differences in the spectra induced respectively from the normal-pulse laser and the long-pulse laser, photodiode signals, time-resolved spectra, plasma parameters, emission from the ambient air, and emission regions are investigated, showing the particular characteristics of the plasma produced by the long-pulse laser. © 2018 Elsevier B.V. All rights reserved.

1. Introduction The Tokyo Electric Power Company (TEPCO) Fukushima Daiichi Nuclear Power Plant was seriously damaged by the tsunami caused by the earthquake on March 11, 2011 in Japan. The nuclear fuels are considered to be melted with the fuel cladding tubes, the control rods, the structural materials, the concretes, etc. The removal of the nuclear fuel debris from the inside of the damaged reactors where the radiation level is extremely high is very challenging. Various approaches have been used to improve the understanding of the reactor conditions, e.g. accident sequence analyses [1–8], evaluation of simulated debris [9–14], cosmic ray muon tomography [15,16], and internal observation with small robots (e.g. ref. [17]). However, at present, the elemental composition of the debris has yet to be confirmed. Fiber-optic laser-induced breakdown spectroscopy (fiber-optic LIBS) is an in-situ analytical technique in which the emission spectroscopy of the laser ablation plasma is performed through an optical fiber cable. Because of the capability of remote analysis and the flexibility of the cable, this technique has been applied to various fields [18–25] including nuclear power plant [19]. The transmission of the laser pulse ⁎ Corresponding authors. E-mail address: [email protected] (A. Matsumoto).

https://doi.org/10.1016/j.sab.2018.01.012 0584-8547/© 2018 Elsevier B.V. All rights reserved.

and the plasma emission through the cable enables us to probe into the damaged reactors without radiation exposure to the field workers and the electronics. Our group has developed a fiber-optic LIBS instrument combined with a radiation-resistant optical fiber and successfully analyzed simulated debris [26]. In practice, the head of the focusing optics should be miniaturized to access the inside of the damaged reactors with the narrow spaces. This limits the performances of the laser focusing and the light collection. To improve the analytical capability, the enhancement of the output energy and the plasma emission is important. Also, the analysis of submerged materials [26] and contaminated water [27] is required, because a part of the damaged reactors is underwater. In a previous work, a gas-flow system was employed [26], which enables us to analyze solid surfaces by ejecting the surrounding water [19,28]. However, the analysis of sediments and suspended particles is difficult to perform with the previous instrument. In this context, the use of a long-pulse laser (~100 ns) instead of a commonly used normal laser (~10 ns) is a promising candidate. The peak power of the long-pulse laser is much lower than that of the normal-pulse laser for a same pulse energy, which allows the delivering of a higher laser energy to samples through the fiber without damaging it. Also, Sakka et al. have reported that a longer pulse duration gives intense and narrow emission lines in the water-confined geometry [29].

38

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

Because of its single-pulse scheme and applicability to high pressure environments [25,30,31], long-pulse LIBS has been widely studied for the analysis underwater (e.g. refs [32–39].). However, the number of reports related to long-pulse LIBS in gaseous phase is still lacking. For example, Yamamoto et al. have evaluated the practicability of an acousto-optic (AO) Q-switched long-pulse laser (150 ns) with a pulse energy of ~10 mJ and a repetition rate up to 6 kHz for the analysis of aluminum, steel, soil samples, and surface contaminations [40]. Elnasharty has studied the effects of the pulse duration on the analysis of aluminum samples using a long-pulse fiber laser (40–200 ns) with a pulse energy of 200 μJ and a repetition rate of 25 kHz [41]. In the present Fukushima Daiichi decommissioning application where the laser beam cannot be tightly focused, a higher pulse energy is important rather than a high repetition rate, which can be achieved by an electro-optic (EO) Q-switched long-pulse laser. The CO2 laser (~100 ns) is also a common long-pulse laser [42–47], but the longer wavelength (10.6 μm) is significantly absorbed by water and special materials are needed for the optics. Since the transmission property of the radiation-resistant fiber is not deteriorated at the near infrared (NIR) region even in the high-radiation fields [48], the wavelength of 1064 nm is a more reasonable choice. Zirconium metal is an important material to evaluate before analyzing more complex systems, because it is a main component of the fuel cladding tubes [10,11,13]. In this context, the emission spectrum of zirconium metal in air is measured using an EO Q-switched long-pulse laser (100 ns) with a wavelength of 1064 nm. The spectrum is compared to that obtained by a normal-pulse laser (6 ns) keeping a same output energy of 12.5 mJ at a spot diameter of 0.35 mm. To understand the differences of the spectra, photodiode signals, time-resolved spectra, plasma parameters, emission from the ambient air, and emission regions are investigated. 2. Experimental Fig. 1 shows the experimental setup for fiber-optic LIBS. In this work, a flash-lamp-pumped EO Q-switched Nd:YAG laser (Continuum, Minilite II) with a pulse duration of 6 ns and a diode-pumped EO Qswitched Nd:YAG laser (OK Lab. Co., Ltd., OKL-LIBSLPL-8000) with a pulse duration of 100 ns were used. The lasers were operated at a wavelength of 1064 nm and a repetition rate of 5 Hz. The pulse duration was measured at the output of the focusing optics. The laser beam to be examined was introduced into a quartz fiber (Mitsubishi Cable Industries, Ltd., PVSMAMSL) (delivery fiber) with a core diameter of 1.0 mm, a numerical aperture (NA) of 0.12, and a length of 2.5 m through a plano-convex lens (focal length: 125 nm) in a home-made coupling optics inside an airtight box. To avoid the damage at the entrance and the inside of the fiber, the focusing position was shifted, i.e. the distance from the lens to the end face of the fiber was 155 mm. Also, to avoid air breakdown, the inside of the box was roughly evacuated using a dry pump (AS ONE corporation, LMP-100) to reach a reduced pressure of 13.3 kPa. Note that the spatial distribution of the laser beam is considered to be averaged out by passing through the

fiber cable and both the pulses show a top-hat beam profile. Thus, the difference of the transverse mode could be negligible in the present experiments. The delivered beam was focused in a normal direction to the sample surface using a focusing optics (OK Lab. Co., Ltd., OKL-FPNC-IA2000) with a focal length of 80.34 mm, a reduction rate about 0.35, a working distance of 12.6 mm, and an outer diameter below 30 mm. The focusing optics was designed for collinear detection with a single fiber, and the chromatic aberration from 500 nm to 1064 nm was corrected. The distance between the focusing optics and the sample was previously adjusted so that the spot size of a He\\Ne laser (wavelength: 632.8 nm) was minimum at the sample surface. The spot diameter was about 0.35 mm. The laser irradiation was performed keeping the output energy at 12.5 mJ for both the pulses (fluence: 13.0 J/cm2, irradiance: 2.2 GW/cm2 (6 ns) and 0.130 GW/cm2 (100 ns)). In the present conditions, the input energy was sufficiently lower than the damage threshold of the delivery fiber for the 6-ns pulse laser, and the output energy was sufficiently higher than the ablation threshold of the sample for the 100-ns pulse laser. As a sample, a zirconium metal plate (The Nilaco Corporation, ZR493402, 99.2%) with a thickness of 0.50 mm was used. The target was cleaned in pure water and ethanol using an ultrasonic cleaner after the surface was polished by an emery paper (#1000). The sample was placed on an automatic rotating stage and the irradiation spot was changed shot-by-shot. Emission spectroscopy was performed using an echelle spectrograph (Catalina Scientific Instruments, LLC, EMU-120/65 UV/VIS/NIR) equipped with an electron multiplying charge coupled device (EMCCD) camera (Raptor Photonics Limited, FA285B-CL) to adapt the numerous number of emission lines from nuclear reactor materials. In echelle spectrograph, horizontally-dispersed high-order diffraction light is vertically dispersed onto the detector for each diffraction order, and those with different wavelength region are connected, which gives high-resolution and wide range spectra. In EMCCD camera, the electrons are multiplied before the readout noise is generated, which enhances the signal without disturbing the spectral resolution. Although the focusing optics is designed for the collinear detection, the plasma emission was collected at the angle of about 30 degrees from the target surface using a light collector composed of two quartz plano-convex lenses (focal length: 40 mm, diameter of 25.4 mm) to observe the ultraviolet (UV) light. The collected light was delivered to the spectrograph through the fiber (SN402) (detection fiber) with a core diameter of 0.4 mm, which is an attachment of the spectrograph. In the spectral measurement, the delay time, the exposure time, the EMCCD gain, and the number of accumulation were set respectively to 1.0 μs, 50 μs, 2800 and 100. Note that 20% of the peak height at the falling edge of the laser pulse was defined as the delay time origin (td = 0 ns) to investigate the plasma evolution after the laser-plasma interaction. The recorded images were converted to intensity curves within a wavelength range from 190 to 1100 nm and a step width (interval between adjacent pixels) of 0.015 nm. In this work, a grating cassette HR1: UV/VIS/NIR with a slit width of 14 μm and a slit height of 41 μm

Fig. 1. Experimental setup for fiber-optic LIBS.

