Scaling of the divertor power spreading (S-factor) in open and closed divertor operation in JET and ASDEX Upgrade

An experimental database of the power spreading into the target private flux region (S-factor) from ASDEX Upgrade and JET outer target data has been gathered and analysed. The power spreading, combined with the Scrape-Off Layer heat flux decay length λq, determines the power exhaust channel width at...

Full description

Saved in:
Bibliographic Details
Published in:Journal of nuclear materials 2015-08, Vol.463, p.49-54
Main Authors: Scarabosio, A., Eich, T., Hoppe, F., Paradela, I., Sieglin, B., Reimold, F., Rack, M., Groth, M., Wischmeier, M., Arnoux, G., Balboa, I., Marsen, S.
Format: Article
Language:English
Subjects:
Channels
Magnetic fields
Nuclear engineering
Nuclear power generation
Plasma density
Spreading
Texts
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An experimental database of the power spreading into the target private flux region (S-factor) from ASDEX Upgrade and JET outer target data has been gathered and analysed. The power spreading, combined with the Scrape-Off Layer heat flux decay length λq, determines the power exhaust channel width at the divertor target. In terms of main plasma parameters, we find that the poloidal magnetic field plays a major role S∼Bpol-1 in both L-mode and H-mode plasmas, similarly to λq. Combining JET and AUG data in H-mode with open divertor we obtain a strong and beneficial major radius dependency and no plasma density influence is found. For closed AUG divII configurations in L-mode, however, larger upstream ne clearly broadens S. Interestingly, the scaling of S with divertor parameters, reveals that only target Te is needed to account for its variation. SOLPS simulations with fixed perpendicular transport reproduce most of the observed phenomenology, suggesting that the observed larger S in closed versus open divertor configuration is simply due to different target Te generated by larger parallel gradients.
ISSN:0022-3115
1873-4820
DOI:10.1016/j.jnucmat.2014.11.076