- Email: [email protected]
Degradation of optical reflectivity of in-vessel mirror materials by helium bombardment
Journal of Nuclear Materials 417 (2011) 838–841
Contents lists available at ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Degradation of optical reflectivity of in-vessel mirror materials by helium bombardment Shin Kajita a,⇑, Tsubasa Saeki b, Noriyasu Ohno b, Masayuki Tokitani c, Takaki Hatae d, Wataru Sakaguchi b a
EcoTopia Science Institute, Nagoya Univ., Nagoya 464-8603, Japan Graduate School of Eng., Nagoya Univ., Nagoya 464-8603, Japan c National Institute for Fusion Science, Toki 509-5292, Japan d Japan Atomic Energy Agency, Mukoyama 801-1, Naka, Ibaraki 311-0193, Japan b
a r t i c l e
i n f o
Article history: Available online 25 December 2010
a b s t r a c t The effect of helium irradiation on in-vessel mirror materials, i.e. molybdenum and rhodium, are investigated experimentally. By the exposure to helium plasmas with low incident ion energy (50 eV) at different surface temperatures, the optical reflectivity of molybdenum and rhodium decreases significantly. From the surface analysis, it is shown that fiberlike nanostructure is formed on molybdenum surface when the surface temperature is high (at 1500 K), while rough surface is observed when the surface temperature is low (<1000 K). The decrease in the optical reflectivity is significant particularly for short wavelength ranges, typically, less than 300 nm. The results indicate that the helium irradiation should be taken into account for in-vessel mirror materials for the optical diagnostics in ITER. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction In future experimental fusion devices, many in-vessel mirrors will be used for optical diagnostics. It is generally recognized that metallic mirrors are used for them, and candidate materials for the first mirror are molybdenum, tungsten, copper, stainless steel, and rhodium [1,2]. Although it has been found that neutron irradiation does not have a significant influence on the optical reflectivity [3], there is concern that the reflectivity decreases by the exposure to charge-exchange neutrals, which cause erosion and deposition [4,5]. Until now, irradiation effects of hydrogen isotopes on the reflectivity change have been experimentally investigated [6,5]. However, it is pointed out that the exposure to helium may cause more serious problems by forming blisters [7], helium bubbles [8], and nanostructures [9,10]. Since these damages would result in a considerable decrease in optical reflectivity [11], it is important to investigate the helium-irradiation effects. Helium irradiation to tungsten has been extensively explored by using linear devices and ion beams because tungsten is also the candidate material for the divertor plate in ITER. From those experiments, it was revealed that the temperature and incident ion energy are key factors to cause damages. Micrometer-sized helium bubbles are easily formed when the surface temperature is higher than 1300–1500 K [12,13]. For high energy helium irradiation, typically >1 keV for displacement, blisters are formed [7]. When the incident ion energy is higher than 20 eV and the surface temperature ⇑ Corresponding author. Tel.: +81 52 789 3144; fax: +81 52 789 3944. E-mail address: [email protected] (S. Kajita). 0022-3115/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jnucmat.2010.12.205
is in the range from 1000 to 2000 K, fiberform structure, which leads to significant decrease in optical reflectivity, is easily formed on tungsten surface [14]. The detailed irradiation for tungsten increases an interest to explore the helium irradiation on other candidate mirror materials such as molybdenum and rhodium. There are several reports about the helium irradiation experiments in a high energy range (>1 keV) [7], and high temperature range [14]. However, it seems that there is no report regarding the effects of helium irradiation in low energy and low temperature ranges using the candidate materials. Note that the surface temperature of the first mirror will be 400–500 K if it is sufficiently cooled, and the energy and flux of the particles are different by the location of mirrors. Concerning the energy of helium, it will be mainly less than 100 eV, but higher charge exchange neutrals may hit the mirrors at some locations. In the present paper, we will show the effect of low-energy helium bombardment to molybdenum and rhodium from the experiments conducted in a linear plasma device. After a brief explanation of the experimental setup in Section 2, irradiation effects at high temperature conditions (higher than 1000 K) are shown in Section 3.1. Although the experiments at high temperature do not directly correspond to the investigation for the first mirror, it will be interesting to compare them with the irradiation at a lower temperature range shown in Section 3.2. Discussion and concluding remarks are given in Section 4. 2. Experimental setup The experiments were performed in the linear divertor simulator NAGDIS (NAGoya DIvertor Simulator)-I [15]. Helium plasma
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
839
was generated by DC arc discharge. We used powder metallurgy molybdenum (Nilaco Co.) and rhodium coated steel (Valtech Co.) for specimens. The molybdenum samples were mechanically polished by sandpapers and alumina suspension before the irradiation. For rhodium coating, a barrier layer of nickel is deposited on steel (NSS431DP2) first, and then, rhodium was deposited on the nickel layer. The layer thicknesses of the rhodium and nickel were approximately 50 nm and several lm, respectively. The sizes of the Mo and Rh samples are, respectively, 10 10 0.5 and 25 25 2 mm. The specimens were exposed to the helium plasma in the NAGDIS-I. The surface temperature and incident ion energy were controlled by changing the helium flux to the specimen. The incident ion energy was controlled by electrically biasing the specimen, and was 50 eV; the effect of the high energy particles is beyond the scope of this paper. After the helium plasma irradiation, the samples were analyzed by SEM (Scanning Electron Microscope) and AFM (Atomic Force Microscope). The wavelength dependence of optical reflectivity was measured with a spectrophotometer (Nihon Bunkosya: ARV-47S) for the wavelength from 200 to 900 nm; the angle of incident and reflection from normal direction on sample surface was in 7.5°. The measured reflectivity corresponds to the specular reflectivity. Since the first mirror will be used for transmitting the emission from the plasmas to optical fibers, which is located far-field area, the specular reflectivity rather than diffusive reflectivity may be important as a tool for plasma diagnostics.
