Modeling of erosion and deposition by the Monte Carlo codes EDDY and ERO

A (CH4)-C-13 injection experiment with a roof-like test limiter in TEXTOR is modeled by EDDY and ERO for the purpose of code-code and code-experiment benchmarking. The modeled 2D patterns of CH emission light of both codes fairly agree with the experiment. The C-13 deposition patterns on the limiter surface are reproduced assuming negligible effective sticking (S = 0) for returning hydrocarbons. Assuming S = 1, both codes produce very peaked patterns in contrast to the experiment. The measured very small C-13 deposition efficiency (similar to 0.1%) is reproduced, assuming negligible sticking of returning hydrocarbons (S = 0) and high re-erosion yield of redeposited C-13 and C-12 atoms (impurity carbons in background plasma). At the early stage of (13)CH4 injection, the modeled C-13 deposition efficiency is high (e. g. 7% in the case of EDDY), whereas with an increase in exposure time, the efficiency strongly decreases due to re-erosion of C-13 concentrated near the surface. With a further increase in time, the efficiency reaches a steady-state value (similar to 0.1% for S = 0) accompanied by a depth profile of C-13 in C-12.