- Email: [email protected]
Tungsten migration studies by controlled injection of volatile compounds
Journal of Nuclear Materials 438 (2013) S170–S174
Contents lists available at SciVerse ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Tungsten migration studies by controlled injection of volatile compounds M. Rubel a,⇑, J. Coenen b, D. Ivanova a, S. Möller b, P. Petersson a, S. Brezinsek b, A. Kreter b, V. Philipps b, A. Pospieszczyk b, B. Schweer b a b
Royal Institute of Technology (KTH), Association EURATOM-VR, Stockholm, Sweden IEK-4, Plasma Physics, Forschungszentrum Jülich, Association EURATOM-FZJ, Jülich, Germany
a r t i c l e
i n f o
Article history: Available online 16 January 2013
a b s t r a c t Volatile tungsten hexa-fluoride was locally injected into the TEXTOR tokamak as a marker for material migration studies. The injection was accompanied by puffing N-15 rare isotope as a nitrogen tracer in discharges with edge cooling by impurity seeding. The objective was to assess material balance by qualitative and quantitative determination of a global and local deposition pattern, material mixing effects and fluorine residence in plasma-facing components. Spectroscopy and ex situ ion beam analysis techniques were used. Tungsten was detected on all types of limiter tiles and short-term probes retrieved from the vessel. Over 80% of the injected W was identified. The largest tungsten concentration, 1 1018 cm 2, was in the vicinity of the gas inlet. Co-deposits contained tungsten and a mix of light isotopes: H, D, He-4, B-10, B-11, C-12, C-13, N-14, N-15, O-16 and small quantities of F-19 thus showing that both He and nitrogen are trapped following wall conditioning (He glow) and edge cooling. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Tungsten (W), as a material for plasma-facing components (PFCs), has been used in tokamaks since 1970 but early experience in Doublet-II [1], ORMAK [2] and PLT [3] was not encouraging because of heavy radiation losses [2]. Advantages and drawbacks of high-Z components were reviewed by Tanabe et al. [4] and application of refractory metals was re-considered. An intense experimental program started at TEXTOR with molybdenum (Mo) [4–8] and tungsten [8–24] test limiters, and in ASDEX-Upgrade with tungsten [25–32] in the divertor and then on the entire vessel wall [31,32]. PFC tiles are made either of bulk [8–18] and nano-structured metal [19] or in the form of coatings deposited on graphite [20–32]. As a result of those studies and comparison of high-Z refractory metals with carbon, tungsten was selected for a part or for the entire ITER divertor [33]. To perform a large scale material test the ITER-Like Wall (ILW) Project at the JET tokamak was initiated. The refurbishment was accomplished [34–43] and JET is operated with tungsten in the divertor and beryllium on the main chamber wall [44]. Such broad application of tungsten calls for developing techniques enabling studies of its migration. There are several ways of approaching this goal using markers: (i) enrichment with a stable isotope; (ii) using another refractory metal being a proxy to tungsten deposited as a coating on W or (iii) a volatile compound. Tungsten has five stable isotopes: 180W 0.1%, ⇑ Corresponding author. E-mail address: [email protected] (M. Rubel). 0022-3115/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jnucmat.2013.01.053
182
W 26.3%, 183W 14.3%, 184W 30.7% and 186W 28.6%. The isotope enrichment may be performed for the light (182) or the heaviest (186) one. The isotope separation procedure from tungsten hexafluoride (WF6) based on centrifuge would be very expensive, whereas the outcome could be doubtful because there is no quantitative surface analysis technique capable of reliable distinguishing heavy isotopes of similar masses (resolution problem). The second concept can rely on Mo markers as it is applied to study erosion–deposition phenomena in the tungsten divertor of JET-ILW [45]. The changes in the layer thickness can be measured with low or high energy ion beam analysis (IBA) methods. A potential risk for the assessment is Mo–W inter-diffusion and/or alloying at high temperatures. The last approach based on application of a volatile compound, e.g. tungsten hexafluoride (WF6), can be considered as a method for local repair of coatings on PFC, as proposed in [46]. The use of tungsten PFC requires impurity seeding for improved edge radiation [47–49]. The injection of neon (Ne), argon (Ar), or nitrogen (N2) is performed for that purpose. Among many issues related to the injection is the in-vessel residence of gas by implantation, co-deposition or by compound formation with PFC materials [50–52]. The latter may become important in the case of nitrogen–tungsten combination. To address W transport and the change of PFC surface morphology in the presence of nitrogen dedicated experiments were performed in the TEXTOR tokamak by injection of WF6 and 15N2. The aim was to assess: (a) material balance by qualitative and quantitative determination of a global and local deposition pattern of tungsten and local of nitrogen; (b) material mixing; and (c) fluorine residence in PFC.
