Study of plasma current effect on divertor power footprint widths through experiments and modeling in EAST L-mode plasmas

Abstract

Integrated studies of the scaling of divertor power footprint widths with plasma current in Experimental Advanced Superconducting Tokamak (EAST) L-mode plasmas are carried out through experiments and modeling. The divertor power widths, which consist of the scrape-off layer power decay length λq and the heat spreading S, are calculated from the measurements of divertor Langmuir probes (div-LPs) and scrape-off layer plasma simulation (SOLPS). A strong inverse scaling of both λq (mm) and S (mm) from the div-LP with plasma current Ip (MA) is demonstrated, with the regressions being λq=4.37Ip−1.04±0.075 and S=2.31Ip−0.997±0.047. However, a much weaker inverse scaling of λq with Ip is shown by the measurement of outer mid-plane reciprocating probes, which gives λq,RP=6.56Ip−0.31±0.056. It indicates that the connection length may have played a stronger role in the negative scaling of λq with Ip than radial transport. A similar conclusion can also be drawn from a comparison of the simulation of the plasma current scan with a set of fixed radial transport and another set of coefficient modified by Ip. This indicates that plasma configuration with a longer connection length is a better L-mode operational scenario for diverted tokamaks.

Statistical study of particle flux footprint widths with tungsten divertor in EAST

“…On the theory side, different models [13][14][15] and simulation codes [16][17][18][19][20][21][22] have been derived/employed to understand the underlying physics of the λq scaling. The heuristic driftbased model proposed by Goldston [13] is very successful to explain the Eich scaling and gives insight on the scaling difference of λq between the inner and outer divertors [12,23].…”

Study of power width scaling in scrape-off layer with 2D electrostatic turbulence code based on EAST L-mode discharges

“…(19) in the scan where C neoc,corr is fixed, meaning that q 95 influences the SOL power width mostly through the parallel heat transports. Since q 95 / B À1 p , this result indicates that the scaling dependence on the poloidal magnetic field (or the plasma current) for the SOL power width scalings is essentially the scaling dependence on the ballooning length, which might explain why simulation codes 17,19,21,22 with different radial heat transports but similar formulation of the parallel heat transports can reproduce the inverse scaling dependence on the poloidal magnetic field. In this sense, the numerical scaling that is dependent on the ballooning length is proposed as [see also Fig.…”

“…19 However, the code predicts a much larger k q for ITER (k q,ITER ¼ 5.9 mm), which is explained to be that the divertor heat flux is dominated by the electron blobby transport in ITER and radial blobby turbulence spreads further because of the wider E Â B shear flow layer as a consequence of the larger major radius. 19 Some of the studies 13,20,21 employ the transport code, SOLPS, to investigate k q , and the study of EAST L-mode plasmas suggests that the parallel heat transports in the SOL contribute more to the inverse scaling dependence on the poloidal magnetic field. 21 Besides these 3D codes, the 2D turbulence code, HESEL (hot edge-SOL electrostatic), has also been utilized to reproduce the AUG L-mode scaling [no temperature dependence, see Eq.…”

“…19 Some of the studies 13,20,21 employ the transport code, SOLPS, to investigate k q , and the study of EAST L-mode plasmas suggests that the parallel heat transports in the SOL contribute more to the inverse scaling dependence on the poloidal magnetic field. 21 Besides these 3D codes, the 2D turbulence code, HESEL (hot edge-SOL electrostatic), has also been utilized to reproduce the AUG L-mode scaling [no temperature dependence, see Eq. (15) 7,22 Benefiting from the first-principles perpendicular transports and the low computation time of the HESEL model, a numerical scaling that is dependent on the safety factor at the LCFS q 95 , the electron density at the last closed flux surface (LCFS) n e,LCFS , and the electron temperature at the LCFS T e,LCFS has been proposed for L-mode plasmas 22 k q ¼ 4:7460:92 ð Þ q 0:8660:05…”

Modeling study of divertor particle and heat flux asymmetries for EAST H-mode discharges

“…2 . The coefficients at the pedestal region are set to be small due to the existence of the transport barrier there, which is a common practice in edge plasma simulations 30 , 38 , 39 . The ion thermal diffusivity is set to be the same as the electron thermal diffusivity for simplicity.…”