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

was used to observe wide-range spectra without the gaps at the connection parts of each wavelength part of echelle spectra. The resolving power was about 15,000 for the Hg I line at 564.1 nm. The intensity calibration was performed using a xenon lamp (Hamamatsu Photonics K. K., L7810–02) and a halogen lamp (Labsphere, Inc., USS-600) in combination. Temporal profiles of the laser pulse and the total emission intensity of the plasma were measured using a photodiode (Thorlabs, DET10A) with a rising time of 1 ns and a wavelength range from 200 to 1100 nm. The horizontal distance from the ablation point to the detection point was 5 cm and the height difference between the target surface and the detection point was 5 mm. For the measurement of the laser pulse profile, the scattered light was detected without the ablation by putting a paper in a position slightly deviated from the focal point. In the case of plasma emission, a YAG cut filter was used to avoid the detection of the laser pulse. The photodiode signal was observed using an oscilloscope (Iwatsu Electric Co., Ltd., DS-5354) with a bandwidth of 500 MHz, a rising time of 750 ps, a sampling rate of 1 GS/s, and an average number of 8. Five signals were recorded for each measurement. Time-resolved spectroscopy was also performed with an exposure time of 1.0 μs, which is the minimum value of the EMCCD camera. The delay time was changed from 0.5 μs to 5.0 μs, and five spectra were recorded for each delay. The smearing and the blooming were negligible. In the spectral analysis, multi-peak fitting 2 in Igor Pro 7 (WaveMetrix) was used with a Lorentzian function and a constant baseline. Fast imaging of the plasma emission was performed from the direction horizontal to the target surface using an objective lens (Mitsutoyo, M Plan Apo NIR 10×) and an intensified CCD (ICCD) camera (Andor Technology Ltd., iStar DH334T-18H-03). The objective lens was connected to the ICCD camera without any other imaging system. Then, the magnification rate was 2.0, and the distortion of the image was negligible. The gate width and the gain of ICCD were set to 10 ns and 500, respectively. The delay time was changed from 0.1 μs to 10 μs. Neutral density (ND) filters were used to avoid the saturation. Shadowgraph measurement was also performed using a xenon flash lamp as a back illumination. 3. Results and discussion 3.1. Emission spectra Fig. 2 shows the emission spectra of zirconium metal in air. The spectral features were considerably different depending on the pulse duration. In the case of the normal-pulse laser, the emission at the near UV (NUV) region was relatively intense. In the case of the long-pulse laser, the emission from the visible (VIS) to the NIR region was rather intense. Fig. 3 shows the enlarged views of the spectra. When the longpulse laser was used, the peak height of the ionic lines (Zr II) displayed here were less intense, while the atomic lines (Zr I) at the NIR region

were stronger, and the ZrO radical emission was remarkably enhanced. In Fig. 3c, a (0,0) sub-band of the ZrO beta-system (e3Π2 → a3Δ3) [49,50] is shown. The ZrO molecules can be formed by the reaction between the Zr atoms ejected from the target surface and the oxygen of the ambient air. The oxygen of the natural oxide thin film on the zirconium surface (thickness b2.5 nm [51]) might also be an origin of the ZrO molecules. The enhancement of the atomic emission in the case of the longpulse laser is an important benefit because the transmission property of the radiation-resistant fiber is not deteriorated at the NIR region even in the high-radiation fields [48]. Also, the line overlapping can be avoided at the NIR region even in the case of multi-element samples of nuclear reactor materials with high number of emission lines [26]. It can be confirmed in Fig. 3 that the line density at the NIR region is quite lower than that at the NUV region. Also, the enhancement of the molecular emission is useful to analyze elements difficult to detect with their atomic and ionic lines [52], and to perform isotopic analysis [53–55], which must be required for the analysis of the debris. It has been known that femtosecond laser gives better spectra to analyze the molecules directly ablated from organic samples [56–58]. Long-pulse laser could be a tool to observe the molecular emission from metal targets in air. 3.2. Photodiode signals Fig. 4 shows the temporal profiles of the plasma emission measured by the photodiode. When the pulse duration is shorter, the rising time of the plasma emission is shorter and the peak height is higher. This indicates that the temperature rising of the initial plasma is faster and the maximum temperature is higher. When the pulse duration is longer, the rising time is slower and the peak height lower. Considering the falling edge of the plasma emission, the decay appears to be faster in the case of the normal-pulse laser. This suggests that the plasma is governed by the continuous emission (bremsstrahlung and radiative recombination [59]) and the emission from the multivalent ions [60] which disappear at early time in general. In the case of the long-pulse laser, the decay is slower. This suggests that the atomic and molecular emission which is generally observed at delayed times has a more important relative contribution. Fig. 5 shows the rising edge of the plasma emission in comparison to the laser pulse profile and the tailing part of the long-pulse laser. A first important factor that explains the differences of the plasma is the lasertarget interaction. The plasma emission was detected respectively 2.8 ± 0.7 ns (6-ns pulse laser) and 71.2 ± 2.7 ns (100-ns pulse laser) after the laser beam reached the target surface. When the pulse duration is shorter, the target surface can be quickly heated to a higher temperature that allows an early formation of the plasma. Then, the rate of ejection can be considered as faster. Alternatively, in the case of the long-pulse laser, the kinetic energy of the ejected species can be lower. As shown in Fig. 5, the rising edge of the Q-switched long-pulse laser is gradual,

1.0 intensity (a. u.)

39

6 ns 100 ns

0.5

0 300

400

500 600 wavelength (nm)

700

800

Fig. 2. Emission spectra of zirconium metal in air obtained for (a) a 6-ns pulse laser (red line) and (b) a 100-ns pulse laser (blue line). The delay time and the exposure time were set to 1.0 μs and 50 μs, respectively. The intensities of both the spectra were normalized to the maximum value of the spectrum obtained for the 6-ns pulse laser. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

1.2

(a)

6 ns 100 ns

Zr II

Zr II

0.8

Zr II Zr II Zr II

0.4

1.0 intensity (a. u.)

40

6 ns 100 ns

0.5

0

0 350

352

354

356

0

1.0

500 1000 time (ns)

intensity (a. u.)

(b) Zr I

Zr I

0.5 Zr I

0 807 0.6

809

811

813

(c) (0,0)

0.4

(1,1) (2,2)

0.2 0 572

574 576 578 wavelength (nm)

Fig. 3. Enlarged views of the spectra in Fig. 2. (a) Zr II lines at the NUV region, (b) Zr I lines at the NIR region, and (c) a molecular band of the ZrO beta-system obtained for a 6-ns pulse laser (red line) and a 100-ns pulse laser (blue line). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

unlike the long-pulse fiber laser (the rising edge is steep regardless of the pulse duration [41]). This feature could delay the plasma ignition and cause a mild excitation of the primary plasma. The plasma characteristics could be more susceptible to the pulse duration when the Qswitched laser is used. A second important factor is the overlapping of the plasma emission with the laser irradiation (see Fig. 5), i.e. the laser interaction with the ejected species. The overlapping time of the laser pulse and the plasma

Fig. 4. Temporal profiles of the total emission intensity of the plasma produced on zirconium metal in air by a 6-ns pulse laser (red line) and a 100-ns pulse laser (blue line). The intensities of both the signals were normalized by the peak height of the signal obtained for the 6-ns pulse laser. The time axis was shifted so that the beginning of both the signals was the same. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

emission is 10.8 ± 0.5 ns for the 6-ns pulse laser and 171.4 ± 3.0 ns for the 100-ns pulse laser. The overlapping time was calculated from the beginning of the plasma emission to 20% of the maximum irradiation at the falling edge of the laser pulse. For the shorter pulse duration, the laser light, with a higher energy density, is instantaneously given to the primary plasma. This accelerates the electron movement in the plasma by the inverse bremsstrahlung [61,62], resulting in a higher temperature of the initial plasma. For the longer pulse duration, the laser light, with a lower energy density, is slowly given to the expanding plasma. This suppresses the efficiency of the plasma shielding, and the initial temperature becomes lower. Such a difference is due to the examined range of the pulse duration, which is close to the time scale of the plasma evolution. In the case of femtosecond LIBS, the laser irradiation is finished before the plasma generation (e.g. refs [63–67].). A third important factor is the compression of the ablated species. Since the primary plasma expands at a supersonic speed, the ambient gas molecules are strongly compressed and a shock wave is generated. Then, the successively incoming species are also compressed behind the shock front [42–44,68,69]. When the pulse duration is shorter, the primary plasma can expand faster. Therefore, the compression should be stronger, producing a highly dense plasma. In this environment, the species can collide each other more frequently, i.e. the species can be easily ionized and dissociated. When the pulse duration is longer, the compression is less effective. Thus, the density of the initial plasma is expected to be lower, resulting in lower degrees of ionization and dissociation. Bogaerts and Chen have numerically investigated the effects of the pulse duration (from 0.3 to 30 ns) on the initial stage of the ablation (until 100 ns) of copper target in helium gas with 266 nm laser at a constant fluence (10.6 J/cm2) [70]. Although the experimental parameters are different, some of the theoretical predictions are qualitatively comparable with our considerations above, e.g. the maximum values of surface temperature, evaporation rate, plume temperature, and electron number density become smaller with extending the pulse duration [70]. 3.3. Time-resolved spectra Time-resolved spectra were measured starting from a delay time of 0.5 μs, and the intensities of one ionic, one atomic, and one molecular emission were plotted as a function of the delay time (Fig. 6). The

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

1.5

(a)

6 ns, laser 6 ns, emission

1.0 0.5 0 0 intensity (a. u.)

