Max-Planck-Institut für Plasmaphysik, EURATOM Association
1
Max-Planck-Institut für Plasmaphysik, EURATOM Association Erosion and Deposition of Tungsten as Plasma-facing Material Erosion and Deposition of Tungsten as Plasma-facing Material in ASDEX Upgrade in ASDEX Upgrade X. Gong 1 , K. Krieger 2 , J. Roth b , H. Maier 2 , V. Rohde 2 and ASDEX Upgrade Team 2 1 Institute of Plasma Physics, Academia Sinica, 230031 Hefei, Anhui, China 2 Max-Planck-Institut für Plasmaphysik, IPP-EURATOM Association, 85748 Garching, Germany Introduction In future fusion devices such as ITER, plasma-material interactions will have a considerably stronger influence on machine performance compared to present experiments, in particular due to the much longer discharge duration. Erosion and deposition are crucial issues because of their impact on plasma contamination, the lifetime of plasma facing components and tritium inventories. Carbon based materials are used for plasma-facing components in most of today’s experiments , but tritium codeposition with carbon will lead to a prohibitively high level of radioactive inventory in a fusion reactor. Tungsten is considered as an alternative material for plasma-facing components in ITER 1, due to its high sputtering threshold energy, low sputtering yield and small tritium retention, and has been successfully employed as plasma-facing material (PFM) in ASDEX Upgrade. Tungsten erosion and deposition were analysed by means of RBS and PIXE. The total tungsten erosion flux and the observed lateral variation across single tiles indicate that the dominant erosion process is sputtering by ion impact 3,4. The strong poloidal variation and the toroidal asymmetries of the tungsten erosion at the central column, and the different tungsten deposition on inner wall and divertor tiles, will be presented in this contribution. • Power dispersal, particle and impurities control, lifetime of the PFCs • Tritium retention and inventory removal • Minimisation and control of production of dust • Disruption damage and mitigation issues Be, C and W be considered as candidate PFM in ITER Good thermomechanical,High melting point, Low radiation Tritium inventory, High erosion with chemical erosion, RES Low radiation, No codeposition with Tritium Low melting point, High vapor pressure, High erosion High sputtering threshold energy, Low sputtering yield,Small tritium retention High self-sputtering yield, High radiation The tungsten coated graphite tiles were extended to the almost central column of ASDEX Upgrade in the present campaign 2002.Tungsten erosion and deposition were studied by ion beam analysis. Most of the W- coated tiles at the central column show both regions with net erosion and with local deposition. The maximum W-erosion rate was found to about 1.410 14 atom/cm 2 /s. The strong poloidal variation and the toroidal asymmetries of the tungsten erosion appears to correlate with the pattern of the scrape- off layer flux surfaces intersecting the surface of the inner column tiles. The maximum tungsten deposition was observed at the inner and outer divertor baffle modules, only small amounts of deposited tungsten were found at the divertor dome baffle. Comparison of the total W-erosion with the deposition rates in the divertor shows that only 10% of the eroded tungsten deposited in the divertor region. Summary Requirements for the next fusion device (ITER) Requirements for the next fusion device (ITER) C Be W Tungsten as PFM at the central column of ASDEX Upgrade Tungsten as PFM at the central column of ASDEX Upgrade 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 1999-2001 T14ab T14cd T oroidalC oordinate(m m ) 0 1 2 3 4 T10ab T10cd T11ab T11cd T12ab T12cd T13ab T13cd 0 1 2 3 4 T6cd T7ab T7cd T8ab T8cd T9ab T9cd W A realD ensity (10 17 /cm 2 ) 0 1 2 3 4 T2ab T2cd T3ab T3cd T4ab T4cd T5ab T5cd T6ab 0 1 2 3 4 T1d T1a T1b T1c 0 10 20 30 40 50 60 70 80 90 100 5.0 5.5 6.0 6.5 7.0 T11a T11b T11c T11d T o roid alC oo rdinate(m m ) 0 10 20 30 40 50 60 