Plasma edge physics (plasmas with temperatures in the 1 to 100 eV range, near solid surfaces) has become a key issue in controlled nuclear fusion research. As for the physics of the fully ionised hot plasma core, appropriate dimensionless parameters have been identified: present fusion research acts like wind-channel experiments on downscaled models, with respect to future fusion reactors.
This is not longer possible for the plasma edge region due to dominant effects from atomic and surface processes. Integrated computational models comprising the physics of the plasma flow near boundaries, the atomic and molecular processes and the particle-surface interactions are the only tool to evaluate present experimental results (LHD, JT-60, Tore Supra, JET...) with respect to their relevance for future fusion power experiments (ITER) or a reactor. These models are limited, because at least one important ingredient, the (turbulent) plasma transport across the B-field, is not understood and this is likely to remain so in the future. The goal of present simulation code development is, therefore, to treat all the other components of the model accurately. In particular the atomic and molecular processes which, largely, control the plasma flow and plasma energy content in the important near target region. If the atomic and molecule physics issues can be resolved with sufficient accuracy, then the "anomalous" cross field plasma transport can be isolated as the only remaining unknown,and can be determined experimentally. In particular proton and electron collisions with the hydrogenic molecules H₂, D₂, T₂, DT, and their ions, play a key role in cooling and attenuating the magnetically confined plasma, before it hits exposed target surfaces. The surface released molecules travel in a bath of electrons and hydrogenic ions, with plasma temperatures (in the relevant region) in the 1 to 20 eV range, and plasma densities between 10¹⁹ to 10²¹ m^-3. Typical plasma gradient lengths are in the 1 to 10 cm range. Dissociation of such molecules into several atoms, which can later recombine again to molecules at surfaces, leads to a cascade process for the neutral gas penetration into the plasma. Significant effort has been (and still is being) spent to incorporate the atomic data for the most important impurities in fusion plasmas into the models. Such impurities can be either released by surface processes (hydrocarbons and their ions) or be introduced deliberately to cool the plasma in near target regions (Ar, Ne, N₂, ...). Sample calculations with the code package B2-EIRENE used by the international design team for the ITER prototypical fusion reactor (for the present "engineering design phase") will be discussed, in which not only neutral atom and molecule transport is simulated with Monte Carlo procedures, but also the very recent transport process of radiation transfer is included by the mathematically analogue photon gas simulation. Connection is made here to similar applications (same code) for lighting applications, e.g. for the study high intensity discharge (HID) lamps