3. Results 3.1. He irradiation at high temperature (>1000 K) (Mo) Fig. 1(a–c) shows the helium irradiated Mo surfaces at the surface temperature of (a and b) 1200 K (Mo-1) and (c) 1500 K (Mo-2). The incident helium ion energy were both 50 eV, and the helium ion fluences for Mo-1 and Mo-2 were 1.8 1026 m 2 and 1.1 1027 m 2, respectively. Interestingly, it is seen in Fig. 1a that the surface roughness was considerably different by crystal grains; the surface became significantly rough on some grains, while it was not so rough on some other grains. It is speculated that the difference in roughness may be related with the difference in the crystal orientation. However, the mechanism to arise the difference is yet to be understood. Indeed, it is interesting to note that such a difference by crystal grains has never been observed on a polycrystalline tungsten samples in the experiments by the author’s group. The original surface has scratches due to mechanical polishing as shown later in Fig. 4; however, the rough surface such as in Fig. 1 can not be seen. Fig. 1b shows an enlarged micrograph of Mo-1, representing that there are many pin-holes on the surface and the surface is significantly rough in the central area compared with the surrounding area. The rough surface may probably correspond to the initial formation process of the fiberform nanostructure, which has been observed on tungsten surface [14]. In Fig. 1c, the fiberform nanostructure was formed, and the surface was visually blackened. The blacking indicates that the diffuse reflectivity also decreased by the surface modification. Fig. 2a shows the wavelength dependences of the optical reflectivity of Mo-1 and Mo-2. The reflectivity considerably decreased by the surface modifications due to the helium irradiation. The optical reflectivity was less than 10% for Mo-1, whereas, for Mo-2, it decreased to almost zero. Non-uniformity of Mo-1 surface is effectively averaged for the reflectivity measurement, because the spot size in the measurement is much larger than the crystal grain. Fig. 2b shows the reduction factor of the reflectivity for Mo-1. Here, the reduction factor was defined as the original reflectivity divided
Fig. 1. SEM micrographs of the helium irradiated Mo samples. Mo-1 is shown in (a) and (b), and (c) shows Mo-2. The irradiation temperatures are (a and b) 1200 K (Mo1) and (c) 1500 K (Mo-2). Concerning Mo-1, the surface was rough, and the roughness was different by the crystal grains, while a fiberform nanostructure was formed on Mo-2.
by the decreased reflectivity. At 900 nm, the reduction factor was approximately five; it increased as decreasing the wavelength, and was 30 at 200 nm. In general, optical reflectivity is strongly altered when the size of the structure on the surface is less than several times the wavelength [16]. From Fig. 1b, the size of the structure is less than 1 lm; moreover, there may exist much finer structure on the surface. 3.2. He irradiation at low temperature (<1000 K) (Mo, Rh) The helium irradiation experiments were also conducted at the lower surface temperature. Fig. 3a shows the wavelength dependences of non-irradiated rhodium, helium irradiated rhodium (Rh-1), and helium irradiated molybdenum (Mo-3). The incident ion energy, surface temperature, and helium ion fluence for Rh-1 and Mo-3 were, respectively, 45 eV, 600 K, and 8 1024 m 2 and 45 eV, 850 K, and 3.7 1025 m 2. The exposure to the helium plasmas also decreased the optical reflectivity significantly for both
840
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
60
optical reflectivity [%]
(a)
50 before irradiation (Mo-1,2)
40 30 20
Mo-1
10
Mo-2
0 40
reduction factor
(b)
30 20 Mo-1
10 0 200
300
400
500
600
700
800
900
wavelength [nm] Fig. 2. (a) Wavelength dependences of the optical reflectivity of molybdenum samples before irradiation and after helium irradiations (Mo-1 and Mo-2) are shown. (b) Wavelength dependence of the reduction rate of optical reflectivity for Mo-1. Here, the reduction rate was defined as the reflectivity before irradiation divided by the reflectivity of Mo-1. The irradiation temperatures for Mo-1 and Mo-2 are 1200 and 1500 K, respectively.
