Performance assessment of tightly-baffled long-leg divertor geometries in the ARC reactor concept M. Wigram1, B. LaBombard2, M.V. Umansky3

A. Q. Kuang2, T. Golfinopoulos2, D. Brunner2, J.L. Terry2, M.E. Rensink3, D.G. Whyte2, A. Hubbard2

ARC is a compact demonstration fusion power plant design, producing ITER levels of fusion power (525MW) at a size comparable to JET (major radius R ~ 3m).

This is facilitated by the very favourable B scaling (Pfus/Swall RB4) and by using REBCO superconducting TF coils to provide the high magnetic field (B0 = 9.2T).[1] The HTS magnets support resistive joints/demountable coils, enabling a modular inner vessel design.[2,3]

ARC (Affordable, Robust, Compact) Reactor

References

[1] Sorbom, B. et al, Fusion Engineering and Design, 100, 2015, pp378-405. [2] Hartwig, Z. et al, Fusion Engineering and Design, 87 (3), 2012, p201. [3] Barnard, H. et al, Fusion Engineering and Design, 87 (3), 2012, pp248-262. [4] Kuang, A.Q., et al., Fusion Engineering and Design, 137, 2018, p221. [5] Eich, T. et al, Nuclear Fusion, 53 (9), 2013, 093031. [6] Umansky, M. et al, Physics of Plasmas, 24, 2017, 056112.

UEDGE ARC SOL Model

Further work This is ongoing research, with further study intending to:

• Implement smaller X-point separations of ~1 λq - to maximise XPTD power handling performance.

• Perform detachment power threshold sensitivity to model parameters: (1) divertor leg transport coefficients D and Vconv compared to main chamber SOL, (2) HFS:LFS power sharing, (3) density at the LCFS.

• Implement impurity model for XPTD power scans.

SXD produces stable detachment up to 108MW

Modelled half-domain Super-X Divertor (SXD) configuration with fixed-fraction 0.5% Ne impurity seeding. Scanned over exhaust power PSOL. Bifurcation of solutions (Fig 5): • Cold branch - only accessible by

maintaining a detached solution. Detachment window PSOL = 80-108 MW.

• Hot branch - accessible approaching from a high plate-Te, attached solution. No detachment.

Transport model:

• Parallel transport: flux-limited Braginskii fluid equations.

• Radial transport: Diffusive thermal transport, particle transport given by:

• Values of D, vconv and χe,i (Fig. 3) tuned to produce expected ARC I-mode midplane profiles (Figs. 4).

Boundary conditions:

• Fixed core input power/density. • Extrapolated wall boundaries

Te,i, ni. • Plasma sheath target plates. • 100% recycling on target/wall

boundaries (steady-state operation).

X-point Target Divertor implemented in ARC [4] ARC divertor challenge: reactor-scale power in compact machine.

• ARC divertor has to exhaust 105 MW of power (assuming ~30% core radiation).

• Eich scaling: λq ~ 1/Bpol = 0.4mm - High, intense divertor power loading.[5]

• Compact design - smaller surface area to dissipate exhaust plasma loading.

• Advanced divertor geometry: double-null, long-legged, secondary X-point target divertor (XPTD).

Simulations of the ADX divertor test tokamak show improved performance of long-legged geometries, in particular the X-point target.[6] Aim of study: model the X-point target divertor design for ARC in UEDGE, assess power handling performance and determine operational windows for detached regimes.

1York Plasma Institute, Department of Physics, University of York, Heslington, York, YO10 5DD, UK. 2MIT PSFC, Cambridge, MA 02139, USA. 3LLNL, Livermore, CA 94550, USA

XPTD attains detachment at 80MW without seeding

Figure 2: Proposed ARC X-point target divertor geometry, showing closed (blue) and open SOL (green) magnetic flux surfaces.[4]

Figure 5: Peak outer target Te vs exhaust power PSOL (full domain).

Figure 6: Te plot for detached P=105MW case, with combined plasma/radiation peak power flux densities to boundaries.

Figure 4: Midplane ni and Te,i profiles produced for ARC I-mode model. λn~ 5.5mm, λq~ 0.4mm.

Figure 3: vconv and D/χe,i profiles. Transport barrier implemented by drop in χe,i around separatrix.

Figure 1: ARC reactor CAD model cross-section.[1]

Fusion power:

525MW

TF coils:

B0=9.2T

Vacuum vessel: single,

replaceable component

Magnetic joints allow

vertical maintenance.

Figure 7: ARC XPTD grid in UEDGE (0.57mm X-point separation). Closest approach between divertor X-point and primary X-point flux surface ~0.1m.

Impurities: Fixed-fraction impurity model, specified % of local electron density. Neutrals: Fluid-diffusive model.

Figure 8: Peak outer target Te vs exhaust power PSOL, for SXD (red) and XPTD grids with 1.6mm (green), 0.9 (purple) and 0.57mm (blue) X-point separations, for “downswing” (left) and “upswing” (right) power scans.

Modelled XPTD configuration. Performed exhaust power scans without Ne impurity for both SXD and XPTD, with primary and divertor X-point separations of 1.6mm, 0.9mm and 0.57mm (mapped to midplane). Significant performance benefits observed for XPTD over SXD, but only when X-points are closely mapped. For 0.57mm X-point separation, the detachment power threshold increases up to 74MW for a power scan with reducing PSOL. Starting from low-power, detached state and increasing PSOL, threshold increases further to 80MW—approaching 105MW target. Relating to λq, results indicate the X-point separation needs to be within 1-2 λq for significant performance benefit to be realised.

Acknowledgements: This work has been supported by the Uni. of York, MIT PSFC (US DoE coop. agreement DE-SC0014264), LLNL (DoE Contract DE-AC52-07NA27344) and the UK EPSRC (Training Grant Number EP/LO1663X/1).

Overview ARC SOL/divertor model implemented in UEDGE for Super-X Divertor

(SXD) and X-Point Target Divertor (XPTD) setups.

Power scans performed over PSOL for SXD/XPTD with/without 0.5% Ne impurity.

Significant performance benefit observed for the XPTD geometry over SXD, but only when radial separation of X-point flux surfaces are small (< 2 λq).

Divertor detachment achieved up to 80 MW w/o impurity seeding with small X-point separations (Fig. 8), factor ~2 greater performance than SXD.