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174
2. Experimental Two experiments with the localized WF6 injection were performed at the TEXTOR tokamak prior to the major shutdowns (years 2008 and 2011) connected with the retrieval of tiles for ex situ analysis. They were done using a test limiter lock located at the bottom of the machine. Fig. 1a shows the location of the lock with respect to the nearest blades of the toroidal belt pump limiter ALT-II (Advanced Limiter Test) which is the main PFC of TEXTOR defining the minor radius a = 46 cm. An assembly of the test limiter is shown in Fig. 1b: a roof-shaped block with a polished plate (both made of graphite) with a hole for WF6 puffing. It was placed in the scrape-off-layer (SOL). The first experiment (Exp. I) with the injection of 2 1020 WF6 molecules in 1.2 s long puffs during seven shots was done, on purpose, 50 shots before the machine opening. The second (Exp. II) was performed on the last operation session when puffing of 1.93 1020 W atoms in 13 shots heated by neutral beam injection (NBI) was accompanied by puffing two other markers: Nitrogen-15 from the toroidal inlets (3.46 1021 at) and 13CH4 (1.75 1021 C-13 at) from the upper test limiter. Major parameters for the two experiments are summarized in Table 1, whereas Fig. 2 shows the timing of gas and neutral beam injection. Local spectroscopy measurements were focused on the test limiter where NII (451.4 nm), WI (400.8 nm), CII (426.7 nm), FII (389.8 nm) and De (397.0 nm) lines were recorded. For the determination of the relative nitrogen fluxes (with respect to Hd) NII line 500.1 nm has been used. The fluxes at the SOL were estimated from the S/XB line ratio: nitrogen accounted for 5–10% of the total flux. The W band at 5 nm and FVI line at 53.521 nm were recorded in core plasma. The exposures were followed by ion and electron beam analyses. Nuclear reaction (NRA), Rutherford backscattering spectroscopy (RBS), time-of-flight heavy ion elastic recoil detection (ToF HIERDA), electron probe microanalysis (EPMA) and/or energy dispersive X-ray spectroscopy (EDX) were done on a large number of tiles from the toroidal and inner bumper limiter (a shield for dynamic ergodig divertor, DED) and, on the plates attached to the test limiter. NRA with a 3He+ ion beam was used for deuterium 3 He(d,p)4He and for carbon 3He (13C,p)15N using a standard 2 MeV beam with 1 mm2 spot. For ToF-HIERDA a 26 MeV 127I7+ beam was used. Measurements can only be performed on smooth surfaces, therefore, they were restricted to the polished graphite plate. The information depth in carbon was approximately 400 nm and the depth resolution of about 20 nm. 3. Results and discussion 3.1. Spectroscopy Temporal evolution of tungsten and fluorine fluxes at the test limiter is plotted in Fig. 3 for the first and the last shot with WF6
a
S171
injection in Exp. II. The data are normalized with respect to the intensity of the De line. The main message is that for a given species the shape and intensity of the signals are fairly similar. Some differences observed, as expected, at the beginning of the discharge level already after 0.7 s into the gas puff. Similar behavior was also recorder during Exp. I both at the edge and core plasma. W and F fluxes where at background level when there was a one pulse break between discharges with injection. This weak ‘‘memory’’ effect may indicate that fluorine retention on the test limiter is not pronounced, i.e. species is effectively transported to other PFC or to pumps.
3.2. Deposition profiles The injection of WF6 lead to the formation of a co-deposit on the graphite plate in the vicinity of the gas inlet, as seen in the photo, Fig. 1b, taken after the limiter exposure to plasma in Exp. II. A similar deposition pattern was observed after the first experiment with WF6 injection. The deposited layer, as identified with ToF HIERDA contains a mixture of light and heavy species: H, D, He, 10 B, 11B, 12C, 13C, 14N, 15N, 16O, F and W accompanied by small quantities of Inconel components (Ni, Cr and Fe). The concentration varies, but the greatest amounts are found near the gas inlet: up to 1 1018 W cm 2, N-15 (3 1016 cm 2), F (2 1016 cm 2) and He (1 1017 cm 2). These measurements reveal only a small quantity of fluorine in comparison to other species, especially to tungsten. They also prove and confirm nitrogen retention in deposits [50–52]. Helium originating from regular glow discharge wall conditioning and He-beam diagnostics is identified again [51] in deposit from TEXTOR. Its concentration is fairly high with respect to other gaseous elements in the analyzed layer. On the molybdenum catcher plates attached to the test limiter base the quantities of W, 15N and F are below the detection limit. In the examination of the ALT-II tiles the emphasis was on differences between tungsten deposition patterns on plasma-wetted areas (erosion zone) and in the deposition zone where the eroded species are eventually resting. The standard pattern for deuterium and impurity atoms on ALT-II after entire campaigns has been shown previously [53,54]: tiny amount of species in the erosion zone and a sharp increase (by orders of magnitude) in the deposition region. Fig. 4a shows tungsten deposition profiles for two ALTII tiles retrieved from TEXTOR 50 discharges after the last shot with WF6: one tile located near the injection point and another 180° toroidally apart. Both profiles are fairly similar though there is a certain difference in the total amount of deposited tungsten; more is found closer to the injector. They correspond to the standard pattern on tiles after long-term operation. The profiles in Fig. 4b measured on tiles retrieved just after the experiment reveal irregular behavior. There is an increase into the deposition zone and then
b WF6
Fig. 1. Top view into the TEXTOR vacuum vessel showing the location of the test limiter with respect to ALT-II limiter blades with marked position of corner tiles (a); and test limiter assembly with marked components of the set-up (b).
S172
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174 Table 1 Experimental conditions in tracer experiments with the WF6 injection. Parameter
Experiment I
Experiment II
ALT-II (cm) Test limiter (cm) ne (1019 m 3) WF6 (W atoms) 15 N2 (15N atoms) 13 CH4 (13C atoms) Probe retrieval PFC retrieval
46 47.5 (1.5 cm in SOL) 3.2 2 1020 in 7 shots with 1–2 pulse breaks between injections – – Directly after experiment 50 shots after last W injection
46 48 (2.0 cm in SOL) 3.5 1.93 1020 in 13 shots 3.46 1021 in 14 shots 1.75 1021 in 14 shots Directly after experiment Directly after experiment
Fig. 2. Operation plan of the experiment: timing of NBI heating and injection of marker gases.