(a)
Fig. 4. (a) and (b) show the AFM images of non-irradiated molybdenum and Mo-3, respectively. (c) and (d) show SEM micrographs of non-irradiated rhodium sample and Rh-1, respectively. The irradiation temperatures for Mo-3 and Rh-1 are 850 and 600 K, respectively. By the exposure to the helium plasma, rough surfaces on the scale of from several hundreds of nm to a micrometer are formed.
optical reflectivity [%]
80 before irradiation (Rh-1)
60
Rh-1
40 Mo-3
20
0 6
reduction factor
(b)
5 4 3 Mo-3
2
Rh-1
1 0 200
300
400
500
600
700
800
900
wavelength [nm] Fig. 3. (a) Wavelength dependences of the optical reflectivity of rhodium and molybdenum samples before irradiation and after helium irradiation (Rh-1 and Mo-3) are shown. (b) Wavelength dependence of the reduction rate of optical reflectivity for Rh-1 and Mo-3. The irradiation temperatures for Rh-1 and Mo-3 are 600 and 850 K, respectively.
rhodium and molybdenum even at those temperatures. Fig. 3b shows the reduction factors, which was defined in Section 3.1, for Rh-1 and Mo-3. In both cases, the factor increases as decreasing the wavelength. For Mo-3, the factor increases considerably
particularly in the range of less than 300 nm. For Rh-1, it gradually increases as decreasing the wavelength around 400 nm, and it increases sharply at less than 250 nm. The difference of the factor in the long wavelength range may be caused by the difference in the helium ion fluence. Fig. 4a and b shows the AFM image of a non-irradiated molybdenum sample and Mo-3, respectively. Fig. 4c and d shows the AFM images of a non-irradiated rhodium sample and Rh-1, respectively. Indeed, irregularity with several hundred nm can be seen on the surface of Mo-3, while only scratches due to mechanical polishing are seen before irradiation. On rhodium, the roughness is enhanced, and irregularity with hundreds of nanometer can be seen after the helium plasma irradiation. It can be said that the enhancement of the roughness lead to the decrease of the optical reflectivity for Mo-3 and Rh-1. The formation of the rough structure cannot be attributed to the physical sputtering because the incident ion energy is significantly lower than the sputtering threshold, which is approximately 100 eV for molybdenum [17]. It is inferred that nanometer sized helium bubbles are formed in the surface region, and then, the highly pressurized bubbles change the surface morphology in submicrometer scale in the process of coalescence and swelling. It is recognized for tungsten that helium bubbles can be formed by the interaction with thermal vacancies even if the incident ion energy is significantly lower than the threshold energy for displacement on tungsten atom [14]. It is worthwhile to note that the helium bubbles may be formed even if the surface temperature and incident ion energy are significantly lower than 1000 K and 100 eV, respectively. For rhodium, however, since the thickness of rhodium layer is 50 nm, we cannot exclude the possibility that the coating changed the helium irradiation effects. The irradiation with using a high quality rhodium mirror such as the one which has been developed recently by magnetron sputtering [18] will remain as a future work.
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
4. Discussion and concluding remarks From the helium irradiation at high temperature regime on molybdenum in Section 3.1, following two points have been revealed. (1) Similar as the helium irradiation to tungsten, fiberform nanostructure can be formed on molybdenum surface as well. (2) However, the formation of the structure is not uniform on the surface, indicating that the formation process may be different by grains. The irradiation experiments at low temperature regime shown in Section 3.2 are more important for in-vessel mirror material. Although the temperature was somewhat higher than the typical operational temperature of 400–500 K, the results indicated that the effects of helium bombardment cannot be neglected for in-vessel mirror material when using molybdenum and rhodium. The optical reflectivity decreased significantly with the helium fluence of 1025 m 2 even though the incident ion energy was low, say 50 eV. Even if the helium fluence was on the order of 1024 m 2, the helium bombardment could significantly alter the optical reflectivity in the UV (ultraviolet) range. Particularly, the optical reflectivity in the wavelength range of less than 300–400 nm may be decreased to less than half by the