1.5

10

(b)

20

30

100 ns, laser 100 ns, emission

1.0 0.5 0 0 0.04

200

(c)

400 100 ns, laser

0.02 0 0

1000 time (ns)

2000

Fig. 5. Rising edge of the plasma emission in Fig. 4 (broken line) with the pulse profile (full line) of (a) a 6-ns pulse laser and (b) a 100-ns pulse laser, and (c) the enlarged view of the tailing part of the 100-ns pulse laser. The signals were rescaled so that the peak height of the plasma emission and the laser pulse to be the same.

peak area was evaluated as a line intensity. The Zr emission lines used in this work are listed in Table 1 [71]. Here, the Zr II line at 355.2 nm and the Zr I line at 813.3 nm were selected because these lines were independently observed over a wide time range for both the pulses. As for the molecular emission, the R-head of a (0,0) sub-band of the ZrO gamma-system (d3Φ4 → a3Δ3) at 623.0 nm [50,72] was considered, because it was independently observed with a reasonable intensity. To avoid the background disturbance, the height of the band head from the rising was evaluated as a band head intensity.

41

As shown in Fig. 6, the atomic and molecular emission is much stronger when the 100-ns pulse laser is used. Regarding the ionic line selected, the peak area was larger in the case of the long-pulse laser. The decay rate of the atomic and molecular emission is lower than that of the ionic emission for both the pulses. Interestingly, the decay rate is almost the same regardless of the pulse duration in the case of the ionic emission, but is different for the atomic and molecular emission. The slower decay of the atomic emission in the case of the long-pulse laser may be due to the longer duration of the surface evaporation. Ghalamdaran et al. have simulated that the material removal of a gold surface continues about 1200 ns for a 235-ns pulse laser [73]. The continuous evaporation provides atoms to the plasma that enhances the number of emitters. When the pulse duration is shorter, a part of the target surface can be explosively removed [74–76] before the heat is sufficiently transferred to the target. When the pulse duration is longer, the target surface can be thermally evaporated with the heat transfer to the target. According to Fig. 5, amount of energy given to the target surface is higher in the case of the 100-ns pulse laser, because the plasma generation is delayed more from the laser irradiation and the efficiency of the plasma shielding may be lower. In this context, depth profiles of the craters produced by 100 shots were measured using a confocal laser scanning microscope (Olympus Corporation, OLS3000) with a magnification rate of 20, a pitch of 0.3, a tilt correction, and an isolated point removal (noise level: 5) (see Fig. 7). The crater produced by the 100-ns pulse laser was deeper and rougher. This confirms that a larger amount of the laser energy is consumed to ablate the target and that the surface is thermally evaporated. In terms of the molecular emission, the ZrO bands were observed from the shortest delays for both the pulses. Recently, the AlO emission has been studied [77–81] and the molecular bands show a maximum intensity at a delay time of ~10 μs or later [79,80]. Fig. 8 shows equilibrium constants of possible reactions for the molecular formation, which were calculated based on the NIST-JANAF thermochemical tables [82]. The equilibrium constants of the ZrO formation are much larger than that for the AlO formation at a wide temperature range. This suggests an early observation of the ZrO emission, which can be regarded as a spectral feature in the analysis of zirconium. In the case of the normal-pulse laser, the ZrO emission decreased monotonically unlike the AlO emission at longer delay. Considering the related chemical reactions including the sequential clustering, the decrease of the intensity factor may be more effective than the increase of the ZrO molecules in the present conditions. In the case of the long-pulse laser, the ZrO emission was quite intense and showed the maximum at a delay time of 2.0 μs. Such an observation may be due to a higher number of reactants and hydrodynamic flow of oxygen, as discussed in the section 3.7. 3.4. Plasma parameters 3.4.1. Temperature The excitation temperature of the Zr atoms in the plasma was estimated using the Boltzmann plot method [83]. Here, the Zr I lines at 527.7 nm, 528.0 nm, 529.7 nm, and 530.2 nm were used. In the case of the normal-pulse laser, a lot of atomic lines disappeared in a short time. In the case of the long-pulse laser, many lines were overlapped at the NUV region due to the line broadening and the molecular bands disturbed the atomic lines. To use the same lines for the analysis regardless of the pulse duration and the delay time, four lines with different energy of the upper state were chosen (see Table 1). In addition, these lines were located at the center of the same wavelength part of echelle spectra. This reduces the error of the intensity calibration. Fig. 9 shows examples of the Boltzmann plot obtained at a delay time of 1.0 μs. As shown in the graph, the plot changes approximately linearly. Fig. 10 shows the temperature as a function of the delay time. The temperature at the time of the spectral measurement is higher in the case of the 100ns pulse laser. This indicates that the intensities of the emission lines are enhanced. Also, the line intensity is proportional to exp.(−Eu/kT) [83],

42

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

(a)

6 ns, Zr II 355.2 nm

1.0

4.0

0.5

2.0

0

0 (b)

intensity (a. u.)

100 ns, Zr II 355.2 nm

6 ns, Zr I 813.3 nm

100 ns, Zr I 813.3 nm

1.0

24

0.5

12

0

0 (c)

6 ns, ZrO_623.0 nm

100 ns, ZrO 623.0 nm

1.0

32

0.5

16

0

0 0

1

2

3

4

5

0

1

2

3

4

5

Fig. 6. Time-resolved spectral intensities of (a) the Zr II line at 355.2 nm, (b) the Zr I line at 813.3 nm, and (c) the band head of the ZrO gamma-system at 623.0 nm as a function of the delay time. The spectra of zirconium metal in air were measured at an exposure time of 1.0 μs with a 6-ns pulse laser and a 100-ns pulse laser. For each emission, the intensities obtained for both the pulses were normalized to the intensity at td = 0.5 μs for the 6-ns pulse laser. The error bars correspond to the standard deviations based on five measurements.

where Eu is the energy of the upper level, k is the Boltzmann constant, and T is the temperature. Thus, the intensity change with respect to the temperature change is smaller when the temperature is high. Therefore, the slower decay of the atomic emission in the case of the longpulse laser can also be explained by a higher temperature of the plasma. As expected from Figs. 4 and 5, the temperature of the plasma immediately after the laser irradiation is higher in the case of the 6-ns pulse laser, while the temperature at the time of the spectral measurement is lower. The cooling rate is then faster. When the pulse duration is shorter, the primary plasma can expand faster. Therefore, the plasma

loses larger energy in an adiabatic process, resulting in a rapid decrease of the temperature. The temporal laser pulse profile is asymmetric and the tailing part lasts for a long time in the case of the 100 ns-pulse laser (see Fig. 5). This is a feature of the Q-switched long-pulse laser (the falling edge of

300

6 ns 100 ns

Table 1 Zr emission lines used in this work [71]. The symbols A, E, and g indicate the Einstein coefficient, the energy, and the statistical weight, and the subscripts l and u mean the lower level and the upper level, respectively. Wavelength (nm)

Element

A (s−1)

El (eV)

gl

Eu (eV)

gu

355.2 416.1 416.6 527.7 528.0 529.7 530.2 813.3

Zr II Zr II Zr I Zr I Zr I Zr I Zr I Zr I

3.24 × 107 1.83 × 107 0.84 × 107 0.11 × 107 0.39 × 107 0.17 × 107 1.32 × 107 0.08 × 107

0.09 0.71 0.69 0.54 1.58 0.10 2.20 0.69

8 6 9 3 9 9 9 9

3.58 3.69 3.66 2.89 3.93 3.34 4.54 2.21

8 4 11 5 9 9 9 9

depth

200 100 0 0

200 400 distance

600

Fig. 7. Depth profiles of the craters produced on zirconium metal in air by 100 shots of a 6ns pulse laser and a 100-ns pulse laser. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

(a)

6

10

4.0

Zr(g) + O(g) ZrO(g) Zr(g) + 1/2 O2(g) ZrO(g) Zr(g) + O2(g) ZrO(g) + O(g)

A·gu )

8

10

43

4

10

6 ns 100 ns

2

R = 0.987

0

ln (

-4.0

2

10

2

-8.0

0

R = 0.999

10

8

10

(b)

6

Keq

10

3.0 Al(g) + O(g) AlO(g) Al(g) + 1/2 O2(g) AlO(g) Al(g) + O2(g) AlO(g) + O(g)

3.5 4.0 Eu (eV)

4.5

Fig. 9. Boltzmann plots obtained for a 6-ns pulse laser and a 100-ns pulse laser at a delay time of 1.0 μs.