Fig. 3. Normalized spectroscopy signals of WI and FII lines for the first and the last discharge with the WF6 injection.
sharp decrease. The graph in Fig. 4c reflects the deposition in the area nearest to the injection point in Exp. II, corner Tile 28 marked in Fig. 1a, when ex situ PFC studies were done directly after the WF6 injection. The profile is wave-shaped with a concentration maximum in the central part of the tile, i.e. in the area known as the erosion zone on ALT-II. The same shape of the tungsten distribution has been found on the other corner plate (Tile 15) from the adjacent ALT-II blade, see Fig. 1a. W content on these two corner tiles is greater than measured on other plates from various toroidal locations. This deposition pattern is a net result of global and local transport phenomena. It reflects the migration of tungsten from the net erosion area where it is originally deposited by plasma flux to areas where it rests: deposition–re-erosion–ionization–re-deposition
Fig. 4. Deposition profiles of tungsten on the ALT-II toroidal limiter tiles detected: (a) 50 shots after the WF6 injection, Exp. I; (b) and (c) after Exp. II in areas located at different positions with respect to the gas injector.
cycle. Erosion of tungsten occurs predominantly via physical sputtering. At the edge temperature Te of about 60 eV it is eroded mainly by plasma impurity ions (carbon, oxygen and boron), but the contribution of deuterium to the overall process is not excluded. The majority of sputtered species are neutrals with mean energy of 5–7 eV. At the edge density ne 5 1018 m 3 their mean free path is 2–3 mm (normal to the limiter surface) before they get ionized [8]. As predicted by Naujoks and Behrisch [55], species ionized close to the limiter surface may promptly be re-deposited within the first gyro orbit (Larmor radius is about 2.0 mm under
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174
the experimental conditions). The results of surface analysis explain gradual tungsten migration on ALT-II. They also clearly demonstrate the agreement with the model. The process leads to the ‘‘spill over’’ of tungsten on graphite until the equilibrium between the re-erosion and re-deposition is reached. The other point is the micro-distribution of tungsten especially in the plasma-wetted area. In a backscattered electron image reflecting the contrast between low- and high-Z species dark fields corresponds to carbon and brighter spots related to the accumulation of high-Z elements were recorded. Tungsten was also present as identified with EDX. Metal resides in small pits acting as local shadowed regions. It explains why some tungsten is found in the erosion zone after long operation periods. This is in agreement with previous studies on micro-distribution of fuel and plasma impurities on PFC [54,56,57]. 3.3. Tungsten balance The integrated quantity of tungsten found on the graphite plate near the gas inlet corresponds to 1% and 2–3% of the injected amount in Exp. I and II, respectively. This very small local deposition of W may be attributed to at least two factors: low sticking probability of the injected gas to the plate and/or efficient erosion of the locally deposited tungsten. The large Larmor radius, as discussed in the previous section, would eventually increase the migration of W from the vicinity of the injection point. The amounts deposited locally are small in both cases, but the difference between the two experiments is significant. It may be associated with the differences in the limiter position and plasma density, see Table 1. In Exp. I the limiter was 0.5 cm closer to plasma and the electron density was lower than in Exp. II, thus creating more favorable conditions for erosion. After each experiment tungsten on ALT-II tiles from several toroidal locations was analyzed. The integrated amount corresponded to a small fraction of the injected atoms: 6.5% in the first case and 10.5% in the second. Summing up the amount from the test and toroidal limiter only 8–13% of the species could be identified. Again, smaller amount was detected was after Exp. I when the ALT-II tiles were still in TEXTOR for 50 shots after the WF6 injection. This clearly motivates a search for another residence place of eroded tungsten. For the first time several DED tiles from various locations were studied. The W content ranges from 1 to 25 1015 cm 2. Taking 3 1015 cm 2 as a mean value, the integrated amount on DED is approximately 2 1020 W atoms, i.e. exceeding the amount puffed in Exp. II. It is impossible to assess accurately the amount originating from the last experiment because the tiles were in TEXTOR for 8 years. As determined by RBS, the tiles also contain other metals, e.g. Inconel components. The results lead to a conclusion that the inner bumper is a major residence place for high-Z metals eroded from the TEXTOR liner, various test limiters, and those introduced by injection. Also the deuterium content on the DED tiles is significant: 3–4 1018 cm 2 leading to the integrated amount in the range 2.0–2.5 1023 D atoms accumulated in that region. 4. Concluding remarks The reported tracer experiments, probably the most complex ever done in a tokamak, have resulted in several major contributions to material migration studies. The most important is the direct demonstration of tungsten erosion and its subsequent migration over long distances by re-erosion–ionization–re-deposition steps. To the authors’ knowledge, it is the first direct and so clear prove of the prompt re-deposition model predicted by Naujoks and Behrisch [55], though there was an earlier indication
S173
