Determination of rate coefficients for atoms and molecules (hydrocarbons and hydrogen with its isotopes) by measurement and modelling in the plasma boundary of TEXTOR

The modelling of experimental emission line profiles from neutral helium was again performed by taking into account the influence of proton collisions on the population of the excited states. It was found that these collisions can lead to enhanced electron density values up to a factor of 2 when fitting the observed emission profiles with the new ones with input from the code ATOM-IMP. The reason for this is the increased ionisation probability of the excited levels by the protons. The values for the electron temperatures remained unchanged. A similar campaign for boron as for oxygen was planned and started including a preliminary modelling and a choice of the suitable spectral lines. Preliminary experimental results (e.g. ratios of adjacent B II lines) and a comparison of modelled S/XB data for 10 lines with those from ADAS are provided. Modelling of the S/XB-value for the 400.9nm line of Tungsten I has been performed. The energies for states mentioned were calculated The results were compared with data from Moore tables. In the calculations 4 even and 2 odd configurations were included. The oscillator strengths were calculated by Cowan codes (5 even and 3 odd configurations) and Grasp codes (4 even and 2 odd configurations). There is strong influence of the configuration mixing and deviation from SL-coupling. Cowan and Grasp gf values differ up to a factor of 2. The Cowan results are in better agreement with the experiment than Grasp. The largest gf values are in the range 290nm, the next are in the range 400nm. The excitation cross sections and rate coefficients were calculated by the Normalized Born method. From the data for oscillator strengths one can see that a direct calculation by ATOM is not appropriate. Therefore, a semiempirical approach based on the assumption that the cross section is proportional to the oscillator strength was used. The ionization cross sections and rate coefficients were calculated for the ground level and metastable state, and contributions from 5d and 6s shells were included. The cross section for the metastable states is about 30% greater than for the ground state. σ (5d) is 3 times the one for σ (6s) of about 3 times. The agreement with experimental values from IPP-Berlin and IPP-Garching was surprisingly good. For the evaluation of experiments carried out on JET in order to determine the molecular contribution to the total hydrogen flux via Fulcher-α- band spectroscopy the collisional radiative model (CRMOL) by Thornton Greenland was used. On TEXTOR the conversion factor D/XB taken from CRMOL and relating the number of molecular losses per photon was verified in gas injection experiments. The latest version of the code contains also vibrationally resolved D/XB values and provides reasonable molecular atom to molecule ratios for JET conditions. Spectroscopy of the CD radical is the standard tool for the determination of CD4 particle fluxes. But the conversion between photon and particle flux measurements by means of D/XB values is indirect and depends on the dissociation chain. Several attempts have recently been performed in order to both measure and model the CD-production and its emission on TEXTOR, JET and ASDEX by gas blow experiments into the boundary plasma and divertor surrounding. However, it was found that consistency of the used/measured D/XB values for CH/D from CH4/D4 is still not satisfactory. Surface and geometry effects seem to dominate the values and, therefore, a better modelling by an ERO-code is urgently needed, which includes all rates for normal and higher order hydrocarbons as far as possible. However, the measured D/XB values for C2 look consistent with laboratory experiments