4

10

where I is the line intensity, A is the Einstein coefficient, gu is the statistical weight of the upper level, Z(T) is the partition function at the temperature T, Eu is the energy of the upper level, k is the Boltzmann constant, and the subscripts I and II are respectively related to the atom and ion form. Here, the Zr II line at 416.1 nm and the Zr I line at 416.6 nm were used. Fig. 11 shows the ratio NZr II/NZr I as a function of the delay time. The ratio is higher in the case of the normal-pulse laser even if the temperature is lower (see Fig. 10). Alternatively, the plasma produced by the long-pulse laser shows a lower ionization degree in spite of a higher temperature. Such an observation may be explained by the magnitude relation of the electron number density in the plasma as discussed below.

2

10

0

10

8

10

(c)

1/2 O2(g)

O(g)

6

10

4

10

2

10

0

10 3000

4000 5000 temperature (K)

6000

3.4.3. Electron number density In typical LIBS conditions, line broadening is considered to be mainly due to the Stark effect, and the electron number density can be estimated from the full width at half maximum (FWHM) of emission lines [83]. In the present case, however, it was difficult to obtain the width of the Stark broadening accurately, because the width of Zr lines was not sufficiently larger than the instrumental width. Instead, the Balmer H-alpha line at 656 nm was observed, which is strongly broadened by the Stark effect [87]. Here, the electron number density

the long-pulse fiber laser is steep regardless of the pulse duration [41]), and the slow decrease of the pulse may be a factor to prevent the temperature to drop: a few energy of the laser pulse is given at td N 0. Wang et al. and Cui et al. have performed collinear long (60 μs, operated at free-running mode) and short dual-pulse LIBS and successfully enhanced the signal intensity [84,85]. It has been considered that the preheating of the target surface by the early part of the long pulse and the temperature maintained high by the later part of the long pulse enhance the signal intensity. In the present case, the plasma can be also continuously heated by the slow decreasing part of the long pulse. 3.4.2. Ionization degree The number density ratio NZr II/NZr I in the plasma was estimated using the following eq. [86]:
 NII III AI g u;I Z II ðT Þ Eu;I −Eu;II exp − ¼ NI II AII g u;II Z I ðT Þ kT

ð1Þ

temperature (K)

Fig. 8. Equilibrium constants of possible reactions for the formation of (a) ZrO and (b) AlO molecules, and the (c) dissociation of O2 molecules as a function of temperature [82].

6 ns 100 ns

7000 6000 5000 0

1

2

3

4

5

Fig. 10. Excitation temperature of Zr atoms, estimated from the Boltzmann plots, obtained for a 6-ns pulse laser and a 100-ns pulse laser as a function of the delay time. The error bars correspond to the standard deviations based on five measurements.

44

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

4.0

(a)

6 ns

-3

2.0

17

3.0

cm )

6 ns 100 ns

Ne (10

2.0 1.0

1.0

0 0

1

2

3

4

0

5

Fig. 11. Number density ratio NZr II/NZr I obtained for a 6-ns pulse laser and a 100-ns pulse laser as a function of the delay time. The error bars correspond to the standard deviations based on five measurements.

was estimated using the H-alpha line based on the relation derived by Pardini et al. [87]:

N e ¼ 8:02  1012

10  Δλ3=2 α

1 pffiffiffiffiffiffi aþ b= Ne ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi q 2 a ¼ 4033:8  24:45  ln T b ¼ 1:028  109 þ 174576:3  T α¼

FWHM (nm)

(b)

0.10

0.05

0 ð2Þ

where Ne is the electron number density [cm−3], Δλ is the FWHM [nm], and T is the temperature [K]. Eq. 2 reproduces the tabulated data simulated by Gigosos et al. [88]. Unfortunately, the H-alpha line was not observed in the spectra obtained for the 100-ns pulse laser. The electron density was obtained only for the case of the 6-ns pulse laser at shorter delays. The instrumental width was neglected because the FWHM was much larger, i.e., 1.55 ± 0.08 nm, 0.94 ± 0.02 nm, and 0.51 ± 0.01 nm at td = 0.5, 1.0, and 2.0 μs, respectively. The self-absorption was also considered as negligible. Since the contents of water vapor and H2 in air are relatively small, the amount of the H atoms transferred from the ambient air is expected to be insignificant. Fig. 12 shows the electron number density and the FWHM of the Zr II line at 355.2 nm as a function of the delay time. When the pulse duration is shorter, the FWHM at td N 1.0 μs is almost the same as the instrumental width. However, it can be confirmed that the line width in the case of the 100-ns pulse laser is much larger than that of the 6-ns pulse laser. Assuming that the line broadening is mainly due to the Stark effect, a larger width means that the electron number density is higher [83]. Note that the difference in the temperature does not significantly change the broadening parameter [89,90]. According to the Saha eq. [83], the number density ratio of ions to atoms decreases with increasing the electron number density at a constant temperature, i.e. the probability of the re-combination increases. A higher electron number density in the case of the long-pulse laser can explain the result that the ionization degree is lower even though the temperature is higher (see Figs. 10 and 11). However, in this consideration, electroneutrality of the plasma is not satisfied. It may be considered that the ions stay at the center of the plasma while the electrons spread outside due to the difference of the weight, and the emission from the center of the plasma is mainly detected. In addition, the re-combination of the Zr ions is considered to occur significantly when the temperature is still high in the case of the long-pulse laser. This can also be a factor to

6 ns, Zr II 355.2 nm 100 ns, Zr II 355.2 nm

0

1

2 3 time

4

5

Fig. 12. (a) Electron number density estimated from the Balmer H-alpha line at 656 nm obtained for a 6-ns pulse laser and (b) FWHM of the Zr II line at 355.2 nm obtained for the 6-ns pulse laser and a 100-ns pulse laser as a function of the delay time. The error bars correspond to the standard deviations based on five measurements.

increase the lifetime of the atomic emission, because the Zr atoms produced by the re-combination are also excited and emit the light. The estimation of the temperature is based on the assumption of the local thermodynamic equilibrium (LTE). Generally, to verify such assumption, use is made of the McWhirter criterion, which however is a necessary, but not sufficient, condition to assess LTE in laser-induced plasma [91]. Using the McWhirter expression for a temperature of 8000 K, which is higher than the temperature estimated for our experiment, the first resonance transition of Zr I (ΔE=1.83 eV [71]), and the data reported in Fig. 12, LTE conditions should be warranted. We stress, however, that this conclusion may not be valid and that further timeand space-resolved measurements would be necessary [91]. 3.5. Emission from the ambient air Fig. 13 shows the time-resolved spectra around 868 nm obtained for the 6-ns pulse laser at td = 2.0 μs and for the 100-ns pulse laser at td = 0.5 μs. In the present conditions, the N I lines at 868.0 nm, 868.3 nm, and 868.6 nm were hardly observed for the long-pulse laser even at shorter delay. In contrast, the N I lines were clearly observed for the 6-ns pulse laser even at longer delay. This suggests that, in the case of the normalpulse laser, a larger number of gas atoms are included in the plasma. A similar behavior was observed for the O I lines at 777.2 nm, 777.4 nm, and 777.5 nm. Bai et al. have investigated the spectrally-resolved emission images of the plasma produced on an aluminum target in air with various pulse durations (4 ns, 25 ns, 45 ns) and demonstrated the role of the

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

ionized gas layer [92,93]. When the ambient gas is compressed by the ejected species at the initial stage, the gas molecules are dissociated and ionized. Then, the incoming laser light is absorbed by the ionized gas layer. In the present case, the ionization of the gas molecules is expected to be significant when the pulse duration is shorter. Therefore, a larger amount of the laser energy is absorbed by the ionized gas layer. In the case of the long-pulse laser, the ionization of the gas layer may be insufficient. Thus, the laser light passes through the compressed gas layer and the laser energy is given to the plasma composed of the ejected species. Note that a higher energy is needed to form the ionized gas layer, i.e. the bond energy of N2 (respectively O2) is 9.80 eV (respectively 5.16 eV) [82] and the ionization energy of N I (respectively O I) is 14.5 eV (respectively 13.6 eV), whereas the ionization energy of Zr I is 6.63 eV [94]. Alternatively, the re-combination of ions can preferentially occur for the gas species. Thus, in the case of the normal-pulse laser, the electron number density of the initial plasma may be rapidly reduced, which increases the lifetime of the ionization degree of the Zr species.