following tests of C & W twin limiter in TEXTOR [13]. The transport is gradual (see Fig. 4) but very effective under experimental conditions, i.e. edge Te 60 eV. One may assume that so significant W migration will be at least partly suppressed at low edge temperature and high divertor densities, as expected in ITER [58]. However, this would not eliminate metal erosion followed by prompt redeposition and local formation of modified mixed material layers. Low sticking probability of WF6 to the plate or efficient erosion at the test limiter causes that only a small fraction (1–3%) of the injected W amount is deposited near the gas inlet. Therefore, the concept of injection-based local repair of tungsten coatings could be conceived as a possible option only under low Te and high edge density regime. Low W content detected on the toroidal limiters (6–11%) inspired a search for a major place of its deposition. Surface studies performed for the first time on the DED tiles allowed for the identification of a major deposition area for highZ species eroded from the wall or introduced by injection. In summary, experimental results motivate comprehensive modeling to improve understanding of tungsten transport. Volatile WF6 can be used as a high-Z transport marker without a major risk of leaving large quantities of fluorine sticking (co-deposited) to PFC and then released to plasma during subsequent pulses. The statement is justified by two facts: (i) the memory effect is not pronounced, as proven by spectroscopy and (ii) small quantities of F (maximum 2 1016 cm 2 versus 1 1018 cm 2) are found even near the injection point. The application of the 15N tracer and a non-standard analysis method, i.e. HIERDA, have allowed for conclusive identification of nitrogen deposited on PCF. Though the exact material balance could not determined, the result shows effective co-deposition of toroidally puffed nitrogen. However, earlier experiments in TEXTOR [51] and recent in ASDEX [52] indicate the retention at the level of 30% of the injected amount. Also helium co-deposition and retention has been confirmed [51]. It is associated with He glow discharge. The level is quite significant (1 1017 cm 2). Therefore, it can be recommended to study helium trapping following He-based ion cyclotron wall conditioning [59], i.e. a method considered for a reactor-class machine. Acknowledgments This work, supported partly by the European Communities under the Contracts of Association between EURATOM–VR and EURATOM–FZJ, was carried out within the framework of the European Fusion Development Agreement. The views and opinions expressed herein do not necessarily reflect those of the European Commission. This work was partly funded by the Swedish Research Council (VR) through Contract No. 621-2009-4138. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
R.K. Fischer et al., Phys. Rev. Lett. 39 (1977) 622. J. Sugar, V. Kaufman, Phys. Rev. A 21 (1980) 2096. E. Hinnov et al., Nucl. Fusion 18 (1978) 1305. T. Tanabe, N. Noda, H. Nakamura, J. Nucl. Mater. 196–198 (1992) 11. V. Philipps et al., Nucl. Fusion 34 (1994) 1417. Y. Ueda et al., J. Nucl. Mater. 220–222 (1995) 240. T. Tanabe et al., J. Nucl. Mater. 212–215 (1994) 1370. M. Rubel et al., J. Nucl. Mater. 249 (1997) 116. V. Philipps et al., J. Nucl. Mater. 258–263 (1998) 858. T. Tanabe et al., J. Nucl. Mater. 283–287 (2000) 1128. J. Rapp et al., Nucl. Fusion 37 (1997) 1691. V. Philipps et al., Plasma Phys. Controlled Fusion 42 (2000) B293. M. Rubel et al., J. Nucl. Mater. 283–287 (2000) 1089. T. Hirai et al., Phys. Scr. T103 (2003) 59. T. Hirai et al., J. Nucl. Mater. 313–316 (2003) 67. M. Rubel et al., Phys. Scr. T81 (1999) 61. G. Sergienko et al., Phys. Scr. T128 (2007) 81. G. Sergienko et al., J. Nucl. Mater. 363–365 (2007) 96. Y. Ueda et al., J. Nucl. Mater. 415 (2011) S92.
S174 [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39]
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174 Y. Ueda et al., J. Nucl. Mater. 363–365 (2007) 66. J. Pospieszczyk et al., Nucl. Mater. 290–293 (2000) 947. M. Rubel et al., J. Nucl. Mater. 307–311 (2002) 111. E. Fortuna et al., Phys. Scr. T128 (2007) 162. M. Psoda et al., J. Nucl. Mater. 386–388 (2009) 740. H. Maier et al., J. Nucl. Mater. 258–263 (1998) 921. K. Krieger, H. Maier, R. Neu, J. Nucl. Mater. 266–269 (1999) 207. R. Neu et al., J. Nucl. Mater. 290–293 (2001) 206. K. Krieger, R. Neu, J. Nucl. Mater. 266–269 (1999) 207. K. Krieger et al., J. Nucl. Mater. 337–339 (2005) 10. R. Neu et al., J. Nucl. Mater. 363–365 (2007) 52. V. Rohde et al., J. Nucl. Mater. 415 (2011) S46. R. Neu et al., J. Nucl. Mater. 415 (2011) S322. G. Federici et al., Nucl. Fusion 41 (2001) 1967. G.F. Matthews et al., Phys. Scr. T128 (2007) 137. G.F. Matthews et al., Phys. Scr. T138 (2009) 014030. G.F. Matthews et al., Phys. Scr. T145 (2011) 014001. T. Hirai et al., Phys. Scr. T128 (2007) 144. R. Neu et al., Phys. Scr. T128 (2007) 150. C. Ruset et al., Phys. Scr. T128 (2007) 171.
[40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59]
H. Maier et al., Nucl. Fusion 47 (2007) 222. Ph. Mertens et al., Fusion Eng. Des. 84 (2009) 1289. Ph. Mertens et al., Fusion Eng. Des. 86 (2011) 1801. Ph. Mertens, Phys. Scr. T145 (2011) 014002. G.F. Matthews, This conference. M. Rubel et al., This conference. V. Philipps et al., Phys. Scr. T145 (2011) 0140. A. Kallenbach et al., J. Nucl. Mater. 415 (2011) S19. S. Brezinsek et al., Nucl. Fusion 51 (2011) 073007. R. Neu et al., Plasma Phys. Controlled Fusion 53 (2011) 124040. M. Rubel et al., J. Nucl. Mater. 415 (2011) S223. P. Petersson et al., Nucl. Instrum. Methods B273 (2012) 113. P. Petersson et al., This conference. M. Rubel et al., J. Nucl. Mater. 290–293 (2002) 473. M. Rubel et al., Fusion Eng. Des. 81 (2006) 211. D. Naujoks, R. Behrisch, J. Nucl. Mater. 220–222 (1995) 227. P. Petersson et al., Nucl. Instrum. Method B 268 (2010) 1833. P. Petersson et al., Phys. Scr. T145 (2011) 014014. K. Ikeda et al., Nucl. Fusion 47 (2007) (Special Issue ITER Physics Basis). A. Lyssoivan et al., J. Nucl. Mater. 415 (2011) S1029.
