Total radiated line power coefficients for ions of medium to heavy weight elements, called PLT coefficients in the atomic data and analysis structure, have been improved by algorithmically optimising the selection of configuration sets that underpin the calculation to include the most important radiating transitions driven by both the ground and metastable configurations and to establish and limit the error of truncation. The optimised calculations typically differ from Ptterich by 20%-30% with truncation error ≲5%. Further appraisal of error due to atomic level bundling, atomic structure and collision strength calculation methods has been carried out. It is shown that bundling to configurations is accurate to ≲10% for all ions except those with closedshell ground configurations which give errors up to a factor 2-3. For near neutral, closed-shell ions, plane-wave Born collision strength calculations, which omit spin-change, give substantial error in comparison with distorted-wave calculations of PLT. For highly charged ions, spinsystem breakdown reduces the error in the PLT markedly, typically ≲10%. The error introduced by the atomic structure codes used here, autostructure and the Cowan code, is probably limited to ≲30%.
Optimisation and assessment of theoretical impurity line power coefficients relevant to ITER and DEMO
Giunta A.Writing – Review & Editing
;
2017-01-01
Abstract
Total radiated line power coefficients for ions of medium to heavy weight elements, called PLT coefficients in the atomic data and analysis structure, have been improved by algorithmically optimising the selection of configuration sets that underpin the calculation to include the most important radiating transitions driven by both the ground and metastable configurations and to establish and limit the error of truncation. The optimised calculations typically differ from Ptterich by 20%-30% with truncation error ≲5%. Further appraisal of error due to atomic level bundling, atomic structure and collision strength calculation methods has been carried out. It is shown that bundling to configurations is accurate to ≲10% for all ions except those with closedshell ground configurations which give errors up to a factor 2-3. For near neutral, closed-shell ions, plane-wave Born collision strength calculations, which omit spin-change, give substantial error in comparison with distorted-wave calculations of PLT. For highly charged ions, spinsystem breakdown reduces the error in the PLT markedly, typically ≲10%. The error introduced by the atomic structure codes used here, autostructure and the Cowan code, is probably limited to ≲30%.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