3.6. Emission regions Fig. 14 shows the spatial evolution of the emission regions taken from the direction horizontal to the target surface. Note that the intensity scale is not corrected taking into account the attenuation by the ND filters. The shape of the plasma was considerably different depending on the pulse duration. In the case of the 6-ns pulse laser, the plasma expanded to the horizontal direction. The flat shape is unique to fiberoptic LIBS in which the beam profile is homogeneous and the spot size

1.5

(a)

6 ns, td = 2.0

45

is relatively large [95]. Concerning the bright spot of the emission region, a separation into two locations occurs at td = 0.5 μs and 1.0 μs. When the initial plasma compresses the ambient air, the downward pressure is applied as a reaction force. The ablated species can escape to lower pressure regions (i.e. the horizontal direction), although they originally have the kinetic energy to the upward direction. In the case of the 100-ns pulse laser, the plasma expanded to the vertical direction, and the bright spot was separated from the target surface. With a decrease of the ambient pressure, the vertical expansion of the emission region and the rising of the bright spot become significant [96,97]. The compression is expected to be weaker when the pulse duration is longer, allowing the ablated species to propagate upwards. Also, the laser beam is gradually absorbed by the propagation front of the expanding plasma, which could be a reason of the rising of the bright spot. In addition, the continuous evaporation of the target surface may generate a movement of the previously evaporated species. In this paragraph, the vertical expansion of the emission region in the case of the 100-ns pulse laser is discussed. When the delay time was 10 μs, the bright spot was far away from the target surface, and a relatively high intensity region was observed in the gap between the bright spot and the target surface. Zhou et al. have observed the backward growth of the plasma produced by a 100-ns pulse laser (spot size: ~30 μm) on aluminum and titanium targets in air shortly after the end of the laser pulse [98]. The authors also mentioned that the backward growth is due to the flow of the high-temperature region but not to emission species. Lazic et al. have found a secondary plasma generation in a bubble produced by the laser ablation of an aluminum target in water, which is considered to be due to the reheating of the target surface caused by the backward growth of the plasma [99,100]. In the present case, the bottom of the bright spot might be expanded to the target surface due to the heat flowing in the same manner as the previous cases. This could be an indication of the secondary plasma generation.

NI

1.0

3.7. Molecular formation

intensity (a. u.)

NI

0.5 NI

0 1.5

(b)

100 ns, td = 0.5

1.0 0.5 0 867.50 868.25 869.00 wavelength (nm) Fig. 13. Time-resolved spectra of nitrogen atoms around 868 nm obtained by the irradiation of zirconium metal in air with (a) a 6-ns pulse laser at td = 2.0 μs and (b) a 100-ns pulse laser at td = 0.5 μs. The intensities of both the spectra were normalized to the maximum value of the displayed spectra.

As shown in Fig. 6, in the case of the long-pulse laser, the intensity of the molecular emission is very high. The higher temperature of the plasma produced by the long-pulse laser increases the intensity factor, but is not favorable for the ZrO formation (see Fig. 8). However, the emission intensity can increase when the absolute density of the reactants is higher, even if the equilibrium constant (concentration ratio of products to reactants) is smaller. As a reference, relative density of Zr atoms was investigated from the line intensity and the temperature (see. Fig. 15). In the case of the long-pulse laser, there seems to be a larger number of Zr atoms (one of the reactants) in the plasma. On the other hand, oxygen is needed to form the ZrO molecules. Saad et al. have reported that the incorporation of nitrogen into aluminum plasma causes the formation of AlN molecules and the collisional energy transfer to a specific transition [80]. Harilal et al. have measured the spectrally-resolved emission images of the AlO molecules from an aluminum target in air [81]. They found that the shock wave can be a barrier to impede the formation of the AlO molecules and that the air flow from the side of the plasma leads to the oxidation of the Al atoms. In this context, shadowgraph images were observed at a delay time of 1.0 μs (Fig. 16). When the pulse duration is shorter, the radius of the shock front is larger. Assuming that the time of the shock wave generation is equal to that of the plasma ignition (see Fig. 5), the actual delay in the case of the 6-ns pulse laser is about 160 ns shorter. Thus, the velocity of the shock wave is obviously faster, i.e. the initial shock is stronger, which makes the formation of the ZrO molecules difficult. In the case of the 100-ns pulse laser, the initial shock pressure is expected to be weaker, which makes the oxygen easy to transfer into the plasma. When the 100-ns pulse laser was used, the emission lines of the gas atoms were hardly observed (see Fig. 13). This indicates that the oxygen is mainly transferred by the hydrodynamic flow, and not by the initial

46

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

Fig. 14. Spatial evolution of the plasma produced on zirconium metal in air by (a) a 6-ns pulse laser and (b) a 100-ns pulse laser. The gate width was set to 10 ns. The delay time for the measurement is shown on the top of the images. The intensity scale in units of counts and ND filters used for each delay are shown under the images.

shock process in which the gas species are mixed into the high-temperature plasma. As shown in Fig. 14, at td = 0.1 μs for the 6-ns pulse laser, the bright spot was very close to the target surface, and the plasma rapidly expanded to the horizontal direction. This may disturb the air flow. On the other hand, at td = 0.1 μs for the 100-ns pulse laser, the bright spot was apart from the target surface, and the expansion to the horizontal direction seems to be smaller. According to Bai et al. studies [92,93], the ionized gas layer has been observed mainly at the top of the plasma when the pulse duration is longer, suggesting that the shock pressure is weaker at the side of the plasma. Such conditions can allow the ambient air to flow into the plasma bottom and facilitate the molecular formation. The slower decay of the molecular emission may be explained by the active flow of the oxygen from the ambient air. In the present case, due to the large number of emission lines, it

NZr (a. u.)

15.0

was difficult to separate the wavelength region of each emission species using band-pass filters. To clarify the mechanism of the molecular formation in the long-pulse-produced plasma, spectrally-resolved imaging of the atomic and molecular emission will be needed. 4. Conclusion In this work, the emission spectra of zirconium metal in air obtained for a 6-ns pulse laser and a 100-ns pulse laser (wavelength: 1064 nm, pulse energy: 12.5 mJ, spot diameter: 0.35 mm) were compared as a preliminary study to employ the long-pulse laser to a fiber-optic LIBS instrument designed for the internal survey of the damaged reactors. The spectral features were considerably different depending on the pulse duration. When the long-pulse laser was used, the atomic and

6 ns 100 ns

10.0 5.0 0 0

1

2

3

4

5

Fig. 15. Relative density of Zr atoms obtained for a 6-ns pulse laser and a 100-ns pulse laser as a function of the delay time. The densities obtained for both the pulses were normalized to the density at td = 0.5 μs for the 6-ns pulse laser. The error bars correspond to the standard deviations based on five measurements.

Fig. 16. Shadowgraph images obtained for (a) a 6-ns pulse laser and (b) a 100-ns pulse laser at a delay time of 1.0 μs. The shadowgraph images were taken using a xenon flash lamp as a back illumination during the measurement of the emission images in Fig. 14.

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

molecular emission was remarkably enhanced. The intense emission of the atoms at the NIR region would enable us to identify in the high-radiation fields, nuclear materials with minimum line overlapping. To discuss the spectral differences, photodiode signals, time-resolved spectra, plasma parameters, emission from the ambient air, and emission regions were investigated. The initial state of the plasma deduced from the photodiode signals was significantly different depending on the pulse duration, which can affect the later plasma characteristics. When the pulse duration is longer, kinetic energy of the ejected species, efficiency of the inverse bremsstrahlung, and the compression of the ablated species are expected to be lower, resulting in the initial plasma's lower degrees of ionization, excitation, and dissociation. The time-resolved spectra showed lower decay rate of the atomic and molecular emission in the case of the 100-ns pulse laser. The slower decay of the atomic emission may be explained by continuous evaporation of the target surface, and higher temperature and electron number density at the time of the spectral measurement. The emission from the ambient air was hardly observed when the longpulse laser was used. This indicates that the ionized gas layer is not sufficiently formed at the initial stage and the laser energy is given to the plasma composed of ejected species. Interestingly, the expansion direction of the emission region was completely different depending on the pulse duration. In the case of the 6-ns pulse laser, the plasma expanded to the horizontal direction and the bright spot was close to the target surface. When the pulse duration was extended, the plasma expanded to the vertical direction and the bright spot moved from the target surface. Such a unique geometry can make the oxygen easy to flow into the plasma, facilitating the formation of the molecules at early time. Fiber-optic LIBS could be used in the decommissioning process of the damaged reactors, e.g. assessment of the current situation, real-time monitoring during the removal process of the debris, safe storage of the removed materials, and also the analysis of unknown objects scattered around the reactor buildings. The use of the long-pulse laser could be the simplest way to improve the analytical capabilities under difficult conditions, and the investigation of the plasma characteristics gives interesting insights into the laser ablation process.