Contents lists available at SciVerse ScienceDirect
Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat
Tungsten migration studies by controlled injection of volatile compounds M. Rubel a,⇑, J. Coenen b, D. Ivanova a, S. Möller b, P. Petersson a, S. Brezinsek b, A. Kreter b, V. Philipps b, A. Pospieszczyk b, B. Schweer b a b
Royal Institute of Technology (KTH), Association EURATOM-VR, Stockholm, Sweden IEK-4, Plasma Physics, Forschungszentrum Jülich, Association EURATOM-FZJ, Jülich, Germany
a r t i c l e
i n f o
Article history: Available online 16 January 2013
a b s t r a c t Volatile tungsten hexa-fluoride was locally injected into the TEXTOR tokamak as a marker for material migration studies. The injection was accompanied by puffing N-15 rare isotope as a nitrogen tracer in discharges with edge cooling by impurity seeding. The objective was to assess material balance by qualitative and quantitative determination of a global and local deposition pattern, material mixing effects and fluorine residence in plasma-facing components. Spectroscopy and ex situ ion beam analysis techniques were used. Tungsten was detected on all types of limiter tiles and short-term probes retrieved from the vessel. Over 80% of the injected W was identified. The largest tungsten concentration, 1 1018 cm 2, was in the vicinity of the gas inlet. Co-deposits contained tungsten and a mix of light isotopes: H, D, He-4, B-10, B-11, C-12, C-13, N-14, N-15, O-16 and small quantities of F-19 thus showing that both He and nitrogen are trapped following wall conditioning (He glow) and edge cooling. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Tungsten (W), as a material for plasma-facing components (PFCs), has been used in tokamaks since 1970 but early experience in Doublet-II [1], ORMAK [2] and PLT [3] was not encouraging because of heavy radiation losses [2]. Advantages and drawbacks of high-Z components were reviewed by Tanabe et al. [4] and application of refractory metals was re-considered. An intense experimental program started at TEXTOR with molybdenum (Mo) [4–8] and tungsten [8–24] test limiters, and in ASDEX-Upgrade with tungsten [25–32] in the divertor and then on the entire vessel wall [31,32]. PFC tiles are made either of bulk [8–18] and nano-structured metal [19] or in the form of coatings deposited on graphite [20–32]. As a result of those studies and comparison of high-Z refractory metals with carbon, tungsten was selected for a part or for the entire ITER divertor [33]. To perform a large scale material test the ITER-Like Wall (ILW) Project at the JET tokamak was initiated. The refurbishment was accomplished [34–43] and JET is operated with tungsten in the divertor and beryllium on the main chamber wall [44]. Such broad application of tungsten calls for developing techniques enabling studies of its migration. There are several ways of approaching this goal using markers: (i) enrichment with a stable isotope; (ii) using another refractory metal being a proxy to tungsten deposited as a coating on W or (iii) a volatile compound. Tungsten has five stable isotopes: 180W 0.1%, ⇑ Corresponding author. E-mail address: [email protected] (M. Rubel). 0022-3115/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jnucmat.2013.01.053
182
W 26.3%, 183W 14.3%, 184W 30.7% and 186W 28.6%. The isotope enrichment may be performed for the light (182) or the heaviest (186) one. The isotope separation procedure from tungsten hexafluoride (WF6) based on centrifuge would be very expensive, whereas the outcome could be doubtful because there is no quantitative surface analysis technique capable of reliable distinguishing heavy isotopes of similar masses (resolution problem). The second concept can rely on Mo markers as it is applied to study erosion–deposition phenomena in the tungsten divertor of JET-ILW [45]. The changes in the layer thickness can be measured with low or high energy ion beam analysis (IBA) methods. A potential risk for the assessment is Mo–W inter-diffusion and/or alloying at high temperatures. The last approach based on application of a volatile compound, e.g. tungsten hexafluoride (WF6), can be considered as a method for local repair of coatings on PFC, as proposed in [46]. The use of tungsten PFC requires impurity seeding for improved edge radiation [47–49]. The injection of neon (Ne), argon (Ar), or nitrogen (N2) is performed for that purpose. Among many issues related to the injection is the in-vessel residence of gas by implantation, co-deposition or by compound formation with PFC materials [50–52]. The latter may become important in the case of nitrogen–tungsten combination. To address W transport and the change of PFC surface morphology in the presence of nitrogen dedicated experiments were performed in the TEXTOR tokamak by injection of WF6 and 15N2. The aim was to assess: (a) material balance by qualitative and quantitative determination of a global and local deposition pattern of tungsten and local of nitrogen; (b) material mixing; and (c) fluorine residence in PFC.
