Implementation of an inelastic collision operator into KIPP‐SOLPS coupling and its effects on electron parallel transport in the scrape‐off layer plasmas

Abstract

This paper develops an inelastic collision operator for the Kinetic Code for Plasma Periphery (KIPP) code to investigate the kinetic effects of electron cooling due to inelastic collisions. It is fully tested based on the self-consistent KIPP-SOLPS coupling algorithm by being compared to the ADAS database. The collisional radiative rate coefficients from the ADAS database for deuterium atomic physics can be recovered using the inelastic collision operator with assuming Maxwellian electrons, which shows that the inelastic collision operator works well for various plasma conditions. Across a wide range of plasma conditions in the scrape-off layer, KIPP-SOLPS coupling simulation results with the implementation of an inelastic collision operator are not significantly different from results using a simpler uniform cooling scheme. The uniform scheme is thus recommended rather than including computationally intensive inelastic collision physics. KEYWORDSedge plasmas, fluid, inelastic collision operator, kinetic INTRODUCTIONPower exhaust is one of the critical issues of future fusion devices. Currently, two-dimensional (2D) fluid codes solving Braginskii-like equations [1] are utilized to investigate tokamak edge plasmas, like SOLPS, [2] EDGE2D, [3] and UEDGE. [4] However, the validity of the fluid model is often violated by electron parallel non-local transport [5][6][7] along magnetic field lines in the scrape-off layer. Non-Maxwellian tails of electron distribution functions due to the non-local transport greatly influence electron-related transport coefficients in fluid equations, which are obtained by assuming collisional plasmas. The Kinetic Code for Plasma Periphery (KIPP), [8][9][10][11] which is a continuum kinetic code solving the Vlasov-Fokker-Planck equation [12] with high accuracy of collision terms for electron parallel transport, was coupled with SOLPS based on an iterative coupling algorithm, [13] briefly described in Section 2. The KIPP-SOLPS coupling algorithm enables one to treat electron parallel transport fully kinetically while still retaining all the physics that SOLPS has, which would be time consuming to capture using a kinetic code, and is accurately captured by the fluid code. In previous studies, [13][14][15] the KIPP-SOLPS coupling algorithm was extensively investigated. The atomic physics was treated in SOLPS code, and the corresponding electron cooling due to ionization, line radiation, and recombination-bremsstrahlung radiation was included in KIPP by a uniform power removal scheme, as described in Section 2, which was, however, not realistic, especially for plasmas downstream near the target. Previous kinetic simulations [5,10,[16][17][18] showed that the electron conductive heat flux was found to be carried mostly by the so-called heat-carrying electrons (HCEs), which have a kinetic energy of ∼5.95T e . A reminder that the ionization threshold for deuterium neutrals is 13.6 eV, HCE could be mainly responsible for the deuterium ionization in low temperature regions,

Kinetic effects in parallel electron energy transport channels in the scrape-off layer

“…The fluid case of the subchannel decomposition (not pictured here) for the background in figure 12 does not differ greatly compared to the kinetic case, supporting the finding by Zhao et al [26]. that kinetic effects on atomic rates in equilibria are negligible.…”

Scaling laws for electron kinetic effects in tokamak scrape-off layer plasmas

“…Finally, it is worth commenting that the findings in [12] and [5], that electron-neutral reaction rates are well-approximated by Maxwellian-averaged rates, is replicated in this study. Differences in rate coefficients for deuterium ionisation and line radiation are negligible.…”

“…As and shown in [12] and [5], electron-impact ionisation rates of hydrogen are very well approximated by a Maxwellian distribution. This is unsurprising when considering the distribution shown in figure 5, which is non-Maxwellian in the tail but very close to Maxwellian in the thermal bulk.…”

“…Conditions where ν * u is small and temperature gradients are large may not be described accurately by a fluid model. This effect has been explored in recent years [3][4][5][6][7][8][9][10][11], where it is now well-documented that kinetic suppression of the heat flux can result in steeper temperature gradients and lower target temperatures when compared to a fluid model. Somewhat less understood is the region of operating parameter space where such effects may become important, and the consequences for the overall energy balance at equilibrium (i.e.…”

Ion temperature anisotropy in the tokamak scrape-off layer

“…The parallel electron thermal diffusivity dominates that of ions because for equal temperatures, the electron and ion mean-free paths, λ e,i , are similar, whereas their kinematic thermal diffusivities scale as λ 2 e,i ν e,i , yielding a much higher thermal diffusivity for electrons than ions because ν e /ν i ∼ (m i /m e ) 1/2 , where m e and m i are the electron and ion masses, respectively. Electron kinetic effects on parallel electron conductive heat flux and thermal force in the SOL plasmas have been extensively studied previously in [12][13][14][15][16][17][18][19][20][21][22].…”