Multipactor-induced induced neutral pressure limits on Alcator C-Mod ICRF Performance T. P. Graves, B. LaBombard, S. J. Wukitch, I. H. Hutchinson MIT Plasma Science and Fusion Center American Physical Society 47th Annual Meeting of the Division of Plasma Physics Denver, CO Oct. 24-28, 2005
Performance limitations on C-Mod High power RF necessary for auxiliary heating on fusion experiments Problems associated with high power RF voltage breakdown, impurity and density production, neutral pressure limit ICRF neutral pressure operating limit, J port ~ 0.5 mtorr, E port ~ 1 mtorr No RF restart possible at pressure above the observed limit
Performance limitations on C-Mod High power RF necessary for auxiliary heating on fusion experiments Problems associated with high power RF voltage breakdown, impurity and density production, neutral pressure limit ICRF neutral pressure operating limit, J port ~ 0.5 mtorr, E port ~ 1 mtorr No RF restart possible at pressure above the observed limit Possible cause multipactor discharge
Multipactor on C-Mod and CMX Alcator C-Mod uses ICRF power 40-80 MHz for auxiliary heating Each antenna has short sections of vacuum transmission line susceptible to multipactor discharges Multipactor experiments done on E, J antennas in same configuration Coaxial Multipactor Experiment* Dedicated tabletop experiment to specifically study multipactor discharges in various geometries and conditions *T. Graves et. al. Rev. Sci. Inst. Submitted Oct. 2005
Multipactor Properties T(0 to π) T(π to 2 π) T(2π to 3π) E 1 e - Secondary Electrons E 2 E 3 E field wave A multipactor discharge is a resonant condition for electrons in an alternating E field Vacuum conditions required Electron multiplication from secondary electrons δ(e) > 1 Voltage range for multipactor susceptibility Minimum voltage onset Voltage pushthru due to electron defocusing
Multipactor Properties T(0 to π) T(π to 2 π) T(2π to 3π) E 1 e - Secondary Electrons E 2 E 3 E field wave A multipactor discharge is a resonant condition for electrons in an alternating E field Vacuum conditions required Electron multiplication from secondary electrons δ(e) > 1 Voltage range for multipactor susceptibility Minimum voltage onset Voltage pushthru due to electron defocusing time (sec) Multipactor seen in reflected power and current measurements Multipactor detunes RF circuit by adding reactive component to network Pushthru requires high reflected power tolerance
Multipactor as a function of pressure At 1-2 mtorr, multipactor-induced glow discharge* BELOW PASCHEN! Glow discharge indicated by total drop in circulating power Once glow discharge established, higher voltage cannot be achieved If no multipactor present, breakdown follows Paschen curve Geometry, frequency, voltage, material δ(e) < 1 *F. Hohn et. al. Phys. Plasmas 4 (4), April 1997
Multipactor as a function of pressure At 1-2 mtorr, multipactor-induced glow discharge* BELOW PASCHEN! Glow discharge indicated by total drop in circulating power Once glow discharge established, higher voltage cannot be achieved If no multipactor present, breakdown follows Paschen curve Geometry, frequency, voltage, material δ(e) < 1 Voltage vs. multipactor rise time** If τ v > τ m, multipactor will occur For raised pressure, slow rise time leads to glow *F. Hohn et. al. Phys. Plasmas 4 (4), April 1997 **R. Kishek et. al. Phys. Plasmas 4 (3), March 1997
C-Mod ICRF Multipactor Experiment Low RF Power (500W) scan as a function of background deuterium pressure 1e-4 to 1 mtorr Diagnostics Forward/reflected power at source Forward/reflected power in circulating loop Circulating loop voltage probes Magnetic Field ~ 0.1 T
C-Mod Antenna Multipactor Susceptibility Voltage handling decreases with increasing pressure Total loss of power at glow onset Magnetized J-port multipactor-induced glow discharge at 0.5 mtorr
C-Mod Antenna Multipactor Susceptibility Voltage handling decreases with increasing pressure Total loss of power at glow onset Magnetized J-port multipactor-induced glow discharge at 0.5 mtorr Magnetized E-port multipactor-induced glow discharge at 1 mtorr Same as observed neutral pressure limit!
Multipactor Susceptible Locations Vacuum feedthru (2.8 cm gap), coaxial section possible location of multipactor Previous campaign, this was 24 long Back of antenna box, center conductor thru rectangular cutout (1.6cm gap) Strip line should have magnetic insulation (E perp B) *E Vacuum Coax larger than J
CMX results support C-Mod observations Pressure experiments If multipactor is present, multipactor-induced glow discharge at lowered pressure If multipactor can be prevented via geometry, frequency, or δ < 1 then Paschen breakdown is observed Voltage ramp time experiments If RF voltage is turned on at high power and ramp time is determined by Q factor, often (but not always) Paschen breakdown is observed If RF voltage is modulated by positive ramp (τ v = 1 5 ms), multipactorinduced glow discharge at lowered pressure Electrode material experiments Copper oxide has much larger δ(e) than pure Cu Pure titanium has δ < 1 for all energy, but titanium oxide does not! Titanium Nitride (TiN) can have δ < 1, but exposure to air can raise value, ruining TiN suppression capabilities *, ** *S. Castaneda et al. J. Vac. Sci. Tech. 21 (6) 2003 **V. Baglin et al. 7 th Proc. of EPAC 2000, Vienna, Austria
Summary Multipactor-induced glow discharges cause observed neutral pressure limits on antennas BELOW Paschen limit; no pushthru possible C-Mod multipactor experimental results 0.5 mtorr limit on J; 1.0 mtorr limit on E both consistent with observed antenna operation Data from CMX supports multipactor-induced glow discharge Possible multipactor location in vacuum coax region. In order to avoid multipactor-induced glow discharge below Paschen limit, need material δ < 1, or change operating regime