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174
2. Experimental Two experiments with the localized WF6 injection were performed at the TEXTOR tokamak prior to the major shutdowns (years 2008 and 2011) connected with the retrieval of tiles for ex situ analysis. They were done using a test limiter lock located at the bottom of the machine. Fig. 1a shows the location of the lock with respect to the nearest blades of the toroidal belt pump limiter ALT-II (Advanced Limiter Test) which is the main PFC of TEXTOR defining the minor radius a = 46 cm. An assembly of the test limiter is shown in Fig. 1b: a roof-shaped block with a polished plate (both made of graphite) with a hole for WF6 puffing. It was placed in the scrape-off-layer (SOL). The first experiment (Exp. I) with the injection of 2 1020 WF6 molecules in 1.2 s long puffs during seven shots was done, on purpose, 50 shots before the machine opening. The second (Exp. II) was performed on the last operation session when puffing of 1.93 1020 W atoms in 13 shots heated by neutral beam injection (NBI) was accompanied by puffing two other markers: Nitrogen-15 from the toroidal inlets (3.46 1021 at) and 13CH4 (1.75 1021 C-13 at) from the upper test limiter. Major parameters for the two experiments are summarized in Table 1, whereas Fig. 2 shows the timing of gas and neutral beam injection. Local spectroscopy measurements were focused on the test limiter where NII (451.4 nm), WI (400.8 nm), CII (426.7 nm), FII (389.8 nm) and De (397.0 nm) lines were recorded. For the determination of the relative nitrogen fluxes (with respect to Hd) NII line 500.1 nm has been used. The fluxes at the SOL were estimated from the S/XB line ratio: nitrogen accounted for 5–10% of the total flux. The W band at 5 nm and FVI line at 53.521 nm were recorded in core plasma. The exposures were followed by ion and electron beam analyses. Nuclear reaction (NRA), Rutherford backscattering spectroscopy (RBS), time-of-flight heavy ion elastic recoil detection (ToF HIERDA), electron probe microanalysis (EPMA) and/or energy dispersive X-ray spectroscopy (EDX) were done on a large number of tiles from the toroidal and inner bumper limiter (a shield for dynamic ergodig divertor, DED) and, on the plates attached to the test limiter. NRA with a 3He+ ion beam was used for deuterium 3 He(d,p)4He and for carbon 3He (13C,p)15N using a standard 2 MeV beam with 1 mm2 spot. For ToF-HIERDA a 26 MeV 127I7+ beam was used. Measurements can only be performed on smooth surfaces, therefore, they were restricted to the polished graphite plate. The information depth in carbon was approximately 400 nm and the depth resolution of about 20 nm. 3. Results and discussion 3.1. Spectroscopy Temporal evolution of tungsten and fluorine fluxes at the test limiter is plotted in Fig. 3 for the first and the last shot with WF6
a
S171
injection in Exp. II. The data are normalized with respect to the intensity of the De line. The main message is that for a given species the shape and intensity of the signals are fairly similar. Some differences observed, as expected, at the beginning of the discharge level already after 0.7 s into the gas puff. Similar behavior was also recorder during Exp. I both at the edge and core plasma. W and F fluxes where at background level when there was a one pulse break between discharges with injection. This weak ‘‘memory’’ effect may indicate that fluorine retention on the test limiter is not pronounced, i.e. species is effectively transported to other PFC or to pumps.
3.2. Deposition profiles The injection of WF6 lead to the formation of a co-deposit on the graphite plate in the vicinity of the gas inlet, as seen in the photo, Fig. 1b, taken after the limiter exposure to plasma in Exp. II. A similar deposition pattern was observed after the first experiment with WF6 injection. The deposited layer, as identified with ToF HIERDA contains a mixture of light and heavy species: H, D, He, 10 B, 11B, 12C, 13C, 14N, 15N, 16O, F and W accompanied by small quantities of Inconel components (Ni, Cr and Fe). The concentration varies, but the greatest amounts are found near the gas inlet: up to 1 1018 W cm 2, N-15 (3 1016 cm 2), F (2 1016 cm 2) and He (1 1017 cm 2). These measurements reveal only a small quantity of fluorine in comparison to other species, especially to tungsten. They also prove and confirm nitrogen retention in deposits [50–52]. Helium originating from regular glow discharge wall conditioning and He-beam diagnostics is identified again [51] in deposit from TEXTOR. Its concentration is fairly high with respect to other gaseous elements in the analyzed layer. On the molybdenum catcher plates attached to the test limiter base the quantities of W, 15N and F are below the detection limit. In the examination of the ALT-II tiles the emphasis was on differences between tungsten deposition patterns on plasma-wetted areas (erosion zone) and in the deposition zone where the eroded species are eventually resting. The standard pattern for deuterium and impurity atoms on ALT-II after entire campaigns has been shown previously [53,54]: tiny amount of species in the erosion zone and a sharp increase (by orders of magnitude) in the deposition region. Fig. 4a shows tungsten deposition profiles for two ALTII tiles retrieved from TEXTOR 50 discharges after the last shot with WF6: one tile located near the injection point and another 180° toroidally apart. Both profiles are fairly similar though there is a certain difference in the total amount of deposited tungsten; more is found closer to the injector. They correspond to the standard pattern on tiles after long-term operation. The profiles in Fig. 4b measured on tiles retrieved just after the experiment reveal irregular behavior. There is an increase into the deposition zone and then
b WF6
Fig. 1. Top view into the TEXTOR vacuum vessel showing the location of the test limiter with respect to ALT-II limiter blades with marked position of corner tiles (a); and test limiter assembly with marked components of the set-up (b).
S172
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174 Table 1 Experimental conditions in tracer experiments with the WF6 injection. Parameter
Experiment I
Experiment II
ALT-II (cm) Test limiter (cm) ne (1019 m 3) WF6 (W atoms) 15 N2 (15N atoms) 13 CH4 (13C atoms) Probe retrieval PFC retrieval
46 47.5 (1.5 cm in SOL) 3.2 2 1020 in 7 shots with 1–2 pulse breaks between injections – – Directly after experiment 50 shots after last W injection
46 48 (2.0 cm in SOL) 3.5 1.93 1020 in 13 shots 3.46 1021 in 14 shots 1.75 1021 in 14 shots Directly after experiment Directly after experiment
Fig. 2. Operation plan of the experiment: timing of NBI heating and injection of marker gases.