Acknowledgements This work includes a part of the results of the project “Advanced study on remote and in-situ elemental analysis of molten fuel debris in damaged core by innovative optical spectroscopy”, the Center of World Intelligence Project for Nuclear S&T and Human Resource Development, the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. This work was also supported by JSPS KAKENHI Grant Number JP17K14506. The authors are grateful to Dr. Violeta Lazic, ENEA, Italy, for her fruitful discussion about secondary plasma formation. References [1] Y. Sibamoto, K. Morimaya, Y. Maruyama, T. Yonomoto, A simple mass and heat balance model for estimating plant conditions during the Fukushima Dai-ichi NPP accident, J Nucl Sci Technol 49 (2012) 768–781. [2] Y. Yamanaka, S. Mizokami, M. Watanabe, T. Honda, Update of the first TEPCO MAAP accident analysis of units 1, 2, and 3 at Fukushima Daiichi Nuclear Power Station, Nucl Technol 186 (2014) 263–279. [3] T. Sevón, A MELCOR model of Fukushima Daiichi Unit 1 accident, Ann Nucl Energy 85 (2015) 1–11. [4] M. Hidaka, T. Fujii, T. Sakai, Improvement of molten core–concrete interaction model in debris spreading analysis module with consideration of concrete degradation by heat, J Nucl Sci Technol 53 (2016) 1260–1275. [5] S. Suehiro, J. Sugimoto, A. Hidaka, H. Okada, S. Mizokami, K. Okamoto, Development of the source term PIRT based on findings during Fukushima Daiichi NPPs accident, Nucl Eng Design 286 (2015) 163–174. [6] X. Zheng, H. Itoh, H. Tamaki, Y. Maruyama, An integrated approach to source term uncertainty and sensitivity analyses for nuclear reactor severe accidents, J Nucl Sci Technol 53 (2016) 333–344.

47

[7] H. Lopez, N. Erkan, K. Okamoto, Reactor core isolation cooling system analysis of the Fukushima Daiichi unit 2 accident with RELAP/ScdapSIM, J Nucl Sci Technol 53 (2016) 1899–1905. [8] F. Nagase, R.O. Gauntt, M. Naito, Overview and outcomes of the OECD/NEA benchmark study of the accident at the Fukushima Daiichi nuclear Power Station, Nucl Technol 196 (2016) 499–510. [9] K. Izawa, Y. Uchida, K. Ohkubo, M. Totsuka, H. Sono, K. Tonoike, Infinite multiplication factor of low-enriched UO2–concrete system, J Nucl Sci Technol 49 (2012) 1043–1047. [10] M. Takano, T. Nishi, High temperature reaction between sea salt deposit and (U,Zr) O2 simulated corium debris, J Nucl Mater 443 (2013) 32–39. [11] S. Hirooka, M. Kato, K. Morimoto, A. Komeno, T. Uchida, M. Akashi, Effect of oxygen-to-metal ratio on properties of corium prepared from UO2 and zircaloy-2, J Nucl Mater 437 (2013) 130–134. [12] M. Takano, T. Nishi, N. Shirasu, Characterization of solidified melt among materials of UO2 fuel and B4C control blade, J Nucl Sci Technol 51 (2014) 859–875. [13] K. Morimoto, S. Hirooka, M. Akashi, M. Watanabe, H. Sugata, The influences of Pu and Zr on the melting temperatures of the UO2–PuO2–ZrO2 pseudo-ternary system, J Nucl Sci Technol 52 (2015) 1247–1252. [14] T. Kitagaki, K. Yano, H. Ogino, T. Washiya, Thermodynamic evaluation of the solidification phase of molten core–concrete under estimated Fukushima Daiichi nuclear power plant accident conditions, J Nucl Mater 486 (2017) 206–215. [15] H. Miyadera, K.N. Borozdin, S.J. Greene, Z. Lukić, K. Masuda, E.C. Milner, C.L. Morris, J.O. Perry, Imaging Fukushima Daiichi reactors with muons, AIP Adv 3 (2013), 052133. [16] K. Takamatsu, H. Takegami, C. Ito, K. Suzuki, H. Ohnuma, R. Hino, T. Okumura, Cosmic-ray muon radiography for reactor core observation, Ann Nucl Energy 78 (2015) 166–175. [17] Decommissioning Plan of Fukushima Daiichi Nuclear Power, Tokyo Electric Power Company Holdings, Inc. (TEPCO). http://www.tepco.co.jp/en/decommision/indexe.html (accessed 17.05.28). [18] B.J. Marquardt, S.R. Goode, S. Michael Angel, In situ determination of lead in paint by laser-induced breakdown spectroscopy using a fiber-optic probe, Anal Chem 68 (1996) 977–981. [19] A.I. Whitehouse, J. Young, I.M. Botheroyd, S. Lawson, C.P. Evans, J. Wright, Remote material analysis of nuclear power station steam generator tubes by laser-induced breakdown spectroscopy, Spectrochim Acta Part B 56 (2001) 821–830. [20] A.K. Rai, F.Y. Yueh, J.P. Singh, H. Zhang, High temperature fiber optic laser-induced breakdown spectroscopy sensor for analysis of molten alloy constituents, Rev Sci Instrum 73 (2002) 3589. [21] C. Bohling, K. Hohmann, D. Scheel, C. Bauer, W. Schippers, J. Burgmeier, U. Willer, G. Holl, W. Schade, All-fiber-coupled laser-induced breakdown spectroscopy sensor for hazardous materials analysis, Spectrochim Acta Part B 62 (2007) 1519–1527. [22] B. Bousquet, G. Travaillé, A. Ismaël, L. Canioni, K. Michel-Le Pierrès, E. Brasseur, S. Roy, I. le Hecho, M. Larregieu, S. Tellier, M. Potin-Gautier, T. Boriachon, P. Wazen, A. Diard, S. Belbèze, Development of a mobile system based on laser-induced breakdown spectroscopy and dedicated to in situ analysis of polluted soils, Spectrochim Acta Part B 63 (2008) 1085–1090. [23] F. Fortes, J.J. Laserna, The development of fieldable laser-induced breakdown spectrometer: no limits on the horizon, Spectrochim Acta Part B 65 (12) (2010) 975–990. [24] S. Guirado, F.J. Fortes, V. Lazic, J.J. Laserna, Chemical analysis of archeological materials in submarine environments using laser-induced breakdown spectroscopy. On-site trials in the Mediterranean Sea, Spectrochim Acta Part B 74 (2012) 137–143. [25] B. Thornton, T. Takahashi, T. Sato, T. Sakka, A. Tamura, A. Matsumoto, T. Nozaki, T. Ohki, K. Ohki, Development of a deep-sea laser-induced breakdown spectrometer for in situ multi-element chemical analysis, Deep-Sea Res I Oceanogr Res Pap 95 (2015) 20–36. [26] M. Saeki, A. Iwanade, C. Ito, I. Wakaida, B. Thornton, T. Sakka, H. Ohba, Development of a fiber-coupled laser-induced breakdown spectroscopy instrument for analysis of underwater debris in a nuclear reactor core, J Nucl Sci Technol 51 (2014) 930–938. [27] A. Ruas, A. Matsumoto, H. Ohba, K. Akaoka, I. Wakaida, Application of laser-induced breakdown spectroscopy to zirconium in aqueous solution, Spectrochim Acta Part B 131 (2017) 99–106. [28] D.C.S. Beddows, O. Samek, M. Liška, H.H. Telle, Single-pulse laser-induced breakdown spectroscopy of samples submerged in water using a single-fibre light delivery system, Spectrochim Acta Part B 57 (2002) 1461–1471. [29] T. Sakka, H. Oguchi, S. Masai, K. Hirata, Y.H. Ogata, M. Saeki, H. Ohba, Use of a longduration ns pulse for efficient emission of spectral lines from the laser ablation plume in water, Appl Phys Lett 88 (2006), 061120. [30] B. Thornton, T. Sakka, T. Takahashi, A. Tamura, T. Masamura, A. Matsumoto, Spectroscopic measurements of solids immersed in water at high pressure using a longduration nanosecond laser pulse, Appl Phys Express 6 (2013), 082401 [31] B. Thornton, T. Sakka, T. Masamura, A. Tamura, T. Takahashi, A. Matsumoto, Long duration nano-second single pulse lasers for observation of spectra from bulk liquids at high hydrostatic pressures, Spectrochim Acta Part B 97 (2014) 7–12. [32] T. Sakka, S. Masai, K. Fukami, Y.H. Ogata, Spectral profile of atomic emission lines and effects of pulse duration on laser ablation in liquid, Spectrochim Acta Part B 64 (2009) 981–985. [33] T. Sakka, A. Tamura, A. Matsumoto, K. Fukami, N. Nishi, B. Thornton, Effects of pulse width on nascent laser-induced bubbles for underwater laser-induced breakdown spectroscopy, Spectrochim Acta Part B 97 (2014) 94–98.