helium bombardment. The decrease in the reflectivity was caused by micro-structural evolutions and irregular structures on the surface. It is likely that the morphology change can be attributed to the helium bubble formation in the surface region. However, we have to say that further TEM analysis is necessary to conclude that the surface modification is caused by the helium bubbles. It is noted that the temperature range in this study was slightly higher than the operational temperature in ITER; it is thought that the irradiation at a lower temperature may decrease the size of helium bubbles, and, consequently, the reduction range in the wavelength may be shifted to shorter. Several points raised as follows were not discussed in this paper and remain as future works. Firstly, pure helium plasma was used for the irradiation in this study; in the actual situation, the mirror will be exposed to helium–deuterium mixture particle flux. Thus, it is of importance to investigate the change of optical reflectivity by the mixture irradiations, particularly, in the light of the effect of blistering. Secondly, since higher energy particles probably formed by charge-exchange process, combination of a small fraction of higher energy helium and hydrogen irradiation will also be important because they should form blisters even though the fluence is significantly small [7,19]. Thirdly, in addition to molybdenum
841
and rhodium used in this study, it is interesting to investigate the helium irradiation effects on other candidate materials for invessel mirrors. In addition to the first mirrors, the effects of helium irradiation on the materials for laser transmission mirror such as copper and silver may be interesting because only a slight decrease in reflectivity may become harmful for them [20]. Acknowledgements The authors thank Dr. Y. Kusama and Dr. T. Sugie at Japan Atomic Energy Agency for fruitful discussions. This work was supported by a Grant-in-Aid for Young Scientists (B), 21760690, from JSPS, NIFS Collaborative Research Program (NIFS09KLGP002), and NIFS/ NINS under the project of Formation of International Network for Scientific Collaborations. References [1] A. Litnovsky, V. Voitsenya, A. Costley, A. Donn´e, Nucl. Fusion 47 (2007) 833– 838. [2] A. Litnovsky, V. Voitsenya, T. Sugie, G.D. Temmerman, A. Costley, A. Donne, K. Vukolov, I. Orlovskiy, J. Brooks, J. Allain, V. Kotov, A. Semerok, P.-Y. Thro, T. Akiyama, N. Yoshida, T. Tokunaga, K. Kawahata, Nucl. Fusion 49 (2009) 075014 (8pp). [3] V.S. Voitsenya, V.G. Konovalov, M.F. Becker, O. Motojima, K. Narihara, B. Schunke, Rev. Sci. Instrum. 70 (1999) 2016–2025. [4] V. Voitsenya, A. Bardamid, V. Bondarenko, W. Jacob, V. Konovalov, S. Masuzaki, O. Motojima, D. Orlinskij, V. Poperenko, I. Ryzhkov, A. Sagara, A. Shtan, S. Solodovchenko, M. Vinnichenko, J. Nucl. Mater. 290–293 (2001) 336–340. [5] D.V. Orlinski, V.S. Voitsenya, K.Y. Vukolov, Plasma Dev. Oper. 15 (2007) 33–75. [6] T. Sugie, S. Kasai, M. Taniguchi, M. Nagatsu, T. Nishitani, J. Nucl. Mater. 329– 333 (2004) 1481–1485. [7] A. Ebihara, M. Tokitani, K. Tokunaga, T. Fujiwara, A. Sagara, N. Yoshida, J. Nucl. Mater. 363–365 (2007) 1195–1200. [8] H. Iwakiri, K. Yasunaga, K. Morishita, N. Yoshida, J. Nucl. Mater. 283–287 (2000) 1134–1138. [9] S. Takamura, N. Ohno, D. Nishijima, S. Kajita, Plasma Fusion Res. 1 (2006) 051. [10] S. Kajita, S. Takamura, N. Ohno, D. Nishijima, H. Iwakiri, N. Yoshida, Nucl. Fusion 47 (2007) 1358. [11] W. Sakaguchi, S. Kajita, N. Ohno, M. Takagi, J. Nucl. Mater. 390–391 (2009) 1149. [12] D. Nishijima, M. Ye, N. Ohno, S. Takamura, J. Nucl. Mater. 329–333 (2004) 1029–1033. [13] D. Nishijima, M.Y. Ye, N. Ohno, S. Takamura, J. Nucl. Mater. 313–316 (2003) 97–101. [14] S. Kajita, W. Sakaguchi, N. Ohno, N. Yoshida, T. Saeki, Nucl. Fusion 49 (2009) 095005. [15] S. Masuzaki, N. Ohno, S. Takamura, J. Nucl. Mater. 223 (1995) 286–293. [16] E. Rephaeli, S. Fan, Appl. Phys. Lett. 92 (2008) 211107. [17] D. Duchs, G. Haas, D. Pfirsch, H. Vernickel, J. Nucl. Mater. 53 (1974) 102–106. [18] L. Marot, G.D. Temmerman, V. Thommen, D. Mathys, P. Oelhafen, Surf. Coat. Technol. 2002 (2008) 2837–2843. [19] M. Tokitani, N. Yoshida, M. Miyamoto, Y. Ohtawa, K. Tokunaga, T. Fujiwara, S. Masuzaki, N. Ashikawa, M. Shoji, M. Kobayashi, A. Sagara, N. Noda, H. Yamada, A. Komori, LHD Experimental Group, S. Nagata, B. Tsuchiyai, J. Nucl. Mater. 386–388 (2009) 173–176. [20] S. Kajita, T. Hatae, V.S. Voitsenya, Plasma Fusion Res. 3 (2008) 032.