70 80 90 100 5.0 5.5 6.0 6.5 7.0 W A realD e n sity (10 18 /cm 2 ) T7a T7b T7c T7d 0 10 20 30 40 50 60 70 80 90 100 5.0 5.5 6.0 6.5 7.0 T3a T3b T3c T3d 0 200 400 600 800 1000 1200 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 W A realD ensity(10 17 /cm 2 ) B ottom Top T he leftofS eg13_2A-Seg13_6A R ight M iddle Left O rigina lW P oloidalC oordinate(m m ) During plasma operation with tungsten as PFM, no negative influences on the plasma behaviour were observed, the tungsten concentration in the core plasma remaining below the maximum tolerable limit for fusion reactor relevant discharge scenarios 5,6. W-coated tiles were extended to almost the full central column in 2002. A complete column of thinner W-coated tiles with 60nm thickness was mounted including the limiter regions in experimental campaign 2001. The investigation of the spatial distribution of the tungsten erosion and deposition. References References 1 G. Janeschitz, ITER JCT and HTs, J. Nucl. Mat. 290-293 (2001) 1-11 2 N. Peacock, et al., in: P. Stott, G. Gorini, E. Sindoni (Eds.), Diagnostics for experiment ThermonuclearFusion Reactors, Varenna, (Plenum, New York, 1996) p.291 3 K. Krieger, et al., ICFRM-10, 2001, submitted to J. Nucl. Mat. 4 A.Tabasso, H. Maier, K. Krieger, J. Roth, J. Nucl. Mat. 290-293 (2001) 326-330 5 R. Neu, et al., PPCF, 44 (2002) 811 6 V. Rohde, et al., 28 th EPS, P1-042, 2001 7 K. Krieger, et al.,, PSI-15, O-14, 2002, submitted to J. Nucl. Mat. [8] M.Mayer, Max-Planck-Institut für Plasmaphysik Report IPP 9/113 (1997) 9 N.P. Barradas, C.Jeynes, R.P.Webb, Appl. Phys. Lett. 71(1997) 291 10 A.Geier, et al., this conference Ion Beam Analysis: RBS Ion Beam Analysis: RBS and PIXE and PIXE using a 2.0MeV using a 2.0MeV 4 H H e ion beam and a e ion beam and a 1.5MeV proton beam 1.5MeV proton beam Ion Beam Analysis for W-coated tiles Ion Beam Analysis for W-coated tiles Schematic of the W-coated tile showing the analysis points along poloidal and toroidal directions. The spatial distribution of tungsten deposited on DivII and DivIIb The spatial distribution of tungsten deposited on DivII and DivIIb The distribution of tungsten erosion along toroidal direction The distribution of tungsten erosion along toroidal direction The poloidal distribution of the tungsten erosion The poloidal distribution of the tungsten erosion Date evaluated by SIMNRA5.0 and WiNDF7.0 Date evaluated by SIMNRA5.0 and WiNDF7.0 0 1 2 3 4 5 6 7 8 9 10 11 0 100 200 300 400 500 600 Simnra1 W iNDF1 W iNDF2 Simnra2 E rror(% ) W A realD e nsity (1 0 15 /cm 2 ) Spot Due to W-layer and substrate roughness in W-coated tiles, and the modification of W-coating erosion by impurity ions during plasma exposure. The error of the data evaluated for tungsten RBS spectrum using SIMNRA5.0 and WiNDF7.0 with and without roughness is about 10% 8,9. The maximum W-erosion is found at the left side of tiles from Seg13_2A to Seg13_6A located in the upper regions, about 2.010 17 atoms/cm 2 . The amounts of erosion >>expected values from CX sputtering modelling. The dominant erosion process is sputtering by ion impact, not only due to bombardment by charge- exchange neutrals. The maximum erosion rate is about 1.4x10 14 atom/cm 2 /s for the experiment 2001, which is quite similar to the experiment 1999/2000. The schematic view of the tiles at central column with the local pitch angel of magnetic field lines at the tile surface. The length of the respective yellow line segments corresponds to the fraction sin () of the parallel ion flux reaching the surface, where is the field line angel of incidence against the tile surface. The poloidal variation and the strong toroidal asymmetries of the tungsten erosion. Correlate with the pattern of the scrape-off layer flux surfaces intersecting the surface of the W-coated tiles 7. Three thick W-coated tiles with 1.0µm in segment13_B were also measured after the experiment 2001, direct neighbours of the thinner W-coated tiles . The corresponding maximum W-erosion is found at the