48

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49

[34] T. Takahashi, B. Thornton, K. Ohki, T. Sakka, Calibration-free analysis of immersed brass alloys using long-ns-duration pulse laser-induced breakdown spectroscopy with and without correction for nonstoichiometric ablation, Spectrochim Acta Part B 111 (2015) 8–14. [35] A. Matsumoto, A. Tamura, R. Koda, K. Fukami, Y.H. Ogata, N. Nishi, B. Thornton, T. Sakka, On-site quantitative elemental analysis of metal ions in aqueous solutions by underwater laser-induced breakdown spectroscopy combined with electrodeposition under controlled potential, Anal Chem 87 (2015) 1655–1661. [36] A. Tamura, A. Matsumoto, T. Nakajima, K. Fukami, Y.H. Ogata, N. Nishi, T. Sakka, Effects of temporal laser profile on the emission spectra for underwater laser-induced breakdown spectroscopy: study by short-interval double pulses with different pulse durations, J Appl Phys 117 (2015), 023302 [37] A. Tamura, A. Matsumoto, K. Fukami, N. Nishi, T. Sakka, Simultaneous observation of nascent plasma and bubble induced by laser ablation in water with various pulse durations, J Appl Phys 117 (2015), 173304 [38] A. Matsumoto, A. Tamura, R. Koda, K. Fukami, Y.H. Ogata, N. Nishi, B. Thornton, T. Sakka, A calibration-free approach for on-site multi-element analysis of metal ions in aqueous solutions by electrodeposition-assisted underwater laser-induced breakdown spectroscopy, Spectrochim Acta Part B 118 (2016) 45–55. [39] T. Takahashi, B. Thornton, T. Sato, T. Ohki, K. Ohki, T. Sakka, Temperature based segmentation for spectral data of laser-induced plasmas for quantitative compositional analysis of brass alloys submerged in water, Spectrochim Acta Part B 124 (2016) 87–93. [40] K.Y. Yamamoto, D.A. Cremers, L.E. Foster, M.P. Davies, R.D. Harris, Laser-induced breakdown spectroscopy analysis of solids using a long-pulse (150 ns) Q-switched Nd:YAG laser, Appl Spectrosc 59 (2005) 1082–1097. [41] I.Y. Elnasharty, Study on the influence of laser pulse duration in the long nanosecond regime on the laser induced plasma spectroscopy, Spectrochim Acta Part B 124 (2016) 1–15. [42] H. Kurniawan, M.O. Tjia, M. Barmawi, S. Yokoi, Y. Kimura, K. Kagawa, A time-resolved spectroscopic study on the shock wave plasma induced by the bombardment of a TEA CO2 laser, J Phys D Appl Phys 28 (1995) 879–883. [43] A.M. Marpaung, H. Kurniawan, M.O. Tjia, K. Kagawa, Comprehensive study on the pressure dependence of shock wave plasma generation under TEA CO2 laser bombardment on metal sample, J Phys D Appl Phys 34 (2001) 758–771. [44] S.N. Madjid, I. Kitazima, T. Maruyama, T.J. Lie, M.O. Tjia, Koo H. Kurniawan, K. Kagawa, TEA-CO2 laser-induced shock wave plasma modulated by wires and needles placed in front of the target at low pressure, Appl Spectrosc 57 (2003) 874–877. [45] K. Kagawa, N. Idris, M. Wada, H. Kurniawan, K. Tsuyuki, S. Miura, Carbon analysis for inspecting carbonation of concrete using a TEA CO2 laser-induced plasma, Appl Spectrosc 58 (2004) 887–896. [46] N. Idris, S. Terai, T.J. Lie, H. Kurniawan, T. Kobayashi, T. Maruyama, K. Kagawa, Atomic hydrogen emission induced by TEA CO2 laser bombardment on solid samples at low pressure and its analytical application, Appl Spectrosc 59 (2005) 115–120. [47] A. Khumaeni, M. Ramli, Y. Deguchi, Y.I. Lee, N. Idris, K.H. Kurniawan, T.J. Lie, K. Kagawa, New technique for the direct analysis of food powders confined in a small hole using transversely excited atmospheric CO2 laser-induced gas plasma, Appl Spectrosc 62 (2008) 1344–1348. [48] C. Ito, H. Naito, A. Nishimura, H. Ohba, I. Wakaida, A. Sugiyama, K. Chatani, Development of radiation-resistant optical fiber for application to observation and laser spectroscopy under high radiation dose, J Nucl Sci Technol 51 (2014) 944–950. [49] U. Uhler, The rotational analysis of the β-system of zirconium oxide, Ark f Fysik 8 (1954) 295–304. [50] S.R. Langhoff, C.W. Bauschlicher Jr., Theoretical study of the spectroscopy of ZrO, Astrophys J 349 (1990) 369–375. [51] T.L. Barr, ESCA studies of naturally passivated metal foils, J Vac Sci Technol 14 (1977) 660–665. [52] C. Alvarez-Llamas, J. Pisonero, N. Bordel, A novel approach for quantitative LIBS fluorine analysis using CaF emission in calcium-free samples, J Anal At Spectrom 32 (2017) 162–166. [53] H. Niki, T. Yasuda, I. Kitazima, Measurement technique of boron isotopic ratio by laser-induced breakdown spectroscopy, J Nucl Sci Technol 35 (1998) 34–39. [54] R.E. Russo, A.A. Bol'shakov, X. Mao, C.P. McKay, D.L. Perry, O. Sorkhabi, Laser ablation molecular isotopic spectrometry, Spectrochim Acta Part B 66 (2011) 99–104. [55] A.A. Bol'shakov, X. Mao, J.J. Gonzalez, R.E. Russo, Laser ablation molecular isotopic spectrometry (LAMIS): current state of the art, J Anal At Spectrom 31 (2016) 119–134. [56] M. Baudelet, L. Guyon, J. Yu, J.P. Wolf, T. Amodeo, E. Fréjafon, P. Laloi, Femtosecond time-resolved laser-induced breakdown spectroscopy for detection and identification of bacteria: a comparison to the nanosecond regime, J Appl Phys 99 (2006), 084701 [57] J. Serrano, J. Moros, J.J. Laserna, Molecular signatures in femtosecond laser-induced organic plasmas: comparison with nanosecond laser ablation, Phys Chem Chem Phys 18 (2016) 2398–2408. [58] E. Nageswara Rao, P. Mathi, S. Abdul Kalam, S. Sreedhar, A.K. Singh, B.N. Jagatapc, S. Venugopal Rao, Femtosecond and nanosecond LIBS studies of nitroimidazoles: correlation between molecular structure and LIBS data, J Anal At Spectrom 31 (2016) 737–750. [59] A. De Giacomo, R. Gaudiuso, M. Dell'Aglio, A. Santagata, The role of continuum radiation in laser induced plasma spectroscopy, Spectrochim Acta Part B 65 (2010) 385–394. [60] M. Gaft, L. Nagli, I. Gornushkin, Laser-induced breakdown spectroscopy of Zr in short ultraviolet wavelength range, Spectrochim Acta Part B 85 (2013) 93–99.

