Velocity distribution of carbon atoms released from a tokamak limiter by chemical and physical sputtering

Abstract

The Doppler broadened profile of a C I line in the infrared (λ =909.5 nm, 3P2 to 3P20) has been measured with high spectral resolution (λ/Δλ > 105) in front of a graphite limiter in the TEXTOR tokamak. The line shows a strong Zeeman effect, but by selecting the pi component with a polarizer, the influence of the magnetic field of TEXTOR on the line shape can be practically eliminated, and the profile is determined only by the velocity distribution of the carbon atoms. By a comparison of shots with a 'detached plasma' and shots with injected gas (CO, CH4) it could be shown that the carbon flux from the limiter is mainly determined by chemical sputtering in a 'detached plasma'. In a plasma attached to the limiter, the energy of the released carbon atoms increases with the electron temperature (Te) at the limiter, the typical width of the line gradually increases from about 16 pm to 50 pm (corresponding to velocities of 5 × 105 to 1.5 × 106 cm/s or energies of 1.5 to 13.5 eV), indicating that physical sputtering dominates at high Te and that chemical sputtering then gives only a small contribution (<15%) to the impurity production at the limiter

The role of plasma–wall interactions in thermal instabilities at the tokamak edge

“…This is especially true at the low plasma temperatures relevant to MARFE formation, where the release of carbon impurity is dominated by chemical sputtering. 17,18 Also, the dependence of hydrogen absorption and recycling properties on the wall coating material 19 can be of significant importance.…”

Determination of carbon release mechanisms in the DIII-D divertors from analysis of C I line profiles

“…6 When physical sputtering seems dominant, effective temperatures can reach 5 eV, a value consistent with the energy spectrum of particles ejected directly by deuteron impact at energies of a few hundred eV. The first term on the right-hand side represents the influx of atomic carbon originating from heavy hydrocarbons and the second from the influx of methane.…”

Measurements of the average energy of carbon atoms released from breakup of methane in the main SOL of DIII-D compared with DIVIMP code modeling

“…It is important to establish if this is the case in order to provide confidence in the interpretation of the naturally emitted CI line wavelength profile, resulting from plasma impact on graphite surfaces, as a means of identifying the relative role of physical and chemical sputtering. In other studies [6][7][8], methane has been injected into tokamak plasmas through divertor targets or small openings in limiters. Although the latter studies also reported narrow CI profiles, indicative of rather cold C-atoms, the interpretation of such experiments is more difficult and subject to greater uncertainties.…”

“…In addition, EL do not include four additional dissociative excitation channels for CH þ 4 that JR do, the ones leading to CH þ 2 , CH 2 , CH + , and C + . Since the Frank-Condon energy is distributed between the products of each breakup, in inverse proportion to the mass of the fragments, it had been conjectured [6] that the CI line should be about twice as broad for 12 CD 4 as for 12 CH 4 . As indicated by Table 1, however, there is little difference in the experimental CI profiles resulting from 12 CH 4 and 12 CD 4 puffing.…”