right of W-coated tiles locating at lower regions. The similar spatial distribution of deposited tungsten in both divertor configurations. A comparison of the total W-erosion with the deposition rates on the divertor shows that only 10% of the eroded tungsten deposited on the divertor, One can infer that most of the eroded tungsten migrates via direct transport channels in the outer scrape-off layer regions without penetrating the confined plasma. This is a good agreement with tungsten transport modelling by the DIVIMP code 10. Maximum W-deposition at inner divertor baffle module (nearby heat shield tiles), Similarly increased W-deposition at outer divertor,only small at the dome baffle. During the present experiment, the W deposition is about two times larger than before, due to the additional source of tungsten from the midplane area, which is generally closest to the separatrix and used as ramp-down limiter. ummary of three experimental campaigns with tungsten as PMF in the ASDEX Upgrade ummary of three experimental campaigns with tungsten as PMF in the ASDEX Upgrade Experim ental cam paign W -coated Area (percentoftotal) W -coating thickness Exposure tim e (D ischarges) Position at the centralcolum n Divertor Configuration Nov.1999--- Jun. 2000 1.2 m 2 (13% oftotal) 0.2-0.5 m 4600s (919 discharges) The tw o low est Toroidalrow s D ivII A pr. 2001--- Jul.2001 5.5 m 2 (68% oftotal) 1.0 m 1780s (374 discharges) Fullinnerw allexceptram p- dow n lim iterand areas facing N BI D ivIIb 2002 7.1 m 2 (87% oftotal) 1.0 m 3690s* (623 discharges) Fullinnerw allexcept areasfacing N BI D ivIIb *Stillongoing Red dots are RBS analyses of the W-coated tiles before plasma exposure. Average thickness of the original tungsten layer: Thin tiles: 3.510 17 atoms/cm 2 Thick tiles: 6.4 10 18 atoms/cm 2 Analysis chamber with mainpulator for large W-coated tiles before and after experiment.
description
Max-Planck-Institut für Plasmaphysik, EURATOM Association Erosion and Deposition of Tungsten as Plasma-facing Material in ASDEX Upgrade. X. Gong 1 , K. Krieger 2 , J. Roth b , H. Maier 2 , V. Rohde 2 and ASDEX Upgrade Team 2 - PowerPoint PPT Presentation
Transcript of Max-Planck-Institut für Plasmaphysik, EURATOM Association
Max-Planck-Institut für Plasmaphysik, EURATOM Association
Erosion and Deposition of Tungsten as Plasma-facing Material in ASDEX UpgradeErosion and Deposition of Tungsten as Plasma-facing Material in ASDEX Upgrade X. Gong1, K. Krieger2, J. Rothb, H. Maier2, V. Rohde2 and ASDEX Upgrade Team2
1Institute of Plasma Physics, Academia Sinica, 230031 Hefei, Anhui, China 2Max-Planck-Institut für Plasmaphysik, IPP-EURATOM Association, 85748 Garching, Germany
IntroductionIn future fusion devices such as ITER, plasma-material interactions will have a considerably stronger influence on machine performance compared to present experiments, in particular due to the much longer discharge duration. Erosion and deposition are crucial issues because of their impact on plasma contamination, the lifetime of plasma facing components and tritium inventories. Carbon based materials are used for plasma-facing components in most of today’s experiments , but tritium codeposition with carbon will lead to a prohibitively high level of radioactive inventory in a fusion reactor.Tungsten is considered as an alternative material for plasma-facing components in ITER 1, due to its high sputtering threshold energy, low sputtering yield and small tritium retention, and has been successfully employed as plasma-facing material (PFM) in ASDEX Upgrade. Tungsten erosion and deposition were analysed by means of RBS and PIXE. The total tungsten erosion flux and the observed lateral variation across single tiles indicate that the dominant erosion process is sputtering by ion impact 3,4. The strong poloidal variation and the toroidal asymmetries of the tungsten erosion at the central column, and the different tungsten deposition on inner wall and divertor tiles, will be presented in this contribution.