Contents lists available at ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Degradation of optical reflectivity of in-vessel mirror materials by helium bombardment Shin Kajita a,⇑, Tsubasa Saeki b, Noriyasu Ohno b, Masayuki Tokitani c, Takaki Hatae d, Wataru Sakaguchi b a
EcoTopia Science Institute, Nagoya Univ., Nagoya 464-8603, Japan Graduate School of Eng., Nagoya Univ., Nagoya 464-8603, Japan c National Institute for Fusion Science, Toki 509-5292, Japan d Japan Atomic Energy Agency, Mukoyama 801-1, Naka, Ibaraki 311-0193, Japan b
a r t i c l e
i n f o
Article history: Available online 25 December 2010
a b s t r a c t The effect of helium irradiation on in-vessel mirror materials, i.e. molybdenum and rhodium, are investigated experimentally. By the exposure to helium plasmas with low incident ion energy (50 eV) at different surface temperatures, the optical reflectivity of molybdenum and rhodium decreases significantly. From the surface analysis, it is shown that fiberlike nanostructure is formed on molybdenum surface when the surface temperature is high (at 1500 K), while rough surface is observed when the surface temperature is low (<1000 K). The decrease in the optical reflectivity is significant particularly for short wavelength ranges, typically, less than 300 nm. The results indicate that the helium irradiation should be taken into account for in-vessel mirror materials for the optical diagnostics in ITER. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction In future experimental fusion devices, many in-vessel mirrors will be used for optical diagnostics. It is generally recognized that metallic mirrors are used for them, and candidate materials for the first mirror are molybdenum, tungsten, copper, stainless steel, and rhodium [1,2]. Although it has been found that neutron irradiation does not have a significant influence on the optical reflectivity [3], there is concern that the reflectivity decreases by the exposure to charge-exchange neutrals, which cause erosion and deposition [4,5]. Until now, irradiation effects of hydrogen isotopes on the reflectivity change have been experimentally investigated [6,5]. However, it is pointed out that the exposure to helium may cause more serious problems by forming blisters [7], helium bubbles [8], and nanostructures [9,10]. Since these damages would result in a considerable decrease in optical reflectivity [11], it is important to investigate the helium-irradiation effects. Helium irradiation to tungsten has been extensively explored by using linear devices and ion beams because tungsten is also the candidate material for the divertor plate in ITER. From those experiments, it was revealed that the temperature and incident ion energy are key factors to cause damages. Micrometer-sized helium bubbles are easily formed when the surface temperature is higher than 1300–1500 K [12,13]. For high energy helium irradiation, typically >1 keV for displacement, blisters are formed [7]. When the incident ion energy is higher than 20 eV and the surface temperature ⇑ Corresponding author. Tel.: +81 52 789 3144; fax: +81 52 789 3944. E-mail address: [email protected] (S. Kajita). 0022-3115/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jnucmat.2010.12.205
is in the range from 1000 to 2000 K, fiberform structure, which leads to significant decrease in optical reflectivity, is easily formed on tungsten surface [14]. The detailed irradiation for tungsten increases an interest to explore the helium irradiation on other candidate mirror materials such as molybdenum and rhodium. There are several reports about the helium irradiation experiments in a high energy range (>1 keV) [7], and high temperature range [14]. However, it seems that there is no report regarding the effects of helium irradiation in low energy and low temperature ranges using the candidate materials. Note that the surface temperature of the first mirror will be 400–500 K if it is sufficiently cooled, and the energy and flux of the particles are different by the location of mirrors. Concerning the energy of helium, it will be mainly less than 100 eV, but higher charge exchange neutrals may hit the mirrors at some locations. In the present paper, we will show the effect of low-energy helium bombardment to molybdenum and rhodium from the experiments conducted in a linear plasma device. After a brief explanation of the experimental setup in Section 2, irradiation effects at high temperature conditions (higher than 1000 K) are shown in Section 3.1. Although the experiments at high temperature do not directly correspond to the investigation for the first mirror, it will be interesting to compare them with the irradiation at a lower temperature range shown in Section 3.2. Discussion and concluding remarks are given in Section 4. 2. Experimental setup The experiments were performed in the linear divertor simulator NAGDIS (NAGoya DIvertor Simulator)-I [15]. Helium plasma
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
839
was generated by DC arc discharge. We used powder metallurgy molybdenum (Nilaco Co.) and rhodium coated steel (Valtech Co.) for specimens. The molybdenum samples were mechanically polished by sandpapers and alumina suspension before the irradiation. For rhodium coating, a barrier layer of nickel is deposited on steel (NSS431DP2) first, and then, rhodium was deposited on the nickel layer. The layer thicknesses of the rhodium and nickel were approximately 50 nm and several lm, respectively. The sizes of the Mo and Rh samples are, respectively, 10 10 0.5 and 25 25 2 mm. The specimens were exposed to the helium plasma in the NAGDIS-I. The surface temperature and incident ion energy were controlled by changing the helium flux to the specimen. The incident ion energy was controlled by electrically biasing the specimen, and was 50 eV; the effect of the high energy particles is beyond the scope of this paper. After the helium plasma irradiation, the samples were analyzed by SEM (Scanning Electron Microscope) and AFM (Atomic Force Microscope). The wavelength dependence of optical reflectivity was measured with a spectrophotometer (Nihon Bunkosya: ARV-47S) for the wavelength from 200 to 900 nm; the angle of incident and reflection from normal direction on sample surface was in 7.5°. The measured reflectivity corresponds to the specular reflectivity. Since the first mirror will be used for transmitting the emission from the plasmas to optical fibers, which is located far-field area, the specular reflectivity rather than diffusive reflectivity may be important as a tool for plasma diagnostics.
