ITER Halo Currents Question for US PFC Community

ITER Halo Currents Question for US PFC Community M. Ulrickson For the ITER Team August 6, 2010 Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy’s National Nuclear Security Administration under contract DE-AC 04 -94 AL 85000. 1

Halo Current Generation • During a disruption the plasma moves toward the first wall (divertor also sometimes). • Flux surfaces carrying current are broken by the contact with the wall. • The current flows through the first wall and shield block to the vessel to another shield block and first wall to reconnect with the flux surface. • The halo current scrape-off length is found to be wider than the power scrape-off length on existing devices (even wider than the far scrape-off layer power width. • The DINA code is used to simulate halo currents in ITER. • The portion of the first wall receiving halo current depends on the motion of the plasma. 2

Predicted Halo Current on ITER • The halo current data from DINA has been integrated over each blanket module as a function of time. • Since DINA is a 2 D code the values are representative of a whole toroidal row of modules. • The halo current to a single module is found by dividing by the number of modules in a row (18 or 36), multiplying by the toroidal peaking factor (2) and an uncertainty factor (1. 3). • The variation of halo current flow over a FW panel is determined by mapping flux surfaces onto the shaped FW and using the observed halo scrape-off length. • The average wetted area is about ¼ of the FW surface area. 3

VDE_Up_Slow_Fast_II 2. 50 E+06 2. 00 E+06 1. 50 E+06 FW 9 cw FW 10 cw 1. 00 E+06 FW 11 cw 5. 00 E+05 0. 00 E+00 600 700 800 900 1000 1100 1200 2. 50 E+06 2. 00 E+06 FW 5 ccw FW 6 ccw 1. 50 E+06 FW 7 ccw FW 8 ccw 1. 00 E+06 FW 9 ccw FW 10 ccw 5. 00 E+05 FW 11 ccw 0. 00 E+00 600 700 800 900 1000 4 1100 1200

VDE_Up_Slow_II 3. 50 E+06 3. 00 E+06 2. 50 E+06 FW 6 cw 2. 00 E+06 FW 7 cw FW 9 cw 1. 50 E+06 FW 10 cw 1. 00 E+06 FW 11 cw 5. 00 E+05 0. 00 E+00 500 2. 50 E+06 700 900 1100 1300 1500 1700 1900 2. 00 E+06 FW 4 ccw 1. 50 E+06 FW 5 ccw FW 6 ccw 1. 00 E+06 FW 7 ccw 5. 00 E+05 0. 00 E+00 500 700 900 1100 1300 5 1500 1700 1900

VDE Down_Slow_Fast_II & Slow_II 5. 00 E+06 4. 50 E+06 4. 00 E+06 3. 50 E+06 3. 00 E+06 2. 50 E+06 FW 1 cw 2. 00 E+06 FW 18 ccw 1. 50 E+06 1. 00 E+06 5. 00 E+05 0. 00 E+00 650 700 750 800 4. 50 E+06 4. 00 E+06 3. 50 E+06 3. 00 E+06 FW 1 cw 2. 50 E+06 FW 2 cw 2. 00 E+06 FW 3 cw 1. 50 E+06 FW 18 ccw 1. 00 E+06 5. 00 E+05 0. 00 E+00 0 100 200 300 400 500 6 700 800 900 1000

VDE_Up_Slow 3500000 3000000 2500000 2000000 FW 9 cw FW 10 cw 1500000 FW 11 cw 1000000 500000 0 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 3500000 3000000 2500000 FW 5 ccw 2000000 FW 6 ccw FW 7 ccw 1500000 FW 8 ccw 1000000 FW 9 ccw 500000 0 700 800 900 1000 1100 1200 1300 7 1400 1500 1600 1700

MD_up_Lin 1. 00 E+06 9. 00 E+05 8. 00 E+05 7. 00 E+05 6. 00 E+05 FW 8 cw 5. 00 E+05 FW 9 cw 4. 00 E+05 FW 10 cw 3. 00 E+05 2. 00 E+05 1. 00 E+05 0. 00 E+00 0 5 10 15 20 25 30 35 40 45 50 800000 700000 600000 500000 FW 5 ccw 400000 FW 6 ccw FW 7 ccw 300000 FW 8 ccw 200000 100000 0 0 5 10 15 20 25 30 8 35 40 45 50

Load Contribution • Halo Currents – Flow radially and poloidally to reach the vessel attachments – Toroidal halo current flow generates some loads but they are small compared to the radial and poloidal flow interaction with the toroidal field. – Radial flow generates net poloidal force and toroidal moment (because of the thickness of the SB). – Poloidal flow generates radial force and poloidal moments because of the distribution of halo current on a FW panel. – The halo forces are directly proportional to the halo current. 9

VDE up Linear III 2. 5 E+05 1. 2 E+06 1. 0 E+06 2. 0 E+05 1. 5 E+05 6. 0 E+05 1. 0 E+05 4. 0 E+05 5. 0 E+04 2. 0 E+05 0. 0 E+00 0. 81 0. 82 0. 83 0. 84 0. 85 Time (s) 10 0. 86 0. 87 0. 88 0. 89 0. 9 Current (A) Moment (Nm) 8. 0 E+05 Fpol (N) ty tz Halo

VDE up Exponential III 2. 5 E+05 1. 2 E+06 1. 0 E+06 2. 0 E+05 1. 5 E+05 6. 0 E+05 1. 0 E+05 4. 0 E+05 5. 0 E+04 2. 0 E+05 0. 0 E+00 0. 81 0. 82 0. 83 0. 84 0. 85 Time (s) 11 0. 86 0. 87 0. 88 0. 89 0. 9 Current (A) Moment (Nm) 8. 0 E+05 Fpol (N) ty tz Halo

Halo Current Density Estimate • Given the maximum halo current to a row (4. 5 MA) and 18 modules per row and the peaking factor shown above. • The maximum estimated halo current on a module is 400 k. A • The wetted area is about 0. 25 m 2 • Hence the current density to the FW is 1. 6 MA/m 2 • The lower end of the scale is 400 k. A/m 2 12

Question for the PFC Community • Given the halo current density on the FW, if the plasma has even a modestly low temperature (1 e. V), the power flow to the FW can be quite substantial (MW/m 2). • Since the duration of the halo event can be 10’s of msec to seconds, the heating of the FW can be quite substantial. • Given the long scrape-off length for halo current, components that are shadowed by being recessed from the FW horizon may also see substantial power due to halo current. • When I asked people on the ITER STAC call to find data on the power due to halo current, no data was cited but there was a general feeling no significant temperature rise on the FW could be attributed to halo flow. • What would be the prediction of the power flow due to halo currents using our edge codes? 13

Discussion • Strategy for answering the question – Step 1 – Step 2 14