• Power dispersal, particle and impurities control, lifetime of the PFCs• Tritium retention and inventory removal• Minimisation and control of production of dust• Disruption damage and mitigation issues
Be, C and W be considered as candidate PFM in ITERGood thermomechanical,High melting point, Low radiationTritium inventory, High erosion with chemical erosion, RESLow radiation, No codeposition with TritiumLow melting point, High vapor pressure, High erosionHigh sputtering threshold energy, Low sputtering yield,Small tritium retentionHigh self-sputtering yield, High radiation
The tungsten coated graphite tiles were extended to the almost central column of ASDEX Upgrade in the present campaign 2002.Tungsten erosion and deposition were studied by ion beam analysis. Most of the W-coated tiles at the central column show both regions with net erosion and with local deposition. The maximum W-erosion rate was found to about 1.41014atom/cm2/s. The strong poloidal variation and the toroidal asymmetries of the tungsten erosion appears to correlate with the pattern of the scrape-off layer flux surfaces intersecting the surface of the inner column tiles. The maximum tungsten deposition was observed at the inner and outer divertor baffle modules, only small amounts of deposited tungsten were found at the divertor dome baffle. Comparison of the total W-erosion with the deposition rates in the divertor shows that only 10% of the eroded tungsten deposited in the divertor region.
Summary
Requirements for the next fusion device (ITER)Requirements for the next fusion device (ITER)
C
Be
W
Tungsten as PFM at the central column of ASDEX UpgradeTungsten as PFM at the central column of ASDEX Upgrade
0 10 20 30 40 50 60 70 80 90 1000
5
10
15
1999-2001
T14ab T14cd
Toroidal Coordinate(mm)
01234
T10ab T10cd T11ab T11cd T12ab T12cd T13ab T13cd
01234
T6cd T7ab T7cd T8ab T8cd T9ab T9cd
W A
real
Den
sity
(10
17/c
m2 )
01234
T2ab T2cd T3ab T3cd T4ab T4cd T5ab T5cd T6ab
01234
T1d T1a T1b T1c
0 10 20 30 40 50 60 70 80 90 1005.0
5.5
6.0
6.5
7.0
T11a T11b T11c T11d
Toroidal Coordinate(mm)
0 10 20 30 40 50 60 70 80 90 1005.0
5.5
6.0
6.5
7.0
W A
rea
l Den
sity
(10
18/c
m2 )
T7a T7b T7c T7d
0 10 20 30 40 50 60 70 80 90 1005.0
5.5
6.0
6.5
7.0
T3a T3b T3c T3d
0 200 400 600 800 1000 12000.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
W
Are
al D
ensi
ty(1
017/c
m2 )
BottomTop
The left of Seg13_2A-Seg13_6A
Right Middle Left Original W
Poloidal Coordinate(mm)
During plasma operation with tungsten as PFM, no negative influences on the plasma behaviour were observed, the tungsten concentration in the core plasma remaining below the maximum tolerable limit for fusion reactor relevant discharge scenarios 5,6.
W-coated tiles were extended to almost the full central column in 2002. A complete column of thinner W-coated tiles with 60nm thickness was mounted including the limiter regions in experimental campaign 2001. The investigation of the spatial distribution of the tungsten erosion and deposition.
ReferencesReferences1 G. Janeschitz, ITER JCT and HTs, J. Nucl. Mat. 290-293 (2001) 1-112 N. Peacock, et al., in: P. Stott, G. Gorini, E. Sindoni (Eds.), Diagnostics for experiment ThermonuclearFusion Reactors, Varenna, (Plenum, New York, 1996) p.2913 K. Krieger, et al., ICFRM-10, 2001, submitted to J. Nucl. Mat.4 A.Tabasso, H. Maier, K. Krieger, J. Roth, J. Nucl. Mat. 290-293 (2001) 326-3305 R. Neu, et al., PPCF, 44 (2002) 811 6 V. Rohde, et al., 28th EPS, P1-042, 20017 K. Krieger, et al.,, PSI-15, O-14, 2002, submitted to J. Nucl. Mat.[8] M.Mayer, Max-Planck-Institut für Plasmaphysik Report IPP 9/113 (1997) 9 N.P. Barradas, C.Jeynes, R.P.Webb, Appl. Phys. Lett. 71(1997) 29110 A.Geier, et al., this conference
Ion Beam Analysis: RBS and Ion Beam Analysis: RBS and PIXE PIXE using a 2.0MeV using a 2.0MeV 44HHe ion e ion beam and a 1.5MeV proton beam and a 1.5MeV proton beambeam
Ion Beam Analysis for W-coated tilesIon Beam Analysis for W-coated tiles
Schematic of the W-coated tile showing the analysis points along poloidal and toroidal directions.