3. Results 3.1. He irradiation at high temperature (>1000 K) (Mo) Fig. 1(a–c) shows the helium irradiated Mo surfaces at the surface temperature of (a and b) 1200 K (Mo-1) and (c) 1500 K (Mo-2). The incident helium ion energy were both 50 eV, and the helium ion fluences for Mo-1 and Mo-2 were 1.8 1026 m 2 and 1.1 1027 m 2, respectively. Interestingly, it is seen in Fig. 1a that the surface roughness was considerably different by crystal grains; the surface became significantly rough on some grains, while it was not so rough on some other grains. It is speculated that the difference in roughness may be related with the difference in the crystal orientation. However, the mechanism to arise the difference is yet to be understood. Indeed, it is interesting to note that such a difference by crystal grains has never been observed on a polycrystalline tungsten samples in the experiments by the author’s group. The original surface has scratches due to mechanical polishing as shown later in Fig. 4; however, the rough surface such as in Fig. 1 can not be seen. Fig. 1b shows an enlarged micrograph of Mo-1, representing that there are many pin-holes on the surface and the surface is significantly rough in the central area compared with the surrounding area. The rough surface may probably correspond to the initial formation process of the fiberform nanostructure, which has been observed on tungsten surface [14]. In Fig. 1c, the fiberform nanostructure was formed, and the surface was visually blackened. The blacking indicates that the diffuse reflectivity also decreased by the surface modification. Fig. 2a shows the wavelength dependences of the optical reflectivity of Mo-1 and Mo-2. The reflectivity considerably decreased by the surface modifications due to the helium irradiation. The optical reflectivity was less than 10% for Mo-1, whereas, for Mo-2, it decreased to almost zero. Non-uniformity of Mo-1 surface is effectively averaged for the reflectivity measurement, because the spot size in the measurement is much larger than the crystal grain. Fig. 2b shows the reduction factor of the reflectivity for Mo-1. Here, the reduction factor was defined as the original reflectivity divided
Fig. 1. SEM micrographs of the helium irradiated Mo samples. Mo-1 is shown in (a) and (b), and (c) shows Mo-2. The irradiation temperatures are (a and b) 1200 K (Mo1) and (c) 1500 K (Mo-2). Concerning Mo-1, the surface was rough, and the roughness was different by the crystal grains, while a fiberform nanostructure was formed on Mo-2.
by the decreased reflectivity. At 900 nm, the reduction factor was approximately five; it increased as decreasing the wavelength, and was 30 at 200 nm. In general, optical reflectivity is strongly altered when the size of the structure on the surface is less than several times the wavelength [16]. From Fig. 1b, the size of the structure is less than 1 lm; moreover, there may exist much finer structure on the surface. 3.2. He irradiation at low temperature (<1000 K) (Mo, Rh) The helium irradiation experiments were also conducted at the lower surface temperature. Fig. 3a shows the wavelength dependences of non-irradiated rhodium, helium irradiated rhodium (Rh-1), and helium irradiated molybdenum (Mo-3). The incident ion energy, surface temperature, and helium ion fluence for Rh-1 and Mo-3 were, respectively, 45 eV, 600 K, and 8 1024 m 2 and 45 eV, 850 K, and 3.7 1025 m 2. The exposure to the helium plasmas also decreased the optical reflectivity significantly for both
840
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
60
optical reflectivity [%]
(a)
50 before irradiation (Mo-1,2)
40 30 20
Mo-1
10
Mo-2
0 40
reduction factor
(b)
30 20 Mo-1
10 0 200
300
400
500
600
700
800
900
wavelength [nm] Fig. 2. (a) Wavelength dependences of the optical reflectivity of molybdenum samples before irradiation and after helium irradiations (Mo-1 and Mo-2) are shown. (b) Wavelength dependence of the reduction rate of optical reflectivity for Mo-1. Here, the reduction rate was defined as the reflectivity before irradiation divided by the reflectivity of Mo-1. The irradiation temperatures for Mo-1 and Mo-2 are 1200 and 1500 K, respectively.