Fig. 3. Normalized spectroscopy signals of WI and FII lines for the first and the last discharge with the WF6 injection.
sharp decrease. The graph in Fig. 4c reflects the deposition in the area nearest to the injection point in Exp. II, corner Tile 28 marked in Fig. 1a, when ex situ PFC studies were done directly after the WF6 injection. The profile is wave-shaped with a concentration maximum in the central part of the tile, i.e. in the area known as the erosion zone on ALT-II. The same shape of the tungsten distribution has been found on the other corner plate (Tile 15) from the adjacent ALT-II blade, see Fig. 1a. W content on these two corner tiles is greater than measured on other plates from various toroidal locations. This deposition pattern is a net result of global and local transport phenomena. It reflects the migration of tungsten from the net erosion area where it is originally deposited by plasma flux to areas where it rests: deposition–re-erosion–ionization–re-deposition
Fig. 4. Deposition profiles of tungsten on the ALT-II toroidal limiter tiles detected: (a) 50 shots after the WF6 injection, Exp. I; (b) and (c) after Exp. II in areas located at different positions with respect to the gas injector.
cycle. Erosion of tungsten occurs predominantly via physical sputtering. At the edge temperature Te of about 60 eV it is eroded mainly by plasma impurity ions (carbon, oxygen and boron), but the contribution of deuterium to the overall process is not excluded. The majority of sputtered species are neutrals with mean energy of 5–7 eV. At the edge density ne 5 1018 m 3 their mean free path is 2–3 mm (normal to the limiter surface) before they get ionized [8]. As predicted by Naujoks and Behrisch [55], species ionized close to the limiter surface may promptly be re-deposited within the first gyro orbit (Larmor radius is about 2.0 mm under
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174
the experimental conditions). The results of surface analysis explain gradual tungsten migration on ALT-II. They also clearly demonstrate the agreement with the model. The process leads to the ‘‘spill over’’ of tungsten on graphite until the equilibrium between the re-erosion and re-deposition is reached. The other point is the micro-distribution of tungsten especially in the plasma-wetted area. In a backscattered electron image reflecting the contrast between low- and high-Z species dark fields corresponds to carbon and brighter spots related to the accumulation of high-Z elements were recorded. Tungsten was also present as identified with EDX. Metal resides in small pits acting as local shadowed regions. It explains why some tungsten is found in the erosion zone after long operation periods. This is in agreement with previous studies on micro-distribution of fuel and plasma impurities on PFC [54,56,57]. 3.3. Tungsten balance The integrated quantity of tungsten found on the graphite plate near the gas inlet corresponds to 1% and 2–3% of the injected amount in Exp. I and II, respectively. This very small local deposition of W may be attributed to at least two factors: low sticking probability of the injected gas to the plate and/or efficient erosion of the locally deposited tungsten. The large Larmor radius, as discussed in the previous section, would eventually increase the migration of W from the vicinity of the injection point. The amounts deposited locally are small in both cases, but the difference between the two experiments is significant. It may be associated with the differences in the limiter position and plasma density, see Table 1. In Exp. I the limiter was 0.5 cm closer to plasma and the electron density was lower than in Exp. II, thus creating more favorable conditions for erosion. After each experiment tungsten on ALT-II tiles from several toroidal locations was analyzed. The integrated amount corresponded to a small fraction of the injected atoms: 6.5% in the first case and 10.5% in the second. Summing up the amount from the test and toroidal limiter only 8–13% of the species could be identified. Again, smaller amount was detected was after Exp. I when the ALT-II tiles were still in TEXTOR for 50 shots after the WF6 injection. This clearly motivates a search for another residence place of eroded tungsten. For the first time several DED tiles from various locations were studied. The W content ranges from 1 to 25 1015 cm 2. Taking 3 1015 cm 2 as a mean value, the integrated amount on DED is approximately 2 1020 W atoms, i.e. exceeding the amount puffed in Exp. II. It is impossible to assess accurately the amount originating from the last experiment because the tiles were in TEXTOR for 8 years. As determined by RBS, the tiles also contain other metals, e.g. Inconel components. The results lead to a conclusion that the inner bumper is a major residence place for high-Z metals eroded from the TEXTOR liner, various test limiters, and those introduced by injection. Also the deuterium content on the DED tiles is significant: 3–4 1018 cm 2 leading to the integrated amount in the range 2.0–2.5 1023 D atoms accumulated in that region. 4. Concluding remarks The reported tracer experiments, probably the most complex ever done in a tokamak, have resulted in several major contributions to material migration studies. The most important is the direct demonstration of tungsten erosion and its subsequent migration over long distances by re-erosion–ionization–re-deposition steps. To the authors’ knowledge, it is the first direct and so clear prove of the prompt re-deposition model predicted by Naujoks and Behrisch [55], though there was an earlier indication
S173