The spatial distribution of tungsten deposited on DivII and DivIIbThe spatial distribution of tungsten deposited on DivII and DivIIb
The distribution of tungsten erosion along toroidal directionThe distribution of tungsten erosion along toroidal direction
The poloidal distribution of the tungsten erosionThe poloidal distribution of the tungsten erosion Date evaluated by SIMNRA5.0 and WiNDF7.0Date evaluated by SIMNRA5.0 and WiNDF7.0
0 1 2 3 4 5 6 7 8 9 10 110
100
200
300
400
500
600
Simnra1 WiNDF1 WiNDF2 Simnra2 Error(%)W
Are
al D
ensi
ty (
101
5 /cm
2 )
Spot
Due to W-layer and substrate roughness in W-coated tiles, and the modification of W-coating erosion by impurity ions during plasma exposure.The error of the data evaluated for tungsten RBS spectrum using SIMNRA5.0 and WiNDF7.0 with and without roughness is about 10% 8,9.
The maximum W-erosion is found at the left side of tiles from Seg13_2A to Seg13_6A located in the upper regions, about 2.01017atoms/cm2.The amounts of erosion >>expected values from CX sputtering modelling. The dominant erosion process is sputtering by ion impact, not only due to bombardment by charge-exchange neutrals. The maximum erosion rate is about 1.4x1014atom/cm2/s for the experiment 2001, which is quite similar to the experiment 1999/2000.
The schematic view of the tiles at central column with the local pitch angel of magnetic field lines at the tile surface. The length of the respective yellow line segments corresponds to the fraction sin () of the parallel ion flux reaching the surface, where is the field line angel of incidence against the tile surface.The poloidal variation and the strong toroidal asymmetries of the tungsten erosion. Correlate with the pattern of the scrape-off layer flux surfaces intersecting the surface of the W-coated tiles 7.Three thick W-coated tiles with 1.0µm in segment13_B were also measured after the experiment 2001, direct neighbours of the thinner W-coated tiles .The corresponding maximum W-erosion is found at the right of W-coated tiles locating at lower regions.
The similar spatial distribution of deposited tungsten in both divertor configurations. A comparison of the total W-erosion with the deposition rates on the divertor shows that only 10% of the eroded tungsten deposited on the divertor, One can infer that most of the eroded tungsten migrates via direct transport channels in the outer scrape-off layer regions without penetrating the confined plasma. This is a good agreement with tungsten transport modelling by
the DIVIMP code 10.
Maximum W-deposition at inner divertor baffle module (nearby heat shield tiles),Similarly increased W-deposition at outer divertor,only small at the dome baffle.
During the present experiment, the W deposition is about two times larger than before, due to the additional source of tungsten from the midplane area, which is generally closest to the separatrix and used as ramp-down limiter.
Summary of three experimental campaigns with tungsten as PMF in the ASDEX UpgradeSummary of three experimental campaigns with tungsten as PMF in the ASDEX Upgrade
Experimentalcampaign
W-coated Area(percent of total)
W-coatingthickness
Exposure time(Discharges)
Positionat the central column
DivertorConfiguration
Nov.1999---Jun. 2000
1.2 m2
(13% of total)0.2-0.5 m 4600s
(919 discharges)The two lowest Toroidal rows
DivII
Apr. 2001--- Jul.2001
5.5 m2
(68% of total)1.0 m 1780s
(374 discharges)Full inner wall except ramp-
down limiter and areasfacing NBI
DivIIb
2002 7.1 m2
(87% of total)1.0 m 3690s*
(623 discharges)Full inner wall except
areas facing NBIDivIIb
*Still ongoing
Red dots are RBS analyses of the W-coated tiles before plasma exposure. Average thickness of the original tungsten layer:Thin tiles: 3.51017atoms/cm2 Thick tiles: 6.4 1018atoms/cm2
Analysis chamber with mainpulator for large W-coated tiles before and after experiment.