(a)
Fig. 4. (a) and (b) show the AFM images of non-irradiated molybdenum and Mo-3, respectively. (c) and (d) show SEM micrographs of non-irradiated rhodium sample and Rh-1, respectively. The irradiation temperatures for Mo-3 and Rh-1 are 850 and 600 K, respectively. By the exposure to the helium plasma, rough surfaces on the scale of from several hundreds of nm to a micrometer are formed.
optical reflectivity [%]
80 before irradiation (Rh-1)
60
Rh-1
40 Mo-3
20
0 6
reduction factor
(b)
5 4 3 Mo-3
2
Rh-1
1 0 200
300
400
500
600
700
800
900
wavelength [nm] Fig. 3. (a) Wavelength dependences of the optical reflectivity of rhodium and molybdenum samples before irradiation and after helium irradiation (Rh-1 and Mo-3) are shown. (b) Wavelength dependence of the reduction rate of optical reflectivity for Rh-1 and Mo-3. The irradiation temperatures for Rh-1 and Mo-3 are 600 and 850 K, respectively.
rhodium and molybdenum even at those temperatures. Fig. 3b shows the reduction factors, which was defined in Section 3.1, for Rh-1 and Mo-3. In both cases, the factor increases as decreasing the wavelength. For Mo-3, the factor increases considerably
particularly in the range of less than 300 nm. For Rh-1, it gradually increases as decreasing the wavelength around 400 nm, and it increases sharply at less than 250 nm. The difference of the factor in the long wavelength range may be caused by the difference in the helium ion fluence. Fig. 4a and b shows the AFM image of a non-irradiated molybdenum sample and Mo-3, respectively. Fig. 4c and d shows the AFM images of a non-irradiated rhodium sample and Rh-1, respectively. Indeed, irregularity with several hundred nm can be seen on the surface of Mo-3, while only scratches due to mechanical polishing are seen before irradiation. On rhodium, the roughness is enhanced, and irregularity with hundreds of nanometer can be seen after the helium plasma irradiation. It can be said that the enhancement of the roughness lead to the decrease of the optical reflectivity for Mo-3 and Rh-1. The formation of the rough structure cannot be attributed to the physical sputtering because the incident ion energy is significantly lower than the sputtering threshold, which is approximately 100 eV for molybdenum [17]. It is inferred that nanometer sized helium bubbles are formed in the surface region, and then, the highly pressurized bubbles change the surface morphology in submicrometer scale in the process of coalescence and swelling. It is recognized for tungsten that helium bubbles can be formed by the interaction with thermal vacancies even if the incident ion energy is significantly lower than the threshold energy for displacement on tungsten atom [14]. It is worthwhile to note that the helium bubbles may be formed even if the surface temperature and incident ion energy are significantly lower than 1000 K and 100 eV, respectively. For rhodium, however, since the thickness of rhodium layer is 50 nm, we cannot exclude the possibility that the coating changed the helium irradiation effects. The irradiation with using a high quality rhodium mirror such as the one which has been developed recently by magnetron sputtering [18] will remain as a future work.
S. Kajita et al. / Journal of Nuclear Materials 417 (2011) 838–841
4. Discussion and concluding remarks From the helium irradiation at high temperature regime on molybdenum in Section 3.1, following two points have been revealed. (1) Similar as the helium irradiation to tungsten, fiberform nanostructure can be formed on molybdenum surface as well. (2) However, the formation of the structure is not uniform on the surface, indicating that the formation process may be different by grains. The irradiation experiments at low temperature regime shown in Section 3.2 are more important for in-vessel mirror material. Although the temperature was somewhat higher than the typical operational temperature of 400–500 K, the results indicated that the effects of helium bombardment cannot be neglected for in-vessel mirror material when using molybdenum and rhodium. The optical reflectivity decreased significantly with the helium fluence of 1025 m 2 even though the incident ion energy was low, say 50 eV. Even if the helium fluence was on the order of 1024 m 2, the helium bombardment could significantly alter the optical reflectivity in the UV (ultraviolet) range. Particularly, the optical reflectivity in the wavelength range of less than 300–400 nm may be decreased to less than half by