[61] J.J. Chang, B.E. Warner, Laser–plasma interaction during visible-laser ablation of methods, Appl Phys Lett 69 (1996) 473–475. [62] A.E. Hussein, P.K. Diwakar, S.S. Harilal, A. Hassanein, The role of laser wavelength on plasma generation and expansion of ablation plumes in air, J Appl Phys 113 (2013), 143305 [63] A. De Giacomo, M. Dell'Aglio, A. Santagata, R. Teghil, Early stage emission spectroscopy study of metallic titanium plasma induced in air by femtosecond- and nanosecond-laser pulses, Spectrochim Acta Part B 60 (2005) 935–947. [64] L. Fornarini, R. Fantoni, F. Colao, A. Santagata, R. Teghil, A. Elhassan, M.A. Harith, Theoretical modeling of laser ablation of quaternary bronze alloys: case studies comparing femtosecond and nanosecond LIBS experimental data, J Phys Chem A 113 (2009) 14364–14374. [65] B. Verhoff, S.S. Harilal, J.R. Freeman, P.K. Diwakar, A. Hassanein, Dynamics of femtoand nanosecond laser ablation plumes investigated using optical emission spectroscopy, J Appl Phys 112 (2012), 093303 [66] R.E. Russo, X. Mao, J.J. Gonzalez, V. Zorba, J. Yoo, Laser ablation in analytical chemistry, Anal Chem 85 (2013) 6162–6177. [67] T.A. Labutin, V.N. Lednev, A.A. Ilyincd, A.M. Popova, Femtosecond laser-induced breakdown spectroscopy, J Anal At Spectrom 31 (2016) 90–188. [68] K. Kagawa, K. Kawai, M. Tani, T. Kobayashi, XeCl excimer laser-induced shock wave plasma and its application to emission spectrochemical analysis, Appl Spectrosc 48 (1994) 198–205. [69] H. Kurniawan, T.J. Lie, N. Idris, M.O. Tjia, M. Ueda, K. Kagawa, Detection of density jump in laser-induced shock wave plasma using low energy Nd: YAG laser at low pressures of air, J Spectrosc Soc Jpn 50 (2001) 13–18. [70] A. Bogaerts, Z. Chen, Effect of laser parameters on laser ablation and laser-induced plasma formation: a numerical modeling investigation, Spectrochim Acta Part B 60 (2005) 1280–1307. [71] P.L. Smith, C. Heise, J.R. Esmond, R.L. Kurucz, Atomic spectral line database, Atomic Line Data (R.L. Kurucz and B. Bell) Kurucz CD-ROM No. 231995. http://www.cfa. harvard.edu/amp/ampdata/kurucz23/sekur.html (accessed 17.05.28). [72] A. Langerqvist, U. Uhler, R.F. Barbow, The rotational analysis of the α and γ systems of zirconium oxide, Ark f Fysik 8 (1954) 281–293. [73] S. Ghalamdaran, P. Parvin, M.J. Torkamany, J.S. Zadeh, Two-dimensional simulation of laser ablation with 235 nanosecond pulses, J Laser Appl 26 (2014), 012009 [74] X.L. Mao, A.C. Ciocan, R.E. Russo, Preferential vaporization during laser ablation inductively coupled plasma atomic emission spectroscopy, Appl Spectrosc 52 (1998) 913–918. [75] J.H. Yoo, S.H. Jeong, R. Greif, R.E. Russo, Explosive change in crater properties during high power nanosecond laser ablation of silicon, J Appl Phys 88 (2000) 1638–1649. [76] C. Porneala, D.A. Willis, Observation of nanosecond laser-induced phase explosion in aluminum, Appl Phys Lett 89 (2006), 211121 [77] C.G. Parigger, J.O. Hornkohl, Computation of AlO B2Σ+ → X2Σ+ emission spectra, Spectrochim Acta Part B 81 (2011) 404–411. [78] J. Lam, V. Motto-Ros, D. Misiak, C. Dujardin, G. Ledoux, D. Amans, Investigation of local thermodynamic equilibrium in laser-induced plasmas: measurements of rotational and excitation temperatures at long time scales, Spectrochim Acta Part B 101 (2014) 86–92. [79] X. Bai, V. Motto-Ros, W. Lei, L. Zheng, J. Yu, Experimental determination of the temperature range of AlO molecular emission in laser-induced aluminum plasma in air, Spectrochim Acta Part B 99 (2014) 193–200. [80] R. Saad, D. L'Hermite, B. Bousquet, Unexpected temporal evolution of atomic spectral lines of aluminum in a laser induced breakdown spectroscopy experiment, Spectrochim Acta Part B 101 (2014) 330–334. [81] S.S. Harilal, B.E. Brumfield, B. Cannon, M.C. Phillips, Shock wave mediated plasma chemistry of AlO formation in laser ablation plumes, Anal Chem 88 (2016) 2296–2302. [82] NIST-JANAF Thermochemical Tables, National Institute of Standards and Technology (NIST), http://kinetics.nist.gov/janaf/ (accessed 17.05.28). [83] A.W. Miziolek, V. Palleschi, I. Schechter (Eds.), Laser Induced Breakdown Spectroscopy, Fundamentals and Applications, Cambridge University Press, Cambridge, 2006. [84] M. Cui, Y. Deguchi, Z. Wang, F. Yuki, R. Liu, F.J. Shiou, S. Zhao, Performance improvement of LIBS for steel sample measurement using a collinear long-short dual-pulse method, Spectrochim Acta Part B 107 (2015) 132–138. [85] Z. Wang, Y. Deguchi, R. Liu, A. Ikutomo, Z. Zhang, D. Chong, J. Yan, J. Liu, F.J. Shiou, Emission characteristics from laser-induced plasma using collinear long and short dual-pulse LIBS, Appl Spectrosc, in press. (DOI: https://doi.org/10.1177/ 0003702817693239). [86] D.W. Hahn, N. Omenetto, Laser-induced breakdown spectroscopy (LIBS), part I: review of basic diagnostics and plasma-particle interactions: still-challenging issues within the analytical plasma community, Appl Spectrosc 64 (2010) 335–366. [87] L. Pardini, S. Legnaioli, G. Lorenzetti, V. Palleschi, R. Gaudiuso, A. De Giacomo, D.M. Diaz Pace, F. Anabitarte Garcia, G. De Holanda Cavalcanti, C. Parigger, On the determination of plasma electron number density from Stark broadened hydrogen Balmer series lines in laser-induced breakdown spectroscopy experiments, Spectrochim Acta Part B 88 (2013) 98–103. [88] M.A. Gigosos, M.A. González, V. Cardeñoso, Computer simulated Balmer-alpha, -beta and -gamma Stark line profiles for non-equilibrium plasmas diagnostics, Spectrochim Acta Part B 58 (2003) 1489–1504. [89] L.Č. Popović, M.S. Dimitrijević, Stark broadening parameter tables for Xe II, Sc II, Y II and Zr II, Bull Astron Belgrade 155 (1997) 159–163. [90] L.Č. Popović, N. Milovanović, M.S. Dimitrijević, The electron-impact broadening effect in hot star atmospheres: the case of singly- and doubly-ionized zirconium, Astron Astrophys 365 (2001) 656–659.

A. Matsumoto et al. / Spectrochimica Acta Part B 142 (2018) 37–49 [91] G. Cristoforetti, A. De Giacomo, M. Dell'Aglio, S. Legnaioli, E. Tognoni, V. Palleschi, N. Omenetto, Local thermodynamic equilibrium in laser-induced breakdown spectroscopy: beyond the McWhirter criterion, Spectrochim Acta Part B 65 (2010) 86–95. [92] X. Bai, Q. Ma, V. Motto-Ros, J. Yu, D. Sabourdy, L. Nguyen, A. Jalocha, Convoluted effect of laser fluence and pulse duration on the property of a nanosecond laser-induced plasma into an argon ambient gas at the atmospheric pressure, J Appl Phys 113 (2013), 013304 [93] X. Bai, Q. Ma, V. Motto-Ros, J. Yu, D. Sabourdy, L. Nguyen, A. Jalocha, Experimental study of laser-induced plasma: influence of laser fluence and pulse duration, Spectrochim Acta Part B 87 (2013) 27–35. [94] Atomic Spectra Database Version 5, National Institute of Standards and Technology (NIST), https://www.nist.gov/pml/atomic-spectra-database, Accessed date: 17 May 2018. [95] Q. Zeng, L. Guo, X. Li, C. He, M. Shen, K. Li, J. Duan, X. Zeng, Y. Lu, Laser-induced breakdown spectroscopy using laser pulses delivered by optical fibers for analyzing Mn and Ti elements in pig iron, J Anal At Spectrom 30 (2015) 403–409.

49

[96] S. Lafane, T. Kerdja, S. Abdelli-Messaci, S. Malek, M. Maaza, Laser-induced plasma study by fast imaging for Sm1−xNdxNiO3 thin film deposition, Appl Phys A Mater Sci Process 98 (2010) 375–383. [97] N. Farid, S.S. Harilal, H. Ding, A. Hassanein, The emission features and expansion dynamics of nanosecond laser ablation plumes under varying ambient pressures, J Appl Phys 115 (2014), 033107 [98] Y. Zhou, S. Tao, B. Wu, Backward growth of plasma induced by long nanosecond laser pulse ablation, Appl Phys Lett 99 (2011), 051106 [99] M.R. Gavrilović, M. Cvejić, V. Lazic, S. Jovićević, Secondary plasma formation after single pulse laser ablation underwater and its advantages for laser induced breakdown spectroscopy (LIBS), Phys Chem Chem Phys 18 (2016) 14629–14637. [100] M.R. Gavrilović, V. Lazic, S. Jovićević, Influence of the target material on secondary plasma formation underwater and its laser induced breakdown spectroscopy (LIBS) signal, J Anal At Spectrom 32 (2017) 345–353.