following tests of C & W twin limiter in TEXTOR [13]. The transport is gradual (see Fig. 4) but very effective under experimental conditions, i.e. edge Te 60 eV. One may assume that so significant W migration will be at least partly suppressed at low edge temperature and high divertor densities, as expected in ITER [58]. However, this would not eliminate metal erosion followed by prompt redeposition and local formation of modified mixed material layers. Low sticking probability of WF6 to the plate or efficient erosion at the test limiter causes that only a small fraction (1–3%) of the injected W amount is deposited near the gas inlet. Therefore, the concept of injection-based local repair of tungsten coatings could be conceived as a possible option only under low Te and high edge density regime. Low W content detected on the toroidal limiters (6–11%) inspired a search for a major place of its deposition. Surface studies performed for the first time on the DED tiles allowed for the identification of a major deposition area for highZ species eroded from the wall or introduced by injection. In summary, experimental results motivate comprehensive modeling to improve understanding of tungsten transport. Volatile WF6 can be used as a high-Z transport marker without a major risk of leaving large quantities of fluorine sticking (co-deposited) to PFC and then released to plasma during subsequent pulses. The statement is justified by two facts: (i) the memory effect is not pronounced, as proven by spectroscopy and (ii) small quantities of F (maximum 2 1016 cm 2 versus 1 1018 cm 2) are found even near the injection point. The application of the 15N tracer and a non-standard analysis method, i.e. HIERDA, have allowed for conclusive identification of nitrogen deposited on PCF. Though the exact material balance could not determined, the result shows effective co-deposition of toroidally puffed nitrogen. However, earlier experiments in TEXTOR [51] and recent in ASDEX [52] indicate the retention at the level of 30% of the injected amount. Also helium co-deposition and retention has been confirmed [51]. It is associated with He glow discharge. The level is quite significant (1 1017 cm 2). Therefore, it can be recommended to study helium trapping following He-based ion cyclotron wall conditioning [59], i.e. a method considered for a reactor-class machine. Acknowledgments This work, supported partly by the European Communities under the Contracts of Association between EURATOM–VR and EURATOM–FZJ, was carried out within the framework of the European Fusion Development Agreement. The views and opinions expressed herein do not necessarily reflect those of the European Commission. This work was partly funded by the Swedish Research Council (VR) through Contract No. 621-2009-4138. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
R.K. Fischer et al., Phys. Rev. Lett. 39 (1977) 622. J. Sugar, V. Kaufman, Phys. Rev. A 21 (1980) 2096. E. Hinnov et al., Nucl. Fusion 18 (1978) 1305. T. Tanabe, N. Noda, H. Nakamura, J. Nucl. Mater. 196–198 (1992) 11. V. Philipps et al., Nucl. Fusion 34 (1994) 1417. Y. Ueda et al., J. Nucl. Mater. 220–222 (1995) 240. T. Tanabe et al., J. Nucl. Mater. 212–215 (1994) 1370. M. Rubel et al., J. Nucl. Mater. 249 (1997) 116. V. Philipps et al., J. Nucl. Mater. 258–263 (1998) 858. T. Tanabe et al., J. Nucl. Mater. 283–287 (2000) 1128. J. Rapp et al., Nucl. Fusion 37 (1997) 1691. V. Philipps et al., Plasma Phys. Controlled Fusion 42 (2000) B293. M. Rubel et al., J. Nucl. Mater. 283–287 (2000) 1089. T. Hirai et al., Phys. Scr. T103 (2003) 59. T. Hirai et al., J. Nucl. Mater. 313–316 (2003) 67. M. Rubel et al., Phys. Scr. T81 (1999) 61. G. Sergienko et al., Phys. Scr. T128 (2007) 81. G. Sergienko et al., J. Nucl. Mater. 363–365 (2007) 96. Y. Ueda et al., J. Nucl. Mater. 415 (2011) S92.
S174 [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39]
M. Rubel et al. / Journal of Nuclear Materials 438 (2013) S170–S174 Y. Ueda et al., J. Nucl. Mater. 363–365 (2007) 66. J. Pospieszczyk et al., Nucl. Mater. 290–293 (2000) 947. M. Rubel et al., J. Nucl. Mater. 307–311 (2002) 111. E. Fortuna et al., Phys. Scr. T128 (2007) 162. M. Psoda et al., J. Nucl. Mater. 386–388 (2009) 740. H. Maier et al., J. Nucl. Mater. 258–263 (1998) 921. K. Krieger, H. Maier, R. Neu, J. Nucl. Mater. 266–269 (1999) 207. R. Neu et al., J. Nucl. Mater. 290–293 (2001) 206. K. Krieger, R. Neu, J. Nucl. Mater. 266–269 (1999) 207. K. Krieger et al., J. Nucl. Mater. 337–339 (2005) 10. R. Neu et al., J. Nucl. Mater. 363–365 (2007) 52. V. Rohde et al., J. Nucl. Mater. 415 (2011) S46. R. Neu et al., J. Nucl. Mater. 415 (2011) S322. G. Federici et al., Nucl. Fusion 41 (2001) 1967. G.F. Matthews et al., Phys. Scr. T128 (2007) 137. G.F. Matthews et al., Phys. Scr. T138 (2009) 014030. G.F. Matthews et al., Phys. Scr. T145 (2011) 014001. T. Hirai et al., Phys. Scr. T128 (2007) 144. R. Neu et al., Phys. Scr. T128 (2007) 150. C. Ruset et al., Phys. Scr. T128 (2007) 171.
[40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59]
H. Maier et al., Nucl. Fusion 47 (2007) 222. Ph. Mertens et al., Fusion Eng. Des. 84 (2009) 1289. Ph. Mertens et al., Fusion Eng. Des. 86 (2011) 1801. Ph. Mertens, Phys. Scr. T145 (2011) 014002. G.F. Matthews, This conference. M. Rubel et al., This conference. V. Philipps et al., Phys. Scr. T145 (2011) 0140. A. Kallenbach et al., J. Nucl. Mater. 415 (2011) S19. S. Brezinsek et al., Nucl. Fusion 51 (2011) 073007. R. Neu et al., Plasma Phys. Controlled Fusion 53 (2011) 124040. M. Rubel et al., J. Nucl. Mater. 415 (2011) S223. P. Petersson et al., Nucl. Instrum. Methods B273 (2012) 113. P. Petersson et al., This conference. M. Rubel et al., J. Nucl. Mater. 290–293 (2002) 473. M. Rubel et al., Fusion Eng. Des. 81 (2006) 211. D. Naujoks, R. Behrisch, J. Nucl. Mater. 220–222 (1995) 227. P. Petersson et al., Nucl. Instrum. Method B 268 (2010) 1833. P. Petersson et al., Phys. Scr. T145 (2011) 014014. K. Ikeda et al., Nucl. Fusion 47 (2007) (Special Issue ITER Physics Basis). A. Lyssoivan et al., J. Nucl. Mater. 415 (2011) S1029.