the helium bombardment. The decrease in the reflectivity was caused by micro-structural evolutions and irregular structures on the surface. It is likely that the morphology change can be attributed to the helium bubble formation in the surface region. However, we have to say that further TEM analysis is necessary to conclude that the surface modification is caused by the helium bubbles. It is noted that the temperature range in this study was slightly higher than the operational temperature in ITER; it is thought that the irradiation at a lower temperature may decrease the size of helium bubbles, and, consequently, the reduction range in the wavelength may be shifted to shorter. Several points raised as follows were not discussed in this paper and remain as future works. Firstly, pure helium plasma was used for the irradiation in this study; in the actual situation, the mirror will be exposed to helium–deuterium mixture particle flux. Thus, it is of importance to investigate the change of optical reflectivity by the mixture irradiations, particularly, in the light of the effect of blistering. Secondly, since higher energy particles probably formed by charge-exchange process, combination of a small fraction of higher energy helium and hydrogen irradiation will also be important because they should form blisters even though the fluence is significantly small [7,19]. Thirdly, in addition to molybdenum
841
and rhodium used in this study, it is interesting to investigate the helium irradiation effects on other candidate materials for invessel mirrors. In addition to the first mirrors, the effects of helium irradiation on the materials for laser transmission mirror such as copper and silver may be interesting because only a slight decrease in reflectivity may become harmful for them [20]. Acknowledgements The authors thank Dr. Y. Kusama and Dr. T. Sugie at Japan Atomic Energy Agency for fruitful discussions. This work was supported by a Grant-in-Aid for Young Scientists (B), 21760690, from JSPS, NIFS Collaborative Research Program (NIFS09KLGP002), and NIFS/ NINS under the project of Formation of International Network for Scientific Collaborations. References [1] A. Litnovsky, V. Voitsenya, A. Costley, A. Donn´e, Nucl. Fusion 47 (2007) 833– 838. [2] A. Litnovsky, V. Voitsenya, T. Sugie, G.D. Temmerman, A. Costley, A. Donne, K. Vukolov, I. Orlovskiy, J. Brooks, J. Allain, V. Kotov, A. Semerok, P.-Y. Thro, T. Akiyama, N. Yoshida, T. Tokunaga, K. Kawahata, Nucl. Fusion 49 (2009) 075014 (8pp). [3] V.S. Voitsenya, V.G. Konovalov, M.F. Becker, O. Motojima, K. Narihara, B. Schunke, Rev. Sci. Instrum. 70 (1999) 2016–2025. [4] V. Voitsenya, A. Bardamid, V. Bondarenko, W. Jacob, V. Konovalov, S. Masuzaki, O. Motojima, D. Orlinskij, V. Poperenko, I. Ryzhkov, A. Sagara, A. Shtan, S. Solodovchenko, M. Vinnichenko, J. Nucl. Mater. 290–293 (2001) 336–340. [5] D.V. Orlinski, V.S. Voitsenya, K.Y. Vukolov, Plasma Dev. Oper. 15 (2007) 33–75. [6] T. Sugie, S. Kasai, M. Taniguchi, M. Nagatsu, T. Nishitani, J. Nucl. Mater. 329– 333 (2004) 1481–1485. [7] A. Ebihara, M. Tokitani, K. Tokunaga, T. Fujiwara, A. Sagara, N. Yoshida, J. Nucl. Mater. 363–365 (2007) 1195–1200. [8] H. Iwakiri, K. Yasunaga, K. Morishita, N. Yoshida, J. Nucl. Mater. 283–287 (2000) 1134–1138. [9] S. Takamura, N. Ohno, D. Nishijima, S. Kajita, Plasma Fusion Res. 1 (2006) 051. [10] S. Kajita, S. Takamura, N. Ohno, D. Nishijima, H. Iwakiri, N. Yoshida, Nucl. Fusion 47 (2007) 1358. [11] W. Sakaguchi, S. Kajita, N. Ohno, M. Takagi, J. Nucl. Mater. 390–391 (2009) 1149. [12] D. Nishijima, M. Ye, N. Ohno, S. Takamura, J. Nucl. Mater. 329–333 (2004) 1029–1033. [13] D. Nishijima, M.Y. Ye, N. Ohno, S. Takamura, J. Nucl. Mater. 313–316 (2003) 97–101. [14] S. Kajita, W. Sakaguchi, N. Ohno, N. Yoshida, T. Saeki, Nucl. Fusion 49 (2009) 095005. [15] S. Masuzaki, N. Ohno, S. Takamura, J. Nucl. Mater. 223 (1995) 286–293. [16] E. Rephaeli, S. Fan, Appl. Phys. Lett. 92 (2008) 211107. [17] D. Duchs, G. Haas, D. Pfirsch, H. Vernickel, J. Nucl. Mater. 53 (1974) 102–106. [18] L. Marot, G.D. Temmerman, V. Thommen, D. Mathys, P. Oelhafen, Surf. Coat. Technol. 2002 (2008) 2837–2843. [19] M. Tokitani, N. Yoshida, M. Miyamoto, Y. Ohtawa, K. Tokunaga, T. Fujiwara, S. Masuzaki, N. Ashikawa, M. Shoji, M. Kobayashi, A. Sagara, N. Noda, H. Yamada, A. Komori, LHD Experimental Group, S. Nagata, B. Tsuchiyai, J. Nucl. Mater. 386–388 (2009) 173–176. [20] S. Kajita, T. Hatae, V.S. Voitsenya, Plasma Fusion Res. 3 (2008) 032.